From 4d28d3f32e7f213e300b24bc61c3f0ac9d6e1ab6 Mon Sep 17 00:00:00 2001
From: Alon Zakai <azakai@google.com>
Date: Thu, 19 Dec 2019 09:04:08 -0800
Subject: DWARF parsing and writing support using LLVM (#2520)
This imports LLVM code for DWARF handling. That code has the
Apache 2 license like us. It's also the same code used to
emit DWARF in the common toolchain, so it seems like a safe choice.
This adds two passes: --dwarfdump which runs the same code LLVM
runs for llvm-dwarfdump. This shows we can parse it ok, and will
be useful for debugging. And --dwarfupdate writes out the DWARF
sections (unchanged from what we read, so it just roundtrips - for
updating we need #2515).
This puts LLVM in thirdparty which is added here.
All the LLVM code is behind USE_LLVM_DWARF, which is on
by default, but off in JS for now, as it increases code size by 20%.
This current approach imports the LLVM files directly. This is not
how they are intended to be used, so it required a bunch of
local changes - more than I expected actually, for the platform-specific
stuff. For now this seems to work, so it may be good enough, but
in the long term we may want to switch to linking against libllvm.
A downside to doing that is that binaryen users would need to
have an LLVM build, and even in the waterfall builds we'd have a
problem - while we ship LLVM there anyhow, we constantly update
it, which means that binaryen would need to be on latest llvm all
the time too (which otherwise, given DWARF is quite stable, we
might not need to constantly update).
An even larger issue is that as I did this work I learned about how
DWARF works in LLVM, and while the reading code is easy to
reuse, the writing code is trickier. The main code path is heavily
integrated with the MC layer, which we don't have - we might want
to create a "fake MC layer" for that, but it sounds hard. Instead,
there is the YAML path which is used mostly for testing, and which
can convert DWARF to and from YAML and from binary. Using
the non-YAML parts there, we can convert binary DWARF to
the YAML layer's nice Info data, then convert that to binary. This
works, however, this is not the path LLVM uses normally, and it
supports only some basic DWARF sections - I had to add ranges
support, in fact. So if we need more complex things, we may end
up needing to use the MC layer approach, or consider some other
DWARF library. However, hopefully that should not affect the core
binaryen code which just calls a library for DWARF stuff.
Helps #2400
---
third_party/.clang-format | 5 +
third_party/.clang-tidy | 1 +
third_party/CMakeLists.txt | 3 +
third_party/llvm-project/Binary.cpp | 72 +
third_party/llvm-project/CMakeLists.txt | 76 +
third_party/llvm-project/ConvertUTF.cpp | 738 ++++++
third_party/llvm-project/DJB.cpp | 82 +
.../llvm-project/DWARFAbbreviationDeclaration.cpp | 215 ++
third_party/llvm-project/DWARFAcceleratorTable.cpp | 889 ++++++++
third_party/llvm-project/DWARFAddressRange.cpp | 28 +
third_party/llvm-project/DWARFCompileUnit.cpp | 38 +
third_party/llvm-project/DWARFContext.cpp | 1881 +++++++++++++++
third_party/llvm-project/DWARFDataExtractor.cpp | 100 +
third_party/llvm-project/DWARFDebugAbbrev.cpp | 138 ++
third_party/llvm-project/DWARFDebugAddr.cpp | 182 ++
third_party/llvm-project/DWARFDebugArangeSet.cpp | 111 +
third_party/llvm-project/DWARFDebugAranges.cpp | 122 +
third_party/llvm-project/DWARFDebugFrame.cpp | 554 +++++
third_party/llvm-project/DWARFDebugInfoEntry.cpp | 69 +
third_party/llvm-project/DWARFDebugLine.cpp | 1181 ++++++++++
third_party/llvm-project/DWARFDebugLoc.cpp | 326 +++
third_party/llvm-project/DWARFDebugMacro.cpp | 106 +
third_party/llvm-project/DWARFDebugPubTable.cpp | 69 +
third_party/llvm-project/DWARFDebugRangeList.cpp | 95 +
third_party/llvm-project/DWARFDebugRnglists.cpp | 245 ++
third_party/llvm-project/DWARFDie.cpp | 747 ++++++
third_party/llvm-project/DWARFEmitter.cpp | 419 ++++
third_party/llvm-project/DWARFExpression.cpp | 344 +++
third_party/llvm-project/DWARFFormValue.cpp | 720 ++++++
third_party/llvm-project/DWARFGdbIndex.cpp | 199 ++
third_party/llvm-project/DWARFListTable.cpp | 116 +
third_party/llvm-project/DWARFTypeUnit.cpp | 49 +
third_party/llvm-project/DWARFUnit.cpp | 904 ++++++++
third_party/llvm-project/DWARFUnitIndex.cpp | 200 ++
third_party/llvm-project/DWARFVerifier.cpp | 1493 ++++++++++++
third_party/llvm-project/DWARFVisitor.cpp | 177 ++
third_party/llvm-project/DWARFVisitor.h | 96 +
third_party/llvm-project/DWARFYAML.cpp | 175 ++
third_party/llvm-project/DataExtractor.cpp | 218 ++
third_party/llvm-project/Debug.cpp | 157 ++
third_party/llvm-project/Dwarf.cpp | 759 ++++++
third_party/llvm-project/Error.cpp | 168 ++
third_party/llvm-project/Error.h | 53 +
third_party/llvm-project/ErrorHandling.cpp | 305 +++
third_party/llvm-project/FormatVariadic.cpp | 155 ++
third_party/llvm-project/Hashing.cpp | 28 +
third_party/llvm-project/LEB128.cpp | 43 +
third_party/llvm-project/LineIterator.cpp | 93 +
third_party/llvm-project/MCRegisterInfo.cpp | 124 +
third_party/llvm-project/MD5.cpp | 283 +++
third_party/llvm-project/MemoryBuffer.cpp | 355 +++
third_party/llvm-project/NativeFormatting.cpp | 263 +++
third_party/llvm-project/ObjectFile.cpp | 112 +
third_party/llvm-project/Optional.cpp | 14 +
third_party/llvm-project/Path.cpp | 1262 ++++++++++
third_party/llvm-project/ScopedPrinter.cpp | 46 +
third_party/llvm-project/SmallVector.cpp | 65 +
third_party/llvm-project/SourceMgr.cpp | 503 ++++
third_party/llvm-project/StringMap.cpp | 261 +++
third_party/llvm-project/StringRef.cpp | 519 +++++
third_party/llvm-project/SymbolicFile.cpp | 45 +
third_party/llvm-project/Twine.cpp | 184 ++
third_party/llvm-project/UnicodeCaseFold.cpp | 742 ++++++
third_party/llvm-project/Unix/Path.inc | 1213 ++++++++++
third_party/llvm-project/Unix/Unix.h | 114 +
third_party/llvm-project/WithColor.cpp | 83 +
third_party/llvm-project/YAMLParser.cpp | 2408 ++++++++++++++++++++
third_party/llvm-project/YAMLTraits.cpp | 1087 +++++++++
third_party/llvm-project/dwarf2yaml.cpp | 385 ++++
.../llvm-project/include/llvm-c/DataTypes.h | 90 +
.../include/llvm-c/DisassemblerTypes.h | 160 ++
third_party/llvm-project/include/llvm-c/Error.h | 69 +
.../llvm-project/include/llvm-c/ErrorHandling.h | 49 +
third_party/llvm-project/include/llvm-c/Types.h | 179 ++
.../llvm-project/include/llvm/ADT/APFloat.h | 1290 +++++++++++
third_party/llvm-project/include/llvm/ADT/APInt.h | 2258 ++++++++++++++++++
third_party/llvm-project/include/llvm/ADT/APSInt.h | 353 +++
.../llvm-project/include/llvm/ADT/AllocatorList.h | 240 ++
.../llvm-project/include/llvm/ADT/ArrayRef.h | 540 +++++
.../llvm-project/include/llvm/ADT/BitmaskEnum.h | 152 ++
.../llvm-project/include/llvm/ADT/DenseMap.h | 1275 +++++++++++
.../llvm-project/include/llvm/ADT/DenseMapInfo.h | 300 +++
.../llvm-project/include/llvm/ADT/DenseSet.h | 283 +++
.../llvm-project/include/llvm/ADT/EpochTracker.h | 98 +
.../llvm-project/include/llvm/ADT/FoldingSet.h | 761 +++++++
.../llvm-project/include/llvm/ADT/FunctionExtras.h | 292 +++
.../llvm-project/include/llvm/ADT/Hashing.h | 659 ++++++
.../llvm-project/include/llvm/ADT/MapVector.h | 239 ++
third_party/llvm-project/include/llvm/ADT/None.h | 26 +
.../llvm-project/include/llvm/ADT/Optional.h | 429 ++++
.../llvm-project/include/llvm/ADT/PointerIntPair.h | 243 ++
.../llvm-project/include/llvm/ADT/PointerUnion.h | 309 +++
.../llvm-project/include/llvm/ADT/STLExtras.h | 1578 +++++++++++++
.../llvm-project/include/llvm/ADT/SmallPtrSet.h | 511 +++++
.../llvm-project/include/llvm/ADT/SmallSet.h | 278 +++
.../llvm-project/include/llvm/ADT/SmallString.h | 296 +++
.../llvm-project/include/llvm/ADT/SmallVector.h | 930 ++++++++
.../llvm-project/include/llvm/ADT/StringExtras.h | 401 ++++
.../llvm-project/include/llvm/ADT/StringMap.h | 593 +++++
.../llvm-project/include/llvm/ADT/StringRef.h | 920 ++++++++
.../llvm-project/include/llvm/ADT/StringSet.h | 58 +
.../llvm-project/include/llvm/ADT/StringSwitch.h | 196 ++
third_party/llvm-project/include/llvm/ADT/Triple.h | 881 +++++++
third_party/llvm-project/include/llvm/ADT/Twine.h | 544 +++++
third_party/llvm-project/include/llvm/ADT/bit.h | 58 +
.../llvm-project/include/llvm/ADT/edit_distance.h | 102 +
.../include/llvm/ADT/fallible_iterator.h | 243 ++
.../llvm-project/include/llvm/ADT/ilist_base.h | 92 +
.../llvm-project/include/llvm/ADT/ilist_iterator.h | 198 ++
.../llvm-project/include/llvm/ADT/ilist_node.h | 305 +++
.../include/llvm/ADT/ilist_node_base.h | 52 +
.../include/llvm/ADT/ilist_node_options.h | 131 ++
.../llvm-project/include/llvm/ADT/iterator.h | 366 +++
.../llvm-project/include/llvm/ADT/iterator_range.h | 69 +
.../llvm-project/include/llvm/ADT/simple_ilist.h | 314 +++
.../llvm-project/include/llvm/BinaryFormat/COFF.h | 729 ++++++
.../include/llvm/BinaryFormat/Dwarf.def | 964 ++++++++
.../llvm-project/include/llvm/BinaryFormat/Dwarf.h | 677 ++++++
.../include/llvm/BinaryFormat/DynamicTags.def | 244 ++
.../llvm-project/include/llvm/BinaryFormat/ELF.h | 1573 +++++++++++++
.../llvm/BinaryFormat/ELFRelocs/AArch64.def | 221 ++
.../include/llvm/BinaryFormat/ELFRelocs/AMDGPU.def | 17 +
.../include/llvm/BinaryFormat/ELFRelocs/ARC.def | 74 +
.../include/llvm/BinaryFormat/ELFRelocs/ARM.def | 141 ++
.../include/llvm/BinaryFormat/ELFRelocs/AVR.def | 41 +
.../include/llvm/BinaryFormat/ELFRelocs/BPF.def | 8 +
.../llvm/BinaryFormat/ELFRelocs/Hexagon.def | 106 +
.../include/llvm/BinaryFormat/ELFRelocs/Lanai.def | 19 +
.../include/llvm/BinaryFormat/ELFRelocs/MSP430.def | 16 +
.../include/llvm/BinaryFormat/ELFRelocs/Mips.def | 117 +
.../llvm/BinaryFormat/ELFRelocs/PowerPC.def | 156 ++
.../llvm/BinaryFormat/ELFRelocs/PowerPC64.def | 195 ++
.../include/llvm/BinaryFormat/ELFRelocs/RISCV.def | 59 +
.../include/llvm/BinaryFormat/ELFRelocs/Sparc.def | 89 +
.../llvm/BinaryFormat/ELFRelocs/SystemZ.def | 71 +
.../include/llvm/BinaryFormat/ELFRelocs/i386.def | 47 +
.../include/llvm/BinaryFormat/ELFRelocs/x86_64.def | 45 +
.../include/llvm/BinaryFormat/MachO.def | 119 +
.../llvm-project/include/llvm/BinaryFormat/MachO.h | 2006 ++++++++++++++++
.../llvm-project/include/llvm/BinaryFormat/Magic.h | 77 +
.../llvm-project/include/llvm/BinaryFormat/Wasm.h | 391 ++++
.../include/llvm/BinaryFormat/WasmRelocs.def | 17 +
.../include/llvm/Config/abi-breaking.h | 3 +
.../llvm-project/include/llvm/Config/config.h | 4 +
.../llvm-project/include/llvm/Config/llvm-config.h | 9 +
.../include/llvm/DebugInfo/DIContext.h | 302 +++
.../DebugInfo/DWARF/DWARFAbbreviationDeclaration.h | 183 ++
.../llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h | 599 +++++
.../llvm/DebugInfo/DWARF/DWARFAddressRange.h | 61 +
.../include/llvm/DebugInfo/DWARF/DWARFAttribute.h | 49 +
.../llvm/DebugInfo/DWARF/DWARFCompileUnit.h | 38 +
.../include/llvm/DebugInfo/DWARF/DWARFContext.h | 382 ++++
.../llvm/DebugInfo/DWARF/DWARFDataExtractor.h | 64 +
.../llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h | 87 +
.../include/llvm/DebugInfo/DWARF/DWARFDebugAddr.h | 97 +
.../llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h | 75 +
.../llvm/DebugInfo/DWARF/DWARFDebugAranges.h | 84 +
.../include/llvm/DebugInfo/DWARF/DWARFDebugFrame.h | 308 +++
.../llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h | 62 +
.../include/llvm/DebugInfo/DWARF/DWARFDebugLine.h | 392 ++++
.../include/llvm/DebugInfo/DWARF/DWARFDebugLoc.h | 121 +
.../include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h | 62 +
.../llvm/DebugInfo/DWARF/DWARFDebugPubTable.h | 80 +
.../llvm/DebugInfo/DWARF/DWARFDebugRangeList.h | 83 +
.../llvm/DebugInfo/DWARF/DWARFDebugRnglists.h | 64 +
.../include/llvm/DebugInfo/DWARF/DWARFDie.h | 468 ++++
.../include/llvm/DebugInfo/DWARF/DWARFExpression.h | 159 ++
.../include/llvm/DebugInfo/DWARF/DWARFFormValue.h | 321 +++
.../include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h | 82 +
.../include/llvm/DebugInfo/DWARF/DWARFListTable.h | 290 +++
.../include/llvm/DebugInfo/DWARF/DWARFObject.h | 85 +
.../include/llvm/DebugInfo/DWARF/DWARFRelocMap.h | 38 +
.../include/llvm/DebugInfo/DWARF/DWARFSection.h | 27 +
.../include/llvm/DebugInfo/DWARF/DWARFTypeUnit.h | 46 +
.../include/llvm/DebugInfo/DWARF/DWARFUnit.h | 527 +++++
.../include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h | 105 +
.../include/llvm/DebugInfo/DWARF/DWARFVerifier.h | 336 +++
third_party/llvm-project/include/llvm/LICENSE.TXT | 279 +++
.../llvm-project/include/llvm/MC/LaneBitmask.h | 101 +
third_party/llvm-project/include/llvm/MC/MCExpr.h | 616 +++++
third_party/llvm-project/include/llvm/MC/MCFixup.h | 202 ++
.../llvm-project/include/llvm/MC/MCFragment.h | 663 ++++++
third_party/llvm-project/include/llvm/MC/MCInst.h | 225 ++
.../llvm-project/include/llvm/MC/MCRegister.h | 110 +
.../llvm-project/include/llvm/MC/MCRegisterInfo.h | 721 ++++++
.../llvm-project/include/llvm/MC/MCSymbol.h | 441 ++++
.../llvm-project/include/llvm/MC/MCSymbolWasm.h | 110 +
.../include/llvm/MC/SubtargetFeature.h | 236 ++
.../llvm-project/include/llvm/Object/Archive.h | 286 +++
.../llvm-project/include/llvm/Object/Binary.h | 237 ++
.../llvm-project/include/llvm/Object/COFF.h | 1269 +++++++++++
.../include/llvm/Object/CVDebugRecord.h | 54 +
.../include/llvm/Object/Decompressor.h | 66 +
third_party/llvm-project/include/llvm/Object/ELF.h | 723 ++++++
.../include/llvm/Object/ELFObjectFile.h | 1214 ++++++++++
.../llvm-project/include/llvm/Object/ELFTypes.h | 763 +++++++
.../llvm-project/include/llvm/Object/Error.h | 92 +
.../llvm-project/include/llvm/Object/MachO.h | 737 ++++++
.../llvm-project/include/llvm/Object/Minidump.h | 216 ++
.../llvm-project/include/llvm/Object/ObjectFile.h | 579 +++++
.../include/llvm/Object/RelocationResolver.h | 42 +
.../include/llvm/Object/SymbolicFile.h | 214 ++
.../llvm-project/include/llvm/Object/Wasm.h | 356 +++
.../include/llvm/ObjectYAML/DWARFEmitter.h | 50 +
.../include/llvm/ObjectYAML/DWARFYAML.h | 320 +++
.../include/llvm/ObjectYAML/ObjectYAML.h | 41 +
.../include/llvm/Support/AArch64TargetParser.def | 151 ++
.../include/llvm/Support/AArch64TargetParser.h | 129 ++
.../include/llvm/Support/ARMAttributeParser.h | 141 ++
.../include/llvm/Support/ARMBuildAttributes.h | 253 ++
.../include/llvm/Support/ARMTargetParser.def | 293 +++
.../include/llvm/Support/ARMTargetParser.h | 267 +++
.../llvm-project/include/llvm/Support/AlignOf.h | 55 +
.../llvm-project/include/llvm/Support/Alignment.h | 403 ++++
.../llvm-project/include/llvm/Support/Allocator.h | 523 +++++
.../include/llvm/Support/BinaryByteStream.h | 273 +++
.../include/llvm/Support/BinaryStream.h | 101 +
.../include/llvm/Support/BinaryStreamError.h | 47 +
.../include/llvm/Support/CBindingWrapping.h | 46 +
.../llvm-project/include/llvm/Support/Casting.h | 408 ++++
.../llvm-project/include/llvm/Support/Chrono.h | 171 ++
.../llvm-project/include/llvm/Support/CodeGen.h | 67 +
.../include/llvm/Support/CommandLine.h | 1 +
.../llvm-project/include/llvm/Support/Compiler.h | 585 +++++
.../llvm-project/include/llvm/Support/ConvertUTF.h | 306 +++
.../llvm-project/include/llvm/Support/DJB.h | 32 +
.../include/llvm/Support/DataExtractor.h | 575 +++++
.../llvm-project/include/llvm/Support/DataTypes.h | 16 +
.../llvm-project/include/llvm/Support/Debug.h | 126 +
.../llvm-project/include/llvm/Support/Endian.h | 429 ++++
.../llvm-project/include/llvm/Support/Errc.h | 86 +
.../llvm-project/include/llvm/Support/Errno.h | 46 +
.../llvm-project/include/llvm/Support/Error.h | 1351 +++++++++++
.../include/llvm/Support/ErrorHandling.h | 143 ++
.../llvm-project/include/llvm/Support/ErrorOr.h | 278 +++
.../include/llvm/Support/FileOutputBuffer.h | 88 +
.../llvm-project/include/llvm/Support/FileSystem.h | 1444 ++++++++++++
.../llvm-project/include/llvm/Support/Format.h | 257 +++
.../include/llvm/Support/FormatAdapters.h | 108 +
.../include/llvm/Support/FormatCommon.h | 76 +
.../include/llvm/Support/FormatProviders.h | 422 ++++
.../include/llvm/Support/FormatVariadic.h | 264 +++
.../include/llvm/Support/FormatVariadicDetails.h | 164 ++
.../include/llvm/Support/FormattedStream.h | 161 ++
.../llvm-project/include/llvm/Support/Host.h | 70 +
.../llvm-project/include/llvm/Support/LEB128.h | 198 ++
.../llvm-project/include/llvm/Support/LICENSE.TXT | 6 +
.../include/llvm/Support/LineIterator.h | 87 +
.../llvm-project/include/llvm/Support/Locale.h | 17 +
.../llvm-project/include/llvm/Support/MD5.h | 122 +
.../include/llvm/Support/ManagedStatic.h | 1 +
.../llvm-project/include/llvm/Support/MathExtras.h | 951 ++++++++
.../llvm-project/include/llvm/Support/MemAlloc.h | 66 +
.../include/llvm/Support/MemoryBuffer.h | 286 +++
.../include/llvm/Support/NativeFormatting.h | 48 +
.../llvm-project/include/llvm/Support/Path.h | 464 ++++
.../include/llvm/Support/PointerLikeTypeTraits.h | 149 ++
.../llvm-project/include/llvm/Support/Printable.h | 51 +
.../llvm-project/include/llvm/Support/Process.h | 209 ++
.../llvm-project/include/llvm/Support/Program.h | 208 ++
.../llvm-project/include/llvm/Support/Regex.h | 109 +
.../include/llvm/Support/ReverseIteration.h | 19 +
.../llvm-project/include/llvm/Support/SMLoc.h | 64 +
.../include/llvm/Support/ScopedPrinter.h | 388 ++++
.../llvm-project/include/llvm/Support/Signals.h | 90 +
.../include/llvm/Support/SmallVectorMemoryBuffer.h | 65 +
.../llvm-project/include/llvm/Support/SourceMgr.h | 310 +++
.../include/llvm/Support/SwapByteOrder.h | 160 ++
.../include/llvm/Support/TargetParser.h | 174 ++
.../include/llvm/Support/TargetRegistry.h | 1209 ++++++++++
.../llvm-project/include/llvm/Support/Threading.h | 0
.../llvm-project/include/llvm/Support/TypeSize.h | 201 ++
.../llvm-project/include/llvm/Support/Unicode.h | 70 +
.../include/llvm/Support/UnicodeCharRanges.h | 103 +
.../include/llvm/Support/WindowsError.h | 18 +
.../llvm-project/include/llvm/Support/WithColor.h | 118 +
.../include/llvm/Support/X86TargetParser.def | 173 ++
.../llvm-project/include/llvm/Support/YAMLParser.h | 619 +++++
.../llvm-project/include/llvm/Support/YAMLTraits.h | 2043 +++++++++++++++++
.../include/llvm/Support/circular_raw_ostream.h | 159 ++
.../include/llvm/Support/raw_ostream.h | 587 +++++
.../include/llvm/Support/type_traits.h | 192 ++
.../llvm/include/llvm/DebugInfo/DWARFContext.h | 382 ++++
third_party/llvm-project/include/llvm/readme.txt | 15 +
third_party/llvm-project/obj2yaml_Error.cpp | 61 +
third_party/llvm-project/raw_ostream.cpp | 712 ++++++
third_party/llvm-project/readme.txt | 19 +
287 files changed, 90576 insertions(+)
create mode 100644 third_party/.clang-format
create mode 100644 third_party/.clang-tidy
create mode 100644 third_party/CMakeLists.txt
create mode 100644 third_party/llvm-project/Binary.cpp
create mode 100644 third_party/llvm-project/CMakeLists.txt
create mode 100644 third_party/llvm-project/ConvertUTF.cpp
create mode 100644 third_party/llvm-project/DJB.cpp
create mode 100644 third_party/llvm-project/DWARFAbbreviationDeclaration.cpp
create mode 100644 third_party/llvm-project/DWARFAcceleratorTable.cpp
create mode 100644 third_party/llvm-project/DWARFAddressRange.cpp
create mode 100644 third_party/llvm-project/DWARFCompileUnit.cpp
create mode 100644 third_party/llvm-project/DWARFContext.cpp
create mode 100644 third_party/llvm-project/DWARFDataExtractor.cpp
create mode 100644 third_party/llvm-project/DWARFDebugAbbrev.cpp
create mode 100644 third_party/llvm-project/DWARFDebugAddr.cpp
create mode 100644 third_party/llvm-project/DWARFDebugArangeSet.cpp
create mode 100644 third_party/llvm-project/DWARFDebugAranges.cpp
create mode 100644 third_party/llvm-project/DWARFDebugFrame.cpp
create mode 100644 third_party/llvm-project/DWARFDebugInfoEntry.cpp
create mode 100644 third_party/llvm-project/DWARFDebugLine.cpp
create mode 100644 third_party/llvm-project/DWARFDebugLoc.cpp
create mode 100644 third_party/llvm-project/DWARFDebugMacro.cpp
create mode 100644 third_party/llvm-project/DWARFDebugPubTable.cpp
create mode 100644 third_party/llvm-project/DWARFDebugRangeList.cpp
create mode 100644 third_party/llvm-project/DWARFDebugRnglists.cpp
create mode 100644 third_party/llvm-project/DWARFDie.cpp
create mode 100644 third_party/llvm-project/DWARFEmitter.cpp
create mode 100644 third_party/llvm-project/DWARFExpression.cpp
create mode 100644 third_party/llvm-project/DWARFFormValue.cpp
create mode 100644 third_party/llvm-project/DWARFGdbIndex.cpp
create mode 100644 third_party/llvm-project/DWARFListTable.cpp
create mode 100644 third_party/llvm-project/DWARFTypeUnit.cpp
create mode 100644 third_party/llvm-project/DWARFUnit.cpp
create mode 100644 third_party/llvm-project/DWARFUnitIndex.cpp
create mode 100644 third_party/llvm-project/DWARFVerifier.cpp
create mode 100644 third_party/llvm-project/DWARFVisitor.cpp
create mode 100644 third_party/llvm-project/DWARFVisitor.h
create mode 100644 third_party/llvm-project/DWARFYAML.cpp
create mode 100644 third_party/llvm-project/DataExtractor.cpp
create mode 100644 third_party/llvm-project/Debug.cpp
create mode 100644 third_party/llvm-project/Dwarf.cpp
create mode 100644 third_party/llvm-project/Error.cpp
create mode 100644 third_party/llvm-project/Error.h
create mode 100644 third_party/llvm-project/ErrorHandling.cpp
create mode 100644 third_party/llvm-project/FormatVariadic.cpp
create mode 100644 third_party/llvm-project/Hashing.cpp
create mode 100644 third_party/llvm-project/LEB128.cpp
create mode 100644 third_party/llvm-project/LineIterator.cpp
create mode 100644 third_party/llvm-project/MCRegisterInfo.cpp
create mode 100644 third_party/llvm-project/MD5.cpp
create mode 100644 third_party/llvm-project/MemoryBuffer.cpp
create mode 100644 third_party/llvm-project/NativeFormatting.cpp
create mode 100644 third_party/llvm-project/ObjectFile.cpp
create mode 100644 third_party/llvm-project/Optional.cpp
create mode 100644 third_party/llvm-project/Path.cpp
create mode 100644 third_party/llvm-project/ScopedPrinter.cpp
create mode 100644 third_party/llvm-project/SmallVector.cpp
create mode 100644 third_party/llvm-project/SourceMgr.cpp
create mode 100644 third_party/llvm-project/StringMap.cpp
create mode 100644 third_party/llvm-project/StringRef.cpp
create mode 100644 third_party/llvm-project/SymbolicFile.cpp
create mode 100644 third_party/llvm-project/Twine.cpp
create mode 100644 third_party/llvm-project/UnicodeCaseFold.cpp
create mode 100644 third_party/llvm-project/Unix/Path.inc
create mode 100644 third_party/llvm-project/Unix/Unix.h
create mode 100644 third_party/llvm-project/WithColor.cpp
create mode 100644 third_party/llvm-project/YAMLParser.cpp
create mode 100644 third_party/llvm-project/YAMLTraits.cpp
create mode 100644 third_party/llvm-project/dwarf2yaml.cpp
create mode 100644 third_party/llvm-project/include/llvm-c/DataTypes.h
create mode 100644 third_party/llvm-project/include/llvm-c/DisassemblerTypes.h
create mode 100644 third_party/llvm-project/include/llvm-c/Error.h
create mode 100644 third_party/llvm-project/include/llvm-c/ErrorHandling.h
create mode 100644 third_party/llvm-project/include/llvm-c/Types.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/APFloat.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/APInt.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/APSInt.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/AllocatorList.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ArrayRef.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/BitmaskEnum.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/DenseMap.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/DenseMapInfo.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/DenseSet.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/EpochTracker.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/FoldingSet.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/FunctionExtras.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/Hashing.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/MapVector.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/None.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/Optional.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/PointerIntPair.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/PointerUnion.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/STLExtras.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/SmallPtrSet.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/SmallSet.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/SmallString.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/SmallVector.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/StringExtras.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/StringMap.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/StringRef.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/StringSet.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/StringSwitch.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/Triple.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/Twine.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/bit.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/edit_distance.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/fallible_iterator.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ilist_base.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ilist_iterator.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ilist_node.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ilist_node_base.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/ilist_node_options.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/iterator.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/iterator_range.h
create mode 100644 third_party/llvm-project/include/llvm/ADT/simple_ilist.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/COFF.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/DynamicTags.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELF.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AArch64.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AMDGPU.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARC.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARM.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AVR.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/BPF.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Hexagon.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Lanai.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/MSP430.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Mips.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC64.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/RISCV.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Sparc.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/SystemZ.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/i386.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/x86_64.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/MachO.def
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/MachO.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/Magic.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/Wasm.h
create mode 100644 third_party/llvm-project/include/llvm/BinaryFormat/WasmRelocs.def
create mode 100644 third_party/llvm-project/include/llvm/Config/abi-breaking.h
create mode 100644 third_party/llvm-project/include/llvm/Config/config.h
create mode 100644 third_party/llvm-project/include/llvm/Config/llvm-config.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DIContext.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAddressRange.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAttribute.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFCompileUnit.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFContext.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDataExtractor.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAddr.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugFrame.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLine.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLoc.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugPubTable.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRnglists.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDie.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFExpression.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFFormValue.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFListTable.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFObject.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFSection.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFTypeUnit.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnit.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h
create mode 100644 third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
create mode 100644 third_party/llvm-project/include/llvm/LICENSE.TXT
create mode 100644 third_party/llvm-project/include/llvm/MC/LaneBitmask.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCExpr.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCFixup.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCFragment.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCInst.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCRegister.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCRegisterInfo.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCSymbol.h
create mode 100644 third_party/llvm-project/include/llvm/MC/MCSymbolWasm.h
create mode 100644 third_party/llvm-project/include/llvm/MC/SubtargetFeature.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Archive.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Binary.h
create mode 100644 third_party/llvm-project/include/llvm/Object/COFF.h
create mode 100644 third_party/llvm-project/include/llvm/Object/CVDebugRecord.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Decompressor.h
create mode 100644 third_party/llvm-project/include/llvm/Object/ELF.h
create mode 100644 third_party/llvm-project/include/llvm/Object/ELFObjectFile.h
create mode 100644 third_party/llvm-project/include/llvm/Object/ELFTypes.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Error.h
create mode 100644 third_party/llvm-project/include/llvm/Object/MachO.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Minidump.h
create mode 100644 third_party/llvm-project/include/llvm/Object/ObjectFile.h
create mode 100644 third_party/llvm-project/include/llvm/Object/RelocationResolver.h
create mode 100644 third_party/llvm-project/include/llvm/Object/SymbolicFile.h
create mode 100644 third_party/llvm-project/include/llvm/Object/Wasm.h
create mode 100644 third_party/llvm-project/include/llvm/ObjectYAML/DWARFEmitter.h
create mode 100644 third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
create mode 100644 third_party/llvm-project/include/llvm/ObjectYAML/ObjectYAML.h
create mode 100644 third_party/llvm-project/include/llvm/Support/AArch64TargetParser.def
create mode 100644 third_party/llvm-project/include/llvm/Support/AArch64TargetParser.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ARMAttributeParser.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ARMBuildAttributes.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ARMTargetParser.def
create mode 100644 third_party/llvm-project/include/llvm/Support/ARMTargetParser.h
create mode 100644 third_party/llvm-project/include/llvm/Support/AlignOf.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Alignment.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Allocator.h
create mode 100644 third_party/llvm-project/include/llvm/Support/BinaryByteStream.h
create mode 100644 third_party/llvm-project/include/llvm/Support/BinaryStream.h
create mode 100644 third_party/llvm-project/include/llvm/Support/BinaryStreamError.h
create mode 100644 third_party/llvm-project/include/llvm/Support/CBindingWrapping.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Casting.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Chrono.h
create mode 100644 third_party/llvm-project/include/llvm/Support/CodeGen.h
create mode 100644 third_party/llvm-project/include/llvm/Support/CommandLine.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Compiler.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ConvertUTF.h
create mode 100644 third_party/llvm-project/include/llvm/Support/DJB.h
create mode 100644 third_party/llvm-project/include/llvm/Support/DataExtractor.h
create mode 100644 third_party/llvm-project/include/llvm/Support/DataTypes.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Debug.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Endian.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Errc.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Errno.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Error.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ErrorHandling.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ErrorOr.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FileOutputBuffer.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FileSystem.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Format.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormatAdapters.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormatCommon.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormatProviders.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormatVariadic.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormatVariadicDetails.h
create mode 100644 third_party/llvm-project/include/llvm/Support/FormattedStream.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Host.h
create mode 100644 third_party/llvm-project/include/llvm/Support/LEB128.h
create mode 100644 third_party/llvm-project/include/llvm/Support/LICENSE.TXT
create mode 100644 third_party/llvm-project/include/llvm/Support/LineIterator.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Locale.h
create mode 100644 third_party/llvm-project/include/llvm/Support/MD5.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ManagedStatic.h
create mode 100644 third_party/llvm-project/include/llvm/Support/MathExtras.h
create mode 100644 third_party/llvm-project/include/llvm/Support/MemAlloc.h
create mode 100644 third_party/llvm-project/include/llvm/Support/MemoryBuffer.h
create mode 100644 third_party/llvm-project/include/llvm/Support/NativeFormatting.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Path.h
create mode 100644 third_party/llvm-project/include/llvm/Support/PointerLikeTypeTraits.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Printable.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Process.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Program.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Regex.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ReverseIteration.h
create mode 100644 third_party/llvm-project/include/llvm/Support/SMLoc.h
create mode 100644 third_party/llvm-project/include/llvm/Support/ScopedPrinter.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Signals.h
create mode 100644 third_party/llvm-project/include/llvm/Support/SmallVectorMemoryBuffer.h
create mode 100644 third_party/llvm-project/include/llvm/Support/SourceMgr.h
create mode 100644 third_party/llvm-project/include/llvm/Support/SwapByteOrder.h
create mode 100644 third_party/llvm-project/include/llvm/Support/TargetParser.h
create mode 100644 third_party/llvm-project/include/llvm/Support/TargetRegistry.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Threading.h
create mode 100644 third_party/llvm-project/include/llvm/Support/TypeSize.h
create mode 100644 third_party/llvm-project/include/llvm/Support/Unicode.h
create mode 100644 third_party/llvm-project/include/llvm/Support/UnicodeCharRanges.h
create mode 100644 third_party/llvm-project/include/llvm/Support/WindowsError.h
create mode 100644 third_party/llvm-project/include/llvm/Support/WithColor.h
create mode 100644 third_party/llvm-project/include/llvm/Support/X86TargetParser.def
create mode 100644 third_party/llvm-project/include/llvm/Support/YAMLParser.h
create mode 100644 third_party/llvm-project/include/llvm/Support/YAMLTraits.h
create mode 100644 third_party/llvm-project/include/llvm/Support/circular_raw_ostream.h
create mode 100644 third_party/llvm-project/include/llvm/Support/raw_ostream.h
create mode 100644 third_party/llvm-project/include/llvm/Support/type_traits.h
create mode 100644 third_party/llvm-project/include/llvm/include/llvm/DebugInfo/DWARFContext.h
create mode 100644 third_party/llvm-project/include/llvm/readme.txt
create mode 100644 third_party/llvm-project/obj2yaml_Error.cpp
create mode 100644 third_party/llvm-project/raw_ostream.cpp
create mode 100644 third_party/llvm-project/readme.txt
(limited to 'third_party')
diff --git a/third_party/.clang-format b/third_party/.clang-format
new file mode 100644
index 000000000..fe98bac05
--- /dev/null
+++ b/third_party/.clang-format
@@ -0,0 +1,5 @@
+---
+Language: Cpp
+DisableFormat: true
+SortIncludes: false
+---
diff --git a/third_party/.clang-tidy b/third_party/.clang-tidy
new file mode 100644
index 000000000..aa4dd1be0
--- /dev/null
+++ b/third_party/.clang-tidy
@@ -0,0 +1 @@
+Checks: ''
diff --git a/third_party/CMakeLists.txt b/third_party/CMakeLists.txt
new file mode 100644
index 000000000..40d5b09bb
--- /dev/null
+++ b/third_party/CMakeLists.txt
@@ -0,0 +1,3 @@
+if(BUILD_LLVM_DWARF)
+ add_subdirectory(llvm-project)
+endif()
diff --git a/third_party/llvm-project/Binary.cpp b/third_party/llvm-project/Binary.cpp
new file mode 100644
index 000000000..37fabac34
--- /dev/null
+++ b/third_party/llvm-project/Binary.cpp
@@ -0,0 +1,72 @@
+//===- Binary.cpp - A generic binary file ---------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the Binary class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Object/Binary.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Magic.h"
+#include "llvm/Object/Archive.h"
+#include "llvm/Object/Error.h"
+#if 0 // XXX BINARYEN
+#include "llvm/Object/MachOUniversal.h"
+#include "llvm/Object/Minidump.h"
+#endif
+#include "llvm/Object/ObjectFile.h"
+#if 0 // XXX BINARYEN
+#include "llvm/Object/TapiUniversal.h"
+#include "llvm/Object/WindowsResource.h"
+#endif
+#include "llvm/Support/Error.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <algorithm>
+#include <memory>
+#include <system_error>
+
+using namespace llvm;
+using namespace object;
+
+Binary::~Binary() = default;
+
+Binary::Binary(unsigned int Type, MemoryBufferRef Source)
+ : TypeID(Type), Data(Source) {}
+
+StringRef Binary::getData() const { return Data.getBuffer(); }
+
+StringRef Binary::getFileName() const { return Data.getBufferIdentifier(); }
+
+MemoryBufferRef Binary::getMemoryBufferRef() const { return Data; }
+
+Expected<std::unique_ptr<Binary>> object::createBinary(MemoryBufferRef Buffer,
+ LLVMContext *Context) {
+ llvm_unreachable("createBinary");
+}
+
+#if 0 // XXX BINARYEN
+Expected<OwningBinary<Binary>> object::createBinary(StringRef Path) {
+ ErrorOr<std::unique_ptr<MemoryBuffer>> FileOrErr =
+ MemoryBuffer::getFileOrSTDIN(Path, /*FileSize=*/-1,
+ /*RequiresNullTerminator=*/false);
+ if (std::error_code EC = FileOrErr.getError())
+ return errorCodeToError(EC);
+ std::unique_ptr<MemoryBuffer> &Buffer = FileOrErr.get();
+
+ Expected<std::unique_ptr<Binary>> BinOrErr =
+ createBinary(Buffer->getMemBufferRef());
+ if (!BinOrErr)
+ return BinOrErr.takeError();
+ std::unique_ptr<Binary> &Bin = BinOrErr.get();
+
+ return OwningBinary<Binary>(std::move(Bin), std::move(Buffer));
+}
+#endif
diff --git a/third_party/llvm-project/CMakeLists.txt b/third_party/llvm-project/CMakeLists.txt
new file mode 100644
index 000000000..d338dd6e0
--- /dev/null
+++ b/third_party/llvm-project/CMakeLists.txt
@@ -0,0 +1,76 @@
+# LLVM sources have warnings, ignore them. Otherwise, use options as close
+# as possible to LLVM's defaults.
+foreach(variable CMAKE_C_FLAGS CMAKE_CXX_FLAGS)
+ if(NOT MSVC)
+ set(${variable} "${${variable}} -w -std=c++14 -D_GNU_SOURCE -D_DEBUG -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS")
+ else()
+ set(${variable} "${${variable}} /w -std:c++14 -D_CRT_SECURE_NO_DEPRECATE -D_CRT_SECURE_NO_WARNINGS -D_CRT_NONSTDC_NO_DEPRECATE -D_CRT_NONSTDC_NO_WARNINGS -D_SCL_SECURE_NO_DEPRECATE -D_SCL_SECURE_NO_WARNINGS -DUNICODE -D_UNICODE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS")
+ endif()
+endforeach(variable)
+
+SET(llvm_dwarf_SOURCES
+ Binary.cpp
+ ConvertUTF.cpp
+ DataExtractor.cpp
+ Debug.cpp
+ DJB.cpp
+ Dwarf.cpp
+ dwarf2yaml.cpp
+ DWARFAbbreviationDeclaration.cpp
+ DWARFAcceleratorTable.cpp
+ DWARFAddressRange.cpp
+ DWARFCompileUnit.cpp
+ DWARFContext.cpp
+ DWARFDataExtractor.cpp
+ DWARFDebugAbbrev.cpp
+ DWARFDebugAddr.cpp
+ DWARFDebugAranges.cpp
+ DWARFDebugArangeSet.cpp
+ DWARFDebugFrame.cpp
+ DWARFDebugInfoEntry.cpp
+ DWARFDebugLine.cpp
+ DWARFDebugLoc.cpp
+ DWARFDebugMacro.cpp
+ DWARFDebugPubTable.cpp
+ DWARFDebugRangeList.cpp
+ DWARFDebugRnglists.cpp
+ DWARFDie.cpp
+ DWARFEmitter.cpp
+ DWARFExpression.cpp
+ DWARFFormValue.cpp
+ DWARFGdbIndex.cpp
+ DWARFListTable.cpp
+ DWARFTypeUnit.cpp
+ DWARFUnit.cpp
+ DWARFUnitIndex.cpp
+ DWARFVerifier.cpp
+ DWARFVisitor.cpp
+ DWARFYAML.cpp
+ Error.cpp
+ ErrorHandling.cpp
+ FormatVariadic.cpp
+ Hashing.cpp
+ LEB128.cpp
+ LineIterator.cpp
+ MCRegisterInfo.cpp
+ MD5.cpp
+ MemoryBuffer.cpp
+ NativeFormatting.cpp
+ ObjectFile.cpp
+ obj2yaml_Error.cpp
+ Optional.cpp
+ Path.cpp
+ raw_ostream.cpp
+ ScopedPrinter.cpp
+ SmallVector.cpp
+ SourceMgr.cpp
+ StringMap.cpp
+ StringRef.cpp
+ SymbolicFile.cpp
+ Twine.cpp
+ UnicodeCaseFold.cpp
+ WithColor.cpp
+ YAMLParser.cpp # XXX needed?
+ YAMLTraits.cpp
+)
+ADD_LIBRARY(llvm_dwarf OBJECT ${llvm_dwarf_SOURCES})
diff --git a/third_party/llvm-project/ConvertUTF.cpp b/third_party/llvm-project/ConvertUTF.cpp
new file mode 100644
index 000000000..e24a918c5
--- /dev/null
+++ b/third_party/llvm-project/ConvertUTF.cpp
@@ -0,0 +1,738 @@
+/*===--- ConvertUTF.c - Universal Character Names conversions ---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===------------------------------------------------------------------------=*/
+/*
+ * Copyright 2001-2004 Unicode, Inc.
+ *
+ * Disclaimer
+ *
+ * This source code is provided as is by Unicode, Inc. No claims are
+ * made as to fitness for any particular purpose. No warranties of any
+ * kind are expressed or implied. The recipient agrees to determine
+ * applicability of information provided. If this file has been
+ * purchased on magnetic or optical media from Unicode, Inc., the
+ * sole remedy for any claim will be exchange of defective media
+ * within 90 days of receipt.
+ *
+ * Limitations on Rights to Redistribute This Code
+ *
+ * Unicode, Inc. hereby grants the right to freely use the information
+ * supplied in this file in the creation of products supporting the
+ * Unicode Standard, and to make copies of this file in any form
+ * for internal or external distribution as long as this notice
+ * remains attached.
+ */
+
+/* ---------------------------------------------------------------------
+
+ Conversions between UTF32, UTF-16, and UTF-8. Source code file.
+ Author: Mark E. Davis, 1994.
+ Rev History: Rick McGowan, fixes & updates May 2001.
+ Sept 2001: fixed const & error conditions per
+ mods suggested by S. Parent & A. Lillich.
+ June 2002: Tim Dodd added detection and handling of incomplete
+ source sequences, enhanced error detection, added casts
+ to eliminate compiler warnings.
+ July 2003: slight mods to back out aggressive FFFE detection.
+ Jan 2004: updated switches in from-UTF8 conversions.
+ Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions.
+
+ See the header file "ConvertUTF.h" for complete documentation.
+
+------------------------------------------------------------------------ */
+
+#include "llvm/Support/ConvertUTF.h"
+#ifdef CVTUTF_DEBUG
+#include <stdio.h>
+#endif
+#include <assert.h>
+
+/*
+ * This code extensively uses fall-through switches.
+ * Keep the compiler from warning about that.
+ */
+#if defined(__clang__) && defined(__has_warning)
+# if __has_warning("-Wimplicit-fallthrough")
+# define ConvertUTF_DISABLE_WARNINGS \
+ _Pragma("clang diagnostic push") \
+ _Pragma("clang diagnostic ignored \"-Wimplicit-fallthrough\"")
+# define ConvertUTF_RESTORE_WARNINGS \
+ _Pragma("clang diagnostic pop")
+# endif
+#elif defined(__GNUC__) && __GNUC__ > 6
+# define ConvertUTF_DISABLE_WARNINGS \
+ _Pragma("GCC diagnostic push") \
+ _Pragma("GCC diagnostic ignored \"-Wimplicit-fallthrough\"")
+# define ConvertUTF_RESTORE_WARNINGS \
+ _Pragma("GCC diagnostic pop")
+#endif
+#ifndef ConvertUTF_DISABLE_WARNINGS
+# define ConvertUTF_DISABLE_WARNINGS
+#endif
+#ifndef ConvertUTF_RESTORE_WARNINGS
+# define ConvertUTF_RESTORE_WARNINGS
+#endif
+
+ConvertUTF_DISABLE_WARNINGS
+
+namespace llvm {
+
+static const int halfShift = 10; /* used for shifting by 10 bits */
+
+static const UTF32 halfBase = 0x0010000UL;
+static const UTF32 halfMask = 0x3FFUL;
+
+#define UNI_SUR_HIGH_START (UTF32)0xD800
+#define UNI_SUR_HIGH_END (UTF32)0xDBFF
+#define UNI_SUR_LOW_START (UTF32)0xDC00
+#define UNI_SUR_LOW_END (UTF32)0xDFFF
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Index into the table below with the first byte of a UTF-8 sequence to
+ * get the number of trailing bytes that are supposed to follow it.
+ * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
+ * left as-is for anyone who may want to do such conversion, which was
+ * allowed in earlier algorithms.
+ */
+static const char trailingBytesForUTF8[256] = {
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+ 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5
+};
+
+/*
+ * Magic values subtracted from a buffer value during UTF8 conversion.
+ * This table contains as many values as there might be trailing bytes
+ * in a UTF-8 sequence.
+ */
+static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL,
+ 0x03C82080UL, 0xFA082080UL, 0x82082080UL };
+
+/*
+ * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed
+ * into the first byte, depending on how many bytes follow. There are
+ * as many entries in this table as there are UTF-8 sequence types.
+ * (I.e., one byte sequence, two byte... etc.). Remember that sequencs
+ * for *legal* UTF-8 will be 4 or fewer bytes total.
+ */
+static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
+
+/* --------------------------------------------------------------------- */
+
+/* The interface converts a whole buffer to avoid function-call overhead.
+ * Constants have been gathered. Loops & conditionals have been removed as
+ * much as possible for efficiency, in favor of drop-through switches.
+ * (See "Note A" at the bottom of the file for equivalent code.)
+ * If your compiler supports it, the "isLegalUTF8" call can be turned
+ * into an inline function.
+ */
+
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF32toUTF16 (
+ const UTF32** sourceStart, const UTF32* sourceEnd,
+ UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
+ ConversionResult result = conversionOK;
+ const UTF32* source = *sourceStart;
+ UTF16* target = *targetStart;
+ while (source < sourceEnd) {
+ UTF32 ch;
+ if (target >= targetEnd) {
+ result = targetExhausted; break;
+ }
+ ch = *source++;
+ if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
+ /* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+ if (flags == strictConversion) {
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ } else {
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ } else {
+ *target++ = (UTF16)ch; /* normal case */
+ }
+ } else if (ch > UNI_MAX_LEGAL_UTF32) {
+ if (flags == strictConversion) {
+ result = sourceIllegal;
+ } else {
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ } else {
+ /* target is a character in range 0xFFFF - 0x10FFFF. */
+ if (target + 1 >= targetEnd) {
+ --source; /* Back up source pointer! */
+ result = targetExhausted; break;
+ }
+ ch -= halfBase;
+ *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
+ *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
+ }
+ }
+ *sourceStart = source;
+ *targetStart = target;
+ return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF16toUTF32 (
+ const UTF16** sourceStart, const UTF16* sourceEnd,
+ UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
+ ConversionResult result = conversionOK;
+ const UTF16* source = *sourceStart;
+ UTF32* target = *targetStart;
+ UTF32 ch, ch2;
+ while (source < sourceEnd) {
+ const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
+ ch = *source++;
+ /* If we have a surrogate pair, convert to UTF32 first. */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
+ /* If the 16 bits following the high surrogate are in the source buffer... */
+ if (source < sourceEnd) {
+ ch2 = *source;
+ /* If it's a low surrogate, convert to UTF32. */
+ if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
+ ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
+ + (ch2 - UNI_SUR_LOW_START) + halfBase;
+ ++source;
+ } else if (flags == strictConversion) { /* it's an unpaired high surrogate */
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ }
+ } else { /* We don't have the 16 bits following the high surrogate. */
+ --source; /* return to the high surrogate */
+ result = sourceExhausted;
+ break;
+ }
+ } else if (flags == strictConversion) {
+ /* UTF-16 surrogate values are illegal in UTF-32 */
+ if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ }
+ }
+ if (target >= targetEnd) {
+ source = oldSource; /* Back up source pointer! */
+ result = targetExhausted; break;
+ }
+ *target++ = ch;
+ }
+ *sourceStart = source;
+ *targetStart = target;
+#ifdef CVTUTF_DEBUG
+if (result == sourceIllegal) {
+ fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2);
+ fflush(stderr);
+}
+#endif
+ return result;
+}
+ConversionResult ConvertUTF16toUTF8 (
+ const UTF16** sourceStart, const UTF16* sourceEnd,
+ UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
+ ConversionResult result = conversionOK;
+ const UTF16* source = *sourceStart;
+ UTF8* target = *targetStart;
+ while (source < sourceEnd) {
+ UTF32 ch;
+ unsigned short bytesToWrite = 0;
+ const UTF32 byteMask = 0xBF;
+ const UTF32 byteMark = 0x80;
+ const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
+ ch = *source++;
+ /* If we have a surrogate pair, convert to UTF32 first. */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
+ /* If the 16 bits following the high surrogate are in the source buffer... */
+ if (source < sourceEnd) {
+ UTF32 ch2 = *source;
+ /* If it's a low surrogate, convert to UTF32. */
+ if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
+ ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
+ + (ch2 - UNI_SUR_LOW_START) + halfBase;
+ ++source;
+ } else if (flags == strictConversion) { /* it's an unpaired high surrogate */
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ }
+ } else { /* We don't have the 16 bits following the high surrogate. */
+ --source; /* return to the high surrogate */
+ result = sourceExhausted;
+ break;
+ }
+ } else if (flags == strictConversion) {
+ /* UTF-16 surrogate values are illegal in UTF-32 */
+ if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ }
+ }
+ /* Figure out how many bytes the result will require */
+ if (ch < (UTF32)0x80) { bytesToWrite = 1;
+ } else if (ch < (UTF32)0x800) { bytesToWrite = 2;
+ } else if (ch < (UTF32)0x10000) { bytesToWrite = 3;
+ } else if (ch < (UTF32)0x110000) { bytesToWrite = 4;
+ } else { bytesToWrite = 3;
+ ch = UNI_REPLACEMENT_CHAR;
+ }
+
+ target += bytesToWrite;
+ if (target > targetEnd) {
+ source = oldSource; /* Back up source pointer! */
+ target -= bytesToWrite; result = targetExhausted; break;
+ }
+ switch (bytesToWrite) { /* note: everything falls through. */
+ case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 1: *--target = (UTF8)(ch | firstByteMark[bytesToWrite]);
+ }
+ target += bytesToWrite;
+ }
+ *sourceStart = source;
+ *targetStart = target;
+ return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF32toUTF8 (
+ const UTF32** sourceStart, const UTF32* sourceEnd,
+ UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
+ ConversionResult result = conversionOK;
+ const UTF32* source = *sourceStart;
+ UTF8* target = *targetStart;
+ while (source < sourceEnd) {
+ UTF32 ch;
+ unsigned short bytesToWrite = 0;
+ const UTF32 byteMask = 0xBF;
+ const UTF32 byteMark = 0x80;
+ ch = *source++;
+ if (flags == strictConversion ) {
+ /* UTF-16 surrogate values are illegal in UTF-32 */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+ --source; /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ }
+ }
+ /*
+ * Figure out how many bytes the result will require. Turn any
+ * illegally large UTF32 things (> Plane 17) into replacement chars.
+ */
+ if (ch < (UTF32)0x80) { bytesToWrite = 1;
+ } else if (ch < (UTF32)0x800) { bytesToWrite = 2;
+ } else if (ch < (UTF32)0x10000) { bytesToWrite = 3;
+ } else if (ch <= UNI_MAX_LEGAL_UTF32) { bytesToWrite = 4;
+ } else { bytesToWrite = 3;
+ ch = UNI_REPLACEMENT_CHAR;
+ result = sourceIllegal;
+ }
+
+ target += bytesToWrite;
+ if (target > targetEnd) {
+ --source; /* Back up source pointer! */
+ target -= bytesToWrite; result = targetExhausted; break;
+ }
+ switch (bytesToWrite) { /* note: everything falls through. */
+ case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+ case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]);
+ }
+ target += bytesToWrite;
+ }
+ *sourceStart = source;
+ *targetStart = target;
+ return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Utility routine to tell whether a sequence of bytes is legal UTF-8.
+ * This must be called with the length pre-determined by the first byte.
+ * If not calling this from ConvertUTF8to*, then the length can be set by:
+ * length = trailingBytesForUTF8[*source]+1;
+ * and the sequence is illegal right away if there aren't that many bytes
+ * available.
+ * If presented with a length > 4, this returns false. The Unicode
+ * definition of UTF-8 goes up to 4-byte sequences.
+ */
+
+static Boolean isLegalUTF8(const UTF8 *source, int length) {
+ UTF8 a;
+ const UTF8 *srcptr = source+length;
+ switch (length) {
+ default: return false;
+ /* Everything else falls through when "true"... */
+ case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+ case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+ case 2: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+
+ switch (*source) {
+ /* no fall-through in this inner switch */
+ case 0xE0: if (a < 0xA0) return false; break;
+ case 0xED: if (a > 0x9F) return false; break;
+ case 0xF0: if (a < 0x90) return false; break;
+ case 0xF4: if (a > 0x8F) return false; break;
+ default: if (a < 0x80) return false;
+ }
+
+ case 1: if (*source >= 0x80 && *source < 0xC2) return false;
+ }
+ if (*source > 0xF4) return false;
+ return true;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Exported function to return whether a UTF-8 sequence is legal or not.
+ * This is not used here; it's just exported.
+ */
+Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) {
+ int length = trailingBytesForUTF8[*source]+1;
+ if (length > sourceEnd - source) {
+ return false;
+ }
+ return isLegalUTF8(source, length);
+}
+
+/* --------------------------------------------------------------------- */
+
+static unsigned
+findMaximalSubpartOfIllFormedUTF8Sequence(const UTF8 *source,
+ const UTF8 *sourceEnd) {
+ UTF8 b1, b2, b3;
+
+ assert(!isLegalUTF8Sequence(source, sourceEnd));
+
+ /*
+ * Unicode 6.3.0, D93b:
+ *
+ * Maximal subpart of an ill-formed subsequence: The longest code unit
+ * subsequence starting at an unconvertible offset that is either:
+ * a. the initial subsequence of a well-formed code unit sequence, or
+ * b. a subsequence of length one.
+ */
+
+ if (source == sourceEnd)
+ return 0;
+
+ /*
+ * Perform case analysis. See Unicode 6.3.0, Table 3-7. Well-Formed UTF-8
+ * Byte Sequences.
+ */
+
+ b1 = *source;
+ ++source;
+ if (b1 >= 0xC2 && b1 <= 0xDF) {
+ /*
+ * First byte is valid, but we know that this code unit sequence is
+ * invalid, so the maximal subpart has to end after the first byte.
+ */
+ return 1;
+ }
+
+ if (source == sourceEnd)
+ return 1;
+
+ b2 = *source;
+ ++source;
+
+ if (b1 == 0xE0) {
+ return (b2 >= 0xA0 && b2 <= 0xBF) ? 2 : 1;
+ }
+ if (b1 >= 0xE1 && b1 <= 0xEC) {
+ return (b2 >= 0x80 && b2 <= 0xBF) ? 2 : 1;
+ }
+ if (b1 == 0xED) {
+ return (b2 >= 0x80 && b2 <= 0x9F) ? 2 : 1;
+ }
+ if (b1 >= 0xEE && b1 <= 0xEF) {
+ return (b2 >= 0x80 && b2 <= 0xBF) ? 2 : 1;
+ }
+ if (b1 == 0xF0) {
+ if (b2 >= 0x90 && b2 <= 0xBF) {
+ if (source == sourceEnd)
+ return 2;
+
+ b3 = *source;
+ return (b3 >= 0x80 && b3 <= 0xBF) ? 3 : 2;
+ }
+ return 1;
+ }
+ if (b1 >= 0xF1 && b1 <= 0xF3) {
+ if (b2 >= 0x80 && b2 <= 0xBF) {
+ if (source == sourceEnd)
+ return 2;
+
+ b3 = *source;
+ return (b3 >= 0x80 && b3 <= 0xBF) ? 3 : 2;
+ }
+ return 1;
+ }
+ if (b1 == 0xF4) {
+ if (b2 >= 0x80 && b2 <= 0x8F) {
+ if (source == sourceEnd)
+ return 2;
+
+ b3 = *source;
+ return (b3 >= 0x80 && b3 <= 0xBF) ? 3 : 2;
+ }
+ return 1;
+ }
+
+ assert((b1 >= 0x80 && b1 <= 0xC1) || b1 >= 0xF5);
+ /*
+ * There are no valid sequences that start with these bytes. Maximal subpart
+ * is defined to have length 1 in these cases.
+ */
+ return 1;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Exported function to return the total number of bytes in a codepoint
+ * represented in UTF-8, given the value of the first byte.
+ */
+unsigned getNumBytesForUTF8(UTF8 first) {
+ return trailingBytesForUTF8[first] + 1;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Exported function to return whether a UTF-8 string is legal or not.
+ * This is not used here; it's just exported.
+ */
+Boolean isLegalUTF8String(const UTF8 **source, const UTF8 *sourceEnd) {
+ while (*source != sourceEnd) {
+ int length = trailingBytesForUTF8[**source] + 1;
+ if (length > sourceEnd - *source || !isLegalUTF8(*source, length))
+ return false;
+ *source += length;
+ }
+ return true;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF8toUTF16 (
+ const UTF8** sourceStart, const UTF8* sourceEnd,
+ UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
+ ConversionResult result = conversionOK;
+ const UTF8* source = *sourceStart;
+ UTF16* target = *targetStart;
+ while (source < sourceEnd) {
+ UTF32 ch = 0;
+ unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
+ if (extraBytesToRead >= sourceEnd - source) {
+ result = sourceExhausted; break;
+ }
+ /* Do this check whether lenient or strict */
+ if (!isLegalUTF8(source, extraBytesToRead+1)) {
+ result = sourceIllegal;
+ break;
+ }
+ /*
+ * The cases all fall through. See "Note A" below.
+ */
+ switch (extraBytesToRead) {
+ case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
+ case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
+ case 3: ch += *source++; ch <<= 6;
+ case 2: ch += *source++; ch <<= 6;
+ case 1: ch += *source++; ch <<= 6;
+ case 0: ch += *source++;
+ }
+ ch -= offsetsFromUTF8[extraBytesToRead];
+
+ if (target >= targetEnd) {
+ source -= (extraBytesToRead+1); /* Back up source pointer! */
+ result = targetExhausted; break;
+ }
+ if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
+ /* UTF-16 surrogate values are illegal in UTF-32 */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+ if (flags == strictConversion) {
+ source -= (extraBytesToRead+1); /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ } else {
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ } else {
+ *target++ = (UTF16)ch; /* normal case */
+ }
+ } else if (ch > UNI_MAX_UTF16) {
+ if (flags == strictConversion) {
+ result = sourceIllegal;
+ source -= (extraBytesToRead+1); /* return to the start */
+ break; /* Bail out; shouldn't continue */
+ } else {
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ } else {
+ /* target is a character in range 0xFFFF - 0x10FFFF. */
+ if (target + 1 >= targetEnd) {
+ source -= (extraBytesToRead+1); /* Back up source pointer! */
+ result = targetExhausted; break;
+ }
+ ch -= halfBase;
+ *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
+ *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
+ }
+ }
+ *sourceStart = source;
+ *targetStart = target;
+ return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+static ConversionResult ConvertUTF8toUTF32Impl(
+ const UTF8** sourceStart, const UTF8* sourceEnd,
+ UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags,
+ Boolean InputIsPartial) {
+ ConversionResult result = conversionOK;
+ const UTF8* source = *sourceStart;
+ UTF32* target = *targetStart;
+ while (source < sourceEnd) {
+ UTF32 ch = 0;
+ unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
+ if (extraBytesToRead >= sourceEnd - source) {
+ if (flags == strictConversion || InputIsPartial) {
+ result = sourceExhausted;
+ break;
+ } else {
+ result = sourceIllegal;
+
+ /*
+ * Replace the maximal subpart of ill-formed sequence with
+ * replacement character.
+ */
+ source += findMaximalSubpartOfIllFormedUTF8Sequence(source,
+ sourceEnd);
+ *target++ = UNI_REPLACEMENT_CHAR;
+ continue;
+ }
+ }
+ if (target >= targetEnd) {
+ result = targetExhausted; break;
+ }
+
+ /* Do this check whether lenient or strict */
+ if (!isLegalUTF8(source, extraBytesToRead+1)) {
+ result = sourceIllegal;
+ if (flags == strictConversion) {
+ /* Abort conversion. */
+ break;
+ } else {
+ /*
+ * Replace the maximal subpart of ill-formed sequence with
+ * replacement character.
+ */
+ source += findMaximalSubpartOfIllFormedUTF8Sequence(source,
+ sourceEnd);
+ *target++ = UNI_REPLACEMENT_CHAR;
+ continue;
+ }
+ }
+ /*
+ * The cases all fall through. See "Note A" below.
+ */
+ switch (extraBytesToRead) {
+ case 5: ch += *source++; ch <<= 6;
+ case 4: ch += *source++; ch <<= 6;
+ case 3: ch += *source++; ch <<= 6;
+ case 2: ch += *source++; ch <<= 6;
+ case 1: ch += *source++; ch <<= 6;
+ case 0: ch += *source++;
+ }
+ ch -= offsetsFromUTF8[extraBytesToRead];
+
+ if (ch <= UNI_MAX_LEGAL_UTF32) {
+ /*
+ * UTF-16 surrogate values are illegal in UTF-32, and anything
+ * over Plane 17 (> 0x10FFFF) is illegal.
+ */
+ if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+ if (flags == strictConversion) {
+ source -= (extraBytesToRead+1); /* return to the illegal value itself */
+ result = sourceIllegal;
+ break;
+ } else {
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ } else {
+ *target++ = ch;
+ }
+ } else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
+ result = sourceIllegal;
+ *target++ = UNI_REPLACEMENT_CHAR;
+ }
+ }
+ *sourceStart = source;
+ *targetStart = target;
+ return result;
+}
+
+ConversionResult ConvertUTF8toUTF32Partial(const UTF8 **sourceStart,
+ const UTF8 *sourceEnd,
+ UTF32 **targetStart,
+ UTF32 *targetEnd,
+ ConversionFlags flags) {
+ return ConvertUTF8toUTF32Impl(sourceStart, sourceEnd, targetStart, targetEnd,
+ flags, /*InputIsPartial=*/true);
+}
+
+ConversionResult ConvertUTF8toUTF32(const UTF8 **sourceStart,
+ const UTF8 *sourceEnd, UTF32 **targetStart,
+ UTF32 *targetEnd, ConversionFlags flags) {
+ return ConvertUTF8toUTF32Impl(sourceStart, sourceEnd, targetStart, targetEnd,
+ flags, /*InputIsPartial=*/false);
+}
+
+/* ---------------------------------------------------------------------
+
+ Note A.
+ The fall-through switches in UTF-8 reading code save a
+ temp variable, some decrements & conditionals. The switches
+ are equivalent to the following loop:
+ {
+ int tmpBytesToRead = extraBytesToRead+1;
+ do {
+ ch += *source++;
+ --tmpBytesToRead;
+ if (tmpBytesToRead) ch <<= 6;
+ } while (tmpBytesToRead > 0);
+ }
+ In UTF-8 writing code, the switches on "bytesToWrite" are
+ similarly unrolled loops.
+
+ --------------------------------------------------------------------- */
+
+} // namespace llvm
+
+ConvertUTF_RESTORE_WARNINGS
diff --git a/third_party/llvm-project/DJB.cpp b/third_party/llvm-project/DJB.cpp
new file mode 100644
index 000000000..f06af7dfd
--- /dev/null
+++ b/third_party/llvm-project/DJB.cpp
@@ -0,0 +1,82 @@
+//===-- Support/DJB.cpp ---DJB Hash -----------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains support for the DJ Bernstein hash function.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/DJB.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ConvertUTF.h"
+#include "llvm/Support/Unicode.h"
+
+using namespace llvm;
+
+static UTF32 chopOneUTF32(StringRef &Buffer) {
+ UTF32 C;
+ const UTF8 *const Begin8Const =
+ reinterpret_cast<const UTF8 *>(Buffer.begin());
+ const UTF8 *Begin8 = Begin8Const;
+ UTF32 *Begin32 = &C;
+
+ // In lenient mode we will always end up with a "reasonable" value in C for
+ // non-empty input.
+ assert(!Buffer.empty());
+ ConvertUTF8toUTF32(&Begin8, reinterpret_cast<const UTF8 *>(Buffer.end()),
+ &Begin32, &C + 1, lenientConversion);
+ Buffer = Buffer.drop_front(Begin8 - Begin8Const);
+ return C;
+}
+
+static StringRef toUTF8(UTF32 C, MutableArrayRef<UTF8> Storage) {
+ const UTF32 *Begin32 = &C;
+ UTF8 *Begin8 = Storage.begin();
+
+ // The case-folded output should always be a valid unicode character, so use
+ // strict mode here.
+ ConversionResult CR = ConvertUTF32toUTF8(&Begin32, &C + 1, &Begin8,
+ Storage.end(), strictConversion);
+ assert(CR == conversionOK && "Case folding produced invalid char?");
+ (void)CR;
+ return StringRef(reinterpret_cast<char *>(Storage.begin()),
+ Begin8 - Storage.begin());
+}
+
+static UTF32 foldCharDwarf(UTF32 C) {
+ // DWARF v5 addition to the unicode folding rules.
+ // Fold "Latin Small Letter Dotless I" and "Latin Capital Letter I With Dot
+ // Above" into "i".
+ if (C == 0x130 || C == 0x131)
+ return 'i';
+ return sys::unicode::foldCharSimple(C);
+}
+
+static Optional<uint32_t> fastCaseFoldingDjbHash(StringRef Buffer, uint32_t H) {
+ bool AllASCII = true;
+ for (unsigned char C : Buffer) {
+ H = H * 33 + ('A' <= C && C <= 'Z' ? C - 'A' + 'a' : C);
+ AllASCII &= C <= 0x7f;
+ }
+ if (AllASCII)
+ return H;
+ return None;
+}
+
+uint32_t llvm::caseFoldingDjbHash(StringRef Buffer, uint32_t H) {
+ if (Optional<uint32_t> Result = fastCaseFoldingDjbHash(Buffer, H))
+ return *Result;
+
+ std::array<UTF8, UNI_MAX_UTF8_BYTES_PER_CODE_POINT> Storage;
+ while (!Buffer.empty()) {
+ UTF32 C = foldCharDwarf(chopOneUTF32(Buffer));
+ StringRef Folded = toUTF8(C, Storage);
+ H = djbHash(Folded, H);
+ }
+ return H;
+}
diff --git a/third_party/llvm-project/DWARFAbbreviationDeclaration.cpp b/third_party/llvm-project/DWARFAbbreviationDeclaration.cpp
new file mode 100644
index 000000000..abbea3a86
--- /dev/null
+++ b/third_party/llvm-project/DWARFAbbreviationDeclaration.cpp
@@ -0,0 +1,215 @@
+//===- DWARFAbbreviationDeclaration.cpp -----------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
+
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstddef>
+#include <cstdint>
+
+using namespace llvm;
+using namespace dwarf;
+
+void DWARFAbbreviationDeclaration::clear() {
+ Code = 0;
+ Tag = DW_TAG_null;
+ CodeByteSize = 0;
+ HasChildren = false;
+ AttributeSpecs.clear();
+ FixedAttributeSize.reset();
+}
+
+DWARFAbbreviationDeclaration::DWARFAbbreviationDeclaration() {
+ clear();
+}
+
+bool
+DWARFAbbreviationDeclaration::extract(DataExtractor Data,
+ uint64_t* OffsetPtr) {
+ clear();
+ const uint64_t Offset = *OffsetPtr;
+ Code = Data.getULEB128(OffsetPtr);
+ if (Code == 0) {
+ return false;
+ }
+ CodeByteSize = *OffsetPtr - Offset;
+ Tag = static_cast<llvm::dwarf::Tag>(Data.getULEB128(OffsetPtr));
+ if (Tag == DW_TAG_null) {
+ clear();
+ return false;
+ }
+ uint8_t ChildrenByte = Data.getU8(OffsetPtr);
+ HasChildren = (ChildrenByte == DW_CHILDREN_yes);
+ // Assign a value to our optional FixedAttributeSize member variable. If
+ // this member variable still has a value after the while loop below, then
+ // all attribute data in this abbreviation declaration has a fixed byte size.
+ FixedAttributeSize = FixedSizeInfo();
+
+ // Read all of the abbreviation attributes and forms.
+ while (true) {
+ auto A = static_cast<Attribute>(Data.getULEB128(OffsetPtr));
+ auto F = static_cast<Form>(Data.getULEB128(OffsetPtr));
+ if (A && F) {
+ bool IsImplicitConst = (F == DW_FORM_implicit_const);
+ if (IsImplicitConst) {
+ int64_t V = Data.getSLEB128(OffsetPtr);
+ AttributeSpecs.push_back(AttributeSpec(A, F, V));
+ continue;
+ }
+ Optional<uint8_t> ByteSize;
+ // If this abbrevation still has a fixed byte size, then update the
+ // FixedAttributeSize as needed.
+ switch (F) {
+ case DW_FORM_addr:
+ if (FixedAttributeSize)
+ ++FixedAttributeSize->NumAddrs;
+ break;
+
+ case DW_FORM_ref_addr:
+ if (FixedAttributeSize)
+ ++FixedAttributeSize->NumRefAddrs;
+ break;
+
+ case DW_FORM_strp:
+ case DW_FORM_GNU_ref_alt:
+ case DW_FORM_GNU_strp_alt:
+ case DW_FORM_line_strp:
+ case DW_FORM_sec_offset:
+ case DW_FORM_strp_sup:
+ if (FixedAttributeSize)
+ ++FixedAttributeSize->NumDwarfOffsets;
+ break;
+
+ default:
+ // The form has a byte size that doesn't depend on Params.
+ // If it's a fixed size, keep track of it.
+ if ((ByteSize = dwarf::getFixedFormByteSize(F, dwarf::FormParams()))) {
+ if (FixedAttributeSize)
+ FixedAttributeSize->NumBytes += *ByteSize;
+ break;
+ }
+ // Indicate we no longer have a fixed byte size for this
+ // abbreviation by clearing the FixedAttributeSize optional value
+ // so it doesn't have a value.
+ FixedAttributeSize.reset();
+ break;
+ }
+ // Record this attribute and its fixed size if it has one.
+ AttributeSpecs.push_back(AttributeSpec(A, F, ByteSize));
+ } else if (A == 0 && F == 0) {
+ // We successfully reached the end of this abbreviation declaration
+ // since both attribute and form are zero.
+ break;
+ } else {
+ // Attribute and form pairs must either both be non-zero, in which case
+ // they are added to the abbreviation declaration, or both be zero to
+ // terminate the abbrevation declaration. In this case only one was
+ // zero which is an error.
+ clear();
+ return false;
+ }
+ }
+ return true;
+}
+
+void DWARFAbbreviationDeclaration::dump(raw_ostream &OS) const {
+ OS << '[' << getCode() << "] ";
+ OS << formatv("{0}", getTag());
+ OS << "\tDW_CHILDREN_" << (hasChildren() ? "yes" : "no") << '\n';
+ for (const AttributeSpec &Spec : AttributeSpecs) {
+ OS << formatv("\t{0}\t{1}", Spec.Attr, Spec.Form);
+ if (Spec.isImplicitConst())
+ OS << '\t' << Spec.getImplicitConstValue();
+ OS << '\n';
+ }
+ OS << '\n';
+}
+
+Optional<uint32_t>
+DWARFAbbreviationDeclaration::findAttributeIndex(dwarf::Attribute Attr) const {
+ for (uint32_t i = 0, e = AttributeSpecs.size(); i != e; ++i) {
+ if (AttributeSpecs[i].Attr == Attr)
+ return i;
+ }
+ return None;
+}
+
+Optional<DWARFFormValue> DWARFAbbreviationDeclaration::getAttributeValue(
+ const uint64_t DIEOffset, const dwarf::Attribute Attr,
+ const DWARFUnit &U) const {
+ Optional<uint32_t> MatchAttrIndex = findAttributeIndex(Attr);
+ if (!MatchAttrIndex)
+ return None;
+
+ auto DebugInfoData = U.getDebugInfoExtractor();
+
+ // Add the byte size of ULEB that for the abbrev Code so we can start
+ // skipping the attribute data.
+ uint64_t Offset = DIEOffset + CodeByteSize;
+ uint32_t AttrIndex = 0;
+ for (const auto &Spec : AttributeSpecs) {
+ if (*MatchAttrIndex == AttrIndex) {
+ // We have arrived at the attribute to extract, extract if from Offset.
+ if (Spec.isImplicitConst())
+ return DWARFFormValue::createFromSValue(Spec.Form,
+ Spec.getImplicitConstValue());
+
+ DWARFFormValue FormValue(Spec.Form);
+ if (FormValue.extractValue(DebugInfoData, &Offset, U.getFormParams(), &U))
+ return FormValue;
+ }
+ // March Offset along until we get to the attribute we want.
+ if (auto FixedSize = Spec.getByteSize(U))
+ Offset += *FixedSize;
+ else
+ DWARFFormValue::skipValue(Spec.Form, DebugInfoData, &Offset,
+ U.getFormParams());
+ ++AttrIndex;
+ }
+ return None;
+}
+
+size_t DWARFAbbreviationDeclaration::FixedSizeInfo::getByteSize(
+ const DWARFUnit &U) const {
+ size_t ByteSize = NumBytes;
+ if (NumAddrs)
+ ByteSize += NumAddrs * U.getAddressByteSize();
+ if (NumRefAddrs)
+ ByteSize += NumRefAddrs * U.getRefAddrByteSize();
+ if (NumDwarfOffsets)
+ ByteSize += NumDwarfOffsets * U.getDwarfOffsetByteSize();
+ return ByteSize;
+}
+
+Optional<int64_t> DWARFAbbreviationDeclaration::AttributeSpec::getByteSize(
+ const DWARFUnit &U) const {
+ if (isImplicitConst())
+ return 0;
+ if (ByteSize.HasByteSize)
+ return ByteSize.ByteSize;
+ Optional<int64_t> S;
+ auto FixedByteSize = dwarf::getFixedFormByteSize(Form, U.getFormParams());
+ if (FixedByteSize)
+ S = *FixedByteSize;
+ return S;
+}
+
+Optional<size_t> DWARFAbbreviationDeclaration::getFixedAttributesByteSize(
+ const DWARFUnit &U) const {
+ if (FixedAttributeSize)
+ return FixedAttributeSize->getByteSize(U);
+ return None;
+}
diff --git a/third_party/llvm-project/DWARFAcceleratorTable.cpp b/third_party/llvm-project/DWARFAcceleratorTable.cpp
new file mode 100644
index 000000000..875f5e998
--- /dev/null
+++ b/third_party/llvm-project/DWARFAcceleratorTable.cpp
@@ -0,0 +1,889 @@
+//===- DWARFAcceleratorTable.cpp ------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h"
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DJB.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/ScopedPrinter.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstddef>
+#include <cstdint>
+#include <utility>
+
+using namespace llvm;
+
+namespace {
+struct Atom {
+ unsigned Value;
+};
+
+static raw_ostream &operator<<(raw_ostream &OS, const Atom &A) {
+ StringRef Str = dwarf::AtomTypeString(A.Value);
+ if (!Str.empty())
+ return OS << Str;
+ return OS << "DW_ATOM_unknown_" << format("%x", A.Value);
+}
+} // namespace
+
+static Atom formatAtom(unsigned Atom) { return {Atom}; }
+
+DWARFAcceleratorTable::~DWARFAcceleratorTable() = default;
+
+Error AppleAcceleratorTable::extract() {
+ uint64_t Offset = 0;
+
+ // Check that we can at least read the header.
+ if (!AccelSection.isValidOffset(offsetof(Header, HeaderDataLength) + 4))
+ return createStringError(errc::illegal_byte_sequence,
+ "Section too small: cannot read header.");
+
+ Hdr.Magic = AccelSection.getU32(&Offset);
+ Hdr.Version = AccelSection.getU16(&Offset);
+ Hdr.HashFunction = AccelSection.getU16(&Offset);
+ Hdr.BucketCount = AccelSection.getU32(&Offset);
+ Hdr.HashCount = AccelSection.getU32(&Offset);
+ Hdr.HeaderDataLength = AccelSection.getU32(&Offset);
+
+ // Check that we can read all the hashes and offsets from the
+ // section (see SourceLevelDebugging.rst for the structure of the index).
+ // We need to substract one because we're checking for an *offset* which is
+ // equal to the size for an empty table and hence pointer after the section.
+ if (!AccelSection.isValidOffset(sizeof(Hdr) + Hdr.HeaderDataLength +
+ Hdr.BucketCount * 4 + Hdr.HashCount * 8 - 1))
+ return createStringError(
+ errc::illegal_byte_sequence,
+ "Section too small: cannot read buckets and hashes.");
+
+ HdrData.DIEOffsetBase = AccelSection.getU32(&Offset);
+ uint32_t NumAtoms = AccelSection.getU32(&Offset);
+
+ for (unsigned i = 0; i < NumAtoms; ++i) {
+ uint16_t AtomType = AccelSection.getU16(&Offset);
+ auto AtomForm = static_cast<dwarf::Form>(AccelSection.getU16(&Offset));
+ HdrData.Atoms.push_back(std::make_pair(AtomType, AtomForm));
+ }
+
+ IsValid = true;
+ return Error::success();
+}
+
+uint32_t AppleAcceleratorTable::getNumBuckets() { return Hdr.BucketCount; }
+uint32_t AppleAcceleratorTable::getNumHashes() { return Hdr.HashCount; }
+uint32_t AppleAcceleratorTable::getSizeHdr() { return sizeof(Hdr); }
+uint32_t AppleAcceleratorTable::getHeaderDataLength() {
+ return Hdr.HeaderDataLength;
+}
+
+ArrayRef<std::pair<AppleAcceleratorTable::HeaderData::AtomType,
+ AppleAcceleratorTable::HeaderData::Form>>
+AppleAcceleratorTable::getAtomsDesc() {
+ return HdrData.Atoms;
+}
+
+bool AppleAcceleratorTable::validateForms() {
+ for (auto Atom : getAtomsDesc()) {
+ DWARFFormValue FormValue(Atom.second);
+ switch (Atom.first) {
+ case dwarf::DW_ATOM_die_offset:
+ case dwarf::DW_ATOM_die_tag:
+ case dwarf::DW_ATOM_type_flags:
+ if ((!FormValue.isFormClass(DWARFFormValue::FC_Constant) &&
+ !FormValue.isFormClass(DWARFFormValue::FC_Flag)) ||
+ FormValue.getForm() == dwarf::DW_FORM_sdata)
+ return false;
+ break;
+ default:
+ break;
+ }
+ }
+ return true;
+}
+
+std::pair<uint64_t, dwarf::Tag>
+AppleAcceleratorTable::readAtoms(uint64_t *HashDataOffset) {
+ uint64_t DieOffset = dwarf::DW_INVALID_OFFSET;
+ dwarf::Tag DieTag = dwarf::DW_TAG_null;
+ dwarf::FormParams FormParams = {Hdr.Version, 0, dwarf::DwarfFormat::DWARF32};
+
+ for (auto Atom : getAtomsDesc()) {
+ DWARFFormValue FormValue(Atom.second);
+ FormValue.extractValue(AccelSection, HashDataOffset, FormParams);
+ switch (Atom.first) {
+ case dwarf::DW_ATOM_die_offset:
+ DieOffset = *FormValue.getAsUnsignedConstant();
+ break;
+ case dwarf::DW_ATOM_die_tag:
+ DieTag = (dwarf::Tag)*FormValue.getAsUnsignedConstant();
+ break;
+ default:
+ break;
+ }
+ }
+ return {DieOffset, DieTag};
+}
+
+void AppleAcceleratorTable::Header::dump(ScopedPrinter &W) const {
+ DictScope HeaderScope(W, "Header");
+ W.printHex("Magic", Magic);
+ W.printHex("Version", Version);
+ W.printHex("Hash function", HashFunction);
+ W.printNumber("Bucket count", BucketCount);
+ W.printNumber("Hashes count", HashCount);
+ W.printNumber("HeaderData length", HeaderDataLength);
+}
+
+Optional<uint64_t> AppleAcceleratorTable::HeaderData::extractOffset(
+ Optional<DWARFFormValue> Value) const {
+ if (!Value)
+ return None;
+
+ switch (Value->getForm()) {
+ case dwarf::DW_FORM_ref1:
+ case dwarf::DW_FORM_ref2:
+ case dwarf::DW_FORM_ref4:
+ case dwarf::DW_FORM_ref8:
+ case dwarf::DW_FORM_ref_udata:
+ return Value->getRawUValue() + DIEOffsetBase;
+ default:
+ return Value->getAsSectionOffset();
+ }
+}
+
+bool AppleAcceleratorTable::dumpName(ScopedPrinter &W,
+ SmallVectorImpl<DWARFFormValue> &AtomForms,
+ uint64_t *DataOffset) const {
+ dwarf::FormParams FormParams = {Hdr.Version, 0, dwarf::DwarfFormat::DWARF32};
+ uint64_t NameOffset = *DataOffset;
+ if (!AccelSection.isValidOffsetForDataOfSize(*DataOffset, 4)) {
+ W.printString("Incorrectly terminated list.");
+ return false;
+ }
+ uint64_t StringOffset = AccelSection.getRelocatedValue(4, DataOffset);
+ if (!StringOffset)
+ return false; // End of list
+
+ DictScope NameScope(W, ("Name@0x" + Twine::utohexstr(NameOffset)).str());
+ W.startLine() << format("String: 0x%08" PRIx64, StringOffset);
+ W.getOStream() << " \"" << StringSection.getCStr(&StringOffset) << "\"\n";
+
+ unsigned NumData = AccelSection.getU32(DataOffset);
+ for (unsigned Data = 0; Data < NumData; ++Data) {
+ ListScope DataScope(W, ("Data " + Twine(Data)).str());
+ unsigned i = 0;
+ for (auto &Atom : AtomForms) {
+ W.startLine() << format("Atom[%d]: ", i);
+ if (Atom.extractValue(AccelSection, DataOffset, FormParams)) {
+ Atom.dump(W.getOStream());
+ if (Optional<uint64_t> Val = Atom.getAsUnsignedConstant()) {
+ StringRef Str = dwarf::AtomValueString(HdrData.Atoms[i].first, *Val);
+ if (!Str.empty())
+ W.getOStream() << " (" << Str << ")";
+ }
+ } else
+ W.getOStream() << "Error extracting the value";
+ W.getOStream() << "\n";
+ i++;
+ }
+ }
+ return true; // more entries follow
+}
+
+LLVM_DUMP_METHOD void AppleAcceleratorTable::dump(raw_ostream &OS) const {
+ if (!IsValid)
+ return;
+
+ ScopedPrinter W(OS);
+
+ Hdr.dump(W);
+
+ W.printNumber("DIE offset base", HdrData.DIEOffsetBase);
+ W.printNumber("Number of atoms", uint64_t(HdrData.Atoms.size()));
+ SmallVector<DWARFFormValue, 3> AtomForms;
+ {
+ ListScope AtomsScope(W, "Atoms");
+ unsigned i = 0;
+ for (const auto &Atom : HdrData.Atoms) {
+ DictScope AtomScope(W, ("Atom " + Twine(i++)).str());
+ W.startLine() << "Type: " << formatAtom(Atom.first) << '\n';
+ W.startLine() << "Form: " << formatv("{0}", Atom.second) << '\n';
+ AtomForms.push_back(DWARFFormValue(Atom.second));
+ }
+ }
+
+ // Now go through the actual tables and dump them.
+ uint64_t Offset = sizeof(Hdr) + Hdr.HeaderDataLength;
+ uint64_t HashesBase = Offset + Hdr.BucketCount * 4;
+ uint64_t OffsetsBase = HashesBase + Hdr.HashCount * 4;
+
+ for (unsigned Bucket = 0; Bucket < Hdr.BucketCount; ++Bucket) {
+ unsigned Index = AccelSection.getU32(&Offset);
+
+ ListScope BucketScope(W, ("Bucket " + Twine(Bucket)).str());
+ if (Index == UINT32_MAX) {
+ W.printString("EMPTY");
+ continue;
+ }
+
+ for (unsigned HashIdx = Index; HashIdx < Hdr.HashCount; ++HashIdx) {
+ uint64_t HashOffset = HashesBase + HashIdx*4;
+ uint64_t OffsetsOffset = OffsetsBase + HashIdx*4;
+ uint32_t Hash = AccelSection.getU32(&HashOffset);
+
+ if (Hash % Hdr.BucketCount != Bucket)
+ break;
+
+ uint64_t DataOffset = AccelSection.getU32(&OffsetsOffset);
+ ListScope HashScope(W, ("Hash 0x" + Twine::utohexstr(Hash)).str());
+ if (!AccelSection.isValidOffset(DataOffset)) {
+ W.printString("Invalid section offset");
+ continue;
+ }
+ while (dumpName(W, AtomForms, &DataOffset))
+ /*empty*/;
+ }
+ }
+}
+
+AppleAcceleratorTable::Entry::Entry(
+ const AppleAcceleratorTable::HeaderData &HdrData)
+ : HdrData(&HdrData) {
+ Values.reserve(HdrData.Atoms.size());
+ for (const auto &Atom : HdrData.Atoms)
+ Values.push_back(DWARFFormValue(Atom.second));
+}
+
+void AppleAcceleratorTable::Entry::extract(
+ const AppleAcceleratorTable &AccelTable, uint64_t *Offset) {
+
+ dwarf::FormParams FormParams = {AccelTable.Hdr.Version, 0,
+ dwarf::DwarfFormat::DWARF32};
+ for (auto &Atom : Values)
+ Atom.extractValue(AccelTable.AccelSection, Offset, FormParams);
+}
+
+Optional<DWARFFormValue>
+AppleAcceleratorTable::Entry::lookup(HeaderData::AtomType Atom) const {
+ assert(HdrData && "Dereferencing end iterator?");
+ assert(HdrData->Atoms.size() == Values.size());
+ for (const auto &Tuple : zip_first(HdrData->Atoms, Values)) {
+ if (std::get<0>(Tuple).first == Atom)
+ return std::get<1>(Tuple);
+ }
+ return None;
+}
+
+Optional<uint64_t> AppleAcceleratorTable::Entry::getDIESectionOffset() const {
+ return HdrData->extractOffset(lookup(dwarf::DW_ATOM_die_offset));
+}
+
+Optional<uint64_t> AppleAcceleratorTable::Entry::getCUOffset() const {
+ return HdrData->extractOffset(lookup(dwarf::DW_ATOM_cu_offset));
+}
+
+Optional<dwarf::Tag> AppleAcceleratorTable::Entry::getTag() const {
+ Optional<DWARFFormValue> Tag = lookup(dwarf::DW_ATOM_die_tag);
+ if (!Tag)
+ return None;
+ if (Optional<uint64_t> Value = Tag->getAsUnsignedConstant())
+ return dwarf::Tag(*Value);
+ return None;
+}
+
+AppleAcceleratorTable::ValueIterator::ValueIterator(
+ const AppleAcceleratorTable &AccelTable, uint64_t Offset)
+ : AccelTable(&AccelTable), Current(AccelTable.HdrData), DataOffset(Offset) {
+ if (!AccelTable.AccelSection.isValidOffsetForDataOfSize(DataOffset, 4))
+ return;
+
+ // Read the first entry.
+ NumData = AccelTable.AccelSection.getU32(&DataOffset);
+ Next();
+}
+
+void AppleAcceleratorTable::ValueIterator::Next() {
+ assert(NumData > 0 && "attempted to increment iterator past the end");
+ auto &AccelSection = AccelTable->AccelSection;
+ if (Data >= NumData ||
+ !AccelSection.isValidOffsetForDataOfSize(DataOffset, 4)) {
+ NumData = 0;
+ DataOffset = 0;
+ return;
+ }
+ Current.extract(*AccelTable, &DataOffset);
+ ++Data;
+}
+
+iterator_range<AppleAcceleratorTable::ValueIterator>
+AppleAcceleratorTable::equal_range(StringRef Key) const {
+ if (!IsValid)
+ return make_range(ValueIterator(), ValueIterator());
+
+ // Find the bucket.
+ unsigned HashValue = djbHash(Key);
+ unsigned Bucket = HashValue % Hdr.BucketCount;
+ uint64_t BucketBase = sizeof(Hdr) + Hdr.HeaderDataLength;
+ uint64_t HashesBase = BucketBase + Hdr.BucketCount * 4;
+ uint64_t OffsetsBase = HashesBase + Hdr.HashCount * 4;
+
+ uint64_t BucketOffset = BucketBase + Bucket * 4;
+ unsigned Index = AccelSection.getU32(&BucketOffset);
+
+ // Search through all hashes in the bucket.
+ for (unsigned HashIdx = Index; HashIdx < Hdr.HashCount; ++HashIdx) {
+ uint64_t HashOffset = HashesBase + HashIdx * 4;
+ uint64_t OffsetsOffset = OffsetsBase + HashIdx * 4;
+ uint32_t Hash = AccelSection.getU32(&HashOffset);
+
+ if (Hash % Hdr.BucketCount != Bucket)
+ // We are already in the next bucket.
+ break;
+
+ uint64_t DataOffset = AccelSection.getU32(&OffsetsOffset);
+ uint64_t StringOffset = AccelSection.getRelocatedValue(4, &DataOffset);
+ if (!StringOffset)
+ break;
+
+ // Finally, compare the key.
+ if (Key == StringSection.getCStr(&StringOffset))
+ return make_range({*this, DataOffset}, ValueIterator());
+ }
+ return make_range(ValueIterator(), ValueIterator());
+}
+
+void DWARFDebugNames::Header::dump(ScopedPrinter &W) const {
+ DictScope HeaderScope(W, "Header");
+ W.printHex("Length", UnitLength);
+ W.printNumber("Version", Version);
+ W.printHex("Padding", Padding);
+ W.printNumber("CU count", CompUnitCount);
+ W.printNumber("Local TU count", LocalTypeUnitCount);
+ W.printNumber("Foreign TU count", ForeignTypeUnitCount);
+ W.printNumber("Bucket count", BucketCount);
+ W.printNumber("Name count", NameCount);
+ W.printHex("Abbreviations table size", AbbrevTableSize);
+ W.startLine() << "Augmentation: '" << AugmentationString << "'\n";
+}
+
+Error DWARFDebugNames::Header::extract(const DWARFDataExtractor &AS,
+ uint64_t *Offset) {
+ // Check that we can read the fixed-size part.
+ if (!AS.isValidOffset(*Offset + sizeof(HeaderPOD) - 1))
+ return createStringError(errc::illegal_byte_sequence,
+ "Section too small: cannot read header.");
+
+ UnitLength = AS.getU32(Offset);
+ Version = AS.getU16(Offset);
+ Padding = AS.getU16(Offset);
+ CompUnitCount = AS.getU32(Offset);
+ LocalTypeUnitCount = AS.getU32(Offset);
+ ForeignTypeUnitCount = AS.getU32(Offset);
+ BucketCount = AS.getU32(Offset);
+ NameCount = AS.getU32(Offset);
+ AbbrevTableSize = AS.getU32(Offset);
+ AugmentationStringSize = alignTo(AS.getU32(Offset), 4);
+
+ if (!AS.isValidOffsetForDataOfSize(*Offset, AugmentationStringSize))
+ return createStringError(
+ errc::illegal_byte_sequence,
+ "Section too small: cannot read header augmentation.");
+ AugmentationString.resize(AugmentationStringSize);
+ AS.getU8(Offset, reinterpret_cast<uint8_t *>(AugmentationString.data()),
+ AugmentationStringSize);
+ return Error::success();
+}
+
+void DWARFDebugNames::Abbrev::dump(ScopedPrinter &W) const {
+ DictScope AbbrevScope(W, ("Abbreviation 0x" + Twine::utohexstr(Code)).str());
+ W.startLine() << formatv("Tag: {0}\n", Tag);
+
+ for (const auto &Attr : Attributes)
+ W.startLine() << formatv("{0}: {1}\n", Attr.Index, Attr.Form);
+}
+
+static constexpr DWARFDebugNames::AttributeEncoding sentinelAttrEnc() {
+ return {dwarf::Index(0), dwarf::Form(0)};
+}
+
+static bool isSentinel(const DWARFDebugNames::AttributeEncoding &AE) {
+ return AE == sentinelAttrEnc();
+}
+
+static DWARFDebugNames::Abbrev sentinelAbbrev() {
+ return DWARFDebugNames::Abbrev(0, dwarf::Tag(0), {});
+}
+
+static bool isSentinel(const DWARFDebugNames::Abbrev &Abbr) {
+ return Abbr.Code == 0;
+}
+
+DWARFDebugNames::Abbrev DWARFDebugNames::AbbrevMapInfo::getEmptyKey() {
+ return sentinelAbbrev();
+}
+
+DWARFDebugNames::Abbrev DWARFDebugNames::AbbrevMapInfo::getTombstoneKey() {
+ return DWARFDebugNames::Abbrev(~0, dwarf::Tag(0), {});
+}
+
+Expected<DWARFDebugNames::AttributeEncoding>
+DWARFDebugNames::NameIndex::extractAttributeEncoding(uint64_t *Offset) {
+ if (*Offset >= EntriesBase) {
+ return createStringError(errc::illegal_byte_sequence,
+ "Incorrectly terminated abbreviation table.");
+ }
+
+ uint32_t Index = Section.AccelSection.getULEB128(Offset);
+ uint32_t Form = Section.AccelSection.getULEB128(Offset);
+ return AttributeEncoding(dwarf::Index(Index), dwarf::Form(Form));
+}
+
+Expected<std::vector<DWARFDebugNames::AttributeEncoding>>
+DWARFDebugNames::NameIndex::extractAttributeEncodings(uint64_t *Offset) {
+ std::vector<AttributeEncoding> Result;
+ for (;;) {
+ auto AttrEncOr = extractAttributeEncoding(Offset);
+ if (!AttrEncOr)
+ return AttrEncOr.takeError();
+ if (isSentinel(*AttrEncOr))
+ return std::move(Result);
+
+ Result.emplace_back(*AttrEncOr);
+ }
+}
+
+Expected<DWARFDebugNames::Abbrev>
+DWARFDebugNames::NameIndex::extractAbbrev(uint64_t *Offset) {
+ if (*Offset >= EntriesBase) {
+ return createStringError(errc::illegal_byte_sequence,
+ "Incorrectly terminated abbreviation table.");
+ }
+
+ uint32_t Code = Section.AccelSection.getULEB128(Offset);
+ if (Code == 0)
+ return sentinelAbbrev();
+
+ uint32_t Tag = Section.AccelSection.getULEB128(Offset);
+ auto AttrEncOr = extractAttributeEncodings(Offset);
+ if (!AttrEncOr)
+ return AttrEncOr.takeError();
+ return Abbrev(Code, dwarf::Tag(Tag), std::move(*AttrEncOr));
+}
+
+Error DWARFDebugNames::NameIndex::extract() {
+ const DWARFDataExtractor &AS = Section.AccelSection;
+ uint64_t Offset = Base;
+ if (Error E = Hdr.extract(AS, &Offset))
+ return E;
+
+ CUsBase = Offset;
+ Offset += Hdr.CompUnitCount * 4;
+ Offset += Hdr.LocalTypeUnitCount * 4;
+ Offset += Hdr.ForeignTypeUnitCount * 8;
+ BucketsBase = Offset;
+ Offset += Hdr.BucketCount * 4;
+ HashesBase = Offset;
+ if (Hdr.BucketCount > 0)
+ Offset += Hdr.NameCount * 4;
+ StringOffsetsBase = Offset;
+ Offset += Hdr.NameCount * 4;
+ EntryOffsetsBase = Offset;
+ Offset += Hdr.NameCount * 4;
+
+ if (!AS.isValidOffsetForDataOfSize(Offset, Hdr.AbbrevTableSize))
+ return createStringError(errc::illegal_byte_sequence,
+ "Section too small: cannot read abbreviations.");
+
+ EntriesBase = Offset + Hdr.AbbrevTableSize;
+
+ for (;;) {
+ auto AbbrevOr = extractAbbrev(&Offset);
+ if (!AbbrevOr)
+ return AbbrevOr.takeError();
+ if (isSentinel(*AbbrevOr))
+ return Error::success();
+
+ if (!Abbrevs.insert(std::move(*AbbrevOr)).second)
+ return createStringError(errc::invalid_argument,
+ "Duplicate abbreviation code.");
+ }
+}
+
+DWARFDebugNames::Entry::Entry(const NameIndex &NameIdx, const Abbrev &Abbr)
+ : NameIdx(&NameIdx), Abbr(&Abbr) {
+ // This merely creates form values. It is up to the caller
+ // (NameIndex::getEntry) to populate them.
+ Values.reserve(Abbr.Attributes.size());
+ for (const auto &Attr : Abbr.Attributes)
+ Values.emplace_back(Attr.Form);
+}
+
+Optional<DWARFFormValue>
+DWARFDebugNames::Entry::lookup(dwarf::Index Index) const {
+ assert(Abbr->Attributes.size() == Values.size());
+ for (const auto &Tuple : zip_first(Abbr->Attributes, Values)) {
+ if (std::get<0>(Tuple).Index == Index)
+ return std::get<1>(Tuple);
+ }
+ return None;
+}
+
+Optional<uint64_t> DWARFDebugNames::Entry::getDIEUnitOffset() const {
+ if (Optional<DWARFFormValue> Off = lookup(dwarf::DW_IDX_die_offset))
+ return Off->getAsReferenceUVal();
+ return None;
+}
+
+Optional<uint64_t> DWARFDebugNames::Entry::getCUIndex() const {
+ if (Optional<DWARFFormValue> Off = lookup(dwarf::DW_IDX_compile_unit))
+ return Off->getAsUnsignedConstant();
+ // In a per-CU index, the entries without a DW_IDX_compile_unit attribute
+ // implicitly refer to the single CU.
+ if (NameIdx->getCUCount() == 1)
+ return 0;
+ return None;
+}
+
+Optional<uint64_t> DWARFDebugNames::Entry::getCUOffset() const {
+ Optional<uint64_t> Index = getCUIndex();
+ if (!Index || *Index >= NameIdx->getCUCount())
+ return None;
+ return NameIdx->getCUOffset(*Index);
+}
+
+void DWARFDebugNames::Entry::dump(ScopedPrinter &W) const {
+ W.printHex("Abbrev", Abbr->Code);
+ W.startLine() << formatv("Tag: {0}\n", Abbr->Tag);
+ assert(Abbr->Attributes.size() == Values.size());
+ for (const auto &Tuple : zip_first(Abbr->Attributes, Values)) {
+ W.startLine() << formatv("{0}: ", std::get<0>(Tuple).Index);
+ std::get<1>(Tuple).dump(W.getOStream());
+ W.getOStream() << '\n';
+ }
+}
+
+char DWARFDebugNames::SentinelError::ID;
+std::error_code DWARFDebugNames::SentinelError::convertToErrorCode() const {
+ return inconvertibleErrorCode();
+}
+
+uint64_t DWARFDebugNames::NameIndex::getCUOffset(uint32_t CU) const {
+ assert(CU < Hdr.CompUnitCount);
+ uint64_t Offset = CUsBase + 4 * CU;
+ return Section.AccelSection.getRelocatedValue(4, &Offset);
+}
+
+uint64_t DWARFDebugNames::NameIndex::getLocalTUOffset(uint32_t TU) const {
+ assert(TU < Hdr.LocalTypeUnitCount);
+ uint64_t Offset = CUsBase + 4 * (Hdr.CompUnitCount + TU);
+ return Section.AccelSection.getRelocatedValue(4, &Offset);
+}
+
+uint64_t DWARFDebugNames::NameIndex::getForeignTUSignature(uint32_t TU) const {
+ assert(TU < Hdr.ForeignTypeUnitCount);
+ uint64_t Offset =
+ CUsBase + 4 * (Hdr.CompUnitCount + Hdr.LocalTypeUnitCount) + 8 * TU;
+ return Section.AccelSection.getU64(&Offset);
+}
+
+Expected<DWARFDebugNames::Entry>
+DWARFDebugNames::NameIndex::getEntry(uint64_t *Offset) const {
+ const DWARFDataExtractor &AS = Section.AccelSection;
+ if (!AS.isValidOffset(*Offset))
+ return createStringError(errc::illegal_byte_sequence,
+ "Incorrectly terminated entry list.");
+
+ uint32_t AbbrevCode = AS.getULEB128(Offset);
+ if (AbbrevCode == 0)
+ return make_error<SentinelError>();
+
+ const auto AbbrevIt = Abbrevs.find_as(AbbrevCode);
+ if (AbbrevIt == Abbrevs.end())
+ return createStringError(errc::invalid_argument, "Invalid abbreviation.");
+
+ Entry E(*this, *AbbrevIt);
+
+ dwarf::FormParams FormParams = {Hdr.Version, 0, dwarf::DwarfFormat::DWARF32};
+ for (auto &Value : E.Values) {
+ if (!Value.extractValue(AS, Offset, FormParams))
+ return createStringError(errc::io_error,
+ "Error extracting index attribute values.");
+ }
+ return std::move(E);
+}
+
+DWARFDebugNames::NameTableEntry
+DWARFDebugNames::NameIndex::getNameTableEntry(uint32_t Index) const {
+ assert(0 < Index && Index <= Hdr.NameCount);
+ uint64_t StringOffsetOffset = StringOffsetsBase + 4 * (Index - 1);
+ uint64_t EntryOffsetOffset = EntryOffsetsBase + 4 * (Index - 1);
+ const DWARFDataExtractor &AS = Section.AccelSection;
+
+ uint64_t StringOffset = AS.getRelocatedValue(4, &StringOffsetOffset);
+ uint64_t EntryOffset = AS.getU32(&EntryOffsetOffset);
+ EntryOffset += EntriesBase;
+ return {Section.StringSection, Index, StringOffset, EntryOffset};
+}
+
+uint32_t
+DWARFDebugNames::NameIndex::getBucketArrayEntry(uint32_t Bucket) const {
+ assert(Bucket < Hdr.BucketCount);
+ uint64_t BucketOffset = BucketsBase + 4 * Bucket;
+ return Section.AccelSection.getU32(&BucketOffset);
+}
+
+uint32_t DWARFDebugNames::NameIndex::getHashArrayEntry(uint32_t Index) const {
+ assert(0 < Index && Index <= Hdr.NameCount);
+ uint64_t HashOffset = HashesBase + 4 * (Index - 1);
+ return Section.AccelSection.getU32(&HashOffset);
+}
+
+// Returns true if we should continue scanning for entries, false if this is the
+// last (sentinel) entry). In case of a parsing error we also return false, as
+// it's not possible to recover this entry list (but the other lists may still
+// parse OK).
+bool DWARFDebugNames::NameIndex::dumpEntry(ScopedPrinter &W,
+ uint64_t *Offset) const {
+ uint64_t EntryId = *Offset;
+ auto EntryOr = getEntry(Offset);
+ if (!EntryOr) {
+ handleAllErrors(EntryOr.takeError(), [](const SentinelError &) {},
+ [&W](const ErrorInfoBase &EI) { EI.log(W.startLine()); });
+ return false;
+ }
+
+ DictScope EntryScope(W, ("Entry @ 0x" + Twine::utohexstr(EntryId)).str());
+ EntryOr->dump(W);
+ return true;
+}
+
+void DWARFDebugNames::NameIndex::dumpName(ScopedPrinter &W,
+ const NameTableEntry &NTE,
+ Optional<uint32_t> Hash) const {
+ DictScope NameScope(W, ("Name " + Twine(NTE.getIndex())).str());
+ if (Hash)
+ W.printHex("Hash", *Hash);
+
+ W.startLine() << format("String: 0x%08" PRIx64, NTE.getStringOffset());
+ W.getOStream() << " \"" << NTE.getString() << "\"\n";
+
+ uint64_t EntryOffset = NTE.getEntryOffset();
+ while (dumpEntry(W, &EntryOffset))
+ /*empty*/;
+}
+
+void DWARFDebugNames::NameIndex::dumpCUs(ScopedPrinter &W) const {
+ ListScope CUScope(W, "Compilation Unit offsets");
+ for (uint32_t CU = 0; CU < Hdr.CompUnitCount; ++CU)
+ W.startLine() << format("CU[%u]: 0x%08" PRIx64 "\n", CU, getCUOffset(CU));
+}
+
+void DWARFDebugNames::NameIndex::dumpLocalTUs(ScopedPrinter &W) const {
+ if (Hdr.LocalTypeUnitCount == 0)
+ return;
+
+ ListScope TUScope(W, "Local Type Unit offsets");
+ for (uint32_t TU = 0; TU < Hdr.LocalTypeUnitCount; ++TU)
+ W.startLine() << format("LocalTU[%u]: 0x%08" PRIx64 "\n", TU,
+ getLocalTUOffset(TU));
+}
+
+void DWARFDebugNames::NameIndex::dumpForeignTUs(ScopedPrinter &W) const {
+ if (Hdr.ForeignTypeUnitCount == 0)
+ return;
+
+ ListScope TUScope(W, "Foreign Type Unit signatures");
+ for (uint32_t TU = 0; TU < Hdr.ForeignTypeUnitCount; ++TU) {
+ W.startLine() << format("ForeignTU[%u]: 0x%016" PRIx64 "\n", TU,
+ getForeignTUSignature(TU));
+ }
+}
+
+void DWARFDebugNames::NameIndex::dumpAbbreviations(ScopedPrinter &W) const {
+ ListScope AbbrevsScope(W, "Abbreviations");
+ for (const auto &Abbr : Abbrevs)
+ Abbr.dump(W);
+}
+
+void DWARFDebugNames::NameIndex::dumpBucket(ScopedPrinter &W,
+ uint32_t Bucket) const {
+ ListScope BucketScope(W, ("Bucket " + Twine(Bucket)).str());
+ uint32_t Index = getBucketArrayEntry(Bucket);
+ if (Index == 0) {
+ W.printString("EMPTY");
+ return;
+ }
+ if (Index > Hdr.NameCount) {
+ W.printString("Name index is invalid");
+ return;
+ }
+
+ for (; Index <= Hdr.NameCount; ++Index) {
+ uint32_t Hash = getHashArrayEntry(Index);
+ if (Hash % Hdr.BucketCount != Bucket)
+ break;
+
+ dumpName(W, getNameTableEntry(Index), Hash);
+ }
+}
+
+LLVM_DUMP_METHOD void DWARFDebugNames::NameIndex::dump(ScopedPrinter &W) const {
+ DictScope UnitScope(W, ("Name Index @ 0x" + Twine::utohexstr(Base)).str());
+ Hdr.dump(W);
+ dumpCUs(W);
+ dumpLocalTUs(W);
+ dumpForeignTUs(W);
+ dumpAbbreviations(W);
+
+ if (Hdr.BucketCount > 0) {
+ for (uint32_t Bucket = 0; Bucket < Hdr.BucketCount; ++Bucket)
+ dumpBucket(W, Bucket);
+ return;
+ }
+
+ W.startLine() << "Hash table not present\n";
+ for (NameTableEntry NTE : *this)
+ dumpName(W, NTE, None);
+}
+
+Error DWARFDebugNames::extract() {
+ uint64_t Offset = 0;
+ while (AccelSection.isValidOffset(Offset)) {
+ NameIndex Next(*this, Offset);
+ if (Error E = Next.extract())
+ return E;
+ Offset = Next.getNextUnitOffset();
+ NameIndices.push_back(std::move(Next));
+ }
+ return Error::success();
+}
+
+iterator_range<DWARFDebugNames::ValueIterator>
+DWARFDebugNames::NameIndex::equal_range(StringRef Key) const {
+ return make_range(ValueIterator(*this, Key), ValueIterator());
+}
+
+LLVM_DUMP_METHOD void DWARFDebugNames::dump(raw_ostream &OS) const {
+ ScopedPrinter W(OS);
+ for (const NameIndex &NI : NameIndices)
+ NI.dump(W);
+}
+
+Optional<uint64_t>
+DWARFDebugNames::ValueIterator::findEntryOffsetInCurrentIndex() {
+ const Header &Hdr = CurrentIndex->Hdr;
+ if (Hdr.BucketCount == 0) {
+ // No Hash Table, We need to search through all names in the Name Index.
+ for (NameTableEntry NTE : *CurrentIndex) {
+ if (NTE.getString() == Key)
+ return NTE.getEntryOffset();
+ }
+ return None;
+ }
+
+ // The Name Index has a Hash Table, so use that to speed up the search.
+ // Compute the Key Hash, if it has not been done already.
+ if (!Hash)
+ Hash = caseFoldingDjbHash(Key);
+ uint32_t Bucket = *Hash % Hdr.BucketCount;
+ uint32_t Index = CurrentIndex->getBucketArrayEntry(Bucket);
+ if (Index == 0)
+ return None; // Empty bucket
+
+ for (; Index <= Hdr.NameCount; ++Index) {
+ uint32_t Hash = CurrentIndex->getHashArrayEntry(Index);
+ if (Hash % Hdr.BucketCount != Bucket)
+ return None; // End of bucket
+
+ NameTableEntry NTE = CurrentIndex->getNameTableEntry(Index);
+ if (NTE.getString() == Key)
+ return NTE.getEntryOffset();
+ }
+ return None;
+}
+
+bool DWARFDebugNames::ValueIterator::getEntryAtCurrentOffset() {
+ auto EntryOr = CurrentIndex->getEntry(&DataOffset);
+ if (!EntryOr) {
+ consumeError(EntryOr.takeError());
+ return false;
+ }
+ CurrentEntry = std::move(*EntryOr);
+ return true;
+}
+
+bool DWARFDebugNames::ValueIterator::findInCurrentIndex() {
+ Optional<uint64_t> Offset = findEntryOffsetInCurrentIndex();
+ if (!Offset)
+ return false;
+ DataOffset = *Offset;
+ return getEntryAtCurrentOffset();
+}
+
+void DWARFDebugNames::ValueIterator::searchFromStartOfCurrentIndex() {
+ for (const NameIndex *End = CurrentIndex->Section.NameIndices.end();
+ CurrentIndex != End; ++CurrentIndex) {
+ if (findInCurrentIndex())
+ return;
+ }
+ setEnd();
+}
+
+void DWARFDebugNames::ValueIterator::next() {
+ assert(CurrentIndex && "Incrementing an end() iterator?");
+
+ // First try the next entry in the current Index.
+ if (getEntryAtCurrentOffset())
+ return;
+
+ // If we're a local iterator or we have reached the last Index, we're done.
+ if (IsLocal || CurrentIndex == &CurrentIndex->Section.NameIndices.back()) {
+ setEnd();
+ return;
+ }
+
+ // Otherwise, try the next index.
+ ++CurrentIndex;
+ searchFromStartOfCurrentIndex();
+}
+
+DWARFDebugNames::ValueIterator::ValueIterator(const DWARFDebugNames &AccelTable,
+ StringRef Key)
+ : CurrentIndex(AccelTable.NameIndices.begin()), IsLocal(false), Key(Key) {
+ searchFromStartOfCurrentIndex();
+}
+
+DWARFDebugNames::ValueIterator::ValueIterator(
+ const DWARFDebugNames::NameIndex &NI, StringRef Key)
+ : CurrentIndex(&NI), IsLocal(true), Key(Key) {
+ if (!findInCurrentIndex())
+ setEnd();
+}
+
+iterator_range<DWARFDebugNames::ValueIterator>
+DWARFDebugNames::equal_range(StringRef Key) const {
+ if (NameIndices.empty())
+ return make_range(ValueIterator(), ValueIterator());
+ return make_range(ValueIterator(*this, Key), ValueIterator());
+}
+
+const DWARFDebugNames::NameIndex *
+DWARFDebugNames::getCUNameIndex(uint64_t CUOffset) {
+ if (CUToNameIndex.size() == 0 && NameIndices.size() > 0) {
+ for (const auto &NI : *this) {
+ for (uint32_t CU = 0; CU < NI.getCUCount(); ++CU)
+ CUToNameIndex.try_emplace(NI.getCUOffset(CU), &NI);
+ }
+ }
+ return CUToNameIndex.lookup(CUOffset);
+}
diff --git a/third_party/llvm-project/DWARFAddressRange.cpp b/third_party/llvm-project/DWARFAddressRange.cpp
new file mode 100644
index 000000000..ef6da08d3
--- /dev/null
+++ b/third_party/llvm-project/DWARFAddressRange.cpp
@@ -0,0 +1,28 @@
+//===- DWARFDebugAranges.cpp ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFAddressRange.h"
+
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+void DWARFAddressRange::dump(raw_ostream &OS, uint32_t AddressSize,
+ DIDumpOptions DumpOpts) const {
+
+ OS << (DumpOpts.DisplayRawContents ? " " : "[");
+ OS << format("0x%*.*" PRIx64 ", ", AddressSize * 2, AddressSize * 2, LowPC)
+ << format("0x%*.*" PRIx64, AddressSize * 2, AddressSize * 2, HighPC);
+ OS << (DumpOpts.DisplayRawContents ? "" : ")");
+}
+
+raw_ostream &llvm::operator<<(raw_ostream &OS, const DWARFAddressRange &R) {
+ R.dump(OS, /* AddressSize */ 8);
+ return OS;
+}
diff --git a/third_party/llvm-project/DWARFCompileUnit.cpp b/third_party/llvm-project/DWARFCompileUnit.cpp
new file mode 100644
index 000000000..f59e49268
--- /dev/null
+++ b/third_party/llvm-project/DWARFCompileUnit.cpp
@@ -0,0 +1,38 @@
+//===-- DWARFCompileUnit.cpp ----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+void DWARFCompileUnit::dump(raw_ostream &OS, DIDumpOptions DumpOpts) {
+ OS << format("0x%08" PRIx64, getOffset()) << ": Compile Unit:"
+ << " length = " << format("0x%08" PRIx64, getLength())
+ << " version = " << format("0x%04x", getVersion());
+ if (getVersion() >= 5)
+ OS << " unit_type = " << dwarf::UnitTypeString(getUnitType());
+ OS << " abbr_offset = "
+ << format("0x%04" PRIx64, getAbbreviations()->getOffset())
+ << " addr_size = " << format("0x%02x", getAddressByteSize());
+ if (getVersion() >= 5 && getUnitType() != dwarf::DW_UT_compile)
+ OS << " DWO_id = " << format("0x%016" PRIx64, *getDWOId());
+ OS << " (next unit at " << format("0x%08" PRIx64, getNextUnitOffset())
+ << ")\n";
+
+ if (DWARFDie CUDie = getUnitDIE(false))
+ CUDie.dump(OS, 0, DumpOpts);
+ else
+ OS << "<compile unit can't be parsed!>\n\n";
+}
+
+// VTable anchor.
+DWARFCompileUnit::~DWARFCompileUnit() = default;
diff --git a/third_party/llvm-project/DWARFContext.cpp b/third_party/llvm-project/DWARFContext.cpp
new file mode 100644
index 000000000..da865a2c3
--- /dev/null
+++ b/third_party/llvm-project/DWARFContext.cpp
@@ -0,0 +1,1881 @@
+//===- DWARFContext.cpp ---------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/StringSwitch.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAddr.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAranges.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugFrame.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLoc.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugMacro.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugPubTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRangeList.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRnglists.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFGdbIndex.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/DebugInfo/DWARF/DWARFVerifier.h"
+#include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/Object/Decompressor.h"
+#include "llvm/Object/MachO.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Object/RelocationResolver.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/LEB128.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/Path.h"
+#include "llvm/Support/TargetRegistry.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cstdint>
+#include <deque>
+#include <map>
+#include <string>
+#include <utility>
+#include <vector>
+
+using namespace llvm;
+using namespace dwarf;
+using namespace object;
+
+#define DEBUG_TYPE "dwarf"
+
+using DWARFLineTable = DWARFDebugLine::LineTable;
+using FileLineInfoKind = DILineInfoSpecifier::FileLineInfoKind;
+using FunctionNameKind = DILineInfoSpecifier::FunctionNameKind;
+
+DWARFContext::DWARFContext(std::unique_ptr<const DWARFObject> DObj,
+ std::string DWPName)
+ : DIContext(CK_DWARF), DWPName(std::move(DWPName)), DObj(std::move(DObj)) {}
+
+DWARFContext::~DWARFContext() = default;
+
+/// Dump the UUID load command.
+static void dumpUUID(raw_ostream &OS, const ObjectFile &Obj) {
+#if 0 // XXX BINARYEN
+ auto *MachO = dyn_cast<MachOObjectFile>(&Obj);
+ if (!MachO)
+ return;
+ for (auto LC : MachO->load_commands()) {
+ raw_ostream::uuid_t UUID;
+ if (LC.C.cmd == MachO::LC_UUID) {
+ if (LC.C.cmdsize < sizeof(UUID) + sizeof(LC.C)) {
+ OS << "error: UUID load command is too short.\n";
+ return;
+ }
+ OS << "UUID: ";
+ memcpy(&UUID, LC.Ptr+sizeof(LC.C), sizeof(UUID));
+ OS.write_uuid(UUID);
+ Triple T = MachO->getArchTriple();
+ OS << " (" << T.getArchName() << ')';
+ OS << ' ' << MachO->getFileName() << '\n';
+ }
+ }
+#endif
+}
+
+using ContributionCollection =
+ std::vector<Optional<StrOffsetsContributionDescriptor>>;
+
+// Collect all the contributions to the string offsets table from all units,
+// sort them by their starting offsets and remove duplicates.
+static ContributionCollection
+collectContributionData(DWARFContext::unit_iterator_range Units) {
+ ContributionCollection Contributions;
+ for (const auto &U : Units)
+ if (const auto &C = U->getStringOffsetsTableContribution())
+ Contributions.push_back(C);
+ // Sort the contributions so that any invalid ones are placed at
+ // the start of the contributions vector. This way they are reported
+ // first.
+ llvm::sort(Contributions,
+ [](const Optional<StrOffsetsContributionDescriptor> &L,
+ const Optional<StrOffsetsContributionDescriptor> &R) {
+ if (L && R)
+ return L->Base < R->Base;
+ return R.hasValue();
+ });
+
+ // Uniquify contributions, as it is possible that units (specifically
+ // type units in dwo or dwp files) share contributions. We don't want
+ // to report them more than once.
+ Contributions.erase(
+ std::unique(Contributions.begin(), Contributions.end(),
+ [](const Optional<StrOffsetsContributionDescriptor> &L,
+ const Optional<StrOffsetsContributionDescriptor> &R) {
+ if (L && R)
+ return L->Base == R->Base && L->Size == R->Size;
+ return false;
+ }),
+ Contributions.end());
+ return Contributions;
+}
+
+static void dumpDWARFv5StringOffsetsSection(
+ raw_ostream &OS, StringRef SectionName, const DWARFObject &Obj,
+ const DWARFSection &StringOffsetsSection, StringRef StringSection,
+ DWARFContext::unit_iterator_range Units, bool LittleEndian) {
+ auto Contributions = collectContributionData(Units);
+ DWARFDataExtractor StrOffsetExt(Obj, StringOffsetsSection, LittleEndian, 0);
+ DataExtractor StrData(StringSection, LittleEndian, 0);
+ uint64_t SectionSize = StringOffsetsSection.Data.size();
+ uint64_t Offset = 0;
+ for (auto &Contribution : Contributions) {
+ // Report an ill-formed contribution.
+ if (!Contribution) {
+ OS << "error: invalid contribution to string offsets table in section ."
+ << SectionName << ".\n";
+ return;
+ }
+
+ dwarf::DwarfFormat Format = Contribution->getFormat();
+ uint16_t Version = Contribution->getVersion();
+ uint64_t ContributionHeader = Contribution->Base;
+ // In DWARF v5 there is a contribution header that immediately precedes
+ // the string offsets base (the location we have previously retrieved from
+ // the CU DIE's DW_AT_str_offsets attribute). The header is located either
+ // 8 or 16 bytes before the base, depending on the contribution's format.
+ if (Version >= 5)
+ ContributionHeader -= Format == DWARF32 ? 8 : 16;
+
+ // Detect overlapping contributions.
+ if (Offset > ContributionHeader) {
+ WithColor::error()
+ << "overlapping contributions to string offsets table in section ."
+ << SectionName << ".\n";
+ return;
+ }
+ // Report a gap in the table.
+ if (Offset < ContributionHeader) {
+ OS << format("0x%8.8" PRIx64 ": Gap, length = ", Offset);
+ OS << (ContributionHeader - Offset) << "\n";
+ }
+ OS << format("0x%8.8" PRIx64 ": ", ContributionHeader);
+ // In DWARF v5 the contribution size in the descriptor does not equal
+ // the originally encoded length (it does not contain the length of the
+ // version field and the padding, a total of 4 bytes). Add them back in
+ // for reporting.
+ OS << "Contribution size = " << (Contribution->Size + (Version < 5 ? 0 : 4))
+ << ", Format = " << (Format == DWARF32 ? "DWARF32" : "DWARF64")
+ << ", Version = " << Version << "\n";
+
+ Offset = Contribution->Base;
+ unsigned EntrySize = Contribution->getDwarfOffsetByteSize();
+ while (Offset - Contribution->Base < Contribution->Size) {
+ OS << format("0x%8.8" PRIx64 ": ", Offset);
+ uint64_t StringOffset =
+ StrOffsetExt.getRelocatedValue(EntrySize, &Offset);
+ OS << format("%8.8" PRIx64 " ", StringOffset);
+ const char *S = StrData.getCStr(&StringOffset);
+ if (S)
+ OS << format("\"%s\"", S);
+ OS << "\n";
+ }
+ }
+ // Report a gap at the end of the table.
+ if (Offset < SectionSize) {
+ OS << format("0x%8.8" PRIx64 ": Gap, length = ", Offset);
+ OS << (SectionSize - Offset) << "\n";
+ }
+}
+
+// Dump a DWARF string offsets section. This may be a DWARF v5 formatted
+// string offsets section, where each compile or type unit contributes a
+// number of entries (string offsets), with each contribution preceded by
+// a header containing size and version number. Alternatively, it may be a
+// monolithic series of string offsets, as generated by the pre-DWARF v5
+// implementation of split DWARF.
+static void dumpStringOffsetsSection(raw_ostream &OS, StringRef SectionName,
+ const DWARFObject &Obj,
+ const DWARFSection &StringOffsetsSection,
+ StringRef StringSection,
+ DWARFContext::unit_iterator_range Units,
+ bool LittleEndian, unsigned MaxVersion) {
+ // If we have at least one (compile or type) unit with DWARF v5 or greater,
+ // we assume that the section is formatted like a DWARF v5 string offsets
+ // section.
+ if (MaxVersion >= 5)
+ dumpDWARFv5StringOffsetsSection(OS, SectionName, Obj, StringOffsetsSection,
+ StringSection, Units, LittleEndian);
+ else {
+ DataExtractor strOffsetExt(StringOffsetsSection.Data, LittleEndian, 0);
+ uint64_t offset = 0;
+ uint64_t size = StringOffsetsSection.Data.size();
+ // Ensure that size is a multiple of the size of an entry.
+ if (size & ((uint64_t)(sizeof(uint32_t) - 1))) {
+ OS << "error: size of ." << SectionName << " is not a multiple of "
+ << sizeof(uint32_t) << ".\n";
+ size &= -(uint64_t)sizeof(uint32_t);
+ }
+ DataExtractor StrData(StringSection, LittleEndian, 0);
+ while (offset < size) {
+ OS << format("0x%8.8" PRIx64 ": ", offset);
+ uint64_t StringOffset = strOffsetExt.getU32(&offset);
+ OS << format("%8.8" PRIx64 " ", StringOffset);
+ const char *S = StrData.getCStr(&StringOffset);
+ if (S)
+ OS << format("\"%s\"", S);
+ OS << "\n";
+ }
+ }
+}
+
+// Dump the .debug_addr section.
+static void dumpAddrSection(raw_ostream &OS, DWARFDataExtractor &AddrData,
+ DIDumpOptions DumpOpts, uint16_t Version,
+ uint8_t AddrSize) {
+ uint64_t Offset = 0;
+ while (AddrData.isValidOffset(Offset)) {
+ DWARFDebugAddrTable AddrTable;
+ uint64_t TableOffset = Offset;
+ if (Error Err = AddrTable.extract(AddrData, &Offset, Version, AddrSize,
+ DWARFContext::dumpWarning)) {
+ WithColor::error() << toString(std::move(Err)) << '\n';
+ // Keep going after an error, if we can, assuming that the length field
+ // could be read. If it couldn't, stop reading the section.
+ if (!AddrTable.hasValidLength())
+ break;
+ Offset = TableOffset + AddrTable.getLength();
+ } else {
+ AddrTable.dump(OS, DumpOpts);
+ }
+ }
+}
+
+// Dump the .debug_rnglists or .debug_rnglists.dwo section (DWARF v5).
+static void dumpRnglistsSection(
+ raw_ostream &OS, DWARFDataExtractor &rnglistData,
+ llvm::function_ref<Optional<object::SectionedAddress>(uint32_t)>
+ LookupPooledAddress,
+ DIDumpOptions DumpOpts) {
+ uint64_t Offset = 0;
+ while (rnglistData.isValidOffset(Offset)) {
+ llvm::DWARFDebugRnglistTable Rnglists;
+ uint64_t TableOffset = Offset;
+ if (Error Err = Rnglists.extract(rnglistData, &Offset)) {
+ WithColor::error() << toString(std::move(Err)) << '\n';
+ uint64_t Length = Rnglists.length();
+ // Keep going after an error, if we can, assuming that the length field
+ // could be read. If it couldn't, stop reading the section.
+ if (Length == 0)
+ break;
+ Offset = TableOffset + Length;
+ } else {
+ Rnglists.dump(OS, LookupPooledAddress, DumpOpts);
+ }
+ }
+}
+
+static void dumpLoclistsSection(raw_ostream &OS, DIDumpOptions DumpOpts,
+ DWARFDataExtractor Data,
+ const MCRegisterInfo *MRI,
+ Optional<uint64_t> DumpOffset) {
+ uint64_t Offset = 0;
+
+ while (Data.isValidOffset(Offset)) {
+ DWARFListTableHeader Header(".debug_loclists", "locations");
+ if (Error E = Header.extract(Data, &Offset)) {
+ WithColor::error() << toString(std::move(E)) << '\n';
+ return;
+ }
+
+ Header.dump(OS, DumpOpts);
+
+ uint64_t EndOffset = Header.length() + Header.getHeaderOffset();
+ Data.setAddressSize(Header.getAddrSize());
+ if (DumpOffset) {
+ if (DumpOffset >= Offset && DumpOffset < EndOffset) {
+ Offset = *DumpOffset;
+ DWARFDebugLoclists::dumpLocationList(Data, &Offset, Header.getVersion(),
+ OS, /*BaseAddr=*/0, MRI, nullptr,
+ DumpOpts, /*Indent=*/0);
+ OS << "\n";
+ return;
+ }
+ } else {
+ DWARFDebugLoclists::dumpRange(Data, Offset, EndOffset - Offset,
+ Header.getVersion(), OS, 0, MRI, DumpOpts);
+ }
+ Offset = EndOffset;
+ }
+}
+
+void DWARFContext::dump(
+ raw_ostream &OS, DIDumpOptions DumpOpts,
+ std::array<Optional<uint64_t>, DIDT_ID_Count> DumpOffsets) {
+ uint64_t DumpType = DumpOpts.DumpType;
+
+ StringRef Extension = sys::path::extension(DObj->getFileName());
+ bool IsDWO = (Extension == ".dwo") || (Extension == ".dwp");
+
+#if 0 // XXX BINARYEN
+ // Print UUID header.
+ const auto *ObjFile = DObj->getFile();
+ if (DumpType & DIDT_UUID)
+ dumpUUID(OS, *ObjFile);
+#endif
+
+ // Print a header for each explicitly-requested section.
+ // Otherwise just print one for non-empty sections.
+ // Only print empty .dwo section headers when dumping a .dwo file.
+ bool Explicit = DumpType != DIDT_All && !IsDWO;
+ bool ExplicitDWO = Explicit && IsDWO;
+ auto shouldDump = [&](bool Explicit, const char *Name, unsigned ID,
+ StringRef Section) -> Optional<uint64_t> * {
+
+ unsigned Mask = 1U << ID;
+ bool Should = (DumpType & Mask) && (Explicit || !Section.empty());
+ if (!Should)
+ return nullptr;
+ OS << "\n" << Name << " contents:\n";
+ return &DumpOffsets[ID];
+ };
+
+ // Dump individual sections.
+ if (shouldDump(Explicit, ".debug_abbrev", DIDT_ID_DebugAbbrev,
+ DObj->getAbbrevSection()))
+ getDebugAbbrev()->dump(OS);
+ if (shouldDump(ExplicitDWO, ".debug_abbrev.dwo", DIDT_ID_DebugAbbrev,
+ DObj->getAbbrevDWOSection()))
+ getDebugAbbrevDWO()->dump(OS);
+
+ auto dumpDebugInfo = [&](const char *Name, unit_iterator_range Units) {
+ OS << '\n' << Name << " contents:\n";
+ if (auto DumpOffset = DumpOffsets[DIDT_ID_DebugInfo])
+ for (const auto &U : Units)
+ U->getDIEForOffset(DumpOffset.getValue())
+ .dump(OS, 0, DumpOpts.noImplicitRecursion());
+ else
+ for (const auto &U : Units)
+ U->dump(OS, DumpOpts);
+ };
+ if ((DumpType & DIDT_DebugInfo)) {
+ if (Explicit || getNumCompileUnits())
+ dumpDebugInfo(".debug_info", info_section_units());
+ if (ExplicitDWO || getNumDWOCompileUnits())
+ dumpDebugInfo(".debug_info.dwo", dwo_info_section_units());
+ }
+
+ auto dumpDebugType = [&](const char *Name, unit_iterator_range Units) {
+ OS << '\n' << Name << " contents:\n";
+ for (const auto &U : Units)
+ if (auto DumpOffset = DumpOffsets[DIDT_ID_DebugTypes])
+ U->getDIEForOffset(*DumpOffset)
+ .dump(OS, 0, DumpOpts.noImplicitRecursion());
+ else
+ U->dump(OS, DumpOpts);
+ };
+ if ((DumpType & DIDT_DebugTypes)) {
+ if (Explicit || getNumTypeUnits())
+ dumpDebugType(".debug_types", types_section_units());
+ if (ExplicitDWO || getNumDWOTypeUnits())
+ dumpDebugType(".debug_types.dwo", dwo_types_section_units());
+ }
+
+ if (const auto *Off = shouldDump(Explicit, ".debug_loc", DIDT_ID_DebugLoc,
+ DObj->getLocSection().Data)) {
+ getDebugLoc()->dump(OS, getRegisterInfo(), DumpOpts, *Off);
+ }
+ if (const auto *Off =
+ shouldDump(Explicit, ".debug_loclists", DIDT_ID_DebugLoclists,
+ DObj->getLoclistsSection().Data)) {
+ DWARFDataExtractor Data(*DObj, DObj->getLoclistsSection(), isLittleEndian(),
+ 0);
+ dumpLoclistsSection(OS, DumpOpts, Data, getRegisterInfo(), *Off);
+ }
+ if (const auto *Off =
+ shouldDump(ExplicitDWO, ".debug_loc.dwo", DIDT_ID_DebugLoc,
+ DObj->getLocDWOSection().Data)) {
+ DWARFDataExtractor Data(*DObj, DObj->getLocDWOSection(), isLittleEndian(),
+ 4);
+ if (*Off) {
+ uint64_t Offset = **Off;
+ DWARFDebugLoclists::dumpLocationList(Data, &Offset, /*Version=*/4, OS,
+ /*BaseAddr=*/0, getRegisterInfo(),
+ nullptr, DumpOpts, /*Indent=*/0);
+ OS << "\n";
+ } else {
+ DWARFDebugLoclists::dumpRange(Data, 0, Data.getData().size(),
+ /*Version=*/4, OS, /*BaseAddr=*/0,
+ getRegisterInfo(), DumpOpts);
+ }
+ }
+
+ if (const auto *Off = shouldDump(Explicit, ".debug_frame", DIDT_ID_DebugFrame,
+ DObj->getFrameSection().Data))
+ getDebugFrame()->dump(OS, getRegisterInfo(), *Off);
+
+ if (const auto *Off = shouldDump(Explicit, ".eh_frame", DIDT_ID_DebugFrame,
+ DObj->getEHFrameSection().Data))
+ getEHFrame()->dump(OS, getRegisterInfo(), *Off);
+
+ if (DumpType & DIDT_DebugMacro) {
+ if (Explicit || !getDebugMacro()->empty()) {
+ OS << "\n.debug_macinfo contents:\n";
+ getDebugMacro()->dump(OS);
+ }
+ }
+
+ if (shouldDump(Explicit, ".debug_aranges", DIDT_ID_DebugAranges,
+ DObj->getArangesSection())) {
+ uint64_t offset = 0;
+ DataExtractor arangesData(DObj->getArangesSection(), isLittleEndian(), 0);
+ DWARFDebugArangeSet set;
+ while (set.extract(arangesData, &offset))
+ set.dump(OS);
+ }
+
+ auto DumpLineSection = [&](DWARFDebugLine::SectionParser Parser,
+ DIDumpOptions DumpOpts,
+ Optional<uint64_t> DumpOffset) {
+ while (!Parser.done()) {
+ if (DumpOffset && Parser.getOffset() != *DumpOffset) {
+ Parser.skip(dumpWarning);
+ continue;
+ }
+ OS << "debug_line[" << format("0x%8.8" PRIx64, Parser.getOffset())
+ << "]\n";
+ if (DumpOpts.Verbose) {
+ Parser.parseNext(dumpWarning, dumpWarning, &OS);
+ } else {
+ DWARFDebugLine::LineTable LineTable =
+ Parser.parseNext(dumpWarning, dumpWarning);
+ LineTable.dump(OS, DumpOpts);
+ }
+ }
+ };
+
+ if (const auto *Off = shouldDump(Explicit, ".debug_line", DIDT_ID_DebugLine,
+ DObj->getLineSection().Data)) {
+ DWARFDataExtractor LineData(*DObj, DObj->getLineSection(), isLittleEndian(),
+ 0);
+ DWARFDebugLine::SectionParser Parser(LineData, *this, compile_units(),
+ type_units());
+ DumpLineSection(Parser, DumpOpts, *Off);
+ }
+
+ if (const auto *Off =
+ shouldDump(ExplicitDWO, ".debug_line.dwo", DIDT_ID_DebugLine,
+ DObj->getLineDWOSection().Data)) {
+ DWARFDataExtractor LineData(*DObj, DObj->getLineDWOSection(),
+ isLittleEndian(), 0);
+ DWARFDebugLine::SectionParser Parser(LineData, *this, dwo_compile_units(),
+ dwo_type_units());
+ DumpLineSection(Parser, DumpOpts, *Off);
+ }
+
+ if (shouldDump(Explicit, ".debug_cu_index", DIDT_ID_DebugCUIndex,
+ DObj->getCUIndexSection())) {
+ getCUIndex().dump(OS);
+ }
+
+ if (shouldDump(Explicit, ".debug_tu_index", DIDT_ID_DebugTUIndex,
+ DObj->getTUIndexSection())) {
+ getTUIndex().dump(OS);
+ }
+
+ if (shouldDump(Explicit, ".debug_str", DIDT_ID_DebugStr,
+ DObj->getStrSection())) {
+ DataExtractor strData(DObj->getStrSection(), isLittleEndian(), 0);
+ uint64_t offset = 0;
+ uint64_t strOffset = 0;
+ while (const char *s = strData.getCStr(&offset)) {
+ OS << format("0x%8.8" PRIx64 ": \"%s\"\n", strOffset, s);
+ strOffset = offset;
+ }
+ }
+ if (shouldDump(ExplicitDWO, ".debug_str.dwo", DIDT_ID_DebugStr,
+ DObj->getStrDWOSection())) {
+ DataExtractor strDWOData(DObj->getStrDWOSection(), isLittleEndian(), 0);
+ uint64_t offset = 0;
+ uint64_t strDWOOffset = 0;
+ while (const char *s = strDWOData.getCStr(&offset)) {
+ OS << format("0x%8.8" PRIx64 ": \"%s\"\n", strDWOOffset, s);
+ strDWOOffset = offset;
+ }
+ }
+ if (shouldDump(Explicit, ".debug_line_str", DIDT_ID_DebugLineStr,
+ DObj->getLineStrSection())) {
+ DataExtractor strData(DObj->getLineStrSection(), isLittleEndian(), 0);
+ uint64_t offset = 0;
+ uint64_t strOffset = 0;
+ while (const char *s = strData.getCStr(&offset)) {
+ OS << format("0x%8.8" PRIx64 ": \"", strOffset);
+ OS.write_escaped(s);
+ OS << "\"\n";
+ strOffset = offset;
+ }
+ }
+
+ if (shouldDump(Explicit, ".debug_addr", DIDT_ID_DebugAddr,
+ DObj->getAddrSection().Data)) {
+ DWARFDataExtractor AddrData(*DObj, DObj->getAddrSection(),
+ isLittleEndian(), 0);
+ dumpAddrSection(OS, AddrData, DumpOpts, getMaxVersion(), getCUAddrSize());
+ }
+
+ if (shouldDump(Explicit, ".debug_ranges", DIDT_ID_DebugRanges,
+ DObj->getRangesSection().Data)) {
+ uint8_t savedAddressByteSize = getCUAddrSize();
+ DWARFDataExtractor rangesData(*DObj, DObj->getRangesSection(),
+ isLittleEndian(), savedAddressByteSize);
+ uint64_t offset = 0;
+ DWARFDebugRangeList rangeList;
+ while (rangesData.isValidOffset(offset)) {
+ if (Error E = rangeList.extract(rangesData, &offset)) {
+ WithColor::error() << toString(std::move(E)) << '\n';
+ break;
+ }
+ rangeList.dump(OS);
+ }
+ }
+
+ auto LookupPooledAddress = [&](uint32_t Index) -> Optional<SectionedAddress> {
+ const auto &CUs = compile_units();
+ auto I = CUs.begin();
+ if (I == CUs.end())
+ return None;
+ return (*I)->getAddrOffsetSectionItem(Index);
+ };
+
+ if (shouldDump(Explicit, ".debug_rnglists", DIDT_ID_DebugRnglists,
+ DObj->getRnglistsSection().Data)) {
+ DWARFDataExtractor RnglistData(*DObj, DObj->getRnglistsSection(),
+ isLittleEndian(), 0);
+ dumpRnglistsSection(OS, RnglistData, LookupPooledAddress, DumpOpts);
+ }
+
+ if (shouldDump(ExplicitDWO, ".debug_rnglists.dwo", DIDT_ID_DebugRnglists,
+ DObj->getRnglistsDWOSection().Data)) {
+ DWARFDataExtractor RnglistData(*DObj, DObj->getRnglistsDWOSection(),
+ isLittleEndian(), 0);
+ dumpRnglistsSection(OS, RnglistData, LookupPooledAddress, DumpOpts);
+ }
+
+ if (shouldDump(Explicit, ".debug_pubnames", DIDT_ID_DebugPubnames,
+ DObj->getPubnamesSection().Data))
+ DWARFDebugPubTable(*DObj, DObj->getPubnamesSection(), isLittleEndian(), false)
+ .dump(OS);
+
+ if (shouldDump(Explicit, ".debug_pubtypes", DIDT_ID_DebugPubtypes,
+ DObj->getPubtypesSection().Data))
+ DWARFDebugPubTable(*DObj, DObj->getPubtypesSection(), isLittleEndian(), false)
+ .dump(OS);
+
+ if (shouldDump(Explicit, ".debug_gnu_pubnames", DIDT_ID_DebugGnuPubnames,
+ DObj->getGnuPubnamesSection().Data))
+ DWARFDebugPubTable(*DObj, DObj->getGnuPubnamesSection(), isLittleEndian(),
+ true /* GnuStyle */)
+ .dump(OS);
+
+ if (shouldDump(Explicit, ".debug_gnu_pubtypes", DIDT_ID_DebugGnuPubtypes,
+ DObj->getGnuPubtypesSection().Data))
+ DWARFDebugPubTable(*DObj, DObj->getGnuPubtypesSection(), isLittleEndian(),
+ true /* GnuStyle */)
+ .dump(OS);
+
+ if (shouldDump(Explicit, ".debug_str_offsets", DIDT_ID_DebugStrOffsets,
+ DObj->getStrOffsetsSection().Data))
+ dumpStringOffsetsSection(OS, "debug_str_offsets", *DObj,
+ DObj->getStrOffsetsSection(),
+ DObj->getStrSection(), normal_units(),
+ isLittleEndian(), getMaxVersion());
+ if (shouldDump(ExplicitDWO, ".debug_str_offsets.dwo", DIDT_ID_DebugStrOffsets,
+ DObj->getStrOffsetsDWOSection().Data))
+ dumpStringOffsetsSection(OS, "debug_str_offsets.dwo", *DObj,
+ DObj->getStrOffsetsDWOSection(),
+ DObj->getStrDWOSection(), dwo_units(),
+ isLittleEndian(), getMaxDWOVersion());
+
+ if (shouldDump(Explicit, ".gdb_index", DIDT_ID_GdbIndex,
+ DObj->getGdbIndexSection())) {
+ getGdbIndex().dump(OS);
+ }
+
+ if (shouldDump(Explicit, ".apple_names", DIDT_ID_AppleNames,
+ DObj->getAppleNamesSection().Data))
+ getAppleNames().dump(OS);
+
+ if (shouldDump(Explicit, ".apple_types", DIDT_ID_AppleTypes,
+ DObj->getAppleTypesSection().Data))
+ getAppleTypes().dump(OS);
+
+ if (shouldDump(Explicit, ".apple_namespaces", DIDT_ID_AppleNamespaces,
+ DObj->getAppleNamespacesSection().Data))
+ getAppleNamespaces().dump(OS);
+
+ if (shouldDump(Explicit, ".apple_objc", DIDT_ID_AppleObjC,
+ DObj->getAppleObjCSection().Data))
+ getAppleObjC().dump(OS);
+ if (shouldDump(Explicit, ".debug_names", DIDT_ID_DebugNames,
+ DObj->getNamesSection().Data))
+ getDebugNames().dump(OS);
+}
+
+DWARFCompileUnit *DWARFContext::getDWOCompileUnitForHash(uint64_t Hash) {
+ parseDWOUnits(LazyParse);
+
+ if (const auto &CUI = getCUIndex()) {
+ if (const auto *R = CUI.getFromHash(Hash))
+ return dyn_cast_or_null<DWARFCompileUnit>(
+ DWOUnits.getUnitForIndexEntry(*R));
+ return nullptr;
+ }
+
+ // If there's no index, just search through the CUs in the DWO - there's
+ // probably only one unless this is something like LTO - though an in-process
+ // built/cached lookup table could be used in that case to improve repeated
+ // lookups of different CUs in the DWO.
+ for (const auto &DWOCU : dwo_compile_units()) {
+ // Might not have parsed DWO ID yet.
+ if (!DWOCU->getDWOId()) {
+ if (Optional<uint64_t> DWOId =
+ toUnsigned(DWOCU->getUnitDIE().find(DW_AT_GNU_dwo_id)))
+ DWOCU->setDWOId(*DWOId);
+ else
+ // No DWO ID?
+ continue;
+ }
+ if (DWOCU->getDWOId() == Hash)
+ return dyn_cast<DWARFCompileUnit>(DWOCU.get());
+ }
+ return nullptr;
+}
+
+DWARFDie DWARFContext::getDIEForOffset(uint64_t Offset) {
+ parseNormalUnits();
+ if (auto *CU = NormalUnits.getUnitForOffset(Offset))
+ return CU->getDIEForOffset(Offset);
+ return DWARFDie();
+}
+
+bool DWARFContext::verify(raw_ostream &OS, DIDumpOptions DumpOpts) {
+ bool Success = true;
+ DWARFVerifier verifier(OS, *this, DumpOpts);
+
+ Success &= verifier.handleDebugAbbrev();
+ if (DumpOpts.DumpType & DIDT_DebugInfo)
+ Success &= verifier.handleDebugInfo();
+ if (DumpOpts.DumpType & DIDT_DebugLine)
+ Success &= verifier.handleDebugLine();
+ Success &= verifier.handleAccelTables();
+ return Success;
+}
+
+const DWARFUnitIndex &DWARFContext::getCUIndex() {
+ if (CUIndex)
+ return *CUIndex;
+
+ DataExtractor CUIndexData(DObj->getCUIndexSection(), isLittleEndian(), 0);
+
+ CUIndex = std::make_unique<DWARFUnitIndex>(DW_SECT_INFO);
+ CUIndex->parse(CUIndexData);
+ return *CUIndex;
+}
+
+const DWARFUnitIndex &DWARFContext::getTUIndex() {
+ if (TUIndex)
+ return *TUIndex;
+
+ DataExtractor TUIndexData(DObj->getTUIndexSection(), isLittleEndian(), 0);
+
+ TUIndex = std::make_unique<DWARFUnitIndex>(DW_SECT_TYPES);
+ TUIndex->parse(TUIndexData);
+ return *TUIndex;
+}
+
+DWARFGdbIndex &DWARFContext::getGdbIndex() {
+ if (GdbIndex)
+ return *GdbIndex;
+
+ DataExtractor GdbIndexData(DObj->getGdbIndexSection(), true /*LE*/, 0);
+ GdbIndex = std::make_unique<DWARFGdbIndex>();
+ GdbIndex->parse(GdbIndexData);
+ return *GdbIndex;
+}
+
+const DWARFDebugAbbrev *DWARFContext::getDebugAbbrev() {
+ if (Abbrev)
+ return Abbrev.get();
+
+ DataExtractor abbrData(DObj->getAbbrevSection(), isLittleEndian(), 0);
+
+ Abbrev.reset(new DWARFDebugAbbrev());
+ Abbrev->extract(abbrData);
+ return Abbrev.get();
+}
+
+const DWARFDebugAbbrev *DWARFContext::getDebugAbbrevDWO() {
+ if (AbbrevDWO)
+ return AbbrevDWO.get();
+
+ DataExtractor abbrData(DObj->getAbbrevDWOSection(), isLittleEndian(), 0);
+ AbbrevDWO.reset(new DWARFDebugAbbrev());
+ AbbrevDWO->extract(abbrData);
+ return AbbrevDWO.get();
+}
+
+const DWARFDebugLoc *DWARFContext::getDebugLoc() {
+ if (Loc)
+ return Loc.get();
+
+ Loc.reset(new DWARFDebugLoc);
+ // Assume all units have the same address byte size.
+ if (getNumCompileUnits()) {
+ DWARFDataExtractor LocData(*DObj, DObj->getLocSection(), isLittleEndian(),
+ getUnitAtIndex(0)->getAddressByteSize());
+ Loc->parse(LocData);
+ }
+ return Loc.get();
+}
+
+const DWARFDebugAranges *DWARFContext::getDebugAranges() {
+ if (Aranges)
+ return Aranges.get();
+
+ Aranges.reset(new DWARFDebugAranges());
+ Aranges->generate(this);
+ return Aranges.get();
+}
+
+const DWARFDebugFrame *DWARFContext::getDebugFrame() {
+ if (DebugFrame)
+ return DebugFrame.get();
+
+ // There's a "bug" in the DWARFv3 standard with respect to the target address
+ // size within debug frame sections. While DWARF is supposed to be independent
+ // of its container, FDEs have fields with size being "target address size",
+ // which isn't specified in DWARF in general. It's only specified for CUs, but
+ // .eh_frame can appear without a .debug_info section. Follow the example of
+ // other tools (libdwarf) and extract this from the container (ObjectFile
+ // provides this information). This problem is fixed in DWARFv4
+ // See this dwarf-discuss discussion for more details:
+ // http://lists.dwarfstd.org/htdig.cgi/dwarf-discuss-dwarfstd.org/2011-December/001173.html
+ DWARFDataExtractor debugFrameData(*DObj, DObj->getFrameSection(),
+ isLittleEndian(), DObj->getAddressSize());
+ DebugFrame.reset(new DWARFDebugFrame(getArch(), false /* IsEH */));
+ DebugFrame->parse(debugFrameData);
+ return DebugFrame.get();
+}
+
+const DWARFDebugFrame *DWARFContext::getEHFrame() {
+ if (EHFrame)
+ return EHFrame.get();
+
+ DWARFDataExtractor debugFrameData(*DObj, DObj->getEHFrameSection(),
+ isLittleEndian(), DObj->getAddressSize());
+ DebugFrame.reset(new DWARFDebugFrame(getArch(), true /* IsEH */));
+ DebugFrame->parse(debugFrameData);
+ return DebugFrame.get();
+}
+
+const DWARFDebugMacro *DWARFContext::getDebugMacro() {
+ if (Macro)
+ return Macro.get();
+
+ DataExtractor MacinfoData(DObj->getMacinfoSection(), isLittleEndian(), 0);
+ Macro.reset(new DWARFDebugMacro());
+ Macro->parse(MacinfoData);
+ return Macro.get();
+}
+
+template <typename T>
+static T &getAccelTable(std::unique_ptr<T> &Cache, const DWARFObject &Obj,
+ const DWARFSection &Section, StringRef StringSection,
+ bool IsLittleEndian) {
+ if (Cache)
+ return *Cache;
+ DWARFDataExtractor AccelSection(Obj, Section, IsLittleEndian, 0);
+ DataExtractor StrData(StringSection, IsLittleEndian, 0);
+ Cache.reset(new T(AccelSection, StrData));
+ if (Error E = Cache->extract())
+ llvm::consumeError(std::move(E));
+ return *Cache;
+}
+
+const DWARFDebugNames &DWARFContext::getDebugNames() {
+ return getAccelTable(Names, *DObj, DObj->getNamesSection(),
+ DObj->getStrSection(), isLittleEndian());
+}
+
+const AppleAcceleratorTable &DWARFContext::getAppleNames() {
+ return getAccelTable(AppleNames, *DObj, DObj->getAppleNamesSection(),
+ DObj->getStrSection(), isLittleEndian());
+}
+
+const AppleAcceleratorTable &DWARFContext::getAppleTypes() {
+ return getAccelTable(AppleTypes, *DObj, DObj->getAppleTypesSection(),
+ DObj->getStrSection(), isLittleEndian());
+}
+
+const AppleAcceleratorTable &DWARFContext::getAppleNamespaces() {
+ return getAccelTable(AppleNamespaces, *DObj,
+ DObj->getAppleNamespacesSection(),
+ DObj->getStrSection(), isLittleEndian());
+}
+
+const AppleAcceleratorTable &DWARFContext::getAppleObjC() {
+ return getAccelTable(AppleObjC, *DObj, DObj->getAppleObjCSection(),
+ DObj->getStrSection(), isLittleEndian());
+}
+
+const DWARFDebugLine::LineTable *
+DWARFContext::getLineTableForUnit(DWARFUnit *U) {
+ Expected<const DWARFDebugLine::LineTable *> ExpectedLineTable =
+ getLineTableForUnit(U, dumpWarning);
+ if (!ExpectedLineTable) {
+ dumpWarning(ExpectedLineTable.takeError());
+ return nullptr;
+ }
+ return *ExpectedLineTable;
+}
+
+Expected<const DWARFDebugLine::LineTable *> DWARFContext::getLineTableForUnit(
+ DWARFUnit *U, std::function<void(Error)> RecoverableErrorCallback) {
+ if (!Line)
+ Line.reset(new DWARFDebugLine);
+
+ auto UnitDIE = U->getUnitDIE();
+ if (!UnitDIE)
+ return nullptr;
+
+ auto Offset = toSectionOffset(UnitDIE.find(DW_AT_stmt_list));
+ if (!Offset)
+ return nullptr; // No line table for this compile unit.
+
+ uint64_t stmtOffset = *Offset + U->getLineTableOffset();
+ // See if the line table is cached.
+ if (const DWARFLineTable *lt = Line->getLineTable(stmtOffset))
+ return lt;
+
+ // Make sure the offset is good before we try to parse.
+ if (stmtOffset >= U->getLineSection().Data.size())
+ return nullptr;
+
+ // We have to parse it first.
+ DWARFDataExtractor lineData(*DObj, U->getLineSection(), isLittleEndian(),
+ U->getAddressByteSize());
+ return Line->getOrParseLineTable(lineData, stmtOffset, *this, U,
+ RecoverableErrorCallback);
+}
+
+void DWARFContext::parseNormalUnits() {
+ if (!NormalUnits.empty())
+ return;
+ DObj->forEachInfoSections([&](const DWARFSection &S) {
+ NormalUnits.addUnitsForSection(*this, S, DW_SECT_INFO);
+ });
+ NormalUnits.finishedInfoUnits();
+ DObj->forEachTypesSections([&](const DWARFSection &S) {
+ NormalUnits.addUnitsForSection(*this, S, DW_SECT_TYPES);
+ });
+}
+
+void DWARFContext::parseDWOUnits(bool Lazy) {
+ if (!DWOUnits.empty())
+ return;
+ DObj->forEachInfoDWOSections([&](const DWARFSection &S) {
+ DWOUnits.addUnitsForDWOSection(*this, S, DW_SECT_INFO, Lazy);
+ });
+ DWOUnits.finishedInfoUnits();
+ DObj->forEachTypesDWOSections([&](const DWARFSection &S) {
+ DWOUnits.addUnitsForDWOSection(*this, S, DW_SECT_TYPES, Lazy);
+ });
+}
+
+DWARFCompileUnit *DWARFContext::getCompileUnitForOffset(uint64_t Offset) {
+ parseNormalUnits();
+ return dyn_cast_or_null<DWARFCompileUnit>(
+ NormalUnits.getUnitForOffset(Offset));
+}
+
+DWARFCompileUnit *DWARFContext::getCompileUnitForAddress(uint64_t Address) {
+ // First, get the offset of the compile unit.
+ uint64_t CUOffset = getDebugAranges()->findAddress(Address);
+ // Retrieve the compile unit.
+ return getCompileUnitForOffset(CUOffset);
+}
+
+DWARFContext::DIEsForAddress DWARFContext::getDIEsForAddress(uint64_t Address) {
+ DIEsForAddress Result;
+
+ DWARFCompileUnit *CU = getCompileUnitForAddress(Address);
+ if (!CU)
+ return Result;
+
+ Result.CompileUnit = CU;
+ Result.FunctionDIE = CU->getSubroutineForAddress(Address);
+
+ std::vector<DWARFDie> Worklist;
+ Worklist.push_back(Result.FunctionDIE);
+ while (!Worklist.empty()) {
+ DWARFDie DIE = Worklist.back();
+ Worklist.pop_back();
+
+ if (!DIE.isValid())
+ continue;
+
+ if (DIE.getTag() == DW_TAG_lexical_block &&
+ DIE.addressRangeContainsAddress(Address)) {
+ Result.BlockDIE = DIE;
+ break;
+ }
+
+ for (auto Child : DIE)
+ Worklist.push_back(Child);
+ }
+
+ return Result;
+}
+
+/// TODO: change input parameter from "uint64_t Address"
+/// into "SectionedAddress Address"
+static bool getFunctionNameAndStartLineForAddress(DWARFCompileUnit *CU,
+ uint64_t Address,
+ FunctionNameKind Kind,
+ std::string &FunctionName,
+ uint32_t &StartLine) {
+ // The address may correspond to instruction in some inlined function,
+ // so we have to build the chain of inlined functions and take the
+ // name of the topmost function in it.
+ SmallVector<DWARFDie, 4> InlinedChain;
+ CU->getInlinedChainForAddress(Address, InlinedChain);
+ if (InlinedChain.empty())
+ return false;
+
+ const DWARFDie &DIE = InlinedChain[0];
+ bool FoundResult = false;
+ const char *Name = nullptr;
+ if (Kind != FunctionNameKind::None && (Name = DIE.getSubroutineName(Kind))) {
+ FunctionName = Name;
+ FoundResult = true;
+ }
+ if (auto DeclLineResult = DIE.getDeclLine()) {
+ StartLine = DeclLineResult;
+ FoundResult = true;
+ }
+
+ return FoundResult;
+}
+
+static Optional<uint64_t> getTypeSize(DWARFDie Type, uint64_t PointerSize) {
+ if (auto SizeAttr = Type.find(DW_AT_byte_size))
+ if (Optional<uint64_t> Size = SizeAttr->getAsUnsignedConstant())
+ return Size;
+
+ switch (Type.getTag()) {
+ case DW_TAG_pointer_type:
+ case DW_TAG_reference_type:
+ case DW_TAG_rvalue_reference_type:
+ return PointerSize;
+ case DW_TAG_ptr_to_member_type: {
+ if (DWARFDie BaseType = Type.getAttributeValueAsReferencedDie(DW_AT_type))
+ if (BaseType.getTag() == DW_TAG_subroutine_type)
+ return 2 * PointerSize;
+ return PointerSize;
+ }
+ case DW_TAG_const_type:
+ case DW_TAG_volatile_type:
+ case DW_TAG_restrict_type:
+ case DW_TAG_typedef: {
+ if (DWARFDie BaseType = Type.getAttributeValueAsReferencedDie(DW_AT_type))
+ return getTypeSize(BaseType, PointerSize);
+ break;
+ }
+ case DW_TAG_array_type: {
+ DWARFDie BaseType = Type.getAttributeValueAsReferencedDie(DW_AT_type);
+ if (!BaseType)
+ return Optional<uint64_t>();
+ Optional<uint64_t> BaseSize = getTypeSize(BaseType, PointerSize);
+ if (!BaseSize)
+ return Optional<uint64_t>();
+ uint64_t Size = *BaseSize;
+ for (DWARFDie Child : Type) {
+ if (Child.getTag() != DW_TAG_subrange_type)
+ continue;
+
+ if (auto ElemCountAttr = Child.find(DW_AT_count))
+ if (Optional<uint64_t> ElemCount =
+ ElemCountAttr->getAsUnsignedConstant())
+ Size *= *ElemCount;
+ if (auto UpperBoundAttr = Child.find(DW_AT_upper_bound))
+ if (Optional<int64_t> UpperBound =
+ UpperBoundAttr->getAsSignedConstant()) {
+ int64_t LowerBound = 0;
+ if (auto LowerBoundAttr = Child.find(DW_AT_lower_bound))
+ LowerBound = LowerBoundAttr->getAsSignedConstant().getValueOr(0);
+ Size *= *UpperBound - LowerBound + 1;
+ }
+ }
+ return Size;
+ }
+ default:
+ break;
+ }
+ return Optional<uint64_t>();
+}
+
+void DWARFContext::addLocalsForDie(DWARFCompileUnit *CU, DWARFDie Subprogram,
+ DWARFDie Die, std::vector<DILocal> &Result) {
+ if (Die.getTag() == DW_TAG_variable ||
+ Die.getTag() == DW_TAG_formal_parameter) {
+ DILocal Local;
+ if (auto NameAttr = Subprogram.find(DW_AT_name))
+ if (Optional<const char *> Name = NameAttr->getAsCString())
+ Local.FunctionName = *Name;
+ if (auto LocationAttr = Die.find(DW_AT_location))
+ if (Optional<ArrayRef<uint8_t>> Location = LocationAttr->getAsBlock())
+ if (!Location->empty() && (*Location)[0] == DW_OP_fbreg)
+ Local.FrameOffset =
+ decodeSLEB128(Location->data() + 1, nullptr, Location->end());
+ if (auto TagOffsetAttr = Die.find(DW_AT_LLVM_tag_offset))
+ Local.TagOffset = TagOffsetAttr->getAsUnsignedConstant();
+
+ if (auto Origin =
+ Die.getAttributeValueAsReferencedDie(DW_AT_abstract_origin))
+ Die = Origin;
+ if (auto NameAttr = Die.find(DW_AT_name))
+ if (Optional<const char *> Name = NameAttr->getAsCString())
+ Local.Name = *Name;
+ if (auto Type = Die.getAttributeValueAsReferencedDie(DW_AT_type))
+ Local.Size = getTypeSize(Type, getCUAddrSize());
+ if (auto DeclFileAttr = Die.find(DW_AT_decl_file)) {
+ if (const auto *LT = CU->getContext().getLineTableForUnit(CU))
+ LT->getFileNameByIndex(
+ DeclFileAttr->getAsUnsignedConstant().getValue(),
+ CU->getCompilationDir(),
+ DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath,
+ Local.DeclFile);
+ }
+ if (auto DeclLineAttr = Die.find(DW_AT_decl_line))
+ Local.DeclLine = DeclLineAttr->getAsUnsignedConstant().getValue();
+
+ Result.push_back(Local);
+ return;
+ }
+
+ if (Die.getTag() == DW_TAG_inlined_subroutine)
+ if (auto Origin =
+ Die.getAttributeValueAsReferencedDie(DW_AT_abstract_origin))
+ Subprogram = Origin;
+
+ for (auto Child : Die)
+ addLocalsForDie(CU, Subprogram, Child, Result);
+}
+
+std::vector<DILocal>
+DWARFContext::getLocalsForAddress(object::SectionedAddress Address) {
+ std::vector<DILocal> Result;
+ DWARFCompileUnit *CU = getCompileUnitForAddress(Address.Address);
+ if (!CU)
+ return Result;
+
+ DWARFDie Subprogram = CU->getSubroutineForAddress(Address.Address);
+ if (Subprogram.isValid())
+ addLocalsForDie(CU, Subprogram, Subprogram, Result);
+ return Result;
+}
+
+DILineInfo DWARFContext::getLineInfoForAddress(object::SectionedAddress Address,
+ DILineInfoSpecifier Spec) {
+ DILineInfo Result;
+
+ DWARFCompileUnit *CU = getCompileUnitForAddress(Address.Address);
+ if (!CU)
+ return Result;
+
+ getFunctionNameAndStartLineForAddress(CU, Address.Address, Spec.FNKind,
+ Result.FunctionName, Result.StartLine);
+ if (Spec.FLIKind != FileLineInfoKind::None) {
+ if (const DWARFLineTable *LineTable = getLineTableForUnit(CU)) {
+ LineTable->getFileLineInfoForAddress(
+ {Address.Address, Address.SectionIndex}, CU->getCompilationDir(),
+ Spec.FLIKind, Result);
+ }
+ }
+ return Result;
+}
+
+DILineInfoTable DWARFContext::getLineInfoForAddressRange(
+ object::SectionedAddress Address, uint64_t Size, DILineInfoSpecifier Spec) {
+ DILineInfoTable Lines;
+ DWARFCompileUnit *CU = getCompileUnitForAddress(Address.Address);
+ if (!CU)
+ return Lines;
+
+ uint32_t StartLine = 0;
+ std::string FunctionName(DILineInfo::BadString);
+ getFunctionNameAndStartLineForAddress(CU, Address.Address, Spec.FNKind,
+ FunctionName, StartLine);
+
+ // If the Specifier says we don't need FileLineInfo, just
+ // return the top-most function at the starting address.
+ if (Spec.FLIKind == FileLineInfoKind::None) {
+ DILineInfo Result;
+ Result.FunctionName = FunctionName;
+ Result.StartLine = StartLine;
+ Lines.push_back(std::make_pair(Address.Address, Result));
+ return Lines;
+ }
+
+ const DWARFLineTable *LineTable = getLineTableForUnit(CU);
+
+ // Get the index of row we're looking for in the line table.
+ std::vector<uint32_t> RowVector;
+ if (!LineTable->lookupAddressRange({Address.Address, Address.SectionIndex},
+ Size, RowVector)) {
+ return Lines;
+ }
+
+ for (uint32_t RowIndex : RowVector) {
+ // Take file number and line/column from the row.
+ const DWARFDebugLine::Row &Row = LineTable->Rows[RowIndex];
+ DILineInfo Result;
+ LineTable->getFileNameByIndex(Row.File, CU->getCompilationDir(),
+ Spec.FLIKind, Result.FileName);
+ Result.FunctionName = FunctionName;
+ Result.Line = Row.Line;
+ Result.Column = Row.Column;
+ Result.StartLine = StartLine;
+ Lines.push_back(std::make_pair(Row.Address.Address, Result));
+ }
+
+ return Lines;
+}
+
+DIInliningInfo
+DWARFContext::getInliningInfoForAddress(object::SectionedAddress Address,
+ DILineInfoSpecifier Spec) {
+ DIInliningInfo InliningInfo;
+
+ DWARFCompileUnit *CU = getCompileUnitForAddress(Address.Address);
+ if (!CU)
+ return InliningInfo;
+
+ const DWARFLineTable *LineTable = nullptr;
+ SmallVector<DWARFDie, 4> InlinedChain;
+ CU->getInlinedChainForAddress(Address.Address, InlinedChain);
+ if (InlinedChain.size() == 0) {
+ // If there is no DIE for address (e.g. it is in unavailable .dwo file),
+ // try to at least get file/line info from symbol table.
+ if (Spec.FLIKind != FileLineInfoKind::None) {
+ DILineInfo Frame;
+ LineTable = getLineTableForUnit(CU);
+ if (LineTable && LineTable->getFileLineInfoForAddress(
+ {Address.Address, Address.SectionIndex},
+ CU->getCompilationDir(), Spec.FLIKind, Frame))
+ InliningInfo.addFrame(Frame);
+ }
+ return InliningInfo;
+ }
+
+ uint32_t CallFile = 0, CallLine = 0, CallColumn = 0, CallDiscriminator = 0;
+ for (uint32_t i = 0, n = InlinedChain.size(); i != n; i++) {
+ DWARFDie &FunctionDIE = InlinedChain[i];
+ DILineInfo Frame;
+ // Get function name if necessary.
+ if (const char *Name = FunctionDIE.getSubroutineName(Spec.FNKind))
+ Frame.FunctionName = Name;
+ if (auto DeclLineResult = FunctionDIE.getDeclLine())
+ Frame.StartLine = DeclLineResult;
+ if (Spec.FLIKind != FileLineInfoKind::None) {
+ if (i == 0) {
+ // For the topmost frame, initialize the line table of this
+ // compile unit and fetch file/line info from it.
+ LineTable = getLineTableForUnit(CU);
+ // For the topmost routine, get file/line info from line table.
+ if (LineTable)
+ LineTable->getFileLineInfoForAddress(
+ {Address.Address, Address.SectionIndex}, CU->getCompilationDir(),
+ Spec.FLIKind, Frame);
+ } else {
+ // Otherwise, use call file, call line and call column from
+ // previous DIE in inlined chain.
+ if (LineTable)
+ LineTable->getFileNameByIndex(CallFile, CU->getCompilationDir(),
+ Spec.FLIKind, Frame.FileName);
+ Frame.Line = CallLine;
+ Frame.Column = CallColumn;
+ Frame.Discriminator = CallDiscriminator;
+ }
+ // Get call file/line/column of a current DIE.
+ if (i + 1 < n) {
+ FunctionDIE.getCallerFrame(CallFile, CallLine, CallColumn,
+ CallDiscriminator);
+ }
+ }
+ InliningInfo.addFrame(Frame);
+ }
+ return InliningInfo;
+}
+
+std::shared_ptr<DWARFContext>
+DWARFContext::getDWOContext(StringRef AbsolutePath) {
+#if 0
+ if (auto S = DWP.lock()) {
+ DWARFContext *Ctxt = S->Context.get();
+ return std::shared_ptr<DWARFContext>(std::move(S), Ctxt);
+ }
+
+ std::weak_ptr<DWOFile> *Entry = &DWOFiles[AbsolutePath];
+
+ if (auto S = Entry->lock()) {
+ DWARFContext *Ctxt = S->Context.get();
+ return std::shared_ptr<DWARFContext>(std::move(S), Ctxt);
+ }
+
+ Expected<OwningBinary<ObjectFile>> Obj = [&] {
+ if (!CheckedForDWP) {
+ SmallString<128> DWPName;
+ auto Obj = object::ObjectFile::createObjectFile(
+ this->DWPName.empty()
+ ? (DObj->getFileName() + ".dwp").toStringRef(DWPName)
+ : StringRef(this->DWPName));
+ if (Obj) {
+ Entry = &DWP;
+ return Obj;
+ } else {
+ CheckedForDWP = true;
+ // TODO: Should this error be handled (maybe in a high verbosity mode)
+ // before falling back to .dwo files?
+ consumeError(Obj.takeError());
+ }
+ }
+
+ return object::ObjectFile::createObjectFile(AbsolutePath);
+ }();
+
+ if (!Obj) {
+ // TODO: Actually report errors helpfully.
+ consumeError(Obj.takeError());
+ return nullptr;
+ }
+
+ auto S = std::make_shared<DWOFile>();
+ S->File = std::move(Obj.get());
+ S->Context = DWARFContext::create(*S->File.getBinary());
+ *Entry = S;
+ auto *Ctxt = S->Context.get();
+ return std::shared_ptr<DWARFContext>(std::move(S), Ctxt);
+#else
+ llvm_unreachable("XXX BINARYEN DWO");
+#endif
+}
+
+static Error createError(const Twine &Reason, llvm::Error E) {
+ return make_error<StringError>(Reason + toString(std::move(E)),
+ inconvertibleErrorCode());
+}
+
+/// SymInfo contains information about symbol: it's address
+/// and section index which is -1LL for absolute symbols.
+struct SymInfo {
+ uint64_t Address;
+ uint64_t SectionIndex;
+};
+
+/// Returns the address of symbol relocation used against and a section index.
+/// Used for futher relocations computation. Symbol's section load address is
+static Expected<SymInfo> getSymbolInfo(const object::ObjectFile &Obj,
+ const RelocationRef &Reloc,
+ const LoadedObjectInfo *L,
+ std::map<SymbolRef, SymInfo> &Cache) {
+ SymInfo Ret = {0, (uint64_t)-1LL};
+ object::section_iterator RSec = Obj.section_end();
+ object::symbol_iterator Sym = Reloc.getSymbol();
+
+ std::map<SymbolRef, SymInfo>::iterator CacheIt = Cache.end();
+ // First calculate the address of the symbol or section as it appears
+ // in the object file
+ if (Sym != Obj.symbol_end()) {
+ bool New;
+ std::tie(CacheIt, New) = Cache.insert({*Sym, {0, 0}});
+ if (!New)
+ return CacheIt->second;
+
+ Expected<uint64_t> SymAddrOrErr = Sym->getAddress();
+ if (!SymAddrOrErr)
+ return createError("failed to compute symbol address: ",
+ SymAddrOrErr.takeError());
+
+ // Also remember what section this symbol is in for later
+ auto SectOrErr = Sym->getSection();
+ if (!SectOrErr)
+ return createError("failed to get symbol section: ",
+ SectOrErr.takeError());
+
+ RSec = *SectOrErr;
+ Ret.Address = *SymAddrOrErr;
+#if 0 // XXX BINARYEN
+ } else if (auto *MObj = dyn_cast<MachOObjectFile>(&Obj)) {
+ RSec = MObj->getRelocationSection(Reloc.getRawDataRefImpl());
+ Ret.Address = RSec->getAddress();
+#endif
+ }
+
+ if (RSec != Obj.section_end())
+ Ret.SectionIndex = RSec->getIndex();
+
+ // If we are given load addresses for the sections, we need to adjust:
+ // SymAddr = (Address of Symbol Or Section in File) -
+ // (Address of Section in File) +
+ // (Load Address of Section)
+ // RSec is now either the section being targeted or the section
+ // containing the symbol being targeted. In either case,
+ // we need to perform the same computation.
+ if (L && RSec != Obj.section_end())
+ if (uint64_t SectionLoadAddress = L->getSectionLoadAddress(*RSec))
+ Ret.Address += SectionLoadAddress - RSec->getAddress();
+
+ if (CacheIt != Cache.end())
+ CacheIt->second = Ret;
+
+ return Ret;
+}
+
+#if 0 // XXX BINARYEN
+static bool isRelocScattered(const object::ObjectFile &Obj,
+ const RelocationRef &Reloc) {
+ const MachOObjectFile *MachObj = dyn_cast<MachOObjectFile>(&Obj);
+ if (!MachObj)
+ return false;
+ // MachO also has relocations that point to sections and
+ // scattered relocations.
+ auto RelocInfo = MachObj->getRelocation(Reloc.getRawDataRefImpl());
+ return MachObj->isRelocationScattered(RelocInfo);
+}
+#endif
+
+ErrorPolicy DWARFContext::defaultErrorHandler(Error E) {
+ WithColor::error() << toString(std::move(E)) << '\n';
+ return ErrorPolicy::Continue;
+}
+
+namespace {
+struct DWARFSectionMap final : public DWARFSection {
+ RelocAddrMap Relocs;
+};
+
+class DWARFObjInMemory final : public DWARFObject {
+ bool IsLittleEndian;
+ uint8_t AddressSize;
+ StringRef FileName;
+ const object::ObjectFile *Obj = nullptr;
+ std::vector<SectionName> SectionNames;
+
+ using InfoSectionMap = MapVector<object::SectionRef, DWARFSectionMap,
+ std::map<object::SectionRef, unsigned>>;
+
+ InfoSectionMap InfoSections;
+ InfoSectionMap TypesSections;
+ InfoSectionMap InfoDWOSections;
+ InfoSectionMap TypesDWOSections;
+
+ DWARFSectionMap LocSection;
+ DWARFSectionMap LoclistsSection;
+ DWARFSectionMap LineSection;
+ DWARFSectionMap RangesSection;
+ DWARFSectionMap RnglistsSection;
+ DWARFSectionMap StrOffsetsSection;
+ DWARFSectionMap LineDWOSection;
+ DWARFSectionMap FrameSection;
+ DWARFSectionMap EHFrameSection;
+ DWARFSectionMap LocDWOSection;
+ DWARFSectionMap StrOffsetsDWOSection;
+ DWARFSectionMap RangesDWOSection;
+ DWARFSectionMap RnglistsDWOSection;
+ DWARFSectionMap AddrSection;
+ DWARFSectionMap AppleNamesSection;
+ DWARFSectionMap AppleTypesSection;
+ DWARFSectionMap AppleNamespacesSection;
+ DWARFSectionMap AppleObjCSection;
+ DWARFSectionMap NamesSection;
+ DWARFSectionMap PubnamesSection;
+ DWARFSectionMap PubtypesSection;
+ DWARFSectionMap GnuPubnamesSection;
+ DWARFSectionMap GnuPubtypesSection;
+
+ DWARFSectionMap *mapNameToDWARFSection(StringRef Name) {
+ return StringSwitch<DWARFSectionMap *>(Name)
+ .Case("debug_loc", &LocSection)
+ .Case("debug_loclists", &LoclistsSection)
+ .Case("debug_line", &LineSection)
+ .Case("debug_frame", &FrameSection)
+ .Case("eh_frame", &EHFrameSection)
+ .Case("debug_str_offsets", &StrOffsetsSection)
+ .Case("debug_ranges", &RangesSection)
+ .Case("debug_rnglists", &RnglistsSection)
+ .Case("debug_loc.dwo", &LocDWOSection)
+ .Case("debug_line.dwo", &LineDWOSection)
+ .Case("debug_names", &NamesSection)
+ .Case("debug_rnglists.dwo", &RnglistsDWOSection)
+ .Case("debug_str_offsets.dwo", &StrOffsetsDWOSection)
+ .Case("debug_addr", &AddrSection)
+ .Case("apple_names", &AppleNamesSection)
+ .Case("debug_pubnames", &PubnamesSection)
+ .Case("debug_pubtypes", &PubtypesSection)
+ .Case("debug_gnu_pubnames", &GnuPubnamesSection)
+ .Case("debug_gnu_pubtypes", &GnuPubtypesSection)
+ .Case("apple_types", &AppleTypesSection)
+ .Case("apple_namespaces", &AppleNamespacesSection)
+ .Case("apple_namespac", &AppleNamespacesSection)
+ .Case("apple_objc", &AppleObjCSection)
+ .Default(nullptr);
+ }
+
+ StringRef AbbrevSection;
+ StringRef ArangesSection;
+ StringRef StrSection;
+ StringRef MacinfoSection;
+ StringRef AbbrevDWOSection;
+ StringRef StrDWOSection;
+ StringRef CUIndexSection;
+ StringRef GdbIndexSection;
+ StringRef TUIndexSection;
+ StringRef LineStrSection;
+
+ // A deque holding section data whose iterators are not invalidated when
+ // new decompressed sections are inserted at the end.
+ std::deque<SmallString<0>> UncompressedSections;
+
+ StringRef *mapSectionToMember(StringRef Name) {
+ if (DWARFSection *Sec = mapNameToDWARFSection(Name))
+ return &Sec->Data;
+ return StringSwitch<StringRef *>(Name)
+ .Case("debug_abbrev", &AbbrevSection)
+ .Case("debug_aranges", &ArangesSection)
+ .Case("debug_str", &StrSection)
+ .Case("debug_macinfo", &MacinfoSection)
+ .Case("debug_abbrev.dwo", &AbbrevDWOSection)
+ .Case("debug_str.dwo", &StrDWOSection)
+ .Case("debug_cu_index", &CUIndexSection)
+ .Case("debug_tu_index", &TUIndexSection)
+ .Case("gdb_index", &GdbIndexSection)
+ .Case("debug_line_str", &LineStrSection)
+ // Any more debug info sections go here.
+ .Default(nullptr);
+ }
+
+ /// If Sec is compressed section, decompresses and updates its contents
+ /// provided by Data. Otherwise leaves it unchanged.
+ Error maybeDecompress(const object::SectionRef &Sec, StringRef Name,
+ StringRef &Data) {
+#if 1 // XXX BINARYEN
+ errs() << "maybeDecompress?\n";
+ return Error::success();
+#else
+ if (!Decompressor::isCompressed(Sec))
+ return Error::success();
+
+ Expected<Decompressor> Decompressor =
+ Decompressor::create(Name, Data, IsLittleEndian, AddressSize == 8);
+ if (!Decompressor)
+ return Decompressor.takeError();
+
+ SmallString<0> Out;
+ if (auto Err = Decompressor->resizeAndDecompress(Out))
+ return Err;
+
+ UncompressedSections.push_back(std::move(Out));
+ Data = UncompressedSections.back();
+
+ return Error::success();
+#endif
+ }
+
+public:
+ DWARFObjInMemory(const StringMap<std::unique_ptr<MemoryBuffer>> &Sections,
+ uint8_t AddrSize, bool IsLittleEndian)
+ : IsLittleEndian(IsLittleEndian) {
+ for (const auto &SecIt : Sections) {
+ if (StringRef *SectionData = mapSectionToMember(SecIt.first()))
+ *SectionData = SecIt.second->getBuffer();
+ else if (SecIt.first() == "debug_info")
+ // Find debug_info and debug_types data by section rather than name as
+ // there are multiple, comdat grouped, of these sections.
+ InfoSections[SectionRef()].Data = SecIt.second->getBuffer();
+ else if (SecIt.first() == "debug_info.dwo")
+ InfoDWOSections[SectionRef()].Data = SecIt.second->getBuffer();
+ else if (SecIt.first() == "debug_types")
+ TypesSections[SectionRef()].Data = SecIt.second->getBuffer();
+ else if (SecIt.first() == "debug_types.dwo")
+ TypesDWOSections[SectionRef()].Data = SecIt.second->getBuffer();
+ }
+ }
+ DWARFObjInMemory(const object::ObjectFile &Obj, const LoadedObjectInfo *L,
+ function_ref<ErrorPolicy(Error)> HandleError)
+ : IsLittleEndian(Obj.isLittleEndian()),
+ AddressSize(Obj.getBytesInAddress()), FileName(Obj.getFileName()),
+ Obj(&Obj) {
+
+ StringMap<unsigned> SectionAmountMap;
+ for (const SectionRef &Section : Obj.sections()) {
+ StringRef Name;
+ if (auto NameOrErr = Section.getName())
+ Name = *NameOrErr;
+ else
+ consumeError(NameOrErr.takeError());
+
+ ++SectionAmountMap[Name];
+ SectionNames.push_back({ Name, true });
+
+ // Skip BSS and Virtual sections, they aren't interesting.
+ if (Section.isBSS() || Section.isVirtual())
+ continue;
+
+ // Skip sections stripped by dsymutil.
+ if (Section.isStripped())
+ continue;
+
+ StringRef Data;
+ Expected<section_iterator> SecOrErr = Section.getRelocatedSection();
+ if (!SecOrErr) {
+ ErrorPolicy EP = HandleError(createError(
+ "failed to get relocated section: ", SecOrErr.takeError()));
+ if (EP == ErrorPolicy::Halt)
+ return;
+ continue;
+ }
+
+ // Try to obtain an already relocated version of this section.
+ // Else use the unrelocated section from the object file. We'll have to
+ // apply relocations ourselves later.
+ section_iterator RelocatedSection = *SecOrErr;
+ if (!L || !L->getLoadedSectionContents(*RelocatedSection, Data)) {
+ Expected<StringRef> E = Section.getContents();
+ if (E)
+ Data = *E;
+ else
+ // maybeDecompress below will error.
+ consumeError(E.takeError());
+ }
+
+ if (auto Err = maybeDecompress(Section, Name, Data)) {
+ ErrorPolicy EP = HandleError(createError(
+ "failed to decompress '" + Name + "', ", std::move(Err)));
+ if (EP == ErrorPolicy::Halt)
+ return;
+ continue;
+ }
+
+ // Compressed sections names in GNU style starts from ".z",
+ // at this point section is decompressed and we drop compression prefix.
+ Name = Name.substr(
+ Name.find_first_not_of("._z")); // Skip ".", "z" and "_" prefixes.
+
+ // Map platform specific debug section names to DWARF standard section
+ // names.
+ Name = Obj.mapDebugSectionName(Name);
+
+ if (StringRef *SectionData = mapSectionToMember(Name)) {
+ *SectionData = Data;
+ if (Name == "debug_ranges") {
+ // FIXME: Use the other dwo range section when we emit it.
+ RangesDWOSection.Data = Data;
+ }
+ } else if (Name == "debug_info") {
+ // Find debug_info and debug_types data by section rather than name as
+ // there are multiple, comdat grouped, of these sections.
+ InfoSections[Section].Data = Data;
+ } else if (Name == "debug_info.dwo") {
+ InfoDWOSections[Section].Data = Data;
+ } else if (Name == "debug_types") {
+ TypesSections[Section].Data = Data;
+ } else if (Name == "debug_types.dwo") {
+ TypesDWOSections[Section].Data = Data;
+ }
+
+ if (RelocatedSection == Obj.section_end())
+ continue;
+
+ StringRef RelSecName;
+ if (auto NameOrErr = RelocatedSection->getName())
+ RelSecName = *NameOrErr;
+ else
+ consumeError(NameOrErr.takeError());
+
+ // If the section we're relocating was relocated already by the JIT,
+ // then we used the relocated version above, so we do not need to process
+ // relocations for it now.
+ StringRef RelSecData;
+ if (L && L->getLoadedSectionContents(*RelocatedSection, RelSecData))
+ continue;
+
+ // In Mach-o files, the relocations do not need to be applied if
+ // there is no load offset to apply. The value read at the
+ // relocation point already factors in the section address
+ // (actually applying the relocations will produce wrong results
+ // as the section address will be added twice).
+ if (!L && isa<MachOObjectFile>(&Obj))
+ continue;
+
+ RelSecName = RelSecName.substr(
+ RelSecName.find_first_not_of("._z")); // Skip . and _ prefixes.
+
+ // TODO: Add support for relocations in other sections as needed.
+ // Record relocations for the debug_info and debug_line sections.
+ DWARFSectionMap *Sec = mapNameToDWARFSection(RelSecName);
+ RelocAddrMap *Map = Sec ? &Sec->Relocs : nullptr;
+ if (!Map) {
+ // Find debug_info and debug_types relocs by section rather than name
+ // as there are multiple, comdat grouped, of these sections.
+ if (RelSecName == "debug_info")
+ Map = &static_cast<DWARFSectionMap &>(InfoSections[*RelocatedSection])
+ .Relocs;
+ else if (RelSecName == "debug_info.dwo")
+ Map = &static_cast<DWARFSectionMap &>(
+ InfoDWOSections[*RelocatedSection])
+ .Relocs;
+ else if (RelSecName == "debug_types")
+ Map =
+ &static_cast<DWARFSectionMap &>(TypesSections[*RelocatedSection])
+ .Relocs;
+ else if (RelSecName == "debug_types.dwo")
+ Map = &static_cast<DWARFSectionMap &>(
+ TypesDWOSections[*RelocatedSection])
+ .Relocs;
+ else
+ continue;
+ }
+
+ if (Section.relocation_begin() == Section.relocation_end())
+ continue;
+
+#if 1 // XXX BINARYEN
+ errs() << "relocation?\n";
+#else
+ // Symbol to [address, section index] cache mapping.
+ std::map<SymbolRef, SymInfo> AddrCache;
+ bool (*Supports)(uint64_t);
+ RelocationResolver Resolver;
+ std::tie(Supports, Resolver) = getRelocationResolver(Obj);
+ for (const RelocationRef &Reloc : Section.relocations()) {
+ // FIXME: it's not clear how to correctly handle scattered
+ // relocations.
+ if (isRelocScattered(Obj, Reloc))
+ continue;
+
+ Expected<SymInfo> SymInfoOrErr =
+ getSymbolInfo(Obj, Reloc, L, AddrCache);
+ if (!SymInfoOrErr) {
+ if (HandleError(SymInfoOrErr.takeError()) == ErrorPolicy::Halt)
+ return;
+ continue;
+ }
+
+ // Check if Resolver can handle this relocation type early so as not to
+ // handle invalid cases in DWARFDataExtractor.
+ //
+ // TODO Don't store Resolver in every RelocAddrEntry.
+ if (Supports && Supports(Reloc.getType())) {
+ auto I = Map->try_emplace(
+ Reloc.getOffset(),
+ RelocAddrEntry{SymInfoOrErr->SectionIndex, Reloc,
+ SymInfoOrErr->Address,
+ Optional<object::RelocationRef>(), 0, Resolver});
+ // If we didn't successfully insert that's because we already had a
+ // relocation for that offset. Store it as a second relocation in the
+ // same RelocAddrEntry instead.
+ if (!I.second) {
+ RelocAddrEntry &entry = I.first->getSecond();
+ if (entry.Reloc2) {
+ ErrorPolicy EP = HandleError(createError(
+ "At most two relocations per offset are supported"));
+ if (EP == ErrorPolicy::Halt)
+ return;
+ }
+ entry.Reloc2 = Reloc;
+ entry.SymbolValue2 = SymInfoOrErr->Address;
+ }
+ } else {
+ SmallString<32> Type;
+ Reloc.getTypeName(Type);
+ ErrorPolicy EP = HandleError(
+ createError("failed to compute relocation: " + Type + ", ",
+ errorCodeToError(object_error::parse_failed)));
+ if (EP == ErrorPolicy::Halt)
+ return;
+ }
+ }
+#endif
+ }
+
+ for (SectionName &S : SectionNames)
+ if (SectionAmountMap[S.Name] > 1)
+ S.IsNameUnique = false;
+ }
+
+ Optional<RelocAddrEntry> find(const DWARFSection &S,
+ uint64_t Pos) const override {
+ auto &Sec = static_cast<const DWARFSectionMap &>(S);
+ RelocAddrMap::const_iterator AI = Sec.Relocs.find(Pos);
+ if (AI == Sec.Relocs.end())
+ return None;
+ return AI->second;
+ }
+
+ const object::ObjectFile *getFile() const override { return Obj; }
+
+ ArrayRef<SectionName> getSectionNames() const override {
+ return SectionNames;
+ }
+
+ bool isLittleEndian() const override { return IsLittleEndian; }
+ StringRef getAbbrevDWOSection() const override { return AbbrevDWOSection; }
+ const DWARFSection &getLineDWOSection() const override {
+ return LineDWOSection;
+ }
+ const DWARFSection &getLocDWOSection() const override {
+ return LocDWOSection;
+ }
+ StringRef getStrDWOSection() const override { return StrDWOSection; }
+ const DWARFSection &getStrOffsetsDWOSection() const override {
+ return StrOffsetsDWOSection;
+ }
+ const DWARFSection &getRangesDWOSection() const override {
+ return RangesDWOSection;
+ }
+ const DWARFSection &getRnglistsDWOSection() const override {
+ return RnglistsDWOSection;
+ }
+ const DWARFSection &getAddrSection() const override { return AddrSection; }
+ StringRef getCUIndexSection() const override { return CUIndexSection; }
+ StringRef getGdbIndexSection() const override { return GdbIndexSection; }
+ StringRef getTUIndexSection() const override { return TUIndexSection; }
+
+ // DWARF v5
+ const DWARFSection &getStrOffsetsSection() const override {
+ return StrOffsetsSection;
+ }
+ StringRef getLineStrSection() const override { return LineStrSection; }
+
+ // Sections for DWARF5 split dwarf proposal.
+ void forEachInfoDWOSections(
+ function_ref<void(const DWARFSection &)> F) const override {
+ for (auto &P : InfoDWOSections)
+ F(P.second);
+ }
+ void forEachTypesDWOSections(
+ function_ref<void(const DWARFSection &)> F) const override {
+ for (auto &P : TypesDWOSections)
+ F(P.second);
+ }
+
+ StringRef getAbbrevSection() const override { return AbbrevSection; }
+ const DWARFSection &getLocSection() const override { return LocSection; }
+ const DWARFSection &getLoclistsSection() const override { return LoclistsSection; }
+ StringRef getArangesSection() const override { return ArangesSection; }
+ const DWARFSection &getFrameSection() const override {
+ return FrameSection;
+ }
+ const DWARFSection &getEHFrameSection() const override {
+ return EHFrameSection;
+ }
+ const DWARFSection &getLineSection() const override { return LineSection; }
+ StringRef getStrSection() const override { return StrSection; }
+ const DWARFSection &getRangesSection() const override { return RangesSection; }
+ const DWARFSection &getRnglistsSection() const override {
+ return RnglistsSection;
+ }
+ StringRef getMacinfoSection() const override { return MacinfoSection; }
+ const DWARFSection &getPubnamesSection() const override { return PubnamesSection; }
+ const DWARFSection &getPubtypesSection() const override { return PubtypesSection; }
+ const DWARFSection &getGnuPubnamesSection() const override {
+ return GnuPubnamesSection;
+ }
+ const DWARFSection &getGnuPubtypesSection() const override {
+ return GnuPubtypesSection;
+ }
+ const DWARFSection &getAppleNamesSection() const override {
+ return AppleNamesSection;
+ }
+ const DWARFSection &getAppleTypesSection() const override {
+ return AppleTypesSection;
+ }
+ const DWARFSection &getAppleNamespacesSection() const override {
+ return AppleNamespacesSection;
+ }
+ const DWARFSection &getAppleObjCSection() const override {
+ return AppleObjCSection;
+ }
+ const DWARFSection &getNamesSection() const override {
+ return NamesSection;
+ }
+
+ StringRef getFileName() const override { return FileName; }
+ uint8_t getAddressSize() const override { return AddressSize; }
+ void forEachInfoSections(
+ function_ref<void(const DWARFSection &)> F) const override {
+ for (auto &P : InfoSections)
+ F(P.second);
+ }
+ void forEachTypesSections(
+ function_ref<void(const DWARFSection &)> F) const override {
+ for (auto &P : TypesSections)
+ F(P.second);
+ }
+};
+} // namespace
+
+#if 0 // XXX BINARYEN
+std::unique_ptr<DWARFContext>
+DWARFContext::create(const object::ObjectFile &Obj, const LoadedObjectInfo *L,
+ function_ref<ErrorPolicy(Error)> HandleError,
+ std::string DWPName) {
+ auto DObj = std::make_unique<DWARFObjInMemory>(Obj, L, HandleError);
+ return std::make_unique<DWARFContext>(std::move(DObj), std::move(DWPName));
+}
+#endif
+
+std::unique_ptr<DWARFContext>
+DWARFContext::create(const StringMap<std::unique_ptr<MemoryBuffer>> &Sections,
+ uint8_t AddrSize, bool isLittleEndian) {
+ auto DObj =
+ std::make_unique<DWARFObjInMemory>(Sections, AddrSize, isLittleEndian);
+ return std::make_unique<DWARFContext>(std::move(DObj), "");
+}
+
+Error DWARFContext::loadRegisterInfo(const object::ObjectFile &Obj) {
+ llvm_unreachable("loadRegisterInfo"); // XXX BINARYEN
+}
+
+uint8_t DWARFContext::getCUAddrSize() {
+ // In theory, different compile units may have different address byte
+ // sizes, but for simplicity we just use the address byte size of the
+ // last compile unit. In practice the address size field is repeated across
+ // various DWARF headers (at least in version 5) to make it easier to dump
+ // them independently, not to enable varying the address size.
+ uint8_t Addr = 0;
+ for (const auto &CU : compile_units()) {
+ Addr = CU->getAddressByteSize();
+ break;
+ }
+ return Addr;
+}
+
+void DWARFContext::dumpWarning(Error Warning) {
+ handleAllErrors(std::move(Warning), [](ErrorInfoBase &Info) {
+ WithColor::warning() << Info.message() << '\n';
+ });
+}
diff --git a/third_party/llvm-project/DWARFDataExtractor.cpp b/third_party/llvm-project/DWARFDataExtractor.cpp
new file mode 100644
index 000000000..53e676bc7
--- /dev/null
+++ b/third_party/llvm-project/DWARFDataExtractor.cpp
@@ -0,0 +1,100 @@
+//===- DWARFDataExtractor.cpp ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+
+using namespace llvm;
+
+uint64_t DWARFDataExtractor::getRelocatedValue(uint32_t Size, uint64_t *Off,
+ uint64_t *SecNdx,
+ Error *Err) const {
+ if (SecNdx)
+ *SecNdx = object::SectionedAddress::UndefSection;
+ if (!Section)
+ return getUnsigned(Off, Size, Err);
+ Optional<RelocAddrEntry> E = Obj->find(*Section, *Off);
+ uint64_t A = getUnsigned(Off, Size, Err);
+ if (!E)
+ return A;
+ if (SecNdx)
+ *SecNdx = E->SectionIndex;
+ uint64_t R = E->Resolver(E->Reloc, E->SymbolValue, A);
+ if (E->Reloc2)
+ R = E->Resolver(*E->Reloc2, E->SymbolValue2, R);
+ return R;
+}
+
+Optional<uint64_t>
+DWARFDataExtractor::getEncodedPointer(uint64_t *Offset, uint8_t Encoding,
+ uint64_t PCRelOffset) const {
+ if (Encoding == dwarf::DW_EH_PE_omit)
+ return None;
+
+ uint64_t Result = 0;
+ uint64_t OldOffset = *Offset;
+ // First get value
+ switch (Encoding & 0x0F) {
+ case dwarf::DW_EH_PE_absptr:
+ switch (getAddressSize()) {
+ case 2:
+ case 4:
+ case 8:
+ Result = getUnsigned(Offset, getAddressSize());
+ break;
+ default:
+ return None;
+ }
+ break;
+ case dwarf::DW_EH_PE_uleb128:
+ Result = getULEB128(Offset);
+ break;
+ case dwarf::DW_EH_PE_sleb128:
+ Result = getSLEB128(Offset);
+ break;
+ case dwarf::DW_EH_PE_udata2:
+ Result = getUnsigned(Offset, 2);
+ break;
+ case dwarf::DW_EH_PE_udata4:
+ Result = getUnsigned(Offset, 4);
+ break;
+ case dwarf::DW_EH_PE_udata8:
+ Result = getUnsigned(Offset, 8);
+ break;
+ case dwarf::DW_EH_PE_sdata2:
+ Result = getSigned(Offset, 2);
+ break;
+ case dwarf::DW_EH_PE_sdata4:
+ Result = getSigned(Offset, 4);
+ break;
+ case dwarf::DW_EH_PE_sdata8:
+ Result = getSigned(Offset, 8);
+ break;
+ default:
+ return None;
+ }
+ // Then add relative offset, if required
+ switch (Encoding & 0x70) {
+ case dwarf::DW_EH_PE_absptr:
+ // do nothing
+ break;
+ case dwarf::DW_EH_PE_pcrel:
+ Result += PCRelOffset;
+ break;
+ case dwarf::DW_EH_PE_datarel:
+ case dwarf::DW_EH_PE_textrel:
+ case dwarf::DW_EH_PE_funcrel:
+ case dwarf::DW_EH_PE_aligned:
+ default:
+ *Offset = OldOffset;
+ return None;
+ }
+
+ return Result;
+}
diff --git a/third_party/llvm-project/DWARFDebugAbbrev.cpp b/third_party/llvm-project/DWARFDebugAbbrev.cpp
new file mode 100644
index 000000000..4afac2f99
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugAbbrev.cpp
@@ -0,0 +1,138 @@
+//===- DWARFDebugAbbrev.cpp -----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cinttypes>
+#include <cstdint>
+
+using namespace llvm;
+
+DWARFAbbreviationDeclarationSet::DWARFAbbreviationDeclarationSet() {
+ clear();
+}
+
+void DWARFAbbreviationDeclarationSet::clear() {
+ Offset = 0;
+ FirstAbbrCode = 0;
+ Decls.clear();
+}
+
+bool DWARFAbbreviationDeclarationSet::extract(DataExtractor Data,
+ uint64_t *OffsetPtr) {
+ clear();
+ const uint64_t BeginOffset = *OffsetPtr;
+ Offset = BeginOffset;
+ DWARFAbbreviationDeclaration AbbrDecl;
+ uint32_t PrevAbbrCode = 0;
+ while (AbbrDecl.extract(Data, OffsetPtr)) {
+ if (FirstAbbrCode == 0) {
+ FirstAbbrCode = AbbrDecl.getCode();
+ } else {
+ if (PrevAbbrCode + 1 != AbbrDecl.getCode()) {
+ // Codes are not consecutive, can't do O(1) lookups.
+ FirstAbbrCode = UINT32_MAX;
+ }
+ }
+ PrevAbbrCode = AbbrDecl.getCode();
+ Decls.push_back(std::move(AbbrDecl));
+ }
+ return BeginOffset != *OffsetPtr;
+}
+
+void DWARFAbbreviationDeclarationSet::dump(raw_ostream &OS) const {
+ for (const auto &Decl : Decls)
+ Decl.dump(OS);
+}
+
+const DWARFAbbreviationDeclaration *
+DWARFAbbreviationDeclarationSet::getAbbreviationDeclaration(
+ uint32_t AbbrCode) const {
+ if (FirstAbbrCode == UINT32_MAX) {
+ for (const auto &Decl : Decls) {
+ if (Decl.getCode() == AbbrCode)
+ return &Decl;
+ }
+ return nullptr;
+ }
+ if (AbbrCode < FirstAbbrCode || AbbrCode >= FirstAbbrCode + Decls.size())
+ return nullptr;
+ return &Decls[AbbrCode - FirstAbbrCode];
+}
+
+DWARFDebugAbbrev::DWARFDebugAbbrev() { clear(); }
+
+void DWARFDebugAbbrev::clear() {
+ AbbrDeclSets.clear();
+ PrevAbbrOffsetPos = AbbrDeclSets.end();
+}
+
+void DWARFDebugAbbrev::extract(DataExtractor Data) {
+ clear();
+ this->Data = Data;
+}
+
+void DWARFDebugAbbrev::parse() const {
+ if (!Data)
+ return;
+ uint64_t Offset = 0;
+ auto I = AbbrDeclSets.begin();
+ while (Data->isValidOffset(Offset)) {
+ while (I != AbbrDeclSets.end() && I->first < Offset)
+ ++I;
+ uint64_t CUAbbrOffset = Offset;
+ DWARFAbbreviationDeclarationSet AbbrDecls;
+ if (!AbbrDecls.extract(*Data, &Offset))
+ break;
+ AbbrDeclSets.insert(I, std::make_pair(CUAbbrOffset, std::move(AbbrDecls)));
+ }
+ Data = None;
+}
+
+void DWARFDebugAbbrev::dump(raw_ostream &OS) const {
+ parse();
+
+ if (AbbrDeclSets.empty()) {
+ OS << "< EMPTY >\n";
+ return;
+ }
+
+ for (const auto &I : AbbrDeclSets) {
+ OS << format("Abbrev table for offset: 0x%8.8" PRIx64 "\n", I.first);
+ I.second.dump(OS);
+ }
+}
+
+const DWARFAbbreviationDeclarationSet*
+DWARFDebugAbbrev::getAbbreviationDeclarationSet(uint64_t CUAbbrOffset) const {
+ const auto End = AbbrDeclSets.end();
+ if (PrevAbbrOffsetPos != End && PrevAbbrOffsetPos->first == CUAbbrOffset) {
+ return &(PrevAbbrOffsetPos->second);
+ }
+
+ const auto Pos = AbbrDeclSets.find(CUAbbrOffset);
+ if (Pos != End) {
+ PrevAbbrOffsetPos = Pos;
+ return &(Pos->second);
+ }
+
+ if (Data && CUAbbrOffset < Data->getData().size()) {
+ uint64_t Offset = CUAbbrOffset;
+ DWARFAbbreviationDeclarationSet AbbrDecls;
+ if (!AbbrDecls.extract(*Data, &Offset))
+ return nullptr;
+ PrevAbbrOffsetPos =
+ AbbrDeclSets.insert(std::make_pair(CUAbbrOffset, std::move(AbbrDecls)))
+ .first;
+ return &PrevAbbrOffsetPos->second;
+ }
+
+ return nullptr;
+}
diff --git a/third_party/llvm-project/DWARFDebugAddr.cpp b/third_party/llvm-project/DWARFDebugAddr.cpp
new file mode 100644
index 000000000..f71543799
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugAddr.cpp
@@ -0,0 +1,182 @@
+//===- DWARFDebugAddr.cpp -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugAddr.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+
+using namespace llvm;
+
+void DWARFDebugAddrTable::clear() {
+ HeaderData = {};
+ Addrs.clear();
+ invalidateLength();
+}
+
+Error DWARFDebugAddrTable::extract(DWARFDataExtractor Data,
+ uint64_t *OffsetPtr,
+ uint16_t Version,
+ uint8_t AddrSize,
+ std::function<void(Error)> WarnCallback) {
+ clear();
+ HeaderOffset = *OffsetPtr;
+ // Read and verify the length field.
+ if (!Data.isValidOffsetForDataOfSize(*OffsetPtr, sizeof(uint32_t)))
+ return createStringError(errc::invalid_argument,
+ "section is not large enough to contain a "
+ ".debug_addr table length at offset 0x%"
+ PRIx64, *OffsetPtr);
+ uint16_t UnitVersion;
+ if (Version == 0) {
+ WarnCallback(createStringError(errc::invalid_argument,
+ "DWARF version is not defined in CU,"
+ " assuming version 5"));
+ UnitVersion = 5;
+ } else {
+ UnitVersion = Version;
+ }
+ // TODO: Add support for DWARF64.
+ Format = dwarf::DwarfFormat::DWARF32;
+ if (UnitVersion >= 5) {
+ HeaderData.Length = Data.getU32(OffsetPtr);
+ if (HeaderData.Length == dwarf::DW_LENGTH_DWARF64) {
+ invalidateLength();
+ return createStringError(errc::not_supported,
+ "DWARF64 is not supported in .debug_addr at offset 0x%" PRIx64,
+ HeaderOffset);
+ }
+ if (HeaderData.Length + sizeof(uint32_t) < sizeof(Header)) {
+ uint32_t TmpLength = getLength();
+ invalidateLength();
+ return createStringError(errc::invalid_argument,
+ ".debug_addr table at offset 0x%" PRIx64
+ " has too small length (0x%" PRIx32
+ ") to contain a complete header",
+ HeaderOffset, TmpLength);
+ }
+ uint64_t End = HeaderOffset + getLength();
+ if (!Data.isValidOffsetForDataOfSize(HeaderOffset, End - HeaderOffset)) {
+ uint32_t TmpLength = getLength();
+ invalidateLength();
+ return createStringError(errc::invalid_argument,
+ "section is not large enough to contain a .debug_addr table "
+ "of length 0x%" PRIx32 " at offset 0x%" PRIx64,
+ TmpLength, HeaderOffset);
+ }
+
+ HeaderData.Version = Data.getU16(OffsetPtr);
+ HeaderData.AddrSize = Data.getU8(OffsetPtr);
+ HeaderData.SegSize = Data.getU8(OffsetPtr);
+ DataSize = getDataSize();
+ } else {
+ HeaderData.Version = UnitVersion;
+ HeaderData.AddrSize = AddrSize;
+ // TODO: Support for non-zero SegSize.
+ HeaderData.SegSize = 0;
+ DataSize = Data.size();
+ }
+
+ // Perform basic validation of the remaining header fields.
+
+ // We support DWARF version 5 for now as well as pre-DWARF5
+ // implementations of .debug_addr table, which doesn't contain a header
+ // and consists only of a series of addresses.
+ if (HeaderData.Version > 5) {
+ return createStringError(errc::not_supported, "version %" PRIu16
+ " of .debug_addr section at offset 0x%" PRIx64 " is not supported",
+ HeaderData.Version, HeaderOffset);
+ }
+ // FIXME: For now we just treat version mismatch as an error,
+ // however the correct way to associate a .debug_addr table
+ // with a .debug_info table is to look at the DW_AT_addr_base
+ // attribute in the info table.
+ if (HeaderData.Version != UnitVersion)
+ return createStringError(errc::invalid_argument,
+ ".debug_addr table at offset 0x%" PRIx64
+ " has version %" PRIu16
+ " which is different from the version suggested"
+ " by the DWARF unit header: %" PRIu16,
+ HeaderOffset, HeaderData.Version, UnitVersion);
+ if (HeaderData.AddrSize != 4 && HeaderData.AddrSize != 8)
+ return createStringError(errc::not_supported,
+ ".debug_addr table at offset 0x%" PRIx64
+ " has unsupported address size %" PRIu8,
+ HeaderOffset, HeaderData.AddrSize);
+ if (HeaderData.AddrSize != AddrSize && AddrSize != 0)
+ return createStringError(errc::invalid_argument,
+ ".debug_addr table at offset 0x%" PRIx64
+ " has address size %" PRIu8
+ " which is different from CU address size %" PRIu8,
+ HeaderOffset, HeaderData.AddrSize, AddrSize);
+
+ // TODO: add support for non-zero segment selector size.
+ if (HeaderData.SegSize != 0)
+ return createStringError(errc::not_supported,
+ ".debug_addr table at offset 0x%" PRIx64
+ " has unsupported segment selector size %" PRIu8,
+ HeaderOffset, HeaderData.SegSize);
+ if (DataSize % HeaderData.AddrSize != 0) {
+ invalidateLength();
+ return createStringError(errc::invalid_argument,
+ ".debug_addr table at offset 0x%" PRIx64
+ " contains data of size %" PRIu32
+ " which is not a multiple of addr size %" PRIu8,
+ HeaderOffset, DataSize, HeaderData.AddrSize);
+ }
+ Data.setAddressSize(HeaderData.AddrSize);
+ uint32_t AddrCount = DataSize / HeaderData.AddrSize;
+ for (uint32_t I = 0; I < AddrCount; ++I)
+ if (HeaderData.AddrSize == 4)
+ Addrs.push_back(Data.getU32(OffsetPtr));
+ else
+ Addrs.push_back(Data.getU64(OffsetPtr));
+ return Error::success();
+}
+
+void DWARFDebugAddrTable::dump(raw_ostream &OS, DIDumpOptions DumpOpts) const {
+ if (DumpOpts.Verbose)
+ OS << format("0x%8.8" PRIx32 ": ", HeaderOffset);
+ OS << format("Addr Section: length = 0x%8.8" PRIx32
+ ", version = 0x%4.4" PRIx16 ", "
+ "addr_size = 0x%2.2" PRIx8 ", seg_size = 0x%2.2" PRIx8 "\n",
+ HeaderData.Length, HeaderData.Version, HeaderData.AddrSize,
+ HeaderData.SegSize);
+
+ if (Addrs.size() > 0) {
+ const char *AddrFmt = (HeaderData.AddrSize == 4) ? "0x%8.8" PRIx64 "\n"
+ : "0x%16.16" PRIx64 "\n";
+ OS << "Addrs: [\n";
+ for (uint64_t Addr : Addrs)
+ OS << format(AddrFmt, Addr);
+ OS << "]\n";
+ }
+}
+
+Expected<uint64_t> DWARFDebugAddrTable::getAddrEntry(uint32_t Index) const {
+ if (Index < Addrs.size())
+ return Addrs[Index];
+ return createStringError(errc::invalid_argument,
+ "Index %" PRIu32 " is out of range of the "
+ ".debug_addr table at offset 0x%" PRIx64,
+ Index, HeaderOffset);
+}
+
+uint32_t DWARFDebugAddrTable::getLength() const {
+ if (HeaderData.Length == 0)
+ return 0;
+ // TODO: DWARF64 support.
+ return HeaderData.Length + sizeof(uint32_t);
+}
+
+uint32_t DWARFDebugAddrTable::getDataSize() const {
+ if (DataSize != 0)
+ return DataSize;
+ if (getLength() == 0)
+ return 0;
+ return getLength() - getHeaderSize();
+}
diff --git a/third_party/llvm-project/DWARFDebugArangeSet.cpp b/third_party/llvm-project/DWARFDebugArangeSet.cpp
new file mode 100644
index 000000000..200b2d52a
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugArangeSet.cpp
@@ -0,0 +1,111 @@
+//===- DWARFDebugArangeSet.cpp --------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cassert>
+#include <cinttypes>
+#include <cstdint>
+#include <cstring>
+
+using namespace llvm;
+
+void DWARFDebugArangeSet::Descriptor::dump(raw_ostream &OS,
+ uint32_t AddressSize) const {
+ OS << format("[0x%*.*" PRIx64 ", ", AddressSize * 2, AddressSize * 2, Address)
+ << format(" 0x%*.*" PRIx64 ")", AddressSize * 2, AddressSize * 2,
+ getEndAddress());
+}
+
+void DWARFDebugArangeSet::clear() {
+ Offset = -1ULL;
+ std::memset(&HeaderData, 0, sizeof(Header));
+ ArangeDescriptors.clear();
+}
+
+bool
+DWARFDebugArangeSet::extract(DataExtractor data, uint64_t *offset_ptr) {
+ if (data.isValidOffset(*offset_ptr)) {
+ ArangeDescriptors.clear();
+ Offset = *offset_ptr;
+
+ // 7.20 Address Range Table
+ //
+ // Each set of entries in the table of address ranges contained in
+ // the .debug_aranges section begins with a header consisting of: a
+ // 4-byte length containing the length of the set of entries for this
+ // compilation unit, not including the length field itself; a 2-byte
+ // version identifier containing the value 2 for DWARF Version 2; a
+ // 4-byte offset into the.debug_infosection; a 1-byte unsigned integer
+ // containing the size in bytes of an address (or the offset portion of
+ // an address for segmented addressing) on the target system; and a
+ // 1-byte unsigned integer containing the size in bytes of a segment
+ // descriptor on the target system. This header is followed by a series
+ // of tuples. Each tuple consists of an address and a length, each in
+ // the size appropriate for an address on the target architecture.
+ HeaderData.Length = data.getU32(offset_ptr);
+ HeaderData.Version = data.getU16(offset_ptr);
+ HeaderData.CuOffset = data.getU32(offset_ptr);
+ HeaderData.AddrSize = data.getU8(offset_ptr);
+ HeaderData.SegSize = data.getU8(offset_ptr);
+
+ // Perform basic validation of the header fields.
+ if (!data.isValidOffsetForDataOfSize(Offset, HeaderData.Length) ||
+ (HeaderData.AddrSize != 4 && HeaderData.AddrSize != 8)) {
+ clear();
+ return false;
+ }
+
+ // The first tuple following the header in each set begins at an offset
+ // that is a multiple of the size of a single tuple (that is, twice the
+ // size of an address). The header is padded, if necessary, to the
+ // appropriate boundary.
+ const uint32_t header_size = *offset_ptr - Offset;
+ const uint32_t tuple_size = HeaderData.AddrSize * 2;
+ uint32_t first_tuple_offset = 0;
+ while (first_tuple_offset < header_size)
+ first_tuple_offset += tuple_size;
+
+ *offset_ptr = Offset + first_tuple_offset;
+
+ Descriptor arangeDescriptor;
+
+ static_assert(sizeof(arangeDescriptor.Address) ==
+ sizeof(arangeDescriptor.Length),
+ "Different datatypes for addresses and sizes!");
+ assert(sizeof(arangeDescriptor.Address) >= HeaderData.AddrSize);
+
+ while (data.isValidOffset(*offset_ptr)) {
+ arangeDescriptor.Address = data.getUnsigned(offset_ptr, HeaderData.AddrSize);
+ arangeDescriptor.Length = data.getUnsigned(offset_ptr, HeaderData.AddrSize);
+
+ // Each set of tuples is terminated by a 0 for the address and 0
+ // for the length.
+ if (arangeDescriptor.Address || arangeDescriptor.Length)
+ ArangeDescriptors.push_back(arangeDescriptor);
+ else
+ break; // We are done if we get a zero address and length
+ }
+
+ return !ArangeDescriptors.empty();
+ }
+ return false;
+}
+
+void DWARFDebugArangeSet::dump(raw_ostream &OS) const {
+ OS << format("Address Range Header: length = 0x%8.8x, version = 0x%4.4x, ",
+ HeaderData.Length, HeaderData.Version)
+ << format("cu_offset = 0x%8.8x, addr_size = 0x%2.2x, seg_size = 0x%2.2x\n",
+ HeaderData.CuOffset, HeaderData.AddrSize, HeaderData.SegSize);
+
+ for (const auto &Desc : ArangeDescriptors) {
+ Desc.dump(OS, HeaderData.AddrSize);
+ OS << '\n';
+ }
+}
diff --git a/third_party/llvm-project/DWARFDebugAranges.cpp b/third_party/llvm-project/DWARFDebugAranges.cpp
new file mode 100644
index 000000000..ca6043109
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugAranges.cpp
@@ -0,0 +1,122 @@
+//===- DWARFDebugAranges.cpp ----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugAranges.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/WithColor.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+#include <set>
+#include <vector>
+
+using namespace llvm;
+
+void DWARFDebugAranges::extract(DataExtractor DebugArangesData) {
+ if (!DebugArangesData.isValidOffset(0))
+ return;
+ uint64_t Offset = 0;
+ DWARFDebugArangeSet Set;
+
+ while (Set.extract(DebugArangesData, &Offset)) {
+ uint64_t CUOffset = Set.getCompileUnitDIEOffset();
+ for (const auto &Desc : Set.descriptors()) {
+ uint64_t LowPC = Desc.Address;
+ uint64_t HighPC = Desc.getEndAddress();
+ appendRange(CUOffset, LowPC, HighPC);
+ }
+ ParsedCUOffsets.insert(CUOffset);
+ }
+}
+
+void DWARFDebugAranges::generate(DWARFContext *CTX) {
+ clear();
+ if (!CTX)
+ return;
+
+ // Extract aranges from .debug_aranges section.
+ DataExtractor ArangesData(CTX->getDWARFObj().getArangesSection(),
+ CTX->isLittleEndian(), 0);
+ extract(ArangesData);
+
+ // Generate aranges from DIEs: even if .debug_aranges section is present,
+ // it may describe only a small subset of compilation units, so we need to
+ // manually build aranges for the rest of them.
+ for (const auto &CU : CTX->compile_units()) {
+ uint64_t CUOffset = CU->getOffset();
+ if (ParsedCUOffsets.insert(CUOffset).second) {
+ Expected<DWARFAddressRangesVector> CURanges = CU->collectAddressRanges();
+ if (!CURanges)
+ WithColor::error() << toString(CURanges.takeError()) << '\n';
+ else
+ for (const auto &R : *CURanges)
+ appendRange(CUOffset, R.LowPC, R.HighPC);
+ }
+ }
+
+ construct();
+}
+
+void DWARFDebugAranges::clear() {
+ Endpoints.clear();
+ Aranges.clear();
+ ParsedCUOffsets.clear();
+}
+
+void DWARFDebugAranges::appendRange(uint64_t CUOffset, uint64_t LowPC,
+ uint64_t HighPC) {
+ if (LowPC >= HighPC)
+ return;
+ Endpoints.emplace_back(LowPC, CUOffset, true);
+ Endpoints.emplace_back(HighPC, CUOffset, false);
+}
+
+void DWARFDebugAranges::construct() {
+ std::multiset<uint64_t> ValidCUs; // Maintain the set of CUs describing
+ // a current address range.
+ llvm::sort(Endpoints);
+ uint64_t PrevAddress = -1ULL;
+ for (const auto &E : Endpoints) {
+ if (PrevAddress < E.Address && !ValidCUs.empty()) {
+ // If the address range between two endpoints is described by some
+ // CU, first try to extend the last range in Aranges. If we can't
+ // do it, start a new range.
+ if (!Aranges.empty() && Aranges.back().HighPC() == PrevAddress &&
+ ValidCUs.find(Aranges.back().CUOffset) != ValidCUs.end()) {
+ Aranges.back().setHighPC(E.Address);
+ } else {
+ Aranges.emplace_back(PrevAddress, E.Address, *ValidCUs.begin());
+ }
+ }
+ // Update the set of valid CUs.
+ if (E.IsRangeStart) {
+ ValidCUs.insert(E.CUOffset);
+ } else {
+ auto CUPos = ValidCUs.find(E.CUOffset);
+ assert(CUPos != ValidCUs.end());
+ ValidCUs.erase(CUPos);
+ }
+ PrevAddress = E.Address;
+ }
+ assert(ValidCUs.empty());
+
+ // Endpoints are not needed now.
+ Endpoints.clear();
+ Endpoints.shrink_to_fit();
+}
+
+uint32_t DWARFDebugAranges::findAddress(uint64_t Address) const {
+ RangeCollIterator It =
+ partition_point(Aranges, [=](Range R) { return R.HighPC() <= Address; });
+ if (It != Aranges.end() && It->LowPC <= Address)
+ return It->CUOffset;
+ return -1U;
+}
diff --git a/third_party/llvm-project/DWARFDebugFrame.cpp b/third_party/llvm-project/DWARFDebugFrame.cpp
new file mode 100644
index 000000000..81b00f657
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugFrame.cpp
@@ -0,0 +1,554 @@
+//===- DWARFDebugFrame.h - Parsing of .debug_frame ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugFrame.h"
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cinttypes>
+#include <cstdint>
+#include <string>
+#include <vector>
+
+using namespace llvm;
+using namespace dwarf;
+
+
+// See DWARF standard v3, section 7.23
+const uint8_t DWARF_CFI_PRIMARY_OPCODE_MASK = 0xc0;
+const uint8_t DWARF_CFI_PRIMARY_OPERAND_MASK = 0x3f;
+
+Error CFIProgram::parse(DWARFDataExtractor Data, uint64_t *Offset,
+ uint64_t EndOffset) {
+ while (*Offset < EndOffset) {
+ uint8_t Opcode = Data.getRelocatedValue(1, Offset);
+ // Some instructions have a primary opcode encoded in the top bits.
+ uint8_t Primary = Opcode & DWARF_CFI_PRIMARY_OPCODE_MASK;
+
+ if (Primary) {
+ // If it's a primary opcode, the first operand is encoded in the bottom
+ // bits of the opcode itself.
+ uint64_t Op1 = Opcode & DWARF_CFI_PRIMARY_OPERAND_MASK;
+ switch (Primary) {
+ default:
+ return createStringError(errc::illegal_byte_sequence,
+ "Invalid primary CFI opcode 0x%" PRIx8,
+ Primary);
+ case DW_CFA_advance_loc:
+ case DW_CFA_restore:
+ addInstruction(Primary, Op1);
+ break;
+ case DW_CFA_offset:
+ addInstruction(Primary, Op1, Data.getULEB128(Offset));
+ break;
+ }
+ } else {
+ // Extended opcode - its value is Opcode itself.
+ switch (Opcode) {
+ default:
+ return createStringError(errc::illegal_byte_sequence,
+ "Invalid extended CFI opcode 0x%" PRIx8,
+ Opcode);
+ case DW_CFA_nop:
+ case DW_CFA_remember_state:
+ case DW_CFA_restore_state:
+ case DW_CFA_GNU_window_save:
+ // No operands
+ addInstruction(Opcode);
+ break;
+ case DW_CFA_set_loc:
+ // Operands: Address
+ addInstruction(Opcode, Data.getRelocatedAddress(Offset));
+ break;
+ case DW_CFA_advance_loc1:
+ // Operands: 1-byte delta
+ addInstruction(Opcode, Data.getRelocatedValue(1, Offset));
+ break;
+ case DW_CFA_advance_loc2:
+ // Operands: 2-byte delta
+ addInstruction(Opcode, Data.getRelocatedValue(2, Offset));
+ break;
+ case DW_CFA_advance_loc4:
+ // Operands: 4-byte delta
+ addInstruction(Opcode, Data.getRelocatedValue(4, Offset));
+ break;
+ case DW_CFA_restore_extended:
+ case DW_CFA_undefined:
+ case DW_CFA_same_value:
+ case DW_CFA_def_cfa_register:
+ case DW_CFA_def_cfa_offset:
+ case DW_CFA_GNU_args_size:
+ // Operands: ULEB128
+ addInstruction(Opcode, Data.getULEB128(Offset));
+ break;
+ case DW_CFA_def_cfa_offset_sf:
+ // Operands: SLEB128
+ addInstruction(Opcode, Data.getSLEB128(Offset));
+ break;
+ case DW_CFA_offset_extended:
+ case DW_CFA_register:
+ case DW_CFA_def_cfa:
+ case DW_CFA_val_offset: {
+ // Operands: ULEB128, ULEB128
+ // Note: We can not embed getULEB128 directly into function
+ // argument list. getULEB128 changes Offset and order of evaluation
+ // for arguments is unspecified.
+ auto op1 = Data.getULEB128(Offset);
+ auto op2 = Data.getULEB128(Offset);
+ addInstruction(Opcode, op1, op2);
+ break;
+ }
+ case DW_CFA_offset_extended_sf:
+ case DW_CFA_def_cfa_sf:
+ case DW_CFA_val_offset_sf: {
+ // Operands: ULEB128, SLEB128
+ // Note: see comment for the previous case
+ auto op1 = Data.getULEB128(Offset);
+ auto op2 = (uint64_t)Data.getSLEB128(Offset);
+ addInstruction(Opcode, op1, op2);
+ break;
+ }
+ case DW_CFA_def_cfa_expression: {
+ uint32_t ExprLength = Data.getULEB128(Offset);
+ addInstruction(Opcode, 0);
+ DataExtractor Extractor(
+ Data.getData().slice(*Offset, *Offset + ExprLength),
+ Data.isLittleEndian(), Data.getAddressSize());
+ Instructions.back().Expression = DWARFExpression(
+ Extractor, Data.getAddressSize(), dwarf::DWARF_VERSION);
+ *Offset += ExprLength;
+ break;
+ }
+ case DW_CFA_expression:
+ case DW_CFA_val_expression: {
+ auto RegNum = Data.getULEB128(Offset);
+ auto BlockLength = Data.getULEB128(Offset);
+ addInstruction(Opcode, RegNum, 0);
+ DataExtractor Extractor(
+ Data.getData().slice(*Offset, *Offset + BlockLength),
+ Data.isLittleEndian(), Data.getAddressSize());
+ Instructions.back().Expression = DWARFExpression(
+ Extractor, Data.getAddressSize(), dwarf::DWARF_VERSION);
+ *Offset += BlockLength;
+ break;
+ }
+ }
+ }
+ }
+
+ return Error::success();
+}
+
+namespace {
+
+
+} // end anonymous namespace
+
+ArrayRef<CFIProgram::OperandType[2]> CFIProgram::getOperandTypes() {
+ static OperandType OpTypes[DW_CFA_restore+1][2];
+ static bool Initialized = false;
+ if (Initialized) {
+ return ArrayRef<OperandType[2]>(&OpTypes[0], DW_CFA_restore+1);
+ }
+ Initialized = true;
+
+#define DECLARE_OP2(OP, OPTYPE0, OPTYPE1) \
+ do { \
+ OpTypes[OP][0] = OPTYPE0; \
+ OpTypes[OP][1] = OPTYPE1; \
+ } while (false)
+#define DECLARE_OP1(OP, OPTYPE0) DECLARE_OP2(OP, OPTYPE0, OT_None)
+#define DECLARE_OP0(OP) DECLARE_OP1(OP, OT_None)
+
+ DECLARE_OP1(DW_CFA_set_loc, OT_Address);
+ DECLARE_OP1(DW_CFA_advance_loc, OT_FactoredCodeOffset);
+ DECLARE_OP1(DW_CFA_advance_loc1, OT_FactoredCodeOffset);
+ DECLARE_OP1(DW_CFA_advance_loc2, OT_FactoredCodeOffset);
+ DECLARE_OP1(DW_CFA_advance_loc4, OT_FactoredCodeOffset);
+ DECLARE_OP1(DW_CFA_MIPS_advance_loc8, OT_FactoredCodeOffset);
+ DECLARE_OP2(DW_CFA_def_cfa, OT_Register, OT_Offset);
+ DECLARE_OP2(DW_CFA_def_cfa_sf, OT_Register, OT_SignedFactDataOffset);
+ DECLARE_OP1(DW_CFA_def_cfa_register, OT_Register);
+ DECLARE_OP1(DW_CFA_def_cfa_offset, OT_Offset);
+ DECLARE_OP1(DW_CFA_def_cfa_offset_sf, OT_SignedFactDataOffset);
+ DECLARE_OP1(DW_CFA_def_cfa_expression, OT_Expression);
+ DECLARE_OP1(DW_CFA_undefined, OT_Register);
+ DECLARE_OP1(DW_CFA_same_value, OT_Register);
+ DECLARE_OP2(DW_CFA_offset, OT_Register, OT_UnsignedFactDataOffset);
+ DECLARE_OP2(DW_CFA_offset_extended, OT_Register, OT_UnsignedFactDataOffset);
+ DECLARE_OP2(DW_CFA_offset_extended_sf, OT_Register, OT_SignedFactDataOffset);
+ DECLARE_OP2(DW_CFA_val_offset, OT_Register, OT_UnsignedFactDataOffset);
+ DECLARE_OP2(DW_CFA_val_offset_sf, OT_Register, OT_SignedFactDataOffset);
+ DECLARE_OP2(DW_CFA_register, OT_Register, OT_Register);
+ DECLARE_OP2(DW_CFA_expression, OT_Register, OT_Expression);
+ DECLARE_OP2(DW_CFA_val_expression, OT_Register, OT_Expression);
+ DECLARE_OP1(DW_CFA_restore, OT_Register);
+ DECLARE_OP1(DW_CFA_restore_extended, OT_Register);
+ DECLARE_OP0(DW_CFA_remember_state);
+ DECLARE_OP0(DW_CFA_restore_state);
+ DECLARE_OP0(DW_CFA_GNU_window_save);
+ DECLARE_OP1(DW_CFA_GNU_args_size, OT_Offset);
+ DECLARE_OP0(DW_CFA_nop);
+
+#undef DECLARE_OP0
+#undef DECLARE_OP1
+#undef DECLARE_OP2
+
+ return ArrayRef<OperandType[2]>(&OpTypes[0], DW_CFA_restore+1);
+}
+
+/// Print \p Opcode's operand number \p OperandIdx which has value \p Operand.
+void CFIProgram::printOperand(raw_ostream &OS, const MCRegisterInfo *MRI,
+ bool IsEH, const Instruction &Instr,
+ unsigned OperandIdx, uint64_t Operand) const {
+ assert(OperandIdx < 2);
+ uint8_t Opcode = Instr.Opcode;
+ OperandType Type = getOperandTypes()[Opcode][OperandIdx];
+
+ switch (Type) {
+ case OT_Unset: {
+ OS << " Unsupported " << (OperandIdx ? "second" : "first") << " operand to";
+ auto OpcodeName = CallFrameString(Opcode, Arch);
+ if (!OpcodeName.empty())
+ OS << " " << OpcodeName;
+ else
+ OS << format(" Opcode %x", Opcode);
+ break;
+ }
+ case OT_None:
+ break;
+ case OT_Address:
+ OS << format(" %" PRIx64, Operand);
+ break;
+ case OT_Offset:
+ // The offsets are all encoded in a unsigned form, but in practice
+ // consumers use them signed. It's most certainly legacy due to
+ // the lack of signed variants in the first Dwarf standards.
+ OS << format(" %+" PRId64, int64_t(Operand));
+ break;
+ case OT_FactoredCodeOffset: // Always Unsigned
+ if (CodeAlignmentFactor)
+ OS << format(" %" PRId64, Operand * CodeAlignmentFactor);
+ else
+ OS << format(" %" PRId64 "*code_alignment_factor" , Operand);
+ break;
+ case OT_SignedFactDataOffset:
+ if (DataAlignmentFactor)
+ OS << format(" %" PRId64, int64_t(Operand) * DataAlignmentFactor);
+ else
+ OS << format(" %" PRId64 "*data_alignment_factor" , int64_t(Operand));
+ break;
+ case OT_UnsignedFactDataOffset:
+ if (DataAlignmentFactor)
+ OS << format(" %" PRId64, Operand * DataAlignmentFactor);
+ else
+ OS << format(" %" PRId64 "*data_alignment_factor" , Operand);
+ break;
+ case OT_Register:
+ OS << format(" reg%" PRId64, Operand);
+ break;
+ case OT_Expression:
+ assert(Instr.Expression && "missing DWARFExpression object");
+ OS << " ";
+ Instr.Expression->print(OS, MRI, nullptr, IsEH);
+ break;
+ }
+}
+
+void CFIProgram::dump(raw_ostream &OS, const MCRegisterInfo *MRI, bool IsEH,
+ unsigned IndentLevel) const {
+ for (const auto &Instr : Instructions) {
+ uint8_t Opcode = Instr.Opcode;
+ if (Opcode & DWARF_CFI_PRIMARY_OPCODE_MASK)
+ Opcode &= DWARF_CFI_PRIMARY_OPCODE_MASK;
+ OS.indent(2 * IndentLevel);
+ OS << CallFrameString(Opcode, Arch) << ":";
+ for (unsigned i = 0; i < Instr.Ops.size(); ++i)
+ printOperand(OS, MRI, IsEH, Instr, i, Instr.Ops[i]);
+ OS << '\n';
+ }
+}
+
+void CIE::dump(raw_ostream &OS, const MCRegisterInfo *MRI, bool IsEH) const {
+ OS << format("%08x %08x %08x CIE", (uint32_t)Offset, (uint32_t)Length,
+ DW_CIE_ID)
+ << "\n";
+ OS << format(" Version: %d\n", Version);
+ OS << " Augmentation: \"" << Augmentation << "\"\n";
+ if (Version >= 4) {
+ OS << format(" Address size: %u\n", (uint32_t)AddressSize);
+ OS << format(" Segment desc size: %u\n",
+ (uint32_t)SegmentDescriptorSize);
+ }
+ OS << format(" Code alignment factor: %u\n", (uint32_t)CodeAlignmentFactor);
+ OS << format(" Data alignment factor: %d\n", (int32_t)DataAlignmentFactor);
+ OS << format(" Return address column: %d\n", (int32_t)ReturnAddressRegister);
+ if (Personality)
+ OS << format(" Personality Address: %016" PRIx64 "\n", *Personality);
+ if (!AugmentationData.empty()) {
+ OS << " Augmentation data: ";
+ for (uint8_t Byte : AugmentationData)
+ OS << ' ' << hexdigit(Byte >> 4) << hexdigit(Byte & 0xf);
+ OS << "\n";
+ }
+ OS << "\n";
+ CFIs.dump(OS, MRI, IsEH);
+ OS << "\n";
+}
+
+void FDE::dump(raw_ostream &OS, const MCRegisterInfo *MRI, bool IsEH) const {
+ OS << format("%08x %08x %08x FDE ", (uint32_t)Offset, (uint32_t)Length,
+ (int32_t)LinkedCIEOffset);
+ OS << format("cie=%08x pc=%08x...%08x\n", (int32_t)LinkedCIEOffset,
+ (uint32_t)InitialLocation,
+ (uint32_t)InitialLocation + (uint32_t)AddressRange);
+ if (LSDAAddress)
+ OS << format(" LSDA Address: %016" PRIx64 "\n", *LSDAAddress);
+ CFIs.dump(OS, MRI, IsEH);
+ OS << "\n";
+}
+
+DWARFDebugFrame::DWARFDebugFrame(Triple::ArchType Arch,
+ bool IsEH, uint64_t EHFrameAddress)
+ : Arch(Arch), IsEH(IsEH), EHFrameAddress(EHFrameAddress) {}
+
+DWARFDebugFrame::~DWARFDebugFrame() = default;
+
+static void LLVM_ATTRIBUTE_UNUSED dumpDataAux(DataExtractor Data,
+ uint64_t Offset, int Length) {
+ errs() << "DUMP: ";
+ for (int i = 0; i < Length; ++i) {
+ uint8_t c = Data.getU8(&Offset);
+ errs().write_hex(c); errs() << " ";
+ }
+ errs() << "\n";
+}
+
+// This is a workaround for old compilers which do not allow
+// noreturn attribute usage in lambdas. Once the support for those
+// compilers are phased out, we can remove this and return back to
+// a ReportError lambda: [StartOffset](const char *ErrorMsg).
+static void LLVM_ATTRIBUTE_NORETURN ReportError(uint64_t StartOffset,
+ const char *ErrorMsg) {
+ std::string Str;
+ raw_string_ostream OS(Str);
+ OS << format(ErrorMsg, StartOffset);
+ OS.flush();
+ report_fatal_error(Str);
+}
+
+void DWARFDebugFrame::parse(DWARFDataExtractor Data) {
+ uint64_t Offset = 0;
+ DenseMap<uint64_t, CIE *> CIEs;
+
+ while (Data.isValidOffset(Offset)) {
+ uint64_t StartOffset = Offset;
+
+ bool IsDWARF64 = false;
+ uint64_t Length = Data.getRelocatedValue(4, &Offset);
+ uint64_t Id;
+
+ if (Length == dwarf::DW_LENGTH_DWARF64) {
+ // DWARF-64 is distinguished by the first 32 bits of the initial length
+ // field being 0xffffffff. Then, the next 64 bits are the actual entry
+ // length.
+ IsDWARF64 = true;
+ Length = Data.getRelocatedValue(8, &Offset);
+ }
+
+ // At this point, Offset points to the next field after Length.
+ // Length is the structure size excluding itself. Compute an offset one
+ // past the end of the structure (needed to know how many instructions to
+ // read).
+ uint64_t StartStructureOffset = Offset;
+ uint64_t EndStructureOffset = Offset + Length;
+
+ // The Id field's size depends on the DWARF format
+ Id = Data.getUnsigned(&Offset, (IsDWARF64 && !IsEH) ? 8 : 4);
+ bool IsCIE =
+ ((IsDWARF64 && Id == DW64_CIE_ID) || Id == DW_CIE_ID || (IsEH && !Id));
+
+ if (IsCIE) {
+ uint8_t Version = Data.getU8(&Offset);
+ const char *Augmentation = Data.getCStr(&Offset);
+ StringRef AugmentationString(Augmentation ? Augmentation : "");
+ uint8_t AddressSize = Version < 4 ? Data.getAddressSize() :
+ Data.getU8(&Offset);
+ Data.setAddressSize(AddressSize);
+ uint8_t SegmentDescriptorSize = Version < 4 ? 0 : Data.getU8(&Offset);
+ uint64_t CodeAlignmentFactor = Data.getULEB128(&Offset);
+ int64_t DataAlignmentFactor = Data.getSLEB128(&Offset);
+ uint64_t ReturnAddressRegister =
+ Version == 1 ? Data.getU8(&Offset) : Data.getULEB128(&Offset);
+
+ // Parse the augmentation data for EH CIEs
+ StringRef AugmentationData("");
+ uint32_t FDEPointerEncoding = DW_EH_PE_absptr;
+ uint32_t LSDAPointerEncoding = DW_EH_PE_omit;
+ Optional<uint64_t> Personality;
+ Optional<uint32_t> PersonalityEncoding;
+ if (IsEH) {
+ Optional<uint64_t> AugmentationLength;
+ uint64_t StartAugmentationOffset;
+ uint64_t EndAugmentationOffset;
+
+ // Walk the augmentation string to get all the augmentation data.
+ for (unsigned i = 0, e = AugmentationString.size(); i != e; ++i) {
+ switch (AugmentationString[i]) {
+ default:
+ ReportError(
+ StartOffset,
+ "Unknown augmentation character in entry at %" PRIx64);
+ case 'L':
+ LSDAPointerEncoding = Data.getU8(&Offset);
+ break;
+ case 'P': {
+ if (Personality)
+ ReportError(StartOffset,
+ "Duplicate personality in entry at %" PRIx64);
+ PersonalityEncoding = Data.getU8(&Offset);
+ Personality = Data.getEncodedPointer(
+ &Offset, *PersonalityEncoding,
+ EHFrameAddress ? EHFrameAddress + Offset : 0);
+ break;
+ }
+ case 'R':
+ FDEPointerEncoding = Data.getU8(&Offset);
+ break;
+ case 'S':
+ // Current frame is a signal trampoline.
+ break;
+ case 'z':
+ if (i)
+ ReportError(StartOffset,
+ "'z' must be the first character at %" PRIx64);
+ // Parse the augmentation length first. We only parse it if
+ // the string contains a 'z'.
+ AugmentationLength = Data.getULEB128(&Offset);
+ StartAugmentationOffset = Offset;
+ EndAugmentationOffset = Offset + *AugmentationLength;
+ break;
+ case 'B':
+ // B-Key is used for signing functions associated with this
+ // augmentation string
+ break;
+ }
+ }
+
+ if (AugmentationLength.hasValue()) {
+ if (Offset != EndAugmentationOffset)
+ ReportError(StartOffset,
+ "Parsing augmentation data at %" PRIx64 " failed");
+
+ AugmentationData = Data.getData().slice(StartAugmentationOffset,
+ EndAugmentationOffset);
+ }
+ }
+
+ auto Cie = std::make_unique<CIE>(
+ StartOffset, Length, Version, AugmentationString, AddressSize,
+ SegmentDescriptorSize, CodeAlignmentFactor, DataAlignmentFactor,
+ ReturnAddressRegister, AugmentationData, FDEPointerEncoding,
+ LSDAPointerEncoding, Personality, PersonalityEncoding, Arch);
+ CIEs[StartOffset] = Cie.get();
+ Entries.emplace_back(std::move(Cie));
+ } else {
+ // FDE
+ uint64_t CIEPointer = Id;
+ uint64_t InitialLocation = 0;
+ uint64_t AddressRange = 0;
+ Optional<uint64_t> LSDAAddress;
+ CIE *Cie = CIEs[IsEH ? (StartStructureOffset - CIEPointer) : CIEPointer];
+
+ if (IsEH) {
+ // The address size is encoded in the CIE we reference.
+ if (!Cie)
+ ReportError(StartOffset, "Parsing FDE data at %" PRIx64
+ " failed due to missing CIE");
+
+ if (auto Val = Data.getEncodedPointer(
+ &Offset, Cie->getFDEPointerEncoding(),
+ EHFrameAddress ? EHFrameAddress + Offset : 0)) {
+ InitialLocation = *Val;
+ }
+ if (auto Val = Data.getEncodedPointer(
+ &Offset, Cie->getFDEPointerEncoding(), 0)) {
+ AddressRange = *Val;
+ }
+
+ StringRef AugmentationString = Cie->getAugmentationString();
+ if (!AugmentationString.empty()) {
+ // Parse the augmentation length and data for this FDE.
+ uint64_t AugmentationLength = Data.getULEB128(&Offset);
+
+ uint64_t EndAugmentationOffset = Offset + AugmentationLength;
+
+ // Decode the LSDA if the CIE augmentation string said we should.
+ if (Cie->getLSDAPointerEncoding() != DW_EH_PE_omit) {
+ LSDAAddress = Data.getEncodedPointer(
+ &Offset, Cie->getLSDAPointerEncoding(),
+ EHFrameAddress ? Offset + EHFrameAddress : 0);
+ }
+
+ if (Offset != EndAugmentationOffset)
+ ReportError(StartOffset,
+ "Parsing augmentation data at %" PRIx64 " failed");
+ }
+ } else {
+ InitialLocation = Data.getRelocatedAddress(&Offset);
+ AddressRange = Data.getRelocatedAddress(&Offset);
+ }
+
+ Entries.emplace_back(new FDE(StartOffset, Length, CIEPointer,
+ InitialLocation, AddressRange,
+ Cie, LSDAAddress, Arch));
+ }
+
+ if (Error E =
+ Entries.back()->cfis().parse(Data, &Offset, EndStructureOffset)) {
+ report_fatal_error(toString(std::move(E)));
+ }
+
+ if (Offset != EndStructureOffset)
+ ReportError(StartOffset,
+ "Parsing entry instructions at %" PRIx64 " failed");
+ }
+}
+
+FrameEntry *DWARFDebugFrame::getEntryAtOffset(uint64_t Offset) const {
+ auto It = partition_point(Entries, [=](const std::unique_ptr<FrameEntry> &E) {
+ return E->getOffset() < Offset;
+ });
+ if (It != Entries.end() && (*It)->getOffset() == Offset)
+ return It->get();
+ return nullptr;
+}
+
+void DWARFDebugFrame::dump(raw_ostream &OS, const MCRegisterInfo *MRI,
+ Optional<uint64_t> Offset) const {
+ if (Offset) {
+ if (auto *Entry = getEntryAtOffset(*Offset))
+ Entry->dump(OS, MRI, IsEH);
+ return;
+ }
+
+ OS << "\n";
+ for (const auto &Entry : Entries)
+ Entry->dump(OS, MRI, IsEH);
+}
diff --git a/third_party/llvm-project/DWARFDebugInfoEntry.cpp b/third_party/llvm-project/DWARFDebugInfoEntry.cpp
new file mode 100644
index 000000000..87eab34d5
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugInfoEntry.cpp
@@ -0,0 +1,69 @@
+//===- DWARFDebugInfoEntry.cpp --------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstddef>
+#include <cstdint>
+
+using namespace llvm;
+using namespace dwarf;
+
+bool DWARFDebugInfoEntry::extractFast(const DWARFUnit &U,
+ uint64_t *OffsetPtr) {
+ DWARFDataExtractor DebugInfoData = U.getDebugInfoExtractor();
+ const uint64_t UEndOffset = U.getNextUnitOffset();
+ return extractFast(U, OffsetPtr, DebugInfoData, UEndOffset, 0);
+}
+
+bool DWARFDebugInfoEntry::extractFast(const DWARFUnit &U, uint64_t *OffsetPtr,
+ const DWARFDataExtractor &DebugInfoData,
+ uint64_t UEndOffset, uint32_t D) {
+ Offset = *OffsetPtr;
+ Depth = D;
+ if (Offset >= UEndOffset || !DebugInfoData.isValidOffset(Offset))
+ return false;
+ uint64_t AbbrCode = DebugInfoData.getULEB128(OffsetPtr);
+ if (0 == AbbrCode) {
+ // NULL debug tag entry.
+ AbbrevDecl = nullptr;
+ return true;
+ }
+ AbbrevDecl = U.getAbbreviations()->getAbbreviationDeclaration(AbbrCode);
+ if (nullptr == AbbrevDecl) {
+ // Restore the original offset.
+ *OffsetPtr = Offset;
+ return false;
+ }
+ // See if all attributes in this DIE have fixed byte sizes. If so, we can
+ // just add this size to the offset to skip to the next DIE.
+ if (Optional<size_t> FixedSize = AbbrevDecl->getFixedAttributesByteSize(U)) {
+ *OffsetPtr += *FixedSize;
+ return true;
+ }
+
+ // Skip all data in the .debug_info for the attributes
+ for (const auto &AttrSpec : AbbrevDecl->attributes()) {
+ // Check if this attribute has a fixed byte size.
+ if (auto FixedSize = AttrSpec.getByteSize(U)) {
+ // Attribute byte size if fixed, just add the size to the offset.
+ *OffsetPtr += *FixedSize;
+ } else if (!DWARFFormValue::skipValue(AttrSpec.Form, DebugInfoData,
+ OffsetPtr, U.getFormParams())) {
+ // We failed to skip this attribute's value, restore the original offset
+ // and return the failure status.
+ *OffsetPtr = Offset;
+ return false;
+ }
+ }
+ return true;
+}
diff --git a/third_party/llvm-project/DWARFDebugLine.cpp b/third_party/llvm-project/DWARFDebugLine.cpp
new file mode 100644
index 000000000..dbee28ff5
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugLine.cpp
@@ -0,0 +1,1181 @@
+//===- DWARFDebugLine.cpp -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cinttypes>
+#include <cstdint>
+#include <cstdio>
+#include <utility>
+
+using namespace llvm;
+using namespace dwarf;
+
+using FileLineInfoKind = DILineInfoSpecifier::FileLineInfoKind;
+
+namespace {
+
+struct ContentDescriptor {
+ dwarf::LineNumberEntryFormat Type;
+ dwarf::Form Form;
+};
+
+using ContentDescriptors = SmallVector<ContentDescriptor, 4>;
+
+} // end anonmyous namespace
+
+void DWARFDebugLine::ContentTypeTracker::trackContentType(
+ dwarf::LineNumberEntryFormat ContentType) {
+ switch (ContentType) {
+ case dwarf::DW_LNCT_timestamp:
+ HasModTime = true;
+ break;
+ case dwarf::DW_LNCT_size:
+ HasLength = true;
+ break;
+ case dwarf::DW_LNCT_MD5:
+ HasMD5 = true;
+ break;
+ case dwarf::DW_LNCT_LLVM_source:
+ HasSource = true;
+ break;
+ default:
+ // We only care about values we consider optional, and new values may be
+ // added in the vendor extension range, so we do not match exhaustively.
+ break;
+ }
+}
+
+DWARFDebugLine::Prologue::Prologue() { clear(); }
+
+bool DWARFDebugLine::Prologue::hasFileAtIndex(uint64_t FileIndex) const {
+ uint16_t DwarfVersion = getVersion();
+ assert(DwarfVersion != 0 &&
+ "line table prologue has no dwarf version information");
+ if (DwarfVersion >= 5)
+ return FileIndex < FileNames.size();
+ return FileIndex != 0 && FileIndex <= FileNames.size();
+}
+
+const llvm::DWARFDebugLine::FileNameEntry &
+DWARFDebugLine::Prologue::getFileNameEntry(uint64_t Index) const {
+ uint16_t DwarfVersion = getVersion();
+ assert(DwarfVersion != 0 &&
+ "line table prologue has no dwarf version information");
+ // In DWARF v5 the file names are 0-indexed.
+ if (DwarfVersion >= 5)
+ return FileNames[Index];
+ return FileNames[Index - 1];
+}
+
+void DWARFDebugLine::Prologue::clear() {
+ TotalLength = PrologueLength = 0;
+ SegSelectorSize = 0;
+ MinInstLength = MaxOpsPerInst = DefaultIsStmt = LineBase = LineRange = 0;
+ OpcodeBase = 0;
+ FormParams = dwarf::FormParams({0, 0, DWARF32});
+ ContentTypes = ContentTypeTracker();
+ StandardOpcodeLengths.clear();
+ IncludeDirectories.clear();
+ FileNames.clear();
+}
+
+void DWARFDebugLine::Prologue::dump(raw_ostream &OS,
+ DIDumpOptions DumpOptions) const {
+ OS << "Line table prologue:\n"
+ << format(" total_length: 0x%8.8" PRIx64 "\n", TotalLength)
+ << format(" version: %u\n", getVersion());
+ if (getVersion() >= 5)
+ OS << format(" address_size: %u\n", getAddressSize())
+ << format(" seg_select_size: %u\n", SegSelectorSize);
+ OS << format(" prologue_length: 0x%8.8" PRIx64 "\n", PrologueLength)
+ << format(" min_inst_length: %u\n", MinInstLength)
+ << format(getVersion() >= 4 ? "max_ops_per_inst: %u\n" : "", MaxOpsPerInst)
+ << format(" default_is_stmt: %u\n", DefaultIsStmt)
+ << format(" line_base: %i\n", LineBase)
+ << format(" line_range: %u\n", LineRange)
+ << format(" opcode_base: %u\n", OpcodeBase);
+
+ for (uint32_t I = 0; I != StandardOpcodeLengths.size(); ++I)
+ OS << format("standard_opcode_lengths[%s] = %u\n",
+ LNStandardString(I + 1).data(), StandardOpcodeLengths[I]);
+
+ if (!IncludeDirectories.empty()) {
+ // DWARF v5 starts directory indexes at 0.
+ uint32_t DirBase = getVersion() >= 5 ? 0 : 1;
+ for (uint32_t I = 0; I != IncludeDirectories.size(); ++I) {
+ OS << format("include_directories[%3u] = ", I + DirBase);
+ IncludeDirectories[I].dump(OS, DumpOptions);
+ OS << '\n';
+ }
+ }
+
+ if (!FileNames.empty()) {
+ // DWARF v5 starts file indexes at 0.
+ uint32_t FileBase = getVersion() >= 5 ? 0 : 1;
+ for (uint32_t I = 0; I != FileNames.size(); ++I) {
+ const FileNameEntry &FileEntry = FileNames[I];
+ OS << format("file_names[%3u]:\n", I + FileBase);
+ OS << " name: ";
+ FileEntry.Name.dump(OS, DumpOptions);
+ OS << '\n'
+ << format(" dir_index: %" PRIu64 "\n", FileEntry.DirIdx);
+ if (ContentTypes.HasMD5)
+ OS << " md5_checksum: " << FileEntry.Checksum.digest() << '\n';
+ if (ContentTypes.HasModTime)
+ OS << format(" mod_time: 0x%8.8" PRIx64 "\n", FileEntry.ModTime);
+ if (ContentTypes.HasLength)
+ OS << format(" length: 0x%8.8" PRIx64 "\n", FileEntry.Length);
+ if (ContentTypes.HasSource) {
+ OS << " source: ";
+ FileEntry.Source.dump(OS, DumpOptions);
+ OS << '\n';
+ }
+ }
+ }
+}
+
+// Parse v2-v4 directory and file tables.
+static void
+parseV2DirFileTables(const DWARFDataExtractor &DebugLineData,
+ uint64_t *OffsetPtr, uint64_t EndPrologueOffset,
+ DWARFDebugLine::ContentTypeTracker &ContentTypes,
+ std::vector<DWARFFormValue> &IncludeDirectories,
+ std::vector<DWARFDebugLine::FileNameEntry> &FileNames) {
+ while (*OffsetPtr < EndPrologueOffset) {
+ StringRef S = DebugLineData.getCStrRef(OffsetPtr);
+ if (S.empty())
+ break;
+ DWARFFormValue Dir =
+ DWARFFormValue::createFromPValue(dwarf::DW_FORM_string, S.data());
+ IncludeDirectories.push_back(Dir);
+ }
+
+ while (*OffsetPtr < EndPrologueOffset) {
+ StringRef Name = DebugLineData.getCStrRef(OffsetPtr);
+ if (Name.empty())
+ break;
+ DWARFDebugLine::FileNameEntry FileEntry;
+ FileEntry.Name =
+ DWARFFormValue::createFromPValue(dwarf::DW_FORM_string, Name.data());
+ FileEntry.DirIdx = DebugLineData.getULEB128(OffsetPtr);
+ FileEntry.ModTime = DebugLineData.getULEB128(OffsetPtr);
+ FileEntry.Length = DebugLineData.getULEB128(OffsetPtr);
+ FileNames.push_back(FileEntry);
+ }
+
+ ContentTypes.HasModTime = true;
+ ContentTypes.HasLength = true;
+}
+
+// Parse v5 directory/file entry content descriptions.
+// Returns the descriptors, or an error if we did not find a path or ran off
+// the end of the prologue.
+static llvm::Expected<ContentDescriptors>
+parseV5EntryFormat(const DWARFDataExtractor &DebugLineData, uint64_t *OffsetPtr,
+ uint64_t EndPrologueOffset,
+ DWARFDebugLine::ContentTypeTracker *ContentTypes) {
+ ContentDescriptors Descriptors;
+ int FormatCount = DebugLineData.getU8(OffsetPtr);
+ bool HasPath = false;
+ for (int I = 0; I != FormatCount; ++I) {
+ if (*OffsetPtr >= EndPrologueOffset)
+ return createStringError(
+ errc::invalid_argument,
+ "failed to parse entry content descriptions at offset "
+ "0x%8.8" PRIx64
+ " because offset extends beyond the prologue end at offset "
+ "0x%8.8" PRIx64,
+ *OffsetPtr, EndPrologueOffset);
+ ContentDescriptor Descriptor;
+ Descriptor.Type =
+ dwarf::LineNumberEntryFormat(DebugLineData.getULEB128(OffsetPtr));
+ Descriptor.Form = dwarf::Form(DebugLineData.getULEB128(OffsetPtr));
+ if (Descriptor.Type == dwarf::DW_LNCT_path)
+ HasPath = true;
+ if (ContentTypes)
+ ContentTypes->trackContentType(Descriptor.Type);
+ Descriptors.push_back(Descriptor);
+ }
+
+ if (!HasPath)
+ return createStringError(errc::invalid_argument,
+ "failed to parse entry content descriptions"
+ " because no path was found");
+ return Descriptors;
+}
+
+static Error
+parseV5DirFileTables(const DWARFDataExtractor &DebugLineData,
+ uint64_t *OffsetPtr, uint64_t EndPrologueOffset,
+ const dwarf::FormParams &FormParams,
+ const DWARFContext &Ctx, const DWARFUnit *U,
+ DWARFDebugLine::ContentTypeTracker &ContentTypes,
+ std::vector<DWARFFormValue> &IncludeDirectories,
+ std::vector<DWARFDebugLine::FileNameEntry> &FileNames) {
+ // Get the directory entry description.
+ llvm::Expected<ContentDescriptors> DirDescriptors =
+ parseV5EntryFormat(DebugLineData, OffsetPtr, EndPrologueOffset, nullptr);
+ if (!DirDescriptors)
+ return DirDescriptors.takeError();
+
+ // Get the directory entries, according to the format described above.
+ int DirEntryCount = DebugLineData.getU8(OffsetPtr);
+ for (int I = 0; I != DirEntryCount; ++I) {
+ if (*OffsetPtr >= EndPrologueOffset)
+ return createStringError(
+ errc::invalid_argument,
+ "failed to parse directory entry at offset "
+ "0x%8.8" PRIx64
+ " because offset extends beyond the prologue end at offset "
+ "0x%8.8" PRIx64,
+ *OffsetPtr, EndPrologueOffset);
+ for (auto Descriptor : *DirDescriptors) {
+ DWARFFormValue Value(Descriptor.Form);
+ switch (Descriptor.Type) {
+ case DW_LNCT_path:
+ if (!Value.extractValue(DebugLineData, OffsetPtr, FormParams, &Ctx, U))
+ return createStringError(errc::invalid_argument,
+ "failed to parse directory entry because "
+ "extracting the form value failed.");
+ IncludeDirectories.push_back(Value);
+ break;
+ default:
+ if (!Value.skipValue(DebugLineData, OffsetPtr, FormParams))
+ return createStringError(errc::invalid_argument,
+ "failed to parse directory entry because "
+ "skipping the form value failed.");
+ }
+ }
+ }
+
+ // Get the file entry description.
+ llvm::Expected<ContentDescriptors> FileDescriptors = parseV5EntryFormat(
+ DebugLineData, OffsetPtr, EndPrologueOffset, &ContentTypes);
+ if (!FileDescriptors)
+ return FileDescriptors.takeError();
+
+ // Get the file entries, according to the format described above.
+ int FileEntryCount = DebugLineData.getU8(OffsetPtr);
+ for (int I = 0; I != FileEntryCount; ++I) {
+ if (*OffsetPtr >= EndPrologueOffset)
+ return createStringError(
+ errc::invalid_argument,
+ "failed to parse file entry at offset "
+ "0x%8.8" PRIx64
+ " because offset extends beyond the prologue end at offset "
+ "0x%8.8" PRIx64,
+ *OffsetPtr, EndPrologueOffset);
+ DWARFDebugLine::FileNameEntry FileEntry;
+ for (auto Descriptor : *FileDescriptors) {
+ DWARFFormValue Value(Descriptor.Form);
+ if (!Value.extractValue(DebugLineData, OffsetPtr, FormParams, &Ctx, U))
+ return createStringError(errc::invalid_argument,
+ "failed to parse file entry because "
+ "extracting the form value failed.");
+ switch (Descriptor.Type) {
+ case DW_LNCT_path:
+ FileEntry.Name = Value;
+ break;
+ case DW_LNCT_LLVM_source:
+ FileEntry.Source = Value;
+ break;
+ case DW_LNCT_directory_index:
+ FileEntry.DirIdx = Value.getAsUnsignedConstant().getValue();
+ break;
+ case DW_LNCT_timestamp:
+ FileEntry.ModTime = Value.getAsUnsignedConstant().getValue();
+ break;
+ case DW_LNCT_size:
+ FileEntry.Length = Value.getAsUnsignedConstant().getValue();
+ break;
+ case DW_LNCT_MD5:
+ if (!Value.getAsBlock() || Value.getAsBlock().getValue().size() != 16)
+ return createStringError(
+ errc::invalid_argument,
+ "failed to parse file entry because the MD5 hash is invalid");
+ std::uninitialized_copy_n(Value.getAsBlock().getValue().begin(), 16,
+ FileEntry.Checksum.Bytes.begin());
+ break;
+ default:
+ break;
+ }
+ }
+ FileNames.push_back(FileEntry);
+ }
+ return Error::success();
+}
+
+Error DWARFDebugLine::Prologue::parse(const DWARFDataExtractor &DebugLineData,
+ uint64_t *OffsetPtr,
+ const DWARFContext &Ctx,
+ const DWARFUnit *U) {
+ const uint64_t PrologueOffset = *OffsetPtr;
+
+ clear();
+ TotalLength = DebugLineData.getRelocatedValue(4, OffsetPtr);
+ if (TotalLength == dwarf::DW_LENGTH_DWARF64) {
+ FormParams.Format = dwarf::DWARF64;
+ TotalLength = DebugLineData.getU64(OffsetPtr);
+ } else if (TotalLength >= dwarf::DW_LENGTH_lo_reserved) {
+ return createStringError(errc::invalid_argument,
+ "parsing line table prologue at offset 0x%8.8" PRIx64
+ " unsupported reserved unit length found of value 0x%8.8" PRIx64,
+ PrologueOffset, TotalLength);
+ }
+ FormParams.Version = DebugLineData.getU16(OffsetPtr);
+ if (getVersion() < 2)
+ return createStringError(errc::not_supported,
+ "parsing line table prologue at offset 0x%8.8" PRIx64
+ " found unsupported version 0x%2.2" PRIx16,
+ PrologueOffset, getVersion());
+
+ if (getVersion() >= 5) {
+ FormParams.AddrSize = DebugLineData.getU8(OffsetPtr);
+ assert((DebugLineData.getAddressSize() == 0 ||
+ DebugLineData.getAddressSize() == getAddressSize()) &&
+ "Line table header and data extractor disagree");
+ SegSelectorSize = DebugLineData.getU8(OffsetPtr);
+ }
+
+ PrologueLength =
+ DebugLineData.getRelocatedValue(sizeofPrologueLength(), OffsetPtr);
+ const uint64_t EndPrologueOffset = PrologueLength + *OffsetPtr;
+ MinInstLength = DebugLineData.getU8(OffsetPtr);
+ if (getVersion() >= 4)
+ MaxOpsPerInst = DebugLineData.getU8(OffsetPtr);
+ DefaultIsStmt = DebugLineData.getU8(OffsetPtr);
+ LineBase = DebugLineData.getU8(OffsetPtr);
+ LineRange = DebugLineData.getU8(OffsetPtr);
+ OpcodeBase = DebugLineData.getU8(OffsetPtr);
+
+ StandardOpcodeLengths.reserve(OpcodeBase - 1);
+ for (uint32_t I = 1; I < OpcodeBase; ++I) {
+ uint8_t OpLen = DebugLineData.getU8(OffsetPtr);
+ StandardOpcodeLengths.push_back(OpLen);
+ }
+
+ if (getVersion() >= 5) {
+ if (Error e = parseV5DirFileTables(
+ DebugLineData, OffsetPtr, EndPrologueOffset, FormParams, Ctx, U,
+ ContentTypes, IncludeDirectories, FileNames)) {
+ return joinErrors(
+ createStringError(
+ errc::invalid_argument,
+ "parsing line table prologue at 0x%8.8" PRIx64
+ " found an invalid directory or file table description at"
+ " 0x%8.8" PRIx64,
+ PrologueOffset, *OffsetPtr),
+ std::move(e));
+ }
+ } else
+ parseV2DirFileTables(DebugLineData, OffsetPtr, EndPrologueOffset,
+ ContentTypes, IncludeDirectories, FileNames);
+
+ if (*OffsetPtr != EndPrologueOffset)
+ return createStringError(errc::invalid_argument,
+ "parsing line table prologue at 0x%8.8" PRIx64
+ " should have ended at 0x%8.8" PRIx64
+ " but it ended at 0x%8.8" PRIx64,
+ PrologueOffset, EndPrologueOffset, *OffsetPtr);
+ return Error::success();
+}
+
+DWARFDebugLine::Row::Row(bool DefaultIsStmt) { reset(DefaultIsStmt); }
+
+void DWARFDebugLine::Row::postAppend() {
+ Discriminator = 0;
+ BasicBlock = false;
+ PrologueEnd = false;
+ EpilogueBegin = false;
+}
+
+void DWARFDebugLine::Row::reset(bool DefaultIsStmt) {
+ Address.Address = 0;
+ Address.SectionIndex = object::SectionedAddress::UndefSection;
+ Line = 1;
+ Column = 0;
+ File = 1;
+ Isa = 0;
+ Discriminator = 0;
+ IsStmt = DefaultIsStmt;
+ BasicBlock = false;
+ EndSequence = false;
+ PrologueEnd = false;
+ EpilogueBegin = false;
+}
+
+void DWARFDebugLine::Row::dumpTableHeader(raw_ostream &OS) {
+ OS << "Address Line Column File ISA Discriminator Flags\n"
+ << "------------------ ------ ------ ------ --- ------------- "
+ "-------------\n";
+}
+
+void DWARFDebugLine::Row::dump(raw_ostream &OS) const {
+ OS << format("0x%16.16" PRIx64 " %6u %6u", Address.Address, Line, Column)
+ << format(" %6u %3u %13u ", File, Isa, Discriminator)
+ << (IsStmt ? " is_stmt" : "") << (BasicBlock ? " basic_block" : "")
+ << (PrologueEnd ? " prologue_end" : "")
+ << (EpilogueBegin ? " epilogue_begin" : "")
+ << (EndSequence ? " end_sequence" : "") << '\n';
+}
+
+DWARFDebugLine::Sequence::Sequence() { reset(); }
+
+void DWARFDebugLine::Sequence::reset() {
+ LowPC = 0;
+ HighPC = 0;
+ SectionIndex = object::SectionedAddress::UndefSection;
+ FirstRowIndex = 0;
+ LastRowIndex = 0;
+ Empty = true;
+}
+
+DWARFDebugLine::LineTable::LineTable() { clear(); }
+
+void DWARFDebugLine::LineTable::dump(raw_ostream &OS,
+ DIDumpOptions DumpOptions) const {
+ Prologue.dump(OS, DumpOptions);
+ OS << '\n';
+
+ if (!Rows.empty()) {
+ Row::dumpTableHeader(OS);
+ for (const Row &R : Rows) {
+ R.dump(OS);
+ }
+ }
+}
+
+void DWARFDebugLine::LineTable::clear() {
+ Prologue.clear();
+ Rows.clear();
+ Sequences.clear();
+}
+
+DWARFDebugLine::ParsingState::ParsingState(struct LineTable *LT)
+ : LineTable(LT) {
+ resetRowAndSequence();
+}
+
+void DWARFDebugLine::ParsingState::resetRowAndSequence() {
+ Row.reset(LineTable->Prologue.DefaultIsStmt);
+ Sequence.reset();
+}
+
+void DWARFDebugLine::ParsingState::appendRowToMatrix() {
+ unsigned RowNumber = LineTable->Rows.size();
+ if (Sequence.Empty) {
+ // Record the beginning of instruction sequence.
+ Sequence.Empty = false;
+ Sequence.LowPC = Row.Address.Address;
+ Sequence.FirstRowIndex = RowNumber;
+ }
+ LineTable->appendRow(Row);
+ if (Row.EndSequence) {
+ // Record the end of instruction sequence.
+ Sequence.HighPC = Row.Address.Address;
+ Sequence.LastRowIndex = RowNumber + 1;
+ Sequence.SectionIndex = Row.Address.SectionIndex;
+ if (Sequence.isValid())
+ LineTable->appendSequence(Sequence);
+ Sequence.reset();
+ }
+ Row.postAppend();
+}
+
+const DWARFDebugLine::LineTable *
+DWARFDebugLine::getLineTable(uint64_t Offset) const {
+ LineTableConstIter Pos = LineTableMap.find(Offset);
+ if (Pos != LineTableMap.end())
+ return &Pos->second;
+ return nullptr;
+}
+
+Expected<const DWARFDebugLine::LineTable *> DWARFDebugLine::getOrParseLineTable(
+ DWARFDataExtractor &DebugLineData, uint64_t Offset, const DWARFContext &Ctx,
+ const DWARFUnit *U, std::function<void(Error)> RecoverableErrorCallback) {
+ if (!DebugLineData.isValidOffset(Offset))
+ return createStringError(errc::invalid_argument, "offset 0x%8.8" PRIx64
+ " is not a valid debug line section offset",
+ Offset);
+
+ std::pair<LineTableIter, bool> Pos =
+ LineTableMap.insert(LineTableMapTy::value_type(Offset, LineTable()));
+ LineTable *LT = &Pos.first->second;
+ if (Pos.second) {
+ if (Error Err =
+ LT->parse(DebugLineData, &Offset, Ctx, U, RecoverableErrorCallback))
+ return std::move(Err);
+ return LT;
+ }
+ return LT;
+}
+
+Error DWARFDebugLine::LineTable::parse(
+ DWARFDataExtractor &DebugLineData, uint64_t *OffsetPtr,
+ const DWARFContext &Ctx, const DWARFUnit *U,
+ std::function<void(Error)> RecoverableErrorCallback, raw_ostream *OS) {
+ const uint64_t DebugLineOffset = *OffsetPtr;
+
+ clear();
+
+ Error PrologueErr = Prologue.parse(DebugLineData, OffsetPtr, Ctx, U);
+
+ if (OS) {
+ // The presence of OS signals verbose dumping.
+ DIDumpOptions DumpOptions;
+ DumpOptions.Verbose = true;
+ Prologue.dump(*OS, DumpOptions);
+ }
+
+ if (PrologueErr)
+ return PrologueErr;
+
+ const uint64_t EndOffset =
+ DebugLineOffset + Prologue.TotalLength + Prologue.sizeofTotalLength();
+
+ // See if we should tell the data extractor the address size.
+ if (DebugLineData.getAddressSize() == 0)
+ DebugLineData.setAddressSize(Prologue.getAddressSize());
+ else
+ assert(Prologue.getAddressSize() == 0 ||
+ Prologue.getAddressSize() == DebugLineData.getAddressSize());
+
+ ParsingState State(this);
+
+ while (*OffsetPtr < EndOffset) {
+ if (OS)
+ *OS << format("0x%08.08" PRIx64 ": ", *OffsetPtr);
+
+ uint8_t Opcode = DebugLineData.getU8(OffsetPtr);
+
+ if (OS)
+ *OS << format("%02.02" PRIx8 " ", Opcode);
+
+ if (Opcode == 0) {
+ // Extended Opcodes always start with a zero opcode followed by
+ // a uleb128 length so you can skip ones you don't know about
+ uint64_t Len = DebugLineData.getULEB128(OffsetPtr);
+ uint64_t ExtOffset = *OffsetPtr;
+
+ // Tolerate zero-length; assume length is correct and soldier on.
+ if (Len == 0) {
+ if (OS)
+ *OS << "Badly formed extended line op (length 0)\n";
+ continue;
+ }
+
+ uint8_t SubOpcode = DebugLineData.getU8(OffsetPtr);
+ if (OS)
+ *OS << LNExtendedString(SubOpcode);
+ switch (SubOpcode) {
+ case DW_LNE_end_sequence:
+ // Set the end_sequence register of the state machine to true and
+ // append a row to the matrix using the current values of the
+ // state-machine registers. Then reset the registers to the initial
+ // values specified above. Every statement program sequence must end
+ // with a DW_LNE_end_sequence instruction which creates a row whose
+ // address is that of the byte after the last target machine instruction
+ // of the sequence.
+ State.Row.EndSequence = true;
+ State.appendRowToMatrix();
+ if (OS) {
+ *OS << "\n";
+ OS->indent(12);
+ State.Row.dump(*OS);
+ }
+ State.resetRowAndSequence();
+ break;
+
+ case DW_LNE_set_address:
+ // Takes a single relocatable address as an operand. The size of the
+ // operand is the size appropriate to hold an address on the target
+ // machine. Set the address register to the value given by the
+ // relocatable address. All of the other statement program opcodes
+ // that affect the address register add a delta to it. This instruction
+ // stores a relocatable value into it instead.
+ //
+ // Make sure the extractor knows the address size. If not, infer it
+ // from the size of the operand.
+ if (DebugLineData.getAddressSize() == 0)
+ DebugLineData.setAddressSize(Len - 1);
+ else if (DebugLineData.getAddressSize() != Len - 1) {
+ return createStringError(errc::invalid_argument,
+ "mismatching address size at offset 0x%8.8" PRIx64
+ " expected 0x%2.2" PRIx8 " found 0x%2.2" PRIx64,
+ ExtOffset, DebugLineData.getAddressSize(),
+ Len - 1);
+ }
+ State.Row.Address.Address = DebugLineData.getRelocatedAddress(
+ OffsetPtr, &State.Row.Address.SectionIndex);
+ if (OS)
+ *OS << format(" (0x%16.16" PRIx64 ")", State.Row.Address.Address);
+ break;
+
+ case DW_LNE_define_file:
+ // Takes 4 arguments. The first is a null terminated string containing
+ // a source file name. The second is an unsigned LEB128 number
+ // representing the directory index of the directory in which the file
+ // was found. The third is an unsigned LEB128 number representing the
+ // time of last modification of the file. The fourth is an unsigned
+ // LEB128 number representing the length in bytes of the file. The time
+ // and length fields may contain LEB128(0) if the information is not
+ // available.
+ //
+ // The directory index represents an entry in the include_directories
+ // section of the statement program prologue. The index is LEB128(0)
+ // if the file was found in the current directory of the compilation,
+ // LEB128(1) if it was found in the first directory in the
+ // include_directories section, and so on. The directory index is
+ // ignored for file names that represent full path names.
+ //
+ // The files are numbered, starting at 1, in the order in which they
+ // appear; the names in the prologue come before names defined by
+ // the DW_LNE_define_file instruction. These numbers are used in the
+ // the file register of the state machine.
+ {
+ FileNameEntry FileEntry;
+ const char *Name = DebugLineData.getCStr(OffsetPtr);
+ FileEntry.Name =
+ DWARFFormValue::createFromPValue(dwarf::DW_FORM_string, Name);
+ FileEntry.DirIdx = DebugLineData.getULEB128(OffsetPtr);
+ FileEntry.ModTime = DebugLineData.getULEB128(OffsetPtr);
+ FileEntry.Length = DebugLineData.getULEB128(OffsetPtr);
+ Prologue.FileNames.push_back(FileEntry);
+ if (OS)
+ *OS << " (" << Name << ", dir=" << FileEntry.DirIdx << ", mod_time="
+ << format("(0x%16.16" PRIx64 ")", FileEntry.ModTime)
+ << ", length=" << FileEntry.Length << ")";
+ }
+ break;
+
+ case DW_LNE_set_discriminator:
+ State.Row.Discriminator = DebugLineData.getULEB128(OffsetPtr);
+ if (OS)
+ *OS << " (" << State.Row.Discriminator << ")";
+ break;
+
+ default:
+ if (OS)
+ *OS << format("Unrecognized extended op 0x%02.02" PRIx8, SubOpcode)
+ << format(" length %" PRIx64, Len);
+ // Len doesn't include the zero opcode byte or the length itself, but
+ // it does include the sub_opcode, so we have to adjust for that.
+ (*OffsetPtr) += Len - 1;
+ break;
+ }
+ // Make sure the stated and parsed lengths are the same.
+ // Otherwise we have an unparseable line-number program.
+ if (*OffsetPtr - ExtOffset != Len)
+ return createStringError(errc::illegal_byte_sequence,
+ "unexpected line op length at offset 0x%8.8" PRIx64
+ " expected 0x%2.2" PRIx64 " found 0x%2.2" PRIx64,
+ ExtOffset, Len, *OffsetPtr - ExtOffset);
+ } else if (Opcode < Prologue.OpcodeBase) {
+ if (OS)
+ *OS << LNStandardString(Opcode);
+ switch (Opcode) {
+ // Standard Opcodes
+ case DW_LNS_copy:
+ // Takes no arguments. Append a row to the matrix using the
+ // current values of the state-machine registers.
+ if (OS) {
+ *OS << "\n";
+ OS->indent(12);
+ State.Row.dump(*OS);
+ *OS << "\n";
+ }
+ State.appendRowToMatrix();
+ break;
+
+ case DW_LNS_advance_pc:
+ // Takes a single unsigned LEB128 operand, multiplies it by the
+ // min_inst_length field of the prologue, and adds the
+ // result to the address register of the state machine.
+ {
+ uint64_t AddrOffset =
+ DebugLineData.getULEB128(OffsetPtr) * Prologue.MinInstLength;
+ State.Row.Address.Address += AddrOffset;
+ if (OS)
+ *OS << " (" << AddrOffset << ")";
+ }
+ break;
+
+ case DW_LNS_advance_line:
+ // Takes a single signed LEB128 operand and adds that value to
+ // the line register of the state machine.
+ State.Row.Line += DebugLineData.getSLEB128(OffsetPtr);
+ if (OS)
+ *OS << " (" << State.Row.Line << ")";
+ break;
+
+ case DW_LNS_set_file:
+ // Takes a single unsigned LEB128 operand and stores it in the file
+ // register of the state machine.
+ State.Row.File = DebugLineData.getULEB128(OffsetPtr);
+ if (OS)
+ *OS << " (" << State.Row.File << ")";
+ break;
+
+ case DW_LNS_set_column:
+ // Takes a single unsigned LEB128 operand and stores it in the
+ // column register of the state machine.
+ State.Row.Column = DebugLineData.getULEB128(OffsetPtr);
+ if (OS)
+ *OS << " (" << State.Row.Column << ")";
+ break;
+
+ case DW_LNS_negate_stmt:
+ // Takes no arguments. Set the is_stmt register of the state
+ // machine to the logical negation of its current value.
+ State.Row.IsStmt = !State.Row.IsStmt;
+ break;
+
+ case DW_LNS_set_basic_block:
+ // Takes no arguments. Set the basic_block register of the
+ // state machine to true
+ State.Row.BasicBlock = true;
+ break;
+
+ case DW_LNS_const_add_pc:
+ // Takes no arguments. Add to the address register of the state
+ // machine the address increment value corresponding to special
+ // opcode 255. The motivation for DW_LNS_const_add_pc is this:
+ // when the statement program needs to advance the address by a
+ // small amount, it can use a single special opcode, which occupies
+ // a single byte. When it needs to advance the address by up to
+ // twice the range of the last special opcode, it can use
+ // DW_LNS_const_add_pc followed by a special opcode, for a total
+ // of two bytes. Only if it needs to advance the address by more
+ // than twice that range will it need to use both DW_LNS_advance_pc
+ // and a special opcode, requiring three or more bytes.
+ {
+ uint8_t AdjustOpcode = 255 - Prologue.OpcodeBase;
+ uint64_t AddrOffset =
+ (AdjustOpcode / Prologue.LineRange) * Prologue.MinInstLength;
+ State.Row.Address.Address += AddrOffset;
+ if (OS)
+ *OS
+ << format(" (0x%16.16" PRIx64 ")", AddrOffset);
+ }
+ break;
+
+ case DW_LNS_fixed_advance_pc:
+ // Takes a single uhalf operand. Add to the address register of
+ // the state machine the value of the (unencoded) operand. This
+ // is the only extended opcode that takes an argument that is not
+ // a variable length number. The motivation for DW_LNS_fixed_advance_pc
+ // is this: existing assemblers cannot emit DW_LNS_advance_pc or
+ // special opcodes because they cannot encode LEB128 numbers or
+ // judge when the computation of a special opcode overflows and
+ // requires the use of DW_LNS_advance_pc. Such assemblers, however,
+ // can use DW_LNS_fixed_advance_pc instead, sacrificing compression.
+ {
+ uint16_t PCOffset = DebugLineData.getRelocatedValue(2, OffsetPtr);
+ State.Row.Address.Address += PCOffset;
+ if (OS)
+ *OS
+ << format(" (0x%4.4" PRIx16 ")", PCOffset);
+ }
+ break;
+
+ case DW_LNS_set_prologue_end:
+ // Takes no arguments. Set the prologue_end register of the
+ // state machine to true
+ State.Row.PrologueEnd = true;
+ break;
+
+ case DW_LNS_set_epilogue_begin:
+ // Takes no arguments. Set the basic_block register of the
+ // state machine to true
+ State.Row.EpilogueBegin = true;
+ break;
+
+ case DW_LNS_set_isa:
+ // Takes a single unsigned LEB128 operand and stores it in the
+ // column register of the state machine.
+ State.Row.Isa = DebugLineData.getULEB128(OffsetPtr);
+ if (OS)
+ *OS << " (" << State.Row.Isa << ")";
+ break;
+
+ default:
+ // Handle any unknown standard opcodes here. We know the lengths
+ // of such opcodes because they are specified in the prologue
+ // as a multiple of LEB128 operands for each opcode.
+ {
+ assert(Opcode - 1U < Prologue.StandardOpcodeLengths.size());
+ uint8_t OpcodeLength = Prologue.StandardOpcodeLengths[Opcode - 1];
+ for (uint8_t I = 0; I < OpcodeLength; ++I) {
+ uint64_t Value = DebugLineData.getULEB128(OffsetPtr);
+ if (OS)
+ *OS << format("Skipping ULEB128 value: 0x%16.16" PRIx64 ")\n",
+ Value);
+ }
+ }
+ break;
+ }
+ } else {
+ // Special Opcodes
+
+ // A special opcode value is chosen based on the amount that needs
+ // to be added to the line and address registers. The maximum line
+ // increment for a special opcode is the value of the line_base
+ // field in the header, plus the value of the line_range field,
+ // minus 1 (line base + line range - 1). If the desired line
+ // increment is greater than the maximum line increment, a standard
+ // opcode must be used instead of a special opcode. The "address
+ // advance" is calculated by dividing the desired address increment
+ // by the minimum_instruction_length field from the header. The
+ // special opcode is then calculated using the following formula:
+ //
+ // opcode = (desired line increment - line_base) +
+ // (line_range * address advance) + opcode_base
+ //
+ // If the resulting opcode is greater than 255, a standard opcode
+ // must be used instead.
+ //
+ // To decode a special opcode, subtract the opcode_base from the
+ // opcode itself to give the adjusted opcode. The amount to
+ // increment the address register is the result of the adjusted
+ // opcode divided by the line_range multiplied by the
+ // minimum_instruction_length field from the header. That is:
+ //
+ // address increment = (adjusted opcode / line_range) *
+ // minimum_instruction_length
+ //
+ // The amount to increment the line register is the line_base plus
+ // the result of the adjusted opcode modulo the line_range. That is:
+ //
+ // line increment = line_base + (adjusted opcode % line_range)
+
+ uint8_t AdjustOpcode = Opcode - Prologue.OpcodeBase;
+ uint64_t AddrOffset =
+ (AdjustOpcode / Prologue.LineRange) * Prologue.MinInstLength;
+ int32_t LineOffset =
+ Prologue.LineBase + (AdjustOpcode % Prologue.LineRange);
+ State.Row.Line += LineOffset;
+ State.Row.Address.Address += AddrOffset;
+
+ if (OS) {
+ *OS << "address += " << AddrOffset << ", line += " << LineOffset
+ << "\n";
+ OS->indent(12);
+ State.Row.dump(*OS);
+ }
+
+ State.appendRowToMatrix();
+ }
+ if(OS)
+ *OS << "\n";
+ }
+
+ if (!State.Sequence.Empty)
+ RecoverableErrorCallback(
+ createStringError(errc::illegal_byte_sequence,
+ "last sequence in debug line table is not terminated!"));
+
+ // Sort all sequences so that address lookup will work faster.
+ if (!Sequences.empty()) {
+ llvm::sort(Sequences, Sequence::orderByHighPC);
+ // Note: actually, instruction address ranges of sequences should not
+ // overlap (in shared objects and executables). If they do, the address
+ // lookup would still work, though, but result would be ambiguous.
+ // We don't report warning in this case. For example,
+ // sometimes .so compiled from multiple object files contains a few
+ // rudimentary sequences for address ranges [0x0, 0xsomething).
+ }
+
+ return Error::success();
+}
+
+uint32_t DWARFDebugLine::LineTable::findRowInSeq(
+ const DWARFDebugLine::Sequence &Seq,
+ object::SectionedAddress Address) const {
+ if (!Seq.containsPC(Address))
+ return UnknownRowIndex;
+ assert(Seq.SectionIndex == Address.SectionIndex);
+ // In some cases, e.g. first instruction in a function, the compiler generates
+ // two entries, both with the same address. We want the last one.
+ //
+ // In general we want a non-empty range: the last row whose address is less
+ // than or equal to Address. This can be computed as upper_bound - 1.
+ DWARFDebugLine::Row Row;
+ Row.Address = Address;
+ RowIter FirstRow = Rows.begin() + Seq.FirstRowIndex;
+ RowIter LastRow = Rows.begin() + Seq.LastRowIndex;
+ assert(FirstRow->Address.Address <= Row.Address.Address &&
+ Row.Address.Address < LastRow[-1].Address.Address);
+ RowIter RowPos = std::upper_bound(FirstRow + 1, LastRow - 1, Row,
+ DWARFDebugLine::Row::orderByAddress) -
+ 1;
+ assert(Seq.SectionIndex == RowPos->Address.SectionIndex);
+ return RowPos - Rows.begin();
+}
+
+uint32_t DWARFDebugLine::LineTable::lookupAddress(
+ object::SectionedAddress Address) const {
+
+ // Search for relocatable addresses
+ uint32_t Result = lookupAddressImpl(Address);
+
+ if (Result != UnknownRowIndex ||
+ Address.SectionIndex == object::SectionedAddress::UndefSection)
+ return Result;
+
+ // Search for absolute addresses
+ Address.SectionIndex = object::SectionedAddress::UndefSection;
+ return lookupAddressImpl(Address);
+}
+
+uint32_t DWARFDebugLine::LineTable::lookupAddressImpl(
+ object::SectionedAddress Address) const {
+ // First, find an instruction sequence containing the given address.
+ DWARFDebugLine::Sequence Sequence;
+ Sequence.SectionIndex = Address.SectionIndex;
+ Sequence.HighPC = Address.Address;
+ SequenceIter It = llvm::upper_bound(Sequences, Sequence,
+ DWARFDebugLine::Sequence::orderByHighPC);
+ if (It == Sequences.end() || It->SectionIndex != Address.SectionIndex)
+ return UnknownRowIndex;
+ return findRowInSeq(*It, Address);
+}
+
+bool DWARFDebugLine::LineTable::lookupAddressRange(
+ object::SectionedAddress Address, uint64_t Size,
+ std::vector<uint32_t> &Result) const {
+
+ // Search for relocatable addresses
+ if (lookupAddressRangeImpl(Address, Size, Result))
+ return true;
+
+ if (Address.SectionIndex == object::SectionedAddress::UndefSection)
+ return false;
+
+ // Search for absolute addresses
+ Address.SectionIndex = object::SectionedAddress::UndefSection;
+ return lookupAddressRangeImpl(Address, Size, Result);
+}
+
+bool DWARFDebugLine::LineTable::lookupAddressRangeImpl(
+ object::SectionedAddress Address, uint64_t Size,
+ std::vector<uint32_t> &Result) const {
+ if (Sequences.empty())
+ return false;
+ uint64_t EndAddr = Address.Address + Size;
+ // First, find an instruction sequence containing the given address.
+ DWARFDebugLine::Sequence Sequence;
+ Sequence.SectionIndex = Address.SectionIndex;
+ Sequence.HighPC = Address.Address;
+ SequenceIter LastSeq = Sequences.end();
+ SequenceIter SeqPos = llvm::upper_bound(
+ Sequences, Sequence, DWARFDebugLine::Sequence::orderByHighPC);
+ if (SeqPos == LastSeq || !SeqPos->containsPC(Address))
+ return false;
+
+ SequenceIter StartPos = SeqPos;
+
+ // Add the rows from the first sequence to the vector, starting with the
+ // index we just calculated
+
+ while (SeqPos != LastSeq && SeqPos->LowPC < EndAddr) {
+ const DWARFDebugLine::Sequence &CurSeq = *SeqPos;
+ // For the first sequence, we need to find which row in the sequence is the
+ // first in our range.
+ uint32_t FirstRowIndex = CurSeq.FirstRowIndex;
+ if (SeqPos == StartPos)
+ FirstRowIndex = findRowInSeq(CurSeq, Address);
+
+ // Figure out the last row in the range.
+ uint32_t LastRowIndex =
+ findRowInSeq(CurSeq, {EndAddr - 1, Address.SectionIndex});
+ if (LastRowIndex == UnknownRowIndex)
+ LastRowIndex = CurSeq.LastRowIndex - 1;
+
+ assert(FirstRowIndex != UnknownRowIndex);
+ assert(LastRowIndex != UnknownRowIndex);
+
+ for (uint32_t I = FirstRowIndex; I <= LastRowIndex; ++I) {
+ Result.push_back(I);
+ }
+
+ ++SeqPos;
+ }
+
+ return true;
+}
+
+Optional<StringRef> DWARFDebugLine::LineTable::getSourceByIndex(uint64_t FileIndex,
+ FileLineInfoKind Kind) const {
+ if (Kind == FileLineInfoKind::None || !Prologue.hasFileAtIndex(FileIndex))
+ return None;
+ const FileNameEntry &Entry = Prologue.getFileNameEntry(FileIndex);
+ if (Optional<const char *> source = Entry.Source.getAsCString())
+ return StringRef(*source);
+ return None;
+}
+
+static bool isPathAbsoluteOnWindowsOrPosix(const Twine &Path) {
+ // Debug info can contain paths from any OS, not necessarily
+ // an OS we're currently running on. Moreover different compilation units can
+ // be compiled on different operating systems and linked together later.
+ return sys::path::is_absolute(Path, sys::path::Style::posix) ||
+ sys::path::is_absolute(Path, sys::path::Style::windows);
+}
+
+bool DWARFDebugLine::Prologue::getFileNameByIndex(
+ uint64_t FileIndex, StringRef CompDir, FileLineInfoKind Kind,
+ std::string &Result, sys::path::Style Style) const {
+ if (Kind == FileLineInfoKind::None || !hasFileAtIndex(FileIndex))
+ return false;
+ const FileNameEntry &Entry = getFileNameEntry(FileIndex);
+ StringRef FileName = Entry.Name.getAsCString().getValue();
+ if (Kind != FileLineInfoKind::AbsoluteFilePath ||
+ isPathAbsoluteOnWindowsOrPosix(FileName)) {
+ Result = FileName;
+ return true;
+ }
+
+ SmallString<16> FilePath;
+ StringRef IncludeDir;
+ // Be defensive about the contents of Entry.
+ if (getVersion() >= 5) {
+ if (Entry.DirIdx < IncludeDirectories.size())
+ IncludeDir = IncludeDirectories[Entry.DirIdx].getAsCString().getValue();
+ } else {
+ if (0 < Entry.DirIdx && Entry.DirIdx <= IncludeDirectories.size())
+ IncludeDir =
+ IncludeDirectories[Entry.DirIdx - 1].getAsCString().getValue();
+
+ // We may still need to append compilation directory of compile unit.
+ // We know that FileName is not absolute, the only way to have an
+ // absolute path at this point would be if IncludeDir is absolute.
+ if (!CompDir.empty() && !isPathAbsoluteOnWindowsOrPosix(IncludeDir))
+ sys::path::append(FilePath, Style, CompDir);
+ }
+
+ // sys::path::append skips empty strings.
+ sys::path::append(FilePath, Style, IncludeDir, FileName);
+ Result = FilePath.str();
+ return true;
+}
+
+bool DWARFDebugLine::LineTable::getFileLineInfoForAddress(
+ object::SectionedAddress Address, const char *CompDir,
+ FileLineInfoKind Kind, DILineInfo &Result) const {
+ // Get the index of row we're looking for in the line table.
+ uint32_t RowIndex = lookupAddress(Address);
+ if (RowIndex == -1U)
+ return false;
+ // Take file number and line/column from the row.
+ const auto &Row = Rows[RowIndex];
+ if (!getFileNameByIndex(Row.File, CompDir, Kind, Result.FileName))
+ return false;
+ Result.Line = Row.Line;
+ Result.Column = Row.Column;
+ Result.Discriminator = Row.Discriminator;
+ Result.Source = getSourceByIndex(Row.File, Kind);
+ return true;
+}
+
+// We want to supply the Unit associated with a .debug_line[.dwo] table when
+// we dump it, if possible, but still dump the table even if there isn't a Unit.
+// Therefore, collect up handles on all the Units that point into the
+// line-table section.
+static DWARFDebugLine::SectionParser::LineToUnitMap
+buildLineToUnitMap(DWARFDebugLine::SectionParser::cu_range CUs,
+ DWARFDebugLine::SectionParser::tu_range TUs) {
+ DWARFDebugLine::SectionParser::LineToUnitMap LineToUnit;
+ for (const auto &CU : CUs)
+ if (auto CUDIE = CU->getUnitDIE())
+ if (auto StmtOffset = toSectionOffset(CUDIE.find(DW_AT_stmt_list)))
+ LineToUnit.insert(std::make_pair(*StmtOffset, &*CU));
+ for (const auto &TU : TUs)
+ if (auto TUDIE = TU->getUnitDIE())
+ if (auto StmtOffset = toSectionOffset(TUDIE.find(DW_AT_stmt_list)))
+ LineToUnit.insert(std::make_pair(*StmtOffset, &*TU));
+ return LineToUnit;
+}
+
+DWARFDebugLine::SectionParser::SectionParser(DWARFDataExtractor &Data,
+ const DWARFContext &C,
+ cu_range CUs, tu_range TUs)
+ : DebugLineData(Data), Context(C) {
+ LineToUnit = buildLineToUnitMap(CUs, TUs);
+ if (!DebugLineData.isValidOffset(Offset))
+ Done = true;
+}
+
+bool DWARFDebugLine::Prologue::totalLengthIsValid() const {
+ return TotalLength == dwarf::DW_LENGTH_DWARF64 ||
+ TotalLength < dwarf::DW_LENGTH_lo_reserved;
+}
+
+DWARFDebugLine::LineTable DWARFDebugLine::SectionParser::parseNext(
+ function_ref<void(Error)> RecoverableErrorCallback,
+ function_ref<void(Error)> UnrecoverableErrorCallback, raw_ostream *OS) {
+ assert(DebugLineData.isValidOffset(Offset) &&
+ "parsing should have terminated");
+ DWARFUnit *U = prepareToParse(Offset);
+ uint64_t OldOffset = Offset;
+ LineTable LT;
+ if (Error Err = LT.parse(DebugLineData, &Offset, Context, U,
+ RecoverableErrorCallback, OS))
+ UnrecoverableErrorCallback(std::move(Err));
+ moveToNextTable(OldOffset, LT.Prologue);
+ return LT;
+}
+
+void DWARFDebugLine::SectionParser::skip(
+ function_ref<void(Error)> ErrorCallback) {
+ assert(DebugLineData.isValidOffset(Offset) &&
+ "parsing should have terminated");
+ DWARFUnit *U = prepareToParse(Offset);
+ uint64_t OldOffset = Offset;
+ LineTable LT;
+ if (Error Err = LT.Prologue.parse(DebugLineData, &Offset, Context, U))
+ ErrorCallback(std::move(Err));
+ moveToNextTable(OldOffset, LT.Prologue);
+}
+
+DWARFUnit *DWARFDebugLine::SectionParser::prepareToParse(uint64_t Offset) {
+ DWARFUnit *U = nullptr;
+ auto It = LineToUnit.find(Offset);
+ if (It != LineToUnit.end())
+ U = It->second;
+ DebugLineData.setAddressSize(U ? U->getAddressByteSize() : 0);
+ return U;
+}
+
+void DWARFDebugLine::SectionParser::moveToNextTable(uint64_t OldOffset,
+ const Prologue &P) {
+ // If the length field is not valid, we don't know where the next table is, so
+ // cannot continue to parse. Mark the parser as done, and leave the Offset
+ // value as it currently is. This will be the end of the bad length field.
+ if (!P.totalLengthIsValid()) {
+ Done = true;
+ return;
+ }
+
+ Offset = OldOffset + P.TotalLength + P.sizeofTotalLength();
+ if (!DebugLineData.isValidOffset(Offset)) {
+ Done = true;
+ }
+}
diff --git a/third_party/llvm-project/DWARFDebugLoc.cpp b/third_party/llvm-project/DWARFDebugLoc.cpp
new file mode 100644
index 000000000..e0faaed03
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugLoc.cpp
@@ -0,0 +1,326 @@
+//===- DWARFDebugLoc.cpp --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugLoc.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cinttypes>
+#include <cstdint>
+
+using namespace llvm;
+
+// When directly dumping the .debug_loc without a compile unit, we have to guess
+// at the DWARF version. This only affects DW_OP_call_ref, which is a rare
+// expression that LLVM doesn't produce. Guessing the wrong version means we
+// won't be able to pretty print expressions in DWARF2 binaries produced by
+// non-LLVM tools.
+static void dumpExpression(raw_ostream &OS, ArrayRef<uint8_t> Data,
+ bool IsLittleEndian, unsigned AddressSize,
+ const MCRegisterInfo *MRI, DWARFUnit *U) {
+ DWARFDataExtractor Extractor(toStringRef(Data), IsLittleEndian, AddressSize);
+ DWARFExpression(Extractor, dwarf::DWARF_VERSION, AddressSize).print(OS, MRI, U);
+}
+
+void DWARFDebugLoc::LocationList::dump(raw_ostream &OS, uint64_t BaseAddress,
+ bool IsLittleEndian,
+ unsigned AddressSize,
+ const MCRegisterInfo *MRI, DWARFUnit *U,
+ DIDumpOptions DumpOpts,
+ unsigned Indent) const {
+ for (const Entry &E : Entries) {
+ OS << '\n';
+ OS.indent(Indent);
+ OS << format("[0x%*.*" PRIx64 ", ", AddressSize * 2, AddressSize * 2,
+ BaseAddress + E.Begin);
+ OS << format(" 0x%*.*" PRIx64 ")", AddressSize * 2, AddressSize * 2,
+ BaseAddress + E.End);
+ OS << ": ";
+
+ dumpExpression(OS, E.Loc, IsLittleEndian, AddressSize, MRI, U);
+ }
+}
+
+DWARFDebugLoc::LocationList const *
+DWARFDebugLoc::getLocationListAtOffset(uint64_t Offset) const {
+ auto It = partition_point(
+ Locations, [=](const LocationList &L) { return L.Offset < Offset; });
+ if (It != Locations.end() && It->Offset == Offset)
+ return &(*It);
+ return nullptr;
+}
+
+void DWARFDebugLoc::dump(raw_ostream &OS, const MCRegisterInfo *MRI, DIDumpOptions DumpOpts,
+ Optional<uint64_t> Offset) const {
+ auto DumpLocationList = [&](const LocationList &L) {
+ OS << format("0x%8.8" PRIx64 ": ", L.Offset);
+ L.dump(OS, 0, IsLittleEndian, AddressSize, MRI, nullptr, DumpOpts, 12);
+ OS << "\n";
+ };
+
+ if (Offset) {
+ if (auto *L = getLocationListAtOffset(*Offset))
+ DumpLocationList(*L);
+ return;
+ }
+
+ for (const LocationList &L : Locations) {
+ DumpLocationList(L);
+ if (&L != &Locations.back())
+ OS << '\n';
+ }
+}
+
+Expected<DWARFDebugLoc::LocationList>
+DWARFDebugLoc::parseOneLocationList(const DWARFDataExtractor &Data,
+ uint64_t *Offset) {
+ LocationList LL;
+ LL.Offset = *Offset;
+ AddressSize = Data.getAddressSize();
+ DataExtractor::Cursor C(*Offset);
+
+ // 2.6.2 Location Lists
+ // A location list entry consists of:
+ while (true) {
+ Entry E;
+
+ // 1. A beginning address offset. ...
+ E.Begin = Data.getRelocatedAddress(C);
+
+ // 2. An ending address offset. ...
+ E.End = Data.getRelocatedAddress(C);
+
+ if (Error Err = C.takeError())
+ return std::move(Err);
+
+ // The end of any given location list is marked by an end of list entry,
+ // which consists of a 0 for the beginning address offset and a 0 for the
+ // ending address offset.
+ if (E.Begin == 0 && E.End == 0) {
+ *Offset = C.tell();
+ return LL;
+ }
+
+ if (E.Begin != (AddressSize == 4 ? -1U : -1ULL)) {
+ unsigned Bytes = Data.getU16(C);
+ // A single location description describing the location of the object...
+ Data.getU8(C, E.Loc, Bytes);
+ }
+
+ LL.Entries.push_back(std::move(E));
+ }
+}
+
+void DWARFDebugLoc::parse(const DWARFDataExtractor &data) {
+ IsLittleEndian = data.isLittleEndian();
+ AddressSize = data.getAddressSize();
+
+ uint64_t Offset = 0;
+ while (Offset < data.getData().size()) {
+ if (auto LL = parseOneLocationList(data, &Offset))
+ Locations.push_back(std::move(*LL));
+ else {
+ logAllUnhandledErrors(LL.takeError(), WithColor::error());
+ break;
+ }
+ }
+}
+
+Error DWARFDebugLoclists::visitLocationList(
+ const DWARFDataExtractor &Data, uint64_t *Offset, uint16_t Version,
+ llvm::function_ref<bool(const Entry &)> F) {
+
+ DataExtractor::Cursor C(*Offset);
+ bool Continue = true;
+ while (Continue) {
+ Entry E;
+ E.Offset = C.tell();
+ E.Kind = Data.getU8(C);
+ switch (E.Kind) {
+ case dwarf::DW_LLE_end_of_list:
+ break;
+ case dwarf::DW_LLE_base_addressx:
+ E.Value0 = Data.getULEB128(C);
+ break;
+ case dwarf::DW_LLE_startx_length:
+ E.Value0 = Data.getULEB128(C);
+ // Pre-DWARF 5 has different interpretation of the length field. We have
+ // to support both pre- and standartized styles for the compatibility.
+ if (Version < 5)
+ E.Value1 = Data.getU32(C);
+ else
+ E.Value1 = Data.getULEB128(C);
+ break;
+ case dwarf::DW_LLE_offset_pair:
+ E.Value0 = Data.getULEB128(C);
+ E.Value1 = Data.getULEB128(C);
+ break;
+ case dwarf::DW_LLE_base_address:
+ E.Value0 = Data.getRelocatedAddress(C);
+ break;
+ case dwarf::DW_LLE_start_length:
+ E.Value0 = Data.getRelocatedAddress(C);
+ E.Value1 = Data.getULEB128(C);
+ break;
+ case dwarf::DW_LLE_startx_endx:
+ case dwarf::DW_LLE_default_location:
+ case dwarf::DW_LLE_start_end:
+ default:
+ cantFail(C.takeError());
+ return createStringError(errc::illegal_byte_sequence,
+ "LLE of kind %x not supported", (int)E.Kind);
+ }
+
+ if (E.Kind != dwarf::DW_LLE_base_address &&
+ E.Kind != dwarf::DW_LLE_base_addressx &&
+ E.Kind != dwarf::DW_LLE_end_of_list) {
+ unsigned Bytes = Version >= 5 ? Data.getULEB128(C) : Data.getU16(C);
+ // A single location description describing the location of the object...
+ Data.getU8(C, E.Loc, Bytes);
+ }
+
+ if (!C)
+ return C.takeError();
+ Continue = F(E) && E.Kind != dwarf::DW_LLE_end_of_list;
+ }
+ *Offset = C.tell();
+ return Error::success();
+}
+
+bool DWARFDebugLoclists::dumpLocationList(const DWARFDataExtractor &Data,
+ uint64_t *Offset, uint16_t Version,
+ raw_ostream &OS, uint64_t BaseAddr,
+ const MCRegisterInfo *MRI,
+ DWARFUnit *U, DIDumpOptions DumpOpts,
+ unsigned Indent) {
+ size_t MaxEncodingStringLength = 0;
+ if (DumpOpts.Verbose) {
+#define HANDLE_DW_LLE(ID, NAME) \
+ MaxEncodingStringLength = std::max(MaxEncodingStringLength, \
+ dwarf::LocListEncodingString(ID).size());
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+
+ OS << format("0x%8.8" PRIx64 ": ", *Offset);
+ Error E = visitLocationList(Data, Offset, Version, [&](const Entry &E) {
+ E.dump(OS, BaseAddr, Data.isLittleEndian(), Data.getAddressSize(), MRI, U,
+ DumpOpts, Indent, MaxEncodingStringLength);
+ return true;
+ });
+ if (E) {
+ OS << "\n";
+ OS.indent(Indent);
+ OS << "error: " << toString(std::move(E));
+ return false;
+ }
+ return true;
+}
+
+void DWARFDebugLoclists::Entry::dump(raw_ostream &OS, uint64_t &BaseAddr,
+ bool IsLittleEndian, unsigned AddressSize,
+ const MCRegisterInfo *MRI, DWARFUnit *U,
+ DIDumpOptions DumpOpts, unsigned Indent,
+ size_t MaxEncodingStringLength) const {
+ if (DumpOpts.Verbose) {
+ OS << "\n";
+ OS.indent(Indent);
+ auto EncodingString = dwarf::LocListEncodingString(Kind);
+ // Unsupported encodings should have been reported during parsing.
+ assert(!EncodingString.empty() && "Unknown loclist entry encoding");
+ OS << format("%s%*c", EncodingString.data(),
+ MaxEncodingStringLength - EncodingString.size() + 1, '(');
+ switch (Kind) {
+ case dwarf::DW_LLE_startx_length:
+ case dwarf::DW_LLE_start_length:
+ case dwarf::DW_LLE_offset_pair:
+ OS << format("0x%*.*" PRIx64 ", 0x%*.*" PRIx64, AddressSize * 2,
+ AddressSize * 2, Value0, AddressSize * 2, AddressSize * 2,
+ Value1);
+ break;
+ case dwarf::DW_LLE_base_addressx:
+ case dwarf::DW_LLE_base_address:
+ OS << format("0x%*.*" PRIx64, AddressSize * 2, AddressSize * 2,
+ Value0);
+ break;
+ case dwarf::DW_LLE_end_of_list:
+ break;
+ }
+ OS << ')';
+ }
+ auto PrintPrefix = [&] {
+ OS << "\n";
+ OS.indent(Indent);
+ if (DumpOpts.Verbose)
+ OS << format("%*s", MaxEncodingStringLength, (const char *)"=> ");
+ };
+ switch (Kind) {
+ case dwarf::DW_LLE_startx_length:
+ PrintPrefix();
+ OS << "Addr idx " << Value0 << " (w/ length " << Value1 << "): ";
+ break;
+ case dwarf::DW_LLE_start_length:
+ PrintPrefix();
+ DWARFAddressRange(Value0, Value0 + Value1)
+ .dump(OS, AddressSize, DumpOpts);
+ OS << ": ";
+ break;
+ case dwarf::DW_LLE_offset_pair:
+ PrintPrefix();
+ DWARFAddressRange(BaseAddr + Value0, BaseAddr + Value1)
+ .dump(OS, AddressSize, DumpOpts);
+ OS << ": ";
+ break;
+ case dwarf::DW_LLE_base_addressx:
+ if (!DumpOpts.Verbose)
+ return;
+ break;
+ case dwarf::DW_LLE_end_of_list:
+ if (!DumpOpts.Verbose)
+ return;
+ break;
+ case dwarf::DW_LLE_base_address:
+ BaseAddr = Value0;
+ if (!DumpOpts.Verbose)
+ return;
+ break;
+ default:
+ llvm_unreachable("unreachable locations list kind");
+ }
+
+ dumpExpression(OS, Loc, IsLittleEndian, AddressSize, MRI, U);
+}
+
+void DWARFDebugLoclists::dumpRange(const DWARFDataExtractor &Data,
+ uint64_t StartOffset, uint64_t Size,
+ uint16_t Version, raw_ostream &OS,
+ uint64_t BaseAddr, const MCRegisterInfo *MRI,
+ DIDumpOptions DumpOpts) {
+ if (!Data.isValidOffsetForDataOfSize(StartOffset, Size)) {
+ OS << "Invalid dump range\n";
+ return;
+ }
+ uint64_t Offset = StartOffset;
+ StringRef Separator;
+ bool CanContinue = true;
+ while (CanContinue && Offset < StartOffset + Size) {
+ OS << Separator;
+ Separator = "\n";
+
+ CanContinue = dumpLocationList(Data, &Offset, Version, OS, BaseAddr, MRI,
+ nullptr, DumpOpts, /*Indent=*/12);
+ OS << '\n';
+ }
+}
diff --git a/third_party/llvm-project/DWARFDebugMacro.cpp b/third_party/llvm-project/DWARFDebugMacro.cpp
new file mode 100644
index 000000000..8cb259ebc
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugMacro.cpp
@@ -0,0 +1,106 @@
+//===- DWARFDebugMacro.cpp ------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugMacro.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstdint>
+
+using namespace llvm;
+using namespace dwarf;
+
+void DWARFDebugMacro::dump(raw_ostream &OS) const {
+ unsigned IndLevel = 0;
+ for (const auto &Macros : MacroLists) {
+ for (const Entry &E : Macros) {
+ // There should not be DW_MACINFO_end_file when IndLevel is Zero. However,
+ // this check handles the case of corrupted ".debug_macinfo" section.
+ if (IndLevel > 0)
+ IndLevel -= (E.Type == DW_MACINFO_end_file);
+ // Print indentation.
+ for (unsigned I = 0; I < IndLevel; I++)
+ OS << " ";
+ IndLevel += (E.Type == DW_MACINFO_start_file);
+
+ WithColor(OS, HighlightColor::Macro).get() << MacinfoString(E.Type);
+ switch (E.Type) {
+ default:
+ // Got a corrupted ".debug_macinfo" section (invalid macinfo type).
+ break;
+ case DW_MACINFO_define:
+ case DW_MACINFO_undef:
+ OS << " - lineno: " << E.Line;
+ OS << " macro: " << E.MacroStr;
+ break;
+ case DW_MACINFO_start_file:
+ OS << " - lineno: " << E.Line;
+ OS << " filenum: " << E.File;
+ break;
+ case DW_MACINFO_end_file:
+ break;
+ case DW_MACINFO_vendor_ext:
+ OS << " - constant: " << E.ExtConstant;
+ OS << " string: " << E.ExtStr;
+ break;
+ }
+ OS << "\n";
+ }
+ OS << "\n";
+ }
+}
+
+void DWARFDebugMacro::parse(DataExtractor data) {
+ uint64_t Offset = 0;
+ MacroList *M = nullptr;
+ while (data.isValidOffset(Offset)) {
+ if (!M) {
+ MacroLists.emplace_back();
+ M = &MacroLists.back();
+ }
+ // A macro list entry consists of:
+ M->emplace_back();
+ Entry &E = M->back();
+ // 1. Macinfo type
+ E.Type = data.getULEB128(&Offset);
+
+ if (E.Type == 0) {
+ // Reached end of a ".debug_macinfo" section contribution.
+ continue;
+ }
+
+ switch (E.Type) {
+ default:
+ // Got a corrupted ".debug_macinfo" section (invalid macinfo type).
+ // Push the corrupted entry to the list and halt parsing.
+ E.Type = DW_MACINFO_invalid;
+ return;
+ case DW_MACINFO_define:
+ case DW_MACINFO_undef:
+ // 2. Source line
+ E.Line = data.getULEB128(&Offset);
+ // 3. Macro string
+ E.MacroStr = data.getCStr(&Offset);
+ break;
+ case DW_MACINFO_start_file:
+ // 2. Source line
+ E.Line = data.getULEB128(&Offset);
+ // 3. Source file id
+ E.File = data.getULEB128(&Offset);
+ break;
+ case DW_MACINFO_end_file:
+ break;
+ case DW_MACINFO_vendor_ext:
+ // 2. Vendor extension constant
+ E.ExtConstant = data.getULEB128(&Offset);
+ // 3. Vendor extension string
+ E.ExtStr = data.getCStr(&Offset);
+ break;
+ }
+ }
+}
diff --git a/third_party/llvm-project/DWARFDebugPubTable.cpp b/third_party/llvm-project/DWARFDebugPubTable.cpp
new file mode 100644
index 000000000..ab71b239c
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugPubTable.cpp
@@ -0,0 +1,69 @@
+//===- DWARFDebugPubTable.cpp ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugPubTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstdint>
+
+using namespace llvm;
+using namespace dwarf;
+
+DWARFDebugPubTable::DWARFDebugPubTable(const DWARFObject &Obj,
+ const DWARFSection &Sec,
+ bool LittleEndian, bool GnuStyle)
+ : GnuStyle(GnuStyle) {
+ DWARFDataExtractor PubNames(Obj, Sec, LittleEndian, 0);
+ uint64_t Offset = 0;
+ while (PubNames.isValidOffset(Offset)) {
+ Sets.push_back({});
+ Set &SetData = Sets.back();
+
+ SetData.Length = PubNames.getU32(&Offset);
+ SetData.Version = PubNames.getU16(&Offset);
+ SetData.Offset = PubNames.getRelocatedValue(4, &Offset);
+ SetData.Size = PubNames.getU32(&Offset);
+
+ while (Offset < Sec.Data.size()) {
+ uint32_t DieRef = PubNames.getU32(&Offset);
+ if (DieRef == 0)
+ break;
+ uint8_t IndexEntryValue = GnuStyle ? PubNames.getU8(&Offset) : 0;
+ StringRef Name = PubNames.getCStrRef(&Offset);
+ SetData.Entries.push_back(
+ {DieRef, PubIndexEntryDescriptor(IndexEntryValue), Name});
+ }
+ }
+}
+
+void DWARFDebugPubTable::dump(raw_ostream &OS) const {
+ for (const Set &S : Sets) {
+ OS << "length = " << format("0x%08x", S.Length);
+ OS << " version = " << format("0x%04x", S.Version);
+ OS << " unit_offset = " << format("0x%08" PRIx64, S.Offset);
+ OS << " unit_size = " << format("0x%08x", S.Size) << '\n';
+ OS << (GnuStyle ? "Offset Linkage Kind Name\n"
+ : "Offset Name\n");
+
+ for (const Entry &E : S.Entries) {
+ OS << format("0x%8.8" PRIx64 " ", E.SecOffset);
+ if (GnuStyle) {
+ StringRef EntryLinkage =
+ GDBIndexEntryLinkageString(E.Descriptor.Linkage);
+ StringRef EntryKind = dwarf::GDBIndexEntryKindString(E.Descriptor.Kind);
+ OS << format("%-8s", EntryLinkage.data()) << ' '
+ << format("%-8s", EntryKind.data()) << ' ';
+ }
+ OS << '\"' << E.Name << "\"\n";
+ }
+ }
+}
diff --git a/third_party/llvm-project/DWARFDebugRangeList.cpp b/third_party/llvm-project/DWARFDebugRangeList.cpp
new file mode 100644
index 000000000..1a1857d8c
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugRangeList.cpp
@@ -0,0 +1,95 @@
+//===- DWARFDebugRangesList.cpp -------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugRangeList.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cinttypes>
+#include <cstdint>
+
+using namespace llvm;
+
+void DWARFDebugRangeList::clear() {
+ Offset = -1ULL;
+ AddressSize = 0;
+ Entries.clear();
+}
+
+Error DWARFDebugRangeList::extract(const DWARFDataExtractor &data,
+ uint64_t *offset_ptr) {
+ clear();
+ if (!data.isValidOffset(*offset_ptr))
+ return createStringError(errc::invalid_argument,
+ "invalid range list offset 0x%" PRIx64, *offset_ptr);
+
+ AddressSize = data.getAddressSize();
+ if (AddressSize != 4 && AddressSize != 8)
+ return createStringError(errc::invalid_argument,
+ "invalid address size: %" PRIu8, AddressSize);
+ Offset = *offset_ptr;
+ while (true) {
+ RangeListEntry Entry;
+ Entry.SectionIndex = -1ULL;
+
+ uint64_t prev_offset = *offset_ptr;
+ Entry.StartAddress = data.getRelocatedAddress(offset_ptr);
+ Entry.EndAddress =
+ data.getRelocatedAddress(offset_ptr, &Entry.SectionIndex);
+
+ // Check that both values were extracted correctly.
+ if (*offset_ptr != prev_offset + 2 * AddressSize) {
+ clear();
+ return createStringError(errc::invalid_argument,
+ "invalid range list entry at offset 0x%" PRIx64,
+ prev_offset);
+ }
+ if (Entry.isEndOfListEntry())
+ break;
+ Entries.push_back(Entry);
+ }
+ return Error::success();
+}
+
+void DWARFDebugRangeList::dump(raw_ostream &OS) const {
+ for (const RangeListEntry &RLE : Entries) {
+ const char *format_str =
+ (AddressSize == 4 ? "%08" PRIx64 " %08" PRIx64 " %08" PRIx64 "\n"
+ : "%08" PRIx64 " %016" PRIx64 " %016" PRIx64 "\n");
+ OS << format(format_str, Offset, RLE.StartAddress, RLE.EndAddress);
+ }
+ OS << format("%08" PRIx64 " <End of list>\n", Offset);
+}
+
+DWARFAddressRangesVector DWARFDebugRangeList::getAbsoluteRanges(
+ llvm::Optional<object::SectionedAddress> BaseAddr) const {
+ DWARFAddressRangesVector Res;
+ for (const RangeListEntry &RLE : Entries) {
+ if (RLE.isBaseAddressSelectionEntry(AddressSize)) {
+ BaseAddr = {RLE.EndAddress, RLE.SectionIndex};
+ continue;
+ }
+
+ DWARFAddressRange E;
+ E.LowPC = RLE.StartAddress;
+ E.HighPC = RLE.EndAddress;
+ E.SectionIndex = RLE.SectionIndex;
+ // Base address of a range list entry is determined by the closest preceding
+ // base address selection entry in the same range list. It defaults to the
+ // base address of the compilation unit if there is no such entry.
+ if (BaseAddr) {
+ E.LowPC += BaseAddr->Address;
+ E.HighPC += BaseAddr->Address;
+ if (E.SectionIndex == -1ULL)
+ E.SectionIndex = BaseAddr->SectionIndex;
+ }
+ Res.push_back(E);
+ }
+ return Res;
+}
diff --git a/third_party/llvm-project/DWARFDebugRnglists.cpp b/third_party/llvm-project/DWARFDebugRnglists.cpp
new file mode 100644
index 000000000..f6785b89e
--- /dev/null
+++ b/third_party/llvm-project/DWARFDebugRnglists.cpp
@@ -0,0 +1,245 @@
+//===- DWARFDebugRnglists.cpp ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDebugRnglists.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+Error RangeListEntry::extract(DWARFDataExtractor Data, uint64_t End,
+ uint64_t *OffsetPtr) {
+ Offset = *OffsetPtr;
+ SectionIndex = -1ULL;
+ // The caller should guarantee that we have at least 1 byte available, so
+ // we just assert instead of revalidate.
+ assert(*OffsetPtr < End &&
+ "not enough space to extract a rangelist encoding");
+ uint8_t Encoding = Data.getU8(OffsetPtr);
+
+ switch (Encoding) {
+ case dwarf::DW_RLE_end_of_list:
+ Value0 = Value1 = 0;
+ break;
+ // TODO: Support other encodings.
+ case dwarf::DW_RLE_base_addressx: {
+ uint64_t PreviousOffset = *OffsetPtr - 1;
+ Value0 = Data.getULEB128(OffsetPtr);
+ if (End < *OffsetPtr)
+ return createStringError(
+ errc::invalid_argument,
+ "read past end of table when reading "
+ "DW_RLE_base_addressx encoding at offset 0x%" PRIx64,
+ PreviousOffset);
+ break;
+ }
+ case dwarf::DW_RLE_startx_endx:
+ return createStringError(errc::not_supported,
+ "unsupported rnglists encoding DW_RLE_startx_endx at "
+ "offset 0x%" PRIx64,
+ *OffsetPtr - 1);
+ case dwarf::DW_RLE_startx_length: {
+ uint64_t PreviousOffset = *OffsetPtr - 1;
+ Value0 = Data.getULEB128(OffsetPtr);
+ Value1 = Data.getULEB128(OffsetPtr);
+ if (End < *OffsetPtr)
+ return createStringError(
+ errc::invalid_argument,
+ "read past end of table when reading "
+ "DW_RLE_startx_length encoding at offset 0x%" PRIx64,
+ PreviousOffset);
+ break;
+ }
+ case dwarf::DW_RLE_offset_pair: {
+ uint64_t PreviousOffset = *OffsetPtr - 1;
+ Value0 = Data.getULEB128(OffsetPtr);
+ Value1 = Data.getULEB128(OffsetPtr);
+ if (End < *OffsetPtr)
+ return createStringError(errc::invalid_argument,
+ "read past end of table when reading "
+ "DW_RLE_offset_pair encoding at offset 0x%" PRIx64,
+ PreviousOffset);
+ break;
+ }
+ case dwarf::DW_RLE_base_address: {
+ if ((End - *OffsetPtr) < Data.getAddressSize())
+ return createStringError(errc::invalid_argument,
+ "insufficient space remaining in table for "
+ "DW_RLE_base_address encoding at offset 0x%" PRIx64,
+ *OffsetPtr - 1);
+ Value0 = Data.getRelocatedAddress(OffsetPtr, &SectionIndex);
+ break;
+ }
+ case dwarf::DW_RLE_start_end: {
+ if ((End - *OffsetPtr) < unsigned(Data.getAddressSize() * 2))
+ return createStringError(errc::invalid_argument,
+ "insufficient space remaining in table for "
+ "DW_RLE_start_end encoding "
+ "at offset 0x%" PRIx64,
+ *OffsetPtr - 1);
+ Value0 = Data.getRelocatedAddress(OffsetPtr, &SectionIndex);
+ Value1 = Data.getRelocatedAddress(OffsetPtr);
+ break;
+ }
+ case dwarf::DW_RLE_start_length: {
+ uint64_t PreviousOffset = *OffsetPtr - 1;
+ Value0 = Data.getRelocatedAddress(OffsetPtr, &SectionIndex);
+ Value1 = Data.getULEB128(OffsetPtr);
+ if (End < *OffsetPtr)
+ return createStringError(errc::invalid_argument,
+ "read past end of table when reading "
+ "DW_RLE_start_length encoding at offset 0x%" PRIx64,
+ PreviousOffset);
+ break;
+ }
+ default:
+ return createStringError(errc::not_supported,
+ "unknown rnglists encoding 0x%" PRIx32
+ " at offset 0x%" PRIx64,
+ uint32_t(Encoding), *OffsetPtr - 1);
+ }
+
+ EntryKind = Encoding;
+ return Error::success();
+}
+
+DWARFAddressRangesVector DWARFDebugRnglist::getAbsoluteRanges(
+ llvm::Optional<object::SectionedAddress> BaseAddr, DWARFUnit &U) const {
+ DWARFAddressRangesVector Res;
+ for (const RangeListEntry &RLE : Entries) {
+ if (RLE.EntryKind == dwarf::DW_RLE_end_of_list)
+ break;
+ if (RLE.EntryKind == dwarf::DW_RLE_base_addressx) {
+ BaseAddr = U.getAddrOffsetSectionItem(RLE.Value0);
+ if (!BaseAddr)
+ BaseAddr = {RLE.Value0, -1ULL};
+ continue;
+ }
+ if (RLE.EntryKind == dwarf::DW_RLE_base_address) {
+ BaseAddr = {RLE.Value0, RLE.SectionIndex};
+ continue;
+ }
+
+ DWARFAddressRange E;
+ E.SectionIndex = RLE.SectionIndex;
+ if (BaseAddr && E.SectionIndex == -1ULL)
+ E.SectionIndex = BaseAddr->SectionIndex;
+
+ switch (RLE.EntryKind) {
+ case dwarf::DW_RLE_offset_pair:
+ E.LowPC = RLE.Value0;
+ E.HighPC = RLE.Value1;
+ if (BaseAddr) {
+ E.LowPC += BaseAddr->Address;
+ E.HighPC += BaseAddr->Address;
+ }
+ break;
+ case dwarf::DW_RLE_start_end:
+ E.LowPC = RLE.Value0;
+ E.HighPC = RLE.Value1;
+ break;
+ case dwarf::DW_RLE_start_length:
+ E.LowPC = RLE.Value0;
+ E.HighPC = E.LowPC + RLE.Value1;
+ break;
+ case dwarf::DW_RLE_startx_length: {
+ auto Start = U.getAddrOffsetSectionItem(RLE.Value0);
+ if (!Start)
+ Start = {0, -1ULL};
+ E.SectionIndex = Start->SectionIndex;
+ E.LowPC = Start->Address;
+ E.HighPC = E.LowPC + RLE.Value1;
+ break;
+ }
+ default:
+ // Unsupported encodings should have been reported during extraction,
+ // so we should not run into any here.
+ llvm_unreachable("Unsupported range list encoding");
+ }
+ Res.push_back(E);
+ }
+ return Res;
+}
+
+void RangeListEntry::dump(
+ raw_ostream &OS, uint8_t AddrSize, uint8_t MaxEncodingStringLength,
+ uint64_t &CurrentBase, DIDumpOptions DumpOpts,
+ llvm::function_ref<Optional<object::SectionedAddress>(uint32_t)>
+ LookupPooledAddress) const {
+ auto PrintRawEntry = [](raw_ostream &OS, const RangeListEntry &Entry,
+ uint8_t AddrSize, DIDumpOptions DumpOpts) {
+ if (DumpOpts.Verbose) {
+ DumpOpts.DisplayRawContents = true;
+ DWARFAddressRange(Entry.Value0, Entry.Value1)
+ .dump(OS, AddrSize, DumpOpts);
+ OS << " => ";
+ }
+ };
+
+ if (DumpOpts.Verbose) {
+ // Print the section offset in verbose mode.
+ OS << format("0x%8.8" PRIx64 ":", Offset);
+ auto EncodingString = dwarf::RangeListEncodingString(EntryKind);
+ // Unsupported encodings should have been reported during parsing.
+ assert(!EncodingString.empty() && "Unknown range entry encoding");
+ OS << format(" [%s%*c", EncodingString.data(),
+ MaxEncodingStringLength - EncodingString.size() + 1, ']');
+ if (EntryKind != dwarf::DW_RLE_end_of_list)
+ OS << ": ";
+ }
+
+ switch (EntryKind) {
+ case dwarf::DW_RLE_end_of_list:
+ OS << (DumpOpts.Verbose ? "" : "<End of list>");
+ break;
+ case dwarf::DW_RLE_base_addressx: {
+ if (auto SA = LookupPooledAddress(Value0))
+ CurrentBase = SA->Address;
+ else
+ CurrentBase = Value0;
+ if (!DumpOpts.Verbose)
+ return;
+ OS << format(" 0x%*.*" PRIx64, AddrSize * 2, AddrSize * 2, Value0);
+ break;
+ }
+ case dwarf::DW_RLE_base_address:
+ // In non-verbose mode we do not print anything for this entry.
+ CurrentBase = Value0;
+ if (!DumpOpts.Verbose)
+ return;
+ OS << format(" 0x%*.*" PRIx64, AddrSize * 2, AddrSize * 2, Value0);
+ break;
+ case dwarf::DW_RLE_start_length:
+ PrintRawEntry(OS, *this, AddrSize, DumpOpts);
+ DWARFAddressRange(Value0, Value0 + Value1).dump(OS, AddrSize, DumpOpts);
+ break;
+ case dwarf::DW_RLE_offset_pair:
+ PrintRawEntry(OS, *this, AddrSize, DumpOpts);
+ DWARFAddressRange(Value0 + CurrentBase, Value1 + CurrentBase)
+ .dump(OS, AddrSize, DumpOpts);
+ break;
+ case dwarf::DW_RLE_start_end:
+ DWARFAddressRange(Value0, Value1).dump(OS, AddrSize, DumpOpts);
+ break;
+ case dwarf::DW_RLE_startx_length: {
+ PrintRawEntry(OS, *this, AddrSize, DumpOpts);
+ uint64_t Start = 0;
+ if (auto SA = LookupPooledAddress(Value0))
+ Start = SA->Address;
+ DWARFAddressRange(Start, Start + Value1).dump(OS, AddrSize, DumpOpts);
+ break;
+ }
+ default:
+ llvm_unreachable("Unsupported range list encoding");
+ }
+ OS << "\n";
+}
diff --git a/third_party/llvm-project/DWARFDie.cpp b/third_party/llvm-project/DWARFDie.cpp
new file mode 100644
index 000000000..ebc8cc565
--- /dev/null
+++ b/third_party/llvm-project/DWARFDie.cpp
@@ -0,0 +1,747 @@
+//===- DWARFDie.cpp -------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallSet.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRangeList.h"
+#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatAdapters.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cinttypes>
+#include <cstdint>
+#include <string>
+#include <utility>
+
+using namespace llvm;
+using namespace dwarf;
+using namespace object;
+
+static void dumpApplePropertyAttribute(raw_ostream &OS, uint64_t Val) {
+ OS << " (";
+ do {
+ uint64_t Shift = countTrailingZeros(Val);
+ assert(Shift < 64 && "undefined behavior");
+ uint64_t Bit = 1ULL << Shift;
+ auto PropName = ApplePropertyString(Bit);
+ if (!PropName.empty())
+ OS << PropName;
+ else
+ OS << format("DW_APPLE_PROPERTY_0x%" PRIx64, Bit);
+ if (!(Val ^= Bit))
+ break;
+ OS << ", ";
+ } while (true);
+ OS << ")";
+}
+
+static void dumpRanges(const DWARFObject &Obj, raw_ostream &OS,
+ const DWARFAddressRangesVector &Ranges,
+ unsigned AddressSize, unsigned Indent,
+ const DIDumpOptions &DumpOpts) {
+ if (!DumpOpts.ShowAddresses)
+ return;
+
+ ArrayRef<SectionName> SectionNames;
+ if (DumpOpts.Verbose)
+ SectionNames = Obj.getSectionNames();
+
+ for (const DWARFAddressRange &R : Ranges) {
+ OS << '\n';
+ OS.indent(Indent);
+ R.dump(OS, AddressSize);
+
+ DWARFFormValue::dumpAddressSection(Obj, OS, DumpOpts, R.SectionIndex);
+ }
+}
+
+static void dumpLocation(raw_ostream &OS, DWARFFormValue &FormValue,
+ DWARFUnit *U, unsigned Indent,
+ DIDumpOptions DumpOpts) {
+ DWARFContext &Ctx = U->getContext();
+ const DWARFObject &Obj = Ctx.getDWARFObj();
+ const MCRegisterInfo *MRI = Ctx.getRegisterInfo();
+ if (FormValue.isFormClass(DWARFFormValue::FC_Block) ||
+ FormValue.isFormClass(DWARFFormValue::FC_Exprloc)) {
+ ArrayRef<uint8_t> Expr = *FormValue.getAsBlock();
+ DataExtractor Data(StringRef((const char *)Expr.data(), Expr.size()),
+ Ctx.isLittleEndian(), 0);
+ DWARFExpression(Data, U->getVersion(), U->getAddressByteSize())
+ .print(OS, MRI, U);
+ return;
+ }
+
+ if (FormValue.isFormClass(DWARFFormValue::FC_SectionOffset)) {
+ uint64_t Offset = *FormValue.getAsSectionOffset();
+ uint64_t BaseAddr = 0;
+ if (Optional<object::SectionedAddress> BA = U->getBaseAddress())
+ BaseAddr = BA->Address;
+ auto LLDumpOpts = DumpOpts;
+ LLDumpOpts.Verbose = false;
+
+ if (!U->isDWOUnit() && !U->getLocSection()->Data.empty()) {
+ DWARFDebugLoc DebugLoc;
+ DWARFDataExtractor Data(Obj, *U->getLocSection(), Ctx.isLittleEndian(),
+ Obj.getAddressSize());
+
+ FormValue.dump(OS, DumpOpts);
+ OS << ": ";
+
+ if (Expected<DWARFDebugLoc::LocationList> LL =
+ DebugLoc.parseOneLocationList(Data, &Offset)) {
+ LL->dump(OS, BaseAddr, Ctx.isLittleEndian(), Obj.getAddressSize(), MRI,
+ U, LLDumpOpts, Indent);
+ } else {
+ OS << '\n';
+ OS.indent(Indent);
+ OS << formatv("error extracting location list: {0}",
+ fmt_consume(LL.takeError()));
+ }
+ return;
+ }
+
+ bool UseLocLists = !U->isDWOUnit();
+ auto Data =
+ UseLocLists
+ ? DWARFDataExtractor(Obj, Obj.getLoclistsSection(),
+ Ctx.isLittleEndian(), Obj.getAddressSize())
+ : DWARFDataExtractor(U->getLocSectionData(), Ctx.isLittleEndian(),
+ Obj.getAddressSize());
+
+ if (!Data.getData().empty()) {
+ // Old-style location list were used in DWARF v4 (.debug_loc.dwo section).
+ // Modern locations list (.debug_loclists) are used starting from v5.
+ // Ideally we should take the version from the .debug_loclists section
+ // header, but using CU's version for simplicity.
+ DWARFDebugLoclists::dumpLocationList(
+ Data, &Offset, UseLocLists ? U->getVersion() : 4, OS, BaseAddr, MRI,
+ U, LLDumpOpts, Indent);
+ }
+ return;
+ }
+
+ FormValue.dump(OS, DumpOpts);
+}
+
+/// Dump the name encoded in the type tag.
+static void dumpTypeTagName(raw_ostream &OS, dwarf::Tag T) {
+ StringRef TagStr = TagString(T);
+ if (!TagStr.startswith("DW_TAG_") || !TagStr.endswith("_type"))
+ return;
+ OS << TagStr.substr(7, TagStr.size() - 12) << " ";
+}
+
+static void dumpArrayType(raw_ostream &OS, const DWARFDie &D) {
+ Optional<uint64_t> Bound;
+ for (const DWARFDie &C : D.children())
+ if (C.getTag() == DW_TAG_subrange_type) {
+ Optional<uint64_t> LB;
+ Optional<uint64_t> Count;
+ Optional<uint64_t> UB;
+ Optional<unsigned> DefaultLB;
+ if (Optional<DWARFFormValue> L = C.find(DW_AT_lower_bound))
+ LB = L->getAsUnsignedConstant();
+ if (Optional<DWARFFormValue> CountV = C.find(DW_AT_count))
+ Count = CountV->getAsUnsignedConstant();
+ if (Optional<DWARFFormValue> UpperV = C.find(DW_AT_upper_bound))
+ UB = UpperV->getAsUnsignedConstant();
+ if (Optional<DWARFFormValue> LV =
+ D.getDwarfUnit()->getUnitDIE().find(DW_AT_language))
+ if (Optional<uint64_t> LC = LV->getAsUnsignedConstant())
+ if ((DefaultLB =
+ LanguageLowerBound(static_cast<dwarf::SourceLanguage>(*LC))))
+ if (LB && *LB == *DefaultLB)
+ LB = None;
+ if (!LB && !Count && !UB)
+ OS << "[]";
+ else if (!LB && (Count || UB) && DefaultLB)
+ OS << '[' << (Count ? *Count : *UB - *DefaultLB + 1) << ']';
+ else {
+ OS << "[[";
+ if (LB)
+ OS << *LB;
+ else
+ OS << '?';
+ OS << ", ";
+ if (Count)
+ if (LB)
+ OS << *LB + *Count;
+ else
+ OS << "? + " << *Count;
+ else if (UB)
+ OS << *UB + 1;
+ else
+ OS << '?';
+ OS << ")]";
+ }
+ }
+}
+
+/// Recursively dump the DIE type name when applicable.
+static void dumpTypeName(raw_ostream &OS, const DWARFDie &D) {
+ if (!D.isValid())
+ return;
+
+ if (const char *Name = D.getName(DINameKind::LinkageName)) {
+ OS << Name;
+ return;
+ }
+
+ // FIXME: We should have pretty printers per language. Currently we print
+ // everything as if it was C++ and fall back to the TAG type name.
+ const dwarf::Tag T = D.getTag();
+ switch (T) {
+ case DW_TAG_array_type:
+ case DW_TAG_pointer_type:
+ case DW_TAG_ptr_to_member_type:
+ case DW_TAG_reference_type:
+ case DW_TAG_rvalue_reference_type:
+ case DW_TAG_subroutine_type:
+ break;
+ default:
+ dumpTypeTagName(OS, T);
+ }
+
+ // Follow the DW_AT_type if possible.
+ DWARFDie TypeDie = D.getAttributeValueAsReferencedDie(DW_AT_type);
+ dumpTypeName(OS, TypeDie);
+
+ switch (T) {
+ case DW_TAG_subroutine_type: {
+ if (!TypeDie)
+ OS << "void";
+ OS << '(';
+ bool First = true;
+ for (const DWARFDie &C : D.children()) {
+ if (C.getTag() == DW_TAG_formal_parameter) {
+ if (!First)
+ OS << ", ";
+ First = false;
+ dumpTypeName(OS, C.getAttributeValueAsReferencedDie(DW_AT_type));
+ }
+ }
+ OS << ')';
+ break;
+ }
+ case DW_TAG_array_type: {
+ dumpArrayType(OS, D);
+ break;
+ }
+ case DW_TAG_pointer_type:
+ OS << '*';
+ break;
+ case DW_TAG_ptr_to_member_type:
+ if (DWARFDie Cont =
+ D.getAttributeValueAsReferencedDie(DW_AT_containing_type)) {
+ dumpTypeName(OS << ' ', Cont);
+ OS << "::";
+ }
+ OS << '*';
+ break;
+ case DW_TAG_reference_type:
+ OS << '&';
+ break;
+ case DW_TAG_rvalue_reference_type:
+ OS << "&&";
+ break;
+ default:
+ break;
+ }
+}
+
+static void dumpAttribute(raw_ostream &OS, const DWARFDie &Die,
+ uint64_t *OffsetPtr, dwarf::Attribute Attr,
+ dwarf::Form Form, unsigned Indent,
+ DIDumpOptions DumpOpts) {
+ if (!Die.isValid())
+ return;
+ const char BaseIndent[] = " ";
+ OS << BaseIndent;
+ OS.indent(Indent + 2);
+ WithColor(OS, HighlightColor::Attribute) << formatv("{0}", Attr);
+
+ if (DumpOpts.Verbose || DumpOpts.ShowForm)
+ OS << formatv(" [{0}]", Form);
+
+ DWARFUnit *U = Die.getDwarfUnit();
+ DWARFFormValue FormValue = DWARFFormValue::createFromUnit(Form, U, OffsetPtr);
+
+ OS << "\t(";
+
+ StringRef Name;
+ std::string File;
+ auto Color = HighlightColor::Enumerator;
+ if (Attr == DW_AT_decl_file || Attr == DW_AT_call_file) {
+ Color = HighlightColor::String;
+ if (const auto *LT = U->getContext().getLineTableForUnit(U))
+ if (LT->getFileNameByIndex(
+ FormValue.getAsUnsignedConstant().getValue(),
+ U->getCompilationDir(),
+ DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, File)) {
+ File = '"' + File + '"';
+ Name = File;
+ }
+ } else if (Optional<uint64_t> Val = FormValue.getAsUnsignedConstant())
+ Name = AttributeValueString(Attr, *Val);
+
+ if (!Name.empty())
+ WithColor(OS, Color) << Name;
+ else if (Attr == DW_AT_decl_line || Attr == DW_AT_call_line)
+ OS << *FormValue.getAsUnsignedConstant();
+ else if (Attr == DW_AT_high_pc && !DumpOpts.ShowForm && !DumpOpts.Verbose &&
+ FormValue.getAsUnsignedConstant()) {
+ if (DumpOpts.ShowAddresses) {
+ // Print the actual address rather than the offset.
+ uint64_t LowPC, HighPC, Index;
+ if (Die.getLowAndHighPC(LowPC, HighPC, Index))
+ OS << format("0x%016" PRIx64, HighPC);
+ else
+ FormValue.dump(OS, DumpOpts);
+ }
+ } else if (DWARFAttribute::mayHaveLocationDescription(Attr))
+ dumpLocation(OS, FormValue, U, sizeof(BaseIndent) + Indent + 4, DumpOpts);
+ else
+ FormValue.dump(OS, DumpOpts);
+
+ std::string Space = DumpOpts.ShowAddresses ? " " : "";
+
+ // We have dumped the attribute raw value. For some attributes
+ // having both the raw value and the pretty-printed value is
+ // interesting. These attributes are handled below.
+ if (Attr == DW_AT_specification || Attr == DW_AT_abstract_origin) {
+ if (const char *Name =
+ Die.getAttributeValueAsReferencedDie(FormValue).getName(
+ DINameKind::LinkageName))
+ OS << Space << "\"" << Name << '\"';
+ } else if (Attr == DW_AT_type) {
+ OS << Space << "\"";
+ dumpTypeName(OS, Die.getAttributeValueAsReferencedDie(FormValue));
+ OS << '"';
+ } else if (Attr == DW_AT_APPLE_property_attribute) {
+ if (Optional<uint64_t> OptVal = FormValue.getAsUnsignedConstant())
+ dumpApplePropertyAttribute(OS, *OptVal);
+ } else if (Attr == DW_AT_ranges) {
+ const DWARFObject &Obj = Die.getDwarfUnit()->getContext().getDWARFObj();
+ // For DW_FORM_rnglistx we need to dump the offset separately, since
+ // we have only dumped the index so far.
+ if (FormValue.getForm() == DW_FORM_rnglistx)
+ if (auto RangeListOffset =
+ U->getRnglistOffset(*FormValue.getAsSectionOffset())) {
+ DWARFFormValue FV = DWARFFormValue::createFromUValue(
+ dwarf::DW_FORM_sec_offset, *RangeListOffset);
+ FV.dump(OS, DumpOpts);
+ }
+ if (auto RangesOrError = Die.getAddressRanges())
+ dumpRanges(Obj, OS, RangesOrError.get(), U->getAddressByteSize(),
+ sizeof(BaseIndent) + Indent + 4, DumpOpts);
+ else
+ WithColor::error() << "decoding address ranges: "
+ << toString(RangesOrError.takeError()) << '\n';
+ }
+
+ OS << ")\n";
+}
+
+bool DWARFDie::isSubprogramDIE() const { return getTag() == DW_TAG_subprogram; }
+
+bool DWARFDie::isSubroutineDIE() const {
+ auto Tag = getTag();
+ return Tag == DW_TAG_subprogram || Tag == DW_TAG_inlined_subroutine;
+}
+
+Optional<DWARFFormValue> DWARFDie::find(dwarf::Attribute Attr) const {
+ if (!isValid())
+ return None;
+ auto AbbrevDecl = getAbbreviationDeclarationPtr();
+ if (AbbrevDecl)
+ return AbbrevDecl->getAttributeValue(getOffset(), Attr, *U);
+ return None;
+}
+
+Optional<DWARFFormValue>
+DWARFDie::find(ArrayRef<dwarf::Attribute> Attrs) const {
+ if (!isValid())
+ return None;
+ auto AbbrevDecl = getAbbreviationDeclarationPtr();
+ if (AbbrevDecl) {
+ for (auto Attr : Attrs) {
+ if (auto Value = AbbrevDecl->getAttributeValue(getOffset(), Attr, *U))
+ return Value;
+ }
+ }
+ return None;
+}
+
+Optional<DWARFFormValue>
+DWARFDie::findRecursively(ArrayRef<dwarf::Attribute> Attrs) const {
+ std::vector<DWARFDie> Worklist;
+ Worklist.push_back(*this);
+
+ // Keep track if DIEs already seen to prevent infinite recursion.
+ // Empirically we rarely see a depth of more than 3 when dealing with valid
+ // DWARF. This corresponds to following the DW_AT_abstract_origin and
+ // DW_AT_specification just once.
+ SmallSet<DWARFDie, 3> Seen;
+ Seen.insert(*this);
+
+ while (!Worklist.empty()) {
+ DWARFDie Die = Worklist.back();
+ Worklist.pop_back();
+
+ if (!Die.isValid())
+ continue;
+
+ if (auto Value = Die.find(Attrs))
+ return Value;
+
+ if (auto D = Die.getAttributeValueAsReferencedDie(DW_AT_abstract_origin))
+ if (Seen.insert(D).second)
+ Worklist.push_back(D);
+
+ if (auto D = Die.getAttributeValueAsReferencedDie(DW_AT_specification))
+ if (Seen.insert(D).second)
+ Worklist.push_back(D);
+ }
+
+ return None;
+}
+
+DWARFDie
+DWARFDie::getAttributeValueAsReferencedDie(dwarf::Attribute Attr) const {
+ if (Optional<DWARFFormValue> F = find(Attr))
+ return getAttributeValueAsReferencedDie(*F);
+ return DWARFDie();
+}
+
+DWARFDie
+DWARFDie::getAttributeValueAsReferencedDie(const DWARFFormValue &V) const {
+ if (auto SpecRef = V.getAsRelativeReference()) {
+ if (SpecRef->Unit)
+ return SpecRef->Unit->getDIEForOffset(SpecRef->Unit->getOffset() + SpecRef->Offset);
+ if (auto SpecUnit = U->getUnitVector().getUnitForOffset(SpecRef->Offset))
+ return SpecUnit->getDIEForOffset(SpecRef->Offset);
+ }
+ return DWARFDie();
+}
+
+Optional<uint64_t> DWARFDie::getRangesBaseAttribute() const {
+ return toSectionOffset(find({DW_AT_rnglists_base, DW_AT_GNU_ranges_base}));
+}
+
+Optional<uint64_t> DWARFDie::getHighPC(uint64_t LowPC) const {
+ if (auto FormValue = find(DW_AT_high_pc)) {
+ if (auto Address = FormValue->getAsAddress()) {
+ // High PC is an address.
+ return Address;
+ }
+ if (auto Offset = FormValue->getAsUnsignedConstant()) {
+ // High PC is an offset from LowPC.
+ return LowPC + *Offset;
+ }
+ }
+ return None;
+}
+
+bool DWARFDie::getLowAndHighPC(uint64_t &LowPC, uint64_t &HighPC,
+ uint64_t &SectionIndex) const {
+ auto F = find(DW_AT_low_pc);
+ auto LowPcAddr = toSectionedAddress(F);
+ if (!LowPcAddr)
+ return false;
+ if (auto HighPcAddr = getHighPC(LowPcAddr->Address)) {
+ LowPC = LowPcAddr->Address;
+ HighPC = *HighPcAddr;
+ SectionIndex = LowPcAddr->SectionIndex;
+ return true;
+ }
+ return false;
+}
+
+Expected<DWARFAddressRangesVector> DWARFDie::getAddressRanges() const {
+ if (isNULL())
+ return DWARFAddressRangesVector();
+ // Single range specified by low/high PC.
+ uint64_t LowPC, HighPC, Index;
+ if (getLowAndHighPC(LowPC, HighPC, Index))
+ return DWARFAddressRangesVector{{LowPC, HighPC, Index}};
+
+ Optional<DWARFFormValue> Value = find(DW_AT_ranges);
+ if (Value) {
+ if (Value->getForm() == DW_FORM_rnglistx)
+ return U->findRnglistFromIndex(*Value->getAsSectionOffset());
+ return U->findRnglistFromOffset(*Value->getAsSectionOffset());
+ }
+ return DWARFAddressRangesVector();
+}
+
+void DWARFDie::collectChildrenAddressRanges(
+ DWARFAddressRangesVector &Ranges) const {
+ if (isNULL())
+ return;
+ if (isSubprogramDIE()) {
+ if (auto DIERangesOrError = getAddressRanges())
+ Ranges.insert(Ranges.end(), DIERangesOrError.get().begin(),
+ DIERangesOrError.get().end());
+ else
+ llvm::consumeError(DIERangesOrError.takeError());
+ }
+
+ for (auto Child : children())
+ Child.collectChildrenAddressRanges(Ranges);
+}
+
+bool DWARFDie::addressRangeContainsAddress(const uint64_t Address) const {
+ auto RangesOrError = getAddressRanges();
+ if (!RangesOrError) {
+ llvm::consumeError(RangesOrError.takeError());
+ return false;
+ }
+
+ for (const auto &R : RangesOrError.get())
+ if (R.LowPC <= Address && Address < R.HighPC)
+ return true;
+ return false;
+}
+
+const char *DWARFDie::getSubroutineName(DINameKind Kind) const {
+ if (!isSubroutineDIE())
+ return nullptr;
+ return getName(Kind);
+}
+
+const char *DWARFDie::getName(DINameKind Kind) const {
+ if (!isValid() || Kind == DINameKind::None)
+ return nullptr;
+ // Try to get mangled name only if it was asked for.
+ if (Kind == DINameKind::LinkageName) {
+ if (auto Name = dwarf::toString(
+ findRecursively({DW_AT_MIPS_linkage_name, DW_AT_linkage_name}),
+ nullptr))
+ return Name;
+ }
+ if (auto Name = dwarf::toString(findRecursively(DW_AT_name), nullptr))
+ return Name;
+ return nullptr;
+}
+
+uint64_t DWARFDie::getDeclLine() const {
+ return toUnsigned(findRecursively(DW_AT_decl_line), 0);
+}
+
+void DWARFDie::getCallerFrame(uint32_t &CallFile, uint32_t &CallLine,
+ uint32_t &CallColumn,
+ uint32_t &CallDiscriminator) const {
+ CallFile = toUnsigned(find(DW_AT_call_file), 0);
+ CallLine = toUnsigned(find(DW_AT_call_line), 0);
+ CallColumn = toUnsigned(find(DW_AT_call_column), 0);
+ CallDiscriminator = toUnsigned(find(DW_AT_GNU_discriminator), 0);
+}
+
+/// Helper to dump a DIE with all of its parents, but no siblings.
+static unsigned dumpParentChain(DWARFDie Die, raw_ostream &OS, unsigned Indent,
+ DIDumpOptions DumpOpts, unsigned Depth = 0) {
+ if (!Die)
+ return Indent;
+ if (DumpOpts.ParentRecurseDepth > 0 && Depth >= DumpOpts.ParentRecurseDepth)
+ return Indent;
+ Indent = dumpParentChain(Die.getParent(), OS, Indent, DumpOpts, Depth + 1);
+ Die.dump(OS, Indent, DumpOpts);
+ return Indent + 2;
+}
+
+void DWARFDie::dump(raw_ostream &OS, unsigned Indent,
+ DIDumpOptions DumpOpts) const {
+ if (!isValid())
+ return;
+ DWARFDataExtractor debug_info_data = U->getDebugInfoExtractor();
+ const uint64_t Offset = getOffset();
+ uint64_t offset = Offset;
+ if (DumpOpts.ShowParents) {
+ DIDumpOptions ParentDumpOpts = DumpOpts;
+ ParentDumpOpts.ShowParents = false;
+ ParentDumpOpts.ShowChildren = false;
+ Indent = dumpParentChain(getParent(), OS, Indent, ParentDumpOpts);
+ }
+
+ if (debug_info_data.isValidOffset(offset)) {
+ uint32_t abbrCode = debug_info_data.getULEB128(&offset);
+ if (DumpOpts.ShowAddresses)
+ WithColor(OS, HighlightColor::Address).get()
+ << format("\n0x%8.8" PRIx64 ": ", Offset);
+
+ if (abbrCode) {
+ auto AbbrevDecl = getAbbreviationDeclarationPtr();
+ if (AbbrevDecl) {
+ WithColor(OS, HighlightColor::Tag).get().indent(Indent)
+ << formatv("{0}", getTag());
+ if (DumpOpts.Verbose)
+ OS << format(" [%u] %c", abbrCode,
+ AbbrevDecl->hasChildren() ? '*' : ' ');
+ OS << '\n';
+
+ // Dump all data in the DIE for the attributes.
+ for (const auto &AttrSpec : AbbrevDecl->attributes()) {
+ if (AttrSpec.Form == DW_FORM_implicit_const) {
+ // We are dumping .debug_info section ,
+ // implicit_const attribute values are not really stored here,
+ // but in .debug_abbrev section. So we just skip such attrs.
+ continue;
+ }
+ dumpAttribute(OS, *this, &offset, AttrSpec.Attr, AttrSpec.Form,
+ Indent, DumpOpts);
+ }
+
+ DWARFDie child = getFirstChild();
+ if (DumpOpts.ShowChildren && DumpOpts.ChildRecurseDepth > 0 && child) {
+ DumpOpts.ChildRecurseDepth--;
+ DIDumpOptions ChildDumpOpts = DumpOpts;
+ ChildDumpOpts.ShowParents = false;
+ while (child) {
+ child.dump(OS, Indent + 2, ChildDumpOpts);
+ child = child.getSibling();
+ }
+ }
+ } else {
+ OS << "Abbreviation code not found in 'debug_abbrev' class for code: "
+ << abbrCode << '\n';
+ }
+ } else {
+ OS.indent(Indent) << "NULL\n";
+ }
+ }
+}
+
+LLVM_DUMP_METHOD void DWARFDie::dump() const { dump(llvm::errs(), 0); }
+
+DWARFDie DWARFDie::getParent() const {
+ if (isValid())
+ return U->getParent(Die);
+ return DWARFDie();
+}
+
+DWARFDie DWARFDie::getSibling() const {
+ if (isValid())
+ return U->getSibling(Die);
+ return DWARFDie();
+}
+
+DWARFDie DWARFDie::getPreviousSibling() const {
+ if (isValid())
+ return U->getPreviousSibling(Die);
+ return DWARFDie();
+}
+
+DWARFDie DWARFDie::getFirstChild() const {
+ if (isValid())
+ return U->getFirstChild(Die);
+ return DWARFDie();
+}
+
+DWARFDie DWARFDie::getLastChild() const {
+ if (isValid())
+ return U->getLastChild(Die);
+ return DWARFDie();
+}
+
+iterator_range<DWARFDie::attribute_iterator> DWARFDie::attributes() const {
+ return make_range(attribute_iterator(*this, false),
+ attribute_iterator(*this, true));
+}
+
+DWARFDie::attribute_iterator::attribute_iterator(DWARFDie D, bool End)
+ : Die(D), Index(0) {
+ auto AbbrDecl = Die.getAbbreviationDeclarationPtr();
+ assert(AbbrDecl && "Must have abbreviation declaration");
+ if (End) {
+ // This is the end iterator so we set the index to the attribute count.
+ Index = AbbrDecl->getNumAttributes();
+ } else {
+ // This is the begin iterator so we extract the value for this->Index.
+ AttrValue.Offset = D.getOffset() + AbbrDecl->getCodeByteSize();
+ updateForIndex(*AbbrDecl, 0);
+ }
+}
+
+void DWARFDie::attribute_iterator::updateForIndex(
+ const DWARFAbbreviationDeclaration &AbbrDecl, uint32_t I) {
+ Index = I;
+ // AbbrDecl must be valid before calling this function.
+ auto NumAttrs = AbbrDecl.getNumAttributes();
+ if (Index < NumAttrs) {
+ AttrValue.Attr = AbbrDecl.getAttrByIndex(Index);
+ // Add the previous byte size of any previous attribute value.
+ AttrValue.Offset += AttrValue.ByteSize;
+ uint64_t ParseOffset = AttrValue.Offset;
+ auto U = Die.getDwarfUnit();
+ assert(U && "Die must have valid DWARF unit");
+ AttrValue.Value = DWARFFormValue::createFromUnit(
+ AbbrDecl.getFormByIndex(Index), U, &ParseOffset);
+ AttrValue.ByteSize = ParseOffset - AttrValue.Offset;
+ } else {
+ assert(Index == NumAttrs && "Indexes should be [0, NumAttrs) only");
+ AttrValue = {};
+ }
+}
+
+DWARFDie::attribute_iterator &DWARFDie::attribute_iterator::operator++() {
+ if (auto AbbrDecl = Die.getAbbreviationDeclarationPtr())
+ updateForIndex(*AbbrDecl, Index + 1);
+ return *this;
+}
+
+bool DWARFAttribute::mayHaveLocationDescription(dwarf::Attribute Attr) {
+ switch (Attr) {
+ // From the DWARF v5 specification.
+ case DW_AT_location:
+ case DW_AT_byte_size:
+ case DW_AT_bit_size:
+ case DW_AT_string_length:
+ case DW_AT_lower_bound:
+ case DW_AT_return_addr:
+ case DW_AT_bit_stride:
+ case DW_AT_upper_bound:
+ case DW_AT_count:
+ case DW_AT_data_member_location:
+ case DW_AT_frame_base:
+ case DW_AT_segment:
+ case DW_AT_static_link:
+ case DW_AT_use_location:
+ case DW_AT_vtable_elem_location:
+ case DW_AT_allocated:
+ case DW_AT_associated:
+ case DW_AT_byte_stride:
+ case DW_AT_rank:
+ case DW_AT_call_value:
+ case DW_AT_call_origin:
+ case DW_AT_call_target:
+ case DW_AT_call_target_clobbered:
+ case DW_AT_call_data_location:
+ case DW_AT_call_data_value:
+ // Extensions.
+ case DW_AT_GNU_call_site_value:
+ case DW_AT_GNU_call_site_target:
+ return true;
+ default:
+ return false;
+ }
+}
diff --git a/third_party/llvm-project/DWARFEmitter.cpp b/third_party/llvm-project/DWARFEmitter.cpp
new file mode 100644
index 000000000..dbaa06111
--- /dev/null
+++ b/third_party/llvm-project/DWARFEmitter.cpp
@@ -0,0 +1,419 @@
+//===- DWARFEmitter - Convert YAML to DWARF binary data -------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// The DWARF component of yaml2obj. Provided as library code for tests.
+///
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ObjectYAML/DWARFEmitter.h"
+#include "DWARFVisitor.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ObjectYAML/DWARFYAML.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Host.h"
+#include "llvm/Support/LEB128.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/SwapByteOrder.h"
+#include "llvm/Support/YAMLTraits.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <memory>
+#include <string>
+#include <vector>
+
+using namespace llvm;
+
+template <typename T>
+static void writeInteger(T Integer, raw_ostream &OS, bool IsLittleEndian) {
+ if (IsLittleEndian != sys::IsLittleEndianHost)
+ sys::swapByteOrder(Integer);
+ OS.write(reinterpret_cast<char *>(&Integer), sizeof(T));
+}
+
+static void writeVariableSizedInteger(uint64_t Integer, size_t Size,
+ raw_ostream &OS, bool IsLittleEndian) {
+ if (8 == Size)
+ writeInteger((uint64_t)Integer, OS, IsLittleEndian);
+ else if (4 == Size)
+ writeInteger((uint32_t)Integer, OS, IsLittleEndian);
+ else if (2 == Size)
+ writeInteger((uint16_t)Integer, OS, IsLittleEndian);
+ else if (1 == Size)
+ writeInteger((uint8_t)Integer, OS, IsLittleEndian);
+ else
+ assert(false && "Invalid integer write size.");
+}
+
+static void ZeroFillBytes(raw_ostream &OS, size_t Size) {
+ std::vector<uint8_t> FillData;
+ FillData.insert(FillData.begin(), Size, 0);
+ OS.write(reinterpret_cast<char *>(FillData.data()), Size);
+}
+
+static void writeInitialLength(const DWARFYAML::InitialLength &Length,
+ raw_ostream &OS, bool IsLittleEndian) {
+ writeInteger((uint32_t)Length.TotalLength, OS, IsLittleEndian);
+ if (Length.isDWARF64())
+ writeInteger((uint64_t)Length.TotalLength64, OS, IsLittleEndian);
+}
+
+void DWARFYAML::EmitDebugStr(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ for (auto Str : DI.DebugStrings) {
+ OS.write(Str.data(), Str.size());
+ OS.write('\0');
+ }
+}
+
+void DWARFYAML::EmitDebugAbbrev(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ for (auto AbbrevDecl : DI.AbbrevDecls) {
+ encodeULEB128(AbbrevDecl.Code, OS);
+ encodeULEB128(AbbrevDecl.Tag, OS);
+ OS.write(AbbrevDecl.Children);
+ for (auto Attr : AbbrevDecl.Attributes) {
+ encodeULEB128(Attr.Attribute, OS);
+ encodeULEB128(Attr.Form, OS);
+ if (Attr.Form == dwarf::DW_FORM_implicit_const)
+ encodeSLEB128(Attr.Value, OS);
+ }
+ encodeULEB128(0, OS);
+ encodeULEB128(0, OS);
+ }
+}
+
+void DWARFYAML::EmitDebugAranges(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ for (auto Range : DI.ARanges) {
+ auto HeaderStart = OS.tell();
+ writeInitialLength(Range.Length, OS, DI.IsLittleEndian);
+ writeInteger((uint16_t)Range.Version, OS, DI.IsLittleEndian);
+ writeInteger((uint32_t)Range.CuOffset, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)Range.AddrSize, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)Range.SegSize, OS, DI.IsLittleEndian);
+
+ auto HeaderSize = OS.tell() - HeaderStart;
+ auto FirstDescriptor = alignTo(HeaderSize, Range.AddrSize * 2);
+ ZeroFillBytes(OS, FirstDescriptor - HeaderSize);
+
+ for (auto Descriptor : Range.Descriptors) {
+ writeVariableSizedInteger(Descriptor.Address, Range.AddrSize, OS,
+ DI.IsLittleEndian);
+ writeVariableSizedInteger(Descriptor.Length, Range.AddrSize, OS,
+ DI.IsLittleEndian);
+ }
+ ZeroFillBytes(OS, Range.AddrSize * 2);
+ }
+}
+
+// XXX BINARYEN
+void DWARFYAML::EmitDebugRanges(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ // As DwarfStreamer.cpp says, "The debug_range section
+ // format is totally trivial, consisting just of pairs of address
+ // sized addresses describing the ranges." and apparently it ends
+ // with a null termination of a pair of zeros
+ for (auto Range : DI.Ranges) {
+ writeInteger((uint32_t)Range.Start, OS, DI.IsLittleEndian);
+ writeInteger((uint32_t)Range.End, OS, DI.IsLittleEndian);
+ }
+}
+
+void DWARFYAML::EmitPubSection(raw_ostream &OS,
+ const DWARFYAML::PubSection &Sect,
+ bool IsLittleEndian) {
+ writeInitialLength(Sect.Length, OS, IsLittleEndian);
+ writeInteger((uint16_t)Sect.Version, OS, IsLittleEndian);
+ writeInteger((uint32_t)Sect.UnitOffset, OS, IsLittleEndian);
+ writeInteger((uint32_t)Sect.UnitSize, OS, IsLittleEndian);
+ for (auto Entry : Sect.Entries) {
+ writeInteger((uint32_t)Entry.DieOffset, OS, IsLittleEndian);
+ if (Sect.IsGNUStyle)
+ writeInteger((uint32_t)Entry.Descriptor, OS, IsLittleEndian);
+ OS.write(Entry.Name.data(), Entry.Name.size());
+ OS.write('\0');
+ }
+}
+
+namespace {
+/// An extension of the DWARFYAML::ConstVisitor which writes compile
+/// units and DIEs to a stream.
+class DumpVisitor : public DWARFYAML::ConstVisitor {
+ raw_ostream &OS;
+
+protected:
+ void onStartCompileUnit(const DWARFYAML::Unit &CU) override {
+ writeInitialLength(CU.Length, OS, DebugInfo.IsLittleEndian);
+ writeInteger((uint16_t)CU.Version, OS, DebugInfo.IsLittleEndian);
+ if(CU.Version >= 5) {
+ writeInteger((uint8_t)CU.Type, OS, DebugInfo.IsLittleEndian);
+ writeInteger((uint8_t)CU.AddrSize, OS, DebugInfo.IsLittleEndian);
+ writeInteger((uint32_t)CU.AbbrOffset, OS, DebugInfo.IsLittleEndian);
+ }else {
+ writeInteger((uint32_t)CU.AbbrOffset, OS, DebugInfo.IsLittleEndian);
+ writeInteger((uint8_t)CU.AddrSize, OS, DebugInfo.IsLittleEndian);
+ }
+ }
+
+ void onStartDIE(const DWARFYAML::Unit &CU,
+ const DWARFYAML::Entry &DIE) override {
+ encodeULEB128(DIE.AbbrCode, OS);
+ }
+
+ void onValue(const uint8_t U) override {
+ writeInteger(U, OS, DebugInfo.IsLittleEndian);
+ }
+
+ void onValue(const uint16_t U) override {
+ writeInteger(U, OS, DebugInfo.IsLittleEndian);
+ }
+
+ void onValue(const uint32_t U) override {
+ writeInteger(U, OS, DebugInfo.IsLittleEndian);
+ }
+
+ void onValue(const uint64_t U, const bool LEB = false) override {
+ if (LEB)
+ encodeULEB128(U, OS);
+ else
+ writeInteger(U, OS, DebugInfo.IsLittleEndian);
+ }
+
+ void onValue(const int64_t S, const bool LEB = false) override {
+ if (LEB)
+ encodeSLEB128(S, OS);
+ else
+ writeInteger(S, OS, DebugInfo.IsLittleEndian);
+ }
+
+ void onValue(const StringRef String) override {
+ OS.write(String.data(), String.size());
+ OS.write('\0');
+ }
+
+ void onValue(const MemoryBufferRef MBR) override {
+ OS.write(MBR.getBufferStart(), MBR.getBufferSize());
+ }
+
+public:
+ DumpVisitor(const DWARFYAML::Data &DI, raw_ostream &Out)
+ : DWARFYAML::ConstVisitor(DI), OS(Out) {}
+};
+} // namespace
+
+void DWARFYAML::EmitDebugInfo(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ DumpVisitor Visitor(DI, OS);
+ Visitor.traverseDebugInfo();
+}
+
+static void EmitFileEntry(raw_ostream &OS, const DWARFYAML::File &File) {
+ OS.write(File.Name.data(), File.Name.size());
+ OS.write('\0');
+ encodeULEB128(File.DirIdx, OS);
+ encodeULEB128(File.ModTime, OS);
+ encodeULEB128(File.Length, OS);
+}
+
+void DWARFYAML::EmitDebugLine(raw_ostream &OS, const DWARFYAML::Data &DI) {
+ for (const auto &LineTable : DI.DebugLines) {
+ writeInitialLength(LineTable.Length, OS, DI.IsLittleEndian);
+ uint64_t SizeOfPrologueLength = LineTable.Length.isDWARF64() ? 8 : 4;
+ writeInteger((uint16_t)LineTable.Version, OS, DI.IsLittleEndian);
+ writeVariableSizedInteger(LineTable.PrologueLength, SizeOfPrologueLength,
+ OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)LineTable.MinInstLength, OS, DI.IsLittleEndian);
+ if (LineTable.Version >= 4)
+ writeInteger((uint8_t)LineTable.MaxOpsPerInst, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)LineTable.DefaultIsStmt, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)LineTable.LineBase, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)LineTable.LineRange, OS, DI.IsLittleEndian);
+ writeInteger((uint8_t)LineTable.OpcodeBase, OS, DI.IsLittleEndian);
+
+ for (auto OpcodeLength : LineTable.StandardOpcodeLengths)
+ writeInteger((uint8_t)OpcodeLength, OS, DI.IsLittleEndian);
+
+ for (auto IncludeDir : LineTable.IncludeDirs) {
+ OS.write(IncludeDir.data(), IncludeDir.size());
+ OS.write('\0');
+ }
+ OS.write('\0');
+
+ for (auto File : LineTable.Files)
+ EmitFileEntry(OS, File);
+ OS.write('\0');
+
+ for (auto Op : LineTable.Opcodes) {
+ writeInteger((uint8_t)Op.Opcode, OS, DI.IsLittleEndian);
+ if (Op.Opcode == 0) {
+ encodeULEB128(Op.ExtLen, OS);
+ writeInteger((uint8_t)Op.SubOpcode, OS, DI.IsLittleEndian);
+ switch (Op.SubOpcode) {
+ case dwarf::DW_LNE_set_address:
+ case dwarf::DW_LNE_set_discriminator:
+ writeVariableSizedInteger(Op.Data, DI.CompileUnits[0].AddrSize, OS,
+ DI.IsLittleEndian);
+ break;
+ case dwarf::DW_LNE_define_file:
+ EmitFileEntry(OS, Op.FileEntry);
+ break;
+ case dwarf::DW_LNE_end_sequence:
+ break;
+ default:
+ for (auto OpByte : Op.UnknownOpcodeData)
+ writeInteger((uint8_t)OpByte, OS, DI.IsLittleEndian);
+ }
+ } else if (Op.Opcode < LineTable.OpcodeBase) {
+ switch (Op.Opcode) {
+ case dwarf::DW_LNS_copy:
+ case dwarf::DW_LNS_negate_stmt:
+ case dwarf::DW_LNS_set_basic_block:
+ case dwarf::DW_LNS_const_add_pc:
+ case dwarf::DW_LNS_set_prologue_end:
+ case dwarf::DW_LNS_set_epilogue_begin:
+ break;
+
+ case dwarf::DW_LNS_advance_pc:
+ case dwarf::DW_LNS_set_file:
+ case dwarf::DW_LNS_set_column:
+ case dwarf::DW_LNS_set_isa:
+ encodeULEB128(Op.Data, OS);
+ break;
+
+ case dwarf::DW_LNS_advance_line:
+ encodeSLEB128(Op.SData, OS);
+ break;
+
+ case dwarf::DW_LNS_fixed_advance_pc:
+ writeInteger((uint16_t)Op.Data, OS, DI.IsLittleEndian);
+ break;
+
+ default:
+ for (auto OpData : Op.StandardOpcodeData) {
+ encodeULEB128(OpData, OS);
+ }
+ }
+ }
+ }
+ }
+}
+
+using EmitFuncType = void (*)(raw_ostream &, const DWARFYAML::Data &);
+
+static void
+EmitDebugSectionImpl(const DWARFYAML::Data &DI, EmitFuncType EmitFunc,
+ StringRef Sec,
+ StringMap<std::unique_ptr<MemoryBuffer>> &OutputBuffers) {
+ std::string Data;
+ raw_string_ostream DebugInfoStream(Data);
+ EmitFunc(DebugInfoStream, DI);
+ DebugInfoStream.flush();
+ if (!Data.empty())
+ OutputBuffers[Sec] = MemoryBuffer::getMemBufferCopy(Data);
+}
+
+namespace {
+class DIEFixupVisitor : public DWARFYAML::Visitor {
+ uint64_t Length;
+
+public:
+ DIEFixupVisitor(DWARFYAML::Data &DI) : DWARFYAML::Visitor(DI){};
+
+private:
+ virtual void onStartCompileUnit(DWARFYAML::Unit &CU) { Length = 7; }
+
+ virtual void onEndCompileUnit(DWARFYAML::Unit &CU) {
+ CU.Length.setLength(Length);
+ }
+
+ virtual void onStartDIE(DWARFYAML::Unit &CU, DWARFYAML::Entry &DIE) {
+ Length += getULEB128Size(DIE.AbbrCode);
+ }
+
+ virtual void onValue(const uint8_t U) { Length += 1; }
+ virtual void onValue(const uint16_t U) { Length += 2; }
+ virtual void onValue(const uint32_t U) { Length += 4; }
+ virtual void onValue(const uint64_t U, const bool LEB = false) {
+ if (LEB)
+ Length += getULEB128Size(U);
+ else
+ Length += 8;
+ }
+ virtual void onValue(const int64_t S, const bool LEB = false) {
+ if (LEB)
+ Length += getSLEB128Size(S);
+ else
+ Length += 8;
+ }
+ virtual void onValue(const StringRef String) { Length += String.size() + 1; }
+
+ virtual void onValue(const MemoryBufferRef MBR) {
+ Length += MBR.getBufferSize();
+ }
+};
+} // namespace
+
+Expected<StringMap<std::unique_ptr<MemoryBuffer>>>
+DWARFYAML::EmitDebugSections(StringRef YAMLString, bool ApplyFixups,
+ bool IsLittleEndian) {
+ yaml::Input YIn(YAMLString);
+
+ DWARFYAML::Data DI;
+ DI.IsLittleEndian = IsLittleEndian;
+ YIn >> DI;
+ if (YIn.error())
+ return errorCodeToError(YIn.error());
+
+ if (ApplyFixups) {
+ DIEFixupVisitor DIFixer(DI);
+ DIFixer.traverseDebugInfo();
+ }
+
+ StringMap<std::unique_ptr<MemoryBuffer>> DebugSections;
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugInfo, "debug_info",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugLine, "debug_line",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugStr, "debug_str",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugAbbrev, "debug_abbrev",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugAranges, "debug_aranges",
+ DebugSections);
+ return std::move(DebugSections);
+}
+
+// XXX BINARYEN <--
+namespace llvm {
+namespace DWARFYAML {
+StringMap<std::unique_ptr<MemoryBuffer>>
+EmitDebugSections(llvm::DWARFYAML::Data &DI, bool ApplyFixups) {
+ if (ApplyFixups) {
+ DIEFixupVisitor DIFixer(DI);
+ DIFixer.traverseDebugInfo();
+ }
+
+ StringMap<std::unique_ptr<MemoryBuffer>> DebugSections;
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugInfo, "debug_info",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugLine, "debug_line",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugStr, "debug_str",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugAbbrev, "debug_abbrev",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugAranges, "debug_aranges",
+ DebugSections);
+ EmitDebugSectionImpl(DI, &DWARFYAML::EmitDebugRanges, "debug_ranges",
+ DebugSections); // XXX BINARYEN
+ return std::move(DebugSections);
+}
+} // namespace DWARFYAML
+} // namespace llvm
+// XXX BINARYEN -->
diff --git a/third_party/llvm-project/DWARFExpression.cpp b/third_party/llvm-project/DWARFExpression.cpp
new file mode 100644
index 000000000..5009b1b7b
--- /dev/null
+++ b/third_party/llvm-project/DWARFExpression.cpp
@@ -0,0 +1,344 @@
+//===-- DWARFExpression.cpp -----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/Support/Format.h"
+#include <cassert>
+#include <cstdint>
+#include <vector>
+
+using namespace llvm;
+using namespace dwarf;
+
+namespace llvm {
+
+typedef std::vector<DWARFExpression::Operation::Description> DescVector;
+
+static DescVector getDescriptions() {
+ DescVector Descriptions;
+ typedef DWARFExpression::Operation Op;
+ typedef Op::Description Desc;
+
+ Descriptions.resize(0xff);
+ Descriptions[DW_OP_addr] = Desc(Op::Dwarf2, Op::SizeAddr);
+ Descriptions[DW_OP_deref] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_const1u] = Desc(Op::Dwarf2, Op::Size1);
+ Descriptions[DW_OP_const1s] = Desc(Op::Dwarf2, Op::SignedSize1);
+ Descriptions[DW_OP_const2u] = Desc(Op::Dwarf2, Op::Size2);
+ Descriptions[DW_OP_const2s] = Desc(Op::Dwarf2, Op::SignedSize2);
+ Descriptions[DW_OP_const4u] = Desc(Op::Dwarf2, Op::Size4);
+ Descriptions[DW_OP_const4s] = Desc(Op::Dwarf2, Op::SignedSize4);
+ Descriptions[DW_OP_const8u] = Desc(Op::Dwarf2, Op::Size8);
+ Descriptions[DW_OP_const8s] = Desc(Op::Dwarf2, Op::SignedSize8);
+ Descriptions[DW_OP_constu] = Desc(Op::Dwarf2, Op::SizeLEB);
+ Descriptions[DW_OP_consts] = Desc(Op::Dwarf2, Op::SignedSizeLEB);
+ Descriptions[DW_OP_dup] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_drop] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_over] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_pick] = Desc(Op::Dwarf2, Op::Size1);
+ Descriptions[DW_OP_swap] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_rot] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_xderef] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_abs] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_and] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_div] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_minus] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_mod] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_mul] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_neg] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_not] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_or] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_plus] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_plus_uconst] = Desc(Op::Dwarf2, Op::SizeLEB);
+ Descriptions[DW_OP_shl] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_shr] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_shra] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_xor] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_skip] = Desc(Op::Dwarf2, Op::SignedSize2);
+ Descriptions[DW_OP_bra] = Desc(Op::Dwarf2, Op::SignedSize2);
+ Descriptions[DW_OP_eq] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_ge] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_gt] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_le] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_lt] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_ne] = Desc(Op::Dwarf2);
+ for (uint16_t LA = DW_OP_lit0; LA <= DW_OP_lit31; ++LA)
+ Descriptions[LA] = Desc(Op::Dwarf2);
+ for (uint16_t LA = DW_OP_reg0; LA <= DW_OP_reg31; ++LA)
+ Descriptions[LA] = Desc(Op::Dwarf2);
+ for (uint16_t LA = DW_OP_breg0; LA <= DW_OP_breg31; ++LA)
+ Descriptions[LA] = Desc(Op::Dwarf2, Op::SignedSizeLEB);
+ Descriptions[DW_OP_regx] = Desc(Op::Dwarf2, Op::SizeLEB);
+ Descriptions[DW_OP_fbreg] = Desc(Op::Dwarf2, Op::SignedSizeLEB);
+ Descriptions[DW_OP_bregx] = Desc(Op::Dwarf2, Op::SizeLEB, Op::SignedSizeLEB);
+ Descriptions[DW_OP_piece] = Desc(Op::Dwarf2, Op::SizeLEB);
+ Descriptions[DW_OP_deref_size] = Desc(Op::Dwarf2, Op::Size1);
+ Descriptions[DW_OP_xderef_size] = Desc(Op::Dwarf2, Op::Size1);
+ Descriptions[DW_OP_nop] = Desc(Op::Dwarf2);
+ Descriptions[DW_OP_push_object_address] = Desc(Op::Dwarf3);
+ Descriptions[DW_OP_call2] = Desc(Op::Dwarf3, Op::Size2);
+ Descriptions[DW_OP_call4] = Desc(Op::Dwarf3, Op::Size4);
+ Descriptions[DW_OP_call_ref] = Desc(Op::Dwarf3, Op::SizeRefAddr);
+ Descriptions[DW_OP_form_tls_address] = Desc(Op::Dwarf3);
+ Descriptions[DW_OP_call_frame_cfa] = Desc(Op::Dwarf3);
+ Descriptions[DW_OP_bit_piece] = Desc(Op::Dwarf3, Op::SizeLEB, Op::SizeLEB);
+ Descriptions[DW_OP_implicit_value] =
+ Desc(Op::Dwarf3, Op::SizeLEB, Op::SizeBlock);
+ Descriptions[DW_OP_stack_value] = Desc(Op::Dwarf3);
+ Descriptions[DW_OP_GNU_push_tls_address] = Desc(Op::Dwarf3);
+ Descriptions[DW_OP_addrx] = Desc(Op::Dwarf4, Op::SizeLEB);
+ Descriptions[DW_OP_GNU_addr_index] = Desc(Op::Dwarf4, Op::SizeLEB);
+ Descriptions[DW_OP_GNU_const_index] = Desc(Op::Dwarf4, Op::SizeLEB);
+ Descriptions[DW_OP_GNU_entry_value] = Desc(Op::Dwarf4, Op::SizeLEB);
+
+ Descriptions[DW_OP_convert] = Desc(Op::Dwarf5, Op::BaseTypeRef);
+ Descriptions[DW_OP_entry_value] = Desc(Op::Dwarf5, Op::SizeLEB);
+
+ return Descriptions;
+}
+
+static DWARFExpression::Operation::Description getOpDesc(unsigned OpCode) {
+ // FIXME: Make this constexpr once all compilers are smart enough to do it.
+ static DescVector Descriptions = getDescriptions();
+ // Handle possible corrupted or unsupported operation.
+ if (OpCode >= Descriptions.size())
+ return {};
+ return Descriptions[OpCode];
+}
+
+static uint8_t getRefAddrSize(uint8_t AddrSize, uint16_t Version) {
+ return (Version == 2) ? AddrSize : 4;
+}
+
+bool DWARFExpression::Operation::extract(DataExtractor Data, uint16_t Version,
+ uint8_t AddressSize, uint64_t Offset) {
+ Opcode = Data.getU8(&Offset);
+
+ Desc = getOpDesc(Opcode);
+ if (Desc.Version == Operation::DwarfNA) {
+ EndOffset = Offset;
+ return false;
+ }
+
+ for (unsigned Operand = 0; Operand < 2; ++Operand) {
+ unsigned Size = Desc.Op[Operand];
+ unsigned Signed = Size & Operation::SignBit;
+
+ if (Size == Operation::SizeNA)
+ break;
+
+ switch (Size & ~Operation::SignBit) {
+ case Operation::Size1:
+ Operands[Operand] = Data.getU8(&Offset);
+ if (Signed)
+ Operands[Operand] = (int8_t)Operands[Operand];
+ break;
+ case Operation::Size2:
+ Operands[Operand] = Data.getU16(&Offset);
+ if (Signed)
+ Operands[Operand] = (int16_t)Operands[Operand];
+ break;
+ case Operation::Size4:
+ Operands[Operand] = Data.getU32(&Offset);
+ if (Signed)
+ Operands[Operand] = (int32_t)Operands[Operand];
+ break;
+ case Operation::Size8:
+ Operands[Operand] = Data.getU64(&Offset);
+ break;
+ case Operation::SizeAddr:
+ if (AddressSize == 8) {
+ Operands[Operand] = Data.getU64(&Offset);
+ } else if (AddressSize == 4) {
+ Operands[Operand] = Data.getU32(&Offset);
+ } else {
+ assert(AddressSize == 2);
+ Operands[Operand] = Data.getU16(&Offset);
+ }
+ break;
+ case Operation::SizeRefAddr:
+ if (getRefAddrSize(AddressSize, Version) == 8) {
+ Operands[Operand] = Data.getU64(&Offset);
+ } else if (getRefAddrSize(AddressSize, Version) == 4) {
+ Operands[Operand] = Data.getU32(&Offset);
+ } else {
+ assert(getRefAddrSize(AddressSize, Version) == 2);
+ Operands[Operand] = Data.getU16(&Offset);
+ }
+ break;
+ case Operation::SizeLEB:
+ if (Signed)
+ Operands[Operand] = Data.getSLEB128(&Offset);
+ else
+ Operands[Operand] = Data.getULEB128(&Offset);
+ break;
+ case Operation::BaseTypeRef:
+ Operands[Operand] = Data.getULEB128(&Offset);
+ break;
+ case Operation::SizeBlock:
+ // We need a size, so this cannot be the first operand
+ if (Operand == 0)
+ return false;
+ // Store the offset of the block as the value.
+ Operands[Operand] = Offset;
+ Offset += Operands[Operand - 1];
+ break;
+ default:
+ llvm_unreachable("Unknown DWARFExpression Op size");
+ }
+
+ OperandEndOffsets[Operand] = Offset;
+ }
+
+ EndOffset = Offset;
+ return true;
+}
+
+static bool prettyPrintRegisterOp(raw_ostream &OS, uint8_t Opcode,
+ uint64_t Operands[2],
+ const MCRegisterInfo *MRI, bool isEH) {
+ if (!MRI)
+ return false;
+
+ uint64_t DwarfRegNum;
+ unsigned OpNum = 0;
+
+ if (Opcode == DW_OP_bregx || Opcode == DW_OP_regx)
+ DwarfRegNum = Operands[OpNum++];
+ else if (Opcode >= DW_OP_breg0 && Opcode < DW_OP_bregx)
+ DwarfRegNum = Opcode - DW_OP_breg0;
+ else
+ DwarfRegNum = Opcode - DW_OP_reg0;
+
+ if (Optional<unsigned> LLVMRegNum = MRI->getLLVMRegNum(DwarfRegNum, isEH)) {
+ if (const char *RegName = MRI->getName(*LLVMRegNum)) {
+ if ((Opcode >= DW_OP_breg0 && Opcode <= DW_OP_breg31) ||
+ Opcode == DW_OP_bregx)
+ OS << format(" %s%+" PRId64, RegName, Operands[OpNum]);
+ else
+ OS << ' ' << RegName;
+ return true;
+ }
+ }
+
+ return false;
+}
+
+bool DWARFExpression::Operation::print(raw_ostream &OS,
+ const DWARFExpression *Expr,
+ const MCRegisterInfo *RegInfo,
+ DWARFUnit *U,
+ bool isEH) {
+ if (Error) {
+ OS << "<decoding error>";
+ return false;
+ }
+
+ StringRef Name = OperationEncodingString(Opcode);
+ assert(!Name.empty() && "DW_OP has no name!");
+ OS << Name;
+
+ if ((Opcode >= DW_OP_breg0 && Opcode <= DW_OP_breg31) ||
+ (Opcode >= DW_OP_reg0 && Opcode <= DW_OP_reg31) ||
+ Opcode == DW_OP_bregx || Opcode == DW_OP_regx)
+ if (prettyPrintRegisterOp(OS, Opcode, Operands, RegInfo, isEH))
+ return true;
+
+ for (unsigned Operand = 0; Operand < 2; ++Operand) {
+ unsigned Size = Desc.Op[Operand];
+ unsigned Signed = Size & Operation::SignBit;
+
+ if (Size == Operation::SizeNA)
+ break;
+
+ if (Size == Operation::BaseTypeRef && U) {
+ auto Die = U->getDIEForOffset(U->getOffset() + Operands[Operand]);
+ if (Die && Die.getTag() == dwarf::DW_TAG_base_type) {
+ OS << format(" (0x%08" PRIx64 ")", U->getOffset() + Operands[Operand]);
+ if (auto Name = Die.find(dwarf::DW_AT_name))
+ OS << " \"" << Name->getAsCString() << "\"";
+ } else {
+ OS << format(" <invalid base_type ref: 0x%" PRIx64 ">",
+ Operands[Operand]);
+ }
+ } else if (Size == Operation::SizeBlock) {
+ uint64_t Offset = Operands[Operand];
+ for (unsigned i = 0; i < Operands[Operand - 1]; ++i)
+ OS << format(" 0x%02x", Expr->Data.getU8(&Offset));
+ } else {
+ if (Signed)
+ OS << format(" %+" PRId64, (int64_t)Operands[Operand]);
+ else if (Opcode != DW_OP_entry_value &&
+ Opcode != DW_OP_GNU_entry_value)
+ OS << format(" 0x%" PRIx64, Operands[Operand]);
+ }
+ }
+ return true;
+}
+
+void DWARFExpression::print(raw_ostream &OS, const MCRegisterInfo *RegInfo,
+ DWARFUnit *U, bool IsEH) const {
+ uint32_t EntryValExprSize = 0;
+ for (auto &Op : *this) {
+ if (!Op.print(OS, this, RegInfo, U, IsEH)) {
+ uint64_t FailOffset = Op.getEndOffset();
+ while (FailOffset < Data.getData().size())
+ OS << format(" %02x", Data.getU8(&FailOffset));
+ return;
+ }
+
+ if (Op.getCode() == DW_OP_entry_value ||
+ Op.getCode() == DW_OP_GNU_entry_value) {
+ OS << "(";
+ EntryValExprSize = Op.getRawOperand(0);
+ continue;
+ }
+
+ if (EntryValExprSize) {
+ EntryValExprSize--;
+ if (EntryValExprSize == 0)
+ OS << ")";
+ }
+
+ if (Op.getEndOffset() < Data.getData().size())
+ OS << ", ";
+ }
+}
+
+bool DWARFExpression::Operation::verify(DWARFUnit *U) {
+
+ for (unsigned Operand = 0; Operand < 2; ++Operand) {
+ unsigned Size = Desc.Op[Operand];
+
+ if (Size == Operation::SizeNA)
+ break;
+
+ if (Size == Operation::BaseTypeRef) {
+ auto Die = U->getDIEForOffset(U->getOffset() + Operands[Operand]);
+ if (!Die || Die.getTag() != dwarf::DW_TAG_base_type) {
+ Error = true;
+ return false;
+ }
+ }
+ }
+
+ return true;
+}
+
+bool DWARFExpression::verify(DWARFUnit *U) {
+ for (auto &Op : *this)
+ if (!Op.verify(U))
+ return false;
+
+ return true;
+}
+
+} // namespace llvm
diff --git a/third_party/llvm-project/DWARFFormValue.cpp b/third_party/llvm-project/DWARFFormValue.cpp
new file mode 100644
index 000000000..26090638b
--- /dev/null
+++ b/third_party/llvm-project/DWARFFormValue.cpp
@@ -0,0 +1,720 @@
+//===- DWARFFormValue.cpp -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cinttypes>
+#include <cstdint>
+#include <limits>
+
+using namespace llvm;
+using namespace dwarf;
+
+static const DWARFFormValue::FormClass DWARF5FormClasses[] = {
+ DWARFFormValue::FC_Unknown, // 0x0
+ DWARFFormValue::FC_Address, // 0x01 DW_FORM_addr
+ DWARFFormValue::FC_Unknown, // 0x02 unused
+ DWARFFormValue::FC_Block, // 0x03 DW_FORM_block2
+ DWARFFormValue::FC_Block, // 0x04 DW_FORM_block4
+ DWARFFormValue::FC_Constant, // 0x05 DW_FORM_data2
+ // --- These can be FC_SectionOffset in DWARF3 and below:
+ DWARFFormValue::FC_Constant, // 0x06 DW_FORM_data4
+ DWARFFormValue::FC_Constant, // 0x07 DW_FORM_data8
+ // ---
+ DWARFFormValue::FC_String, // 0x08 DW_FORM_string
+ DWARFFormValue::FC_Block, // 0x09 DW_FORM_block
+ DWARFFormValue::FC_Block, // 0x0a DW_FORM_block1
+ DWARFFormValue::FC_Constant, // 0x0b DW_FORM_data1
+ DWARFFormValue::FC_Flag, // 0x0c DW_FORM_flag
+ DWARFFormValue::FC_Constant, // 0x0d DW_FORM_sdata
+ DWARFFormValue::FC_String, // 0x0e DW_FORM_strp
+ DWARFFormValue::FC_Constant, // 0x0f DW_FORM_udata
+ DWARFFormValue::FC_Reference, // 0x10 DW_FORM_ref_addr
+ DWARFFormValue::FC_Reference, // 0x11 DW_FORM_ref1
+ DWARFFormValue::FC_Reference, // 0x12 DW_FORM_ref2
+ DWARFFormValue::FC_Reference, // 0x13 DW_FORM_ref4
+ DWARFFormValue::FC_Reference, // 0x14 DW_FORM_ref8
+ DWARFFormValue::FC_Reference, // 0x15 DW_FORM_ref_udata
+ DWARFFormValue::FC_Indirect, // 0x16 DW_FORM_indirect
+ DWARFFormValue::FC_SectionOffset, // 0x17 DW_FORM_sec_offset
+ DWARFFormValue::FC_Exprloc, // 0x18 DW_FORM_exprloc
+ DWARFFormValue::FC_Flag, // 0x19 DW_FORM_flag_present
+ DWARFFormValue::FC_String, // 0x1a DW_FORM_strx
+ DWARFFormValue::FC_Address, // 0x1b DW_FORM_addrx
+ DWARFFormValue::FC_Reference, // 0x1c DW_FORM_ref_sup4
+ DWARFFormValue::FC_String, // 0x1d DW_FORM_strp_sup
+ DWARFFormValue::FC_Constant, // 0x1e DW_FORM_data16
+ DWARFFormValue::FC_String, // 0x1f DW_FORM_line_strp
+ DWARFFormValue::FC_Reference, // 0x20 DW_FORM_ref_sig8
+ DWARFFormValue::FC_Constant, // 0x21 DW_FORM_implicit_const
+ DWARFFormValue::FC_SectionOffset, // 0x22 DW_FORM_loclistx
+ DWARFFormValue::FC_SectionOffset, // 0x23 DW_FORM_rnglistx
+ DWARFFormValue::FC_Reference, // 0x24 DW_FORM_ref_sup8
+ DWARFFormValue::FC_String, // 0x25 DW_FORM_strx1
+ DWARFFormValue::FC_String, // 0x26 DW_FORM_strx2
+ DWARFFormValue::FC_String, // 0x27 DW_FORM_strx3
+ DWARFFormValue::FC_String, // 0x28 DW_FORM_strx4
+ DWARFFormValue::FC_Address, // 0x29 DW_FORM_addrx1
+ DWARFFormValue::FC_Address, // 0x2a DW_FORM_addrx2
+ DWARFFormValue::FC_Address, // 0x2b DW_FORM_addrx3
+ DWARFFormValue::FC_Address, // 0x2c DW_FORM_addrx4
+
+};
+
+DWARFFormValue DWARFFormValue::createFromSValue(dwarf::Form F, int64_t V) {
+ return DWARFFormValue(F, ValueType(V));
+}
+
+DWARFFormValue DWARFFormValue::createFromUValue(dwarf::Form F, uint64_t V) {
+ return DWARFFormValue(F, ValueType(V));
+}
+
+DWARFFormValue DWARFFormValue::createFromPValue(dwarf::Form F, const char *V) {
+ return DWARFFormValue(F, ValueType(V));
+}
+
+DWARFFormValue DWARFFormValue::createFromBlockValue(dwarf::Form F,
+ ArrayRef<uint8_t> D) {
+ ValueType V;
+ V.uval = D.size();
+ V.data = D.data();
+ return DWARFFormValue(F, V);
+}
+
+DWARFFormValue DWARFFormValue::createFromUnit(dwarf::Form F, const DWARFUnit *U,
+ uint64_t *OffsetPtr) {
+ DWARFFormValue FormValue(F);
+ FormValue.extractValue(U->getDebugInfoExtractor(), OffsetPtr,
+ U->getFormParams(), U);
+ return FormValue;
+}
+
+bool DWARFFormValue::skipValue(dwarf::Form Form, DataExtractor DebugInfoData,
+ uint64_t *OffsetPtr,
+ const dwarf::FormParams Params) {
+ bool Indirect = false;
+ do {
+ switch (Form) {
+ // Blocks of inlined data that have a length field and the data bytes
+ // inlined in the .debug_info.
+ case DW_FORM_exprloc:
+ case DW_FORM_block: {
+ uint64_t size = DebugInfoData.getULEB128(OffsetPtr);
+ *OffsetPtr += size;
+ return true;
+ }
+ case DW_FORM_block1: {
+ uint8_t size = DebugInfoData.getU8(OffsetPtr);
+ *OffsetPtr += size;
+ return true;
+ }
+ case DW_FORM_block2: {
+ uint16_t size = DebugInfoData.getU16(OffsetPtr);
+ *OffsetPtr += size;
+ return true;
+ }
+ case DW_FORM_block4: {
+ uint32_t size = DebugInfoData.getU32(OffsetPtr);
+ *OffsetPtr += size;
+ return true;
+ }
+
+ // Inlined NULL terminated C-strings.
+ case DW_FORM_string:
+ DebugInfoData.getCStr(OffsetPtr);
+ return true;
+
+ case DW_FORM_addr:
+ case DW_FORM_ref_addr:
+ case DW_FORM_flag_present:
+ case DW_FORM_data1:
+ case DW_FORM_data2:
+ case DW_FORM_data4:
+ case DW_FORM_data8:
+ case DW_FORM_data16:
+ case DW_FORM_flag:
+ case DW_FORM_ref1:
+ case DW_FORM_ref2:
+ case DW_FORM_ref4:
+ case DW_FORM_ref8:
+ case DW_FORM_ref_sig8:
+ case DW_FORM_ref_sup4:
+ case DW_FORM_ref_sup8:
+ case DW_FORM_strx1:
+ case DW_FORM_strx2:
+ case DW_FORM_strx4:
+ case DW_FORM_addrx1:
+ case DW_FORM_addrx2:
+ case DW_FORM_addrx4:
+ case DW_FORM_sec_offset:
+ case DW_FORM_strp:
+ case DW_FORM_strp_sup:
+ case DW_FORM_line_strp:
+ case DW_FORM_GNU_ref_alt:
+ case DW_FORM_GNU_strp_alt:
+ if (Optional<uint8_t> FixedSize =
+ dwarf::getFixedFormByteSize(Form, Params)) {
+ *OffsetPtr += *FixedSize;
+ return true;
+ }
+ return false;
+
+ // signed or unsigned LEB 128 values.
+ case DW_FORM_sdata:
+ DebugInfoData.getSLEB128(OffsetPtr);
+ return true;
+
+ case DW_FORM_udata:
+ case DW_FORM_ref_udata:
+ case DW_FORM_strx:
+ case DW_FORM_addrx:
+ case DW_FORM_loclistx:
+ case DW_FORM_rnglistx:
+ case DW_FORM_GNU_addr_index:
+ case DW_FORM_GNU_str_index:
+ DebugInfoData.getULEB128(OffsetPtr);
+ return true;
+
+ case DW_FORM_indirect:
+ Indirect = true;
+ Form = static_cast<dwarf::Form>(DebugInfoData.getULEB128(OffsetPtr));
+ break;
+
+ default:
+ return false;
+ }
+ } while (Indirect);
+ return true;
+}
+
+bool DWARFFormValue::isFormClass(DWARFFormValue::FormClass FC) const {
+ // First, check DWARF5 form classes.
+ if (Form < makeArrayRef(DWARF5FormClasses).size() &&
+ DWARF5FormClasses[Form] == FC)
+ return true;
+ // Check more forms from extensions and proposals.
+ switch (Form) {
+ case DW_FORM_GNU_ref_alt:
+ return (FC == FC_Reference);
+ case DW_FORM_GNU_addr_index:
+ return (FC == FC_Address);
+ case DW_FORM_GNU_str_index:
+ case DW_FORM_GNU_strp_alt:
+ return (FC == FC_String);
+ default:
+ break;
+ }
+
+ if (FC == FC_SectionOffset) {
+ if (Form == DW_FORM_strp || Form == DW_FORM_line_strp)
+ return true;
+ // In DWARF3 DW_FORM_data4 and DW_FORM_data8 served also as a section
+ // offset. If we don't have a DWARFUnit, default to the old behavior.
+ if (Form == DW_FORM_data4 || Form == DW_FORM_data8)
+ return !U || U->getVersion() <= 3;
+ }
+
+ return false;
+}
+
+bool DWARFFormValue::extractValue(const DWARFDataExtractor &Data,
+ uint64_t *OffsetPtr, dwarf::FormParams FP,
+ const DWARFContext *Ctx,
+ const DWARFUnit *CU) {
+ if (!Ctx && CU)
+ Ctx = &CU->getContext();
+ C = Ctx;
+ U = CU;
+ bool Indirect = false;
+ bool IsBlock = false;
+ Value.data = nullptr;
+ // Read the value for the form into value and follow and DW_FORM_indirect
+ // instances we run into
+ do {
+ Indirect = false;
+ switch (Form) {
+ case DW_FORM_addr:
+ case DW_FORM_ref_addr: {
+ uint16_t Size =
+ (Form == DW_FORM_addr) ? FP.AddrSize : FP.getRefAddrByteSize();
+ Value.uval = Data.getRelocatedValue(Size, OffsetPtr, &Value.SectionIndex);
+ break;
+ }
+ case DW_FORM_exprloc:
+ case DW_FORM_block:
+ Value.uval = Data.getULEB128(OffsetPtr);
+ IsBlock = true;
+ break;
+ case DW_FORM_block1:
+ Value.uval = Data.getU8(OffsetPtr);
+ IsBlock = true;
+ break;
+ case DW_FORM_block2:
+ Value.uval = Data.getU16(OffsetPtr);
+ IsBlock = true;
+ break;
+ case DW_FORM_block4:
+ Value.uval = Data.getU32(OffsetPtr);
+ IsBlock = true;
+ break;
+ case DW_FORM_data1:
+ case DW_FORM_ref1:
+ case DW_FORM_flag:
+ case DW_FORM_strx1:
+ case DW_FORM_addrx1:
+ Value.uval = Data.getU8(OffsetPtr);
+ break;
+ case DW_FORM_data2:
+ case DW_FORM_ref2:
+ case DW_FORM_strx2:
+ case DW_FORM_addrx2:
+ Value.uval = Data.getU16(OffsetPtr);
+ break;
+ case DW_FORM_strx3:
+ Value.uval = Data.getU24(OffsetPtr);
+ break;
+ case DW_FORM_data4:
+ case DW_FORM_ref4:
+ case DW_FORM_ref_sup4:
+ case DW_FORM_strx4:
+ case DW_FORM_addrx4:
+ Value.uval = Data.getRelocatedValue(4, OffsetPtr);
+ break;
+ case DW_FORM_data8:
+ case DW_FORM_ref8:
+ case DW_FORM_ref_sup8:
+ Value.uval = Data.getRelocatedValue(8, OffsetPtr);
+ break;
+ case DW_FORM_data16:
+ // Treat this like a 16-byte block.
+ Value.uval = 16;
+ IsBlock = true;
+ break;
+ case DW_FORM_sdata:
+ Value.sval = Data.getSLEB128(OffsetPtr);
+ break;
+ case DW_FORM_udata:
+ case DW_FORM_ref_udata:
+ case DW_FORM_rnglistx:
+ Value.uval = Data.getULEB128(OffsetPtr);
+ break;
+ case DW_FORM_string:
+ Value.cstr = Data.getCStr(OffsetPtr);
+ break;
+ case DW_FORM_indirect:
+ Form = static_cast<dwarf::Form>(Data.getULEB128(OffsetPtr));
+ Indirect = true;
+ break;
+ case DW_FORM_strp:
+ case DW_FORM_sec_offset:
+ case DW_FORM_GNU_ref_alt:
+ case DW_FORM_GNU_strp_alt:
+ case DW_FORM_line_strp:
+ case DW_FORM_strp_sup: {
+ Value.uval =
+ Data.getRelocatedValue(FP.getDwarfOffsetByteSize(), OffsetPtr);
+ break;
+ }
+ case DW_FORM_flag_present:
+ Value.uval = 1;
+ break;
+ case DW_FORM_ref_sig8:
+ Value.uval = Data.getU64(OffsetPtr);
+ break;
+ case DW_FORM_GNU_addr_index:
+ case DW_FORM_GNU_str_index:
+ case DW_FORM_addrx:
+ case DW_FORM_strx:
+ Value.uval = Data.getULEB128(OffsetPtr);
+ break;
+ default:
+ // DWARFFormValue::skipValue() will have caught this and caused all
+ // DWARF DIEs to fail to be parsed, so this code is not be reachable.
+ llvm_unreachable("unsupported form");
+ }
+ } while (Indirect);
+
+ if (IsBlock) {
+ StringRef Str = Data.getData().substr(*OffsetPtr, Value.uval);
+ Value.data = nullptr;
+ if (!Str.empty()) {
+ Value.data = Str.bytes_begin();
+ *OffsetPtr += Value.uval;
+ }
+ }
+
+ return true;
+}
+
+void DWARFFormValue::dumpSectionedAddress(raw_ostream &OS,
+ DIDumpOptions DumpOpts,
+ object::SectionedAddress SA) const {
+ OS << format("0x%016" PRIx64, SA.Address);
+ dumpAddressSection(U->getContext().getDWARFObj(), OS, DumpOpts,
+ SA.SectionIndex);
+}
+
+void DWARFFormValue::dumpAddressSection(const DWARFObject &Obj, raw_ostream &OS,
+ DIDumpOptions DumpOpts,
+ uint64_t SectionIndex) {
+ if (!DumpOpts.Verbose || SectionIndex == -1ULL)
+ return;
+ ArrayRef<SectionName> SectionNames = Obj.getSectionNames();
+ const auto &SecRef = SectionNames[SectionIndex];
+
+ OS << " \"" << SecRef.Name << '\"';
+
+ // Print section index if name is not unique.
+ if (!SecRef.IsNameUnique)
+ OS << format(" [%" PRIu64 "]", SectionIndex);
+}
+
+void DWARFFormValue::dump(raw_ostream &OS, DIDumpOptions DumpOpts) const {
+ uint64_t UValue = Value.uval;
+ bool CURelativeOffset = false;
+ raw_ostream &AddrOS = DumpOpts.ShowAddresses
+ ? WithColor(OS, HighlightColor::Address).get()
+ : nulls();
+ switch (Form) {
+ case DW_FORM_addr:
+ dumpSectionedAddress(AddrOS, DumpOpts, {Value.uval, Value.SectionIndex});
+ break;
+ case DW_FORM_addrx:
+ case DW_FORM_addrx1:
+ case DW_FORM_addrx2:
+ case DW_FORM_addrx3:
+ case DW_FORM_addrx4:
+ case DW_FORM_GNU_addr_index: {
+ if (U == nullptr) {
+ OS << "<invalid dwarf unit>";
+ break;
+ }
+ Optional<object::SectionedAddress> A = U->getAddrOffsetSectionItem(UValue);
+ if (!A || DumpOpts.Verbose)
+ AddrOS << format("indexed (%8.8x) address = ", (uint32_t)UValue);
+ if (A)
+ dumpSectionedAddress(AddrOS, DumpOpts, *A);
+ else
+ OS << "<no .debug_addr section>";
+ break;
+ }
+ case DW_FORM_flag_present:
+ OS << "true";
+ break;
+ case DW_FORM_flag:
+ case DW_FORM_data1:
+ OS << format("0x%02x", (uint8_t)UValue);
+ break;
+ case DW_FORM_data2:
+ OS << format("0x%04x", (uint16_t)UValue);
+ break;
+ case DW_FORM_data4:
+ OS << format("0x%08x", (uint32_t)UValue);
+ break;
+ case DW_FORM_ref_sig8:
+ AddrOS << format("0x%016" PRIx64, UValue);
+ break;
+ case DW_FORM_data8:
+ OS << format("0x%016" PRIx64, UValue);
+ break;
+ case DW_FORM_data16:
+ OS << format_bytes(ArrayRef<uint8_t>(Value.data, 16), None, 16, 16);
+ break;
+ case DW_FORM_string:
+ OS << '"';
+ OS.write_escaped(Value.cstr);
+ OS << '"';
+ break;
+ case DW_FORM_exprloc:
+ case DW_FORM_block:
+ case DW_FORM_block1:
+ case DW_FORM_block2:
+ case DW_FORM_block4:
+ if (UValue > 0) {
+ switch (Form) {
+ case DW_FORM_exprloc:
+ case DW_FORM_block:
+ AddrOS << format("<0x%" PRIx64 "> ", UValue);
+ break;
+ case DW_FORM_block1:
+ AddrOS << format("<0x%2.2x> ", (uint8_t)UValue);
+ break;
+ case DW_FORM_block2:
+ AddrOS << format("<0x%4.4x> ", (uint16_t)UValue);
+ break;
+ case DW_FORM_block4:
+ AddrOS << format("<0x%8.8x> ", (uint32_t)UValue);
+ break;
+ default:
+ break;
+ }
+
+ const uint8_t *DataPtr = Value.data;
+ if (DataPtr) {
+ // UValue contains size of block
+ const uint8_t *EndDataPtr = DataPtr + UValue;
+ while (DataPtr < EndDataPtr) {
+ AddrOS << format("%2.2x ", *DataPtr);
+ ++DataPtr;
+ }
+ } else
+ OS << "NULL";
+ }
+ break;
+
+ case DW_FORM_sdata:
+ OS << Value.sval;
+ break;
+ case DW_FORM_udata:
+ OS << Value.uval;
+ break;
+ case DW_FORM_strp:
+ if (DumpOpts.Verbose)
+ OS << format(" .debug_str[0x%8.8x] = ", (uint32_t)UValue);
+ dumpString(OS);
+ break;
+ case DW_FORM_line_strp:
+ if (DumpOpts.Verbose)
+ OS << format(" .debug_line_str[0x%8.8x] = ", (uint32_t)UValue);
+ dumpString(OS);
+ break;
+ case DW_FORM_strx:
+ case DW_FORM_strx1:
+ case DW_FORM_strx2:
+ case DW_FORM_strx3:
+ case DW_FORM_strx4:
+ case DW_FORM_GNU_str_index:
+ if (DumpOpts.Verbose)
+ OS << format("indexed (%8.8x) string = ", (uint32_t)UValue);
+ dumpString(OS);
+ break;
+ case DW_FORM_GNU_strp_alt:
+ if (DumpOpts.Verbose)
+ OS << format("alt indirect string, offset: 0x%" PRIx64 "", UValue);
+ dumpString(OS);
+ break;
+ case DW_FORM_ref_addr:
+ AddrOS << format("0x%016" PRIx64, UValue);
+ break;
+ case DW_FORM_ref1:
+ CURelativeOffset = true;
+ if (DumpOpts.Verbose)
+ AddrOS << format("cu + 0x%2.2x", (uint8_t)UValue);
+ break;
+ case DW_FORM_ref2:
+ CURelativeOffset = true;
+ if (DumpOpts.Verbose)
+ AddrOS << format("cu + 0x%4.4x", (uint16_t)UValue);
+ break;
+ case DW_FORM_ref4:
+ CURelativeOffset = true;
+ if (DumpOpts.Verbose)
+ AddrOS << format("cu + 0x%4.4x", (uint32_t)UValue);
+ break;
+ case DW_FORM_ref8:
+ CURelativeOffset = true;
+ if (DumpOpts.Verbose)
+ AddrOS << format("cu + 0x%8.8" PRIx64, UValue);
+ break;
+ case DW_FORM_ref_udata:
+ CURelativeOffset = true;
+ if (DumpOpts.Verbose)
+ AddrOS << format("cu + 0x%" PRIx64, UValue);
+ break;
+ case DW_FORM_GNU_ref_alt:
+ AddrOS << format("<alt 0x%" PRIx64 ">", UValue);
+ break;
+
+ // All DW_FORM_indirect attributes should be resolved prior to calling
+ // this function
+ case DW_FORM_indirect:
+ OS << "DW_FORM_indirect";
+ break;
+
+ case DW_FORM_rnglistx:
+ OS << format("indexed (0x%x) rangelist = ", (uint32_t)UValue);
+ break;
+
+ // Should be formatted to 64-bit for DWARF64.
+ case DW_FORM_sec_offset:
+ AddrOS << format("0x%08x", (uint32_t)UValue);
+ break;
+
+ default:
+ OS << format("DW_FORM(0x%4.4x)", Form);
+ break;
+ }
+
+ if (CURelativeOffset) {
+ if (DumpOpts.Verbose)
+ OS << " => {";
+ if (DumpOpts.ShowAddresses)
+ WithColor(OS, HighlightColor::Address).get()
+ << format("0x%8.8" PRIx64, UValue + (U ? U->getOffset() : 0));
+ if (DumpOpts.Verbose)
+ OS << "}";
+ }
+}
+
+void DWARFFormValue::dumpString(raw_ostream &OS) const {
+ Optional<const char *> DbgStr = getAsCString();
+ if (DbgStr.hasValue()) {
+ auto COS = WithColor(OS, HighlightColor::String);
+ COS.get() << '"';
+ COS.get().write_escaped(DbgStr.getValue());
+ COS.get() << '"';
+ }
+}
+
+Optional<const char *> DWARFFormValue::getAsCString() const {
+ if (!isFormClass(FC_String))
+ return None;
+ if (Form == DW_FORM_string)
+ return Value.cstr;
+ // FIXME: Add support for DW_FORM_GNU_strp_alt
+ if (Form == DW_FORM_GNU_strp_alt || C == nullptr)
+ return None;
+ uint64_t Offset = Value.uval;
+ if (Form == DW_FORM_line_strp) {
+ // .debug_line_str is tracked in the Context.
+ if (const char *Str = C->getLineStringExtractor().getCStr(&Offset))
+ return Str;
+ return None;
+ }
+ if (Form == DW_FORM_GNU_str_index || Form == DW_FORM_strx ||
+ Form == DW_FORM_strx1 || Form == DW_FORM_strx2 || Form == DW_FORM_strx3 ||
+ Form == DW_FORM_strx4) {
+ if (!U)
+ return None;
+ Optional<uint64_t> StrOffset = U->getStringOffsetSectionItem(Offset);
+ if (!StrOffset)
+ return None;
+ Offset = *StrOffset;
+ }
+ // Prefer the Unit's string extractor, because for .dwo it will point to
+ // .debug_str.dwo, while the Context's extractor always uses .debug_str.
+ if (U) {
+ if (const char *Str = U->getStringExtractor().getCStr(&Offset))
+ return Str;
+ return None;
+ }
+ if (const char *Str = C->getStringExtractor().getCStr(&Offset))
+ return Str;
+ return None;
+}
+
+Optional<uint64_t> DWARFFormValue::getAsAddress() const {
+ if (auto SA = getAsSectionedAddress())
+ return SA->Address;
+ return None;
+}
+
+Optional<object::SectionedAddress>
+DWARFFormValue::getAsSectionedAddress() const {
+ if (!isFormClass(FC_Address))
+ return None;
+ if (Form == DW_FORM_GNU_addr_index || Form == DW_FORM_addrx) {
+ uint32_t Index = Value.uval;
+ if (!U)
+ return None;
+ Optional<object::SectionedAddress> SA = U->getAddrOffsetSectionItem(Index);
+ if (!SA)
+ return None;
+ return SA;
+ }
+ return {{Value.uval, Value.SectionIndex}};
+}
+
+Optional<uint64_t> DWARFFormValue::getAsReference() const {
+ if (auto R = getAsRelativeReference())
+ return R->Unit ? R->Unit->getOffset() + R->Offset : R->Offset;
+ return None;
+}
+
+Optional<DWARFFormValue::UnitOffset> DWARFFormValue::getAsRelativeReference() const {
+ if (!isFormClass(FC_Reference))
+ return None;
+ switch (Form) {
+ case DW_FORM_ref1:
+ case DW_FORM_ref2:
+ case DW_FORM_ref4:
+ case DW_FORM_ref8:
+ case DW_FORM_ref_udata:
+ if (!U)
+ return None;
+ return UnitOffset{const_cast<DWARFUnit*>(U), Value.uval};
+ case DW_FORM_ref_addr:
+ case DW_FORM_ref_sig8:
+ case DW_FORM_GNU_ref_alt:
+ return UnitOffset{nullptr, Value.uval};
+ default:
+ return None;
+ }
+}
+
+Optional<uint64_t> DWARFFormValue::getAsSectionOffset() const {
+ if (!isFormClass(FC_SectionOffset))
+ return None;
+ return Value.uval;
+}
+
+Optional<uint64_t> DWARFFormValue::getAsUnsignedConstant() const {
+ if ((!isFormClass(FC_Constant) && !isFormClass(FC_Flag)) ||
+ Form == DW_FORM_sdata)
+ return None;
+ return Value.uval;
+}
+
+Optional<int64_t> DWARFFormValue::getAsSignedConstant() const {
+ if ((!isFormClass(FC_Constant) && !isFormClass(FC_Flag)) ||
+ (Form == DW_FORM_udata &&
+ uint64_t(std::numeric_limits<int64_t>::max()) < Value.uval))
+ return None;
+ switch (Form) {
+ case DW_FORM_data4:
+ return int32_t(Value.uval);
+ case DW_FORM_data2:
+ return int16_t(Value.uval);
+ case DW_FORM_data1:
+ return int8_t(Value.uval);
+ case DW_FORM_sdata:
+ case DW_FORM_data8:
+ default:
+ return Value.sval;
+ }
+}
+
+Optional<ArrayRef<uint8_t>> DWARFFormValue::getAsBlock() const {
+ if (!isFormClass(FC_Block) && !isFormClass(FC_Exprloc) &&
+ Form != DW_FORM_data16)
+ return None;
+ return makeArrayRef(Value.data, Value.uval);
+}
+
+Optional<uint64_t> DWARFFormValue::getAsCStringOffset() const {
+ if (!isFormClass(FC_String) && Form == DW_FORM_string)
+ return None;
+ return Value.uval;
+}
+
+Optional<uint64_t> DWARFFormValue::getAsReferenceUVal() const {
+ if (!isFormClass(FC_Reference))
+ return None;
+ return Value.uval;
+}
diff --git a/third_party/llvm-project/DWARFGdbIndex.cpp b/third_party/llvm-project/DWARFGdbIndex.cpp
new file mode 100644
index 000000000..252b58e5a
--- /dev/null
+++ b/third_party/llvm-project/DWARFGdbIndex.cpp
@@ -0,0 +1,199 @@
+//===- DWARFGdbIndex.cpp --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFGdbIndex.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cinttypes>
+#include <cstdint>
+#include <utility>
+
+using namespace llvm;
+
+// .gdb_index section format reference:
+// https://sourceware.org/gdb/onlinedocs/gdb/Index-Section-Format.html
+
+void DWARFGdbIndex::dumpCUList(raw_ostream &OS) const {
+ OS << format("\n CU list offset = 0x%x, has %" PRId64 " entries:",
+ CuListOffset, (uint64_t)CuList.size())
+ << '\n';
+ uint32_t I = 0;
+ for (const CompUnitEntry &CU : CuList)
+ OS << format(" %d: Offset = 0x%llx, Length = 0x%llx\n", I++, CU.Offset,
+ CU.Length);
+}
+
+void DWARFGdbIndex::dumpTUList(raw_ostream &OS) const {
+ OS << formatv("\n Types CU list offset = {0:x}, has {1} entries:\n",
+ TuListOffset, TuList.size());
+ uint32_t I = 0;
+ for (const TypeUnitEntry &TU : TuList)
+ OS << formatv(" {0}: offset = {1:x8}, type_offset = {2:x8}, "
+ "type_signature = {3:x16}\n",
+ I++, TU.Offset, TU.TypeOffset, TU.TypeSignature);
+}
+
+void DWARFGdbIndex::dumpAddressArea(raw_ostream &OS) const {
+ OS << format("\n Address area offset = 0x%x, has %" PRId64 " entries:",
+ AddressAreaOffset, (uint64_t)AddressArea.size())
+ << '\n';
+ for (const AddressEntry &Addr : AddressArea)
+ OS << format(
+ " Low/High address = [0x%llx, 0x%llx) (Size: 0x%llx), CU id = %d\n",
+ Addr.LowAddress, Addr.HighAddress, Addr.HighAddress - Addr.LowAddress,
+ Addr.CuIndex);
+}
+
+void DWARFGdbIndex::dumpSymbolTable(raw_ostream &OS) const {
+ OS << format("\n Symbol table offset = 0x%x, size = %" PRId64
+ ", filled slots:",
+ SymbolTableOffset, (uint64_t)SymbolTable.size())
+ << '\n';
+ uint32_t I = -1;
+ for (const SymTableEntry &E : SymbolTable) {
+ ++I;
+ if (!E.NameOffset && !E.VecOffset)
+ continue;
+
+ OS << format(" %d: Name offset = 0x%x, CU vector offset = 0x%x\n", I,
+ E.NameOffset, E.VecOffset);
+
+ StringRef Name = ConstantPoolStrings.substr(
+ ConstantPoolOffset - StringPoolOffset + E.NameOffset);
+
+ auto CuVector = std::find_if(
+ ConstantPoolVectors.begin(), ConstantPoolVectors.end(),
+ [&](const std::pair<uint32_t, SmallVector<uint32_t, 0>> &V) {
+ return V.first == E.VecOffset;
+ });
+ assert(CuVector != ConstantPoolVectors.end() && "Invalid symbol table");
+ uint32_t CuVectorId = CuVector - ConstantPoolVectors.begin();
+ OS << format(" String name: %s, CU vector index: %d\n", Name.data(),
+ CuVectorId);
+ }
+}
+
+void DWARFGdbIndex::dumpConstantPool(raw_ostream &OS) const {
+ OS << format("\n Constant pool offset = 0x%x, has %" PRId64 " CU vectors:",
+ ConstantPoolOffset, (uint64_t)ConstantPoolVectors.size());
+ uint32_t I = 0;
+ for (const auto &V : ConstantPoolVectors) {
+ OS << format("\n %d(0x%x): ", I++, V.first);
+ for (uint32_t Val : V.second)
+ OS << format("0x%x ", Val);
+ }
+ OS << '\n';
+}
+
+void DWARFGdbIndex::dump(raw_ostream &OS) {
+ if (HasError) {
+ OS << "\n<error parsing>\n";
+ return;
+ }
+
+ if (HasContent) {
+ OS << " Version = " << Version << '\n';
+ dumpCUList(OS);
+ dumpTUList(OS);
+ dumpAddressArea(OS);
+ dumpSymbolTable(OS);
+ dumpConstantPool(OS);
+ }
+}
+
+bool DWARFGdbIndex::parseImpl(DataExtractor Data) {
+ uint64_t Offset = 0;
+
+ // Only version 7 is supported at this moment.
+ Version = Data.getU32(&Offset);
+ if (Version != 7)
+ return false;
+
+ CuListOffset = Data.getU32(&Offset);
+ TuListOffset = Data.getU32(&Offset);
+ AddressAreaOffset = Data.getU32(&Offset);
+ SymbolTableOffset = Data.getU32(&Offset);
+ ConstantPoolOffset = Data.getU32(&Offset);
+
+ if (Offset != CuListOffset)
+ return false;
+
+ uint32_t CuListSize = (TuListOffset - CuListOffset) / 16;
+ CuList.reserve(CuListSize);
+ for (uint32_t i = 0; i < CuListSize; ++i) {
+ uint64_t CuOffset = Data.getU64(&Offset);
+ uint64_t CuLength = Data.getU64(&Offset);
+ CuList.push_back({CuOffset, CuLength});
+ }
+
+ // CU Types are no longer needed as DWARF skeleton type units never made it
+ // into the standard.
+ uint32_t TuListSize = (AddressAreaOffset - TuListOffset) / 24;
+ TuList.resize(TuListSize);
+ for (uint32_t I = 0; I < TuListSize; ++I) {
+ uint64_t CuOffset = Data.getU64(&Offset);
+ uint64_t TypeOffset = Data.getU64(&Offset);
+ uint64_t Signature = Data.getU64(&Offset);
+ TuList[I] = {CuOffset, TypeOffset, Signature};
+ }
+
+ uint32_t AddressAreaSize = (SymbolTableOffset - AddressAreaOffset) / 20;
+ AddressArea.reserve(AddressAreaSize);
+ for (uint32_t i = 0; i < AddressAreaSize; ++i) {
+ uint64_t LowAddress = Data.getU64(&Offset);
+ uint64_t HighAddress = Data.getU64(&Offset);
+ uint32_t CuIndex = Data.getU32(&Offset);
+ AddressArea.push_back({LowAddress, HighAddress, CuIndex});
+ }
+
+ // The symbol table. This is an open addressed hash table. The size of the
+ // hash table is always a power of 2.
+ // Each slot in the hash table consists of a pair of offset_type values. The
+ // first value is the offset of the symbol's name in the constant pool. The
+ // second value is the offset of the CU vector in the constant pool.
+ // If both values are 0, then this slot in the hash table is empty. This is ok
+ // because while 0 is a valid constant pool index, it cannot be a valid index
+ // for both a string and a CU vector.
+ uint32_t SymTableSize = (ConstantPoolOffset - SymbolTableOffset) / 8;
+ SymbolTable.reserve(SymTableSize);
+ uint32_t CuVectorsTotal = 0;
+ for (uint32_t i = 0; i < SymTableSize; ++i) {
+ uint32_t NameOffset = Data.getU32(&Offset);
+ uint32_t CuVecOffset = Data.getU32(&Offset);
+ SymbolTable.push_back({NameOffset, CuVecOffset});
+ if (NameOffset || CuVecOffset)
+ ++CuVectorsTotal;
+ }
+
+ // The constant pool. CU vectors are stored first, followed by strings.
+ // The first value is the number of CU indices in the vector. Each subsequent
+ // value is the index and symbol attributes of a CU in the CU list.
+ for (uint32_t i = 0; i < CuVectorsTotal; ++i) {
+ ConstantPoolVectors.emplace_back(0, SmallVector<uint32_t, 0>());
+ auto &Vec = ConstantPoolVectors.back();
+ Vec.first = Offset - ConstantPoolOffset;
+
+ uint32_t Num = Data.getU32(&Offset);
+ for (uint32_t j = 0; j < Num; ++j)
+ Vec.second.push_back(Data.getU32(&Offset));
+ }
+
+ ConstantPoolStrings = Data.getData().drop_front(Offset);
+ StringPoolOffset = Offset;
+ return true;
+}
+
+void DWARFGdbIndex::parse(DataExtractor Data) {
+ HasContent = !Data.getData().empty();
+ HasError = HasContent && !parseImpl(Data);
+}
diff --git a/third_party/llvm-project/DWARFListTable.cpp b/third_party/llvm-project/DWARFListTable.cpp
new file mode 100644
index 000000000..269ea9f79
--- /dev/null
+++ b/third_party/llvm-project/DWARFListTable.cpp
@@ -0,0 +1,116 @@
+//===- DWARFListTable.cpp ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFListTable.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+Error DWARFListTableHeader::extract(DWARFDataExtractor Data,
+ uint64_t *OffsetPtr) {
+ HeaderOffset = *OffsetPtr;
+ // Read and verify the length field.
+ if (!Data.isValidOffsetForDataOfSize(*OffsetPtr, sizeof(uint32_t)))
+ return createStringError(errc::invalid_argument,
+ "section is not large enough to contain a "
+ "%s table length at offset 0x%" PRIx64,
+ SectionName.data(), *OffsetPtr);
+ Format = dwarf::DwarfFormat::DWARF32;
+ uint8_t OffsetByteSize = 4;
+ HeaderData.Length = Data.getRelocatedValue(4, OffsetPtr);
+ if (HeaderData.Length == dwarf::DW_LENGTH_DWARF64) {
+ Format = dwarf::DwarfFormat::DWARF64;
+ OffsetByteSize = 8;
+ HeaderData.Length = Data.getU64(OffsetPtr);
+ } else if (HeaderData.Length >= dwarf::DW_LENGTH_lo_reserved) {
+ return createStringError(errc::invalid_argument,
+ "%s table at offset 0x%" PRIx64
+ " has unsupported reserved unit length of value 0x%8.8" PRIx64,
+ SectionName.data(), HeaderOffset, HeaderData.Length);
+ }
+ uint64_t FullLength =
+ HeaderData.Length + dwarf::getUnitLengthFieldByteSize(Format);
+ assert(FullLength == length());
+ if (FullLength < getHeaderSize(Format))
+ return createStringError(errc::invalid_argument,
+ "%s table at offset 0x%" PRIx64
+ " has too small length (0x%" PRIx64
+ ") to contain a complete header",
+ SectionName.data(), HeaderOffset, FullLength);
+ uint64_t End = HeaderOffset + FullLength;
+ if (!Data.isValidOffsetForDataOfSize(HeaderOffset, FullLength))
+ return createStringError(errc::invalid_argument,
+ "section is not large enough to contain a %s table "
+ "of length 0x%" PRIx64 " at offset 0x%" PRIx64,
+ SectionName.data(), FullLength, HeaderOffset);
+
+ HeaderData.Version = Data.getU16(OffsetPtr);
+ HeaderData.AddrSize = Data.getU8(OffsetPtr);
+ HeaderData.SegSize = Data.getU8(OffsetPtr);
+ HeaderData.OffsetEntryCount = Data.getU32(OffsetPtr);
+
+ // Perform basic validation of the remaining header fields.
+ if (HeaderData.Version != 5)
+ return createStringError(errc::invalid_argument,
+ "unrecognised %s table version %" PRIu16
+ " in table at offset 0x%" PRIx64,
+ SectionName.data(), HeaderData.Version, HeaderOffset);
+ if (HeaderData.AddrSize != 4 && HeaderData.AddrSize != 8)
+ return createStringError(errc::not_supported,
+ "%s table at offset 0x%" PRIx64
+ " has unsupported address size %" PRIu8,
+ SectionName.data(), HeaderOffset, HeaderData.AddrSize);
+ if (HeaderData.SegSize != 0)
+ return createStringError(errc::not_supported,
+ "%s table at offset 0x%" PRIx64
+ " has unsupported segment selector size %" PRIu8,
+ SectionName.data(), HeaderOffset, HeaderData.SegSize);
+ if (End < HeaderOffset + getHeaderSize(Format) +
+ HeaderData.OffsetEntryCount * OffsetByteSize)
+ return createStringError(errc::invalid_argument,
+ "%s table at offset 0x%" PRIx64 " has more offset entries (%" PRIu32
+ ") than there is space for",
+ SectionName.data(), HeaderOffset, HeaderData.OffsetEntryCount);
+ Data.setAddressSize(HeaderData.AddrSize);
+ for (uint32_t I = 0; I < HeaderData.OffsetEntryCount; ++I)
+ Offsets.push_back(Data.getRelocatedValue(OffsetByteSize, OffsetPtr));
+ return Error::success();
+}
+
+void DWARFListTableHeader::dump(raw_ostream &OS, DIDumpOptions DumpOpts) const {
+ if (DumpOpts.Verbose)
+ OS << format("0x%8.8" PRIx64 ": ", HeaderOffset);
+ OS << format(
+ "%s list header: length = 0x%8.8" PRIx64 ", version = 0x%4.4" PRIx16 ", "
+ "addr_size = 0x%2.2" PRIx8 ", seg_size = 0x%2.2" PRIx8
+ ", offset_entry_count = "
+ "0x%8.8" PRIx32 "\n",
+ ListTypeString.data(), HeaderData.Length, HeaderData.Version,
+ HeaderData.AddrSize, HeaderData.SegSize, HeaderData.OffsetEntryCount);
+
+ if (HeaderData.OffsetEntryCount > 0) {
+ OS << "offsets: [";
+ for (const auto &Off : Offsets) {
+ OS << format("\n0x%8.8" PRIx64, Off);
+ if (DumpOpts.Verbose)
+ OS << format(" => 0x%8.8" PRIx64,
+ Off + HeaderOffset + getHeaderSize(Format));
+ }
+ OS << "\n]\n";
+ }
+}
+
+uint64_t DWARFListTableHeader::length() const {
+ if (HeaderData.Length == 0)
+ return 0;
+ return HeaderData.Length + dwarf::getUnitLengthFieldByteSize(Format);
+}
diff --git a/third_party/llvm-project/DWARFTypeUnit.cpp b/third_party/llvm-project/DWARFTypeUnit.cpp
new file mode 100644
index 000000000..bb81090ba
--- /dev/null
+++ b/third_party/llvm-project/DWARFTypeUnit.cpp
@@ -0,0 +1,49 @@
+//===- DWARFTypeUnit.cpp --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFTypeUnit.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cinttypes>
+
+using namespace llvm;
+
+void DWARFTypeUnit::dump(raw_ostream &OS, DIDumpOptions DumpOpts) {
+ DWARFDie TD = getDIEForOffset(getTypeOffset() + getOffset());
+ const char *Name = TD.getName(DINameKind::ShortName);
+
+ if (DumpOpts.SummarizeTypes) {
+ OS << "name = '" << Name << "'"
+ << " type_signature = " << format("0x%016" PRIx64, getTypeHash())
+ << " length = " << format("0x%08" PRIx64, getLength()) << '\n';
+ return;
+ }
+
+ OS << format("0x%08" PRIx64, getOffset()) << ": Type Unit:"
+ << " length = " << format("0x%08" PRIx64, getLength())
+ << " version = " << format("0x%04x", getVersion());
+ if (getVersion() >= 5)
+ OS << " unit_type = " << dwarf::UnitTypeString(getUnitType());
+ OS << " abbr_offset = "
+ << format("0x%04" PRIx64, getAbbreviations()->getOffset())
+ << " addr_size = " << format("0x%02x", getAddressByteSize())
+ << " name = '" << Name << "'"
+ << " type_signature = " << format("0x%016" PRIx64, getTypeHash())
+ << " type_offset = " << format("0x%04" PRIx64, getTypeOffset())
+ << " (next unit at " << format("0x%08" PRIx64, getNextUnitOffset())
+ << ")\n";
+
+ if (DWARFDie TU = getUnitDIE(false))
+ TU.dump(OS, 0, DumpOpts);
+ else
+ OS << "<type unit can't be parsed!>\n\n";
+}
diff --git a/third_party/llvm-project/DWARFUnit.cpp b/third_party/llvm-project/DWARFUnit.cpp
new file mode 100644
index 000000000..a56402a70
--- /dev/null
+++ b/third_party/llvm-project/DWARFUnit.cpp
@@ -0,0 +1,904 @@
+//===- DWARFUnit.cpp ------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRnglists.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFTypeUnit.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Path.h"
+#include "llvm/Support/WithColor.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstdio>
+#include <utility>
+#include <vector>
+
+using namespace llvm;
+using namespace dwarf;
+
+void DWARFUnitVector::addUnitsForSection(DWARFContext &C,
+ const DWARFSection &Section,
+ DWARFSectionKind SectionKind) {
+ const DWARFObject &D = C.getDWARFObj();
+ addUnitsImpl(C, D, Section, C.getDebugAbbrev(), &D.getRangesSection(),
+ &D.getLocSection(), D.getStrSection(),
+ D.getStrOffsetsSection(), &D.getAddrSection(),
+ D.getLineSection(), D.isLittleEndian(), false, false,
+ SectionKind);
+}
+
+void DWARFUnitVector::addUnitsForDWOSection(DWARFContext &C,
+ const DWARFSection &DWOSection,
+ DWARFSectionKind SectionKind,
+ bool Lazy) {
+ const DWARFObject &D = C.getDWARFObj();
+ addUnitsImpl(C, D, DWOSection, C.getDebugAbbrevDWO(), &D.getRangesDWOSection(),
+ &D.getLocDWOSection(), D.getStrDWOSection(),
+ D.getStrOffsetsDWOSection(), &D.getAddrSection(),
+ D.getLineDWOSection(), C.isLittleEndian(), true, Lazy,
+ SectionKind);
+}
+
+void DWARFUnitVector::addUnitsImpl(
+ DWARFContext &Context, const DWARFObject &Obj, const DWARFSection &Section,
+ const DWARFDebugAbbrev *DA, const DWARFSection *RS,
+ const DWARFSection *LocSection, StringRef SS, const DWARFSection &SOS,
+ const DWARFSection *AOS, const DWARFSection &LS, bool LE, bool IsDWO,
+ bool Lazy, DWARFSectionKind SectionKind) {
+ DWARFDataExtractor Data(Obj, Section, LE, 0);
+ // Lazy initialization of Parser, now that we have all section info.
+ if (!Parser) {
+ Parser = [=, &Context, &Obj, &Section, &SOS,
+ &LS](uint64_t Offset, DWARFSectionKind SectionKind,
+ const DWARFSection *CurSection,
+ const DWARFUnitIndex::Entry *IndexEntry)
+ -> std::unique_ptr<DWARFUnit> {
+ const DWARFSection &InfoSection = CurSection ? *CurSection : Section;
+ DWARFDataExtractor Data(Obj, InfoSection, LE, 0);
+ if (!Data.isValidOffset(Offset))
+ return nullptr;
+ const DWARFUnitIndex *Index = nullptr;
+ if (IsDWO)
+ Index = &getDWARFUnitIndex(Context, SectionKind);
+ DWARFUnitHeader Header;
+ if (!Header.extract(Context, Data, &Offset, SectionKind, Index,
+ IndexEntry))
+ return nullptr;
+ std::unique_ptr<DWARFUnit> U;
+ if (Header.isTypeUnit())
+ U = std::make_unique<DWARFTypeUnit>(Context, InfoSection, Header, DA,
+ RS, LocSection, SS, SOS, AOS, LS,
+ LE, IsDWO, *this);
+ else
+ U = std::make_unique<DWARFCompileUnit>(Context, InfoSection, Header,
+ DA, RS, LocSection, SS, SOS,
+ AOS, LS, LE, IsDWO, *this);
+ return U;
+ };
+ }
+ if (Lazy)
+ return;
+ // Find a reasonable insertion point within the vector. We skip over
+ // (a) units from a different section, (b) units from the same section
+ // but with lower offset-within-section. This keeps units in order
+ // within a section, although not necessarily within the object file,
+ // even if we do lazy parsing.
+ auto I = this->begin();
+ uint64_t Offset = 0;
+ while (Data.isValidOffset(Offset)) {
+ if (I != this->end() &&
+ (&(*I)->getInfoSection() != &Section || (*I)->getOffset() == Offset)) {
+ ++I;
+ continue;
+ }
+ auto U = Parser(Offset, SectionKind, &Section, nullptr);
+ // If parsing failed, we're done with this section.
+ if (!U)
+ break;
+ Offset = U->getNextUnitOffset();
+ I = std::next(this->insert(I, std::move(U)));
+ }
+}
+
+DWARFUnit *DWARFUnitVector::addUnit(std::unique_ptr<DWARFUnit> Unit) {
+ auto I = std::upper_bound(begin(), end(), Unit,
+ [](const std::unique_ptr<DWARFUnit> &LHS,
+ const std::unique_ptr<DWARFUnit> &RHS) {
+ return LHS->getOffset() < RHS->getOffset();
+ });
+ return this->insert(I, std::move(Unit))->get();
+}
+
+DWARFUnit *DWARFUnitVector::getUnitForOffset(uint64_t Offset) const {
+ auto end = begin() + getNumInfoUnits();
+ auto *CU =
+ std::upper_bound(begin(), end, Offset,
+ [](uint64_t LHS, const std::unique_ptr<DWARFUnit> &RHS) {
+ return LHS < RHS->getNextUnitOffset();
+ });
+ if (CU != end && (*CU)->getOffset() <= Offset)
+ return CU->get();
+ return nullptr;
+}
+
+DWARFUnit *
+DWARFUnitVector::getUnitForIndexEntry(const DWARFUnitIndex::Entry &E) {
+ const auto *CUOff = E.getOffset(DW_SECT_INFO);
+ if (!CUOff)
+ return nullptr;
+
+ auto Offset = CUOff->Offset;
+ auto end = begin() + getNumInfoUnits();
+
+ auto *CU =
+ std::upper_bound(begin(), end, CUOff->Offset,
+ [](uint64_t LHS, const std::unique_ptr<DWARFUnit> &RHS) {
+ return LHS < RHS->getNextUnitOffset();
+ });
+ if (CU != end && (*CU)->getOffset() <= Offset)
+ return CU->get();
+
+ if (!Parser)
+ return nullptr;
+
+ auto U = Parser(Offset, DW_SECT_INFO, nullptr, &E);
+ if (!U)
+ U = nullptr;
+
+ auto *NewCU = U.get();
+ this->insert(CU, std::move(U));
+ ++NumInfoUnits;
+ return NewCU;
+}
+
+DWARFUnit::DWARFUnit(DWARFContext &DC, const DWARFSection &Section,
+ const DWARFUnitHeader &Header, const DWARFDebugAbbrev *DA,
+ const DWARFSection *RS, const DWARFSection *LocSection,
+ StringRef SS, const DWARFSection &SOS,
+ const DWARFSection *AOS, const DWARFSection &LS, bool LE,
+ bool IsDWO, const DWARFUnitVector &UnitVector)
+ : Context(DC), InfoSection(Section), Header(Header), Abbrev(DA),
+ RangeSection(RS), LocSection(LocSection), LineSection(LS),
+ StringSection(SS), StringOffsetSection(SOS), AddrOffsetSection(AOS),
+ isLittleEndian(LE), IsDWO(IsDWO), UnitVector(UnitVector) {
+ clear();
+ // For split DWARF we only need to keep track of the location list section's
+ // data (no relocations), and if we are reading a package file, we need to
+ // adjust the location list data based on the index entries.
+ if (IsDWO) {
+ LocSectionData = LocSection->Data;
+ if (auto *IndexEntry = Header.getIndexEntry())
+ if (const auto *C = IndexEntry->getOffset(DW_SECT_LOC))
+ LocSectionData = LocSectionData.substr(C->Offset, C->Length);
+ }
+}
+
+DWARFUnit::~DWARFUnit() = default;
+
+DWARFDataExtractor DWARFUnit::getDebugInfoExtractor() const {
+ return DWARFDataExtractor(Context.getDWARFObj(), InfoSection, isLittleEndian,
+ getAddressByteSize());
+}
+
+Optional<object::SectionedAddress>
+DWARFUnit::getAddrOffsetSectionItem(uint32_t Index) const {
+ if (IsDWO) {
+ auto R = Context.info_section_units();
+ auto I = R.begin();
+ // Surprising if a DWO file has more than one skeleton unit in it - this
+ // probably shouldn't be valid, but if a use case is found, here's where to
+ // support it (probably have to linearly search for the matching skeleton CU
+ // here)
+ if (I != R.end() && std::next(I) == R.end())
+ return (*I)->getAddrOffsetSectionItem(Index);
+ }
+ uint64_t Offset = AddrOffsetSectionBase + Index * getAddressByteSize();
+ if (AddrOffsetSection->Data.size() < Offset + getAddressByteSize())
+ return None;
+ DWARFDataExtractor DA(Context.getDWARFObj(), *AddrOffsetSection,
+ isLittleEndian, getAddressByteSize());
+ uint64_t Section;
+ uint64_t Address = DA.getRelocatedAddress(&Offset, &Section);
+ return {{Address, Section}};
+}
+
+Optional<uint64_t> DWARFUnit::getStringOffsetSectionItem(uint32_t Index) const {
+ if (!StringOffsetsTableContribution)
+ return None;
+ unsigned ItemSize = getDwarfStringOffsetsByteSize();
+ uint64_t Offset = getStringOffsetsBase() + Index * ItemSize;
+ if (StringOffsetSection.Data.size() < Offset + ItemSize)
+ return None;
+ DWARFDataExtractor DA(Context.getDWARFObj(), StringOffsetSection,
+ isLittleEndian, 0);
+ return DA.getRelocatedValue(ItemSize, &Offset);
+}
+
+bool DWARFUnitHeader::extract(DWARFContext &Context,
+ const DWARFDataExtractor &debug_info,
+ uint64_t *offset_ptr,
+ DWARFSectionKind SectionKind,
+ const DWARFUnitIndex *Index,
+ const DWARFUnitIndex::Entry *Entry) {
+ Offset = *offset_ptr;
+ IndexEntry = Entry;
+ if (!IndexEntry && Index)
+ IndexEntry = Index->getFromOffset(*offset_ptr);
+ Length = debug_info.getRelocatedValue(4, offset_ptr);
+ FormParams.Format = DWARF32;
+ if (Length == dwarf::DW_LENGTH_DWARF64) {
+ Length = debug_info.getU64(offset_ptr);
+ FormParams.Format = DWARF64;
+ }
+ FormParams.Version = debug_info.getU16(offset_ptr);
+ if (FormParams.Version >= 5) {
+ UnitType = debug_info.getU8(offset_ptr);
+ FormParams.AddrSize = debug_info.getU8(offset_ptr);
+ AbbrOffset = debug_info.getRelocatedValue(FormParams.getDwarfOffsetByteSize(), offset_ptr);
+ } else {
+ AbbrOffset = debug_info.getRelocatedValue(FormParams.getDwarfOffsetByteSize(), offset_ptr);
+ FormParams.AddrSize = debug_info.getU8(offset_ptr);
+ // Fake a unit type based on the section type. This isn't perfect,
+ // but distinguishing compile and type units is generally enough.
+ if (SectionKind == DW_SECT_TYPES)
+ UnitType = DW_UT_type;
+ else
+ UnitType = DW_UT_compile;
+ }
+ if (IndexEntry) {
+ if (AbbrOffset)
+ return false;
+ auto *UnitContrib = IndexEntry->getOffset();
+ if (!UnitContrib || UnitContrib->Length != (Length + 4))
+ return false;
+ auto *AbbrEntry = IndexEntry->getOffset(DW_SECT_ABBREV);
+ if (!AbbrEntry)
+ return false;
+ AbbrOffset = AbbrEntry->Offset;
+ }
+ if (isTypeUnit()) {
+ TypeHash = debug_info.getU64(offset_ptr);
+ TypeOffset =
+ debug_info.getUnsigned(offset_ptr, FormParams.getDwarfOffsetByteSize());
+ } else if (UnitType == DW_UT_split_compile || UnitType == DW_UT_skeleton)
+ DWOId = debug_info.getU64(offset_ptr);
+
+ // Header fields all parsed, capture the size of this unit header.
+ assert(*offset_ptr - Offset <= 255 && "unexpected header size");
+ Size = uint8_t(*offset_ptr - Offset);
+
+ // Type offset is unit-relative; should be after the header and before
+ // the end of the current unit.
+ bool TypeOffsetOK =
+ !isTypeUnit()
+ ? true
+ : TypeOffset >= Size &&
+ TypeOffset < getLength() + getUnitLengthFieldByteSize();
+ bool LengthOK = debug_info.isValidOffset(getNextUnitOffset() - 1);
+ bool VersionOK = DWARFContext::isSupportedVersion(getVersion());
+ bool AddrSizeOK = getAddressByteSize() == 4 || getAddressByteSize() == 8;
+
+ if (!LengthOK || !VersionOK || !AddrSizeOK || !TypeOffsetOK)
+ return false;
+
+ // Keep track of the highest DWARF version we encounter across all units.
+ Context.setMaxVersionIfGreater(getVersion());
+ return true;
+}
+
+// Parse the rangelist table header, including the optional array of offsets
+// following it (DWARF v5 and later).
+static Expected<DWARFDebugRnglistTable>
+parseRngListTableHeader(DWARFDataExtractor &DA, uint64_t Offset,
+ DwarfFormat Format) {
+ // We are expected to be called with Offset 0 or pointing just past the table
+ // header. Correct Offset in the latter case so that it points to the start
+ // of the header.
+ if (Offset > 0) {
+ uint64_t HeaderSize = DWARFListTableHeader::getHeaderSize(Format);
+ if (Offset < HeaderSize)
+ return createStringError(errc::invalid_argument, "Did not detect a valid"
+ " range list table with base = 0x%" PRIx64 "\n",
+ Offset);
+ Offset -= HeaderSize;
+ }
+ llvm::DWARFDebugRnglistTable Table;
+ if (Error E = Table.extractHeaderAndOffsets(DA, &Offset))
+ return std::move(E);
+ return Table;
+}
+
+Error DWARFUnit::extractRangeList(uint64_t RangeListOffset,
+ DWARFDebugRangeList &RangeList) const {
+ // Require that compile unit is extracted.
+ assert(!DieArray.empty());
+ DWARFDataExtractor RangesData(Context.getDWARFObj(), *RangeSection,
+ isLittleEndian, getAddressByteSize());
+ uint64_t ActualRangeListOffset = RangeSectionBase + RangeListOffset;
+ return RangeList.extract(RangesData, &ActualRangeListOffset);
+}
+
+void DWARFUnit::clear() {
+ Abbrevs = nullptr;
+ BaseAddr.reset();
+ RangeSectionBase = 0;
+ AddrOffsetSectionBase = 0;
+ clearDIEs(false);
+ DWO.reset();
+}
+
+const char *DWARFUnit::getCompilationDir() {
+ return dwarf::toString(getUnitDIE().find(DW_AT_comp_dir), nullptr);
+}
+
+void DWARFUnit::extractDIEsToVector(
+ bool AppendCUDie, bool AppendNonCUDies,
+ std::vector<DWARFDebugInfoEntry> &Dies) const {
+ if (!AppendCUDie && !AppendNonCUDies)
+ return;
+
+ // Set the offset to that of the first DIE and calculate the start of the
+ // next compilation unit header.
+ uint64_t DIEOffset = getOffset() + getHeaderSize();
+ uint64_t NextCUOffset = getNextUnitOffset();
+ DWARFDebugInfoEntry DIE;
+ DWARFDataExtractor DebugInfoData = getDebugInfoExtractor();
+ uint32_t Depth = 0;
+ bool IsCUDie = true;
+
+ while (DIE.extractFast(*this, &DIEOffset, DebugInfoData, NextCUOffset,
+ Depth)) {
+ if (IsCUDie) {
+ if (AppendCUDie)
+ Dies.push_back(DIE);
+ if (!AppendNonCUDies)
+ break;
+ // The average bytes per DIE entry has been seen to be
+ // around 14-20 so let's pre-reserve the needed memory for
+ // our DIE entries accordingly.
+ Dies.reserve(Dies.size() + getDebugInfoSize() / 14);
+ IsCUDie = false;
+ } else {
+ Dies.push_back(DIE);
+ }
+
+ if (const DWARFAbbreviationDeclaration *AbbrDecl =
+ DIE.getAbbreviationDeclarationPtr()) {
+ // Normal DIE
+ if (AbbrDecl->hasChildren())
+ ++Depth;
+ } else {
+ // NULL DIE.
+ if (Depth > 0)
+ --Depth;
+ if (Depth == 0)
+ break; // We are done with this compile unit!
+ }
+ }
+
+ // Give a little bit of info if we encounter corrupt DWARF (our offset
+ // should always terminate at or before the start of the next compilation
+ // unit header).
+ if (DIEOffset > NextCUOffset)
+ WithColor::warning() << format("DWARF compile unit extends beyond its "
+ "bounds cu 0x%8.8" PRIx64 " "
+ "at 0x%8.8" PRIx64 "\n",
+ getOffset(), DIEOffset);
+}
+
+void DWARFUnit::extractDIEsIfNeeded(bool CUDieOnly) {
+ if (Error e = tryExtractDIEsIfNeeded(CUDieOnly))
+ WithColor::error() << toString(std::move(e));
+}
+
+Error DWARFUnit::tryExtractDIEsIfNeeded(bool CUDieOnly) {
+ if ((CUDieOnly && !DieArray.empty()) ||
+ DieArray.size() > 1)
+ return Error::success(); // Already parsed.
+
+ bool HasCUDie = !DieArray.empty();
+ extractDIEsToVector(!HasCUDie, !CUDieOnly, DieArray);
+
+ if (DieArray.empty())
+ return Error::success();
+
+ // If CU DIE was just parsed, copy several attribute values from it.
+ if (HasCUDie)
+ return Error::success();
+
+ DWARFDie UnitDie(this, &DieArray[0]);
+ if (Optional<uint64_t> DWOId = toUnsigned(UnitDie.find(DW_AT_GNU_dwo_id)))
+ Header.setDWOId(*DWOId);
+ if (!IsDWO) {
+ assert(AddrOffsetSectionBase == 0);
+ assert(RangeSectionBase == 0);
+ AddrOffsetSectionBase = toSectionOffset(UnitDie.find(DW_AT_addr_base), 0);
+ if (!AddrOffsetSectionBase)
+ AddrOffsetSectionBase =
+ toSectionOffset(UnitDie.find(DW_AT_GNU_addr_base), 0);
+ RangeSectionBase = toSectionOffset(UnitDie.find(DW_AT_rnglists_base), 0);
+ }
+
+ // In general, in DWARF v5 and beyond we derive the start of the unit's
+ // contribution to the string offsets table from the unit DIE's
+ // DW_AT_str_offsets_base attribute. Split DWARF units do not use this
+ // attribute, so we assume that there is a contribution to the string
+ // offsets table starting at offset 0 of the debug_str_offsets.dwo section.
+ // In both cases we need to determine the format of the contribution,
+ // which may differ from the unit's format.
+ DWARFDataExtractor DA(Context.getDWARFObj(), StringOffsetSection,
+ isLittleEndian, 0);
+ if (IsDWO || getVersion() >= 5) {
+ auto StringOffsetOrError =
+ IsDWO ? determineStringOffsetsTableContributionDWO(DA)
+ : determineStringOffsetsTableContribution(DA);
+ if (!StringOffsetOrError)
+ return createStringError(errc::invalid_argument,
+ "invalid reference to or invalid content in "
+ ".debug_str_offsets[.dwo]: " +
+ toString(StringOffsetOrError.takeError()));
+
+ StringOffsetsTableContribution = *StringOffsetOrError;
+ }
+
+ // DWARF v5 uses the .debug_rnglists and .debug_rnglists.dwo sections to
+ // describe address ranges.
+ if (getVersion() >= 5) {
+ if (IsDWO)
+ setRangesSection(&Context.getDWARFObj().getRnglistsDWOSection(), 0);
+ else
+ setRangesSection(&Context.getDWARFObj().getRnglistsSection(),
+ toSectionOffset(UnitDie.find(DW_AT_rnglists_base), 0));
+ if (RangeSection->Data.size()) {
+ // Parse the range list table header. Individual range lists are
+ // extracted lazily.
+ DWARFDataExtractor RangesDA(Context.getDWARFObj(), *RangeSection,
+ isLittleEndian, 0);
+ auto TableOrError = parseRngListTableHeader(RangesDA, RangeSectionBase,
+ Header.getFormat());
+ if (!TableOrError)
+ return createStringError(errc::invalid_argument,
+ "parsing a range list table: " +
+ toString(TableOrError.takeError()));
+
+ RngListTable = TableOrError.get();
+
+ // In a split dwarf unit, there is no DW_AT_rnglists_base attribute.
+ // Adjust RangeSectionBase to point past the table header.
+ if (IsDWO && RngListTable)
+ RangeSectionBase = RngListTable->getHeaderSize();
+ }
+ }
+
+ // Don't fall back to DW_AT_GNU_ranges_base: it should be ignored for
+ // skeleton CU DIE, so that DWARF users not aware of it are not broken.
+ return Error::success();
+}
+
+bool DWARFUnit::parseDWO() {
+ if (IsDWO)
+ return false;
+ if (DWO.get())
+ return false;
+ DWARFDie UnitDie = getUnitDIE();
+ if (!UnitDie)
+ return false;
+ auto DWOFileName = dwarf::toString(UnitDie.find(DW_AT_GNU_dwo_name));
+ if (!DWOFileName)
+ return false;
+ auto CompilationDir = dwarf::toString(UnitDie.find(DW_AT_comp_dir));
+ SmallString<16> AbsolutePath;
+ if (sys::path::is_relative(*DWOFileName) && CompilationDir &&
+ *CompilationDir) {
+ sys::path::append(AbsolutePath, *CompilationDir);
+ }
+ sys::path::append(AbsolutePath, *DWOFileName);
+ auto DWOId = getDWOId();
+ if (!DWOId)
+ return false;
+ auto DWOContext = Context.getDWOContext(AbsolutePath);
+ if (!DWOContext)
+ return false;
+
+ DWARFCompileUnit *DWOCU = DWOContext->getDWOCompileUnitForHash(*DWOId);
+ if (!DWOCU)
+ return false;
+ DWO = std::shared_ptr<DWARFCompileUnit>(std::move(DWOContext), DWOCU);
+ // Share .debug_addr and .debug_ranges section with compile unit in .dwo
+ DWO->setAddrOffsetSection(AddrOffsetSection, AddrOffsetSectionBase);
+ if (getVersion() >= 5) {
+ DWO->setRangesSection(&Context.getDWARFObj().getRnglistsDWOSection(), 0);
+ DWARFDataExtractor RangesDA(Context.getDWARFObj(), *RangeSection,
+ isLittleEndian, 0);
+ if (auto TableOrError = parseRngListTableHeader(RangesDA, RangeSectionBase,
+ Header.getFormat()))
+ DWO->RngListTable = TableOrError.get();
+ else
+ WithColor::error() << "parsing a range list table: "
+ << toString(TableOrError.takeError())
+ << '\n';
+ if (DWO->RngListTable)
+ DWO->RangeSectionBase = DWO->RngListTable->getHeaderSize();
+ } else {
+ auto DWORangesBase = UnitDie.getRangesBaseAttribute();
+ DWO->setRangesSection(RangeSection, DWORangesBase ? *DWORangesBase : 0);
+ }
+
+ return true;
+}
+
+void DWARFUnit::clearDIEs(bool KeepCUDie) {
+ if (DieArray.size() > (unsigned)KeepCUDie) {
+ DieArray.resize((unsigned)KeepCUDie);
+ DieArray.shrink_to_fit();
+ }
+}
+
+Expected<DWARFAddressRangesVector>
+DWARFUnit::findRnglistFromOffset(uint64_t Offset) {
+ if (getVersion() <= 4) {
+ DWARFDebugRangeList RangeList;
+ if (Error E = extractRangeList(Offset, RangeList))
+ return std::move(E);
+ return RangeList.getAbsoluteRanges(getBaseAddress());
+ }
+ if (RngListTable) {
+ DWARFDataExtractor RangesData(Context.getDWARFObj(), *RangeSection,
+ isLittleEndian, RngListTable->getAddrSize());
+ auto RangeListOrError = RngListTable->findList(RangesData, Offset);
+ if (RangeListOrError)
+ return RangeListOrError.get().getAbsoluteRanges(getBaseAddress(), *this);
+ return RangeListOrError.takeError();
+ }
+
+ return createStringError(errc::invalid_argument,
+ "missing or invalid range list table");
+}
+
+Expected<DWARFAddressRangesVector>
+DWARFUnit::findRnglistFromIndex(uint32_t Index) {
+ if (auto Offset = getRnglistOffset(Index))
+ return findRnglistFromOffset(*Offset + RangeSectionBase);
+
+ if (RngListTable)
+ return createStringError(errc::invalid_argument,
+ "invalid range list table index %d", Index);
+
+ return createStringError(errc::invalid_argument,
+ "missing or invalid range list table");
+}
+
+Expected<DWARFAddressRangesVector> DWARFUnit::collectAddressRanges() {
+ DWARFDie UnitDie = getUnitDIE();
+ if (!UnitDie)
+ return createStringError(errc::invalid_argument, "No unit DIE");
+
+ // First, check if unit DIE describes address ranges for the whole unit.
+ auto CUDIERangesOrError = UnitDie.getAddressRanges();
+ if (!CUDIERangesOrError)
+ return createStringError(errc::invalid_argument,
+ "decoding address ranges: %s",
+ toString(CUDIERangesOrError.takeError()).c_str());
+ return *CUDIERangesOrError;
+}
+
+void DWARFUnit::updateAddressDieMap(DWARFDie Die) {
+ if (Die.isSubroutineDIE()) {
+ auto DIERangesOrError = Die.getAddressRanges();
+ if (DIERangesOrError) {
+ for (const auto &R : DIERangesOrError.get()) {
+ // Ignore 0-sized ranges.
+ if (R.LowPC == R.HighPC)
+ continue;
+ auto B = AddrDieMap.upper_bound(R.LowPC);
+ if (B != AddrDieMap.begin() && R.LowPC < (--B)->second.first) {
+ // The range is a sub-range of existing ranges, we need to split the
+ // existing range.
+ if (R.HighPC < B->second.first)
+ AddrDieMap[R.HighPC] = B->second;
+ if (R.LowPC > B->first)
+ AddrDieMap[B->first].first = R.LowPC;
+ }
+ AddrDieMap[R.LowPC] = std::make_pair(R.HighPC, Die);
+ }
+ } else
+ llvm::consumeError(DIERangesOrError.takeError());
+ }
+ // Parent DIEs are added to the AddrDieMap prior to the Children DIEs to
+ // simplify the logic to update AddrDieMap. The child's range will always
+ // be equal or smaller than the parent's range. With this assumption, when
+ // adding one range into the map, it will at most split a range into 3
+ // sub-ranges.
+ for (DWARFDie Child = Die.getFirstChild(); Child; Child = Child.getSibling())
+ updateAddressDieMap(Child);
+}
+
+DWARFDie DWARFUnit::getSubroutineForAddress(uint64_t Address) {
+ extractDIEsIfNeeded(false);
+ if (AddrDieMap.empty())
+ updateAddressDieMap(getUnitDIE());
+ auto R = AddrDieMap.upper_bound(Address);
+ if (R == AddrDieMap.begin())
+ return DWARFDie();
+ // upper_bound's previous item contains Address.
+ --R;
+ if (Address >= R->second.first)
+ return DWARFDie();
+ return R->second.second;
+}
+
+void
+DWARFUnit::getInlinedChainForAddress(uint64_t Address,
+ SmallVectorImpl<DWARFDie> &InlinedChain) {
+ assert(InlinedChain.empty());
+ // Try to look for subprogram DIEs in the DWO file.
+ parseDWO();
+ // First, find the subroutine that contains the given address (the leaf
+ // of inlined chain).
+ DWARFDie SubroutineDIE =
+ (DWO ? *DWO : *this).getSubroutineForAddress(Address);
+
+ if (!SubroutineDIE)
+ return;
+
+ while (!SubroutineDIE.isSubprogramDIE()) {
+ if (SubroutineDIE.getTag() == DW_TAG_inlined_subroutine)
+ InlinedChain.push_back(SubroutineDIE);
+ SubroutineDIE = SubroutineDIE.getParent();
+ }
+ InlinedChain.push_back(SubroutineDIE);
+}
+
+const DWARFUnitIndex &llvm::getDWARFUnitIndex(DWARFContext &Context,
+ DWARFSectionKind Kind) {
+ if (Kind == DW_SECT_INFO)
+ return Context.getCUIndex();
+ assert(Kind == DW_SECT_TYPES);
+ return Context.getTUIndex();
+}
+
+DWARFDie DWARFUnit::getParent(const DWARFDebugInfoEntry *Die) {
+ if (!Die)
+ return DWARFDie();
+ const uint32_t Depth = Die->getDepth();
+ // Unit DIEs always have a depth of zero and never have parents.
+ if (Depth == 0)
+ return DWARFDie();
+ // Depth of 1 always means parent is the compile/type unit.
+ if (Depth == 1)
+ return getUnitDIE();
+ // Look for previous DIE with a depth that is one less than the Die's depth.
+ const uint32_t ParentDepth = Depth - 1;
+ for (uint32_t I = getDIEIndex(Die) - 1; I > 0; --I) {
+ if (DieArray[I].getDepth() == ParentDepth)
+ return DWARFDie(this, &DieArray[I]);
+ }
+ return DWARFDie();
+}
+
+DWARFDie DWARFUnit::getSibling(const DWARFDebugInfoEntry *Die) {
+ if (!Die)
+ return DWARFDie();
+ uint32_t Depth = Die->getDepth();
+ // Unit DIEs always have a depth of zero and never have siblings.
+ if (Depth == 0)
+ return DWARFDie();
+ // NULL DIEs don't have siblings.
+ if (Die->getAbbreviationDeclarationPtr() == nullptr)
+ return DWARFDie();
+
+ // Find the next DIE whose depth is the same as the Die's depth.
+ for (size_t I = getDIEIndex(Die) + 1, EndIdx = DieArray.size(); I < EndIdx;
+ ++I) {
+ if (DieArray[I].getDepth() == Depth)
+ return DWARFDie(this, &DieArray[I]);
+ }
+ return DWARFDie();
+}
+
+DWARFDie DWARFUnit::getPreviousSibling(const DWARFDebugInfoEntry *Die) {
+ if (!Die)
+ return DWARFDie();
+ uint32_t Depth = Die->getDepth();
+ // Unit DIEs always have a depth of zero and never have siblings.
+ if (Depth == 0)
+ return DWARFDie();
+
+ // Find the previous DIE whose depth is the same as the Die's depth.
+ for (size_t I = getDIEIndex(Die); I > 0;) {
+ --I;
+ if (DieArray[I].getDepth() == Depth - 1)
+ return DWARFDie();
+ if (DieArray[I].getDepth() == Depth)
+ return DWARFDie(this, &DieArray[I]);
+ }
+ return DWARFDie();
+}
+
+DWARFDie DWARFUnit::getFirstChild(const DWARFDebugInfoEntry *Die) {
+ if (!Die->hasChildren())
+ return DWARFDie();
+
+ // We do not want access out of bounds when parsing corrupted debug data.
+ size_t I = getDIEIndex(Die) + 1;
+ if (I >= DieArray.size())
+ return DWARFDie();
+ return DWARFDie(this, &DieArray[I]);
+}
+
+DWARFDie DWARFUnit::getLastChild(const DWARFDebugInfoEntry *Die) {
+ if (!Die->hasChildren())
+ return DWARFDie();
+
+ uint32_t Depth = Die->getDepth();
+ for (size_t I = getDIEIndex(Die) + 1, EndIdx = DieArray.size(); I < EndIdx;
+ ++I) {
+ if (DieArray[I].getDepth() == Depth + 1 &&
+ DieArray[I].getTag() == dwarf::DW_TAG_null)
+ return DWARFDie(this, &DieArray[I]);
+ assert(DieArray[I].getDepth() > Depth && "Not processing children?");
+ }
+ return DWARFDie();
+}
+
+const DWARFAbbreviationDeclarationSet *DWARFUnit::getAbbreviations() const {
+ if (!Abbrevs)
+ Abbrevs = Abbrev->getAbbreviationDeclarationSet(Header.getAbbrOffset());
+ return Abbrevs;
+}
+
+llvm::Optional<object::SectionedAddress> DWARFUnit::getBaseAddress() {
+ if (BaseAddr)
+ return BaseAddr;
+
+ DWARFDie UnitDie = getUnitDIE();
+ Optional<DWARFFormValue> PC = UnitDie.find({DW_AT_low_pc, DW_AT_entry_pc});
+ BaseAddr = toSectionedAddress(PC);
+ return BaseAddr;
+}
+
+Expected<StrOffsetsContributionDescriptor>
+StrOffsetsContributionDescriptor::validateContributionSize(
+ DWARFDataExtractor &DA) {
+ uint8_t EntrySize = getDwarfOffsetByteSize();
+ // In order to ensure that we don't read a partial record at the end of
+ // the section we validate for a multiple of the entry size.
+ uint64_t ValidationSize = alignTo(Size, EntrySize);
+ // Guard against overflow.
+ if (ValidationSize >= Size)
+ if (DA.isValidOffsetForDataOfSize((uint32_t)Base, ValidationSize))
+ return *this;
+ return createStringError(errc::invalid_argument, "length exceeds section size");
+}
+
+// Look for a DWARF64-formatted contribution to the string offsets table
+// starting at a given offset and record it in a descriptor.
+static Expected<StrOffsetsContributionDescriptor>
+parseDWARF64StringOffsetsTableHeader(DWARFDataExtractor &DA, uint64_t Offset) {
+ if (!DA.isValidOffsetForDataOfSize(Offset, 16))
+ return createStringError(errc::invalid_argument, "section offset exceeds section size");
+
+ if (DA.getU32(&Offset) != dwarf::DW_LENGTH_DWARF64)
+ return createStringError(errc::invalid_argument, "32 bit contribution referenced from a 64 bit unit");
+
+ uint64_t Size = DA.getU64(&Offset);
+ uint8_t Version = DA.getU16(&Offset);
+ (void)DA.getU16(&Offset); // padding
+ // The encoded length includes the 2-byte version field and the 2-byte
+ // padding, so we need to subtract them out when we populate the descriptor.
+ return StrOffsetsContributionDescriptor(Offset, Size - 4, Version, DWARF64);
+}
+
+// Look for a DWARF32-formatted contribution to the string offsets table
+// starting at a given offset and record it in a descriptor.
+static Expected<StrOffsetsContributionDescriptor>
+parseDWARF32StringOffsetsTableHeader(DWARFDataExtractor &DA, uint64_t Offset) {
+ if (!DA.isValidOffsetForDataOfSize(Offset, 8))
+ return createStringError(errc::invalid_argument, "section offset exceeds section size");
+
+ uint32_t ContributionSize = DA.getU32(&Offset);
+ if (ContributionSize >= dwarf::DW_LENGTH_lo_reserved)
+ return createStringError(errc::invalid_argument, "invalid length");
+
+ uint8_t Version = DA.getU16(&Offset);
+ (void)DA.getU16(&Offset); // padding
+ // The encoded length includes the 2-byte version field and the 2-byte
+ // padding, so we need to subtract them out when we populate the descriptor.
+ return StrOffsetsContributionDescriptor(Offset, ContributionSize - 4, Version,
+ DWARF32);
+}
+
+static Expected<StrOffsetsContributionDescriptor>
+parseDWARFStringOffsetsTableHeader(DWARFDataExtractor &DA,
+ llvm::dwarf::DwarfFormat Format,
+ uint64_t Offset) {
+ StrOffsetsContributionDescriptor Desc;
+ switch (Format) {
+ case dwarf::DwarfFormat::DWARF64: {
+ if (Offset < 16)
+ return createStringError(errc::invalid_argument, "insufficient space for 64 bit header prefix");
+ auto DescOrError = parseDWARF64StringOffsetsTableHeader(DA, Offset - 16);
+ if (!DescOrError)
+ return DescOrError.takeError();
+ Desc = *DescOrError;
+ break;
+ }
+ case dwarf::DwarfFormat::DWARF32: {
+ if (Offset < 8)
+ return createStringError(errc::invalid_argument, "insufficient space for 32 bit header prefix");
+ auto DescOrError = parseDWARF32StringOffsetsTableHeader(DA, Offset - 8);
+ if (!DescOrError)
+ return DescOrError.takeError();
+ Desc = *DescOrError;
+ break;
+ }
+ }
+ return Desc.validateContributionSize(DA);
+}
+
+Expected<Optional<StrOffsetsContributionDescriptor>>
+DWARFUnit::determineStringOffsetsTableContribution(DWARFDataExtractor &DA) {
+ uint64_t Offset;
+ if (IsDWO) {
+ Offset = 0;
+ if (DA.getData().data() == nullptr)
+ return None;
+ } else {
+ auto OptOffset = toSectionOffset(getUnitDIE().find(DW_AT_str_offsets_base));
+ if (!OptOffset)
+ return None;
+ Offset = *OptOffset;
+ }
+ auto DescOrError = parseDWARFStringOffsetsTableHeader(DA, Header.getFormat(), Offset);
+ if (!DescOrError)
+ return DescOrError.takeError();
+ return *DescOrError;
+}
+
+Expected<Optional<StrOffsetsContributionDescriptor>>
+DWARFUnit::determineStringOffsetsTableContributionDWO(DWARFDataExtractor & DA) {
+ uint64_t Offset = 0;
+ auto IndexEntry = Header.getIndexEntry();
+ const auto *C =
+ IndexEntry ? IndexEntry->getOffset(DW_SECT_STR_OFFSETS) : nullptr;
+ if (C)
+ Offset = C->Offset;
+ if (getVersion() >= 5) {
+ if (DA.getData().data() == nullptr)
+ return None;
+ Offset += Header.getFormat() == dwarf::DwarfFormat::DWARF32 ? 8 : 16;
+ // Look for a valid contribution at the given offset.
+ auto DescOrError = parseDWARFStringOffsetsTableHeader(DA, Header.getFormat(), Offset);
+ if (!DescOrError)
+ return DescOrError.takeError();
+ return *DescOrError;
+ }
+ // Prior to DWARF v5, we derive the contribution size from the
+ // index table (in a package file). In a .dwo file it is simply
+ // the length of the string offsets section.
+ if (!IndexEntry)
+ return {
+ Optional<StrOffsetsContributionDescriptor>(
+ {0, StringOffsetSection.Data.size(), 4, DWARF32})};
+ if (C)
+ return {Optional<StrOffsetsContributionDescriptor>(
+ {C->Offset, C->Length, 4, DWARF32})};
+ return None;
+}
diff --git a/third_party/llvm-project/DWARFUnitIndex.cpp b/third_party/llvm-project/DWARFUnitIndex.cpp
new file mode 100644
index 000000000..f29c1e6cc
--- /dev/null
+++ b/third_party/llvm-project/DWARFUnitIndex.cpp
@@ -0,0 +1,200 @@
+//===- DWARFUnitIndex.cpp -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cinttypes>
+#include <cstdint>
+
+using namespace llvm;
+
+bool DWARFUnitIndex::Header::parse(DataExtractor IndexData,
+ uint64_t *OffsetPtr) {
+ if (!IndexData.isValidOffsetForDataOfSize(*OffsetPtr, 16))
+ return false;
+ Version = IndexData.getU32(OffsetPtr);
+ NumColumns = IndexData.getU32(OffsetPtr);
+ NumUnits = IndexData.getU32(OffsetPtr);
+ NumBuckets = IndexData.getU32(OffsetPtr);
+ return Version <= 2;
+}
+
+void DWARFUnitIndex::Header::dump(raw_ostream &OS) const {
+ OS << format("version = %u slots = %u\n\n", Version, NumBuckets);
+}
+
+bool DWARFUnitIndex::parse(DataExtractor IndexData) {
+ bool b = parseImpl(IndexData);
+ if (!b) {
+ // Make sure we don't try to dump anything
+ Header.NumBuckets = 0;
+ // Release any partially initialized data.
+ ColumnKinds.reset();
+ Rows.reset();
+ }
+ return b;
+}
+
+bool DWARFUnitIndex::parseImpl(DataExtractor IndexData) {
+ uint64_t Offset = 0;
+ if (!Header.parse(IndexData, &Offset))
+ return false;
+
+ if (!IndexData.isValidOffsetForDataOfSize(
+ Offset, Header.NumBuckets * (8 + 4) +
+ (2 * Header.NumUnits + 1) * 4 * Header.NumColumns))
+ return false;
+
+ Rows = std::make_unique<Entry[]>(Header.NumBuckets);
+ auto Contribs =
+ std::make_unique<Entry::SectionContribution *[]>(Header.NumUnits);
+ ColumnKinds = std::make_unique<DWARFSectionKind[]>(Header.NumColumns);
+
+ // Read Hash Table of Signatures
+ for (unsigned i = 0; i != Header.NumBuckets; ++i)
+ Rows[i].Signature = IndexData.getU64(&Offset);
+
+ // Read Parallel Table of Indexes
+ for (unsigned i = 0; i != Header.NumBuckets; ++i) {
+ auto Index = IndexData.getU32(&Offset);
+ if (!Index)
+ continue;
+ Rows[i].Index = this;
+ Rows[i].Contributions =
+ std::make_unique<Entry::SectionContribution[]>(Header.NumColumns);
+ Contribs[Index - 1] = Rows[i].Contributions.get();
+ }
+
+ // Read the Column Headers
+ for (unsigned i = 0; i != Header.NumColumns; ++i) {
+ ColumnKinds[i] = static_cast<DWARFSectionKind>(IndexData.getU32(&Offset));
+ if (ColumnKinds[i] == InfoColumnKind) {
+ if (InfoColumn != -1)
+ return false;
+ InfoColumn = i;
+ }
+ }
+
+ if (InfoColumn == -1)
+ return false;
+
+ // Read Table of Section Offsets
+ for (unsigned i = 0; i != Header.NumUnits; ++i) {
+ auto *Contrib = Contribs[i];
+ for (unsigned i = 0; i != Header.NumColumns; ++i)
+ Contrib[i].Offset = IndexData.getU32(&Offset);
+ }
+
+ // Read Table of Section Sizes
+ for (unsigned i = 0; i != Header.NumUnits; ++i) {
+ auto *Contrib = Contribs[i];
+ for (unsigned i = 0; i != Header.NumColumns; ++i)
+ Contrib[i].Length = IndexData.getU32(&Offset);
+ }
+
+ return true;
+}
+
+StringRef DWARFUnitIndex::getColumnHeader(DWARFSectionKind DS) {
+#define CASE(DS) \
+ case DW_SECT_##DS: \
+ return #DS;
+ switch (DS) {
+ CASE(INFO);
+ CASE(TYPES);
+ CASE(ABBREV);
+ CASE(LINE);
+ CASE(LOC);
+ CASE(STR_OFFSETS);
+ CASE(MACINFO);
+ CASE(MACRO);
+ }
+ llvm_unreachable("unknown DWARFSectionKind");
+}
+
+void DWARFUnitIndex::dump(raw_ostream &OS) const {
+ if (!*this)
+ return;
+
+ Header.dump(OS);
+ OS << "Index Signature ";
+ for (unsigned i = 0; i != Header.NumColumns; ++i)
+ OS << ' ' << left_justify(getColumnHeader(ColumnKinds[i]), 24);
+ OS << "\n----- ------------------";
+ for (unsigned i = 0; i != Header.NumColumns; ++i)
+ OS << " ------------------------";
+ OS << '\n';
+ for (unsigned i = 0; i != Header.NumBuckets; ++i) {
+ auto &Row = Rows[i];
+ if (auto *Contribs = Row.Contributions.get()) {
+ OS << format("%5u 0x%016" PRIx64 " ", i + 1, Row.Signature);
+ for (unsigned i = 0; i != Header.NumColumns; ++i) {
+ auto &Contrib = Contribs[i];
+ OS << format("[0x%08x, 0x%08x) ", Contrib.Offset,
+ Contrib.Offset + Contrib.Length);
+ }
+ OS << '\n';
+ }
+ }
+}
+
+const DWARFUnitIndex::Entry::SectionContribution *
+DWARFUnitIndex::Entry::getOffset(DWARFSectionKind Sec) const {
+ uint32_t i = 0;
+ for (; i != Index->Header.NumColumns; ++i)
+ if (Index->ColumnKinds[i] == Sec)
+ return &Contributions[i];
+ return nullptr;
+}
+
+const DWARFUnitIndex::Entry::SectionContribution *
+DWARFUnitIndex::Entry::getOffset() const {
+ return &Contributions[Index->InfoColumn];
+}
+
+const DWARFUnitIndex::Entry *
+DWARFUnitIndex::getFromOffset(uint32_t Offset) const {
+ if (OffsetLookup.empty()) {
+ for (uint32_t i = 0; i != Header.NumBuckets; ++i)
+ if (Rows[i].Contributions)
+ OffsetLookup.push_back(&Rows[i]);
+ llvm::sort(OffsetLookup, [&](Entry *E1, Entry *E2) {
+ return E1->Contributions[InfoColumn].Offset <
+ E2->Contributions[InfoColumn].Offset;
+ });
+ }
+ auto I = partition_point(OffsetLookup, [&](Entry *E2) {
+ return E2->Contributions[InfoColumn].Offset <= Offset;
+ });
+ if (I == OffsetLookup.begin())
+ return nullptr;
+ --I;
+ const auto *E = *I;
+ const auto &InfoContrib = E->Contributions[InfoColumn];
+ if ((InfoContrib.Offset + InfoContrib.Length) <= Offset)
+ return nullptr;
+ return E;
+}
+
+const DWARFUnitIndex::Entry *DWARFUnitIndex::getFromHash(uint64_t S) const {
+ uint64_t Mask = Header.NumBuckets - 1;
+
+ auto H = S & Mask;
+ auto HP = ((S >> 32) & Mask) | 1;
+ while (Rows[H].getSignature() != S && Rows[H].getSignature() != 0)
+ H = (H + HP) & Mask;
+
+ if (Rows[H].getSignature() != S)
+ return nullptr;
+
+ return &Rows[H];
+}
diff --git a/third_party/llvm-project/DWARFVerifier.cpp b/third_party/llvm-project/DWARFVerifier.cpp
new file mode 100644
index 000000000..bf499b6ee
--- /dev/null
+++ b/third_party/llvm-project/DWARFVerifier.cpp
@@ -0,0 +1,1493 @@
+//===- DWARFVerifier.cpp --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+#include "llvm/DebugInfo/DWARF/DWARFVerifier.h"
+#include "llvm/ADT/SmallSet.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/Support/DJB.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <map>
+#include <set>
+#include <vector>
+
+using namespace llvm;
+using namespace dwarf;
+using namespace object;
+
+DWARFVerifier::DieRangeInfo::address_range_iterator
+DWARFVerifier::DieRangeInfo::insert(const DWARFAddressRange &R) {
+ auto Begin = Ranges.begin();
+ auto End = Ranges.end();
+ auto Pos = std::lower_bound(Begin, End, R);
+
+ if (Pos != End) {
+ if (Pos->intersects(R))
+ return std::move(Pos);
+ if (Pos != Begin) {
+ auto Iter = Pos - 1;
+ if (Iter->intersects(R))
+ return std::move(Iter);
+ }
+ }
+
+ Ranges.insert(Pos, R);
+ return Ranges.end();
+}
+
+DWARFVerifier::DieRangeInfo::die_range_info_iterator
+DWARFVerifier::DieRangeInfo::insert(const DieRangeInfo &RI) {
+ auto End = Children.end();
+ auto Iter = Children.begin();
+ while (Iter != End) {
+ if (Iter->intersects(RI))
+ return Iter;
+ ++Iter;
+ }
+ Children.insert(RI);
+ return Children.end();
+}
+
+bool DWARFVerifier::DieRangeInfo::contains(const DieRangeInfo &RHS) const {
+ auto I1 = Ranges.begin(), E1 = Ranges.end();
+ auto I2 = RHS.Ranges.begin(), E2 = RHS.Ranges.end();
+ if (I2 == E2)
+ return true;
+
+ DWARFAddressRange R = *I2;
+ while (I1 != E1) {
+ bool Covered = I1->LowPC <= R.LowPC;
+ if (R.LowPC == R.HighPC || (Covered && R.HighPC <= I1->HighPC)) {
+ if (++I2 == E2)
+ return true;
+ R = *I2;
+ continue;
+ }
+ if (!Covered)
+ return false;
+ if (R.LowPC < I1->HighPC)
+ R.LowPC = I1->HighPC;
+ ++I1;
+ }
+ return false;
+}
+
+bool DWARFVerifier::DieRangeInfo::intersects(const DieRangeInfo &RHS) const {
+ auto I1 = Ranges.begin(), E1 = Ranges.end();
+ auto I2 = RHS.Ranges.begin(), E2 = RHS.Ranges.end();
+ while (I1 != E1 && I2 != E2) {
+ if (I1->intersects(*I2))
+ return true;
+ if (I1->LowPC < I2->LowPC)
+ ++I1;
+ else
+ ++I2;
+ }
+ return false;
+}
+
+bool DWARFVerifier::verifyUnitHeader(const DWARFDataExtractor DebugInfoData,
+ uint64_t *Offset, unsigned UnitIndex,
+ uint8_t &UnitType, bool &isUnitDWARF64) {
+ uint64_t AbbrOffset, Length;
+ uint8_t AddrSize = 0;
+ uint16_t Version;
+ bool Success = true;
+
+ bool ValidLength = false;
+ bool ValidVersion = false;
+ bool ValidAddrSize = false;
+ bool ValidType = true;
+ bool ValidAbbrevOffset = true;
+
+ uint64_t OffsetStart = *Offset;
+ Length = DebugInfoData.getU32(Offset);
+ if (Length == dwarf::DW_LENGTH_DWARF64) {
+ Length = DebugInfoData.getU64(Offset);
+ isUnitDWARF64 = true;
+ }
+ Version = DebugInfoData.getU16(Offset);
+
+ if (Version >= 5) {
+ UnitType = DebugInfoData.getU8(Offset);
+ AddrSize = DebugInfoData.getU8(Offset);
+ AbbrOffset = isUnitDWARF64 ? DebugInfoData.getU64(Offset) : DebugInfoData.getU32(Offset);
+ ValidType = dwarf::isUnitType(UnitType);
+ } else {
+ UnitType = 0;
+ AbbrOffset = isUnitDWARF64 ? DebugInfoData.getU64(Offset) : DebugInfoData.getU32(Offset);
+ AddrSize = DebugInfoData.getU8(Offset);
+ }
+
+ if (!DCtx.getDebugAbbrev()->getAbbreviationDeclarationSet(AbbrOffset))
+ ValidAbbrevOffset = false;
+
+ ValidLength = DebugInfoData.isValidOffset(OffsetStart + Length + 3);
+ ValidVersion = DWARFContext::isSupportedVersion(Version);
+ ValidAddrSize = AddrSize == 4 || AddrSize == 8;
+ if (!ValidLength || !ValidVersion || !ValidAddrSize || !ValidAbbrevOffset ||
+ !ValidType) {
+ Success = false;
+ error() << format("Units[%d] - start offset: 0x%08" PRIx64 " \n", UnitIndex,
+ OffsetStart);
+ if (!ValidLength)
+ note() << "The length for this unit is too "
+ "large for the .debug_info provided.\n";
+ if (!ValidVersion)
+ note() << "The 16 bit unit header version is not valid.\n";
+ if (!ValidType)
+ note() << "The unit type encoding is not valid.\n";
+ if (!ValidAbbrevOffset)
+ note() << "The offset into the .debug_abbrev section is "
+ "not valid.\n";
+ if (!ValidAddrSize)
+ note() << "The address size is unsupported.\n";
+ }
+ *Offset = OffsetStart + Length + (isUnitDWARF64 ? 12 : 4);
+ return Success;
+}
+
+unsigned DWARFVerifier::verifyUnitContents(DWARFUnit &Unit) {
+ unsigned NumUnitErrors = 0;
+ unsigned NumDies = Unit.getNumDIEs();
+ for (unsigned I = 0; I < NumDies; ++I) {
+ auto Die = Unit.getDIEAtIndex(I);
+
+ if (Die.getTag() == DW_TAG_null)
+ continue;
+
+ for (auto AttrValue : Die.attributes()) {
+ NumUnitErrors += verifyDebugInfoAttribute(Die, AttrValue);
+ NumUnitErrors += verifyDebugInfoForm(Die, AttrValue);
+ }
+
+ NumUnitErrors += verifyDebugInfoCallSite(Die);
+ }
+
+ DWARFDie Die = Unit.getUnitDIE(/* ExtractUnitDIEOnly = */ false);
+ if (!Die) {
+ error() << "Compilation unit without DIE.\n";
+ NumUnitErrors++;
+ return NumUnitErrors;
+ }
+
+ if (!dwarf::isUnitType(Die.getTag())) {
+ error() << "Compilation unit root DIE is not a unit DIE: "
+ << dwarf::TagString(Die.getTag()) << ".\n";
+ NumUnitErrors++;
+ }
+
+ uint8_t UnitType = Unit.getUnitType();
+ if (!DWARFUnit::isMatchingUnitTypeAndTag(UnitType, Die.getTag())) {
+ error() << "Compilation unit type (" << dwarf::UnitTypeString(UnitType)
+ << ") and root DIE (" << dwarf::TagString(Die.getTag())
+ << ") do not match.\n";
+ NumUnitErrors++;
+ }
+
+ DieRangeInfo RI;
+ NumUnitErrors += verifyDieRanges(Die, RI);
+
+ return NumUnitErrors;
+}
+
+unsigned DWARFVerifier::verifyDebugInfoCallSite(const DWARFDie &Die) {
+ if (Die.getTag() != DW_TAG_call_site && Die.getTag() != DW_TAG_GNU_call_site)
+ return 0;
+
+ DWARFDie Curr = Die.getParent();
+ for (; Curr.isValid() && !Curr.isSubprogramDIE(); Curr = Die.getParent()) {
+ if (Curr.getTag() == DW_TAG_inlined_subroutine) {
+ error() << "Call site entry nested within inlined subroutine:";
+ Curr.dump(OS);
+ return 1;
+ }
+ }
+
+ if (!Curr.isValid()) {
+ error() << "Call site entry not nested within a valid subprogram:";
+ Die.dump(OS);
+ return 1;
+ }
+
+ Optional<DWARFFormValue> CallAttr =
+ Curr.find({DW_AT_call_all_calls, DW_AT_call_all_source_calls,
+ DW_AT_call_all_tail_calls, DW_AT_GNU_all_call_sites,
+ DW_AT_GNU_all_source_call_sites,
+ DW_AT_GNU_all_tail_call_sites});
+ if (!CallAttr) {
+ error() << "Subprogram with call site entry has no DW_AT_call attribute:";
+ Curr.dump(OS);
+ Die.dump(OS, /*indent*/ 1);
+ return 1;
+ }
+
+ return 0;
+}
+
+unsigned DWARFVerifier::verifyAbbrevSection(const DWARFDebugAbbrev *Abbrev) {
+ unsigned NumErrors = 0;
+ if (Abbrev) {
+ const DWARFAbbreviationDeclarationSet *AbbrDecls =
+ Abbrev->getAbbreviationDeclarationSet(0);
+ for (auto AbbrDecl : *AbbrDecls) {
+ SmallDenseSet<uint16_t> AttributeSet;
+ for (auto Attribute : AbbrDecl.attributes()) {
+ auto Result = AttributeSet.insert(Attribute.Attr);
+ if (!Result.second) {
+ error() << "Abbreviation declaration contains multiple "
+ << AttributeString(Attribute.Attr) << " attributes.\n";
+ AbbrDecl.dump(OS);
+ ++NumErrors;
+ }
+ }
+ }
+ }
+ return NumErrors;
+}
+
+bool DWARFVerifier::handleDebugAbbrev() {
+ OS << "Verifying .debug_abbrev...\n";
+
+ const DWARFObject &DObj = DCtx.getDWARFObj();
+ unsigned NumErrors = 0;
+ if (!DObj.getAbbrevSection().empty())
+ NumErrors += verifyAbbrevSection(DCtx.getDebugAbbrev());
+ if (!DObj.getAbbrevDWOSection().empty())
+ NumErrors += verifyAbbrevSection(DCtx.getDebugAbbrevDWO());
+
+ return NumErrors == 0;
+}
+
+unsigned DWARFVerifier::verifyUnitSection(const DWARFSection &S,
+ DWARFSectionKind SectionKind) {
+ const DWARFObject &DObj = DCtx.getDWARFObj();
+ DWARFDataExtractor DebugInfoData(DObj, S, DCtx.isLittleEndian(), 0);
+ unsigned NumDebugInfoErrors = 0;
+ uint64_t OffsetStart = 0, Offset = 0, UnitIdx = 0;
+ uint8_t UnitType = 0;
+ bool isUnitDWARF64 = false;
+ bool isHeaderChainValid = true;
+ bool hasDIE = DebugInfoData.isValidOffset(Offset);
+ DWARFUnitVector TypeUnitVector;
+ DWARFUnitVector CompileUnitVector;
+ while (hasDIE) {
+ OffsetStart = Offset;
+ if (!verifyUnitHeader(DebugInfoData, &Offset, UnitIdx, UnitType,
+ isUnitDWARF64)) {
+ isHeaderChainValid = false;
+ if (isUnitDWARF64)
+ break;
+ } else {
+ DWARFUnitHeader Header;
+ Header.extract(DCtx, DebugInfoData, &OffsetStart, SectionKind);
+ DWARFUnit *Unit;
+ switch (UnitType) {
+ case dwarf::DW_UT_type:
+ case dwarf::DW_UT_split_type: {
+ Unit = TypeUnitVector.addUnit(std::make_unique<DWARFTypeUnit>(
+ DCtx, S, Header, DCtx.getDebugAbbrev(), &DObj.getRangesSection(),
+ &DObj.getLocSection(), DObj.getStrSection(),
+ DObj.getStrOffsetsSection(), &DObj.getAppleObjCSection(),
+ DObj.getLineSection(), DCtx.isLittleEndian(), false,
+ TypeUnitVector));
+ break;
+ }
+ case dwarf::DW_UT_skeleton:
+ case dwarf::DW_UT_split_compile:
+ case dwarf::DW_UT_compile:
+ case dwarf::DW_UT_partial:
+ // UnitType = 0 means that we are verifying a compile unit in DWARF v4.
+ case 0: {
+ Unit = CompileUnitVector.addUnit(std::make_unique<DWARFCompileUnit>(
+ DCtx, S, Header, DCtx.getDebugAbbrev(), &DObj.getRangesSection(),
+ &DObj.getLocSection(), DObj.getStrSection(),
+ DObj.getStrOffsetsSection(), &DObj.getAppleObjCSection(),
+ DObj.getLineSection(), DCtx.isLittleEndian(), false,
+ CompileUnitVector));
+ break;
+ }
+ default: { llvm_unreachable("Invalid UnitType."); }
+ }
+ NumDebugInfoErrors += verifyUnitContents(*Unit);
+ }
+ hasDIE = DebugInfoData.isValidOffset(Offset);
+ ++UnitIdx;
+ }
+ if (UnitIdx == 0 && !hasDIE) {
+ warn() << "Section is empty.\n";
+ isHeaderChainValid = true;
+ }
+ if (!isHeaderChainValid)
+ ++NumDebugInfoErrors;
+ NumDebugInfoErrors += verifyDebugInfoReferences();
+ return NumDebugInfoErrors;
+}
+
+bool DWARFVerifier::handleDebugInfo() {
+ const DWARFObject &DObj = DCtx.getDWARFObj();
+ unsigned NumErrors = 0;
+
+ OS << "Verifying .debug_info Unit Header Chain...\n";
+ DObj.forEachInfoSections([&](const DWARFSection &S) {
+ NumErrors += verifyUnitSection(S, DW_SECT_INFO);
+ });
+
+ OS << "Verifying .debug_types Unit Header Chain...\n";
+ DObj.forEachTypesSections([&](const DWARFSection &S) {
+ NumErrors += verifyUnitSection(S, DW_SECT_TYPES);
+ });
+ return NumErrors == 0;
+}
+
+unsigned DWARFVerifier::verifyDieRanges(const DWARFDie &Die,
+ DieRangeInfo &ParentRI) {
+ unsigned NumErrors = 0;
+
+ if (!Die.isValid())
+ return NumErrors;
+
+ auto RangesOrError = Die.getAddressRanges();
+ if (!RangesOrError) {
+ // FIXME: Report the error.
+ ++NumErrors;
+ llvm::consumeError(RangesOrError.takeError());
+ return NumErrors;
+ }
+
+ DWARFAddressRangesVector Ranges = RangesOrError.get();
+ // Build RI for this DIE and check that ranges within this DIE do not
+ // overlap.
+ DieRangeInfo RI(Die);
+
+ // TODO support object files better
+ //
+ // Some object file formats (i.e. non-MachO) support COMDAT. ELF in
+ // particular does so by placing each function into a section. The DWARF data
+ // for the function at that point uses a section relative DW_FORM_addrp for
+ // the DW_AT_low_pc and a DW_FORM_data4 for the offset as the DW_AT_high_pc.
+ // In such a case, when the Die is the CU, the ranges will overlap, and we
+ // will flag valid conflicting ranges as invalid.
+ //
+ // For such targets, we should read the ranges from the CU and partition them
+ // by the section id. The ranges within a particular section should be
+ // disjoint, although the ranges across sections may overlap. We would map
+ // the child die to the entity that it references and the section with which
+ // it is associated. The child would then be checked against the range
+ // information for the associated section.
+ //
+ // For now, simply elide the range verification for the CU DIEs if we are
+ // processing an object file.
+
+ if (!IsObjectFile || IsMachOObject || Die.getTag() != DW_TAG_compile_unit) {
+ for (auto Range : Ranges) {
+ if (!Range.valid()) {
+ ++NumErrors;
+ error() << "Invalid address range " << Range << "\n";
+ continue;
+ }
+
+ // Verify that ranges don't intersect.
+ const auto IntersectingRange = RI.insert(Range);
+ if (IntersectingRange != RI.Ranges.end()) {
+ ++NumErrors;
+ error() << "DIE has overlapping address ranges: " << Range << " and "
+ << *IntersectingRange << "\n";
+ break;
+ }
+ }
+ }
+
+ // Verify that children don't intersect.
+ const auto IntersectingChild = ParentRI.insert(RI);
+ if (IntersectingChild != ParentRI.Children.end()) {
+ ++NumErrors;
+ error() << "DIEs have overlapping address ranges:";
+ dump(Die);
+ dump(IntersectingChild->Die) << '\n';
+ }
+
+ // Verify that ranges are contained within their parent.
+ bool ShouldBeContained = !Ranges.empty() && !ParentRI.Ranges.empty() &&
+ !(Die.getTag() == DW_TAG_subprogram &&
+ ParentRI.Die.getTag() == DW_TAG_subprogram);
+ if (ShouldBeContained && !ParentRI.contains(RI)) {
+ ++NumErrors;
+ error() << "DIE address ranges are not contained in its parent's ranges:";
+ dump(ParentRI.Die);
+ dump(Die, 2) << '\n';
+ }
+
+ // Recursively check children.
+ for (DWARFDie Child : Die)
+ NumErrors += verifyDieRanges(Child, RI);
+
+ return NumErrors;
+}
+
+unsigned DWARFVerifier::verifyDebugInfoAttribute(const DWARFDie &Die,
+ DWARFAttribute &AttrValue) {
+ unsigned NumErrors = 0;
+ auto ReportError = [&](const Twine &TitleMsg) {
+ ++NumErrors;
+ error() << TitleMsg << '\n';
+ dump(Die) << '\n';
+ };
+
+ const DWARFObject &DObj = DCtx.getDWARFObj();
+ const auto Attr = AttrValue.Attr;
+ switch (Attr) {
+ case DW_AT_ranges:
+ // Make sure the offset in the DW_AT_ranges attribute is valid.
+ if (auto SectionOffset = AttrValue.Value.getAsSectionOffset()) {
+ if (*SectionOffset >= DObj.getRangesSection().Data.size())
+ ReportError("DW_AT_ranges offset is beyond .debug_ranges bounds:");
+ break;
+ }
+ ReportError("DIE has invalid DW_AT_ranges encoding:");
+ break;
+ case DW_AT_stmt_list:
+ // Make sure the offset in the DW_AT_stmt_list attribute is valid.
+ if (auto SectionOffset = AttrValue.Value.getAsSectionOffset()) {
+ if (*SectionOffset >= DObj.getLineSection().Data.size())
+ ReportError("DW_AT_stmt_list offset is beyond .debug_line bounds: " +
+ llvm::formatv("{0:x8}", *SectionOffset));
+ break;
+ }
+ ReportError("DIE has invalid DW_AT_stmt_list encoding:");
+ break;
+ case DW_AT_location: {
+ auto VerifyLocationExpr = [&](ArrayRef<uint8_t> D) {
+ DWARFUnit *U = Die.getDwarfUnit();
+ DataExtractor Data(toStringRef(D), DCtx.isLittleEndian(), 0);
+ DWARFExpression Expression(Data, U->getVersion(),
+ U->getAddressByteSize());
+ bool Error = llvm::any_of(Expression, [](DWARFExpression::Operation &Op) {
+ return Op.isError();
+ });
+ if (Error || !Expression.verify(U))
+ ReportError("DIE contains invalid DWARF expression:");
+ };
+ if (Optional<ArrayRef<uint8_t>> Expr = AttrValue.Value.getAsBlock()) {
+ // Verify inlined location.
+ VerifyLocationExpr(*Expr);
+ } else if (auto LocOffset = AttrValue.Value.getAsSectionOffset()) {
+ // Verify location list.
+ if (auto DebugLoc = DCtx.getDebugLoc())
+ if (auto LocList = DebugLoc->getLocationListAtOffset(*LocOffset))
+ for (const auto &Entry : LocList->Entries)
+ VerifyLocationExpr(Entry.Loc);
+ }
+ break;
+ }
+ case DW_AT_specification:
+ case DW_AT_abstract_origin: {
+ if (auto ReferencedDie = Die.getAttributeValueAsReferencedDie(Attr)) {
+ auto DieTag = Die.getTag();
+ auto RefTag = ReferencedDie.getTag();
+ if (DieTag == RefTag)
+ break;
+ if (DieTag == DW_TAG_inlined_subroutine && RefTag == DW_TAG_subprogram)
+ break;
+ if (DieTag == DW_TAG_variable && RefTag == DW_TAG_member)
+ break;
+ // This might be reference to a function declaration.
+ if (DieTag == DW_TAG_GNU_call_site && RefTag == DW_TAG_subprogram)
+ break;
+ ReportError("DIE with tag " + TagString(DieTag) + " has " +
+ AttributeString(Attr) +
+ " that points to DIE with "
+ "incompatible tag " +
+ TagString(RefTag));
+ }
+ break;
+ }
+ case DW_AT_type: {
+ DWARFDie TypeDie = Die.getAttributeValueAsReferencedDie(DW_AT_type);
+ if (TypeDie && !isType(TypeDie.getTag())) {
+ ReportError("DIE has " + AttributeString(Attr) +
+ " with incompatible tag " + TagString(TypeDie.getTag()));
+ }
+ break;
+ }
+ default:
+ break;
+ }
+ return NumErrors;
+}
+
+unsigned DWARFVerifier::verifyDebugInfoForm(const DWARFDie &Die,
+ DWARFAttribute &AttrValue) {
+ const DWARFObject &DObj = DCtx.getDWARFObj();
+ auto DieCU = Die.getDwarfUnit();
+ unsigned NumErrors = 0;
+ const auto Form = AttrValue.Value.getForm();
+ switch (Form) {
+ case DW_FORM_ref1:
+ case DW_FORM_ref2:
+ case DW_FORM_ref4:
+ case DW_FORM_ref8:
+ case DW_FORM_ref_udata: {
+ // Verify all CU relative references are valid CU offsets.
+ Optional<uint64_t> RefVal = AttrValue.Value.getAsReference();
+ assert(RefVal);
+ if (RefVal) {
+ auto CUSize = DieCU->getNextUnitOffset() - DieCU->getOffset();
+ auto CUOffset = AttrValue.Value.getRawUValue();
+ if (CUOffset >= CUSize) {
+ ++NumErrors;
+ error() << FormEncodingString(Form) << " CU offset "
+ << format("0x%08" PRIx64, CUOffset)
+ << " is invalid (must be less than CU size of "
+ << format("0x%08" PRIx64, CUSize) << "):\n";
+ Die.dump(OS, 0, DumpOpts);
+ dump(Die) << '\n';
+ } else {
+ // Valid reference, but we will verify it points to an actual
+ // DIE later.
+ ReferenceToDIEOffsets[*RefVal].insert(Die.getOffset());
+ }
+ }
+ break;
+ }
+ case DW_FORM_ref_addr: {
+ // Verify all absolute DIE references have valid offsets in the
+ // .debug_info section.
+ Optional<uint64_t> RefVal = AttrValue.Value.getAsReference();
+ assert(RefVal);
+ if (RefVal) {
+ if (*RefVal >= DieCU->getInfoSection().Data.size()) {
+ ++NumErrors;
+ error() << "DW_FORM_ref_addr offset beyond .debug_info "
+ "bounds:\n";
+ dump(Die) << '\n';
+ } else {
+ // Valid reference, but we will verify it points to an actual
+ // DIE later.
+ ReferenceToDIEOffsets[*RefVal].insert(Die.getOffset());
+ }
+ }
+ break;
+ }
+ case DW_FORM_strp: {
+ auto SecOffset = AttrValue.Value.getAsSectionOffset();
+ assert(SecOffset); // DW_FORM_strp is a section offset.
+ if (SecOffset && *SecOffset >= DObj.getStrSection().size()) {
+ ++NumErrors;
+ error() << "DW_FORM_strp offset beyond .debug_str bounds:\n";
+ dump(Die) << '\n';
+ }
+ break;
+ }
+ case DW_FORM_strx:
+ case DW_FORM_strx1:
+ case DW_FORM_strx2:
+ case DW_FORM_strx3:
+ case DW_FORM_strx4: {
+ auto Index = AttrValue.Value.getRawUValue();
+ auto DieCU = Die.getDwarfUnit();
+ // Check that we have a valid DWARF v5 string offsets table.
+ if (!DieCU->getStringOffsetsTableContribution()) {
+ ++NumErrors;
+ error() << FormEncodingString(Form)
+ << " used without a valid string offsets table:\n";
+ dump(Die) << '\n';
+ break;
+ }
+ // Check that the index is within the bounds of the section.
+ unsigned ItemSize = DieCU->getDwarfStringOffsetsByteSize();
+ // Use a 64-bit type to calculate the offset to guard against overflow.
+ uint64_t Offset =
+ (uint64_t)DieCU->getStringOffsetsBase() + Index * ItemSize;
+ if (DObj.getStrOffsetsSection().Data.size() < Offset + ItemSize) {
+ ++NumErrors;
+ error() << FormEncodingString(Form) << " uses index "
+ << format("%" PRIu64, Index) << ", which is too large:\n";
+ dump(Die) << '\n';
+ break;
+ }
+ // Check that the string offset is valid.
+ uint64_t StringOffset = *DieCU->getStringOffsetSectionItem(Index);
+ if (StringOffset >= DObj.getStrSection().size()) {
+ ++NumErrors;
+ error() << FormEncodingString(Form) << " uses index "
+ << format("%" PRIu64, Index)
+ << ", but the referenced string"
+ " offset is beyond .debug_str bounds:\n";
+ dump(Die) << '\n';
+ }
+ break;
+ }
+ default:
+ break;
+ }
+ return NumErrors;
+}
+
+unsigned DWARFVerifier::verifyDebugInfoReferences() {
+ // Take all references and make sure they point to an actual DIE by
+ // getting the DIE by offset and emitting an error
+ OS << "Verifying .debug_info references...\n";
+ unsigned NumErrors = 0;
+ for (const std::pair<uint64_t, std::set<uint64_t>> &Pair :
+ ReferenceToDIEOffsets) {
+ if (DCtx.getDIEForOffset(Pair.first))
+ continue;
+ ++NumErrors;
+ error() << "invalid DIE reference " << format("0x%08" PRIx64, Pair.first)
+ << ". Offset is in between DIEs:\n";
+ for (auto Offset : Pair.second)
+ dump(DCtx.getDIEForOffset(Offset)) << '\n';
+ OS << "\n";
+ }
+ return NumErrors;
+}
+
+void DWARFVerifier::verifyDebugLineStmtOffsets() {
+ std::map<uint64_t, DWARFDie> StmtListToDie;
+ for (const auto &CU : DCtx.compile_units()) {
+ auto Die = CU->getUnitDIE();
+ // Get the attribute value as a section offset. No need to produce an
+ // error here if the encoding isn't correct because we validate this in
+ // the .debug_info verifier.
+ auto StmtSectionOffset = toSectionOffset(Die.find(DW_AT_stmt_list));
+ if (!StmtSectionOffset)
+ continue;
+ const uint64_t LineTableOffset = *StmtSectionOffset;
+ auto LineTable = DCtx.getLineTableForUnit(CU.get());
+ if (LineTableOffset < DCtx.getDWARFObj().getLineSection().Data.size()) {
+ if (!LineTable) {
+ ++NumDebugLineErrors;
+ error() << ".debug_line[" << format("0x%08" PRIx64, LineTableOffset)
+ << "] was not able to be parsed for CU:\n";
+ dump(Die) << '\n';
+ continue;
+ }
+ } else {
+ // Make sure we don't get a valid line table back if the offset is wrong.
+ assert(LineTable == nullptr);
+ // Skip this line table as it isn't valid. No need to create an error
+ // here because we validate this in the .debug_info verifier.
+ continue;
+ }
+ auto Iter = StmtListToDie.find(LineTableOffset);
+ if (Iter != StmtListToDie.end()) {
+ ++NumDebugLineErrors;
+ error() << "two compile unit DIEs, "
+ << format("0x%08" PRIx64, Iter->second.getOffset()) << " and "
+ << format("0x%08" PRIx64, Die.getOffset())
+ << ", have the same DW_AT_stmt_list section offset:\n";
+ dump(Iter->second);
+ dump(Die) << '\n';
+ // Already verified this line table before, no need to do it again.
+ continue;
+ }
+ StmtListToDie[LineTableOffset] = Die;
+ }
+}
+
+void DWARFVerifier::verifyDebugLineRows() {
+ for (const auto &CU : DCtx.compile_units()) {
+ auto Die = CU->getUnitDIE();
+ auto LineTable = DCtx.getLineTableForUnit(CU.get());
+ // If there is no line table we will have created an error in the
+ // .debug_info verifier or in verifyDebugLineStmtOffsets().
+ if (!LineTable)
+ continue;
+
+ // Verify prologue.
+ uint32_t MaxDirIndex = LineTable->Prologue.IncludeDirectories.size();
+ uint32_t FileIndex = 1;
+ StringMap<uint16_t> FullPathMap;
+ for (const auto &FileName : LineTable->Prologue.FileNames) {
+ // Verify directory index.
+ if (FileName.DirIdx > MaxDirIndex) {
+ ++NumDebugLineErrors;
+ error() << ".debug_line["
+ << format("0x%08" PRIx64,
+ *toSectionOffset(Die.find(DW_AT_stmt_list)))
+ << "].prologue.file_names[" << FileIndex
+ << "].dir_idx contains an invalid index: " << FileName.DirIdx
+ << "\n";
+ }
+
+ // Check file paths for duplicates.
+ std::string FullPath;
+ const bool HasFullPath = LineTable->getFileNameByIndex(
+ FileIndex, CU->getCompilationDir(),
+ DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, FullPath);
+ assert(HasFullPath && "Invalid index?");
+ (void)HasFullPath;
+ auto It = FullPathMap.find(FullPath);
+ if (It == FullPathMap.end())
+ FullPathMap[FullPath] = FileIndex;
+ else if (It->second != FileIndex) {
+ warn() << ".debug_line["
+ << format("0x%08" PRIx64,
+ *toSectionOffset(Die.find(DW_AT_stmt_list)))
+ << "].prologue.file_names[" << FileIndex
+ << "] is a duplicate of file_names[" << It->second << "]\n";
+ }
+
+ FileIndex++;
+ }
+
+ // Verify rows.
+ uint64_t PrevAddress = 0;
+ uint32_t RowIndex = 0;
+ for (const auto &Row : LineTable->Rows) {
+ // Verify row address.
+ if (Row.Address.Address < PrevAddress) {
+ ++NumDebugLineErrors;
+ error() << ".debug_line["
+ << format("0x%08" PRIx64,
+ *toSectionOffset(Die.find(DW_AT_stmt_list)))
+ << "] row[" << RowIndex
+ << "] decreases in address from previous row:\n";
+
+ DWARFDebugLine::Row::dumpTableHeader(OS);
+ if (RowIndex > 0)
+ LineTable->Rows[RowIndex - 1].dump(OS);
+ Row.dump(OS);
+ OS << '\n';
+ }
+
+ // Verify file index.
+ if (!LineTable->hasFileAtIndex(Row.File)) {
+ ++NumDebugLineErrors;
+ bool isDWARF5 = LineTable->Prologue.getVersion() >= 5;
+ error() << ".debug_line["
+ << format("0x%08" PRIx64,
+ *toSectionOffset(Die.find(DW_AT_stmt_list)))
+ << "][" << RowIndex << "] has invalid file index " << Row.File
+ << " (valid values are [" << (isDWARF5 ? "0," : "1,")
+ << LineTable->Prologue.FileNames.size()
+ << (isDWARF5 ? ")" : "]") << "):\n";
+ DWARFDebugLine::Row::dumpTableHeader(OS);
+ Row.dump(OS);
+ OS << '\n';
+ }
+ if (Row.EndSequence)
+ PrevAddress = 0;
+ else
+ PrevAddress = Row.Address.Address;
+ ++RowIndex;
+ }
+ }
+}
+
+DWARFVerifier::DWARFVerifier(raw_ostream &S, DWARFContext &D,
+ DIDumpOptions DumpOpts)
+ : OS(S), DCtx(D), DumpOpts(std::move(DumpOpts)), IsObjectFile(false),
+ IsMachOObject(false) {
+ if (const auto *F = DCtx.getDWARFObj().getFile()) {
+ IsObjectFile = F->isRelocatableObject();
+ IsMachOObject = F->isMachO();
+ }
+}
+
+bool DWARFVerifier::handleDebugLine() {
+ NumDebugLineErrors = 0;
+ OS << "Verifying .debug_line...\n";
+ verifyDebugLineStmtOffsets();
+ verifyDebugLineRows();
+ return NumDebugLineErrors == 0;
+}
+
+unsigned DWARFVerifier::verifyAppleAccelTable(const DWARFSection *AccelSection,
+ DataExtractor *StrData,
+ const char *SectionName) {
+ unsigned NumErrors = 0;
+ DWARFDataExtractor AccelSectionData(DCtx.getDWARFObj(), *AccelSection,
+ DCtx.isLittleEndian(), 0);
+ AppleAcceleratorTable AccelTable(AccelSectionData, *StrData);
+
+ OS << "Verifying " << SectionName << "...\n";
+
+ // Verify that the fixed part of the header is not too short.
+ if (!AccelSectionData.isValidOffset(AccelTable.getSizeHdr())) {
+ error() << "Section is too small to fit a section header.\n";
+ return 1;
+ }
+
+ // Verify that the section is not too short.
+ if (Error E = AccelTable.extract()) {
+ error() << toString(std::move(E)) << '\n';
+ return 1;
+ }
+
+ // Verify that all buckets have a valid hash index or are empty.
+ uint32_t NumBuckets = AccelTable.getNumBuckets();
+ uint32_t NumHashes = AccelTable.getNumHashes();
+
+ uint64_t BucketsOffset =
+ AccelTable.getSizeHdr() + AccelTable.getHeaderDataLength();
+ uint64_t HashesBase = BucketsOffset + NumBuckets * 4;
+ uint64_t OffsetsBase = HashesBase + NumHashes * 4;
+ for (uint32_t BucketIdx = 0; BucketIdx < NumBuckets; ++BucketIdx) {
+ uint32_t HashIdx = AccelSectionData.getU32(&BucketsOffset);
+ if (HashIdx >= NumHashes && HashIdx != UINT32_MAX) {
+ error() << format("Bucket[%d] has invalid hash index: %u.\n", BucketIdx,
+ HashIdx);
+ ++NumErrors;
+ }
+ }
+ uint32_t NumAtoms = AccelTable.getAtomsDesc().size();
+ if (NumAtoms == 0) {
+ error() << "No atoms: failed to read HashData.\n";
+ return 1;
+ }
+ if (!AccelTable.validateForms()) {
+ error() << "Unsupported form: failed to read HashData.\n";
+ return 1;
+ }
+
+ for (uint32_t HashIdx = 0; HashIdx < NumHashes; ++HashIdx) {
+ uint64_t HashOffset = HashesBase + 4 * HashIdx;
+ uint64_t DataOffset = OffsetsBase + 4 * HashIdx;
+ uint32_t Hash = AccelSectionData.getU32(&HashOffset);
+ uint64_t HashDataOffset = AccelSectionData.getU32(&DataOffset);
+ if (!AccelSectionData.isValidOffsetForDataOfSize(HashDataOffset,
+ sizeof(uint64_t))) {
+ error() << format("Hash[%d] has invalid HashData offset: "
+ "0x%08" PRIx64 ".\n",
+ HashIdx, HashDataOffset);
+ ++NumErrors;
+ }
+
+ uint64_t StrpOffset;
+ uint64_t StringOffset;
+ uint32_t StringCount = 0;
+ uint64_t Offset;
+ unsigned Tag;
+ while ((StrpOffset = AccelSectionData.getU32(&HashDataOffset)) != 0) {
+ const uint32_t NumHashDataObjects =
+ AccelSectionData.getU32(&HashDataOffset);
+ for (uint32_t HashDataIdx = 0; HashDataIdx < NumHashDataObjects;
+ ++HashDataIdx) {
+ std::tie(Offset, Tag) = AccelTable.readAtoms(&HashDataOffset);
+ auto Die = DCtx.getDIEForOffset(Offset);
+ if (!Die) {
+ const uint32_t BucketIdx =
+ NumBuckets ? (Hash % NumBuckets) : UINT32_MAX;
+ StringOffset = StrpOffset;
+ const char *Name = StrData->getCStr(&StringOffset);
+ if (!Name)
+ Name = "<NULL>";
+
+ error() << format(
+ "%s Bucket[%d] Hash[%d] = 0x%08x "
+ "Str[%u] = 0x%08" PRIx64 " DIE[%d] = 0x%08" PRIx64 " "
+ "is not a valid DIE offset for \"%s\".\n",
+ SectionName, BucketIdx, HashIdx, Hash, StringCount, StrpOffset,
+ HashDataIdx, Offset, Name);
+
+ ++NumErrors;
+ continue;
+ }
+ if ((Tag != dwarf::DW_TAG_null) && (Die.getTag() != Tag)) {
+ error() << "Tag " << dwarf::TagString(Tag)
+ << " in accelerator table does not match Tag "
+ << dwarf::TagString(Die.getTag()) << " of DIE[" << HashDataIdx
+ << "].\n";
+ ++NumErrors;
+ }
+ }
+ ++StringCount;
+ }
+ }
+ return NumErrors;
+}
+
+unsigned
+DWARFVerifier::verifyDebugNamesCULists(const DWARFDebugNames &AccelTable) {
+ // A map from CU offset to the (first) Name Index offset which claims to index
+ // this CU.
+ DenseMap<uint64_t, uint64_t> CUMap;
+ const uint64_t NotIndexed = std::numeric_limits<uint64_t>::max();
+
+ CUMap.reserve(DCtx.getNumCompileUnits());
+ for (const auto &CU : DCtx.compile_units())
+ CUMap[CU->getOffset()] = NotIndexed;
+
+ unsigned NumErrors = 0;
+ for (const DWARFDebugNames::NameIndex &NI : AccelTable) {
+ if (NI.getCUCount() == 0) {
+ error() << formatv("Name Index @ {0:x} does not index any CU\n",
+ NI.getUnitOffset());
+ ++NumErrors;
+ continue;
+ }
+ for (uint32_t CU = 0, End = NI.getCUCount(); CU < End; ++CU) {
+ uint64_t Offset = NI.getCUOffset(CU);
+ auto Iter = CUMap.find(Offset);
+
+ if (Iter == CUMap.end()) {
+ error() << formatv(
+ "Name Index @ {0:x} references a non-existing CU @ {1:x}\n",
+ NI.getUnitOffset(), Offset);
+ ++NumErrors;
+ continue;
+ }
+
+ if (Iter->second != NotIndexed) {
+ error() << formatv("Name Index @ {0:x} references a CU @ {1:x}, but "
+ "this CU is already indexed by Name Index @ {2:x}\n",
+ NI.getUnitOffset(), Offset, Iter->second);
+ continue;
+ }
+ Iter->second = NI.getUnitOffset();
+ }
+ }
+
+ for (const auto &KV : CUMap) {
+ if (KV.second == NotIndexed)
+ warn() << formatv("CU @ {0:x} not covered by any Name Index\n", KV.first);
+ }
+
+ return NumErrors;
+}
+
+unsigned
+DWARFVerifier::verifyNameIndexBuckets(const DWARFDebugNames::NameIndex &NI,
+ const DataExtractor &StrData) {
+ struct BucketInfo {
+ uint32_t Bucket;
+ uint32_t Index;
+
+ constexpr BucketInfo(uint32_t Bucket, uint32_t Index)
+ : Bucket(Bucket), Index(Index) {}
+ bool operator<(const BucketInfo &RHS) const { return Index < RHS.Index; };
+ };
+
+ uint32_t NumErrors = 0;
+ if (NI.getBucketCount() == 0) {
+ warn() << formatv("Name Index @ {0:x} does not contain a hash table.\n",
+ NI.getUnitOffset());
+ return NumErrors;
+ }
+
+ // Build up a list of (Bucket, Index) pairs. We use this later to verify that
+ // each Name is reachable from the appropriate bucket.
+ std::vector<BucketInfo> BucketStarts;
+ BucketStarts.reserve(NI.getBucketCount() + 1);
+ for (uint32_t Bucket = 0, End = NI.getBucketCount(); Bucket < End; ++Bucket) {
+ uint32_t Index = NI.getBucketArrayEntry(Bucket);
+ if (Index > NI.getNameCount()) {
+ error() << formatv("Bucket {0} of Name Index @ {1:x} contains invalid "
+ "value {2}. Valid range is [0, {3}].\n",
+ Bucket, NI.getUnitOffset(), Index, NI.getNameCount());
+ ++NumErrors;
+ continue;
+ }
+ if (Index > 0)
+ BucketStarts.emplace_back(Bucket, Index);
+ }
+
+ // If there were any buckets with invalid values, skip further checks as they
+ // will likely produce many errors which will only confuse the actual root
+ // problem.
+ if (NumErrors > 0)
+ return NumErrors;
+
+ // Sort the list in the order of increasing "Index" entries.
+ array_pod_sort(BucketStarts.begin(), BucketStarts.end());
+
+ // Insert a sentinel entry at the end, so we can check that the end of the
+ // table is covered in the loop below.
+ BucketStarts.emplace_back(NI.getBucketCount(), NI.getNameCount() + 1);
+
+ // Loop invariant: NextUncovered is the (1-based) index of the first Name
+ // which is not reachable by any of the buckets we processed so far (and
+ // hasn't been reported as uncovered).
+ uint32_t NextUncovered = 1;
+ for (const BucketInfo &B : BucketStarts) {
+ // Under normal circumstances B.Index be equal to NextUncovered, but it can
+ // be less if a bucket points to names which are already known to be in some
+ // bucket we processed earlier. In that case, we won't trigger this error,
+ // but report the mismatched hash value error instead. (We know the hash
+ // will not match because we have already verified that the name's hash
+ // puts it into the previous bucket.)
+ if (B.Index > NextUncovered) {
+ error() << formatv("Name Index @ {0:x}: Name table entries [{1}, {2}] "
+ "are not covered by the hash table.\n",
+ NI.getUnitOffset(), NextUncovered, B.Index - 1);
+ ++NumErrors;
+ }
+ uint32_t Idx = B.Index;
+
+ // The rest of the checks apply only to non-sentinel entries.
+ if (B.Bucket == NI.getBucketCount())
+ break;
+
+ // This triggers if a non-empty bucket points to a name with a mismatched
+ // hash. Clients are likely to interpret this as an empty bucket, because a
+ // mismatched hash signals the end of a bucket, but if this is indeed an
+ // empty bucket, the producer should have signalled this by marking the
+ // bucket as empty.
+ uint32_t FirstHash = NI.getHashArrayEntry(Idx);
+ if (FirstHash % NI.getBucketCount() != B.Bucket) {
+ error() << formatv(
+ "Name Index @ {0:x}: Bucket {1} is not empty but points to a "
+ "mismatched hash value {2:x} (belonging to bucket {3}).\n",
+ NI.getUnitOffset(), B.Bucket, FirstHash,
+ FirstHash % NI.getBucketCount());
+ ++NumErrors;
+ }
+
+ // This find the end of this bucket and also verifies that all the hashes in
+ // this bucket are correct by comparing the stored hashes to the ones we
+ // compute ourselves.
+ while (Idx <= NI.getNameCount()) {
+ uint32_t Hash = NI.getHashArrayEntry(Idx);
+ if (Hash % NI.getBucketCount() != B.Bucket)
+ break;
+
+ const char *Str = NI.getNameTableEntry(Idx).getString();
+ if (caseFoldingDjbHash(Str) != Hash) {
+ error() << formatv("Name Index @ {0:x}: String ({1}) at index {2} "
+ "hashes to {3:x}, but "
+ "the Name Index hash is {4:x}\n",
+ NI.getUnitOffset(), Str, Idx,
+ caseFoldingDjbHash(Str), Hash);
+ ++NumErrors;
+ }
+
+ ++Idx;
+ }
+ NextUncovered = std::max(NextUncovered, Idx);
+ }
+ return NumErrors;
+}
+
+unsigned DWARFVerifier::verifyNameIndexAttribute(
+ const DWARFDebugNames::NameIndex &NI, const DWARFDebugNames::Abbrev &Abbr,
+ DWARFDebugNames::AttributeEncoding AttrEnc) {
+ StringRef FormName = dwarf::FormEncodingString(AttrEnc.Form);
+ if (FormName.empty()) {
+ error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x}: {2} uses an "
+ "unknown form: {3}.\n",
+ NI.getUnitOffset(), Abbr.Code, AttrEnc.Index,
+ AttrEnc.Form);
+ return 1;
+ }
+
+ if (AttrEnc.Index == DW_IDX_type_hash) {
+ if (AttrEnc.Form != dwarf::DW_FORM_data8) {
+ error() << formatv(
+ "NameIndex @ {0:x}: Abbreviation {1:x}: DW_IDX_type_hash "
+ "uses an unexpected form {2} (should be {3}).\n",
+ NI.getUnitOffset(), Abbr.Code, AttrEnc.Form, dwarf::DW_FORM_data8);
+ return 1;
+ }
+ }
+
+ // A list of known index attributes and their expected form classes.
+ // DW_IDX_type_hash is handled specially in the check above, as it has a
+ // specific form (not just a form class) we should expect.
+ struct FormClassTable {
+ dwarf::Index Index;
+ DWARFFormValue::FormClass Class;
+ StringLiteral ClassName;
+ };
+ static constexpr FormClassTable Table[] = {
+ {dwarf::DW_IDX_compile_unit, DWARFFormValue::FC_Constant, {"constant"}},
+ {dwarf::DW_IDX_type_unit, DWARFFormValue::FC_Constant, {"constant"}},
+ {dwarf::DW_IDX_die_offset, DWARFFormValue::FC_Reference, {"reference"}},
+ {dwarf::DW_IDX_parent, DWARFFormValue::FC_Constant, {"constant"}},
+ };
+
+ ArrayRef<FormClassTable> TableRef(Table);
+ auto Iter = find_if(TableRef, [AttrEnc](const FormClassTable &T) {
+ return T.Index == AttrEnc.Index;
+ });
+ if (Iter == TableRef.end()) {
+ warn() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} contains an "
+ "unknown index attribute: {2}.\n",
+ NI.getUnitOffset(), Abbr.Code, AttrEnc.Index);
+ return 0;
+ }
+
+ if (!DWARFFormValue(AttrEnc.Form).isFormClass(Iter->Class)) {
+ error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x}: {2} uses an "
+ "unexpected form {3} (expected form class {4}).\n",
+ NI.getUnitOffset(), Abbr.Code, AttrEnc.Index,
+ AttrEnc.Form, Iter->ClassName);
+ return 1;
+ }
+ return 0;
+}
+
+unsigned
+DWARFVerifier::verifyNameIndexAbbrevs(const DWARFDebugNames::NameIndex &NI) {
+ if (NI.getLocalTUCount() + NI.getForeignTUCount() > 0) {
+ warn() << formatv("Name Index @ {0:x}: Verifying indexes of type units is "
+ "not currently supported.\n",
+ NI.getUnitOffset());
+ return 0;
+ }
+
+ unsigned NumErrors = 0;
+ for (const auto &Abbrev : NI.getAbbrevs()) {
+ StringRef TagName = dwarf::TagString(Abbrev.Tag);
+ if (TagName.empty()) {
+ warn() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} references an "
+ "unknown tag: {2}.\n",
+ NI.getUnitOffset(), Abbrev.Code, Abbrev.Tag);
+ }
+ SmallSet<unsigned, 5> Attributes;
+ for (const auto &AttrEnc : Abbrev.Attributes) {
+ if (!Attributes.insert(AttrEnc.Index).second) {
+ error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} contains "
+ "multiple {2} attributes.\n",
+ NI.getUnitOffset(), Abbrev.Code, AttrEnc.Index);
+ ++NumErrors;
+ continue;
+ }
+ NumErrors += verifyNameIndexAttribute(NI, Abbrev, AttrEnc);
+ }
+
+ if (NI.getCUCount() > 1 && !Attributes.count(dwarf::DW_IDX_compile_unit)) {
+ error() << formatv("NameIndex @ {0:x}: Indexing multiple compile units "
+ "and abbreviation {1:x} has no {2} attribute.\n",
+ NI.getUnitOffset(), Abbrev.Code,
+ dwarf::DW_IDX_compile_unit);
+ ++NumErrors;
+ }
+ if (!Attributes.count(dwarf::DW_IDX_die_offset)) {
+ error() << formatv(
+ "NameIndex @ {0:x}: Abbreviation {1:x} has no {2} attribute.\n",
+ NI.getUnitOffset(), Abbrev.Code, dwarf::DW_IDX_die_offset);
+ ++NumErrors;
+ }
+ }
+ return NumErrors;
+}
+
+static SmallVector<StringRef, 2> getNames(const DWARFDie &DIE,
+ bool IncludeLinkageName = true) {
+ SmallVector<StringRef, 2> Result;
+ if (const char *Str = DIE.getName(DINameKind::ShortName))
+ Result.emplace_back(Str);
+ else if (DIE.getTag() == dwarf::DW_TAG_namespace)
+ Result.emplace_back("(anonymous namespace)");
+
+ if (IncludeLinkageName) {
+ if (const char *Str = DIE.getName(DINameKind::LinkageName)) {
+ if (Result.empty() || Result[0] != Str)
+ Result.emplace_back(Str);
+ }
+ }
+
+ return Result;
+}
+
+unsigned DWARFVerifier::verifyNameIndexEntries(
+ const DWARFDebugNames::NameIndex &NI,
+ const DWARFDebugNames::NameTableEntry &NTE) {
+ // Verifying type unit indexes not supported.
+ if (NI.getLocalTUCount() + NI.getForeignTUCount() > 0)
+ return 0;
+
+ const char *CStr = NTE.getString();
+ if (!CStr) {
+ error() << formatv(
+ "Name Index @ {0:x}: Unable to get string associated with name {1}.\n",
+ NI.getUnitOffset(), NTE.getIndex());
+ return 1;
+ }
+ StringRef Str(CStr);
+
+ unsigned NumErrors = 0;
+ unsigned NumEntries = 0;
+ uint64_t EntryID = NTE.getEntryOffset();
+ uint64_t NextEntryID = EntryID;
+ Expected<DWARFDebugNames::Entry> EntryOr = NI.getEntry(&NextEntryID);
+ for (; EntryOr; ++NumEntries, EntryID = NextEntryID,
+ EntryOr = NI.getEntry(&NextEntryID)) {
+ uint32_t CUIndex = *EntryOr->getCUIndex();
+ if (CUIndex > NI.getCUCount()) {
+ error() << formatv("Name Index @ {0:x}: Entry @ {1:x} contains an "
+ "invalid CU index ({2}).\n",
+ NI.getUnitOffset(), EntryID, CUIndex);
+ ++NumErrors;
+ continue;
+ }
+ uint64_t CUOffset = NI.getCUOffset(CUIndex);
+ uint64_t DIEOffset = CUOffset + *EntryOr->getDIEUnitOffset();
+ DWARFDie DIE = DCtx.getDIEForOffset(DIEOffset);
+ if (!DIE) {
+ error() << formatv("Name Index @ {0:x}: Entry @ {1:x} references a "
+ "non-existing DIE @ {2:x}.\n",
+ NI.getUnitOffset(), EntryID, DIEOffset);
+ ++NumErrors;
+ continue;
+ }
+ if (DIE.getDwarfUnit()->getOffset() != CUOffset) {
+ error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched CU of "
+ "DIE @ {2:x}: index - {3:x}; debug_info - {4:x}.\n",
+ NI.getUnitOffset(), EntryID, DIEOffset, CUOffset,
+ DIE.getDwarfUnit()->getOffset());
+ ++NumErrors;
+ }
+ if (DIE.getTag() != EntryOr->tag()) {
+ error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched Tag of "
+ "DIE @ {2:x}: index - {3}; debug_info - {4}.\n",
+ NI.getUnitOffset(), EntryID, DIEOffset, EntryOr->tag(),
+ DIE.getTag());
+ ++NumErrors;
+ }
+
+ auto EntryNames = getNames(DIE);
+ if (!is_contained(EntryNames, Str)) {
+ error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched Name "
+ "of DIE @ {2:x}: index - {3}; debug_info - {4}.\n",
+ NI.getUnitOffset(), EntryID, DIEOffset, Str,
+ make_range(EntryNames.begin(), EntryNames.end()));
+ ++NumErrors;
+ }
+ }
+ handleAllErrors(EntryOr.takeError(),
+ [&](const DWARFDebugNames::SentinelError &) {
+ if (NumEntries > 0)
+ return;
+ error() << formatv("Name Index @ {0:x}: Name {1} ({2}) is "
+ "not associated with any entries.\n",
+ NI.getUnitOffset(), NTE.getIndex(), Str);
+ ++NumErrors;
+ },
+ [&](const ErrorInfoBase &Info) {
+ error()
+ << formatv("Name Index @ {0:x}: Name {1} ({2}): {3}\n",
+ NI.getUnitOffset(), NTE.getIndex(), Str,
+ Info.message());
+ ++NumErrors;
+ });
+ return NumErrors;
+}
+
+static bool isVariableIndexable(const DWARFDie &Die, DWARFContext &DCtx) {
+ Optional<DWARFFormValue> Location = Die.findRecursively(DW_AT_location);
+ if (!Location)
+ return false;
+
+ auto ContainsInterestingOperators = [&](ArrayRef<uint8_t> D) {
+ DWARFUnit *U = Die.getDwarfUnit();
+ DataExtractor Data(toStringRef(D), DCtx.isLittleEndian(), U->getAddressByteSize());
+ DWARFExpression Expression(Data, U->getVersion(), U->getAddressByteSize());
+ return any_of(Expression, [](DWARFExpression::Operation &Op) {
+ return !Op.isError() && (Op.getCode() == DW_OP_addr ||
+ Op.getCode() == DW_OP_form_tls_address ||
+ Op.getCode() == DW_OP_GNU_push_tls_address);
+ });
+ };
+
+ if (Optional<ArrayRef<uint8_t>> Expr = Location->getAsBlock()) {
+ // Inlined location.
+ if (ContainsInterestingOperators(*Expr))
+ return true;
+ } else if (Optional<uint64_t> Offset = Location->getAsSectionOffset()) {
+ // Location list.
+ if (const DWARFDebugLoc *DebugLoc = DCtx.getDebugLoc()) {
+ if (const DWARFDebugLoc::LocationList *LocList =
+ DebugLoc->getLocationListAtOffset(*Offset)) {
+ if (any_of(LocList->Entries, [&](const DWARFDebugLoc::Entry &E) {
+ return ContainsInterestingOperators(E.Loc);
+ }))
+ return true;
+ }
+ }
+ }
+ return false;
+}
+
+unsigned DWARFVerifier::verifyNameIndexCompleteness(
+ const DWARFDie &Die, const DWARFDebugNames::NameIndex &NI) {
+
+ // First check, if the Die should be indexed. The code follows the DWARF v5
+ // wording as closely as possible.
+
+ // "All non-defining declarations (that is, debugging information entries
+ // with a DW_AT_declaration attribute) are excluded."
+ if (Die.find(DW_AT_declaration))
+ return 0;
+
+ // "DW_TAG_namespace debugging information entries without a DW_AT_name
+ // attribute are included with the name “(anonymous namespace)”.
+ // All other debugging information entries without a DW_AT_name attribute
+ // are excluded."
+ // "If a subprogram or inlined subroutine is included, and has a
+ // DW_AT_linkage_name attribute, there will be an additional index entry for
+ // the linkage name."
+ auto IncludeLinkageName = Die.getTag() == DW_TAG_subprogram ||
+ Die.getTag() == DW_TAG_inlined_subroutine;
+ auto EntryNames = getNames(Die, IncludeLinkageName);
+ if (EntryNames.empty())
+ return 0;
+
+ // We deviate from the specification here, which says:
+ // "The name index must contain an entry for each debugging information entry
+ // that defines a named subprogram, label, variable, type, or namespace,
+ // subject to ..."
+ // Instead whitelisting all TAGs representing a "type" or a "subprogram", to
+ // make sure we catch any missing items, we instead blacklist all TAGs that we
+ // know shouldn't be indexed.
+ switch (Die.getTag()) {
+ // Compile units and modules have names but shouldn't be indexed.
+ case DW_TAG_compile_unit:
+ case DW_TAG_module:
+ return 0;
+
+ // Function and template parameters are not globally visible, so we shouldn't
+ // index them.
+ case DW_TAG_formal_parameter:
+ case DW_TAG_template_value_parameter:
+ case DW_TAG_template_type_parameter:
+ case DW_TAG_GNU_template_parameter_pack:
+ case DW_TAG_GNU_template_template_param:
+ return 0;
+
+ // Object members aren't globally visible.
+ case DW_TAG_member:
+ return 0;
+
+ // According to a strict reading of the specification, enumerators should not
+ // be indexed (and LLVM currently does not do that). However, this causes
+ // problems for the debuggers, so we may need to reconsider this.
+ case DW_TAG_enumerator:
+ return 0;
+
+ // Imported declarations should not be indexed according to the specification
+ // and LLVM currently does not do that.
+ case DW_TAG_imported_declaration:
+ return 0;
+
+ // "DW_TAG_subprogram, DW_TAG_inlined_subroutine, and DW_TAG_label debugging
+ // information entries without an address attribute (DW_AT_low_pc,
+ // DW_AT_high_pc, DW_AT_ranges, or DW_AT_entry_pc) are excluded."
+ case DW_TAG_subprogram:
+ case DW_TAG_inlined_subroutine:
+ case DW_TAG_label:
+ if (Die.findRecursively(
+ {DW_AT_low_pc, DW_AT_high_pc, DW_AT_ranges, DW_AT_entry_pc}))
+ break;
+ return 0;
+
+ // "DW_TAG_variable debugging information entries with a DW_AT_location
+ // attribute that includes a DW_OP_addr or DW_OP_form_tls_address operator are
+ // included; otherwise, they are excluded."
+ //
+ // LLVM extension: We also add DW_OP_GNU_push_tls_address to this list.
+ case DW_TAG_variable:
+ if (isVariableIndexable(Die, DCtx))
+ break;
+ return 0;
+
+ default:
+ break;
+ }
+
+ // Now we know that our Die should be present in the Index. Let's check if
+ // that's the case.
+ unsigned NumErrors = 0;
+ uint64_t DieUnitOffset = Die.getOffset() - Die.getDwarfUnit()->getOffset();
+ for (StringRef Name : EntryNames) {
+ if (none_of(NI.equal_range(Name), [&](const DWARFDebugNames::Entry &E) {
+ return E.getDIEUnitOffset() == DieUnitOffset;
+ })) {
+ error() << formatv("Name Index @ {0:x}: Entry for DIE @ {1:x} ({2}) with "
+ "name {3} missing.\n",
+ NI.getUnitOffset(), Die.getOffset(), Die.getTag(),
+ Name);
+ ++NumErrors;
+ }
+ }
+ return NumErrors;
+}
+
+unsigned DWARFVerifier::verifyDebugNames(const DWARFSection &AccelSection,
+ const DataExtractor &StrData) {
+ unsigned NumErrors = 0;
+ DWARFDataExtractor AccelSectionData(DCtx.getDWARFObj(), AccelSection,
+ DCtx.isLittleEndian(), 0);
+ DWARFDebugNames AccelTable(AccelSectionData, StrData);
+
+ OS << "Verifying .debug_names...\n";
+
+ // This verifies that we can read individual name indices and their
+ // abbreviation tables.
+ if (Error E = AccelTable.extract()) {
+ error() << toString(std::move(E)) << '\n';
+ return 1;
+ }
+
+ NumErrors += verifyDebugNamesCULists(AccelTable);
+ for (const auto &NI : AccelTable)
+ NumErrors += verifyNameIndexBuckets(NI, StrData);
+ for (const auto &NI : AccelTable)
+ NumErrors += verifyNameIndexAbbrevs(NI);
+
+ // Don't attempt Entry validation if any of the previous checks found errors
+ if (NumErrors > 0)
+ return NumErrors;
+ for (const auto &NI : AccelTable)
+ for (DWARFDebugNames::NameTableEntry NTE : NI)
+ NumErrors += verifyNameIndexEntries(NI, NTE);
+
+ if (NumErrors > 0)
+ return NumErrors;
+
+ for (const std::unique_ptr<DWARFUnit> &U : DCtx.compile_units()) {
+ if (const DWARFDebugNames::NameIndex *NI =
+ AccelTable.getCUNameIndex(U->getOffset())) {
+ auto *CU = cast<DWARFCompileUnit>(U.get());
+ for (const DWARFDebugInfoEntry &Die : CU->dies())
+ NumErrors += verifyNameIndexCompleteness(DWARFDie(CU, &Die), *NI);
+ }
+ }
+ return NumErrors;
+}
+
+bool DWARFVerifier::handleAccelTables() {
+ const DWARFObject &D = DCtx.getDWARFObj();
+ DataExtractor StrData(D.getStrSection(), DCtx.isLittleEndian(), 0);
+ unsigned NumErrors = 0;
+ if (!D.getAppleNamesSection().Data.empty())
+ NumErrors += verifyAppleAccelTable(&D.getAppleNamesSection(), &StrData,
+ ".apple_names");
+ if (!D.getAppleTypesSection().Data.empty())
+ NumErrors += verifyAppleAccelTable(&D.getAppleTypesSection(), &StrData,
+ ".apple_types");
+ if (!D.getAppleNamespacesSection().Data.empty())
+ NumErrors += verifyAppleAccelTable(&D.getAppleNamespacesSection(), &StrData,
+ ".apple_namespaces");
+ if (!D.getAppleObjCSection().Data.empty())
+ NumErrors += verifyAppleAccelTable(&D.getAppleObjCSection(), &StrData,
+ ".apple_objc");
+
+ if (!D.getNamesSection().Data.empty())
+ NumErrors += verifyDebugNames(D.getNamesSection(), StrData);
+ return NumErrors == 0;
+}
+
+raw_ostream &DWARFVerifier::error() const { return WithColor::error(OS); }
+
+raw_ostream &DWARFVerifier::warn() const { return WithColor::warning(OS); }
+
+raw_ostream &DWARFVerifier::note() const { return WithColor::note(OS); }
+
+raw_ostream &DWARFVerifier::dump(const DWARFDie &Die, unsigned indent) const {
+ Die.dump(OS, indent, DumpOpts);
+ return OS;
+}
diff --git a/third_party/llvm-project/DWARFVisitor.cpp b/third_party/llvm-project/DWARFVisitor.cpp
new file mode 100644
index 000000000..ecb5967ac
--- /dev/null
+++ b/third_party/llvm-project/DWARFVisitor.cpp
@@ -0,0 +1,177 @@
+//===--- DWARFVisitor.cpp ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+//===----------------------------------------------------------------------===//
+
+#include "DWARFVisitor.h"
+#include "llvm/ObjectYAML/DWARFYAML.h"
+
+using namespace llvm;
+
+template <typename T>
+void DWARFYAML::VisitorImpl<T>::onVariableSizeValue(uint64_t U, unsigned Size) {
+ switch (Size) {
+ case 8:
+ onValue((uint64_t)U);
+ break;
+ case 4:
+ onValue((uint32_t)U);
+ break;
+ case 2:
+ onValue((uint16_t)U);
+ break;
+ case 1:
+ onValue((uint8_t)U);
+ break;
+ default:
+ llvm_unreachable("Invalid integer write size.");
+ }
+}
+
+static unsigned getOffsetSize(const DWARFYAML::Unit &Unit) {
+ return Unit.Length.isDWARF64() ? 8 : 4;
+}
+
+static unsigned getRefSize(const DWARFYAML::Unit &Unit) {
+ if (Unit.Version == 2)
+ return Unit.AddrSize;
+ return getOffsetSize(Unit);
+}
+
+template <typename T> void DWARFYAML::VisitorImpl<T>::traverseDebugInfo() {
+ for (auto &Unit : DebugInfo.CompileUnits) {
+ onStartCompileUnit(Unit);
+ auto FirstAbbrevCode = Unit.Entries[0].AbbrCode;
+
+ for (auto &Entry : Unit.Entries) {
+ onStartDIE(Unit, Entry);
+ if (Entry.AbbrCode == 0u)
+ continue;
+ auto &Abbrev = DebugInfo.AbbrevDecls[Entry.AbbrCode - FirstAbbrevCode];
+ auto FormVal = Entry.Values.begin();
+ auto AbbrForm = Abbrev.Attributes.begin();
+ for (;
+ FormVal != Entry.Values.end() && AbbrForm != Abbrev.Attributes.end();
+ ++FormVal, ++AbbrForm) {
+ onForm(*AbbrForm, *FormVal);
+ dwarf::Form Form = AbbrForm->Form;
+ bool Indirect;
+ do {
+ Indirect = false;
+ switch (Form) {
+ case dwarf::DW_FORM_addr:
+ onVariableSizeValue(FormVal->Value, Unit.AddrSize);
+ break;
+ case dwarf::DW_FORM_ref_addr:
+ onVariableSizeValue(FormVal->Value, getRefSize(Unit));
+ break;
+ case dwarf::DW_FORM_exprloc:
+ case dwarf::DW_FORM_block:
+ onValue((uint64_t)FormVal->BlockData.size(), true);
+ onValue(
+ MemoryBufferRef(StringRef((const char *)&FormVal->BlockData[0],
+ FormVal->BlockData.size()),
+ ""));
+ break;
+ case dwarf::DW_FORM_block1: {
+ auto writeSize = FormVal->BlockData.size();
+ onValue((uint8_t)writeSize);
+ onValue(
+ MemoryBufferRef(StringRef((const char *)&FormVal->BlockData[0],
+ FormVal->BlockData.size()),
+ ""));
+ break;
+ }
+ case dwarf::DW_FORM_block2: {
+ auto writeSize = FormVal->BlockData.size();
+ onValue((uint16_t)writeSize);
+ onValue(
+ MemoryBufferRef(StringRef((const char *)&FormVal->BlockData[0],
+ FormVal->BlockData.size()),
+ ""));
+ break;
+ }
+ case dwarf::DW_FORM_block4: {
+ auto writeSize = FormVal->BlockData.size();
+ onValue((uint32_t)writeSize);
+ onValue(
+ MemoryBufferRef(StringRef((const char *)&FormVal->BlockData[0],
+ FormVal->BlockData.size()),
+ ""));
+ break;
+ }
+ case dwarf::DW_FORM_data1:
+ case dwarf::DW_FORM_ref1:
+ case dwarf::DW_FORM_flag:
+ case dwarf::DW_FORM_strx1:
+ case dwarf::DW_FORM_addrx1:
+ onValue((uint8_t)FormVal->Value);
+ break;
+ case dwarf::DW_FORM_data2:
+ case dwarf::DW_FORM_ref2:
+ case dwarf::DW_FORM_strx2:
+ case dwarf::DW_FORM_addrx2:
+ onValue((uint16_t)FormVal->Value);
+ break;
+ case dwarf::DW_FORM_data4:
+ case dwarf::DW_FORM_ref4:
+ case dwarf::DW_FORM_ref_sup4:
+ case dwarf::DW_FORM_strx4:
+ case dwarf::DW_FORM_addrx4:
+ onValue((uint32_t)FormVal->Value);
+ break;
+ case dwarf::DW_FORM_data8:
+ case dwarf::DW_FORM_ref8:
+ case dwarf::DW_FORM_ref_sup8:
+ onValue((uint64_t)FormVal->Value);
+ break;
+ case dwarf::DW_FORM_sdata:
+ onValue((int64_t)FormVal->Value, true);
+ break;
+ case dwarf::DW_FORM_udata:
+ case dwarf::DW_FORM_ref_udata:
+ onValue((uint64_t)FormVal->Value, true);
+ break;
+ case dwarf::DW_FORM_string:
+ onValue(FormVal->CStr);
+ break;
+ case dwarf::DW_FORM_indirect:
+ onValue((uint64_t)FormVal->Value, true);
+ Indirect = true;
+ Form = static_cast<dwarf::Form>((uint64_t)FormVal->Value);
+ ++FormVal;
+ break;
+ case dwarf::DW_FORM_strp:
+ case dwarf::DW_FORM_sec_offset:
+ case dwarf::DW_FORM_GNU_ref_alt:
+ case dwarf::DW_FORM_GNU_strp_alt:
+ case dwarf::DW_FORM_line_strp:
+ case dwarf::DW_FORM_strp_sup:
+ onVariableSizeValue(FormVal->Value, getOffsetSize(Unit));
+ break;
+ case dwarf::DW_FORM_ref_sig8:
+ onValue((uint64_t)FormVal->Value);
+ break;
+ case dwarf::DW_FORM_GNU_addr_index:
+ case dwarf::DW_FORM_GNU_str_index:
+ onValue((uint64_t)FormVal->Value, true);
+ break;
+ default:
+ break;
+ }
+ } while (Indirect);
+ }
+ onEndDIE(Unit, Entry);
+ }
+ onEndCompileUnit(Unit);
+ }
+}
+
+// Explicitly instantiate the two template expansions.
+template class DWARFYAML::VisitorImpl<DWARFYAML::Data>;
+template class DWARFYAML::VisitorImpl<const DWARFYAML::Data>;
diff --git a/third_party/llvm-project/DWARFVisitor.h b/third_party/llvm-project/DWARFVisitor.h
new file mode 100644
index 000000000..50e88aa7a
--- /dev/null
+++ b/third_party/llvm-project/DWARFVisitor.h
@@ -0,0 +1,96 @@
+//===--- DWARFVisitor.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECTYAML_DWARFVISITOR_H
+#define LLVM_OBJECTYAML_DWARFVISITOR_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/MemoryBuffer.h"
+
+namespace llvm {
+
+namespace DWARFYAML {
+
+struct Data;
+struct Unit;
+struct Entry;
+struct FormValue;
+struct AttributeAbbrev;
+
+/// A class to visits DWARFYAML Compile Units and DIEs in preorder.
+///
+/// Extensions of this class can either maintain const or non-const references
+/// to the DWARFYAML::Data object.
+template <typename T> class VisitorImpl {
+protected:
+ T &DebugInfo;
+
+ /// Visitor Functions
+ /// @{
+ virtual void onStartCompileUnit(Unit &CU) {}
+ virtual void onEndCompileUnit(Unit &CU) {}
+ virtual void onStartDIE(Unit &CU, Entry &DIE) {}
+ virtual void onEndDIE(Unit &CU, Entry &DIE) {}
+ virtual void onForm(AttributeAbbrev &AttAbbrev, FormValue &Value) {}
+ /// @}
+
+ /// Const Visitor Functions
+ /// @{
+ virtual void onStartCompileUnit(const Unit &CU) {}
+ virtual void onEndCompileUnit(const Unit &CU) {}
+ virtual void onStartDIE(const Unit &CU, const Entry &DIE) {}
+ virtual void onEndDIE(const Unit &CU, const Entry &DIE) {}
+ virtual void onForm(const AttributeAbbrev &AttAbbrev,
+ const FormValue &Value) {}
+ /// @}
+
+ /// Value visitors
+ /// @{
+ virtual void onValue(const uint8_t U) {}
+ virtual void onValue(const uint16_t U) {}
+ virtual void onValue(const uint32_t U) {}
+ virtual void onValue(const uint64_t U, const bool LEB = false) {}
+ virtual void onValue(const int64_t S, const bool LEB = false) {}
+ virtual void onValue(const StringRef String) {}
+ virtual void onValue(const MemoryBufferRef MBR) {}
+ /// @}
+
+public:
+ VisitorImpl(T &DI) : DebugInfo(DI) {}
+
+ virtual ~VisitorImpl() {}
+
+ void traverseDebugInfo();
+
+private:
+ void onVariableSizeValue(uint64_t U, unsigned Size);
+};
+
+// Making the visior instantiations extern and explicit in the cpp file. This
+// prevents them from being instantiated in every compile unit that uses the
+// visitors.
+extern template class VisitorImpl<DWARFYAML::Data>;
+extern template class VisitorImpl<const DWARFYAML::Data>;
+
+class Visitor : public VisitorImpl<Data> {
+public:
+ Visitor(Data &DI) : VisitorImpl<Data>(DI) {}
+};
+
+class ConstVisitor : public VisitorImpl<const Data> {
+public:
+ ConstVisitor(const Data &DI) : VisitorImpl<const Data>(DI) {}
+};
+
+} // namespace DWARFYAML
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/DWARFYAML.cpp b/third_party/llvm-project/DWARFYAML.cpp
new file mode 100644
index 000000000..bb3b1422e
--- /dev/null
+++ b/third_party/llvm-project/DWARFYAML.cpp
@@ -0,0 +1,175 @@
+//===- DWARFYAML.cpp - DWARF YAMLIO implementation ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines classes for handling the YAML representation of DWARF Debug
+// Info.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ObjectYAML/DWARFYAML.h"
+
+namespace llvm {
+
+bool DWARFYAML::Data::isEmpty() const {
+ return 0 == DebugStrings.size() + AbbrevDecls.size();
+}
+
+namespace yaml {
+
+void MappingTraits<DWARFYAML::Data>::mapping(IO &IO, DWARFYAML::Data &DWARF) {
+ auto oldContext = IO.getContext();
+ IO.setContext(&DWARF);
+ IO.mapOptional("debug_str", DWARF.DebugStrings);
+ IO.mapOptional("debug_abbrev", DWARF.AbbrevDecls);
+ if (!DWARF.ARanges.empty() || !IO.outputting())
+ IO.mapOptional("debug_aranges", DWARF.ARanges);
+ if (!DWARF.PubNames.Entries.empty() || !IO.outputting())
+ IO.mapOptional("debug_pubnames", DWARF.PubNames);
+ if (!DWARF.PubTypes.Entries.empty() || !IO.outputting())
+ IO.mapOptional("debug_pubtypes", DWARF.PubTypes);
+ if (!DWARF.GNUPubNames.Entries.empty() || !IO.outputting())
+ IO.mapOptional("debug_gnu_pubnames", DWARF.GNUPubNames);
+ if (!DWARF.GNUPubTypes.Entries.empty() || !IO.outputting())
+ IO.mapOptional("debug_gnu_pubtypes", DWARF.GNUPubTypes);
+ IO.mapOptional("debug_info", DWARF.CompileUnits);
+ IO.mapOptional("debug_line", DWARF.DebugLines);
+ IO.setContext(&oldContext);
+}
+
+void MappingTraits<DWARFYAML::Abbrev>::mapping(IO &IO,
+ DWARFYAML::Abbrev &Abbrev) {
+ IO.mapRequired("Code", Abbrev.Code);
+ IO.mapRequired("Tag", Abbrev.Tag);
+ IO.mapRequired("Children", Abbrev.Children);
+ IO.mapRequired("Attributes", Abbrev.Attributes);
+}
+
+void MappingTraits<DWARFYAML::AttributeAbbrev>::mapping(
+ IO &IO, DWARFYAML::AttributeAbbrev &AttAbbrev) {
+ IO.mapRequired("Attribute", AttAbbrev.Attribute);
+ IO.mapRequired("Form", AttAbbrev.Form);
+ if(AttAbbrev.Form == dwarf::DW_FORM_implicit_const)
+ IO.mapRequired("Value", AttAbbrev.Value);
+}
+
+void MappingTraits<DWARFYAML::ARangeDescriptor>::mapping(
+ IO &IO, DWARFYAML::ARangeDescriptor &Descriptor) {
+ IO.mapRequired("Address", Descriptor.Address);
+ IO.mapRequired("Length", Descriptor.Length);
+}
+
+void MappingTraits<DWARFYAML::ARange>::mapping(IO &IO,
+ DWARFYAML::ARange &Range) {
+ IO.mapRequired("Length", Range.Length);
+ IO.mapRequired("Version", Range.Version);
+ IO.mapRequired("CuOffset", Range.CuOffset);
+ IO.mapRequired("AddrSize", Range.AddrSize);
+ IO.mapRequired("SegSize", Range.SegSize);
+ IO.mapRequired("Descriptors", Range.Descriptors);
+}
+
+void MappingTraits<DWARFYAML::PubEntry>::mapping(IO &IO,
+ DWARFYAML::PubEntry &Entry) {
+ IO.mapRequired("DieOffset", Entry.DieOffset);
+ if (reinterpret_cast<DWARFYAML::PubSection *>(IO.getContext())->IsGNUStyle)
+ IO.mapRequired("Descriptor", Entry.Descriptor);
+ IO.mapRequired("Name", Entry.Name);
+}
+
+void MappingTraits<DWARFYAML::PubSection>::mapping(
+ IO &IO, DWARFYAML::PubSection &Section) {
+ auto OldContext = IO.getContext();
+ IO.setContext(&Section);
+
+ IO.mapRequired("Length", Section.Length);
+ IO.mapRequired("Version", Section.Version);
+ IO.mapRequired("UnitOffset", Section.UnitOffset);
+ IO.mapRequired("UnitSize", Section.UnitSize);
+ IO.mapRequired("Entries", Section.Entries);
+
+ IO.setContext(OldContext);
+}
+
+void MappingTraits<DWARFYAML::Unit>::mapping(IO &IO, DWARFYAML::Unit &Unit) {
+ IO.mapRequired("Length", Unit.Length);
+ IO.mapRequired("Version", Unit.Version);
+ if (Unit.Version >= 5)
+ IO.mapRequired("UnitType", Unit.Type);
+ IO.mapRequired("AbbrOffset", Unit.AbbrOffset);
+ IO.mapRequired("AddrSize", Unit.AddrSize);
+ IO.mapOptional("Entries", Unit.Entries);
+}
+
+void MappingTraits<DWARFYAML::Entry>::mapping(IO &IO, DWARFYAML::Entry &Entry) {
+ IO.mapRequired("AbbrCode", Entry.AbbrCode);
+ IO.mapRequired("Values", Entry.Values);
+}
+
+void MappingTraits<DWARFYAML::FormValue>::mapping(
+ IO &IO, DWARFYAML::FormValue &FormValue) {
+ IO.mapOptional("Value", FormValue.Value);
+ if (!FormValue.CStr.empty() || !IO.outputting())
+ IO.mapOptional("CStr", FormValue.CStr);
+ if (!FormValue.BlockData.empty() || !IO.outputting())
+ IO.mapOptional("BlockData", FormValue.BlockData);
+}
+
+void MappingTraits<DWARFYAML::File>::mapping(IO &IO, DWARFYAML::File &File) {
+ IO.mapRequired("Name", File.Name);
+ IO.mapRequired("DirIdx", File.DirIdx);
+ IO.mapRequired("ModTime", File.ModTime);
+ IO.mapRequired("Length", File.Length);
+}
+
+void MappingTraits<DWARFYAML::LineTableOpcode>::mapping(
+ IO &IO, DWARFYAML::LineTableOpcode &LineTableOpcode) {
+ IO.mapRequired("Opcode", LineTableOpcode.Opcode);
+ if (LineTableOpcode.Opcode == dwarf::DW_LNS_extended_op) {
+ IO.mapRequired("ExtLen", LineTableOpcode.ExtLen);
+ IO.mapRequired("SubOpcode", LineTableOpcode.SubOpcode);
+ }
+
+ if (!LineTableOpcode.UnknownOpcodeData.empty() || !IO.outputting())
+ IO.mapOptional("UnknownOpcodeData", LineTableOpcode.UnknownOpcodeData);
+ if (!LineTableOpcode.UnknownOpcodeData.empty() || !IO.outputting())
+ IO.mapOptional("StandardOpcodeData", LineTableOpcode.StandardOpcodeData);
+ if (!LineTableOpcode.FileEntry.Name.empty() || !IO.outputting())
+ IO.mapOptional("FileEntry", LineTableOpcode.FileEntry);
+ if (LineTableOpcode.Opcode == dwarf::DW_LNS_advance_line || !IO.outputting())
+ IO.mapOptional("SData", LineTableOpcode.SData);
+ IO.mapOptional("Data", LineTableOpcode.Data);
+}
+
+void MappingTraits<DWARFYAML::LineTable>::mapping(
+ IO &IO, DWARFYAML::LineTable &LineTable) {
+ IO.mapRequired("Length", LineTable.Length);
+ IO.mapRequired("Version", LineTable.Version);
+ IO.mapRequired("PrologueLength", LineTable.PrologueLength);
+ IO.mapRequired("MinInstLength", LineTable.MinInstLength);
+ if(LineTable.Version >= 4)
+ IO.mapRequired("MaxOpsPerInst", LineTable.MaxOpsPerInst);
+ IO.mapRequired("DefaultIsStmt", LineTable.DefaultIsStmt);
+ IO.mapRequired("LineBase", LineTable.LineBase);
+ IO.mapRequired("LineRange", LineTable.LineRange);
+ IO.mapRequired("OpcodeBase", LineTable.OpcodeBase);
+ IO.mapRequired("StandardOpcodeLengths", LineTable.StandardOpcodeLengths);
+ IO.mapRequired("IncludeDirs", LineTable.IncludeDirs);
+ IO.mapRequired("Files", LineTable.Files);
+ IO.mapRequired("Opcodes", LineTable.Opcodes);
+}
+
+void MappingTraits<DWARFYAML::InitialLength>::mapping(
+ IO &IO, DWARFYAML::InitialLength &InitialLength) {
+ IO.mapRequired("TotalLength", InitialLength.TotalLength);
+ if (InitialLength.isDWARF64())
+ IO.mapRequired("TotalLength64", InitialLength.TotalLength64);
+}
+
+} // end namespace yaml
+
+} // end namespace llvm
diff --git a/third_party/llvm-project/DataExtractor.cpp b/third_party/llvm-project/DataExtractor.cpp
new file mode 100644
index 000000000..a98297cdb
--- /dev/null
+++ b/third_party/llvm-project/DataExtractor.cpp
@@ -0,0 +1,218 @@
+//===-- DataExtractor.cpp -------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Host.h"
+#include "llvm/Support/LEB128.h"
+#include "llvm/Support/SwapByteOrder.h"
+
+using namespace llvm;
+
+static void unexpectedEndReached(Error *E) {
+ if (E)
+ *E = createStringError(errc::illegal_byte_sequence,
+ "unexpected end of data");
+}
+
+static bool isError(Error *E) { return E && *E; }
+
+template <typename T>
+static T getU(uint64_t *offset_ptr, const DataExtractor *de,
+ bool isLittleEndian, const char *Data, llvm::Error *Err) {
+ ErrorAsOutParameter ErrAsOut(Err);
+ T val = 0;
+ if (isError(Err))
+ return val;
+
+ uint64_t offset = *offset_ptr;
+ if (!de->isValidOffsetForDataOfSize(offset, sizeof(T))) {
+ unexpectedEndReached(Err);
+ return val;
+ }
+ std::memcpy(&val, &Data[offset], sizeof(val));
+ if (sys::IsLittleEndianHost != isLittleEndian)
+ sys::swapByteOrder(val);
+
+ // Advance the offset
+ *offset_ptr += sizeof(val);
+ return val;
+}
+
+template <typename T>
+static T *getUs(uint64_t *offset_ptr, T *dst, uint32_t count,
+ const DataExtractor *de, bool isLittleEndian, const char *Data,
+ llvm::Error *Err) {
+ ErrorAsOutParameter ErrAsOut(Err);
+ if (isError(Err))
+ return nullptr;
+
+ uint64_t offset = *offset_ptr;
+
+ if (!de->isValidOffsetForDataOfSize(offset, sizeof(*dst) * count)) {
+ unexpectedEndReached(Err);
+ return nullptr;
+ }
+ for (T *value_ptr = dst, *end = dst + count; value_ptr != end;
+ ++value_ptr, offset += sizeof(*dst))
+ *value_ptr = getU<T>(offset_ptr, de, isLittleEndian, Data, Err);
+ // Advance the offset
+ *offset_ptr = offset;
+ // Return a non-NULL pointer to the converted data as an indicator of
+ // success
+ return dst;
+}
+
+uint8_t DataExtractor::getU8(uint64_t *offset_ptr, llvm::Error *Err) const {
+ return getU<uint8_t>(offset_ptr, this, IsLittleEndian, Data.data(), Err);
+}
+
+uint8_t *
+DataExtractor::getU8(uint64_t *offset_ptr, uint8_t *dst, uint32_t count) const {
+ return getUs<uint8_t>(offset_ptr, dst, count, this, IsLittleEndian,
+ Data.data(), nullptr);
+}
+
+uint8_t *DataExtractor::getU8(Cursor &C, uint8_t *Dst, uint32_t Count) const {
+ return getUs<uint8_t>(&C.Offset, Dst, Count, this, IsLittleEndian,
+ Data.data(), &C.Err);
+}
+
+uint16_t DataExtractor::getU16(uint64_t *offset_ptr, llvm::Error *Err) const {
+ return getU<uint16_t>(offset_ptr, this, IsLittleEndian, Data.data(), Err);
+}
+
+uint16_t *DataExtractor::getU16(uint64_t *offset_ptr, uint16_t *dst,
+ uint32_t count) const {
+ return getUs<uint16_t>(offset_ptr, dst, count, this, IsLittleEndian,
+ Data.data(), nullptr);
+}
+
+uint32_t DataExtractor::getU24(uint64_t *offset_ptr) const {
+ uint24_t ExtractedVal =
+ getU<uint24_t>(offset_ptr, this, IsLittleEndian, Data.data(), nullptr);
+ // The 3 bytes are in the correct byte order for the host.
+ return ExtractedVal.getAsUint32(sys::IsLittleEndianHost);
+}
+
+uint32_t DataExtractor::getU32(uint64_t *offset_ptr, llvm::Error *Err) const {
+ return getU<uint32_t>(offset_ptr, this, IsLittleEndian, Data.data(), Err);
+}
+
+uint32_t *DataExtractor::getU32(uint64_t *offset_ptr, uint32_t *dst,
+ uint32_t count) const {
+ return getUs<uint32_t>(offset_ptr, dst, count, this, IsLittleEndian,
+ Data.data(), nullptr);
+}
+
+uint64_t DataExtractor::getU64(uint64_t *offset_ptr, llvm::Error *Err) const {
+ return getU<uint64_t>(offset_ptr, this, IsLittleEndian, Data.data(), Err);
+}
+
+uint64_t *DataExtractor::getU64(uint64_t *offset_ptr, uint64_t *dst,
+ uint32_t count) const {
+ return getUs<uint64_t>(offset_ptr, dst, count, this, IsLittleEndian,
+ Data.data(), nullptr);
+}
+
+uint64_t DataExtractor::getUnsigned(uint64_t *offset_ptr, uint32_t byte_size,
+ llvm::Error *Err) const {
+ switch (byte_size) {
+ case 1:
+ return getU8(offset_ptr, Err);
+ case 2:
+ return getU16(offset_ptr, Err);
+ case 4:
+ return getU32(offset_ptr, Err);
+ case 8:
+ return getU64(offset_ptr, Err);
+ }
+ llvm_unreachable("getUnsigned unhandled case!");
+}
+
+int64_t
+DataExtractor::getSigned(uint64_t *offset_ptr, uint32_t byte_size) const {
+ switch (byte_size) {
+ case 1:
+ return (int8_t)getU8(offset_ptr);
+ case 2:
+ return (int16_t)getU16(offset_ptr);
+ case 4:
+ return (int32_t)getU32(offset_ptr);
+ case 8:
+ return (int64_t)getU64(offset_ptr);
+ }
+ llvm_unreachable("getSigned unhandled case!");
+}
+
+const char *DataExtractor::getCStr(uint64_t *offset_ptr) const {
+ uint64_t offset = *offset_ptr;
+ StringRef::size_type pos = Data.find('\0', offset);
+ if (pos != StringRef::npos) {
+ *offset_ptr = pos + 1;
+ return Data.data() + offset;
+ }
+ return nullptr;
+}
+
+StringRef DataExtractor::getCStrRef(uint64_t *offset_ptr) const {
+ uint64_t Start = *offset_ptr;
+ StringRef::size_type Pos = Data.find('\0', Start);
+ if (Pos != StringRef::npos) {
+ *offset_ptr = Pos + 1;
+ return StringRef(Data.data() + Start, Pos - Start);
+ }
+ return StringRef();
+}
+
+uint64_t DataExtractor::getULEB128(uint64_t *offset_ptr,
+ llvm::Error *Err) const {
+ assert(*offset_ptr <= Data.size());
+ ErrorAsOutParameter ErrAsOut(Err);
+ if (isError(Err))
+ return 0;
+
+ const char *error;
+ unsigned bytes_read;
+ uint64_t result = decodeULEB128(
+ reinterpret_cast<const uint8_t *>(Data.data() + *offset_ptr), &bytes_read,
+ reinterpret_cast<const uint8_t *>(Data.data() + Data.size()), &error);
+ if (error) {
+ if (Err)
+ *Err = createStringError(errc::illegal_byte_sequence, error);
+ return 0;
+ }
+ *offset_ptr += bytes_read;
+ return result;
+}
+
+int64_t DataExtractor::getSLEB128(uint64_t *offset_ptr) const {
+ assert(*offset_ptr <= Data.size());
+
+ const char *error;
+ unsigned bytes_read;
+ int64_t result = decodeSLEB128(
+ reinterpret_cast<const uint8_t *>(Data.data() + *offset_ptr), &bytes_read,
+ reinterpret_cast<const uint8_t *>(Data.data() + Data.size()), &error);
+ if (error)
+ return 0;
+ *offset_ptr += bytes_read;
+ return result;
+}
+
+void DataExtractor::skip(Cursor &C, uint64_t Length) const {
+ ErrorAsOutParameter ErrAsOut(&C.Err);
+ if (isError(&C.Err))
+ return;
+
+ if (isValidOffsetForDataOfSize(C.Offset, Length))
+ C.Offset += Length;
+ else
+ unexpectedEndReached(&C.Err);
+}
diff --git a/third_party/llvm-project/Debug.cpp b/third_party/llvm-project/Debug.cpp
new file mode 100644
index 000000000..9d90546a6
--- /dev/null
+++ b/third_party/llvm-project/Debug.cpp
@@ -0,0 +1,157 @@
+//===-- Debug.cpp - An easy way to add debug output to your code ----------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a handy way of adding debugging information to your
+// code, without it being enabled all of the time, and without having to add
+// command line options to enable it.
+//
+// In particular, just wrap your code with the LLVM_DEBUG() macro, and it will
+// be enabled automatically if you specify '-debug' on the command-line.
+// Alternatively, you can also use the SET_DEBUG_TYPE("foo") macro to specify
+// that your debug code belongs to class "foo". Then, on the command line, you
+// can specify '-debug-only=foo' to enable JUST the debug information for the
+// foo class.
+//
+// When compiling without assertions, the -debug-* options and all code in
+// LLVM_DEBUG() statements disappears, so it does not affect the runtime of the
+// code.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/ManagedStatic.h"
+#include "llvm/Support/Signals.h"
+#include "llvm/Support/circular_raw_ostream.h"
+#include "llvm/Support/raw_ostream.h"
+
+#undef isCurrentDebugType
+#undef setCurrentDebugType
+#undef setCurrentDebugTypes
+
+using namespace llvm;
+
+// Even though LLVM might be built with NDEBUG, define symbols that the code
+// built without NDEBUG can depend on via the llvm/Support/Debug.h header.
+namespace llvm {
+/// Exported boolean set by the -debug option.
+bool DebugFlag = false;
+
+#if 0 // XXX BINARYEN
+static ManagedStatic<std::vector<std::string>> CurrentDebugType;
+#endif
+
+/// Return true if the specified string is the debug type
+/// specified on the command line, or if none was specified on the command line
+/// with the -debug-only=X option.
+bool isCurrentDebugType(const char *DebugType) {
+ llvm_unreachable("debug type");
+}
+
+/// Set the current debug type, as if the -debug-only=X
+/// option were specified. Note that DebugFlag also needs to be set to true for
+/// debug output to be produced.
+///
+void setCurrentDebugTypes(const char **Types, unsigned Count);
+
+void setCurrentDebugType(const char *Type) {
+ setCurrentDebugTypes(&Type, 1);
+}
+
+void setCurrentDebugTypes(const char **Types, unsigned Count) {
+ llvm_unreachable("set debug type");
+}
+} // namespace llvm
+
+// All Debug.h functionality is a no-op in NDEBUG mode.
+#if 0 // XXX BINARYEN ndef NDEBUG
+
+// -debug - Command line option to enable the DEBUG statements in the passes.
+// This flag may only be enabled in debug builds.
+static cl::opt<bool, true>
+Debug("debug", cl::desc("Enable debug output"), cl::Hidden,
+ cl::location(DebugFlag));
+
+// -debug-buffer-size - Buffer the last N characters of debug output
+//until program termination.
+static cl::opt<unsigned>
+DebugBufferSize("debug-buffer-size",
+ cl::desc("Buffer the last N characters of debug output "
+ "until program termination. "
+ "[default 0 -- immediate print-out]"),
+ cl::Hidden,
+ cl::init(0));
+
+namespace {
+
+struct DebugOnlyOpt {
+ void operator=(const std::string &Val) const {
+ if (Val.empty())
+ return;
+ DebugFlag = true;
+ SmallVector<StringRef,8> dbgTypes;
+ StringRef(Val).split(dbgTypes, ',', -1, false);
+ for (auto dbgType : dbgTypes)
+ CurrentDebugType->push_back(dbgType);
+ }
+};
+
+}
+
+static DebugOnlyOpt DebugOnlyOptLoc;
+
+static cl::opt<DebugOnlyOpt, true, cl::parser<std::string> >
+DebugOnly("debug-only", cl::desc("Enable a specific type of debug output (comma separated list of types)"),
+ cl::Hidden, cl::ZeroOrMore, cl::value_desc("debug string"),
+ cl::location(DebugOnlyOptLoc), cl::ValueRequired);
+// Signal handlers - dump debug output on termination.
+static void debug_user_sig_handler(void *Cookie) {
+ // This is a bit sneaky. Since this is under #ifndef NDEBUG, we
+ // know that debug mode is enabled and dbgs() really is a
+ // circular_raw_ostream. If NDEBUG is defined, then dbgs() ==
+ // errs() but this will never be invoked.
+ llvm::circular_raw_ostream &dbgout =
+ static_cast<circular_raw_ostream &>(llvm::dbgs());
+ dbgout.flushBufferWithBanner();
+}
+
+/// dbgs - Return a circular-buffered debug stream.
+raw_ostream &llvm::dbgs() {
+ // Do one-time initialization in a thread-safe way.
+ static struct dbgstream {
+ circular_raw_ostream strm;
+
+ dbgstream() :
+ strm(errs(), "*** Debug Log Output ***\n",
+ (!EnableDebugBuffering || !DebugFlag) ? 0 : DebugBufferSize) {
+ if (EnableDebugBuffering && DebugFlag && DebugBufferSize != 0)
+ // TODO: Add a handler for SIGUSER1-type signals so the user can
+ // force a debug dump.
+ sys::AddSignalHandler(&debug_user_sig_handler, nullptr);
+ // Otherwise we've already set the debug stream buffer size to
+ // zero, disabling buffering so it will output directly to errs().
+ }
+ } thestrm;
+
+ return thestrm.strm;
+}
+
+#else
+// Avoid "has no symbols" warning.
+namespace llvm {
+ /// dbgs - Return errs().
+ raw_ostream &dbgs() {
+ return errs();
+ }
+}
+
+#endif
+
+/// EnableDebugBuffering - Turn on signal handler installation.
+///
+bool llvm::EnableDebugBuffering = false;
diff --git a/third_party/llvm-project/Dwarf.cpp b/third_party/llvm-project/Dwarf.cpp
new file mode 100644
index 000000000..9ca331741
--- /dev/null
+++ b/third_party/llvm-project/Dwarf.cpp
@@ -0,0 +1,759 @@
+//===-- llvm/BinaryFormat/Dwarf.cpp - Dwarf Framework ------------*- C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains support for generic dwarf information.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/ADT/StringSwitch.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/Support/ErrorHandling.h"
+
+using namespace llvm;
+using namespace dwarf;
+
+StringRef llvm::dwarf::TagString(unsigned Tag) {
+ switch (Tag) {
+ default:
+ return StringRef();
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) \
+ case DW_TAG_##NAME: \
+ return "DW_TAG_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::getTag(StringRef TagString) {
+ return StringSwitch<unsigned>(TagString)
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) \
+ .Case("DW_TAG_" #NAME, DW_TAG_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Default(DW_TAG_invalid);
+}
+
+unsigned llvm::dwarf::TagVersion(dwarf::Tag Tag) {
+ switch (Tag) {
+ default:
+ return 0;
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) \
+ case DW_TAG_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::TagVendor(dwarf::Tag Tag) {
+ switch (Tag) {
+ default:
+ return 0;
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) \
+ case DW_TAG_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::ChildrenString(unsigned Children) {
+ switch (Children) {
+ case DW_CHILDREN_no:
+ return "DW_CHILDREN_no";
+ case DW_CHILDREN_yes:
+ return "DW_CHILDREN_yes";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::AttributeString(unsigned Attribute) {
+ switch (Attribute) {
+ default:
+ return StringRef();
+#define HANDLE_DW_AT(ID, NAME, VERSION, VENDOR) \
+ case DW_AT_##NAME: \
+ return "DW_AT_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::AttributeVersion(dwarf::Attribute Attribute) {
+ switch (Attribute) {
+ default:
+ return 0;
+#define HANDLE_DW_AT(ID, NAME, VERSION, VENDOR) \
+ case DW_AT_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::AttributeVendor(dwarf::Attribute Attribute) {
+ switch (Attribute) {
+ default:
+ return 0;
+#define HANDLE_DW_AT(ID, NAME, VERSION, VENDOR) \
+ case DW_AT_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::FormEncodingString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_FORM(ID, NAME, VERSION, VENDOR) \
+ case DW_FORM_##NAME: \
+ return "DW_FORM_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::FormVersion(dwarf::Form Form) {
+ switch (Form) {
+ default:
+ return 0;
+#define HANDLE_DW_FORM(ID, NAME, VERSION, VENDOR) \
+ case DW_FORM_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::FormVendor(dwarf::Form Form) {
+ switch (Form) {
+ default:
+ return 0;
+#define HANDLE_DW_FORM(ID, NAME, VERSION, VENDOR) \
+ case DW_FORM_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::OperationEncodingString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR) \
+ case DW_OP_##NAME: \
+ return "DW_OP_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ case DW_OP_LLVM_convert:
+ return "DW_OP_LLVM_convert";
+ case DW_OP_LLVM_fragment:
+ return "DW_OP_LLVM_fragment";
+ case DW_OP_LLVM_tag_offset:
+ return "DW_OP_LLVM_tag_offset";
+ case DW_OP_LLVM_entry_value:
+ return "DW_OP_LLVM_entry_value";
+ }
+}
+
+unsigned llvm::dwarf::getOperationEncoding(StringRef OperationEncodingString) {
+ return StringSwitch<unsigned>(OperationEncodingString)
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR) \
+ .Case("DW_OP_" #NAME, DW_OP_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Case("DW_OP_LLVM_convert", DW_OP_LLVM_convert)
+ .Case("DW_OP_LLVM_fragment", DW_OP_LLVM_fragment)
+ .Case("DW_OP_LLVM_tag_offset", DW_OP_LLVM_tag_offset)
+ .Case("DW_OP_LLVM_entry_value", DW_OP_LLVM_entry_value)
+ .Default(0);
+}
+
+unsigned llvm::dwarf::OperationVersion(dwarf::LocationAtom Op) {
+ switch (Op) {
+ default:
+ return 0;
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR) \
+ case DW_OP_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::OperationVendor(dwarf::LocationAtom Op) {
+ switch (Op) {
+ default:
+ return 0;
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR) \
+ case DW_OP_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::AttributeEncodingString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR) \
+ case DW_ATE_##NAME: \
+ return "DW_ATE_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::getAttributeEncoding(StringRef EncodingString) {
+ return StringSwitch<unsigned>(EncodingString)
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR) \
+ .Case("DW_ATE_" #NAME, DW_ATE_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Default(0);
+}
+
+unsigned llvm::dwarf::AttributeEncodingVersion(dwarf::TypeKind ATE) {
+ switch (ATE) {
+ default:
+ return 0;
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR) \
+ case DW_ATE_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::AttributeEncodingVendor(dwarf::TypeKind ATE) {
+ switch (ATE) {
+ default:
+ return 0;
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR) \
+ case DW_ATE_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::DecimalSignString(unsigned Sign) {
+ switch (Sign) {
+ case DW_DS_unsigned:
+ return "DW_DS_unsigned";
+ case DW_DS_leading_overpunch:
+ return "DW_DS_leading_overpunch";
+ case DW_DS_trailing_overpunch:
+ return "DW_DS_trailing_overpunch";
+ case DW_DS_leading_separate:
+ return "DW_DS_leading_separate";
+ case DW_DS_trailing_separate:
+ return "DW_DS_trailing_separate";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::EndianityString(unsigned Endian) {
+ switch (Endian) {
+ case DW_END_default:
+ return "DW_END_default";
+ case DW_END_big:
+ return "DW_END_big";
+ case DW_END_little:
+ return "DW_END_little";
+ case DW_END_lo_user:
+ return "DW_END_lo_user";
+ case DW_END_hi_user:
+ return "DW_END_hi_user";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::AccessibilityString(unsigned Access) {
+ switch (Access) {
+ // Accessibility codes
+ case DW_ACCESS_public:
+ return "DW_ACCESS_public";
+ case DW_ACCESS_protected:
+ return "DW_ACCESS_protected";
+ case DW_ACCESS_private:
+ return "DW_ACCESS_private";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::DefaultedMemberString(unsigned DefaultedEncodings) {
+ switch (DefaultedEncodings) {
+ // Defaulted Member Encodings codes
+ case DW_DEFAULTED_no:
+ return "DW_DEFAULTED_no";
+ case DW_DEFAULTED_in_class:
+ return "DW_DEFAULTED_in_class";
+ case DW_DEFAULTED_out_of_class:
+ return "DW_DEFAULTED_out_of_class";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::VisibilityString(unsigned Visibility) {
+ switch (Visibility) {
+ case DW_VIS_local:
+ return "DW_VIS_local";
+ case DW_VIS_exported:
+ return "DW_VIS_exported";
+ case DW_VIS_qualified:
+ return "DW_VIS_qualified";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::VirtualityString(unsigned Virtuality) {
+ switch (Virtuality) {
+ default:
+ return StringRef();
+#define HANDLE_DW_VIRTUALITY(ID, NAME) \
+ case DW_VIRTUALITY_##NAME: \
+ return "DW_VIRTUALITY_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::getVirtuality(StringRef VirtualityString) {
+ return StringSwitch<unsigned>(VirtualityString)
+#define HANDLE_DW_VIRTUALITY(ID, NAME) \
+ .Case("DW_VIRTUALITY_" #NAME, DW_VIRTUALITY_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Default(DW_VIRTUALITY_invalid);
+}
+
+StringRef llvm::dwarf::LanguageString(unsigned Language) {
+ switch (Language) {
+ default:
+ return StringRef();
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ case DW_LANG_##NAME: \
+ return "DW_LANG_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::getLanguage(StringRef LanguageString) {
+ return StringSwitch<unsigned>(LanguageString)
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ .Case("DW_LANG_" #NAME, DW_LANG_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Default(0);
+}
+
+unsigned llvm::dwarf::LanguageVersion(dwarf::SourceLanguage Lang) {
+ switch (Lang) {
+ default:
+ return 0;
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ case DW_LANG_##NAME: \
+ return VERSION;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::LanguageVendor(dwarf::SourceLanguage Lang) {
+ switch (Lang) {
+ default:
+ return 0;
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ case DW_LANG_##NAME: \
+ return DWARF_VENDOR_##VENDOR;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+Optional<unsigned> llvm::dwarf::LanguageLowerBound(dwarf::SourceLanguage Lang) {
+ switch (Lang) {
+ default:
+ return None;
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ case DW_LANG_##NAME: \
+ return LOWER_BOUND;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::CaseString(unsigned Case) {
+ switch (Case) {
+ case DW_ID_case_sensitive:
+ return "DW_ID_case_sensitive";
+ case DW_ID_up_case:
+ return "DW_ID_up_case";
+ case DW_ID_down_case:
+ return "DW_ID_down_case";
+ case DW_ID_case_insensitive:
+ return "DW_ID_case_insensitive";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::ConventionString(unsigned CC) {
+ switch (CC) {
+ default:
+ return StringRef();
+#define HANDLE_DW_CC(ID, NAME) \
+ case DW_CC_##NAME: \
+ return "DW_CC_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+unsigned llvm::dwarf::getCallingConvention(StringRef CCString) {
+ return StringSwitch<unsigned>(CCString)
+#define HANDLE_DW_CC(ID, NAME) .Case("DW_CC_" #NAME, DW_CC_##NAME)
+#include "llvm/BinaryFormat/Dwarf.def"
+ .Default(0);
+}
+
+StringRef llvm::dwarf::InlineCodeString(unsigned Code) {
+ switch (Code) {
+ case DW_INL_not_inlined:
+ return "DW_INL_not_inlined";
+ case DW_INL_inlined:
+ return "DW_INL_inlined";
+ case DW_INL_declared_not_inlined:
+ return "DW_INL_declared_not_inlined";
+ case DW_INL_declared_inlined:
+ return "DW_INL_declared_inlined";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::ArrayOrderString(unsigned Order) {
+ switch (Order) {
+ case DW_ORD_row_major:
+ return "DW_ORD_row_major";
+ case DW_ORD_col_major:
+ return "DW_ORD_col_major";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::LNStandardString(unsigned Standard) {
+ switch (Standard) {
+ default:
+ return StringRef();
+#define HANDLE_DW_LNS(ID, NAME) \
+ case DW_LNS_##NAME: \
+ return "DW_LNS_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::LNExtendedString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_LNE(ID, NAME) \
+ case DW_LNE_##NAME: \
+ return "DW_LNE_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::MacinfoString(unsigned Encoding) {
+ switch (Encoding) {
+ // Macinfo Type Encodings
+ case DW_MACINFO_define:
+ return "DW_MACINFO_define";
+ case DW_MACINFO_undef:
+ return "DW_MACINFO_undef";
+ case DW_MACINFO_start_file:
+ return "DW_MACINFO_start_file";
+ case DW_MACINFO_end_file:
+ return "DW_MACINFO_end_file";
+ case DW_MACINFO_vendor_ext:
+ return "DW_MACINFO_vendor_ext";
+ case DW_MACINFO_invalid:
+ return "DW_MACINFO_invalid";
+ }
+ return StringRef();
+}
+
+unsigned llvm::dwarf::getMacinfo(StringRef MacinfoString) {
+ return StringSwitch<unsigned>(MacinfoString)
+ .Case("DW_MACINFO_define", DW_MACINFO_define)
+ .Case("DW_MACINFO_undef", DW_MACINFO_undef)
+ .Case("DW_MACINFO_start_file", DW_MACINFO_start_file)
+ .Case("DW_MACINFO_end_file", DW_MACINFO_end_file)
+ .Case("DW_MACINFO_vendor_ext", DW_MACINFO_vendor_ext)
+ .Default(DW_MACINFO_invalid);
+}
+
+StringRef llvm::dwarf::RangeListEncodingString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_RLE(ID, NAME) \
+ case DW_RLE_##NAME: \
+ return "DW_RLE_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::LocListEncodingString(unsigned Encoding) {
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_LLE(ID, NAME) \
+ case DW_LLE_##NAME: \
+ return "DW_LLE_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::CallFrameString(unsigned Encoding,
+ Triple::ArchType Arch) {
+ assert(Arch != llvm::Triple::ArchType::UnknownArch);
+#define SELECT_AARCH64 (Arch == llvm::Triple::aarch64_be || Arch == llvm::Triple::aarch64)
+#define SELECT_MIPS64 Arch == llvm::Triple::mips64
+#define SELECT_SPARC (Arch == llvm::Triple::sparc || Arch == llvm::Triple::sparcv9)
+#define SELECT_X86 (Arch == llvm::Triple::x86 || Arch == llvm::Triple::x86_64)
+#define HANDLE_DW_CFA(ID, NAME)
+#define HANDLE_DW_CFA_PRED(ID, NAME, PRED) \
+ if (ID == Encoding && PRED) \
+ return "DW_CFA_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+
+ switch (Encoding) {
+ default:
+ return StringRef();
+#define HANDLE_DW_CFA_PRED(ID, NAME, PRED)
+#define HANDLE_DW_CFA(ID, NAME) \
+ case DW_CFA_##NAME: \
+ return "DW_CFA_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+
+#undef SELECT_X86
+#undef SELECT_SPARC
+#undef SELECT_MIPS64
+#undef SELECT_AARCH64
+ }
+}
+
+StringRef llvm::dwarf::ApplePropertyString(unsigned Prop) {
+ switch (Prop) {
+ default:
+ return StringRef();
+#define HANDLE_DW_APPLE_PROPERTY(ID, NAME) \
+ case DW_APPLE_PROPERTY_##NAME: \
+ return "DW_APPLE_PROPERTY_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::UnitTypeString(unsigned UT) {
+ switch (UT) {
+ default:
+ return StringRef();
+#define HANDLE_DW_UT(ID, NAME) \
+ case DW_UT_##NAME: \
+ return "DW_UT_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+StringRef llvm::dwarf::AtomTypeString(unsigned AT) {
+ switch (AT) {
+ case dwarf::DW_ATOM_null:
+ return "DW_ATOM_null";
+ case dwarf::DW_ATOM_die_offset:
+ return "DW_ATOM_die_offset";
+ case DW_ATOM_cu_offset:
+ return "DW_ATOM_cu_offset";
+ case DW_ATOM_die_tag:
+ return "DW_ATOM_die_tag";
+ case DW_ATOM_type_flags:
+ case DW_ATOM_type_type_flags:
+ return "DW_ATOM_type_flags";
+ case DW_ATOM_qual_name_hash:
+ return "DW_ATOM_qual_name_hash";
+ }
+ return StringRef();
+}
+
+StringRef llvm::dwarf::GDBIndexEntryKindString(GDBIndexEntryKind Kind) {
+ switch (Kind) {
+ case GIEK_NONE:
+ return "NONE";
+ case GIEK_TYPE:
+ return "TYPE";
+ case GIEK_VARIABLE:
+ return "VARIABLE";
+ case GIEK_FUNCTION:
+ return "FUNCTION";
+ case GIEK_OTHER:
+ return "OTHER";
+ case GIEK_UNUSED5:
+ return "UNUSED5";
+ case GIEK_UNUSED6:
+ return "UNUSED6";
+ case GIEK_UNUSED7:
+ return "UNUSED7";
+ }
+ llvm_unreachable("Unknown GDBIndexEntryKind value");
+}
+
+StringRef
+llvm::dwarf::GDBIndexEntryLinkageString(GDBIndexEntryLinkage Linkage) {
+ switch (Linkage) {
+ case GIEL_EXTERNAL:
+ return "EXTERNAL";
+ case GIEL_STATIC:
+ return "STATIC";
+ }
+ llvm_unreachable("Unknown GDBIndexEntryLinkage value");
+}
+
+StringRef llvm::dwarf::AttributeValueString(uint16_t Attr, unsigned Val) {
+ switch (Attr) {
+ case DW_AT_accessibility:
+ return AccessibilityString(Val);
+ case DW_AT_virtuality:
+ return VirtualityString(Val);
+ case DW_AT_language:
+ return LanguageString(Val);
+ case DW_AT_encoding:
+ return AttributeEncodingString(Val);
+ case DW_AT_decimal_sign:
+ return DecimalSignString(Val);
+ case DW_AT_endianity:
+ return EndianityString(Val);
+ case DW_AT_visibility:
+ return VisibilityString(Val);
+ case DW_AT_identifier_case:
+ return CaseString(Val);
+ case DW_AT_calling_convention:
+ return ConventionString(Val);
+ case DW_AT_inline:
+ return InlineCodeString(Val);
+ case DW_AT_ordering:
+ return ArrayOrderString(Val);
+ case DW_AT_APPLE_runtime_class:
+ return LanguageString(Val);
+ case DW_AT_defaulted:
+ return DefaultedMemberString(Val);
+ }
+
+ return StringRef();
+}
+
+StringRef llvm::dwarf::AtomValueString(uint16_t Atom, unsigned Val) {
+ switch (Atom) {
+ case DW_ATOM_null:
+ return "NULL";
+ case DW_ATOM_die_tag:
+ return TagString(Val);
+ }
+
+ return StringRef();
+}
+
+StringRef llvm::dwarf::IndexString(unsigned Idx) {
+ switch (Idx) {
+ default:
+ return StringRef();
+#define HANDLE_DW_IDX(ID, NAME) \
+ case DW_IDX_##NAME: \
+ return "DW_IDX_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+Optional<uint8_t> llvm::dwarf::getFixedFormByteSize(dwarf::Form Form,
+ FormParams Params) {
+ switch (Form) {
+ case DW_FORM_addr:
+ if (Params)
+ return Params.AddrSize;
+ return None;
+
+ case DW_FORM_block: // ULEB128 length L followed by L bytes.
+ case DW_FORM_block1: // 1 byte length L followed by L bytes.
+ case DW_FORM_block2: // 2 byte length L followed by L bytes.
+ case DW_FORM_block4: // 4 byte length L followed by L bytes.
+ case DW_FORM_string: // C-string with null terminator.
+ case DW_FORM_sdata: // SLEB128.
+ case DW_FORM_udata: // ULEB128.
+ case DW_FORM_ref_udata: // ULEB128.
+ case DW_FORM_indirect: // ULEB128.
+ case DW_FORM_exprloc: // ULEB128 length L followed by L bytes.
+ case DW_FORM_strx: // ULEB128.
+ case DW_FORM_addrx: // ULEB128.
+ case DW_FORM_loclistx: // ULEB128.
+ case DW_FORM_rnglistx: // ULEB128.
+ case DW_FORM_GNU_addr_index: // ULEB128.
+ case DW_FORM_GNU_str_index: // ULEB128.
+ return None;
+
+ case DW_FORM_ref_addr:
+ if (Params)
+ return Params.getRefAddrByteSize();
+ return None;
+
+ case DW_FORM_flag:
+ case DW_FORM_data1:
+ case DW_FORM_ref1:
+ case DW_FORM_strx1:
+ case DW_FORM_addrx1:
+ return 1;
+
+ case DW_FORM_data2:
+ case DW_FORM_ref2:
+ case DW_FORM_strx2:
+ case DW_FORM_addrx2:
+ return 2;
+
+ case DW_FORM_strx3:
+ return 3;
+
+ case DW_FORM_data4:
+ case DW_FORM_ref4:
+ case DW_FORM_ref_sup4:
+ case DW_FORM_strx4:
+ case DW_FORM_addrx4:
+ return 4;
+
+ case DW_FORM_strp:
+ case DW_FORM_GNU_ref_alt:
+ case DW_FORM_GNU_strp_alt:
+ case DW_FORM_line_strp:
+ case DW_FORM_sec_offset:
+ case DW_FORM_strp_sup:
+ if (Params)
+ return Params.getDwarfOffsetByteSize();
+ return None;
+
+ case DW_FORM_data8:
+ case DW_FORM_ref8:
+ case DW_FORM_ref_sig8:
+ case DW_FORM_ref_sup8:
+ return 8;
+
+ case DW_FORM_flag_present:
+ return 0;
+
+ case DW_FORM_data16:
+ return 16;
+
+ case DW_FORM_implicit_const:
+ // The implicit value is stored in the abbreviation as a SLEB128, and
+ // there no data in debug info.
+ return 0;
+
+ default:
+ break;
+ }
+ return None;
+}
+
+bool llvm::dwarf::isValidFormForVersion(Form F, unsigned Version,
+ bool ExtensionsOk) {
+ if (FormVendor(F) == DWARF_VENDOR_DWARF) {
+ unsigned FV = FormVersion(F);
+ return FV > 0 && FV <= Version;
+ }
+ return ExtensionsOk;
+}
+
+constexpr char llvm::dwarf::EnumTraits<Attribute>::Type[];
+constexpr char llvm::dwarf::EnumTraits<Form>::Type[];
+constexpr char llvm::dwarf::EnumTraits<Index>::Type[];
+constexpr char llvm::dwarf::EnumTraits<Tag>::Type[];
diff --git a/third_party/llvm-project/Error.cpp b/third_party/llvm-project/Error.cpp
new file mode 100644
index 000000000..bb13c836d
--- /dev/null
+++ b/third_party/llvm-project/Error.cpp
@@ -0,0 +1,168 @@
+//===----- lib/Support/Error.cpp - Error and associated utilities ---------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/Error.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ManagedStatic.h"
+#include <system_error>
+
+using namespace llvm;
+
+namespace {
+
+ enum class ErrorErrorCode : int {
+ MultipleErrors = 1,
+ FileError,
+ InconvertibleError
+ };
+
+ // FIXME: This class is only here to support the transition to llvm::Error. It
+ // will be removed once this transition is complete. Clients should prefer to
+ // deal with the Error value directly, rather than converting to error_code.
+ class ErrorErrorCategory : public std::error_category {
+ public:
+ const char *name() const noexcept override { return "Error"; }
+
+ std::string message(int condition) const override {
+ switch (static_cast<ErrorErrorCode>(condition)) {
+ case ErrorErrorCode::MultipleErrors:
+ return "Multiple errors";
+ case ErrorErrorCode::InconvertibleError:
+ return "Inconvertible error value. An error has occurred that could "
+ "not be converted to a known std::error_code. Please file a "
+ "bug.";
+ case ErrorErrorCode::FileError:
+ return "A file error occurred.";
+ }
+ llvm_unreachable("Unhandled error code");
+ }
+ };
+
+}
+
+#if 0 // XXX BINARYEN
+static ManagedStatic<ErrorErrorCategory> ErrorErrorCat;
+#endif
+
+namespace llvm {
+
+void ErrorInfoBase::anchor() {}
+char ErrorInfoBase::ID = 0;
+char ErrorList::ID = 0;
+void ECError::anchor() {}
+char ECError::ID = 0;
+char StringError::ID = 0;
+char FileError::ID = 0;
+
+void logAllUnhandledErrors(Error E, raw_ostream &OS, Twine ErrorBanner) {
+ if (!E)
+ return;
+ OS << ErrorBanner;
+ handleAllErrors(std::move(E), [&](const ErrorInfoBase &EI) {
+ EI.log(OS);
+ OS << "\n";
+ });
+}
+
+
+std::error_code ErrorList::convertToErrorCode() const {
+ llvm_unreachable("convert error code");
+}
+
+std::error_code inconvertibleErrorCode() {
+ llvm_unreachable("inconvertible error code");
+}
+
+std::error_code FileError::convertToErrorCode() const {
+ llvm_unreachable("(file) convert error code");
+}
+
+Error errorCodeToError(std::error_code EC) {
+ if (!EC)
+ return Error::success();
+ return Error(std::make_unique<ECError>(ECError(EC)));
+}
+
+std::error_code errorToErrorCode(Error Err) {
+ std::error_code EC;
+ handleAllErrors(std::move(Err), [&](const ErrorInfoBase &EI) {
+ EC = EI.convertToErrorCode();
+ });
+ if (EC == inconvertibleErrorCode())
+ report_fatal_error(EC.message());
+ return EC;
+}
+
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+void Error::fatalUncheckedError() const {
+ dbgs() << "Program aborted due to an unhandled Error:\n";
+ if (getPtr())
+ getPtr()->log(dbgs());
+ else
+ dbgs() << "Error value was Success. (Note: Success values must still be "
+ "checked prior to being destroyed).\n";
+ abort();
+}
+#endif
+
+StringError::StringError(std::error_code EC, const Twine &S)
+ : Msg(S.str()), EC(EC) {}
+
+StringError::StringError(const Twine &S, std::error_code EC)
+ : Msg(S.str()), EC(EC), PrintMsgOnly(true) {}
+
+void StringError::log(raw_ostream &OS) const {
+ if (PrintMsgOnly) {
+ OS << Msg;
+ } else {
+ OS << EC.message();
+ if (!Msg.empty())
+ OS << (" " + Msg);
+ }
+}
+
+std::error_code StringError::convertToErrorCode() const {
+ return EC;
+}
+
+Error createStringError(std::error_code EC, char const *Msg) {
+ return make_error<StringError>(Msg, EC);
+}
+
+void report_fatal_error(Error Err, bool GenCrashDiag) {
+ assert(Err && "report_fatal_error called with success value");
+ std::string ErrMsg;
+ {
+ raw_string_ostream ErrStream(ErrMsg);
+ logAllUnhandledErrors(std::move(Err), ErrStream);
+ }
+ report_fatal_error(ErrMsg);
+}
+
+} // end namespace llvm
+
+LLVMErrorTypeId LLVMGetErrorTypeId(LLVMErrorRef Err) {
+ return reinterpret_cast<ErrorInfoBase *>(Err)->dynamicClassID();
+}
+
+void LLVMConsumeError(LLVMErrorRef Err) { consumeError(unwrap(Err)); }
+
+char *LLVMGetErrorMessage(LLVMErrorRef Err) {
+ std::string Tmp = toString(unwrap(Err));
+ char *ErrMsg = new char[Tmp.size() + 1];
+ memcpy(ErrMsg, Tmp.data(), Tmp.size());
+ ErrMsg[Tmp.size()] = '\0';
+ return ErrMsg;
+}
+
+void LLVMDisposeErrorMessage(char *ErrMsg) { delete[] ErrMsg; }
+
+LLVMErrorTypeId LLVMGetStringErrorTypeId() {
+ return reinterpret_cast<void *>(&StringError::ID);
+}
diff --git a/third_party/llvm-project/Error.h b/third_party/llvm-project/Error.h
new file mode 100644
index 000000000..8985338c9
--- /dev/null
+++ b/third_party/llvm-project/Error.h
@@ -0,0 +1,53 @@
+//===- Error.h - system_error extensions for obj2yaml -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_TOOLS_OBJ2YAML_ERROR_H
+#define LLVM_TOOLS_OBJ2YAML_ERROR_H
+
+#include "llvm/Support/Error.h"
+
+#include <system_error>
+
+namespace llvm {
+const std::error_category &obj2yaml_category();
+
+enum class obj2yaml_error {
+ success = 0,
+ file_not_found,
+ unrecognized_file_format,
+ unsupported_obj_file_format,
+ not_implemented
+};
+
+inline std::error_code make_error_code(obj2yaml_error e) {
+ return std::error_code(static_cast<int>(e), obj2yaml_category());
+}
+
+class Obj2YamlError : public ErrorInfo<Obj2YamlError> {
+public:
+ static char ID;
+ Obj2YamlError(obj2yaml_error C) : Code(C) {}
+ Obj2YamlError(std::string ErrMsg) : ErrMsg(std::move(ErrMsg)) {}
+ Obj2YamlError(obj2yaml_error C, std::string ErrMsg)
+ : ErrMsg(std::move(ErrMsg)), Code(C) {}
+ void log(raw_ostream &OS) const override;
+ const std::string &getErrorMessage() const { return ErrMsg; }
+ std::error_code convertToErrorCode() const override;
+
+private:
+ std::string ErrMsg;
+ obj2yaml_error Code = obj2yaml_error::success;
+};
+
+} // namespace llvm
+
+namespace std {
+template <> struct is_error_code_enum<llvm::obj2yaml_error> : std::true_type {};
+}
+
+#endif
diff --git a/third_party/llvm-project/ErrorHandling.cpp b/third_party/llvm-project/ErrorHandling.cpp
new file mode 100644
index 000000000..d7f546eec
--- /dev/null
+++ b/third_party/llvm-project/ErrorHandling.cpp
@@ -0,0 +1,305 @@
+//===- lib/Support/ErrorHandling.cpp - Callbacks for errors ---------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines an API used to indicate fatal error conditions. Non-fatal
+// errors (most of them) should be handled through LLVMContext.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm-c/ErrorHandling.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Config/config.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Signals.h"
+#include "llvm/Support/Threading.h"
+#include "llvm/Support/WindowsError.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cassert>
+#include <cstdlib>
+#include <mutex>
+#include <new>
+#include <iostream> // XXX BINARYEN
+
+#if defined(HAVE_UNISTD_H)
+# include <unistd.h>
+#endif
+#if defined(_MSC_VER)
+# include <io.h>
+# include <fcntl.h>
+#endif
+
+using namespace llvm;
+
+static fatal_error_handler_t ErrorHandler = nullptr;
+static void *ErrorHandlerUserData = nullptr;
+
+static fatal_error_handler_t BadAllocErrorHandler = nullptr;
+static void *BadAllocErrorHandlerUserData = nullptr;
+
+#if LLVM_ENABLE_THREADS == 1
+// Mutexes to synchronize installing error handlers and calling error handlers.
+// Do not use ManagedStatic, or that may allocate memory while attempting to
+// report an OOM.
+//
+// This usage of std::mutex has to be conditionalized behind ifdefs because
+// of this script:
+// compiler-rt/lib/sanitizer_common/symbolizer/scripts/build_symbolizer.sh
+// That script attempts to statically link the LLVM symbolizer library with the
+// STL and hide all of its symbols with 'opt -internalize'. To reduce size, it
+// cuts out the threading portions of the hermetic copy of libc++ that it
+// builds. We can remove these ifdefs if that script goes away.
+static std::mutex ErrorHandlerMutex;
+static std::mutex BadAllocErrorHandlerMutex;
+#endif
+
+void llvm::install_fatal_error_handler(fatal_error_handler_t handler,
+ void *user_data) {
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(ErrorHandlerMutex);
+#endif
+ assert(!ErrorHandler && "Error handler already registered!\n");
+ ErrorHandler = handler;
+ ErrorHandlerUserData = user_data;
+}
+
+void llvm::remove_fatal_error_handler() {
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(ErrorHandlerMutex);
+#endif
+ ErrorHandler = nullptr;
+ ErrorHandlerUserData = nullptr;
+}
+
+void llvm::report_fatal_error(const char *Reason, bool GenCrashDiag) {
+ report_fatal_error(Twine(Reason), GenCrashDiag);
+}
+
+void llvm::report_fatal_error(const std::string &Reason, bool GenCrashDiag) {
+ report_fatal_error(Twine(Reason), GenCrashDiag);
+}
+
+void llvm::report_fatal_error(StringRef Reason, bool GenCrashDiag) {
+ report_fatal_error(Twine(Reason), GenCrashDiag);
+}
+
+void llvm::report_fatal_error(const Twine &Reason, bool GenCrashDiag) {
+ Reason.dump();
+#if 0 // XXX BINARYEN
+ llvm::fatal_error_handler_t handler = nullptr;
+ void* handlerData = nullptr;
+ {
+ // Only acquire the mutex while reading the handler, so as not to invoke a
+ // user-supplied callback under a lock.
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(ErrorHandlerMutex);
+#endif
+ handler = ErrorHandler;
+ handlerData = ErrorHandlerUserData;
+ }
+
+ if (handler) {
+ handler(handlerData, Reason.str(), GenCrashDiag);
+ } else {
+ // Blast the result out to stderr. We don't try hard to make sure this
+ // succeeds (e.g. handling EINTR) and we can't use errs() here because
+ // raw ostreams can call report_fatal_error.
+ SmallVector<char, 64> Buffer;
+ raw_svector_ostream OS(Buffer);
+ OS << "LLVM ERROR: " << Reason << "\n";
+ ssize_t written = ::write(2, MessageStr.data(), MessageStr.size());
+ StringRef MessageStr = OS.str();
+ (void)written; // If something went wrong, we deliberately just give up.
+ }
+
+ // If we reached here, we are failing ungracefully. Run the interrupt handlers
+ // to make sure any special cleanups get done, in particular that we remove
+ // files registered with RemoveFileOnSignal.
+ sys::RunInterruptHandlers();
+#else
+ Reason.dump();
+#endif
+
+ exit(1);
+}
+
+void llvm::install_bad_alloc_error_handler(fatal_error_handler_t handler,
+ void *user_data) {
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(BadAllocErrorHandlerMutex);
+#endif
+ assert(!ErrorHandler && "Bad alloc error handler already registered!\n");
+ BadAllocErrorHandler = handler;
+ BadAllocErrorHandlerUserData = user_data;
+}
+
+void llvm::remove_bad_alloc_error_handler() {
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(BadAllocErrorHandlerMutex);
+#endif
+ BadAllocErrorHandler = nullptr;
+ BadAllocErrorHandlerUserData = nullptr;
+}
+
+void llvm::report_bad_alloc_error(const char *Reason, bool GenCrashDiag) {
+ fatal_error_handler_t Handler = nullptr;
+ void *HandlerData = nullptr;
+ {
+ // Only acquire the mutex while reading the handler, so as not to invoke a
+ // user-supplied callback under a lock.
+#if LLVM_ENABLE_THREADS == 1
+ std::lock_guard<std::mutex> Lock(BadAllocErrorHandlerMutex);
+#endif
+ Handler = BadAllocErrorHandler;
+ HandlerData = BadAllocErrorHandlerUserData;
+ }
+
+ if (Handler) {
+ Handler(HandlerData, Reason, GenCrashDiag);
+ llvm_unreachable("bad alloc handler should not return");
+ }
+
+#ifdef LLVM_ENABLE_EXCEPTIONS
+ // If exceptions are enabled, make OOM in malloc look like OOM in new.
+ throw std::bad_alloc();
+#else
+ // Don't call the normal error handler. It may allocate memory. Directly write
+ // an OOM to stderr and abort.
+ char OOMMessage[] = "LLVM ERROR: out of memory\n";
+ ssize_t written = ::write(2, OOMMessage, strlen(OOMMessage));
+ (void)written;
+ abort();
+#endif
+}
+
+#ifdef LLVM_ENABLE_EXCEPTIONS
+// Do not set custom new handler if exceptions are enabled. In this case OOM
+// errors are handled by throwing 'std::bad_alloc'.
+void llvm::install_out_of_memory_new_handler() {
+}
+#else
+// Causes crash on allocation failure. It is called prior to the handler set by
+// 'install_bad_alloc_error_handler'.
+static void out_of_memory_new_handler() {
+ llvm::report_bad_alloc_error("Allocation failed");
+}
+
+// Installs new handler that causes crash on allocation failure. It is called by
+// InitLLVM.
+void llvm::install_out_of_memory_new_handler() {
+ std::new_handler old = std::set_new_handler(out_of_memory_new_handler);
+ (void)old;
+ assert(old == nullptr && "new-handler already installed");
+}
+#endif
+
+void llvm::llvm_unreachable_internal(const char *msg, const char *file,
+ unsigned line) {
+ // This code intentionally doesn't call the ErrorHandler callback, because
+ // llvm_unreachable is intended to be used to indicate "impossible"
+ // situations, and not legitimate runtime errors.
+ // XXX BINARYEN: use cout
+ if (msg)
+ std::cout << msg << "\n";
+ std::cout << "UNREACHABLE executed";
+ if (file)
+ std::cout << " at " << file << ":" << line;
+ std::cout << "!\n";
+ abort();
+#ifdef LLVM_BUILTIN_UNREACHABLE
+ // Windows systems and possibly others don't declare abort() to be noreturn,
+ // so use the unreachable builtin to avoid a Clang self-host warning.
+ LLVM_BUILTIN_UNREACHABLE;
+#endif
+}
+
+static void bindingsErrorHandler(void *user_data, const std::string& reason,
+ bool gen_crash_diag) {
+ LLVMFatalErrorHandler handler =
+ LLVM_EXTENSION reinterpret_cast<LLVMFatalErrorHandler>(user_data);
+ handler(reason.c_str());
+}
+
+void LLVMInstallFatalErrorHandler(LLVMFatalErrorHandler Handler) {
+ install_fatal_error_handler(bindingsErrorHandler,
+ LLVM_EXTENSION reinterpret_cast<void *>(Handler));
+}
+
+void LLVMResetFatalErrorHandler() {
+ remove_fatal_error_handler();
+}
+
+#ifdef _WIN32
+
+#include <winerror.h>
+
+// I'd rather not double the line count of the following.
+#define MAP_ERR_TO_COND(x, y) \
+ case x: \
+ return make_error_code(errc::y)
+
+std::error_code llvm::mapWindowsError(unsigned EV) {
+ switch (EV) {
+ MAP_ERR_TO_COND(ERROR_ACCESS_DENIED, permission_denied);
+ MAP_ERR_TO_COND(ERROR_ALREADY_EXISTS, file_exists);
+ MAP_ERR_TO_COND(ERROR_BAD_UNIT, no_such_device);
+ MAP_ERR_TO_COND(ERROR_BUFFER_OVERFLOW, filename_too_long);
+ MAP_ERR_TO_COND(ERROR_BUSY, device_or_resource_busy);
+ MAP_ERR_TO_COND(ERROR_BUSY_DRIVE, device_or_resource_busy);
+ MAP_ERR_TO_COND(ERROR_CANNOT_MAKE, permission_denied);
+ MAP_ERR_TO_COND(ERROR_CANTOPEN, io_error);
+ MAP_ERR_TO_COND(ERROR_CANTREAD, io_error);
+ MAP_ERR_TO_COND(ERROR_CANTWRITE, io_error);
+ MAP_ERR_TO_COND(ERROR_CURRENT_DIRECTORY, permission_denied);
+ MAP_ERR_TO_COND(ERROR_DEV_NOT_EXIST, no_such_device);
+ MAP_ERR_TO_COND(ERROR_DEVICE_IN_USE, device_or_resource_busy);
+ MAP_ERR_TO_COND(ERROR_DIR_NOT_EMPTY, directory_not_empty);
+ MAP_ERR_TO_COND(ERROR_DIRECTORY, invalid_argument);
+ MAP_ERR_TO_COND(ERROR_DISK_FULL, no_space_on_device);
+ MAP_ERR_TO_COND(ERROR_FILE_EXISTS, file_exists);
+ MAP_ERR_TO_COND(ERROR_FILE_NOT_FOUND, no_such_file_or_directory);
+ MAP_ERR_TO_COND(ERROR_HANDLE_DISK_FULL, no_space_on_device);
+ MAP_ERR_TO_COND(ERROR_INVALID_ACCESS, permission_denied);
+ MAP_ERR_TO_COND(ERROR_INVALID_DRIVE, no_such_device);
+ MAP_ERR_TO_COND(ERROR_INVALID_FUNCTION, function_not_supported);
+ MAP_ERR_TO_COND(ERROR_INVALID_HANDLE, invalid_argument);
+ MAP_ERR_TO_COND(ERROR_INVALID_NAME, invalid_argument);
+ MAP_ERR_TO_COND(ERROR_LOCK_VIOLATION, no_lock_available);
+ MAP_ERR_TO_COND(ERROR_LOCKED, no_lock_available);
+ MAP_ERR_TO_COND(ERROR_NEGATIVE_SEEK, invalid_argument);
+ MAP_ERR_TO_COND(ERROR_NOACCESS, permission_denied);
+ MAP_ERR_TO_COND(ERROR_NOT_ENOUGH_MEMORY, not_enough_memory);
+ MAP_ERR_TO_COND(ERROR_NOT_READY, resource_unavailable_try_again);
+ MAP_ERR_TO_COND(ERROR_OPEN_FAILED, io_error);
+ MAP_ERR_TO_COND(ERROR_OPEN_FILES, device_or_resource_busy);
+ MAP_ERR_TO_COND(ERROR_OUTOFMEMORY, not_enough_memory);
+ MAP_ERR_TO_COND(ERROR_PATH_NOT_FOUND, no_such_file_or_directory);
+ MAP_ERR_TO_COND(ERROR_BAD_NETPATH, no_such_file_or_directory);
+ MAP_ERR_TO_COND(ERROR_READ_FAULT, io_error);
+ MAP_ERR_TO_COND(ERROR_RETRY, resource_unavailable_try_again);
+ MAP_ERR_TO_COND(ERROR_SEEK, io_error);
+ MAP_ERR_TO_COND(ERROR_SHARING_VIOLATION, permission_denied);
+ MAP_ERR_TO_COND(ERROR_TOO_MANY_OPEN_FILES, too_many_files_open);
+ MAP_ERR_TO_COND(ERROR_WRITE_FAULT, io_error);
+ MAP_ERR_TO_COND(ERROR_WRITE_PROTECT, permission_denied);
+ MAP_ERR_TO_COND(WSAEACCES, permission_denied);
+ MAP_ERR_TO_COND(WSAEBADF, bad_file_descriptor);
+ MAP_ERR_TO_COND(WSAEFAULT, bad_address);
+ MAP_ERR_TO_COND(WSAEINTR, interrupted);
+ MAP_ERR_TO_COND(WSAEINVAL, invalid_argument);
+ MAP_ERR_TO_COND(WSAEMFILE, too_many_files_open);
+ MAP_ERR_TO_COND(WSAENAMETOOLONG, filename_too_long);
+ default:
+ return std::error_code(EV, std::system_category());
+ }
+}
+
+#endif
diff --git a/third_party/llvm-project/FormatVariadic.cpp b/third_party/llvm-project/FormatVariadic.cpp
new file mode 100644
index 000000000..f9e89f69b
--- /dev/null
+++ b/third_party/llvm-project/FormatVariadic.cpp
@@ -0,0 +1,155 @@
+//===- FormatVariadic.cpp - Format string parsing and analysis ----*-C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/FormatVariadic.h"
+
+using namespace llvm;
+
+static Optional<AlignStyle> translateLocChar(char C) {
+ switch (C) {
+ case '-':
+ return AlignStyle::Left;
+ case '=':
+ return AlignStyle::Center;
+ case '+':
+ return AlignStyle::Right;
+ default:
+ return None;
+ }
+ LLVM_BUILTIN_UNREACHABLE;
+}
+
+bool formatv_object_base::consumeFieldLayout(StringRef &Spec, AlignStyle &Where,
+ size_t &Align, char &Pad) {
+ Where = AlignStyle::Right;
+ Align = 0;
+ Pad = ' ';
+ if (Spec.empty())
+ return true;
+
+ if (Spec.size() > 1) {
+ // A maximum of 2 characters at the beginning can be used for something
+ // other
+ // than the width.
+ // If Spec[1] is a loc char, then Spec[0] is a pad char and Spec[2:...]
+ // contains the width.
+ // Otherwise, if Spec[0] is a loc char, then Spec[1:...] contains the width.
+ // Otherwise, Spec[0:...] contains the width.
+ if (auto Loc = translateLocChar(Spec[1])) {
+ Pad = Spec[0];
+ Where = *Loc;
+ Spec = Spec.drop_front(2);
+ } else if (auto Loc = translateLocChar(Spec[0])) {
+ Where = *Loc;
+ Spec = Spec.drop_front(1);
+ }
+ }
+
+ bool Failed = Spec.consumeInteger(0, Align);
+ return !Failed;
+}
+
+Optional<ReplacementItem>
+formatv_object_base::parseReplacementItem(StringRef Spec) {
+ StringRef RepString = Spec.trim("{}");
+
+ // If the replacement sequence does not start with a non-negative integer,
+ // this is an error.
+ char Pad = ' ';
+ std::size_t Align = 0;
+ AlignStyle Where = AlignStyle::Right;
+ StringRef Options;
+ size_t Index = 0;
+ RepString = RepString.trim();
+ if (RepString.consumeInteger(0, Index)) {
+ assert(false && "Invalid replacement sequence index!");
+ return ReplacementItem{};
+ }
+ RepString = RepString.trim();
+ if (!RepString.empty() && RepString.front() == ',') {
+ RepString = RepString.drop_front();
+ if (!consumeFieldLayout(RepString, Where, Align, Pad))
+ assert(false && "Invalid replacement field layout specification!");
+ }
+ RepString = RepString.trim();
+ if (!RepString.empty() && RepString.front() == ':') {
+ Options = RepString.drop_front().trim();
+ RepString = StringRef();
+ }
+ RepString = RepString.trim();
+ if (!RepString.empty()) {
+ assert(false && "Unexpected characters found in replacement string!");
+ }
+
+ return ReplacementItem{Spec, Index, Align, Where, Pad, Options};
+}
+
+std::pair<ReplacementItem, StringRef>
+formatv_object_base::splitLiteralAndReplacement(StringRef Fmt) {
+ std::size_t From = 0;
+ while (From < Fmt.size() && From != StringRef::npos) {
+ std::size_t BO = Fmt.find_first_of('{', From);
+ // Everything up until the first brace is a literal.
+ if (BO != 0)
+ return std::make_pair(ReplacementItem{Fmt.substr(0, BO)}, Fmt.substr(BO));
+
+ StringRef Braces =
+ Fmt.drop_front(BO).take_while([](char C) { return C == '{'; });
+ // If there is more than one brace, then some of them are escaped. Treat
+ // these as replacements.
+ if (Braces.size() > 1) {
+ size_t NumEscapedBraces = Braces.size() / 2;
+ StringRef Middle = Fmt.substr(BO, NumEscapedBraces);
+ StringRef Right = Fmt.drop_front(BO + NumEscapedBraces * 2);
+ return std::make_pair(ReplacementItem{Middle}, Right);
+ }
+ // An unterminated open brace is undefined. We treat the rest of the string
+ // as a literal replacement, but we assert to indicate that this is
+ // undefined and that we consider it an error.
+ std::size_t BC = Fmt.find_first_of('}', BO);
+ if (BC == StringRef::npos) {
+ assert(
+ false &&
+ "Unterminated brace sequence. Escape with {{ for a literal brace.");
+ return std::make_pair(ReplacementItem{Fmt}, StringRef());
+ }
+
+ // Even if there is a closing brace, if there is another open brace before
+ // this closing brace, treat this portion as literal, and try again with the
+ // next one.
+ std::size_t BO2 = Fmt.find_first_of('{', BO + 1);
+ if (BO2 < BC)
+ return std::make_pair(ReplacementItem{Fmt.substr(0, BO2)},
+ Fmt.substr(BO2));
+
+ StringRef Spec = Fmt.slice(BO + 1, BC);
+ StringRef Right = Fmt.substr(BC + 1);
+
+ auto RI = parseReplacementItem(Spec);
+ if (RI.hasValue())
+ return std::make_pair(*RI, Right);
+
+ // If there was an error parsing the replacement item, treat it as an
+ // invalid replacement spec, and just continue.
+ From = BC + 1;
+ }
+ return std::make_pair(ReplacementItem{Fmt}, StringRef());
+}
+
+std::vector<ReplacementItem>
+formatv_object_base::parseFormatString(StringRef Fmt) {
+ std::vector<ReplacementItem> Replacements;
+ ReplacementItem I;
+ while (!Fmt.empty()) {
+ std::tie(I, Fmt) = splitLiteralAndReplacement(Fmt);
+ if (I.Type != ReplacementType::Empty)
+ Replacements.push_back(I);
+ }
+ return Replacements;
+}
+
+void detail::format_adapter::anchor() { }
diff --git a/third_party/llvm-project/Hashing.cpp b/third_party/llvm-project/Hashing.cpp
new file mode 100644
index 000000000..1b20a6704
--- /dev/null
+++ b/third_party/llvm-project/Hashing.cpp
@@ -0,0 +1,28 @@
+//===-------------- lib/Support/Hashing.cpp -------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides implementation bits for the LLVM common hashing
+// infrastructure. Documentation and most of the other information is in the
+// header file.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/Hashing.h"
+
+using namespace llvm;
+
+// Provide a definition and static initializer for the fixed seed. This
+// initializer should always be zero to ensure its value can never appear to be
+// non-zero, even during dynamic initialization.
+uint64_t llvm::hashing::detail::fixed_seed_override = 0;
+
+// Implement the function for forced setting of the fixed seed.
+// FIXME: Use atomic operations here so that there is no data race.
+void llvm::set_fixed_execution_hash_seed(uint64_t fixed_value) {
+ hashing::detail::fixed_seed_override = fixed_value;
+}
diff --git a/third_party/llvm-project/LEB128.cpp b/third_party/llvm-project/LEB128.cpp
new file mode 100644
index 000000000..d41b673e9
--- /dev/null
+++ b/third_party/llvm-project/LEB128.cpp
@@ -0,0 +1,43 @@
+//===- LEB128.cpp - LEB128 utility functions implementation -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements some utility functions for encoding SLEB128 and
+// ULEB128 values.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/LEB128.h"
+
+namespace llvm {
+
+/// Utility function to get the size of the ULEB128-encoded value.
+unsigned getULEB128Size(uint64_t Value) {
+ unsigned Size = 0;
+ do {
+ Value >>= 7;
+ Size += sizeof(int8_t);
+ } while (Value);
+ return Size;
+}
+
+/// Utility function to get the size of the SLEB128-encoded value.
+unsigned getSLEB128Size(int64_t Value) {
+ unsigned Size = 0;
+ int Sign = Value >> (8 * sizeof(Value) - 1);
+ bool IsMore;
+
+ do {
+ unsigned Byte = Value & 0x7f;
+ Value >>= 7;
+ IsMore = Value != Sign || ((Byte ^ Sign) & 0x40) != 0;
+ Size += sizeof(int8_t);
+ } while (IsMore);
+ return Size;
+}
+
+} // namespace llvm
diff --git a/third_party/llvm-project/LineIterator.cpp b/third_party/llvm-project/LineIterator.cpp
new file mode 100644
index 000000000..164436a2c
--- /dev/null
+++ b/third_party/llvm-project/LineIterator.cpp
@@ -0,0 +1,93 @@
+//===- LineIterator.cpp - Implementation of line iteration ----------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/LineIterator.h"
+#include "llvm/Support/MemoryBuffer.h"
+
+using namespace llvm;
+
+static bool isAtLineEnd(const char *P) {
+ if (*P == '\n')
+ return true;
+ if (*P == '\r' && *(P + 1) == '\n')
+ return true;
+ return false;
+}
+
+static bool skipIfAtLineEnd(const char *&P) {
+ if (*P == '\n') {
+ ++P;
+ return true;
+ }
+ if (*P == '\r' && *(P + 1) == '\n') {
+ P += 2;
+ return true;
+ }
+ return false;
+}
+
+line_iterator::line_iterator(const MemoryBuffer &Buffer, bool SkipBlanks,
+ char CommentMarker)
+ : Buffer(Buffer.getBufferSize() ? &Buffer : nullptr),
+ CommentMarker(CommentMarker), SkipBlanks(SkipBlanks), LineNumber(1),
+ CurrentLine(Buffer.getBufferSize() ? Buffer.getBufferStart() : nullptr,
+ 0) {
+ // Ensure that if we are constructed on a non-empty memory buffer that it is
+ // a null terminated buffer.
+ if (Buffer.getBufferSize()) {
+ assert(Buffer.getBufferEnd()[0] == '\0');
+ // Make sure we don't skip a leading newline if we're keeping blanks
+ if (SkipBlanks || !isAtLineEnd(Buffer.getBufferStart()))
+ advance();
+ }
+}
+
+void line_iterator::advance() {
+ assert(Buffer && "Cannot advance past the end!");
+
+ const char *Pos = CurrentLine.end();
+ assert(Pos == Buffer->getBufferStart() || isAtLineEnd(Pos) || *Pos == '\0');
+
+ if (skipIfAtLineEnd(Pos))
+ ++LineNumber;
+ if (!SkipBlanks && isAtLineEnd(Pos)) {
+ // Nothing to do for a blank line.
+ } else if (CommentMarker == '\0') {
+ // If we're not stripping comments, this is simpler.
+ while (skipIfAtLineEnd(Pos))
+ ++LineNumber;
+ } else {
+ // Skip comments and count line numbers, which is a bit more complex.
+ for (;;) {
+ if (isAtLineEnd(Pos) && !SkipBlanks)
+ break;
+ if (*Pos == CommentMarker)
+ do {
+ ++Pos;
+ } while (*Pos != '\0' && !isAtLineEnd(Pos));
+ if (!skipIfAtLineEnd(Pos))
+ break;
+ ++LineNumber;
+ }
+ }
+
+ if (*Pos == '\0') {
+ // We've hit the end of the buffer, reset ourselves to the end state.
+ Buffer = nullptr;
+ CurrentLine = StringRef();
+ return;
+ }
+
+ // Measure the line.
+ size_t Length = 0;
+ while (Pos[Length] != '\0' && !isAtLineEnd(&Pos[Length])) {
+ ++Length;
+ }
+
+ CurrentLine = StringRef(Pos, Length);
+}
diff --git a/third_party/llvm-project/MCRegisterInfo.cpp b/third_party/llvm-project/MCRegisterInfo.cpp
new file mode 100644
index 000000000..d491c0eb7
--- /dev/null
+++ b/third_party/llvm-project/MCRegisterInfo.cpp
@@ -0,0 +1,124 @@
+//===- MC/MCRegisterInfo.cpp - Target Register Description ----------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements MCRegisterInfo functions.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+
+using namespace llvm;
+
+MCRegister
+MCRegisterInfo::getMatchingSuperReg(MCRegister Reg, unsigned SubIdx,
+ const MCRegisterClass *RC) const {
+ for (MCSuperRegIterator Supers(Reg, this); Supers.isValid(); ++Supers)
+ if (RC->contains(*Supers) && Reg == getSubReg(*Supers, SubIdx))
+ return *Supers;
+ return 0;
+}
+
+MCRegister MCRegisterInfo::getSubReg(MCRegister Reg, unsigned Idx) const {
+ assert(Idx && Idx < getNumSubRegIndices() &&
+ "This is not a subregister index");
+ // Get a pointer to the corresponding SubRegIndices list. This list has the
+ // name of each sub-register in the same order as MCSubRegIterator.
+ const uint16_t *SRI = SubRegIndices + get(Reg).SubRegIndices;
+ for (MCSubRegIterator Subs(Reg, this); Subs.isValid(); ++Subs, ++SRI)
+ if (*SRI == Idx)
+ return *Subs;
+ return 0;
+}
+
+unsigned MCRegisterInfo::getSubRegIndex(MCRegister Reg,
+ MCRegister SubReg) const {
+ assert(SubReg && SubReg < getNumRegs() && "This is not a register");
+ // Get a pointer to the corresponding SubRegIndices list. This list has the
+ // name of each sub-register in the same order as MCSubRegIterator.
+ const uint16_t *SRI = SubRegIndices + get(Reg).SubRegIndices;
+ for (MCSubRegIterator Subs(Reg, this); Subs.isValid(); ++Subs, ++SRI)
+ if (*Subs == SubReg)
+ return *SRI;
+ return 0;
+}
+
+unsigned MCRegisterInfo::getSubRegIdxSize(unsigned Idx) const {
+ assert(Idx && Idx < getNumSubRegIndices() &&
+ "This is not a subregister index");
+ return SubRegIdxRanges[Idx].Size;
+}
+
+unsigned MCRegisterInfo::getSubRegIdxOffset(unsigned Idx) const {
+ assert(Idx && Idx < getNumSubRegIndices() &&
+ "This is not a subregister index");
+ return SubRegIdxRanges[Idx].Offset;
+}
+
+int MCRegisterInfo::getDwarfRegNum(MCRegister RegNum, bool isEH) const {
+ const DwarfLLVMRegPair *M = isEH ? EHL2DwarfRegs : L2DwarfRegs;
+ unsigned Size = isEH ? EHL2DwarfRegsSize : L2DwarfRegsSize;
+
+ if (!M)
+ return -1;
+ DwarfLLVMRegPair Key = { RegNum, 0 };
+ const DwarfLLVMRegPair *I = std::lower_bound(M, M+Size, Key);
+ if (I == M+Size || I->FromReg != RegNum)
+ return -1;
+ return I->ToReg;
+}
+
+Optional<unsigned> MCRegisterInfo::getLLVMRegNum(unsigned RegNum,
+ bool isEH) const {
+ const DwarfLLVMRegPair *M = isEH ? EHDwarf2LRegs : Dwarf2LRegs;
+ unsigned Size = isEH ? EHDwarf2LRegsSize : Dwarf2LRegsSize;
+
+ if (!M)
+ return None;
+ DwarfLLVMRegPair Key = { RegNum, 0 };
+ const DwarfLLVMRegPair *I = std::lower_bound(M, M+Size, Key);
+ if (I != M + Size && I->FromReg == RegNum)
+ return I->ToReg;
+ return None;
+}
+
+int MCRegisterInfo::getDwarfRegNumFromDwarfEHRegNum(unsigned RegNum) const {
+ // On ELF platforms, DWARF EH register numbers are the same as DWARF
+ // other register numbers. On Darwin x86, they differ and so need to be
+ // mapped. The .cfi_* directives accept integer literals as well as
+ // register names and should generate exactly what the assembly code
+ // asked for, so there might be DWARF/EH register numbers that don't have
+ // a corresponding LLVM register number at all. So if we can't map the
+ // EH register number to an LLVM register number, assume it's just a
+ // valid DWARF register number as is.
+ if (Optional<unsigned> LRegNum = getLLVMRegNum(RegNum, true))
+ return getDwarfRegNum(*LRegNum, false);
+ return RegNum;
+}
+
+int MCRegisterInfo::getSEHRegNum(MCRegister RegNum) const {
+ const DenseMap<MCRegister, int>::const_iterator I = L2SEHRegs.find(RegNum);
+ if (I == L2SEHRegs.end()) return (int)RegNum;
+ return I->second;
+}
+
+int MCRegisterInfo::getCodeViewRegNum(MCRegister RegNum) const {
+ if (L2CVRegs.empty())
+ report_fatal_error("target does not implement codeview register mapping");
+ const DenseMap<MCRegister, int>::const_iterator I = L2CVRegs.find(RegNum);
+ if (I == L2CVRegs.end())
+ report_fatal_error("unknown codeview register " + (RegNum < getNumRegs()
+ ? getName(RegNum)
+ : Twine(RegNum)));
+ return I->second;
+}
diff --git a/third_party/llvm-project/MD5.cpp b/third_party/llvm-project/MD5.cpp
new file mode 100644
index 000000000..9b02f6291
--- /dev/null
+++ b/third_party/llvm-project/MD5.cpp
@@ -0,0 +1,283 @@
+/*
+ * This code is derived from (original license follows):
+ *
+ * This is an OpenSSL-compatible implementation of the RSA Data Security, Inc.
+ * MD5 Message-Digest Algorithm (RFC 1321).
+ *
+ * Homepage:
+ * http://openwall.info/wiki/people/solar/software/public-domain-source-code/md5
+ *
+ * Author:
+ * Alexander Peslyak, better known as Solar Designer <solar at openwall.com>
+ *
+ * This software was written by Alexander Peslyak in 2001. No copyright is
+ * claimed, and the software is hereby placed in the public domain.
+ * In case this attempt to disclaim copyright and place the software in the
+ * public domain is deemed null and void, then the software is
+ * Copyright (c) 2001 Alexander Peslyak and it is hereby released to the
+ * general public under the following terms:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * There's ABSOLUTELY NO WARRANTY, express or implied.
+ *
+ * (This is a heavily cut-down "BSD license".)
+ *
+ * This differs from Colin Plumb's older public domain implementation in that
+ * no exactly 32-bit integer data type is required (any 32-bit or wider
+ * unsigned integer data type will do), there's no compile-time endianness
+ * configuration, and the function prototypes match OpenSSL's. No code from
+ * Colin Plumb's implementation has been reused; this comment merely compares
+ * the properties of the two independent implementations.
+ *
+ * The primary goals of this implementation are portability and ease of use.
+ * It is meant to be fast, but not as fast as possible. Some known
+ * optimizations are not included to reduce source code size and avoid
+ * compile-time configuration.
+ */
+
+#include "llvm/Support/MD5.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <array>
+#include <cstdint>
+#include <cstring>
+
+// The basic MD5 functions.
+
+// F and G are optimized compared to their RFC 1321 definitions for
+// architectures that lack an AND-NOT instruction, just like in Colin Plumb's
+// implementation.
+#define F(x, y, z) ((z) ^ ((x) & ((y) ^ (z))))
+#define G(x, y, z) ((y) ^ ((z) & ((x) ^ (y))))
+#define H(x, y, z) ((x) ^ (y) ^ (z))
+#define I(x, y, z) ((y) ^ ((x) | ~(z)))
+
+// The MD5 transformation for all four rounds.
+#define STEP(f, a, b, c, d, x, t, s) \
+ (a) += f((b), (c), (d)) + (x) + (t); \
+ (a) = (((a) << (s)) | (((a) & 0xffffffff) >> (32 - (s)))); \
+ (a) += (b);
+
+// SET reads 4 input bytes in little-endian byte order and stores them
+// in a properly aligned word in host byte order.
+#define SET(n) \
+ (block[(n)] = \
+ (MD5_u32plus) ptr[(n) * 4] | ((MD5_u32plus) ptr[(n) * 4 + 1] << 8) | \
+ ((MD5_u32plus) ptr[(n) * 4 + 2] << 16) | \
+ ((MD5_u32plus) ptr[(n) * 4 + 3] << 24))
+#define GET(n) (block[(n)])
+
+using namespace llvm;
+
+/// This processes one or more 64-byte data blocks, but does NOT update
+///the bit counters. There are no alignment requirements.
+const uint8_t *MD5::body(ArrayRef<uint8_t> Data) {
+ const uint8_t *ptr;
+ MD5_u32plus a, b, c, d;
+ MD5_u32plus saved_a, saved_b, saved_c, saved_d;
+ unsigned long Size = Data.size();
+
+ ptr = Data.data();
+
+ a = this->a;
+ b = this->b;
+ c = this->c;
+ d = this->d;
+
+ do {
+ saved_a = a;
+ saved_b = b;
+ saved_c = c;
+ saved_d = d;
+
+ // Round 1
+ STEP(F, a, b, c, d, SET(0), 0xd76aa478, 7)
+ STEP(F, d, a, b, c, SET(1), 0xe8c7b756, 12)
+ STEP(F, c, d, a, b, SET(2), 0x242070db, 17)
+ STEP(F, b, c, d, a, SET(3), 0xc1bdceee, 22)
+ STEP(F, a, b, c, d, SET(4), 0xf57c0faf, 7)
+ STEP(F, d, a, b, c, SET(5), 0x4787c62a, 12)
+ STEP(F, c, d, a, b, SET(6), 0xa8304613, 17)
+ STEP(F, b, c, d, a, SET(7), 0xfd469501, 22)
+ STEP(F, a, b, c, d, SET(8), 0x698098d8, 7)
+ STEP(F, d, a, b, c, SET(9), 0x8b44f7af, 12)
+ STEP(F, c, d, a, b, SET(10), 0xffff5bb1, 17)
+ STEP(F, b, c, d, a, SET(11), 0x895cd7be, 22)
+ STEP(F, a, b, c, d, SET(12), 0x6b901122, 7)
+ STEP(F, d, a, b, c, SET(13), 0xfd987193, 12)
+ STEP(F, c, d, a, b, SET(14), 0xa679438e, 17)
+ STEP(F, b, c, d, a, SET(15), 0x49b40821, 22)
+
+ // Round 2
+ STEP(G, a, b, c, d, GET(1), 0xf61e2562, 5)
+ STEP(G, d, a, b, c, GET(6), 0xc040b340, 9)
+ STEP(G, c, d, a, b, GET(11), 0x265e5a51, 14)
+ STEP(G, b, c, d, a, GET(0), 0xe9b6c7aa, 20)
+ STEP(G, a, b, c, d, GET(5), 0xd62f105d, 5)
+ STEP(G, d, a, b, c, GET(10), 0x02441453, 9)
+ STEP(G, c, d, a, b, GET(15), 0xd8a1e681, 14)
+ STEP(G, b, c, d, a, GET(4), 0xe7d3fbc8, 20)
+ STEP(G, a, b, c, d, GET(9), 0x21e1cde6, 5)
+ STEP(G, d, a, b, c, GET(14), 0xc33707d6, 9)
+ STEP(G, c, d, a, b, GET(3), 0xf4d50d87, 14)
+ STEP(G, b, c, d, a, GET(8), 0x455a14ed, 20)
+ STEP(G, a, b, c, d, GET(13), 0xa9e3e905, 5)
+ STEP(G, d, a, b, c, GET(2), 0xfcefa3f8, 9)
+ STEP(G, c, d, a, b, GET(7), 0x676f02d9, 14)
+ STEP(G, b, c, d, a, GET(12), 0x8d2a4c8a, 20)
+
+ // Round 3
+ STEP(H, a, b, c, d, GET(5), 0xfffa3942, 4)
+ STEP(H, d, a, b, c, GET(8), 0x8771f681, 11)
+ STEP(H, c, d, a, b, GET(11), 0x6d9d6122, 16)
+ STEP(H, b, c, d, a, GET(14), 0xfde5380c, 23)
+ STEP(H, a, b, c, d, GET(1), 0xa4beea44, 4)
+ STEP(H, d, a, b, c, GET(4), 0x4bdecfa9, 11)
+ STEP(H, c, d, a, b, GET(7), 0xf6bb4b60, 16)
+ STEP(H, b, c, d, a, GET(10), 0xbebfbc70, 23)
+ STEP(H, a, b, c, d, GET(13), 0x289b7ec6, 4)
+ STEP(H, d, a, b, c, GET(0), 0xeaa127fa, 11)
+ STEP(H, c, d, a, b, GET(3), 0xd4ef3085, 16)
+ STEP(H, b, c, d, a, GET(6), 0x04881d05, 23)
+ STEP(H, a, b, c, d, GET(9), 0xd9d4d039, 4)
+ STEP(H, d, a, b, c, GET(12), 0xe6db99e5, 11)
+ STEP(H, c, d, a, b, GET(15), 0x1fa27cf8, 16)
+ STEP(H, b, c, d, a, GET(2), 0xc4ac5665, 23)
+
+ // Round 4
+ STEP(I, a, b, c, d, GET(0), 0xf4292244, 6)
+ STEP(I, d, a, b, c, GET(7), 0x432aff97, 10)
+ STEP(I, c, d, a, b, GET(14), 0xab9423a7, 15)
+ STEP(I, b, c, d, a, GET(5), 0xfc93a039, 21)
+ STEP(I, a, b, c, d, GET(12), 0x655b59c3, 6)
+ STEP(I, d, a, b, c, GET(3), 0x8f0ccc92, 10)
+ STEP(I, c, d, a, b, GET(10), 0xffeff47d, 15)
+ STEP(I, b, c, d, a, GET(1), 0x85845dd1, 21)
+ STEP(I, a, b, c, d, GET(8), 0x6fa87e4f, 6)
+ STEP(I, d, a, b, c, GET(15), 0xfe2ce6e0, 10)
+ STEP(I, c, d, a, b, GET(6), 0xa3014314, 15)
+ STEP(I, b, c, d, a, GET(13), 0x4e0811a1, 21)
+ STEP(I, a, b, c, d, GET(4), 0xf7537e82, 6)
+ STEP(I, d, a, b, c, GET(11), 0xbd3af235, 10)
+ STEP(I, c, d, a, b, GET(2), 0x2ad7d2bb, 15)
+ STEP(I, b, c, d, a, GET(9), 0xeb86d391, 21)
+
+ a += saved_a;
+ b += saved_b;
+ c += saved_c;
+ d += saved_d;
+
+ ptr += 64;
+ } while (Size -= 64);
+
+ this->a = a;
+ this->b = b;
+ this->c = c;
+ this->d = d;
+
+ return ptr;
+}
+
+MD5::MD5() = default;
+
+/// Incrementally add the bytes in \p Data to the hash.
+void MD5::update(ArrayRef<uint8_t> Data) {
+ MD5_u32plus saved_lo;
+ unsigned long used, free;
+ const uint8_t *Ptr = Data.data();
+ unsigned long Size = Data.size();
+
+ saved_lo = lo;
+ if ((lo = (saved_lo + Size) & 0x1fffffff) < saved_lo)
+ hi++;
+ hi += Size >> 29;
+
+ used = saved_lo & 0x3f;
+
+ if (used) {
+ free = 64 - used;
+
+ if (Size < free) {
+ memcpy(&buffer[used], Ptr, Size);
+ return;
+ }
+
+ memcpy(&buffer[used], Ptr, free);
+ Ptr = Ptr + free;
+ Size -= free;
+ body(makeArrayRef(buffer, 64));
+ }
+
+ if (Size >= 64) {
+ Ptr = body(makeArrayRef(Ptr, Size & ~(unsigned long) 0x3f));
+ Size &= 0x3f;
+ }
+
+ memcpy(buffer, Ptr, Size);
+}
+
+/// Add the bytes in the StringRef \p Str to the hash.
+// Note that this isn't a string and so this won't include any trailing NULL
+// bytes.
+void MD5::update(StringRef Str) {
+ ArrayRef<uint8_t> SVal((const uint8_t *)Str.data(), Str.size());
+ update(SVal);
+}
+
+/// Finish the hash and place the resulting hash into \p result.
+/// \param Result is assumed to be a minimum of 16-bytes in size.
+void MD5::final(MD5Result &Result) {
+ unsigned long used, free;
+
+ used = lo & 0x3f;
+
+ buffer[used++] = 0x80;
+
+ free = 64 - used;
+
+ if (free < 8) {
+ memset(&buffer[used], 0, free);
+ body(makeArrayRef(buffer, 64));
+ used = 0;
+ free = 64;
+ }
+
+ memset(&buffer[used], 0, free - 8);
+
+ lo <<= 3;
+ support::endian::write32le(&buffer[56], lo);
+ support::endian::write32le(&buffer[60], hi);
+
+ body(makeArrayRef(buffer, 64));
+
+ support::endian::write32le(&Result[0], a);
+ support::endian::write32le(&Result[4], b);
+ support::endian::write32le(&Result[8], c);
+ support::endian::write32le(&Result[12], d);
+}
+
+SmallString<32> MD5::MD5Result::digest() const {
+ SmallString<32> Str;
+ raw_svector_ostream Res(Str);
+ for (int i = 0; i < 16; ++i)
+ Res << format("%.2x", Bytes[i]);
+ return Str;
+}
+
+void MD5::stringifyResult(MD5Result &Result, SmallString<32> &Str) {
+ Str = Result.digest();
+}
+
+std::array<uint8_t, 16> MD5::hash(ArrayRef<uint8_t> Data) {
+ MD5 Hash;
+ Hash.update(Data);
+ MD5::MD5Result Res;
+ Hash.final(Res);
+
+ return Res;
+}
diff --git a/third_party/llvm-project/MemoryBuffer.cpp b/third_party/llvm-project/MemoryBuffer.cpp
new file mode 100644
index 000000000..79b93fd90
--- /dev/null
+++ b/third_party/llvm-project/MemoryBuffer.cpp
@@ -0,0 +1,355 @@
+//===--- MemoryBuffer.cpp - Memory Buffer implementation ------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the MemoryBuffer interface.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/Config/config.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Errno.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
+#include "llvm/Support/Program.h"
+#include "llvm/Support/SmallVectorMemoryBuffer.h"
+#include <cassert>
+#include <cerrno>
+#include <cstring>
+#include <new>
+#include <sys/types.h>
+#include <system_error>
+#if !defined(_MSC_VER) && !defined(__MINGW32__)
+#include <unistd.h>
+#else
+#include <io.h>
+#endif
+using namespace llvm;
+
+//===----------------------------------------------------------------------===//
+// MemoryBuffer implementation itself.
+//===----------------------------------------------------------------------===//
+
+MemoryBuffer::~MemoryBuffer() { }
+
+/// init - Initialize this MemoryBuffer as a reference to externally allocated
+/// memory, memory that we know is already null terminated.
+void MemoryBuffer::init(const char *BufStart, const char *BufEnd,
+ bool RequiresNullTerminator) {
+ assert((!RequiresNullTerminator || BufEnd[0] == 0) &&
+ "Buffer is not null terminated!");
+ BufferStart = BufStart;
+ BufferEnd = BufEnd;
+}
+
+//===----------------------------------------------------------------------===//
+// MemoryBufferMem implementation.
+//===----------------------------------------------------------------------===//
+
+/// CopyStringRef - Copies contents of a StringRef into a block of memory and
+/// null-terminates it.
+static void CopyStringRef(char *Memory, StringRef Data) {
+ if (!Data.empty())
+ memcpy(Memory, Data.data(), Data.size());
+ Memory[Data.size()] = 0; // Null terminate string.
+}
+
+namespace {
+struct NamedBufferAlloc {
+ const Twine &Name;
+ NamedBufferAlloc(const Twine &Name) : Name(Name) {}
+};
+}
+
+void *operator new(size_t N, const NamedBufferAlloc &Alloc) {
+ SmallString<256> NameBuf;
+ StringRef NameRef = Alloc.Name.toStringRef(NameBuf);
+
+ char *Mem = static_cast<char *>(operator new(N + NameRef.size() + 1));
+ CopyStringRef(Mem + N, NameRef);
+ return Mem;
+}
+
+namespace {
+/// MemoryBufferMem - Named MemoryBuffer pointing to a block of memory.
+template<typename MB>
+class MemoryBufferMem : public MB {
+public:
+ MemoryBufferMem(StringRef InputData, bool RequiresNullTerminator) {
+ MemoryBuffer::init(InputData.begin(), InputData.end(),
+ RequiresNullTerminator);
+ }
+
+ /// Disable sized deallocation for MemoryBufferMem, because it has
+ /// tail-allocated data.
+ void operator delete(void *p) { ::operator delete(p); }
+
+ StringRef getBufferIdentifier() const override {
+ // The name is stored after the class itself.
+ return StringRef(reinterpret_cast<const char *>(this + 1));
+ }
+
+ MemoryBuffer::BufferKind getBufferKind() const override {
+ return MemoryBuffer::MemoryBuffer_Malloc;
+ }
+};
+}
+
+template <typename MB>
+static ErrorOr<std::unique_ptr<MB>>
+getFileAux(const Twine &Filename, int64_t FileSize, uint64_t MapSize,
+ uint64_t Offset, bool RequiresNullTerminator, bool IsVolatile);
+
+std::unique_ptr<MemoryBuffer>
+MemoryBuffer::getMemBuffer(StringRef InputData, StringRef BufferName,
+ bool RequiresNullTerminator) {
+ auto *Ret = new (NamedBufferAlloc(BufferName))
+ MemoryBufferMem<MemoryBuffer>(InputData, RequiresNullTerminator);
+ return std::unique_ptr<MemoryBuffer>(Ret);
+}
+
+std::unique_ptr<MemoryBuffer>
+MemoryBuffer::getMemBuffer(MemoryBufferRef Ref, bool RequiresNullTerminator) {
+ return std::unique_ptr<MemoryBuffer>(getMemBuffer(
+ Ref.getBuffer(), Ref.getBufferIdentifier(), RequiresNullTerminator));
+}
+
+static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+getMemBufferCopyImpl(StringRef InputData, const Twine &BufferName) {
+ auto Buf = WritableMemoryBuffer::getNewUninitMemBuffer(InputData.size(), BufferName);
+ if (!Buf)
+ return make_error_code(errc::not_enough_memory);
+ memcpy(Buf->getBufferStart(), InputData.data(), InputData.size());
+ return std::move(Buf);
+}
+
+std::unique_ptr<MemoryBuffer>
+MemoryBuffer::getMemBufferCopy(StringRef InputData, const Twine &BufferName) {
+ auto Buf = getMemBufferCopyImpl(InputData, BufferName);
+ if (Buf)
+ return std::move(*Buf);
+ return nullptr;
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getFileOrSTDIN(const Twine &Filename, int64_t FileSize,
+ bool RequiresNullTerminator) {
+ SmallString<256> NameBuf;
+ StringRef NameRef = Filename.toStringRef(NameBuf);
+
+ if (NameRef == "-")
+ return getSTDIN();
+ return getFile(Filename, FileSize, RequiresNullTerminator);
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getFileSlice(const Twine &FilePath, uint64_t MapSize,
+ uint64_t Offset, bool IsVolatile) {
+ return getFileAux<MemoryBuffer>(FilePath, -1, MapSize, Offset, false,
+ IsVolatile);
+}
+
+//===----------------------------------------------------------------------===//
+// MemoryBuffer::getFile implementation.
+//===----------------------------------------------------------------------===//
+
+#if 0 // XXX BINARYEN
+namespace {
+/// Memory maps a file descriptor using sys::fs::mapped_file_region.
+///
+/// This handles converting the offset into a legal offset on the platform.
+template<typename MB>
+class MemoryBufferMMapFile : public MB {
+ sys::fs::mapped_file_region MFR;
+
+ static uint64_t getLegalMapOffset(uint64_t Offset) {
+ return Offset & ~(sys::fs::mapped_file_region::alignment() - 1);
+ }
+
+ static uint64_t getLegalMapSize(uint64_t Len, uint64_t Offset) {
+ return Len + (Offset - getLegalMapOffset(Offset));
+ }
+
+ const char *getStart(uint64_t Len, uint64_t Offset) {
+ return MFR.const_data() + (Offset - getLegalMapOffset(Offset));
+ }
+
+public:
+ MemoryBufferMMapFile(bool RequiresNullTerminator, sys::fs::file_t FD, uint64_t Len,
+ uint64_t Offset, std::error_code &EC)
+ : MFR(FD, MB::Mapmode, getLegalMapSize(Len, Offset),
+ getLegalMapOffset(Offset), EC) {
+ if (!EC) {
+ const char *Start = getStart(Len, Offset);
+ MemoryBuffer::init(Start, Start + Len, RequiresNullTerminator);
+ }
+ }
+
+ /// Disable sized deallocation for MemoryBufferMMapFile, because it has
+ /// tail-allocated data.
+ void operator delete(void *p) { ::operator delete(p); }
+
+ StringRef getBufferIdentifier() const override {
+ // The name is stored after the class itself.
+ return StringRef(reinterpret_cast<const char *>(this + 1));
+ }
+
+ MemoryBuffer::BufferKind getBufferKind() const override {
+ return MemoryBuffer::MemoryBuffer_MMap;
+ }
+};
+}
+#endif
+
+static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+getMemoryBufferForStream(sys::fs::file_t FD, const Twine &BufferName) {
+ llvm_unreachable("getMemoryBufferForStream");
+}
+
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getFile(const Twine &Filename, int64_t FileSize,
+ bool RequiresNullTerminator, bool IsVolatile) {
+ return getFileAux<MemoryBuffer>(Filename, FileSize, FileSize, 0,
+ RequiresNullTerminator, IsVolatile);
+}
+
+template <typename MB>
+static ErrorOr<std::unique_ptr<MB>>
+getOpenFileImpl(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
+ uint64_t MapSize, int64_t Offset, bool RequiresNullTerminator,
+ bool IsVolatile);
+
+template <typename MB>
+static ErrorOr<std::unique_ptr<MB>>
+getFileAux(const Twine &Filename, int64_t FileSize, uint64_t MapSize,
+ uint64_t Offset, bool RequiresNullTerminator, bool IsVolatile) {
+ llvm_unreachable("getFileAux");
+}
+
+ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+WritableMemoryBuffer::getFile(const Twine &Filename, int64_t FileSize,
+ bool IsVolatile) {
+ return getFileAux<WritableMemoryBuffer>(Filename, FileSize, FileSize, 0,
+ /*RequiresNullTerminator*/ false,
+ IsVolatile);
+}
+
+ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+WritableMemoryBuffer::getFileSlice(const Twine &Filename, uint64_t MapSize,
+ uint64_t Offset, bool IsVolatile) {
+ return getFileAux<WritableMemoryBuffer>(Filename, -1, MapSize, Offset, false,
+ IsVolatile);
+}
+
+std::unique_ptr<WritableMemoryBuffer>
+WritableMemoryBuffer::getNewUninitMemBuffer(size_t Size, const Twine &BufferName) {
+ using MemBuffer = MemoryBufferMem<WritableMemoryBuffer>;
+ // Allocate space for the MemoryBuffer, the data and the name. It is important
+ // that MemoryBuffer and data are aligned so PointerIntPair works with them.
+ // TODO: Is 16-byte alignment enough? We copy small object files with large
+ // alignment expectations into this buffer.
+ SmallString<256> NameBuf;
+ StringRef NameRef = BufferName.toStringRef(NameBuf);
+ size_t AlignedStringLen = alignTo(sizeof(MemBuffer) + NameRef.size() + 1, 16);
+ size_t RealLen = AlignedStringLen + Size + 1;
+ char *Mem = static_cast<char*>(operator new(RealLen, std::nothrow));
+ if (!Mem)
+ return nullptr;
+
+ // The name is stored after the class itself.
+ CopyStringRef(Mem + sizeof(MemBuffer), NameRef);
+
+ // The buffer begins after the name and must be aligned.
+ char *Buf = Mem + AlignedStringLen;
+ Buf[Size] = 0; // Null terminate buffer.
+
+ auto *Ret = new (Mem) MemBuffer(StringRef(Buf, Size), true);
+ return std::unique_ptr<WritableMemoryBuffer>(Ret);
+}
+
+std::unique_ptr<WritableMemoryBuffer>
+WritableMemoryBuffer::getNewMemBuffer(size_t Size, const Twine &BufferName) {
+ auto SB = WritableMemoryBuffer::getNewUninitMemBuffer(Size, BufferName);
+ if (!SB)
+ return nullptr;
+ memset(SB->getBufferStart(), 0, Size);
+ return SB;
+}
+
+static bool shouldUseMmap(sys::fs::file_t FD,
+ size_t FileSize,
+ size_t MapSize,
+ off_t Offset,
+ bool RequiresNullTerminator,
+ int PageSize,
+ bool IsVolatile) {
+ // XXX BINARYEn
+ return false;
+}
+
+static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
+getReadWriteFile(const Twine &Filename, uint64_t FileSize, uint64_t MapSize,
+ uint64_t Offset) {
+ llvm_unreachable("getReadWriteFile");
+}
+
+ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
+WriteThroughMemoryBuffer::getFile(const Twine &Filename, int64_t FileSize) {
+ return getReadWriteFile(Filename, FileSize, FileSize, 0);
+}
+
+/// Map a subrange of the specified file as a WritableMemoryBuffer.
+ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
+WriteThroughMemoryBuffer::getFileSlice(const Twine &Filename, uint64_t MapSize,
+ uint64_t Offset) {
+ return getReadWriteFile(Filename, -1, MapSize, Offset);
+}
+
+template <typename MB>
+static ErrorOr<std::unique_ptr<MB>>
+getOpenFileImpl(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
+ uint64_t MapSize, int64_t Offset, bool RequiresNullTerminator,
+ bool IsVolatile) {
+ llvm_unreachable("getOpenFileImpl");
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getOpenFile(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
+ bool RequiresNullTerminator, bool IsVolatile) {
+ return getOpenFileImpl<MemoryBuffer>(FD, Filename, FileSize, FileSize, 0,
+ RequiresNullTerminator, IsVolatile);
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getOpenFileSlice(sys::fs::file_t FD, const Twine &Filename, uint64_t MapSize,
+ int64_t Offset, bool IsVolatile) {
+ assert(MapSize != uint64_t(-1));
+ return getOpenFileImpl<MemoryBuffer>(FD, Filename, -1, MapSize, Offset, false,
+ IsVolatile);
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>> MemoryBuffer::getSTDIN() {
+ llvm_unreachable("getSTDIN");
+}
+
+ErrorOr<std::unique_ptr<MemoryBuffer>>
+MemoryBuffer::getFileAsStream(const Twine &Filename) {
+ llvm_unreachable("getFileAsStream");
+}
+
+MemoryBufferRef MemoryBuffer::getMemBufferRef() const {
+ StringRef Data = getBuffer();
+ StringRef Identifier = getBufferIdentifier();
+ return MemoryBufferRef(Data, Identifier);
+}
+
+SmallVectorMemoryBuffer::~SmallVectorMemoryBuffer() {}
diff --git a/third_party/llvm-project/NativeFormatting.cpp b/third_party/llvm-project/NativeFormatting.cpp
new file mode 100644
index 000000000..3731e0c56
--- /dev/null
+++ b/third_party/llvm-project/NativeFormatting.cpp
@@ -0,0 +1,263 @@
+//===- NativeFormatting.cpp - Low level formatting helpers -------*- C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/NativeFormatting.h"
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/Support/Format.h"
+
+#include <float.h>
+
+using namespace llvm;
+
+template<typename T, std::size_t N>
+static int format_to_buffer(T Value, char (&Buffer)[N]) {
+ char *EndPtr = std::end(Buffer);
+ char *CurPtr = EndPtr;
+
+ do {
+ *--CurPtr = '0' + char(Value % 10);
+ Value /= 10;
+ } while (Value);
+ return EndPtr - CurPtr;
+}
+
+static void writeWithCommas(raw_ostream &S, ArrayRef<char> Buffer) {
+ assert(!Buffer.empty());
+
+ ArrayRef<char> ThisGroup;
+ int InitialDigits = ((Buffer.size() - 1) % 3) + 1;
+ ThisGroup = Buffer.take_front(InitialDigits);
+ S.write(ThisGroup.data(), ThisGroup.size());
+
+ Buffer = Buffer.drop_front(InitialDigits);
+ assert(Buffer.size() % 3 == 0);
+ while (!Buffer.empty()) {
+ S << ',';
+ ThisGroup = Buffer.take_front(3);
+ S.write(ThisGroup.data(), 3);
+ Buffer = Buffer.drop_front(3);
+ }
+}
+
+template <typename T>
+static void write_unsigned_impl(raw_ostream &S, T N, size_t MinDigits,
+ IntegerStyle Style, bool IsNegative) {
+ static_assert(std::is_unsigned<T>::value, "Value is not unsigned!");
+
+ char NumberBuffer[128];
+ std::memset(NumberBuffer, '0', sizeof(NumberBuffer));
+
+ size_t Len = 0;
+ Len = format_to_buffer(N, NumberBuffer);
+
+ if (IsNegative)
+ S << '-';
+
+ if (Len < MinDigits && Style != IntegerStyle::Number) {
+ for (size_t I = Len; I < MinDigits; ++I)
+ S << '0';
+ }
+
+ if (Style == IntegerStyle::Number) {
+ writeWithCommas(S, ArrayRef<char>(std::end(NumberBuffer) - Len, Len));
+ } else {
+ S.write(std::end(NumberBuffer) - Len, Len);
+ }
+}
+
+template <typename T>
+static void write_unsigned(raw_ostream &S, T N, size_t MinDigits,
+ IntegerStyle Style, bool IsNegative = false) {
+ // Output using 32-bit div/mod if possible.
+ if (N == static_cast<uint32_t>(N))
+ write_unsigned_impl(S, static_cast<uint32_t>(N), MinDigits, Style,
+ IsNegative);
+ else
+ write_unsigned_impl(S, N, MinDigits, Style, IsNegative);
+}
+
+template <typename T>
+static void write_signed(raw_ostream &S, T N, size_t MinDigits,
+ IntegerStyle Style) {
+ static_assert(std::is_signed<T>::value, "Value is not signed!");
+
+ using UnsignedT = typename std::make_unsigned<T>::type;
+
+ if (N >= 0) {
+ write_unsigned(S, static_cast<UnsignedT>(N), MinDigits, Style);
+ return;
+ }
+
+ UnsignedT UN = -(UnsignedT)N;
+ write_unsigned(S, UN, MinDigits, Style, true);
+}
+
+void llvm::write_integer(raw_ostream &S, unsigned int N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_unsigned(S, N, MinDigits, Style);
+}
+
+void llvm::write_integer(raw_ostream &S, int N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_signed(S, N, MinDigits, Style);
+}
+
+void llvm::write_integer(raw_ostream &S, unsigned long N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_unsigned(S, N, MinDigits, Style);
+}
+
+void llvm::write_integer(raw_ostream &S, long N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_signed(S, N, MinDigits, Style);
+}
+
+void llvm::write_integer(raw_ostream &S, unsigned long long N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_unsigned(S, N, MinDigits, Style);
+}
+
+void llvm::write_integer(raw_ostream &S, long long N, size_t MinDigits,
+ IntegerStyle Style) {
+ write_signed(S, N, MinDigits, Style);
+}
+
+void llvm::write_hex(raw_ostream &S, uint64_t N, HexPrintStyle Style,
+ Optional<size_t> Width) {
+ const size_t kMaxWidth = 128u;
+
+ size_t W = std::min(kMaxWidth, Width.getValueOr(0u));
+
+ unsigned Nibbles = (64 - countLeadingZeros(N) + 3) / 4;
+ bool Prefix = (Style == HexPrintStyle::PrefixLower ||
+ Style == HexPrintStyle::PrefixUpper);
+ bool Upper =
+ (Style == HexPrintStyle::Upper || Style == HexPrintStyle::PrefixUpper);
+ unsigned PrefixChars = Prefix ? 2 : 0;
+ unsigned NumChars =
+ std::max(static_cast<unsigned>(W), std::max(1u, Nibbles) + PrefixChars);
+
+ char NumberBuffer[kMaxWidth];
+ ::memset(NumberBuffer, '0', llvm::array_lengthof(NumberBuffer));
+ if (Prefix)
+ NumberBuffer[1] = 'x';
+ char *EndPtr = NumberBuffer + NumChars;
+ char *CurPtr = EndPtr;
+ while (N) {
+ unsigned char x = static_cast<unsigned char>(N) % 16;
+ *--CurPtr = hexdigit(x, !Upper);
+ N /= 16;
+ }
+
+ S.write(NumberBuffer, NumChars);
+}
+
+void llvm::write_double(raw_ostream &S, double N, FloatStyle Style,
+ Optional<size_t> Precision) {
+ size_t Prec = Precision.getValueOr(getDefaultPrecision(Style));
+
+ if (std::isnan(N)) {
+ S << "nan";
+ return;
+ } else if (std::isinf(N)) {
+ S << "INF";
+ return;
+ }
+
+ char Letter;
+ if (Style == FloatStyle::Exponent)
+ Letter = 'e';
+ else if (Style == FloatStyle::ExponentUpper)
+ Letter = 'E';
+ else
+ Letter = 'f';
+
+ SmallString<8> Spec;
+ llvm::raw_svector_ostream Out(Spec);
+ Out << "%." << Prec << Letter;
+
+ if (Style == FloatStyle::Exponent || Style == FloatStyle::ExponentUpper) {
+#ifdef _WIN32
+// On MSVCRT and compatible, output of %e is incompatible to Posix
+// by default. Number of exponent digits should be at least 2. "%+03d"
+// FIXME: Implement our formatter to here or Support/Format.h!
+#if defined(__MINGW32__)
+ // FIXME: It should be generic to C++11.
+ if (N == 0.0 && std::signbit(N)) {
+ char NegativeZero[] = "-0.000000e+00";
+ if (Style == FloatStyle::ExponentUpper)
+ NegativeZero[strlen(NegativeZero) - 4] = 'E';
+ S << NegativeZero;
+ return;
+ }
+#else
+ int fpcl = _fpclass(N);
+
+ // negative zero
+ if (fpcl == _FPCLASS_NZ) {
+ char NegativeZero[] = "-0.000000e+00";
+ if (Style == FloatStyle::ExponentUpper)
+ NegativeZero[strlen(NegativeZero) - 4] = 'E';
+ S << NegativeZero;
+ return;
+ }
+#endif
+
+ char buf[32];
+ unsigned len;
+ len = format(Spec.c_str(), N).snprint(buf, sizeof(buf));
+ if (len <= sizeof(buf) - 2) {
+ if (len >= 5 && (buf[len - 5] == 'e' || buf[len - 5] == 'E') &&
+ buf[len - 3] == '0') {
+ int cs = buf[len - 4];
+ if (cs == '+' || cs == '-') {
+ int c1 = buf[len - 2];
+ int c0 = buf[len - 1];
+ if (isdigit(static_cast<unsigned char>(c1)) &&
+ isdigit(static_cast<unsigned char>(c0))) {
+ // Trim leading '0': "...e+012" -> "...e+12\0"
+ buf[len - 3] = c1;
+ buf[len - 2] = c0;
+ buf[--len] = 0;
+ }
+ }
+ }
+ S << buf;
+ return;
+ }
+#endif
+ }
+
+ if (Style == FloatStyle::Percent)
+ N *= 100.0;
+
+ char Buf[32];
+ format(Spec.c_str(), N).snprint(Buf, sizeof(Buf));
+ S << Buf;
+ if (Style == FloatStyle::Percent)
+ S << '%';
+}
+
+bool llvm::isPrefixedHexStyle(HexPrintStyle S) {
+ return (S == HexPrintStyle::PrefixLower || S == HexPrintStyle::PrefixUpper);
+}
+
+size_t llvm::getDefaultPrecision(FloatStyle Style) {
+ switch (Style) {
+ case FloatStyle::Exponent:
+ case FloatStyle::ExponentUpper:
+ return 6; // Number of decimal places.
+ case FloatStyle::Fixed:
+ case FloatStyle::Percent:
+ return 2; // Number of decimal places.
+ }
+ LLVM_BUILTIN_UNREACHABLE;
+}
diff --git a/third_party/llvm-project/ObjectFile.cpp b/third_party/llvm-project/ObjectFile.cpp
new file mode 100644
index 000000000..3ee649fd3
--- /dev/null
+++ b/third_party/llvm-project/ObjectFile.cpp
@@ -0,0 +1,112 @@
+//===- ObjectFile.cpp - File format independent object file ---------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines a file format independent ObjectFile class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Magic.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/COFF.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Object/MachO.h"
+#include "llvm/Object/Wasm.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cstdint>
+#include <memory>
+#include <system_error>
+
+using namespace llvm;
+using namespace object;
+
+void ObjectFile::anchor() {}
+
+ObjectFile::ObjectFile(unsigned int Type, MemoryBufferRef Source)
+ : SymbolicFile(Type, Source) {}
+
+bool SectionRef::containsSymbol(SymbolRef S) const {
+ llvm_unreachable("containsSymbol"); // XXX BINARYEN
+}
+
+uint64_t ObjectFile::getSymbolValue(DataRefImpl Ref) const {
+ uint32_t Flags = getSymbolFlags(Ref);
+ if (Flags & SymbolRef::SF_Undefined)
+ return 0;
+ if (Flags & SymbolRef::SF_Common)
+ return getCommonSymbolSize(Ref);
+ return getSymbolValueImpl(Ref);
+}
+
+Error ObjectFile::printSymbolName(raw_ostream &OS, DataRefImpl Symb) const {
+ Expected<StringRef> Name = getSymbolName(Symb);
+ if (!Name)
+ return Name.takeError();
+ OS << *Name;
+ return Error::success();
+}
+
+uint32_t ObjectFile::getSymbolAlignment(DataRefImpl DRI) const { return 0; }
+
+bool ObjectFile::isSectionBitcode(DataRefImpl Sec) const {
+ Expected<StringRef> NameOrErr = getSectionName(Sec);
+ if (NameOrErr)
+ return *NameOrErr == ".llvmbc";
+ consumeError(NameOrErr.takeError());
+ return false;
+}
+
+bool ObjectFile::isSectionStripped(DataRefImpl Sec) const { return false; }
+
+bool ObjectFile::isBerkeleyText(DataRefImpl Sec) const {
+ return isSectionText(Sec);
+}
+
+bool ObjectFile::isBerkeleyData(DataRefImpl Sec) const {
+ return isSectionData(Sec);
+}
+
+Expected<section_iterator>
+ObjectFile::getRelocatedSection(DataRefImpl Sec) const {
+ return section_iterator(SectionRef(Sec, this));
+}
+
+Triple ObjectFile::makeTriple() const {
+ llvm_unreachable("makeTriple"); // XXX BINARYEN
+}
+
+Expected<std::unique_ptr<ObjectFile>>
+ObjectFile::createObjectFile(MemoryBufferRef Object, file_magic Type) {
+ llvm_unreachable("createObjectFile"); // XXX BINARYEN
+}
+
+#if 0 // XXX BINARYEN
+Expected<OwningBinary<ObjectFile>>
+ObjectFile::createObjectFile(StringRef ObjectPath) {
+ ErrorOr<std::unique_ptr<MemoryBuffer>> FileOrErr =
+ MemoryBuffer::getFile(ObjectPath);
+ if (std::error_code EC = FileOrErr.getError())
+ return errorCodeToError(EC);
+ std::unique_ptr<MemoryBuffer> Buffer = std::move(FileOrErr.get());
+
+ Expected<std::unique_ptr<ObjectFile>> ObjOrErr =
+ createObjectFile(Buffer->getMemBufferRef());
+ if (Error Err = ObjOrErr.takeError())
+ return std::move(Err);
+ std::unique_ptr<ObjectFile> Obj = std::move(ObjOrErr.get());
+
+ return OwningBinary<ObjectFile>(std::move(Obj), std::move(Buffer));
+}
+#endif
diff --git a/third_party/llvm-project/Optional.cpp b/third_party/llvm-project/Optional.cpp
new file mode 100644
index 000000000..2425739c8
--- /dev/null
+++ b/third_party/llvm-project/Optional.cpp
@@ -0,0 +1,14 @@
+//===- Optional.cpp - Optional values ---------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/Support/raw_ostream.h"
+
+llvm::raw_ostream &llvm::operator<<(raw_ostream &OS, NoneType) {
+ return OS << "None";
+}
diff --git a/third_party/llvm-project/Path.cpp b/third_party/llvm-project/Path.cpp
new file mode 100644
index 000000000..e4d890a98
--- /dev/null
+++ b/third_party/llvm-project/Path.cpp
@@ -0,0 +1,1262 @@
+//===-- Path.cpp - Implement OS Path Concept ------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the operating system Path API.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/Path.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/Config/llvm-config.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/Process.h"
+#include "llvm/Support/Signals.h"
+#include <cctype>
+#include <cstring>
+
+#if !defined(_MSC_VER) && !defined(__MINGW32__)
+#include <unistd.h>
+#else
+#include <io.h>
+#endif
+
+using namespace llvm;
+using namespace llvm::support::endian;
+
+namespace {
+ using llvm::StringRef;
+ using llvm::sys::path::is_separator;
+ using llvm::sys::path::Style;
+
+ inline Style real_style(Style style) {
+#ifdef _WIN32
+ return (style == Style::posix) ? Style::posix : Style::windows;
+#else
+ return (style == Style::windows) ? Style::windows : Style::posix;
+#endif
+ }
+
+ inline const char *separators(Style style) {
+ if (real_style(style) == Style::windows)
+ return "\\/";
+ return "/";
+ }
+
+ inline char preferred_separator(Style style) {
+ if (real_style(style) == Style::windows)
+ return '\\';
+ return '/';
+ }
+
+ StringRef find_first_component(StringRef path, Style style) {
+ // Look for this first component in the following order.
+ // * empty (in this case we return an empty string)
+ // * either C: or {//,\\}net.
+ // * {/,\}
+ // * {file,directory}name
+
+ if (path.empty())
+ return path;
+
+ if (real_style(style) == Style::windows) {
+ // C:
+ if (path.size() >= 2 &&
+ std::isalpha(static_cast<unsigned char>(path[0])) && path[1] == ':')
+ return path.substr(0, 2);
+ }
+
+ // //net
+ if ((path.size() > 2) && is_separator(path[0], style) &&
+ path[0] == path[1] && !is_separator(path[2], style)) {
+ // Find the next directory separator.
+ size_t end = path.find_first_of(separators(style), 2);
+ return path.substr(0, end);
+ }
+
+ // {/,\}
+ if (is_separator(path[0], style))
+ return path.substr(0, 1);
+
+ // * {file,directory}name
+ size_t end = path.find_first_of(separators(style));
+ return path.substr(0, end);
+ }
+
+ // Returns the first character of the filename in str. For paths ending in
+ // '/', it returns the position of the '/'.
+ size_t filename_pos(StringRef str, Style style) {
+ if (str.size() > 0 && is_separator(str[str.size() - 1], style))
+ return str.size() - 1;
+
+ size_t pos = str.find_last_of(separators(style), str.size() - 1);
+
+ if (real_style(style) == Style::windows) {
+ if (pos == StringRef::npos)
+ pos = str.find_last_of(':', str.size() - 2);
+ }
+
+ if (pos == StringRef::npos || (pos == 1 && is_separator(str[0], style)))
+ return 0;
+
+ return pos + 1;
+ }
+
+ // Returns the position of the root directory in str. If there is no root
+ // directory in str, it returns StringRef::npos.
+ size_t root_dir_start(StringRef str, Style style) {
+ // case "c:/"
+ if (real_style(style) == Style::windows) {
+ if (str.size() > 2 && str[1] == ':' && is_separator(str[2], style))
+ return 2;
+ }
+
+ // case "//net"
+ if (str.size() > 3 && is_separator(str[0], style) && str[0] == str[1] &&
+ !is_separator(str[2], style)) {
+ return str.find_first_of(separators(style), 2);
+ }
+
+ // case "/"
+ if (str.size() > 0 && is_separator(str[0], style))
+ return 0;
+
+ return StringRef::npos;
+ }
+
+ // Returns the position past the end of the "parent path" of path. The parent
+ // path will not end in '/', unless the parent is the root directory. If the
+ // path has no parent, 0 is returned.
+ size_t parent_path_end(StringRef path, Style style) {
+ size_t end_pos = filename_pos(path, style);
+
+ bool filename_was_sep =
+ path.size() > 0 && is_separator(path[end_pos], style);
+
+ // Skip separators until we reach root dir (or the start of the string).
+ size_t root_dir_pos = root_dir_start(path, style);
+ while (end_pos > 0 &&
+ (root_dir_pos == StringRef::npos || end_pos > root_dir_pos) &&
+ is_separator(path[end_pos - 1], style))
+ --end_pos;
+
+ if (end_pos == root_dir_pos && !filename_was_sep) {
+ // We've reached the root dir and the input path was *not* ending in a
+ // sequence of slashes. Include the root dir in the parent path.
+ return root_dir_pos + 1;
+ }
+
+ // Otherwise, just include before the last slash.
+ return end_pos;
+ }
+} // end unnamed namespace
+
+enum FSEntity {
+ FS_Dir,
+ FS_File,
+ FS_Name
+};
+
+static std::error_code
+createUniqueEntity(const Twine &Model, int &ResultFD,
+ SmallVectorImpl<char> &ResultPath, bool MakeAbsolute,
+ unsigned Mode, FSEntity Type,
+ sys::fs::OpenFlags Flags = sys::fs::OF_None) {
+
+ // Limit the number of attempts we make, so that we don't infinite loop. E.g.
+ // "permission denied" could be for a specific file (so we retry with a
+ // different name) or for the whole directory (retry would always fail).
+ // Checking which is racy, so we try a number of times, then give up.
+ std::error_code EC;
+ for (int Retries = 128; Retries > 0; --Retries) {
+ sys::fs::createUniquePath(Model, ResultPath, MakeAbsolute);
+ // Try to open + create the file.
+ switch (Type) {
+ case FS_File: {
+ EC = sys::fs::openFileForReadWrite(Twine(ResultPath.begin()), ResultFD,
+ sys::fs::CD_CreateNew, Flags, Mode);
+ if (EC) {
+ // errc::permission_denied happens on Windows when we try to open a file
+ // that has been marked for deletion.
+ if (EC == errc::file_exists || EC == errc::permission_denied)
+ continue;
+ return EC;
+ }
+
+ return std::error_code();
+ }
+
+ case FS_Name: {
+ EC = sys::fs::access(ResultPath.begin(), sys::fs::AccessMode::Exist);
+ if (EC == errc::no_such_file_or_directory)
+ return std::error_code();
+ if (EC)
+ return EC;
+ continue;
+ }
+
+ case FS_Dir: {
+ EC = sys::fs::create_directory(ResultPath.begin(), false);
+ if (EC) {
+ if (EC == errc::file_exists)
+ continue;
+ return EC;
+ }
+ return std::error_code();
+ }
+ }
+ llvm_unreachable("Invalid Type");
+ }
+ return EC;
+}
+
+namespace llvm {
+namespace sys {
+namespace path {
+
+const_iterator begin(StringRef path, Style style) {
+ const_iterator i;
+ i.Path = path;
+ i.Component = find_first_component(path, style);
+ i.Position = 0;
+ i.S = style;
+ return i;
+}
+
+const_iterator end(StringRef path) {
+ const_iterator i;
+ i.Path = path;
+ i.Position = path.size();
+ return i;
+}
+
+const_iterator &const_iterator::operator++() {
+ assert(Position < Path.size() && "Tried to increment past end!");
+
+ // Increment Position to past the current component
+ Position += Component.size();
+
+ // Check for end.
+ if (Position == Path.size()) {
+ Component = StringRef();
+ return *this;
+ }
+
+ // Both POSIX and Windows treat paths that begin with exactly two separators
+ // specially.
+ bool was_net = Component.size() > 2 && is_separator(Component[0], S) &&
+ Component[1] == Component[0] && !is_separator(Component[2], S);
+
+ // Handle separators.
+ if (is_separator(Path[Position], S)) {
+ // Root dir.
+ if (was_net ||
+ // c:/
+ (real_style(S) == Style::windows && Component.endswith(":"))) {
+ Component = Path.substr(Position, 1);
+ return *this;
+ }
+
+ // Skip extra separators.
+ while (Position != Path.size() && is_separator(Path[Position], S)) {
+ ++Position;
+ }
+
+ // Treat trailing '/' as a '.', unless it is the root dir.
+ if (Position == Path.size() && Component != "/") {
+ --Position;
+ Component = ".";
+ return *this;
+ }
+ }
+
+ // Find next component.
+ size_t end_pos = Path.find_first_of(separators(S), Position);
+ Component = Path.slice(Position, end_pos);
+
+ return *this;
+}
+
+bool const_iterator::operator==(const const_iterator &RHS) const {
+ return Path.begin() == RHS.Path.begin() && Position == RHS.Position;
+}
+
+ptrdiff_t const_iterator::operator-(const const_iterator &RHS) const {
+ return Position - RHS.Position;
+}
+
+reverse_iterator rbegin(StringRef Path, Style style) {
+ reverse_iterator I;
+ I.Path = Path;
+ I.Position = Path.size();
+ I.S = style;
+ ++I;
+ return I;
+}
+
+reverse_iterator rend(StringRef Path) {
+ reverse_iterator I;
+ I.Path = Path;
+ I.Component = Path.substr(0, 0);
+ I.Position = 0;
+ return I;
+}
+
+reverse_iterator &reverse_iterator::operator++() {
+ size_t root_dir_pos = root_dir_start(Path, S);
+
+ // Skip separators unless it's the root directory.
+ size_t end_pos = Position;
+ while (end_pos > 0 && (end_pos - 1) != root_dir_pos &&
+ is_separator(Path[end_pos - 1], S))
+ --end_pos;
+
+ // Treat trailing '/' as a '.', unless it is the root dir.
+ if (Position == Path.size() && !Path.empty() &&
+ is_separator(Path.back(), S) &&
+ (root_dir_pos == StringRef::npos || end_pos - 1 > root_dir_pos)) {
+ --Position;
+ Component = ".";
+ return *this;
+ }
+
+ // Find next separator.
+ size_t start_pos = filename_pos(Path.substr(0, end_pos), S);
+ Component = Path.slice(start_pos, end_pos);
+ Position = start_pos;
+ return *this;
+}
+
+bool reverse_iterator::operator==(const reverse_iterator &RHS) const {
+ return Path.begin() == RHS.Path.begin() && Component == RHS.Component &&
+ Position == RHS.Position;
+}
+
+ptrdiff_t reverse_iterator::operator-(const reverse_iterator &RHS) const {
+ return Position - RHS.Position;
+}
+
+StringRef root_path(StringRef path, Style style) {
+ const_iterator b = begin(path, style), pos = b, e = end(path);
+ if (b != e) {
+ bool has_net =
+ b->size() > 2 && is_separator((*b)[0], style) && (*b)[1] == (*b)[0];
+ bool has_drive = (real_style(style) == Style::windows) && b->endswith(":");
+
+ if (has_net || has_drive) {
+ if ((++pos != e) && is_separator((*pos)[0], style)) {
+ // {C:/,//net/}, so get the first two components.
+ return path.substr(0, b->size() + pos->size());
+ } else {
+ // just {C:,//net}, return the first component.
+ return *b;
+ }
+ }
+
+ // POSIX style root directory.
+ if (is_separator((*b)[0], style)) {
+ return *b;
+ }
+ }
+
+ return StringRef();
+}
+
+StringRef root_name(StringRef path, Style style) {
+ const_iterator b = begin(path, style), e = end(path);
+ if (b != e) {
+ bool has_net =
+ b->size() > 2 && is_separator((*b)[0], style) && (*b)[1] == (*b)[0];
+ bool has_drive = (real_style(style) == Style::windows) && b->endswith(":");
+
+ if (has_net || has_drive) {
+ // just {C:,//net}, return the first component.
+ return *b;
+ }
+ }
+
+ // No path or no name.
+ return StringRef();
+}
+
+StringRef root_directory(StringRef path, Style style) {
+ const_iterator b = begin(path, style), pos = b, e = end(path);
+ if (b != e) {
+ bool has_net =
+ b->size() > 2 && is_separator((*b)[0], style) && (*b)[1] == (*b)[0];
+ bool has_drive = (real_style(style) == Style::windows) && b->endswith(":");
+
+ if ((has_net || has_drive) &&
+ // {C:,//net}, skip to the next component.
+ (++pos != e) && is_separator((*pos)[0], style)) {
+ return *pos;
+ }
+
+ // POSIX style root directory.
+ if (!has_net && is_separator((*b)[0], style)) {
+ return *b;
+ }
+ }
+
+ // No path or no root.
+ return StringRef();
+}
+
+StringRef relative_path(StringRef path, Style style) {
+ StringRef root = root_path(path, style);
+ return path.substr(root.size());
+}
+
+void append(SmallVectorImpl<char> &path, Style style, const Twine &a,
+ const Twine &b, const Twine &c, const Twine &d) {
+ SmallString<32> a_storage;
+ SmallString<32> b_storage;
+ SmallString<32> c_storage;
+ SmallString<32> d_storage;
+
+ SmallVector<StringRef, 4> components;
+ if (!a.isTriviallyEmpty()) components.push_back(a.toStringRef(a_storage));
+ if (!b.isTriviallyEmpty()) components.push_back(b.toStringRef(b_storage));
+ if (!c.isTriviallyEmpty()) components.push_back(c.toStringRef(c_storage));
+ if (!d.isTriviallyEmpty()) components.push_back(d.toStringRef(d_storage));
+
+ for (auto &component : components) {
+ bool path_has_sep =
+ !path.empty() && is_separator(path[path.size() - 1], style);
+ if (path_has_sep) {
+ // Strip separators from beginning of component.
+ size_t loc = component.find_first_not_of(separators(style));
+ StringRef c = component.substr(loc);
+
+ // Append it.
+ path.append(c.begin(), c.end());
+ continue;
+ }
+
+ bool component_has_sep =
+ !component.empty() && is_separator(component[0], style);
+ if (!component_has_sep &&
+ !(path.empty() || has_root_name(component, style))) {
+ // Add a separator.
+ path.push_back(preferred_separator(style));
+ }
+
+ path.append(component.begin(), component.end());
+ }
+}
+
+void append(SmallVectorImpl<char> &path, const Twine &a, const Twine &b,
+ const Twine &c, const Twine &d) {
+ append(path, Style::native, a, b, c, d);
+}
+
+void append(SmallVectorImpl<char> &path, const_iterator begin,
+ const_iterator end, Style style) {
+ for (; begin != end; ++begin)
+ path::append(path, style, *begin);
+}
+
+StringRef parent_path(StringRef path, Style style) {
+ size_t end_pos = parent_path_end(path, style);
+ if (end_pos == StringRef::npos)
+ return StringRef();
+ else
+ return path.substr(0, end_pos);
+}
+
+void remove_filename(SmallVectorImpl<char> &path, Style style) {
+ size_t end_pos = parent_path_end(StringRef(path.begin(), path.size()), style);
+ if (end_pos != StringRef::npos)
+ path.set_size(end_pos);
+}
+
+void replace_extension(SmallVectorImpl<char> &path, const Twine &extension,
+ Style style) {
+ StringRef p(path.begin(), path.size());
+ SmallString<32> ext_storage;
+ StringRef ext = extension.toStringRef(ext_storage);
+
+ // Erase existing extension.
+ size_t pos = p.find_last_of('.');
+ if (pos != StringRef::npos && pos >= filename_pos(p, style))
+ path.set_size(pos);
+
+ // Append '.' if needed.
+ if (ext.size() > 0 && ext[0] != '.')
+ path.push_back('.');
+
+ // Append extension.
+ path.append(ext.begin(), ext.end());
+}
+
+void replace_path_prefix(SmallVectorImpl<char> &Path,
+ const StringRef &OldPrefix, const StringRef &NewPrefix,
+ Style style) {
+ if (OldPrefix.empty() && NewPrefix.empty())
+ return;
+
+ StringRef OrigPath(Path.begin(), Path.size());
+ if (!OrigPath.startswith(OldPrefix))
+ return;
+
+ // If prefixes have the same size we can simply copy the new one over.
+ if (OldPrefix.size() == NewPrefix.size()) {
+ llvm::copy(NewPrefix, Path.begin());
+ return;
+ }
+
+ StringRef RelPath = OrigPath.substr(OldPrefix.size());
+ SmallString<256> NewPath;
+ path::append(NewPath, style, NewPrefix);
+ path::append(NewPath, style, RelPath);
+ Path.swap(NewPath);
+}
+
+void native(const Twine &path, SmallVectorImpl<char> &result, Style style) {
+ assert((!path.isSingleStringRef() ||
+ path.getSingleStringRef().data() != result.data()) &&
+ "path and result are not allowed to overlap!");
+ // Clear result.
+ result.clear();
+ path.toVector(result);
+ native(result, style);
+}
+
+void native(SmallVectorImpl<char> &Path, Style style) {
+ if (Path.empty())
+ return;
+ if (real_style(style) == Style::windows) {
+ std::replace(Path.begin(), Path.end(), '/', '\\');
+ if (Path[0] == '~' && (Path.size() == 1 || is_separator(Path[1], style))) {
+ llvm_unreachable("BINARYEN native");
+#if 0
+ SmallString<128> PathHome;
+ home_directory(PathHome);
+ PathHome.append(Path.begin() + 1, Path.end());
+ Path = PathHome;
+#endif
+ }
+ } else {
+ for (auto PI = Path.begin(), PE = Path.end(); PI < PE; ++PI) {
+ if (*PI == '\\') {
+ auto PN = PI + 1;
+ if (PN < PE && *PN == '\\')
+ ++PI; // increment once, the for loop will move over the escaped slash
+ else
+ *PI = '/';
+ }
+ }
+ }
+}
+
+std::string convert_to_slash(StringRef path, Style style) {
+ if (real_style(style) != Style::windows)
+ return path;
+
+ std::string s = path.str();
+ std::replace(s.begin(), s.end(), '\\', '/');
+ return s;
+}
+
+StringRef filename(StringRef path, Style style) { return *rbegin(path, style); }
+
+StringRef stem(StringRef path, Style style) {
+ StringRef fname = filename(path, style);
+ size_t pos = fname.find_last_of('.');
+ if (pos == StringRef::npos)
+ return fname;
+ else
+ if ((fname.size() == 1 && fname == ".") ||
+ (fname.size() == 2 && fname == ".."))
+ return fname;
+ else
+ return fname.substr(0, pos);
+}
+
+StringRef extension(StringRef path, Style style) {
+ StringRef fname = filename(path, style);
+ size_t pos = fname.find_last_of('.');
+ if (pos == StringRef::npos)
+ return StringRef();
+ else
+ if ((fname.size() == 1 && fname == ".") ||
+ (fname.size() == 2 && fname == ".."))
+ return StringRef();
+ else
+ return fname.substr(pos);
+}
+
+bool is_separator(char value, Style style) {
+ if (value == '/')
+ return true;
+ if (real_style(style) == Style::windows)
+ return value == '\\';
+ return false;
+}
+
+StringRef get_separator(Style style) {
+ if (real_style(style) == Style::windows)
+ return "\\";
+ return "/";
+}
+
+bool has_root_name(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !root_name(p, style).empty();
+}
+
+bool has_root_directory(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !root_directory(p, style).empty();
+}
+
+bool has_root_path(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !root_path(p, style).empty();
+}
+
+bool has_relative_path(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !relative_path(p, style).empty();
+}
+
+bool has_filename(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !filename(p, style).empty();
+}
+
+bool has_parent_path(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !parent_path(p, style).empty();
+}
+
+bool has_stem(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !stem(p, style).empty();
+}
+
+bool has_extension(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ return !extension(p, style).empty();
+}
+
+bool is_absolute(const Twine &path, Style style) {
+ SmallString<128> path_storage;
+ StringRef p = path.toStringRef(path_storage);
+
+ bool rootDir = has_root_directory(p, style);
+ bool rootName =
+ (real_style(style) != Style::windows) || has_root_name(p, style);
+
+ return rootDir && rootName;
+}
+
+bool is_relative(const Twine &path, Style style) {
+ return !is_absolute(path, style);
+}
+
+StringRef remove_leading_dotslash(StringRef Path, Style style) {
+ // Remove leading "./" (or ".//" or "././" etc.)
+ while (Path.size() > 2 && Path[0] == '.' && is_separator(Path[1], style)) {
+ Path = Path.substr(2);
+ while (Path.size() > 0 && is_separator(Path[0], style))
+ Path = Path.substr(1);
+ }
+ return Path;
+}
+
+static SmallString<256> remove_dots(StringRef path, bool remove_dot_dot,
+ Style style) {
+ SmallVector<StringRef, 16> components;
+
+ // Skip the root path, then look for traversal in the components.
+ StringRef rel = path::relative_path(path, style);
+ for (StringRef C :
+ llvm::make_range(path::begin(rel, style), path::end(rel))) {
+ if (C == ".")
+ continue;
+ // Leading ".." will remain in the path unless it's at the root.
+ if (remove_dot_dot && C == "..") {
+ if (!components.empty() && components.back() != "..") {
+ components.pop_back();
+ continue;
+ }
+ if (path::is_absolute(path, style))
+ continue;
+ }
+ components.push_back(C);
+ }
+
+ SmallString<256> buffer = path::root_path(path, style);
+ for (StringRef C : components)
+ path::append(buffer, style, C);
+ return buffer;
+}
+
+bool remove_dots(SmallVectorImpl<char> &path, bool remove_dot_dot,
+ Style style) {
+ StringRef p(path.data(), path.size());
+
+ SmallString<256> result = remove_dots(p, remove_dot_dot, style);
+ if (result == path)
+ return false;
+
+ path.swap(result);
+ return true;
+}
+
+} // end namespace path
+
+#if 0 // XXX BINARYEN
+
+namespace fs {
+
+std::error_code getUniqueID(const Twine Path, UniqueID &Result) {
+ file_status Status;
+ std::error_code EC = status(Path, Status);
+ if (EC)
+ return EC;
+ Result = Status.getUniqueID();
+ return std::error_code();
+}
+
+void createUniquePath(const Twine &Model, SmallVectorImpl<char> &ResultPath,
+ bool MakeAbsolute) {
+ SmallString<128> ModelStorage;
+ Model.toVector(ModelStorage);
+
+ if (MakeAbsolute) {
+ // Make model absolute by prepending a temp directory if it's not already.
+ if (!sys::path::is_absolute(Twine(ModelStorage))) {
+ SmallString<128> TDir;
+ sys::path::system_temp_directory(true, TDir);
+ sys::path::append(TDir, Twine(ModelStorage));
+ ModelStorage.swap(TDir);
+ }
+ }
+
+ ResultPath = ModelStorage;
+ ResultPath.push_back(0);
+ ResultPath.pop_back();
+
+ // Replace '%' with random chars.
+ for (unsigned i = 0, e = ModelStorage.size(); i != e; ++i) {
+ if (ModelStorage[i] == '%')
+ ResultPath[i] = "0123456789abcdef"[sys::Process::GetRandomNumber() & 15];
+ }
+}
+
+std::error_code createUniqueFile(const Twine &Model, int &ResultFd,
+ SmallVectorImpl<char> &ResultPath,
+ unsigned Mode) {
+ return createUniqueEntity(Model, ResultFd, ResultPath, false, Mode, FS_File);
+}
+
+static std::error_code createUniqueFile(const Twine &Model, int &ResultFd,
+ SmallVectorImpl<char> &ResultPath,
+ unsigned Mode, OpenFlags Flags) {
+ return createUniqueEntity(Model, ResultFd, ResultPath, false, Mode, FS_File,
+ Flags);
+}
+
+std::error_code createUniqueFile(const Twine &Model,
+ SmallVectorImpl<char> &ResultPath,
+ unsigned Mode) {
+ int FD;
+ auto EC = createUniqueFile(Model, FD, ResultPath, Mode);
+ if (EC)
+ return EC;
+ // FD is only needed to avoid race conditions. Close it right away.
+ close(FD);
+ return EC;
+}
+
+static std::error_code
+createTemporaryFile(const Twine &Model, int &ResultFD,
+ llvm::SmallVectorImpl<char> &ResultPath, FSEntity Type) {
+ SmallString<128> Storage;
+ StringRef P = Model.toNullTerminatedStringRef(Storage);
+ assert(P.find_first_of(separators(Style::native)) == StringRef::npos &&
+ "Model must be a simple filename.");
+ // Use P.begin() so that createUniqueEntity doesn't need to recreate Storage.
+ return createUniqueEntity(P.begin(), ResultFD, ResultPath, true,
+ owner_read | owner_write, Type);
+}
+
+static std::error_code
+createTemporaryFile(const Twine &Prefix, StringRef Suffix, int &ResultFD,
+ llvm::SmallVectorImpl<char> &ResultPath, FSEntity Type) {
+ const char *Middle = Suffix.empty() ? "-%%%%%%" : "-%%%%%%.";
+ return createTemporaryFile(Prefix + Middle + Suffix, ResultFD, ResultPath,
+ Type);
+}
+
+std::error_code createTemporaryFile(const Twine &Prefix, StringRef Suffix,
+ int &ResultFD,
+ SmallVectorImpl<char> &ResultPath) {
+ return createTemporaryFile(Prefix, Suffix, ResultFD, ResultPath, FS_File);
+}
+
+std::error_code createTemporaryFile(const Twine &Prefix, StringRef Suffix,
+ SmallVectorImpl<char> &ResultPath) {
+ int FD;
+ auto EC = createTemporaryFile(Prefix, Suffix, FD, ResultPath);
+ if (EC)
+ return EC;
+ // FD is only needed to avoid race conditions. Close it right away.
+ close(FD);
+ return EC;
+}
+
+
+// This is a mkdtemp with a different pattern. We use createUniqueEntity mostly
+// for consistency. We should try using mkdtemp.
+std::error_code createUniqueDirectory(const Twine &Prefix,
+ SmallVectorImpl<char> &ResultPath) {
+ int Dummy;
+ return createUniqueEntity(Prefix + "-%%%%%%", Dummy, ResultPath, true, 0,
+ FS_Dir);
+}
+
+std::error_code
+getPotentiallyUniqueFileName(const Twine &Model,
+ SmallVectorImpl<char> &ResultPath) {
+ int Dummy;
+ return createUniqueEntity(Model, Dummy, ResultPath, false, 0, FS_Name);
+}
+
+std::error_code
+getPotentiallyUniqueTempFileName(const Twine &Prefix, StringRef Suffix,
+ SmallVectorImpl<char> &ResultPath) {
+ int Dummy;
+ return createTemporaryFile(Prefix, Suffix, Dummy, ResultPath, FS_Name);
+}
+
+void make_absolute(const Twine ¤t_directory,
+ SmallVectorImpl<char> &path) {
+ StringRef p(path.data(), path.size());
+
+ bool rootDirectory = path::has_root_directory(p);
+ bool rootName = path::has_root_name(p);
+
+ // Already absolute.
+ if ((rootName || real_style(Style::native) != Style::windows) &&
+ rootDirectory)
+ return;
+
+ // All of the following conditions will need the current directory.
+ SmallString<128> current_dir;
+ current_directory.toVector(current_dir);
+
+ // Relative path. Prepend the current directory.
+ if (!rootName && !rootDirectory) {
+ // Append path to the current directory.
+ path::append(current_dir, p);
+ // Set path to the result.
+ path.swap(current_dir);
+ return;
+ }
+
+ if (!rootName && rootDirectory) {
+ StringRef cdrn = path::root_name(current_dir);
+ SmallString<128> curDirRootName(cdrn.begin(), cdrn.end());
+ path::append(curDirRootName, p);
+ // Set path to the result.
+ path.swap(curDirRootName);
+ return;
+ }
+
+ if (rootName && !rootDirectory) {
+ StringRef pRootName = path::root_name(p);
+ StringRef bRootDirectory = path::root_directory(current_dir);
+ StringRef bRelativePath = path::relative_path(current_dir);
+ StringRef pRelativePath = path::relative_path(p);
+
+ SmallString<128> res;
+ path::append(res, pRootName, bRootDirectory, bRelativePath, pRelativePath);
+ path.swap(res);
+ return;
+ }
+
+ llvm_unreachable("All rootName and rootDirectory combinations should have "
+ "occurred above!");
+}
+
+std::error_code make_absolute(SmallVectorImpl<char> &path) {
+ if (path::is_absolute(path))
+ return {};
+
+ SmallString<128> current_dir;
+ if (std::error_code ec = current_path(current_dir))
+ return ec;
+
+ make_absolute(current_dir, path);
+ return {};
+}
+
+std::error_code create_directories(const Twine &Path, bool IgnoreExisting,
+ perms Perms) {
+ SmallString<128> PathStorage;
+ StringRef P = Path.toStringRef(PathStorage);
+
+ // Be optimistic and try to create the directory
+ std::error_code EC = create_directory(P, IgnoreExisting, Perms);
+ // If we succeeded, or had any error other than the parent not existing, just
+ // return it.
+ if (EC != errc::no_such_file_or_directory)
+ return EC;
+
+ // We failed because of a no_such_file_or_directory, try to create the
+ // parent.
+ StringRef Parent = path::parent_path(P);
+ if (Parent.empty())
+ return EC;
+
+ if ((EC = create_directories(Parent, IgnoreExisting, Perms)))
+ return EC;
+
+ return create_directory(P, IgnoreExisting, Perms);
+}
+
+static std::error_code copy_file_internal(int ReadFD, int WriteFD) {
+ const size_t BufSize = 4096;
+ char *Buf = new char[BufSize];
+ int BytesRead = 0, BytesWritten = 0;
+ for (;;) {
+ BytesRead = read(ReadFD, Buf, BufSize);
+ if (BytesRead <= 0)
+ break;
+ while (BytesRead) {
+ BytesWritten = write(WriteFD, Buf, BytesRead);
+ if (BytesWritten < 0)
+ break;
+ BytesRead -= BytesWritten;
+ }
+ if (BytesWritten < 0)
+ break;
+ }
+ delete[] Buf;
+
+ if (BytesRead < 0 || BytesWritten < 0)
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+}
+
+#ifndef __APPLE__
+std::error_code copy_file(const Twine &From, const Twine &To) {
+ int ReadFD, WriteFD;
+ if (std::error_code EC = openFileForRead(From, ReadFD, OF_None))
+ return EC;
+ if (std::error_code EC =
+ openFileForWrite(To, WriteFD, CD_CreateAlways, OF_None)) {
+ close(ReadFD);
+ return EC;
+ }
+
+ std::error_code EC = copy_file_internal(ReadFD, WriteFD);
+
+ close(ReadFD);
+ close(WriteFD);
+
+ return EC;
+}
+#endif
+
+std::error_code copy_file(const Twine &From, int ToFD) {
+ int ReadFD;
+ if (std::error_code EC = openFileForRead(From, ReadFD, OF_None))
+ return EC;
+
+ std::error_code EC = copy_file_internal(ReadFD, ToFD);
+
+ close(ReadFD);
+
+ return EC;
+}
+
+ErrorOr<MD5::MD5Result> md5_contents(int FD) {
+ MD5 Hash;
+
+ constexpr size_t BufSize = 4096;
+ std::vector<uint8_t> Buf(BufSize);
+ int BytesRead = 0;
+ for (;;) {
+ BytesRead = read(FD, Buf.data(), BufSize);
+ if (BytesRead <= 0)
+ break;
+ Hash.update(makeArrayRef(Buf.data(), BytesRead));
+ }
+
+ if (BytesRead < 0)
+ return std::error_code(errno, std::generic_category());
+ MD5::MD5Result Result;
+ Hash.final(Result);
+ return Result;
+}
+
+ErrorOr<MD5::MD5Result> md5_contents(const Twine &Path) {
+ int FD;
+ if (auto EC = openFileForRead(Path, FD, OF_None))
+ return EC;
+
+ auto Result = md5_contents(FD);
+ close(FD);
+ return Result;
+}
+
+bool exists(const basic_file_status &status) {
+ return status_known(status) && status.type() != file_type::file_not_found;
+}
+
+bool status_known(const basic_file_status &s) {
+ return s.type() != file_type::status_error;
+}
+
+file_type get_file_type(const Twine &Path, bool Follow) {
+ file_status st;
+ if (status(Path, st, Follow))
+ return file_type::status_error;
+ return st.type();
+}
+
+bool is_directory(const basic_file_status &status) {
+ return status.type() == file_type::directory_file;
+}
+
+std::error_code is_directory(const Twine &path, bool &result) {
+ file_status st;
+ if (std::error_code ec = status(path, st))
+ return ec;
+ result = is_directory(st);
+ return std::error_code();
+}
+
+bool is_regular_file(const basic_file_status &status) {
+ return status.type() == file_type::regular_file;
+}
+
+std::error_code is_regular_file(const Twine &path, bool &result) {
+ file_status st;
+ if (std::error_code ec = status(path, st))
+ return ec;
+ result = is_regular_file(st);
+ return std::error_code();
+}
+
+bool is_symlink_file(const basic_file_status &status) {
+ return status.type() == file_type::symlink_file;
+}
+
+std::error_code is_symlink_file(const Twine &path, bool &result) {
+ file_status st;
+ if (std::error_code ec = status(path, st, false))
+ return ec;
+ result = is_symlink_file(st);
+ return std::error_code();
+}
+
+bool is_other(const basic_file_status &status) {
+ return exists(status) &&
+ !is_regular_file(status) &&
+ !is_directory(status);
+}
+
+std::error_code is_other(const Twine &Path, bool &Result) {
+ file_status FileStatus;
+ if (std::error_code EC = status(Path, FileStatus))
+ return EC;
+ Result = is_other(FileStatus);
+ return std::error_code();
+}
+
+void directory_entry::replace_filename(const Twine &Filename, file_type Type,
+ basic_file_status Status) {
+ SmallString<128> PathStr = path::parent_path(Path);
+ path::append(PathStr, Filename);
+ this->Path = PathStr.str();
+ this->Type = Type;
+ this->Status = Status;
+}
+
+ErrorOr<perms> getPermissions(const Twine &Path) {
+ file_status Status;
+ if (std::error_code EC = status(Path, Status))
+ return EC;
+
+ return Status.permissions();
+}
+
+} // end namespace fs
+
+#endif // XXX BINARYEN
+
+} // end namespace sys
+} // end namespace llvm
+
+// Include the truly platform-specific parts.
+#if 0 // XXX BINARYEN - we don't need platform-specific parts.
+#if defined(LLVM_ON_UNIX)
+#include "Unix/Path.inc"
+#endif
+#if defined(_WIN32)
+#include "Windows/Path.inc"
+#endif
+#endif
+
+#if 0 // XXX BINARYEN
+
+namespace llvm {
+namespace sys {
+namespace fs {
+TempFile::TempFile(StringRef Name, int FD) : TmpName(Name), FD(FD) {}
+TempFile::TempFile(TempFile &&Other) { *this = std::move(Other); }
+TempFile &TempFile::operator=(TempFile &&Other) {
+ TmpName = std::move(Other.TmpName);
+ FD = Other.FD;
+ Other.Done = true;
+ Other.FD = -1;
+ return *this;
+}
+
+TempFile::~TempFile() { assert(Done); }
+
+Error TempFile::discard() {
+ Done = true;
+ if (FD != -1 && close(FD) == -1) {
+ std::error_code EC = std::error_code(errno, std::generic_category());
+ return errorCodeToError(EC);
+ }
+ FD = -1;
+
+#ifdef _WIN32
+ // On windows closing will remove the file.
+ TmpName = "";
+ return Error::success();
+#else
+ // Always try to close and remove.
+ std::error_code RemoveEC;
+ if (!TmpName.empty()) {
+ RemoveEC = fs::remove(TmpName);
+ sys::DontRemoveFileOnSignal(TmpName);
+ if (!RemoveEC)
+ TmpName = "";
+ }
+ return errorCodeToError(RemoveEC);
+#endif
+}
+
+Error TempFile::keep(const Twine &Name) {
+ assert(!Done);
+ Done = true;
+ // Always try to close and rename.
+#ifdef _WIN32
+ // If we can't cancel the delete don't rename.
+ auto H = reinterpret_cast<HANDLE>(_get_osfhandle(FD));
+ std::error_code RenameEC = setDeleteDisposition(H, false);
+ if (!RenameEC) {
+ RenameEC = rename_fd(FD, Name);
+ // If rename failed because it's cross-device, copy instead
+ if (RenameEC ==
+ std::error_code(ERROR_NOT_SAME_DEVICE, std::system_category())) {
+ RenameEC = copy_file(TmpName, Name);
+ setDeleteDisposition(H, true);
+ }
+ }
+
+ // If we can't rename, discard the temporary file.
+ if (RenameEC)
+ setDeleteDisposition(H, true);
+#else
+ std::error_code RenameEC = fs::rename(TmpName, Name);
+ if (RenameEC) {
+ // If we can't rename, try to copy to work around cross-device link issues.
+ RenameEC = sys::fs::copy_file(TmpName, Name);
+ // If we can't rename or copy, discard the temporary file.
+ if (RenameEC)
+ remove(TmpName);
+ }
+ sys::DontRemoveFileOnSignal(TmpName);
+#endif
+
+ if (!RenameEC)
+ TmpName = "";
+
+ if (close(FD) == -1) {
+ std::error_code EC(errno, std::generic_category());
+ return errorCodeToError(EC);
+ }
+ FD = -1;
+
+ return errorCodeToError(RenameEC);
+}
+
+Error TempFile::keep() {
+ assert(!Done);
+ Done = true;
+
+#ifdef _WIN32
+ auto H = reinterpret_cast<HANDLE>(_get_osfhandle(FD));
+ if (std::error_code EC = setDeleteDisposition(H, false))
+ return errorCodeToError(EC);
+#else
+ sys::DontRemoveFileOnSignal(TmpName);
+#endif
+
+ TmpName = "";
+
+ if (close(FD) == -1) {
+ std::error_code EC(errno, std::generic_category());
+ return errorCodeToError(EC);
+ }
+ FD = -1;
+
+ return Error::success();
+}
+
+Expected<TempFile> TempFile::create(const Twine &Model, unsigned Mode) {
+ int FD;
+ SmallString<128> ResultPath;
+ if (std::error_code EC =
+ createUniqueFile(Model, FD, ResultPath, Mode, OF_Delete))
+ return errorCodeToError(EC);
+
+ TempFile Ret(ResultPath, FD);
+#ifndef _WIN32
+ if (sys::RemoveFileOnSignal(ResultPath)) {
+ // Make sure we delete the file when RemoveFileOnSignal fails.
+ consumeError(Ret.discard());
+ std::error_code EC(errc::operation_not_permitted);
+ return errorCodeToError(EC);
+ }
+#endif
+ return std::move(Ret);
+}
+}
+
+} // end namsspace sys
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/ScopedPrinter.cpp b/third_party/llvm-project/ScopedPrinter.cpp
new file mode 100644
index 000000000..981dfbff5
--- /dev/null
+++ b/third_party/llvm-project/ScopedPrinter.cpp
@@ -0,0 +1,46 @@
+#include "llvm/Support/ScopedPrinter.h"
+
+#include "llvm/Support/Format.h"
+#include <cctype>
+
+using namespace llvm::support;
+
+namespace llvm {
+
+raw_ostream &operator<<(raw_ostream &OS, const HexNumber &Value) {
+ OS << "0x" << to_hexString(Value.Value);
+ return OS;
+}
+
+const std::string to_hexString(uint64_t Value, bool UpperCase) {
+ std::string number;
+ llvm::raw_string_ostream stream(number);
+ stream << format_hex_no_prefix(Value, 1, UpperCase);
+ return stream.str();
+}
+
+void ScopedPrinter::printBinaryImpl(StringRef Label, StringRef Str,
+ ArrayRef<uint8_t> Data, bool Block,
+ uint32_t StartOffset) {
+ if (Data.size() > 16)
+ Block = true;
+
+ if (Block) {
+ startLine() << Label;
+ if (!Str.empty())
+ OS << ": " << Str;
+ OS << " (\n";
+ if (!Data.empty())
+ OS << format_bytes_with_ascii(Data, StartOffset, 16, 4,
+ (IndentLevel + 1) * 2, true)
+ << "\n";
+ startLine() << ")\n";
+ } else {
+ startLine() << Label << ":";
+ if (!Str.empty())
+ OS << " " << Str;
+ OS << " (" << format_bytes(Data, None, Data.size(), 1, 0, true) << ")\n";
+ }
+}
+
+} // namespace llvm
diff --git a/third_party/llvm-project/SmallVector.cpp b/third_party/llvm-project/SmallVector.cpp
new file mode 100644
index 000000000..36f0a81f6
--- /dev/null
+++ b/third_party/llvm-project/SmallVector.cpp
@@ -0,0 +1,65 @@
+//===- llvm/ADT/SmallVector.cpp - 'Normally small' vectors ----------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the SmallVector class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/SmallVector.h"
+using namespace llvm;
+
+// Check that no bytes are wasted and everything is well-aligned.
+namespace {
+struct Struct16B {
+ alignas(16) void *X;
+};
+struct Struct32B {
+ alignas(32) void *X;
+};
+}
+static_assert(sizeof(SmallVector<void *, 0>) ==
+ sizeof(unsigned) * 2 + sizeof(void *),
+ "wasted space in SmallVector size 0");
+static_assert(alignof(SmallVector<Struct16B, 0>) >= alignof(Struct16B),
+ "wrong alignment for 16-byte aligned T");
+static_assert(alignof(SmallVector<Struct32B, 0>) >= alignof(Struct32B),
+ "wrong alignment for 32-byte aligned T");
+static_assert(sizeof(SmallVector<Struct16B, 0>) >= alignof(Struct16B),
+ "missing padding for 16-byte aligned T");
+static_assert(sizeof(SmallVector<Struct32B, 0>) >= alignof(Struct32B),
+ "missing padding for 32-byte aligned T");
+static_assert(sizeof(SmallVector<void *, 1>) ==
+ sizeof(unsigned) * 2 + sizeof(void *) * 2,
+ "wasted space in SmallVector size 1");
+
+/// grow_pod - This is an implementation of the grow() method which only works
+/// on POD-like datatypes and is out of line to reduce code duplication.
+void SmallVectorBase::grow_pod(void *FirstEl, size_t MinCapacity,
+ size_t TSize) {
+ // Ensure we can fit the new capacity in 32 bits.
+ if (MinCapacity > UINT32_MAX)
+ report_bad_alloc_error("SmallVector capacity overflow during allocation");
+
+ size_t NewCapacity = 2 * capacity() + 1; // Always grow.
+ NewCapacity =
+ std::min(std::max(NewCapacity, MinCapacity), size_t(UINT32_MAX));
+
+ void *NewElts;
+ if (BeginX == FirstEl) {
+ NewElts = safe_malloc(NewCapacity * TSize);
+
+ // Copy the elements over. No need to run dtors on PODs.
+ memcpy(NewElts, this->BeginX, size() * TSize);
+ } else {
+ // If this wasn't grown from the inline copy, grow the allocated space.
+ NewElts = safe_realloc(this->BeginX, NewCapacity * TSize);
+ }
+
+ this->BeginX = NewElts;
+ this->Capacity = NewCapacity;
+}
diff --git a/third_party/llvm-project/SourceMgr.cpp b/third_party/llvm-project/SourceMgr.cpp
new file mode 100644
index 000000000..34ddf5e78
--- /dev/null
+++ b/third_party/llvm-project/SourceMgr.cpp
@@ -0,0 +1,503 @@
+//===- SourceMgr.cpp - Manager for Simple Source Buffers & Diagnostics ----===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the SourceMgr class. This class is used as a simple
+// substrate for diagnostics, #include handling, and other low level things for
+// simple parsers.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/SourceMgr.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/Locale.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/Path.h"
+#include "llvm/Support/SMLoc.h"
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <limits>
+#include <memory>
+#include <string>
+#include <utility>
+
+using namespace llvm;
+
+static const size_t TabStop = 8;
+
+unsigned SourceMgr::AddIncludeFile(const std::string &Filename,
+ SMLoc IncludeLoc,
+ std::string &IncludedFile) {
+ IncludedFile = Filename;
+ ErrorOr<std::unique_ptr<MemoryBuffer>> NewBufOrErr =
+ MemoryBuffer::getFile(IncludedFile);
+
+ // If the file didn't exist directly, see if it's in an include path.
+ for (unsigned i = 0, e = IncludeDirectories.size(); i != e && !NewBufOrErr;
+ ++i) {
+ IncludedFile =
+ IncludeDirectories[i] + sys::path::get_separator().data() + Filename;
+ NewBufOrErr = MemoryBuffer::getFile(IncludedFile);
+ }
+
+ if (!NewBufOrErr)
+ return 0;
+
+ return AddNewSourceBuffer(std::move(*NewBufOrErr), IncludeLoc);
+}
+
+unsigned SourceMgr::FindBufferContainingLoc(SMLoc Loc) const {
+ for (unsigned i = 0, e = Buffers.size(); i != e; ++i)
+ if (Loc.getPointer() >= Buffers[i].Buffer->getBufferStart() &&
+ // Use <= here so that a pointer to the null at the end of the buffer
+ // is included as part of the buffer.
+ Loc.getPointer() <= Buffers[i].Buffer->getBufferEnd())
+ return i + 1;
+ return 0;
+}
+
+template <typename T>
+unsigned SourceMgr::SrcBuffer::getLineNumber(const char *Ptr) const {
+
+ // Ensure OffsetCache is allocated and populated with offsets of all the
+ // '\n' bytes.
+ std::vector<T> *Offsets = nullptr;
+ if (OffsetCache.isNull()) {
+ Offsets = new std::vector<T>();
+ OffsetCache = Offsets;
+ size_t Sz = Buffer->getBufferSize();
+ assert(Sz <= std::numeric_limits<T>::max());
+ StringRef S = Buffer->getBuffer();
+ for (size_t N = 0; N < Sz; ++N) {
+ if (S[N] == '\n') {
+ Offsets->push_back(static_cast<T>(N));
+ }
+ }
+ } else {
+ Offsets = OffsetCache.get<std::vector<T> *>();
+ }
+
+ const char *BufStart = Buffer->getBufferStart();
+ assert(Ptr >= BufStart && Ptr <= Buffer->getBufferEnd());
+ ptrdiff_t PtrDiff = Ptr - BufStart;
+ assert(PtrDiff >= 0 && static_cast<size_t>(PtrDiff) <= std::numeric_limits<T>::max());
+ T PtrOffset = static_cast<T>(PtrDiff);
+
+ // llvm::lower_bound gives the number of EOL before PtrOffset. Add 1 to get
+ // the line number.
+ return llvm::lower_bound(*Offsets, PtrOffset) - Offsets->begin() + 1;
+}
+
+SourceMgr::SrcBuffer::SrcBuffer(SourceMgr::SrcBuffer &&Other)
+ : Buffer(std::move(Other.Buffer)),
+ OffsetCache(Other.OffsetCache),
+ IncludeLoc(Other.IncludeLoc) {
+ Other.OffsetCache = nullptr;
+}
+
+SourceMgr::SrcBuffer::~SrcBuffer() {
+ if (!OffsetCache.isNull()) {
+ if (OffsetCache.is<std::vector<uint8_t>*>())
+ delete OffsetCache.get<std::vector<uint8_t>*>();
+ else if (OffsetCache.is<std::vector<uint16_t>*>())
+ delete OffsetCache.get<std::vector<uint16_t>*>();
+ else if (OffsetCache.is<std::vector<uint32_t>*>())
+ delete OffsetCache.get<std::vector<uint32_t>*>();
+ else
+ delete OffsetCache.get<std::vector<uint64_t>*>();
+ OffsetCache = nullptr;
+ }
+}
+
+std::pair<unsigned, unsigned>
+SourceMgr::getLineAndColumn(SMLoc Loc, unsigned BufferID) const {
+ if (!BufferID)
+ BufferID = FindBufferContainingLoc(Loc);
+ assert(BufferID && "Invalid Location!");
+
+ auto &SB = getBufferInfo(BufferID);
+ const char *Ptr = Loc.getPointer();
+
+ size_t Sz = SB.Buffer->getBufferSize();
+ unsigned LineNo;
+ if (Sz <= std::numeric_limits<uint8_t>::max())
+ LineNo = SB.getLineNumber<uint8_t>(Ptr);
+ else if (Sz <= std::numeric_limits<uint16_t>::max())
+ LineNo = SB.getLineNumber<uint16_t>(Ptr);
+ else if (Sz <= std::numeric_limits<uint32_t>::max())
+ LineNo = SB.getLineNumber<uint32_t>(Ptr);
+ else
+ LineNo = SB.getLineNumber<uint64_t>(Ptr);
+
+ const char *BufStart = SB.Buffer->getBufferStart();
+ size_t NewlineOffs = StringRef(BufStart, Ptr-BufStart).find_last_of("\n\r");
+ if (NewlineOffs == StringRef::npos) NewlineOffs = ~(size_t)0;
+ return std::make_pair(LineNo, Ptr-BufStart-NewlineOffs);
+}
+
+void SourceMgr::PrintIncludeStack(SMLoc IncludeLoc, raw_ostream &OS) const {
+ if (IncludeLoc == SMLoc()) return; // Top of stack.
+
+ unsigned CurBuf = FindBufferContainingLoc(IncludeLoc);
+ assert(CurBuf && "Invalid or unspecified location!");
+
+ PrintIncludeStack(getBufferInfo(CurBuf).IncludeLoc, OS);
+
+ OS << "Included from "
+ << getBufferInfo(CurBuf).Buffer->getBufferIdentifier()
+ << ":" << FindLineNumber(IncludeLoc, CurBuf) << ":\n";
+}
+
+SMDiagnostic SourceMgr::GetMessage(SMLoc Loc, SourceMgr::DiagKind Kind,
+ const Twine &Msg,
+ ArrayRef<SMRange> Ranges,
+ ArrayRef<SMFixIt> FixIts) const {
+ // First thing to do: find the current buffer containing the specified
+ // location to pull out the source line.
+ SmallVector<std::pair<unsigned, unsigned>, 4> ColRanges;
+ std::pair<unsigned, unsigned> LineAndCol;
+ StringRef BufferID = "<unknown>";
+ std::string LineStr;
+
+ if (Loc.isValid()) {
+ unsigned CurBuf = FindBufferContainingLoc(Loc);
+ assert(CurBuf && "Invalid or unspecified location!");
+
+ const MemoryBuffer *CurMB = getMemoryBuffer(CurBuf);
+ BufferID = CurMB->getBufferIdentifier();
+
+ // Scan backward to find the start of the line.
+ const char *LineStart = Loc.getPointer();
+ const char *BufStart = CurMB->getBufferStart();
+ while (LineStart != BufStart && LineStart[-1] != '\n' &&
+ LineStart[-1] != '\r')
+ --LineStart;
+
+ // Get the end of the line.
+ const char *LineEnd = Loc.getPointer();
+ const char *BufEnd = CurMB->getBufferEnd();
+ while (LineEnd != BufEnd && LineEnd[0] != '\n' && LineEnd[0] != '\r')
+ ++LineEnd;
+ LineStr = std::string(LineStart, LineEnd);
+
+ // Convert any ranges to column ranges that only intersect the line of the
+ // location.
+ for (unsigned i = 0, e = Ranges.size(); i != e; ++i) {
+ SMRange R = Ranges[i];
+ if (!R.isValid()) continue;
+
+ // If the line doesn't contain any part of the range, then ignore it.
+ if (R.Start.getPointer() > LineEnd || R.End.getPointer() < LineStart)
+ continue;
+
+ // Ignore pieces of the range that go onto other lines.
+ if (R.Start.getPointer() < LineStart)
+ R.Start = SMLoc::getFromPointer(LineStart);
+ if (R.End.getPointer() > LineEnd)
+ R.End = SMLoc::getFromPointer(LineEnd);
+
+ // Translate from SMLoc ranges to column ranges.
+ // FIXME: Handle multibyte characters.
+ ColRanges.push_back(std::make_pair(R.Start.getPointer()-LineStart,
+ R.End.getPointer()-LineStart));
+ }
+
+ LineAndCol = getLineAndColumn(Loc, CurBuf);
+ }
+
+ return SMDiagnostic(*this, Loc, BufferID, LineAndCol.first,
+ LineAndCol.second-1, Kind, Msg.str(),
+ LineStr, ColRanges, FixIts);
+}
+
+void SourceMgr::PrintMessage(raw_ostream &OS, const SMDiagnostic &Diagnostic,
+ bool ShowColors) const {
+ // Report the message with the diagnostic handler if present.
+ if (DiagHandler) {
+ DiagHandler(Diagnostic, DiagContext);
+ return;
+ }
+
+ if (Diagnostic.getLoc().isValid()) {
+ unsigned CurBuf = FindBufferContainingLoc(Diagnostic.getLoc());
+ assert(CurBuf && "Invalid or unspecified location!");
+ PrintIncludeStack(getBufferInfo(CurBuf).IncludeLoc, OS);
+ }
+
+ Diagnostic.print(nullptr, OS, ShowColors);
+}
+
+void SourceMgr::PrintMessage(raw_ostream &OS, SMLoc Loc,
+ SourceMgr::DiagKind Kind,
+ const Twine &Msg, ArrayRef<SMRange> Ranges,
+ ArrayRef<SMFixIt> FixIts, bool ShowColors) const {
+ PrintMessage(OS, GetMessage(Loc, Kind, Msg, Ranges, FixIts), ShowColors);
+}
+
+void SourceMgr::PrintMessage(SMLoc Loc, SourceMgr::DiagKind Kind,
+ const Twine &Msg, ArrayRef<SMRange> Ranges,
+ ArrayRef<SMFixIt> FixIts, bool ShowColors) const {
+ PrintMessage(errs(), Loc, Kind, Msg, Ranges, FixIts, ShowColors);
+}
+
+//===----------------------------------------------------------------------===//
+// SMDiagnostic Implementation
+//===----------------------------------------------------------------------===//
+
+SMDiagnostic::SMDiagnostic(const SourceMgr &sm, SMLoc L, StringRef FN,
+ int Line, int Col, SourceMgr::DiagKind Kind,
+ StringRef Msg, StringRef LineStr,
+ ArrayRef<std::pair<unsigned,unsigned>> Ranges,
+ ArrayRef<SMFixIt> Hints)
+ : SM(&sm), Loc(L), Filename(FN), LineNo(Line), ColumnNo(Col), Kind(Kind),
+ Message(Msg), LineContents(LineStr), Ranges(Ranges.vec()),
+ FixIts(Hints.begin(), Hints.end()) {
+ llvm::sort(FixIts);
+}
+
+static void buildFixItLine(std::string &CaretLine, std::string &FixItLine,
+ ArrayRef<SMFixIt> FixIts, ArrayRef<char> SourceLine){
+ if (FixIts.empty())
+ return;
+
+ const char *LineStart = SourceLine.begin();
+ const char *LineEnd = SourceLine.end();
+
+ size_t PrevHintEndCol = 0;
+
+ for (ArrayRef<SMFixIt>::iterator I = FixIts.begin(), E = FixIts.end();
+ I != E; ++I) {
+ // If the fixit contains a newline or tab, ignore it.
+ if (I->getText().find_first_of("\n\r\t") != StringRef::npos)
+ continue;
+
+ SMRange R = I->getRange();
+
+ // If the line doesn't contain any part of the range, then ignore it.
+ if (R.Start.getPointer() > LineEnd || R.End.getPointer() < LineStart)
+ continue;
+
+ // Translate from SMLoc to column.
+ // Ignore pieces of the range that go onto other lines.
+ // FIXME: Handle multibyte characters in the source line.
+ unsigned FirstCol;
+ if (R.Start.getPointer() < LineStart)
+ FirstCol = 0;
+ else
+ FirstCol = R.Start.getPointer() - LineStart;
+
+ // If we inserted a long previous hint, push this one forwards, and add
+ // an extra space to show that this is not part of the previous
+ // completion. This is sort of the best we can do when two hints appear
+ // to overlap.
+ //
+ // Note that if this hint is located immediately after the previous
+ // hint, no space will be added, since the location is more important.
+ unsigned HintCol = FirstCol;
+ if (HintCol < PrevHintEndCol)
+ HintCol = PrevHintEndCol + 1;
+
+#if 0 // XXX BINARYEN
+ // FIXME: This assertion is intended to catch unintended use of multibyte
+ // characters in fixits. If we decide to do this, we'll have to track
+ // separate byte widths for the source and fixit lines.
+ assert((size_t)sys::locale::columnWidth(I->getText()) ==
+ I->getText().size());
+#endif
+
+ // This relies on one byte per column in our fixit hints.
+ unsigned LastColumnModified = HintCol + I->getText().size();
+ if (LastColumnModified > FixItLine.size())
+ FixItLine.resize(LastColumnModified, ' ');
+
+ std::copy(I->getText().begin(), I->getText().end(),
+ FixItLine.begin() + HintCol);
+
+ PrevHintEndCol = LastColumnModified;
+
+ // For replacements, mark the removal range with '~'.
+ // FIXME: Handle multibyte characters in the source line.
+ unsigned LastCol;
+ if (R.End.getPointer() >= LineEnd)
+ LastCol = LineEnd - LineStart;
+ else
+ LastCol = R.End.getPointer() - LineStart;
+
+ std::fill(&CaretLine[FirstCol], &CaretLine[LastCol], '~');
+ }
+}
+
+static void printSourceLine(raw_ostream &S, StringRef LineContents) {
+ // Print out the source line one character at a time, so we can expand tabs.
+ for (unsigned i = 0, e = LineContents.size(), OutCol = 0; i != e; ++i) {
+ size_t NextTab = LineContents.find('\t', i);
+ // If there were no tabs left, print the rest, we are done.
+ if (NextTab == StringRef::npos) {
+ S << LineContents.drop_front(i);
+ break;
+ }
+
+ // Otherwise, print from i to NextTab.
+ S << LineContents.slice(i, NextTab);
+ OutCol += NextTab - i;
+ i = NextTab;
+
+ // If we have a tab, emit at least one space, then round up to 8 columns.
+ do {
+ S << ' ';
+ ++OutCol;
+ } while ((OutCol % TabStop) != 0);
+ }
+ S << '\n';
+}
+
+static bool isNonASCII(char c) {
+ return c & 0x80;
+}
+
+void SMDiagnostic::print(const char *ProgName, raw_ostream &OS,
+ bool ShowColors, bool ShowKindLabel) const {
+ {
+ WithColor S(OS, raw_ostream::SAVEDCOLOR, true, false, !ShowColors);
+
+ if (ProgName && ProgName[0])
+ S << ProgName << ": ";
+
+ if (!Filename.empty()) {
+ if (Filename == "-")
+ S << "<stdin>";
+ else
+ S << Filename;
+
+ if (LineNo != -1) {
+ S << ':' << LineNo;
+ if (ColumnNo != -1)
+ S << ':' << (ColumnNo + 1);
+ }
+ S << ": ";
+ }
+ }
+
+ if (ShowKindLabel) {
+ switch (Kind) {
+ case SourceMgr::DK_Error:
+ WithColor::error(OS, "", !ShowColors);
+ break;
+ case SourceMgr::DK_Warning:
+ WithColor::warning(OS, "", !ShowColors);
+ break;
+ case SourceMgr::DK_Note:
+ WithColor::note(OS, "", !ShowColors);
+ break;
+ case SourceMgr::DK_Remark:
+ WithColor::remark(OS, "", !ShowColors);
+ break;
+ }
+ }
+
+ WithColor(OS, raw_ostream::SAVEDCOLOR, true, false, !ShowColors)
+ << Message << '\n';
+
+ if (LineNo == -1 || ColumnNo == -1)
+ return;
+
+ // FIXME: If there are multibyte or multi-column characters in the source, all
+ // our ranges will be wrong. To do this properly, we'll need a byte-to-column
+ // map like Clang's TextDiagnostic. For now, we'll just handle tabs by
+ // expanding them later, and bail out rather than show incorrect ranges and
+ // misaligned fixits for any other odd characters.
+ if (find_if(LineContents, isNonASCII) != LineContents.end()) {
+ printSourceLine(OS, LineContents);
+ return;
+ }
+ size_t NumColumns = LineContents.size();
+
+ // Build the line with the caret and ranges.
+ std::string CaretLine(NumColumns+1, ' ');
+
+ // Expand any ranges.
+ for (unsigned r = 0, e = Ranges.size(); r != e; ++r) {
+ std::pair<unsigned, unsigned> R = Ranges[r];
+ std::fill(&CaretLine[R.first],
+ &CaretLine[std::min((size_t)R.second, CaretLine.size())],
+ '~');
+ }
+
+ // Add any fix-its.
+ // FIXME: Find the beginning of the line properly for multibyte characters.
+ std::string FixItInsertionLine;
+ buildFixItLine(CaretLine, FixItInsertionLine, FixIts,
+ makeArrayRef(Loc.getPointer() - ColumnNo,
+ LineContents.size()));
+
+ // Finally, plop on the caret.
+ if (unsigned(ColumnNo) <= NumColumns)
+ CaretLine[ColumnNo] = '^';
+ else
+ CaretLine[NumColumns] = '^';
+
+ // ... and remove trailing whitespace so the output doesn't wrap for it. We
+ // know that the line isn't completely empty because it has the caret in it at
+ // least.
+ CaretLine.erase(CaretLine.find_last_not_of(' ')+1);
+
+ printSourceLine(OS, LineContents);
+
+ {
+ WithColor S(OS, raw_ostream::GREEN, true, false, !ShowColors);
+
+ // Print out the caret line, matching tabs in the source line.
+ for (unsigned i = 0, e = CaretLine.size(), OutCol = 0; i != e; ++i) {
+ if (i >= LineContents.size() || LineContents[i] != '\t') {
+ S << CaretLine[i];
+ ++OutCol;
+ continue;
+ }
+
+ // Okay, we have a tab. Insert the appropriate number of characters.
+ do {
+ S << CaretLine[i];
+ ++OutCol;
+ } while ((OutCol % TabStop) != 0);
+ }
+ S << '\n';
+ }
+
+ // Print out the replacement line, matching tabs in the source line.
+ if (FixItInsertionLine.empty())
+ return;
+
+ for (size_t i = 0, e = FixItInsertionLine.size(), OutCol = 0; i < e; ++i) {
+ if (i >= LineContents.size() || LineContents[i] != '\t') {
+ OS << FixItInsertionLine[i];
+ ++OutCol;
+ continue;
+ }
+
+ // Okay, we have a tab. Insert the appropriate number of characters.
+ do {
+ OS << FixItInsertionLine[i];
+ // FIXME: This is trying not to break up replacements, but then to re-sync
+ // with the tabs between replacements. This will fail, though, if two
+ // fix-it replacements are exactly adjacent, or if a fix-it contains a
+ // space. Really we should be precomputing column widths, which we'll
+ // need anyway for multibyte chars.
+ if (FixItInsertionLine[i] != ' ')
+ ++i;
+ ++OutCol;
+ } while (((OutCol % TabStop) != 0) && i != e);
+ }
+ OS << '\n';
+}
diff --git a/third_party/llvm-project/StringMap.cpp b/third_party/llvm-project/StringMap.cpp
new file mode 100644
index 000000000..6b5ea020d
--- /dev/null
+++ b/third_party/llvm-project/StringMap.cpp
@@ -0,0 +1,261 @@
+//===--- StringMap.cpp - String Hash table map implementation -------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the StringMap class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DJB.h"
+#include "llvm/Support/MathExtras.h"
+#include <cassert>
+
+using namespace llvm;
+
+/// Returns the number of buckets to allocate to ensure that the DenseMap can
+/// accommodate \p NumEntries without need to grow().
+static unsigned getMinBucketToReserveForEntries(unsigned NumEntries) {
+ // Ensure that "NumEntries * 4 < NumBuckets * 3"
+ if (NumEntries == 0)
+ return 0;
+ // +1 is required because of the strict equality.
+ // For example if NumEntries is 48, we need to return 401.
+ return NextPowerOf2(NumEntries * 4 / 3 + 1);
+}
+
+StringMapImpl::StringMapImpl(unsigned InitSize, unsigned itemSize) {
+ ItemSize = itemSize;
+
+ // If a size is specified, initialize the table with that many buckets.
+ if (InitSize) {
+ // The table will grow when the number of entries reach 3/4 of the number of
+ // buckets. To guarantee that "InitSize" number of entries can be inserted
+ // in the table without growing, we allocate just what is needed here.
+ init(getMinBucketToReserveForEntries(InitSize));
+ return;
+ }
+
+ // Otherwise, initialize it with zero buckets to avoid the allocation.
+ TheTable = nullptr;
+ NumBuckets = 0;
+ NumItems = 0;
+ NumTombstones = 0;
+}
+
+void StringMapImpl::init(unsigned InitSize) {
+ assert((InitSize & (InitSize-1)) == 0 &&
+ "Init Size must be a power of 2 or zero!");
+
+ unsigned NewNumBuckets = InitSize ? InitSize : 16;
+ NumItems = 0;
+ NumTombstones = 0;
+
+ TheTable = static_cast<StringMapEntryBase **>(
+ safe_calloc(NewNumBuckets+1,
+ sizeof(StringMapEntryBase **) + sizeof(unsigned)));
+
+ // Set the member only if TheTable was successfully allocated
+ NumBuckets = NewNumBuckets;
+
+ // Allocate one extra bucket, set it to look filled so the iterators stop at
+ // end.
+ TheTable[NumBuckets] = (StringMapEntryBase*)2;
+}
+
+/// LookupBucketFor - Look up the bucket that the specified string should end
+/// up in. If it already exists as a key in the map, the Item pointer for the
+/// specified bucket will be non-null. Otherwise, it will be null. In either
+/// case, the FullHashValue field of the bucket will be set to the hash value
+/// of the string.
+unsigned StringMapImpl::LookupBucketFor(StringRef Name) {
+ unsigned HTSize = NumBuckets;
+ if (HTSize == 0) { // Hash table unallocated so far?
+ init(16);
+ HTSize = NumBuckets;
+ }
+ unsigned FullHashValue = djbHash(Name, 0);
+ unsigned BucketNo = FullHashValue & (HTSize-1);
+ unsigned *HashTable = (unsigned *)(TheTable + NumBuckets + 1);
+
+ unsigned ProbeAmt = 1;
+ int FirstTombstone = -1;
+ while (true) {
+ StringMapEntryBase *BucketItem = TheTable[BucketNo];
+ // If we found an empty bucket, this key isn't in the table yet, return it.
+ if (LLVM_LIKELY(!BucketItem)) {
+ // If we found a tombstone, we want to reuse the tombstone instead of an
+ // empty bucket. This reduces probing.
+ if (FirstTombstone != -1) {
+ HashTable[FirstTombstone] = FullHashValue;
+ return FirstTombstone;
+ }
+
+ HashTable[BucketNo] = FullHashValue;
+ return BucketNo;
+ }
+
+ if (BucketItem == getTombstoneVal()) {
+ // Skip over tombstones. However, remember the first one we see.
+ if (FirstTombstone == -1) FirstTombstone = BucketNo;
+ } else if (LLVM_LIKELY(HashTable[BucketNo] == FullHashValue)) {
+ // If the full hash value matches, check deeply for a match. The common
+ // case here is that we are only looking at the buckets (for item info
+ // being non-null and for the full hash value) not at the items. This
+ // is important for cache locality.
+
+ // Do the comparison like this because Name isn't necessarily
+ // null-terminated!
+ char *ItemStr = (char*)BucketItem+ItemSize;
+ if (Name == StringRef(ItemStr, BucketItem->getKeyLength())) {
+ // We found a match!
+ return BucketNo;
+ }
+ }
+
+ // Okay, we didn't find the item. Probe to the next bucket.
+ BucketNo = (BucketNo+ProbeAmt) & (HTSize-1);
+
+ // Use quadratic probing, it has fewer clumping artifacts than linear
+ // probing and has good cache behavior in the common case.
+ ++ProbeAmt;
+ }
+}
+
+/// FindKey - Look up the bucket that contains the specified key. If it exists
+/// in the map, return the bucket number of the key. Otherwise return -1.
+/// This does not modify the map.
+int StringMapImpl::FindKey(StringRef Key) const {
+ unsigned HTSize = NumBuckets;
+ if (HTSize == 0) return -1; // Really empty table?
+ unsigned FullHashValue = djbHash(Key, 0);
+ unsigned BucketNo = FullHashValue & (HTSize-1);
+ unsigned *HashTable = (unsigned *)(TheTable + NumBuckets + 1);
+
+ unsigned ProbeAmt = 1;
+ while (true) {
+ StringMapEntryBase *BucketItem = TheTable[BucketNo];
+ // If we found an empty bucket, this key isn't in the table yet, return.
+ if (LLVM_LIKELY(!BucketItem))
+ return -1;
+
+ if (BucketItem == getTombstoneVal()) {
+ // Ignore tombstones.
+ } else if (LLVM_LIKELY(HashTable[BucketNo] == FullHashValue)) {
+ // If the full hash value matches, check deeply for a match. The common
+ // case here is that we are only looking at the buckets (for item info
+ // being non-null and for the full hash value) not at the items. This
+ // is important for cache locality.
+
+ // Do the comparison like this because NameStart isn't necessarily
+ // null-terminated!
+ char *ItemStr = (char*)BucketItem+ItemSize;
+ if (Key == StringRef(ItemStr, BucketItem->getKeyLength())) {
+ // We found a match!
+ return BucketNo;
+ }
+ }
+
+ // Okay, we didn't find the item. Probe to the next bucket.
+ BucketNo = (BucketNo+ProbeAmt) & (HTSize-1);
+
+ // Use quadratic probing, it has fewer clumping artifacts than linear
+ // probing and has good cache behavior in the common case.
+ ++ProbeAmt;
+ }
+}
+
+/// RemoveKey - Remove the specified StringMapEntry from the table, but do not
+/// delete it. This aborts if the value isn't in the table.
+void StringMapImpl::RemoveKey(StringMapEntryBase *V) {
+ const char *VStr = (char*)V + ItemSize;
+ StringMapEntryBase *V2 = RemoveKey(StringRef(VStr, V->getKeyLength()));
+ (void)V2;
+ assert(V == V2 && "Didn't find key?");
+}
+
+/// RemoveKey - Remove the StringMapEntry for the specified key from the
+/// table, returning it. If the key is not in the table, this returns null.
+StringMapEntryBase *StringMapImpl::RemoveKey(StringRef Key) {
+ int Bucket = FindKey(Key);
+ if (Bucket == -1) return nullptr;
+
+ StringMapEntryBase *Result = TheTable[Bucket];
+ TheTable[Bucket] = getTombstoneVal();
+ --NumItems;
+ ++NumTombstones;
+ assert(NumItems + NumTombstones <= NumBuckets);
+
+ return Result;
+}
+
+/// RehashTable - Grow the table, redistributing values into the buckets with
+/// the appropriate mod-of-hashtable-size.
+unsigned StringMapImpl::RehashTable(unsigned BucketNo) {
+ unsigned NewSize;
+ unsigned *HashTable = (unsigned *)(TheTable + NumBuckets + 1);
+
+ // If the hash table is now more than 3/4 full, or if fewer than 1/8 of
+ // the buckets are empty (meaning that many are filled with tombstones),
+ // grow/rehash the table.
+ if (LLVM_UNLIKELY(NumItems * 4 > NumBuckets * 3)) {
+ NewSize = NumBuckets*2;
+ } else if (LLVM_UNLIKELY(NumBuckets - (NumItems + NumTombstones) <=
+ NumBuckets / 8)) {
+ NewSize = NumBuckets;
+ } else {
+ return BucketNo;
+ }
+
+ unsigned NewBucketNo = BucketNo;
+ // Allocate one extra bucket which will always be non-empty. This allows the
+ // iterators to stop at end.
+ auto NewTableArray = static_cast<StringMapEntryBase **>(
+ safe_calloc(NewSize+1, sizeof(StringMapEntryBase *) + sizeof(unsigned)));
+
+ unsigned *NewHashArray = (unsigned *)(NewTableArray + NewSize + 1);
+ NewTableArray[NewSize] = (StringMapEntryBase*)2;
+
+ // Rehash all the items into their new buckets. Luckily :) we already have
+ // the hash values available, so we don't have to rehash any strings.
+ for (unsigned I = 0, E = NumBuckets; I != E; ++I) {
+ StringMapEntryBase *Bucket = TheTable[I];
+ if (Bucket && Bucket != getTombstoneVal()) {
+ // Fast case, bucket available.
+ unsigned FullHash = HashTable[I];
+ unsigned NewBucket = FullHash & (NewSize-1);
+ if (!NewTableArray[NewBucket]) {
+ NewTableArray[FullHash & (NewSize-1)] = Bucket;
+ NewHashArray[FullHash & (NewSize-1)] = FullHash;
+ if (I == BucketNo)
+ NewBucketNo = NewBucket;
+ continue;
+ }
+
+ // Otherwise probe for a spot.
+ unsigned ProbeSize = 1;
+ do {
+ NewBucket = (NewBucket + ProbeSize++) & (NewSize-1);
+ } while (NewTableArray[NewBucket]);
+
+ // Finally found a slot. Fill it in.
+ NewTableArray[NewBucket] = Bucket;
+ NewHashArray[NewBucket] = FullHash;
+ if (I == BucketNo)
+ NewBucketNo = NewBucket;
+ }
+ }
+
+ free(TheTable);
+
+ TheTable = NewTableArray;
+ NumBuckets = NewSize;
+ NumTombstones = 0;
+ return NewBucketNo;
+}
diff --git a/third_party/llvm-project/StringRef.cpp b/third_party/llvm-project/StringRef.cpp
new file mode 100644
index 000000000..08af7eaac
--- /dev/null
+++ b/third_party/llvm-project/StringRef.cpp
@@ -0,0 +1,519 @@
+//===-- StringRef.cpp - Lightweight String References ---------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/APFloat.h"
+#include "llvm/ADT/APInt.h"
+#include "llvm/ADT/Hashing.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/edit_distance.h"
+#include <bitset>
+
+using namespace llvm;
+
+// MSVC emits references to this into the translation units which reference it.
+#ifndef _MSC_VER
+const size_t StringRef::npos;
+#endif
+
+// strncasecmp() is not available on non-POSIX systems, so define an
+// alternative function here.
+static int ascii_strncasecmp(const char *LHS, const char *RHS, size_t Length) {
+ for (size_t I = 0; I < Length; ++I) {
+ unsigned char LHC = toLower(LHS[I]);
+ unsigned char RHC = toLower(RHS[I]);
+ if (LHC != RHC)
+ return LHC < RHC ? -1 : 1;
+ }
+ return 0;
+}
+
+/// compare_lower - Compare strings, ignoring case.
+int StringRef::compare_lower(StringRef RHS) const {
+ if (int Res = ascii_strncasecmp(Data, RHS.Data, std::min(Length, RHS.Length)))
+ return Res;
+ if (Length == RHS.Length)
+ return 0;
+ return Length < RHS.Length ? -1 : 1;
+}
+
+/// Check if this string starts with the given \p Prefix, ignoring case.
+bool StringRef::startswith_lower(StringRef Prefix) const {
+ return Length >= Prefix.Length &&
+ ascii_strncasecmp(Data, Prefix.Data, Prefix.Length) == 0;
+}
+
+/// Check if this string ends with the given \p Suffix, ignoring case.
+bool StringRef::endswith_lower(StringRef Suffix) const {
+ return Length >= Suffix.Length &&
+ ascii_strncasecmp(end() - Suffix.Length, Suffix.Data, Suffix.Length) == 0;
+}
+
+size_t StringRef::find_lower(char C, size_t From) const {
+ char L = toLower(C);
+ return find_if([L](char D) { return toLower(D) == L; }, From);
+}
+
+/// compare_numeric - Compare strings, handle embedded numbers.
+int StringRef::compare_numeric(StringRef RHS) const {
+ for (size_t I = 0, E = std::min(Length, RHS.Length); I != E; ++I) {
+ // Check for sequences of digits.
+ if (isDigit(Data[I]) && isDigit(RHS.Data[I])) {
+ // The longer sequence of numbers is considered larger.
+ // This doesn't really handle prefixed zeros well.
+ size_t J;
+ for (J = I + 1; J != E + 1; ++J) {
+ bool ld = J < Length && isDigit(Data[J]);
+ bool rd = J < RHS.Length && isDigit(RHS.Data[J]);
+ if (ld != rd)
+ return rd ? -1 : 1;
+ if (!rd)
+ break;
+ }
+ // The two number sequences have the same length (J-I), just memcmp them.
+ if (int Res = compareMemory(Data + I, RHS.Data + I, J - I))
+ return Res < 0 ? -1 : 1;
+ // Identical number sequences, continue search after the numbers.
+ I = J - 1;
+ continue;
+ }
+ if (Data[I] != RHS.Data[I])
+ return (unsigned char)Data[I] < (unsigned char)RHS.Data[I] ? -1 : 1;
+ }
+ if (Length == RHS.Length)
+ return 0;
+ return Length < RHS.Length ? -1 : 1;
+}
+
+// Compute the edit distance between the two given strings.
+unsigned StringRef::edit_distance(llvm::StringRef Other,
+ bool AllowReplacements,
+ unsigned MaxEditDistance) const {
+ return llvm::ComputeEditDistance(
+ makeArrayRef(data(), size()),
+ makeArrayRef(Other.data(), Other.size()),
+ AllowReplacements, MaxEditDistance);
+}
+
+//===----------------------------------------------------------------------===//
+// String Operations
+//===----------------------------------------------------------------------===//
+
+std::string StringRef::lower() const {
+ std::string Result(size(), char());
+ for (size_type i = 0, e = size(); i != e; ++i) {
+ Result[i] = toLower(Data[i]);
+ }
+ return Result;
+}
+
+std::string StringRef::upper() const {
+ std::string Result(size(), char());
+ for (size_type i = 0, e = size(); i != e; ++i) {
+ Result[i] = toUpper(Data[i]);
+ }
+ return Result;
+}
+
+//===----------------------------------------------------------------------===//
+// String Searching
+//===----------------------------------------------------------------------===//
+
+
+/// find - Search for the first string \arg Str in the string.
+///
+/// \return - The index of the first occurrence of \arg Str, or npos if not
+/// found.
+size_t StringRef::find(StringRef Str, size_t From) const {
+ if (From > Length)
+ return npos;
+
+ const char *Start = Data + From;
+ size_t Size = Length - From;
+
+ const char *Needle = Str.data();
+ size_t N = Str.size();
+ if (N == 0)
+ return From;
+ if (Size < N)
+ return npos;
+ if (N == 1) {
+ const char *Ptr = (const char *)::memchr(Start, Needle[0], Size);
+ return Ptr == nullptr ? npos : Ptr - Data;
+ }
+
+ const char *Stop = Start + (Size - N + 1);
+
+ // For short haystacks or unsupported needles fall back to the naive algorithm
+ if (Size < 16 || N > 255) {
+ do {
+ if (std::memcmp(Start, Needle, N) == 0)
+ return Start - Data;
+ ++Start;
+ } while (Start < Stop);
+ return npos;
+ }
+
+ // Build the bad char heuristic table, with uint8_t to reduce cache thrashing.
+ uint8_t BadCharSkip[256];
+ std::memset(BadCharSkip, N, 256);
+ for (unsigned i = 0; i != N-1; ++i)
+ BadCharSkip[(uint8_t)Str[i]] = N-1-i;
+
+ do {
+ uint8_t Last = Start[N - 1];
+ if (LLVM_UNLIKELY(Last == (uint8_t)Needle[N - 1]))
+ if (std::memcmp(Start, Needle, N - 1) == 0)
+ return Start - Data;
+
+ // Otherwise skip the appropriate number of bytes.
+ Start += BadCharSkip[Last];
+ } while (Start < Stop);
+
+ return npos;
+}
+
+size_t StringRef::find_lower(StringRef Str, size_t From) const {
+ StringRef This = substr(From);
+ while (This.size() >= Str.size()) {
+ if (This.startswith_lower(Str))
+ return From;
+ This = This.drop_front();
+ ++From;
+ }
+ return npos;
+}
+
+size_t StringRef::rfind_lower(char C, size_t From) const {
+ From = std::min(From, Length);
+ size_t i = From;
+ while (i != 0) {
+ --i;
+ if (toLower(Data[i]) == toLower(C))
+ return i;
+ }
+ return npos;
+}
+
+/// rfind - Search for the last string \arg Str in the string.
+///
+/// \return - The index of the last occurrence of \arg Str, or npos if not
+/// found.
+size_t StringRef::rfind(StringRef Str) const {
+ size_t N = Str.size();
+ if (N > Length)
+ return npos;
+ for (size_t i = Length - N + 1, e = 0; i != e;) {
+ --i;
+ if (substr(i, N).equals(Str))
+ return i;
+ }
+ return npos;
+}
+
+size_t StringRef::rfind_lower(StringRef Str) const {
+ size_t N = Str.size();
+ if (N > Length)
+ return npos;
+ for (size_t i = Length - N + 1, e = 0; i != e;) {
+ --i;
+ if (substr(i, N).equals_lower(Str))
+ return i;
+ }
+ return npos;
+}
+
+/// find_first_of - Find the first character in the string that is in \arg
+/// Chars, or npos if not found.
+///
+/// Note: O(size() + Chars.size())
+StringRef::size_type StringRef::find_first_of(StringRef Chars,
+ size_t From) const {
+ std::bitset<1 << CHAR_BIT> CharBits;
+ for (size_type i = 0; i != Chars.size(); ++i)
+ CharBits.set((unsigned char)Chars[i]);
+
+ for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
+ if (CharBits.test((unsigned char)Data[i]))
+ return i;
+ return npos;
+}
+
+/// find_first_not_of - Find the first character in the string that is not
+/// \arg C or npos if not found.
+StringRef::size_type StringRef::find_first_not_of(char C, size_t From) const {
+ for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
+ if (Data[i] != C)
+ return i;
+ return npos;
+}
+
+/// find_first_not_of - Find the first character in the string that is not
+/// in the string \arg Chars, or npos if not found.
+///
+/// Note: O(size() + Chars.size())
+StringRef::size_type StringRef::find_first_not_of(StringRef Chars,
+ size_t From) const {
+ std::bitset<1 << CHAR_BIT> CharBits;
+ for (size_type i = 0; i != Chars.size(); ++i)
+ CharBits.set((unsigned char)Chars[i]);
+
+ for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
+ if (!CharBits.test((unsigned char)Data[i]))
+ return i;
+ return npos;
+}
+
+/// find_last_of - Find the last character in the string that is in \arg C,
+/// or npos if not found.
+///
+/// Note: O(size() + Chars.size())
+StringRef::size_type StringRef::find_last_of(StringRef Chars,
+ size_t From) const {
+ std::bitset<1 << CHAR_BIT> CharBits;
+ for (size_type i = 0; i != Chars.size(); ++i)
+ CharBits.set((unsigned char)Chars[i]);
+
+ for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
+ if (CharBits.test((unsigned char)Data[i]))
+ return i;
+ return npos;
+}
+
+/// find_last_not_of - Find the last character in the string that is not
+/// \arg C, or npos if not found.
+StringRef::size_type StringRef::find_last_not_of(char C, size_t From) const {
+ for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
+ if (Data[i] != C)
+ return i;
+ return npos;
+}
+
+/// find_last_not_of - Find the last character in the string that is not in
+/// \arg Chars, or npos if not found.
+///
+/// Note: O(size() + Chars.size())
+StringRef::size_type StringRef::find_last_not_of(StringRef Chars,
+ size_t From) const {
+ std::bitset<1 << CHAR_BIT> CharBits;
+ for (size_type i = 0, e = Chars.size(); i != e; ++i)
+ CharBits.set((unsigned char)Chars[i]);
+
+ for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
+ if (!CharBits.test((unsigned char)Data[i]))
+ return i;
+ return npos;
+}
+
+void StringRef::split(SmallVectorImpl<StringRef> &A,
+ StringRef Separator, int MaxSplit,
+ bool KeepEmpty) const {
+ StringRef S = *this;
+
+ // Count down from MaxSplit. When MaxSplit is -1, this will just split
+ // "forever". This doesn't support splitting more than 2^31 times
+ // intentionally; if we ever want that we can make MaxSplit a 64-bit integer
+ // but that seems unlikely to be useful.
+ while (MaxSplit-- != 0) {
+ size_t Idx = S.find(Separator);
+ if (Idx == npos)
+ break;
+
+ // Push this split.
+ if (KeepEmpty || Idx > 0)
+ A.push_back(S.slice(0, Idx));
+
+ // Jump forward.
+ S = S.slice(Idx + Separator.size(), npos);
+ }
+
+ // Push the tail.
+ if (KeepEmpty || !S.empty())
+ A.push_back(S);
+}
+
+void StringRef::split(SmallVectorImpl<StringRef> &A, char Separator,
+ int MaxSplit, bool KeepEmpty) const {
+ StringRef S = *this;
+
+ // Count down from MaxSplit. When MaxSplit is -1, this will just split
+ // "forever". This doesn't support splitting more than 2^31 times
+ // intentionally; if we ever want that we can make MaxSplit a 64-bit integer
+ // but that seems unlikely to be useful.
+ while (MaxSplit-- != 0) {
+ size_t Idx = S.find(Separator);
+ if (Idx == npos)
+ break;
+
+ // Push this split.
+ if (KeepEmpty || Idx > 0)
+ A.push_back(S.slice(0, Idx));
+
+ // Jump forward.
+ S = S.slice(Idx + 1, npos);
+ }
+
+ // Push the tail.
+ if (KeepEmpty || !S.empty())
+ A.push_back(S);
+}
+
+//===----------------------------------------------------------------------===//
+// Helpful Algorithms
+//===----------------------------------------------------------------------===//
+
+/// count - Return the number of non-overlapped occurrences of \arg Str in
+/// the string.
+size_t StringRef::count(StringRef Str) const {
+ size_t Count = 0;
+ size_t N = Str.size();
+ if (N > Length)
+ return 0;
+ for (size_t i = 0, e = Length - N + 1; i != e; ++i)
+ if (substr(i, N).equals(Str))
+ ++Count;
+ return Count;
+}
+
+static unsigned GetAutoSenseRadix(StringRef &Str) {
+ if (Str.empty())
+ return 10;
+
+ if (Str.startswith("0x") || Str.startswith("0X")) {
+ Str = Str.substr(2);
+ return 16;
+ }
+
+ if (Str.startswith("0b") || Str.startswith("0B")) {
+ Str = Str.substr(2);
+ return 2;
+ }
+
+ if (Str.startswith("0o")) {
+ Str = Str.substr(2);
+ return 8;
+ }
+
+ if (Str[0] == '0' && Str.size() > 1 && isDigit(Str[1])) {
+ Str = Str.substr(1);
+ return 8;
+ }
+
+ return 10;
+}
+
+bool llvm::consumeUnsignedInteger(StringRef &Str, unsigned Radix,
+ unsigned long long &Result) {
+ // Autosense radix if not specified.
+ if (Radix == 0)
+ Radix = GetAutoSenseRadix(Str);
+
+ // Empty strings (after the radix autosense) are invalid.
+ if (Str.empty()) return true;
+
+ // Parse all the bytes of the string given this radix. Watch for overflow.
+ StringRef Str2 = Str;
+ Result = 0;
+ while (!Str2.empty()) {
+ unsigned CharVal;
+ if (Str2[0] >= '0' && Str2[0] <= '9')
+ CharVal = Str2[0] - '0';
+ else if (Str2[0] >= 'a' && Str2[0] <= 'z')
+ CharVal = Str2[0] - 'a' + 10;
+ else if (Str2[0] >= 'A' && Str2[0] <= 'Z')
+ CharVal = Str2[0] - 'A' + 10;
+ else
+ break;
+
+ // If the parsed value is larger than the integer radix, we cannot
+ // consume any more characters.
+ if (CharVal >= Radix)
+ break;
+
+ // Add in this character.
+ unsigned long long PrevResult = Result;
+ Result = Result * Radix + CharVal;
+
+ // Check for overflow by shifting back and seeing if bits were lost.
+ if (Result / Radix < PrevResult)
+ return true;
+
+ Str2 = Str2.substr(1);
+ }
+
+ // We consider the operation a failure if no characters were consumed
+ // successfully.
+ if (Str.size() == Str2.size())
+ return true;
+
+ Str = Str2;
+ return false;
+}
+
+bool llvm::consumeSignedInteger(StringRef &Str, unsigned Radix,
+ long long &Result) {
+ unsigned long long ULLVal;
+
+ // Handle positive strings first.
+ if (Str.empty() || Str.front() != '-') {
+ if (consumeUnsignedInteger(Str, Radix, ULLVal) ||
+ // Check for value so large it overflows a signed value.
+ (long long)ULLVal < 0)
+ return true;
+ Result = ULLVal;
+ return false;
+ }
+
+ // Get the positive part of the value.
+ StringRef Str2 = Str.drop_front(1);
+ if (consumeUnsignedInteger(Str2, Radix, ULLVal) ||
+ // Reject values so large they'd overflow as negative signed, but allow
+ // "-0". This negates the unsigned so that the negative isn't undefined
+ // on signed overflow.
+ (long long)-ULLVal > 0)
+ return true;
+
+ Str = Str2;
+ Result = -ULLVal;
+ return false;
+}
+
+/// GetAsUnsignedInteger - Workhorse method that converts a integer character
+/// sequence of radix up to 36 to an unsigned long long value.
+bool llvm::getAsUnsignedInteger(StringRef Str, unsigned Radix,
+ unsigned long long &Result) {
+ if (consumeUnsignedInteger(Str, Radix, Result))
+ return true;
+
+ // For getAsUnsignedInteger, we require the whole string to be consumed or
+ // else we consider it a failure.
+ return !Str.empty();
+}
+
+bool llvm::getAsSignedInteger(StringRef Str, unsigned Radix,
+ long long &Result) {
+ if (consumeSignedInteger(Str, Radix, Result))
+ return true;
+
+ // For getAsSignedInteger, we require the whole string to be consumed or else
+ // we consider it a failure.
+ return !Str.empty();
+}
+
+bool StringRef::getAsInteger(unsigned Radix, APInt &Result) const {
+ llvm_unreachable("getAsinteger");
+}
+
+bool StringRef::getAsDouble(double &Result, bool AllowInexact) const {
+ llvm_unreachable("getAsDouble");
+}
+
+// Implementation of StringRef hashing.
+hash_code llvm::hash_value(StringRef S) {
+ return hash_combine_range(S.begin(), S.end());
+}
diff --git a/third_party/llvm-project/SymbolicFile.cpp b/third_party/llvm-project/SymbolicFile.cpp
new file mode 100644
index 000000000..a3d080c73
--- /dev/null
+++ b/third_party/llvm-project/SymbolicFile.cpp
@@ -0,0 +1,45 @@
+//===- SymbolicFile.cpp - Interface that only provides symbols ------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines a file format independent SymbolicFile class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Object/SymbolicFile.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Magic.h"
+#if 0 // XXX BINARYEN
+#include "llvm/Object/COFFImportFile.h"
+#endif
+#include "llvm/Object/Error.h"
+#if 0 // XXX BINARYEN
+#include "llvm/Object/IRObjectFile.h"
+#endif
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <algorithm>
+#include <memory>
+
+using namespace llvm;
+using namespace object;
+
+SymbolicFile::SymbolicFile(unsigned int Type, MemoryBufferRef Source)
+ : Binary(Type, Source) {}
+
+SymbolicFile::~SymbolicFile() = default;
+
+Expected<std::unique_ptr<SymbolicFile>>
+SymbolicFile::createSymbolicFile(MemoryBufferRef Object, file_magic Type,
+ LLVMContext *Context) {
+ llvm_unreachable("createSymbolicFile");
+}
diff --git a/third_party/llvm-project/Twine.cpp b/third_party/llvm-project/Twine.cpp
new file mode 100644
index 000000000..fbbcd8848
--- /dev/null
+++ b/third_party/llvm-project/Twine.cpp
@@ -0,0 +1,184 @@
+//===-- Twine.cpp - Fast Temporary String Concatenation -------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/Twine.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/Config/llvm-config.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/raw_ostream.h"
+using namespace llvm;
+
+std::string Twine::str() const {
+ // If we're storing only a std::string, just return it.
+ if (LHSKind == StdStringKind && RHSKind == EmptyKind)
+ return *LHS.stdString;
+
+ // If we're storing a formatv_object, we can avoid an extra copy by formatting
+ // it immediately and returning the result.
+ if (LHSKind == FormatvObjectKind && RHSKind == EmptyKind)
+ return LHS.formatvObject->str();
+
+ // Otherwise, flatten and copy the contents first.
+ SmallString<256> Vec;
+ return toStringRef(Vec).str();
+}
+
+void Twine::toVector(SmallVectorImpl<char> &Out) const {
+ raw_svector_ostream OS(Out);
+ print(OS);
+}
+
+StringRef Twine::toNullTerminatedStringRef(SmallVectorImpl<char> &Out) const {
+ if (isUnary()) {
+ switch (getLHSKind()) {
+ case CStringKind:
+ // Already null terminated, yay!
+ return StringRef(LHS.cString);
+ case StdStringKind: {
+ const std::string *str = LHS.stdString;
+ return StringRef(str->c_str(), str->size());
+ }
+ default:
+ break;
+ }
+ }
+ toVector(Out);
+ Out.push_back(0);
+ Out.pop_back();
+ return StringRef(Out.data(), Out.size());
+}
+
+void Twine::printOneChild(raw_ostream &OS, Child Ptr,
+ NodeKind Kind) const {
+ switch (Kind) {
+ case Twine::NullKind: break;
+ case Twine::EmptyKind: break;
+ case Twine::TwineKind:
+ Ptr.twine->print(OS);
+ break;
+ case Twine::CStringKind:
+ OS << Ptr.cString;
+ break;
+ case Twine::StdStringKind:
+ OS << *Ptr.stdString;
+ break;
+ case Twine::StringRefKind:
+ OS << *Ptr.stringRef;
+ break;
+ case Twine::SmallStringKind:
+ OS << *Ptr.smallString;
+ break;
+ case Twine::FormatvObjectKind:
+ OS << *Ptr.formatvObject;
+ break;
+ case Twine::CharKind:
+ OS << Ptr.character;
+ break;
+ case Twine::DecUIKind:
+ OS << Ptr.decUI;
+ break;
+ case Twine::DecIKind:
+ OS << Ptr.decI;
+ break;
+ case Twine::DecULKind:
+ OS << *Ptr.decUL;
+ break;
+ case Twine::DecLKind:
+ OS << *Ptr.decL;
+ break;
+ case Twine::DecULLKind:
+ OS << *Ptr.decULL;
+ break;
+ case Twine::DecLLKind:
+ OS << *Ptr.decLL;
+ break;
+ case Twine::UHexKind:
+ OS.write_hex(*Ptr.uHex);
+ break;
+ }
+}
+
+void Twine::printOneChildRepr(raw_ostream &OS, Child Ptr,
+ NodeKind Kind) const {
+ switch (Kind) {
+ case Twine::NullKind:
+ OS << "null"; break;
+ case Twine::EmptyKind:
+ OS << "empty"; break;
+ case Twine::TwineKind:
+ OS << "rope:";
+ Ptr.twine->printRepr(OS);
+ break;
+ case Twine::CStringKind:
+ OS << "cstring:\""
+ << Ptr.cString << "\"";
+ break;
+ case Twine::StdStringKind:
+ OS << "std::string:\""
+ << Ptr.stdString << "\"";
+ break;
+ case Twine::StringRefKind:
+ OS << "stringref:\""
+ << Ptr.stringRef << "\"";
+ break;
+ case Twine::SmallStringKind:
+ OS << "smallstring:\"" << *Ptr.smallString << "\"";
+ break;
+ case Twine::FormatvObjectKind:
+ OS << "formatv:\"" << *Ptr.formatvObject << "\"";
+ break;
+ case Twine::CharKind:
+ OS << "char:\"" << Ptr.character << "\"";
+ break;
+ case Twine::DecUIKind:
+ OS << "decUI:\"" << Ptr.decUI << "\"";
+ break;
+ case Twine::DecIKind:
+ OS << "decI:\"" << Ptr.decI << "\"";
+ break;
+ case Twine::DecULKind:
+ OS << "decUL:\"" << *Ptr.decUL << "\"";
+ break;
+ case Twine::DecLKind:
+ OS << "decL:\"" << *Ptr.decL << "\"";
+ break;
+ case Twine::DecULLKind:
+ OS << "decULL:\"" << *Ptr.decULL << "\"";
+ break;
+ case Twine::DecLLKind:
+ OS << "decLL:\"" << *Ptr.decLL << "\"";
+ break;
+ case Twine::UHexKind:
+ OS << "uhex:\"" << Ptr.uHex << "\"";
+ break;
+ }
+}
+
+void Twine::print(raw_ostream &OS) const {
+ printOneChild(OS, LHS, getLHSKind());
+ printOneChild(OS, RHS, getRHSKind());
+}
+
+void Twine::printRepr(raw_ostream &OS) const {
+ OS << "(Twine ";
+ printOneChildRepr(OS, LHS, getLHSKind());
+ OS << " ";
+ printOneChildRepr(OS, RHS, getRHSKind());
+ OS << ")";
+}
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+LLVM_DUMP_METHOD void Twine::dump() const {
+ print(dbgs());
+}
+
+LLVM_DUMP_METHOD void Twine::dumpRepr() const {
+ printRepr(dbgs());
+}
+#endif
diff --git a/third_party/llvm-project/UnicodeCaseFold.cpp b/third_party/llvm-project/UnicodeCaseFold.cpp
new file mode 100644
index 000000000..9070a7c28
--- /dev/null
+++ b/third_party/llvm-project/UnicodeCaseFold.cpp
@@ -0,0 +1,742 @@
+//===---------- Support/UnicodeCaseFold.cpp -------------------------------===//
+//
+// This file was generated by utils/unicode-case-fold.py from the Unicode
+// case folding database at
+// http://www.unicode.org/Public/9.0.0/ucd/CaseFolding.txt
+//
+// To regenerate this file, run:
+// utils/unicode-case-fold.py
+// "http://www.unicode.org/Public/9.0.0/ucd/CaseFolding.txt"
+// > lib/Support/UnicodeCaseFold.cpp
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/Unicode.h"
+
+int llvm::sys::unicode::foldCharSimple(int C) {
+ if (C < 0x0041)
+ return C;
+ // 26 characters
+ if (C <= 0x005a)
+ return C + 32;
+ // MICRO SIGN
+ if (C == 0x00b5)
+ return 0x03bc;
+ if (C < 0x00c0)
+ return C;
+ // 23 characters
+ if (C <= 0x00d6)
+ return C + 32;
+ if (C < 0x00d8)
+ return C;
+ // 7 characters
+ if (C <= 0x00de)
+ return C + 32;
+ if (C < 0x0100)
+ return C;
+ // 24 characters
+ if (C <= 0x012e)
+ return C | 1;
+ if (C < 0x0132)
+ return C;
+ // 3 characters
+ if (C <= 0x0136)
+ return C | 1;
+ if (C < 0x0139)
+ return C;
+ // 8 characters
+ if (C <= 0x0147 && C % 2 == 1)
+ return C + 1;
+ if (C < 0x014a)
+ return C;
+ // 23 characters
+ if (C <= 0x0176)
+ return C | 1;
+ // LATIN CAPITAL LETTER Y WITH DIAERESIS
+ if (C == 0x0178)
+ return 0x00ff;
+ if (C < 0x0179)
+ return C;
+ // 3 characters
+ if (C <= 0x017d && C % 2 == 1)
+ return C + 1;
+ // LATIN SMALL LETTER LONG S
+ if (C == 0x017f)
+ return 0x0073;
+ // LATIN CAPITAL LETTER B WITH HOOK
+ if (C == 0x0181)
+ return 0x0253;
+ if (C < 0x0182)
+ return C;
+ // 2 characters
+ if (C <= 0x0184)
+ return C | 1;
+ // LATIN CAPITAL LETTER OPEN O
+ if (C == 0x0186)
+ return 0x0254;
+ // LATIN CAPITAL LETTER C WITH HOOK
+ if (C == 0x0187)
+ return 0x0188;
+ if (C < 0x0189)
+ return C;
+ // 2 characters
+ if (C <= 0x018a)
+ return C + 205;
+ // LATIN CAPITAL LETTER D WITH TOPBAR
+ if (C == 0x018b)
+ return 0x018c;
+ // LATIN CAPITAL LETTER REVERSED E
+ if (C == 0x018e)
+ return 0x01dd;
+ // LATIN CAPITAL LETTER SCHWA
+ if (C == 0x018f)
+ return 0x0259;
+ // LATIN CAPITAL LETTER OPEN E
+ if (C == 0x0190)
+ return 0x025b;
+ // LATIN CAPITAL LETTER F WITH HOOK
+ if (C == 0x0191)
+ return 0x0192;
+ // LATIN CAPITAL LETTER G WITH HOOK
+ if (C == 0x0193)
+ return 0x0260;
+ // LATIN CAPITAL LETTER GAMMA
+ if (C == 0x0194)
+ return 0x0263;
+ // LATIN CAPITAL LETTER IOTA
+ if (C == 0x0196)
+ return 0x0269;
+ // LATIN CAPITAL LETTER I WITH STROKE
+ if (C == 0x0197)
+ return 0x0268;
+ // LATIN CAPITAL LETTER K WITH HOOK
+ if (C == 0x0198)
+ return 0x0199;
+ // LATIN CAPITAL LETTER TURNED M
+ if (C == 0x019c)
+ return 0x026f;
+ // LATIN CAPITAL LETTER N WITH LEFT HOOK
+ if (C == 0x019d)
+ return 0x0272;
+ // LATIN CAPITAL LETTER O WITH MIDDLE TILDE
+ if (C == 0x019f)
+ return 0x0275;
+ if (C < 0x01a0)
+ return C;
+ // 3 characters
+ if (C <= 0x01a4)
+ return C | 1;
+ // LATIN LETTER YR
+ if (C == 0x01a6)
+ return 0x0280;
+ // LATIN CAPITAL LETTER TONE TWO
+ if (C == 0x01a7)
+ return 0x01a8;
+ // LATIN CAPITAL LETTER ESH
+ if (C == 0x01a9)
+ return 0x0283;
+ // LATIN CAPITAL LETTER T WITH HOOK
+ if (C == 0x01ac)
+ return 0x01ad;
+ // LATIN CAPITAL LETTER T WITH RETROFLEX HOOK
+ if (C == 0x01ae)
+ return 0x0288;
+ // LATIN CAPITAL LETTER U WITH HORN
+ if (C == 0x01af)
+ return 0x01b0;
+ if (C < 0x01b1)
+ return C;
+ // 2 characters
+ if (C <= 0x01b2)
+ return C + 217;
+ if (C < 0x01b3)
+ return C;
+ // 2 characters
+ if (C <= 0x01b5 && C % 2 == 1)
+ return C + 1;
+ // LATIN CAPITAL LETTER EZH
+ if (C == 0x01b7)
+ return 0x0292;
+ if (C < 0x01b8)
+ return C;
+ // 2 characters
+ if (C <= 0x01bc && C % 4 == 0)
+ return C + 1;
+ // LATIN CAPITAL LETTER DZ WITH CARON
+ if (C == 0x01c4)
+ return 0x01c6;
+ // LATIN CAPITAL LETTER D WITH SMALL LETTER Z WITH CARON
+ if (C == 0x01c5)
+ return 0x01c6;
+ // LATIN CAPITAL LETTER LJ
+ if (C == 0x01c7)
+ return 0x01c9;
+ // LATIN CAPITAL LETTER L WITH SMALL LETTER J
+ if (C == 0x01c8)
+ return 0x01c9;
+ // LATIN CAPITAL LETTER NJ
+ if (C == 0x01ca)
+ return 0x01cc;
+ if (C < 0x01cb)
+ return C;
+ // 9 characters
+ if (C <= 0x01db && C % 2 == 1)
+ return C + 1;
+ if (C < 0x01de)
+ return C;
+ // 9 characters
+ if (C <= 0x01ee)
+ return C | 1;
+ // LATIN CAPITAL LETTER DZ
+ if (C == 0x01f1)
+ return 0x01f3;
+ if (C < 0x01f2)
+ return C;
+ // 2 characters
+ if (C <= 0x01f4)
+ return C | 1;
+ // LATIN CAPITAL LETTER HWAIR
+ if (C == 0x01f6)
+ return 0x0195;
+ // LATIN CAPITAL LETTER WYNN
+ if (C == 0x01f7)
+ return 0x01bf;
+ if (C < 0x01f8)
+ return C;
+ // 20 characters
+ if (C <= 0x021e)
+ return C | 1;
+ // LATIN CAPITAL LETTER N WITH LONG RIGHT LEG
+ if (C == 0x0220)
+ return 0x019e;
+ if (C < 0x0222)
+ return C;
+ // 9 characters
+ if (C <= 0x0232)
+ return C | 1;
+ // LATIN CAPITAL LETTER A WITH STROKE
+ if (C == 0x023a)
+ return 0x2c65;
+ // LATIN CAPITAL LETTER C WITH STROKE
+ if (C == 0x023b)
+ return 0x023c;
+ // LATIN CAPITAL LETTER L WITH BAR
+ if (C == 0x023d)
+ return 0x019a;
+ // LATIN CAPITAL LETTER T WITH DIAGONAL STROKE
+ if (C == 0x023e)
+ return 0x2c66;
+ // LATIN CAPITAL LETTER GLOTTAL STOP
+ if (C == 0x0241)
+ return 0x0242;
+ // LATIN CAPITAL LETTER B WITH STROKE
+ if (C == 0x0243)
+ return 0x0180;
+ // LATIN CAPITAL LETTER U BAR
+ if (C == 0x0244)
+ return 0x0289;
+ // LATIN CAPITAL LETTER TURNED V
+ if (C == 0x0245)
+ return 0x028c;
+ if (C < 0x0246)
+ return C;
+ // 5 characters
+ if (C <= 0x024e)
+ return C | 1;
+ // COMBINING GREEK YPOGEGRAMMENI
+ if (C == 0x0345)
+ return 0x03b9;
+ if (C < 0x0370)
+ return C;
+ // 2 characters
+ if (C <= 0x0372)
+ return C | 1;
+ // GREEK CAPITAL LETTER PAMPHYLIAN DIGAMMA
+ if (C == 0x0376)
+ return 0x0377;
+ // GREEK CAPITAL LETTER YOT
+ if (C == 0x037f)
+ return 0x03f3;
+ // GREEK CAPITAL LETTER ALPHA WITH TONOS
+ if (C == 0x0386)
+ return 0x03ac;
+ if (C < 0x0388)
+ return C;
+ // 3 characters
+ if (C <= 0x038a)
+ return C + 37;
+ // GREEK CAPITAL LETTER OMICRON WITH TONOS
+ if (C == 0x038c)
+ return 0x03cc;
+ if (C < 0x038e)
+ return C;
+ // 2 characters
+ if (C <= 0x038f)
+ return C + 63;
+ if (C < 0x0391)
+ return C;
+ // 17 characters
+ if (C <= 0x03a1)
+ return C + 32;
+ if (C < 0x03a3)
+ return C;
+ // 9 characters
+ if (C <= 0x03ab)
+ return C + 32;
+ // GREEK SMALL LETTER FINAL SIGMA
+ if (C == 0x03c2)
+ return 0x03c3;
+ // GREEK CAPITAL KAI SYMBOL
+ if (C == 0x03cf)
+ return 0x03d7;
+ // GREEK BETA SYMBOL
+ if (C == 0x03d0)
+ return 0x03b2;
+ // GREEK THETA SYMBOL
+ if (C == 0x03d1)
+ return 0x03b8;
+ // GREEK PHI SYMBOL
+ if (C == 0x03d5)
+ return 0x03c6;
+ // GREEK PI SYMBOL
+ if (C == 0x03d6)
+ return 0x03c0;
+ if (C < 0x03d8)
+ return C;
+ // 12 characters
+ if (C <= 0x03ee)
+ return C | 1;
+ // GREEK KAPPA SYMBOL
+ if (C == 0x03f0)
+ return 0x03ba;
+ // GREEK RHO SYMBOL
+ if (C == 0x03f1)
+ return 0x03c1;
+ // GREEK CAPITAL THETA SYMBOL
+ if (C == 0x03f4)
+ return 0x03b8;
+ // GREEK LUNATE EPSILON SYMBOL
+ if (C == 0x03f5)
+ return 0x03b5;
+ // GREEK CAPITAL LETTER SHO
+ if (C == 0x03f7)
+ return 0x03f8;
+ // GREEK CAPITAL LUNATE SIGMA SYMBOL
+ if (C == 0x03f9)
+ return 0x03f2;
+ // GREEK CAPITAL LETTER SAN
+ if (C == 0x03fa)
+ return 0x03fb;
+ if (C < 0x03fd)
+ return C;
+ // 3 characters
+ if (C <= 0x03ff)
+ return C + -130;
+ if (C < 0x0400)
+ return C;
+ // 16 characters
+ if (C <= 0x040f)
+ return C + 80;
+ if (C < 0x0410)
+ return C;
+ // 32 characters
+ if (C <= 0x042f)
+ return C + 32;
+ if (C < 0x0460)
+ return C;
+ // 17 characters
+ if (C <= 0x0480)
+ return C | 1;
+ if (C < 0x048a)
+ return C;
+ // 27 characters
+ if (C <= 0x04be)
+ return C | 1;
+ // CYRILLIC LETTER PALOCHKA
+ if (C == 0x04c0)
+ return 0x04cf;
+ if (C < 0x04c1)
+ return C;
+ // 7 characters
+ if (C <= 0x04cd && C % 2 == 1)
+ return C + 1;
+ if (C < 0x04d0)
+ return C;
+ // 48 characters
+ if (C <= 0x052e)
+ return C | 1;
+ if (C < 0x0531)
+ return C;
+ // 38 characters
+ if (C <= 0x0556)
+ return C + 48;
+ if (C < 0x10a0)
+ return C;
+ // 38 characters
+ if (C <= 0x10c5)
+ return C + 7264;
+ if (C < 0x10c7)
+ return C;
+ // 2 characters
+ if (C <= 0x10cd && C % 6 == 5)
+ return C + 7264;
+ if (C < 0x13f8)
+ return C;
+ // 6 characters
+ if (C <= 0x13fd)
+ return C + -8;
+ // CYRILLIC SMALL LETTER ROUNDED VE
+ if (C == 0x1c80)
+ return 0x0432;
+ // CYRILLIC SMALL LETTER LONG-LEGGED DE
+ if (C == 0x1c81)
+ return 0x0434;
+ // CYRILLIC SMALL LETTER NARROW O
+ if (C == 0x1c82)
+ return 0x043e;
+ if (C < 0x1c83)
+ return C;
+ // 2 characters
+ if (C <= 0x1c84)
+ return C + -6210;
+ // CYRILLIC SMALL LETTER THREE-LEGGED TE
+ if (C == 0x1c85)
+ return 0x0442;
+ // CYRILLIC SMALL LETTER TALL HARD SIGN
+ if (C == 0x1c86)
+ return 0x044a;
+ // CYRILLIC SMALL LETTER TALL YAT
+ if (C == 0x1c87)
+ return 0x0463;
+ // CYRILLIC SMALL LETTER UNBLENDED UK
+ if (C == 0x1c88)
+ return 0xa64b;
+ if (C < 0x1e00)
+ return C;
+ // 75 characters
+ if (C <= 0x1e94)
+ return C | 1;
+ // LATIN SMALL LETTER LONG S WITH DOT ABOVE
+ if (C == 0x1e9b)
+ return 0x1e61;
+ // LATIN CAPITAL LETTER SHARP S
+ if (C == 0x1e9e)
+ return 0x00df;
+ if (C < 0x1ea0)
+ return C;
+ // 48 characters
+ if (C <= 0x1efe)
+ return C | 1;
+ if (C < 0x1f08)
+ return C;
+ // 8 characters
+ if (C <= 0x1f0f)
+ return C + -8;
+ if (C < 0x1f18)
+ return C;
+ // 6 characters
+ if (C <= 0x1f1d)
+ return C + -8;
+ if (C < 0x1f28)
+ return C;
+ // 8 characters
+ if (C <= 0x1f2f)
+ return C + -8;
+ if (C < 0x1f38)
+ return C;
+ // 8 characters
+ if (C <= 0x1f3f)
+ return C + -8;
+ if (C < 0x1f48)
+ return C;
+ // 6 characters
+ if (C <= 0x1f4d)
+ return C + -8;
+ if (C < 0x1f59)
+ return C;
+ // 4 characters
+ if (C <= 0x1f5f && C % 2 == 1)
+ return C + -8;
+ if (C < 0x1f68)
+ return C;
+ // 8 characters
+ if (C <= 0x1f6f)
+ return C + -8;
+ if (C < 0x1f88)
+ return C;
+ // 8 characters
+ if (C <= 0x1f8f)
+ return C + -8;
+ if (C < 0x1f98)
+ return C;
+ // 8 characters
+ if (C <= 0x1f9f)
+ return C + -8;
+ if (C < 0x1fa8)
+ return C;
+ // 8 characters
+ if (C <= 0x1faf)
+ return C + -8;
+ if (C < 0x1fb8)
+ return C;
+ // 2 characters
+ if (C <= 0x1fb9)
+ return C + -8;
+ if (C < 0x1fba)
+ return C;
+ // 2 characters
+ if (C <= 0x1fbb)
+ return C + -74;
+ // GREEK CAPITAL LETTER ALPHA WITH PROSGEGRAMMENI
+ if (C == 0x1fbc)
+ return 0x1fb3;
+ // GREEK PROSGEGRAMMENI
+ if (C == 0x1fbe)
+ return 0x03b9;
+ if (C < 0x1fc8)
+ return C;
+ // 4 characters
+ if (C <= 0x1fcb)
+ return C + -86;
+ // GREEK CAPITAL LETTER ETA WITH PROSGEGRAMMENI
+ if (C == 0x1fcc)
+ return 0x1fc3;
+ if (C < 0x1fd8)
+ return C;
+ // 2 characters
+ if (C <= 0x1fd9)
+ return C + -8;
+ if (C < 0x1fda)
+ return C;
+ // 2 characters
+ if (C <= 0x1fdb)
+ return C + -100;
+ if (C < 0x1fe8)
+ return C;
+ // 2 characters
+ if (C <= 0x1fe9)
+ return C + -8;
+ if (C < 0x1fea)
+ return C;
+ // 2 characters
+ if (C <= 0x1feb)
+ return C + -112;
+ // GREEK CAPITAL LETTER RHO WITH DASIA
+ if (C == 0x1fec)
+ return 0x1fe5;
+ if (C < 0x1ff8)
+ return C;
+ // 2 characters
+ if (C <= 0x1ff9)
+ return C + -128;
+ if (C < 0x1ffa)
+ return C;
+ // 2 characters
+ if (C <= 0x1ffb)
+ return C + -126;
+ // GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI
+ if (C == 0x1ffc)
+ return 0x1ff3;
+ // OHM SIGN
+ if (C == 0x2126)
+ return 0x03c9;
+ // KELVIN SIGN
+ if (C == 0x212a)
+ return 0x006b;
+ // ANGSTROM SIGN
+ if (C == 0x212b)
+ return 0x00e5;
+ // TURNED CAPITAL F
+ if (C == 0x2132)
+ return 0x214e;
+ if (C < 0x2160)
+ return C;
+ // 16 characters
+ if (C <= 0x216f)
+ return C + 16;
+ // ROMAN NUMERAL REVERSED ONE HUNDRED
+ if (C == 0x2183)
+ return 0x2184;
+ if (C < 0x24b6)
+ return C;
+ // 26 characters
+ if (C <= 0x24cf)
+ return C + 26;
+ if (C < 0x2c00)
+ return C;
+ // 47 characters
+ if (C <= 0x2c2e)
+ return C + 48;
+ // LATIN CAPITAL LETTER L WITH DOUBLE BAR
+ if (C == 0x2c60)
+ return 0x2c61;
+ // LATIN CAPITAL LETTER L WITH MIDDLE TILDE
+ if (C == 0x2c62)
+ return 0x026b;
+ // LATIN CAPITAL LETTER P WITH STROKE
+ if (C == 0x2c63)
+ return 0x1d7d;
+ // LATIN CAPITAL LETTER R WITH TAIL
+ if (C == 0x2c64)
+ return 0x027d;
+ if (C < 0x2c67)
+ return C;
+ // 3 characters
+ if (C <= 0x2c6b && C % 2 == 1)
+ return C + 1;
+ // LATIN CAPITAL LETTER ALPHA
+ if (C == 0x2c6d)
+ return 0x0251;
+ // LATIN CAPITAL LETTER M WITH HOOK
+ if (C == 0x2c6e)
+ return 0x0271;
+ // LATIN CAPITAL LETTER TURNED A
+ if (C == 0x2c6f)
+ return 0x0250;
+ // LATIN CAPITAL LETTER TURNED ALPHA
+ if (C == 0x2c70)
+ return 0x0252;
+ if (C < 0x2c72)
+ return C;
+ // 2 characters
+ if (C <= 0x2c75 && C % 3 == 2)
+ return C + 1;
+ if (C < 0x2c7e)
+ return C;
+ // 2 characters
+ if (C <= 0x2c7f)
+ return C + -10815;
+ if (C < 0x2c80)
+ return C;
+ // 50 characters
+ if (C <= 0x2ce2)
+ return C | 1;
+ if (C < 0x2ceb)
+ return C;
+ // 2 characters
+ if (C <= 0x2ced && C % 2 == 1)
+ return C + 1;
+ if (C < 0x2cf2)
+ return C;
+ // 2 characters
+ if (C <= 0xa640 && C % 31054 == 11506)
+ return C + 1;
+ if (C < 0xa642)
+ return C;
+ // 22 characters
+ if (C <= 0xa66c)
+ return C | 1;
+ if (C < 0xa680)
+ return C;
+ // 14 characters
+ if (C <= 0xa69a)
+ return C | 1;
+ if (C < 0xa722)
+ return C;
+ // 7 characters
+ if (C <= 0xa72e)
+ return C | 1;
+ if (C < 0xa732)
+ return C;
+ // 31 characters
+ if (C <= 0xa76e)
+ return C | 1;
+ if (C < 0xa779)
+ return C;
+ // 2 characters
+ if (C <= 0xa77b && C % 2 == 1)
+ return C + 1;
+ // LATIN CAPITAL LETTER INSULAR G
+ if (C == 0xa77d)
+ return 0x1d79;
+ if (C < 0xa77e)
+ return C;
+ // 5 characters
+ if (C <= 0xa786)
+ return C | 1;
+ // LATIN CAPITAL LETTER SALTILLO
+ if (C == 0xa78b)
+ return 0xa78c;
+ // LATIN CAPITAL LETTER TURNED H
+ if (C == 0xa78d)
+ return 0x0265;
+ if (C < 0xa790)
+ return C;
+ // 2 characters
+ if (C <= 0xa792)
+ return C | 1;
+ if (C < 0xa796)
+ return C;
+ // 10 characters
+ if (C <= 0xa7a8)
+ return C | 1;
+ // LATIN CAPITAL LETTER H WITH HOOK
+ if (C == 0xa7aa)
+ return 0x0266;
+ // LATIN CAPITAL LETTER REVERSED OPEN E
+ if (C == 0xa7ab)
+ return 0x025c;
+ // LATIN CAPITAL LETTER SCRIPT G
+ if (C == 0xa7ac)
+ return 0x0261;
+ // LATIN CAPITAL LETTER L WITH BELT
+ if (C == 0xa7ad)
+ return 0x026c;
+ // LATIN CAPITAL LETTER SMALL CAPITAL I
+ if (C == 0xa7ae)
+ return 0x026a;
+ // LATIN CAPITAL LETTER TURNED K
+ if (C == 0xa7b0)
+ return 0x029e;
+ // LATIN CAPITAL LETTER TURNED T
+ if (C == 0xa7b1)
+ return 0x0287;
+ // LATIN CAPITAL LETTER J WITH CROSSED-TAIL
+ if (C == 0xa7b2)
+ return 0x029d;
+ // LATIN CAPITAL LETTER CHI
+ if (C == 0xa7b3)
+ return 0xab53;
+ if (C < 0xa7b4)
+ return C;
+ // 2 characters
+ if (C <= 0xa7b6)
+ return C | 1;
+ if (C < 0xab70)
+ return C;
+ // 80 characters
+ if (C <= 0xabbf)
+ return C + -38864;
+ if (C < 0xff21)
+ return C;
+ // 26 characters
+ if (C <= 0xff3a)
+ return C + 32;
+ if (C < 0x10400)
+ return C;
+ // 40 characters
+ if (C <= 0x10427)
+ return C + 40;
+ if (C < 0x104b0)
+ return C;
+ // 36 characters
+ if (C <= 0x104d3)
+ return C + 40;
+ if (C < 0x10c80)
+ return C;
+ // 51 characters
+ if (C <= 0x10cb2)
+ return C + 64;
+ if (C < 0x118a0)
+ return C;
+ // 32 characters
+ if (C <= 0x118bf)
+ return C + 32;
+ if (C < 0x1e900)
+ return C;
+ // 34 characters
+ if (C <= 0x1e921)
+ return C + 34;
+
+ return C;
+}
diff --git a/third_party/llvm-project/Unix/Path.inc b/third_party/llvm-project/Unix/Path.inc
new file mode 100644
index 000000000..b86766d4b
--- /dev/null
+++ b/third_party/llvm-project/Unix/Path.inc
@@ -0,0 +1,1213 @@
+//===- llvm/Support/Unix/Path.inc - Unix Path Implementation ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the Unix specific implementation of the Path API.
+//
+//===----------------------------------------------------------------------===//
+
+//===----------------------------------------------------------------------===//
+//=== WARNING: Implementation here must contain only generic UNIX code that
+//=== is guaranteed to work on *all* UNIX variants.
+//===----------------------------------------------------------------------===//
+
+#include "Unix.h"
+#include <limits.h>
+#include <stdio.h>
+#if HAVE_SYS_STAT_H
+#include <sys/stat.h>
+#endif
+#if HAVE_FCNTL_H
+#include <fcntl.h>
+#endif
+#ifdef HAVE_UNISTD_H
+#include <unistd.h>
+#endif
+#ifdef HAVE_SYS_MMAN_H
+#include <sys/mman.h>
+#endif
+
+#include <dirent.h>
+#include <pwd.h>
+
+#ifdef __APPLE__
+#include <mach-o/dyld.h>
+#include <sys/attr.h>
+#include <copyfile.h>
+#elif defined(__DragonFly__)
+#include <sys/mount.h>
+#endif
+
+// Both stdio.h and cstdio are included via different paths and
+// stdcxx's cstdio doesn't include stdio.h, so it doesn't #undef the macros
+// either.
+#undef ferror
+#undef feof
+
+// For GNU Hurd
+#if defined(__GNU__) && !defined(PATH_MAX)
+# define PATH_MAX 4096
+# define MAXPATHLEN 4096
+#endif
+
+#include <sys/types.h>
+#if !defined(__APPLE__) && !defined(__OpenBSD__) && !defined(__FreeBSD__) && \
+ !defined(__linux__) && !defined(__FreeBSD_kernel__) && !defined(_AIX)
+#include <sys/statvfs.h>
+#define STATVFS statvfs
+#define FSTATVFS fstatvfs
+#define STATVFS_F_FRSIZE(vfs) vfs.f_frsize
+#else
+#if defined(__OpenBSD__) || defined(__FreeBSD__)
+#include <sys/mount.h>
+#include <sys/param.h>
+#elif defined(__linux__)
+#if defined(HAVE_LINUX_MAGIC_H)
+#include <linux/magic.h>
+#else
+#if defined(HAVE_LINUX_NFS_FS_H)
+#include <linux/nfs_fs.h>
+#endif
+#if defined(HAVE_LINUX_SMB_H)
+#include <linux/smb.h>
+#endif
+#endif
+#include <sys/vfs.h>
+#elif defined(_AIX)
+#include <sys/statfs.h>
+
+// <sys/vmount.h> depends on `uint` to be a typedef from <sys/types.h> to
+// `uint_t`; however, <sys/types.h> does not always declare `uint`. We provide
+// the typedef prior to including <sys/vmount.h> to work around this issue.
+typedef uint_t uint;
+#include <sys/vmount.h>
+#else
+#include <sys/mount.h>
+#endif
+#define STATVFS statfs
+#define FSTATVFS fstatfs
+#define STATVFS_F_FRSIZE(vfs) static_cast<uint64_t>(vfs.f_bsize)
+#endif
+
+#if defined(__NetBSD__) || defined(__DragonFly__) || defined(__GNU__)
+#define STATVFS_F_FLAG(vfs) (vfs).f_flag
+#else
+#define STATVFS_F_FLAG(vfs) (vfs).f_flags
+#endif
+
+using namespace llvm;
+
+namespace llvm {
+namespace sys {
+namespace fs {
+
+const file_t kInvalidFile = -1;
+
+#if defined(__FreeBSD__) || defined(__NetBSD__) || defined(__OpenBSD__) || \
+ defined(__minix) || defined(__FreeBSD_kernel__) || defined(__linux__) || \
+ defined(__CYGWIN__) || defined(__DragonFly__) || defined(_AIX) || defined(__GNU__)
+static int
+test_dir(char ret[PATH_MAX], const char *dir, const char *bin)
+{
+ struct stat sb;
+ char fullpath[PATH_MAX];
+
+ int chars = snprintf(fullpath, PATH_MAX, "%s/%s", dir, bin);
+ // We cannot write PATH_MAX characters because the string will be terminated
+ // with a null character. Fail if truncation happened.
+ if (chars >= PATH_MAX)
+ return 1;
+ if (!realpath(fullpath, ret))
+ return 1;
+ if (stat(fullpath, &sb) != 0)
+ return 1;
+
+ return 0;
+}
+
+static char *
+getprogpath(char ret[PATH_MAX], const char *bin)
+{
+ /* First approach: absolute path. */
+ if (bin[0] == '/') {
+ if (test_dir(ret, "/", bin) == 0)
+ return ret;
+ return nullptr;
+ }
+
+ /* Second approach: relative path. */
+ if (strchr(bin, '/')) {
+ char cwd[PATH_MAX];
+ if (!getcwd(cwd, PATH_MAX))
+ return nullptr;
+ if (test_dir(ret, cwd, bin) == 0)
+ return ret;
+ return nullptr;
+ }
+
+ /* Third approach: $PATH */
+ char *pv;
+ if ((pv = getenv("PATH")) == nullptr)
+ return nullptr;
+ char *s = strdup(pv);
+ if (!s)
+ return nullptr;
+ char *state;
+ for (char *t = strtok_r(s, ":", &state); t != nullptr;
+ t = strtok_r(nullptr, ":", &state)) {
+ if (test_dir(ret, t, bin) == 0) {
+ free(s);
+ return ret;
+ }
+ }
+ free(s);
+ return nullptr;
+}
+#endif // __FreeBSD__ || __NetBSD__ || __FreeBSD_kernel__
+
+/// GetMainExecutable - Return the path to the main executable, given the
+/// value of argv[0] from program startup.
+std::string getMainExecutable(const char *argv0, void *MainAddr) {
+#if defined(__APPLE__)
+ // On OS X the executable path is saved to the stack by dyld. Reading it
+ // from there is much faster than calling dladdr, especially for large
+ // binaries with symbols.
+ char exe_path[MAXPATHLEN];
+ uint32_t size = sizeof(exe_path);
+ if (_NSGetExecutablePath(exe_path, &size) == 0) {
+ char link_path[MAXPATHLEN];
+ if (realpath(exe_path, link_path))
+ return link_path;
+ }
+#elif defined(__FreeBSD__) || defined(__NetBSD__) || defined(__OpenBSD__) || \
+ defined(__minix) || defined(__DragonFly__) || \
+ defined(__FreeBSD_kernel__) || defined(_AIX)
+ const char *curproc = "/proc/curproc/file";
+ char exe_path[PATH_MAX];
+ // /proc is not mounted by default under FreeBSD, but gives more accurate
+ // information than argv[0] when it is.
+ if (sys::fs::exists(curproc)) {
+ ssize_t len = readlink(curproc, exe_path, sizeof(exe_path));
+ if (len > 0) {
+ // Null terminate the string for realpath. readlink never null
+ // terminates its output.
+ len = std::min(len, ssize_t(sizeof(exe_path) - 1));
+ exe_path[len] = '\0';
+ return exe_path;
+ }
+ }
+ // If we don't have procfs mounted, fall back to argv[0]
+ if (getprogpath(exe_path, argv0) != NULL)
+ return exe_path;
+#elif defined(__linux__) || defined(__CYGWIN__) || defined(__gnu_hurd__)
+ char exe_path[MAXPATHLEN];
+ const char *aPath = "/proc/self/exe";
+ if (sys::fs::exists(aPath)) {
+ // /proc is not always mounted under Linux (chroot for example).
+ ssize_t len = readlink(aPath, exe_path, sizeof(exe_path));
+ if (len < 0)
+ return "";
+
+ // Null terminate the string for realpath. readlink never null
+ // terminates its output.
+ len = std::min(len, ssize_t(sizeof(exe_path) - 1));
+ exe_path[len] = '\0';
+
+ // On Linux, /proc/self/exe always looks through symlinks. However, on
+ // GNU/Hurd, /proc/self/exe is a symlink to the path that was used to start
+ // the program, and not the eventual binary file. Therefore, call realpath
+ // so this behaves the same on all platforms.
+#if _POSIX_VERSION >= 200112 || defined(__GLIBC__)
+ if (char *real_path = realpath(exe_path, NULL)) {
+ std::string ret = std::string(real_path);
+ free(real_path);
+ return ret;
+ }
+#else
+ char real_path[MAXPATHLEN];
+ if (realpath(exe_path, real_path))
+ return std::string(real_path);
+#endif
+ }
+ // Fall back to the classical detection.
+ if (getprogpath(exe_path, argv0))
+ return exe_path;
+#elif defined(HAVE_DLFCN_H) && defined(HAVE_DLADDR)
+ // Use dladdr to get executable path if available.
+ Dl_info DLInfo;
+ int err = dladdr(MainAddr, &DLInfo);
+ if (err == 0)
+ return "";
+
+ // If the filename is a symlink, we need to resolve and return the location of
+ // the actual executable.
+ char link_path[MAXPATHLEN];
+ if (realpath(DLInfo.dli_fname, link_path))
+ return link_path;
+#else
+#error GetMainExecutable is not implemented on this host yet.
+#endif
+ return "";
+}
+
+TimePoint<> basic_file_status::getLastAccessedTime() const {
+ return toTimePoint(fs_st_atime, fs_st_atime_nsec);
+}
+
+TimePoint<> basic_file_status::getLastModificationTime() const {
+ return toTimePoint(fs_st_mtime, fs_st_mtime_nsec);
+}
+
+UniqueID file_status::getUniqueID() const {
+ return UniqueID(fs_st_dev, fs_st_ino);
+}
+
+uint32_t file_status::getLinkCount() const {
+ return fs_st_nlinks;
+}
+
+ErrorOr<space_info> disk_space(const Twine &Path) {
+ struct STATVFS Vfs;
+ if (::STATVFS(const_cast<char *>(Path.str().c_str()), &Vfs))
+ return std::error_code(errno, std::generic_category());
+ auto FrSize = STATVFS_F_FRSIZE(Vfs);
+ space_info SpaceInfo;
+ SpaceInfo.capacity = static_cast<uint64_t>(Vfs.f_blocks) * FrSize;
+ SpaceInfo.free = static_cast<uint64_t>(Vfs.f_bfree) * FrSize;
+ SpaceInfo.available = static_cast<uint64_t>(Vfs.f_bavail) * FrSize;
+ return SpaceInfo;
+}
+
+std::error_code current_path(SmallVectorImpl<char> &result) {
+ result.clear();
+
+ const char *pwd = ::getenv("PWD");
+ llvm::sys::fs::file_status PWDStatus, DotStatus;
+ if (pwd && llvm::sys::path::is_absolute(pwd) &&
+ !llvm::sys::fs::status(pwd, PWDStatus) &&
+ !llvm::sys::fs::status(".", DotStatus) &&
+ PWDStatus.getUniqueID() == DotStatus.getUniqueID()) {
+ result.append(pwd, pwd + strlen(pwd));
+ return std::error_code();
+ }
+
+#ifdef MAXPATHLEN
+ result.reserve(MAXPATHLEN);
+#else
+// For GNU Hurd
+ result.reserve(1024);
+#endif
+
+ while (true) {
+ if (::getcwd(result.data(), result.capacity()) == nullptr) {
+ // See if there was a real error.
+ if (errno != ENOMEM)
+ return std::error_code(errno, std::generic_category());
+ // Otherwise there just wasn't enough space.
+ result.reserve(result.capacity() * 2);
+ } else
+ break;
+ }
+
+ result.set_size(strlen(result.data()));
+ return std::error_code();
+}
+
+std::error_code set_current_path(const Twine &path) {
+ SmallString<128> path_storage;
+ StringRef p = path.toNullTerminatedStringRef(path_storage);
+
+ if (::chdir(p.begin()) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ return std::error_code();
+}
+
+std::error_code create_directory(const Twine &path, bool IgnoreExisting,
+ perms Perms) {
+ SmallString<128> path_storage;
+ StringRef p = path.toNullTerminatedStringRef(path_storage);
+
+ if (::mkdir(p.begin(), Perms) == -1) {
+ if (errno != EEXIST || !IgnoreExisting)
+ return std::error_code(errno, std::generic_category());
+ }
+
+ return std::error_code();
+}
+
+// Note that we are using symbolic link because hard links are not supported by
+// all filesystems (SMB doesn't).
+std::error_code create_link(const Twine &to, const Twine &from) {
+ // Get arguments.
+ SmallString<128> from_storage;
+ SmallString<128> to_storage;
+ StringRef f = from.toNullTerminatedStringRef(from_storage);
+ StringRef t = to.toNullTerminatedStringRef(to_storage);
+
+ if (::symlink(t.begin(), f.begin()) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ return std::error_code();
+}
+
+std::error_code create_hard_link(const Twine &to, const Twine &from) {
+ // Get arguments.
+ SmallString<128> from_storage;
+ SmallString<128> to_storage;
+ StringRef f = from.toNullTerminatedStringRef(from_storage);
+ StringRef t = to.toNullTerminatedStringRef(to_storage);
+
+ if (::link(t.begin(), f.begin()) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ return std::error_code();
+}
+
+std::error_code remove(const Twine &path, bool IgnoreNonExisting) {
+ SmallString<128> path_storage;
+ StringRef p = path.toNullTerminatedStringRef(path_storage);
+
+ struct stat buf;
+ if (lstat(p.begin(), &buf) != 0) {
+ if (errno != ENOENT || !IgnoreNonExisting)
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+ }
+
+ // Note: this check catches strange situations. In all cases, LLVM should
+ // only be involved in the creation and deletion of regular files. This
+ // check ensures that what we're trying to erase is a regular file. It
+ // effectively prevents LLVM from erasing things like /dev/null, any block
+ // special file, or other things that aren't "regular" files.
+ if (!S_ISREG(buf.st_mode) && !S_ISDIR(buf.st_mode) && !S_ISLNK(buf.st_mode))
+ return make_error_code(errc::operation_not_permitted);
+
+ if (::remove(p.begin()) == -1) {
+ if (errno != ENOENT || !IgnoreNonExisting)
+ return std::error_code(errno, std::generic_category());
+ }
+
+ return std::error_code();
+}
+
+static bool is_local_impl(struct STATVFS &Vfs) {
+#if defined(__linux__) || defined(__GNU__)
+#ifndef NFS_SUPER_MAGIC
+#define NFS_SUPER_MAGIC 0x6969
+#endif
+#ifndef SMB_SUPER_MAGIC
+#define SMB_SUPER_MAGIC 0x517B
+#endif
+#ifndef CIFS_MAGIC_NUMBER
+#define CIFS_MAGIC_NUMBER 0xFF534D42
+#endif
+#ifdef __GNU__
+ switch ((uint32_t)Vfs.__f_type) {
+#else
+ switch ((uint32_t)Vfs.f_type) {
+#endif
+ case NFS_SUPER_MAGIC:
+ case SMB_SUPER_MAGIC:
+ case CIFS_MAGIC_NUMBER:
+ return false;
+ default:
+ return true;
+ }
+#elif defined(__CYGWIN__)
+ // Cygwin doesn't expose this information; would need to use Win32 API.
+ return false;
+#elif defined(__Fuchsia__)
+ // Fuchsia doesn't yet support remote filesystem mounts.
+ return true;
+#elif defined(__EMSCRIPTEN__)
+ // Emscripten doesn't currently support remote filesystem mounts.
+ return true;
+#elif defined(__HAIKU__)
+ // Haiku doesn't expose this information.
+ return false;
+#elif defined(__sun)
+ // statvfs::f_basetype contains a null-terminated FSType name of the mounted target
+ StringRef fstype(Vfs.f_basetype);
+ // NFS is the only non-local fstype??
+ return !fstype.equals("nfs");
+#elif defined(_AIX)
+ // Call mntctl; try more than twice in case of timing issues with a concurrent
+ // mount.
+ int Ret;
+ size_t BufSize = 2048u;
+ std::unique_ptr<char[]> Buf;
+ int Tries = 3;
+ while (Tries--) {
+ Buf = std::make_unique<char[]>(BufSize);
+ Ret = mntctl(MCTL_QUERY, BufSize, Buf.get());
+ if (Ret != 0)
+ break;
+ BufSize = *reinterpret_cast<unsigned int *>(Buf.get());
+ Buf.reset();
+ }
+
+ if (Ret == -1)
+ // There was an error; "remote" is the conservative answer.
+ return false;
+
+ // Look for the correct vmount entry.
+ char *CurObjPtr = Buf.get();
+ while (Ret--) {
+ struct vmount *Vp = reinterpret_cast<struct vmount *>(CurObjPtr);
+ static_assert(sizeof(Vfs.f_fsid) == sizeof(Vp->vmt_fsid),
+ "fsid length mismatch");
+ if (memcmp(&Vfs.f_fsid, &Vp->vmt_fsid, sizeof Vfs.f_fsid) == 0)
+ return (Vp->vmt_flags & MNT_REMOTE) == 0;
+
+ CurObjPtr += Vp->vmt_length;
+ }
+
+ // vmount entry not found; "remote" is the conservative answer.
+ return false;
+#else
+ return !!(STATVFS_F_FLAG(Vfs) & MNT_LOCAL);
+#endif
+}
+
+std::error_code is_local(const Twine &Path, bool &Result) {
+ struct STATVFS Vfs;
+ if (::STATVFS(const_cast<char *>(Path.str().c_str()), &Vfs))
+ return std::error_code(errno, std::generic_category());
+
+ Result = is_local_impl(Vfs);
+ return std::error_code();
+}
+
+std::error_code is_local(int FD, bool &Result) {
+ struct STATVFS Vfs;
+ if (::FSTATVFS(FD, &Vfs))
+ return std::error_code(errno, std::generic_category());
+
+ Result = is_local_impl(Vfs);
+ return std::error_code();
+}
+
+std::error_code rename(const Twine &from, const Twine &to) {
+ // Get arguments.
+ SmallString<128> from_storage;
+ SmallString<128> to_storage;
+ StringRef f = from.toNullTerminatedStringRef(from_storage);
+ StringRef t = to.toNullTerminatedStringRef(to_storage);
+
+ if (::rename(f.begin(), t.begin()) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ return std::error_code();
+}
+
+std::error_code resize_file(int FD, uint64_t Size) {
+#if defined(HAVE_POSIX_FALLOCATE)
+ // If we have posix_fallocate use it. Unlike ftruncate it always allocates
+ // space, so we get an error if the disk is full.
+ if (int Err = ::posix_fallocate(FD, 0, Size)) {
+#ifdef _AIX
+ constexpr int NotSupportedError = ENOTSUP;
+#else
+ constexpr int NotSupportedError = EOPNOTSUPP;
+#endif
+ if (Err != EINVAL && Err != NotSupportedError)
+ return std::error_code(Err, std::generic_category());
+ }
+#endif
+ // Use ftruncate as a fallback. It may or may not allocate space. At least on
+ // OS X with HFS+ it does.
+ if (::ftruncate(FD, Size) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ return std::error_code();
+}
+
+static int convertAccessMode(AccessMode Mode) {
+ switch (Mode) {
+ case AccessMode::Exist:
+ return F_OK;
+ case AccessMode::Write:
+ return W_OK;
+ case AccessMode::Execute:
+ return R_OK | X_OK; // scripts also need R_OK.
+ }
+ llvm_unreachable("invalid enum");
+}
+
+std::error_code access(const Twine &Path, AccessMode Mode) {
+ SmallString<128> PathStorage;
+ StringRef P = Path.toNullTerminatedStringRef(PathStorage);
+
+ if (::access(P.begin(), convertAccessMode(Mode)) == -1)
+ return std::error_code(errno, std::generic_category());
+
+ if (Mode == AccessMode::Execute) {
+ // Don't say that directories are executable.
+ struct stat buf;
+ if (0 != stat(P.begin(), &buf))
+ return errc::permission_denied;
+ if (!S_ISREG(buf.st_mode))
+ return errc::permission_denied;
+ }
+
+ return std::error_code();
+}
+
+bool can_execute(const Twine &Path) {
+ return !access(Path, AccessMode::Execute);
+}
+
+bool equivalent(file_status A, file_status B) {
+ assert(status_known(A) && status_known(B));
+ return A.fs_st_dev == B.fs_st_dev &&
+ A.fs_st_ino == B.fs_st_ino;
+}
+
+std::error_code equivalent(const Twine &A, const Twine &B, bool &result) {
+ file_status fsA, fsB;
+ if (std::error_code ec = status(A, fsA))
+ return ec;
+ if (std::error_code ec = status(B, fsB))
+ return ec;
+ result = equivalent(fsA, fsB);
+ return std::error_code();
+}
+
+static void expandTildeExpr(SmallVectorImpl<char> &Path) {
+ StringRef PathStr(Path.begin(), Path.size());
+ if (PathStr.empty() || !PathStr.startswith("~"))
+ return;
+
+ PathStr = PathStr.drop_front();
+ StringRef Expr =
+ PathStr.take_until([](char c) { return path::is_separator(c); });
+ StringRef Remainder = PathStr.substr(Expr.size() + 1);
+ SmallString<128> Storage;
+ if (Expr.empty()) {
+ // This is just ~/..., resolve it to the current user's home dir.
+ if (!path::home_directory(Storage)) {
+ // For some reason we couldn't get the home directory. Just exit.
+ return;
+ }
+
+ // Overwrite the first character and insert the rest.
+ Path[0] = Storage[0];
+ Path.insert(Path.begin() + 1, Storage.begin() + 1, Storage.end());
+ return;
+ }
+
+ // This is a string of the form ~username/, look up this user's entry in the
+ // password database.
+ struct passwd *Entry = nullptr;
+ std::string User = Expr.str();
+ Entry = ::getpwnam(User.c_str());
+
+ if (!Entry) {
+ // Unable to look up the entry, just return back the original path.
+ return;
+ }
+
+ Storage = Remainder;
+ Path.clear();
+ Path.append(Entry->pw_dir, Entry->pw_dir + strlen(Entry->pw_dir));
+ llvm::sys::path::append(Path, Storage);
+}
+
+
+void expand_tilde(const Twine &path, SmallVectorImpl<char> &dest) {
+ dest.clear();
+ if (path.isTriviallyEmpty())
+ return;
+
+ path.toVector(dest);
+ expandTildeExpr(dest);
+
+ return;
+}
+
+static file_type typeForMode(mode_t Mode) {
+ if (S_ISDIR(Mode))
+ return file_type::directory_file;
+ else if (S_ISREG(Mode))
+ return file_type::regular_file;
+ else if (S_ISBLK(Mode))
+ return file_type::block_file;
+ else if (S_ISCHR(Mode))
+ return file_type::character_file;
+ else if (S_ISFIFO(Mode))
+ return file_type::fifo_file;
+ else if (S_ISSOCK(Mode))
+ return file_type::socket_file;
+ else if (S_ISLNK(Mode))
+ return file_type::symlink_file;
+ return file_type::type_unknown;
+}
+
+static std::error_code fillStatus(int StatRet, const struct stat &Status,
+ file_status &Result) {
+ if (StatRet != 0) {
+ std::error_code EC(errno, std::generic_category());
+ if (EC == errc::no_such_file_or_directory)
+ Result = file_status(file_type::file_not_found);
+ else
+ Result = file_status(file_type::status_error);
+ return EC;
+ }
+
+ uint32_t atime_nsec, mtime_nsec;
+#if defined(HAVE_STRUCT_STAT_ST_MTIMESPEC_TV_NSEC)
+ atime_nsec = Status.st_atimespec.tv_nsec;
+ mtime_nsec = Status.st_mtimespec.tv_nsec;
+#elif defined(HAVE_STRUCT_STAT_ST_MTIM_TV_NSEC)
+ atime_nsec = Status.st_atim.tv_nsec;
+ mtime_nsec = Status.st_mtim.tv_nsec;
+#else
+ atime_nsec = mtime_nsec = 0;
+#endif
+
+ perms Perms = static_cast<perms>(Status.st_mode) & all_perms;
+ Result = file_status(typeForMode(Status.st_mode), Perms, Status.st_dev,
+ Status.st_nlink, Status.st_ino,
+ Status.st_atime, atime_nsec, Status.st_mtime, mtime_nsec,
+ Status.st_uid, Status.st_gid, Status.st_size);
+
+ return std::error_code();
+}
+
+std::error_code status(const Twine &Path, file_status &Result, bool Follow) {
+ SmallString<128> PathStorage;
+ StringRef P = Path.toNullTerminatedStringRef(PathStorage);
+
+ struct stat Status;
+ int StatRet = (Follow ? ::stat : ::lstat)(P.begin(), &Status);
+ return fillStatus(StatRet, Status, Result);
+}
+
+std::error_code status(int FD, file_status &Result) {
+ struct stat Status;
+ int StatRet = ::fstat(FD, &Status);
+ return fillStatus(StatRet, Status, Result);
+}
+
+unsigned getUmask() {
+ // Chose arbitary new mask and reset the umask to the old mask.
+ // umask(2) never fails so ignore the return of the second call.
+ unsigned Mask = ::umask(0);
+ (void) ::umask(Mask);
+ return Mask;
+}
+
+std::error_code setPermissions(const Twine &Path, perms Permissions) {
+ SmallString<128> PathStorage;
+ StringRef P = Path.toNullTerminatedStringRef(PathStorage);
+
+ if (::chmod(P.begin(), Permissions))
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+}
+
+std::error_code setPermissions(int FD, perms Permissions) {
+ if (::fchmod(FD, Permissions))
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+}
+
+std::error_code setLastAccessAndModificationTime(int FD, TimePoint<> AccessTime,
+ TimePoint<> ModificationTime) {
+#if defined(HAVE_FUTIMENS)
+ timespec Times[2];
+ Times[0] = sys::toTimeSpec(AccessTime);
+ Times[1] = sys::toTimeSpec(ModificationTime);
+ if (::futimens(FD, Times))
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+#elif defined(HAVE_FUTIMES)
+ timeval Times[2];
+ Times[0] = sys::toTimeVal(
+ std::chrono::time_point_cast<std::chrono::microseconds>(AccessTime));
+ Times[1] =
+ sys::toTimeVal(std::chrono::time_point_cast<std::chrono::microseconds>(
+ ModificationTime));
+ if (::futimes(FD, Times))
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+#else
+#warning Missing futimes() and futimens()
+ return make_error_code(errc::function_not_supported);
+#endif
+}
+
+std::error_code mapped_file_region::init(int FD, uint64_t Offset,
+ mapmode Mode) {
+ assert(Size != 0);
+
+ int flags = (Mode == readwrite) ? MAP_SHARED : MAP_PRIVATE;
+ int prot = (Mode == readonly) ? PROT_READ : (PROT_READ | PROT_WRITE);
+#if defined(__APPLE__)
+ //----------------------------------------------------------------------
+ // Newer versions of MacOSX have a flag that will allow us to read from
+ // binaries whose code signature is invalid without crashing by using
+ // the MAP_RESILIENT_CODESIGN flag. Also if a file from removable media
+ // is mapped we can avoid crashing and return zeroes to any pages we try
+ // to read if the media becomes unavailable by using the
+ // MAP_RESILIENT_MEDIA flag. These flags are only usable when mapping
+ // with PROT_READ, so take care not to specify them otherwise.
+ //----------------------------------------------------------------------
+ if (Mode == readonly) {
+#if defined(MAP_RESILIENT_CODESIGN)
+ flags |= MAP_RESILIENT_CODESIGN;
+#endif
+#if defined(MAP_RESILIENT_MEDIA)
+ flags |= MAP_RESILIENT_MEDIA;
+#endif
+ }
+#endif // #if defined (__APPLE__)
+
+ Mapping = ::mmap(nullptr, Size, prot, flags, FD, Offset);
+ if (Mapping == MAP_FAILED)
+ return std::error_code(errno, std::generic_category());
+ return std::error_code();
+}
+
+mapped_file_region::mapped_file_region(int fd, mapmode mode, size_t length,
+ uint64_t offset, std::error_code &ec)
+ : Size(length), Mapping(), Mode(mode) {
+ (void)Mode;
+ ec = init(fd, offset, mode);
+ if (ec)
+ Mapping = nullptr;
+}
+
+mapped_file_region::~mapped_file_region() {
+ if (Mapping)
+ ::munmap(Mapping, Size);
+}
+
+size_t mapped_file_region::size() const {
+ assert(Mapping && "Mapping failed but used anyway!");
+ return Size;
+}
+
+char *mapped_file_region::data() const {
+ assert(Mapping && "Mapping failed but used anyway!");
+ return reinterpret_cast<char*>(Mapping);
+}
+
+const char *mapped_file_region::const_data() const {
+ assert(Mapping && "Mapping failed but used anyway!");
+ return reinterpret_cast<const char*>(Mapping);
+}
+
+int mapped_file_region::alignment() {
+ return Process::getPageSizeEstimate();
+}
+
+std::error_code detail::directory_iterator_construct(detail::DirIterState &it,
+ StringRef path,
+ bool follow_symlinks) {
+ SmallString<128> path_null(path);
+ DIR *directory = ::opendir(path_null.c_str());
+ if (!directory)
+ return std::error_code(errno, std::generic_category());
+
+ it.IterationHandle = reinterpret_cast<intptr_t>(directory);
+ // Add something for replace_filename to replace.
+ path::append(path_null, ".");
+ it.CurrentEntry = directory_entry(path_null.str(), follow_symlinks);
+ return directory_iterator_increment(it);
+}
+
+std::error_code detail::directory_iterator_destruct(detail::DirIterState &it) {
+ if (it.IterationHandle)
+ ::closedir(reinterpret_cast<DIR *>(it.IterationHandle));
+ it.IterationHandle = 0;
+ it.CurrentEntry = directory_entry();
+ return std::error_code();
+}
+
+static file_type direntType(dirent* Entry) {
+ // Most platforms provide the file type in the dirent: Linux/BSD/Mac.
+ // The DTTOIF macro lets us reuse our status -> type conversion.
+ // Note that while glibc provides a macro to see if this is supported,
+ // _DIRENT_HAVE_D_TYPE, it's not defined on BSD/Mac, so we test for the
+ // d_type-to-mode_t conversion macro instead.
+#if defined(DTTOIF)
+ return typeForMode(DTTOIF(Entry->d_type));
+#else
+ // Other platforms such as Solaris require a stat() to get the type.
+ return file_type::type_unknown;
+#endif
+}
+
+std::error_code detail::directory_iterator_increment(detail::DirIterState &It) {
+ errno = 0;
+ dirent *CurDir = ::readdir(reinterpret_cast<DIR *>(It.IterationHandle));
+ if (CurDir == nullptr && errno != 0) {
+ return std::error_code(errno, std::generic_category());
+ } else if (CurDir != nullptr) {
+ StringRef Name(CurDir->d_name);
+ if ((Name.size() == 1 && Name[0] == '.') ||
+ (Name.size() == 2 && Name[0] == '.' && Name[1] == '.'))
+ return directory_iterator_increment(It);
+ It.CurrentEntry.replace_filename(Name, direntType(CurDir));
+ } else
+ return directory_iterator_destruct(It);
+
+ return std::error_code();
+}
+
+ErrorOr<basic_file_status> directory_entry::status() const {
+ file_status s;
+ if (auto EC = fs::status(Path, s, FollowSymlinks))
+ return EC;
+ return s;
+}
+
+#if !defined(F_GETPATH)
+static bool hasProcSelfFD() {
+ // If we have a /proc filesystem mounted, we can quickly establish the
+ // real name of the file with readlink
+ static const bool Result = (::access("/proc/self/fd", R_OK) == 0);
+ return Result;
+}
+#endif
+
+static int nativeOpenFlags(CreationDisposition Disp, OpenFlags Flags,
+ FileAccess Access) {
+ int Result = 0;
+ if (Access == FA_Read)
+ Result |= O_RDONLY;
+ else if (Access == FA_Write)
+ Result |= O_WRONLY;
+ else if (Access == (FA_Read | FA_Write))
+ Result |= O_RDWR;
+
+ // This is for compatibility with old code that assumed OF_Append implied
+ // would open an existing file. See Windows/Path.inc for a longer comment.
+ if (Flags & OF_Append)
+ Disp = CD_OpenAlways;
+
+ if (Disp == CD_CreateNew) {
+ Result |= O_CREAT; // Create if it doesn't exist.
+ Result |= O_EXCL; // Fail if it does.
+ } else if (Disp == CD_CreateAlways) {
+ Result |= O_CREAT; // Create if it doesn't exist.
+ Result |= O_TRUNC; // Truncate if it does.
+ } else if (Disp == CD_OpenAlways) {
+ Result |= O_CREAT; // Create if it doesn't exist.
+ } else if (Disp == CD_OpenExisting) {
+ // Nothing special, just don't add O_CREAT and we get these semantics.
+ }
+
+ if (Flags & OF_Append)
+ Result |= O_APPEND;
+
+#ifdef O_CLOEXEC
+ if (!(Flags & OF_ChildInherit))
+ Result |= O_CLOEXEC;
+#endif
+
+ return Result;
+}
+
+std::error_code openFile(const Twine &Name, int &ResultFD,
+ CreationDisposition Disp, FileAccess Access,
+ OpenFlags Flags, unsigned Mode) {
+ int OpenFlags = nativeOpenFlags(Disp, Flags, Access);
+
+ SmallString<128> Storage;
+ StringRef P = Name.toNullTerminatedStringRef(Storage);
+ // Call ::open in a lambda to avoid overload resolution in RetryAfterSignal
+ // when open is overloaded, such as in Bionic.
+ auto Open = [&]() { return ::open(P.begin(), OpenFlags, Mode); };
+ if ((ResultFD = sys::RetryAfterSignal(-1, Open)) < 0)
+ return std::error_code(errno, std::generic_category());
+#ifndef O_CLOEXEC
+ if (!(Flags & OF_ChildInherit)) {
+ int r = fcntl(ResultFD, F_SETFD, FD_CLOEXEC);
+ (void)r;
+ assert(r == 0 && "fcntl(F_SETFD, FD_CLOEXEC) failed");
+ }
+#endif
+ return std::error_code();
+}
+
+Expected<int> openNativeFile(const Twine &Name, CreationDisposition Disp,
+ FileAccess Access, OpenFlags Flags,
+ unsigned Mode) {
+
+ int FD;
+ std::error_code EC = openFile(Name, FD, Disp, Access, Flags, Mode);
+ if (EC)
+ return errorCodeToError(EC);
+ return FD;
+}
+
+std::error_code openFileForRead(const Twine &Name, int &ResultFD,
+ OpenFlags Flags,
+ SmallVectorImpl<char> *RealPath) {
+ std::error_code EC =
+ openFile(Name, ResultFD, CD_OpenExisting, FA_Read, Flags, 0666);
+ if (EC)
+ return EC;
+
+ // Attempt to get the real name of the file, if the user asked
+ if(!RealPath)
+ return std::error_code();
+ RealPath->clear();
+#if defined(F_GETPATH)
+ // When F_GETPATH is availble, it is the quickest way to get
+ // the real path name.
+ char Buffer[MAXPATHLEN];
+ if (::fcntl(ResultFD, F_GETPATH, Buffer) != -1)
+ RealPath->append(Buffer, Buffer + strlen(Buffer));
+#else
+ char Buffer[PATH_MAX];
+ if (hasProcSelfFD()) {
+ char ProcPath[64];
+ snprintf(ProcPath, sizeof(ProcPath), "/proc/self/fd/%d", ResultFD);
+ ssize_t CharCount = ::readlink(ProcPath, Buffer, sizeof(Buffer));
+ if (CharCount > 0)
+ RealPath->append(Buffer, Buffer + CharCount);
+ } else {
+ SmallString<128> Storage;
+ StringRef P = Name.toNullTerminatedStringRef(Storage);
+
+ // Use ::realpath to get the real path name
+ if (::realpath(P.begin(), Buffer) != nullptr)
+ RealPath->append(Buffer, Buffer + strlen(Buffer));
+ }
+#endif
+ return std::error_code();
+}
+
+Expected<file_t> openNativeFileForRead(const Twine &Name, OpenFlags Flags,
+ SmallVectorImpl<char> *RealPath) {
+ file_t ResultFD;
+ std::error_code EC = openFileForRead(Name, ResultFD, Flags, RealPath);
+ if (EC)
+ return errorCodeToError(EC);
+ return ResultFD;
+}
+
+file_t getStdinHandle() { return 0; }
+file_t getStdoutHandle() { return 1; }
+file_t getStderrHandle() { return 2; }
+
+Expected<size_t> readNativeFile(file_t FD, MutableArrayRef<char> Buf) {
+ ssize_t NumRead =
+ sys::RetryAfterSignal(-1, ::read, FD, Buf.data(), Buf.size());
+ if (ssize_t(NumRead) == -1)
+ return errorCodeToError(std::error_code(errno, std::generic_category()));
+ return NumRead;
+}
+
+Expected<size_t> readNativeFileSlice(file_t FD, MutableArrayRef<char> Buf,
+ uint64_t Offset) {
+#ifdef HAVE_PREAD
+ ssize_t NumRead =
+ sys::RetryAfterSignal(-1, ::pread, FD, Buf.data(), Buf.size(), Offset);
+#else
+ if (lseek(FD, Offset, SEEK_SET) == -1)
+ return errorCodeToError(std::error_code(errno, std::generic_category()));
+ ssize_t NumRead =
+ sys::RetryAfterSignal(-1, ::read, FD, Buf.data(), Buf.size());
+#endif
+ if (NumRead == -1)
+ return errorCodeToError(std::error_code(errno, std::generic_category()));
+ return NumRead;
+}
+
+std::error_code closeFile(file_t &F) {
+ file_t TmpF = F;
+ F = kInvalidFile;
+ return Process::SafelyCloseFileDescriptor(TmpF);
+}
+
+template <typename T>
+static std::error_code remove_directories_impl(const T &Entry,
+ bool IgnoreErrors) {
+ std::error_code EC;
+ directory_iterator Begin(Entry, EC, false);
+ directory_iterator End;
+ while (Begin != End) {
+ auto &Item = *Begin;
+ ErrorOr<basic_file_status> st = Item.status();
+ if (!st && !IgnoreErrors)
+ return st.getError();
+
+ if (is_directory(*st)) {
+ EC = remove_directories_impl(Item, IgnoreErrors);
+ if (EC && !IgnoreErrors)
+ return EC;
+ }
+
+ EC = fs::remove(Item.path(), true);
+ if (EC && !IgnoreErrors)
+ return EC;
+
+ Begin.increment(EC);
+ if (EC && !IgnoreErrors)
+ return EC;
+ }
+ return std::error_code();
+}
+
+std::error_code remove_directories(const Twine &path, bool IgnoreErrors) {
+ auto EC = remove_directories_impl(path, IgnoreErrors);
+ if (EC && !IgnoreErrors)
+ return EC;
+ EC = fs::remove(path, true);
+ if (EC && !IgnoreErrors)
+ return EC;
+ return std::error_code();
+}
+
+std::error_code real_path(const Twine &path, SmallVectorImpl<char> &dest,
+ bool expand_tilde) {
+ dest.clear();
+ if (path.isTriviallyEmpty())
+ return std::error_code();
+
+ if (expand_tilde) {
+ SmallString<128> Storage;
+ path.toVector(Storage);
+ expandTildeExpr(Storage);
+ return real_path(Storage, dest, false);
+ }
+
+ SmallString<128> Storage;
+ StringRef P = path.toNullTerminatedStringRef(Storage);
+ char Buffer[PATH_MAX];
+ if (::realpath(P.begin(), Buffer) == nullptr)
+ return std::error_code(errno, std::generic_category());
+ dest.append(Buffer, Buffer + strlen(Buffer));
+ return std::error_code();
+}
+
+} // end namespace fs
+
+namespace path {
+
+bool home_directory(SmallVectorImpl<char> &result) {
+ char *RequestedDir = getenv("HOME");
+ if (!RequestedDir) {
+ struct passwd *pw = getpwuid(getuid());
+ if (pw && pw->pw_dir)
+ RequestedDir = pw->pw_dir;
+ }
+ if (!RequestedDir)
+ return false;
+
+ result.clear();
+ result.append(RequestedDir, RequestedDir + strlen(RequestedDir));
+ return true;
+}
+
+static bool getDarwinConfDir(bool TempDir, SmallVectorImpl<char> &Result) {
+ #if defined(_CS_DARWIN_USER_TEMP_DIR) && defined(_CS_DARWIN_USER_CACHE_DIR)
+ // On Darwin, use DARWIN_USER_TEMP_DIR or DARWIN_USER_CACHE_DIR.
+ // macros defined in <unistd.h> on darwin >= 9
+ int ConfName = TempDir ? _CS_DARWIN_USER_TEMP_DIR
+ : _CS_DARWIN_USER_CACHE_DIR;
+ size_t ConfLen = confstr(ConfName, nullptr, 0);
+ if (ConfLen > 0) {
+ do {
+ Result.resize(ConfLen);
+ ConfLen = confstr(ConfName, Result.data(), Result.size());
+ } while (ConfLen > 0 && ConfLen != Result.size());
+
+ if (ConfLen > 0) {
+ assert(Result.back() == 0);
+ Result.pop_back();
+ return true;
+ }
+
+ Result.clear();
+ }
+ #endif
+ return false;
+}
+
+static const char *getEnvTempDir() {
+ // Check whether the temporary directory is specified by an environment
+ // variable.
+ const char *EnvironmentVariables[] = {"TMPDIR", "TMP", "TEMP", "TEMPDIR"};
+ for (const char *Env : EnvironmentVariables) {
+ if (const char *Dir = std::getenv(Env))
+ return Dir;
+ }
+
+ return nullptr;
+}
+
+static const char *getDefaultTempDir(bool ErasedOnReboot) {
+#ifdef P_tmpdir
+ if ((bool)P_tmpdir)
+ return P_tmpdir;
+#endif
+
+ if (ErasedOnReboot)
+ return "/tmp";
+ return "/var/tmp";
+}
+
+void system_temp_directory(bool ErasedOnReboot, SmallVectorImpl<char> &Result) {
+ Result.clear();
+
+ if (ErasedOnReboot) {
+ // There is no env variable for the cache directory.
+ if (const char *RequestedDir = getEnvTempDir()) {
+ Result.append(RequestedDir, RequestedDir + strlen(RequestedDir));
+ return;
+ }
+ }
+
+ if (getDarwinConfDir(ErasedOnReboot, Result))
+ return;
+
+ const char *RequestedDir = getDefaultTempDir(ErasedOnReboot);
+ Result.append(RequestedDir, RequestedDir + strlen(RequestedDir));
+}
+
+} // end namespace path
+
+namespace fs {
+
+#ifdef __APPLE__
+/// This implementation tries to perform an APFS CoW clone of the file,
+/// which can be much faster and uses less space.
+/// Unfortunately fcopyfile(3) does not support COPYFILE_CLONE, so the
+/// file descriptor variant of this function still uses the default
+/// implementation.
+std::error_code copy_file(const Twine &From, const Twine &To) {
+ uint32_t Flag = COPYFILE_DATA;
+#if __has_builtin(__builtin_available) && defined(COPYFILE_CLONE)
+ if (__builtin_available(macos 10.12, *)) {
+ bool IsSymlink;
+ if (std::error_code Error = is_symlink_file(From, IsSymlink))
+ return Error;
+ // COPYFILE_CLONE clones the symlink instead of following it
+ // and returns EEXISTS if the target file already exists.
+ if (!IsSymlink && !exists(To))
+ Flag = COPYFILE_CLONE;
+ }
+#endif
+ int Status =
+ copyfile(From.str().c_str(), To.str().c_str(), /* State */ NULL, Flag);
+
+ if (Status == 0)
+ return std::error_code();
+ return std::error_code(errno, std::generic_category());
+}
+#endif // __APPLE__
+
+} // end namespace fs
+
+} // end namespace sys
+} // end namespace llvm
diff --git a/third_party/llvm-project/Unix/Unix.h b/third_party/llvm-project/Unix/Unix.h
new file mode 100644
index 000000000..1fc9a414f
--- /dev/null
+++ b/third_party/llvm-project/Unix/Unix.h
@@ -0,0 +1,114 @@
+//===- llvm/Support/Unix/Unix.h - Common Unix Include File -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines things specific to Unix implementations.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIB_SUPPORT_UNIX_UNIX_H
+#define LLVM_LIB_SUPPORT_UNIX_UNIX_H
+
+//===----------------------------------------------------------------------===//
+//=== WARNING: Implementation here must contain only generic UNIX code that
+//=== is guaranteed to work on all UNIX variants.
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Config/config.h"
+#include "llvm/Support/Chrono.h"
+#include "llvm/Support/Errno.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <algorithm>
+#include <assert.h>
+#include <cerrno>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <string>
+#include <sys/types.h>
+#include <sys/wait.h>
+
+#ifdef HAVE_UNISTD_H
+#include <unistd.h>
+#endif
+
+#ifdef HAVE_SYS_PARAM_H
+#include <sys/param.h>
+#endif
+
+#ifdef HAVE_SYS_TIME_H
+# include <sys/time.h>
+#endif
+#include <time.h>
+
+#ifdef HAVE_DLFCN_H
+# include <dlfcn.h>
+#endif
+
+#ifdef HAVE_FCNTL_H
+# include <fcntl.h>
+#endif
+
+/// This function builds an error message into \p ErrMsg using the \p prefix
+/// string and the Unix error number given by \p errnum. If errnum is -1, the
+/// default then the value of errno is used.
+/// Make an error message
+///
+/// If the error number can be converted to a string, it will be
+/// separated from prefix by ": ".
+static inline bool MakeErrMsg(
+ std::string* ErrMsg, const std::string& prefix, int errnum = -1) {
+ if (!ErrMsg)
+ return true;
+ if (errnum == -1)
+ errnum = errno;
+ *ErrMsg = prefix + ": " + llvm::sys::StrError(errnum);
+ return true;
+}
+
+// Include StrError(errnum) in a fatal error message.
+LLVM_ATTRIBUTE_NORETURN static inline void ReportErrnumFatal(const char *Msg,
+ int errnum) {
+ std::string ErrMsg;
+ MakeErrMsg(&ErrMsg, Msg, errnum);
+ llvm::report_fatal_error(ErrMsg);
+}
+
+namespace llvm {
+namespace sys {
+
+/// Convert a struct timeval to a duration. Note that timeval can be used both
+/// as a time point and a duration. Be sure to check what the input represents.
+inline std::chrono::microseconds toDuration(const struct timeval &TV) {
+ return std::chrono::seconds(TV.tv_sec) +
+ std::chrono::microseconds(TV.tv_usec);
+}
+
+/// Convert a time point to struct timespec.
+inline struct timespec toTimeSpec(TimePoint<> TP) {
+ using namespace std::chrono;
+
+ struct timespec RetVal;
+ RetVal.tv_sec = toTimeT(TP);
+ RetVal.tv_nsec = (TP.time_since_epoch() % seconds(1)).count();
+ return RetVal;
+}
+
+/// Convert a time point to struct timeval.
+inline struct timeval toTimeVal(TimePoint<std::chrono::microseconds> TP) {
+ using namespace std::chrono;
+
+ struct timeval RetVal;
+ RetVal.tv_sec = toTimeT(TP);
+ RetVal.tv_usec = (TP.time_since_epoch() % seconds(1)).count();
+ return RetVal;
+}
+
+} // namespace sys
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/WithColor.cpp b/third_party/llvm-project/WithColor.cpp
new file mode 100644
index 000000000..9c5cbf462
--- /dev/null
+++ b/third_party/llvm-project/WithColor.cpp
@@ -0,0 +1,83 @@
+//===- WithColor.cpp ------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/WithColor.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+#if 0 // XXX BINARYEN
+cl::OptionCategory llvm::ColorCategory("Color Options");
+
+static cl::opt<cl::boolOrDefault>
+ UseColor("color", cl::cat(ColorCategory),
+ cl::desc("Use colors in output (default=autodetect)"),
+ cl::init(cl::BOU_UNSET));
+#endif
+
+WithColor::WithColor(raw_ostream &OS, HighlightColor Color, bool DisableColors)
+ : OS(OS), DisableColors(DisableColors) {
+ // Detect color from terminal type unless the user passed the --color option.
+ // XXX BINARYEN - no color support
+}
+
+raw_ostream &WithColor::error() { return error(errs()); }
+
+raw_ostream &WithColor::warning() { return warning(errs()); }
+
+raw_ostream &WithColor::note() { return note(errs()); }
+
+raw_ostream &WithColor::remark() { return remark(errs()); }
+
+raw_ostream &WithColor::error(raw_ostream &OS, StringRef Prefix,
+ bool DisableColors) {
+ if (!Prefix.empty())
+ OS << Prefix << ": ";
+ return WithColor(OS, HighlightColor::Error, DisableColors).get()
+ << "error: ";
+}
+
+raw_ostream &WithColor::warning(raw_ostream &OS, StringRef Prefix,
+ bool DisableColors) {
+ if (!Prefix.empty())
+ OS << Prefix << ": ";
+ return WithColor(OS, HighlightColor::Warning, DisableColors).get()
+ << "warning: ";
+}
+
+raw_ostream &WithColor::note(raw_ostream &OS, StringRef Prefix,
+ bool DisableColors) {
+ if (!Prefix.empty())
+ OS << Prefix << ": ";
+ return WithColor(OS, HighlightColor::Note, DisableColors).get() << "note: ";
+}
+
+raw_ostream &WithColor::remark(raw_ostream &OS, StringRef Prefix,
+ bool DisableColors) {
+ if (!Prefix.empty())
+ OS << Prefix << ": ";
+ return WithColor(OS, HighlightColor::Remark, DisableColors).get()
+ << "remark: ";
+}
+
+bool WithColor::colorsEnabled() {
+ return false; // XXX BINARYEN
+}
+
+WithColor &WithColor::changeColor(raw_ostream::Colors Color, bool Bold,
+ bool BG) {
+ // XXX BINARYEN
+ return *this;
+}
+
+WithColor &WithColor::resetColor() {
+ // XXX BINARYEN
+ return *this;
+}
+
+WithColor::~WithColor() { resetColor(); }
diff --git a/third_party/llvm-project/YAMLParser.cpp b/third_party/llvm-project/YAMLParser.cpp
new file mode 100644
index 000000000..1b85eba07
--- /dev/null
+++ b/third_party/llvm-project/YAMLParser.cpp
@@ -0,0 +1,2408 @@
+//===- YAMLParser.cpp - Simple YAML parser --------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a YAML parser.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/YAMLParser.h"
+#include "llvm/ADT/AllocatorList.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/SMLoc.h"
+#include "llvm/Support/SourceMgr.h"
+#include "llvm/Support/Unicode.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <map>
+#include <memory>
+#include <string>
+#include <system_error>
+#include <utility>
+
+using namespace llvm;
+using namespace yaml;
+
+enum UnicodeEncodingForm {
+ UEF_UTF32_LE, ///< UTF-32 Little Endian
+ UEF_UTF32_BE, ///< UTF-32 Big Endian
+ UEF_UTF16_LE, ///< UTF-16 Little Endian
+ UEF_UTF16_BE, ///< UTF-16 Big Endian
+ UEF_UTF8, ///< UTF-8 or ascii.
+ UEF_Unknown ///< Not a valid Unicode encoding.
+};
+
+/// EncodingInfo - Holds the encoding type and length of the byte order mark if
+/// it exists. Length is in {0, 2, 3, 4}.
+using EncodingInfo = std::pair<UnicodeEncodingForm, unsigned>;
+
+/// getUnicodeEncoding - Reads up to the first 4 bytes to determine the Unicode
+/// encoding form of \a Input.
+///
+/// @param Input A string of length 0 or more.
+/// @returns An EncodingInfo indicating the Unicode encoding form of the input
+/// and how long the byte order mark is if one exists.
+static EncodingInfo getUnicodeEncoding(StringRef Input) {
+ if (Input.empty())
+ return std::make_pair(UEF_Unknown, 0);
+
+ switch (uint8_t(Input[0])) {
+ case 0x00:
+ if (Input.size() >= 4) {
+ if ( Input[1] == 0
+ && uint8_t(Input[2]) == 0xFE
+ && uint8_t(Input[3]) == 0xFF)
+ return std::make_pair(UEF_UTF32_BE, 4);
+ if (Input[1] == 0 && Input[2] == 0 && Input[3] != 0)
+ return std::make_pair(UEF_UTF32_BE, 0);
+ }
+
+ if (Input.size() >= 2 && Input[1] != 0)
+ return std::make_pair(UEF_UTF16_BE, 0);
+ return std::make_pair(UEF_Unknown, 0);
+ case 0xFF:
+ if ( Input.size() >= 4
+ && uint8_t(Input[1]) == 0xFE
+ && Input[2] == 0
+ && Input[3] == 0)
+ return std::make_pair(UEF_UTF32_LE, 4);
+
+ if (Input.size() >= 2 && uint8_t(Input[1]) == 0xFE)
+ return std::make_pair(UEF_UTF16_LE, 2);
+ return std::make_pair(UEF_Unknown, 0);
+ case 0xFE:
+ if (Input.size() >= 2 && uint8_t(Input[1]) == 0xFF)
+ return std::make_pair(UEF_UTF16_BE, 2);
+ return std::make_pair(UEF_Unknown, 0);
+ case 0xEF:
+ if ( Input.size() >= 3
+ && uint8_t(Input[1]) == 0xBB
+ && uint8_t(Input[2]) == 0xBF)
+ return std::make_pair(UEF_UTF8, 3);
+ return std::make_pair(UEF_Unknown, 0);
+ }
+
+ // It could still be utf-32 or utf-16.
+ if (Input.size() >= 4 && Input[1] == 0 && Input[2] == 0 && Input[3] == 0)
+ return std::make_pair(UEF_UTF32_LE, 0);
+
+ if (Input.size() >= 2 && Input[1] == 0)
+ return std::make_pair(UEF_UTF16_LE, 0);
+
+ return std::make_pair(UEF_UTF8, 0);
+}
+
+/// Pin the vtables to this file.
+void Node::anchor() {}
+void NullNode::anchor() {}
+void ScalarNode::anchor() {}
+void BlockScalarNode::anchor() {}
+void KeyValueNode::anchor() {}
+void MappingNode::anchor() {}
+void SequenceNode::anchor() {}
+void AliasNode::anchor() {}
+
+namespace llvm {
+namespace yaml {
+
+/// Token - A single YAML token.
+struct Token {
+ enum TokenKind {
+ TK_Error, // Uninitialized token.
+ TK_StreamStart,
+ TK_StreamEnd,
+ TK_VersionDirective,
+ TK_TagDirective,
+ TK_DocumentStart,
+ TK_DocumentEnd,
+ TK_BlockEntry,
+ TK_BlockEnd,
+ TK_BlockSequenceStart,
+ TK_BlockMappingStart,
+ TK_FlowEntry,
+ TK_FlowSequenceStart,
+ TK_FlowSequenceEnd,
+ TK_FlowMappingStart,
+ TK_FlowMappingEnd,
+ TK_Key,
+ TK_Value,
+ TK_Scalar,
+ TK_BlockScalar,
+ TK_Alias,
+ TK_Anchor,
+ TK_Tag
+ } Kind = TK_Error;
+
+ /// A string of length 0 or more whose begin() points to the logical location
+ /// of the token in the input.
+ StringRef Range;
+
+ /// The value of a block scalar node.
+ std::string Value;
+
+ Token() = default;
+};
+
+} // end namespace yaml
+} // end namespace llvm
+
+using TokenQueueT = BumpPtrList<Token>;
+
+namespace {
+
+/// This struct is used to track simple keys.
+///
+/// Simple keys are handled by creating an entry in SimpleKeys for each Token
+/// which could legally be the start of a simple key. When peekNext is called,
+/// if the Token To be returned is referenced by a SimpleKey, we continue
+/// tokenizing until that potential simple key has either been found to not be
+/// a simple key (we moved on to the next line or went further than 1024 chars).
+/// Or when we run into a Value, and then insert a Key token (and possibly
+/// others) before the SimpleKey's Tok.
+struct SimpleKey {
+ TokenQueueT::iterator Tok;
+ unsigned Column = 0;
+ unsigned Line = 0;
+ unsigned FlowLevel = 0;
+ bool IsRequired = false;
+
+ bool operator ==(const SimpleKey &Other) {
+ return Tok == Other.Tok;
+ }
+};
+
+} // end anonymous namespace
+
+/// The Unicode scalar value of a UTF-8 minimal well-formed code unit
+/// subsequence and the subsequence's length in code units (uint8_t).
+/// A length of 0 represents an error.
+using UTF8Decoded = std::pair<uint32_t, unsigned>;
+
+static UTF8Decoded decodeUTF8(StringRef Range) {
+ StringRef::iterator Position= Range.begin();
+ StringRef::iterator End = Range.end();
+ // 1 byte: [0x00, 0x7f]
+ // Bit pattern: 0xxxxxxx
+ if ((*Position & 0x80) == 0) {
+ return std::make_pair(*Position, 1);
+ }
+ // 2 bytes: [0x80, 0x7ff]
+ // Bit pattern: 110xxxxx 10xxxxxx
+ if (Position + 1 != End &&
+ ((*Position & 0xE0) == 0xC0) &&
+ ((*(Position + 1) & 0xC0) == 0x80)) {
+ uint32_t codepoint = ((*Position & 0x1F) << 6) |
+ (*(Position + 1) & 0x3F);
+ if (codepoint >= 0x80)
+ return std::make_pair(codepoint, 2);
+ }
+ // 3 bytes: [0x8000, 0xffff]
+ // Bit pattern: 1110xxxx 10xxxxxx 10xxxxxx
+ if (Position + 2 != End &&
+ ((*Position & 0xF0) == 0xE0) &&
+ ((*(Position + 1) & 0xC0) == 0x80) &&
+ ((*(Position + 2) & 0xC0) == 0x80)) {
+ uint32_t codepoint = ((*Position & 0x0F) << 12) |
+ ((*(Position + 1) & 0x3F) << 6) |
+ (*(Position + 2) & 0x3F);
+ // Codepoints between 0xD800 and 0xDFFF are invalid, as
+ // they are high / low surrogate halves used by UTF-16.
+ if (codepoint >= 0x800 &&
+ (codepoint < 0xD800 || codepoint > 0xDFFF))
+ return std::make_pair(codepoint, 3);
+ }
+ // 4 bytes: [0x10000, 0x10FFFF]
+ // Bit pattern: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
+ if (Position + 3 != End &&
+ ((*Position & 0xF8) == 0xF0) &&
+ ((*(Position + 1) & 0xC0) == 0x80) &&
+ ((*(Position + 2) & 0xC0) == 0x80) &&
+ ((*(Position + 3) & 0xC0) == 0x80)) {
+ uint32_t codepoint = ((*Position & 0x07) << 18) |
+ ((*(Position + 1) & 0x3F) << 12) |
+ ((*(Position + 2) & 0x3F) << 6) |
+ (*(Position + 3) & 0x3F);
+ if (codepoint >= 0x10000 && codepoint <= 0x10FFFF)
+ return std::make_pair(codepoint, 4);
+ }
+ return std::make_pair(0, 0);
+}
+
+namespace llvm {
+namespace yaml {
+
+/// Scans YAML tokens from a MemoryBuffer.
+class Scanner {
+public:
+ Scanner(StringRef Input, SourceMgr &SM, bool ShowColors = true,
+ std::error_code *EC = nullptr);
+ Scanner(MemoryBufferRef Buffer, SourceMgr &SM_, bool ShowColors = true,
+ std::error_code *EC = nullptr);
+
+ /// Parse the next token and return it without popping it.
+ Token &peekNext();
+
+ /// Parse the next token and pop it from the queue.
+ Token getNext();
+
+ void printError(SMLoc Loc, SourceMgr::DiagKind Kind, const Twine &Message,
+ ArrayRef<SMRange> Ranges = None) {
+ SM.PrintMessage(Loc, Kind, Message, Ranges, /* FixIts= */ None, ShowColors);
+ }
+
+ void setError(const Twine &Message, StringRef::iterator Position) {
+ if (Current >= End)
+ Current = End - 1;
+
+ // propagate the error if possible
+ if (EC)
+ *EC = make_error_code(std::errc::invalid_argument);
+
+ // Don't print out more errors after the first one we encounter. The rest
+ // are just the result of the first, and have no meaning.
+ if (!Failed)
+ printError(SMLoc::getFromPointer(Current), SourceMgr::DK_Error, Message);
+ Failed = true;
+ }
+
+ void setError(const Twine &Message) {
+ setError(Message, Current);
+ }
+
+ /// Returns true if an error occurred while parsing.
+ bool failed() {
+ return Failed;
+ }
+
+private:
+ void init(MemoryBufferRef Buffer);
+
+ StringRef currentInput() {
+ return StringRef(Current, End - Current);
+ }
+
+ /// Decode a UTF-8 minimal well-formed code unit subsequence starting
+ /// at \a Position.
+ ///
+ /// If the UTF-8 code units starting at Position do not form a well-formed
+ /// code unit subsequence, then the Unicode scalar value is 0, and the length
+ /// is 0.
+ UTF8Decoded decodeUTF8(StringRef::iterator Position) {
+ return ::decodeUTF8(StringRef(Position, End - Position));
+ }
+
+ // The following functions are based on the gramar rules in the YAML spec. The
+ // style of the function names it meant to closely match how they are written
+ // in the spec. The number within the [] is the number of the grammar rule in
+ // the spec.
+ //
+ // See 4.2 [Production Naming Conventions] for the meaning of the prefixes.
+ //
+ // c-
+ // A production starting and ending with a special character.
+ // b-
+ // A production matching a single line break.
+ // nb-
+ // A production starting and ending with a non-break character.
+ // s-
+ // A production starting and ending with a white space character.
+ // ns-
+ // A production starting and ending with a non-space character.
+ // l-
+ // A production matching complete line(s).
+
+ /// Skip a single nb-char[27] starting at Position.
+ ///
+ /// A nb-char is 0x9 | [0x20-0x7E] | 0x85 | [0xA0-0xD7FF] | [0xE000-0xFEFE]
+ /// | [0xFF00-0xFFFD] | [0x10000-0x10FFFF]
+ ///
+ /// @returns The code unit after the nb-char, or Position if it's not an
+ /// nb-char.
+ StringRef::iterator skip_nb_char(StringRef::iterator Position);
+
+ /// Skip a single b-break[28] starting at Position.
+ ///
+ /// A b-break is 0xD 0xA | 0xD | 0xA
+ ///
+ /// @returns The code unit after the b-break, or Position if it's not a
+ /// b-break.
+ StringRef::iterator skip_b_break(StringRef::iterator Position);
+
+ /// Skip a single s-space[31] starting at Position.
+ ///
+ /// An s-space is 0x20
+ ///
+ /// @returns The code unit after the s-space, or Position if it's not a
+ /// s-space.
+ StringRef::iterator skip_s_space(StringRef::iterator Position);
+
+ /// Skip a single s-white[33] starting at Position.
+ ///
+ /// A s-white is 0x20 | 0x9
+ ///
+ /// @returns The code unit after the s-white, or Position if it's not a
+ /// s-white.
+ StringRef::iterator skip_s_white(StringRef::iterator Position);
+
+ /// Skip a single ns-char[34] starting at Position.
+ ///
+ /// A ns-char is nb-char - s-white
+ ///
+ /// @returns The code unit after the ns-char, or Position if it's not a
+ /// ns-char.
+ StringRef::iterator skip_ns_char(StringRef::iterator Position);
+
+ using SkipWhileFunc = StringRef::iterator (Scanner::*)(StringRef::iterator);
+
+ /// Skip minimal well-formed code unit subsequences until Func
+ /// returns its input.
+ ///
+ /// @returns The code unit after the last minimal well-formed code unit
+ /// subsequence that Func accepted.
+ StringRef::iterator skip_while( SkipWhileFunc Func
+ , StringRef::iterator Position);
+
+ /// Skip minimal well-formed code unit subsequences until Func returns its
+ /// input.
+ void advanceWhile(SkipWhileFunc Func);
+
+ /// Scan ns-uri-char[39]s starting at Cur.
+ ///
+ /// This updates Cur and Column while scanning.
+ void scan_ns_uri_char();
+
+ /// Consume a minimal well-formed code unit subsequence starting at
+ /// \a Cur. Return false if it is not the same Unicode scalar value as
+ /// \a Expected. This updates \a Column.
+ bool consume(uint32_t Expected);
+
+ /// Skip \a Distance UTF-8 code units. Updates \a Cur and \a Column.
+ void skip(uint32_t Distance);
+
+ /// Return true if the minimal well-formed code unit subsequence at
+ /// Pos is whitespace or a new line
+ bool isBlankOrBreak(StringRef::iterator Position);
+
+ /// Consume a single b-break[28] if it's present at the current position.
+ ///
+ /// Return false if the code unit at the current position isn't a line break.
+ bool consumeLineBreakIfPresent();
+
+ /// If IsSimpleKeyAllowed, create and push_back a new SimpleKey.
+ void saveSimpleKeyCandidate( TokenQueueT::iterator Tok
+ , unsigned AtColumn
+ , bool IsRequired);
+
+ /// Remove simple keys that can no longer be valid simple keys.
+ ///
+ /// Invalid simple keys are not on the current line or are further than 1024
+ /// columns back.
+ void removeStaleSimpleKeyCandidates();
+
+ /// Remove all simple keys on FlowLevel \a Level.
+ void removeSimpleKeyCandidatesOnFlowLevel(unsigned Level);
+
+ /// Unroll indentation in \a Indents back to \a Col. Creates BlockEnd
+ /// tokens if needed.
+ bool unrollIndent(int ToColumn);
+
+ /// Increase indent to \a Col. Creates \a Kind token at \a InsertPoint
+ /// if needed.
+ bool rollIndent( int ToColumn
+ , Token::TokenKind Kind
+ , TokenQueueT::iterator InsertPoint);
+
+ /// Skip a single-line comment when the comment starts at the current
+ /// position of the scanner.
+ void skipComment();
+
+ /// Skip whitespace and comments until the start of the next token.
+ void scanToNextToken();
+
+ /// Must be the first token generated.
+ bool scanStreamStart();
+
+ /// Generate tokens needed to close out the stream.
+ bool scanStreamEnd();
+
+ /// Scan a %BLAH directive.
+ bool scanDirective();
+
+ /// Scan a ... or ---.
+ bool scanDocumentIndicator(bool IsStart);
+
+ /// Scan a [ or { and generate the proper flow collection start token.
+ bool scanFlowCollectionStart(bool IsSequence);
+
+ /// Scan a ] or } and generate the proper flow collection end token.
+ bool scanFlowCollectionEnd(bool IsSequence);
+
+ /// Scan the , that separates entries in a flow collection.
+ bool scanFlowEntry();
+
+ /// Scan the - that starts block sequence entries.
+ bool scanBlockEntry();
+
+ /// Scan an explicit ? indicating a key.
+ bool scanKey();
+
+ /// Scan an explicit : indicating a value.
+ bool scanValue();
+
+ /// Scan a quoted scalar.
+ bool scanFlowScalar(bool IsDoubleQuoted);
+
+ /// Scan an unquoted scalar.
+ bool scanPlainScalar();
+
+ /// Scan an Alias or Anchor starting with * or &.
+ bool scanAliasOrAnchor(bool IsAlias);
+
+ /// Scan a block scalar starting with | or >.
+ bool scanBlockScalar(bool IsLiteral);
+
+ /// Scan a chomping indicator in a block scalar header.
+ char scanBlockChompingIndicator();
+
+ /// Scan an indentation indicator in a block scalar header.
+ unsigned scanBlockIndentationIndicator();
+
+ /// Scan a block scalar header.
+ ///
+ /// Return false if an error occurred.
+ bool scanBlockScalarHeader(char &ChompingIndicator, unsigned &IndentIndicator,
+ bool &IsDone);
+
+ /// Look for the indentation level of a block scalar.
+ ///
+ /// Return false if an error occurred.
+ bool findBlockScalarIndent(unsigned &BlockIndent, unsigned BlockExitIndent,
+ unsigned &LineBreaks, bool &IsDone);
+
+ /// Scan the indentation of a text line in a block scalar.
+ ///
+ /// Return false if an error occurred.
+ bool scanBlockScalarIndent(unsigned BlockIndent, unsigned BlockExitIndent,
+ bool &IsDone);
+
+ /// Scan a tag of the form !stuff.
+ bool scanTag();
+
+ /// Dispatch to the next scanning function based on \a *Cur.
+ bool fetchMoreTokens();
+
+ /// The SourceMgr used for diagnostics and buffer management.
+ SourceMgr &SM;
+
+ /// The original input.
+ MemoryBufferRef InputBuffer;
+
+ /// The current position of the scanner.
+ StringRef::iterator Current;
+
+ /// The end of the input (one past the last character).
+ StringRef::iterator End;
+
+ /// Current YAML indentation level in spaces.
+ int Indent;
+
+ /// Current column number in Unicode code points.
+ unsigned Column;
+
+ /// Current line number.
+ unsigned Line;
+
+ /// How deep we are in flow style containers. 0 Means at block level.
+ unsigned FlowLevel;
+
+ /// Are we at the start of the stream?
+ bool IsStartOfStream;
+
+ /// Can the next token be the start of a simple key?
+ bool IsSimpleKeyAllowed;
+
+ /// True if an error has occurred.
+ bool Failed;
+
+ /// Should colors be used when printing out the diagnostic messages?
+ bool ShowColors;
+
+ /// Queue of tokens. This is required to queue up tokens while looking
+ /// for the end of a simple key. And for cases where a single character
+ /// can produce multiple tokens (e.g. BlockEnd).
+ TokenQueueT TokenQueue;
+
+ /// Indentation levels.
+ SmallVector<int, 4> Indents;
+
+ /// Potential simple keys.
+ SmallVector<SimpleKey, 4> SimpleKeys;
+
+ std::error_code *EC;
+};
+
+} // end namespace yaml
+} // end namespace llvm
+
+/// encodeUTF8 - Encode \a UnicodeScalarValue in UTF-8 and append it to result.
+static void encodeUTF8( uint32_t UnicodeScalarValue
+ , SmallVectorImpl<char> &Result) {
+ if (UnicodeScalarValue <= 0x7F) {
+ Result.push_back(UnicodeScalarValue & 0x7F);
+ } else if (UnicodeScalarValue <= 0x7FF) {
+ uint8_t FirstByte = 0xC0 | ((UnicodeScalarValue & 0x7C0) >> 6);
+ uint8_t SecondByte = 0x80 | (UnicodeScalarValue & 0x3F);
+ Result.push_back(FirstByte);
+ Result.push_back(SecondByte);
+ } else if (UnicodeScalarValue <= 0xFFFF) {
+ uint8_t FirstByte = 0xE0 | ((UnicodeScalarValue & 0xF000) >> 12);
+ uint8_t SecondByte = 0x80 | ((UnicodeScalarValue & 0xFC0) >> 6);
+ uint8_t ThirdByte = 0x80 | (UnicodeScalarValue & 0x3F);
+ Result.push_back(FirstByte);
+ Result.push_back(SecondByte);
+ Result.push_back(ThirdByte);
+ } else if (UnicodeScalarValue <= 0x10FFFF) {
+ uint8_t FirstByte = 0xF0 | ((UnicodeScalarValue & 0x1F0000) >> 18);
+ uint8_t SecondByte = 0x80 | ((UnicodeScalarValue & 0x3F000) >> 12);
+ uint8_t ThirdByte = 0x80 | ((UnicodeScalarValue & 0xFC0) >> 6);
+ uint8_t FourthByte = 0x80 | (UnicodeScalarValue & 0x3F);
+ Result.push_back(FirstByte);
+ Result.push_back(SecondByte);
+ Result.push_back(ThirdByte);
+ Result.push_back(FourthByte);
+ }
+}
+
+bool yaml::dumpTokens(StringRef Input, raw_ostream &OS) {
+ SourceMgr SM;
+ Scanner scanner(Input, SM);
+ while (true) {
+ Token T = scanner.getNext();
+ switch (T.Kind) {
+ case Token::TK_StreamStart:
+ OS << "Stream-Start: ";
+ break;
+ case Token::TK_StreamEnd:
+ OS << "Stream-End: ";
+ break;
+ case Token::TK_VersionDirective:
+ OS << "Version-Directive: ";
+ break;
+ case Token::TK_TagDirective:
+ OS << "Tag-Directive: ";
+ break;
+ case Token::TK_DocumentStart:
+ OS << "Document-Start: ";
+ break;
+ case Token::TK_DocumentEnd:
+ OS << "Document-End: ";
+ break;
+ case Token::TK_BlockEntry:
+ OS << "Block-Entry: ";
+ break;
+ case Token::TK_BlockEnd:
+ OS << "Block-End: ";
+ break;
+ case Token::TK_BlockSequenceStart:
+ OS << "Block-Sequence-Start: ";
+ break;
+ case Token::TK_BlockMappingStart:
+ OS << "Block-Mapping-Start: ";
+ break;
+ case Token::TK_FlowEntry:
+ OS << "Flow-Entry: ";
+ break;
+ case Token::TK_FlowSequenceStart:
+ OS << "Flow-Sequence-Start: ";
+ break;
+ case Token::TK_FlowSequenceEnd:
+ OS << "Flow-Sequence-End: ";
+ break;
+ case Token::TK_FlowMappingStart:
+ OS << "Flow-Mapping-Start: ";
+ break;
+ case Token::TK_FlowMappingEnd:
+ OS << "Flow-Mapping-End: ";
+ break;
+ case Token::TK_Key:
+ OS << "Key: ";
+ break;
+ case Token::TK_Value:
+ OS << "Value: ";
+ break;
+ case Token::TK_Scalar:
+ OS << "Scalar: ";
+ break;
+ case Token::TK_BlockScalar:
+ OS << "Block Scalar: ";
+ break;
+ case Token::TK_Alias:
+ OS << "Alias: ";
+ break;
+ case Token::TK_Anchor:
+ OS << "Anchor: ";
+ break;
+ case Token::TK_Tag:
+ OS << "Tag: ";
+ break;
+ case Token::TK_Error:
+ break;
+ }
+ OS << T.Range << "\n";
+ if (T.Kind == Token::TK_StreamEnd)
+ break;
+ else if (T.Kind == Token::TK_Error)
+ return false;
+ }
+ return true;
+}
+
+bool yaml::scanTokens(StringRef Input) {
+ SourceMgr SM;
+ Scanner scanner(Input, SM);
+ while (true) {
+ Token T = scanner.getNext();
+ if (T.Kind == Token::TK_StreamEnd)
+ break;
+ else if (T.Kind == Token::TK_Error)
+ return false;
+ }
+ return true;
+}
+
+std::string yaml::escape(StringRef Input, bool EscapePrintable) {
+ llvm_unreachable("BYN yaml::escape");
+}
+
+Scanner::Scanner(StringRef Input, SourceMgr &sm, bool ShowColors,
+ std::error_code *EC)
+ : SM(sm), ShowColors(ShowColors), EC(EC) {
+ init(MemoryBufferRef(Input, "YAML"));
+}
+
+Scanner::Scanner(MemoryBufferRef Buffer, SourceMgr &SM_, bool ShowColors,
+ std::error_code *EC)
+ : SM(SM_), ShowColors(ShowColors), EC(EC) {
+ init(Buffer);
+}
+
+void Scanner::init(MemoryBufferRef Buffer) {
+ InputBuffer = Buffer;
+ Current = InputBuffer.getBufferStart();
+ End = InputBuffer.getBufferEnd();
+ Indent = -1;
+ Column = 0;
+ Line = 0;
+ FlowLevel = 0;
+ IsStartOfStream = true;
+ IsSimpleKeyAllowed = true;
+ Failed = false;
+ std::unique_ptr<MemoryBuffer> InputBufferOwner =
+ MemoryBuffer::getMemBuffer(Buffer);
+ SM.AddNewSourceBuffer(std::move(InputBufferOwner), SMLoc());
+}
+
+Token &Scanner::peekNext() {
+ // If the current token is a possible simple key, keep parsing until we
+ // can confirm.
+ bool NeedMore = false;
+ while (true) {
+ if (TokenQueue.empty() || NeedMore) {
+ if (!fetchMoreTokens()) {
+ TokenQueue.clear();
+ SimpleKeys.clear();
+ TokenQueue.push_back(Token());
+ return TokenQueue.front();
+ }
+ }
+ assert(!TokenQueue.empty() &&
+ "fetchMoreTokens lied about getting tokens!");
+
+ removeStaleSimpleKeyCandidates();
+ SimpleKey SK;
+ SK.Tok = TokenQueue.begin();
+ if (!is_contained(SimpleKeys, SK))
+ break;
+ else
+ NeedMore = true;
+ }
+ return TokenQueue.front();
+}
+
+Token Scanner::getNext() {
+ Token Ret = peekNext();
+ // TokenQueue can be empty if there was an error getting the next token.
+ if (!TokenQueue.empty())
+ TokenQueue.pop_front();
+
+ // There cannot be any referenced Token's if the TokenQueue is empty. So do a
+ // quick deallocation of them all.
+ if (TokenQueue.empty())
+ TokenQueue.resetAlloc();
+
+ return Ret;
+}
+
+StringRef::iterator Scanner::skip_nb_char(StringRef::iterator Position) {
+ if (Position == End)
+ return Position;
+ // Check 7 bit c-printable - b-char.
+ if ( *Position == 0x09
+ || (*Position >= 0x20 && *Position <= 0x7E))
+ return Position + 1;
+
+ // Check for valid UTF-8.
+ if (uint8_t(*Position) & 0x80) {
+ UTF8Decoded u8d = decodeUTF8(Position);
+ if ( u8d.second != 0
+ && u8d.first != 0xFEFF
+ && ( u8d.first == 0x85
+ || ( u8d.first >= 0xA0
+ && u8d.first <= 0xD7FF)
+ || ( u8d.first >= 0xE000
+ && u8d.first <= 0xFFFD)
+ || ( u8d.first >= 0x10000
+ && u8d.first <= 0x10FFFF)))
+ return Position + u8d.second;
+ }
+ return Position;
+}
+
+StringRef::iterator Scanner::skip_b_break(StringRef::iterator Position) {
+ if (Position == End)
+ return Position;
+ if (*Position == 0x0D) {
+ if (Position + 1 != End && *(Position + 1) == 0x0A)
+ return Position + 2;
+ return Position + 1;
+ }
+
+ if (*Position == 0x0A)
+ return Position + 1;
+ return Position;
+}
+
+StringRef::iterator Scanner::skip_s_space(StringRef::iterator Position) {
+ if (Position == End)
+ return Position;
+ if (*Position == ' ')
+ return Position + 1;
+ return Position;
+}
+
+StringRef::iterator Scanner::skip_s_white(StringRef::iterator Position) {
+ if (Position == End)
+ return Position;
+ if (*Position == ' ' || *Position == '\t')
+ return Position + 1;
+ return Position;
+}
+
+StringRef::iterator Scanner::skip_ns_char(StringRef::iterator Position) {
+ if (Position == End)
+ return Position;
+ if (*Position == ' ' || *Position == '\t')
+ return Position;
+ return skip_nb_char(Position);
+}
+
+StringRef::iterator Scanner::skip_while( SkipWhileFunc Func
+ , StringRef::iterator Position) {
+ while (true) {
+ StringRef::iterator i = (this->*Func)(Position);
+ if (i == Position)
+ break;
+ Position = i;
+ }
+ return Position;
+}
+
+void Scanner::advanceWhile(SkipWhileFunc Func) {
+ auto Final = skip_while(Func, Current);
+ Column += Final - Current;
+ Current = Final;
+}
+
+static bool is_ns_hex_digit(const char C) {
+ return (C >= '0' && C <= '9')
+ || (C >= 'a' && C <= 'z')
+ || (C >= 'A' && C <= 'Z');
+}
+
+static bool is_ns_word_char(const char C) {
+ return C == '-'
+ || (C >= 'a' && C <= 'z')
+ || (C >= 'A' && C <= 'Z');
+}
+
+void Scanner::scan_ns_uri_char() {
+ while (true) {
+ if (Current == End)
+ break;
+ if (( *Current == '%'
+ && Current + 2 < End
+ && is_ns_hex_digit(*(Current + 1))
+ && is_ns_hex_digit(*(Current + 2)))
+ || is_ns_word_char(*Current)
+ || StringRef(Current, 1).find_first_of("#;/?:@&=+$,_.!~*'()[]")
+ != StringRef::npos) {
+ ++Current;
+ ++Column;
+ } else
+ break;
+ }
+}
+
+bool Scanner::consume(uint32_t Expected) {
+ if (Expected >= 0x80) {
+ setError("Cannot consume non-ascii characters");
+ return false;
+ }
+ if (Current == End)
+ return false;
+ if (uint8_t(*Current) >= 0x80) {
+ setError("Cannot consume non-ascii characters");
+ return false;
+ }
+ if (uint8_t(*Current) == Expected) {
+ ++Current;
+ ++Column;
+ return true;
+ }
+ return false;
+}
+
+void Scanner::skip(uint32_t Distance) {
+ Current += Distance;
+ Column += Distance;
+ assert(Current <= End && "Skipped past the end");
+}
+
+bool Scanner::isBlankOrBreak(StringRef::iterator Position) {
+ if (Position == End)
+ return false;
+ return *Position == ' ' || *Position == '\t' || *Position == '\r' ||
+ *Position == '\n';
+}
+
+bool Scanner::consumeLineBreakIfPresent() {
+ auto Next = skip_b_break(Current);
+ if (Next == Current)
+ return false;
+ Column = 0;
+ ++Line;
+ Current = Next;
+ return true;
+}
+
+void Scanner::saveSimpleKeyCandidate( TokenQueueT::iterator Tok
+ , unsigned AtColumn
+ , bool IsRequired) {
+ if (IsSimpleKeyAllowed) {
+ SimpleKey SK;
+ SK.Tok = Tok;
+ SK.Line = Line;
+ SK.Column = AtColumn;
+ SK.IsRequired = IsRequired;
+ SK.FlowLevel = FlowLevel;
+ SimpleKeys.push_back(SK);
+ }
+}
+
+void Scanner::removeStaleSimpleKeyCandidates() {
+ for (SmallVectorImpl<SimpleKey>::iterator i = SimpleKeys.begin();
+ i != SimpleKeys.end();) {
+ if (i->Line != Line || i->Column + 1024 < Column) {
+ if (i->IsRequired)
+ setError( "Could not find expected : for simple key"
+ , i->Tok->Range.begin());
+ i = SimpleKeys.erase(i);
+ } else
+ ++i;
+ }
+}
+
+void Scanner::removeSimpleKeyCandidatesOnFlowLevel(unsigned Level) {
+ if (!SimpleKeys.empty() && (SimpleKeys.end() - 1)->FlowLevel == Level)
+ SimpleKeys.pop_back();
+}
+
+bool Scanner::unrollIndent(int ToColumn) {
+ Token T;
+ // Indentation is ignored in flow.
+ if (FlowLevel != 0)
+ return true;
+
+ while (Indent > ToColumn) {
+ T.Kind = Token::TK_BlockEnd;
+ T.Range = StringRef(Current, 1);
+ TokenQueue.push_back(T);
+ Indent = Indents.pop_back_val();
+ }
+
+ return true;
+}
+
+bool Scanner::rollIndent( int ToColumn
+ , Token::TokenKind Kind
+ , TokenQueueT::iterator InsertPoint) {
+ if (FlowLevel)
+ return true;
+ if (Indent < ToColumn) {
+ Indents.push_back(Indent);
+ Indent = ToColumn;
+
+ Token T;
+ T.Kind = Kind;
+ T.Range = StringRef(Current, 0);
+ TokenQueue.insert(InsertPoint, T);
+ }
+ return true;
+}
+
+void Scanner::skipComment() {
+ if (*Current != '#')
+ return;
+ while (true) {
+ // This may skip more than one byte, thus Column is only incremented
+ // for code points.
+ StringRef::iterator I = skip_nb_char(Current);
+ if (I == Current)
+ break;
+ Current = I;
+ ++Column;
+ }
+}
+
+void Scanner::scanToNextToken() {
+ while (true) {
+ while (*Current == ' ' || *Current == '\t') {
+ skip(1);
+ }
+
+ skipComment();
+
+ // Skip EOL.
+ StringRef::iterator i = skip_b_break(Current);
+ if (i == Current)
+ break;
+ Current = i;
+ ++Line;
+ Column = 0;
+ // New lines may start a simple key.
+ if (!FlowLevel)
+ IsSimpleKeyAllowed = true;
+ }
+}
+
+bool Scanner::scanStreamStart() {
+ IsStartOfStream = false;
+
+ EncodingInfo EI = getUnicodeEncoding(currentInput());
+
+ Token T;
+ T.Kind = Token::TK_StreamStart;
+ T.Range = StringRef(Current, EI.second);
+ TokenQueue.push_back(T);
+ Current += EI.second;
+ return true;
+}
+
+bool Scanner::scanStreamEnd() {
+ // Force an ending new line if one isn't present.
+ if (Column != 0) {
+ Column = 0;
+ ++Line;
+ }
+
+ unrollIndent(-1);
+ SimpleKeys.clear();
+ IsSimpleKeyAllowed = false;
+
+ Token T;
+ T.Kind = Token::TK_StreamEnd;
+ T.Range = StringRef(Current, 0);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanDirective() {
+ // Reset the indentation level.
+ unrollIndent(-1);
+ SimpleKeys.clear();
+ IsSimpleKeyAllowed = false;
+
+ StringRef::iterator Start = Current;
+ consume('%');
+ StringRef::iterator NameStart = Current;
+ Current = skip_while(&Scanner::skip_ns_char, Current);
+ StringRef Name(NameStart, Current - NameStart);
+ Current = skip_while(&Scanner::skip_s_white, Current);
+
+ Token T;
+ if (Name == "YAML") {
+ Current = skip_while(&Scanner::skip_ns_char, Current);
+ T.Kind = Token::TK_VersionDirective;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+ return true;
+ } else if(Name == "TAG") {
+ Current = skip_while(&Scanner::skip_ns_char, Current);
+ Current = skip_while(&Scanner::skip_s_white, Current);
+ Current = skip_while(&Scanner::skip_ns_char, Current);
+ T.Kind = Token::TK_TagDirective;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+ return true;
+ }
+ return false;
+}
+
+bool Scanner::scanDocumentIndicator(bool IsStart) {
+ unrollIndent(-1);
+ SimpleKeys.clear();
+ IsSimpleKeyAllowed = false;
+
+ Token T;
+ T.Kind = IsStart ? Token::TK_DocumentStart : Token::TK_DocumentEnd;
+ T.Range = StringRef(Current, 3);
+ skip(3);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanFlowCollectionStart(bool IsSequence) {
+ Token T;
+ T.Kind = IsSequence ? Token::TK_FlowSequenceStart
+ : Token::TK_FlowMappingStart;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+
+ // [ and { may begin a simple key.
+ saveSimpleKeyCandidate(--TokenQueue.end(), Column - 1, false);
+
+ // And may also be followed by a simple key.
+ IsSimpleKeyAllowed = true;
+ ++FlowLevel;
+ return true;
+}
+
+bool Scanner::scanFlowCollectionEnd(bool IsSequence) {
+ removeSimpleKeyCandidatesOnFlowLevel(FlowLevel);
+ IsSimpleKeyAllowed = false;
+ Token T;
+ T.Kind = IsSequence ? Token::TK_FlowSequenceEnd
+ : Token::TK_FlowMappingEnd;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+ if (FlowLevel)
+ --FlowLevel;
+ return true;
+}
+
+bool Scanner::scanFlowEntry() {
+ removeSimpleKeyCandidatesOnFlowLevel(FlowLevel);
+ IsSimpleKeyAllowed = true;
+ Token T;
+ T.Kind = Token::TK_FlowEntry;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanBlockEntry() {
+ rollIndent(Column, Token::TK_BlockSequenceStart, TokenQueue.end());
+ removeSimpleKeyCandidatesOnFlowLevel(FlowLevel);
+ IsSimpleKeyAllowed = true;
+ Token T;
+ T.Kind = Token::TK_BlockEntry;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanKey() {
+ if (!FlowLevel)
+ rollIndent(Column, Token::TK_BlockMappingStart, TokenQueue.end());
+
+ removeSimpleKeyCandidatesOnFlowLevel(FlowLevel);
+ IsSimpleKeyAllowed = !FlowLevel;
+
+ Token T;
+ T.Kind = Token::TK_Key;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanValue() {
+ // If the previous token could have been a simple key, insert the key token
+ // into the token queue.
+ if (!SimpleKeys.empty()) {
+ SimpleKey SK = SimpleKeys.pop_back_val();
+ Token T;
+ T.Kind = Token::TK_Key;
+ T.Range = SK.Tok->Range;
+ TokenQueueT::iterator i, e;
+ for (i = TokenQueue.begin(), e = TokenQueue.end(); i != e; ++i) {
+ if (i == SK.Tok)
+ break;
+ }
+ if (i == e) {
+ Failed = true;
+ return false;
+ }
+ i = TokenQueue.insert(i, T);
+
+ // We may also need to add a Block-Mapping-Start token.
+ rollIndent(SK.Column, Token::TK_BlockMappingStart, i);
+
+ IsSimpleKeyAllowed = false;
+ } else {
+ if (!FlowLevel)
+ rollIndent(Column, Token::TK_BlockMappingStart, TokenQueue.end());
+ IsSimpleKeyAllowed = !FlowLevel;
+ }
+
+ Token T;
+ T.Kind = Token::TK_Value;
+ T.Range = StringRef(Current, 1);
+ skip(1);
+ TokenQueue.push_back(T);
+ return true;
+}
+
+// Forbidding inlining improves performance by roughly 20%.
+// FIXME: Remove once llvm optimizes this to the faster version without hints.
+LLVM_ATTRIBUTE_NOINLINE static bool
+wasEscaped(StringRef::iterator First, StringRef::iterator Position);
+
+// Returns whether a character at 'Position' was escaped with a leading '\'.
+// 'First' specifies the position of the first character in the string.
+static bool wasEscaped(StringRef::iterator First,
+ StringRef::iterator Position) {
+ assert(Position - 1 >= First);
+ StringRef::iterator I = Position - 1;
+ // We calculate the number of consecutive '\'s before the current position
+ // by iterating backwards through our string.
+ while (I >= First && *I == '\\') --I;
+ // (Position - 1 - I) now contains the number of '\'s before the current
+ // position. If it is odd, the character at 'Position' was escaped.
+ return (Position - 1 - I) % 2 == 1;
+}
+
+bool Scanner::scanFlowScalar(bool IsDoubleQuoted) {
+ StringRef::iterator Start = Current;
+ unsigned ColStart = Column;
+ if (IsDoubleQuoted) {
+ do {
+ ++Current;
+ while (Current != End && *Current != '"')
+ ++Current;
+ // Repeat until the previous character was not a '\' or was an escaped
+ // backslash.
+ } while ( Current != End
+ && *(Current - 1) == '\\'
+ && wasEscaped(Start + 1, Current));
+ } else {
+ skip(1);
+ while (true) {
+ // Skip a ' followed by another '.
+ if (Current + 1 < End && *Current == '\'' && *(Current + 1) == '\'') {
+ skip(2);
+ continue;
+ } else if (*Current == '\'')
+ break;
+ StringRef::iterator i = skip_nb_char(Current);
+ if (i == Current) {
+ i = skip_b_break(Current);
+ if (i == Current)
+ break;
+ Current = i;
+ Column = 0;
+ ++Line;
+ } else {
+ if (i == End)
+ break;
+ Current = i;
+ ++Column;
+ }
+ }
+ }
+
+ if (Current == End) {
+ setError("Expected quote at end of scalar", Current);
+ return false;
+ }
+
+ skip(1); // Skip ending quote.
+ Token T;
+ T.Kind = Token::TK_Scalar;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+
+ saveSimpleKeyCandidate(--TokenQueue.end(), ColStart, false);
+
+ IsSimpleKeyAllowed = false;
+
+ return true;
+}
+
+bool Scanner::scanPlainScalar() {
+ StringRef::iterator Start = Current;
+ unsigned ColStart = Column;
+ unsigned LeadingBlanks = 0;
+ assert(Indent >= -1 && "Indent must be >= -1 !");
+ unsigned indent = static_cast<unsigned>(Indent + 1);
+ while (true) {
+ if (*Current == '#')
+ break;
+
+ while (!isBlankOrBreak(Current)) {
+ if ( FlowLevel && *Current == ':'
+ && !(isBlankOrBreak(Current + 1) || *(Current + 1) == ',')) {
+ setError("Found unexpected ':' while scanning a plain scalar", Current);
+ return false;
+ }
+
+ // Check for the end of the plain scalar.
+ if ( (*Current == ':' && isBlankOrBreak(Current + 1))
+ || ( FlowLevel
+ && (StringRef(Current, 1).find_first_of(",:?[]{}")
+ != StringRef::npos)))
+ break;
+
+ StringRef::iterator i = skip_nb_char(Current);
+ if (i == Current)
+ break;
+ Current = i;
+ ++Column;
+ }
+
+ // Are we at the end?
+ if (!isBlankOrBreak(Current))
+ break;
+
+ // Eat blanks.
+ StringRef::iterator Tmp = Current;
+ while (isBlankOrBreak(Tmp)) {
+ StringRef::iterator i = skip_s_white(Tmp);
+ if (i != Tmp) {
+ if (LeadingBlanks && (Column < indent) && *Tmp == '\t') {
+ setError("Found invalid tab character in indentation", Tmp);
+ return false;
+ }
+ Tmp = i;
+ ++Column;
+ } else {
+ i = skip_b_break(Tmp);
+ if (!LeadingBlanks)
+ LeadingBlanks = 1;
+ Tmp = i;
+ Column = 0;
+ ++Line;
+ }
+ }
+
+ if (!FlowLevel && Column < indent)
+ break;
+
+ Current = Tmp;
+ }
+ if (Start == Current) {
+ setError("Got empty plain scalar", Start);
+ return false;
+ }
+ Token T;
+ T.Kind = Token::TK_Scalar;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+
+ // Plain scalars can be simple keys.
+ saveSimpleKeyCandidate(--TokenQueue.end(), ColStart, false);
+
+ IsSimpleKeyAllowed = false;
+
+ return true;
+}
+
+bool Scanner::scanAliasOrAnchor(bool IsAlias) {
+ StringRef::iterator Start = Current;
+ unsigned ColStart = Column;
+ skip(1);
+ while(true) {
+ if ( *Current == '[' || *Current == ']'
+ || *Current == '{' || *Current == '}'
+ || *Current == ','
+ || *Current == ':')
+ break;
+ StringRef::iterator i = skip_ns_char(Current);
+ if (i == Current)
+ break;
+ Current = i;
+ ++Column;
+ }
+
+ if (Start == Current) {
+ setError("Got empty alias or anchor", Start);
+ return false;
+ }
+
+ Token T;
+ T.Kind = IsAlias ? Token::TK_Alias : Token::TK_Anchor;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+
+ // Alias and anchors can be simple keys.
+ saveSimpleKeyCandidate(--TokenQueue.end(), ColStart, false);
+
+ IsSimpleKeyAllowed = false;
+
+ return true;
+}
+
+char Scanner::scanBlockChompingIndicator() {
+ char Indicator = ' ';
+ if (Current != End && (*Current == '+' || *Current == '-')) {
+ Indicator = *Current;
+ skip(1);
+ }
+ return Indicator;
+}
+
+/// Get the number of line breaks after chomping.
+///
+/// Return the number of trailing line breaks to emit, depending on
+/// \p ChompingIndicator.
+static unsigned getChompedLineBreaks(char ChompingIndicator,
+ unsigned LineBreaks, StringRef Str) {
+ if (ChompingIndicator == '-') // Strip all line breaks.
+ return 0;
+ if (ChompingIndicator == '+') // Keep all line breaks.
+ return LineBreaks;
+ // Clip trailing lines.
+ return Str.empty() ? 0 : 1;
+}
+
+unsigned Scanner::scanBlockIndentationIndicator() {
+ unsigned Indent = 0;
+ if (Current != End && (*Current >= '1' && *Current <= '9')) {
+ Indent = unsigned(*Current - '0');
+ skip(1);
+ }
+ return Indent;
+}
+
+bool Scanner::scanBlockScalarHeader(char &ChompingIndicator,
+ unsigned &IndentIndicator, bool &IsDone) {
+ auto Start = Current;
+
+ ChompingIndicator = scanBlockChompingIndicator();
+ IndentIndicator = scanBlockIndentationIndicator();
+ // Check for the chomping indicator once again.
+ if (ChompingIndicator == ' ')
+ ChompingIndicator = scanBlockChompingIndicator();
+ Current = skip_while(&Scanner::skip_s_white, Current);
+ skipComment();
+
+ if (Current == End) { // EOF, we have an empty scalar.
+ Token T;
+ T.Kind = Token::TK_BlockScalar;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+ IsDone = true;
+ return true;
+ }
+
+ if (!consumeLineBreakIfPresent()) {
+ setError("Expected a line break after block scalar header", Current);
+ return false;
+ }
+ return true;
+}
+
+bool Scanner::findBlockScalarIndent(unsigned &BlockIndent,
+ unsigned BlockExitIndent,
+ unsigned &LineBreaks, bool &IsDone) {
+ unsigned MaxAllSpaceLineCharacters = 0;
+ StringRef::iterator LongestAllSpaceLine;
+
+ while (true) {
+ advanceWhile(&Scanner::skip_s_space);
+ if (skip_nb_char(Current) != Current) {
+ // This line isn't empty, so try and find the indentation.
+ if (Column <= BlockExitIndent) { // End of the block literal.
+ IsDone = true;
+ return true;
+ }
+ // We found the block's indentation.
+ BlockIndent = Column;
+ if (MaxAllSpaceLineCharacters > BlockIndent) {
+ setError(
+ "Leading all-spaces line must be smaller than the block indent",
+ LongestAllSpaceLine);
+ return false;
+ }
+ return true;
+ }
+ if (skip_b_break(Current) != Current &&
+ Column > MaxAllSpaceLineCharacters) {
+ // Record the longest all-space line in case it's longer than the
+ // discovered block indent.
+ MaxAllSpaceLineCharacters = Column;
+ LongestAllSpaceLine = Current;
+ }
+
+ // Check for EOF.
+ if (Current == End) {
+ IsDone = true;
+ return true;
+ }
+
+ if (!consumeLineBreakIfPresent()) {
+ IsDone = true;
+ return true;
+ }
+ ++LineBreaks;
+ }
+ return true;
+}
+
+bool Scanner::scanBlockScalarIndent(unsigned BlockIndent,
+ unsigned BlockExitIndent, bool &IsDone) {
+ // Skip the indentation.
+ while (Column < BlockIndent) {
+ auto I = skip_s_space(Current);
+ if (I == Current)
+ break;
+ Current = I;
+ ++Column;
+ }
+
+ if (skip_nb_char(Current) == Current)
+ return true;
+
+ if (Column <= BlockExitIndent) { // End of the block literal.
+ IsDone = true;
+ return true;
+ }
+
+ if (Column < BlockIndent) {
+ if (Current != End && *Current == '#') { // Trailing comment.
+ IsDone = true;
+ return true;
+ }
+ setError("A text line is less indented than the block scalar", Current);
+ return false;
+ }
+ return true; // A normal text line.
+}
+
+bool Scanner::scanBlockScalar(bool IsLiteral) {
+ // Eat '|' or '>'
+ assert(*Current == '|' || *Current == '>');
+ skip(1);
+
+ char ChompingIndicator;
+ unsigned BlockIndent;
+ bool IsDone = false;
+ if (!scanBlockScalarHeader(ChompingIndicator, BlockIndent, IsDone))
+ return false;
+ if (IsDone)
+ return true;
+
+ auto Start = Current;
+ unsigned BlockExitIndent = Indent < 0 ? 0 : (unsigned)Indent;
+ unsigned LineBreaks = 0;
+ if (BlockIndent == 0) {
+ if (!findBlockScalarIndent(BlockIndent, BlockExitIndent, LineBreaks,
+ IsDone))
+ return false;
+ }
+
+ // Scan the block's scalars body.
+ SmallString<256> Str;
+ while (!IsDone) {
+ if (!scanBlockScalarIndent(BlockIndent, BlockExitIndent, IsDone))
+ return false;
+ if (IsDone)
+ break;
+
+ // Parse the current line.
+ auto LineStart = Current;
+ advanceWhile(&Scanner::skip_nb_char);
+ if (LineStart != Current) {
+ Str.append(LineBreaks, '\n');
+ Str.append(StringRef(LineStart, Current - LineStart));
+ LineBreaks = 0;
+ }
+
+ // Check for EOF.
+ if (Current == End)
+ break;
+
+ if (!consumeLineBreakIfPresent())
+ break;
+ ++LineBreaks;
+ }
+
+ if (Current == End && !LineBreaks)
+ // Ensure that there is at least one line break before the end of file.
+ LineBreaks = 1;
+ Str.append(getChompedLineBreaks(ChompingIndicator, LineBreaks, Str), '\n');
+
+ // New lines may start a simple key.
+ if (!FlowLevel)
+ IsSimpleKeyAllowed = true;
+
+ Token T;
+ T.Kind = Token::TK_BlockScalar;
+ T.Range = StringRef(Start, Current - Start);
+ T.Value = Str.str().str();
+ TokenQueue.push_back(T);
+ return true;
+}
+
+bool Scanner::scanTag() {
+ StringRef::iterator Start = Current;
+ unsigned ColStart = Column;
+ skip(1); // Eat !.
+ if (Current == End || isBlankOrBreak(Current)); // An empty tag.
+ else if (*Current == '<') {
+ skip(1);
+ scan_ns_uri_char();
+ if (!consume('>'))
+ return false;
+ } else {
+ // FIXME: Actually parse the c-ns-shorthand-tag rule.
+ Current = skip_while(&Scanner::skip_ns_char, Current);
+ }
+
+ Token T;
+ T.Kind = Token::TK_Tag;
+ T.Range = StringRef(Start, Current - Start);
+ TokenQueue.push_back(T);
+
+ // Tags can be simple keys.
+ saveSimpleKeyCandidate(--TokenQueue.end(), ColStart, false);
+
+ IsSimpleKeyAllowed = false;
+
+ return true;
+}
+
+bool Scanner::fetchMoreTokens() {
+ if (IsStartOfStream)
+ return scanStreamStart();
+
+ scanToNextToken();
+
+ if (Current == End)
+ return scanStreamEnd();
+
+ removeStaleSimpleKeyCandidates();
+
+ unrollIndent(Column);
+
+ if (Column == 0 && *Current == '%')
+ return scanDirective();
+
+ if (Column == 0 && Current + 4 <= End
+ && *Current == '-'
+ && *(Current + 1) == '-'
+ && *(Current + 2) == '-'
+ && (Current + 3 == End || isBlankOrBreak(Current + 3)))
+ return scanDocumentIndicator(true);
+
+ if (Column == 0 && Current + 4 <= End
+ && *Current == '.'
+ && *(Current + 1) == '.'
+ && *(Current + 2) == '.'
+ && (Current + 3 == End || isBlankOrBreak(Current + 3)))
+ return scanDocumentIndicator(false);
+
+ if (*Current == '[')
+ return scanFlowCollectionStart(true);
+
+ if (*Current == '{')
+ return scanFlowCollectionStart(false);
+
+ if (*Current == ']')
+ return scanFlowCollectionEnd(true);
+
+ if (*Current == '}')
+ return scanFlowCollectionEnd(false);
+
+ if (*Current == ',')
+ return scanFlowEntry();
+
+ if (*Current == '-' && isBlankOrBreak(Current + 1))
+ return scanBlockEntry();
+
+ if (*Current == '?' && (FlowLevel || isBlankOrBreak(Current + 1)))
+ return scanKey();
+
+ if (*Current == ':' && (FlowLevel || isBlankOrBreak(Current + 1)))
+ return scanValue();
+
+ if (*Current == '*')
+ return scanAliasOrAnchor(true);
+
+ if (*Current == '&')
+ return scanAliasOrAnchor(false);
+
+ if (*Current == '!')
+ return scanTag();
+
+ if (*Current == '|' && !FlowLevel)
+ return scanBlockScalar(true);
+
+ if (*Current == '>' && !FlowLevel)
+ return scanBlockScalar(false);
+
+ if (*Current == '\'')
+ return scanFlowScalar(false);
+
+ if (*Current == '"')
+ return scanFlowScalar(true);
+
+ // Get a plain scalar.
+ StringRef FirstChar(Current, 1);
+ if (!(isBlankOrBreak(Current)
+ || FirstChar.find_first_of("-?:,[]{}#&*!|>'\"%@`") != StringRef::npos)
+ || (*Current == '-' && !isBlankOrBreak(Current + 1))
+ || (!FlowLevel && (*Current == '?' || *Current == ':')
+ && isBlankOrBreak(Current + 1))
+ || (!FlowLevel && *Current == ':'
+ && Current + 2 < End
+ && *(Current + 1) == ':'
+ && !isBlankOrBreak(Current + 2)))
+ return scanPlainScalar();
+
+ setError("Unrecognized character while tokenizing.");
+ return false;
+}
+
+Stream::Stream(StringRef Input, SourceMgr &SM, bool ShowColors,
+ std::error_code *EC)
+ : scanner(new Scanner(Input, SM, ShowColors, EC)), CurrentDoc() {}
+
+Stream::Stream(MemoryBufferRef InputBuffer, SourceMgr &SM, bool ShowColors,
+ std::error_code *EC)
+ : scanner(new Scanner(InputBuffer, SM, ShowColors, EC)), CurrentDoc() {}
+
+Stream::~Stream() = default;
+
+bool Stream::failed() { return scanner->failed(); }
+
+void Stream::printError(Node *N, const Twine &Msg) {
+ SMRange Range = N ? N->getSourceRange() : SMRange();
+ scanner->printError( Range.Start
+ , SourceMgr::DK_Error
+ , Msg
+ , Range);
+}
+
+document_iterator Stream::begin() {
+ if (CurrentDoc)
+ report_fatal_error("Can only iterate over the stream once");
+
+ // Skip Stream-Start.
+ scanner->getNext();
+
+ CurrentDoc.reset(new Document(*this));
+ return document_iterator(CurrentDoc);
+}
+
+document_iterator Stream::end() {
+ return document_iterator();
+}
+
+void Stream::skip() {
+ for (document_iterator i = begin(), e = end(); i != e; ++i)
+ i->skip();
+}
+
+Node::Node(unsigned int Type, std::unique_ptr<Document> &D, StringRef A,
+ StringRef T)
+ : Doc(D), TypeID(Type), Anchor(A), Tag(T) {
+ SMLoc Start = SMLoc::getFromPointer(peekNext().Range.begin());
+ SourceRange = SMRange(Start, Start);
+}
+
+std::string Node::getVerbatimTag() const {
+ StringRef Raw = getRawTag();
+ if (!Raw.empty() && Raw != "!") {
+ std::string Ret;
+ if (Raw.find_last_of('!') == 0) {
+ Ret = Doc->getTagMap().find("!")->second;
+ Ret += Raw.substr(1);
+ return Ret;
+ } else if (Raw.startswith("!!")) {
+ Ret = Doc->getTagMap().find("!!")->second;
+ Ret += Raw.substr(2);
+ return Ret;
+ } else {
+ StringRef TagHandle = Raw.substr(0, Raw.find_last_of('!') + 1);
+ std::map<StringRef, StringRef>::const_iterator It =
+ Doc->getTagMap().find(TagHandle);
+ if (It != Doc->getTagMap().end())
+ Ret = It->second;
+ else {
+ Token T;
+ T.Kind = Token::TK_Tag;
+ T.Range = TagHandle;
+ setError(Twine("Unknown tag handle ") + TagHandle, T);
+ }
+ Ret += Raw.substr(Raw.find_last_of('!') + 1);
+ return Ret;
+ }
+ }
+
+ switch (getType()) {
+ case NK_Null:
+ return "tag:yaml.org,2002:null";
+ case NK_Scalar:
+ case NK_BlockScalar:
+ // TODO: Tag resolution.
+ return "tag:yaml.org,2002:str";
+ case NK_Mapping:
+ return "tag:yaml.org,2002:map";
+ case NK_Sequence:
+ return "tag:yaml.org,2002:seq";
+ }
+
+ return "";
+}
+
+Token &Node::peekNext() {
+ return Doc->peekNext();
+}
+
+Token Node::getNext() {
+ return Doc->getNext();
+}
+
+Node *Node::parseBlockNode() {
+ return Doc->parseBlockNode();
+}
+
+BumpPtrAllocator &Node::getAllocator() {
+ return Doc->NodeAllocator;
+}
+
+void Node::setError(const Twine &Msg, Token &Tok) const {
+ Doc->setError(Msg, Tok);
+}
+
+bool Node::failed() const {
+ return Doc->failed();
+}
+
+StringRef ScalarNode::getValue(SmallVectorImpl<char> &Storage) const {
+ // TODO: Handle newlines properly. We need to remove leading whitespace.
+ if (Value[0] == '"') { // Double quoted.
+ // Pull off the leading and trailing "s.
+ StringRef UnquotedValue = Value.substr(1, Value.size() - 2);
+ // Search for characters that would require unescaping the value.
+ StringRef::size_type i = UnquotedValue.find_first_of("\\\r\n");
+ if (i != StringRef::npos)
+ return unescapeDoubleQuoted(UnquotedValue, i, Storage);
+ return UnquotedValue;
+ } else if (Value[0] == '\'') { // Single quoted.
+ // Pull off the leading and trailing 's.
+ StringRef UnquotedValue = Value.substr(1, Value.size() - 2);
+ StringRef::size_type i = UnquotedValue.find('\'');
+ if (i != StringRef::npos) {
+ // We're going to need Storage.
+ Storage.clear();
+ Storage.reserve(UnquotedValue.size());
+ for (; i != StringRef::npos; i = UnquotedValue.find('\'')) {
+ StringRef Valid(UnquotedValue.begin(), i);
+ Storage.insert(Storage.end(), Valid.begin(), Valid.end());
+ Storage.push_back('\'');
+ UnquotedValue = UnquotedValue.substr(i + 2);
+ }
+ Storage.insert(Storage.end(), UnquotedValue.begin(), UnquotedValue.end());
+ return StringRef(Storage.begin(), Storage.size());
+ }
+ return UnquotedValue;
+ }
+ // Plain or block.
+ return Value.rtrim(' ');
+}
+
+StringRef ScalarNode::unescapeDoubleQuoted( StringRef UnquotedValue
+ , StringRef::size_type i
+ , SmallVectorImpl<char> &Storage)
+ const {
+ // Use Storage to build proper value.
+ Storage.clear();
+ Storage.reserve(UnquotedValue.size());
+ for (; i != StringRef::npos; i = UnquotedValue.find_first_of("\\\r\n")) {
+ // Insert all previous chars into Storage.
+ StringRef Valid(UnquotedValue.begin(), i);
+ Storage.insert(Storage.end(), Valid.begin(), Valid.end());
+ // Chop off inserted chars.
+ UnquotedValue = UnquotedValue.substr(i);
+
+ assert(!UnquotedValue.empty() && "Can't be empty!");
+
+ // Parse escape or line break.
+ switch (UnquotedValue[0]) {
+ case '\r':
+ case '\n':
+ Storage.push_back('\n');
+ if ( UnquotedValue.size() > 1
+ && (UnquotedValue[1] == '\r' || UnquotedValue[1] == '\n'))
+ UnquotedValue = UnquotedValue.substr(1);
+ UnquotedValue = UnquotedValue.substr(1);
+ break;
+ default:
+ if (UnquotedValue.size() == 1) {
+ Token T;
+ T.Range = StringRef(UnquotedValue.begin(), 1);
+ setError("Unrecognized escape code", T);
+ return "";
+ }
+ UnquotedValue = UnquotedValue.substr(1);
+ switch (UnquotedValue[0]) {
+ default: {
+ Token T;
+ T.Range = StringRef(UnquotedValue.begin(), 1);
+ setError("Unrecognized escape code", T);
+ return "";
+ }
+ case '\r':
+ case '\n':
+ // Remove the new line.
+ if ( UnquotedValue.size() > 1
+ && (UnquotedValue[1] == '\r' || UnquotedValue[1] == '\n'))
+ UnquotedValue = UnquotedValue.substr(1);
+ // If this was just a single byte newline, it will get skipped
+ // below.
+ break;
+ case '0':
+ Storage.push_back(0x00);
+ break;
+ case 'a':
+ Storage.push_back(0x07);
+ break;
+ case 'b':
+ Storage.push_back(0x08);
+ break;
+ case 't':
+ case 0x09:
+ Storage.push_back(0x09);
+ break;
+ case 'n':
+ Storage.push_back(0x0A);
+ break;
+ case 'v':
+ Storage.push_back(0x0B);
+ break;
+ case 'f':
+ Storage.push_back(0x0C);
+ break;
+ case 'r':
+ Storage.push_back(0x0D);
+ break;
+ case 'e':
+ Storage.push_back(0x1B);
+ break;
+ case ' ':
+ Storage.push_back(0x20);
+ break;
+ case '"':
+ Storage.push_back(0x22);
+ break;
+ case '/':
+ Storage.push_back(0x2F);
+ break;
+ case '\\':
+ Storage.push_back(0x5C);
+ break;
+ case 'N':
+ encodeUTF8(0x85, Storage);
+ break;
+ case '_':
+ encodeUTF8(0xA0, Storage);
+ break;
+ case 'L':
+ encodeUTF8(0x2028, Storage);
+ break;
+ case 'P':
+ encodeUTF8(0x2029, Storage);
+ break;
+ case 'x': {
+ if (UnquotedValue.size() < 3)
+ // TODO: Report error.
+ break;
+ unsigned int UnicodeScalarValue;
+ if (UnquotedValue.substr(1, 2).getAsInteger(16, UnicodeScalarValue))
+ // TODO: Report error.
+ UnicodeScalarValue = 0xFFFD;
+ encodeUTF8(UnicodeScalarValue, Storage);
+ UnquotedValue = UnquotedValue.substr(2);
+ break;
+ }
+ case 'u': {
+ if (UnquotedValue.size() < 5)
+ // TODO: Report error.
+ break;
+ unsigned int UnicodeScalarValue;
+ if (UnquotedValue.substr(1, 4).getAsInteger(16, UnicodeScalarValue))
+ // TODO: Report error.
+ UnicodeScalarValue = 0xFFFD;
+ encodeUTF8(UnicodeScalarValue, Storage);
+ UnquotedValue = UnquotedValue.substr(4);
+ break;
+ }
+ case 'U': {
+ if (UnquotedValue.size() < 9)
+ // TODO: Report error.
+ break;
+ unsigned int UnicodeScalarValue;
+ if (UnquotedValue.substr(1, 8).getAsInteger(16, UnicodeScalarValue))
+ // TODO: Report error.
+ UnicodeScalarValue = 0xFFFD;
+ encodeUTF8(UnicodeScalarValue, Storage);
+ UnquotedValue = UnquotedValue.substr(8);
+ break;
+ }
+ }
+ UnquotedValue = UnquotedValue.substr(1);
+ }
+ }
+ Storage.insert(Storage.end(), UnquotedValue.begin(), UnquotedValue.end());
+ return StringRef(Storage.begin(), Storage.size());
+}
+
+Node *KeyValueNode::getKey() {
+ if (Key)
+ return Key;
+ // Handle implicit null keys.
+ {
+ Token &t = peekNext();
+ if ( t.Kind == Token::TK_BlockEnd
+ || t.Kind == Token::TK_Value
+ || t.Kind == Token::TK_Error) {
+ return Key = new (getAllocator()) NullNode(Doc);
+ }
+ if (t.Kind == Token::TK_Key)
+ getNext(); // skip TK_Key.
+ }
+
+ // Handle explicit null keys.
+ Token &t = peekNext();
+ if (t.Kind == Token::TK_BlockEnd || t.Kind == Token::TK_Value) {
+ return Key = new (getAllocator()) NullNode(Doc);
+ }
+
+ // We've got a normal key.
+ return Key = parseBlockNode();
+}
+
+Node *KeyValueNode::getValue() {
+ if (Value)
+ return Value;
+
+ if (Node* Key = getKey())
+ Key->skip();
+ else {
+ setError("Null key in Key Value.", peekNext());
+ return Value = new (getAllocator()) NullNode(Doc);
+ }
+
+ if (failed())
+ return Value = new (getAllocator()) NullNode(Doc);
+
+ // Handle implicit null values.
+ {
+ Token &t = peekNext();
+ if ( t.Kind == Token::TK_BlockEnd
+ || t.Kind == Token::TK_FlowMappingEnd
+ || t.Kind == Token::TK_Key
+ || t.Kind == Token::TK_FlowEntry
+ || t.Kind == Token::TK_Error) {
+ return Value = new (getAllocator()) NullNode(Doc);
+ }
+
+ if (t.Kind != Token::TK_Value) {
+ setError("Unexpected token in Key Value.", t);
+ return Value = new (getAllocator()) NullNode(Doc);
+ }
+ getNext(); // skip TK_Value.
+ }
+
+ // Handle explicit null values.
+ Token &t = peekNext();
+ if (t.Kind == Token::TK_BlockEnd || t.Kind == Token::TK_Key) {
+ return Value = new (getAllocator()) NullNode(Doc);
+ }
+
+ // We got a normal value.
+ return Value = parseBlockNode();
+}
+
+void MappingNode::increment() {
+ if (failed()) {
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ return;
+ }
+ if (CurrentEntry) {
+ CurrentEntry->skip();
+ if (Type == MT_Inline) {
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ return;
+ }
+ }
+ Token T = peekNext();
+ if (T.Kind == Token::TK_Key || T.Kind == Token::TK_Scalar) {
+ // KeyValueNode eats the TK_Key. That way it can detect null keys.
+ CurrentEntry = new (getAllocator()) KeyValueNode(Doc);
+ } else if (Type == MT_Block) {
+ switch (T.Kind) {
+ case Token::TK_BlockEnd:
+ getNext();
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ default:
+ setError("Unexpected token. Expected Key or Block End", T);
+ LLVM_FALLTHROUGH;
+ case Token::TK_Error:
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ } else {
+ switch (T.Kind) {
+ case Token::TK_FlowEntry:
+ // Eat the flow entry and recurse.
+ getNext();
+ return increment();
+ case Token::TK_FlowMappingEnd:
+ getNext();
+ LLVM_FALLTHROUGH;
+ case Token::TK_Error:
+ // Set this to end iterator.
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ default:
+ setError( "Unexpected token. Expected Key, Flow Entry, or Flow "
+ "Mapping End."
+ , T);
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ }
+}
+
+void SequenceNode::increment() {
+ if (failed()) {
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ return;
+ }
+ if (CurrentEntry)
+ CurrentEntry->skip();
+ Token T = peekNext();
+ if (SeqType == ST_Block) {
+ switch (T.Kind) {
+ case Token::TK_BlockEntry:
+ getNext();
+ CurrentEntry = parseBlockNode();
+ if (!CurrentEntry) { // An error occurred.
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ break;
+ case Token::TK_BlockEnd:
+ getNext();
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ default:
+ setError( "Unexpected token. Expected Block Entry or Block End."
+ , T);
+ LLVM_FALLTHROUGH;
+ case Token::TK_Error:
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ } else if (SeqType == ST_Indentless) {
+ switch (T.Kind) {
+ case Token::TK_BlockEntry:
+ getNext();
+ CurrentEntry = parseBlockNode();
+ if (!CurrentEntry) { // An error occurred.
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ break;
+ default:
+ case Token::TK_Error:
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ }
+ } else if (SeqType == ST_Flow) {
+ switch (T.Kind) {
+ case Token::TK_FlowEntry:
+ // Eat the flow entry and recurse.
+ getNext();
+ WasPreviousTokenFlowEntry = true;
+ return increment();
+ case Token::TK_FlowSequenceEnd:
+ getNext();
+ LLVM_FALLTHROUGH;
+ case Token::TK_Error:
+ // Set this to end iterator.
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ case Token::TK_StreamEnd:
+ case Token::TK_DocumentEnd:
+ case Token::TK_DocumentStart:
+ setError("Could not find closing ]!", T);
+ // Set this to end iterator.
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ default:
+ if (!WasPreviousTokenFlowEntry) {
+ setError("Expected , between entries!", T);
+ IsAtEnd = true;
+ CurrentEntry = nullptr;
+ break;
+ }
+ // Otherwise it must be a flow entry.
+ CurrentEntry = parseBlockNode();
+ if (!CurrentEntry) {
+ IsAtEnd = true;
+ }
+ WasPreviousTokenFlowEntry = false;
+ break;
+ }
+ }
+}
+
+Document::Document(Stream &S) : stream(S), Root(nullptr) {
+ // Tag maps starts with two default mappings.
+ TagMap["!"] = "!";
+ TagMap["!!"] = "tag:yaml.org,2002:";
+
+ if (parseDirectives())
+ expectToken(Token::TK_DocumentStart);
+ Token &T = peekNext();
+ if (T.Kind == Token::TK_DocumentStart)
+ getNext();
+}
+
+bool Document::skip() {
+ if (stream.scanner->failed())
+ return false;
+ if (!Root)
+ getRoot();
+ Root->skip();
+ Token &T = peekNext();
+ if (T.Kind == Token::TK_StreamEnd)
+ return false;
+ if (T.Kind == Token::TK_DocumentEnd) {
+ getNext();
+ return skip();
+ }
+ return true;
+}
+
+Token &Document::peekNext() {
+ return stream.scanner->peekNext();
+}
+
+Token Document::getNext() {
+ return stream.scanner->getNext();
+}
+
+void Document::setError(const Twine &Message, Token &Location) const {
+ stream.scanner->setError(Message, Location.Range.begin());
+}
+
+bool Document::failed() const {
+ return stream.scanner->failed();
+}
+
+Node *Document::parseBlockNode() {
+ Token T = peekNext();
+ // Handle properties.
+ Token AnchorInfo;
+ Token TagInfo;
+parse_property:
+ switch (T.Kind) {
+ case Token::TK_Alias:
+ getNext();
+ return new (NodeAllocator) AliasNode(stream.CurrentDoc, T.Range.substr(1));
+ case Token::TK_Anchor:
+ if (AnchorInfo.Kind == Token::TK_Anchor) {
+ setError("Already encountered an anchor for this node!", T);
+ return nullptr;
+ }
+ AnchorInfo = getNext(); // Consume TK_Anchor.
+ T = peekNext();
+ goto parse_property;
+ case Token::TK_Tag:
+ if (TagInfo.Kind == Token::TK_Tag) {
+ setError("Already encountered a tag for this node!", T);
+ return nullptr;
+ }
+ TagInfo = getNext(); // Consume TK_Tag.
+ T = peekNext();
+ goto parse_property;
+ default:
+ break;
+ }
+
+ switch (T.Kind) {
+ case Token::TK_BlockEntry:
+ // We got an unindented BlockEntry sequence. This is not terminated with
+ // a BlockEnd.
+ // Don't eat the TK_BlockEntry, SequenceNode needs it.
+ return new (NodeAllocator) SequenceNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , SequenceNode::ST_Indentless);
+ case Token::TK_BlockSequenceStart:
+ getNext();
+ return new (NodeAllocator)
+ SequenceNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , SequenceNode::ST_Block);
+ case Token::TK_BlockMappingStart:
+ getNext();
+ return new (NodeAllocator)
+ MappingNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , MappingNode::MT_Block);
+ case Token::TK_FlowSequenceStart:
+ getNext();
+ return new (NodeAllocator)
+ SequenceNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , SequenceNode::ST_Flow);
+ case Token::TK_FlowMappingStart:
+ getNext();
+ return new (NodeAllocator)
+ MappingNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , MappingNode::MT_Flow);
+ case Token::TK_Scalar:
+ getNext();
+ return new (NodeAllocator)
+ ScalarNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , T.Range);
+ case Token::TK_BlockScalar: {
+ getNext();
+ StringRef NullTerminatedStr(T.Value.c_str(), T.Value.length() + 1);
+ StringRef StrCopy = NullTerminatedStr.copy(NodeAllocator).drop_back();
+ return new (NodeAllocator)
+ BlockScalarNode(stream.CurrentDoc, AnchorInfo.Range.substr(1),
+ TagInfo.Range, StrCopy, T.Range);
+ }
+ case Token::TK_Key:
+ // Don't eat the TK_Key, KeyValueNode expects it.
+ return new (NodeAllocator)
+ MappingNode( stream.CurrentDoc
+ , AnchorInfo.Range.substr(1)
+ , TagInfo.Range
+ , MappingNode::MT_Inline);
+ case Token::TK_DocumentStart:
+ case Token::TK_DocumentEnd:
+ case Token::TK_StreamEnd:
+ default:
+ // TODO: Properly handle tags. "[!!str ]" should resolve to !!str "", not
+ // !!null null.
+ return new (NodeAllocator) NullNode(stream.CurrentDoc);
+ case Token::TK_FlowMappingEnd:
+ case Token::TK_FlowSequenceEnd:
+ case Token::TK_FlowEntry: {
+ if (Root && (isa<MappingNode>(Root) || isa<SequenceNode>(Root)))
+ return new (NodeAllocator) NullNode(stream.CurrentDoc);
+
+ setError("Unexpected token", T);
+ return nullptr;
+ }
+ case Token::TK_Error:
+ return nullptr;
+ }
+ llvm_unreachable("Control flow shouldn't reach here.");
+ return nullptr;
+}
+
+bool Document::parseDirectives() {
+ bool isDirective = false;
+ while (true) {
+ Token T = peekNext();
+ if (T.Kind == Token::TK_TagDirective) {
+ parseTAGDirective();
+ isDirective = true;
+ } else if (T.Kind == Token::TK_VersionDirective) {
+ parseYAMLDirective();
+ isDirective = true;
+ } else
+ break;
+ }
+ return isDirective;
+}
+
+void Document::parseYAMLDirective() {
+ getNext(); // Eat %YAML <version>
+}
+
+void Document::parseTAGDirective() {
+ Token Tag = getNext(); // %TAG <handle> <prefix>
+ StringRef T = Tag.Range;
+ // Strip %TAG
+ T = T.substr(T.find_first_of(" \t")).ltrim(" \t");
+ std::size_t HandleEnd = T.find_first_of(" \t");
+ StringRef TagHandle = T.substr(0, HandleEnd);
+ StringRef TagPrefix = T.substr(HandleEnd).ltrim(" \t");
+ TagMap[TagHandle] = TagPrefix;
+}
+
+bool Document::expectToken(int TK) {
+ Token T = getNext();
+ if (T.Kind != TK) {
+ setError("Unexpected token", T);
+ return false;
+ }
+ return true;
+}
diff --git a/third_party/llvm-project/YAMLTraits.cpp b/third_party/llvm-project/YAMLTraits.cpp
new file mode 100644
index 000000000..5f0cedc71
--- /dev/null
+++ b/third_party/llvm-project/YAMLTraits.cpp
@@ -0,0 +1,1087 @@
+//===- lib/Support/YAMLTraits.cpp -----------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/YAMLTraits.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/LineIterator.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/Unicode.h"
+#include "llvm/Support/YAMLParser.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+#include <cstdlib>
+#include <cstring>
+#include <string>
+#include <vector>
+
+using namespace llvm;
+using namespace yaml;
+
+//===----------------------------------------------------------------------===//
+// IO
+//===----------------------------------------------------------------------===//
+
+IO::IO(void *Context) : Ctxt(Context) {}
+
+IO::~IO() = default;
+
+void *IO::getContext() const {
+ return Ctxt;
+}
+
+void IO::setContext(void *Context) {
+ Ctxt = Context;
+}
+
+//===----------------------------------------------------------------------===//
+// Input
+//===----------------------------------------------------------------------===//
+
+Input::Input(StringRef InputContent, void *Ctxt,
+ SourceMgr::DiagHandlerTy DiagHandler, void *DiagHandlerCtxt)
+ : IO(Ctxt), Strm(new Stream(InputContent, SrcMgr, false, &EC)) {
+ if (DiagHandler)
+ SrcMgr.setDiagHandler(DiagHandler, DiagHandlerCtxt);
+ DocIterator = Strm->begin();
+}
+
+Input::Input(MemoryBufferRef Input, void *Ctxt,
+ SourceMgr::DiagHandlerTy DiagHandler, void *DiagHandlerCtxt)
+ : IO(Ctxt), Strm(new Stream(Input, SrcMgr, false, &EC)) {
+ if (DiagHandler)
+ SrcMgr.setDiagHandler(DiagHandler, DiagHandlerCtxt);
+ DocIterator = Strm->begin();
+}
+
+Input::~Input() = default;
+
+std::error_code Input::error() { return EC; }
+
+// Pin the vtables to this file.
+void Input::HNode::anchor() {}
+void Input::EmptyHNode::anchor() {}
+void Input::ScalarHNode::anchor() {}
+void Input::MapHNode::anchor() {}
+void Input::SequenceHNode::anchor() {}
+
+bool Input::outputting() const {
+ return false;
+}
+
+bool Input::setCurrentDocument() {
+ if (DocIterator != Strm->end()) {
+ Node *N = DocIterator->getRoot();
+ if (!N) {
+ EC = make_error_code(errc::invalid_argument);
+ return false;
+ }
+
+ if (isa<NullNode>(N)) {
+ // Empty files are allowed and ignored
+ ++DocIterator;
+ return setCurrentDocument();
+ }
+ TopNode = createHNodes(N);
+ CurrentNode = TopNode.get();
+ return true;
+ }
+ return false;
+}
+
+bool Input::nextDocument() {
+ return ++DocIterator != Strm->end();
+}
+
+const Node *Input::getCurrentNode() const {
+ return CurrentNode ? CurrentNode->_node : nullptr;
+}
+
+bool Input::mapTag(StringRef Tag, bool Default) {
+ // CurrentNode can be null if setCurrentDocument() was unable to
+ // parse the document because it was invalid or empty.
+ if (!CurrentNode)
+ return false;
+
+ std::string foundTag = CurrentNode->_node->getVerbatimTag();
+ if (foundTag.empty()) {
+ // If no tag found and 'Tag' is the default, say it was found.
+ return Default;
+ }
+ // Return true iff found tag matches supplied tag.
+ return Tag.equals(foundTag);
+}
+
+void Input::beginMapping() {
+ if (EC)
+ return;
+ // CurrentNode can be null if the document is empty.
+ MapHNode *MN = dyn_cast_or_null<MapHNode>(CurrentNode);
+ if (MN) {
+ MN->ValidKeys.clear();
+ }
+}
+
+std::vector<StringRef> Input::keys() {
+ MapHNode *MN = dyn_cast<MapHNode>(CurrentNode);
+ std::vector<StringRef> Ret;
+ if (!MN) {
+ setError(CurrentNode, "not a mapping");
+ return Ret;
+ }
+ for (auto &P : MN->Mapping)
+ Ret.push_back(P.first());
+ return Ret;
+}
+
+bool Input::preflightKey(const char *Key, bool Required, bool, bool &UseDefault,
+ void *&SaveInfo) {
+ UseDefault = false;
+ if (EC)
+ return false;
+
+ // CurrentNode is null for empty documents, which is an error in case required
+ // nodes are present.
+ if (!CurrentNode) {
+ if (Required)
+ EC = make_error_code(errc::invalid_argument);
+ return false;
+ }
+
+ MapHNode *MN = dyn_cast<MapHNode>(CurrentNode);
+ if (!MN) {
+ if (Required || !isa<EmptyHNode>(CurrentNode))
+ setError(CurrentNode, "not a mapping");
+ return false;
+ }
+ MN->ValidKeys.push_back(Key);
+ HNode *Value = MN->Mapping[Key].get();
+ if (!Value) {
+ if (Required)
+ setError(CurrentNode, Twine("missing required key '") + Key + "'");
+ else
+ UseDefault = true;
+ return false;
+ }
+ SaveInfo = CurrentNode;
+ CurrentNode = Value;
+ return true;
+}
+
+void Input::postflightKey(void *saveInfo) {
+ CurrentNode = reinterpret_cast<HNode *>(saveInfo);
+}
+
+void Input::endMapping() {
+ if (EC)
+ return;
+ // CurrentNode can be null if the document is empty.
+ MapHNode *MN = dyn_cast_or_null<MapHNode>(CurrentNode);
+ if (!MN)
+ return;
+ for (const auto &NN : MN->Mapping) {
+ if (!is_contained(MN->ValidKeys, NN.first())) {
+ setError(NN.second.get(), Twine("unknown key '") + NN.first() + "'");
+ break;
+ }
+ }
+}
+
+void Input::beginFlowMapping() { beginMapping(); }
+
+void Input::endFlowMapping() { endMapping(); }
+
+unsigned Input::beginSequence() {
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode))
+ return SQ->Entries.size();
+ if (isa<EmptyHNode>(CurrentNode))
+ return 0;
+ // Treat case where there's a scalar "null" value as an empty sequence.
+ if (ScalarHNode *SN = dyn_cast<ScalarHNode>(CurrentNode)) {
+ if (isNull(SN->value()))
+ return 0;
+ }
+ // Any other type of HNode is an error.
+ setError(CurrentNode, "not a sequence");
+ return 0;
+}
+
+void Input::endSequence() {
+}
+
+bool Input::preflightElement(unsigned Index, void *&SaveInfo) {
+ if (EC)
+ return false;
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode)) {
+ SaveInfo = CurrentNode;
+ CurrentNode = SQ->Entries[Index].get();
+ return true;
+ }
+ return false;
+}
+
+void Input::postflightElement(void *SaveInfo) {
+ CurrentNode = reinterpret_cast<HNode *>(SaveInfo);
+}
+
+unsigned Input::beginFlowSequence() { return beginSequence(); }
+
+bool Input::preflightFlowElement(unsigned index, void *&SaveInfo) {
+ if (EC)
+ return false;
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode)) {
+ SaveInfo = CurrentNode;
+ CurrentNode = SQ->Entries[index].get();
+ return true;
+ }
+ return false;
+}
+
+void Input::postflightFlowElement(void *SaveInfo) {
+ CurrentNode = reinterpret_cast<HNode *>(SaveInfo);
+}
+
+void Input::endFlowSequence() {
+}
+
+void Input::beginEnumScalar() {
+ ScalarMatchFound = false;
+}
+
+bool Input::matchEnumScalar(const char *Str, bool) {
+ if (ScalarMatchFound)
+ return false;
+ if (ScalarHNode *SN = dyn_cast<ScalarHNode>(CurrentNode)) {
+ if (SN->value().equals(Str)) {
+ ScalarMatchFound = true;
+ return true;
+ }
+ }
+ return false;
+}
+
+bool Input::matchEnumFallback() {
+ if (ScalarMatchFound)
+ return false;
+ ScalarMatchFound = true;
+ return true;
+}
+
+void Input::endEnumScalar() {
+ if (!ScalarMatchFound) {
+ setError(CurrentNode, "unknown enumerated scalar");
+ }
+}
+
+bool Input::beginBitSetScalar(bool &DoClear) {
+ BitValuesUsed.clear();
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode)) {
+ BitValuesUsed.insert(BitValuesUsed.begin(), SQ->Entries.size(), false);
+ } else {
+ setError(CurrentNode, "expected sequence of bit values");
+ }
+ DoClear = true;
+ return true;
+}
+
+bool Input::bitSetMatch(const char *Str, bool) {
+ if (EC)
+ return false;
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode)) {
+ unsigned Index = 0;
+ for (auto &N : SQ->Entries) {
+ if (ScalarHNode *SN = dyn_cast<ScalarHNode>(N.get())) {
+ if (SN->value().equals(Str)) {
+ BitValuesUsed[Index] = true;
+ return true;
+ }
+ } else {
+ setError(CurrentNode, "unexpected scalar in sequence of bit values");
+ }
+ ++Index;
+ }
+ } else {
+ setError(CurrentNode, "expected sequence of bit values");
+ }
+ return false;
+}
+
+void Input::endBitSetScalar() {
+ if (EC)
+ return;
+ if (SequenceHNode *SQ = dyn_cast<SequenceHNode>(CurrentNode)) {
+ assert(BitValuesUsed.size() == SQ->Entries.size());
+ for (unsigned i = 0; i < SQ->Entries.size(); ++i) {
+ if (!BitValuesUsed[i]) {
+ setError(SQ->Entries[i].get(), "unknown bit value");
+ return;
+ }
+ }
+ }
+}
+
+void Input::scalarString(StringRef &S, QuotingType) {
+ if (ScalarHNode *SN = dyn_cast<ScalarHNode>(CurrentNode)) {
+ S = SN->value();
+ } else {
+ setError(CurrentNode, "unexpected scalar");
+ }
+}
+
+void Input::blockScalarString(StringRef &S) { scalarString(S, QuotingType::None); }
+
+void Input::scalarTag(std::string &Tag) {
+ Tag = CurrentNode->_node->getVerbatimTag();
+}
+
+void Input::setError(HNode *hnode, const Twine &message) {
+ assert(hnode && "HNode must not be NULL");
+ setError(hnode->_node, message);
+}
+
+NodeKind Input::getNodeKind() {
+ if (isa<ScalarHNode>(CurrentNode))
+ return NodeKind::Scalar;
+ else if (isa<MapHNode>(CurrentNode))
+ return NodeKind::Map;
+ else if (isa<SequenceHNode>(CurrentNode))
+ return NodeKind::Sequence;
+ llvm_unreachable("Unsupported node kind");
+}
+
+void Input::setError(Node *node, const Twine &message) {
+ Strm->printError(node, message);
+ EC = make_error_code(errc::invalid_argument);
+}
+
+std::unique_ptr<Input::HNode> Input::createHNodes(Node *N) {
+ SmallString<128> StringStorage;
+ if (ScalarNode *SN = dyn_cast<ScalarNode>(N)) {
+ StringRef KeyStr = SN->getValue(StringStorage);
+ if (!StringStorage.empty()) {
+ // Copy string to permanent storage
+ KeyStr = StringStorage.str().copy(StringAllocator);
+ }
+ return std::make_unique<ScalarHNode>(N, KeyStr);
+ } else if (BlockScalarNode *BSN = dyn_cast<BlockScalarNode>(N)) {
+ StringRef ValueCopy = BSN->getValue().copy(StringAllocator);
+ return std::make_unique<ScalarHNode>(N, ValueCopy);
+ } else if (SequenceNode *SQ = dyn_cast<SequenceNode>(N)) {
+ auto SQHNode = std::make_unique<SequenceHNode>(N);
+ for (Node &SN : *SQ) {
+ auto Entry = createHNodes(&SN);
+ if (EC)
+ break;
+ SQHNode->Entries.push_back(std::move(Entry));
+ }
+ return std::move(SQHNode);
+ } else if (MappingNode *Map = dyn_cast<MappingNode>(N)) {
+ auto mapHNode = std::make_unique<MapHNode>(N);
+ for (KeyValueNode &KVN : *Map) {
+ Node *KeyNode = KVN.getKey();
+ ScalarNode *Key = dyn_cast_or_null<ScalarNode>(KeyNode);
+ Node *Value = KVN.getValue();
+ if (!Key || !Value) {
+ if (!Key)
+ setError(KeyNode, "Map key must be a scalar");
+ if (!Value)
+ setError(KeyNode, "Map value must not be empty");
+ break;
+ }
+ StringStorage.clear();
+ StringRef KeyStr = Key->getValue(StringStorage);
+ if (!StringStorage.empty()) {
+ // Copy string to permanent storage
+ KeyStr = StringStorage.str().copy(StringAllocator);
+ }
+ auto ValueHNode = createHNodes(Value);
+ if (EC)
+ break;
+ mapHNode->Mapping[KeyStr] = std::move(ValueHNode);
+ }
+ return std::move(mapHNode);
+ } else if (isa<NullNode>(N)) {
+ return std::make_unique<EmptyHNode>(N);
+ } else {
+ setError(N, "unknown node kind");
+ return nullptr;
+ }
+}
+
+void Input::setError(const Twine &Message) {
+ setError(CurrentNode, Message);
+}
+
+bool Input::canElideEmptySequence() {
+ return false;
+}
+
+//===----------------------------------------------------------------------===//
+// Output
+//===----------------------------------------------------------------------===//
+
+Output::Output(raw_ostream &yout, void *context, int WrapColumn)
+ : IO(context), Out(yout), WrapColumn(WrapColumn) {}
+
+Output::~Output() = default;
+
+bool Output::outputting() const {
+ return true;
+}
+
+void Output::beginMapping() {
+ StateStack.push_back(inMapFirstKey);
+ PaddingBeforeContainer = Padding;
+ Padding = "\n";
+}
+
+bool Output::mapTag(StringRef Tag, bool Use) {
+ if (Use) {
+ // If this tag is being written inside a sequence we should write the start
+ // of the sequence before writing the tag, otherwise the tag won't be
+ // attached to the element in the sequence, but rather the sequence itself.
+ bool SequenceElement = false;
+ if (StateStack.size() > 1) {
+ auto &E = StateStack[StateStack.size() - 2];
+ SequenceElement = inSeqAnyElement(E) || inFlowSeqAnyElement(E);
+ }
+ if (SequenceElement && StateStack.back() == inMapFirstKey) {
+ newLineCheck();
+ } else {
+ output(" ");
+ }
+ output(Tag);
+ if (SequenceElement) {
+ // If we're writing the tag during the first element of a map, the tag
+ // takes the place of the first element in the sequence.
+ if (StateStack.back() == inMapFirstKey) {
+ StateStack.pop_back();
+ StateStack.push_back(inMapOtherKey);
+ }
+ // Tags inside maps in sequences should act as keys in the map from a
+ // formatting perspective, so we always want a newline in a sequence.
+ Padding = "\n";
+ }
+ }
+ return Use;
+}
+
+void Output::endMapping() {
+ // If we did not map anything, we should explicitly emit an empty map
+ if (StateStack.back() == inMapFirstKey) {
+ Padding = PaddingBeforeContainer;
+ newLineCheck();
+ output("{}");
+ Padding = "\n";
+ }
+ StateStack.pop_back();
+}
+
+std::vector<StringRef> Output::keys() {
+ report_fatal_error("invalid call");
+}
+
+bool Output::preflightKey(const char *Key, bool Required, bool SameAsDefault,
+ bool &UseDefault, void *&) {
+ UseDefault = false;
+ if (Required || !SameAsDefault || WriteDefaultValues) {
+ auto State = StateStack.back();
+ if (State == inFlowMapFirstKey || State == inFlowMapOtherKey) {
+ flowKey(Key);
+ } else {
+ newLineCheck();
+ paddedKey(Key);
+ }
+ return true;
+ }
+ return false;
+}
+
+void Output::postflightKey(void *) {
+ if (StateStack.back() == inMapFirstKey) {
+ StateStack.pop_back();
+ StateStack.push_back(inMapOtherKey);
+ } else if (StateStack.back() == inFlowMapFirstKey) {
+ StateStack.pop_back();
+ StateStack.push_back(inFlowMapOtherKey);
+ }
+}
+
+void Output::beginFlowMapping() {
+ StateStack.push_back(inFlowMapFirstKey);
+ newLineCheck();
+ ColumnAtMapFlowStart = Column;
+ output("{ ");
+}
+
+void Output::endFlowMapping() {
+ StateStack.pop_back();
+ outputUpToEndOfLine(" }");
+}
+
+void Output::beginDocuments() {
+ outputUpToEndOfLine("---");
+}
+
+bool Output::preflightDocument(unsigned index) {
+ if (index > 0)
+ outputUpToEndOfLine("\n---");
+ return true;
+}
+
+void Output::postflightDocument() {
+}
+
+void Output::endDocuments() {
+ output("\n...\n");
+}
+
+unsigned Output::beginSequence() {
+ StateStack.push_back(inSeqFirstElement);
+ PaddingBeforeContainer = Padding;
+ Padding = "\n";
+ return 0;
+}
+
+void Output::endSequence() {
+ // If we did not emit anything, we should explicitly emit an empty sequence
+ if (StateStack.back() == inSeqFirstElement) {
+ Padding = PaddingBeforeContainer;
+ newLineCheck();
+ output("[]");
+ Padding = "\n";
+ }
+ StateStack.pop_back();
+}
+
+bool Output::preflightElement(unsigned, void *&) {
+ return true;
+}
+
+void Output::postflightElement(void *) {
+ if (StateStack.back() == inSeqFirstElement) {
+ StateStack.pop_back();
+ StateStack.push_back(inSeqOtherElement);
+ } else if (StateStack.back() == inFlowSeqFirstElement) {
+ StateStack.pop_back();
+ StateStack.push_back(inFlowSeqOtherElement);
+ }
+}
+
+unsigned Output::beginFlowSequence() {
+ StateStack.push_back(inFlowSeqFirstElement);
+ newLineCheck();
+ ColumnAtFlowStart = Column;
+ output("[ ");
+ NeedFlowSequenceComma = false;
+ return 0;
+}
+
+void Output::endFlowSequence() {
+ StateStack.pop_back();
+ outputUpToEndOfLine(" ]");
+}
+
+bool Output::preflightFlowElement(unsigned, void *&) {
+ if (NeedFlowSequenceComma)
+ output(", ");
+ if (WrapColumn && Column > WrapColumn) {
+ output("\n");
+ for (int i = 0; i < ColumnAtFlowStart; ++i)
+ output(" ");
+ Column = ColumnAtFlowStart;
+ output(" ");
+ }
+ return true;
+}
+
+void Output::postflightFlowElement(void *) {
+ NeedFlowSequenceComma = true;
+}
+
+void Output::beginEnumScalar() {
+ EnumerationMatchFound = false;
+}
+
+bool Output::matchEnumScalar(const char *Str, bool Match) {
+ if (Match && !EnumerationMatchFound) {
+ newLineCheck();
+ outputUpToEndOfLine(Str);
+ EnumerationMatchFound = true;
+ }
+ return false;
+}
+
+bool Output::matchEnumFallback() {
+ if (EnumerationMatchFound)
+ return false;
+ EnumerationMatchFound = true;
+ return true;
+}
+
+void Output::endEnumScalar() {
+ if (!EnumerationMatchFound)
+ llvm_unreachable("bad runtime enum value");
+}
+
+bool Output::beginBitSetScalar(bool &DoClear) {
+ newLineCheck();
+ output("[ ");
+ NeedBitValueComma = false;
+ DoClear = false;
+ return true;
+}
+
+bool Output::bitSetMatch(const char *Str, bool Matches) {
+ if (Matches) {
+ if (NeedBitValueComma)
+ output(", ");
+ output(Str);
+ NeedBitValueComma = true;
+ }
+ return false;
+}
+
+void Output::endBitSetScalar() {
+ outputUpToEndOfLine(" ]");
+}
+
+void Output::scalarString(StringRef &S, QuotingType MustQuote) {
+ newLineCheck();
+ if (S.empty()) {
+ // Print '' for the empty string because leaving the field empty is not
+ // allowed.
+ outputUpToEndOfLine("''");
+ return;
+ }
+ if (MustQuote == QuotingType::None) {
+ // Only quote if we must.
+ outputUpToEndOfLine(S);
+ return;
+ }
+
+ const char *const Quote = MustQuote == QuotingType::Single ? "'" : "\"";
+ output(Quote); // Starting quote.
+
+ // When using double-quoted strings (and only in that case), non-printable characters may be
+ // present, and will be escaped using a variety of unicode-scalar and special short-form
+ // escapes. This is handled in yaml::escape.
+ if (MustQuote == QuotingType::Double) {
+ output(yaml::escape(S, /* EscapePrintable= */ false));
+ outputUpToEndOfLine(Quote);
+ return;
+ }
+
+ unsigned i = 0;
+ unsigned j = 0;
+ unsigned End = S.size();
+ const char *Base = S.data();
+
+ // When using single-quoted strings, any single quote ' must be doubled to be escaped.
+ while (j < End) {
+ if (S[j] == '\'') { // Escape quotes.
+ output(StringRef(&Base[i], j - i)); // "flush".
+ output(StringLiteral("''")); // Print it as ''
+ i = j + 1;
+ }
+ ++j;
+ }
+ output(StringRef(&Base[i], j - i));
+ outputUpToEndOfLine(Quote); // Ending quote.
+}
+
+void Output::blockScalarString(StringRef &S) {
+ if (!StateStack.empty())
+ newLineCheck();
+ output(" |");
+ outputNewLine();
+
+ unsigned Indent = StateStack.empty() ? 1 : StateStack.size();
+
+ auto Buffer = MemoryBuffer::getMemBuffer(S, "", false);
+ for (line_iterator Lines(*Buffer, false); !Lines.is_at_end(); ++Lines) {
+ for (unsigned I = 0; I < Indent; ++I) {
+ output(" ");
+ }
+ output(*Lines);
+ outputNewLine();
+ }
+}
+
+void Output::scalarTag(std::string &Tag) {
+ if (Tag.empty())
+ return;
+ newLineCheck();
+ output(Tag);
+ output(" ");
+}
+
+void Output::setError(const Twine &message) {
+}
+
+bool Output::canElideEmptySequence() {
+ // Normally, with an optional key/value where the value is an empty sequence,
+ // the whole key/value can be not written. But, that produces wrong yaml
+ // if the key/value is the only thing in the map and the map is used in
+ // a sequence. This detects if the this sequence is the first key/value
+ // in map that itself is embedded in a sequnce.
+ if (StateStack.size() < 2)
+ return true;
+ if (StateStack.back() != inMapFirstKey)
+ return true;
+ return !inSeqAnyElement(StateStack[StateStack.size() - 2]);
+}
+
+void Output::output(StringRef s) {
+ Column += s.size();
+ Out << s;
+}
+
+void Output::outputUpToEndOfLine(StringRef s) {
+ output(s);
+ if (StateStack.empty() || (!inFlowSeqAnyElement(StateStack.back()) &&
+ !inFlowMapAnyKey(StateStack.back())))
+ Padding = "\n";
+}
+
+void Output::outputNewLine() {
+ Out << "\n";
+ Column = 0;
+}
+
+// if seq at top, indent as if map, then add "- "
+// if seq in middle, use "- " if firstKey, else use " "
+//
+
+void Output::newLineCheck() {
+ if (Padding != "\n") {
+ output(Padding);
+ Padding = {};
+ return;
+ }
+ outputNewLine();
+ Padding = {};
+
+ if (StateStack.size() == 0)
+ return;
+
+ unsigned Indent = StateStack.size() - 1;
+ bool OutputDash = false;
+
+ if (StateStack.back() == inSeqFirstElement ||
+ StateStack.back() == inSeqOtherElement) {
+ OutputDash = true;
+ } else if ((StateStack.size() > 1) &&
+ ((StateStack.back() == inMapFirstKey) ||
+ inFlowSeqAnyElement(StateStack.back()) ||
+ (StateStack.back() == inFlowMapFirstKey)) &&
+ inSeqAnyElement(StateStack[StateStack.size() - 2])) {
+ --Indent;
+ OutputDash = true;
+ }
+
+ for (unsigned i = 0; i < Indent; ++i) {
+ output(" ");
+ }
+ if (OutputDash) {
+ output("- ");
+ }
+
+}
+
+void Output::paddedKey(StringRef key) {
+ output(key);
+ output(":");
+ const char *spaces = " ";
+ if (key.size() < strlen(spaces))
+ Padding = &spaces[key.size()];
+ else
+ Padding = " ";
+}
+
+void Output::flowKey(StringRef Key) {
+ if (StateStack.back() == inFlowMapOtherKey)
+ output(", ");
+ if (WrapColumn && Column > WrapColumn) {
+ output("\n");
+ for (int I = 0; I < ColumnAtMapFlowStart; ++I)
+ output(" ");
+ Column = ColumnAtMapFlowStart;
+ output(" ");
+ }
+ output(Key);
+ output(": ");
+}
+
+NodeKind Output::getNodeKind() { report_fatal_error("invalid call"); }
+
+bool Output::inSeqAnyElement(InState State) {
+ return State == inSeqFirstElement || State == inSeqOtherElement;
+}
+
+bool Output::inFlowSeqAnyElement(InState State) {
+ return State == inFlowSeqFirstElement || State == inFlowSeqOtherElement;
+}
+
+bool Output::inMapAnyKey(InState State) {
+ return State == inMapFirstKey || State == inMapOtherKey;
+}
+
+bool Output::inFlowMapAnyKey(InState State) {
+ return State == inFlowMapFirstKey || State == inFlowMapOtherKey;
+}
+
+//===----------------------------------------------------------------------===//
+// traits for built-in types
+//===----------------------------------------------------------------------===//
+
+void ScalarTraits<bool>::output(const bool &Val, void *, raw_ostream &Out) {
+ Out << (Val ? "true" : "false");
+}
+
+StringRef ScalarTraits<bool>::input(StringRef Scalar, void *, bool &Val) {
+ if (Scalar.equals("true")) {
+ Val = true;
+ return StringRef();
+ } else if (Scalar.equals("false")) {
+ Val = false;
+ return StringRef();
+ }
+ return "invalid boolean";
+}
+
+void ScalarTraits<StringRef>::output(const StringRef &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<StringRef>::input(StringRef Scalar, void *,
+ StringRef &Val) {
+ Val = Scalar;
+ return StringRef();
+}
+
+void ScalarTraits<std::string>::output(const std::string &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<std::string>::input(StringRef Scalar, void *,
+ std::string &Val) {
+ Val = Scalar.str();
+ return StringRef();
+}
+
+void ScalarTraits<uint8_t>::output(const uint8_t &Val, void *,
+ raw_ostream &Out) {
+ // use temp uin32_t because ostream thinks uint8_t is a character
+ uint32_t Num = Val;
+ Out << Num;
+}
+
+StringRef ScalarTraits<uint8_t>::input(StringRef Scalar, void *, uint8_t &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid number";
+ if (n > 0xFF)
+ return "out of range number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<uint16_t>::output(const uint16_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<uint16_t>::input(StringRef Scalar, void *,
+ uint16_t &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid number";
+ if (n > 0xFFFF)
+ return "out of range number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<uint32_t>::output(const uint32_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<uint32_t>::input(StringRef Scalar, void *,
+ uint32_t &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid number";
+ if (n > 0xFFFFFFFFUL)
+ return "out of range number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<uint64_t>::output(const uint64_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<uint64_t>::input(StringRef Scalar, void *,
+ uint64_t &Val) {
+ unsigned long long N;
+ if (getAsUnsignedInteger(Scalar, 0, N))
+ return "invalid number";
+ Val = N;
+ return StringRef();
+}
+
+void ScalarTraits<int8_t>::output(const int8_t &Val, void *, raw_ostream &Out) {
+ // use temp in32_t because ostream thinks int8_t is a character
+ int32_t Num = Val;
+ Out << Num;
+}
+
+StringRef ScalarTraits<int8_t>::input(StringRef Scalar, void *, int8_t &Val) {
+ long long N;
+ if (getAsSignedInteger(Scalar, 0, N))
+ return "invalid number";
+ if ((N > 127) || (N < -128))
+ return "out of range number";
+ Val = N;
+ return StringRef();
+}
+
+void ScalarTraits<int16_t>::output(const int16_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<int16_t>::input(StringRef Scalar, void *, int16_t &Val) {
+ long long N;
+ if (getAsSignedInteger(Scalar, 0, N))
+ return "invalid number";
+ if ((N > INT16_MAX) || (N < INT16_MIN))
+ return "out of range number";
+ Val = N;
+ return StringRef();
+}
+
+void ScalarTraits<int32_t>::output(const int32_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<int32_t>::input(StringRef Scalar, void *, int32_t &Val) {
+ long long N;
+ if (getAsSignedInteger(Scalar, 0, N))
+ return "invalid number";
+ if ((N > INT32_MAX) || (N < INT32_MIN))
+ return "out of range number";
+ Val = N;
+ return StringRef();
+}
+
+void ScalarTraits<int64_t>::output(const int64_t &Val, void *,
+ raw_ostream &Out) {
+ Out << Val;
+}
+
+StringRef ScalarTraits<int64_t>::input(StringRef Scalar, void *, int64_t &Val) {
+ long long N;
+ if (getAsSignedInteger(Scalar, 0, N))
+ return "invalid number";
+ Val = N;
+ return StringRef();
+}
+
+void ScalarTraits<double>::output(const double &Val, void *, raw_ostream &Out) {
+ Out << format("%g", Val);
+}
+
+StringRef ScalarTraits<double>::input(StringRef Scalar, void *, double &Val) {
+ if (to_float(Scalar, Val))
+ return StringRef();
+ return "invalid floating point number";
+}
+
+void ScalarTraits<float>::output(const float &Val, void *, raw_ostream &Out) {
+ Out << format("%g", Val);
+}
+
+StringRef ScalarTraits<float>::input(StringRef Scalar, void *, float &Val) {
+ if (to_float(Scalar, Val))
+ return StringRef();
+ return "invalid floating point number";
+}
+
+void ScalarTraits<Hex8>::output(const Hex8 &Val, void *, raw_ostream &Out) {
+ uint8_t Num = Val;
+ Out << format("0x%02X", Num);
+}
+
+StringRef ScalarTraits<Hex8>::input(StringRef Scalar, void *, Hex8 &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid hex8 number";
+ if (n > 0xFF)
+ return "out of range hex8 number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<Hex16>::output(const Hex16 &Val, void *, raw_ostream &Out) {
+ uint16_t Num = Val;
+ Out << format("0x%04X", Num);
+}
+
+StringRef ScalarTraits<Hex16>::input(StringRef Scalar, void *, Hex16 &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid hex16 number";
+ if (n > 0xFFFF)
+ return "out of range hex16 number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<Hex32>::output(const Hex32 &Val, void *, raw_ostream &Out) {
+ uint32_t Num = Val;
+ Out << format("0x%08X", Num);
+}
+
+StringRef ScalarTraits<Hex32>::input(StringRef Scalar, void *, Hex32 &Val) {
+ unsigned long long n;
+ if (getAsUnsignedInteger(Scalar, 0, n))
+ return "invalid hex32 number";
+ if (n > 0xFFFFFFFFUL)
+ return "out of range hex32 number";
+ Val = n;
+ return StringRef();
+}
+
+void ScalarTraits<Hex64>::output(const Hex64 &Val, void *, raw_ostream &Out) {
+ uint64_t Num = Val;
+ Out << format("0x%016llX", Num);
+}
+
+StringRef ScalarTraits<Hex64>::input(StringRef Scalar, void *, Hex64 &Val) {
+ unsigned long long Num;
+ if (getAsUnsignedInteger(Scalar, 0, Num))
+ return "invalid hex64 number";
+ Val = Num;
+ return StringRef();
+}
diff --git a/third_party/llvm-project/dwarf2yaml.cpp b/third_party/llvm-project/dwarf2yaml.cpp
new file mode 100644
index 000000000..bdaf1e429
--- /dev/null
+++ b/third_party/llvm-project/dwarf2yaml.cpp
@@ -0,0 +1,385 @@
+//===------ dwarf2yaml.cpp - obj2yaml conversion tool -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "Error.h"
+#include "llvm/DebugInfo/DWARF/DWARFContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/ObjectYAML/DWARFYAML.h"
+
+#include <algorithm>
+
+using namespace llvm;
+
+void dumpInitialLength(DataExtractor &Data, uint64_t &Offset,
+ DWARFYAML::InitialLength &InitialLength) {
+ InitialLength.TotalLength = Data.getU32(&Offset);
+ if (InitialLength.isDWARF64())
+ InitialLength.TotalLength64 = Data.getU64(&Offset);
+}
+
+void dumpDebugAbbrev(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ auto AbbrevSetPtr = DCtx.getDebugAbbrev();
+ if (AbbrevSetPtr) {
+ for (auto AbbrvDeclSet : *AbbrevSetPtr) {
+ for (auto AbbrvDecl : AbbrvDeclSet.second) {
+ DWARFYAML::Abbrev Abbrv;
+ Abbrv.Code = AbbrvDecl.getCode();
+ Abbrv.Tag = AbbrvDecl.getTag();
+ Abbrv.Children = AbbrvDecl.hasChildren() ? dwarf::DW_CHILDREN_yes
+ : dwarf::DW_CHILDREN_no;
+ for (auto Attribute : AbbrvDecl.attributes()) {
+ DWARFYAML::AttributeAbbrev AttAbrv;
+ AttAbrv.Attribute = Attribute.Attr;
+ AttAbrv.Form = Attribute.Form;
+ if (AttAbrv.Form == dwarf::DW_FORM_implicit_const)
+ AttAbrv.Value = Attribute.getImplicitConstValue();
+ Abbrv.Attributes.push_back(AttAbrv);
+ }
+ Y.AbbrevDecls.push_back(Abbrv);
+ }
+ }
+ }
+}
+
+void dumpDebugStrings(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ StringRef RemainingTable = DCtx.getDWARFObj().getStrSection();
+ while (RemainingTable.size() > 0) {
+ auto SymbolPair = RemainingTable.split('\0');
+ RemainingTable = SymbolPair.second;
+ Y.DebugStrings.push_back(SymbolPair.first);
+ }
+}
+
+void dumpDebugARanges(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ DataExtractor ArangesData(DCtx.getDWARFObj().getArangesSection(),
+ DCtx.isLittleEndian(), 0);
+ uint64_t Offset = 0;
+ DWARFDebugArangeSet Set;
+
+ while (Set.extract(ArangesData, &Offset)) {
+ DWARFYAML::ARange Range;
+ Range.Length.setLength(Set.getHeader().Length);
+ Range.Version = Set.getHeader().Version;
+ Range.CuOffset = Set.getHeader().CuOffset;
+ Range.AddrSize = Set.getHeader().AddrSize;
+ Range.SegSize = Set.getHeader().SegSize;
+ for (auto Descriptor : Set.descriptors()) {
+ DWARFYAML::ARangeDescriptor Desc;
+ Desc.Address = Descriptor.Address;
+ Desc.Length = Descriptor.Length;
+ Range.Descriptors.push_back(Desc);
+ }
+ Y.ARanges.push_back(Range);
+ }
+}
+
+void dumpDebugRanges(DWARFContext &DCtx, DWARFYAML::Data &Y) { // XXX BINARYEN
+ uint8_t savedAddressByteSize = 4;
+ DWARFDataExtractor rangesData(DCtx.getDWARFObj(), DCtx.getDWARFObj().getRangesSection(),
+ DCtx.isLittleEndian(), savedAddressByteSize);
+ uint64_t offset = 0;
+ DWARFDebugRangeList rangeList;
+ while (rangesData.isValidOffset(offset)) {
+ if (Error E = rangeList.extract(rangesData, &offset)) {
+ errs() << toString(std::move(E)) << '\n';
+ break;
+ }
+ for (auto& entry : rangeList.getEntries()) {
+ DWARFYAML::Range range;
+ range.Start = entry.StartAddress;
+ range.End = entry.EndAddress;
+ range.SectionIndex = entry.SectionIndex;
+ Y.Ranges.push_back(range);
+ }
+ DWARFYAML::Range range;
+ range.Start = 0;
+ range.End = 0;
+ range.SectionIndex = -1;
+ Y.Ranges.push_back(range);
+ }
+}
+
+void dumpPubSection(DWARFContext &DCtx, DWARFYAML::PubSection &Y,
+ DWARFSection Section) {
+ DWARFDataExtractor PubSectionData(DCtx.getDWARFObj(), Section,
+ DCtx.isLittleEndian(), 0);
+ uint64_t Offset = 0;
+ dumpInitialLength(PubSectionData, Offset, Y.Length);
+ Y.Version = PubSectionData.getU16(&Offset);
+ Y.UnitOffset = PubSectionData.getU32(&Offset);
+ Y.UnitSize = PubSectionData.getU32(&Offset);
+ while (Offset < Y.Length.getLength()) {
+ DWARFYAML::PubEntry NewEntry;
+ NewEntry.DieOffset = PubSectionData.getU32(&Offset);
+ if (Y.IsGNUStyle)
+ NewEntry.Descriptor = PubSectionData.getU8(&Offset);
+ NewEntry.Name = PubSectionData.getCStr(&Offset);
+ Y.Entries.push_back(NewEntry);
+ }
+}
+
+void dumpDebugPubSections(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ const DWARFObject &D = DCtx.getDWARFObj();
+ Y.PubNames.IsGNUStyle = false;
+ dumpPubSection(DCtx, Y.PubNames, D.getPubnamesSection());
+
+ Y.PubTypes.IsGNUStyle = false;
+ dumpPubSection(DCtx, Y.PubTypes, D.getPubtypesSection());
+
+ Y.GNUPubNames.IsGNUStyle = true;
+ dumpPubSection(DCtx, Y.GNUPubNames, D.getGnuPubnamesSection());
+
+ Y.GNUPubTypes.IsGNUStyle = true;
+ dumpPubSection(DCtx, Y.GNUPubTypes, D.getGnuPubtypesSection());
+}
+
+void dumpDebugInfo(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ for (const auto &CU : DCtx.compile_units()) {
+ DWARFYAML::Unit NewUnit;
+ NewUnit.Length.setLength(CU->getLength());
+ NewUnit.Version = CU->getVersion();
+ if(NewUnit.Version >= 5)
+ NewUnit.Type = (dwarf::UnitType)CU->getUnitType();
+ NewUnit.AbbrOffset = CU->getAbbreviations()->getOffset();
+ NewUnit.AddrSize = CU->getAddressByteSize();
+ for (auto DIE : CU->dies()) {
+ DWARFYAML::Entry NewEntry;
+ DataExtractor EntryData = CU->getDebugInfoExtractor();
+ uint64_t offset = DIE.getOffset();
+
+ assert(EntryData.isValidOffset(offset) && "Invalid DIE Offset");
+ if (!EntryData.isValidOffset(offset))
+ continue;
+
+ NewEntry.AbbrCode = EntryData.getULEB128(&offset);
+
+ auto AbbrevDecl = DIE.getAbbreviationDeclarationPtr();
+ if (AbbrevDecl) {
+ for (const auto &AttrSpec : AbbrevDecl->attributes()) {
+ DWARFYAML::FormValue NewValue;
+ NewValue.Value = 0xDEADBEEFDEADBEEF;
+ DWARFDie DIEWrapper(CU.get(), &DIE);
+ auto FormValue = DIEWrapper.find(AttrSpec.Attr);
+ if (!FormValue)
+ return;
+ auto Form = FormValue.getValue().getForm();
+ bool indirect = false;
+ do {
+ indirect = false;
+ switch (Form) {
+ case dwarf::DW_FORM_addr:
+ case dwarf::DW_FORM_GNU_addr_index:
+ if (auto Val = FormValue.getValue().getAsAddress())
+ NewValue.Value = Val.getValue();
+ break;
+ case dwarf::DW_FORM_ref_addr:
+ case dwarf::DW_FORM_ref1:
+ case dwarf::DW_FORM_ref2:
+ case dwarf::DW_FORM_ref4:
+ case dwarf::DW_FORM_ref8:
+ case dwarf::DW_FORM_ref_udata:
+ case dwarf::DW_FORM_ref_sig8:
+ if (auto Val = FormValue.getValue().getAsReferenceUVal())
+ NewValue.Value = Val.getValue();
+ break;
+ case dwarf::DW_FORM_exprloc:
+ case dwarf::DW_FORM_block:
+ case dwarf::DW_FORM_block1:
+ case dwarf::DW_FORM_block2:
+ case dwarf::DW_FORM_block4:
+ if (auto Val = FormValue.getValue().getAsBlock()) {
+ auto BlockData = Val.getValue();
+ std::copy(BlockData.begin(), BlockData.end(),
+ std::back_inserter(NewValue.BlockData));
+ }
+ NewValue.Value = NewValue.BlockData.size();
+ break;
+ case dwarf::DW_FORM_data1:
+ case dwarf::DW_FORM_flag:
+ case dwarf::DW_FORM_data2:
+ case dwarf::DW_FORM_data4:
+ case dwarf::DW_FORM_data8:
+ case dwarf::DW_FORM_sdata:
+ case dwarf::DW_FORM_udata:
+ case dwarf::DW_FORM_ref_sup4:
+ case dwarf::DW_FORM_ref_sup8:
+ if (auto Val = FormValue.getValue().getAsUnsignedConstant())
+ NewValue.Value = Val.getValue();
+ break;
+ case dwarf::DW_FORM_string:
+ if (auto Val = FormValue.getValue().getAsCString())
+ NewValue.CStr = Val.getValue();
+ break;
+ case dwarf::DW_FORM_indirect:
+ indirect = true;
+ if (auto Val = FormValue.getValue().getAsUnsignedConstant()) {
+ NewValue.Value = Val.getValue();
+ NewEntry.Values.push_back(NewValue);
+ Form = static_cast<dwarf::Form>(Val.getValue());
+ }
+ break;
+ case dwarf::DW_FORM_strp:
+ case dwarf::DW_FORM_sec_offset:
+ case dwarf::DW_FORM_GNU_ref_alt:
+ case dwarf::DW_FORM_GNU_strp_alt:
+ case dwarf::DW_FORM_line_strp:
+ case dwarf::DW_FORM_strp_sup:
+ case dwarf::DW_FORM_GNU_str_index:
+ case dwarf::DW_FORM_strx:
+ if (auto Val = FormValue.getValue().getAsCStringOffset())
+ NewValue.Value = Val.getValue();
+ break;
+ case dwarf::DW_FORM_flag_present:
+ NewValue.Value = 1;
+ break;
+ default:
+ break;
+ }
+ } while (indirect);
+ NewEntry.Values.push_back(NewValue);
+ }
+ }
+
+ NewUnit.Entries.push_back(NewEntry);
+ }
+ Y.CompileUnits.push_back(NewUnit);
+ }
+}
+
+bool dumpFileEntry(DataExtractor &Data, uint64_t &Offset,
+ DWARFYAML::File &File) {
+ File.Name = Data.getCStr(&Offset);
+ if (File.Name.empty())
+ return false;
+ File.DirIdx = Data.getULEB128(&Offset);
+ File.ModTime = Data.getULEB128(&Offset);
+ File.Length = Data.getULEB128(&Offset);
+ return true;
+}
+
+void dumpDebugLines(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ for (const auto &CU : DCtx.compile_units()) {
+ auto CUDIE = CU->getUnitDIE();
+ if (!CUDIE)
+ continue;
+ if (auto StmtOffset =
+ dwarf::toSectionOffset(CUDIE.find(dwarf::DW_AT_stmt_list))) {
+ DWARFYAML::LineTable DebugLines;
+ DataExtractor LineData(DCtx.getDWARFObj().getLineSection().Data,
+ DCtx.isLittleEndian(), CU->getAddressByteSize());
+ uint64_t Offset = *StmtOffset;
+ dumpInitialLength(LineData, Offset, DebugLines.Length);
+ uint64_t LineTableLength = DebugLines.Length.getLength();
+ uint64_t SizeOfPrologueLength = DebugLines.Length.isDWARF64() ? 8 : 4;
+ DebugLines.Version = LineData.getU16(&Offset);
+ DebugLines.PrologueLength =
+ LineData.getUnsigned(&Offset, SizeOfPrologueLength);
+ const uint64_t EndPrologue = DebugLines.PrologueLength + Offset;
+
+ DebugLines.MinInstLength = LineData.getU8(&Offset);
+ if (DebugLines.Version >= 4)
+ DebugLines.MaxOpsPerInst = LineData.getU8(&Offset);
+ DebugLines.DefaultIsStmt = LineData.getU8(&Offset);
+ DebugLines.LineBase = LineData.getU8(&Offset);
+ DebugLines.LineRange = LineData.getU8(&Offset);
+ DebugLines.OpcodeBase = LineData.getU8(&Offset);
+
+ DebugLines.StandardOpcodeLengths.reserve(DebugLines.OpcodeBase - 1);
+ for (uint8_t i = 1; i < DebugLines.OpcodeBase; ++i)
+ DebugLines.StandardOpcodeLengths.push_back(LineData.getU8(&Offset));
+
+ while (Offset < EndPrologue) {
+ StringRef Dir = LineData.getCStr(&Offset);
+ if (!Dir.empty())
+ DebugLines.IncludeDirs.push_back(Dir);
+ else
+ break;
+ }
+
+ while (Offset < EndPrologue) {
+ DWARFYAML::File TmpFile;
+ if (dumpFileEntry(LineData, Offset, TmpFile))
+ DebugLines.Files.push_back(TmpFile);
+ else
+ break;
+ }
+
+ const uint64_t LineEnd =
+ LineTableLength + *StmtOffset + SizeOfPrologueLength;
+ while (Offset < LineEnd) {
+ DWARFYAML::LineTableOpcode NewOp = {};
+ NewOp.Opcode = (dwarf::LineNumberOps)LineData.getU8(&Offset);
+ if (NewOp.Opcode == 0) {
+ auto StartExt = Offset;
+ NewOp.ExtLen = LineData.getULEB128(&Offset);
+ NewOp.SubOpcode =
+ (dwarf::LineNumberExtendedOps)LineData.getU8(&Offset);
+ switch (NewOp.SubOpcode) {
+ case dwarf::DW_LNE_set_address:
+ case dwarf::DW_LNE_set_discriminator:
+ NewOp.Data = LineData.getAddress(&Offset);
+ break;
+ case dwarf::DW_LNE_define_file:
+ dumpFileEntry(LineData, Offset, NewOp.FileEntry);
+ break;
+ case dwarf::DW_LNE_end_sequence:
+ break;
+ default:
+ while (Offset < StartExt + NewOp.ExtLen)
+ NewOp.UnknownOpcodeData.push_back(LineData.getU8(&Offset));
+ }
+ } else if (NewOp.Opcode < DebugLines.OpcodeBase) {
+ switch (NewOp.Opcode) {
+ case dwarf::DW_LNS_copy:
+ case dwarf::DW_LNS_negate_stmt:
+ case dwarf::DW_LNS_set_basic_block:
+ case dwarf::DW_LNS_const_add_pc:
+ case dwarf::DW_LNS_set_prologue_end:
+ case dwarf::DW_LNS_set_epilogue_begin:
+ break;
+
+ case dwarf::DW_LNS_advance_pc:
+ case dwarf::DW_LNS_set_file:
+ case dwarf::DW_LNS_set_column:
+ case dwarf::DW_LNS_set_isa:
+ NewOp.Data = LineData.getULEB128(&Offset);
+ break;
+
+ case dwarf::DW_LNS_advance_line:
+ NewOp.SData = LineData.getSLEB128(&Offset);
+ break;
+
+ case dwarf::DW_LNS_fixed_advance_pc:
+ NewOp.Data = LineData.getU16(&Offset);
+ break;
+
+ default:
+ for (uint8_t i = 0;
+ i < DebugLines.StandardOpcodeLengths[NewOp.Opcode - 1]; ++i)
+ NewOp.StandardOpcodeData.push_back(LineData.getULEB128(&Offset));
+ }
+ }
+ DebugLines.Opcodes.push_back(NewOp);
+ }
+ Y.DebugLines.push_back(DebugLines);
+ }
+ }
+}
+
+std::error_code dwarf2yaml(DWARFContext &DCtx, DWARFYAML::Data &Y) {
+ Y.IsLittleEndian = true; // XXX BINARYEN
+ dumpDebugAbbrev(DCtx, Y);
+ dumpDebugStrings(DCtx, Y);
+ dumpDebugARanges(DCtx, Y);
+ dumpDebugRanges(DCtx, Y); // XXX BINARYEN
+ dumpDebugPubSections(DCtx, Y);
+ dumpDebugInfo(DCtx, Y);
+ dumpDebugLines(DCtx, Y);
+ return obj2yaml_error::success;
+}
diff --git a/third_party/llvm-project/include/llvm-c/DataTypes.h b/third_party/llvm-project/include/llvm-c/DataTypes.h
new file mode 100644
index 000000000..893b22b49
--- /dev/null
+++ b/third_party/llvm-project/include/llvm-c/DataTypes.h
@@ -0,0 +1,90 @@
+/*===-- include/llvm-c/DataTypes.h - Define fixed size types ------*- C -*-===*\
+|* *|
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM *|
+|* Exceptions. *|
+|* See https://llvm.org/LICENSE.txt for license information. *|
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception *|
+|* *|
+|*===----------------------------------------------------------------------===*|
+|* *|
+|* This file contains definitions to figure out the size of _HOST_ data types.*|
+|* This file is important because different host OS's define different macros,*|
+|* which makes portability tough. This file exports the following *|
+|* definitions: *|
+|* *|
+|* [u]int(32|64)_t : typedefs for signed and unsigned 32/64 bit system types*|
+|* [U]INT(8|16|32|64)_(MIN|MAX) : Constants for the min and max values. *|
+|* *|
+|* No library is required when using these functions. *|
+|* *|
+|*===----------------------------------------------------------------------===*/
+
+/* Please leave this file C-compatible. */
+
+#ifndef LLVM_C_DATATYPES_H
+#define LLVM_C_DATATYPES_H
+
+#ifdef __cplusplus
+#include <cmath>
+#else
+#include <math.h>
+#endif
+
+#include <inttypes.h>
+#include <stdint.h>
+
+#ifndef _MSC_VER
+
+#if !defined(UINT32_MAX)
+# error "The standard header <cstdint> is not C++11 compliant. Must #define "\
+ "__STDC_LIMIT_MACROS before #including llvm-c/DataTypes.h"
+#endif
+
+#if !defined(UINT32_C)
+# error "The standard header <cstdint> is not C++11 compliant. Must #define "\
+ "__STDC_CONSTANT_MACROS before #including llvm-c/DataTypes.h"
+#endif
+
+/* Note that <inttypes.h> includes <stdint.h>, if this is a C99 system. */
+#include <sys/types.h>
+
+#ifdef _AIX
+// GCC is strict about defining large constants: they must have LL modifier.
+#undef INT64_MAX
+#undef INT64_MIN
+#endif
+
+#else /* _MSC_VER */
+#ifdef __cplusplus
+#include <cstddef>
+#include <cstdlib>
+#else
+#include <stddef.h>
+#include <stdlib.h>
+#endif
+#include <sys/types.h>
+
+#if defined(_WIN64)
+typedef signed __int64 ssize_t;
+#else
+typedef signed int ssize_t;
+#endif /* _WIN64 */
+
+#endif /* _MSC_VER */
+
+/* Set defaults for constants which we cannot find. */
+#if !defined(INT64_MAX)
+# define INT64_MAX 9223372036854775807LL
+#endif
+#if !defined(INT64_MIN)
+# define INT64_MIN ((-INT64_MAX)-1)
+#endif
+#if !defined(UINT64_MAX)
+# define UINT64_MAX 0xffffffffffffffffULL
+#endif
+
+#ifndef HUGE_VALF
+#define HUGE_VALF (float)HUGE_VAL
+#endif
+
+#endif /* LLVM_C_DATATYPES_H */
diff --git a/third_party/llvm-project/include/llvm-c/DisassemblerTypes.h b/third_party/llvm-project/include/llvm-c/DisassemblerTypes.h
new file mode 100644
index 000000000..389e5ee45
--- /dev/null
+++ b/third_party/llvm-project/include/llvm-c/DisassemblerTypes.h
@@ -0,0 +1,160 @@
+/*===-- llvm-c/DisassemblerTypedefs.h -----------------------------*- C -*-===*\
+|* *|
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM *|
+|* Exceptions. *|
+|* See https://llvm.org/LICENSE.txt for license information. *|
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception *|
+|* *|
+|*===----------------------------------------------------------------------===*/
+
+#ifndef LLVM_DISASSEMBLER_TYPES_H
+#define LLVM_DISASSEMBLER_TYPES_H
+
+#include "llvm-c/DataTypes.h"
+#ifdef __cplusplus
+#include <cstddef>
+#else
+#include <stddef.h>
+#endif
+
+/**
+ * An opaque reference to a disassembler context.
+ */
+typedef void *LLVMDisasmContextRef;
+
+/**
+ * The type for the operand information call back function. This is called to
+ * get the symbolic information for an operand of an instruction. Typically
+ * this is from the relocation information, symbol table, etc. That block of
+ * information is saved when the disassembler context is created and passed to
+ * the call back in the DisInfo parameter. The instruction containing operand
+ * is at the PC parameter. For some instruction sets, there can be more than
+ * one operand with symbolic information. To determine the symbolic operand
+ * information for each operand, the bytes for the specific operand in the
+ * instruction are specified by the Offset parameter and its byte widith is the
+ * size parameter. For instructions sets with fixed widths and one symbolic
+ * operand per instruction, the Offset parameter will be zero and Size parameter
+ * will be the instruction width. The information is returned in TagBuf and is
+ * Triple specific with its specific information defined by the value of
+ * TagType for that Triple. If symbolic information is returned the function
+ * returns 1, otherwise it returns 0.
+ */
+typedef int (*LLVMOpInfoCallback)(void *DisInfo, uint64_t PC,
+ uint64_t Offset, uint64_t Size,
+ int TagType, void *TagBuf);
+
+/**
+ * The initial support in LLVM MC for the most general form of a relocatable
+ * expression is "AddSymbol - SubtractSymbol + Offset". For some Darwin targets
+ * this full form is encoded in the relocation information so that AddSymbol and
+ * SubtractSymbol can be link edited independent of each other. Many other
+ * platforms only allow a relocatable expression of the form AddSymbol + Offset
+ * to be encoded.
+ *
+ * The LLVMOpInfoCallback() for the TagType value of 1 uses the struct
+ * LLVMOpInfo1. The value of the relocatable expression for the operand,
+ * including any PC adjustment, is passed in to the call back in the Value
+ * field. The symbolic information about the operand is returned using all
+ * the fields of the structure with the Offset of the relocatable expression
+ * returned in the Value field. It is possible that some symbols in the
+ * relocatable expression were assembly temporary symbols, for example
+ * "Ldata - LpicBase + constant", and only the Values of the symbols without
+ * symbol names are present in the relocation information. The VariantKind
+ * type is one of the Target specific #defines below and is used to print
+ * operands like "_foo@GOT", ":lower16:_foo", etc.
+ */
+struct LLVMOpInfoSymbol1 {
+ uint64_t Present; /* 1 if this symbol is present */
+ const char *Name; /* symbol name if not NULL */
+ uint64_t Value; /* symbol value if name is NULL */
+};
+
+struct LLVMOpInfo1 {
+ struct LLVMOpInfoSymbol1 AddSymbol;
+ struct LLVMOpInfoSymbol1 SubtractSymbol;
+ uint64_t Value;
+ uint64_t VariantKind;
+};
+
+/**
+ * The operand VariantKinds for symbolic disassembly.
+ */
+#define LLVMDisassembler_VariantKind_None 0 /* all targets */
+
+/**
+ * The ARM target VariantKinds.
+ */
+#define LLVMDisassembler_VariantKind_ARM_HI16 1 /* :upper16: */
+#define LLVMDisassembler_VariantKind_ARM_LO16 2 /* :lower16: */
+
+/**
+ * The ARM64 target VariantKinds.
+ */
+#define LLVMDisassembler_VariantKind_ARM64_PAGE 1 /* @page */
+#define LLVMDisassembler_VariantKind_ARM64_PAGEOFF 2 /* @pageoff */
+#define LLVMDisassembler_VariantKind_ARM64_GOTPAGE 3 /* @gotpage */
+#define LLVMDisassembler_VariantKind_ARM64_GOTPAGEOFF 4 /* @gotpageoff */
+#define LLVMDisassembler_VariantKind_ARM64_TLVP 5 /* @tvlppage */
+#define LLVMDisassembler_VariantKind_ARM64_TLVOFF 6 /* @tvlppageoff */
+
+/**
+ * The type for the symbol lookup function. This may be called by the
+ * disassembler for things like adding a comment for a PC plus a constant
+ * offset load instruction to use a symbol name instead of a load address value.
+ * It is passed the block information is saved when the disassembler context is
+ * created and the ReferenceValue to look up as a symbol. If no symbol is found
+ * for the ReferenceValue NULL is returned. The ReferenceType of the
+ * instruction is passed indirectly as is the PC of the instruction in
+ * ReferencePC. If the output reference can be determined its type is returned
+ * indirectly in ReferenceType along with ReferenceName if any, or that is set
+ * to NULL.
+ */
+typedef const char *(*LLVMSymbolLookupCallback)(void *DisInfo,
+ uint64_t ReferenceValue,
+ uint64_t *ReferenceType,
+ uint64_t ReferencePC,
+ const char **ReferenceName);
+/**
+ * The reference types on input and output.
+ */
+/* No input reference type or no output reference type. */
+#define LLVMDisassembler_ReferenceType_InOut_None 0
+
+/* The input reference is from a branch instruction. */
+#define LLVMDisassembler_ReferenceType_In_Branch 1
+/* The input reference is from a PC relative load instruction. */
+#define LLVMDisassembler_ReferenceType_In_PCrel_Load 2
+
+/* The input reference is from an ARM64::ADRP instruction. */
+#define LLVMDisassembler_ReferenceType_In_ARM64_ADRP 0x100000001
+/* The input reference is from an ARM64::ADDXri instruction. */
+#define LLVMDisassembler_ReferenceType_In_ARM64_ADDXri 0x100000002
+/* The input reference is from an ARM64::LDRXui instruction. */
+#define LLVMDisassembler_ReferenceType_In_ARM64_LDRXui 0x100000003
+/* The input reference is from an ARM64::LDRXl instruction. */
+#define LLVMDisassembler_ReferenceType_In_ARM64_LDRXl 0x100000004
+/* The input reference is from an ARM64::ADR instruction. */
+#define LLVMDisassembler_ReferenceType_In_ARM64_ADR 0x100000005
+
+/* The output reference is to as symbol stub. */
+#define LLVMDisassembler_ReferenceType_Out_SymbolStub 1
+/* The output reference is to a symbol address in a literal pool. */
+#define LLVMDisassembler_ReferenceType_Out_LitPool_SymAddr 2
+/* The output reference is to a cstring address in a literal pool. */
+#define LLVMDisassembler_ReferenceType_Out_LitPool_CstrAddr 3
+
+/* The output reference is to a Objective-C CoreFoundation string. */
+#define LLVMDisassembler_ReferenceType_Out_Objc_CFString_Ref 4
+/* The output reference is to a Objective-C message. */
+#define LLVMDisassembler_ReferenceType_Out_Objc_Message 5
+/* The output reference is to a Objective-C message ref. */
+#define LLVMDisassembler_ReferenceType_Out_Objc_Message_Ref 6
+/* The output reference is to a Objective-C selector ref. */
+#define LLVMDisassembler_ReferenceType_Out_Objc_Selector_Ref 7
+/* The output reference is to a Objective-C class ref. */
+#define LLVMDisassembler_ReferenceType_Out_Objc_Class_Ref 8
+
+/* The output reference is to a C++ symbol name. */
+#define LLVMDisassembler_ReferenceType_DeMangled_Name 9
+
+#endif
diff --git a/third_party/llvm-project/include/llvm-c/Error.h b/third_party/llvm-project/include/llvm-c/Error.h
new file mode 100644
index 000000000..52943063c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm-c/Error.h
@@ -0,0 +1,69 @@
+/*===------- llvm-c/Error.h - llvm::Error class C Interface -------*- C -*-===*\
+|* *|
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM *|
+|* Exceptions. *|
+|* See https://llvm.org/LICENSE.txt for license information. *|
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception *|
+|* *|
+|*===----------------------------------------------------------------------===*|
+|* *|
+|* This file defines the C interface to LLVM's Error class. *|
+|* *|
+\*===----------------------------------------------------------------------===*/
+
+#ifndef LLVM_C_ERROR_H
+#define LLVM_C_ERROR_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define LLVMErrorSuccess 0
+
+/**
+ * Opaque reference to an error instance. Null serves as the 'success' value.
+ */
+typedef struct LLVMOpaqueError *LLVMErrorRef;
+
+/**
+ * Error type identifier.
+ */
+typedef const void *LLVMErrorTypeId;
+
+/**
+ * Returns the type id for the given error instance, which must be a failure
+ * value (i.e. non-null).
+ */
+LLVMErrorTypeId LLVMGetErrorTypeId(LLVMErrorRef Err);
+
+/**
+ * Dispose of the given error without handling it. This operation consumes the
+ * error, and the given LLVMErrorRef value is not usable once this call returns.
+ * Note: This method *only* needs to be called if the error is not being passed
+ * to some other consuming operation, e.g. LLVMGetErrorMessage.
+ */
+void LLVMConsumeError(LLVMErrorRef Err);
+
+/**
+ * Returns the given string's error message. This operation consumes the error,
+ * and the given LLVMErrorRef value is not usable once this call returns.
+ * The caller is responsible for disposing of the string by calling
+ * LLVMDisposeErrorMessage.
+ */
+char *LLVMGetErrorMessage(LLVMErrorRef Err);
+
+/**
+ * Dispose of the given error message.
+ */
+void LLVMDisposeErrorMessage(char *ErrMsg);
+
+/**
+ * Returns the type id for llvm StringError.
+ */
+LLVMErrorTypeId LLVMGetStringErrorTypeId(void);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/third_party/llvm-project/include/llvm-c/ErrorHandling.h b/third_party/llvm-project/include/llvm-c/ErrorHandling.h
new file mode 100644
index 000000000..4927349d8
--- /dev/null
+++ b/third_party/llvm-project/include/llvm-c/ErrorHandling.h
@@ -0,0 +1,49 @@
+/*===-- llvm-c/ErrorHandling.h - Error Handling C Interface -------*- C -*-===*\
+|* *|
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM *|
+|* Exceptions. *|
+|* See https://llvm.org/LICENSE.txt for license information. *|
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception *|
+|* *|
+|*===----------------------------------------------------------------------===*|
+|* *|
+|* This file defines the C interface to LLVM's error handling mechanism. *|
+|* *|
+\*===----------------------------------------------------------------------===*/
+
+#ifndef LLVM_C_ERROR_HANDLING_H
+#define LLVM_C_ERROR_HANDLING_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef void (*LLVMFatalErrorHandler)(const char *Reason);
+
+/**
+ * Install a fatal error handler. By default, if LLVM detects a fatal error, it
+ * will call exit(1). This may not be appropriate in many contexts. For example,
+ * doing exit(1) will bypass many crash reporting/tracing system tools. This
+ * function allows you to install a callback that will be invoked prior to the
+ * call to exit(1).
+ */
+void LLVMInstallFatalErrorHandler(LLVMFatalErrorHandler Handler);
+
+/**
+ * Reset the fatal error handler. This resets LLVM's fatal error handling
+ * behavior to the default.
+ */
+void LLVMResetFatalErrorHandler(void);
+
+/**
+ * Enable LLVM's built-in stack trace code. This intercepts the OS's crash
+ * signals and prints which component of LLVM you were in at the time if the
+ * crash.
+ */
+void LLVMEnablePrettyStackTrace(void);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/third_party/llvm-project/include/llvm-c/Types.h b/third_party/llvm-project/include/llvm-c/Types.h
new file mode 100644
index 000000000..612c7d3ef
--- /dev/null
+++ b/third_party/llvm-project/include/llvm-c/Types.h
@@ -0,0 +1,179 @@
+/*===-- llvm-c/Support.h - C Interface Types declarations ---------*- C -*-===*\
+|* *|
+|* Part of the LLVM Project, under the Apache License v2.0 with LLVM *|
+|* Exceptions. *|
+|* See https://llvm.org/LICENSE.txt for license information. *|
+|* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception *|
+|* *|
+|*===----------------------------------------------------------------------===*|
+|* *|
+|* This file defines types used by the C interface to LLVM. *|
+|* *|
+\*===----------------------------------------------------------------------===*/
+
+#ifndef LLVM_C_TYPES_H
+#define LLVM_C_TYPES_H
+
+#include "llvm-c/DataTypes.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * @defgroup LLVMCSupportTypes Types and Enumerations
+ *
+ * @{
+ */
+
+typedef int LLVMBool;
+
+/* Opaque types. */
+
+/**
+ * LLVM uses a polymorphic type hierarchy which C cannot represent, therefore
+ * parameters must be passed as base types. Despite the declared types, most
+ * of the functions provided operate only on branches of the type hierarchy.
+ * The declared parameter names are descriptive and specify which type is
+ * required. Additionally, each type hierarchy is documented along with the
+ * functions that operate upon it. For more detail, refer to LLVM's C++ code.
+ * If in doubt, refer to Core.cpp, which performs parameter downcasts in the
+ * form unwrap<RequiredType>(Param).
+ */
+
+/**
+ * Used to pass regions of memory through LLVM interfaces.
+ *
+ * @see llvm::MemoryBuffer
+ */
+typedef struct LLVMOpaqueMemoryBuffer *LLVMMemoryBufferRef;
+
+/**
+ * The top-level container for all LLVM global data. See the LLVMContext class.
+ */
+typedef struct LLVMOpaqueContext *LLVMContextRef;
+
+/**
+ * The top-level container for all other LLVM Intermediate Representation (IR)
+ * objects.
+ *
+ * @see llvm::Module
+ */
+typedef struct LLVMOpaqueModule *LLVMModuleRef;
+
+/**
+ * Each value in the LLVM IR has a type, an LLVMTypeRef.
+ *
+ * @see llvm::Type
+ */
+typedef struct LLVMOpaqueType *LLVMTypeRef;
+
+/**
+ * Represents an individual value in LLVM IR.
+ *
+ * This models llvm::Value.
+ */
+typedef struct LLVMOpaqueValue *LLVMValueRef;
+
+/**
+ * Represents a basic block of instructions in LLVM IR.
+ *
+ * This models llvm::BasicBlock.
+ */
+typedef struct LLVMOpaqueBasicBlock *LLVMBasicBlockRef;
+
+/**
+ * Represents an LLVM Metadata.
+ *
+ * This models llvm::Metadata.
+ */
+typedef struct LLVMOpaqueMetadata *LLVMMetadataRef;
+
+/**
+ * Represents an LLVM Named Metadata Node.
+ *
+ * This models llvm::NamedMDNode.
+ */
+typedef struct LLVMOpaqueNamedMDNode *LLVMNamedMDNodeRef;
+
+/**
+ * Represents an entry in a Global Object's metadata attachments.
+ *
+ * This models std::pair<unsigned, MDNode *>
+ */
+typedef struct LLVMOpaqueValueMetadataEntry LLVMValueMetadataEntry;
+
+/**
+ * Represents an LLVM basic block builder.
+ *
+ * This models llvm::IRBuilder.
+ */
+typedef struct LLVMOpaqueBuilder *LLVMBuilderRef;
+
+/**
+ * Represents an LLVM debug info builder.
+ *
+ * This models llvm::DIBuilder.
+ */
+typedef struct LLVMOpaqueDIBuilder *LLVMDIBuilderRef;
+
+/**
+ * Interface used to provide a module to JIT or interpreter.
+ * This is now just a synonym for llvm::Module, but we have to keep using the
+ * different type to keep binary compatibility.
+ */
+typedef struct LLVMOpaqueModuleProvider *LLVMModuleProviderRef;
+
+/** @see llvm::PassManagerBase */
+typedef struct LLVMOpaquePassManager *LLVMPassManagerRef;
+
+/** @see llvm::PassRegistry */
+typedef struct LLVMOpaquePassRegistry *LLVMPassRegistryRef;
+
+/**
+ * Used to get the users and usees of a Value.
+ *
+ * @see llvm::Use */
+typedef struct LLVMOpaqueUse *LLVMUseRef;
+
+/**
+ * Used to represent an attributes.
+ *
+ * @see llvm::Attribute
+ */
+typedef struct LLVMOpaqueAttributeRef *LLVMAttributeRef;
+
+/**
+ * @see llvm::DiagnosticInfo
+ */
+typedef struct LLVMOpaqueDiagnosticInfo *LLVMDiagnosticInfoRef;
+
+/**
+ * @see llvm::Comdat
+ */
+typedef struct LLVMComdat *LLVMComdatRef;
+
+/**
+ * @see llvm::Module::ModuleFlagEntry
+ */
+typedef struct LLVMOpaqueModuleFlagEntry LLVMModuleFlagEntry;
+
+/**
+ * @see llvm::JITEventListener
+ */
+typedef struct LLVMOpaqueJITEventListener *LLVMJITEventListenerRef;
+
+/**
+ * @see llvm::object::Binary
+ */
+typedef struct LLVMOpaqueBinary *LLVMBinaryRef;
+
+/**
+ * @}
+ */
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/APFloat.h b/third_party/llvm-project/include/llvm/ADT/APFloat.h
new file mode 100644
index 000000000..1c4969733
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/APFloat.h
@@ -0,0 +1,1290 @@
+//===- llvm/ADT/APFloat.h - Arbitrary Precision Floating Point ---*- C++ -*-==//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// \brief
+/// This file declares a class to represent arbitrary precision floating point
+/// values and provide a variety of arithmetic operations on them.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_APFLOAT_H
+#define LLVM_ADT_APFLOAT_H
+
+#include "llvm/ADT/APInt.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <memory>
+
+#define APFLOAT_DISPATCH_ON_SEMANTICS(METHOD_CALL) \
+ do { \
+ if (usesLayout<IEEEFloat>(getSemantics())) \
+ return U.IEEE.METHOD_CALL; \
+ if (usesLayout<DoubleAPFloat>(getSemantics())) \
+ return U.Double.METHOD_CALL; \
+ llvm_unreachable("Unexpected semantics"); \
+ } while (false)
+
+namespace llvm {
+
+struct fltSemantics;
+class APSInt;
+class StringRef;
+class APFloat;
+class raw_ostream;
+
+template <typename T> class SmallVectorImpl;
+
+/// Enum that represents what fraction of the LSB truncated bits of an fp number
+/// represent.
+///
+/// This essentially combines the roles of guard and sticky bits.
+enum lostFraction { // Example of truncated bits:
+ lfExactlyZero, // 000000
+ lfLessThanHalf, // 0xxxxx x's not all zero
+ lfExactlyHalf, // 100000
+ lfMoreThanHalf // 1xxxxx x's not all zero
+};
+
+/// A self-contained host- and target-independent arbitrary-precision
+/// floating-point software implementation.
+///
+/// APFloat uses bignum integer arithmetic as provided by static functions in
+/// the APInt class. The library will work with bignum integers whose parts are
+/// any unsigned type at least 16 bits wide, but 64 bits is recommended.
+///
+/// Written for clarity rather than speed, in particular with a view to use in
+/// the front-end of a cross compiler so that target arithmetic can be correctly
+/// performed on the host. Performance should nonetheless be reasonable,
+/// particularly for its intended use. It may be useful as a base
+/// implementation for a run-time library during development of a faster
+/// target-specific one.
+///
+/// All 5 rounding modes in the IEEE-754R draft are handled correctly for all
+/// implemented operations. Currently implemented operations are add, subtract,
+/// multiply, divide, fused-multiply-add, conversion-to-float,
+/// conversion-to-integer and conversion-from-integer. New rounding modes
+/// (e.g. away from zero) can be added with three or four lines of code.
+///
+/// Four formats are built-in: IEEE single precision, double precision,
+/// quadruple precision, and x87 80-bit extended double (when operating with
+/// full extended precision). Adding a new format that obeys IEEE semantics
+/// only requires adding two lines of code: a declaration and definition of the
+/// format.
+///
+/// All operations return the status of that operation as an exception bit-mask,
+/// so multiple operations can be done consecutively with their results or-ed
+/// together. The returned status can be useful for compiler diagnostics; e.g.,
+/// inexact, underflow and overflow can be easily diagnosed on constant folding,
+/// and compiler optimizers can determine what exceptions would be raised by
+/// folding operations and optimize, or perhaps not optimize, accordingly.
+///
+/// At present, underflow tininess is detected after rounding; it should be
+/// straight forward to add support for the before-rounding case too.
+///
+/// The library reads hexadecimal floating point numbers as per C99, and
+/// correctly rounds if necessary according to the specified rounding mode.
+/// Syntax is required to have been validated by the caller. It also converts
+/// floating point numbers to hexadecimal text as per the C99 %a and %A
+/// conversions. The output precision (or alternatively the natural minimal
+/// precision) can be specified; if the requested precision is less than the
+/// natural precision the output is correctly rounded for the specified rounding
+/// mode.
+///
+/// It also reads decimal floating point numbers and correctly rounds according
+/// to the specified rounding mode.
+///
+/// Conversion to decimal text is not currently implemented.
+///
+/// Non-zero finite numbers are represented internally as a sign bit, a 16-bit
+/// signed exponent, and the significand as an array of integer parts. After
+/// normalization of a number of precision P the exponent is within the range of
+/// the format, and if the number is not denormal the P-th bit of the
+/// significand is set as an explicit integer bit. For denormals the most
+/// significant bit is shifted right so that the exponent is maintained at the
+/// format's minimum, so that the smallest denormal has just the least
+/// significant bit of the significand set. The sign of zeroes and infinities
+/// is significant; the exponent and significand of such numbers is not stored,
+/// but has a known implicit (deterministic) value: 0 for the significands, 0
+/// for zero exponent, all 1 bits for infinity exponent. For NaNs the sign and
+/// significand are deterministic, although not really meaningful, and preserved
+/// in non-conversion operations. The exponent is implicitly all 1 bits.
+///
+/// APFloat does not provide any exception handling beyond default exception
+/// handling. We represent Signaling NaNs via IEEE-754R 2008 6.2.1 should clause
+/// by encoding Signaling NaNs with the first bit of its trailing significand as
+/// 0.
+///
+/// TODO
+/// ====
+///
+/// Some features that may or may not be worth adding:
+///
+/// Binary to decimal conversion (hard).
+///
+/// Optional ability to detect underflow tininess before rounding.
+///
+/// New formats: x87 in single and double precision mode (IEEE apart from
+/// extended exponent range) (hard).
+///
+/// New operations: sqrt, IEEE remainder, C90 fmod, nexttoward.
+///
+
+// This is the common type definitions shared by APFloat and its internal
+// implementation classes. This struct should not define any non-static data
+// members.
+struct APFloatBase {
+ typedef APInt::WordType integerPart;
+ static const unsigned integerPartWidth = APInt::APINT_BITS_PER_WORD;
+
+ /// A signed type to represent a floating point numbers unbiased exponent.
+ typedef signed short ExponentType;
+
+ /// \name Floating Point Semantics.
+ /// @{
+ enum Semantics {
+ S_IEEEhalf,
+ S_IEEEsingle,
+ S_IEEEdouble,
+ S_x87DoubleExtended,
+ S_IEEEquad,
+ S_PPCDoubleDouble
+ };
+
+ static const llvm::fltSemantics &EnumToSemantics(Semantics S);
+ static Semantics SemanticsToEnum(const llvm::fltSemantics &Sem);
+
+ static const fltSemantics &IEEEhalf() LLVM_READNONE;
+ static const fltSemantics &IEEEsingle() LLVM_READNONE;
+ static const fltSemantics &IEEEdouble() LLVM_READNONE;
+ static const fltSemantics &IEEEquad() LLVM_READNONE;
+ static const fltSemantics &PPCDoubleDouble() LLVM_READNONE;
+ static const fltSemantics &x87DoubleExtended() LLVM_READNONE;
+
+ /// A Pseudo fltsemantic used to construct APFloats that cannot conflict with
+ /// anything real.
+ static const fltSemantics &Bogus() LLVM_READNONE;
+
+ /// @}
+
+ /// IEEE-754R 5.11: Floating Point Comparison Relations.
+ enum cmpResult {
+ cmpLessThan,
+ cmpEqual,
+ cmpGreaterThan,
+ cmpUnordered
+ };
+
+ /// IEEE-754R 4.3: Rounding-direction attributes.
+ enum roundingMode {
+ rmNearestTiesToEven,
+ rmTowardPositive,
+ rmTowardNegative,
+ rmTowardZero,
+ rmNearestTiesToAway
+ };
+
+ /// IEEE-754R 7: Default exception handling.
+ ///
+ /// opUnderflow or opOverflow are always returned or-ed with opInexact.
+ ///
+ /// APFloat models this behavior specified by IEEE-754:
+ /// "For operations producing results in floating-point format, the default
+ /// result of an operation that signals the invalid operation exception
+ /// shall be a quiet NaN."
+ enum opStatus {
+ opOK = 0x00,
+ opInvalidOp = 0x01,
+ opDivByZero = 0x02,
+ opOverflow = 0x04,
+ opUnderflow = 0x08,
+ opInexact = 0x10
+ };
+
+ /// Category of internally-represented number.
+ enum fltCategory {
+ fcInfinity,
+ fcNaN,
+ fcNormal,
+ fcZero
+ };
+
+ /// Convenience enum used to construct an uninitialized APFloat.
+ enum uninitializedTag {
+ uninitialized
+ };
+
+ /// Enumeration of \c ilogb error results.
+ enum IlogbErrorKinds {
+ IEK_Zero = INT_MIN + 1,
+ IEK_NaN = INT_MIN,
+ IEK_Inf = INT_MAX
+ };
+
+ static unsigned int semanticsPrecision(const fltSemantics &);
+ static ExponentType semanticsMinExponent(const fltSemantics &);
+ static ExponentType semanticsMaxExponent(const fltSemantics &);
+ static unsigned int semanticsSizeInBits(const fltSemantics &);
+
+ /// Returns the size of the floating point number (in bits) in the given
+ /// semantics.
+ static unsigned getSizeInBits(const fltSemantics &Sem);
+};
+
+namespace detail {
+
+class IEEEFloat final : public APFloatBase {
+public:
+ /// \name Constructors
+ /// @{
+
+ IEEEFloat(const fltSemantics &); // Default construct to 0.0
+ IEEEFloat(const fltSemantics &, integerPart);
+ IEEEFloat(const fltSemantics &, uninitializedTag);
+ IEEEFloat(const fltSemantics &, const APInt &);
+ explicit IEEEFloat(double d);
+ explicit IEEEFloat(float f);
+ IEEEFloat(const IEEEFloat &);
+ IEEEFloat(IEEEFloat &&);
+ ~IEEEFloat();
+
+ /// @}
+
+ /// Returns whether this instance allocated memory.
+ bool needsCleanup() const { return partCount() > 1; }
+
+ /// \name Convenience "constructors"
+ /// @{
+
+ /// @}
+
+ /// \name Arithmetic
+ /// @{
+
+ opStatus add(const IEEEFloat &, roundingMode);
+ opStatus subtract(const IEEEFloat &, roundingMode);
+ opStatus multiply(const IEEEFloat &, roundingMode);
+ opStatus divide(const IEEEFloat &, roundingMode);
+ /// IEEE remainder.
+ opStatus remainder(const IEEEFloat &);
+ /// C fmod, or llvm frem.
+ opStatus mod(const IEEEFloat &);
+ opStatus fusedMultiplyAdd(const IEEEFloat &, const IEEEFloat &, roundingMode);
+ opStatus roundToIntegral(roundingMode);
+ /// IEEE-754R 5.3.1: nextUp/nextDown.
+ opStatus next(bool nextDown);
+
+ /// @}
+
+ /// \name Sign operations.
+ /// @{
+
+ void changeSign();
+
+ /// @}
+
+ /// \name Conversions
+ /// @{
+
+ opStatus convert(const fltSemantics &, roundingMode, bool *);
+ opStatus convertToInteger(MutableArrayRef<integerPart>, unsigned int, bool,
+ roundingMode, bool *) const;
+ opStatus convertFromAPInt(const APInt &, bool, roundingMode);
+ opStatus convertFromSignExtendedInteger(const integerPart *, unsigned int,
+ bool, roundingMode);
+ opStatus convertFromZeroExtendedInteger(const integerPart *, unsigned int,
+ bool, roundingMode);
+ opStatus convertFromString(StringRef, roundingMode);
+ APInt bitcastToAPInt() const;
+ double convertToDouble() const;
+ float convertToFloat() const;
+
+ /// @}
+
+ /// The definition of equality is not straightforward for floating point, so
+ /// we won't use operator==. Use one of the following, or write whatever it
+ /// is you really mean.
+ bool operator==(const IEEEFloat &) const = delete;
+
+ /// IEEE comparison with another floating point number (NaNs compare
+ /// unordered, 0==-0).
+ cmpResult compare(const IEEEFloat &) const;
+
+ /// Bitwise comparison for equality (QNaNs compare equal, 0!=-0).
+ bool bitwiseIsEqual(const IEEEFloat &) const;
+
+ /// Write out a hexadecimal representation of the floating point value to DST,
+ /// which must be of sufficient size, in the C99 form [-]0xh.hhhhp[+-]d.
+ /// Return the number of characters written, excluding the terminating NUL.
+ unsigned int convertToHexString(char *dst, unsigned int hexDigits,
+ bool upperCase, roundingMode) const;
+
+ /// \name IEEE-754R 5.7.2 General operations.
+ /// @{
+
+ /// IEEE-754R isSignMinus: Returns true if and only if the current value is
+ /// negative.
+ ///
+ /// This applies to zeros and NaNs as well.
+ bool isNegative() const { return sign; }
+
+ /// IEEE-754R isNormal: Returns true if and only if the current value is normal.
+ ///
+ /// This implies that the current value of the float is not zero, subnormal,
+ /// infinite, or NaN following the definition of normality from IEEE-754R.
+ bool isNormal() const { return !isDenormal() && isFiniteNonZero(); }
+
+ /// Returns true if and only if the current value is zero, subnormal, or
+ /// normal.
+ ///
+ /// This means that the value is not infinite or NaN.
+ bool isFinite() const { return !isNaN() && !isInfinity(); }
+
+ /// Returns true if and only if the float is plus or minus zero.
+ bool isZero() const { return category == fcZero; }
+
+ /// IEEE-754R isSubnormal(): Returns true if and only if the float is a
+ /// denormal.
+ bool isDenormal() const;
+
+ /// IEEE-754R isInfinite(): Returns true if and only if the float is infinity.
+ bool isInfinity() const { return category == fcInfinity; }
+
+ /// Returns true if and only if the float is a quiet or signaling NaN.
+ bool isNaN() const { return category == fcNaN; }
+
+ /// Returns true if and only if the float is a signaling NaN.
+ bool isSignaling() const;
+
+ /// @}
+
+ /// \name Simple Queries
+ /// @{
+
+ fltCategory getCategory() const { return category; }
+ const fltSemantics &getSemantics() const { return *semantics; }
+ bool isNonZero() const { return category != fcZero; }
+ bool isFiniteNonZero() const { return isFinite() && !isZero(); }
+ bool isPosZero() const { return isZero() && !isNegative(); }
+ bool isNegZero() const { return isZero() && isNegative(); }
+
+ /// Returns true if and only if the number has the smallest possible non-zero
+ /// magnitude in the current semantics.
+ bool isSmallest() const;
+
+ /// Returns true if and only if the number has the largest possible finite
+ /// magnitude in the current semantics.
+ bool isLargest() const;
+
+ /// Returns true if and only if the number is an exact integer.
+ bool isInteger() const;
+
+ /// @}
+
+ IEEEFloat &operator=(const IEEEFloat &);
+ IEEEFloat &operator=(IEEEFloat &&);
+
+ /// Overload to compute a hash code for an APFloat value.
+ ///
+ /// Note that the use of hash codes for floating point values is in general
+ /// frought with peril. Equality is hard to define for these values. For
+ /// example, should negative and positive zero hash to different codes? Are
+ /// they equal or not? This hash value implementation specifically
+ /// emphasizes producing different codes for different inputs in order to
+ /// be used in canonicalization and memoization. As such, equality is
+ /// bitwiseIsEqual, and 0 != -0.
+ friend hash_code hash_value(const IEEEFloat &Arg);
+
+ /// Converts this value into a decimal string.
+ ///
+ /// \param FormatPrecision The maximum number of digits of
+ /// precision to output. If there are fewer digits available,
+ /// zero padding will not be used unless the value is
+ /// integral and small enough to be expressed in
+ /// FormatPrecision digits. 0 means to use the natural
+ /// precision of the number.
+ /// \param FormatMaxPadding The maximum number of zeros to
+ /// consider inserting before falling back to scientific
+ /// notation. 0 means to always use scientific notation.
+ ///
+ /// \param TruncateZero Indicate whether to remove the trailing zero in
+ /// fraction part or not. Also setting this parameter to false forcing
+ /// producing of output more similar to default printf behavior.
+ /// Specifically the lower e is used as exponent delimiter and exponent
+ /// always contains no less than two digits.
+ ///
+ /// Number Precision MaxPadding Result
+ /// ------ --------- ---------- ------
+ /// 1.01E+4 5 2 10100
+ /// 1.01E+4 4 2 1.01E+4
+ /// 1.01E+4 5 1 1.01E+4
+ /// 1.01E-2 5 2 0.0101
+ /// 1.01E-2 4 2 0.0101
+ /// 1.01E-2 4 1 1.01E-2
+ void toString(SmallVectorImpl<char> &Str, unsigned FormatPrecision = 0,
+ unsigned FormatMaxPadding = 3, bool TruncateZero = true) const;
+
+ /// If this value has an exact multiplicative inverse, store it in inv and
+ /// return true.
+ bool getExactInverse(APFloat *inv) const;
+
+ /// Returns the exponent of the internal representation of the APFloat.
+ ///
+ /// Because the radix of APFloat is 2, this is equivalent to floor(log2(x)).
+ /// For special APFloat values, this returns special error codes:
+ ///
+ /// NaN -> \c IEK_NaN
+ /// 0 -> \c IEK_Zero
+ /// Inf -> \c IEK_Inf
+ ///
+ friend int ilogb(const IEEEFloat &Arg);
+
+ /// Returns: X * 2^Exp for integral exponents.
+ friend IEEEFloat scalbn(IEEEFloat X, int Exp, roundingMode);
+
+ friend IEEEFloat frexp(const IEEEFloat &X, int &Exp, roundingMode);
+
+ /// \name Special value setters.
+ /// @{
+
+ void makeLargest(bool Neg = false);
+ void makeSmallest(bool Neg = false);
+ void makeNaN(bool SNaN = false, bool Neg = false,
+ const APInt *fill = nullptr);
+ void makeInf(bool Neg = false);
+ void makeZero(bool Neg = false);
+ void makeQuiet();
+
+ /// Returns the smallest (by magnitude) normalized finite number in the given
+ /// semantics.
+ ///
+ /// \param Negative - True iff the number should be negative
+ void makeSmallestNormalized(bool Negative = false);
+
+ /// @}
+
+ cmpResult compareAbsoluteValue(const IEEEFloat &) const;
+
+private:
+ /// \name Simple Queries
+ /// @{
+
+ integerPart *significandParts();
+ const integerPart *significandParts() const;
+ unsigned int partCount() const;
+
+ /// @}
+
+ /// \name Significand operations.
+ /// @{
+
+ integerPart addSignificand(const IEEEFloat &);
+ integerPart subtractSignificand(const IEEEFloat &, integerPart);
+ lostFraction addOrSubtractSignificand(const IEEEFloat &, bool subtract);
+ lostFraction multiplySignificand(const IEEEFloat &, const IEEEFloat *);
+ lostFraction divideSignificand(const IEEEFloat &);
+ void incrementSignificand();
+ void initialize(const fltSemantics *);
+ void shiftSignificandLeft(unsigned int);
+ lostFraction shiftSignificandRight(unsigned int);
+ unsigned int significandLSB() const;
+ unsigned int significandMSB() const;
+ void zeroSignificand();
+ /// Return true if the significand excluding the integral bit is all ones.
+ bool isSignificandAllOnes() const;
+ /// Return true if the significand excluding the integral bit is all zeros.
+ bool isSignificandAllZeros() const;
+
+ /// @}
+
+ /// \name Arithmetic on special values.
+ /// @{
+
+ opStatus addOrSubtractSpecials(const IEEEFloat &, bool subtract);
+ opStatus divideSpecials(const IEEEFloat &);
+ opStatus multiplySpecials(const IEEEFloat &);
+ opStatus modSpecials(const IEEEFloat &);
+
+ /// @}
+
+ /// \name Miscellany
+ /// @{
+
+ bool convertFromStringSpecials(StringRef str);
+ opStatus normalize(roundingMode, lostFraction);
+ opStatus addOrSubtract(const IEEEFloat &, roundingMode, bool subtract);
+ opStatus handleOverflow(roundingMode);
+ bool roundAwayFromZero(roundingMode, lostFraction, unsigned int) const;
+ opStatus convertToSignExtendedInteger(MutableArrayRef<integerPart>,
+ unsigned int, bool, roundingMode,
+ bool *) const;
+ opStatus convertFromUnsignedParts(const integerPart *, unsigned int,
+ roundingMode);
+ opStatus convertFromHexadecimalString(StringRef, roundingMode);
+ opStatus convertFromDecimalString(StringRef, roundingMode);
+ char *convertNormalToHexString(char *, unsigned int, bool,
+ roundingMode) const;
+ opStatus roundSignificandWithExponent(const integerPart *, unsigned int, int,
+ roundingMode);
+
+ /// @}
+
+ APInt convertHalfAPFloatToAPInt() const;
+ APInt convertFloatAPFloatToAPInt() const;
+ APInt convertDoubleAPFloatToAPInt() const;
+ APInt convertQuadrupleAPFloatToAPInt() const;
+ APInt convertF80LongDoubleAPFloatToAPInt() const;
+ APInt convertPPCDoubleDoubleAPFloatToAPInt() const;
+ void initFromAPInt(const fltSemantics *Sem, const APInt &api);
+ void initFromHalfAPInt(const APInt &api);
+ void initFromFloatAPInt(const APInt &api);
+ void initFromDoubleAPInt(const APInt &api);
+ void initFromQuadrupleAPInt(const APInt &api);
+ void initFromF80LongDoubleAPInt(const APInt &api);
+ void initFromPPCDoubleDoubleAPInt(const APInt &api);
+
+ void assign(const IEEEFloat &);
+ void copySignificand(const IEEEFloat &);
+ void freeSignificand();
+
+ /// Note: this must be the first data member.
+ /// The semantics that this value obeys.
+ const fltSemantics *semantics;
+
+ /// A binary fraction with an explicit integer bit.
+ ///
+ /// The significand must be at least one bit wider than the target precision.
+ union Significand {
+ integerPart part;
+ integerPart *parts;
+ } significand;
+
+ /// The signed unbiased exponent of the value.
+ ExponentType exponent;
+
+ /// What kind of floating point number this is.
+ ///
+ /// Only 2 bits are required, but VisualStudio incorrectly sign extends it.
+ /// Using the extra bit keeps it from failing under VisualStudio.
+ fltCategory category : 3;
+
+ /// Sign bit of the number.
+ unsigned int sign : 1;
+};
+
+hash_code hash_value(const IEEEFloat &Arg);
+int ilogb(const IEEEFloat &Arg);
+IEEEFloat scalbn(IEEEFloat X, int Exp, IEEEFloat::roundingMode);
+IEEEFloat frexp(const IEEEFloat &Val, int &Exp, IEEEFloat::roundingMode RM);
+
+// This mode implements more precise float in terms of two APFloats.
+// The interface and layout is designed for arbitray underlying semantics,
+// though currently only PPCDoubleDouble semantics are supported, whose
+// corresponding underlying semantics are IEEEdouble.
+class DoubleAPFloat final : public APFloatBase {
+ // Note: this must be the first data member.
+ const fltSemantics *Semantics;
+ std::unique_ptr<APFloat[]> Floats;
+
+ opStatus addImpl(const APFloat &a, const APFloat &aa, const APFloat &c,
+ const APFloat &cc, roundingMode RM);
+
+ opStatus addWithSpecial(const DoubleAPFloat &LHS, const DoubleAPFloat &RHS,
+ DoubleAPFloat &Out, roundingMode RM);
+
+public:
+ DoubleAPFloat(const fltSemantics &S);
+ DoubleAPFloat(const fltSemantics &S, uninitializedTag);
+ DoubleAPFloat(const fltSemantics &S, integerPart);
+ DoubleAPFloat(const fltSemantics &S, const APInt &I);
+ DoubleAPFloat(const fltSemantics &S, APFloat &&First, APFloat &&Second);
+ DoubleAPFloat(const DoubleAPFloat &RHS);
+ DoubleAPFloat(DoubleAPFloat &&RHS);
+
+ DoubleAPFloat &operator=(const DoubleAPFloat &RHS);
+
+ DoubleAPFloat &operator=(DoubleAPFloat &&RHS) {
+ if (this != &RHS) {
+ this->~DoubleAPFloat();
+ new (this) DoubleAPFloat(std::move(RHS));
+ }
+ return *this;
+ }
+
+ bool needsCleanup() const { return Floats != nullptr; }
+
+ APFloat &getFirst() { return Floats[0]; }
+ const APFloat &getFirst() const { return Floats[0]; }
+ APFloat &getSecond() { return Floats[1]; }
+ const APFloat &getSecond() const { return Floats[1]; }
+
+ opStatus add(const DoubleAPFloat &RHS, roundingMode RM);
+ opStatus subtract(const DoubleAPFloat &RHS, roundingMode RM);
+ opStatus multiply(const DoubleAPFloat &RHS, roundingMode RM);
+ opStatus divide(const DoubleAPFloat &RHS, roundingMode RM);
+ opStatus remainder(const DoubleAPFloat &RHS);
+ opStatus mod(const DoubleAPFloat &RHS);
+ opStatus fusedMultiplyAdd(const DoubleAPFloat &Multiplicand,
+ const DoubleAPFloat &Addend, roundingMode RM);
+ opStatus roundToIntegral(roundingMode RM);
+ void changeSign();
+ cmpResult compareAbsoluteValue(const DoubleAPFloat &RHS) const;
+
+ fltCategory getCategory() const;
+ bool isNegative() const;
+
+ void makeInf(bool Neg);
+ void makeZero(bool Neg);
+ void makeLargest(bool Neg);
+ void makeSmallest(bool Neg);
+ void makeSmallestNormalized(bool Neg);
+ void makeNaN(bool SNaN, bool Neg, const APInt *fill);
+
+ cmpResult compare(const DoubleAPFloat &RHS) const;
+ bool bitwiseIsEqual(const DoubleAPFloat &RHS) const;
+ APInt bitcastToAPInt() const;
+ opStatus convertFromString(StringRef, roundingMode);
+ opStatus next(bool nextDown);
+
+ opStatus convertToInteger(MutableArrayRef<integerPart> Input,
+ unsigned int Width, bool IsSigned, roundingMode RM,
+ bool *IsExact) const;
+ opStatus convertFromAPInt(const APInt &Input, bool IsSigned, roundingMode RM);
+ opStatus convertFromSignExtendedInteger(const integerPart *Input,
+ unsigned int InputSize, bool IsSigned,
+ roundingMode RM);
+ opStatus convertFromZeroExtendedInteger(const integerPart *Input,
+ unsigned int InputSize, bool IsSigned,
+ roundingMode RM);
+ unsigned int convertToHexString(char *DST, unsigned int HexDigits,
+ bool UpperCase, roundingMode RM) const;
+
+ bool isDenormal() const;
+ bool isSmallest() const;
+ bool isLargest() const;
+ bool isInteger() const;
+
+ void toString(SmallVectorImpl<char> &Str, unsigned FormatPrecision,
+ unsigned FormatMaxPadding, bool TruncateZero = true) const;
+
+ bool getExactInverse(APFloat *inv) const;
+
+ friend int ilogb(const DoubleAPFloat &Arg);
+ friend DoubleAPFloat scalbn(DoubleAPFloat X, int Exp, roundingMode);
+ friend DoubleAPFloat frexp(const DoubleAPFloat &X, int &Exp, roundingMode);
+ friend hash_code hash_value(const DoubleAPFloat &Arg);
+};
+
+hash_code hash_value(const DoubleAPFloat &Arg);
+
+} // End detail namespace
+
+// This is a interface class that is currently forwarding functionalities from
+// detail::IEEEFloat.
+class APFloat : public APFloatBase {
+ typedef detail::IEEEFloat IEEEFloat;
+ typedef detail::DoubleAPFloat DoubleAPFloat;
+
+ static_assert(std::is_standard_layout<IEEEFloat>::value, "");
+
+ union Storage {
+ const fltSemantics *semantics;
+ IEEEFloat IEEE;
+ DoubleAPFloat Double;
+
+ explicit Storage(IEEEFloat F, const fltSemantics &S);
+ explicit Storage(DoubleAPFloat F, const fltSemantics &S)
+ : Double(std::move(F)) {
+ assert(&S == &PPCDoubleDouble());
+ }
+
+ template <typename... ArgTypes>
+ Storage(const fltSemantics &Semantics, ArgTypes &&... Args) {
+ if (usesLayout<IEEEFloat>(Semantics)) {
+ new (&IEEE) IEEEFloat(Semantics, std::forward<ArgTypes>(Args)...);
+ return;
+ }
+ if (usesLayout<DoubleAPFloat>(Semantics)) {
+ new (&Double) DoubleAPFloat(Semantics, std::forward<ArgTypes>(Args)...);
+ return;
+ }
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ ~Storage() {
+ if (usesLayout<IEEEFloat>(*semantics)) {
+ IEEE.~IEEEFloat();
+ return;
+ }
+ if (usesLayout<DoubleAPFloat>(*semantics)) {
+ Double.~DoubleAPFloat();
+ return;
+ }
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ Storage(const Storage &RHS) {
+ if (usesLayout<IEEEFloat>(*RHS.semantics)) {
+ new (this) IEEEFloat(RHS.IEEE);
+ return;
+ }
+ if (usesLayout<DoubleAPFloat>(*RHS.semantics)) {
+ new (this) DoubleAPFloat(RHS.Double);
+ return;
+ }
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ Storage(Storage &&RHS) {
+ if (usesLayout<IEEEFloat>(*RHS.semantics)) {
+ new (this) IEEEFloat(std::move(RHS.IEEE));
+ return;
+ }
+ if (usesLayout<DoubleAPFloat>(*RHS.semantics)) {
+ new (this) DoubleAPFloat(std::move(RHS.Double));
+ return;
+ }
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ Storage &operator=(const Storage &RHS) {
+ if (usesLayout<IEEEFloat>(*semantics) &&
+ usesLayout<IEEEFloat>(*RHS.semantics)) {
+ IEEE = RHS.IEEE;
+ } else if (usesLayout<DoubleAPFloat>(*semantics) &&
+ usesLayout<DoubleAPFloat>(*RHS.semantics)) {
+ Double = RHS.Double;
+ } else if (this != &RHS) {
+ this->~Storage();
+ new (this) Storage(RHS);
+ }
+ return *this;
+ }
+
+ Storage &operator=(Storage &&RHS) {
+ if (usesLayout<IEEEFloat>(*semantics) &&
+ usesLayout<IEEEFloat>(*RHS.semantics)) {
+ IEEE = std::move(RHS.IEEE);
+ } else if (usesLayout<DoubleAPFloat>(*semantics) &&
+ usesLayout<DoubleAPFloat>(*RHS.semantics)) {
+ Double = std::move(RHS.Double);
+ } else if (this != &RHS) {
+ this->~Storage();
+ new (this) Storage(std::move(RHS));
+ }
+ return *this;
+ }
+ } U;
+
+ template <typename T> static bool usesLayout(const fltSemantics &Semantics) {
+ static_assert(std::is_same<T, IEEEFloat>::value ||
+ std::is_same<T, DoubleAPFloat>::value, "");
+ if (std::is_same<T, DoubleAPFloat>::value) {
+ return &Semantics == &PPCDoubleDouble();
+ }
+ return &Semantics != &PPCDoubleDouble();
+ }
+
+ IEEEFloat &getIEEE() {
+ if (usesLayout<IEEEFloat>(*U.semantics))
+ return U.IEEE;
+ if (usesLayout<DoubleAPFloat>(*U.semantics))
+ return U.Double.getFirst().U.IEEE;
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ const IEEEFloat &getIEEE() const {
+ if (usesLayout<IEEEFloat>(*U.semantics))
+ return U.IEEE;
+ if (usesLayout<DoubleAPFloat>(*U.semantics))
+ return U.Double.getFirst().U.IEEE;
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ void makeZero(bool Neg) { APFLOAT_DISPATCH_ON_SEMANTICS(makeZero(Neg)); }
+
+ void makeInf(bool Neg) { APFLOAT_DISPATCH_ON_SEMANTICS(makeInf(Neg)); }
+
+ void makeNaN(bool SNaN, bool Neg, const APInt *fill) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(makeNaN(SNaN, Neg, fill));
+ }
+
+ void makeLargest(bool Neg) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(makeLargest(Neg));
+ }
+
+ void makeSmallest(bool Neg) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(makeSmallest(Neg));
+ }
+
+ void makeSmallestNormalized(bool Neg) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(makeSmallestNormalized(Neg));
+ }
+
+ // FIXME: This is due to clang 3.3 (or older version) always checks for the
+ // default constructor in an array aggregate initialization, even if no
+ // elements in the array is default initialized.
+ APFloat() : U(IEEEdouble()) {
+ llvm_unreachable("This is a workaround for old clang.");
+ }
+
+ explicit APFloat(IEEEFloat F, const fltSemantics &S) : U(std::move(F), S) {}
+ explicit APFloat(DoubleAPFloat F, const fltSemantics &S)
+ : U(std::move(F), S) {}
+
+ cmpResult compareAbsoluteValue(const APFloat &RHS) const {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only compare APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.compareAbsoluteValue(RHS.U.IEEE);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.compareAbsoluteValue(RHS.U.Double);
+ llvm_unreachable("Unexpected semantics");
+ }
+
+public:
+ APFloat(const fltSemantics &Semantics) : U(Semantics) {}
+ APFloat(const fltSemantics &Semantics, StringRef S);
+ APFloat(const fltSemantics &Semantics, integerPart I) : U(Semantics, I) {}
+ // TODO: Remove this constructor. This isn't faster than the first one.
+ APFloat(const fltSemantics &Semantics, uninitializedTag)
+ : U(Semantics, uninitialized) {}
+ APFloat(const fltSemantics &Semantics, const APInt &I) : U(Semantics, I) {}
+ explicit APFloat(double d) : U(IEEEFloat(d), IEEEdouble()) {}
+ explicit APFloat(float f) : U(IEEEFloat(f), IEEEsingle()) {}
+ APFloat(const APFloat &RHS) = default;
+ APFloat(APFloat &&RHS) = default;
+
+ ~APFloat() = default;
+
+ bool needsCleanup() const { APFLOAT_DISPATCH_ON_SEMANTICS(needsCleanup()); }
+
+ /// Factory for Positive and Negative Zero.
+ ///
+ /// \param Negative True iff the number should be negative.
+ static APFloat getZero(const fltSemantics &Sem, bool Negative = false) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeZero(Negative);
+ return Val;
+ }
+
+ /// Factory for Positive and Negative Infinity.
+ ///
+ /// \param Negative True iff the number should be negative.
+ static APFloat getInf(const fltSemantics &Sem, bool Negative = false) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeInf(Negative);
+ return Val;
+ }
+
+ /// Factory for NaN values.
+ ///
+ /// \param Negative - True iff the NaN generated should be negative.
+ /// \param payload - The unspecified fill bits for creating the NaN, 0 by
+ /// default. The value is truncated as necessary.
+ static APFloat getNaN(const fltSemantics &Sem, bool Negative = false,
+ uint64_t payload = 0) {
+ if (payload) {
+ APInt intPayload(64, payload);
+ return getQNaN(Sem, Negative, &intPayload);
+ } else {
+ return getQNaN(Sem, Negative, nullptr);
+ }
+ }
+
+ /// Factory for QNaN values.
+ static APFloat getQNaN(const fltSemantics &Sem, bool Negative = false,
+ const APInt *payload = nullptr) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeNaN(false, Negative, payload);
+ return Val;
+ }
+
+ /// Factory for SNaN values.
+ static APFloat getSNaN(const fltSemantics &Sem, bool Negative = false,
+ const APInt *payload = nullptr) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeNaN(true, Negative, payload);
+ return Val;
+ }
+
+ /// Returns the largest finite number in the given semantics.
+ ///
+ /// \param Negative - True iff the number should be negative
+ static APFloat getLargest(const fltSemantics &Sem, bool Negative = false) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeLargest(Negative);
+ return Val;
+ }
+
+ /// Returns the smallest (by magnitude) finite number in the given semantics.
+ /// Might be denormalized, which implies a relative loss of precision.
+ ///
+ /// \param Negative - True iff the number should be negative
+ static APFloat getSmallest(const fltSemantics &Sem, bool Negative = false) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeSmallest(Negative);
+ return Val;
+ }
+
+ /// Returns the smallest (by magnitude) normalized finite number in the given
+ /// semantics.
+ ///
+ /// \param Negative - True iff the number should be negative
+ static APFloat getSmallestNormalized(const fltSemantics &Sem,
+ bool Negative = false) {
+ APFloat Val(Sem, uninitialized);
+ Val.makeSmallestNormalized(Negative);
+ return Val;
+ }
+
+ /// Returns a float which is bitcasted from an all one value int.
+ ///
+ /// \param BitWidth - Select float type
+ /// \param isIEEE - If 128 bit number, select between PPC and IEEE
+ static APFloat getAllOnesValue(unsigned BitWidth, bool isIEEE = false);
+
+ /// Used to insert APFloat objects, or objects that contain APFloat objects,
+ /// into FoldingSets.
+ void Profile(FoldingSetNodeID &NID) const;
+
+ opStatus add(const APFloat &RHS, roundingMode RM) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.add(RHS.U.IEEE, RM);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.add(RHS.U.Double, RM);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus subtract(const APFloat &RHS, roundingMode RM) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.subtract(RHS.U.IEEE, RM);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.subtract(RHS.U.Double, RM);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus multiply(const APFloat &RHS, roundingMode RM) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.multiply(RHS.U.IEEE, RM);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.multiply(RHS.U.Double, RM);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus divide(const APFloat &RHS, roundingMode RM) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.divide(RHS.U.IEEE, RM);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.divide(RHS.U.Double, RM);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus remainder(const APFloat &RHS) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.remainder(RHS.U.IEEE);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.remainder(RHS.U.Double);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus mod(const APFloat &RHS) {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only call on two APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.mod(RHS.U.IEEE);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.mod(RHS.U.Double);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus fusedMultiplyAdd(const APFloat &Multiplicand, const APFloat &Addend,
+ roundingMode RM) {
+ assert(&getSemantics() == &Multiplicand.getSemantics() &&
+ "Should only call on APFloats with the same semantics");
+ assert(&getSemantics() == &Addend.getSemantics() &&
+ "Should only call on APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.fusedMultiplyAdd(Multiplicand.U.IEEE, Addend.U.IEEE, RM);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.fusedMultiplyAdd(Multiplicand.U.Double, Addend.U.Double,
+ RM);
+ llvm_unreachable("Unexpected semantics");
+ }
+ opStatus roundToIntegral(roundingMode RM) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(roundToIntegral(RM));
+ }
+
+ // TODO: bool parameters are not readable and a source of bugs.
+ // Do something.
+ opStatus next(bool nextDown) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(next(nextDown));
+ }
+
+ /// Add two APFloats, rounding ties to the nearest even.
+ /// No error checking.
+ APFloat operator+(const APFloat &RHS) const {
+ APFloat Result(*this);
+ (void)Result.add(RHS, rmNearestTiesToEven);
+ return Result;
+ }
+
+ /// Subtract two APFloats, rounding ties to the nearest even.
+ /// No error checking.
+ APFloat operator-(const APFloat &RHS) const {
+ APFloat Result(*this);
+ (void)Result.subtract(RHS, rmNearestTiesToEven);
+ return Result;
+ }
+
+ /// Multiply two APFloats, rounding ties to the nearest even.
+ /// No error checking.
+ APFloat operator*(const APFloat &RHS) const {
+ APFloat Result(*this);
+ (void)Result.multiply(RHS, rmNearestTiesToEven);
+ return Result;
+ }
+
+ /// Divide the first APFloat by the second, rounding ties to the nearest even.
+ /// No error checking.
+ APFloat operator/(const APFloat &RHS) const {
+ APFloat Result(*this);
+ (void)Result.divide(RHS, rmNearestTiesToEven);
+ return Result;
+ }
+
+ void changeSign() { APFLOAT_DISPATCH_ON_SEMANTICS(changeSign()); }
+ void clearSign() {
+ if (isNegative())
+ changeSign();
+ }
+ void copySign(const APFloat &RHS) {
+ if (isNegative() != RHS.isNegative())
+ changeSign();
+ }
+
+ /// A static helper to produce a copy of an APFloat value with its sign
+ /// copied from some other APFloat.
+ static APFloat copySign(APFloat Value, const APFloat &Sign) {
+ Value.copySign(Sign);
+ return Value;
+ }
+
+ opStatus convert(const fltSemantics &ToSemantics, roundingMode RM,
+ bool *losesInfo);
+ opStatus convertToInteger(MutableArrayRef<integerPart> Input,
+ unsigned int Width, bool IsSigned, roundingMode RM,
+ bool *IsExact) const {
+ APFLOAT_DISPATCH_ON_SEMANTICS(
+ convertToInteger(Input, Width, IsSigned, RM, IsExact));
+ }
+ opStatus convertToInteger(APSInt &Result, roundingMode RM,
+ bool *IsExact) const;
+ opStatus convertFromAPInt(const APInt &Input, bool IsSigned,
+ roundingMode RM) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(convertFromAPInt(Input, IsSigned, RM));
+ }
+ opStatus convertFromSignExtendedInteger(const integerPart *Input,
+ unsigned int InputSize, bool IsSigned,
+ roundingMode RM) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(
+ convertFromSignExtendedInteger(Input, InputSize, IsSigned, RM));
+ }
+ opStatus convertFromZeroExtendedInteger(const integerPart *Input,
+ unsigned int InputSize, bool IsSigned,
+ roundingMode RM) {
+ APFLOAT_DISPATCH_ON_SEMANTICS(
+ convertFromZeroExtendedInteger(Input, InputSize, IsSigned, RM));
+ }
+ opStatus convertFromString(StringRef, roundingMode);
+ APInt bitcastToAPInt() const {
+ APFLOAT_DISPATCH_ON_SEMANTICS(bitcastToAPInt());
+ }
+ double convertToDouble() const { return getIEEE().convertToDouble(); }
+ float convertToFloat() const { return getIEEE().convertToFloat(); }
+
+ bool operator==(const APFloat &) const = delete;
+
+ cmpResult compare(const APFloat &RHS) const {
+ assert(&getSemantics() == &RHS.getSemantics() &&
+ "Should only compare APFloats with the same semantics");
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.compare(RHS.U.IEEE);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.compare(RHS.U.Double);
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ bool bitwiseIsEqual(const APFloat &RHS) const {
+ if (&getSemantics() != &RHS.getSemantics())
+ return false;
+ if (usesLayout<IEEEFloat>(getSemantics()))
+ return U.IEEE.bitwiseIsEqual(RHS.U.IEEE);
+ if (usesLayout<DoubleAPFloat>(getSemantics()))
+ return U.Double.bitwiseIsEqual(RHS.U.Double);
+ llvm_unreachable("Unexpected semantics");
+ }
+
+ /// We don't rely on operator== working on double values, as
+ /// it returns true for things that are clearly not equal, like -0.0 and 0.0.
+ /// As such, this method can be used to do an exact bit-for-bit comparison of
+ /// two floating point values.
+ ///
+ /// We leave the version with the double argument here because it's just so
+ /// convenient to write "2.0" and the like. Without this function we'd
+ /// have to duplicate its logic everywhere it's called.
+ bool isExactlyValue(double V) const {
+ bool ignored;
+ APFloat Tmp(V);
+ Tmp.convert(getSemantics(), APFloat::rmNearestTiesToEven, &ignored);
+ return bitwiseIsEqual(Tmp);
+ }
+
+ unsigned int convertToHexString(char *DST, unsigned int HexDigits,
+ bool UpperCase, roundingMode RM) const {
+ APFLOAT_DISPATCH_ON_SEMANTICS(
+ convertToHexString(DST, HexDigits, UpperCase, RM));
+ }
+
+ bool isZero() const { return getCategory() == fcZero; }
+ bool isInfinity() const { return getCategory() == fcInfinity; }
+ bool isNaN() const { return getCategory() == fcNaN; }
+
+ bool isNegative() const { return getIEEE().isNegative(); }
+ bool isDenormal() const { APFLOAT_DISPATCH_ON_SEMANTICS(isDenormal()); }
+ bool isSignaling() const { return getIEEE().isSignaling(); }
+
+ bool isNormal() const { return !isDenormal() && isFiniteNonZero(); }
+ bool isFinite() const { return !isNaN() && !isInfinity(); }
+
+ fltCategory getCategory() const { return getIEEE().getCategory(); }
+ const fltSemantics &getSemantics() const { return *U.semantics; }
+ bool isNonZero() const { return !isZero(); }
+ bool isFiniteNonZero() const { return isFinite() && !isZero(); }
+ bool isPosZero() const { return isZero() && !isNegative(); }
+ bool isNegZero() const { return isZero() && isNegative(); }
+ bool isSmallest() const { APFLOAT_DISPATCH_ON_SEMANTICS(isSmallest()); }
+ bool isLargest() const { APFLOAT_DISPATCH_ON_SEMANTICS(isLargest()); }
+ bool isInteger() const { APFLOAT_DISPATCH_ON_SEMANTICS(isInteger()); }
+
+ APFloat &operator=(const APFloat &RHS) = default;
+ APFloat &operator=(APFloat &&RHS) = default;
+
+ void toString(SmallVectorImpl<char> &Str, unsigned FormatPrecision = 0,
+ unsigned FormatMaxPadding = 3, bool TruncateZero = true) const {
+ APFLOAT_DISPATCH_ON_SEMANTICS(
+ toString(Str, FormatPrecision, FormatMaxPadding, TruncateZero));
+ }
+
+ void print(raw_ostream &) const;
+ void dump() const;
+
+ bool getExactInverse(APFloat *inv) const {
+ APFLOAT_DISPATCH_ON_SEMANTICS(getExactInverse(inv));
+ }
+
+ friend hash_code hash_value(const APFloat &Arg);
+ friend int ilogb(const APFloat &Arg) { return ilogb(Arg.getIEEE()); }
+ friend APFloat scalbn(APFloat X, int Exp, roundingMode RM);
+ friend APFloat frexp(const APFloat &X, int &Exp, roundingMode RM);
+ friend IEEEFloat;
+ friend DoubleAPFloat;
+};
+
+/// See friend declarations above.
+///
+/// These additional declarations are required in order to compile LLVM with IBM
+/// xlC compiler.
+hash_code hash_value(const APFloat &Arg);
+inline APFloat scalbn(APFloat X, int Exp, APFloat::roundingMode RM) {
+ if (APFloat::usesLayout<detail::IEEEFloat>(X.getSemantics()))
+ return APFloat(scalbn(X.U.IEEE, Exp, RM), X.getSemantics());
+ if (APFloat::usesLayout<detail::DoubleAPFloat>(X.getSemantics()))
+ return APFloat(scalbn(X.U.Double, Exp, RM), X.getSemantics());
+ llvm_unreachable("Unexpected semantics");
+}
+
+/// Equivalent of C standard library function.
+///
+/// While the C standard says Exp is an unspecified value for infinity and nan,
+/// this returns INT_MAX for infinities, and INT_MIN for NaNs.
+inline APFloat frexp(const APFloat &X, int &Exp, APFloat::roundingMode RM) {
+ if (APFloat::usesLayout<detail::IEEEFloat>(X.getSemantics()))
+ return APFloat(frexp(X.U.IEEE, Exp, RM), X.getSemantics());
+ if (APFloat::usesLayout<detail::DoubleAPFloat>(X.getSemantics()))
+ return APFloat(frexp(X.U.Double, Exp, RM), X.getSemantics());
+ llvm_unreachable("Unexpected semantics");
+}
+/// Returns the absolute value of the argument.
+inline APFloat abs(APFloat X) {
+ X.clearSign();
+ return X;
+}
+
+/// Returns the negated value of the argument.
+inline APFloat neg(APFloat X) {
+ X.changeSign();
+ return X;
+}
+
+/// Implements IEEE minNum semantics. Returns the smaller of the 2 arguments if
+/// both are not NaN. If either argument is a NaN, returns the other argument.
+LLVM_READONLY
+inline APFloat minnum(const APFloat &A, const APFloat &B) {
+ if (A.isNaN())
+ return B;
+ if (B.isNaN())
+ return A;
+ return (B.compare(A) == APFloat::cmpLessThan) ? B : A;
+}
+
+/// Implements IEEE maxNum semantics. Returns the larger of the 2 arguments if
+/// both are not NaN. If either argument is a NaN, returns the other argument.
+LLVM_READONLY
+inline APFloat maxnum(const APFloat &A, const APFloat &B) {
+ if (A.isNaN())
+ return B;
+ if (B.isNaN())
+ return A;
+ return (A.compare(B) == APFloat::cmpLessThan) ? B : A;
+}
+
+/// Implements IEEE 754-2018 minimum semantics. Returns the smaller of 2
+/// arguments, propagating NaNs and treating -0 as less than +0.
+LLVM_READONLY
+inline APFloat minimum(const APFloat &A, const APFloat &B) {
+ if (A.isNaN())
+ return A;
+ if (B.isNaN())
+ return B;
+ if (A.isZero() && B.isZero() && (A.isNegative() != B.isNegative()))
+ return A.isNegative() ? A : B;
+ return (B.compare(A) == APFloat::cmpLessThan) ? B : A;
+}
+
+/// Implements IEEE 754-2018 maximum semantics. Returns the larger of 2
+/// arguments, propagating NaNs and treating -0 as less than +0.
+LLVM_READONLY
+inline APFloat maximum(const APFloat &A, const APFloat &B) {
+ if (A.isNaN())
+ return A;
+ if (B.isNaN())
+ return B;
+ if (A.isZero() && B.isZero() && (A.isNegative() != B.isNegative()))
+ return A.isNegative() ? B : A;
+ return (A.compare(B) == APFloat::cmpLessThan) ? B : A;
+}
+
+} // namespace llvm
+
+#undef APFLOAT_DISPATCH_ON_SEMANTICS
+#endif // LLVM_ADT_APFLOAT_H
diff --git a/third_party/llvm-project/include/llvm/ADT/APInt.h b/third_party/llvm-project/include/llvm/ADT/APInt.h
new file mode 100644
index 000000000..60a0db7e9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/APInt.h
@@ -0,0 +1,2258 @@
+//===-- llvm/ADT/APInt.h - For Arbitrary Precision Integer -----*- C++ -*--===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file implements a class to represent arbitrary precision
+/// integral constant values and operations on them.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_APINT_H
+#define LLVM_ADT_APINT_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/MathExtras.h"
+#include <cassert>
+#include <climits>
+#include <cstring>
+#include <string>
+
+namespace llvm {
+class FoldingSetNodeID;
+class StringRef;
+class hash_code;
+class raw_ostream;
+
+template <typename T> class SmallVectorImpl;
+template <typename T> class ArrayRef;
+template <typename T> class Optional;
+
+class APInt;
+
+inline APInt operator-(APInt);
+
+//===----------------------------------------------------------------------===//
+// APInt Class
+//===----------------------------------------------------------------------===//
+
+/// Class for arbitrary precision integers.
+///
+/// APInt is a functional replacement for common case unsigned integer type like
+/// "unsigned", "unsigned long" or "uint64_t", but also allows non-byte-width
+/// integer sizes and large integer value types such as 3-bits, 15-bits, or more
+/// than 64-bits of precision. APInt provides a variety of arithmetic operators
+/// and methods to manipulate integer values of any bit-width. It supports both
+/// the typical integer arithmetic and comparison operations as well as bitwise
+/// manipulation.
+///
+/// The class has several invariants worth noting:
+/// * All bit, byte, and word positions are zero-based.
+/// * Once the bit width is set, it doesn't change except by the Truncate,
+/// SignExtend, or ZeroExtend operations.
+/// * All binary operators must be on APInt instances of the same bit width.
+/// Attempting to use these operators on instances with different bit
+/// widths will yield an assertion.
+/// * The value is stored canonically as an unsigned value. For operations
+/// where it makes a difference, there are both signed and unsigned variants
+/// of the operation. For example, sdiv and udiv. However, because the bit
+/// widths must be the same, operations such as Mul and Add produce the same
+/// results regardless of whether the values are interpreted as signed or
+/// not.
+/// * In general, the class tries to follow the style of computation that LLVM
+/// uses in its IR. This simplifies its use for LLVM.
+///
+class LLVM_NODISCARD APInt {
+public:
+ typedef uint64_t WordType;
+
+ /// This enum is used to hold the constants we needed for APInt.
+ enum : unsigned {
+ /// Byte size of a word.
+ APINT_WORD_SIZE = sizeof(WordType),
+ /// Bits in a word.
+ APINT_BITS_PER_WORD = APINT_WORD_SIZE * CHAR_BIT
+ };
+
+ enum class Rounding {
+ DOWN,
+ TOWARD_ZERO,
+ UP,
+ };
+
+ static const WordType WORDTYPE_MAX = ~WordType(0);
+
+private:
+ /// This union is used to store the integer value. When the
+ /// integer bit-width <= 64, it uses VAL, otherwise it uses pVal.
+ union {
+ uint64_t VAL; ///< Used to store the <= 64 bits integer value.
+ uint64_t *pVal; ///< Used to store the >64 bits integer value.
+ } U;
+
+ unsigned BitWidth; ///< The number of bits in this APInt.
+
+ friend struct DenseMapAPIntKeyInfo;
+
+ friend class APSInt;
+
+ /// Fast internal constructor
+ ///
+ /// This constructor is used only internally for speed of construction of
+ /// temporaries. It is unsafe for general use so it is not public.
+ APInt(uint64_t *val, unsigned bits) : BitWidth(bits) {
+ U.pVal = val;
+ }
+
+ /// Determine if this APInt just has one word to store value.
+ ///
+ /// \returns true if the number of bits <= 64, false otherwise.
+ bool isSingleWord() const { return BitWidth <= APINT_BITS_PER_WORD; }
+
+ /// Determine which word a bit is in.
+ ///
+ /// \returns the word position for the specified bit position.
+ static unsigned whichWord(unsigned bitPosition) {
+ return bitPosition / APINT_BITS_PER_WORD;
+ }
+
+ /// Determine which bit in a word a bit is in.
+ ///
+ /// \returns the bit position in a word for the specified bit position
+ /// in the APInt.
+ static unsigned whichBit(unsigned bitPosition) {
+ return bitPosition % APINT_BITS_PER_WORD;
+ }
+
+ /// Get a single bit mask.
+ ///
+ /// \returns a uint64_t with only bit at "whichBit(bitPosition)" set
+ /// This method generates and returns a uint64_t (word) mask for a single
+ /// bit at a specific bit position. This is used to mask the bit in the
+ /// corresponding word.
+ static uint64_t maskBit(unsigned bitPosition) {
+ return 1ULL << whichBit(bitPosition);
+ }
+
+ /// Clear unused high order bits
+ ///
+ /// This method is used internally to clear the top "N" bits in the high order
+ /// word that are not used by the APInt. This is needed after the most
+ /// significant word is assigned a value to ensure that those bits are
+ /// zero'd out.
+ APInt &clearUnusedBits() {
+ // Compute how many bits are used in the final word
+ unsigned WordBits = ((BitWidth-1) % APINT_BITS_PER_WORD) + 1;
+
+ // Mask out the high bits.
+ uint64_t mask = WORDTYPE_MAX >> (APINT_BITS_PER_WORD - WordBits);
+ if (isSingleWord())
+ U.VAL &= mask;
+ else
+ U.pVal[getNumWords() - 1] &= mask;
+ return *this;
+ }
+
+ /// Get the word corresponding to a bit position
+ /// \returns the corresponding word for the specified bit position.
+ uint64_t getWord(unsigned bitPosition) const {
+ return isSingleWord() ? U.VAL : U.pVal[whichWord(bitPosition)];
+ }
+
+ /// Utility method to change the bit width of this APInt to new bit width,
+ /// allocating and/or deallocating as necessary. There is no guarantee on the
+ /// value of any bits upon return. Caller should populate the bits after.
+ void reallocate(unsigned NewBitWidth);
+
+ /// Convert a char array into an APInt
+ ///
+ /// \param radix 2, 8, 10, 16, or 36
+ /// Converts a string into a number. The string must be non-empty
+ /// and well-formed as a number of the given base. The bit-width
+ /// must be sufficient to hold the result.
+ ///
+ /// This is used by the constructors that take string arguments.
+ ///
+ /// StringRef::getAsInteger is superficially similar but (1) does
+ /// not assume that the string is well-formed and (2) grows the
+ /// result to hold the input.
+ void fromString(unsigned numBits, StringRef str, uint8_t radix);
+
+ /// An internal division function for dividing APInts.
+ ///
+ /// This is used by the toString method to divide by the radix. It simply
+ /// provides a more convenient form of divide for internal use since KnuthDiv
+ /// has specific constraints on its inputs. If those constraints are not met
+ /// then it provides a simpler form of divide.
+ static void divide(const WordType *LHS, unsigned lhsWords,
+ const WordType *RHS, unsigned rhsWords, WordType *Quotient,
+ WordType *Remainder);
+
+ /// out-of-line slow case for inline constructor
+ void initSlowCase(uint64_t val, bool isSigned);
+
+ /// shared code between two array constructors
+ void initFromArray(ArrayRef<uint64_t> array);
+
+ /// out-of-line slow case for inline copy constructor
+ void initSlowCase(const APInt &that);
+
+ /// out-of-line slow case for shl
+ void shlSlowCase(unsigned ShiftAmt);
+
+ /// out-of-line slow case for lshr.
+ void lshrSlowCase(unsigned ShiftAmt);
+
+ /// out-of-line slow case for ashr.
+ void ashrSlowCase(unsigned ShiftAmt);
+
+ /// out-of-line slow case for operator=
+ void AssignSlowCase(const APInt &RHS);
+
+ /// out-of-line slow case for operator==
+ bool EqualSlowCase(const APInt &RHS) const LLVM_READONLY;
+
+ /// out-of-line slow case for countLeadingZeros
+ unsigned countLeadingZerosSlowCase() const LLVM_READONLY;
+
+ /// out-of-line slow case for countLeadingOnes.
+ unsigned countLeadingOnesSlowCase() const LLVM_READONLY;
+
+ /// out-of-line slow case for countTrailingZeros.
+ unsigned countTrailingZerosSlowCase() const LLVM_READONLY;
+
+ /// out-of-line slow case for countTrailingOnes
+ unsigned countTrailingOnesSlowCase() const LLVM_READONLY;
+
+ /// out-of-line slow case for countPopulation
+ unsigned countPopulationSlowCase() const LLVM_READONLY;
+
+ /// out-of-line slow case for intersects.
+ bool intersectsSlowCase(const APInt &RHS) const LLVM_READONLY;
+
+ /// out-of-line slow case for isSubsetOf.
+ bool isSubsetOfSlowCase(const APInt &RHS) const LLVM_READONLY;
+
+ /// out-of-line slow case for setBits.
+ void setBitsSlowCase(unsigned loBit, unsigned hiBit);
+
+ /// out-of-line slow case for flipAllBits.
+ void flipAllBitsSlowCase();
+
+ /// out-of-line slow case for operator&=.
+ void AndAssignSlowCase(const APInt& RHS);
+
+ /// out-of-line slow case for operator|=.
+ void OrAssignSlowCase(const APInt& RHS);
+
+ /// out-of-line slow case for operator^=.
+ void XorAssignSlowCase(const APInt& RHS);
+
+ /// Unsigned comparison. Returns -1, 0, or 1 if this APInt is less than, equal
+ /// to, or greater than RHS.
+ int compare(const APInt &RHS) const LLVM_READONLY;
+
+ /// Signed comparison. Returns -1, 0, or 1 if this APInt is less than, equal
+ /// to, or greater than RHS.
+ int compareSigned(const APInt &RHS) const LLVM_READONLY;
+
+public:
+ /// \name Constructors
+ /// @{
+
+ /// Create a new APInt of numBits width, initialized as val.
+ ///
+ /// If isSigned is true then val is treated as if it were a signed value
+ /// (i.e. as an int64_t) and the appropriate sign extension to the bit width
+ /// will be done. Otherwise, no sign extension occurs (high order bits beyond
+ /// the range of val are zero filled).
+ ///
+ /// \param numBits the bit width of the constructed APInt
+ /// \param val the initial value of the APInt
+ /// \param isSigned how to treat signedness of val
+ APInt(unsigned numBits, uint64_t val, bool isSigned = false)
+ : BitWidth(numBits) {
+ assert(BitWidth && "bitwidth too small");
+ if (isSingleWord()) {
+ U.VAL = val;
+ clearUnusedBits();
+ } else {
+ initSlowCase(val, isSigned);
+ }
+ }
+
+ /// Construct an APInt of numBits width, initialized as bigVal[].
+ ///
+ /// Note that bigVal.size() can be smaller or larger than the corresponding
+ /// bit width but any extraneous bits will be dropped.
+ ///
+ /// \param numBits the bit width of the constructed APInt
+ /// \param bigVal a sequence of words to form the initial value of the APInt
+ APInt(unsigned numBits, ArrayRef<uint64_t> bigVal);
+
+ /// Equivalent to APInt(numBits, ArrayRef<uint64_t>(bigVal, numWords)), but
+ /// deprecated because this constructor is prone to ambiguity with the
+ /// APInt(unsigned, uint64_t, bool) constructor.
+ ///
+ /// If this overload is ever deleted, care should be taken to prevent calls
+ /// from being incorrectly captured by the APInt(unsigned, uint64_t, bool)
+ /// constructor.
+ APInt(unsigned numBits, unsigned numWords, const uint64_t bigVal[]);
+
+ /// Construct an APInt from a string representation.
+ ///
+ /// This constructor interprets the string \p str in the given radix. The
+ /// interpretation stops when the first character that is not suitable for the
+ /// radix is encountered, or the end of the string. Acceptable radix values
+ /// are 2, 8, 10, 16, and 36. It is an error for the value implied by the
+ /// string to require more bits than numBits.
+ ///
+ /// \param numBits the bit width of the constructed APInt
+ /// \param str the string to be interpreted
+ /// \param radix the radix to use for the conversion
+ APInt(unsigned numBits, StringRef str, uint8_t radix);
+
+ /// Simply makes *this a copy of that.
+ /// Copy Constructor.
+ APInt(const APInt &that) : BitWidth(that.BitWidth) {
+ if (isSingleWord())
+ U.VAL = that.U.VAL;
+ else
+ initSlowCase(that);
+ }
+
+ /// Move Constructor.
+ APInt(APInt &&that) : BitWidth(that.BitWidth) {
+ memcpy(&U, &that.U, sizeof(U));
+ that.BitWidth = 0;
+ }
+
+ /// Destructor.
+ ~APInt() {
+ if (needsCleanup())
+ delete[] U.pVal;
+ }
+
+ /// Default constructor that creates an uninteresting APInt
+ /// representing a 1-bit zero value.
+ ///
+ /// This is useful for object deserialization (pair this with the static
+ /// method Read).
+ explicit APInt() : BitWidth(1) { U.VAL = 0; }
+
+ /// Returns whether this instance allocated memory.
+ bool needsCleanup() const { return !isSingleWord(); }
+
+ /// Used to insert APInt objects, or objects that contain APInt objects, into
+ /// FoldingSets.
+ void Profile(FoldingSetNodeID &id) const;
+
+ /// @}
+ /// \name Value Tests
+ /// @{
+
+ /// Determine sign of this APInt.
+ ///
+ /// This tests the high bit of this APInt to determine if it is set.
+ ///
+ /// \returns true if this APInt is negative, false otherwise
+ bool isNegative() const { return (*this)[BitWidth - 1]; }
+
+ /// Determine if this APInt Value is non-negative (>= 0)
+ ///
+ /// This tests the high bit of the APInt to determine if it is unset.
+ bool isNonNegative() const { return !isNegative(); }
+
+ /// Determine if sign bit of this APInt is set.
+ ///
+ /// This tests the high bit of this APInt to determine if it is set.
+ ///
+ /// \returns true if this APInt has its sign bit set, false otherwise.
+ bool isSignBitSet() const { return (*this)[BitWidth-1]; }
+
+ /// Determine if sign bit of this APInt is clear.
+ ///
+ /// This tests the high bit of this APInt to determine if it is clear.
+ ///
+ /// \returns true if this APInt has its sign bit clear, false otherwise.
+ bool isSignBitClear() const { return !isSignBitSet(); }
+
+ /// Determine if this APInt Value is positive.
+ ///
+ /// This tests if the value of this APInt is positive (> 0). Note
+ /// that 0 is not a positive value.
+ ///
+ /// \returns true if this APInt is positive.
+ bool isStrictlyPositive() const { return isNonNegative() && !isNullValue(); }
+
+ /// Determine if all bits are set
+ ///
+ /// This checks to see if the value has all bits of the APInt are set or not.
+ bool isAllOnesValue() const {
+ if (isSingleWord())
+ return U.VAL == WORDTYPE_MAX >> (APINT_BITS_PER_WORD - BitWidth);
+ return countTrailingOnesSlowCase() == BitWidth;
+ }
+
+ /// Determine if all bits are clear
+ ///
+ /// This checks to see if the value has all bits of the APInt are clear or
+ /// not.
+ bool isNullValue() const { return !*this; }
+
+ /// Determine if this is a value of 1.
+ ///
+ /// This checks to see if the value of this APInt is one.
+ bool isOneValue() const {
+ if (isSingleWord())
+ return U.VAL == 1;
+ return countLeadingZerosSlowCase() == BitWidth - 1;
+ }
+
+ /// Determine if this is the largest unsigned value.
+ ///
+ /// This checks to see if the value of this APInt is the maximum unsigned
+ /// value for the APInt's bit width.
+ bool isMaxValue() const { return isAllOnesValue(); }
+
+ /// Determine if this is the largest signed value.
+ ///
+ /// This checks to see if the value of this APInt is the maximum signed
+ /// value for the APInt's bit width.
+ bool isMaxSignedValue() const {
+ if (isSingleWord())
+ return U.VAL == ((WordType(1) << (BitWidth - 1)) - 1);
+ return !isNegative() && countTrailingOnesSlowCase() == BitWidth - 1;
+ }
+
+ /// Determine if this is the smallest unsigned value.
+ ///
+ /// This checks to see if the value of this APInt is the minimum unsigned
+ /// value for the APInt's bit width.
+ bool isMinValue() const { return isNullValue(); }
+
+ /// Determine if this is the smallest signed value.
+ ///
+ /// This checks to see if the value of this APInt is the minimum signed
+ /// value for the APInt's bit width.
+ bool isMinSignedValue() const {
+ if (isSingleWord())
+ return U.VAL == (WordType(1) << (BitWidth - 1));
+ return isNegative() && countTrailingZerosSlowCase() == BitWidth - 1;
+ }
+
+ /// Check if this APInt has an N-bits unsigned integer value.
+ bool isIntN(unsigned N) const {
+ assert(N && "N == 0 ???");
+ return getActiveBits() <= N;
+ }
+
+ /// Check if this APInt has an N-bits signed integer value.
+ bool isSignedIntN(unsigned N) const {
+ assert(N && "N == 0 ???");
+ return getMinSignedBits() <= N;
+ }
+
+ /// Check if this APInt's value is a power of two greater than zero.
+ ///
+ /// \returns true if the argument APInt value is a power of two > 0.
+ bool isPowerOf2() const {
+ if (isSingleWord())
+ return isPowerOf2_64(U.VAL);
+ return countPopulationSlowCase() == 1;
+ }
+
+ /// Check if the APInt's value is returned by getSignMask.
+ ///
+ /// \returns true if this is the value returned by getSignMask.
+ bool isSignMask() const { return isMinSignedValue(); }
+
+ /// Convert APInt to a boolean value.
+ ///
+ /// This converts the APInt to a boolean value as a test against zero.
+ bool getBoolValue() const { return !!*this; }
+
+ /// If this value is smaller than the specified limit, return it, otherwise
+ /// return the limit value. This causes the value to saturate to the limit.
+ uint64_t getLimitedValue(uint64_t Limit = UINT64_MAX) const {
+ return ugt(Limit) ? Limit : getZExtValue();
+ }
+
+ /// Check if the APInt consists of a repeated bit pattern.
+ ///
+ /// e.g. 0x01010101 satisfies isSplat(8).
+ /// \param SplatSizeInBits The size of the pattern in bits. Must divide bit
+ /// width without remainder.
+ bool isSplat(unsigned SplatSizeInBits) const;
+
+ /// \returns true if this APInt value is a sequence of \param numBits ones
+ /// starting at the least significant bit with the remainder zero.
+ bool isMask(unsigned numBits) const {
+ assert(numBits != 0 && "numBits must be non-zero");
+ assert(numBits <= BitWidth && "numBits out of range");
+ if (isSingleWord())
+ return U.VAL == (WORDTYPE_MAX >> (APINT_BITS_PER_WORD - numBits));
+ unsigned Ones = countTrailingOnesSlowCase();
+ return (numBits == Ones) &&
+ ((Ones + countLeadingZerosSlowCase()) == BitWidth);
+ }
+
+ /// \returns true if this APInt is a non-empty sequence of ones starting at
+ /// the least significant bit with the remainder zero.
+ /// Ex. isMask(0x0000FFFFU) == true.
+ bool isMask() const {
+ if (isSingleWord())
+ return isMask_64(U.VAL);
+ unsigned Ones = countTrailingOnesSlowCase();
+ return (Ones > 0) && ((Ones + countLeadingZerosSlowCase()) == BitWidth);
+ }
+
+ /// Return true if this APInt value contains a sequence of ones with
+ /// the remainder zero.
+ bool isShiftedMask() const {
+ if (isSingleWord())
+ return isShiftedMask_64(U.VAL);
+ unsigned Ones = countPopulationSlowCase();
+ unsigned LeadZ = countLeadingZerosSlowCase();
+ return (Ones + LeadZ + countTrailingZeros()) == BitWidth;
+ }
+
+ /// @}
+ /// \name Value Generators
+ /// @{
+
+ /// Gets maximum unsigned value of APInt for specific bit width.
+ static APInt getMaxValue(unsigned numBits) {
+ return getAllOnesValue(numBits);
+ }
+
+ /// Gets maximum signed value of APInt for a specific bit width.
+ static APInt getSignedMaxValue(unsigned numBits) {
+ APInt API = getAllOnesValue(numBits);
+ API.clearBit(numBits - 1);
+ return API;
+ }
+
+ /// Gets minimum unsigned value of APInt for a specific bit width.
+ static APInt getMinValue(unsigned numBits) { return APInt(numBits, 0); }
+
+ /// Gets minimum signed value of APInt for a specific bit width.
+ static APInt getSignedMinValue(unsigned numBits) {
+ APInt API(numBits, 0);
+ API.setBit(numBits - 1);
+ return API;
+ }
+
+ /// Get the SignMask for a specific bit width.
+ ///
+ /// This is just a wrapper function of getSignedMinValue(), and it helps code
+ /// readability when we want to get a SignMask.
+ static APInt getSignMask(unsigned BitWidth) {
+ return getSignedMinValue(BitWidth);
+ }
+
+ /// Get the all-ones value.
+ ///
+ /// \returns the all-ones value for an APInt of the specified bit-width.
+ static APInt getAllOnesValue(unsigned numBits) {
+ return APInt(numBits, WORDTYPE_MAX, true);
+ }
+
+ /// Get the '0' value.
+ ///
+ /// \returns the '0' value for an APInt of the specified bit-width.
+ static APInt getNullValue(unsigned numBits) { return APInt(numBits, 0); }
+
+ /// Compute an APInt containing numBits highbits from this APInt.
+ ///
+ /// Get an APInt with the same BitWidth as this APInt, just zero mask
+ /// the low bits and right shift to the least significant bit.
+ ///
+ /// \returns the high "numBits" bits of this APInt.
+ APInt getHiBits(unsigned numBits) const;
+
+ /// Compute an APInt containing numBits lowbits from this APInt.
+ ///
+ /// Get an APInt with the same BitWidth as this APInt, just zero mask
+ /// the high bits.
+ ///
+ /// \returns the low "numBits" bits of this APInt.
+ APInt getLoBits(unsigned numBits) const;
+
+ /// Return an APInt with exactly one bit set in the result.
+ static APInt getOneBitSet(unsigned numBits, unsigned BitNo) {
+ APInt Res(numBits, 0);
+ Res.setBit(BitNo);
+ return Res;
+ }
+
+ /// Get a value with a block of bits set.
+ ///
+ /// Constructs an APInt value that has a contiguous range of bits set. The
+ /// bits from loBit (inclusive) to hiBit (exclusive) will be set. All other
+ /// bits will be zero. For example, with parameters(32, 0, 16) you would get
+ /// 0x0000FFFF. If hiBit is less than loBit then the set bits "wrap". For
+ /// example, with parameters (32, 28, 4), you would get 0xF000000F.
+ ///
+ /// \param numBits the intended bit width of the result
+ /// \param loBit the index of the lowest bit set.
+ /// \param hiBit the index of the highest bit set.
+ ///
+ /// \returns An APInt value with the requested bits set.
+ static APInt getBitsSet(unsigned numBits, unsigned loBit, unsigned hiBit) {
+ APInt Res(numBits, 0);
+ Res.setBits(loBit, hiBit);
+ return Res;
+ }
+
+ /// Get a value with upper bits starting at loBit set.
+ ///
+ /// Constructs an APInt value that has a contiguous range of bits set. The
+ /// bits from loBit (inclusive) to numBits (exclusive) will be set. All other
+ /// bits will be zero. For example, with parameters(32, 12) you would get
+ /// 0xFFFFF000.
+ ///
+ /// \param numBits the intended bit width of the result
+ /// \param loBit the index of the lowest bit to set.
+ ///
+ /// \returns An APInt value with the requested bits set.
+ static APInt getBitsSetFrom(unsigned numBits, unsigned loBit) {
+ APInt Res(numBits, 0);
+ Res.setBitsFrom(loBit);
+ return Res;
+ }
+
+ /// Get a value with high bits set
+ ///
+ /// Constructs an APInt value that has the top hiBitsSet bits set.
+ ///
+ /// \param numBits the bitwidth of the result
+ /// \param hiBitsSet the number of high-order bits set in the result.
+ static APInt getHighBitsSet(unsigned numBits, unsigned hiBitsSet) {
+ APInt Res(numBits, 0);
+ Res.setHighBits(hiBitsSet);
+ return Res;
+ }
+
+ /// Get a value with low bits set
+ ///
+ /// Constructs an APInt value that has the bottom loBitsSet bits set.
+ ///
+ /// \param numBits the bitwidth of the result
+ /// \param loBitsSet the number of low-order bits set in the result.
+ static APInt getLowBitsSet(unsigned numBits, unsigned loBitsSet) {
+ APInt Res(numBits, 0);
+ Res.setLowBits(loBitsSet);
+ return Res;
+ }
+
+ /// Return a value containing V broadcasted over NewLen bits.
+ static APInt getSplat(unsigned NewLen, const APInt &V);
+
+ /// Determine if two APInts have the same value, after zero-extending
+ /// one of them (if needed!) to ensure that the bit-widths match.
+ static bool isSameValue(const APInt &I1, const APInt &I2) {
+ if (I1.getBitWidth() == I2.getBitWidth())
+ return I1 == I2;
+
+ if (I1.getBitWidth() > I2.getBitWidth())
+ return I1 == I2.zext(I1.getBitWidth());
+
+ return I1.zext(I2.getBitWidth()) == I2;
+ }
+
+ /// Overload to compute a hash_code for an APInt value.
+ friend hash_code hash_value(const APInt &Arg);
+
+ /// This function returns a pointer to the internal storage of the APInt.
+ /// This is useful for writing out the APInt in binary form without any
+ /// conversions.
+ const uint64_t *getRawData() const {
+ if (isSingleWord())
+ return &U.VAL;
+ return &U.pVal[0];
+ }
+
+ /// @}
+ /// \name Unary Operators
+ /// @{
+
+ /// Postfix increment operator.
+ ///
+ /// Increments *this by 1.
+ ///
+ /// \returns a new APInt value representing the original value of *this.
+ const APInt operator++(int) {
+ APInt API(*this);
+ ++(*this);
+ return API;
+ }
+
+ /// Prefix increment operator.
+ ///
+ /// \returns *this incremented by one
+ APInt &operator++();
+
+ /// Postfix decrement operator.
+ ///
+ /// Decrements *this by 1.
+ ///
+ /// \returns a new APInt value representing the original value of *this.
+ const APInt operator--(int) {
+ APInt API(*this);
+ --(*this);
+ return API;
+ }
+
+ /// Prefix decrement operator.
+ ///
+ /// \returns *this decremented by one.
+ APInt &operator--();
+
+ /// Logical negation operator.
+ ///
+ /// Performs logical negation operation on this APInt.
+ ///
+ /// \returns true if *this is zero, false otherwise.
+ bool operator!() const {
+ if (isSingleWord())
+ return U.VAL == 0;
+ return countLeadingZerosSlowCase() == BitWidth;
+ }
+
+ /// @}
+ /// \name Assignment Operators
+ /// @{
+
+ /// Copy assignment operator.
+ ///
+ /// \returns *this after assignment of RHS.
+ APInt &operator=(const APInt &RHS) {
+ // If the bitwidths are the same, we can avoid mucking with memory
+ if (isSingleWord() && RHS.isSingleWord()) {
+ U.VAL = RHS.U.VAL;
+ BitWidth = RHS.BitWidth;
+ return clearUnusedBits();
+ }
+
+ AssignSlowCase(RHS);
+ return *this;
+ }
+
+ /// Move assignment operator.
+ APInt &operator=(APInt &&that) {
+#ifdef _MSC_VER
+ // The MSVC std::shuffle implementation still does self-assignment.
+ if (this == &that)
+ return *this;
+#endif
+ assert(this != &that && "Self-move not supported");
+ if (!isSingleWord())
+ delete[] U.pVal;
+
+ // Use memcpy so that type based alias analysis sees both VAL and pVal
+ // as modified.
+ memcpy(&U, &that.U, sizeof(U));
+
+ BitWidth = that.BitWidth;
+ that.BitWidth = 0;
+
+ return *this;
+ }
+
+ /// Assignment operator.
+ ///
+ /// The RHS value is assigned to *this. If the significant bits in RHS exceed
+ /// the bit width, the excess bits are truncated. If the bit width is larger
+ /// than 64, the value is zero filled in the unspecified high order bits.
+ ///
+ /// \returns *this after assignment of RHS value.
+ APInt &operator=(uint64_t RHS) {
+ if (isSingleWord()) {
+ U.VAL = RHS;
+ clearUnusedBits();
+ } else {
+ U.pVal[0] = RHS;
+ memset(U.pVal+1, 0, (getNumWords() - 1) * APINT_WORD_SIZE);
+ }
+ return *this;
+ }
+
+ /// Bitwise AND assignment operator.
+ ///
+ /// Performs a bitwise AND operation on this APInt and RHS. The result is
+ /// assigned to *this.
+ ///
+ /// \returns *this after ANDing with RHS.
+ APInt &operator&=(const APInt &RHS) {
+ assert(BitWidth == RHS.BitWidth && "Bit widths must be the same");
+ if (isSingleWord())
+ U.VAL &= RHS.U.VAL;
+ else
+ AndAssignSlowCase(RHS);
+ return *this;
+ }
+
+ /// Bitwise AND assignment operator.
+ ///
+ /// Performs a bitwise AND operation on this APInt and RHS. RHS is
+ /// logically zero-extended or truncated to match the bit-width of
+ /// the LHS.
+ APInt &operator&=(uint64_t RHS) {
+ if (isSingleWord()) {
+ U.VAL &= RHS;
+ return *this;
+ }
+ U.pVal[0] &= RHS;
+ memset(U.pVal+1, 0, (getNumWords() - 1) * APINT_WORD_SIZE);
+ return *this;
+ }
+
+ /// Bitwise OR assignment operator.
+ ///
+ /// Performs a bitwise OR operation on this APInt and RHS. The result is
+ /// assigned *this;
+ ///
+ /// \returns *this after ORing with RHS.
+ APInt &operator|=(const APInt &RHS) {
+ assert(BitWidth == RHS.BitWidth && "Bit widths must be the same");
+ if (isSingleWord())
+ U.VAL |= RHS.U.VAL;
+ else
+ OrAssignSlowCase(RHS);
+ return *this;
+ }
+
+ /// Bitwise OR assignment operator.
+ ///
+ /// Performs a bitwise OR operation on this APInt and RHS. RHS is
+ /// logically zero-extended or truncated to match the bit-width of
+ /// the LHS.
+ APInt &operator|=(uint64_t RHS) {
+ if (isSingleWord()) {
+ U.VAL |= RHS;
+ clearUnusedBits();
+ } else {
+ U.pVal[0] |= RHS;
+ }
+ return *this;
+ }
+
+ /// Bitwise XOR assignment operator.
+ ///
+ /// Performs a bitwise XOR operation on this APInt and RHS. The result is
+ /// assigned to *this.
+ ///
+ /// \returns *this after XORing with RHS.
+ APInt &operator^=(const APInt &RHS) {
+ assert(BitWidth == RHS.BitWidth && "Bit widths must be the same");
+ if (isSingleWord())
+ U.VAL ^= RHS.U.VAL;
+ else
+ XorAssignSlowCase(RHS);
+ return *this;
+ }
+
+ /// Bitwise XOR assignment operator.
+ ///
+ /// Performs a bitwise XOR operation on this APInt and RHS. RHS is
+ /// logically zero-extended or truncated to match the bit-width of
+ /// the LHS.
+ APInt &operator^=(uint64_t RHS) {
+ if (isSingleWord()) {
+ U.VAL ^= RHS;
+ clearUnusedBits();
+ } else {
+ U.pVal[0] ^= RHS;
+ }
+ return *this;
+ }
+
+ /// Multiplication assignment operator.
+ ///
+ /// Multiplies this APInt by RHS and assigns the result to *this.
+ ///
+ /// \returns *this
+ APInt &operator*=(const APInt &RHS);
+ APInt &operator*=(uint64_t RHS);
+
+ /// Addition assignment operator.
+ ///
+ /// Adds RHS to *this and assigns the result to *this.
+ ///
+ /// \returns *this
+ APInt &operator+=(const APInt &RHS);
+ APInt &operator+=(uint64_t RHS);
+
+ /// Subtraction assignment operator.
+ ///
+ /// Subtracts RHS from *this and assigns the result to *this.
+ ///
+ /// \returns *this
+ APInt &operator-=(const APInt &RHS);
+ APInt &operator-=(uint64_t RHS);
+
+ /// Left-shift assignment function.
+ ///
+ /// Shifts *this left by shiftAmt and assigns the result to *this.
+ ///
+ /// \returns *this after shifting left by ShiftAmt
+ APInt &operator<<=(unsigned ShiftAmt) {
+ assert(ShiftAmt <= BitWidth && "Invalid shift amount");
+ if (isSingleWord()) {
+ if (ShiftAmt == BitWidth)
+ U.VAL = 0;
+ else
+ U.VAL <<= ShiftAmt;
+ return clearUnusedBits();
+ }
+ shlSlowCase(ShiftAmt);
+ return *this;
+ }
+
+ /// Left-shift assignment function.
+ ///
+ /// Shifts *this left by shiftAmt and assigns the result to *this.
+ ///
+ /// \returns *this after shifting left by ShiftAmt
+ APInt &operator<<=(const APInt &ShiftAmt);
+
+ /// @}
+ /// \name Binary Operators
+ /// @{
+
+ /// Multiplication operator.
+ ///
+ /// Multiplies this APInt by RHS and returns the result.
+ APInt operator*(const APInt &RHS) const;
+
+ /// Left logical shift operator.
+ ///
+ /// Shifts this APInt left by \p Bits and returns the result.
+ APInt operator<<(unsigned Bits) const { return shl(Bits); }
+
+ /// Left logical shift operator.
+ ///
+ /// Shifts this APInt left by \p Bits and returns the result.
+ APInt operator<<(const APInt &Bits) const { return shl(Bits); }
+
+ /// Arithmetic right-shift function.
+ ///
+ /// Arithmetic right-shift this APInt by shiftAmt.
+ APInt ashr(unsigned ShiftAmt) const {
+ APInt R(*this);
+ R.ashrInPlace(ShiftAmt);
+ return R;
+ }
+
+ /// Arithmetic right-shift this APInt by ShiftAmt in place.
+ void ashrInPlace(unsigned ShiftAmt) {
+ assert(ShiftAmt <= BitWidth && "Invalid shift amount");
+ if (isSingleWord()) {
+ int64_t SExtVAL = SignExtend64(U.VAL, BitWidth);
+ if (ShiftAmt == BitWidth)
+ U.VAL = SExtVAL >> (APINT_BITS_PER_WORD - 1); // Fill with sign bit.
+ else
+ U.VAL = SExtVAL >> ShiftAmt;
+ clearUnusedBits();
+ return;
+ }
+ ashrSlowCase(ShiftAmt);
+ }
+
+ /// Logical right-shift function.
+ ///
+ /// Logical right-shift this APInt by shiftAmt.
+ APInt lshr(unsigned shiftAmt) const {
+ APInt R(*this);
+ R.lshrInPlace(shiftAmt);
+ return R;
+ }
+
+ /// Logical right-shift this APInt by ShiftAmt in place.
+ void lshrInPlace(unsigned ShiftAmt) {
+ assert(ShiftAmt <= BitWidth && "Invalid shift amount");
+ if (isSingleWord()) {
+ if (ShiftAmt == BitWidth)
+ U.VAL = 0;
+ else
+ U.VAL >>= ShiftAmt;
+ return;
+ }
+ lshrSlowCase(ShiftAmt);
+ }
+
+ /// Left-shift function.
+ ///
+ /// Left-shift this APInt by shiftAmt.
+ APInt shl(unsigned shiftAmt) const {
+ APInt R(*this);
+ R <<= shiftAmt;
+ return R;
+ }
+
+ /// Rotate left by rotateAmt.
+ APInt rotl(unsigned rotateAmt) const;
+
+ /// Rotate right by rotateAmt.
+ APInt rotr(unsigned rotateAmt) const;
+
+ /// Arithmetic right-shift function.
+ ///
+ /// Arithmetic right-shift this APInt by shiftAmt.
+ APInt ashr(const APInt &ShiftAmt) const {
+ APInt R(*this);
+ R.ashrInPlace(ShiftAmt);
+ return R;
+ }
+
+ /// Arithmetic right-shift this APInt by shiftAmt in place.
+ void ashrInPlace(const APInt &shiftAmt);
+
+ /// Logical right-shift function.
+ ///
+ /// Logical right-shift this APInt by shiftAmt.
+ APInt lshr(const APInt &ShiftAmt) const {
+ APInt R(*this);
+ R.lshrInPlace(ShiftAmt);
+ return R;
+ }
+
+ /// Logical right-shift this APInt by ShiftAmt in place.
+ void lshrInPlace(const APInt &ShiftAmt);
+
+ /// Left-shift function.
+ ///
+ /// Left-shift this APInt by shiftAmt.
+ APInt shl(const APInt &ShiftAmt) const {
+ APInt R(*this);
+ R <<= ShiftAmt;
+ return R;
+ }
+
+ /// Rotate left by rotateAmt.
+ APInt rotl(const APInt &rotateAmt) const;
+
+ /// Rotate right by rotateAmt.
+ APInt rotr(const APInt &rotateAmt) const;
+
+ /// Unsigned division operation.
+ ///
+ /// Perform an unsigned divide operation on this APInt by RHS. Both this and
+ /// RHS are treated as unsigned quantities for purposes of this division.
+ ///
+ /// \returns a new APInt value containing the division result, rounded towards
+ /// zero.
+ APInt udiv(const APInt &RHS) const;
+ APInt udiv(uint64_t RHS) const;
+
+ /// Signed division function for APInt.
+ ///
+ /// Signed divide this APInt by APInt RHS.
+ ///
+ /// The result is rounded towards zero.
+ APInt sdiv(const APInt &RHS) const;
+ APInt sdiv(int64_t RHS) const;
+
+ /// Unsigned remainder operation.
+ ///
+ /// Perform an unsigned remainder operation on this APInt with RHS being the
+ /// divisor. Both this and RHS are treated as unsigned quantities for purposes
+ /// of this operation. Note that this is a true remainder operation and not a
+ /// modulo operation because the sign follows the sign of the dividend which
+ /// is *this.
+ ///
+ /// \returns a new APInt value containing the remainder result
+ APInt urem(const APInt &RHS) const;
+ uint64_t urem(uint64_t RHS) const;
+
+ /// Function for signed remainder operation.
+ ///
+ /// Signed remainder operation on APInt.
+ APInt srem(const APInt &RHS) const;
+ int64_t srem(int64_t RHS) const;
+
+ /// Dual division/remainder interface.
+ ///
+ /// Sometimes it is convenient to divide two APInt values and obtain both the
+ /// quotient and remainder. This function does both operations in the same
+ /// computation making it a little more efficient. The pair of input arguments
+ /// may overlap with the pair of output arguments. It is safe to call
+ /// udivrem(X, Y, X, Y), for example.
+ static void udivrem(const APInt &LHS, const APInt &RHS, APInt &Quotient,
+ APInt &Remainder);
+ static void udivrem(const APInt &LHS, uint64_t RHS, APInt &Quotient,
+ uint64_t &Remainder);
+
+ static void sdivrem(const APInt &LHS, const APInt &RHS, APInt &Quotient,
+ APInt &Remainder);
+ static void sdivrem(const APInt &LHS, int64_t RHS, APInt &Quotient,
+ int64_t &Remainder);
+
+ // Operations that return overflow indicators.
+ APInt sadd_ov(const APInt &RHS, bool &Overflow) const;
+ APInt uadd_ov(const APInt &RHS, bool &Overflow) const;
+ APInt ssub_ov(const APInt &RHS, bool &Overflow) const;
+ APInt usub_ov(const APInt &RHS, bool &Overflow) const;
+ APInt sdiv_ov(const APInt &RHS, bool &Overflow) const;
+ APInt smul_ov(const APInt &RHS, bool &Overflow) const;
+ APInt umul_ov(const APInt &RHS, bool &Overflow) const;
+ APInt sshl_ov(const APInt &Amt, bool &Overflow) const;
+ APInt ushl_ov(const APInt &Amt, bool &Overflow) const;
+
+ // Operations that saturate
+ APInt sadd_sat(const APInt &RHS) const;
+ APInt uadd_sat(const APInt &RHS) const;
+ APInt ssub_sat(const APInt &RHS) const;
+ APInt usub_sat(const APInt &RHS) const;
+ APInt smul_sat(const APInt &RHS) const;
+ APInt umul_sat(const APInt &RHS) const;
+ APInt sshl_sat(const APInt &RHS) const;
+ APInt ushl_sat(const APInt &RHS) const;
+
+ /// Array-indexing support.
+ ///
+ /// \returns the bit value at bitPosition
+ bool operator[](unsigned bitPosition) const {
+ assert(bitPosition < getBitWidth() && "Bit position out of bounds!");
+ return (maskBit(bitPosition) & getWord(bitPosition)) != 0;
+ }
+
+ /// @}
+ /// \name Comparison Operators
+ /// @{
+
+ /// Equality operator.
+ ///
+ /// Compares this APInt with RHS for the validity of the equality
+ /// relationship.
+ bool operator==(const APInt &RHS) const {
+ assert(BitWidth == RHS.BitWidth && "Comparison requires equal bit widths");
+ if (isSingleWord())
+ return U.VAL == RHS.U.VAL;
+ return EqualSlowCase(RHS);
+ }
+
+ /// Equality operator.
+ ///
+ /// Compares this APInt with a uint64_t for the validity of the equality
+ /// relationship.
+ ///
+ /// \returns true if *this == Val
+ bool operator==(uint64_t Val) const {
+ return (isSingleWord() || getActiveBits() <= 64) && getZExtValue() == Val;
+ }
+
+ /// Equality comparison.
+ ///
+ /// Compares this APInt with RHS for the validity of the equality
+ /// relationship.
+ ///
+ /// \returns true if *this == Val
+ bool eq(const APInt &RHS) const { return (*this) == RHS; }
+
+ /// Inequality operator.
+ ///
+ /// Compares this APInt with RHS for the validity of the inequality
+ /// relationship.
+ ///
+ /// \returns true if *this != Val
+ bool operator!=(const APInt &RHS) const { return !((*this) == RHS); }
+
+ /// Inequality operator.
+ ///
+ /// Compares this APInt with a uint64_t for the validity of the inequality
+ /// relationship.
+ ///
+ /// \returns true if *this != Val
+ bool operator!=(uint64_t Val) const { return !((*this) == Val); }
+
+ /// Inequality comparison
+ ///
+ /// Compares this APInt with RHS for the validity of the inequality
+ /// relationship.
+ ///
+ /// \returns true if *this != Val
+ bool ne(const APInt &RHS) const { return !((*this) == RHS); }
+
+ /// Unsigned less than comparison
+ ///
+ /// Regards both *this and RHS as unsigned quantities and compares them for
+ /// the validity of the less-than relationship.
+ ///
+ /// \returns true if *this < RHS when both are considered unsigned.
+ bool ult(const APInt &RHS) const { return compare(RHS) < 0; }
+
+ /// Unsigned less than comparison
+ ///
+ /// Regards both *this as an unsigned quantity and compares it with RHS for
+ /// the validity of the less-than relationship.
+ ///
+ /// \returns true if *this < RHS when considered unsigned.
+ bool ult(uint64_t RHS) const {
+ // Only need to check active bits if not a single word.
+ return (isSingleWord() || getActiveBits() <= 64) && getZExtValue() < RHS;
+ }
+
+ /// Signed less than comparison
+ ///
+ /// Regards both *this and RHS as signed quantities and compares them for
+ /// validity of the less-than relationship.
+ ///
+ /// \returns true if *this < RHS when both are considered signed.
+ bool slt(const APInt &RHS) const { return compareSigned(RHS) < 0; }
+
+ /// Signed less than comparison
+ ///
+ /// Regards both *this as a signed quantity and compares it with RHS for
+ /// the validity of the less-than relationship.
+ ///
+ /// \returns true if *this < RHS when considered signed.
+ bool slt(int64_t RHS) const {
+ return (!isSingleWord() && getMinSignedBits() > 64) ? isNegative()
+ : getSExtValue() < RHS;
+ }
+
+ /// Unsigned less or equal comparison
+ ///
+ /// Regards both *this and RHS as unsigned quantities and compares them for
+ /// validity of the less-or-equal relationship.
+ ///
+ /// \returns true if *this <= RHS when both are considered unsigned.
+ bool ule(const APInt &RHS) const { return compare(RHS) <= 0; }
+
+ /// Unsigned less or equal comparison
+ ///
+ /// Regards both *this as an unsigned quantity and compares it with RHS for
+ /// the validity of the less-or-equal relationship.
+ ///
+ /// \returns true if *this <= RHS when considered unsigned.
+ bool ule(uint64_t RHS) const { return !ugt(RHS); }
+
+ /// Signed less or equal comparison
+ ///
+ /// Regards both *this and RHS as signed quantities and compares them for
+ /// validity of the less-or-equal relationship.
+ ///
+ /// \returns true if *this <= RHS when both are considered signed.
+ bool sle(const APInt &RHS) const { return compareSigned(RHS) <= 0; }
+
+ /// Signed less or equal comparison
+ ///
+ /// Regards both *this as a signed quantity and compares it with RHS for the
+ /// validity of the less-or-equal relationship.
+ ///
+ /// \returns true if *this <= RHS when considered signed.
+ bool sle(uint64_t RHS) const { return !sgt(RHS); }
+
+ /// Unsigned greater than comparison
+ ///
+ /// Regards both *this and RHS as unsigned quantities and compares them for
+ /// the validity of the greater-than relationship.
+ ///
+ /// \returns true if *this > RHS when both are considered unsigned.
+ bool ugt(const APInt &RHS) const { return !ule(RHS); }
+
+ /// Unsigned greater than comparison
+ ///
+ /// Regards both *this as an unsigned quantity and compares it with RHS for
+ /// the validity of the greater-than relationship.
+ ///
+ /// \returns true if *this > RHS when considered unsigned.
+ bool ugt(uint64_t RHS) const {
+ // Only need to check active bits if not a single word.
+ return (!isSingleWord() && getActiveBits() > 64) || getZExtValue() > RHS;
+ }
+
+ /// Signed greater than comparison
+ ///
+ /// Regards both *this and RHS as signed quantities and compares them for the
+ /// validity of the greater-than relationship.
+ ///
+ /// \returns true if *this > RHS when both are considered signed.
+ bool sgt(const APInt &RHS) const { return !sle(RHS); }
+
+ /// Signed greater than comparison
+ ///
+ /// Regards both *this as a signed quantity and compares it with RHS for
+ /// the validity of the greater-than relationship.
+ ///
+ /// \returns true if *this > RHS when considered signed.
+ bool sgt(int64_t RHS) const {
+ return (!isSingleWord() && getMinSignedBits() > 64) ? !isNegative()
+ : getSExtValue() > RHS;
+ }
+
+ /// Unsigned greater or equal comparison
+ ///
+ /// Regards both *this and RHS as unsigned quantities and compares them for
+ /// validity of the greater-or-equal relationship.
+ ///
+ /// \returns true if *this >= RHS when both are considered unsigned.
+ bool uge(const APInt &RHS) const { return !ult(RHS); }
+
+ /// Unsigned greater or equal comparison
+ ///
+ /// Regards both *this as an unsigned quantity and compares it with RHS for
+ /// the validity of the greater-or-equal relationship.
+ ///
+ /// \returns true if *this >= RHS when considered unsigned.
+ bool uge(uint64_t RHS) const { return !ult(RHS); }
+
+ /// Signed greater or equal comparison
+ ///
+ /// Regards both *this and RHS as signed quantities and compares them for
+ /// validity of the greater-or-equal relationship.
+ ///
+ /// \returns true if *this >= RHS when both are considered signed.
+ bool sge(const APInt &RHS) const { return !slt(RHS); }
+
+ /// Signed greater or equal comparison
+ ///
+ /// Regards both *this as a signed quantity and compares it with RHS for
+ /// the validity of the greater-or-equal relationship.
+ ///
+ /// \returns true if *this >= RHS when considered signed.
+ bool sge(int64_t RHS) const { return !slt(RHS); }
+
+ /// This operation tests if there are any pairs of corresponding bits
+ /// between this APInt and RHS that are both set.
+ bool intersects(const APInt &RHS) const {
+ assert(BitWidth == RHS.BitWidth && "Bit widths must be the same");
+ if (isSingleWord())
+ return (U.VAL & RHS.U.VAL) != 0;
+ return intersectsSlowCase(RHS);
+ }
+
+ /// This operation checks that all bits set in this APInt are also set in RHS.
+ bool isSubsetOf(const APInt &RHS) const {
+ assert(BitWidth == RHS.BitWidth && "Bit widths must be the same");
+ if (isSingleWord())
+ return (U.VAL & ~RHS.U.VAL) == 0;
+ return isSubsetOfSlowCase(RHS);
+ }
+
+ /// @}
+ /// \name Resizing Operators
+ /// @{
+
+ /// Truncate to new width.
+ ///
+ /// Truncate the APInt to a specified width. It is an error to specify a width
+ /// that is greater than or equal to the current width.
+ APInt trunc(unsigned width) const;
+
+ /// Truncate to new width with unsigned saturation.
+ ///
+ /// If the APInt, treated as unsigned integer, can be losslessly truncated to
+ /// the new bitwidth, then return truncated APInt. Else, return max value.
+ APInt truncUSat(unsigned width) const;
+
+ /// Truncate to new width with signed saturation.
+ ///
+ /// If this APInt, treated as signed integer, can be losslessly truncated to
+ /// the new bitwidth, then return truncated APInt. Else, return either
+ /// signed min value if the APInt was negative, or signed max value.
+ APInt truncSSat(unsigned width) const;
+
+ /// Sign extend to a new width.
+ ///
+ /// This operation sign extends the APInt to a new width. If the high order
+ /// bit is set, the fill on the left will be done with 1 bits, otherwise zero.
+ /// It is an error to specify a width that is less than or equal to the
+ /// current width.
+ APInt sext(unsigned width) const;
+
+ /// Zero extend to a new width.
+ ///
+ /// This operation zero extends the APInt to a new width. The high order bits
+ /// are filled with 0 bits. It is an error to specify a width that is less
+ /// than or equal to the current width.
+ APInt zext(unsigned width) const;
+
+ /// Sign extend or truncate to width
+ ///
+ /// Make this APInt have the bit width given by \p width. The value is sign
+ /// extended, truncated, or left alone to make it that width.
+ APInt sextOrTrunc(unsigned width) const;
+
+ /// Zero extend or truncate to width
+ ///
+ /// Make this APInt have the bit width given by \p width. The value is zero
+ /// extended, truncated, or left alone to make it that width.
+ APInt zextOrTrunc(unsigned width) const;
+
+ /// Sign extend or truncate to width
+ ///
+ /// Make this APInt have the bit width given by \p width. The value is sign
+ /// extended, or left alone to make it that width.
+ APInt sextOrSelf(unsigned width) const;
+
+ /// Zero extend or truncate to width
+ ///
+ /// Make this APInt have the bit width given by \p width. The value is zero
+ /// extended, or left alone to make it that width.
+ APInt zextOrSelf(unsigned width) const;
+
+ /// @}
+ /// \name Bit Manipulation Operators
+ /// @{
+
+ /// Set every bit to 1.
+ void setAllBits() {
+ if (isSingleWord())
+ U.VAL = WORDTYPE_MAX;
+ else
+ // Set all the bits in all the words.
+ memset(U.pVal, -1, getNumWords() * APINT_WORD_SIZE);
+ // Clear the unused ones
+ clearUnusedBits();
+ }
+
+ /// Set a given bit to 1.
+ ///
+ /// Set the given bit to 1 whose position is given as "bitPosition".
+ void setBit(unsigned BitPosition) {
+ assert(BitPosition < BitWidth && "BitPosition out of range");
+ WordType Mask = maskBit(BitPosition);
+ if (isSingleWord())
+ U.VAL |= Mask;
+ else
+ U.pVal[whichWord(BitPosition)] |= Mask;
+ }
+
+ /// Set the sign bit to 1.
+ void setSignBit() {
+ setBit(BitWidth - 1);
+ }
+
+ /// Set the bits from loBit (inclusive) to hiBit (exclusive) to 1.
+ void setBits(unsigned loBit, unsigned hiBit) {
+ assert(hiBit <= BitWidth && "hiBit out of range");
+ assert(loBit <= BitWidth && "loBit out of range");
+ assert(loBit <= hiBit && "loBit greater than hiBit");
+ if (loBit == hiBit)
+ return;
+ if (loBit < APINT_BITS_PER_WORD && hiBit <= APINT_BITS_PER_WORD) {
+ uint64_t mask = WORDTYPE_MAX >> (APINT_BITS_PER_WORD - (hiBit - loBit));
+ mask <<= loBit;
+ if (isSingleWord())
+ U.VAL |= mask;
+ else
+ U.pVal[0] |= mask;
+ } else {
+ setBitsSlowCase(loBit, hiBit);
+ }
+ }
+
+ /// Set the top bits starting from loBit.
+ void setBitsFrom(unsigned loBit) {
+ return setBits(loBit, BitWidth);
+ }
+
+ /// Set the bottom loBits bits.
+ void setLowBits(unsigned loBits) {
+ return setBits(0, loBits);
+ }
+
+ /// Set the top hiBits bits.
+ void setHighBits(unsigned hiBits) {
+ return setBits(BitWidth - hiBits, BitWidth);
+ }
+
+ /// Set every bit to 0.
+ void clearAllBits() {
+ if (isSingleWord())
+ U.VAL = 0;
+ else
+ memset(U.pVal, 0, getNumWords() * APINT_WORD_SIZE);
+ }
+
+ /// Set a given bit to 0.
+ ///
+ /// Set the given bit to 0 whose position is given as "bitPosition".
+ void clearBit(unsigned BitPosition) {
+ assert(BitPosition < BitWidth && "BitPosition out of range");
+ WordType Mask = ~maskBit(BitPosition);
+ if (isSingleWord())
+ U.VAL &= Mask;
+ else
+ U.pVal[whichWord(BitPosition)] &= Mask;
+ }
+
+ /// Set bottom loBits bits to 0.
+ void clearLowBits(unsigned loBits) {
+ assert(loBits <= BitWidth && "More bits than bitwidth");
+ APInt Keep = getHighBitsSet(BitWidth, BitWidth - loBits);
+ *this &= Keep;
+ }
+
+ /// Set the sign bit to 0.
+ void clearSignBit() {
+ clearBit(BitWidth - 1);
+ }
+
+ /// Toggle every bit to its opposite value.
+ void flipAllBits() {
+ if (isSingleWord()) {
+ U.VAL ^= WORDTYPE_MAX;
+ clearUnusedBits();
+ } else {
+ flipAllBitsSlowCase();
+ }
+ }
+
+ /// Toggles a given bit to its opposite value.
+ ///
+ /// Toggle a given bit to its opposite value whose position is given
+ /// as "bitPosition".
+ void flipBit(unsigned bitPosition);
+
+ /// Negate this APInt in place.
+ void negate() {
+ flipAllBits();
+ ++(*this);
+ }
+
+ /// Insert the bits from a smaller APInt starting at bitPosition.
+ void insertBits(const APInt &SubBits, unsigned bitPosition);
+ void insertBits(uint64_t SubBits, unsigned bitPosition, unsigned numBits);
+
+ /// Return an APInt with the extracted bits [bitPosition,bitPosition+numBits).
+ APInt extractBits(unsigned numBits, unsigned bitPosition) const;
+ uint64_t extractBitsAsZExtValue(unsigned numBits, unsigned bitPosition) const;
+
+ /// @}
+ /// \name Value Characterization Functions
+ /// @{
+
+ /// Return the number of bits in the APInt.
+ unsigned getBitWidth() const { return BitWidth; }
+
+ /// Get the number of words.
+ ///
+ /// Here one word's bitwidth equals to that of uint64_t.
+ ///
+ /// \returns the number of words to hold the integer value of this APInt.
+ unsigned getNumWords() const { return getNumWords(BitWidth); }
+
+ /// Get the number of words.
+ ///
+ /// *NOTE* Here one word's bitwidth equals to that of uint64_t.
+ ///
+ /// \returns the number of words to hold the integer value with a given bit
+ /// width.
+ static unsigned getNumWords(unsigned BitWidth) {
+ return ((uint64_t)BitWidth + APINT_BITS_PER_WORD - 1) / APINT_BITS_PER_WORD;
+ }
+
+ /// Compute the number of active bits in the value
+ ///
+ /// This function returns the number of active bits which is defined as the
+ /// bit width minus the number of leading zeros. This is used in several
+ /// computations to see how "wide" the value is.
+ unsigned getActiveBits() const { return BitWidth - countLeadingZeros(); }
+
+ /// Compute the number of active words in the value of this APInt.
+ ///
+ /// This is used in conjunction with getActiveData to extract the raw value of
+ /// the APInt.
+ unsigned getActiveWords() const {
+ unsigned numActiveBits = getActiveBits();
+ return numActiveBits ? whichWord(numActiveBits - 1) + 1 : 1;
+ }
+
+ /// Get the minimum bit size for this signed APInt
+ ///
+ /// Computes the minimum bit width for this APInt while considering it to be a
+ /// signed (and probably negative) value. If the value is not negative, this
+ /// function returns the same value as getActiveBits()+1. Otherwise, it
+ /// returns the smallest bit width that will retain the negative value. For
+ /// example, -1 can be written as 0b1 or 0xFFFFFFFFFF. 0b1 is shorter and so
+ /// for -1, this function will always return 1.
+ unsigned getMinSignedBits() const {
+ if (isNegative())
+ return BitWidth - countLeadingOnes() + 1;
+ return getActiveBits() + 1;
+ }
+
+ /// Get zero extended value
+ ///
+ /// This method attempts to return the value of this APInt as a zero extended
+ /// uint64_t. The bitwidth must be <= 64 or the value must fit within a
+ /// uint64_t. Otherwise an assertion will result.
+ uint64_t getZExtValue() const {
+ if (isSingleWord())
+ return U.VAL;
+ assert(getActiveBits() <= 64 && "Too many bits for uint64_t");
+ return U.pVal[0];
+ }
+
+ /// Get sign extended value
+ ///
+ /// This method attempts to return the value of this APInt as a sign extended
+ /// int64_t. The bit width must be <= 64 or the value must fit within an
+ /// int64_t. Otherwise an assertion will result.
+ int64_t getSExtValue() const {
+ if (isSingleWord())
+ return SignExtend64(U.VAL, BitWidth);
+ assert(getMinSignedBits() <= 64 && "Too many bits for int64_t");
+ return int64_t(U.pVal[0]);
+ }
+
+ /// Get bits required for string value.
+ ///
+ /// This method determines how many bits are required to hold the APInt
+ /// equivalent of the string given by \p str.
+ static unsigned getBitsNeeded(StringRef str, uint8_t radix);
+
+ /// The APInt version of the countLeadingZeros functions in
+ /// MathExtras.h.
+ ///
+ /// It counts the number of zeros from the most significant bit to the first
+ /// one bit.
+ ///
+ /// \returns BitWidth if the value is zero, otherwise returns the number of
+ /// zeros from the most significant bit to the first one bits.
+ unsigned countLeadingZeros() const {
+ if (isSingleWord()) {
+ unsigned unusedBits = APINT_BITS_PER_WORD - BitWidth;
+ return llvm::countLeadingZeros(U.VAL) - unusedBits;
+ }
+ return countLeadingZerosSlowCase();
+ }
+
+ /// Count the number of leading one bits.
+ ///
+ /// This function is an APInt version of the countLeadingOnes
+ /// functions in MathExtras.h. It counts the number of ones from the most
+ /// significant bit to the first zero bit.
+ ///
+ /// \returns 0 if the high order bit is not set, otherwise returns the number
+ /// of 1 bits from the most significant to the least
+ unsigned countLeadingOnes() const {
+ if (isSingleWord())
+ return llvm::countLeadingOnes(U.VAL << (APINT_BITS_PER_WORD - BitWidth));
+ return countLeadingOnesSlowCase();
+ }
+
+ /// Computes the number of leading bits of this APInt that are equal to its
+ /// sign bit.
+ unsigned getNumSignBits() const {
+ return isNegative() ? countLeadingOnes() : countLeadingZeros();
+ }
+
+ /// Count the number of trailing zero bits.
+ ///
+ /// This function is an APInt version of the countTrailingZeros
+ /// functions in MathExtras.h. It counts the number of zeros from the least
+ /// significant bit to the first set bit.
+ ///
+ /// \returns BitWidth if the value is zero, otherwise returns the number of
+ /// zeros from the least significant bit to the first one bit.
+ unsigned countTrailingZeros() const {
+ if (isSingleWord())
+ return std::min(unsigned(llvm::countTrailingZeros(U.VAL)), BitWidth);
+ return countTrailingZerosSlowCase();
+ }
+
+ /// Count the number of trailing one bits.
+ ///
+ /// This function is an APInt version of the countTrailingOnes
+ /// functions in MathExtras.h. It counts the number of ones from the least
+ /// significant bit to the first zero bit.
+ ///
+ /// \returns BitWidth if the value is all ones, otherwise returns the number
+ /// of ones from the least significant bit to the first zero bit.
+ unsigned countTrailingOnes() const {
+ if (isSingleWord())
+ return llvm::countTrailingOnes(U.VAL);
+ return countTrailingOnesSlowCase();
+ }
+
+ /// Count the number of bits set.
+ ///
+ /// This function is an APInt version of the countPopulation functions
+ /// in MathExtras.h. It counts the number of 1 bits in the APInt value.
+ ///
+ /// \returns 0 if the value is zero, otherwise returns the number of set bits.
+ unsigned countPopulation() const {
+ if (isSingleWord())
+ return llvm::countPopulation(U.VAL);
+ return countPopulationSlowCase();
+ }
+
+ /// @}
+ /// \name Conversion Functions
+ /// @{
+ void print(raw_ostream &OS, bool isSigned) const;
+
+ /// Converts an APInt to a string and append it to Str. Str is commonly a
+ /// SmallString.
+ void toString(SmallVectorImpl<char> &Str, unsigned Radix, bool Signed,
+ bool formatAsCLiteral = false) const;
+
+ /// Considers the APInt to be unsigned and converts it into a string in the
+ /// radix given. The radix can be 2, 8, 10 16, or 36.
+ void toStringUnsigned(SmallVectorImpl<char> &Str, unsigned Radix = 10) const {
+ toString(Str, Radix, false, false);
+ }
+
+ /// Considers the APInt to be signed and converts it into a string in the
+ /// radix given. The radix can be 2, 8, 10, 16, or 36.
+ void toStringSigned(SmallVectorImpl<char> &Str, unsigned Radix = 10) const {
+ toString(Str, Radix, true, false);
+ }
+
+ /// Return the APInt as a std::string.
+ ///
+ /// Note that this is an inefficient method. It is better to pass in a
+ /// SmallVector/SmallString to the methods above to avoid thrashing the heap
+ /// for the string.
+ std::string toString(unsigned Radix, bool Signed) const;
+
+ /// \returns a byte-swapped representation of this APInt Value.
+ APInt byteSwap() const;
+
+ /// \returns the value with the bit representation reversed of this APInt
+ /// Value.
+ APInt reverseBits() const;
+
+ /// Converts this APInt to a double value.
+ double roundToDouble(bool isSigned) const;
+
+ /// Converts this unsigned APInt to a double value.
+ double roundToDouble() const { return roundToDouble(false); }
+
+ /// Converts this signed APInt to a double value.
+ double signedRoundToDouble() const { return roundToDouble(true); }
+
+ /// Converts APInt bits to a double
+ ///
+ /// The conversion does not do a translation from integer to double, it just
+ /// re-interprets the bits as a double. Note that it is valid to do this on
+ /// any bit width. Exactly 64 bits will be translated.
+ double bitsToDouble() const {
+ return BitsToDouble(getWord(0));
+ }
+
+ /// Converts APInt bits to a float
+ ///
+ /// The conversion does not do a translation from integer to float, it just
+ /// re-interprets the bits as a float. Note that it is valid to do this on
+ /// any bit width. Exactly 32 bits will be translated.
+ float bitsToFloat() const {
+ return BitsToFloat(static_cast<uint32_t>(getWord(0)));
+ }
+
+ /// Converts a double to APInt bits.
+ ///
+ /// The conversion does not do a translation from double to integer, it just
+ /// re-interprets the bits of the double.
+ static APInt doubleToBits(double V) {
+ return APInt(sizeof(double) * CHAR_BIT, DoubleToBits(V));
+ }
+
+ /// Converts a float to APInt bits.
+ ///
+ /// The conversion does not do a translation from float to integer, it just
+ /// re-interprets the bits of the float.
+ static APInt floatToBits(float V) {
+ return APInt(sizeof(float) * CHAR_BIT, FloatToBits(V));
+ }
+
+ /// @}
+ /// \name Mathematics Operations
+ /// @{
+
+ /// \returns the floor log base 2 of this APInt.
+ unsigned logBase2() const { return getActiveBits() - 1; }
+
+ /// \returns the ceil log base 2 of this APInt.
+ unsigned ceilLogBase2() const {
+ APInt temp(*this);
+ --temp;
+ return temp.getActiveBits();
+ }
+
+ /// \returns the nearest log base 2 of this APInt. Ties round up.
+ ///
+ /// NOTE: When we have a BitWidth of 1, we define:
+ ///
+ /// log2(0) = UINT32_MAX
+ /// log2(1) = 0
+ ///
+ /// to get around any mathematical concerns resulting from
+ /// referencing 2 in a space where 2 does no exist.
+ unsigned nearestLogBase2() const {
+ // Special case when we have a bitwidth of 1. If VAL is 1, then we
+ // get 0. If VAL is 0, we get WORDTYPE_MAX which gets truncated to
+ // UINT32_MAX.
+ if (BitWidth == 1)
+ return U.VAL - 1;
+
+ // Handle the zero case.
+ if (isNullValue())
+ return UINT32_MAX;
+
+ // The non-zero case is handled by computing:
+ //
+ // nearestLogBase2(x) = logBase2(x) + x[logBase2(x)-1].
+ //
+ // where x[i] is referring to the value of the ith bit of x.
+ unsigned lg = logBase2();
+ return lg + unsigned((*this)[lg - 1]);
+ }
+
+ /// \returns the log base 2 of this APInt if its an exact power of two, -1
+ /// otherwise
+ int32_t exactLogBase2() const {
+ if (!isPowerOf2())
+ return -1;
+ return logBase2();
+ }
+
+ /// Compute the square root
+ APInt sqrt() const;
+
+ /// Get the absolute value;
+ ///
+ /// If *this is < 0 then return -(*this), otherwise *this;
+ APInt abs() const {
+ if (isNegative())
+ return -(*this);
+ return *this;
+ }
+
+ /// \returns the multiplicative inverse for a given modulo.
+ APInt multiplicativeInverse(const APInt &modulo) const;
+
+ /// @}
+ /// \name Support for division by constant
+ /// @{
+
+ /// Calculate the magic number for signed division by a constant.
+ struct ms;
+ ms magic() const;
+
+ /// Calculate the magic number for unsigned division by a constant.
+ struct mu;
+ mu magicu(unsigned LeadingZeros = 0) const;
+
+ /// @}
+ /// \name Building-block Operations for APInt and APFloat
+ /// @{
+
+ // These building block operations operate on a representation of arbitrary
+ // precision, two's-complement, bignum integer values. They should be
+ // sufficient to implement APInt and APFloat bignum requirements. Inputs are
+ // generally a pointer to the base of an array of integer parts, representing
+ // an unsigned bignum, and a count of how many parts there are.
+
+ /// Sets the least significant part of a bignum to the input value, and zeroes
+ /// out higher parts.
+ static void tcSet(WordType *, WordType, unsigned);
+
+ /// Assign one bignum to another.
+ static void tcAssign(WordType *, const WordType *, unsigned);
+
+ /// Returns true if a bignum is zero, false otherwise.
+ static bool tcIsZero(const WordType *, unsigned);
+
+ /// Extract the given bit of a bignum; returns 0 or 1. Zero-based.
+ static int tcExtractBit(const WordType *, unsigned bit);
+
+ /// Copy the bit vector of width srcBITS from SRC, starting at bit srcLSB, to
+ /// DST, of dstCOUNT parts, such that the bit srcLSB becomes the least
+ /// significant bit of DST. All high bits above srcBITS in DST are
+ /// zero-filled.
+ static void tcExtract(WordType *, unsigned dstCount,
+ const WordType *, unsigned srcBits,
+ unsigned srcLSB);
+
+ /// Set the given bit of a bignum. Zero-based.
+ static void tcSetBit(WordType *, unsigned bit);
+
+ /// Clear the given bit of a bignum. Zero-based.
+ static void tcClearBit(WordType *, unsigned bit);
+
+ /// Returns the bit number of the least or most significant set bit of a
+ /// number. If the input number has no bits set -1U is returned.
+ static unsigned tcLSB(const WordType *, unsigned n);
+ static unsigned tcMSB(const WordType *parts, unsigned n);
+
+ /// Negate a bignum in-place.
+ static void tcNegate(WordType *, unsigned);
+
+ /// DST += RHS + CARRY where CARRY is zero or one. Returns the carry flag.
+ static WordType tcAdd(WordType *, const WordType *,
+ WordType carry, unsigned);
+ /// DST += RHS. Returns the carry flag.
+ static WordType tcAddPart(WordType *, WordType, unsigned);
+
+ /// DST -= RHS + CARRY where CARRY is zero or one. Returns the carry flag.
+ static WordType tcSubtract(WordType *, const WordType *,
+ WordType carry, unsigned);
+ /// DST -= RHS. Returns the carry flag.
+ static WordType tcSubtractPart(WordType *, WordType, unsigned);
+
+ /// DST += SRC * MULTIPLIER + PART if add is true
+ /// DST = SRC * MULTIPLIER + PART if add is false
+ ///
+ /// Requires 0 <= DSTPARTS <= SRCPARTS + 1. If DST overlaps SRC they must
+ /// start at the same point, i.e. DST == SRC.
+ ///
+ /// If DSTPARTS == SRC_PARTS + 1 no overflow occurs and zero is returned.
+ /// Otherwise DST is filled with the least significant DSTPARTS parts of the
+ /// result, and if all of the omitted higher parts were zero return zero,
+ /// otherwise overflow occurred and return one.
+ static int tcMultiplyPart(WordType *dst, const WordType *src,
+ WordType multiplier, WordType carry,
+ unsigned srcParts, unsigned dstParts,
+ bool add);
+
+ /// DST = LHS * RHS, where DST has the same width as the operands and is
+ /// filled with the least significant parts of the result. Returns one if
+ /// overflow occurred, otherwise zero. DST must be disjoint from both
+ /// operands.
+ static int tcMultiply(WordType *, const WordType *, const WordType *,
+ unsigned);
+
+ /// DST = LHS * RHS, where DST has width the sum of the widths of the
+ /// operands. No overflow occurs. DST must be disjoint from both operands.
+ static void tcFullMultiply(WordType *, const WordType *,
+ const WordType *, unsigned, unsigned);
+
+ /// If RHS is zero LHS and REMAINDER are left unchanged, return one.
+ /// Otherwise set LHS to LHS / RHS with the fractional part discarded, set
+ /// REMAINDER to the remainder, return zero. i.e.
+ ///
+ /// OLD_LHS = RHS * LHS + REMAINDER
+ ///
+ /// SCRATCH is a bignum of the same size as the operands and result for use by
+ /// the routine; its contents need not be initialized and are destroyed. LHS,
+ /// REMAINDER and SCRATCH must be distinct.
+ static int tcDivide(WordType *lhs, const WordType *rhs,
+ WordType *remainder, WordType *scratch,
+ unsigned parts);
+
+ /// Shift a bignum left Count bits. Shifted in bits are zero. There are no
+ /// restrictions on Count.
+ static void tcShiftLeft(WordType *, unsigned Words, unsigned Count);
+
+ /// Shift a bignum right Count bits. Shifted in bits are zero. There are no
+ /// restrictions on Count.
+ static void tcShiftRight(WordType *, unsigned Words, unsigned Count);
+
+ /// The obvious AND, OR and XOR and complement operations.
+ static void tcAnd(WordType *, const WordType *, unsigned);
+ static void tcOr(WordType *, const WordType *, unsigned);
+ static void tcXor(WordType *, const WordType *, unsigned);
+ static void tcComplement(WordType *, unsigned);
+
+ /// Comparison (unsigned) of two bignums.
+ static int tcCompare(const WordType *, const WordType *, unsigned);
+
+ /// Increment a bignum in-place. Return the carry flag.
+ static WordType tcIncrement(WordType *dst, unsigned parts) {
+ return tcAddPart(dst, 1, parts);
+ }
+
+ /// Decrement a bignum in-place. Return the borrow flag.
+ static WordType tcDecrement(WordType *dst, unsigned parts) {
+ return tcSubtractPart(dst, 1, parts);
+ }
+
+ /// Set the least significant BITS and clear the rest.
+ static void tcSetLeastSignificantBits(WordType *, unsigned, unsigned bits);
+
+ /// debug method
+ void dump() const;
+
+ /// @}
+};
+
+/// Magic data for optimising signed division by a constant.
+struct APInt::ms {
+ APInt m; ///< magic number
+ unsigned s; ///< shift amount
+};
+
+/// Magic data for optimising unsigned division by a constant.
+struct APInt::mu {
+ APInt m; ///< magic number
+ bool a; ///< add indicator
+ unsigned s; ///< shift amount
+};
+
+inline bool operator==(uint64_t V1, const APInt &V2) { return V2 == V1; }
+
+inline bool operator!=(uint64_t V1, const APInt &V2) { return V2 != V1; }
+
+/// Unary bitwise complement operator.
+///
+/// \returns an APInt that is the bitwise complement of \p v.
+inline APInt operator~(APInt v) {
+ v.flipAllBits();
+ return v;
+}
+
+inline APInt operator&(APInt a, const APInt &b) {
+ a &= b;
+ return a;
+}
+
+inline APInt operator&(const APInt &a, APInt &&b) {
+ b &= a;
+ return std::move(b);
+}
+
+inline APInt operator&(APInt a, uint64_t RHS) {
+ a &= RHS;
+ return a;
+}
+
+inline APInt operator&(uint64_t LHS, APInt b) {
+ b &= LHS;
+ return b;
+}
+
+inline APInt operator|(APInt a, const APInt &b) {
+ a |= b;
+ return a;
+}
+
+inline APInt operator|(const APInt &a, APInt &&b) {
+ b |= a;
+ return std::move(b);
+}
+
+inline APInt operator|(APInt a, uint64_t RHS) {
+ a |= RHS;
+ return a;
+}
+
+inline APInt operator|(uint64_t LHS, APInt b) {
+ b |= LHS;
+ return b;
+}
+
+inline APInt operator^(APInt a, const APInt &b) {
+ a ^= b;
+ return a;
+}
+
+inline APInt operator^(const APInt &a, APInt &&b) {
+ b ^= a;
+ return std::move(b);
+}
+
+inline APInt operator^(APInt a, uint64_t RHS) {
+ a ^= RHS;
+ return a;
+}
+
+inline APInt operator^(uint64_t LHS, APInt b) {
+ b ^= LHS;
+ return b;
+}
+
+inline raw_ostream &operator<<(raw_ostream &OS, const APInt &I) {
+ I.print(OS, true);
+ return OS;
+}
+
+inline APInt operator-(APInt v) {
+ v.negate();
+ return v;
+}
+
+inline APInt operator+(APInt a, const APInt &b) {
+ a += b;
+ return a;
+}
+
+inline APInt operator+(const APInt &a, APInt &&b) {
+ b += a;
+ return std::move(b);
+}
+
+inline APInt operator+(APInt a, uint64_t RHS) {
+ a += RHS;
+ return a;
+}
+
+inline APInt operator+(uint64_t LHS, APInt b) {
+ b += LHS;
+ return b;
+}
+
+inline APInt operator-(APInt a, const APInt &b) {
+ a -= b;
+ return a;
+}
+
+inline APInt operator-(const APInt &a, APInt &&b) {
+ b.negate();
+ b += a;
+ return std::move(b);
+}
+
+inline APInt operator-(APInt a, uint64_t RHS) {
+ a -= RHS;
+ return a;
+}
+
+inline APInt operator-(uint64_t LHS, APInt b) {
+ b.negate();
+ b += LHS;
+ return b;
+}
+
+inline APInt operator*(APInt a, uint64_t RHS) {
+ a *= RHS;
+ return a;
+}
+
+inline APInt operator*(uint64_t LHS, APInt b) {
+ b *= LHS;
+ return b;
+}
+
+
+namespace APIntOps {
+
+/// Determine the smaller of two APInts considered to be signed.
+inline const APInt &smin(const APInt &A, const APInt &B) {
+ return A.slt(B) ? A : B;
+}
+
+/// Determine the larger of two APInts considered to be signed.
+inline const APInt &smax(const APInt &A, const APInt &B) {
+ return A.sgt(B) ? A : B;
+}
+
+/// Determine the smaller of two APInts considered to be signed.
+inline const APInt &umin(const APInt &A, const APInt &B) {
+ return A.ult(B) ? A : B;
+}
+
+/// Determine the larger of two APInts considered to be unsigned.
+inline const APInt &umax(const APInt &A, const APInt &B) {
+ return A.ugt(B) ? A : B;
+}
+
+/// Compute GCD of two unsigned APInt values.
+///
+/// This function returns the greatest common divisor of the two APInt values
+/// using Stein's algorithm.
+///
+/// \returns the greatest common divisor of A and B.
+APInt GreatestCommonDivisor(APInt A, APInt B);
+
+/// Converts the given APInt to a double value.
+///
+/// Treats the APInt as an unsigned value for conversion purposes.
+inline double RoundAPIntToDouble(const APInt &APIVal) {
+ return APIVal.roundToDouble();
+}
+
+/// Converts the given APInt to a double value.
+///
+/// Treats the APInt as a signed value for conversion purposes.
+inline double RoundSignedAPIntToDouble(const APInt &APIVal) {
+ return APIVal.signedRoundToDouble();
+}
+
+/// Converts the given APInt to a float vlalue.
+inline float RoundAPIntToFloat(const APInt &APIVal) {
+ return float(RoundAPIntToDouble(APIVal));
+}
+
+/// Converts the given APInt to a float value.
+///
+/// Treats the APInt as a signed value for conversion purposes.
+inline float RoundSignedAPIntToFloat(const APInt &APIVal) {
+ return float(APIVal.signedRoundToDouble());
+}
+
+/// Converts the given double value into a APInt.
+///
+/// This function convert a double value to an APInt value.
+APInt RoundDoubleToAPInt(double Double, unsigned width);
+
+/// Converts a float value into a APInt.
+///
+/// Converts a float value into an APInt value.
+inline APInt RoundFloatToAPInt(float Float, unsigned width) {
+ return RoundDoubleToAPInt(double(Float), width);
+}
+
+/// Return A unsign-divided by B, rounded by the given rounding mode.
+APInt RoundingUDiv(const APInt &A, const APInt &B, APInt::Rounding RM);
+
+/// Return A sign-divided by B, rounded by the given rounding mode.
+APInt RoundingSDiv(const APInt &A, const APInt &B, APInt::Rounding RM);
+
+/// Let q(n) = An^2 + Bn + C, and BW = bit width of the value range
+/// (e.g. 32 for i32).
+/// This function finds the smallest number n, such that
+/// (a) n >= 0 and q(n) = 0, or
+/// (b) n >= 1 and q(n-1) and q(n), when evaluated in the set of all
+/// integers, belong to two different intervals [Rk, Rk+R),
+/// where R = 2^BW, and k is an integer.
+/// The idea here is to find when q(n) "overflows" 2^BW, while at the
+/// same time "allowing" subtraction. In unsigned modulo arithmetic a
+/// subtraction (treated as addition of negated numbers) would always
+/// count as an overflow, but here we want to allow values to decrease
+/// and increase as long as they are within the same interval.
+/// Specifically, adding of two negative numbers should not cause an
+/// overflow (as long as the magnitude does not exceed the bit width).
+/// On the other hand, given a positive number, adding a negative
+/// number to it can give a negative result, which would cause the
+/// value to go from [-2^BW, 0) to [0, 2^BW). In that sense, zero is
+/// treated as a special case of an overflow.
+///
+/// This function returns None if after finding k that minimizes the
+/// positive solution to q(n) = kR, both solutions are contained between
+/// two consecutive integers.
+///
+/// There are cases where q(n) > T, and q(n+1) < T (assuming evaluation
+/// in arithmetic modulo 2^BW, and treating the values as signed) by the
+/// virtue of *signed* overflow. This function will *not* find such an n,
+/// however it may find a value of n satisfying the inequalities due to
+/// an *unsigned* overflow (if the values are treated as unsigned).
+/// To find a solution for a signed overflow, treat it as a problem of
+/// finding an unsigned overflow with a range with of BW-1.
+///
+/// The returned value may have a different bit width from the input
+/// coefficients.
+Optional<APInt> SolveQuadraticEquationWrap(APInt A, APInt B, APInt C,
+ unsigned RangeWidth);
+
+/// Compare two values, and if they are different, return the position of the
+/// most significant bit that is different in the values.
+Optional<unsigned> GetMostSignificantDifferentBit(const APInt &A,
+ const APInt &B);
+
+} // End of APIntOps namespace
+
+// See friend declaration above. This additional declaration is required in
+// order to compile LLVM with IBM xlC compiler.
+hash_code hash_value(const APInt &Arg);
+
+/// StoreIntToMemory - Fills the StoreBytes bytes of memory starting from Dst
+/// with the integer held in IntVal.
+void StoreIntToMemory(const APInt &IntVal, uint8_t *Dst, unsigned StoreBytes);
+
+/// LoadIntFromMemory - Loads the integer stored in the LoadBytes bytes starting
+/// from Src into IntVal, which is assumed to be wide enough and to hold zero.
+void LoadIntFromMemory(APInt &IntVal, uint8_t *Src, unsigned LoadBytes);
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/APSInt.h b/third_party/llvm-project/include/llvm/ADT/APSInt.h
new file mode 100644
index 000000000..0f991826c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/APSInt.h
@@ -0,0 +1,353 @@
+//===-- llvm/ADT/APSInt.h - Arbitrary Precision Signed Int -----*- C++ -*--===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the APSInt class, which is a simple class that
+// represents an arbitrary sized integer that knows its signedness.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_APSINT_H
+#define LLVM_ADT_APSINT_H
+
+#include "llvm/ADT/APInt.h"
+
+namespace llvm {
+
+class LLVM_NODISCARD APSInt : public APInt {
+ bool IsUnsigned;
+
+public:
+ /// Default constructor that creates an uninitialized APInt.
+ explicit APSInt() : IsUnsigned(false) {}
+
+ /// APSInt ctor - Create an APSInt with the specified width, default to
+ /// unsigned.
+ explicit APSInt(uint32_t BitWidth, bool isUnsigned = true)
+ : APInt(BitWidth, 0), IsUnsigned(isUnsigned) {}
+
+ explicit APSInt(APInt I, bool isUnsigned = true)
+ : APInt(std::move(I)), IsUnsigned(isUnsigned) {}
+
+ /// Construct an APSInt from a string representation.
+ ///
+ /// This constructor interprets the string \p Str using the radix of 10.
+ /// The interpretation stops at the end of the string. The bit width of the
+ /// constructed APSInt is determined automatically.
+ ///
+ /// \param Str the string to be interpreted.
+ explicit APSInt(StringRef Str);
+
+ /// Determine sign of this APSInt.
+ ///
+ /// \returns true if this APSInt is negative, false otherwise
+ bool isNegative() const { return isSigned() && APInt::isNegative(); }
+
+ /// Determine if this APSInt Value is non-negative (>= 0)
+ ///
+ /// \returns true if this APSInt is non-negative, false otherwise
+ bool isNonNegative() const { return !isNegative(); }
+
+ /// Determine if this APSInt Value is positive.
+ ///
+ /// This tests if the value of this APSInt is positive (> 0). Note
+ /// that 0 is not a positive value.
+ ///
+ /// \returns true if this APSInt is positive.
+ bool isStrictlyPositive() const { return isNonNegative() && !isNullValue(); }
+
+ APSInt &operator=(APInt RHS) {
+ // Retain our current sign.
+ APInt::operator=(std::move(RHS));
+ return *this;
+ }
+
+ APSInt &operator=(uint64_t RHS) {
+ // Retain our current sign.
+ APInt::operator=(RHS);
+ return *this;
+ }
+
+ // Query sign information.
+ bool isSigned() const { return !IsUnsigned; }
+ bool isUnsigned() const { return IsUnsigned; }
+ void setIsUnsigned(bool Val) { IsUnsigned = Val; }
+ void setIsSigned(bool Val) { IsUnsigned = !Val; }
+
+ /// toString - Append this APSInt to the specified SmallString.
+ void toString(SmallVectorImpl<char> &Str, unsigned Radix = 10) const {
+ APInt::toString(Str, Radix, isSigned());
+ }
+ /// toString - Converts an APInt to a std::string. This is an inefficient
+ /// method; you should prefer passing in a SmallString instead.
+ std::string toString(unsigned Radix) const {
+ return APInt::toString(Radix, isSigned());
+ }
+ using APInt::toString;
+
+ /// Get the correctly-extended \c int64_t value.
+ int64_t getExtValue() const {
+ assert(getMinSignedBits() <= 64 && "Too many bits for int64_t");
+ return isSigned() ? getSExtValue() : getZExtValue();
+ }
+
+ APSInt trunc(uint32_t width) const {
+ return APSInt(APInt::trunc(width), IsUnsigned);
+ }
+
+ APSInt extend(uint32_t width) const {
+ if (IsUnsigned)
+ return APSInt(zext(width), IsUnsigned);
+ else
+ return APSInt(sext(width), IsUnsigned);
+ }
+
+ APSInt extOrTrunc(uint32_t width) const {
+ if (IsUnsigned)
+ return APSInt(zextOrTrunc(width), IsUnsigned);
+ else
+ return APSInt(sextOrTrunc(width), IsUnsigned);
+ }
+
+ const APSInt &operator%=(const APSInt &RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ if (IsUnsigned)
+ *this = urem(RHS);
+ else
+ *this = srem(RHS);
+ return *this;
+ }
+ const APSInt &operator/=(const APSInt &RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ if (IsUnsigned)
+ *this = udiv(RHS);
+ else
+ *this = sdiv(RHS);
+ return *this;
+ }
+ APSInt operator%(const APSInt &RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? APSInt(urem(RHS), true) : APSInt(srem(RHS), false);
+ }
+ APSInt operator/(const APSInt &RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? APSInt(udiv(RHS), true) : APSInt(sdiv(RHS), false);
+ }
+
+ APSInt operator>>(unsigned Amt) const {
+ return IsUnsigned ? APSInt(lshr(Amt), true) : APSInt(ashr(Amt), false);
+ }
+ APSInt& operator>>=(unsigned Amt) {
+ if (IsUnsigned)
+ lshrInPlace(Amt);
+ else
+ ashrInPlace(Amt);
+ return *this;
+ }
+
+ inline bool operator<(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? ult(RHS) : slt(RHS);
+ }
+ inline bool operator>(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? ugt(RHS) : sgt(RHS);
+ }
+ inline bool operator<=(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? ule(RHS) : sle(RHS);
+ }
+ inline bool operator>=(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return IsUnsigned ? uge(RHS) : sge(RHS);
+ }
+ inline bool operator==(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return eq(RHS);
+ }
+ inline bool operator!=(const APSInt& RHS) const {
+ return !((*this) == RHS);
+ }
+
+ bool operator==(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) == 0;
+ }
+ bool operator!=(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) != 0;
+ }
+ bool operator<=(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) <= 0;
+ }
+ bool operator>=(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) >= 0;
+ }
+ bool operator<(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) < 0;
+ }
+ bool operator>(int64_t RHS) const {
+ return compareValues(*this, get(RHS)) > 0;
+ }
+
+ // The remaining operators just wrap the logic of APInt, but retain the
+ // signedness information.
+
+ APSInt operator<<(unsigned Bits) const {
+ return APSInt(static_cast<const APInt&>(*this) << Bits, IsUnsigned);
+ }
+ APSInt& operator<<=(unsigned Amt) {
+ static_cast<APInt&>(*this) <<= Amt;
+ return *this;
+ }
+
+ APSInt& operator++() {
+ ++(static_cast<APInt&>(*this));
+ return *this;
+ }
+ APSInt& operator--() {
+ --(static_cast<APInt&>(*this));
+ return *this;
+ }
+ APSInt operator++(int) {
+ return APSInt(++static_cast<APInt&>(*this), IsUnsigned);
+ }
+ APSInt operator--(int) {
+ return APSInt(--static_cast<APInt&>(*this), IsUnsigned);
+ }
+ APSInt operator-() const {
+ return APSInt(-static_cast<const APInt&>(*this), IsUnsigned);
+ }
+ APSInt& operator+=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) += RHS;
+ return *this;
+ }
+ APSInt& operator-=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) -= RHS;
+ return *this;
+ }
+ APSInt& operator*=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) *= RHS;
+ return *this;
+ }
+ APSInt& operator&=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) &= RHS;
+ return *this;
+ }
+ APSInt& operator|=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) |= RHS;
+ return *this;
+ }
+ APSInt& operator^=(const APSInt& RHS) {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ static_cast<APInt&>(*this) ^= RHS;
+ return *this;
+ }
+
+ APSInt operator&(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) & RHS, IsUnsigned);
+ }
+
+ APSInt operator|(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) | RHS, IsUnsigned);
+ }
+
+ APSInt operator^(const APSInt &RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) ^ RHS, IsUnsigned);
+ }
+
+ APSInt operator*(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) * RHS, IsUnsigned);
+ }
+ APSInt operator+(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) + RHS, IsUnsigned);
+ }
+ APSInt operator-(const APSInt& RHS) const {
+ assert(IsUnsigned == RHS.IsUnsigned && "Signedness mismatch!");
+ return APSInt(static_cast<const APInt&>(*this) - RHS, IsUnsigned);
+ }
+ APSInt operator~() const {
+ return APSInt(~static_cast<const APInt&>(*this), IsUnsigned);
+ }
+
+ /// getMaxValue - Return the APSInt representing the maximum integer value
+ /// with the given bit width and signedness.
+ static APSInt getMaxValue(uint32_t numBits, bool Unsigned) {
+ return APSInt(Unsigned ? APInt::getMaxValue(numBits)
+ : APInt::getSignedMaxValue(numBits), Unsigned);
+ }
+
+ /// getMinValue - Return the APSInt representing the minimum integer value
+ /// with the given bit width and signedness.
+ static APSInt getMinValue(uint32_t numBits, bool Unsigned) {
+ return APSInt(Unsigned ? APInt::getMinValue(numBits)
+ : APInt::getSignedMinValue(numBits), Unsigned);
+ }
+
+ /// Determine if two APSInts have the same value, zero- or
+ /// sign-extending as needed.
+ static bool isSameValue(const APSInt &I1, const APSInt &I2) {
+ return !compareValues(I1, I2);
+ }
+
+ /// Compare underlying values of two numbers.
+ static int compareValues(const APSInt &I1, const APSInt &I2) {
+ if (I1.getBitWidth() == I2.getBitWidth() && I1.isSigned() == I2.isSigned())
+ return I1.IsUnsigned ? I1.compare(I2) : I1.compareSigned(I2);
+
+ // Check for a bit-width mismatch.
+ if (I1.getBitWidth() > I2.getBitWidth())
+ return compareValues(I1, I2.extend(I1.getBitWidth()));
+ if (I2.getBitWidth() > I1.getBitWidth())
+ return compareValues(I1.extend(I2.getBitWidth()), I2);
+
+ // We have a signedness mismatch. Check for negative values and do an
+ // unsigned compare if both are positive.
+ if (I1.isSigned()) {
+ assert(!I2.isSigned() && "Expected signed mismatch");
+ if (I1.isNegative())
+ return -1;
+ } else {
+ assert(I2.isSigned() && "Expected signed mismatch");
+ if (I2.isNegative())
+ return 1;
+ }
+
+ return I1.compare(I2);
+ }
+
+ static APSInt get(int64_t X) { return APSInt(APInt(64, X), false); }
+ static APSInt getUnsigned(uint64_t X) { return APSInt(APInt(64, X), true); }
+
+ /// Profile - Used to insert APSInt objects, or objects that contain APSInt
+ /// objects, into FoldingSets.
+ void Profile(FoldingSetNodeID& ID) const;
+};
+
+inline bool operator==(int64_t V1, const APSInt &V2) { return V2 == V1; }
+inline bool operator!=(int64_t V1, const APSInt &V2) { return V2 != V1; }
+inline bool operator<=(int64_t V1, const APSInt &V2) { return V2 >= V1; }
+inline bool operator>=(int64_t V1, const APSInt &V2) { return V2 <= V1; }
+inline bool operator<(int64_t V1, const APSInt &V2) { return V2 > V1; }
+inline bool operator>(int64_t V1, const APSInt &V2) { return V2 < V1; }
+
+inline raw_ostream &operator<<(raw_ostream &OS, const APSInt &I) {
+ I.print(OS, I.isSigned());
+ return OS;
+}
+
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/AllocatorList.h b/third_party/llvm-project/include/llvm/ADT/AllocatorList.h
new file mode 100644
index 000000000..405a2e426
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/AllocatorList.h
@@ -0,0 +1,240 @@
+//===- llvm/ADT/AllocatorList.h - Custom allocator list ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ALLOCATORLIST_H
+#define LLVM_ADT_ALLOCATORLIST_H
+
+#include "llvm/ADT/ilist_node.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/simple_ilist.h"
+#include "llvm/Support/Allocator.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <iterator>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// A linked-list with a custom, local allocator.
+///
+/// Expose a std::list-like interface that owns and uses a custom LLVM-style
+/// allocator (e.g., BumpPtrAllocator), leveraging \a simple_ilist for the
+/// implementation details.
+///
+/// Because this list owns the allocator, calling \a splice() with a different
+/// list isn't generally safe. As such, \a splice has been left out of the
+/// interface entirely.
+template <class T, class AllocatorT> class AllocatorList : AllocatorT {
+ struct Node : ilist_node<Node> {
+ Node(Node &&) = delete;
+ Node(const Node &) = delete;
+ Node &operator=(Node &&) = delete;
+ Node &operator=(const Node &) = delete;
+
+ Node(T &&V) : V(std::move(V)) {}
+ Node(const T &V) : V(V) {}
+ template <class... Ts> Node(Ts &&... Vs) : V(std::forward<Ts>(Vs)...) {}
+ T V;
+ };
+
+ using list_type = simple_ilist<Node>;
+
+ list_type List;
+
+ AllocatorT &getAlloc() { return *this; }
+ const AllocatorT &getAlloc() const { return *this; }
+
+ template <class... ArgTs> Node *create(ArgTs &&... Args) {
+ return new (getAlloc()) Node(std::forward<ArgTs>(Args)...);
+ }
+
+ struct Cloner {
+ AllocatorList &AL;
+
+ Cloner(AllocatorList &AL) : AL(AL) {}
+
+ Node *operator()(const Node &N) const { return AL.create(N.V); }
+ };
+
+ struct Disposer {
+ AllocatorList &AL;
+
+ Disposer(AllocatorList &AL) : AL(AL) {}
+
+ void operator()(Node *N) const {
+ N->~Node();
+ AL.getAlloc().Deallocate(N);
+ }
+ };
+
+public:
+ using value_type = T;
+ using pointer = T *;
+ using reference = T &;
+ using const_pointer = const T *;
+ using const_reference = const T &;
+ using size_type = typename list_type::size_type;
+ using difference_type = typename list_type::difference_type;
+
+private:
+ template <class ValueT, class IteratorBase>
+ class IteratorImpl
+ : public iterator_adaptor_base<IteratorImpl<ValueT, IteratorBase>,
+ IteratorBase,
+ std::bidirectional_iterator_tag, ValueT> {
+ template <class OtherValueT, class OtherIteratorBase>
+ friend class IteratorImpl;
+ friend AllocatorList;
+
+ using base_type =
+ iterator_adaptor_base<IteratorImpl<ValueT, IteratorBase>, IteratorBase,
+ std::bidirectional_iterator_tag, ValueT>;
+
+ public:
+ using value_type = ValueT;
+ using pointer = ValueT *;
+ using reference = ValueT &;
+
+ IteratorImpl() = default;
+ IteratorImpl(const IteratorImpl &) = default;
+ IteratorImpl &operator=(const IteratorImpl &) = default;
+
+ explicit IteratorImpl(const IteratorBase &I) : base_type(I) {}
+
+ template <class OtherValueT, class OtherIteratorBase>
+ IteratorImpl(const IteratorImpl<OtherValueT, OtherIteratorBase> &X,
+ typename std::enable_if<std::is_convertible<
+ OtherIteratorBase, IteratorBase>::value>::type * = nullptr)
+ : base_type(X.wrapped()) {}
+
+ ~IteratorImpl() = default;
+
+ reference operator*() const { return base_type::wrapped()->V; }
+ pointer operator->() const { return &operator*(); }
+
+ friend bool operator==(const IteratorImpl &L, const IteratorImpl &R) {
+ return L.wrapped() == R.wrapped();
+ }
+ friend bool operator!=(const IteratorImpl &L, const IteratorImpl &R) {
+ return !(L == R);
+ }
+ };
+
+public:
+ using iterator = IteratorImpl<T, typename list_type::iterator>;
+ using reverse_iterator =
+ IteratorImpl<T, typename list_type::reverse_iterator>;
+ using const_iterator =
+ IteratorImpl<const T, typename list_type::const_iterator>;
+ using const_reverse_iterator =
+ IteratorImpl<const T, typename list_type::const_reverse_iterator>;
+
+ AllocatorList() = default;
+ AllocatorList(AllocatorList &&X)
+ : AllocatorT(std::move(X.getAlloc())), List(std::move(X.List)) {}
+
+ AllocatorList(const AllocatorList &X) {
+ List.cloneFrom(X.List, Cloner(*this), Disposer(*this));
+ }
+
+ AllocatorList &operator=(AllocatorList &&X) {
+ clear(); // Dispose of current nodes explicitly.
+ List = std::move(X.List);
+ getAlloc() = std::move(X.getAlloc());
+ return *this;
+ }
+
+ AllocatorList &operator=(const AllocatorList &X) {
+ List.cloneFrom(X.List, Cloner(*this), Disposer(*this));
+ return *this;
+ }
+
+ ~AllocatorList() { clear(); }
+
+ void swap(AllocatorList &RHS) {
+ List.swap(RHS.List);
+ std::swap(getAlloc(), RHS.getAlloc());
+ }
+
+ bool empty() { return List.empty(); }
+ size_t size() { return List.size(); }
+
+ iterator begin() { return iterator(List.begin()); }
+ iterator end() { return iterator(List.end()); }
+ const_iterator begin() const { return const_iterator(List.begin()); }
+ const_iterator end() const { return const_iterator(List.end()); }
+ reverse_iterator rbegin() { return reverse_iterator(List.rbegin()); }
+ reverse_iterator rend() { return reverse_iterator(List.rend()); }
+ const_reverse_iterator rbegin() const {
+ return const_reverse_iterator(List.rbegin());
+ }
+ const_reverse_iterator rend() const {
+ return const_reverse_iterator(List.rend());
+ }
+
+ T &back() { return List.back().V; }
+ T &front() { return List.front().V; }
+ const T &back() const { return List.back().V; }
+ const T &front() const { return List.front().V; }
+
+ template <class... Ts> iterator emplace(iterator I, Ts &&... Vs) {
+ return iterator(List.insert(I.wrapped(), *create(std::forward<Ts>(Vs)...)));
+ }
+
+ iterator insert(iterator I, T &&V) {
+ return iterator(List.insert(I.wrapped(), *create(std::move(V))));
+ }
+ iterator insert(iterator I, const T &V) {
+ return iterator(List.insert(I.wrapped(), *create(V)));
+ }
+
+ template <class Iterator>
+ void insert(iterator I, Iterator First, Iterator Last) {
+ for (; First != Last; ++First)
+ List.insert(I.wrapped(), *create(*First));
+ }
+
+ iterator erase(iterator I) {
+ return iterator(List.eraseAndDispose(I.wrapped(), Disposer(*this)));
+ }
+
+ iterator erase(iterator First, iterator Last) {
+ return iterator(
+ List.eraseAndDispose(First.wrapped(), Last.wrapped(), Disposer(*this)));
+ }
+
+ void clear() { List.clearAndDispose(Disposer(*this)); }
+ void pop_back() { List.eraseAndDispose(--List.end(), Disposer(*this)); }
+ void pop_front() { List.eraseAndDispose(List.begin(), Disposer(*this)); }
+ void push_back(T &&V) { insert(end(), std::move(V)); }
+ void push_front(T &&V) { insert(begin(), std::move(V)); }
+ void push_back(const T &V) { insert(end(), V); }
+ void push_front(const T &V) { insert(begin(), V); }
+ template <class... Ts> void emplace_back(Ts &&... Vs) {
+ emplace(end(), std::forward<Ts>(Vs)...);
+ }
+ template <class... Ts> void emplace_front(Ts &&... Vs) {
+ emplace(begin(), std::forward<Ts>(Vs)...);
+ }
+
+ /// Reset the underlying allocator.
+ ///
+ /// \pre \c empty()
+ void resetAlloc() {
+ assert(empty() && "Cannot reset allocator if not empty");
+ getAlloc().Reset();
+ }
+};
+
+template <class T> using BumpPtrList = AllocatorList<T, BumpPtrAllocator>;
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ALLOCATORLIST_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ArrayRef.h b/third_party/llvm-project/include/llvm/ADT/ArrayRef.h
new file mode 100644
index 000000000..f6455d3fa
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ArrayRef.h
@@ -0,0 +1,540 @@
+//===- ArrayRef.h - Array Reference Wrapper ---------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ARRAYREF_H
+#define LLVM_ADT_ARRAYREF_H
+
+#include "llvm/ADT/Hashing.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/Support/Compiler.h"
+#include <algorithm>
+#include <array>
+#include <cassert>
+#include <cstddef>
+#include <initializer_list>
+#include <iterator>
+#include <memory>
+#include <type_traits>
+#include <vector>
+
+namespace llvm {
+
+ /// ArrayRef - Represent a constant reference to an array (0 or more elements
+ /// consecutively in memory), i.e. a start pointer and a length. It allows
+ /// various APIs to take consecutive elements easily and conveniently.
+ ///
+ /// This class does not own the underlying data, it is expected to be used in
+ /// situations where the data resides in some other buffer, whose lifetime
+ /// extends past that of the ArrayRef. For this reason, it is not in general
+ /// safe to store an ArrayRef.
+ ///
+ /// This is intended to be trivially copyable, so it should be passed by
+ /// value.
+ template<typename T>
+ class LLVM_NODISCARD ArrayRef {
+ public:
+ using iterator = const T *;
+ using const_iterator = const T *;
+ using size_type = size_t;
+ using reverse_iterator = std::reverse_iterator<iterator>;
+
+ private:
+ /// The start of the array, in an external buffer.
+ const T *Data = nullptr;
+
+ /// The number of elements.
+ size_type Length = 0;
+
+ public:
+ /// @name Constructors
+ /// @{
+
+ /// Construct an empty ArrayRef.
+ /*implicit*/ ArrayRef() = default;
+
+ /// Construct an empty ArrayRef from None.
+ /*implicit*/ ArrayRef(NoneType) {}
+
+ /// Construct an ArrayRef from a single element.
+ /*implicit*/ ArrayRef(const T &OneElt)
+ : Data(&OneElt), Length(1) {}
+
+ /// Construct an ArrayRef from a pointer and length.
+ /*implicit*/ ArrayRef(const T *data, size_t length)
+ : Data(data), Length(length) {}
+
+ /// Construct an ArrayRef from a range.
+ ArrayRef(const T *begin, const T *end)
+ : Data(begin), Length(end - begin) {}
+
+ /// Construct an ArrayRef from a SmallVector. This is templated in order to
+ /// avoid instantiating SmallVectorTemplateCommon<T> whenever we
+ /// copy-construct an ArrayRef.
+ template<typename U>
+ /*implicit*/ ArrayRef(const SmallVectorTemplateCommon<T, U> &Vec)
+ : Data(Vec.data()), Length(Vec.size()) {
+ }
+
+ /// Construct an ArrayRef from a std::vector.
+ template<typename A>
+ /*implicit*/ ArrayRef(const std::vector<T, A> &Vec)
+ : Data(Vec.data()), Length(Vec.size()) {}
+
+ /// Construct an ArrayRef from a std::array
+ template <size_t N>
+ /*implicit*/ constexpr ArrayRef(const std::array<T, N> &Arr)
+ : Data(Arr.data()), Length(N) {}
+
+ /// Construct an ArrayRef from a C array.
+ template <size_t N>
+ /*implicit*/ constexpr ArrayRef(const T (&Arr)[N]) : Data(Arr), Length(N) {}
+
+ /// Construct an ArrayRef from a std::initializer_list.
+ /*implicit*/ ArrayRef(const std::initializer_list<T> &Vec)
+ : Data(Vec.begin() == Vec.end() ? (T*)nullptr : Vec.begin()),
+ Length(Vec.size()) {}
+
+ /// Construct an ArrayRef<const T*> from ArrayRef<T*>. This uses SFINAE to
+ /// ensure that only ArrayRefs of pointers can be converted.
+ template <typename U>
+ ArrayRef(
+ const ArrayRef<U *> &A,
+ typename std::enable_if<
+ std::is_convertible<U *const *, T const *>::value>::type * = nullptr)
+ : Data(A.data()), Length(A.size()) {}
+
+ /// Construct an ArrayRef<const T*> from a SmallVector<T*>. This is
+ /// templated in order to avoid instantiating SmallVectorTemplateCommon<T>
+ /// whenever we copy-construct an ArrayRef.
+ template<typename U, typename DummyT>
+ /*implicit*/ ArrayRef(
+ const SmallVectorTemplateCommon<U *, DummyT> &Vec,
+ typename std::enable_if<
+ std::is_convertible<U *const *, T const *>::value>::type * = nullptr)
+ : Data(Vec.data()), Length(Vec.size()) {
+ }
+
+ /// Construct an ArrayRef<const T*> from std::vector<T*>. This uses SFINAE
+ /// to ensure that only vectors of pointers can be converted.
+ template<typename U, typename A>
+ ArrayRef(const std::vector<U *, A> &Vec,
+ typename std::enable_if<
+ std::is_convertible<U *const *, T const *>::value>::type* = 0)
+ : Data(Vec.data()), Length(Vec.size()) {}
+
+ /// @}
+ /// @name Simple Operations
+ /// @{
+
+ iterator begin() const { return Data; }
+ iterator end() const { return Data + Length; }
+
+ reverse_iterator rbegin() const { return reverse_iterator(end()); }
+ reverse_iterator rend() const { return reverse_iterator(begin()); }
+
+ /// empty - Check if the array is empty.
+ bool empty() const { return Length == 0; }
+
+ const T *data() const { return Data; }
+
+ /// size - Get the array size.
+ size_t size() const { return Length; }
+
+ /// front - Get the first element.
+ const T &front() const {
+ assert(!empty());
+ return Data[0];
+ }
+
+ /// back - Get the last element.
+ const T &back() const {
+ assert(!empty());
+ return Data[Length-1];
+ }
+
+ // copy - Allocate copy in Allocator and return ArrayRef<T> to it.
+ template <typename Allocator> ArrayRef<T> copy(Allocator &A) {
+ T *Buff = A.template Allocate<T>(Length);
+ std::uninitialized_copy(begin(), end(), Buff);
+ return ArrayRef<T>(Buff, Length);
+ }
+
+ /// equals - Check for element-wise equality.
+ bool equals(ArrayRef RHS) const {
+ if (Length != RHS.Length)
+ return false;
+ return std::equal(begin(), end(), RHS.begin());
+ }
+
+ /// slice(n, m) - Chop off the first N elements of the array, and keep M
+ /// elements in the array.
+ ArrayRef<T> slice(size_t N, size_t M) const {
+ assert(N+M <= size() && "Invalid specifier");
+ return ArrayRef<T>(data()+N, M);
+ }
+
+ /// slice(n) - Chop off the first N elements of the array.
+ ArrayRef<T> slice(size_t N) const { return slice(N, size() - N); }
+
+ /// Drop the first \p N elements of the array.
+ ArrayRef<T> drop_front(size_t N = 1) const {
+ assert(size() >= N && "Dropping more elements than exist");
+ return slice(N, size() - N);
+ }
+
+ /// Drop the last \p N elements of the array.
+ ArrayRef<T> drop_back(size_t N = 1) const {
+ assert(size() >= N && "Dropping more elements than exist");
+ return slice(0, size() - N);
+ }
+
+ /// Return a copy of *this with the first N elements satisfying the
+ /// given predicate removed.
+ template <class PredicateT> ArrayRef<T> drop_while(PredicateT Pred) const {
+ return ArrayRef<T>(find_if_not(*this, Pred), end());
+ }
+
+ /// Return a copy of *this with the first N elements not satisfying
+ /// the given predicate removed.
+ template <class PredicateT> ArrayRef<T> drop_until(PredicateT Pred) const {
+ return ArrayRef<T>(find_if(*this, Pred), end());
+ }
+
+ /// Return a copy of *this with only the first \p N elements.
+ ArrayRef<T> take_front(size_t N = 1) const {
+ if (N >= size())
+ return *this;
+ return drop_back(size() - N);
+ }
+
+ /// Return a copy of *this with only the last \p N elements.
+ ArrayRef<T> take_back(size_t N = 1) const {
+ if (N >= size())
+ return *this;
+ return drop_front(size() - N);
+ }
+
+ /// Return the first N elements of this Array that satisfy the given
+ /// predicate.
+ template <class PredicateT> ArrayRef<T> take_while(PredicateT Pred) const {
+ return ArrayRef<T>(begin(), find_if_not(*this, Pred));
+ }
+
+ /// Return the first N elements of this Array that don't satisfy the
+ /// given predicate.
+ template <class PredicateT> ArrayRef<T> take_until(PredicateT Pred) const {
+ return ArrayRef<T>(begin(), find_if(*this, Pred));
+ }
+
+ /// @}
+ /// @name Operator Overloads
+ /// @{
+ const T &operator[](size_t Index) const {
+ assert(Index < Length && "Invalid index!");
+ return Data[Index];
+ }
+
+ /// Disallow accidental assignment from a temporary.
+ ///
+ /// The declaration here is extra complicated so that "arrayRef = {}"
+ /// continues to select the move assignment operator.
+ template <typename U>
+ typename std::enable_if<std::is_same<U, T>::value, ArrayRef<T>>::type &
+ operator=(U &&Temporary) = delete;
+
+ /// Disallow accidental assignment from a temporary.
+ ///
+ /// The declaration here is extra complicated so that "arrayRef = {}"
+ /// continues to select the move assignment operator.
+ template <typename U>
+ typename std::enable_if<std::is_same<U, T>::value, ArrayRef<T>>::type &
+ operator=(std::initializer_list<U>) = delete;
+
+ /// @}
+ /// @name Expensive Operations
+ /// @{
+ std::vector<T> vec() const {
+ return std::vector<T>(Data, Data+Length);
+ }
+
+ /// @}
+ /// @name Conversion operators
+ /// @{
+ operator std::vector<T>() const {
+ return std::vector<T>(Data, Data+Length);
+ }
+
+ /// @}
+ };
+
+ /// MutableArrayRef - Represent a mutable reference to an array (0 or more
+ /// elements consecutively in memory), i.e. a start pointer and a length. It
+ /// allows various APIs to take and modify consecutive elements easily and
+ /// conveniently.
+ ///
+ /// This class does not own the underlying data, it is expected to be used in
+ /// situations where the data resides in some other buffer, whose lifetime
+ /// extends past that of the MutableArrayRef. For this reason, it is not in
+ /// general safe to store a MutableArrayRef.
+ ///
+ /// This is intended to be trivially copyable, so it should be passed by
+ /// value.
+ template<typename T>
+ class LLVM_NODISCARD MutableArrayRef : public ArrayRef<T> {
+ public:
+ using iterator = T *;
+ using reverse_iterator = std::reverse_iterator<iterator>;
+
+ /// Construct an empty MutableArrayRef.
+ /*implicit*/ MutableArrayRef() = default;
+
+ /// Construct an empty MutableArrayRef from None.
+ /*implicit*/ MutableArrayRef(NoneType) : ArrayRef<T>() {}
+
+ /// Construct an MutableArrayRef from a single element.
+ /*implicit*/ MutableArrayRef(T &OneElt) : ArrayRef<T>(OneElt) {}
+
+ /// Construct an MutableArrayRef from a pointer and length.
+ /*implicit*/ MutableArrayRef(T *data, size_t length)
+ : ArrayRef<T>(data, length) {}
+
+ /// Construct an MutableArrayRef from a range.
+ MutableArrayRef(T *begin, T *end) : ArrayRef<T>(begin, end) {}
+
+ /// Construct an MutableArrayRef from a SmallVector.
+ /*implicit*/ MutableArrayRef(SmallVectorImpl<T> &Vec)
+ : ArrayRef<T>(Vec) {}
+
+ /// Construct a MutableArrayRef from a std::vector.
+ /*implicit*/ MutableArrayRef(std::vector<T> &Vec)
+ : ArrayRef<T>(Vec) {}
+
+ /// Construct an ArrayRef from a std::array
+ template <size_t N>
+ /*implicit*/ constexpr MutableArrayRef(std::array<T, N> &Arr)
+ : ArrayRef<T>(Arr) {}
+
+ /// Construct an MutableArrayRef from a C array.
+ template <size_t N>
+ /*implicit*/ constexpr MutableArrayRef(T (&Arr)[N]) : ArrayRef<T>(Arr) {}
+
+ T *data() const { return const_cast<T*>(ArrayRef<T>::data()); }
+
+ iterator begin() const { return data(); }
+ iterator end() const { return data() + this->size(); }
+
+ reverse_iterator rbegin() const { return reverse_iterator(end()); }
+ reverse_iterator rend() const { return reverse_iterator(begin()); }
+
+ /// front - Get the first element.
+ T &front() const {
+ assert(!this->empty());
+ return data()[0];
+ }
+
+ /// back - Get the last element.
+ T &back() const {
+ assert(!this->empty());
+ return data()[this->size()-1];
+ }
+
+ /// slice(n, m) - Chop off the first N elements of the array, and keep M
+ /// elements in the array.
+ MutableArrayRef<T> slice(size_t N, size_t M) const {
+ assert(N + M <= this->size() && "Invalid specifier");
+ return MutableArrayRef<T>(this->data() + N, M);
+ }
+
+ /// slice(n) - Chop off the first N elements of the array.
+ MutableArrayRef<T> slice(size_t N) const {
+ return slice(N, this->size() - N);
+ }
+
+ /// Drop the first \p N elements of the array.
+ MutableArrayRef<T> drop_front(size_t N = 1) const {
+ assert(this->size() >= N && "Dropping more elements than exist");
+ return slice(N, this->size() - N);
+ }
+
+ MutableArrayRef<T> drop_back(size_t N = 1) const {
+ assert(this->size() >= N && "Dropping more elements than exist");
+ return slice(0, this->size() - N);
+ }
+
+ /// Return a copy of *this with the first N elements satisfying the
+ /// given predicate removed.
+ template <class PredicateT>
+ MutableArrayRef<T> drop_while(PredicateT Pred) const {
+ return MutableArrayRef<T>(find_if_not(*this, Pred), end());
+ }
+
+ /// Return a copy of *this with the first N elements not satisfying
+ /// the given predicate removed.
+ template <class PredicateT>
+ MutableArrayRef<T> drop_until(PredicateT Pred) const {
+ return MutableArrayRef<T>(find_if(*this, Pred), end());
+ }
+
+ /// Return a copy of *this with only the first \p N elements.
+ MutableArrayRef<T> take_front(size_t N = 1) const {
+ if (N >= this->size())
+ return *this;
+ return drop_back(this->size() - N);
+ }
+
+ /// Return a copy of *this with only the last \p N elements.
+ MutableArrayRef<T> take_back(size_t N = 1) const {
+ if (N >= this->size())
+ return *this;
+ return drop_front(this->size() - N);
+ }
+
+ /// Return the first N elements of this Array that satisfy the given
+ /// predicate.
+ template <class PredicateT>
+ MutableArrayRef<T> take_while(PredicateT Pred) const {
+ return MutableArrayRef<T>(begin(), find_if_not(*this, Pred));
+ }
+
+ /// Return the first N elements of this Array that don't satisfy the
+ /// given predicate.
+ template <class PredicateT>
+ MutableArrayRef<T> take_until(PredicateT Pred) const {
+ return MutableArrayRef<T>(begin(), find_if(*this, Pred));
+ }
+
+ /// @}
+ /// @name Operator Overloads
+ /// @{
+ T &operator[](size_t Index) const {
+ assert(Index < this->size() && "Invalid index!");
+ return data()[Index];
+ }
+ };
+
+ /// This is a MutableArrayRef that owns its array.
+ template <typename T> class OwningArrayRef : public MutableArrayRef<T> {
+ public:
+ OwningArrayRef() = default;
+ OwningArrayRef(size_t Size) : MutableArrayRef<T>(new T[Size], Size) {}
+
+ OwningArrayRef(ArrayRef<T> Data)
+ : MutableArrayRef<T>(new T[Data.size()], Data.size()) {
+ std::copy(Data.begin(), Data.end(), this->begin());
+ }
+
+ OwningArrayRef(OwningArrayRef &&Other) { *this = std::move(Other); }
+
+ OwningArrayRef &operator=(OwningArrayRef &&Other) {
+ delete[] this->data();
+ this->MutableArrayRef<T>::operator=(Other);
+ Other.MutableArrayRef<T>::operator=(MutableArrayRef<T>());
+ return *this;
+ }
+
+ ~OwningArrayRef() { delete[] this->data(); }
+ };
+
+ /// @name ArrayRef Convenience constructors
+ /// @{
+
+ /// Construct an ArrayRef from a single element.
+ template<typename T>
+ ArrayRef<T> makeArrayRef(const T &OneElt) {
+ return OneElt;
+ }
+
+ /// Construct an ArrayRef from a pointer and length.
+ template<typename T>
+ ArrayRef<T> makeArrayRef(const T *data, size_t length) {
+ return ArrayRef<T>(data, length);
+ }
+
+ /// Construct an ArrayRef from a range.
+ template<typename T>
+ ArrayRef<T> makeArrayRef(const T *begin, const T *end) {
+ return ArrayRef<T>(begin, end);
+ }
+
+ /// Construct an ArrayRef from a SmallVector.
+ template <typename T>
+ ArrayRef<T> makeArrayRef(const SmallVectorImpl<T> &Vec) {
+ return Vec;
+ }
+
+ /// Construct an ArrayRef from a SmallVector.
+ template <typename T, unsigned N>
+ ArrayRef<T> makeArrayRef(const SmallVector<T, N> &Vec) {
+ return Vec;
+ }
+
+ /// Construct an ArrayRef from a std::vector.
+ template<typename T>
+ ArrayRef<T> makeArrayRef(const std::vector<T> &Vec) {
+ return Vec;
+ }
+
+ /// Construct an ArrayRef from a std::array.
+ template <typename T, std::size_t N>
+ ArrayRef<T> makeArrayRef(const std::array<T, N> &Arr) {
+ return Arr;
+ }
+
+ /// Construct an ArrayRef from an ArrayRef (no-op) (const)
+ template <typename T> ArrayRef<T> makeArrayRef(const ArrayRef<T> &Vec) {
+ return Vec;
+ }
+
+ /// Construct an ArrayRef from an ArrayRef (no-op)
+ template <typename T> ArrayRef<T> &makeArrayRef(ArrayRef<T> &Vec) {
+ return Vec;
+ }
+
+ /// Construct an ArrayRef from a C array.
+ template<typename T, size_t N>
+ ArrayRef<T> makeArrayRef(const T (&Arr)[N]) {
+ return ArrayRef<T>(Arr);
+ }
+
+ /// Construct a MutableArrayRef from a single element.
+ template<typename T>
+ MutableArrayRef<T> makeMutableArrayRef(T &OneElt) {
+ return OneElt;
+ }
+
+ /// Construct a MutableArrayRef from a pointer and length.
+ template<typename T>
+ MutableArrayRef<T> makeMutableArrayRef(T *data, size_t length) {
+ return MutableArrayRef<T>(data, length);
+ }
+
+ /// @}
+ /// @name ArrayRef Comparison Operators
+ /// @{
+
+ template<typename T>
+ inline bool operator==(ArrayRef<T> LHS, ArrayRef<T> RHS) {
+ return LHS.equals(RHS);
+ }
+
+ template<typename T>
+ inline bool operator!=(ArrayRef<T> LHS, ArrayRef<T> RHS) {
+ return !(LHS == RHS);
+ }
+
+ /// @}
+
+ template <typename T> hash_code hash_value(ArrayRef<T> S) {
+ return hash_combine_range(S.begin(), S.end());
+ }
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ARRAYREF_H
diff --git a/third_party/llvm-project/include/llvm/ADT/BitmaskEnum.h b/third_party/llvm-project/include/llvm/ADT/BitmaskEnum.h
new file mode 100644
index 000000000..1a18bc721
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/BitmaskEnum.h
@@ -0,0 +1,152 @@
+//===-- llvm/ADT/BitmaskEnum.h ----------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_BITMASKENUM_H
+#define LLVM_ADT_BITMASKENUM_H
+
+#include <cassert>
+#include <type_traits>
+#include <utility>
+
+#include "llvm/Support/MathExtras.h"
+
+/// LLVM_MARK_AS_BITMASK_ENUM lets you opt in an individual enum type so you can
+/// perform bitwise operations on it without putting static_cast everywhere.
+///
+/// \code
+/// enum MyEnum {
+/// E1 = 1, E2 = 2, E3 = 4, E4 = 8,
+/// LLVM_MARK_AS_BITMASK_ENUM(/* LargestValue = */ E4)
+/// };
+///
+/// void Foo() {
+/// MyEnum A = (E1 | E2) & E3 ^ ~E4; // Look, ma: No static_cast!
+/// }
+/// \endcode
+///
+/// Normally when you do a bitwise operation on an enum value, you get back an
+/// instance of the underlying type (e.g. int). But using this macro, bitwise
+/// ops on your enum will return you back instances of the enum. This is
+/// particularly useful for enums which represent a combination of flags.
+///
+/// The parameter to LLVM_MARK_AS_BITMASK_ENUM should be the largest individual
+/// value in your enum.
+///
+/// All of the enum's values must be non-negative.
+#define LLVM_MARK_AS_BITMASK_ENUM(LargestValue) \
+ LLVM_BITMASK_LARGEST_ENUMERATOR = LargestValue
+
+/// LLVM_ENABLE_BITMASK_ENUMS_IN_NAMESPACE() pulls the operator overloads used
+/// by LLVM_MARK_AS_BITMASK_ENUM into the current namespace.
+///
+/// Suppose you have an enum foo::bar::MyEnum. Before using
+/// LLVM_MARK_AS_BITMASK_ENUM on MyEnum, you must put
+/// LLVM_ENABLE_BITMASK_ENUMS_IN_NAMESPACE() somewhere inside namespace foo or
+/// namespace foo::bar. This allows the relevant operator overloads to be found
+/// by ADL.
+///
+/// You don't need to use this macro in namespace llvm; it's done at the bottom
+/// of this file.
+#define LLVM_ENABLE_BITMASK_ENUMS_IN_NAMESPACE() \
+ using ::llvm::BitmaskEnumDetail::operator~; \
+ using ::llvm::BitmaskEnumDetail::operator|; \
+ using ::llvm::BitmaskEnumDetail::operator&; \
+ using ::llvm::BitmaskEnumDetail::operator^; \
+ using ::llvm::BitmaskEnumDetail::operator|=; \
+ using ::llvm::BitmaskEnumDetail::operator&=; \
+ /* Force a semicolon at the end of this macro. */ \
+ using ::llvm::BitmaskEnumDetail::operator^=
+
+namespace llvm {
+
+/// Traits class to determine whether an enum has a
+/// LLVM_BITMASK_LARGEST_ENUMERATOR enumerator.
+template <typename E, typename Enable = void>
+struct is_bitmask_enum : std::false_type {};
+
+template <typename E>
+struct is_bitmask_enum<
+ E, typename std::enable_if<sizeof(E::LLVM_BITMASK_LARGEST_ENUMERATOR) >=
+ 0>::type> : std::true_type {};
+namespace BitmaskEnumDetail {
+
+/// Get a bitmask with 1s in all places up to the high-order bit of E's largest
+/// value.
+template <typename E> typename std::underlying_type<E>::type Mask() {
+ // On overflow, NextPowerOf2 returns zero with the type uint64_t, so
+ // subtracting 1 gives us the mask with all bits set, like we want.
+ return NextPowerOf2(static_cast<typename std::underlying_type<E>::type>(
+ E::LLVM_BITMASK_LARGEST_ENUMERATOR)) -
+ 1;
+}
+
+/// Check that Val is in range for E, and return Val cast to E's underlying
+/// type.
+template <typename E> typename std::underlying_type<E>::type Underlying(E Val) {
+ auto U = static_cast<typename std::underlying_type<E>::type>(Val);
+ assert(U >= 0 && "Negative enum values are not allowed.");
+ assert(U <= Mask<E>() && "Enum value too large (or largest val too small?)");
+ return U;
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E operator~(E Val) {
+ return static_cast<E>(~Underlying(Val) & Mask<E>());
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E operator|(E LHS, E RHS) {
+ return static_cast<E>(Underlying(LHS) | Underlying(RHS));
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E operator&(E LHS, E RHS) {
+ return static_cast<E>(Underlying(LHS) & Underlying(RHS));
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E operator^(E LHS, E RHS) {
+ return static_cast<E>(Underlying(LHS) ^ Underlying(RHS));
+}
+
+// |=, &=, and ^= return a reference to LHS, to match the behavior of the
+// operators on builtin types.
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E &operator|=(E &LHS, E RHS) {
+ LHS = LHS | RHS;
+ return LHS;
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E &operator&=(E &LHS, E RHS) {
+ LHS = LHS & RHS;
+ return LHS;
+}
+
+template <typename E,
+ typename = typename std::enable_if<is_bitmask_enum<E>::value>::type>
+E &operator^=(E &LHS, E RHS) {
+ LHS = LHS ^ RHS;
+ return LHS;
+}
+
+} // namespace BitmaskEnumDetail
+
+// Enable bitmask enums in namespace ::llvm and all nested namespaces.
+LLVM_ENABLE_BITMASK_ENUMS_IN_NAMESPACE();
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/DenseMap.h b/third_party/llvm-project/include/llvm/ADT/DenseMap.h
new file mode 100644
index 000000000..948a6e6bf
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/DenseMap.h
@@ -0,0 +1,1275 @@
+//===- llvm/ADT/DenseMap.h - Dense probed hash table ------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the DenseMap class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_DENSEMAP_H
+#define LLVM_ADT_DENSEMAP_H
+
+#include "llvm/ADT/DenseMapInfo.h"
+#include "llvm/ADT/EpochTracker.h"
+#include "llvm/Support/AlignOf.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/ReverseIteration.h"
+#include "llvm/Support/type_traits.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstring>
+#include <initializer_list>
+#include <iterator>
+#include <new>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+namespace detail {
+
+// We extend a pair to allow users to override the bucket type with their own
+// implementation without requiring two members.
+template <typename KeyT, typename ValueT>
+struct DenseMapPair : public std::pair<KeyT, ValueT> {
+ using std::pair<KeyT, ValueT>::pair;
+
+ KeyT &getFirst() { return std::pair<KeyT, ValueT>::first; }
+ const KeyT &getFirst() const { return std::pair<KeyT, ValueT>::first; }
+ ValueT &getSecond() { return std::pair<KeyT, ValueT>::second; }
+ const ValueT &getSecond() const { return std::pair<KeyT, ValueT>::second; }
+};
+
+} // end namespace detail
+
+template <typename KeyT, typename ValueT,
+ typename KeyInfoT = DenseMapInfo<KeyT>,
+ typename Bucket = llvm::detail::DenseMapPair<KeyT, ValueT>,
+ bool IsConst = false>
+class DenseMapIterator;
+
+template <typename DerivedT, typename KeyT, typename ValueT, typename KeyInfoT,
+ typename BucketT>
+class DenseMapBase : public DebugEpochBase {
+ template <typename T>
+ using const_arg_type_t = typename const_pointer_or_const_ref<T>::type;
+
+public:
+ using size_type = unsigned;
+ using key_type = KeyT;
+ using mapped_type = ValueT;
+ using value_type = BucketT;
+
+ using iterator = DenseMapIterator<KeyT, ValueT, KeyInfoT, BucketT>;
+ using const_iterator =
+ DenseMapIterator<KeyT, ValueT, KeyInfoT, BucketT, true>;
+
+ inline iterator begin() {
+ // When the map is empty, avoid the overhead of advancing/retreating past
+ // empty buckets.
+ if (empty())
+ return end();
+ if (shouldReverseIterate<KeyT>())
+ return makeIterator(getBucketsEnd() - 1, getBuckets(), *this);
+ return makeIterator(getBuckets(), getBucketsEnd(), *this);
+ }
+ inline iterator end() {
+ return makeIterator(getBucketsEnd(), getBucketsEnd(), *this, true);
+ }
+ inline const_iterator begin() const {
+ if (empty())
+ return end();
+ if (shouldReverseIterate<KeyT>())
+ return makeConstIterator(getBucketsEnd() - 1, getBuckets(), *this);
+ return makeConstIterator(getBuckets(), getBucketsEnd(), *this);
+ }
+ inline const_iterator end() const {
+ return makeConstIterator(getBucketsEnd(), getBucketsEnd(), *this, true);
+ }
+
+ LLVM_NODISCARD bool empty() const {
+ return getNumEntries() == 0;
+ }
+ unsigned size() const { return getNumEntries(); }
+
+ /// Grow the densemap so that it can contain at least \p NumEntries items
+ /// before resizing again.
+ void reserve(size_type NumEntries) {
+ auto NumBuckets = getMinBucketToReserveForEntries(NumEntries);
+ incrementEpoch();
+ if (NumBuckets > getNumBuckets())
+ grow(NumBuckets);
+ }
+
+ void clear() {
+ incrementEpoch();
+ if (getNumEntries() == 0 && getNumTombstones() == 0) return;
+
+ // If the capacity of the array is huge, and the # elements used is small,
+ // shrink the array.
+ if (getNumEntries() * 4 < getNumBuckets() && getNumBuckets() > 64) {
+ shrink_and_clear();
+ return;
+ }
+
+ const KeyT EmptyKey = getEmptyKey(), TombstoneKey = getTombstoneKey();
+ if (is_trivially_copyable<KeyT>::value &&
+ is_trivially_copyable<ValueT>::value) {
+ // Use a simpler loop when these are trivial types.
+ for (BucketT *P = getBuckets(), *E = getBucketsEnd(); P != E; ++P)
+ P->getFirst() = EmptyKey;
+ } else {
+ unsigned NumEntries = getNumEntries();
+ for (BucketT *P = getBuckets(), *E = getBucketsEnd(); P != E; ++P) {
+ if (!KeyInfoT::isEqual(P->getFirst(), EmptyKey)) {
+ if (!KeyInfoT::isEqual(P->getFirst(), TombstoneKey)) {
+ P->getSecond().~ValueT();
+ --NumEntries;
+ }
+ P->getFirst() = EmptyKey;
+ }
+ }
+ assert(NumEntries == 0 && "Node count imbalance!");
+ }
+ setNumEntries(0);
+ setNumTombstones(0);
+ }
+
+ /// Return 1 if the specified key is in the map, 0 otherwise.
+ size_type count(const_arg_type_t<KeyT> Val) const {
+ const BucketT *TheBucket;
+ return LookupBucketFor(Val, TheBucket) ? 1 : 0;
+ }
+
+ iterator find(const_arg_type_t<KeyT> Val) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return makeIterator(TheBucket, getBucketsEnd(), *this, true);
+ return end();
+ }
+ const_iterator find(const_arg_type_t<KeyT> Val) const {
+ const BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return makeConstIterator(TheBucket, getBucketsEnd(), *this, true);
+ return end();
+ }
+
+ /// Alternate version of find() which allows a different, and possibly
+ /// less expensive, key type.
+ /// The DenseMapInfo is responsible for supplying methods
+ /// getHashValue(LookupKeyT) and isEqual(LookupKeyT, KeyT) for each key
+ /// type used.
+ template<class LookupKeyT>
+ iterator find_as(const LookupKeyT &Val) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return makeIterator(TheBucket, getBucketsEnd(), *this, true);
+ return end();
+ }
+ template<class LookupKeyT>
+ const_iterator find_as(const LookupKeyT &Val) const {
+ const BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return makeConstIterator(TheBucket, getBucketsEnd(), *this, true);
+ return end();
+ }
+
+ /// lookup - Return the entry for the specified key, or a default
+ /// constructed value if no such entry exists.
+ ValueT lookup(const_arg_type_t<KeyT> Val) const {
+ const BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return TheBucket->getSecond();
+ return ValueT();
+ }
+
+ // Inserts key,value pair into the map if the key isn't already in the map.
+ // If the key is already in the map, it returns false and doesn't update the
+ // value.
+ std::pair<iterator, bool> insert(const std::pair<KeyT, ValueT> &KV) {
+ return try_emplace(KV.first, KV.second);
+ }
+
+ // Inserts key,value pair into the map if the key isn't already in the map.
+ // If the key is already in the map, it returns false and doesn't update the
+ // value.
+ std::pair<iterator, bool> insert(std::pair<KeyT, ValueT> &&KV) {
+ return try_emplace(std::move(KV.first), std::move(KV.second));
+ }
+
+ // Inserts key,value pair into the map if the key isn't already in the map.
+ // The value is constructed in-place if the key is not in the map, otherwise
+ // it is not moved.
+ template <typename... Ts>
+ std::pair<iterator, bool> try_emplace(KeyT &&Key, Ts &&... Args) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Key, TheBucket))
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ false); // Already in map.
+
+ // Otherwise, insert the new element.
+ TheBucket =
+ InsertIntoBucket(TheBucket, std::move(Key), std::forward<Ts>(Args)...);
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ true);
+ }
+
+ // Inserts key,value pair into the map if the key isn't already in the map.
+ // The value is constructed in-place if the key is not in the map, otherwise
+ // it is not moved.
+ template <typename... Ts>
+ std::pair<iterator, bool> try_emplace(const KeyT &Key, Ts &&... Args) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Key, TheBucket))
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ false); // Already in map.
+
+ // Otherwise, insert the new element.
+ TheBucket = InsertIntoBucket(TheBucket, Key, std::forward<Ts>(Args)...);
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ true);
+ }
+
+ /// Alternate version of insert() which allows a different, and possibly
+ /// less expensive, key type.
+ /// The DenseMapInfo is responsible for supplying methods
+ /// getHashValue(LookupKeyT) and isEqual(LookupKeyT, KeyT) for each key
+ /// type used.
+ template <typename LookupKeyT>
+ std::pair<iterator, bool> insert_as(std::pair<KeyT, ValueT> &&KV,
+ const LookupKeyT &Val) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Val, TheBucket))
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ false); // Already in map.
+
+ // Otherwise, insert the new element.
+ TheBucket = InsertIntoBucketWithLookup(TheBucket, std::move(KV.first),
+ std::move(KV.second), Val);
+ return std::make_pair(
+ makeIterator(TheBucket, getBucketsEnd(), *this, true),
+ true);
+ }
+
+ /// insert - Range insertion of pairs.
+ template<typename InputIt>
+ void insert(InputIt I, InputIt E) {
+ for (; I != E; ++I)
+ insert(*I);
+ }
+
+ bool erase(const KeyT &Val) {
+ BucketT *TheBucket;
+ if (!LookupBucketFor(Val, TheBucket))
+ return false; // not in map.
+
+ TheBucket->getSecond().~ValueT();
+ TheBucket->getFirst() = getTombstoneKey();
+ decrementNumEntries();
+ incrementNumTombstones();
+ return true;
+ }
+ void erase(iterator I) {
+ BucketT *TheBucket = &*I;
+ TheBucket->getSecond().~ValueT();
+ TheBucket->getFirst() = getTombstoneKey();
+ decrementNumEntries();
+ incrementNumTombstones();
+ }
+
+ value_type& FindAndConstruct(const KeyT &Key) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Key, TheBucket))
+ return *TheBucket;
+
+ return *InsertIntoBucket(TheBucket, Key);
+ }
+
+ ValueT &operator[](const KeyT &Key) {
+ return FindAndConstruct(Key).second;
+ }
+
+ value_type& FindAndConstruct(KeyT &&Key) {
+ BucketT *TheBucket;
+ if (LookupBucketFor(Key, TheBucket))
+ return *TheBucket;
+
+ return *InsertIntoBucket(TheBucket, std::move(Key));
+ }
+
+ ValueT &operator[](KeyT &&Key) {
+ return FindAndConstruct(std::move(Key)).second;
+ }
+
+ /// isPointerIntoBucketsArray - Return true if the specified pointer points
+ /// somewhere into the DenseMap's array of buckets (i.e. either to a key or
+ /// value in the DenseMap).
+ bool isPointerIntoBucketsArray(const void *Ptr) const {
+ return Ptr >= getBuckets() && Ptr < getBucketsEnd();
+ }
+
+ /// getPointerIntoBucketsArray() - Return an opaque pointer into the buckets
+ /// array. In conjunction with the previous method, this can be used to
+ /// determine whether an insertion caused the DenseMap to reallocate.
+ const void *getPointerIntoBucketsArray() const { return getBuckets(); }
+
+protected:
+ DenseMapBase() = default;
+
+ void destroyAll() {
+ if (getNumBuckets() == 0) // Nothing to do.
+ return;
+
+ const KeyT EmptyKey = getEmptyKey(), TombstoneKey = getTombstoneKey();
+ for (BucketT *P = getBuckets(), *E = getBucketsEnd(); P != E; ++P) {
+ if (!KeyInfoT::isEqual(P->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(P->getFirst(), TombstoneKey))
+ P->getSecond().~ValueT();
+ P->getFirst().~KeyT();
+ }
+ }
+
+ void initEmpty() {
+ setNumEntries(0);
+ setNumTombstones(0);
+
+ assert((getNumBuckets() & (getNumBuckets()-1)) == 0 &&
+ "# initial buckets must be a power of two!");
+ const KeyT EmptyKey = getEmptyKey();
+ for (BucketT *B = getBuckets(), *E = getBucketsEnd(); B != E; ++B)
+ ::new (&B->getFirst()) KeyT(EmptyKey);
+ }
+
+ /// Returns the number of buckets to allocate to ensure that the DenseMap can
+ /// accommodate \p NumEntries without need to grow().
+ unsigned getMinBucketToReserveForEntries(unsigned NumEntries) {
+ // Ensure that "NumEntries * 4 < NumBuckets * 3"
+ if (NumEntries == 0)
+ return 0;
+ // +1 is required because of the strict equality.
+ // For example if NumEntries is 48, we need to return 401.
+ return NextPowerOf2(NumEntries * 4 / 3 + 1);
+ }
+
+ void moveFromOldBuckets(BucketT *OldBucketsBegin, BucketT *OldBucketsEnd) {
+ initEmpty();
+
+ // Insert all the old elements.
+ const KeyT EmptyKey = getEmptyKey();
+ const KeyT TombstoneKey = getTombstoneKey();
+ for (BucketT *B = OldBucketsBegin, *E = OldBucketsEnd; B != E; ++B) {
+ if (!KeyInfoT::isEqual(B->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(B->getFirst(), TombstoneKey)) {
+ // Insert the key/value into the new table.
+ BucketT *DestBucket;
+ bool FoundVal = LookupBucketFor(B->getFirst(), DestBucket);
+ (void)FoundVal; // silence warning.
+ assert(!FoundVal && "Key already in new map?");
+ DestBucket->getFirst() = std::move(B->getFirst());
+ ::new (&DestBucket->getSecond()) ValueT(std::move(B->getSecond()));
+ incrementNumEntries();
+
+ // Free the value.
+ B->getSecond().~ValueT();
+ }
+ B->getFirst().~KeyT();
+ }
+ }
+
+ template <typename OtherBaseT>
+ void copyFrom(
+ const DenseMapBase<OtherBaseT, KeyT, ValueT, KeyInfoT, BucketT> &other) {
+ assert(&other != this);
+ assert(getNumBuckets() == other.getNumBuckets());
+
+ setNumEntries(other.getNumEntries());
+ setNumTombstones(other.getNumTombstones());
+
+ if (is_trivially_copyable<KeyT>::value &&
+ is_trivially_copyable<ValueT>::value)
+ memcpy(reinterpret_cast<void *>(getBuckets()), other.getBuckets(),
+ getNumBuckets() * sizeof(BucketT));
+ else
+ for (size_t i = 0; i < getNumBuckets(); ++i) {
+ ::new (&getBuckets()[i].getFirst())
+ KeyT(other.getBuckets()[i].getFirst());
+ if (!KeyInfoT::isEqual(getBuckets()[i].getFirst(), getEmptyKey()) &&
+ !KeyInfoT::isEqual(getBuckets()[i].getFirst(), getTombstoneKey()))
+ ::new (&getBuckets()[i].getSecond())
+ ValueT(other.getBuckets()[i].getSecond());
+ }
+ }
+
+ static unsigned getHashValue(const KeyT &Val) {
+ return KeyInfoT::getHashValue(Val);
+ }
+
+ template<typename LookupKeyT>
+ static unsigned getHashValue(const LookupKeyT &Val) {
+ return KeyInfoT::getHashValue(Val);
+ }
+
+ static const KeyT getEmptyKey() {
+ static_assert(std::is_base_of<DenseMapBase, DerivedT>::value,
+ "Must pass the derived type to this template!");
+ return KeyInfoT::getEmptyKey();
+ }
+
+ static const KeyT getTombstoneKey() {
+ return KeyInfoT::getTombstoneKey();
+ }
+
+private:
+ iterator makeIterator(BucketT *P, BucketT *E,
+ DebugEpochBase &Epoch,
+ bool NoAdvance=false) {
+ if (shouldReverseIterate<KeyT>()) {
+ BucketT *B = P == getBucketsEnd() ? getBuckets() : P + 1;
+ return iterator(B, E, Epoch, NoAdvance);
+ }
+ return iterator(P, E, Epoch, NoAdvance);
+ }
+
+ const_iterator makeConstIterator(const BucketT *P, const BucketT *E,
+ const DebugEpochBase &Epoch,
+ const bool NoAdvance=false) const {
+ if (shouldReverseIterate<KeyT>()) {
+ const BucketT *B = P == getBucketsEnd() ? getBuckets() : P + 1;
+ return const_iterator(B, E, Epoch, NoAdvance);
+ }
+ return const_iterator(P, E, Epoch, NoAdvance);
+ }
+
+ unsigned getNumEntries() const {
+ return static_cast<const DerivedT *>(this)->getNumEntries();
+ }
+
+ void setNumEntries(unsigned Num) {
+ static_cast<DerivedT *>(this)->setNumEntries(Num);
+ }
+
+ void incrementNumEntries() {
+ setNumEntries(getNumEntries() + 1);
+ }
+
+ void decrementNumEntries() {
+ setNumEntries(getNumEntries() - 1);
+ }
+
+ unsigned getNumTombstones() const {
+ return static_cast<const DerivedT *>(this)->getNumTombstones();
+ }
+
+ void setNumTombstones(unsigned Num) {
+ static_cast<DerivedT *>(this)->setNumTombstones(Num);
+ }
+
+ void incrementNumTombstones() {
+ setNumTombstones(getNumTombstones() + 1);
+ }
+
+ void decrementNumTombstones() {
+ setNumTombstones(getNumTombstones() - 1);
+ }
+
+ const BucketT *getBuckets() const {
+ return static_cast<const DerivedT *>(this)->getBuckets();
+ }
+
+ BucketT *getBuckets() {
+ return static_cast<DerivedT *>(this)->getBuckets();
+ }
+
+ unsigned getNumBuckets() const {
+ return static_cast<const DerivedT *>(this)->getNumBuckets();
+ }
+
+ BucketT *getBucketsEnd() {
+ return getBuckets() + getNumBuckets();
+ }
+
+ const BucketT *getBucketsEnd() const {
+ return getBuckets() + getNumBuckets();
+ }
+
+ void grow(unsigned AtLeast) {
+ static_cast<DerivedT *>(this)->grow(AtLeast);
+ }
+
+ void shrink_and_clear() {
+ static_cast<DerivedT *>(this)->shrink_and_clear();
+ }
+
+ template <typename KeyArg, typename... ValueArgs>
+ BucketT *InsertIntoBucket(BucketT *TheBucket, KeyArg &&Key,
+ ValueArgs &&... Values) {
+ TheBucket = InsertIntoBucketImpl(Key, Key, TheBucket);
+
+ TheBucket->getFirst() = std::forward<KeyArg>(Key);
+ ::new (&TheBucket->getSecond()) ValueT(std::forward<ValueArgs>(Values)...);
+ return TheBucket;
+ }
+
+ template <typename LookupKeyT>
+ BucketT *InsertIntoBucketWithLookup(BucketT *TheBucket, KeyT &&Key,
+ ValueT &&Value, LookupKeyT &Lookup) {
+ TheBucket = InsertIntoBucketImpl(Key, Lookup, TheBucket);
+
+ TheBucket->getFirst() = std::move(Key);
+ ::new (&TheBucket->getSecond()) ValueT(std::move(Value));
+ return TheBucket;
+ }
+
+ template <typename LookupKeyT>
+ BucketT *InsertIntoBucketImpl(const KeyT &Key, const LookupKeyT &Lookup,
+ BucketT *TheBucket) {
+ incrementEpoch();
+
+ // If the load of the hash table is more than 3/4, or if fewer than 1/8 of
+ // the buckets are empty (meaning that many are filled with tombstones),
+ // grow the table.
+ //
+ // The later case is tricky. For example, if we had one empty bucket with
+ // tons of tombstones, failing lookups (e.g. for insertion) would have to
+ // probe almost the entire table until it found the empty bucket. If the
+ // table completely filled with tombstones, no lookup would ever succeed,
+ // causing infinite loops in lookup.
+ unsigned NewNumEntries = getNumEntries() + 1;
+ unsigned NumBuckets = getNumBuckets();
+ if (LLVM_UNLIKELY(NewNumEntries * 4 >= NumBuckets * 3)) {
+ this->grow(NumBuckets * 2);
+ LookupBucketFor(Lookup, TheBucket);
+ NumBuckets = getNumBuckets();
+ } else if (LLVM_UNLIKELY(NumBuckets-(NewNumEntries+getNumTombstones()) <=
+ NumBuckets/8)) {
+ this->grow(NumBuckets);
+ LookupBucketFor(Lookup, TheBucket);
+ }
+ assert(TheBucket);
+
+ // Only update the state after we've grown our bucket space appropriately
+ // so that when growing buckets we have self-consistent entry count.
+ incrementNumEntries();
+
+ // If we are writing over a tombstone, remember this.
+ const KeyT EmptyKey = getEmptyKey();
+ if (!KeyInfoT::isEqual(TheBucket->getFirst(), EmptyKey))
+ decrementNumTombstones();
+
+ return TheBucket;
+ }
+
+ /// LookupBucketFor - Lookup the appropriate bucket for Val, returning it in
+ /// FoundBucket. If the bucket contains the key and a value, this returns
+ /// true, otherwise it returns a bucket with an empty marker or tombstone and
+ /// returns false.
+ template<typename LookupKeyT>
+ bool LookupBucketFor(const LookupKeyT &Val,
+ const BucketT *&FoundBucket) const {
+ const BucketT *BucketsPtr = getBuckets();
+ const unsigned NumBuckets = getNumBuckets();
+
+ if (NumBuckets == 0) {
+ FoundBucket = nullptr;
+ return false;
+ }
+
+ // FoundTombstone - Keep track of whether we find a tombstone while probing.
+ const BucketT *FoundTombstone = nullptr;
+ const KeyT EmptyKey = getEmptyKey();
+ const KeyT TombstoneKey = getTombstoneKey();
+ assert(!KeyInfoT::isEqual(Val, EmptyKey) &&
+ !KeyInfoT::isEqual(Val, TombstoneKey) &&
+ "Empty/Tombstone value shouldn't be inserted into map!");
+
+ unsigned BucketNo = getHashValue(Val) & (NumBuckets-1);
+ unsigned ProbeAmt = 1;
+ while (true) {
+ const BucketT *ThisBucket = BucketsPtr + BucketNo;
+ // Found Val's bucket? If so, return it.
+ if (LLVM_LIKELY(KeyInfoT::isEqual(Val, ThisBucket->getFirst()))) {
+ FoundBucket = ThisBucket;
+ return true;
+ }
+
+ // If we found an empty bucket, the key doesn't exist in the set.
+ // Insert it and return the default value.
+ if (LLVM_LIKELY(KeyInfoT::isEqual(ThisBucket->getFirst(), EmptyKey))) {
+ // If we've already seen a tombstone while probing, fill it in instead
+ // of the empty bucket we eventually probed to.
+ FoundBucket = FoundTombstone ? FoundTombstone : ThisBucket;
+ return false;
+ }
+
+ // If this is a tombstone, remember it. If Val ends up not in the map, we
+ // prefer to return it than something that would require more probing.
+ if (KeyInfoT::isEqual(ThisBucket->getFirst(), TombstoneKey) &&
+ !FoundTombstone)
+ FoundTombstone = ThisBucket; // Remember the first tombstone found.
+
+ // Otherwise, it's a hash collision or a tombstone, continue quadratic
+ // probing.
+ BucketNo += ProbeAmt++;
+ BucketNo &= (NumBuckets-1);
+ }
+ }
+
+ template <typename LookupKeyT>
+ bool LookupBucketFor(const LookupKeyT &Val, BucketT *&FoundBucket) {
+ const BucketT *ConstFoundBucket;
+ bool Result = const_cast<const DenseMapBase *>(this)
+ ->LookupBucketFor(Val, ConstFoundBucket);
+ FoundBucket = const_cast<BucketT *>(ConstFoundBucket);
+ return Result;
+ }
+
+public:
+ /// Return the approximate size (in bytes) of the actual map.
+ /// This is just the raw memory used by DenseMap.
+ /// If entries are pointers to objects, the size of the referenced objects
+ /// are not included.
+ size_t getMemorySize() const {
+ return getNumBuckets() * sizeof(BucketT);
+ }
+};
+
+/// Equality comparison for DenseMap.
+///
+/// Iterates over elements of LHS confirming that each (key, value) pair in LHS
+/// is also in RHS, and that no additional pairs are in RHS.
+/// Equivalent to N calls to RHS.find and N value comparisons. Amortized
+/// complexity is linear, worst case is O(N^2) (if every hash collides).
+template <typename DerivedT, typename KeyT, typename ValueT, typename KeyInfoT,
+ typename BucketT>
+bool operator==(
+ const DenseMapBase<DerivedT, KeyT, ValueT, KeyInfoT, BucketT> &LHS,
+ const DenseMapBase<DerivedT, KeyT, ValueT, KeyInfoT, BucketT> &RHS) {
+ if (LHS.size() != RHS.size())
+ return false;
+
+ for (auto &KV : LHS) {
+ auto I = RHS.find(KV.first);
+ if (I == RHS.end() || I->second != KV.second)
+ return false;
+ }
+
+ return true;
+}
+
+/// Inequality comparison for DenseMap.
+///
+/// Equivalent to !(LHS == RHS). See operator== for performance notes.
+template <typename DerivedT, typename KeyT, typename ValueT, typename KeyInfoT,
+ typename BucketT>
+bool operator!=(
+ const DenseMapBase<DerivedT, KeyT, ValueT, KeyInfoT, BucketT> &LHS,
+ const DenseMapBase<DerivedT, KeyT, ValueT, KeyInfoT, BucketT> &RHS) {
+ return !(LHS == RHS);
+}
+
+template <typename KeyT, typename ValueT,
+ typename KeyInfoT = DenseMapInfo<KeyT>,
+ typename BucketT = llvm::detail::DenseMapPair<KeyT, ValueT>>
+class DenseMap : public DenseMapBase<DenseMap<KeyT, ValueT, KeyInfoT, BucketT>,
+ KeyT, ValueT, KeyInfoT, BucketT> {
+ friend class DenseMapBase<DenseMap, KeyT, ValueT, KeyInfoT, BucketT>;
+
+ // Lift some types from the dependent base class into this class for
+ // simplicity of referring to them.
+ using BaseT = DenseMapBase<DenseMap, KeyT, ValueT, KeyInfoT, BucketT>;
+
+ BucketT *Buckets;
+ unsigned NumEntries;
+ unsigned NumTombstones;
+ unsigned NumBuckets;
+
+public:
+ /// Create a DenseMap wth an optional \p InitialReserve that guarantee that
+ /// this number of elements can be inserted in the map without grow()
+ explicit DenseMap(unsigned InitialReserve = 0) { init(InitialReserve); }
+
+ DenseMap(const DenseMap &other) : BaseT() {
+ init(0);
+ copyFrom(other);
+ }
+
+ DenseMap(DenseMap &&other) : BaseT() {
+ init(0);
+ swap(other);
+ }
+
+ template<typename InputIt>
+ DenseMap(const InputIt &I, const InputIt &E) {
+ init(std::distance(I, E));
+ this->insert(I, E);
+ }
+
+ DenseMap(std::initializer_list<typename BaseT::value_type> Vals) {
+ init(Vals.size());
+ this->insert(Vals.begin(), Vals.end());
+ }
+
+ ~DenseMap() {
+ this->destroyAll();
+ deallocate_buffer(Buckets, sizeof(BucketT) * NumBuckets, alignof(BucketT));
+ }
+
+ void swap(DenseMap& RHS) {
+ this->incrementEpoch();
+ RHS.incrementEpoch();
+ std::swap(Buckets, RHS.Buckets);
+ std::swap(NumEntries, RHS.NumEntries);
+ std::swap(NumTombstones, RHS.NumTombstones);
+ std::swap(NumBuckets, RHS.NumBuckets);
+ }
+
+ DenseMap& operator=(const DenseMap& other) {
+ if (&other != this)
+ copyFrom(other);
+ return *this;
+ }
+
+ DenseMap& operator=(DenseMap &&other) {
+ this->destroyAll();
+ deallocate_buffer(Buckets, sizeof(BucketT) * NumBuckets, alignof(BucketT));
+ init(0);
+ swap(other);
+ return *this;
+ }
+
+ void copyFrom(const DenseMap& other) {
+ this->destroyAll();
+ deallocate_buffer(Buckets, sizeof(BucketT) * NumBuckets, alignof(BucketT));
+ if (allocateBuckets(other.NumBuckets)) {
+ this->BaseT::copyFrom(other);
+ } else {
+ NumEntries = 0;
+ NumTombstones = 0;
+ }
+ }
+
+ void init(unsigned InitNumEntries) {
+ auto InitBuckets = BaseT::getMinBucketToReserveForEntries(InitNumEntries);
+ if (allocateBuckets(InitBuckets)) {
+ this->BaseT::initEmpty();
+ } else {
+ NumEntries = 0;
+ NumTombstones = 0;
+ }
+ }
+
+ void grow(unsigned AtLeast) {
+ unsigned OldNumBuckets = NumBuckets;
+ BucketT *OldBuckets = Buckets;
+
+ allocateBuckets(std::max<unsigned>(64, static_cast<unsigned>(NextPowerOf2(AtLeast-1))));
+ assert(Buckets);
+ if (!OldBuckets) {
+ this->BaseT::initEmpty();
+ return;
+ }
+
+ this->moveFromOldBuckets(OldBuckets, OldBuckets+OldNumBuckets);
+
+ // Free the old table.
+ deallocate_buffer(OldBuckets, sizeof(BucketT) * OldNumBuckets,
+ alignof(BucketT));
+ }
+
+ void shrink_and_clear() {
+ unsigned OldNumBuckets = NumBuckets;
+ unsigned OldNumEntries = NumEntries;
+ this->destroyAll();
+
+ // Reduce the number of buckets.
+ unsigned NewNumBuckets = 0;
+ if (OldNumEntries)
+ NewNumBuckets = std::max(64, 1 << (Log2_32_Ceil(OldNumEntries) + 1));
+ if (NewNumBuckets == NumBuckets) {
+ this->BaseT::initEmpty();
+ return;
+ }
+
+ deallocate_buffer(Buckets, sizeof(BucketT) * OldNumBuckets,
+ alignof(BucketT));
+ init(NewNumBuckets);
+ }
+
+private:
+ unsigned getNumEntries() const {
+ return NumEntries;
+ }
+
+ void setNumEntries(unsigned Num) {
+ NumEntries = Num;
+ }
+
+ unsigned getNumTombstones() const {
+ return NumTombstones;
+ }
+
+ void setNumTombstones(unsigned Num) {
+ NumTombstones = Num;
+ }
+
+ BucketT *getBuckets() const {
+ return Buckets;
+ }
+
+ unsigned getNumBuckets() const {
+ return NumBuckets;
+ }
+
+ bool allocateBuckets(unsigned Num) {
+ NumBuckets = Num;
+ if (NumBuckets == 0) {
+ Buckets = nullptr;
+ return false;
+ }
+
+ Buckets = static_cast<BucketT *>(
+ allocate_buffer(sizeof(BucketT) * NumBuckets, alignof(BucketT)));
+ return true;
+ }
+};
+
+template <typename KeyT, typename ValueT, unsigned InlineBuckets = 4,
+ typename KeyInfoT = DenseMapInfo<KeyT>,
+ typename BucketT = llvm::detail::DenseMapPair<KeyT, ValueT>>
+class SmallDenseMap
+ : public DenseMapBase<
+ SmallDenseMap<KeyT, ValueT, InlineBuckets, KeyInfoT, BucketT>, KeyT,
+ ValueT, KeyInfoT, BucketT> {
+ friend class DenseMapBase<SmallDenseMap, KeyT, ValueT, KeyInfoT, BucketT>;
+
+ // Lift some types from the dependent base class into this class for
+ // simplicity of referring to them.
+ using BaseT = DenseMapBase<SmallDenseMap, KeyT, ValueT, KeyInfoT, BucketT>;
+
+ static_assert(isPowerOf2_64(InlineBuckets),
+ "InlineBuckets must be a power of 2.");
+
+ unsigned Small : 1;
+ unsigned NumEntries : 31;
+ unsigned NumTombstones;
+
+ struct LargeRep {
+ BucketT *Buckets;
+ unsigned NumBuckets;
+ };
+
+ /// A "union" of an inline bucket array and the struct representing
+ /// a large bucket. This union will be discriminated by the 'Small' bit.
+ AlignedCharArrayUnion<BucketT[InlineBuckets], LargeRep> storage;
+
+public:
+ explicit SmallDenseMap(unsigned NumInitBuckets = 0) {
+ init(NumInitBuckets);
+ }
+
+ SmallDenseMap(const SmallDenseMap &other) : BaseT() {
+ init(0);
+ copyFrom(other);
+ }
+
+ SmallDenseMap(SmallDenseMap &&other) : BaseT() {
+ init(0);
+ swap(other);
+ }
+
+ template<typename InputIt>
+ SmallDenseMap(const InputIt &I, const InputIt &E) {
+ init(NextPowerOf2(std::distance(I, E)));
+ this->insert(I, E);
+ }
+
+ ~SmallDenseMap() {
+ this->destroyAll();
+ deallocateBuckets();
+ }
+
+ void swap(SmallDenseMap& RHS) {
+ unsigned TmpNumEntries = RHS.NumEntries;
+ RHS.NumEntries = NumEntries;
+ NumEntries = TmpNumEntries;
+ std::swap(NumTombstones, RHS.NumTombstones);
+
+ const KeyT EmptyKey = this->getEmptyKey();
+ const KeyT TombstoneKey = this->getTombstoneKey();
+ if (Small && RHS.Small) {
+ // If we're swapping inline bucket arrays, we have to cope with some of
+ // the tricky bits of DenseMap's storage system: the buckets are not
+ // fully initialized. Thus we swap every key, but we may have
+ // a one-directional move of the value.
+ for (unsigned i = 0, e = InlineBuckets; i != e; ++i) {
+ BucketT *LHSB = &getInlineBuckets()[i],
+ *RHSB = &RHS.getInlineBuckets()[i];
+ bool hasLHSValue = (!KeyInfoT::isEqual(LHSB->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(LHSB->getFirst(), TombstoneKey));
+ bool hasRHSValue = (!KeyInfoT::isEqual(RHSB->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(RHSB->getFirst(), TombstoneKey));
+ if (hasLHSValue && hasRHSValue) {
+ // Swap together if we can...
+ std::swap(*LHSB, *RHSB);
+ continue;
+ }
+ // Swap separately and handle any assymetry.
+ std::swap(LHSB->getFirst(), RHSB->getFirst());
+ if (hasLHSValue) {
+ ::new (&RHSB->getSecond()) ValueT(std::move(LHSB->getSecond()));
+ LHSB->getSecond().~ValueT();
+ } else if (hasRHSValue) {
+ ::new (&LHSB->getSecond()) ValueT(std::move(RHSB->getSecond()));
+ RHSB->getSecond().~ValueT();
+ }
+ }
+ return;
+ }
+ if (!Small && !RHS.Small) {
+ std::swap(getLargeRep()->Buckets, RHS.getLargeRep()->Buckets);
+ std::swap(getLargeRep()->NumBuckets, RHS.getLargeRep()->NumBuckets);
+ return;
+ }
+
+ SmallDenseMap &SmallSide = Small ? *this : RHS;
+ SmallDenseMap &LargeSide = Small ? RHS : *this;
+
+ // First stash the large side's rep and move the small side across.
+ LargeRep TmpRep = std::move(*LargeSide.getLargeRep());
+ LargeSide.getLargeRep()->~LargeRep();
+ LargeSide.Small = true;
+ // This is similar to the standard move-from-old-buckets, but the bucket
+ // count hasn't actually rotated in this case. So we have to carefully
+ // move construct the keys and values into their new locations, but there
+ // is no need to re-hash things.
+ for (unsigned i = 0, e = InlineBuckets; i != e; ++i) {
+ BucketT *NewB = &LargeSide.getInlineBuckets()[i],
+ *OldB = &SmallSide.getInlineBuckets()[i];
+ ::new (&NewB->getFirst()) KeyT(std::move(OldB->getFirst()));
+ OldB->getFirst().~KeyT();
+ if (!KeyInfoT::isEqual(NewB->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(NewB->getFirst(), TombstoneKey)) {
+ ::new (&NewB->getSecond()) ValueT(std::move(OldB->getSecond()));
+ OldB->getSecond().~ValueT();
+ }
+ }
+
+ // The hard part of moving the small buckets across is done, just move
+ // the TmpRep into its new home.
+ SmallSide.Small = false;
+ new (SmallSide.getLargeRep()) LargeRep(std::move(TmpRep));
+ }
+
+ SmallDenseMap& operator=(const SmallDenseMap& other) {
+ if (&other != this)
+ copyFrom(other);
+ return *this;
+ }
+
+ SmallDenseMap& operator=(SmallDenseMap &&other) {
+ this->destroyAll();
+ deallocateBuckets();
+ init(0);
+ swap(other);
+ return *this;
+ }
+
+ void copyFrom(const SmallDenseMap& other) {
+ this->destroyAll();
+ deallocateBuckets();
+ Small = true;
+ if (other.getNumBuckets() > InlineBuckets) {
+ Small = false;
+ new (getLargeRep()) LargeRep(allocateBuckets(other.getNumBuckets()));
+ }
+ this->BaseT::copyFrom(other);
+ }
+
+ void init(unsigned InitBuckets) {
+ Small = true;
+ if (InitBuckets > InlineBuckets) {
+ Small = false;
+ new (getLargeRep()) LargeRep(allocateBuckets(InitBuckets));
+ }
+ this->BaseT::initEmpty();
+ }
+
+ void grow(unsigned AtLeast) {
+ if (AtLeast >= InlineBuckets)
+ AtLeast = std::max<unsigned>(64, NextPowerOf2(AtLeast-1));
+
+ if (Small) {
+ if (AtLeast < InlineBuckets)
+ return; // Nothing to do.
+
+ // First move the inline buckets into a temporary storage.
+ AlignedCharArrayUnion<BucketT[InlineBuckets]> TmpStorage;
+ BucketT *TmpBegin = reinterpret_cast<BucketT *>(TmpStorage.buffer);
+ BucketT *TmpEnd = TmpBegin;
+
+ // Loop over the buckets, moving non-empty, non-tombstones into the
+ // temporary storage. Have the loop move the TmpEnd forward as it goes.
+ const KeyT EmptyKey = this->getEmptyKey();
+ const KeyT TombstoneKey = this->getTombstoneKey();
+ for (BucketT *P = getBuckets(), *E = P + InlineBuckets; P != E; ++P) {
+ if (!KeyInfoT::isEqual(P->getFirst(), EmptyKey) &&
+ !KeyInfoT::isEqual(P->getFirst(), TombstoneKey)) {
+ assert(size_t(TmpEnd - TmpBegin) < InlineBuckets &&
+ "Too many inline buckets!");
+ ::new (&TmpEnd->getFirst()) KeyT(std::move(P->getFirst()));
+ ::new (&TmpEnd->getSecond()) ValueT(std::move(P->getSecond()));
+ ++TmpEnd;
+ P->getSecond().~ValueT();
+ }
+ P->getFirst().~KeyT();
+ }
+
+ // Now make this map use the large rep, and move all the entries back
+ // into it.
+ Small = false;
+ new (getLargeRep()) LargeRep(allocateBuckets(AtLeast));
+ this->moveFromOldBuckets(TmpBegin, TmpEnd);
+ return;
+ }
+
+ LargeRep OldRep = std::move(*getLargeRep());
+ getLargeRep()->~LargeRep();
+ if (AtLeast <= InlineBuckets) {
+ Small = true;
+ } else {
+ new (getLargeRep()) LargeRep(allocateBuckets(AtLeast));
+ }
+
+ this->moveFromOldBuckets(OldRep.Buckets, OldRep.Buckets+OldRep.NumBuckets);
+
+ // Free the old table.
+ deallocate_buffer(OldRep.Buckets, sizeof(BucketT) * OldRep.NumBuckets,
+ alignof(BucketT));
+ }
+
+ void shrink_and_clear() {
+ unsigned OldSize = this->size();
+ this->destroyAll();
+
+ // Reduce the number of buckets.
+ unsigned NewNumBuckets = 0;
+ if (OldSize) {
+ NewNumBuckets = 1 << (Log2_32_Ceil(OldSize) + 1);
+ if (NewNumBuckets > InlineBuckets && NewNumBuckets < 64u)
+ NewNumBuckets = 64;
+ }
+ if ((Small && NewNumBuckets <= InlineBuckets) ||
+ (!Small && NewNumBuckets == getLargeRep()->NumBuckets)) {
+ this->BaseT::initEmpty();
+ return;
+ }
+
+ deallocateBuckets();
+ init(NewNumBuckets);
+ }
+
+private:
+ unsigned getNumEntries() const {
+ return NumEntries;
+ }
+
+ void setNumEntries(unsigned Num) {
+ // NumEntries is hardcoded to be 31 bits wide.
+ assert(Num < (1U << 31) && "Cannot support more than 1<<31 entries");
+ NumEntries = Num;
+ }
+
+ unsigned getNumTombstones() const {
+ return NumTombstones;
+ }
+
+ void setNumTombstones(unsigned Num) {
+ NumTombstones = Num;
+ }
+
+ const BucketT *getInlineBuckets() const {
+ assert(Small);
+ // Note that this cast does not violate aliasing rules as we assert that
+ // the memory's dynamic type is the small, inline bucket buffer, and the
+ // 'storage.buffer' static type is 'char *'.
+ return reinterpret_cast<const BucketT *>(storage.buffer);
+ }
+
+ BucketT *getInlineBuckets() {
+ return const_cast<BucketT *>(
+ const_cast<const SmallDenseMap *>(this)->getInlineBuckets());
+ }
+
+ const LargeRep *getLargeRep() const {
+ assert(!Small);
+ // Note, same rule about aliasing as with getInlineBuckets.
+ return reinterpret_cast<const LargeRep *>(storage.buffer);
+ }
+
+ LargeRep *getLargeRep() {
+ return const_cast<LargeRep *>(
+ const_cast<const SmallDenseMap *>(this)->getLargeRep());
+ }
+
+ const BucketT *getBuckets() const {
+ return Small ? getInlineBuckets() : getLargeRep()->Buckets;
+ }
+
+ BucketT *getBuckets() {
+ return const_cast<BucketT *>(
+ const_cast<const SmallDenseMap *>(this)->getBuckets());
+ }
+
+ unsigned getNumBuckets() const {
+ return Small ? InlineBuckets : getLargeRep()->NumBuckets;
+ }
+
+ void deallocateBuckets() {
+ if (Small)
+ return;
+
+ deallocate_buffer(getLargeRep()->Buckets,
+ sizeof(BucketT) * getLargeRep()->NumBuckets,
+ alignof(BucketT));
+ getLargeRep()->~LargeRep();
+ }
+
+ LargeRep allocateBuckets(unsigned Num) {
+ assert(Num > InlineBuckets && "Must allocate more buckets than are inline");
+ LargeRep Rep = {static_cast<BucketT *>(allocate_buffer(
+ sizeof(BucketT) * Num, alignof(BucketT))),
+ Num};
+ return Rep;
+ }
+};
+
+template <typename KeyT, typename ValueT, typename KeyInfoT, typename Bucket,
+ bool IsConst>
+class DenseMapIterator : DebugEpochBase::HandleBase {
+ friend class DenseMapIterator<KeyT, ValueT, KeyInfoT, Bucket, true>;
+ friend class DenseMapIterator<KeyT, ValueT, KeyInfoT, Bucket, false>;
+
+ using ConstIterator = DenseMapIterator<KeyT, ValueT, KeyInfoT, Bucket, true>;
+
+public:
+ using difference_type = ptrdiff_t;
+ using value_type =
+ typename std::conditional<IsConst, const Bucket, Bucket>::type;
+ using pointer = value_type *;
+ using reference = value_type &;
+ using iterator_category = std::forward_iterator_tag;
+
+private:
+ pointer Ptr = nullptr;
+ pointer End = nullptr;
+
+public:
+ DenseMapIterator() = default;
+
+ DenseMapIterator(pointer Pos, pointer E, const DebugEpochBase &Epoch,
+ bool NoAdvance = false)
+ : DebugEpochBase::HandleBase(&Epoch), Ptr(Pos), End(E) {
+ assert(isHandleInSync() && "invalid construction!");
+
+ if (NoAdvance) return;
+ if (shouldReverseIterate<KeyT>()) {
+ RetreatPastEmptyBuckets();
+ return;
+ }
+ AdvancePastEmptyBuckets();
+ }
+
+ // Converting ctor from non-const iterators to const iterators. SFINAE'd out
+ // for const iterator destinations so it doesn't end up as a user defined copy
+ // constructor.
+ template <bool IsConstSrc,
+ typename = typename std::enable_if<!IsConstSrc && IsConst>::type>
+ DenseMapIterator(
+ const DenseMapIterator<KeyT, ValueT, KeyInfoT, Bucket, IsConstSrc> &I)
+ : DebugEpochBase::HandleBase(I), Ptr(I.Ptr), End(I.End) {}
+
+ reference operator*() const {
+ assert(isHandleInSync() && "invalid iterator access!");
+ if (shouldReverseIterate<KeyT>())
+ return Ptr[-1];
+ return *Ptr;
+ }
+ pointer operator->() const {
+ assert(isHandleInSync() && "invalid iterator access!");
+ if (shouldReverseIterate<KeyT>())
+ return &(Ptr[-1]);
+ return Ptr;
+ }
+
+ bool operator==(const ConstIterator &RHS) const {
+ assert((!Ptr || isHandleInSync()) && "handle not in sync!");
+ assert((!RHS.Ptr || RHS.isHandleInSync()) && "handle not in sync!");
+ assert(getEpochAddress() == RHS.getEpochAddress() &&
+ "comparing incomparable iterators!");
+ return Ptr == RHS.Ptr;
+ }
+ bool operator!=(const ConstIterator &RHS) const {
+ assert((!Ptr || isHandleInSync()) && "handle not in sync!");
+ assert((!RHS.Ptr || RHS.isHandleInSync()) && "handle not in sync!");
+ assert(getEpochAddress() == RHS.getEpochAddress() &&
+ "comparing incomparable iterators!");
+ return Ptr != RHS.Ptr;
+ }
+
+ inline DenseMapIterator& operator++() { // Preincrement
+ assert(isHandleInSync() && "invalid iterator access!");
+ if (shouldReverseIterate<KeyT>()) {
+ --Ptr;
+ RetreatPastEmptyBuckets();
+ return *this;
+ }
+ ++Ptr;
+ AdvancePastEmptyBuckets();
+ return *this;
+ }
+ DenseMapIterator operator++(int) { // Postincrement
+ assert(isHandleInSync() && "invalid iterator access!");
+ DenseMapIterator tmp = *this; ++*this; return tmp;
+ }
+
+private:
+ void AdvancePastEmptyBuckets() {
+ assert(Ptr <= End);
+ const KeyT Empty = KeyInfoT::getEmptyKey();
+ const KeyT Tombstone = KeyInfoT::getTombstoneKey();
+
+ while (Ptr != End && (KeyInfoT::isEqual(Ptr->getFirst(), Empty) ||
+ KeyInfoT::isEqual(Ptr->getFirst(), Tombstone)))
+ ++Ptr;
+ }
+
+ void RetreatPastEmptyBuckets() {
+ assert(Ptr >= End);
+ const KeyT Empty = KeyInfoT::getEmptyKey();
+ const KeyT Tombstone = KeyInfoT::getTombstoneKey();
+
+ while (Ptr != End && (KeyInfoT::isEqual(Ptr[-1].getFirst(), Empty) ||
+ KeyInfoT::isEqual(Ptr[-1].getFirst(), Tombstone)))
+ --Ptr;
+ }
+};
+
+template <typename KeyT, typename ValueT, typename KeyInfoT>
+inline size_t capacity_in_bytes(const DenseMap<KeyT, ValueT, KeyInfoT> &X) {
+ return X.getMemorySize();
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_DENSEMAP_H
diff --git a/third_party/llvm-project/include/llvm/ADT/DenseMapInfo.h b/third_party/llvm-project/include/llvm/ADT/DenseMapInfo.h
new file mode 100644
index 000000000..bd4c60c8f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/DenseMapInfo.h
@@ -0,0 +1,300 @@
+//===- llvm/ADT/DenseMapInfo.h - Type traits for DenseMap -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines DenseMapInfo traits for DenseMap.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_DENSEMAPINFO_H
+#define LLVM_ADT_DENSEMAPINFO_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/Hashing.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/PointerLikeTypeTraits.h"
+#include "llvm/Support/TypeSize.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+template<typename T>
+struct DenseMapInfo {
+ //static inline T getEmptyKey();
+ //static inline T getTombstoneKey();
+ //static unsigned getHashValue(const T &Val);
+ //static bool isEqual(const T &LHS, const T &RHS);
+};
+
+// Provide DenseMapInfo for all pointers.
+template<typename T>
+struct DenseMapInfo<T*> {
+ static inline T* getEmptyKey() {
+ uintptr_t Val = static_cast<uintptr_t>(-1);
+ Val <<= PointerLikeTypeTraits<T*>::NumLowBitsAvailable;
+ return reinterpret_cast<T*>(Val);
+ }
+
+ static inline T* getTombstoneKey() {
+ uintptr_t Val = static_cast<uintptr_t>(-2);
+ Val <<= PointerLikeTypeTraits<T*>::NumLowBitsAvailable;
+ return reinterpret_cast<T*>(Val);
+ }
+
+ static unsigned getHashValue(const T *PtrVal) {
+ return (unsigned((uintptr_t)PtrVal) >> 4) ^
+ (unsigned((uintptr_t)PtrVal) >> 9);
+ }
+
+ static bool isEqual(const T *LHS, const T *RHS) { return LHS == RHS; }
+};
+
+// Provide DenseMapInfo for chars.
+template<> struct DenseMapInfo<char> {
+ static inline char getEmptyKey() { return ~0; }
+ static inline char getTombstoneKey() { return ~0 - 1; }
+ static unsigned getHashValue(const char& Val) { return Val * 37U; }
+
+ static bool isEqual(const char &LHS, const char &RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for unsigned chars.
+template <> struct DenseMapInfo<unsigned char> {
+ static inline unsigned char getEmptyKey() { return ~0; }
+ static inline unsigned char getTombstoneKey() { return ~0 - 1; }
+ static unsigned getHashValue(const unsigned char &Val) { return Val * 37U; }
+
+ static bool isEqual(const unsigned char &LHS, const unsigned char &RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for unsigned shorts.
+template <> struct DenseMapInfo<unsigned short> {
+ static inline unsigned short getEmptyKey() { return 0xFFFF; }
+ static inline unsigned short getTombstoneKey() { return 0xFFFF - 1; }
+ static unsigned getHashValue(const unsigned short &Val) { return Val * 37U; }
+
+ static bool isEqual(const unsigned short &LHS, const unsigned short &RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for unsigned ints.
+template<> struct DenseMapInfo<unsigned> {
+ static inline unsigned getEmptyKey() { return ~0U; }
+ static inline unsigned getTombstoneKey() { return ~0U - 1; }
+ static unsigned getHashValue(const unsigned& Val) { return Val * 37U; }
+
+ static bool isEqual(const unsigned& LHS, const unsigned& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for unsigned longs.
+template<> struct DenseMapInfo<unsigned long> {
+ static inline unsigned long getEmptyKey() { return ~0UL; }
+ static inline unsigned long getTombstoneKey() { return ~0UL - 1L; }
+
+ static unsigned getHashValue(const unsigned long& Val) {
+ return (unsigned)(Val * 37UL);
+ }
+
+ static bool isEqual(const unsigned long& LHS, const unsigned long& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for unsigned long longs.
+template<> struct DenseMapInfo<unsigned long long> {
+ static inline unsigned long long getEmptyKey() { return ~0ULL; }
+ static inline unsigned long long getTombstoneKey() { return ~0ULL - 1ULL; }
+
+ static unsigned getHashValue(const unsigned long long& Val) {
+ return (unsigned)(Val * 37ULL);
+ }
+
+ static bool isEqual(const unsigned long long& LHS,
+ const unsigned long long& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for shorts.
+template <> struct DenseMapInfo<short> {
+ static inline short getEmptyKey() { return 0x7FFF; }
+ static inline short getTombstoneKey() { return -0x7FFF - 1; }
+ static unsigned getHashValue(const short &Val) { return Val * 37U; }
+ static bool isEqual(const short &LHS, const short &RHS) { return LHS == RHS; }
+};
+
+// Provide DenseMapInfo for ints.
+template<> struct DenseMapInfo<int> {
+ static inline int getEmptyKey() { return 0x7fffffff; }
+ static inline int getTombstoneKey() { return -0x7fffffff - 1; }
+ static unsigned getHashValue(const int& Val) { return (unsigned)(Val * 37U); }
+
+ static bool isEqual(const int& LHS, const int& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for longs.
+template<> struct DenseMapInfo<long> {
+ static inline long getEmptyKey() {
+ return (1UL << (sizeof(long) * 8 - 1)) - 1UL;
+ }
+
+ static inline long getTombstoneKey() { return getEmptyKey() - 1L; }
+
+ static unsigned getHashValue(const long& Val) {
+ return (unsigned)(Val * 37UL);
+ }
+
+ static bool isEqual(const long& LHS, const long& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for long longs.
+template<> struct DenseMapInfo<long long> {
+ static inline long long getEmptyKey() { return 0x7fffffffffffffffLL; }
+ static inline long long getTombstoneKey() { return -0x7fffffffffffffffLL-1; }
+
+ static unsigned getHashValue(const long long& Val) {
+ return (unsigned)(Val * 37ULL);
+ }
+
+ static bool isEqual(const long long& LHS,
+ const long long& RHS) {
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for all pairs whose members have info.
+template<typename T, typename U>
+struct DenseMapInfo<std::pair<T, U>> {
+ using Pair = std::pair<T, U>;
+ using FirstInfo = DenseMapInfo<T>;
+ using SecondInfo = DenseMapInfo<U>;
+
+ static inline Pair getEmptyKey() {
+ return std::make_pair(FirstInfo::getEmptyKey(),
+ SecondInfo::getEmptyKey());
+ }
+
+ static inline Pair getTombstoneKey() {
+ return std::make_pair(FirstInfo::getTombstoneKey(),
+ SecondInfo::getTombstoneKey());
+ }
+
+ static unsigned getHashValue(const Pair& PairVal) {
+ uint64_t key = (uint64_t)FirstInfo::getHashValue(PairVal.first) << 32
+ | (uint64_t)SecondInfo::getHashValue(PairVal.second);
+ key += ~(key << 32);
+ key ^= (key >> 22);
+ key += ~(key << 13);
+ key ^= (key >> 8);
+ key += (key << 3);
+ key ^= (key >> 15);
+ key += ~(key << 27);
+ key ^= (key >> 31);
+ return (unsigned)key;
+ }
+
+ static bool isEqual(const Pair &LHS, const Pair &RHS) {
+ return FirstInfo::isEqual(LHS.first, RHS.first) &&
+ SecondInfo::isEqual(LHS.second, RHS.second);
+ }
+};
+
+// Provide DenseMapInfo for StringRefs.
+template <> struct DenseMapInfo<StringRef> {
+ static inline StringRef getEmptyKey() {
+ return StringRef(reinterpret_cast<const char *>(~static_cast<uintptr_t>(0)),
+ 0);
+ }
+
+ static inline StringRef getTombstoneKey() {
+ return StringRef(reinterpret_cast<const char *>(~static_cast<uintptr_t>(1)),
+ 0);
+ }
+
+ static unsigned getHashValue(StringRef Val) {
+ assert(Val.data() != getEmptyKey().data() && "Cannot hash the empty key!");
+ assert(Val.data() != getTombstoneKey().data() &&
+ "Cannot hash the tombstone key!");
+ return (unsigned)(hash_value(Val));
+ }
+
+ static bool isEqual(StringRef LHS, StringRef RHS) {
+ if (RHS.data() == getEmptyKey().data())
+ return LHS.data() == getEmptyKey().data();
+ if (RHS.data() == getTombstoneKey().data())
+ return LHS.data() == getTombstoneKey().data();
+ return LHS == RHS;
+ }
+};
+
+// Provide DenseMapInfo for ArrayRefs.
+template <typename T> struct DenseMapInfo<ArrayRef<T>> {
+ static inline ArrayRef<T> getEmptyKey() {
+ return ArrayRef<T>(reinterpret_cast<const T *>(~static_cast<uintptr_t>(0)),
+ size_t(0));
+ }
+
+ static inline ArrayRef<T> getTombstoneKey() {
+ return ArrayRef<T>(reinterpret_cast<const T *>(~static_cast<uintptr_t>(1)),
+ size_t(0));
+ }
+
+ static unsigned getHashValue(ArrayRef<T> Val) {
+ assert(Val.data() != getEmptyKey().data() && "Cannot hash the empty key!");
+ assert(Val.data() != getTombstoneKey().data() &&
+ "Cannot hash the tombstone key!");
+ return (unsigned)(hash_value(Val));
+ }
+
+ static bool isEqual(ArrayRef<T> LHS, ArrayRef<T> RHS) {
+ if (RHS.data() == getEmptyKey().data())
+ return LHS.data() == getEmptyKey().data();
+ if (RHS.data() == getTombstoneKey().data())
+ return LHS.data() == getTombstoneKey().data();
+ return LHS == RHS;
+ }
+};
+
+template <> struct DenseMapInfo<hash_code> {
+ static inline hash_code getEmptyKey() { return hash_code(-1); }
+ static inline hash_code getTombstoneKey() { return hash_code(-2); }
+ static unsigned getHashValue(hash_code val) { return val; }
+ static bool isEqual(hash_code LHS, hash_code RHS) { return LHS == RHS; }
+};
+
+template <> struct DenseMapInfo<ElementCount> {
+ static inline ElementCount getEmptyKey() { return {~0U, true}; }
+ static inline ElementCount getTombstoneKey() { return {~0U - 1, false}; }
+ static unsigned getHashValue(const ElementCount& EltCnt) {
+ if (EltCnt.Scalable)
+ return (EltCnt.Min * 37U) - 1U;
+
+ return EltCnt.Min * 37U;
+ }
+
+ static bool isEqual(const ElementCount& LHS, const ElementCount& RHS) {
+ return LHS == RHS;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_DENSEMAPINFO_H
diff --git a/third_party/llvm-project/include/llvm/ADT/DenseSet.h b/third_party/llvm-project/include/llvm/ADT/DenseSet.h
new file mode 100644
index 000000000..9afb715ae
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/DenseSet.h
@@ -0,0 +1,283 @@
+//===- llvm/ADT/DenseSet.h - Dense probed hash table ------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the DenseSet and SmallDenseSet classes.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_DENSESET_H
+#define LLVM_ADT_DENSESET_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/DenseMapInfo.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/type_traits.h"
+#include <algorithm>
+#include <cstddef>
+#include <initializer_list>
+#include <iterator>
+#include <utility>
+
+namespace llvm {
+
+namespace detail {
+
+struct DenseSetEmpty {};
+
+// Use the empty base class trick so we can create a DenseMap where the buckets
+// contain only a single item.
+template <typename KeyT> class DenseSetPair : public DenseSetEmpty {
+ KeyT key;
+
+public:
+ KeyT &getFirst() { return key; }
+ const KeyT &getFirst() const { return key; }
+ DenseSetEmpty &getSecond() { return *this; }
+ const DenseSetEmpty &getSecond() const { return *this; }
+};
+
+/// Base class for DenseSet and DenseSmallSet.
+///
+/// MapTy should be either
+///
+/// DenseMap<ValueT, detail::DenseSetEmpty, ValueInfoT,
+/// detail::DenseSetPair<ValueT>>
+///
+/// or the equivalent SmallDenseMap type. ValueInfoT must implement the
+/// DenseMapInfo "concept".
+template <typename ValueT, typename MapTy, typename ValueInfoT>
+class DenseSetImpl {
+ static_assert(sizeof(typename MapTy::value_type) == sizeof(ValueT),
+ "DenseMap buckets unexpectedly large!");
+ MapTy TheMap;
+
+ template <typename T>
+ using const_arg_type_t = typename const_pointer_or_const_ref<T>::type;
+
+public:
+ using key_type = ValueT;
+ using value_type = ValueT;
+ using size_type = unsigned;
+
+ explicit DenseSetImpl(unsigned InitialReserve = 0) : TheMap(InitialReserve) {}
+
+ DenseSetImpl(std::initializer_list<ValueT> Elems)
+ : DenseSetImpl(PowerOf2Ceil(Elems.size())) {
+ insert(Elems.begin(), Elems.end());
+ }
+
+ bool empty() const { return TheMap.empty(); }
+ size_type size() const { return TheMap.size(); }
+ size_t getMemorySize() const { return TheMap.getMemorySize(); }
+
+ /// Grow the DenseSet so that it has at least Size buckets. Will not shrink
+ /// the Size of the set.
+ void resize(size_t Size) { TheMap.resize(Size); }
+
+ /// Grow the DenseSet so that it can contain at least \p NumEntries items
+ /// before resizing again.
+ void reserve(size_t Size) { TheMap.reserve(Size); }
+
+ void clear() {
+ TheMap.clear();
+ }
+
+ /// Return 1 if the specified key is in the set, 0 otherwise.
+ size_type count(const_arg_type_t<ValueT> V) const {
+ return TheMap.count(V);
+ }
+
+ bool erase(const ValueT &V) {
+ return TheMap.erase(V);
+ }
+
+ void swap(DenseSetImpl &RHS) { TheMap.swap(RHS.TheMap); }
+
+ // Iterators.
+
+ class ConstIterator;
+
+ class Iterator {
+ typename MapTy::iterator I;
+ friend class DenseSetImpl;
+ friend class ConstIterator;
+
+ public:
+ using difference_type = typename MapTy::iterator::difference_type;
+ using value_type = ValueT;
+ using pointer = value_type *;
+ using reference = value_type &;
+ using iterator_category = std::forward_iterator_tag;
+
+ Iterator() = default;
+ Iterator(const typename MapTy::iterator &i) : I(i) {}
+
+ ValueT &operator*() { return I->getFirst(); }
+ const ValueT &operator*() const { return I->getFirst(); }
+ ValueT *operator->() { return &I->getFirst(); }
+ const ValueT *operator->() const { return &I->getFirst(); }
+
+ Iterator& operator++() { ++I; return *this; }
+ Iterator operator++(int) { auto T = *this; ++I; return T; }
+ bool operator==(const ConstIterator& X) const { return I == X.I; }
+ bool operator!=(const ConstIterator& X) const { return I != X.I; }
+ };
+
+ class ConstIterator {
+ typename MapTy::const_iterator I;
+ friend class DenseSetImpl;
+ friend class Iterator;
+
+ public:
+ using difference_type = typename MapTy::const_iterator::difference_type;
+ using value_type = ValueT;
+ using pointer = const value_type *;
+ using reference = const value_type &;
+ using iterator_category = std::forward_iterator_tag;
+
+ ConstIterator() = default;
+ ConstIterator(const Iterator &B) : I(B.I) {}
+ ConstIterator(const typename MapTy::const_iterator &i) : I(i) {}
+
+ const ValueT &operator*() const { return I->getFirst(); }
+ const ValueT *operator->() const { return &I->getFirst(); }
+
+ ConstIterator& operator++() { ++I; return *this; }
+ ConstIterator operator++(int) { auto T = *this; ++I; return T; }
+ bool operator==(const ConstIterator& X) const { return I == X.I; }
+ bool operator!=(const ConstIterator& X) const { return I != X.I; }
+ };
+
+ using iterator = Iterator;
+ using const_iterator = ConstIterator;
+
+ iterator begin() { return Iterator(TheMap.begin()); }
+ iterator end() { return Iterator(TheMap.end()); }
+
+ const_iterator begin() const { return ConstIterator(TheMap.begin()); }
+ const_iterator end() const { return ConstIterator(TheMap.end()); }
+
+ iterator find(const_arg_type_t<ValueT> V) { return Iterator(TheMap.find(V)); }
+ const_iterator find(const_arg_type_t<ValueT> V) const {
+ return ConstIterator(TheMap.find(V));
+ }
+
+ /// Alternative version of find() which allows a different, and possibly less
+ /// expensive, key type.
+ /// The DenseMapInfo is responsible for supplying methods
+ /// getHashValue(LookupKeyT) and isEqual(LookupKeyT, KeyT) for each key type
+ /// used.
+ template <class LookupKeyT>
+ iterator find_as(const LookupKeyT &Val) {
+ return Iterator(TheMap.find_as(Val));
+ }
+ template <class LookupKeyT>
+ const_iterator find_as(const LookupKeyT &Val) const {
+ return ConstIterator(TheMap.find_as(Val));
+ }
+
+ void erase(Iterator I) { return TheMap.erase(I.I); }
+ void erase(ConstIterator CI) { return TheMap.erase(CI.I); }
+
+ std::pair<iterator, bool> insert(const ValueT &V) {
+ detail::DenseSetEmpty Empty;
+ return TheMap.try_emplace(V, Empty);
+ }
+
+ std::pair<iterator, bool> insert(ValueT &&V) {
+ detail::DenseSetEmpty Empty;
+ return TheMap.try_emplace(std::move(V), Empty);
+ }
+
+ /// Alternative version of insert that uses a different (and possibly less
+ /// expensive) key type.
+ template <typename LookupKeyT>
+ std::pair<iterator, bool> insert_as(const ValueT &V,
+ const LookupKeyT &LookupKey) {
+ return TheMap.insert_as({V, detail::DenseSetEmpty()}, LookupKey);
+ }
+ template <typename LookupKeyT>
+ std::pair<iterator, bool> insert_as(ValueT &&V, const LookupKeyT &LookupKey) {
+ return TheMap.insert_as({std::move(V), detail::DenseSetEmpty()}, LookupKey);
+ }
+
+ // Range insertion of values.
+ template<typename InputIt>
+ void insert(InputIt I, InputIt E) {
+ for (; I != E; ++I)
+ insert(*I);
+ }
+};
+
+/// Equality comparison for DenseSet.
+///
+/// Iterates over elements of LHS confirming that each element is also a member
+/// of RHS, and that RHS contains no additional values.
+/// Equivalent to N calls to RHS.count. Amortized complexity is linear, worst
+/// case is O(N^2) (if every hash collides).
+template <typename ValueT, typename MapTy, typename ValueInfoT>
+bool operator==(const DenseSetImpl<ValueT, MapTy, ValueInfoT> &LHS,
+ const DenseSetImpl<ValueT, MapTy, ValueInfoT> &RHS) {
+ if (LHS.size() != RHS.size())
+ return false;
+
+ for (auto &E : LHS)
+ if (!RHS.count(E))
+ return false;
+
+ return true;
+}
+
+/// Inequality comparison for DenseSet.
+///
+/// Equivalent to !(LHS == RHS). See operator== for performance notes.
+template <typename ValueT, typename MapTy, typename ValueInfoT>
+bool operator!=(const DenseSetImpl<ValueT, MapTy, ValueInfoT> &LHS,
+ const DenseSetImpl<ValueT, MapTy, ValueInfoT> &RHS) {
+ return !(LHS == RHS);
+}
+
+} // end namespace detail
+
+/// Implements a dense probed hash-table based set.
+template <typename ValueT, typename ValueInfoT = DenseMapInfo<ValueT>>
+class DenseSet : public detail::DenseSetImpl<
+ ValueT, DenseMap<ValueT, detail::DenseSetEmpty, ValueInfoT,
+ detail::DenseSetPair<ValueT>>,
+ ValueInfoT> {
+ using BaseT =
+ detail::DenseSetImpl<ValueT,
+ DenseMap<ValueT, detail::DenseSetEmpty, ValueInfoT,
+ detail::DenseSetPair<ValueT>>,
+ ValueInfoT>;
+
+public:
+ using BaseT::BaseT;
+};
+
+/// Implements a dense probed hash-table based set with some number of buckets
+/// stored inline.
+template <typename ValueT, unsigned InlineBuckets = 4,
+ typename ValueInfoT = DenseMapInfo<ValueT>>
+class SmallDenseSet
+ : public detail::DenseSetImpl<
+ ValueT, SmallDenseMap<ValueT, detail::DenseSetEmpty, InlineBuckets,
+ ValueInfoT, detail::DenseSetPair<ValueT>>,
+ ValueInfoT> {
+ using BaseT = detail::DenseSetImpl<
+ ValueT, SmallDenseMap<ValueT, detail::DenseSetEmpty, InlineBuckets,
+ ValueInfoT, detail::DenseSetPair<ValueT>>,
+ ValueInfoT>;
+
+public:
+ using BaseT::BaseT;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_DENSESET_H
diff --git a/third_party/llvm-project/include/llvm/ADT/EpochTracker.h b/third_party/llvm-project/include/llvm/ADT/EpochTracker.h
new file mode 100644
index 000000000..a782b4756
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/EpochTracker.h
@@ -0,0 +1,98 @@
+//===- llvm/ADT/EpochTracker.h - ADT epoch tracking --------------*- C++ -*-==//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the DebugEpochBase and DebugEpochBase::HandleBase classes.
+// These can be used to write iterators that are fail-fast when LLVM is built
+// with asserts enabled.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_EPOCH_TRACKER_H
+#define LLVM_ADT_EPOCH_TRACKER_H
+
+#include "llvm/Config/abi-breaking.h"
+
+#include <cstdint>
+
+namespace llvm {
+
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+
+/// A base class for data structure classes wishing to make iterators
+/// ("handles") pointing into themselves fail-fast. When building without
+/// asserts, this class is empty and does nothing.
+///
+/// DebugEpochBase does not by itself track handles pointing into itself. The
+/// expectation is that routines touching the handles will poll on
+/// isHandleInSync at appropriate points to assert that the handle they're using
+/// is still valid.
+///
+class DebugEpochBase {
+ uint64_t Epoch;
+
+public:
+ DebugEpochBase() : Epoch(0) {}
+
+ /// Calling incrementEpoch invalidates all handles pointing into the
+ /// calling instance.
+ void incrementEpoch() { ++Epoch; }
+
+ /// The destructor calls incrementEpoch to make use-after-free bugs
+ /// more likely to crash deterministically.
+ ~DebugEpochBase() { incrementEpoch(); }
+
+ /// A base class for iterator classes ("handles") that wish to poll for
+ /// iterator invalidating modifications in the underlying data structure.
+ /// When LLVM is built without asserts, this class is empty and does nothing.
+ ///
+ /// HandleBase does not track the parent data structure by itself. It expects
+ /// the routines modifying the data structure to call incrementEpoch when they
+ /// make an iterator-invalidating modification.
+ ///
+ class HandleBase {
+ const uint64_t *EpochAddress;
+ uint64_t EpochAtCreation;
+
+ public:
+ HandleBase() : EpochAddress(nullptr), EpochAtCreation(UINT64_MAX) {}
+
+ explicit HandleBase(const DebugEpochBase *Parent)
+ : EpochAddress(&Parent->Epoch), EpochAtCreation(Parent->Epoch) {}
+
+ /// Returns true if the DebugEpochBase this Handle is linked to has
+ /// not called incrementEpoch on itself since the creation of this
+ /// HandleBase instance.
+ bool isHandleInSync() const { return *EpochAddress == EpochAtCreation; }
+
+ /// Returns a pointer to the epoch word stored in the data structure
+ /// this handle points into. Can be used to check if two iterators point
+ /// into the same data structure.
+ const void *getEpochAddress() const { return EpochAddress; }
+ };
+};
+
+#else
+
+class DebugEpochBase {
+public:
+ void incrementEpoch() {}
+
+ class HandleBase {
+ public:
+ HandleBase() = default;
+ explicit HandleBase(const DebugEpochBase *) {}
+ bool isHandleInSync() const { return true; }
+ const void *getEpochAddress() const { return nullptr; }
+ };
+};
+
+#endif // LLVM_ENABLE_ABI_BREAKING_CHECKS
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/FoldingSet.h b/third_party/llvm-project/include/llvm/ADT/FoldingSet.h
new file mode 100644
index 000000000..d5837e51b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/FoldingSet.h
@@ -0,0 +1,761 @@
+//===- llvm/ADT/FoldingSet.h - Uniquing Hash Set ----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines a hash set that can be used to remove duplication of nodes
+// in a graph. This code was originally created by Chris Lattner for use with
+// SelectionDAGCSEMap, but was isolated to provide use across the llvm code set.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_FOLDINGSET_H
+#define LLVM_ADT_FOLDINGSET_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/Support/Allocator.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+/// This folding set used for two purposes:
+/// 1. Given information about a node we want to create, look up the unique
+/// instance of the node in the set. If the node already exists, return
+/// it, otherwise return the bucket it should be inserted into.
+/// 2. Given a node that has already been created, remove it from the set.
+///
+/// This class is implemented as a single-link chained hash table, where the
+/// "buckets" are actually the nodes themselves (the next pointer is in the
+/// node). The last node points back to the bucket to simplify node removal.
+///
+/// Any node that is to be included in the folding set must be a subclass of
+/// FoldingSetNode. The node class must also define a Profile method used to
+/// establish the unique bits of data for the node. The Profile method is
+/// passed a FoldingSetNodeID object which is used to gather the bits. Just
+/// call one of the Add* functions defined in the FoldingSetBase::NodeID class.
+/// NOTE: That the folding set does not own the nodes and it is the
+/// responsibility of the user to dispose of the nodes.
+///
+/// Eg.
+/// class MyNode : public FoldingSetNode {
+/// private:
+/// std::string Name;
+/// unsigned Value;
+/// public:
+/// MyNode(const char *N, unsigned V) : Name(N), Value(V) {}
+/// ...
+/// void Profile(FoldingSetNodeID &ID) const {
+/// ID.AddString(Name);
+/// ID.AddInteger(Value);
+/// }
+/// ...
+/// };
+///
+/// To define the folding set itself use the FoldingSet template;
+///
+/// Eg.
+/// FoldingSet<MyNode> MyFoldingSet;
+///
+/// Four public methods are available to manipulate the folding set;
+///
+/// 1) If you have an existing node that you want add to the set but unsure
+/// that the node might already exist then call;
+///
+/// MyNode *M = MyFoldingSet.GetOrInsertNode(N);
+///
+/// If The result is equal to the input then the node has been inserted.
+/// Otherwise, the result is the node existing in the folding set, and the
+/// input can be discarded (use the result instead.)
+///
+/// 2) If you are ready to construct a node but want to check if it already
+/// exists, then call FindNodeOrInsertPos with a FoldingSetNodeID of the bits to
+/// check;
+///
+/// FoldingSetNodeID ID;
+/// ID.AddString(Name);
+/// ID.AddInteger(Value);
+/// void *InsertPoint;
+///
+/// MyNode *M = MyFoldingSet.FindNodeOrInsertPos(ID, InsertPoint);
+///
+/// If found then M with be non-NULL, else InsertPoint will point to where it
+/// should be inserted using InsertNode.
+///
+/// 3) If you get a NULL result from FindNodeOrInsertPos then you can as a new
+/// node with FindNodeOrInsertPos;
+///
+/// InsertNode(N, InsertPoint);
+///
+/// 4) Finally, if you want to remove a node from the folding set call;
+///
+/// bool WasRemoved = RemoveNode(N);
+///
+/// The result indicates whether the node existed in the folding set.
+
+class FoldingSetNodeID;
+class StringRef;
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetBase - Implements the folding set functionality. The main
+/// structure is an array of buckets. Each bucket is indexed by the hash of
+/// the nodes it contains. The bucket itself points to the nodes contained
+/// in the bucket via a singly linked list. The last node in the list points
+/// back to the bucket to facilitate node removal.
+///
+class FoldingSetBase {
+ virtual void anchor(); // Out of line virtual method.
+
+protected:
+ /// Buckets - Array of bucket chains.
+ void **Buckets;
+
+ /// NumBuckets - Length of the Buckets array. Always a power of 2.
+ unsigned NumBuckets;
+
+ /// NumNodes - Number of nodes in the folding set. Growth occurs when NumNodes
+ /// is greater than twice the number of buckets.
+ unsigned NumNodes;
+
+ explicit FoldingSetBase(unsigned Log2InitSize = 6);
+ FoldingSetBase(FoldingSetBase &&Arg);
+ FoldingSetBase &operator=(FoldingSetBase &&RHS);
+ ~FoldingSetBase();
+
+public:
+ //===--------------------------------------------------------------------===//
+ /// Node - This class is used to maintain the singly linked bucket list in
+ /// a folding set.
+ class Node {
+ private:
+ // NextInFoldingSetBucket - next link in the bucket list.
+ void *NextInFoldingSetBucket = nullptr;
+
+ public:
+ Node() = default;
+
+ // Accessors
+ void *getNextInBucket() const { return NextInFoldingSetBucket; }
+ void SetNextInBucket(void *N) { NextInFoldingSetBucket = N; }
+ };
+
+ /// clear - Remove all nodes from the folding set.
+ void clear();
+
+ /// size - Returns the number of nodes in the folding set.
+ unsigned size() const { return NumNodes; }
+
+ /// empty - Returns true if there are no nodes in the folding set.
+ bool empty() const { return NumNodes == 0; }
+
+ /// reserve - Increase the number of buckets such that adding the
+ /// EltCount-th node won't cause a rebucket operation. reserve is permitted
+ /// to allocate more space than requested by EltCount.
+ void reserve(unsigned EltCount);
+
+ /// capacity - Returns the number of nodes permitted in the folding set
+ /// before a rebucket operation is performed.
+ unsigned capacity() {
+ // We allow a load factor of up to 2.0,
+ // so that means our capacity is NumBuckets * 2
+ return NumBuckets * 2;
+ }
+
+private:
+ /// GrowHashTable - Double the size of the hash table and rehash everything.
+ void GrowHashTable();
+
+ /// GrowBucketCount - resize the hash table and rehash everything.
+ /// NewBucketCount must be a power of two, and must be greater than the old
+ /// bucket count.
+ void GrowBucketCount(unsigned NewBucketCount);
+
+protected:
+ /// GetNodeProfile - Instantiations of the FoldingSet template implement
+ /// this function to gather data bits for the given node.
+ virtual void GetNodeProfile(Node *N, FoldingSetNodeID &ID) const = 0;
+
+ /// NodeEquals - Instantiations of the FoldingSet template implement
+ /// this function to compare the given node with the given ID.
+ virtual bool NodeEquals(Node *N, const FoldingSetNodeID &ID, unsigned IDHash,
+ FoldingSetNodeID &TempID) const=0;
+
+ /// ComputeNodeHash - Instantiations of the FoldingSet template implement
+ /// this function to compute a hash value for the given node.
+ virtual unsigned ComputeNodeHash(Node *N, FoldingSetNodeID &TempID) const = 0;
+
+ // The below methods are protected to encourage subclasses to provide a more
+ // type-safe API.
+
+ /// RemoveNode - Remove a node from the folding set, returning true if one
+ /// was removed or false if the node was not in the folding set.
+ bool RemoveNode(Node *N);
+
+ /// GetOrInsertNode - If there is an existing simple Node exactly
+ /// equal to the specified node, return it. Otherwise, insert 'N' and return
+ /// it instead.
+ Node *GetOrInsertNode(Node *N);
+
+ /// FindNodeOrInsertPos - Look up the node specified by ID. If it exists,
+ /// return it. If not, return the insertion token that will make insertion
+ /// faster.
+ Node *FindNodeOrInsertPos(const FoldingSetNodeID &ID, void *&InsertPos);
+
+ /// InsertNode - Insert the specified node into the folding set, knowing that
+ /// it is not already in the folding set. InsertPos must be obtained from
+ /// FindNodeOrInsertPos.
+ void InsertNode(Node *N, void *InsertPos);
+};
+
+//===----------------------------------------------------------------------===//
+
+/// DefaultFoldingSetTrait - This class provides default implementations
+/// for FoldingSetTrait implementations.
+template<typename T> struct DefaultFoldingSetTrait {
+ static void Profile(const T &X, FoldingSetNodeID &ID) {
+ X.Profile(ID);
+ }
+ static void Profile(T &X, FoldingSetNodeID &ID) {
+ X.Profile(ID);
+ }
+
+ // Equals - Test if the profile for X would match ID, using TempID
+ // to compute a temporary ID if necessary. The default implementation
+ // just calls Profile and does a regular comparison. Implementations
+ // can override this to provide more efficient implementations.
+ static inline bool Equals(T &X, const FoldingSetNodeID &ID, unsigned IDHash,
+ FoldingSetNodeID &TempID);
+
+ // ComputeHash - Compute a hash value for X, using TempID to
+ // compute a temporary ID if necessary. The default implementation
+ // just calls Profile and does a regular hash computation.
+ // Implementations can override this to provide more efficient
+ // implementations.
+ static inline unsigned ComputeHash(T &X, FoldingSetNodeID &TempID);
+};
+
+/// FoldingSetTrait - This trait class is used to define behavior of how
+/// to "profile" (in the FoldingSet parlance) an object of a given type.
+/// The default behavior is to invoke a 'Profile' method on an object, but
+/// through template specialization the behavior can be tailored for specific
+/// types. Combined with the FoldingSetNodeWrapper class, one can add objects
+/// to FoldingSets that were not originally designed to have that behavior.
+template<typename T> struct FoldingSetTrait
+ : public DefaultFoldingSetTrait<T> {};
+
+/// DefaultContextualFoldingSetTrait - Like DefaultFoldingSetTrait, but
+/// for ContextualFoldingSets.
+template<typename T, typename Ctx>
+struct DefaultContextualFoldingSetTrait {
+ static void Profile(T &X, FoldingSetNodeID &ID, Ctx Context) {
+ X.Profile(ID, Context);
+ }
+
+ static inline bool Equals(T &X, const FoldingSetNodeID &ID, unsigned IDHash,
+ FoldingSetNodeID &TempID, Ctx Context);
+ static inline unsigned ComputeHash(T &X, FoldingSetNodeID &TempID,
+ Ctx Context);
+};
+
+/// ContextualFoldingSetTrait - Like FoldingSetTrait, but for
+/// ContextualFoldingSets.
+template<typename T, typename Ctx> struct ContextualFoldingSetTrait
+ : public DefaultContextualFoldingSetTrait<T, Ctx> {};
+
+//===--------------------------------------------------------------------===//
+/// FoldingSetNodeIDRef - This class describes a reference to an interned
+/// FoldingSetNodeID, which can be a useful to store node id data rather
+/// than using plain FoldingSetNodeIDs, since the 32-element SmallVector
+/// is often much larger than necessary, and the possibility of heap
+/// allocation means it requires a non-trivial destructor call.
+class FoldingSetNodeIDRef {
+ const unsigned *Data = nullptr;
+ size_t Size = 0;
+
+public:
+ FoldingSetNodeIDRef() = default;
+ FoldingSetNodeIDRef(const unsigned *D, size_t S) : Data(D), Size(S) {}
+
+ /// ComputeHash - Compute a strong hash value for this FoldingSetNodeIDRef,
+ /// used to lookup the node in the FoldingSetBase.
+ unsigned ComputeHash() const;
+
+ bool operator==(FoldingSetNodeIDRef) const;
+
+ bool operator!=(FoldingSetNodeIDRef RHS) const { return !(*this == RHS); }
+
+ /// Used to compare the "ordering" of two nodes as defined by the
+ /// profiled bits and their ordering defined by memcmp().
+ bool operator<(FoldingSetNodeIDRef) const;
+
+ const unsigned *getData() const { return Data; }
+ size_t getSize() const { return Size; }
+};
+
+//===--------------------------------------------------------------------===//
+/// FoldingSetNodeID - This class is used to gather all the unique data bits of
+/// a node. When all the bits are gathered this class is used to produce a
+/// hash value for the node.
+class FoldingSetNodeID {
+ /// Bits - Vector of all the data bits that make the node unique.
+ /// Use a SmallVector to avoid a heap allocation in the common case.
+ SmallVector<unsigned, 32> Bits;
+
+public:
+ FoldingSetNodeID() = default;
+
+ FoldingSetNodeID(FoldingSetNodeIDRef Ref)
+ : Bits(Ref.getData(), Ref.getData() + Ref.getSize()) {}
+
+ /// Add* - Add various data types to Bit data.
+ void AddPointer(const void *Ptr);
+ void AddInteger(signed I);
+ void AddInteger(unsigned I);
+ void AddInteger(long I);
+ void AddInteger(unsigned long I);
+ void AddInteger(long long I);
+ void AddInteger(unsigned long long I);
+ void AddBoolean(bool B) { AddInteger(B ? 1U : 0U); }
+ void AddString(StringRef String);
+ void AddNodeID(const FoldingSetNodeID &ID);
+
+ template <typename T>
+ inline void Add(const T &x) { FoldingSetTrait<T>::Profile(x, *this); }
+
+ /// clear - Clear the accumulated profile, allowing this FoldingSetNodeID
+ /// object to be used to compute a new profile.
+ inline void clear() { Bits.clear(); }
+
+ /// ComputeHash - Compute a strong hash value for this FoldingSetNodeID, used
+ /// to lookup the node in the FoldingSetBase.
+ unsigned ComputeHash() const;
+
+ /// operator== - Used to compare two nodes to each other.
+ bool operator==(const FoldingSetNodeID &RHS) const;
+ bool operator==(const FoldingSetNodeIDRef RHS) const;
+
+ bool operator!=(const FoldingSetNodeID &RHS) const { return !(*this == RHS); }
+ bool operator!=(const FoldingSetNodeIDRef RHS) const { return !(*this ==RHS);}
+
+ /// Used to compare the "ordering" of two nodes as defined by the
+ /// profiled bits and their ordering defined by memcmp().
+ bool operator<(const FoldingSetNodeID &RHS) const;
+ bool operator<(const FoldingSetNodeIDRef RHS) const;
+
+ /// Intern - Copy this node's data to a memory region allocated from the
+ /// given allocator and return a FoldingSetNodeIDRef describing the
+ /// interned data.
+ FoldingSetNodeIDRef Intern(BumpPtrAllocator &Allocator) const;
+};
+
+// Convenience type to hide the implementation of the folding set.
+using FoldingSetNode = FoldingSetBase::Node;
+template<class T> class FoldingSetIterator;
+template<class T> class FoldingSetBucketIterator;
+
+// Definitions of FoldingSetTrait and ContextualFoldingSetTrait functions, which
+// require the definition of FoldingSetNodeID.
+template<typename T>
+inline bool
+DefaultFoldingSetTrait<T>::Equals(T &X, const FoldingSetNodeID &ID,
+ unsigned /*IDHash*/,
+ FoldingSetNodeID &TempID) {
+ FoldingSetTrait<T>::Profile(X, TempID);
+ return TempID == ID;
+}
+template<typename T>
+inline unsigned
+DefaultFoldingSetTrait<T>::ComputeHash(T &X, FoldingSetNodeID &TempID) {
+ FoldingSetTrait<T>::Profile(X, TempID);
+ return TempID.ComputeHash();
+}
+template<typename T, typename Ctx>
+inline bool
+DefaultContextualFoldingSetTrait<T, Ctx>::Equals(T &X,
+ const FoldingSetNodeID &ID,
+ unsigned /*IDHash*/,
+ FoldingSetNodeID &TempID,
+ Ctx Context) {
+ ContextualFoldingSetTrait<T, Ctx>::Profile(X, TempID, Context);
+ return TempID == ID;
+}
+template<typename T, typename Ctx>
+inline unsigned
+DefaultContextualFoldingSetTrait<T, Ctx>::ComputeHash(T &X,
+ FoldingSetNodeID &TempID,
+ Ctx Context) {
+ ContextualFoldingSetTrait<T, Ctx>::Profile(X, TempID, Context);
+ return TempID.ComputeHash();
+}
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetImpl - An implementation detail that lets us share code between
+/// FoldingSet and ContextualFoldingSet.
+template <class T> class FoldingSetImpl : public FoldingSetBase {
+protected:
+ explicit FoldingSetImpl(unsigned Log2InitSize)
+ : FoldingSetBase(Log2InitSize) {}
+
+ FoldingSetImpl(FoldingSetImpl &&Arg) = default;
+ FoldingSetImpl &operator=(FoldingSetImpl &&RHS) = default;
+ ~FoldingSetImpl() = default;
+
+public:
+ using iterator = FoldingSetIterator<T>;
+
+ iterator begin() { return iterator(Buckets); }
+ iterator end() { return iterator(Buckets+NumBuckets); }
+
+ using const_iterator = FoldingSetIterator<const T>;
+
+ const_iterator begin() const { return const_iterator(Buckets); }
+ const_iterator end() const { return const_iterator(Buckets+NumBuckets); }
+
+ using bucket_iterator = FoldingSetBucketIterator<T>;
+
+ bucket_iterator bucket_begin(unsigned hash) {
+ return bucket_iterator(Buckets + (hash & (NumBuckets-1)));
+ }
+
+ bucket_iterator bucket_end(unsigned hash) {
+ return bucket_iterator(Buckets + (hash & (NumBuckets-1)), true);
+ }
+
+ /// RemoveNode - Remove a node from the folding set, returning true if one
+ /// was removed or false if the node was not in the folding set.
+ bool RemoveNode(T *N) { return FoldingSetBase::RemoveNode(N); }
+
+ /// GetOrInsertNode - If there is an existing simple Node exactly
+ /// equal to the specified node, return it. Otherwise, insert 'N' and
+ /// return it instead.
+ T *GetOrInsertNode(T *N) {
+ return static_cast<T *>(FoldingSetBase::GetOrInsertNode(N));
+ }
+
+ /// FindNodeOrInsertPos - Look up the node specified by ID. If it exists,
+ /// return it. If not, return the insertion token that will make insertion
+ /// faster.
+ T *FindNodeOrInsertPos(const FoldingSetNodeID &ID, void *&InsertPos) {
+ return static_cast<T *>(FoldingSetBase::FindNodeOrInsertPos(ID, InsertPos));
+ }
+
+ /// InsertNode - Insert the specified node into the folding set, knowing that
+ /// it is not already in the folding set. InsertPos must be obtained from
+ /// FindNodeOrInsertPos.
+ void InsertNode(T *N, void *InsertPos) {
+ FoldingSetBase::InsertNode(N, InsertPos);
+ }
+
+ /// InsertNode - Insert the specified node into the folding set, knowing that
+ /// it is not already in the folding set.
+ void InsertNode(T *N) {
+ T *Inserted = GetOrInsertNode(N);
+ (void)Inserted;
+ assert(Inserted == N && "Node already inserted!");
+ }
+};
+
+//===----------------------------------------------------------------------===//
+/// FoldingSet - This template class is used to instantiate a specialized
+/// implementation of the folding set to the node class T. T must be a
+/// subclass of FoldingSetNode and implement a Profile function.
+///
+/// Note that this set type is movable and move-assignable. However, its
+/// moved-from state is not a valid state for anything other than
+/// move-assigning and destroying. This is primarily to enable movable APIs
+/// that incorporate these objects.
+template <class T> class FoldingSet final : public FoldingSetImpl<T> {
+ using Super = FoldingSetImpl<T>;
+ using Node = typename Super::Node;
+
+ /// GetNodeProfile - Each instantiatation of the FoldingSet needs to provide a
+ /// way to convert nodes into a unique specifier.
+ void GetNodeProfile(Node *N, FoldingSetNodeID &ID) const override {
+ T *TN = static_cast<T *>(N);
+ FoldingSetTrait<T>::Profile(*TN, ID);
+ }
+
+ /// NodeEquals - Instantiations may optionally provide a way to compare a
+ /// node with a specified ID.
+ bool NodeEquals(Node *N, const FoldingSetNodeID &ID, unsigned IDHash,
+ FoldingSetNodeID &TempID) const override {
+ T *TN = static_cast<T *>(N);
+ return FoldingSetTrait<T>::Equals(*TN, ID, IDHash, TempID);
+ }
+
+ /// ComputeNodeHash - Instantiations may optionally provide a way to compute a
+ /// hash value directly from a node.
+ unsigned ComputeNodeHash(Node *N, FoldingSetNodeID &TempID) const override {
+ T *TN = static_cast<T *>(N);
+ return FoldingSetTrait<T>::ComputeHash(*TN, TempID);
+ }
+
+public:
+ explicit FoldingSet(unsigned Log2InitSize = 6) : Super(Log2InitSize) {}
+ FoldingSet(FoldingSet &&Arg) = default;
+ FoldingSet &operator=(FoldingSet &&RHS) = default;
+};
+
+//===----------------------------------------------------------------------===//
+/// ContextualFoldingSet - This template class is a further refinement
+/// of FoldingSet which provides a context argument when calling
+/// Profile on its nodes. Currently, that argument is fixed at
+/// initialization time.
+///
+/// T must be a subclass of FoldingSetNode and implement a Profile
+/// function with signature
+/// void Profile(FoldingSetNodeID &, Ctx);
+template <class T, class Ctx>
+class ContextualFoldingSet final : public FoldingSetImpl<T> {
+ // Unfortunately, this can't derive from FoldingSet<T> because the
+ // construction of the vtable for FoldingSet<T> requires
+ // FoldingSet<T>::GetNodeProfile to be instantiated, which in turn
+ // requires a single-argument T::Profile().
+
+ using Super = FoldingSetImpl<T>;
+ using Node = typename Super::Node;
+
+ Ctx Context;
+
+ /// GetNodeProfile - Each instantiatation of the FoldingSet needs to provide a
+ /// way to convert nodes into a unique specifier.
+ void GetNodeProfile(Node *N, FoldingSetNodeID &ID) const override {
+ T *TN = static_cast<T *>(N);
+ ContextualFoldingSetTrait<T, Ctx>::Profile(*TN, ID, Context);
+ }
+
+ bool NodeEquals(Node *N, const FoldingSetNodeID &ID, unsigned IDHash,
+ FoldingSetNodeID &TempID) const override {
+ T *TN = static_cast<T *>(N);
+ return ContextualFoldingSetTrait<T, Ctx>::Equals(*TN, ID, IDHash, TempID,
+ Context);
+ }
+
+ unsigned ComputeNodeHash(Node *N, FoldingSetNodeID &TempID) const override {
+ T *TN = static_cast<T *>(N);
+ return ContextualFoldingSetTrait<T, Ctx>::ComputeHash(*TN, TempID, Context);
+ }
+
+public:
+ explicit ContextualFoldingSet(Ctx Context, unsigned Log2InitSize = 6)
+ : Super(Log2InitSize), Context(Context) {}
+
+ Ctx getContext() const { return Context; }
+};
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetVector - This template class combines a FoldingSet and a vector
+/// to provide the interface of FoldingSet but with deterministic iteration
+/// order based on the insertion order. T must be a subclass of FoldingSetNode
+/// and implement a Profile function.
+template <class T, class VectorT = SmallVector<T*, 8>>
+class FoldingSetVector {
+ FoldingSet<T> Set;
+ VectorT Vector;
+
+public:
+ explicit FoldingSetVector(unsigned Log2InitSize = 6) : Set(Log2InitSize) {}
+
+ using iterator = pointee_iterator<typename VectorT::iterator>;
+
+ iterator begin() { return Vector.begin(); }
+ iterator end() { return Vector.end(); }
+
+ using const_iterator = pointee_iterator<typename VectorT::const_iterator>;
+
+ const_iterator begin() const { return Vector.begin(); }
+ const_iterator end() const { return Vector.end(); }
+
+ /// clear - Remove all nodes from the folding set.
+ void clear() { Set.clear(); Vector.clear(); }
+
+ /// FindNodeOrInsertPos - Look up the node specified by ID. If it exists,
+ /// return it. If not, return the insertion token that will make insertion
+ /// faster.
+ T *FindNodeOrInsertPos(const FoldingSetNodeID &ID, void *&InsertPos) {
+ return Set.FindNodeOrInsertPos(ID, InsertPos);
+ }
+
+ /// GetOrInsertNode - If there is an existing simple Node exactly
+ /// equal to the specified node, return it. Otherwise, insert 'N' and
+ /// return it instead.
+ T *GetOrInsertNode(T *N) {
+ T *Result = Set.GetOrInsertNode(N);
+ if (Result == N) Vector.push_back(N);
+ return Result;
+ }
+
+ /// InsertNode - Insert the specified node into the folding set, knowing that
+ /// it is not already in the folding set. InsertPos must be obtained from
+ /// FindNodeOrInsertPos.
+ void InsertNode(T *N, void *InsertPos) {
+ Set.InsertNode(N, InsertPos);
+ Vector.push_back(N);
+ }
+
+ /// InsertNode - Insert the specified node into the folding set, knowing that
+ /// it is not already in the folding set.
+ void InsertNode(T *N) {
+ Set.InsertNode(N);
+ Vector.push_back(N);
+ }
+
+ /// size - Returns the number of nodes in the folding set.
+ unsigned size() const { return Set.size(); }
+
+ /// empty - Returns true if there are no nodes in the folding set.
+ bool empty() const { return Set.empty(); }
+};
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetIteratorImpl - This is the common iterator support shared by all
+/// folding sets, which knows how to walk the folding set hash table.
+class FoldingSetIteratorImpl {
+protected:
+ FoldingSetNode *NodePtr;
+
+ FoldingSetIteratorImpl(void **Bucket);
+
+ void advance();
+
+public:
+ bool operator==(const FoldingSetIteratorImpl &RHS) const {
+ return NodePtr == RHS.NodePtr;
+ }
+ bool operator!=(const FoldingSetIteratorImpl &RHS) const {
+ return NodePtr != RHS.NodePtr;
+ }
+};
+
+template <class T> class FoldingSetIterator : public FoldingSetIteratorImpl {
+public:
+ explicit FoldingSetIterator(void **Bucket) : FoldingSetIteratorImpl(Bucket) {}
+
+ T &operator*() const {
+ return *static_cast<T*>(NodePtr);
+ }
+
+ T *operator->() const {
+ return static_cast<T*>(NodePtr);
+ }
+
+ inline FoldingSetIterator &operator++() { // Preincrement
+ advance();
+ return *this;
+ }
+ FoldingSetIterator operator++(int) { // Postincrement
+ FoldingSetIterator tmp = *this; ++*this; return tmp;
+ }
+};
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetBucketIteratorImpl - This is the common bucket iterator support
+/// shared by all folding sets, which knows how to walk a particular bucket
+/// of a folding set hash table.
+class FoldingSetBucketIteratorImpl {
+protected:
+ void *Ptr;
+
+ explicit FoldingSetBucketIteratorImpl(void **Bucket);
+
+ FoldingSetBucketIteratorImpl(void **Bucket, bool) : Ptr(Bucket) {}
+
+ void advance() {
+ void *Probe = static_cast<FoldingSetNode*>(Ptr)->getNextInBucket();
+ uintptr_t x = reinterpret_cast<uintptr_t>(Probe) & ~0x1;
+ Ptr = reinterpret_cast<void*>(x);
+ }
+
+public:
+ bool operator==(const FoldingSetBucketIteratorImpl &RHS) const {
+ return Ptr == RHS.Ptr;
+ }
+ bool operator!=(const FoldingSetBucketIteratorImpl &RHS) const {
+ return Ptr != RHS.Ptr;
+ }
+};
+
+template <class T>
+class FoldingSetBucketIterator : public FoldingSetBucketIteratorImpl {
+public:
+ explicit FoldingSetBucketIterator(void **Bucket) :
+ FoldingSetBucketIteratorImpl(Bucket) {}
+
+ FoldingSetBucketIterator(void **Bucket, bool) :
+ FoldingSetBucketIteratorImpl(Bucket, true) {}
+
+ T &operator*() const { return *static_cast<T*>(Ptr); }
+ T *operator->() const { return static_cast<T*>(Ptr); }
+
+ inline FoldingSetBucketIterator &operator++() { // Preincrement
+ advance();
+ return *this;
+ }
+ FoldingSetBucketIterator operator++(int) { // Postincrement
+ FoldingSetBucketIterator tmp = *this; ++*this; return tmp;
+ }
+};
+
+//===----------------------------------------------------------------------===//
+/// FoldingSetNodeWrapper - This template class is used to "wrap" arbitrary
+/// types in an enclosing object so that they can be inserted into FoldingSets.
+template <typename T>
+class FoldingSetNodeWrapper : public FoldingSetNode {
+ T data;
+
+public:
+ template <typename... Ts>
+ explicit FoldingSetNodeWrapper(Ts &&... Args)
+ : data(std::forward<Ts>(Args)...) {}
+
+ void Profile(FoldingSetNodeID &ID) { FoldingSetTrait<T>::Profile(data, ID); }
+
+ T &getValue() { return data; }
+ const T &getValue() const { return data; }
+
+ operator T&() { return data; }
+ operator const T&() const { return data; }
+};
+
+//===----------------------------------------------------------------------===//
+/// FastFoldingSetNode - This is a subclass of FoldingSetNode which stores
+/// a FoldingSetNodeID value rather than requiring the node to recompute it
+/// each time it is needed. This trades space for speed (which can be
+/// significant if the ID is long), and it also permits nodes to drop
+/// information that would otherwise only be required for recomputing an ID.
+class FastFoldingSetNode : public FoldingSetNode {
+ FoldingSetNodeID FastID;
+
+protected:
+ explicit FastFoldingSetNode(const FoldingSetNodeID &ID) : FastID(ID) {}
+
+public:
+ void Profile(FoldingSetNodeID &ID) const { ID.AddNodeID(FastID); }
+};
+
+//===----------------------------------------------------------------------===//
+// Partial specializations of FoldingSetTrait.
+
+template<typename T> struct FoldingSetTrait<T*> {
+ static inline void Profile(T *X, FoldingSetNodeID &ID) {
+ ID.AddPointer(X);
+ }
+};
+template <typename T1, typename T2>
+struct FoldingSetTrait<std::pair<T1, T2>> {
+ static inline void Profile(const std::pair<T1, T2> &P,
+ FoldingSetNodeID &ID) {
+ ID.Add(P.first);
+ ID.Add(P.second);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_FOLDINGSET_H
diff --git a/third_party/llvm-project/include/llvm/ADT/FunctionExtras.h b/third_party/llvm-project/include/llvm/ADT/FunctionExtras.h
new file mode 100644
index 000000000..121aa527a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/FunctionExtras.h
@@ -0,0 +1,292 @@
+//===- FunctionExtras.h - Function type erasure utilities -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+/// This file provides a collection of function (or more generally, callable)
+/// type erasure utilities supplementing those provided by the standard library
+/// in `<function>`.
+///
+/// It provides `unique_function`, which works like `std::function` but supports
+/// move-only callable objects.
+///
+/// Future plans:
+/// - Add a `function` that provides const, volatile, and ref-qualified support,
+/// which doesn't work with `std::function`.
+/// - Provide support for specifying multiple signatures to type erase callable
+/// objects with an overload set, such as those produced by generic lambdas.
+/// - Expand to include a copyable utility that directly replaces std::function
+/// but brings the above improvements.
+///
+/// Note that LLVM's utilities are greatly simplified by not supporting
+/// allocators.
+///
+/// If the standard library ever begins to provide comparable facilities we can
+/// consider switching to those.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_FUNCTION_EXTRAS_H
+#define LLVM_ADT_FUNCTION_EXTRAS_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/ADT/PointerUnion.h"
+#include "llvm/Support/type_traits.h"
+#include <memory>
+
+namespace llvm {
+
+template <typename FunctionT> class unique_function;
+
+template <typename ReturnT, typename... ParamTs>
+class unique_function<ReturnT(ParamTs...)> {
+ static constexpr size_t InlineStorageSize = sizeof(void *) * 3;
+
+ // MSVC has a bug and ICEs if we give it a particular dependent value
+ // expression as part of the `std::conditional` below. To work around this,
+ // we build that into a template struct's constexpr bool.
+ template <typename T> struct IsSizeLessThanThresholdT {
+ static constexpr bool value = sizeof(T) <= (2 * sizeof(void *));
+ };
+
+ // Provide a type function to map parameters that won't observe extra copies
+ // or moves and which are small enough to likely pass in register to values
+ // and all other types to l-value reference types. We use this to compute the
+ // types used in our erased call utility to minimize copies and moves unless
+ // doing so would force things unnecessarily into memory.
+ //
+ // The heuristic used is related to common ABI register passing conventions.
+ // It doesn't have to be exact though, and in one way it is more strict
+ // because we want to still be able to observe either moves *or* copies.
+ template <typename T>
+ using AdjustedParamT = typename std::conditional<
+ !std::is_reference<T>::value &&
+ llvm::is_trivially_copy_constructible<T>::value &&
+ llvm::is_trivially_move_constructible<T>::value &&
+ IsSizeLessThanThresholdT<T>::value,
+ T, T &>::type;
+
+ // The type of the erased function pointer we use as a callback to dispatch to
+ // the stored callable when it is trivial to move and destroy.
+ using CallPtrT = ReturnT (*)(void *CallableAddr,
+ AdjustedParamT<ParamTs>... Params);
+ using MovePtrT = void (*)(void *LHSCallableAddr, void *RHSCallableAddr);
+ using DestroyPtrT = void (*)(void *CallableAddr);
+
+ /// A struct to hold a single trivial callback with sufficient alignment for
+ /// our bitpacking.
+ struct alignas(8) TrivialCallback {
+ CallPtrT CallPtr;
+ };
+
+ /// A struct we use to aggregate three callbacks when we need full set of
+ /// operations.
+ struct alignas(8) NonTrivialCallbacks {
+ CallPtrT CallPtr;
+ MovePtrT MovePtr;
+ DestroyPtrT DestroyPtr;
+ };
+
+ // Create a pointer union between either a pointer to a static trivial call
+ // pointer in a struct or a pointer to a static struct of the call, move, and
+ // destroy pointers.
+ using CallbackPointerUnionT =
+ PointerUnion<TrivialCallback *, NonTrivialCallbacks *>;
+
+ // The main storage buffer. This will either have a pointer to out-of-line
+ // storage or an inline buffer storing the callable.
+ union StorageUnionT {
+ // For out-of-line storage we keep a pointer to the underlying storage and
+ // the size. This is enough to deallocate the memory.
+ struct OutOfLineStorageT {
+ void *StoragePtr;
+ size_t Size;
+ size_t Alignment;
+ } OutOfLineStorage;
+ static_assert(
+ sizeof(OutOfLineStorageT) <= InlineStorageSize,
+ "Should always use all of the out-of-line storage for inline storage!");
+
+ // For in-line storage, we just provide an aligned character buffer. We
+ // provide three pointers worth of storage here.
+ typename std::aligned_storage<InlineStorageSize, alignof(void *)>::type
+ InlineStorage;
+ } StorageUnion;
+
+ // A compressed pointer to either our dispatching callback or our table of
+ // dispatching callbacks and the flag for whether the callable itself is
+ // stored inline or not.
+ PointerIntPair<CallbackPointerUnionT, 1, bool> CallbackAndInlineFlag;
+
+ bool isInlineStorage() const { return CallbackAndInlineFlag.getInt(); }
+
+ bool isTrivialCallback() const {
+ return CallbackAndInlineFlag.getPointer().template is<TrivialCallback *>();
+ }
+
+ CallPtrT getTrivialCallback() const {
+ return CallbackAndInlineFlag.getPointer().template get<TrivialCallback *>()->CallPtr;
+ }
+
+ NonTrivialCallbacks *getNonTrivialCallbacks() const {
+ return CallbackAndInlineFlag.getPointer()
+ .template get<NonTrivialCallbacks *>();
+ }
+
+ void *getInlineStorage() { return &StorageUnion.InlineStorage; }
+
+ void *getOutOfLineStorage() {
+ return StorageUnion.OutOfLineStorage.StoragePtr;
+ }
+ size_t getOutOfLineStorageSize() const {
+ return StorageUnion.OutOfLineStorage.Size;
+ }
+ size_t getOutOfLineStorageAlignment() const {
+ return StorageUnion.OutOfLineStorage.Alignment;
+ }
+
+ void setOutOfLineStorage(void *Ptr, size_t Size, size_t Alignment) {
+ StorageUnion.OutOfLineStorage = {Ptr, Size, Alignment};
+ }
+
+ template <typename CallableT>
+ static ReturnT CallImpl(void *CallableAddr, AdjustedParamT<ParamTs>... Params) {
+ return (*reinterpret_cast<CallableT *>(CallableAddr))(
+ std::forward<ParamTs>(Params)...);
+ }
+
+ template <typename CallableT>
+ static void MoveImpl(void *LHSCallableAddr, void *RHSCallableAddr) noexcept {
+ new (LHSCallableAddr)
+ CallableT(std::move(*reinterpret_cast<CallableT *>(RHSCallableAddr)));
+ }
+
+ template <typename CallableT>
+ static void DestroyImpl(void *CallableAddr) noexcept {
+ reinterpret_cast<CallableT *>(CallableAddr)->~CallableT();
+ }
+
+public:
+ unique_function() = default;
+ unique_function(std::nullptr_t /*null_callable*/) {}
+
+ ~unique_function() {
+ if (!CallbackAndInlineFlag.getPointer())
+ return;
+
+ // Cache this value so we don't re-check it after type-erased operations.
+ bool IsInlineStorage = isInlineStorage();
+
+ if (!isTrivialCallback())
+ getNonTrivialCallbacks()->DestroyPtr(
+ IsInlineStorage ? getInlineStorage() : getOutOfLineStorage());
+
+ if (!IsInlineStorage)
+ deallocate_buffer(getOutOfLineStorage(), getOutOfLineStorageSize(),
+ getOutOfLineStorageAlignment());
+ }
+
+ unique_function(unique_function &&RHS) noexcept {
+ // Copy the callback and inline flag.
+ CallbackAndInlineFlag = RHS.CallbackAndInlineFlag;
+
+ // If the RHS is empty, just copying the above is sufficient.
+ if (!RHS)
+ return;
+
+ if (!isInlineStorage()) {
+ // The out-of-line case is easiest to move.
+ StorageUnion.OutOfLineStorage = RHS.StorageUnion.OutOfLineStorage;
+ } else if (isTrivialCallback()) {
+ // Move is trivial, just memcpy the bytes across.
+ memcpy(getInlineStorage(), RHS.getInlineStorage(), InlineStorageSize);
+ } else {
+ // Non-trivial move, so dispatch to a type-erased implementation.
+ getNonTrivialCallbacks()->MovePtr(getInlineStorage(),
+ RHS.getInlineStorage());
+ }
+
+ // Clear the old callback and inline flag to get back to as-if-null.
+ RHS.CallbackAndInlineFlag = {};
+
+#ifndef NDEBUG
+ // In debug builds, we also scribble across the rest of the storage.
+ memset(RHS.getInlineStorage(), 0xAD, InlineStorageSize);
+#endif
+ }
+
+ unique_function &operator=(unique_function &&RHS) noexcept {
+ if (this == &RHS)
+ return *this;
+
+ // Because we don't try to provide any exception safety guarantees we can
+ // implement move assignment very simply by first destroying the current
+ // object and then move-constructing over top of it.
+ this->~unique_function();
+ new (this) unique_function(std::move(RHS));
+ return *this;
+ }
+
+ template <typename CallableT> unique_function(CallableT Callable) {
+ bool IsInlineStorage = true;
+ void *CallableAddr = getInlineStorage();
+ if (sizeof(CallableT) > InlineStorageSize ||
+ alignof(CallableT) > alignof(decltype(StorageUnion.InlineStorage))) {
+ IsInlineStorage = false;
+ // Allocate out-of-line storage. FIXME: Use an explicit alignment
+ // parameter in C++17 mode.
+ auto Size = sizeof(CallableT);
+ auto Alignment = alignof(CallableT);
+ CallableAddr = allocate_buffer(Size, Alignment);
+ setOutOfLineStorage(CallableAddr, Size, Alignment);
+ }
+
+ // Now move into the storage.
+ new (CallableAddr) CallableT(std::move(Callable));
+
+ // See if we can create a trivial callback. We need the callable to be
+ // trivially moved and trivially destroyed so that we don't have to store
+ // type erased callbacks for those operations.
+ //
+ // FIXME: We should use constexpr if here and below to avoid instantiating
+ // the non-trivial static objects when unnecessary. While the linker should
+ // remove them, it is still wasteful.
+ if (llvm::is_trivially_move_constructible<CallableT>::value &&
+ std::is_trivially_destructible<CallableT>::value) {
+ // We need to create a nicely aligned object. We use a static variable
+ // for this because it is a trivial struct.
+ static TrivialCallback Callback = { &CallImpl<CallableT> };
+
+ CallbackAndInlineFlag = {&Callback, IsInlineStorage};
+ return;
+ }
+
+ // Otherwise, we need to point at an object that contains all the different
+ // type erased behaviors needed. Create a static instance of the struct type
+ // here and then use a pointer to that.
+ static NonTrivialCallbacks Callbacks = {
+ &CallImpl<CallableT>, &MoveImpl<CallableT>, &DestroyImpl<CallableT>};
+
+ CallbackAndInlineFlag = {&Callbacks, IsInlineStorage};
+ }
+
+ ReturnT operator()(ParamTs... Params) {
+ void *CallableAddr =
+ isInlineStorage() ? getInlineStorage() : getOutOfLineStorage();
+
+ return (isTrivialCallback()
+ ? getTrivialCallback()
+ : getNonTrivialCallbacks()->CallPtr)(CallableAddr, Params...);
+ }
+
+ explicit operator bool() const {
+ return (bool)CallbackAndInlineFlag.getPointer();
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_FUNCTION_H
diff --git a/third_party/llvm-project/include/llvm/ADT/Hashing.h b/third_party/llvm-project/include/llvm/ADT/Hashing.h
new file mode 100644
index 000000000..adcc5cf54
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/Hashing.h
@@ -0,0 +1,659 @@
+//===-- llvm/ADT/Hashing.h - Utilities for hashing --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the newly proposed standard C++ interfaces for hashing
+// arbitrary data and building hash functions for user-defined types. This
+// interface was originally proposed in N3333[1] and is currently under review
+// for inclusion in a future TR and/or standard.
+//
+// The primary interfaces provide are comprised of one type and three functions:
+//
+// -- 'hash_code' class is an opaque type representing the hash code for some
+// data. It is the intended product of hashing, and can be used to implement
+// hash tables, checksumming, and other common uses of hashes. It is not an
+// integer type (although it can be converted to one) because it is risky
+// to assume much about the internals of a hash_code. In particular, each
+// execution of the program has a high probability of producing a different
+// hash_code for a given input. Thus their values are not stable to save or
+// persist, and should only be used during the execution for the
+// construction of hashing datastructures.
+//
+// -- 'hash_value' is a function designed to be overloaded for each
+// user-defined type which wishes to be used within a hashing context. It
+// should be overloaded within the user-defined type's namespace and found
+// via ADL. Overloads for primitive types are provided by this library.
+//
+// -- 'hash_combine' and 'hash_combine_range' are functions designed to aid
+// programmers in easily and intuitively combining a set of data into
+// a single hash_code for their object. They should only logically be used
+// within the implementation of a 'hash_value' routine or similar context.
+//
+// Note that 'hash_combine_range' contains very special logic for hashing
+// a contiguous array of integers or pointers. This logic is *extremely* fast,
+// on a modern Intel "Gainestown" Xeon (Nehalem uarch) @2.2 GHz, these were
+// benchmarked at over 6.5 GiB/s for large keys, and <20 cycles/hash for keys
+// under 32-bytes.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_HASHING_H
+#define LLVM_ADT_HASHING_H
+
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/SwapByteOrder.h"
+#include "llvm/Support/type_traits.h"
+#include <algorithm>
+#include <cassert>
+#include <cstring>
+#include <string>
+#include <utility>
+
+namespace llvm {
+
+/// An opaque object representing a hash code.
+///
+/// This object represents the result of hashing some entity. It is intended to
+/// be used to implement hashtables or other hashing-based data structures.
+/// While it wraps and exposes a numeric value, this value should not be
+/// trusted to be stable or predictable across processes or executions.
+///
+/// In order to obtain the hash_code for an object 'x':
+/// \code
+/// using llvm::hash_value;
+/// llvm::hash_code code = hash_value(x);
+/// \endcode
+class hash_code {
+ size_t value;
+
+public:
+ /// Default construct a hash_code.
+ /// Note that this leaves the value uninitialized.
+ hash_code() = default;
+
+ /// Form a hash code directly from a numerical value.
+ hash_code(size_t value) : value(value) {}
+
+ /// Convert the hash code to its numerical value for use.
+ /*explicit*/ operator size_t() const { return value; }
+
+ friend bool operator==(const hash_code &lhs, const hash_code &rhs) {
+ return lhs.value == rhs.value;
+ }
+ friend bool operator!=(const hash_code &lhs, const hash_code &rhs) {
+ return lhs.value != rhs.value;
+ }
+
+ /// Allow a hash_code to be directly run through hash_value.
+ friend size_t hash_value(const hash_code &code) { return code.value; }
+};
+
+/// Compute a hash_code for any integer value.
+///
+/// Note that this function is intended to compute the same hash_code for
+/// a particular value without regard to the pre-promotion type. This is in
+/// contrast to hash_combine which may produce different hash_codes for
+/// differing argument types even if they would implicit promote to a common
+/// type without changing the value.
+template <typename T>
+typename std::enable_if<is_integral_or_enum<T>::value, hash_code>::type
+hash_value(T value);
+
+/// Compute a hash_code for a pointer's address.
+///
+/// N.B.: This hashes the *address*. Not the value and not the type.
+template <typename T> hash_code hash_value(const T *ptr);
+
+/// Compute a hash_code for a pair of objects.
+template <typename T, typename U>
+hash_code hash_value(const std::pair<T, U> &arg);
+
+/// Compute a hash_code for a standard string.
+template <typename T>
+hash_code hash_value(const std::basic_string<T> &arg);
+
+
+/// Override the execution seed with a fixed value.
+///
+/// This hashing library uses a per-execution seed designed to change on each
+/// run with high probability in order to ensure that the hash codes are not
+/// attackable and to ensure that output which is intended to be stable does
+/// not rely on the particulars of the hash codes produced.
+///
+/// That said, there are use cases where it is important to be able to
+/// reproduce *exactly* a specific behavior. To that end, we provide a function
+/// which will forcibly set the seed to a fixed value. This must be done at the
+/// start of the program, before any hashes are computed. Also, it cannot be
+/// undone. This makes it thread-hostile and very hard to use outside of
+/// immediately on start of a simple program designed for reproducible
+/// behavior.
+void set_fixed_execution_hash_seed(uint64_t fixed_value);
+
+
+// All of the implementation details of actually computing the various hash
+// code values are held within this namespace. These routines are included in
+// the header file mainly to allow inlining and constant propagation.
+namespace hashing {
+namespace detail {
+
+inline uint64_t fetch64(const char *p) {
+ uint64_t result;
+ memcpy(&result, p, sizeof(result));
+ if (sys::IsBigEndianHost)
+ sys::swapByteOrder(result);
+ return result;
+}
+
+inline uint32_t fetch32(const char *p) {
+ uint32_t result;
+ memcpy(&result, p, sizeof(result));
+ if (sys::IsBigEndianHost)
+ sys::swapByteOrder(result);
+ return result;
+}
+
+/// Some primes between 2^63 and 2^64 for various uses.
+static const uint64_t k0 = 0xc3a5c85c97cb3127ULL;
+static const uint64_t k1 = 0xb492b66fbe98f273ULL;
+static const uint64_t k2 = 0x9ae16a3b2f90404fULL;
+static const uint64_t k3 = 0xc949d7c7509e6557ULL;
+
+/// Bitwise right rotate.
+/// Normally this will compile to a single instruction, especially if the
+/// shift is a manifest constant.
+inline uint64_t rotate(uint64_t val, size_t shift) {
+ // Avoid shifting by 64: doing so yields an undefined result.
+ return shift == 0 ? val : ((val >> shift) | (val << (64 - shift)));
+}
+
+inline uint64_t shift_mix(uint64_t val) {
+ return val ^ (val >> 47);
+}
+
+inline uint64_t hash_16_bytes(uint64_t low, uint64_t high) {
+ // Murmur-inspired hashing.
+ const uint64_t kMul = 0x9ddfea08eb382d69ULL;
+ uint64_t a = (low ^ high) * kMul;
+ a ^= (a >> 47);
+ uint64_t b = (high ^ a) * kMul;
+ b ^= (b >> 47);
+ b *= kMul;
+ return b;
+}
+
+inline uint64_t hash_1to3_bytes(const char *s, size_t len, uint64_t seed) {
+ uint8_t a = s[0];
+ uint8_t b = s[len >> 1];
+ uint8_t c = s[len - 1];
+ uint32_t y = static_cast<uint32_t>(a) + (static_cast<uint32_t>(b) << 8);
+ uint32_t z = static_cast<uint32_t>(len) + (static_cast<uint32_t>(c) << 2);
+ return shift_mix(y * k2 ^ z * k3 ^ seed) * k2;
+}
+
+inline uint64_t hash_4to8_bytes(const char *s, size_t len, uint64_t seed) {
+ uint64_t a = fetch32(s);
+ return hash_16_bytes(len + (a << 3), seed ^ fetch32(s + len - 4));
+}
+
+inline uint64_t hash_9to16_bytes(const char *s, size_t len, uint64_t seed) {
+ uint64_t a = fetch64(s);
+ uint64_t b = fetch64(s + len - 8);
+ return hash_16_bytes(seed ^ a, rotate(b + len, len)) ^ b;
+}
+
+inline uint64_t hash_17to32_bytes(const char *s, size_t len, uint64_t seed) {
+ uint64_t a = fetch64(s) * k1;
+ uint64_t b = fetch64(s + 8);
+ uint64_t c = fetch64(s + len - 8) * k2;
+ uint64_t d = fetch64(s + len - 16) * k0;
+ return hash_16_bytes(rotate(a - b, 43) + rotate(c ^ seed, 30) + d,
+ a + rotate(b ^ k3, 20) - c + len + seed);
+}
+
+inline uint64_t hash_33to64_bytes(const char *s, size_t len, uint64_t seed) {
+ uint64_t z = fetch64(s + 24);
+ uint64_t a = fetch64(s) + (len + fetch64(s + len - 16)) * k0;
+ uint64_t b = rotate(a + z, 52);
+ uint64_t c = rotate(a, 37);
+ a += fetch64(s + 8);
+ c += rotate(a, 7);
+ a += fetch64(s + 16);
+ uint64_t vf = a + z;
+ uint64_t vs = b + rotate(a, 31) + c;
+ a = fetch64(s + 16) + fetch64(s + len - 32);
+ z = fetch64(s + len - 8);
+ b = rotate(a + z, 52);
+ c = rotate(a, 37);
+ a += fetch64(s + len - 24);
+ c += rotate(a, 7);
+ a += fetch64(s + len - 16);
+ uint64_t wf = a + z;
+ uint64_t ws = b + rotate(a, 31) + c;
+ uint64_t r = shift_mix((vf + ws) * k2 + (wf + vs) * k0);
+ return shift_mix((seed ^ (r * k0)) + vs) * k2;
+}
+
+inline uint64_t hash_short(const char *s, size_t length, uint64_t seed) {
+ if (length >= 4 && length <= 8)
+ return hash_4to8_bytes(s, length, seed);
+ if (length > 8 && length <= 16)
+ return hash_9to16_bytes(s, length, seed);
+ if (length > 16 && length <= 32)
+ return hash_17to32_bytes(s, length, seed);
+ if (length > 32)
+ return hash_33to64_bytes(s, length, seed);
+ if (length != 0)
+ return hash_1to3_bytes(s, length, seed);
+
+ return k2 ^ seed;
+}
+
+/// The intermediate state used during hashing.
+/// Currently, the algorithm for computing hash codes is based on CityHash and
+/// keeps 56 bytes of arbitrary state.
+struct hash_state {
+ uint64_t h0 = 0, h1 = 0, h2 = 0, h3 = 0, h4 = 0, h5 = 0, h6 = 0;
+
+ /// Create a new hash_state structure and initialize it based on the
+ /// seed and the first 64-byte chunk.
+ /// This effectively performs the initial mix.
+ static hash_state create(const char *s, uint64_t seed) {
+ hash_state state = {
+ 0, seed, hash_16_bytes(seed, k1), rotate(seed ^ k1, 49),
+ seed * k1, shift_mix(seed), 0 };
+ state.h6 = hash_16_bytes(state.h4, state.h5);
+ state.mix(s);
+ return state;
+ }
+
+ /// Mix 32-bytes from the input sequence into the 16-bytes of 'a'
+ /// and 'b', including whatever is already in 'a' and 'b'.
+ static void mix_32_bytes(const char *s, uint64_t &a, uint64_t &b) {
+ a += fetch64(s);
+ uint64_t c = fetch64(s + 24);
+ b = rotate(b + a + c, 21);
+ uint64_t d = a;
+ a += fetch64(s + 8) + fetch64(s + 16);
+ b += rotate(a, 44) + d;
+ a += c;
+ }
+
+ /// Mix in a 64-byte buffer of data.
+ /// We mix all 64 bytes even when the chunk length is smaller, but we
+ /// record the actual length.
+ void mix(const char *s) {
+ h0 = rotate(h0 + h1 + h3 + fetch64(s + 8), 37) * k1;
+ h1 = rotate(h1 + h4 + fetch64(s + 48), 42) * k1;
+ h0 ^= h6;
+ h1 += h3 + fetch64(s + 40);
+ h2 = rotate(h2 + h5, 33) * k1;
+ h3 = h4 * k1;
+ h4 = h0 + h5;
+ mix_32_bytes(s, h3, h4);
+ h5 = h2 + h6;
+ h6 = h1 + fetch64(s + 16);
+ mix_32_bytes(s + 32, h5, h6);
+ std::swap(h2, h0);
+ }
+
+ /// Compute the final 64-bit hash code value based on the current
+ /// state and the length of bytes hashed.
+ uint64_t finalize(size_t length) {
+ return hash_16_bytes(hash_16_bytes(h3, h5) + shift_mix(h1) * k1 + h2,
+ hash_16_bytes(h4, h6) + shift_mix(length) * k1 + h0);
+ }
+};
+
+
+/// A global, fixed seed-override variable.
+///
+/// This variable can be set using the \see llvm::set_fixed_execution_seed
+/// function. See that function for details. Do not, under any circumstances,
+/// set or read this variable.
+extern uint64_t fixed_seed_override;
+
+inline uint64_t get_execution_seed() {
+ // FIXME: This needs to be a per-execution seed. This is just a placeholder
+ // implementation. Switching to a per-execution seed is likely to flush out
+ // instability bugs and so will happen as its own commit.
+ //
+ // However, if there is a fixed seed override set the first time this is
+ // called, return that instead of the per-execution seed.
+ const uint64_t seed_prime = 0xff51afd7ed558ccdULL;
+ static uint64_t seed = fixed_seed_override ? fixed_seed_override : seed_prime;
+ return seed;
+}
+
+
+/// Trait to indicate whether a type's bits can be hashed directly.
+///
+/// A type trait which is true if we want to combine values for hashing by
+/// reading the underlying data. It is false if values of this type must
+/// first be passed to hash_value, and the resulting hash_codes combined.
+//
+// FIXME: We want to replace is_integral_or_enum and is_pointer here with
+// a predicate which asserts that comparing the underlying storage of two
+// values of the type for equality is equivalent to comparing the two values
+// for equality. For all the platforms we care about, this holds for integers
+// and pointers, but there are platforms where it doesn't and we would like to
+// support user-defined types which happen to satisfy this property.
+template <typename T> struct is_hashable_data
+ : std::integral_constant<bool, ((is_integral_or_enum<T>::value ||
+ std::is_pointer<T>::value) &&
+ 64 % sizeof(T) == 0)> {};
+
+// Special case std::pair to detect when both types are viable and when there
+// is no alignment-derived padding in the pair. This is a bit of a lie because
+// std::pair isn't truly POD, but it's close enough in all reasonable
+// implementations for our use case of hashing the underlying data.
+template <typename T, typename U> struct is_hashable_data<std::pair<T, U> >
+ : std::integral_constant<bool, (is_hashable_data<T>::value &&
+ is_hashable_data<U>::value &&
+ (sizeof(T) + sizeof(U)) ==
+ sizeof(std::pair<T, U>))> {};
+
+/// Helper to get the hashable data representation for a type.
+/// This variant is enabled when the type itself can be used.
+template <typename T>
+typename std::enable_if<is_hashable_data<T>::value, T>::type
+get_hashable_data(const T &value) {
+ return value;
+}
+/// Helper to get the hashable data representation for a type.
+/// This variant is enabled when we must first call hash_value and use the
+/// result as our data.
+template <typename T>
+typename std::enable_if<!is_hashable_data<T>::value, size_t>::type
+get_hashable_data(const T &value) {
+ using ::llvm::hash_value;
+ return hash_value(value);
+}
+
+/// Helper to store data from a value into a buffer and advance the
+/// pointer into that buffer.
+///
+/// This routine first checks whether there is enough space in the provided
+/// buffer, and if not immediately returns false. If there is space, it
+/// copies the underlying bytes of value into the buffer, advances the
+/// buffer_ptr past the copied bytes, and returns true.
+template <typename T>
+bool store_and_advance(char *&buffer_ptr, char *buffer_end, const T& value,
+ size_t offset = 0) {
+ size_t store_size = sizeof(value) - offset;
+ if (buffer_ptr + store_size > buffer_end)
+ return false;
+ const char *value_data = reinterpret_cast<const char *>(&value);
+ memcpy(buffer_ptr, value_data + offset, store_size);
+ buffer_ptr += store_size;
+ return true;
+}
+
+/// Implement the combining of integral values into a hash_code.
+///
+/// This overload is selected when the value type of the iterator is
+/// integral. Rather than computing a hash_code for each object and then
+/// combining them, this (as an optimization) directly combines the integers.
+template <typename InputIteratorT>
+hash_code hash_combine_range_impl(InputIteratorT first, InputIteratorT last) {
+ const uint64_t seed = get_execution_seed();
+ char buffer[64], *buffer_ptr = buffer;
+ char *const buffer_end = std::end(buffer);
+ while (first != last && store_and_advance(buffer_ptr, buffer_end,
+ get_hashable_data(*first)))
+ ++first;
+ if (first == last)
+ return hash_short(buffer, buffer_ptr - buffer, seed);
+ assert(buffer_ptr == buffer_end);
+
+ hash_state state = state.create(buffer, seed);
+ size_t length = 64;
+ while (first != last) {
+ // Fill up the buffer. We don't clear it, which re-mixes the last round
+ // when only a partial 64-byte chunk is left.
+ buffer_ptr = buffer;
+ while (first != last && store_and_advance(buffer_ptr, buffer_end,
+ get_hashable_data(*first)))
+ ++first;
+
+ // Rotate the buffer if we did a partial fill in order to simulate doing
+ // a mix of the last 64-bytes. That is how the algorithm works when we
+ // have a contiguous byte sequence, and we want to emulate that here.
+ std::rotate(buffer, buffer_ptr, buffer_end);
+
+ // Mix this chunk into the current state.
+ state.mix(buffer);
+ length += buffer_ptr - buffer;
+ };
+
+ return state.finalize(length);
+}
+
+/// Implement the combining of integral values into a hash_code.
+///
+/// This overload is selected when the value type of the iterator is integral
+/// and when the input iterator is actually a pointer. Rather than computing
+/// a hash_code for each object and then combining them, this (as an
+/// optimization) directly combines the integers. Also, because the integers
+/// are stored in contiguous memory, this routine avoids copying each value
+/// and directly reads from the underlying memory.
+template <typename ValueT>
+typename std::enable_if<is_hashable_data<ValueT>::value, hash_code>::type
+hash_combine_range_impl(ValueT *first, ValueT *last) {
+ const uint64_t seed = get_execution_seed();
+ const char *s_begin = reinterpret_cast<const char *>(first);
+ const char *s_end = reinterpret_cast<const char *>(last);
+ const size_t length = std::distance(s_begin, s_end);
+ if (length <= 64)
+ return hash_short(s_begin, length, seed);
+
+ const char *s_aligned_end = s_begin + (length & ~63);
+ hash_state state = state.create(s_begin, seed);
+ s_begin += 64;
+ while (s_begin != s_aligned_end) {
+ state.mix(s_begin);
+ s_begin += 64;
+ }
+ if (length & 63)
+ state.mix(s_end - 64);
+
+ return state.finalize(length);
+}
+
+} // namespace detail
+} // namespace hashing
+
+
+/// Compute a hash_code for a sequence of values.
+///
+/// This hashes a sequence of values. It produces the same hash_code as
+/// 'hash_combine(a, b, c, ...)', but can run over arbitrary sized sequences
+/// and is significantly faster given pointers and types which can be hashed as
+/// a sequence of bytes.
+template <typename InputIteratorT>
+hash_code hash_combine_range(InputIteratorT first, InputIteratorT last) {
+ return ::llvm::hashing::detail::hash_combine_range_impl(first, last);
+}
+
+
+// Implementation details for hash_combine.
+namespace hashing {
+namespace detail {
+
+/// Helper class to manage the recursive combining of hash_combine
+/// arguments.
+///
+/// This class exists to manage the state and various calls involved in the
+/// recursive combining of arguments used in hash_combine. It is particularly
+/// useful at minimizing the code in the recursive calls to ease the pain
+/// caused by a lack of variadic functions.
+struct hash_combine_recursive_helper {
+ char buffer[64] = {};
+ hash_state state;
+ const uint64_t seed;
+
+public:
+ /// Construct a recursive hash combining helper.
+ ///
+ /// This sets up the state for a recursive hash combine, including getting
+ /// the seed and buffer setup.
+ hash_combine_recursive_helper()
+ : seed(get_execution_seed()) {}
+
+ /// Combine one chunk of data into the current in-flight hash.
+ ///
+ /// This merges one chunk of data into the hash. First it tries to buffer
+ /// the data. If the buffer is full, it hashes the buffer into its
+ /// hash_state, empties it, and then merges the new chunk in. This also
+ /// handles cases where the data straddles the end of the buffer.
+ template <typename T>
+ char *combine_data(size_t &length, char *buffer_ptr, char *buffer_end, T data) {
+ if (!store_and_advance(buffer_ptr, buffer_end, data)) {
+ // Check for skew which prevents the buffer from being packed, and do
+ // a partial store into the buffer to fill it. This is only a concern
+ // with the variadic combine because that formation can have varying
+ // argument types.
+ size_t partial_store_size = buffer_end - buffer_ptr;
+ memcpy(buffer_ptr, &data, partial_store_size);
+
+ // If the store fails, our buffer is full and ready to hash. We have to
+ // either initialize the hash state (on the first full buffer) or mix
+ // this buffer into the existing hash state. Length tracks the *hashed*
+ // length, not the buffered length.
+ if (length == 0) {
+ state = state.create(buffer, seed);
+ length = 64;
+ } else {
+ // Mix this chunk into the current state and bump length up by 64.
+ state.mix(buffer);
+ length += 64;
+ }
+ // Reset the buffer_ptr to the head of the buffer for the next chunk of
+ // data.
+ buffer_ptr = buffer;
+
+ // Try again to store into the buffer -- this cannot fail as we only
+ // store types smaller than the buffer.
+ if (!store_and_advance(buffer_ptr, buffer_end, data,
+ partial_store_size))
+ llvm_unreachable("buffer smaller than stored type");
+ }
+ return buffer_ptr;
+ }
+
+ /// Recursive, variadic combining method.
+ ///
+ /// This function recurses through each argument, combining that argument
+ /// into a single hash.
+ template <typename T, typename ...Ts>
+ hash_code combine(size_t length, char *buffer_ptr, char *buffer_end,
+ const T &arg, const Ts &...args) {
+ buffer_ptr = combine_data(length, buffer_ptr, buffer_end, get_hashable_data(arg));
+
+ // Recurse to the next argument.
+ return combine(length, buffer_ptr, buffer_end, args...);
+ }
+
+ /// Base case for recursive, variadic combining.
+ ///
+ /// The base case when combining arguments recursively is reached when all
+ /// arguments have been handled. It flushes the remaining buffer and
+ /// constructs a hash_code.
+ hash_code combine(size_t length, char *buffer_ptr, char *buffer_end) {
+ // Check whether the entire set of values fit in the buffer. If so, we'll
+ // use the optimized short hashing routine and skip state entirely.
+ if (length == 0)
+ return hash_short(buffer, buffer_ptr - buffer, seed);
+
+ // Mix the final buffer, rotating it if we did a partial fill in order to
+ // simulate doing a mix of the last 64-bytes. That is how the algorithm
+ // works when we have a contiguous byte sequence, and we want to emulate
+ // that here.
+ std::rotate(buffer, buffer_ptr, buffer_end);
+
+ // Mix this chunk into the current state.
+ state.mix(buffer);
+ length += buffer_ptr - buffer;
+
+ return state.finalize(length);
+ }
+};
+
+} // namespace detail
+} // namespace hashing
+
+/// Combine values into a single hash_code.
+///
+/// This routine accepts a varying number of arguments of any type. It will
+/// attempt to combine them into a single hash_code. For user-defined types it
+/// attempts to call a \see hash_value overload (via ADL) for the type. For
+/// integer and pointer types it directly combines their data into the
+/// resulting hash_code.
+///
+/// The result is suitable for returning from a user's hash_value
+/// *implementation* for their user-defined type. Consumers of a type should
+/// *not* call this routine, they should instead call 'hash_value'.
+template <typename ...Ts> hash_code hash_combine(const Ts &...args) {
+ // Recursively hash each argument using a helper class.
+ ::llvm::hashing::detail::hash_combine_recursive_helper helper;
+ return helper.combine(0, helper.buffer, helper.buffer + 64, args...);
+}
+
+// Implementation details for implementations of hash_value overloads provided
+// here.
+namespace hashing {
+namespace detail {
+
+/// Helper to hash the value of a single integer.
+///
+/// Overloads for smaller integer types are not provided to ensure consistent
+/// behavior in the presence of integral promotions. Essentially,
+/// "hash_value('4')" and "hash_value('0' + 4)" should be the same.
+inline hash_code hash_integer_value(uint64_t value) {
+ // Similar to hash_4to8_bytes but using a seed instead of length.
+ const uint64_t seed = get_execution_seed();
+ const char *s = reinterpret_cast<const char *>(&value);
+ const uint64_t a = fetch32(s);
+ return hash_16_bytes(seed + (a << 3), fetch32(s + 4));
+}
+
+} // namespace detail
+} // namespace hashing
+
+// Declared and documented above, but defined here so that any of the hashing
+// infrastructure is available.
+template <typename T>
+typename std::enable_if<is_integral_or_enum<T>::value, hash_code>::type
+hash_value(T value) {
+ return ::llvm::hashing::detail::hash_integer_value(
+ static_cast<uint64_t>(value));
+}
+
+// Declared and documented above, but defined here so that any of the hashing
+// infrastructure is available.
+template <typename T> hash_code hash_value(const T *ptr) {
+ return ::llvm::hashing::detail::hash_integer_value(
+ reinterpret_cast<uintptr_t>(ptr));
+}
+
+// Declared and documented above, but defined here so that any of the hashing
+// infrastructure is available.
+template <typename T, typename U>
+hash_code hash_value(const std::pair<T, U> &arg) {
+ return hash_combine(arg.first, arg.second);
+}
+
+// Declared and documented above, but defined here so that any of the hashing
+// infrastructure is available.
+template <typename T>
+hash_code hash_value(const std::basic_string<T> &arg) {
+ return hash_combine_range(arg.begin(), arg.end());
+}
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/MapVector.h b/third_party/llvm-project/include/llvm/ADT/MapVector.h
new file mode 100644
index 000000000..1de1124f4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/MapVector.h
@@ -0,0 +1,239 @@
+//===- llvm/ADT/MapVector.h - Map w/ deterministic value order --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a map that provides insertion order iteration. The
+// interface is purposefully minimal. The key is assumed to be cheap to copy
+// and 2 copies are kept, one for indexing in a DenseMap, one for iteration in
+// a std::vector.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_MAPVECTOR_H
+#define LLVM_ADT_MAPVECTOR_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/SmallVector.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <iterator>
+#include <type_traits>
+#include <utility>
+#include <vector>
+
+namespace llvm {
+
+/// This class implements a map that also provides access to all stored values
+/// in a deterministic order. The values are kept in a std::vector and the
+/// mapping is done with DenseMap from Keys to indexes in that vector.
+template<typename KeyT, typename ValueT,
+ typename MapType = DenseMap<KeyT, unsigned>,
+ typename VectorType = std::vector<std::pair<KeyT, ValueT>>>
+class MapVector {
+ MapType Map;
+ VectorType Vector;
+
+ static_assert(
+ std::is_integral<typename MapType::mapped_type>::value,
+ "The mapped_type of the specified Map must be an integral type");
+
+public:
+ using value_type = typename VectorType::value_type;
+ using size_type = typename VectorType::size_type;
+
+ using iterator = typename VectorType::iterator;
+ using const_iterator = typename VectorType::const_iterator;
+ using reverse_iterator = typename VectorType::reverse_iterator;
+ using const_reverse_iterator = typename VectorType::const_reverse_iterator;
+
+ /// Clear the MapVector and return the underlying vector.
+ VectorType takeVector() {
+ Map.clear();
+ return std::move(Vector);
+ }
+
+ size_type size() const { return Vector.size(); }
+
+ /// Grow the MapVector so that it can contain at least \p NumEntries items
+ /// before resizing again.
+ void reserve(size_type NumEntries) {
+ Map.reserve(NumEntries);
+ Vector.reserve(NumEntries);
+ }
+
+ iterator begin() { return Vector.begin(); }
+ const_iterator begin() const { return Vector.begin(); }
+ iterator end() { return Vector.end(); }
+ const_iterator end() const { return Vector.end(); }
+
+ reverse_iterator rbegin() { return Vector.rbegin(); }
+ const_reverse_iterator rbegin() const { return Vector.rbegin(); }
+ reverse_iterator rend() { return Vector.rend(); }
+ const_reverse_iterator rend() const { return Vector.rend(); }
+
+ bool empty() const {
+ return Vector.empty();
+ }
+
+ std::pair<KeyT, ValueT> &front() { return Vector.front(); }
+ const std::pair<KeyT, ValueT> &front() const { return Vector.front(); }
+ std::pair<KeyT, ValueT> &back() { return Vector.back(); }
+ const std::pair<KeyT, ValueT> &back() const { return Vector.back(); }
+
+ void clear() {
+ Map.clear();
+ Vector.clear();
+ }
+
+ void swap(MapVector &RHS) {
+ std::swap(Map, RHS.Map);
+ std::swap(Vector, RHS.Vector);
+ }
+
+ ValueT &operator[](const KeyT &Key) {
+ std::pair<KeyT, typename MapType::mapped_type> Pair = std::make_pair(Key, 0);
+ std::pair<typename MapType::iterator, bool> Result = Map.insert(Pair);
+ auto &I = Result.first->second;
+ if (Result.second) {
+ Vector.push_back(std::make_pair(Key, ValueT()));
+ I = Vector.size() - 1;
+ }
+ return Vector[I].second;
+ }
+
+ // Returns a copy of the value. Only allowed if ValueT is copyable.
+ ValueT lookup(const KeyT &Key) const {
+ static_assert(std::is_copy_constructible<ValueT>::value,
+ "Cannot call lookup() if ValueT is not copyable.");
+ typename MapType::const_iterator Pos = Map.find(Key);
+ return Pos == Map.end()? ValueT() : Vector[Pos->second].second;
+ }
+
+ std::pair<iterator, bool> insert(const std::pair<KeyT, ValueT> &KV) {
+ std::pair<KeyT, typename MapType::mapped_type> Pair = std::make_pair(KV.first, 0);
+ std::pair<typename MapType::iterator, bool> Result = Map.insert(Pair);
+ auto &I = Result.first->second;
+ if (Result.second) {
+ Vector.push_back(std::make_pair(KV.first, KV.second));
+ I = Vector.size() - 1;
+ return std::make_pair(std::prev(end()), true);
+ }
+ return std::make_pair(begin() + I, false);
+ }
+
+ std::pair<iterator, bool> insert(std::pair<KeyT, ValueT> &&KV) {
+ // Copy KV.first into the map, then move it into the vector.
+ std::pair<KeyT, typename MapType::mapped_type> Pair = std::make_pair(KV.first, 0);
+ std::pair<typename MapType::iterator, bool> Result = Map.insert(Pair);
+ auto &I = Result.first->second;
+ if (Result.second) {
+ Vector.push_back(std::move(KV));
+ I = Vector.size() - 1;
+ return std::make_pair(std::prev(end()), true);
+ }
+ return std::make_pair(begin() + I, false);
+ }
+
+ size_type count(const KeyT &Key) const {
+ typename MapType::const_iterator Pos = Map.find(Key);
+ return Pos == Map.end()? 0 : 1;
+ }
+
+ iterator find(const KeyT &Key) {
+ typename MapType::const_iterator Pos = Map.find(Key);
+ return Pos == Map.end()? Vector.end() :
+ (Vector.begin() + Pos->second);
+ }
+
+ const_iterator find(const KeyT &Key) const {
+ typename MapType::const_iterator Pos = Map.find(Key);
+ return Pos == Map.end()? Vector.end() :
+ (Vector.begin() + Pos->second);
+ }
+
+ /// Remove the last element from the vector.
+ void pop_back() {
+ typename MapType::iterator Pos = Map.find(Vector.back().first);
+ Map.erase(Pos);
+ Vector.pop_back();
+ }
+
+ /// Remove the element given by Iterator.
+ ///
+ /// Returns an iterator to the element following the one which was removed,
+ /// which may be end().
+ ///
+ /// \note This is a deceivingly expensive operation (linear time). It's
+ /// usually better to use \a remove_if() if possible.
+ typename VectorType::iterator erase(typename VectorType::iterator Iterator) {
+ Map.erase(Iterator->first);
+ auto Next = Vector.erase(Iterator);
+ if (Next == Vector.end())
+ return Next;
+
+ // Update indices in the map.
+ size_t Index = Next - Vector.begin();
+ for (auto &I : Map) {
+ assert(I.second != Index && "Index was already erased!");
+ if (I.second > Index)
+ --I.second;
+ }
+ return Next;
+ }
+
+ /// Remove all elements with the key value Key.
+ ///
+ /// Returns the number of elements removed.
+ size_type erase(const KeyT &Key) {
+ auto Iterator = find(Key);
+ if (Iterator == end())
+ return 0;
+ erase(Iterator);
+ return 1;
+ }
+
+ /// Remove the elements that match the predicate.
+ ///
+ /// Erase all elements that match \c Pred in a single pass. Takes linear
+ /// time.
+ template <class Predicate> void remove_if(Predicate Pred);
+};
+
+template <typename KeyT, typename ValueT, typename MapType, typename VectorType>
+template <class Function>
+void MapVector<KeyT, ValueT, MapType, VectorType>::remove_if(Function Pred) {
+ auto O = Vector.begin();
+ for (auto I = O, E = Vector.end(); I != E; ++I) {
+ if (Pred(*I)) {
+ // Erase from the map.
+ Map.erase(I->first);
+ continue;
+ }
+
+ if (I != O) {
+ // Move the value and update the index in the map.
+ *O = std::move(*I);
+ Map[O->first] = O - Vector.begin();
+ }
+ ++O;
+ }
+ // Erase trailing entries in the vector.
+ Vector.erase(O, Vector.end());
+}
+
+/// A MapVector that performs no allocations if smaller than a certain
+/// size.
+template <typename KeyT, typename ValueT, unsigned N>
+struct SmallMapVector
+ : MapVector<KeyT, ValueT, SmallDenseMap<KeyT, unsigned, N>,
+ SmallVector<std::pair<KeyT, ValueT>, N>> {
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_MAPVECTOR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/None.h b/third_party/llvm-project/include/llvm/ADT/None.h
new file mode 100644
index 000000000..004ca0ac5
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/None.h
@@ -0,0 +1,26 @@
+//===-- None.h - Simple null value for implicit construction ------*- C++ -*-=//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides None, an enumerator for use in implicit constructors
+// of various (usually templated) types to make such construction more
+// terse.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_NONE_H
+#define LLVM_ADT_NONE_H
+
+namespace llvm {
+/// A simple null object to allow implicit construction of Optional<T>
+/// and similar types without having to spell out the specialization's name.
+// (constant value 1 in an attempt to workaround MSVC build issue... )
+enum class NoneType { None = 1 };
+const NoneType None = NoneType::None;
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/Optional.h b/third_party/llvm-project/include/llvm/ADT/Optional.h
new file mode 100644
index 000000000..b45a74002
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/Optional.h
@@ -0,0 +1,429 @@
+//===- Optional.h - Simple variant for passing optional values --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides Optional, a template class modeled in the spirit of
+// OCaml's 'opt' variant. The idea is to strongly type whether or not
+// a value can be optional.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_OPTIONAL_H
+#define LLVM_ADT_OPTIONAL_H
+
+#include "llvm/ADT/None.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/type_traits.h"
+#include <cassert>
+#include <memory>
+#include <new>
+#include <utility>
+
+namespace llvm {
+
+class raw_ostream;
+
+namespace optional_detail {
+
+struct in_place_t {};
+
+/// Storage for any type.
+template <typename T, bool = is_trivially_copyable<T>::value>
+class OptionalStorage {
+ union {
+ char empty;
+ T value;
+ };
+ bool hasVal;
+
+public:
+ ~OptionalStorage() { reset(); }
+
+ OptionalStorage() noexcept : empty(), hasVal(false) {}
+
+ OptionalStorage(OptionalStorage const &other) : OptionalStorage() {
+ if (other.hasValue()) {
+ emplace(other.value);
+ }
+ }
+ OptionalStorage(OptionalStorage &&other) : OptionalStorage() {
+ if (other.hasValue()) {
+ emplace(std::move(other.value));
+ }
+ }
+
+ template <class... Args>
+ explicit OptionalStorage(in_place_t, Args &&... args)
+ : value(std::forward<Args>(args)...), hasVal(true) {}
+
+ void reset() noexcept {
+ if (hasVal) {
+ value.~T();
+ hasVal = false;
+ }
+ }
+
+ bool hasValue() const noexcept { return hasVal; }
+
+ T &getValue() LLVM_LVALUE_FUNCTION noexcept {
+ assert(hasVal);
+ return value;
+ }
+ T const &getValue() const LLVM_LVALUE_FUNCTION noexcept {
+ assert(hasVal);
+ return value;
+ }
+#if LLVM_HAS_RVALUE_REFERENCE_THIS
+ T &&getValue() && noexcept {
+ assert(hasVal);
+ return std::move(value);
+ }
+#endif
+
+ template <class... Args> void emplace(Args &&... args) {
+ reset();
+ ::new ((void *)std::addressof(value)) T(std::forward<Args>(args)...);
+ hasVal = true;
+ }
+
+ OptionalStorage &operator=(T const &y) {
+ if (hasValue()) {
+ value = y;
+ } else {
+ ::new ((void *)std::addressof(value)) T(y);
+ hasVal = true;
+ }
+ return *this;
+ }
+ OptionalStorage &operator=(T &&y) {
+ if (hasValue()) {
+ value = std::move(y);
+ } else {
+ ::new ((void *)std::addressof(value)) T(std::move(y));
+ hasVal = true;
+ }
+ return *this;
+ }
+
+ OptionalStorage &operator=(OptionalStorage const &other) {
+ if (other.hasValue()) {
+ if (hasValue()) {
+ value = other.value;
+ } else {
+ ::new ((void *)std::addressof(value)) T(other.value);
+ hasVal = true;
+ }
+ } else {
+ reset();
+ }
+ return *this;
+ }
+
+ OptionalStorage &operator=(OptionalStorage &&other) {
+ if (other.hasValue()) {
+ if (hasValue()) {
+ value = std::move(other.value);
+ } else {
+ ::new ((void *)std::addressof(value)) T(std::move(other.value));
+ hasVal = true;
+ }
+ } else {
+ reset();
+ }
+ return *this;
+ }
+};
+
+template <typename T> class OptionalStorage<T, true> {
+ union {
+ char empty;
+ T value;
+ };
+ bool hasVal = false;
+
+public:
+ ~OptionalStorage() = default;
+
+ OptionalStorage() noexcept : empty{} {}
+
+ OptionalStorage(OptionalStorage const &other) = default;
+ OptionalStorage(OptionalStorage &&other) = default;
+
+ OptionalStorage &operator=(OptionalStorage const &other) = default;
+ OptionalStorage &operator=(OptionalStorage &&other) = default;
+
+ template <class... Args>
+ explicit OptionalStorage(in_place_t, Args &&... args)
+ : value(std::forward<Args>(args)...), hasVal(true) {}
+
+ void reset() noexcept {
+ if (hasVal) {
+ value.~T();
+ hasVal = false;
+ }
+ }
+
+ bool hasValue() const noexcept { return hasVal; }
+
+ T &getValue() LLVM_LVALUE_FUNCTION noexcept {
+ assert(hasVal);
+ return value;
+ }
+ T const &getValue() const LLVM_LVALUE_FUNCTION noexcept {
+ assert(hasVal);
+ return value;
+ }
+#if LLVM_HAS_RVALUE_REFERENCE_THIS
+ T &&getValue() && noexcept {
+ assert(hasVal);
+ return std::move(value);
+ }
+#endif
+
+ template <class... Args> void emplace(Args &&... args) {
+ reset();
+ ::new ((void *)std::addressof(value)) T(std::forward<Args>(args)...);
+ hasVal = true;
+ }
+
+ OptionalStorage &operator=(T const &y) {
+ if (hasValue()) {
+ value = y;
+ } else {
+ ::new ((void *)std::addressof(value)) T(y);
+ hasVal = true;
+ }
+ return *this;
+ }
+ OptionalStorage &operator=(T &&y) {
+ if (hasValue()) {
+ value = std::move(y);
+ } else {
+ ::new ((void *)std::addressof(value)) T(std::move(y));
+ hasVal = true;
+ }
+ return *this;
+ }
+};
+
+} // namespace optional_detail
+
+template <typename T> class Optional {
+ optional_detail::OptionalStorage<T> Storage;
+
+public:
+ using value_type = T;
+
+ constexpr Optional() {}
+ constexpr Optional(NoneType) {}
+
+ Optional(const T &y) : Storage(optional_detail::in_place_t{}, y) {}
+ Optional(const Optional &O) = default;
+
+ Optional(T &&y) : Storage(optional_detail::in_place_t{}, std::move(y)) {}
+ Optional(Optional &&O) = default;
+
+ Optional &operator=(T &&y) {
+ Storage = std::move(y);
+ return *this;
+ }
+ Optional &operator=(Optional &&O) = default;
+
+ /// Create a new object by constructing it in place with the given arguments.
+ template <typename... ArgTypes> void emplace(ArgTypes &&... Args) {
+ Storage.emplace(std::forward<ArgTypes>(Args)...);
+ }
+
+ static inline Optional create(const T *y) {
+ return y ? Optional(*y) : Optional();
+ }
+
+ Optional &operator=(const T &y) {
+ Storage = y;
+ return *this;
+ }
+ Optional &operator=(const Optional &O) = default;
+
+ void reset() { Storage.reset(); }
+
+ const T *getPointer() const { return &Storage.getValue(); }
+ T *getPointer() { return &Storage.getValue(); }
+ const T &getValue() const LLVM_LVALUE_FUNCTION { return Storage.getValue(); }
+ T &getValue() LLVM_LVALUE_FUNCTION { return Storage.getValue(); }
+
+ explicit operator bool() const { return hasValue(); }
+ bool hasValue() const { return Storage.hasValue(); }
+ const T *operator->() const { return getPointer(); }
+ T *operator->() { return getPointer(); }
+ const T &operator*() const LLVM_LVALUE_FUNCTION { return getValue(); }
+ T &operator*() LLVM_LVALUE_FUNCTION { return getValue(); }
+
+ template <typename U>
+ constexpr T getValueOr(U &&value) const LLVM_LVALUE_FUNCTION {
+ return hasValue() ? getValue() : std::forward<U>(value);
+ }
+
+#if LLVM_HAS_RVALUE_REFERENCE_THIS
+ T &&getValue() && { return std::move(Storage.getValue()); }
+ T &&operator*() && { return std::move(Storage.getValue()); }
+
+ template <typename U>
+ T getValueOr(U &&value) && {
+ return hasValue() ? std::move(getValue()) : std::forward<U>(value);
+ }
+#endif
+};
+
+template <typename T, typename U>
+bool operator==(const Optional<T> &X, const Optional<U> &Y) {
+ if (X && Y)
+ return *X == *Y;
+ return X.hasValue() == Y.hasValue();
+}
+
+template <typename T, typename U>
+bool operator!=(const Optional<T> &X, const Optional<U> &Y) {
+ return !(X == Y);
+}
+
+template <typename T, typename U>
+bool operator<(const Optional<T> &X, const Optional<U> &Y) {
+ if (X && Y)
+ return *X < *Y;
+ return X.hasValue() < Y.hasValue();
+}
+
+template <typename T, typename U>
+bool operator<=(const Optional<T> &X, const Optional<U> &Y) {
+ return !(Y < X);
+}
+
+template <typename T, typename U>
+bool operator>(const Optional<T> &X, const Optional<U> &Y) {
+ return Y < X;
+}
+
+template <typename T, typename U>
+bool operator>=(const Optional<T> &X, const Optional<U> &Y) {
+ return !(X < Y);
+}
+
+template<typename T>
+bool operator==(const Optional<T> &X, NoneType) {
+ return !X;
+}
+
+template<typename T>
+bool operator==(NoneType, const Optional<T> &X) {
+ return X == None;
+}
+
+template<typename T>
+bool operator!=(const Optional<T> &X, NoneType) {
+ return !(X == None);
+}
+
+template<typename T>
+bool operator!=(NoneType, const Optional<T> &X) {
+ return X != None;
+}
+
+template <typename T> bool operator<(const Optional<T> &X, NoneType) {
+ return false;
+}
+
+template <typename T> bool operator<(NoneType, const Optional<T> &X) {
+ return X.hasValue();
+}
+
+template <typename T> bool operator<=(const Optional<T> &X, NoneType) {
+ return !(None < X);
+}
+
+template <typename T> bool operator<=(NoneType, const Optional<T> &X) {
+ return !(X < None);
+}
+
+template <typename T> bool operator>(const Optional<T> &X, NoneType) {
+ return None < X;
+}
+
+template <typename T> bool operator>(NoneType, const Optional<T> &X) {
+ return X < None;
+}
+
+template <typename T> bool operator>=(const Optional<T> &X, NoneType) {
+ return None <= X;
+}
+
+template <typename T> bool operator>=(NoneType, const Optional<T> &X) {
+ return X <= None;
+}
+
+template <typename T> bool operator==(const Optional<T> &X, const T &Y) {
+ return X && *X == Y;
+}
+
+template <typename T> bool operator==(const T &X, const Optional<T> &Y) {
+ return Y && X == *Y;
+}
+
+template <typename T> bool operator!=(const Optional<T> &X, const T &Y) {
+ return !(X == Y);
+}
+
+template <typename T> bool operator!=(const T &X, const Optional<T> &Y) {
+ return !(X == Y);
+}
+
+template <typename T> bool operator<(const Optional<T> &X, const T &Y) {
+ return !X || *X < Y;
+}
+
+template <typename T> bool operator<(const T &X, const Optional<T> &Y) {
+ return Y && X < *Y;
+}
+
+template <typename T> bool operator<=(const Optional<T> &X, const T &Y) {
+ return !(Y < X);
+}
+
+template <typename T> bool operator<=(const T &X, const Optional<T> &Y) {
+ return !(Y < X);
+}
+
+template <typename T> bool operator>(const Optional<T> &X, const T &Y) {
+ return Y < X;
+}
+
+template <typename T> bool operator>(const T &X, const Optional<T> &Y) {
+ return Y < X;
+}
+
+template <typename T> bool operator>=(const Optional<T> &X, const T &Y) {
+ return !(X < Y);
+}
+
+template <typename T> bool operator>=(const T &X, const Optional<T> &Y) {
+ return !(X < Y);
+}
+
+raw_ostream &operator<<(raw_ostream &OS, NoneType);
+
+template <typename T, typename = decltype(std::declval<raw_ostream &>()
+ << std::declval<const T &>())>
+raw_ostream &operator<<(raw_ostream &OS, const Optional<T> &O) {
+ if (O)
+ OS << *O;
+ else
+ OS << None;
+ return OS;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_OPTIONAL_H
diff --git a/third_party/llvm-project/include/llvm/ADT/PointerIntPair.h b/third_party/llvm-project/include/llvm/ADT/PointerIntPair.h
new file mode 100644
index 000000000..fa6bf1504
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/PointerIntPair.h
@@ -0,0 +1,243 @@
+//===- llvm/ADT/PointerIntPair.h - Pair for pointer and int -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the PointerIntPair class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_POINTERINTPAIR_H
+#define LLVM_ADT_POINTERINTPAIR_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/PointerLikeTypeTraits.h"
+#include "llvm/Support/type_traits.h"
+#include <cassert>
+#include <cstdint>
+#include <limits>
+
+namespace llvm {
+
+template <typename T> struct DenseMapInfo;
+template <typename PointerT, unsigned IntBits, typename PtrTraits>
+struct PointerIntPairInfo;
+
+/// PointerIntPair - This class implements a pair of a pointer and small
+/// integer. It is designed to represent this in the space required by one
+/// pointer by bitmangling the integer into the low part of the pointer. This
+/// can only be done for small integers: typically up to 3 bits, but it depends
+/// on the number of bits available according to PointerLikeTypeTraits for the
+/// type.
+///
+/// Note that PointerIntPair always puts the IntVal part in the highest bits
+/// possible. For example, PointerIntPair<void*, 1, bool> will put the bit for
+/// the bool into bit #2, not bit #0, which allows the low two bits to be used
+/// for something else. For example, this allows:
+/// PointerIntPair<PointerIntPair<void*, 1, bool>, 1, bool>
+/// ... and the two bools will land in different bits.
+template <typename PointerTy, unsigned IntBits, typename IntType = unsigned,
+ typename PtrTraits = PointerLikeTypeTraits<PointerTy>,
+ typename Info = PointerIntPairInfo<PointerTy, IntBits, PtrTraits>>
+class PointerIntPair {
+ // Used by MSVC visualizer and generally helpful for debugging/visualizing.
+ using InfoTy = Info;
+ intptr_t Value = 0;
+
+public:
+ constexpr PointerIntPair() = default;
+
+ PointerIntPair(PointerTy PtrVal, IntType IntVal) {
+ setPointerAndInt(PtrVal, IntVal);
+ }
+
+ explicit PointerIntPair(PointerTy PtrVal) { initWithPointer(PtrVal); }
+
+ PointerTy getPointer() const { return Info::getPointer(Value); }
+
+ IntType getInt() const { return (IntType)Info::getInt(Value); }
+
+ void setPointer(PointerTy PtrVal) LLVM_LVALUE_FUNCTION {
+ Value = Info::updatePointer(Value, PtrVal);
+ }
+
+ void setInt(IntType IntVal) LLVM_LVALUE_FUNCTION {
+ Value = Info::updateInt(Value, static_cast<intptr_t>(IntVal));
+ }
+
+ void initWithPointer(PointerTy PtrVal) LLVM_LVALUE_FUNCTION {
+ Value = Info::updatePointer(0, PtrVal);
+ }
+
+ void setPointerAndInt(PointerTy PtrVal, IntType IntVal) LLVM_LVALUE_FUNCTION {
+ Value = Info::updateInt(Info::updatePointer(0, PtrVal),
+ static_cast<intptr_t>(IntVal));
+ }
+
+ PointerTy const *getAddrOfPointer() const {
+ return const_cast<PointerIntPair *>(this)->getAddrOfPointer();
+ }
+
+ PointerTy *getAddrOfPointer() {
+ assert(Value == reinterpret_cast<intptr_t>(getPointer()) &&
+ "Can only return the address if IntBits is cleared and "
+ "PtrTraits doesn't change the pointer");
+ return reinterpret_cast<PointerTy *>(&Value);
+ }
+
+ void *getOpaqueValue() const { return reinterpret_cast<void *>(Value); }
+
+ void setFromOpaqueValue(void *Val) LLVM_LVALUE_FUNCTION {
+ Value = reinterpret_cast<intptr_t>(Val);
+ }
+
+ static PointerIntPair getFromOpaqueValue(void *V) {
+ PointerIntPair P;
+ P.setFromOpaqueValue(V);
+ return P;
+ }
+
+ // Allow PointerIntPairs to be created from const void * if and only if the
+ // pointer type could be created from a const void *.
+ static PointerIntPair getFromOpaqueValue(const void *V) {
+ (void)PtrTraits::getFromVoidPointer(V);
+ return getFromOpaqueValue(const_cast<void *>(V));
+ }
+
+ bool operator==(const PointerIntPair &RHS) const {
+ return Value == RHS.Value;
+ }
+
+ bool operator!=(const PointerIntPair &RHS) const {
+ return Value != RHS.Value;
+ }
+
+ bool operator<(const PointerIntPair &RHS) const { return Value < RHS.Value; }
+ bool operator>(const PointerIntPair &RHS) const { return Value > RHS.Value; }
+
+ bool operator<=(const PointerIntPair &RHS) const {
+ return Value <= RHS.Value;
+ }
+
+ bool operator>=(const PointerIntPair &RHS) const {
+ return Value >= RHS.Value;
+ }
+};
+
+// Specialize is_trivially_copyable to avoid limitation of llvm::is_trivially_copyable
+// when compiled with gcc 4.9.
+template <typename PointerTy, unsigned IntBits, typename IntType,
+ typename PtrTraits,
+ typename Info>
+struct is_trivially_copyable<PointerIntPair<PointerTy, IntBits, IntType, PtrTraits, Info>> : std::true_type {
+#ifdef HAVE_STD_IS_TRIVIALLY_COPYABLE
+ static_assert(std::is_trivially_copyable<PointerIntPair<PointerTy, IntBits, IntType, PtrTraits, Info>>::value,
+ "inconsistent behavior between llvm:: and std:: implementation of is_trivially_copyable");
+#endif
+};
+
+
+template <typename PointerT, unsigned IntBits, typename PtrTraits>
+struct PointerIntPairInfo {
+ static_assert(PtrTraits::NumLowBitsAvailable <
+ std::numeric_limits<uintptr_t>::digits,
+ "cannot use a pointer type that has all bits free");
+ static_assert(IntBits <= PtrTraits::NumLowBitsAvailable,
+ "PointerIntPair with integer size too large for pointer");
+ enum : uintptr_t {
+ /// PointerBitMask - The bits that come from the pointer.
+ PointerBitMask =
+ ~(uintptr_t)(((intptr_t)1 << PtrTraits::NumLowBitsAvailable) - 1),
+
+ /// IntShift - The number of low bits that we reserve for other uses, and
+ /// keep zero.
+ IntShift = (uintptr_t)PtrTraits::NumLowBitsAvailable - IntBits,
+
+ /// IntMask - This is the unshifted mask for valid bits of the int type.
+ IntMask = (uintptr_t)(((intptr_t)1 << IntBits) - 1),
+
+ // ShiftedIntMask - This is the bits for the integer shifted in place.
+ ShiftedIntMask = (uintptr_t)(IntMask << IntShift)
+ };
+
+ static PointerT getPointer(intptr_t Value) {
+ return PtrTraits::getFromVoidPointer(
+ reinterpret_cast<void *>(Value & PointerBitMask));
+ }
+
+ static intptr_t getInt(intptr_t Value) {
+ return (Value >> IntShift) & IntMask;
+ }
+
+ static intptr_t updatePointer(intptr_t OrigValue, PointerT Ptr) {
+ intptr_t PtrWord =
+ reinterpret_cast<intptr_t>(PtrTraits::getAsVoidPointer(Ptr));
+ assert((PtrWord & ~PointerBitMask) == 0 &&
+ "Pointer is not sufficiently aligned");
+ // Preserve all low bits, just update the pointer.
+ return PtrWord | (OrigValue & ~PointerBitMask);
+ }
+
+ static intptr_t updateInt(intptr_t OrigValue, intptr_t Int) {
+ intptr_t IntWord = static_cast<intptr_t>(Int);
+ assert((IntWord & ~IntMask) == 0 && "Integer too large for field");
+
+ // Preserve all bits other than the ones we are updating.
+ return (OrigValue & ~ShiftedIntMask) | IntWord << IntShift;
+ }
+};
+
+// Provide specialization of DenseMapInfo for PointerIntPair.
+template <typename PointerTy, unsigned IntBits, typename IntType>
+struct DenseMapInfo<PointerIntPair<PointerTy, IntBits, IntType>> {
+ using Ty = PointerIntPair<PointerTy, IntBits, IntType>;
+
+ static Ty getEmptyKey() {
+ uintptr_t Val = static_cast<uintptr_t>(-1);
+ Val <<= PointerLikeTypeTraits<Ty>::NumLowBitsAvailable;
+ return Ty::getFromOpaqueValue(reinterpret_cast<void *>(Val));
+ }
+
+ static Ty getTombstoneKey() {
+ uintptr_t Val = static_cast<uintptr_t>(-2);
+ Val <<= PointerLikeTypeTraits<PointerTy>::NumLowBitsAvailable;
+ return Ty::getFromOpaqueValue(reinterpret_cast<void *>(Val));
+ }
+
+ static unsigned getHashValue(Ty V) {
+ uintptr_t IV = reinterpret_cast<uintptr_t>(V.getOpaqueValue());
+ return unsigned(IV) ^ unsigned(IV >> 9);
+ }
+
+ static bool isEqual(const Ty &LHS, const Ty &RHS) { return LHS == RHS; }
+};
+
+// Teach SmallPtrSet that PointerIntPair is "basically a pointer".
+template <typename PointerTy, unsigned IntBits, typename IntType,
+ typename PtrTraits>
+struct PointerLikeTypeTraits<
+ PointerIntPair<PointerTy, IntBits, IntType, PtrTraits>> {
+ static inline void *
+ getAsVoidPointer(const PointerIntPair<PointerTy, IntBits, IntType> &P) {
+ return P.getOpaqueValue();
+ }
+
+ static inline PointerIntPair<PointerTy, IntBits, IntType>
+ getFromVoidPointer(void *P) {
+ return PointerIntPair<PointerTy, IntBits, IntType>::getFromOpaqueValue(P);
+ }
+
+ static inline PointerIntPair<PointerTy, IntBits, IntType>
+ getFromVoidPointer(const void *P) {
+ return PointerIntPair<PointerTy, IntBits, IntType>::getFromOpaqueValue(P);
+ }
+
+ enum { NumLowBitsAvailable = PtrTraits::NumLowBitsAvailable - IntBits };
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_POINTERINTPAIR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/PointerUnion.h b/third_party/llvm-project/include/llvm/ADT/PointerUnion.h
new file mode 100644
index 000000000..98c905775
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/PointerUnion.h
@@ -0,0 +1,309 @@
+//===- llvm/ADT/PointerUnion.h - Discriminated Union of 2 Ptrs --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the PointerUnion class, which is a discriminated union of
+// pointer types.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_POINTERUNION_H
+#define LLVM_ADT_POINTERUNION_H
+
+#include "llvm/ADT/DenseMapInfo.h"
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/PointerLikeTypeTraits.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+
+namespace llvm {
+
+template <typename T> struct PointerUnionTypeSelectorReturn {
+ using Return = T;
+};
+
+/// Get a type based on whether two types are the same or not.
+///
+/// For:
+///
+/// \code
+/// using Ret = typename PointerUnionTypeSelector<T1, T2, EQ, NE>::Return;
+/// \endcode
+///
+/// Ret will be EQ type if T1 is same as T2 or NE type otherwise.
+template <typename T1, typename T2, typename RET_EQ, typename RET_NE>
+struct PointerUnionTypeSelector {
+ using Return = typename PointerUnionTypeSelectorReturn<RET_NE>::Return;
+};
+
+template <typename T, typename RET_EQ, typename RET_NE>
+struct PointerUnionTypeSelector<T, T, RET_EQ, RET_NE> {
+ using Return = typename PointerUnionTypeSelectorReturn<RET_EQ>::Return;
+};
+
+template <typename T1, typename T2, typename RET_EQ, typename RET_NE>
+struct PointerUnionTypeSelectorReturn<
+ PointerUnionTypeSelector<T1, T2, RET_EQ, RET_NE>> {
+ using Return =
+ typename PointerUnionTypeSelector<T1, T2, RET_EQ, RET_NE>::Return;
+};
+
+namespace pointer_union_detail {
+ /// Determine the number of bits required to store integers with values < n.
+ /// This is ceil(log2(n)).
+ constexpr int bitsRequired(unsigned n) {
+ return n > 1 ? 1 + bitsRequired((n + 1) / 2) : 0;
+ }
+
+ template <typename... Ts> constexpr int lowBitsAvailable() {
+ return std::min<int>({PointerLikeTypeTraits<Ts>::NumLowBitsAvailable...});
+ }
+
+ /// Find the index of a type in a list of types. TypeIndex<T, Us...>::Index
+ /// is the index of T in Us, or sizeof...(Us) if T does not appear in the
+ /// list.
+ template <typename T, typename ...Us> struct TypeIndex;
+ template <typename T, typename ...Us> struct TypeIndex<T, T, Us...> {
+ static constexpr int Index = 0;
+ };
+ template <typename T, typename U, typename... Us>
+ struct TypeIndex<T, U, Us...> {
+ static constexpr int Index = 1 + TypeIndex<T, Us...>::Index;
+ };
+ template <typename T> struct TypeIndex<T> {
+ static constexpr int Index = 0;
+ };
+
+ /// Find the first type in a list of types.
+ template <typename T, typename...> struct GetFirstType {
+ using type = T;
+ };
+
+ /// Provide PointerLikeTypeTraits for void* that is used by PointerUnion
+ /// for the template arguments.
+ template <typename ...PTs> class PointerUnionUIntTraits {
+ public:
+ static inline void *getAsVoidPointer(void *P) { return P; }
+ static inline void *getFromVoidPointer(void *P) { return P; }
+ static constexpr int NumLowBitsAvailable = lowBitsAvailable<PTs...>();
+ };
+
+ /// Implement assigment in terms of construction.
+ template <typename Derived, typename T> struct AssignableFrom {
+ Derived &operator=(T t) {
+ return static_cast<Derived &>(*this) = Derived(t);
+ }
+ };
+
+ template <typename Derived, typename ValTy, int I, typename ...Types>
+ class PointerUnionMembers;
+
+ template <typename Derived, typename ValTy, int I>
+ class PointerUnionMembers<Derived, ValTy, I> {
+ protected:
+ ValTy Val;
+ PointerUnionMembers() = default;
+ PointerUnionMembers(ValTy Val) : Val(Val) {}
+
+ friend struct PointerLikeTypeTraits<Derived>;
+ };
+
+ template <typename Derived, typename ValTy, int I, typename Type,
+ typename ...Types>
+ class PointerUnionMembers<Derived, ValTy, I, Type, Types...>
+ : public PointerUnionMembers<Derived, ValTy, I + 1, Types...> {
+ using Base = PointerUnionMembers<Derived, ValTy, I + 1, Types...>;
+ public:
+ using Base::Base;
+ PointerUnionMembers() = default;
+ PointerUnionMembers(Type V)
+ : Base(ValTy(const_cast<void *>(
+ PointerLikeTypeTraits<Type>::getAsVoidPointer(V)),
+ I)) {}
+
+ using Base::operator=;
+ Derived &operator=(Type V) {
+ this->Val = ValTy(
+ const_cast<void *>(PointerLikeTypeTraits<Type>::getAsVoidPointer(V)),
+ I);
+ return static_cast<Derived &>(*this);
+ };
+ };
+}
+
+/// A discriminated union of two or more pointer types, with the discriminator
+/// in the low bit of the pointer.
+///
+/// This implementation is extremely efficient in space due to leveraging the
+/// low bits of the pointer, while exposing a natural and type-safe API.
+///
+/// Common use patterns would be something like this:
+/// PointerUnion<int*, float*> P;
+/// P = (int*)0;
+/// printf("%d %d", P.is<int*>(), P.is<float*>()); // prints "1 0"
+/// X = P.get<int*>(); // ok.
+/// Y = P.get<float*>(); // runtime assertion failure.
+/// Z = P.get<double*>(); // compile time failure.
+/// P = (float*)0;
+/// Y = P.get<float*>(); // ok.
+/// X = P.get<int*>(); // runtime assertion failure.
+template <typename... PTs>
+class PointerUnion
+ : public pointer_union_detail::PointerUnionMembers<
+ PointerUnion<PTs...>,
+ PointerIntPair<
+ void *, pointer_union_detail::bitsRequired(sizeof...(PTs)), int,
+ pointer_union_detail::PointerUnionUIntTraits<PTs...>>,
+ 0, PTs...> {
+ // The first type is special because we want to directly cast a pointer to a
+ // default-initialized union to a pointer to the first type. But we don't
+ // want PointerUnion to be a 'template <typename First, typename ...Rest>'
+ // because it's much more convenient to have a name for the whole pack. So
+ // split off the first type here.
+ using First = typename pointer_union_detail::GetFirstType<PTs...>::type;
+ using Base = typename PointerUnion::PointerUnionMembers;
+
+public:
+ PointerUnion() = default;
+
+ PointerUnion(std::nullptr_t) : PointerUnion() {}
+ using Base::Base;
+
+ /// Test if the pointer held in the union is null, regardless of
+ /// which type it is.
+ bool isNull() const { return !this->Val.getPointer(); }
+
+ explicit operator bool() const { return !isNull(); }
+
+ /// Test if the Union currently holds the type matching T.
+ template <typename T> int is() const {
+ constexpr int Index = pointer_union_detail::TypeIndex<T, PTs...>::Index;
+ static_assert(Index < sizeof...(PTs),
+ "PointerUnion::is<T> given type not in the union");
+ return this->Val.getInt() == Index;
+ }
+
+ /// Returns the value of the specified pointer type.
+ ///
+ /// If the specified pointer type is incorrect, assert.
+ template <typename T> T get() const {
+ assert(is<T>() && "Invalid accessor called");
+ return PointerLikeTypeTraits<T>::getFromVoidPointer(this->Val.getPointer());
+ }
+
+ /// Returns the current pointer if it is of the specified pointer type,
+ /// otherwises returns null.
+ template <typename T> T dyn_cast() const {
+ if (is<T>())
+ return get<T>();
+ return T();
+ }
+
+ /// If the union is set to the first pointer type get an address pointing to
+ /// it.
+ First const *getAddrOfPtr1() const {
+ return const_cast<PointerUnion *>(this)->getAddrOfPtr1();
+ }
+
+ /// If the union is set to the first pointer type get an address pointing to
+ /// it.
+ First *getAddrOfPtr1() {
+ assert(is<First>() && "Val is not the first pointer");
+ assert(
+ PointerLikeTypeTraits<First>::getAsVoidPointer(get<First>()) ==
+ this->Val.getPointer() &&
+ "Can't get the address because PointerLikeTypeTraits changes the ptr");
+ return const_cast<First *>(
+ reinterpret_cast<const First *>(this->Val.getAddrOfPointer()));
+ }
+
+ /// Assignment from nullptr which just clears the union.
+ const PointerUnion &operator=(std::nullptr_t) {
+ this->Val.initWithPointer(nullptr);
+ return *this;
+ }
+
+ /// Assignment from elements of the union.
+ using Base::operator=;
+
+ void *getOpaqueValue() const { return this->Val.getOpaqueValue(); }
+ static inline PointerUnion getFromOpaqueValue(void *VP) {
+ PointerUnion V;
+ V.Val = decltype(V.Val)::getFromOpaqueValue(VP);
+ return V;
+ }
+};
+
+template <typename ...PTs>
+bool operator==(PointerUnion<PTs...> lhs, PointerUnion<PTs...> rhs) {
+ return lhs.getOpaqueValue() == rhs.getOpaqueValue();
+}
+
+template <typename ...PTs>
+bool operator!=(PointerUnion<PTs...> lhs, PointerUnion<PTs...> rhs) {
+ return lhs.getOpaqueValue() != rhs.getOpaqueValue();
+}
+
+template <typename ...PTs>
+bool operator<(PointerUnion<PTs...> lhs, PointerUnion<PTs...> rhs) {
+ return lhs.getOpaqueValue() < rhs.getOpaqueValue();
+}
+
+// Teach SmallPtrSet that PointerUnion is "basically a pointer", that has
+// # low bits available = min(PT1bits,PT2bits)-1.
+template <typename ...PTs>
+struct PointerLikeTypeTraits<PointerUnion<PTs...>> {
+ static inline void *getAsVoidPointer(const PointerUnion<PTs...> &P) {
+ return P.getOpaqueValue();
+ }
+
+ static inline PointerUnion<PTs...> getFromVoidPointer(void *P) {
+ return PointerUnion<PTs...>::getFromOpaqueValue(P);
+ }
+
+ // The number of bits available are the min of the pointer types minus the
+ // bits needed for the discriminator.
+ static constexpr int NumLowBitsAvailable = PointerLikeTypeTraits<decltype(
+ PointerUnion<PTs...>::Val)>::NumLowBitsAvailable;
+};
+
+/// A pointer union of three pointer types. See documentation for PointerUnion
+/// for usage.
+template <typename PT1, typename PT2, typename PT3>
+using PointerUnion3 = PointerUnion<PT1, PT2, PT3>;
+
+/// A pointer union of four pointer types. See documentation for PointerUnion
+/// for usage.
+template <typename PT1, typename PT2, typename PT3, typename PT4>
+using PointerUnion4 = PointerUnion<PT1, PT2, PT3, PT4>;
+
+// Teach DenseMap how to use PointerUnions as keys.
+template <typename ...PTs> struct DenseMapInfo<PointerUnion<PTs...>> {
+ using Union = PointerUnion<PTs...>;
+ using FirstInfo =
+ DenseMapInfo<typename pointer_union_detail::GetFirstType<PTs...>::type>;
+
+ static inline Union getEmptyKey() { return Union(FirstInfo::getEmptyKey()); }
+
+ static inline Union getTombstoneKey() {
+ return Union(FirstInfo::getTombstoneKey());
+ }
+
+ static unsigned getHashValue(const Union &UnionVal) {
+ intptr_t key = (intptr_t)UnionVal.getOpaqueValue();
+ return DenseMapInfo<intptr_t>::getHashValue(key);
+ }
+
+ static bool isEqual(const Union &LHS, const Union &RHS) {
+ return LHS == RHS;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_POINTERUNION_H
diff --git a/third_party/llvm-project/include/llvm/ADT/STLExtras.h b/third_party/llvm-project/include/llvm/ADT/STLExtras.h
new file mode 100644
index 000000000..274933bc5
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/STLExtras.h
@@ -0,0 +1,1578 @@
+//===- llvm/ADT/STLExtras.h - Useful STL related functions ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains some templates that are useful if you are working with the
+// STL at all.
+//
+// No library is required when using these functions.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_STLEXTRAS_H
+#define LLVM_ADT_STLEXTRAS_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Config/abi-breaking.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstdlib>
+#include <functional>
+#include <initializer_list>
+#include <iterator>
+#include <limits>
+#include <memory>
+#include <tuple>
+#include <type_traits>
+#include <utility>
+
+#ifdef EXPENSIVE_CHECKS
+#include <random> // for std::mt19937
+#endif
+
+namespace llvm {
+
+// Only used by compiler if both template types are the same. Useful when
+// using SFINAE to test for the existence of member functions.
+template <typename T, T> struct SameType;
+
+namespace detail {
+
+template <typename RangeT>
+using IterOfRange = decltype(std::begin(std::declval<RangeT &>()));
+
+template <typename RangeT>
+using ValueOfRange = typename std::remove_reference<decltype(
+ *std::begin(std::declval<RangeT &>()))>::type;
+
+} // end namespace detail
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <type_traits>
+//===----------------------------------------------------------------------===//
+
+template <typename T>
+struct negation : std::integral_constant<bool, !bool(T::value)> {};
+
+template <typename...> struct conjunction : std::true_type {};
+template <typename B1> struct conjunction<B1> : B1 {};
+template <typename B1, typename... Bn>
+struct conjunction<B1, Bn...>
+ : std::conditional<bool(B1::value), conjunction<Bn...>, B1>::type {};
+
+template <typename T> struct make_const_ptr {
+ using type =
+ typename std::add_pointer<typename std::add_const<T>::type>::type;
+};
+
+template <typename T> struct make_const_ref {
+ using type = typename std::add_lvalue_reference<
+ typename std::add_const<T>::type>::type;
+};
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <functional>
+//===----------------------------------------------------------------------===//
+
+template <class Ty> struct identity {
+ using argument_type = Ty;
+
+ Ty &operator()(Ty &self) const {
+ return self;
+ }
+ const Ty &operator()(const Ty &self) const {
+ return self;
+ }
+};
+
+/// An efficient, type-erasing, non-owning reference to a callable. This is
+/// intended for use as the type of a function parameter that is not used
+/// after the function in question returns.
+///
+/// This class does not own the callable, so it is not in general safe to store
+/// a function_ref.
+template<typename Fn> class function_ref;
+
+template<typename Ret, typename ...Params>
+class function_ref<Ret(Params...)> {
+ Ret (*callback)(intptr_t callable, Params ...params) = nullptr;
+ intptr_t callable;
+
+ template<typename Callable>
+ static Ret callback_fn(intptr_t callable, Params ...params) {
+ return (*reinterpret_cast<Callable*>(callable))(
+ std::forward<Params>(params)...);
+ }
+
+public:
+ function_ref() = default;
+ function_ref(std::nullptr_t) {}
+
+ template <typename Callable>
+ function_ref(Callable &&callable,
+ typename std::enable_if<
+ !std::is_same<typename std::remove_reference<Callable>::type,
+ function_ref>::value>::type * = nullptr)
+ : callback(callback_fn<typename std::remove_reference<Callable>::type>),
+ callable(reinterpret_cast<intptr_t>(&callable)) {}
+
+ Ret operator()(Params ...params) const {
+ return callback(callable, std::forward<Params>(params)...);
+ }
+
+ operator bool() const { return callback; }
+};
+
+// deleter - Very very very simple method that is used to invoke operator
+// delete on something. It is used like this:
+//
+// for_each(V.begin(), B.end(), deleter<Interval>);
+template <class T>
+inline void deleter(T *Ptr) {
+ delete Ptr;
+}
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <iterator>
+//===----------------------------------------------------------------------===//
+
+namespace adl_detail {
+
+using std::begin;
+
+template <typename ContainerTy>
+auto adl_begin(ContainerTy &&container)
+ -> decltype(begin(std::forward<ContainerTy>(container))) {
+ return begin(std::forward<ContainerTy>(container));
+}
+
+using std::end;
+
+template <typename ContainerTy>
+auto adl_end(ContainerTy &&container)
+ -> decltype(end(std::forward<ContainerTy>(container))) {
+ return end(std::forward<ContainerTy>(container));
+}
+
+using std::swap;
+
+template <typename T>
+void adl_swap(T &&lhs, T &&rhs) noexcept(noexcept(swap(std::declval<T>(),
+ std::declval<T>()))) {
+ swap(std::forward<T>(lhs), std::forward<T>(rhs));
+}
+
+} // end namespace adl_detail
+
+template <typename ContainerTy>
+auto adl_begin(ContainerTy &&container)
+ -> decltype(adl_detail::adl_begin(std::forward<ContainerTy>(container))) {
+ return adl_detail::adl_begin(std::forward<ContainerTy>(container));
+}
+
+template <typename ContainerTy>
+auto adl_end(ContainerTy &&container)
+ -> decltype(adl_detail::adl_end(std::forward<ContainerTy>(container))) {
+ return adl_detail::adl_end(std::forward<ContainerTy>(container));
+}
+
+template <typename T>
+void adl_swap(T &&lhs, T &&rhs) noexcept(
+ noexcept(adl_detail::adl_swap(std::declval<T>(), std::declval<T>()))) {
+ adl_detail::adl_swap(std::forward<T>(lhs), std::forward<T>(rhs));
+}
+
+/// Test whether \p RangeOrContainer is empty. Similar to C++17 std::empty.
+template <typename T>
+constexpr bool empty(const T &RangeOrContainer) {
+ return adl_begin(RangeOrContainer) == adl_end(RangeOrContainer);
+}
+
+// mapped_iterator - This is a simple iterator adapter that causes a function to
+// be applied whenever operator* is invoked on the iterator.
+
+template <typename ItTy, typename FuncTy,
+ typename FuncReturnTy =
+ decltype(std::declval<FuncTy>()(*std::declval<ItTy>()))>
+class mapped_iterator
+ : public iterator_adaptor_base<
+ mapped_iterator<ItTy, FuncTy>, ItTy,
+ typename std::iterator_traits<ItTy>::iterator_category,
+ typename std::remove_reference<FuncReturnTy>::type> {
+public:
+ mapped_iterator(ItTy U, FuncTy F)
+ : mapped_iterator::iterator_adaptor_base(std::move(U)), F(std::move(F)) {}
+
+ ItTy getCurrent() { return this->I; }
+
+ FuncReturnTy operator*() { return F(*this->I); }
+
+private:
+ FuncTy F;
+};
+
+// map_iterator - Provide a convenient way to create mapped_iterators, just like
+// make_pair is useful for creating pairs...
+template <class ItTy, class FuncTy>
+inline mapped_iterator<ItTy, FuncTy> map_iterator(ItTy I, FuncTy F) {
+ return mapped_iterator<ItTy, FuncTy>(std::move(I), std::move(F));
+}
+
+template <class ContainerTy, class FuncTy>
+auto map_range(ContainerTy &&C, FuncTy F)
+ -> decltype(make_range(map_iterator(C.begin(), F),
+ map_iterator(C.end(), F))) {
+ return make_range(map_iterator(C.begin(), F), map_iterator(C.end(), F));
+}
+
+/// Helper to determine if type T has a member called rbegin().
+template <typename Ty> class has_rbegin_impl {
+ using yes = char[1];
+ using no = char[2];
+
+ template <typename Inner>
+ static yes& test(Inner *I, decltype(I->rbegin()) * = nullptr);
+
+ template <typename>
+ static no& test(...);
+
+public:
+ static const bool value = sizeof(test<Ty>(nullptr)) == sizeof(yes);
+};
+
+/// Metafunction to determine if T& or T has a member called rbegin().
+template <typename Ty>
+struct has_rbegin : has_rbegin_impl<typename std::remove_reference<Ty>::type> {
+};
+
+// Returns an iterator_range over the given container which iterates in reverse.
+// Note that the container must have rbegin()/rend() methods for this to work.
+template <typename ContainerTy>
+auto reverse(ContainerTy &&C,
+ typename std::enable_if<has_rbegin<ContainerTy>::value>::type * =
+ nullptr) -> decltype(make_range(C.rbegin(), C.rend())) {
+ return make_range(C.rbegin(), C.rend());
+}
+
+// Returns a std::reverse_iterator wrapped around the given iterator.
+template <typename IteratorTy>
+std::reverse_iterator<IteratorTy> make_reverse_iterator(IteratorTy It) {
+ return std::reverse_iterator<IteratorTy>(It);
+}
+
+// Returns an iterator_range over the given container which iterates in reverse.
+// Note that the container must have begin()/end() methods which return
+// bidirectional iterators for this to work.
+template <typename ContainerTy>
+auto reverse(
+ ContainerTy &&C,
+ typename std::enable_if<!has_rbegin<ContainerTy>::value>::type * = nullptr)
+ -> decltype(make_range(llvm::make_reverse_iterator(std::end(C)),
+ llvm::make_reverse_iterator(std::begin(C)))) {
+ return make_range(llvm::make_reverse_iterator(std::end(C)),
+ llvm::make_reverse_iterator(std::begin(C)));
+}
+
+/// An iterator adaptor that filters the elements of given inner iterators.
+///
+/// The predicate parameter should be a callable object that accepts the wrapped
+/// iterator's reference type and returns a bool. When incrementing or
+/// decrementing the iterator, it will call the predicate on each element and
+/// skip any where it returns false.
+///
+/// \code
+/// int A[] = { 1, 2, 3, 4 };
+/// auto R = make_filter_range(A, [](int N) { return N % 2 == 1; });
+/// // R contains { 1, 3 }.
+/// \endcode
+///
+/// Note: filter_iterator_base implements support for forward iteration.
+/// filter_iterator_impl exists to provide support for bidirectional iteration,
+/// conditional on whether the wrapped iterator supports it.
+template <typename WrappedIteratorT, typename PredicateT, typename IterTag>
+class filter_iterator_base
+ : public iterator_adaptor_base<
+ filter_iterator_base<WrappedIteratorT, PredicateT, IterTag>,
+ WrappedIteratorT,
+ typename std::common_type<
+ IterTag, typename std::iterator_traits<
+ WrappedIteratorT>::iterator_category>::type> {
+ using BaseT = iterator_adaptor_base<
+ filter_iterator_base<WrappedIteratorT, PredicateT, IterTag>,
+ WrappedIteratorT,
+ typename std::common_type<
+ IterTag, typename std::iterator_traits<
+ WrappedIteratorT>::iterator_category>::type>;
+
+protected:
+ WrappedIteratorT End;
+ PredicateT Pred;
+
+ void findNextValid() {
+ while (this->I != End && !Pred(*this->I))
+ BaseT::operator++();
+ }
+
+ // Construct the iterator. The begin iterator needs to know where the end
+ // is, so that it can properly stop when it gets there. The end iterator only
+ // needs the predicate to support bidirectional iteration.
+ filter_iterator_base(WrappedIteratorT Begin, WrappedIteratorT End,
+ PredicateT Pred)
+ : BaseT(Begin), End(End), Pred(Pred) {
+ findNextValid();
+ }
+
+public:
+ using BaseT::operator++;
+
+ filter_iterator_base &operator++() {
+ BaseT::operator++();
+ findNextValid();
+ return *this;
+ }
+};
+
+/// Specialization of filter_iterator_base for forward iteration only.
+template <typename WrappedIteratorT, typename PredicateT,
+ typename IterTag = std::forward_iterator_tag>
+class filter_iterator_impl
+ : public filter_iterator_base<WrappedIteratorT, PredicateT, IterTag> {
+ using BaseT = filter_iterator_base<WrappedIteratorT, PredicateT, IterTag>;
+
+public:
+ filter_iterator_impl(WrappedIteratorT Begin, WrappedIteratorT End,
+ PredicateT Pred)
+ : BaseT(Begin, End, Pred) {}
+};
+
+/// Specialization of filter_iterator_base for bidirectional iteration.
+template <typename WrappedIteratorT, typename PredicateT>
+class filter_iterator_impl<WrappedIteratorT, PredicateT,
+ std::bidirectional_iterator_tag>
+ : public filter_iterator_base<WrappedIteratorT, PredicateT,
+ std::bidirectional_iterator_tag> {
+ using BaseT = filter_iterator_base<WrappedIteratorT, PredicateT,
+ std::bidirectional_iterator_tag>;
+ void findPrevValid() {
+ while (!this->Pred(*this->I))
+ BaseT::operator--();
+ }
+
+public:
+ using BaseT::operator--;
+
+ filter_iterator_impl(WrappedIteratorT Begin, WrappedIteratorT End,
+ PredicateT Pred)
+ : BaseT(Begin, End, Pred) {}
+
+ filter_iterator_impl &operator--() {
+ BaseT::operator--();
+ findPrevValid();
+ return *this;
+ }
+};
+
+namespace detail {
+
+template <bool is_bidirectional> struct fwd_or_bidi_tag_impl {
+ using type = std::forward_iterator_tag;
+};
+
+template <> struct fwd_or_bidi_tag_impl<true> {
+ using type = std::bidirectional_iterator_tag;
+};
+
+/// Helper which sets its type member to forward_iterator_tag if the category
+/// of \p IterT does not derive from bidirectional_iterator_tag, and to
+/// bidirectional_iterator_tag otherwise.
+template <typename IterT> struct fwd_or_bidi_tag {
+ using type = typename fwd_or_bidi_tag_impl<std::is_base_of<
+ std::bidirectional_iterator_tag,
+ typename std::iterator_traits<IterT>::iterator_category>::value>::type;
+};
+
+} // namespace detail
+
+/// Defines filter_iterator to a suitable specialization of
+/// filter_iterator_impl, based on the underlying iterator's category.
+template <typename WrappedIteratorT, typename PredicateT>
+using filter_iterator = filter_iterator_impl<
+ WrappedIteratorT, PredicateT,
+ typename detail::fwd_or_bidi_tag<WrappedIteratorT>::type>;
+
+/// Convenience function that takes a range of elements and a predicate,
+/// and return a new filter_iterator range.
+///
+/// FIXME: Currently if RangeT && is a rvalue reference to a temporary, the
+/// lifetime of that temporary is not kept by the returned range object, and the
+/// temporary is going to be dropped on the floor after the make_iterator_range
+/// full expression that contains this function call.
+template <typename RangeT, typename PredicateT>
+iterator_range<filter_iterator<detail::IterOfRange<RangeT>, PredicateT>>
+make_filter_range(RangeT &&Range, PredicateT Pred) {
+ using FilterIteratorT =
+ filter_iterator<detail::IterOfRange<RangeT>, PredicateT>;
+ return make_range(
+ FilterIteratorT(std::begin(std::forward<RangeT>(Range)),
+ std::end(std::forward<RangeT>(Range)), Pred),
+ FilterIteratorT(std::end(std::forward<RangeT>(Range)),
+ std::end(std::forward<RangeT>(Range)), Pred));
+}
+
+/// A pseudo-iterator adaptor that is designed to implement "early increment"
+/// style loops.
+///
+/// This is *not a normal iterator* and should almost never be used directly. It
+/// is intended primarily to be used with range based for loops and some range
+/// algorithms.
+///
+/// The iterator isn't quite an `OutputIterator` or an `InputIterator` but
+/// somewhere between them. The constraints of these iterators are:
+///
+/// - On construction or after being incremented, it is comparable and
+/// dereferencable. It is *not* incrementable.
+/// - After being dereferenced, it is neither comparable nor dereferencable, it
+/// is only incrementable.
+///
+/// This means you can only dereference the iterator once, and you can only
+/// increment it once between dereferences.
+template <typename WrappedIteratorT>
+class early_inc_iterator_impl
+ : public iterator_adaptor_base<early_inc_iterator_impl<WrappedIteratorT>,
+ WrappedIteratorT, std::input_iterator_tag> {
+ using BaseT =
+ iterator_adaptor_base<early_inc_iterator_impl<WrappedIteratorT>,
+ WrappedIteratorT, std::input_iterator_tag>;
+
+ using PointerT = typename std::iterator_traits<WrappedIteratorT>::pointer;
+
+protected:
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ bool IsEarlyIncremented = false;
+#endif
+
+public:
+ early_inc_iterator_impl(WrappedIteratorT I) : BaseT(I) {}
+
+ using BaseT::operator*;
+ typename BaseT::reference operator*() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ assert(!IsEarlyIncremented && "Cannot dereference twice!");
+ IsEarlyIncremented = true;
+#endif
+ return *(this->I)++;
+ }
+
+ using BaseT::operator++;
+ early_inc_iterator_impl &operator++() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ assert(IsEarlyIncremented && "Cannot increment before dereferencing!");
+ IsEarlyIncremented = false;
+#endif
+ return *this;
+ }
+
+ using BaseT::operator==;
+ bool operator==(const early_inc_iterator_impl &RHS) const {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ assert(!IsEarlyIncremented && "Cannot compare after dereferencing!");
+#endif
+ return BaseT::operator==(RHS);
+ }
+};
+
+/// Make a range that does early increment to allow mutation of the underlying
+/// range without disrupting iteration.
+///
+/// The underlying iterator will be incremented immediately after it is
+/// dereferenced, allowing deletion of the current node or insertion of nodes to
+/// not disrupt iteration provided they do not invalidate the *next* iterator --
+/// the current iterator can be invalidated.
+///
+/// This requires a very exact pattern of use that is only really suitable to
+/// range based for loops and other range algorithms that explicitly guarantee
+/// to dereference exactly once each element, and to increment exactly once each
+/// element.
+template <typename RangeT>
+iterator_range<early_inc_iterator_impl<detail::IterOfRange<RangeT>>>
+make_early_inc_range(RangeT &&Range) {
+ using EarlyIncIteratorT =
+ early_inc_iterator_impl<detail::IterOfRange<RangeT>>;
+ return make_range(EarlyIncIteratorT(std::begin(std::forward<RangeT>(Range))),
+ EarlyIncIteratorT(std::end(std::forward<RangeT>(Range))));
+}
+
+// forward declarations required by zip_shortest/zip_first/zip_longest
+template <typename R, typename UnaryPredicate>
+bool all_of(R &&range, UnaryPredicate P);
+template <typename R, typename UnaryPredicate>
+bool any_of(R &&range, UnaryPredicate P);
+
+namespace detail {
+
+using std::declval;
+
+// We have to alias this since inlining the actual type at the usage site
+// in the parameter list of iterator_facade_base<> below ICEs MSVC 2017.
+template<typename... Iters> struct ZipTupleType {
+ using type = std::tuple<decltype(*declval<Iters>())...>;
+};
+
+template <typename ZipType, typename... Iters>
+using zip_traits = iterator_facade_base<
+ ZipType, typename std::common_type<std::bidirectional_iterator_tag,
+ typename std::iterator_traits<
+ Iters>::iterator_category...>::type,
+ // ^ TODO: Implement random access methods.
+ typename ZipTupleType<Iters...>::type,
+ typename std::iterator_traits<typename std::tuple_element<
+ 0, std::tuple<Iters...>>::type>::difference_type,
+ // ^ FIXME: This follows boost::make_zip_iterator's assumption that all
+ // inner iterators have the same difference_type. It would fail if, for
+ // instance, the second field's difference_type were non-numeric while the
+ // first is.
+ typename ZipTupleType<Iters...>::type *,
+ typename ZipTupleType<Iters...>::type>;
+
+template <typename ZipType, typename... Iters>
+struct zip_common : public zip_traits<ZipType, Iters...> {
+ using Base = zip_traits<ZipType, Iters...>;
+ using value_type = typename Base::value_type;
+
+ std::tuple<Iters...> iterators;
+
+protected:
+ template <size_t... Ns> value_type deref(std::index_sequence<Ns...>) const {
+ return value_type(*std::get<Ns>(iterators)...);
+ }
+
+ template <size_t... Ns>
+ decltype(iterators) tup_inc(std::index_sequence<Ns...>) const {
+ return std::tuple<Iters...>(std::next(std::get<Ns>(iterators))...);
+ }
+
+ template <size_t... Ns>
+ decltype(iterators) tup_dec(std::index_sequence<Ns...>) const {
+ return std::tuple<Iters...>(std::prev(std::get<Ns>(iterators))...);
+ }
+
+public:
+ zip_common(Iters &&... ts) : iterators(std::forward<Iters>(ts)...) {}
+
+ value_type operator*() { return deref(std::index_sequence_for<Iters...>{}); }
+
+ const value_type operator*() const {
+ return deref(std::index_sequence_for<Iters...>{});
+ }
+
+ ZipType &operator++() {
+ iterators = tup_inc(std::index_sequence_for<Iters...>{});
+ return *reinterpret_cast<ZipType *>(this);
+ }
+
+ ZipType &operator--() {
+ static_assert(Base::IsBidirectional,
+ "All inner iterators must be at least bidirectional.");
+ iterators = tup_dec(std::index_sequence_for<Iters...>{});
+ return *reinterpret_cast<ZipType *>(this);
+ }
+};
+
+template <typename... Iters>
+struct zip_first : public zip_common<zip_first<Iters...>, Iters...> {
+ using Base = zip_common<zip_first<Iters...>, Iters...>;
+
+ bool operator==(const zip_first<Iters...> &other) const {
+ return std::get<0>(this->iterators) == std::get<0>(other.iterators);
+ }
+
+ zip_first(Iters &&... ts) : Base(std::forward<Iters>(ts)...) {}
+};
+
+template <typename... Iters>
+class zip_shortest : public zip_common<zip_shortest<Iters...>, Iters...> {
+ template <size_t... Ns>
+ bool test(const zip_shortest<Iters...> &other,
+ std::index_sequence<Ns...>) const {
+ return all_of(std::initializer_list<bool>{std::get<Ns>(this->iterators) !=
+ std::get<Ns>(other.iterators)...},
+ identity<bool>{});
+ }
+
+public:
+ using Base = zip_common<zip_shortest<Iters...>, Iters...>;
+
+ zip_shortest(Iters &&... ts) : Base(std::forward<Iters>(ts)...) {}
+
+ bool operator==(const zip_shortest<Iters...> &other) const {
+ return !test(other, std::index_sequence_for<Iters...>{});
+ }
+};
+
+template <template <typename...> class ItType, typename... Args> class zippy {
+public:
+ using iterator = ItType<decltype(std::begin(std::declval<Args>()))...>;
+ using iterator_category = typename iterator::iterator_category;
+ using value_type = typename iterator::value_type;
+ using difference_type = typename iterator::difference_type;
+ using pointer = typename iterator::pointer;
+ using reference = typename iterator::reference;
+
+private:
+ std::tuple<Args...> ts;
+
+ template <size_t... Ns>
+ iterator begin_impl(std::index_sequence<Ns...>) const {
+ return iterator(std::begin(std::get<Ns>(ts))...);
+ }
+ template <size_t... Ns> iterator end_impl(std::index_sequence<Ns...>) const {
+ return iterator(std::end(std::get<Ns>(ts))...);
+ }
+
+public:
+ zippy(Args &&... ts_) : ts(std::forward<Args>(ts_)...) {}
+
+ iterator begin() const {
+ return begin_impl(std::index_sequence_for<Args...>{});
+ }
+ iterator end() const { return end_impl(std::index_sequence_for<Args...>{}); }
+};
+
+} // end namespace detail
+
+/// zip iterator for two or more iteratable types.
+template <typename T, typename U, typename... Args>
+detail::zippy<detail::zip_shortest, T, U, Args...> zip(T &&t, U &&u,
+ Args &&... args) {
+ return detail::zippy<detail::zip_shortest, T, U, Args...>(
+ std::forward<T>(t), std::forward<U>(u), std::forward<Args>(args)...);
+}
+
+/// zip iterator that, for the sake of efficiency, assumes the first iteratee to
+/// be the shortest.
+template <typename T, typename U, typename... Args>
+detail::zippy<detail::zip_first, T, U, Args...> zip_first(T &&t, U &&u,
+ Args &&... args) {
+ return detail::zippy<detail::zip_first, T, U, Args...>(
+ std::forward<T>(t), std::forward<U>(u), std::forward<Args>(args)...);
+}
+
+namespace detail {
+template <typename Iter>
+static Iter next_or_end(const Iter &I, const Iter &End) {
+ if (I == End)
+ return End;
+ return std::next(I);
+}
+
+template <typename Iter>
+static auto deref_or_none(const Iter &I, const Iter &End)
+ -> llvm::Optional<typename std::remove_const<
+ typename std::remove_reference<decltype(*I)>::type>::type> {
+ if (I == End)
+ return None;
+ return *I;
+}
+
+template <typename Iter> struct ZipLongestItemType {
+ using type =
+ llvm::Optional<typename std::remove_const<typename std::remove_reference<
+ decltype(*std::declval<Iter>())>::type>::type>;
+};
+
+template <typename... Iters> struct ZipLongestTupleType {
+ using type = std::tuple<typename ZipLongestItemType<Iters>::type...>;
+};
+
+template <typename... Iters>
+class zip_longest_iterator
+ : public iterator_facade_base<
+ zip_longest_iterator<Iters...>,
+ typename std::common_type<
+ std::forward_iterator_tag,
+ typename std::iterator_traits<Iters>::iterator_category...>::type,
+ typename ZipLongestTupleType<Iters...>::type,
+ typename std::iterator_traits<typename std::tuple_element<
+ 0, std::tuple<Iters...>>::type>::difference_type,
+ typename ZipLongestTupleType<Iters...>::type *,
+ typename ZipLongestTupleType<Iters...>::type> {
+public:
+ using value_type = typename ZipLongestTupleType<Iters...>::type;
+
+private:
+ std::tuple<Iters...> iterators;
+ std::tuple<Iters...> end_iterators;
+
+ template <size_t... Ns>
+ bool test(const zip_longest_iterator<Iters...> &other,
+ std::index_sequence<Ns...>) const {
+ return llvm::any_of(
+ std::initializer_list<bool>{std::get<Ns>(this->iterators) !=
+ std::get<Ns>(other.iterators)...},
+ identity<bool>{});
+ }
+
+ template <size_t... Ns> value_type deref(std::index_sequence<Ns...>) const {
+ return value_type(
+ deref_or_none(std::get<Ns>(iterators), std::get<Ns>(end_iterators))...);
+ }
+
+ template <size_t... Ns>
+ decltype(iterators) tup_inc(std::index_sequence<Ns...>) const {
+ return std::tuple<Iters...>(
+ next_or_end(std::get<Ns>(iterators), std::get<Ns>(end_iterators))...);
+ }
+
+public:
+ zip_longest_iterator(std::pair<Iters &&, Iters &&>... ts)
+ : iterators(std::forward<Iters>(ts.first)...),
+ end_iterators(std::forward<Iters>(ts.second)...) {}
+
+ value_type operator*() { return deref(std::index_sequence_for<Iters...>{}); }
+
+ value_type operator*() const {
+ return deref(std::index_sequence_for<Iters...>{});
+ }
+
+ zip_longest_iterator<Iters...> &operator++() {
+ iterators = tup_inc(std::index_sequence_for<Iters...>{});
+ return *this;
+ }
+
+ bool operator==(const zip_longest_iterator<Iters...> &other) const {
+ return !test(other, std::index_sequence_for<Iters...>{});
+ }
+};
+
+template <typename... Args> class zip_longest_range {
+public:
+ using iterator =
+ zip_longest_iterator<decltype(adl_begin(std::declval<Args>()))...>;
+ using iterator_category = typename iterator::iterator_category;
+ using value_type = typename iterator::value_type;
+ using difference_type = typename iterator::difference_type;
+ using pointer = typename iterator::pointer;
+ using reference = typename iterator::reference;
+
+private:
+ std::tuple<Args...> ts;
+
+ template <size_t... Ns>
+ iterator begin_impl(std::index_sequence<Ns...>) const {
+ return iterator(std::make_pair(adl_begin(std::get<Ns>(ts)),
+ adl_end(std::get<Ns>(ts)))...);
+ }
+
+ template <size_t... Ns> iterator end_impl(std::index_sequence<Ns...>) const {
+ return iterator(std::make_pair(adl_end(std::get<Ns>(ts)),
+ adl_end(std::get<Ns>(ts)))...);
+ }
+
+public:
+ zip_longest_range(Args &&... ts_) : ts(std::forward<Args>(ts_)...) {}
+
+ iterator begin() const {
+ return begin_impl(std::index_sequence_for<Args...>{});
+ }
+ iterator end() const { return end_impl(std::index_sequence_for<Args...>{}); }
+};
+} // namespace detail
+
+/// Iterate over two or more iterators at the same time. Iteration continues
+/// until all iterators reach the end. The llvm::Optional only contains a value
+/// if the iterator has not reached the end.
+template <typename T, typename U, typename... Args>
+detail::zip_longest_range<T, U, Args...> zip_longest(T &&t, U &&u,
+ Args &&... args) {
+ return detail::zip_longest_range<T, U, Args...>(
+ std::forward<T>(t), std::forward<U>(u), std::forward<Args>(args)...);
+}
+
+/// Iterator wrapper that concatenates sequences together.
+///
+/// This can concatenate different iterators, even with different types, into
+/// a single iterator provided the value types of all the concatenated
+/// iterators expose `reference` and `pointer` types that can be converted to
+/// `ValueT &` and `ValueT *` respectively. It doesn't support more
+/// interesting/customized pointer or reference types.
+///
+/// Currently this only supports forward or higher iterator categories as
+/// inputs and always exposes a forward iterator interface.
+template <typename ValueT, typename... IterTs>
+class concat_iterator
+ : public iterator_facade_base<concat_iterator<ValueT, IterTs...>,
+ std::forward_iterator_tag, ValueT> {
+ using BaseT = typename concat_iterator::iterator_facade_base;
+
+ /// We store both the current and end iterators for each concatenated
+ /// sequence in a tuple of pairs.
+ ///
+ /// Note that something like iterator_range seems nice at first here, but the
+ /// range properties are of little benefit and end up getting in the way
+ /// because we need to do mutation on the current iterators.
+ std::tuple<IterTs...> Begins;
+ std::tuple<IterTs...> Ends;
+
+ /// Attempts to increment a specific iterator.
+ ///
+ /// Returns true if it was able to increment the iterator. Returns false if
+ /// the iterator is already at the end iterator.
+ template <size_t Index> bool incrementHelper() {
+ auto &Begin = std::get<Index>(Begins);
+ auto &End = std::get<Index>(Ends);
+ if (Begin == End)
+ return false;
+
+ ++Begin;
+ return true;
+ }
+
+ /// Increments the first non-end iterator.
+ ///
+ /// It is an error to call this with all iterators at the end.
+ template <size_t... Ns> void increment(std::index_sequence<Ns...>) {
+ // Build a sequence of functions to increment each iterator if possible.
+ bool (concat_iterator::*IncrementHelperFns[])() = {
+ &concat_iterator::incrementHelper<Ns>...};
+
+ // Loop over them, and stop as soon as we succeed at incrementing one.
+ for (auto &IncrementHelperFn : IncrementHelperFns)
+ if ((this->*IncrementHelperFn)())
+ return;
+
+ llvm_unreachable("Attempted to increment an end concat iterator!");
+ }
+
+ /// Returns null if the specified iterator is at the end. Otherwise,
+ /// dereferences the iterator and returns the address of the resulting
+ /// reference.
+ template <size_t Index> ValueT *getHelper() const {
+ auto &Begin = std::get<Index>(Begins);
+ auto &End = std::get<Index>(Ends);
+ if (Begin == End)
+ return nullptr;
+
+ return &*Begin;
+ }
+
+ /// Finds the first non-end iterator, dereferences, and returns the resulting
+ /// reference.
+ ///
+ /// It is an error to call this with all iterators at the end.
+ template <size_t... Ns> ValueT &get(std::index_sequence<Ns...>) const {
+ // Build a sequence of functions to get from iterator if possible.
+ ValueT *(concat_iterator::*GetHelperFns[])() const = {
+ &concat_iterator::getHelper<Ns>...};
+
+ // Loop over them, and return the first result we find.
+ for (auto &GetHelperFn : GetHelperFns)
+ if (ValueT *P = (this->*GetHelperFn)())
+ return *P;
+
+ llvm_unreachable("Attempted to get a pointer from an end concat iterator!");
+ }
+
+public:
+ /// Constructs an iterator from a squence of ranges.
+ ///
+ /// We need the full range to know how to switch between each of the
+ /// iterators.
+ template <typename... RangeTs>
+ explicit concat_iterator(RangeTs &&... Ranges)
+ : Begins(std::begin(Ranges)...), Ends(std::end(Ranges)...) {}
+
+ using BaseT::operator++;
+
+ concat_iterator &operator++() {
+ increment(std::index_sequence_for<IterTs...>());
+ return *this;
+ }
+
+ ValueT &operator*() const {
+ return get(std::index_sequence_for<IterTs...>());
+ }
+
+ bool operator==(const concat_iterator &RHS) const {
+ return Begins == RHS.Begins && Ends == RHS.Ends;
+ }
+};
+
+namespace detail {
+
+/// Helper to store a sequence of ranges being concatenated and access them.
+///
+/// This is designed to facilitate providing actual storage when temporaries
+/// are passed into the constructor such that we can use it as part of range
+/// based for loops.
+template <typename ValueT, typename... RangeTs> class concat_range {
+public:
+ using iterator =
+ concat_iterator<ValueT,
+ decltype(std::begin(std::declval<RangeTs &>()))...>;
+
+private:
+ std::tuple<RangeTs...> Ranges;
+
+ template <size_t... Ns> iterator begin_impl(std::index_sequence<Ns...>) {
+ return iterator(std::get<Ns>(Ranges)...);
+ }
+ template <size_t... Ns> iterator end_impl(std::index_sequence<Ns...>) {
+ return iterator(make_range(std::end(std::get<Ns>(Ranges)),
+ std::end(std::get<Ns>(Ranges)))...);
+ }
+
+public:
+ concat_range(RangeTs &&... Ranges)
+ : Ranges(std::forward<RangeTs>(Ranges)...) {}
+
+ iterator begin() { return begin_impl(std::index_sequence_for<RangeTs...>{}); }
+ iterator end() { return end_impl(std::index_sequence_for<RangeTs...>{}); }
+};
+
+} // end namespace detail
+
+/// Concatenated range across two or more ranges.
+///
+/// The desired value type must be explicitly specified.
+template <typename ValueT, typename... RangeTs>
+detail::concat_range<ValueT, RangeTs...> concat(RangeTs &&... Ranges) {
+ static_assert(sizeof...(RangeTs) > 1,
+ "Need more than one range to concatenate!");
+ return detail::concat_range<ValueT, RangeTs...>(
+ std::forward<RangeTs>(Ranges)...);
+}
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <utility>
+//===----------------------------------------------------------------------===//
+
+/// Function object to check whether the first component of a std::pair
+/// compares less than the first component of another std::pair.
+struct less_first {
+ template <typename T> bool operator()(const T &lhs, const T &rhs) const {
+ return lhs.first < rhs.first;
+ }
+};
+
+/// Function object to check whether the second component of a std::pair
+/// compares less than the second component of another std::pair.
+struct less_second {
+ template <typename T> bool operator()(const T &lhs, const T &rhs) const {
+ return lhs.second < rhs.second;
+ }
+};
+
+/// \brief Function object to apply a binary function to the first component of
+/// a std::pair.
+template<typename FuncTy>
+struct on_first {
+ FuncTy func;
+
+ template <typename T>
+ auto operator()(const T &lhs, const T &rhs) const
+ -> decltype(func(lhs.first, rhs.first)) {
+ return func(lhs.first, rhs.first);
+ }
+};
+
+/// Utility type to build an inheritance chain that makes it easy to rank
+/// overload candidates.
+template <int N> struct rank : rank<N - 1> {};
+template <> struct rank<0> {};
+
+/// traits class for checking whether type T is one of any of the given
+/// types in the variadic list.
+template <typename T, typename... Ts> struct is_one_of {
+ static const bool value = false;
+};
+
+template <typename T, typename U, typename... Ts>
+struct is_one_of<T, U, Ts...> {
+ static const bool value =
+ std::is_same<T, U>::value || is_one_of<T, Ts...>::value;
+};
+
+/// traits class for checking whether type T is a base class for all
+/// the given types in the variadic list.
+template <typename T, typename... Ts> struct are_base_of {
+ static const bool value = true;
+};
+
+template <typename T, typename U, typename... Ts>
+struct are_base_of<T, U, Ts...> {
+ static const bool value =
+ std::is_base_of<T, U>::value && are_base_of<T, Ts...>::value;
+};
+
+//===----------------------------------------------------------------------===//
+// Extra additions for arrays
+//===----------------------------------------------------------------------===//
+
+/// Find the length of an array.
+template <class T, std::size_t N>
+constexpr inline size_t array_lengthof(T (&)[N]) {
+ return N;
+}
+
+/// Adapt std::less<T> for array_pod_sort.
+template<typename T>
+inline int array_pod_sort_comparator(const void *P1, const void *P2) {
+ if (std::less<T>()(*reinterpret_cast<const T*>(P1),
+ *reinterpret_cast<const T*>(P2)))
+ return -1;
+ if (std::less<T>()(*reinterpret_cast<const T*>(P2),
+ *reinterpret_cast<const T*>(P1)))
+ return 1;
+ return 0;
+}
+
+/// get_array_pod_sort_comparator - This is an internal helper function used to
+/// get type deduction of T right.
+template<typename T>
+inline int (*get_array_pod_sort_comparator(const T &))
+ (const void*, const void*) {
+ return array_pod_sort_comparator<T>;
+}
+
+/// array_pod_sort - This sorts an array with the specified start and end
+/// extent. This is just like std::sort, except that it calls qsort instead of
+/// using an inlined template. qsort is slightly slower than std::sort, but
+/// most sorts are not performance critical in LLVM and std::sort has to be
+/// template instantiated for each type, leading to significant measured code
+/// bloat. This function should generally be used instead of std::sort where
+/// possible.
+///
+/// This function assumes that you have simple POD-like types that can be
+/// compared with std::less and can be moved with memcpy. If this isn't true,
+/// you should use std::sort.
+///
+/// NOTE: If qsort_r were portable, we could allow a custom comparator and
+/// default to std::less.
+template<class IteratorTy>
+inline void array_pod_sort(IteratorTy Start, IteratorTy End) {
+ // Don't inefficiently call qsort with one element or trigger undefined
+ // behavior with an empty sequence.
+ auto NElts = End - Start;
+ if (NElts <= 1) return;
+#ifdef EXPENSIVE_CHECKS
+ std::mt19937 Generator(std::random_device{}());
+ std::shuffle(Start, End, Generator);
+#endif
+ qsort(&*Start, NElts, sizeof(*Start), get_array_pod_sort_comparator(*Start));
+}
+
+template <class IteratorTy>
+inline void array_pod_sort(
+ IteratorTy Start, IteratorTy End,
+ int (*Compare)(
+ const typename std::iterator_traits<IteratorTy>::value_type *,
+ const typename std::iterator_traits<IteratorTy>::value_type *)) {
+ // Don't inefficiently call qsort with one element or trigger undefined
+ // behavior with an empty sequence.
+ auto NElts = End - Start;
+ if (NElts <= 1) return;
+#ifdef EXPENSIVE_CHECKS
+ std::mt19937 Generator(std::random_device{}());
+ std::shuffle(Start, End, Generator);
+#endif
+ qsort(&*Start, NElts, sizeof(*Start),
+ reinterpret_cast<int (*)(const void *, const void *)>(Compare));
+}
+
+// Provide wrappers to std::sort which shuffle the elements before sorting
+// to help uncover non-deterministic behavior (PR35135).
+template <typename IteratorTy>
+inline void sort(IteratorTy Start, IteratorTy End) {
+#ifdef EXPENSIVE_CHECKS
+ std::mt19937 Generator(std::random_device{}());
+ std::shuffle(Start, End, Generator);
+#endif
+ std::sort(Start, End);
+}
+
+template <typename Container> inline void sort(Container &&C) {
+ llvm::sort(adl_begin(C), adl_end(C));
+}
+
+template <typename IteratorTy, typename Compare>
+inline void sort(IteratorTy Start, IteratorTy End, Compare Comp) {
+#ifdef EXPENSIVE_CHECKS
+ std::mt19937 Generator(std::random_device{}());
+ std::shuffle(Start, End, Generator);
+#endif
+ std::sort(Start, End, Comp);
+}
+
+template <typename Container, typename Compare>
+inline void sort(Container &&C, Compare Comp) {
+ llvm::sort(adl_begin(C), adl_end(C), Comp);
+}
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <algorithm>
+//===----------------------------------------------------------------------===//
+
+/// For a container of pointers, deletes the pointers and then clears the
+/// container.
+template<typename Container>
+void DeleteContainerPointers(Container &C) {
+ for (auto V : C)
+ delete V;
+ C.clear();
+}
+
+/// In a container of pairs (usually a map) whose second element is a pointer,
+/// deletes the second elements and then clears the container.
+template<typename Container>
+void DeleteContainerSeconds(Container &C) {
+ for (auto &V : C)
+ delete V.second;
+ C.clear();
+}
+
+/// Get the size of a range. This is a wrapper function around std::distance
+/// which is only enabled when the operation is O(1).
+template <typename R>
+auto size(R &&Range, typename std::enable_if<
+ std::is_same<typename std::iterator_traits<decltype(
+ Range.begin())>::iterator_category,
+ std::random_access_iterator_tag>::value,
+ void>::type * = nullptr)
+ -> decltype(std::distance(Range.begin(), Range.end())) {
+ return std::distance(Range.begin(), Range.end());
+}
+
+/// Provide wrappers to std::for_each which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+UnaryPredicate for_each(R &&Range, UnaryPredicate P) {
+ return std::for_each(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::all_of which take ranges instead of having to pass
+/// begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+bool all_of(R &&Range, UnaryPredicate P) {
+ return std::all_of(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::any_of which take ranges instead of having to pass
+/// begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+bool any_of(R &&Range, UnaryPredicate P) {
+ return std::any_of(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::none_of which take ranges instead of having to pass
+/// begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+bool none_of(R &&Range, UnaryPredicate P) {
+ return std::none_of(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::find which take ranges instead of having to pass
+/// begin/end explicitly.
+template <typename R, typename T>
+auto find(R &&Range, const T &Val) -> decltype(adl_begin(Range)) {
+ return std::find(adl_begin(Range), adl_end(Range), Val);
+}
+
+/// Provide wrappers to std::find_if which take ranges instead of having to pass
+/// begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+auto find_if(R &&Range, UnaryPredicate P) -> decltype(adl_begin(Range)) {
+ return std::find_if(adl_begin(Range), adl_end(Range), P);
+}
+
+template <typename R, typename UnaryPredicate>
+auto find_if_not(R &&Range, UnaryPredicate P) -> decltype(adl_begin(Range)) {
+ return std::find_if_not(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::remove_if which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+auto remove_if(R &&Range, UnaryPredicate P) -> decltype(adl_begin(Range)) {
+ return std::remove_if(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::copy_if which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename OutputIt, typename UnaryPredicate>
+OutputIt copy_if(R &&Range, OutputIt Out, UnaryPredicate P) {
+ return std::copy_if(adl_begin(Range), adl_end(Range), Out, P);
+}
+
+template <typename R, typename OutputIt>
+OutputIt copy(R &&Range, OutputIt Out) {
+ return std::copy(adl_begin(Range), adl_end(Range), Out);
+}
+
+/// Wrapper function around std::find to detect if an element exists
+/// in a container.
+template <typename R, typename E>
+bool is_contained(R &&Range, const E &Element) {
+ return std::find(adl_begin(Range), adl_end(Range), Element) != adl_end(Range);
+}
+
+/// Wrapper function around std::count to count the number of times an element
+/// \p Element occurs in the given range \p Range.
+template <typename R, typename E>
+auto count(R &&Range, const E &Element) ->
+ typename std::iterator_traits<decltype(adl_begin(Range))>::difference_type {
+ return std::count(adl_begin(Range), adl_end(Range), Element);
+}
+
+/// Wrapper function around std::count_if to count the number of times an
+/// element satisfying a given predicate occurs in a range.
+template <typename R, typename UnaryPredicate>
+auto count_if(R &&Range, UnaryPredicate P) ->
+ typename std::iterator_traits<decltype(adl_begin(Range))>::difference_type {
+ return std::count_if(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Wrapper function around std::transform to apply a function to a range and
+/// store the result elsewhere.
+template <typename R, typename OutputIt, typename UnaryPredicate>
+OutputIt transform(R &&Range, OutputIt d_first, UnaryPredicate P) {
+ return std::transform(adl_begin(Range), adl_end(Range), d_first, P);
+}
+
+/// Provide wrappers to std::partition which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename UnaryPredicate>
+auto partition(R &&Range, UnaryPredicate P) -> decltype(adl_begin(Range)) {
+ return std::partition(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Provide wrappers to std::lower_bound which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename T>
+auto lower_bound(R &&Range, T &&Value) -> decltype(adl_begin(Range)) {
+ return std::lower_bound(adl_begin(Range), adl_end(Range),
+ std::forward<T>(Value));
+}
+
+template <typename R, typename T, typename Compare>
+auto lower_bound(R &&Range, T &&Value, Compare C)
+ -> decltype(adl_begin(Range)) {
+ return std::lower_bound(adl_begin(Range), adl_end(Range),
+ std::forward<T>(Value), C);
+}
+
+/// Provide wrappers to std::upper_bound which take ranges instead of having to
+/// pass begin/end explicitly.
+template <typename R, typename T>
+auto upper_bound(R &&Range, T &&Value) -> decltype(adl_begin(Range)) {
+ return std::upper_bound(adl_begin(Range), adl_end(Range),
+ std::forward<T>(Value));
+}
+
+template <typename R, typename T, typename Compare>
+auto upper_bound(R &&Range, T &&Value, Compare C)
+ -> decltype(adl_begin(Range)) {
+ return std::upper_bound(adl_begin(Range), adl_end(Range),
+ std::forward<T>(Value), C);
+}
+
+template <typename R>
+void stable_sort(R &&Range) {
+ std::stable_sort(adl_begin(Range), adl_end(Range));
+}
+
+template <typename R, typename Compare>
+void stable_sort(R &&Range, Compare C) {
+ std::stable_sort(adl_begin(Range), adl_end(Range), C);
+}
+
+/// Binary search for the first iterator in a range where a predicate is false.
+/// Requires that C is always true below some limit, and always false above it.
+template <typename R, typename Predicate,
+ typename Val = decltype(*adl_begin(std::declval<R>()))>
+auto partition_point(R &&Range, Predicate P) -> decltype(adl_begin(Range)) {
+ return std::partition_point(adl_begin(Range), adl_end(Range), P);
+}
+
+/// Wrapper function around std::equal to detect if all elements
+/// in a container are same.
+template <typename R>
+bool is_splat(R &&Range) {
+ size_t range_size = size(Range);
+ return range_size != 0 && (range_size == 1 ||
+ std::equal(adl_begin(Range) + 1, adl_end(Range), adl_begin(Range)));
+}
+
+/// Given a range of type R, iterate the entire range and return a
+/// SmallVector with elements of the vector. This is useful, for example,
+/// when you want to iterate a range and then sort the results.
+template <unsigned Size, typename R>
+SmallVector<typename std::remove_const<detail::ValueOfRange<R>>::type, Size>
+to_vector(R &&Range) {
+ return {adl_begin(Range), adl_end(Range)};
+}
+
+/// Provide a container algorithm similar to C++ Library Fundamentals v2's
+/// `erase_if` which is equivalent to:
+///
+/// C.erase(remove_if(C, pred), C.end());
+///
+/// This version works for any container with an erase method call accepting
+/// two iterators.
+template <typename Container, typename UnaryPredicate>
+void erase_if(Container &C, UnaryPredicate P) {
+ C.erase(remove_if(C, P), C.end());
+}
+
+/// Given a sequence container Cont, replace the range [ContIt, ContEnd) with
+/// the range [ValIt, ValEnd) (which is not from the same container).
+template<typename Container, typename RandomAccessIterator>
+void replace(Container &Cont, typename Container::iterator ContIt,
+ typename Container::iterator ContEnd, RandomAccessIterator ValIt,
+ RandomAccessIterator ValEnd) {
+ while (true) {
+ if (ValIt == ValEnd) {
+ Cont.erase(ContIt, ContEnd);
+ return;
+ } else if (ContIt == ContEnd) {
+ Cont.insert(ContIt, ValIt, ValEnd);
+ return;
+ }
+ *ContIt++ = *ValIt++;
+ }
+}
+
+/// Given a sequence container Cont, replace the range [ContIt, ContEnd) with
+/// the range R.
+template<typename Container, typename Range = std::initializer_list<
+ typename Container::value_type>>
+void replace(Container &Cont, typename Container::iterator ContIt,
+ typename Container::iterator ContEnd, Range R) {
+ replace(Cont, ContIt, ContEnd, R.begin(), R.end());
+}
+
+//===----------------------------------------------------------------------===//
+// Extra additions to <memory>
+//===----------------------------------------------------------------------===//
+
+struct FreeDeleter {
+ void operator()(void* v) {
+ ::free(v);
+ }
+};
+
+template<typename First, typename Second>
+struct pair_hash {
+ size_t operator()(const std::pair<First, Second> &P) const {
+ return std::hash<First>()(P.first) * 31 + std::hash<Second>()(P.second);
+ }
+};
+
+/// Binary functor that adapts to any other binary functor after dereferencing
+/// operands.
+template <typename T> struct deref {
+ T func;
+
+ // Could be further improved to cope with non-derivable functors and
+ // non-binary functors (should be a variadic template member function
+ // operator()).
+ template <typename A, typename B>
+ auto operator()(A &lhs, B &rhs) const -> decltype(func(*lhs, *rhs)) {
+ assert(lhs);
+ assert(rhs);
+ return func(*lhs, *rhs);
+ }
+};
+
+namespace detail {
+
+template <typename R> class enumerator_iter;
+
+template <typename R> struct result_pair {
+ using value_reference =
+ typename std::iterator_traits<IterOfRange<R>>::reference;
+
+ friend class enumerator_iter<R>;
+
+ result_pair() = default;
+ result_pair(std::size_t Index, IterOfRange<R> Iter)
+ : Index(Index), Iter(Iter) {}
+
+ result_pair<R> &operator=(const result_pair<R> &Other) {
+ Index = Other.Index;
+ Iter = Other.Iter;
+ return *this;
+ }
+
+ std::size_t index() const { return Index; }
+ const value_reference value() const { return *Iter; }
+ value_reference value() { return *Iter; }
+
+private:
+ std::size_t Index = std::numeric_limits<std::size_t>::max();
+ IterOfRange<R> Iter;
+};
+
+template <typename R>
+class enumerator_iter
+ : public iterator_facade_base<
+ enumerator_iter<R>, std::forward_iterator_tag, result_pair<R>,
+ typename std::iterator_traits<IterOfRange<R>>::difference_type,
+ typename std::iterator_traits<IterOfRange<R>>::pointer,
+ typename std::iterator_traits<IterOfRange<R>>::reference> {
+ using result_type = result_pair<R>;
+
+public:
+ explicit enumerator_iter(IterOfRange<R> EndIter)
+ : Result(std::numeric_limits<size_t>::max(), EndIter) {}
+
+ enumerator_iter(std::size_t Index, IterOfRange<R> Iter)
+ : Result(Index, Iter) {}
+
+ result_type &operator*() { return Result; }
+ const result_type &operator*() const { return Result; }
+
+ enumerator_iter<R> &operator++() {
+ assert(Result.Index != std::numeric_limits<size_t>::max());
+ ++Result.Iter;
+ ++Result.Index;
+ return *this;
+ }
+
+ bool operator==(const enumerator_iter<R> &RHS) const {
+ // Don't compare indices here, only iterators. It's possible for an end
+ // iterator to have different indices depending on whether it was created
+ // by calling std::end() versus incrementing a valid iterator.
+ return Result.Iter == RHS.Result.Iter;
+ }
+
+ enumerator_iter<R> &operator=(const enumerator_iter<R> &Other) {
+ Result = Other.Result;
+ return *this;
+ }
+
+private:
+ result_type Result;
+};
+
+template <typename R> class enumerator {
+public:
+ explicit enumerator(R &&Range) : TheRange(std::forward<R>(Range)) {}
+
+ enumerator_iter<R> begin() {
+ return enumerator_iter<R>(0, std::begin(TheRange));
+ }
+
+ enumerator_iter<R> end() {
+ return enumerator_iter<R>(std::end(TheRange));
+ }
+
+private:
+ R TheRange;
+};
+
+} // end namespace detail
+
+/// Given an input range, returns a new range whose values are are pair (A,B)
+/// such that A is the 0-based index of the item in the sequence, and B is
+/// the value from the original sequence. Example:
+///
+/// std::vector<char> Items = {'A', 'B', 'C', 'D'};
+/// for (auto X : enumerate(Items)) {
+/// printf("Item %d - %c\n", X.index(), X.value());
+/// }
+///
+/// Output:
+/// Item 0 - A
+/// Item 1 - B
+/// Item 2 - C
+/// Item 3 - D
+///
+template <typename R> detail::enumerator<R> enumerate(R &&TheRange) {
+ return detail::enumerator<R>(std::forward<R>(TheRange));
+}
+
+namespace detail {
+
+template <typename F, typename Tuple, std::size_t... I>
+auto apply_tuple_impl(F &&f, Tuple &&t, std::index_sequence<I...>)
+ -> decltype(std::forward<F>(f)(std::get<I>(std::forward<Tuple>(t))...)) {
+ return std::forward<F>(f)(std::get<I>(std::forward<Tuple>(t))...);
+}
+
+} // end namespace detail
+
+/// Given an input tuple (a1, a2, ..., an), pass the arguments of the
+/// tuple variadically to f as if by calling f(a1, a2, ..., an) and
+/// return the result.
+template <typename F, typename Tuple>
+auto apply_tuple(F &&f, Tuple &&t) -> decltype(detail::apply_tuple_impl(
+ std::forward<F>(f), std::forward<Tuple>(t),
+ std::make_index_sequence<
+ std::tuple_size<typename std::decay<Tuple>::type>::value>{})) {
+ using Indices = std::make_index_sequence<
+ std::tuple_size<typename std::decay<Tuple>::type>::value>;
+
+ return detail::apply_tuple_impl(std::forward<F>(f), std::forward<Tuple>(t),
+ Indices{});
+}
+
+/// Return true if the sequence [Begin, End) has exactly N items. Runs in O(N)
+/// time. Not meant for use with random-access iterators.
+template <typename IterTy>
+bool hasNItems(
+ IterTy &&Begin, IterTy &&End, unsigned N,
+ typename std::enable_if<
+ !std::is_same<
+ typename std::iterator_traits<typename std::remove_reference<
+ decltype(Begin)>::type>::iterator_category,
+ std::random_access_iterator_tag>::value,
+ void>::type * = nullptr) {
+ for (; N; --N, ++Begin)
+ if (Begin == End)
+ return false; // Too few.
+ return Begin == End;
+}
+
+/// Return true if the sequence [Begin, End) has N or more items. Runs in O(N)
+/// time. Not meant for use with random-access iterators.
+template <typename IterTy>
+bool hasNItemsOrMore(
+ IterTy &&Begin, IterTy &&End, unsigned N,
+ typename std::enable_if<
+ !std::is_same<
+ typename std::iterator_traits<typename std::remove_reference<
+ decltype(Begin)>::type>::iterator_category,
+ std::random_access_iterator_tag>::value,
+ void>::type * = nullptr) {
+ for (; N; --N, ++Begin)
+ if (Begin == End)
+ return false; // Too few.
+ return true;
+}
+
+/// Returns a raw pointer that represents the same address as the argument.
+///
+/// The late bound return should be removed once we move to C++14 to better
+/// align with the C++20 declaration. Also, this implementation can be removed
+/// once we move to C++20 where it's defined as std::to_addres()
+///
+/// The std::pointer_traits<>::to_address(p) variations of these overloads has
+/// not been implemented.
+template <class Ptr> auto to_address(const Ptr &P) -> decltype(P.operator->()) {
+ return P.operator->();
+}
+template <class T> constexpr T *to_address(T *P) { return P; }
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STLEXTRAS_H
diff --git a/third_party/llvm-project/include/llvm/ADT/SmallPtrSet.h b/third_party/llvm-project/include/llvm/ADT/SmallPtrSet.h
new file mode 100644
index 000000000..1d8280063
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/SmallPtrSet.h
@@ -0,0 +1,511 @@
+//===- llvm/ADT/SmallPtrSet.h - 'Normally small' pointer set ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the SmallPtrSet class. See the doxygen comment for
+// SmallPtrSetImplBase for more details on the algorithm used.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_SMALLPTRSET_H
+#define LLVM_ADT_SMALLPTRSET_H
+
+#include "llvm/ADT/EpochTracker.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ReverseIteration.h"
+#include "llvm/Support/type_traits.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdlib>
+#include <cstring>
+#include <initializer_list>
+#include <iterator>
+#include <utility>
+
+namespace llvm {
+
+/// SmallPtrSetImplBase - This is the common code shared among all the
+/// SmallPtrSet<>'s, which is almost everything. SmallPtrSet has two modes, one
+/// for small and one for large sets.
+///
+/// Small sets use an array of pointers allocated in the SmallPtrSet object,
+/// which is treated as a simple array of pointers. When a pointer is added to
+/// the set, the array is scanned to see if the element already exists, if not
+/// the element is 'pushed back' onto the array. If we run out of space in the
+/// array, we grow into the 'large set' case. SmallSet should be used when the
+/// sets are often small. In this case, no memory allocation is used, and only
+/// light-weight and cache-efficient scanning is used.
+///
+/// Large sets use a classic exponentially-probed hash table. Empty buckets are
+/// represented with an illegal pointer value (-1) to allow null pointers to be
+/// inserted. Tombstones are represented with another illegal pointer value
+/// (-2), to allow deletion. The hash table is resized when the table is 3/4 or
+/// more. When this happens, the table is doubled in size.
+///
+class SmallPtrSetImplBase : public DebugEpochBase {
+ friend class SmallPtrSetIteratorImpl;
+
+protected:
+ /// SmallArray - Points to a fixed size set of buckets, used in 'small mode'.
+ const void **SmallArray;
+ /// CurArray - This is the current set of buckets. If equal to SmallArray,
+ /// then the set is in 'small mode'.
+ const void **CurArray;
+ /// CurArraySize - The allocated size of CurArray, always a power of two.
+ unsigned CurArraySize;
+
+ /// Number of elements in CurArray that contain a value or are a tombstone.
+ /// If small, all these elements are at the beginning of CurArray and the rest
+ /// is uninitialized.
+ unsigned NumNonEmpty;
+ /// Number of tombstones in CurArray.
+ unsigned NumTombstones;
+
+ // Helpers to copy and move construct a SmallPtrSet.
+ SmallPtrSetImplBase(const void **SmallStorage,
+ const SmallPtrSetImplBase &that);
+ SmallPtrSetImplBase(const void **SmallStorage, unsigned SmallSize,
+ SmallPtrSetImplBase &&that);
+
+ explicit SmallPtrSetImplBase(const void **SmallStorage, unsigned SmallSize)
+ : SmallArray(SmallStorage), CurArray(SmallStorage),
+ CurArraySize(SmallSize), NumNonEmpty(0), NumTombstones(0) {
+ assert(SmallSize && (SmallSize & (SmallSize-1)) == 0 &&
+ "Initial size must be a power of two!");
+ }
+
+ ~SmallPtrSetImplBase() {
+ if (!isSmall())
+ free(CurArray);
+ }
+
+public:
+ using size_type = unsigned;
+
+ SmallPtrSetImplBase &operator=(const SmallPtrSetImplBase &) = delete;
+
+ LLVM_NODISCARD bool empty() const { return size() == 0; }
+ size_type size() const { return NumNonEmpty - NumTombstones; }
+
+ void clear() {
+ incrementEpoch();
+ // If the capacity of the array is huge, and the # elements used is small,
+ // shrink the array.
+ if (!isSmall()) {
+ if (size() * 4 < CurArraySize && CurArraySize > 32)
+ return shrink_and_clear();
+ // Fill the array with empty markers.
+ memset(CurArray, -1, CurArraySize * sizeof(void *));
+ }
+
+ NumNonEmpty = 0;
+ NumTombstones = 0;
+ }
+
+protected:
+ static void *getTombstoneMarker() { return reinterpret_cast<void*>(-2); }
+
+ static void *getEmptyMarker() {
+ // Note that -1 is chosen to make clear() efficiently implementable with
+ // memset and because it's not a valid pointer value.
+ return reinterpret_cast<void*>(-1);
+ }
+
+ const void **EndPointer() const {
+ return isSmall() ? CurArray + NumNonEmpty : CurArray + CurArraySize;
+ }
+
+ /// insert_imp - This returns true if the pointer was new to the set, false if
+ /// it was already in the set. This is hidden from the client so that the
+ /// derived class can check that the right type of pointer is passed in.
+ std::pair<const void *const *, bool> insert_imp(const void *Ptr) {
+ if (isSmall()) {
+ // Check to see if it is already in the set.
+ const void **LastTombstone = nullptr;
+ for (const void **APtr = SmallArray, **E = SmallArray + NumNonEmpty;
+ APtr != E; ++APtr) {
+ const void *Value = *APtr;
+ if (Value == Ptr)
+ return std::make_pair(APtr, false);
+ if (Value == getTombstoneMarker())
+ LastTombstone = APtr;
+ }
+
+ // Did we find any tombstone marker?
+ if (LastTombstone != nullptr) {
+ *LastTombstone = Ptr;
+ --NumTombstones;
+ incrementEpoch();
+ return std::make_pair(LastTombstone, true);
+ }
+
+ // Nope, there isn't. If we stay small, just 'pushback' now.
+ if (NumNonEmpty < CurArraySize) {
+ SmallArray[NumNonEmpty++] = Ptr;
+ incrementEpoch();
+ return std::make_pair(SmallArray + (NumNonEmpty - 1), true);
+ }
+ // Otherwise, hit the big set case, which will call grow.
+ }
+ return insert_imp_big(Ptr);
+ }
+
+ /// erase_imp - If the set contains the specified pointer, remove it and
+ /// return true, otherwise return false. This is hidden from the client so
+ /// that the derived class can check that the right type of pointer is passed
+ /// in.
+ bool erase_imp(const void * Ptr) {
+ const void *const *P = find_imp(Ptr);
+ if (P == EndPointer())
+ return false;
+
+ const void **Loc = const_cast<const void **>(P);
+ assert(*Loc == Ptr && "broken find!");
+ *Loc = getTombstoneMarker();
+ NumTombstones++;
+ return true;
+ }
+
+ /// Returns the raw pointer needed to construct an iterator. If element not
+ /// found, this will be EndPointer. Otherwise, it will be a pointer to the
+ /// slot which stores Ptr;
+ const void *const * find_imp(const void * Ptr) const {
+ if (isSmall()) {
+ // Linear search for the item.
+ for (const void *const *APtr = SmallArray,
+ *const *E = SmallArray + NumNonEmpty; APtr != E; ++APtr)
+ if (*APtr == Ptr)
+ return APtr;
+ return EndPointer();
+ }
+
+ // Big set case.
+ auto *Bucket = FindBucketFor(Ptr);
+ if (*Bucket == Ptr)
+ return Bucket;
+ return EndPointer();
+ }
+
+private:
+ bool isSmall() const { return CurArray == SmallArray; }
+
+ std::pair<const void *const *, bool> insert_imp_big(const void *Ptr);
+
+ const void * const *FindBucketFor(const void *Ptr) const;
+ void shrink_and_clear();
+
+ /// Grow - Allocate a larger backing store for the buckets and move it over.
+ void Grow(unsigned NewSize);
+
+protected:
+ /// swap - Swaps the elements of two sets.
+ /// Note: This method assumes that both sets have the same small size.
+ void swap(SmallPtrSetImplBase &RHS);
+
+ void CopyFrom(const SmallPtrSetImplBase &RHS);
+ void MoveFrom(unsigned SmallSize, SmallPtrSetImplBase &&RHS);
+
+private:
+ /// Code shared by MoveFrom() and move constructor.
+ void MoveHelper(unsigned SmallSize, SmallPtrSetImplBase &&RHS);
+ /// Code shared by CopyFrom() and copy constructor.
+ void CopyHelper(const SmallPtrSetImplBase &RHS);
+};
+
+/// SmallPtrSetIteratorImpl - This is the common base class shared between all
+/// instances of SmallPtrSetIterator.
+class SmallPtrSetIteratorImpl {
+protected:
+ const void *const *Bucket;
+ const void *const *End;
+
+public:
+ explicit SmallPtrSetIteratorImpl(const void *const *BP, const void*const *E)
+ : Bucket(BP), End(E) {
+ if (shouldReverseIterate()) {
+ RetreatIfNotValid();
+ return;
+ }
+ AdvanceIfNotValid();
+ }
+
+ bool operator==(const SmallPtrSetIteratorImpl &RHS) const {
+ return Bucket == RHS.Bucket;
+ }
+ bool operator!=(const SmallPtrSetIteratorImpl &RHS) const {
+ return Bucket != RHS.Bucket;
+ }
+
+protected:
+ /// AdvanceIfNotValid - If the current bucket isn't valid, advance to a bucket
+ /// that is. This is guaranteed to stop because the end() bucket is marked
+ /// valid.
+ void AdvanceIfNotValid() {
+ assert(Bucket <= End);
+ while (Bucket != End &&
+ (*Bucket == SmallPtrSetImplBase::getEmptyMarker() ||
+ *Bucket == SmallPtrSetImplBase::getTombstoneMarker()))
+ ++Bucket;
+ }
+ void RetreatIfNotValid() {
+ assert(Bucket >= End);
+ while (Bucket != End &&
+ (Bucket[-1] == SmallPtrSetImplBase::getEmptyMarker() ||
+ Bucket[-1] == SmallPtrSetImplBase::getTombstoneMarker())) {
+ --Bucket;
+ }
+ }
+};
+
+/// SmallPtrSetIterator - This implements a const_iterator for SmallPtrSet.
+template <typename PtrTy>
+class SmallPtrSetIterator : public SmallPtrSetIteratorImpl,
+ DebugEpochBase::HandleBase {
+ using PtrTraits = PointerLikeTypeTraits<PtrTy>;
+
+public:
+ using value_type = PtrTy;
+ using reference = PtrTy;
+ using pointer = PtrTy;
+ using difference_type = std::ptrdiff_t;
+ using iterator_category = std::forward_iterator_tag;
+
+ explicit SmallPtrSetIterator(const void *const *BP, const void *const *E,
+ const DebugEpochBase &Epoch)
+ : SmallPtrSetIteratorImpl(BP, E), DebugEpochBase::HandleBase(&Epoch) {}
+
+ // Most methods provided by baseclass.
+
+ const PtrTy operator*() const {
+ assert(isHandleInSync() && "invalid iterator access!");
+ if (shouldReverseIterate()) {
+ assert(Bucket > End);
+ return PtrTraits::getFromVoidPointer(const_cast<void *>(Bucket[-1]));
+ }
+ assert(Bucket < End);
+ return PtrTraits::getFromVoidPointer(const_cast<void*>(*Bucket));
+ }
+
+ inline SmallPtrSetIterator& operator++() { // Preincrement
+ assert(isHandleInSync() && "invalid iterator access!");
+ if (shouldReverseIterate()) {
+ --Bucket;
+ RetreatIfNotValid();
+ return *this;
+ }
+ ++Bucket;
+ AdvanceIfNotValid();
+ return *this;
+ }
+
+ SmallPtrSetIterator operator++(int) { // Postincrement
+ SmallPtrSetIterator tmp = *this;
+ ++*this;
+ return tmp;
+ }
+};
+
+/// RoundUpToPowerOfTwo - This is a helper template that rounds N up to the next
+/// power of two (which means N itself if N is already a power of two).
+template<unsigned N>
+struct RoundUpToPowerOfTwo;
+
+/// RoundUpToPowerOfTwoH - If N is not a power of two, increase it. This is a
+/// helper template used to implement RoundUpToPowerOfTwo.
+template<unsigned N, bool isPowerTwo>
+struct RoundUpToPowerOfTwoH {
+ enum { Val = N };
+};
+template<unsigned N>
+struct RoundUpToPowerOfTwoH<N, false> {
+ enum {
+ // We could just use NextVal = N+1, but this converges faster. N|(N-1) sets
+ // the right-most zero bits to one all at once, e.g. 0b0011000 -> 0b0011111.
+ Val = RoundUpToPowerOfTwo<(N|(N-1)) + 1>::Val
+ };
+};
+
+template<unsigned N>
+struct RoundUpToPowerOfTwo {
+ enum { Val = RoundUpToPowerOfTwoH<N, (N&(N-1)) == 0>::Val };
+};
+
+/// A templated base class for \c SmallPtrSet which provides the
+/// typesafe interface that is common across all small sizes.
+///
+/// This is particularly useful for passing around between interface boundaries
+/// to avoid encoding a particular small size in the interface boundary.
+template <typename PtrType>
+class SmallPtrSetImpl : public SmallPtrSetImplBase {
+ using ConstPtrType = typename add_const_past_pointer<PtrType>::type;
+ using PtrTraits = PointerLikeTypeTraits<PtrType>;
+ using ConstPtrTraits = PointerLikeTypeTraits<ConstPtrType>;
+
+protected:
+ // Constructors that forward to the base.
+ SmallPtrSetImpl(const void **SmallStorage, const SmallPtrSetImpl &that)
+ : SmallPtrSetImplBase(SmallStorage, that) {}
+ SmallPtrSetImpl(const void **SmallStorage, unsigned SmallSize,
+ SmallPtrSetImpl &&that)
+ : SmallPtrSetImplBase(SmallStorage, SmallSize, std::move(that)) {}
+ explicit SmallPtrSetImpl(const void **SmallStorage, unsigned SmallSize)
+ : SmallPtrSetImplBase(SmallStorage, SmallSize) {}
+
+public:
+ using iterator = SmallPtrSetIterator<PtrType>;
+ using const_iterator = SmallPtrSetIterator<PtrType>;
+ using key_type = ConstPtrType;
+ using value_type = PtrType;
+
+ SmallPtrSetImpl(const SmallPtrSetImpl &) = delete;
+
+ /// Inserts Ptr if and only if there is no element in the container equal to
+ /// Ptr. The bool component of the returned pair is true if and only if the
+ /// insertion takes place, and the iterator component of the pair points to
+ /// the element equal to Ptr.
+ std::pair<iterator, bool> insert(PtrType Ptr) {
+ auto p = insert_imp(PtrTraits::getAsVoidPointer(Ptr));
+ return std::make_pair(makeIterator(p.first), p.second);
+ }
+
+ /// erase - If the set contains the specified pointer, remove it and return
+ /// true, otherwise return false.
+ bool erase(PtrType Ptr) {
+ return erase_imp(PtrTraits::getAsVoidPointer(Ptr));
+ }
+ /// count - Return 1 if the specified pointer is in the set, 0 otherwise.
+ size_type count(ConstPtrType Ptr) const { return find(Ptr) != end() ? 1 : 0; }
+ iterator find(ConstPtrType Ptr) const {
+ return makeIterator(find_imp(ConstPtrTraits::getAsVoidPointer(Ptr)));
+ }
+
+ template <typename IterT>
+ void insert(IterT I, IterT E) {
+ for (; I != E; ++I)
+ insert(*I);
+ }
+
+ void insert(std::initializer_list<PtrType> IL) {
+ insert(IL.begin(), IL.end());
+ }
+
+ iterator begin() const {
+ if (shouldReverseIterate())
+ return makeIterator(EndPointer() - 1);
+ return makeIterator(CurArray);
+ }
+ iterator end() const { return makeIterator(EndPointer()); }
+
+private:
+ /// Create an iterator that dereferences to same place as the given pointer.
+ iterator makeIterator(const void *const *P) const {
+ if (shouldReverseIterate())
+ return iterator(P == EndPointer() ? CurArray : P + 1, CurArray, *this);
+ return iterator(P, EndPointer(), *this);
+ }
+};
+
+/// Equality comparison for SmallPtrSet.
+///
+/// Iterates over elements of LHS confirming that each value from LHS is also in
+/// RHS, and that no additional values are in RHS.
+template <typename PtrType>
+bool operator==(const SmallPtrSetImpl<PtrType> &LHS,
+ const SmallPtrSetImpl<PtrType> &RHS) {
+ if (LHS.size() != RHS.size())
+ return false;
+
+ for (const auto *KV : LHS)
+ if (!RHS.count(KV))
+ return false;
+
+ return true;
+}
+
+/// Inequality comparison for SmallPtrSet.
+///
+/// Equivalent to !(LHS == RHS).
+template <typename PtrType>
+bool operator!=(const SmallPtrSetImpl<PtrType> &LHS,
+ const SmallPtrSetImpl<PtrType> &RHS) {
+ return !(LHS == RHS);
+}
+
+/// SmallPtrSet - This class implements a set which is optimized for holding
+/// SmallSize or less elements. This internally rounds up SmallSize to the next
+/// power of two if it is not already a power of two. See the comments above
+/// SmallPtrSetImplBase for details of the algorithm.
+template<class PtrType, unsigned SmallSize>
+class SmallPtrSet : public SmallPtrSetImpl<PtrType> {
+ // In small mode SmallPtrSet uses linear search for the elements, so it is
+ // not a good idea to choose this value too high. You may consider using a
+ // DenseSet<> instead if you expect many elements in the set.
+ static_assert(SmallSize <= 32, "SmallSize should be small");
+
+ using BaseT = SmallPtrSetImpl<PtrType>;
+
+ // Make sure that SmallSize is a power of two, round up if not.
+ enum { SmallSizePowTwo = RoundUpToPowerOfTwo<SmallSize>::Val };
+ /// SmallStorage - Fixed size storage used in 'small mode'.
+ const void *SmallStorage[SmallSizePowTwo];
+
+public:
+ SmallPtrSet() : BaseT(SmallStorage, SmallSizePowTwo) {}
+ SmallPtrSet(const SmallPtrSet &that) : BaseT(SmallStorage, that) {}
+ SmallPtrSet(SmallPtrSet &&that)
+ : BaseT(SmallStorage, SmallSizePowTwo, std::move(that)) {}
+
+ template<typename It>
+ SmallPtrSet(It I, It E) : BaseT(SmallStorage, SmallSizePowTwo) {
+ this->insert(I, E);
+ }
+
+ SmallPtrSet(std::initializer_list<PtrType> IL)
+ : BaseT(SmallStorage, SmallSizePowTwo) {
+ this->insert(IL.begin(), IL.end());
+ }
+
+ SmallPtrSet<PtrType, SmallSize> &
+ operator=(const SmallPtrSet<PtrType, SmallSize> &RHS) {
+ if (&RHS != this)
+ this->CopyFrom(RHS);
+ return *this;
+ }
+
+ SmallPtrSet<PtrType, SmallSize> &
+ operator=(SmallPtrSet<PtrType, SmallSize> &&RHS) {
+ if (&RHS != this)
+ this->MoveFrom(SmallSizePowTwo, std::move(RHS));
+ return *this;
+ }
+
+ SmallPtrSet<PtrType, SmallSize> &
+ operator=(std::initializer_list<PtrType> IL) {
+ this->clear();
+ this->insert(IL.begin(), IL.end());
+ return *this;
+ }
+
+ /// swap - Swaps the elements of two sets.
+ void swap(SmallPtrSet<PtrType, SmallSize> &RHS) {
+ SmallPtrSetImplBase::swap(RHS);
+ }
+};
+
+} // end namespace llvm
+
+namespace std {
+
+ /// Implement std::swap in terms of SmallPtrSet swap.
+ template<class T, unsigned N>
+ inline void swap(llvm::SmallPtrSet<T, N> &LHS, llvm::SmallPtrSet<T, N> &RHS) {
+ LHS.swap(RHS);
+ }
+
+} // end namespace std
+
+#endif // LLVM_ADT_SMALLPTRSET_H
diff --git a/third_party/llvm-project/include/llvm/ADT/SmallSet.h b/third_party/llvm-project/include/llvm/ADT/SmallSet.h
new file mode 100644
index 000000000..a03fa7dd8
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/SmallSet.h
@@ -0,0 +1,278 @@
+//===- llvm/ADT/SmallSet.h - 'Normally small' sets --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the SmallSet class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_SMALLSET_H
+#define LLVM_ADT_SMALLSET_H
+
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/type_traits.h"
+#include <cstddef>
+#include <functional>
+#include <set>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// SmallSetIterator - This class implements a const_iterator for SmallSet by
+/// delegating to the underlying SmallVector or Set iterators.
+template <typename T, unsigned N, typename C>
+class SmallSetIterator
+ : public iterator_facade_base<SmallSetIterator<T, N, C>,
+ std::forward_iterator_tag, T> {
+private:
+ using SetIterTy = typename std::set<T, C>::const_iterator;
+ using VecIterTy = typename SmallVector<T, N>::const_iterator;
+ using SelfTy = SmallSetIterator<T, N, C>;
+
+ /// Iterators to the parts of the SmallSet containing the data. They are set
+ /// depending on isSmall.
+ union {
+ SetIterTy SetIter;
+ VecIterTy VecIter;
+ };
+
+ bool isSmall;
+
+public:
+ SmallSetIterator(SetIterTy SetIter) : SetIter(SetIter), isSmall(false) {}
+
+ SmallSetIterator(VecIterTy VecIter) : VecIter(VecIter), isSmall(true) {}
+
+ // Spell out destructor, copy/move constructor and assignment operators for
+ // MSVC STL, where set<T>::const_iterator is not trivially copy constructible.
+ ~SmallSetIterator() {
+ if (isSmall)
+ VecIter.~VecIterTy();
+ else
+ SetIter.~SetIterTy();
+ }
+
+ SmallSetIterator(const SmallSetIterator &Other) : isSmall(Other.isSmall) {
+ if (isSmall)
+ VecIter = Other.VecIter;
+ else
+ // Use placement new, to make sure SetIter is properly constructed, even
+ // if it is not trivially copy-able (e.g. in MSVC).
+ new (&SetIter) SetIterTy(Other.SetIter);
+ }
+
+ SmallSetIterator(SmallSetIterator &&Other) : isSmall(Other.isSmall) {
+ if (isSmall)
+ VecIter = std::move(Other.VecIter);
+ else
+ // Use placement new, to make sure SetIter is properly constructed, even
+ // if it is not trivially copy-able (e.g. in MSVC).
+ new (&SetIter) SetIterTy(std::move(Other.SetIter));
+ }
+
+ SmallSetIterator& operator=(const SmallSetIterator& Other) {
+ // Call destructor for SetIter, so it gets properly destroyed if it is
+ // not trivially destructible in case we are setting VecIter.
+ if (!isSmall)
+ SetIter.~SetIterTy();
+
+ isSmall = Other.isSmall;
+ if (isSmall)
+ VecIter = Other.VecIter;
+ else
+ new (&SetIter) SetIterTy(Other.SetIter);
+ return *this;
+ }
+
+ SmallSetIterator& operator=(SmallSetIterator&& Other) {
+ // Call destructor for SetIter, so it gets properly destroyed if it is
+ // not trivially destructible in case we are setting VecIter.
+ if (!isSmall)
+ SetIter.~SetIterTy();
+
+ isSmall = Other.isSmall;
+ if (isSmall)
+ VecIter = std::move(Other.VecIter);
+ else
+ new (&SetIter) SetIterTy(std::move(Other.SetIter));
+ return *this;
+ }
+
+ bool operator==(const SmallSetIterator &RHS) const {
+ if (isSmall != RHS.isSmall)
+ return false;
+ if (isSmall)
+ return VecIter == RHS.VecIter;
+ return SetIter == RHS.SetIter;
+ }
+
+ SmallSetIterator &operator++() { // Preincrement
+ if (isSmall)
+ VecIter++;
+ else
+ SetIter++;
+ return *this;
+ }
+
+ const T &operator*() const { return isSmall ? *VecIter : *SetIter; }
+};
+
+/// SmallSet - This maintains a set of unique values, optimizing for the case
+/// when the set is small (less than N). In this case, the set can be
+/// maintained with no mallocs. If the set gets large, we expand to using an
+/// std::set to maintain reasonable lookup times.
+template <typename T, unsigned N, typename C = std::less<T>>
+class SmallSet {
+ /// Use a SmallVector to hold the elements here (even though it will never
+ /// reach its 'large' stage) to avoid calling the default ctors of elements
+ /// we will never use.
+ SmallVector<T, N> Vector;
+ std::set<T, C> Set;
+
+ using VIterator = typename SmallVector<T, N>::const_iterator;
+ using mutable_iterator = typename SmallVector<T, N>::iterator;
+
+ // In small mode SmallPtrSet uses linear search for the elements, so it is
+ // not a good idea to choose this value too high. You may consider using a
+ // DenseSet<> instead if you expect many elements in the set.
+ static_assert(N <= 32, "N should be small");
+
+public:
+ using size_type = size_t;
+ using const_iterator = SmallSetIterator<T, N, C>;
+
+ SmallSet() = default;
+
+ LLVM_NODISCARD bool empty() const {
+ return Vector.empty() && Set.empty();
+ }
+
+ size_type size() const {
+ return isSmall() ? Vector.size() : Set.size();
+ }
+
+ /// count - Return 1 if the element is in the set, 0 otherwise.
+ size_type count(const T &V) const {
+ if (isSmall()) {
+ // Since the collection is small, just do a linear search.
+ return vfind(V) == Vector.end() ? 0 : 1;
+ } else {
+ return Set.count(V);
+ }
+ }
+
+ /// insert - Insert an element into the set if it isn't already there.
+ /// Returns true if the element is inserted (it was not in the set before).
+ /// The first value of the returned pair is unused and provided for
+ /// partial compatibility with the standard library self-associative container
+ /// concept.
+ // FIXME: Add iterators that abstract over the small and large form, and then
+ // return those here.
+ std::pair<NoneType, bool> insert(const T &V) {
+ if (!isSmall())
+ return std::make_pair(None, Set.insert(V).second);
+
+ VIterator I = vfind(V);
+ if (I != Vector.end()) // Don't reinsert if it already exists.
+ return std::make_pair(None, false);
+ if (Vector.size() < N) {
+ Vector.push_back(V);
+ return std::make_pair(None, true);
+ }
+
+ // Otherwise, grow from vector to set.
+ while (!Vector.empty()) {
+ Set.insert(Vector.back());
+ Vector.pop_back();
+ }
+ Set.insert(V);
+ return std::make_pair(None, true);
+ }
+
+ template <typename IterT>
+ void insert(IterT I, IterT E) {
+ for (; I != E; ++I)
+ insert(*I);
+ }
+
+ bool erase(const T &V) {
+ if (!isSmall())
+ return Set.erase(V);
+ for (mutable_iterator I = Vector.begin(), E = Vector.end(); I != E; ++I)
+ if (*I == V) {
+ Vector.erase(I);
+ return true;
+ }
+ return false;
+ }
+
+ void clear() {
+ Vector.clear();
+ Set.clear();
+ }
+
+ const_iterator begin() const {
+ if (isSmall())
+ return {Vector.begin()};
+ return {Set.begin()};
+ }
+
+ const_iterator end() const {
+ if (isSmall())
+ return {Vector.end()};
+ return {Set.end()};
+ }
+
+private:
+ bool isSmall() const { return Set.empty(); }
+
+ VIterator vfind(const T &V) const {
+ for (VIterator I = Vector.begin(), E = Vector.end(); I != E; ++I)
+ if (*I == V)
+ return I;
+ return Vector.end();
+ }
+};
+
+/// If this set is of pointer values, transparently switch over to using
+/// SmallPtrSet for performance.
+template <typename PointeeType, unsigned N>
+class SmallSet<PointeeType*, N> : public SmallPtrSet<PointeeType*, N> {};
+
+/// Equality comparison for SmallSet.
+///
+/// Iterates over elements of LHS confirming that each element is also a member
+/// of RHS, and that RHS contains no additional values.
+/// Equivalent to N calls to RHS.count.
+/// For small-set mode amortized complexity is O(N^2)
+/// For large-set mode amortized complexity is linear, worst case is O(N^2) (if
+/// every hash collides).
+template <typename T, unsigned LN, unsigned RN, typename C>
+bool operator==(const SmallSet<T, LN, C> &LHS, const SmallSet<T, RN, C> &RHS) {
+ if (LHS.size() != RHS.size())
+ return false;
+
+ // All elements in LHS must also be in RHS
+ return all_of(LHS, [&RHS](const T &E) { return RHS.count(E); });
+}
+
+/// Inequality comparison for SmallSet.
+///
+/// Equivalent to !(LHS == RHS). See operator== for performance notes.
+template <typename T, unsigned LN, unsigned RN, typename C>
+bool operator!=(const SmallSet<T, LN, C> &LHS, const SmallSet<T, RN, C> &RHS) {
+ return !(LHS == RHS);
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_SMALLSET_H
diff --git a/third_party/llvm-project/include/llvm/ADT/SmallString.h b/third_party/llvm-project/include/llvm/ADT/SmallString.h
new file mode 100644
index 000000000..898be80d0
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/SmallString.h
@@ -0,0 +1,296 @@
+//===- llvm/ADT/SmallString.h - 'Normally small' strings --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the SmallString class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_SMALLSTRING_H
+#define LLVM_ADT_SMALLSTRING_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include <cstddef>
+
+namespace llvm {
+
+/// SmallString - A SmallString is just a SmallVector with methods and accessors
+/// that make it work better as a string (e.g. operator+ etc).
+template<unsigned InternalLen>
+class SmallString : public SmallVector<char, InternalLen> {
+public:
+ /// Default ctor - Initialize to empty.
+ SmallString() = default;
+
+ /// Initialize from a StringRef.
+ SmallString(StringRef S) : SmallVector<char, InternalLen>(S.begin(), S.end()) {}
+
+ /// Initialize with a range.
+ template<typename ItTy>
+ SmallString(ItTy S, ItTy E) : SmallVector<char, InternalLen>(S, E) {}
+
+ // Note that in order to add new overloads for append & assign, we have to
+ // duplicate the inherited versions so as not to inadvertently hide them.
+
+ /// @}
+ /// @name String Assignment
+ /// @{
+
+ /// Assign from a repeated element.
+ void assign(size_t NumElts, char Elt) {
+ this->SmallVectorImpl<char>::assign(NumElts, Elt);
+ }
+
+ /// Assign from an iterator pair.
+ template<typename in_iter>
+ void assign(in_iter S, in_iter E) {
+ this->clear();
+ SmallVectorImpl<char>::append(S, E);
+ }
+
+ /// Assign from a StringRef.
+ void assign(StringRef RHS) {
+ this->clear();
+ SmallVectorImpl<char>::append(RHS.begin(), RHS.end());
+ }
+
+ /// Assign from a SmallVector.
+ void assign(const SmallVectorImpl<char> &RHS) {
+ this->clear();
+ SmallVectorImpl<char>::append(RHS.begin(), RHS.end());
+ }
+
+ /// @}
+ /// @name String Concatenation
+ /// @{
+
+ /// Append from an iterator pair.
+ template<typename in_iter>
+ void append(in_iter S, in_iter E) {
+ SmallVectorImpl<char>::append(S, E);
+ }
+
+ void append(size_t NumInputs, char Elt) {
+ SmallVectorImpl<char>::append(NumInputs, Elt);
+ }
+
+ /// Append from a StringRef.
+ void append(StringRef RHS) {
+ SmallVectorImpl<char>::append(RHS.begin(), RHS.end());
+ }
+
+ /// Append from a SmallVector.
+ void append(const SmallVectorImpl<char> &RHS) {
+ SmallVectorImpl<char>::append(RHS.begin(), RHS.end());
+ }
+
+ /// @}
+ /// @name String Comparison
+ /// @{
+
+ /// Check for string equality. This is more efficient than compare() when
+ /// the relative ordering of inequal strings isn't needed.
+ bool equals(StringRef RHS) const {
+ return str().equals(RHS);
+ }
+
+ /// Check for string equality, ignoring case.
+ bool equals_lower(StringRef RHS) const {
+ return str().equals_lower(RHS);
+ }
+
+ /// Compare two strings; the result is -1, 0, or 1 if this string is
+ /// lexicographically less than, equal to, or greater than the \p RHS.
+ int compare(StringRef RHS) const {
+ return str().compare(RHS);
+ }
+
+ /// compare_lower - Compare two strings, ignoring case.
+ int compare_lower(StringRef RHS) const {
+ return str().compare_lower(RHS);
+ }
+
+ /// compare_numeric - Compare two strings, treating sequences of digits as
+ /// numbers.
+ int compare_numeric(StringRef RHS) const {
+ return str().compare_numeric(RHS);
+ }
+
+ /// @}
+ /// @name String Predicates
+ /// @{
+
+ /// startswith - Check if this string starts with the given \p Prefix.
+ bool startswith(StringRef Prefix) const {
+ return str().startswith(Prefix);
+ }
+
+ /// endswith - Check if this string ends with the given \p Suffix.
+ bool endswith(StringRef Suffix) const {
+ return str().endswith(Suffix);
+ }
+
+ /// @}
+ /// @name String Searching
+ /// @{
+
+ /// find - Search for the first character \p C in the string.
+ ///
+ /// \return - The index of the first occurrence of \p C, or npos if not
+ /// found.
+ size_t find(char C, size_t From = 0) const {
+ return str().find(C, From);
+ }
+
+ /// Search for the first string \p Str in the string.
+ ///
+ /// \returns The index of the first occurrence of \p Str, or npos if not
+ /// found.
+ size_t find(StringRef Str, size_t From = 0) const {
+ return str().find(Str, From);
+ }
+
+ /// Search for the last character \p C in the string.
+ ///
+ /// \returns The index of the last occurrence of \p C, or npos if not
+ /// found.
+ size_t rfind(char C, size_t From = StringRef::npos) const {
+ return str().rfind(C, From);
+ }
+
+ /// Search for the last string \p Str in the string.
+ ///
+ /// \returns The index of the last occurrence of \p Str, or npos if not
+ /// found.
+ size_t rfind(StringRef Str) const {
+ return str().rfind(Str);
+ }
+
+ /// Find the first character in the string that is \p C, or npos if not
+ /// found. Same as find.
+ size_t find_first_of(char C, size_t From = 0) const {
+ return str().find_first_of(C, From);
+ }
+
+ /// Find the first character in the string that is in \p Chars, or npos if
+ /// not found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ size_t find_first_of(StringRef Chars, size_t From = 0) const {
+ return str().find_first_of(Chars, From);
+ }
+
+ /// Find the first character in the string that is not \p C or npos if not
+ /// found.
+ size_t find_first_not_of(char C, size_t From = 0) const {
+ return str().find_first_not_of(C, From);
+ }
+
+ /// Find the first character in the string that is not in the string
+ /// \p Chars, or npos if not found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ size_t find_first_not_of(StringRef Chars, size_t From = 0) const {
+ return str().find_first_not_of(Chars, From);
+ }
+
+ /// Find the last character in the string that is \p C, or npos if not
+ /// found.
+ size_t find_last_of(char C, size_t From = StringRef::npos) const {
+ return str().find_last_of(C, From);
+ }
+
+ /// Find the last character in the string that is in \p C, or npos if not
+ /// found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ size_t find_last_of(
+ StringRef Chars, size_t From = StringRef::npos) const {
+ return str().find_last_of(Chars, From);
+ }
+
+ /// @}
+ /// @name Helpful Algorithms
+ /// @{
+
+ /// Return the number of occurrences of \p C in the string.
+ size_t count(char C) const {
+ return str().count(C);
+ }
+
+ /// Return the number of non-overlapped occurrences of \p Str in the
+ /// string.
+ size_t count(StringRef Str) const {
+ return str().count(Str);
+ }
+
+ /// @}
+ /// @name Substring Operations
+ /// @{
+
+ /// Return a reference to the substring from [Start, Start + N).
+ ///
+ /// \param Start The index of the starting character in the substring; if
+ /// the index is npos or greater than the length of the string then the
+ /// empty substring will be returned.
+ ///
+ /// \param N The number of characters to included in the substring. If \p N
+ /// exceeds the number of characters remaining in the string, the string
+ /// suffix (starting with \p Start) will be returned.
+ StringRef substr(size_t Start, size_t N = StringRef::npos) const {
+ return str().substr(Start, N);
+ }
+
+ /// Return a reference to the substring from [Start, End).
+ ///
+ /// \param Start The index of the starting character in the substring; if
+ /// the index is npos or greater than the length of the string then the
+ /// empty substring will be returned.
+ ///
+ /// \param End The index following the last character to include in the
+ /// substring. If this is npos, or less than \p Start, or exceeds the
+ /// number of characters remaining in the string, the string suffix
+ /// (starting with \p Start) will be returned.
+ StringRef slice(size_t Start, size_t End) const {
+ return str().slice(Start, End);
+ }
+
+ // Extra methods.
+
+ /// Explicit conversion to StringRef.
+ StringRef str() const { return StringRef(this->begin(), this->size()); }
+
+ // TODO: Make this const, if it's safe...
+ const char* c_str() {
+ this->push_back(0);
+ this->pop_back();
+ return this->data();
+ }
+
+ /// Implicit conversion to StringRef.
+ operator StringRef() const { return str(); }
+
+ // Extra operators.
+ const SmallString &operator=(StringRef RHS) {
+ this->clear();
+ return *this += RHS;
+ }
+
+ SmallString &operator+=(StringRef RHS) {
+ this->append(RHS.begin(), RHS.end());
+ return *this;
+ }
+ SmallString &operator+=(char C) {
+ this->push_back(C);
+ return *this;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_SMALLSTRING_H
diff --git a/third_party/llvm-project/include/llvm/ADT/SmallVector.h b/third_party/llvm-project/include/llvm/ADT/SmallVector.h
new file mode 100644
index 000000000..17586904d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/SmallVector.h
@@ -0,0 +1,930 @@
+//===- llvm/ADT/SmallVector.h - 'Normally small' vectors --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the SmallVector class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_SMALLVECTOR_H
+#define LLVM_ADT_SMALLVECTOR_H
+
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/AlignOf.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/MemAlloc.h"
+#include "llvm/Support/type_traits.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdlib>
+#include <cstring>
+#include <initializer_list>
+#include <iterator>
+#include <memory>
+#include <new>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// This is all the non-templated stuff common to all SmallVectors.
+class SmallVectorBase {
+protected:
+ void *BeginX;
+ unsigned Size = 0, Capacity;
+
+ SmallVectorBase() = delete;
+ SmallVectorBase(void *FirstEl, size_t TotalCapacity)
+ : BeginX(FirstEl), Capacity(TotalCapacity) {}
+
+ /// This is an implementation of the grow() method which only works
+ /// on POD-like data types and is out of line to reduce code duplication.
+ void grow_pod(void *FirstEl, size_t MinCapacity, size_t TSize);
+
+public:
+ size_t size() const { return Size; }
+ size_t capacity() const { return Capacity; }
+
+ LLVM_NODISCARD bool empty() const { return !Size; }
+
+ /// Set the array size to \p N, which the current array must have enough
+ /// capacity for.
+ ///
+ /// This does not construct or destroy any elements in the vector.
+ ///
+ /// Clients can use this in conjunction with capacity() to write past the end
+ /// of the buffer when they know that more elements are available, and only
+ /// update the size later. This avoids the cost of value initializing elements
+ /// which will only be overwritten.
+ void set_size(size_t N) {
+ assert(N <= capacity());
+ Size = N;
+ }
+};
+
+/// Figure out the offset of the first element.
+template <class T, typename = void> struct SmallVectorAlignmentAndSize {
+ AlignedCharArrayUnion<SmallVectorBase> Base;
+ AlignedCharArrayUnion<T> FirstEl;
+};
+
+/// This is the part of SmallVectorTemplateBase which does not depend on whether
+/// the type T is a POD. The extra dummy template argument is used by ArrayRef
+/// to avoid unnecessarily requiring T to be complete.
+template <typename T, typename = void>
+class SmallVectorTemplateCommon : public SmallVectorBase {
+ /// Find the address of the first element. For this pointer math to be valid
+ /// with small-size of 0 for T with lots of alignment, it's important that
+ /// SmallVectorStorage is properly-aligned even for small-size of 0.
+ void *getFirstEl() const {
+ return const_cast<void *>(reinterpret_cast<const void *>(
+ reinterpret_cast<const char *>(this) +
+ offsetof(SmallVectorAlignmentAndSize<T>, FirstEl)));
+ }
+ // Space after 'FirstEl' is clobbered, do not add any instance vars after it.
+
+protected:
+ SmallVectorTemplateCommon(size_t Size)
+ : SmallVectorBase(getFirstEl(), Size) {}
+
+ void grow_pod(size_t MinCapacity, size_t TSize) {
+ SmallVectorBase::grow_pod(getFirstEl(), MinCapacity, TSize);
+ }
+
+ /// Return true if this is a smallvector which has not had dynamic
+ /// memory allocated for it.
+ bool isSmall() const { return BeginX == getFirstEl(); }
+
+ /// Put this vector in a state of being small.
+ void resetToSmall() {
+ BeginX = getFirstEl();
+ Size = Capacity = 0; // FIXME: Setting Capacity to 0 is suspect.
+ }
+
+public:
+ using size_type = size_t;
+ using difference_type = ptrdiff_t;
+ using value_type = T;
+ using iterator = T *;
+ using const_iterator = const T *;
+
+ using const_reverse_iterator = std::reverse_iterator<const_iterator>;
+ using reverse_iterator = std::reverse_iterator<iterator>;
+
+ using reference = T &;
+ using const_reference = const T &;
+ using pointer = T *;
+ using const_pointer = const T *;
+
+ // forward iterator creation methods.
+ iterator begin() { return (iterator)this->BeginX; }
+ const_iterator begin() const { return (const_iterator)this->BeginX; }
+ iterator end() { return begin() + size(); }
+ const_iterator end() const { return begin() + size(); }
+
+ // reverse iterator creation methods.
+ reverse_iterator rbegin() { return reverse_iterator(end()); }
+ const_reverse_iterator rbegin() const{ return const_reverse_iterator(end()); }
+ reverse_iterator rend() { return reverse_iterator(begin()); }
+ const_reverse_iterator rend() const { return const_reverse_iterator(begin());}
+
+ size_type size_in_bytes() const { return size() * sizeof(T); }
+ size_type max_size() const { return size_type(-1) / sizeof(T); }
+
+ size_t capacity_in_bytes() const { return capacity() * sizeof(T); }
+
+ /// Return a pointer to the vector's buffer, even if empty().
+ pointer data() { return pointer(begin()); }
+ /// Return a pointer to the vector's buffer, even if empty().
+ const_pointer data() const { return const_pointer(begin()); }
+
+ reference operator[](size_type idx) {
+ assert(idx < size());
+ return begin()[idx];
+ }
+ const_reference operator[](size_type idx) const {
+ assert(idx < size());
+ return begin()[idx];
+ }
+
+ reference front() {
+ assert(!empty());
+ return begin()[0];
+ }
+ const_reference front() const {
+ assert(!empty());
+ return begin()[0];
+ }
+
+ reference back() {
+ assert(!empty());
+ return end()[-1];
+ }
+ const_reference back() const {
+ assert(!empty());
+ return end()[-1];
+ }
+};
+
+/// SmallVectorTemplateBase<TriviallyCopyable = false> - This is where we put method
+/// implementations that are designed to work with non-POD-like T's.
+template <typename T, bool = is_trivially_copyable<T>::value>
+class SmallVectorTemplateBase : public SmallVectorTemplateCommon<T> {
+protected:
+ SmallVectorTemplateBase(size_t Size) : SmallVectorTemplateCommon<T>(Size) {}
+
+ static void destroy_range(T *S, T *E) {
+ while (S != E) {
+ --E;
+ E->~T();
+ }
+ }
+
+ /// Move the range [I, E) into the uninitialized memory starting with "Dest",
+ /// constructing elements as needed.
+ template<typename It1, typename It2>
+ static void uninitialized_move(It1 I, It1 E, It2 Dest) {
+ std::uninitialized_copy(std::make_move_iterator(I),
+ std::make_move_iterator(E), Dest);
+ }
+
+ /// Copy the range [I, E) onto the uninitialized memory starting with "Dest",
+ /// constructing elements as needed.
+ template<typename It1, typename It2>
+ static void uninitialized_copy(It1 I, It1 E, It2 Dest) {
+ std::uninitialized_copy(I, E, Dest);
+ }
+
+ /// Grow the allocated memory (without initializing new elements), doubling
+ /// the size of the allocated memory. Guarantees space for at least one more
+ /// element, or MinSize more elements if specified.
+ void grow(size_t MinSize = 0);
+
+public:
+ void push_back(const T &Elt) {
+ if (LLVM_UNLIKELY(this->size() >= this->capacity()))
+ this->grow();
+ ::new ((void*) this->end()) T(Elt);
+ this->set_size(this->size() + 1);
+ }
+
+ void push_back(T &&Elt) {
+ if (LLVM_UNLIKELY(this->size() >= this->capacity()))
+ this->grow();
+ ::new ((void*) this->end()) T(::std::move(Elt));
+ this->set_size(this->size() + 1);
+ }
+
+ void pop_back() {
+ this->set_size(this->size() - 1);
+ this->end()->~T();
+ }
+};
+
+// Define this out-of-line to dissuade the C++ compiler from inlining it.
+template <typename T, bool TriviallyCopyable>
+void SmallVectorTemplateBase<T, TriviallyCopyable>::grow(size_t MinSize) {
+ if (MinSize > UINT32_MAX)
+ report_bad_alloc_error("SmallVector capacity overflow during allocation");
+
+ // Always grow, even from zero.
+ size_t NewCapacity = size_t(NextPowerOf2(this->capacity() + 2));
+ NewCapacity = std::min(std::max(NewCapacity, MinSize), size_t(UINT32_MAX));
+ T *NewElts = static_cast<T*>(llvm::safe_malloc(NewCapacity*sizeof(T)));
+
+ // Move the elements over.
+ this->uninitialized_move(this->begin(), this->end(), NewElts);
+
+ // Destroy the original elements.
+ destroy_range(this->begin(), this->end());
+
+ // If this wasn't grown from the inline copy, deallocate the old space.
+ if (!this->isSmall())
+ free(this->begin());
+
+ this->BeginX = NewElts;
+ this->Capacity = NewCapacity;
+}
+
+/// SmallVectorTemplateBase<TriviallyCopyable = true> - This is where we put
+/// method implementations that are designed to work with POD-like T's.
+template <typename T>
+class SmallVectorTemplateBase<T, true> : public SmallVectorTemplateCommon<T> {
+protected:
+ SmallVectorTemplateBase(size_t Size) : SmallVectorTemplateCommon<T>(Size) {}
+
+ // No need to do a destroy loop for POD's.
+ static void destroy_range(T *, T *) {}
+
+ /// Move the range [I, E) onto the uninitialized memory
+ /// starting with "Dest", constructing elements into it as needed.
+ template<typename It1, typename It2>
+ static void uninitialized_move(It1 I, It1 E, It2 Dest) {
+ // Just do a copy.
+ uninitialized_copy(I, E, Dest);
+ }
+
+ /// Copy the range [I, E) onto the uninitialized memory
+ /// starting with "Dest", constructing elements into it as needed.
+ template<typename It1, typename It2>
+ static void uninitialized_copy(It1 I, It1 E, It2 Dest) {
+ // Arbitrary iterator types; just use the basic implementation.
+ std::uninitialized_copy(I, E, Dest);
+ }
+
+ /// Copy the range [I, E) onto the uninitialized memory
+ /// starting with "Dest", constructing elements into it as needed.
+ template <typename T1, typename T2>
+ static void uninitialized_copy(
+ T1 *I, T1 *E, T2 *Dest,
+ typename std::enable_if<std::is_same<typename std::remove_const<T1>::type,
+ T2>::value>::type * = nullptr) {
+ // Use memcpy for PODs iterated by pointers (which includes SmallVector
+ // iterators): std::uninitialized_copy optimizes to memmove, but we can
+ // use memcpy here. Note that I and E are iterators and thus might be
+ // invalid for memcpy if they are equal.
+ if (I != E)
+ memcpy(reinterpret_cast<void *>(Dest), I, (E - I) * sizeof(T));
+ }
+
+ /// Double the size of the allocated memory, guaranteeing space for at
+ /// least one more element or MinSize if specified.
+ void grow(size_t MinSize = 0) { this->grow_pod(MinSize, sizeof(T)); }
+
+public:
+ void push_back(const T &Elt) {
+ if (LLVM_UNLIKELY(this->size() >= this->capacity()))
+ this->grow();
+ memcpy(reinterpret_cast<void *>(this->end()), &Elt, sizeof(T));
+ this->set_size(this->size() + 1);
+ }
+
+ void pop_back() { this->set_size(this->size() - 1); }
+};
+
+/// This class consists of common code factored out of the SmallVector class to
+/// reduce code duplication based on the SmallVector 'N' template parameter.
+template <typename T>
+class SmallVectorImpl : public SmallVectorTemplateBase<T> {
+ using SuperClass = SmallVectorTemplateBase<T>;
+
+public:
+ using iterator = typename SuperClass::iterator;
+ using const_iterator = typename SuperClass::const_iterator;
+ using reference = typename SuperClass::reference;
+ using size_type = typename SuperClass::size_type;
+
+protected:
+ // Default ctor - Initialize to empty.
+ explicit SmallVectorImpl(unsigned N)
+ : SmallVectorTemplateBase<T>(N) {}
+
+public:
+ SmallVectorImpl(const SmallVectorImpl &) = delete;
+
+ ~SmallVectorImpl() {
+ // Subclass has already destructed this vector's elements.
+ // If this wasn't grown from the inline copy, deallocate the old space.
+ if (!this->isSmall())
+ free(this->begin());
+ }
+
+ void clear() {
+ this->destroy_range(this->begin(), this->end());
+ this->Size = 0;
+ }
+
+ void resize(size_type N) {
+ if (N < this->size()) {
+ this->destroy_range(this->begin()+N, this->end());
+ this->set_size(N);
+ } else if (N > this->size()) {
+ if (this->capacity() < N)
+ this->grow(N);
+ for (auto I = this->end(), E = this->begin() + N; I != E; ++I)
+ new (&*I) T();
+ this->set_size(N);
+ }
+ }
+
+ void resize(size_type N, const T &NV) {
+ if (N < this->size()) {
+ this->destroy_range(this->begin()+N, this->end());
+ this->set_size(N);
+ } else if (N > this->size()) {
+ if (this->capacity() < N)
+ this->grow(N);
+ std::uninitialized_fill(this->end(), this->begin()+N, NV);
+ this->set_size(N);
+ }
+ }
+
+ void reserve(size_type N) {
+ if (this->capacity() < N)
+ this->grow(N);
+ }
+
+ LLVM_NODISCARD T pop_back_val() {
+ T Result = ::std::move(this->back());
+ this->pop_back();
+ return Result;
+ }
+
+ void swap(SmallVectorImpl &RHS);
+
+ /// Add the specified range to the end of the SmallVector.
+ template <typename in_iter,
+ typename = typename std::enable_if<std::is_convertible<
+ typename std::iterator_traits<in_iter>::iterator_category,
+ std::input_iterator_tag>::value>::type>
+ void append(in_iter in_start, in_iter in_end) {
+ size_type NumInputs = std::distance(in_start, in_end);
+ if (NumInputs > this->capacity() - this->size())
+ this->grow(this->size()+NumInputs);
+
+ this->uninitialized_copy(in_start, in_end, this->end());
+ this->set_size(this->size() + NumInputs);
+ }
+
+ /// Append \p NumInputs copies of \p Elt to the end.
+ void append(size_type NumInputs, const T &Elt) {
+ if (NumInputs > this->capacity() - this->size())
+ this->grow(this->size()+NumInputs);
+
+ std::uninitialized_fill_n(this->end(), NumInputs, Elt);
+ this->set_size(this->size() + NumInputs);
+ }
+
+ void append(std::initializer_list<T> IL) {
+ append(IL.begin(), IL.end());
+ }
+
+ // FIXME: Consider assigning over existing elements, rather than clearing &
+ // re-initializing them - for all assign(...) variants.
+
+ void assign(size_type NumElts, const T &Elt) {
+ clear();
+ if (this->capacity() < NumElts)
+ this->grow(NumElts);
+ this->set_size(NumElts);
+ std::uninitialized_fill(this->begin(), this->end(), Elt);
+ }
+
+ template <typename in_iter,
+ typename = typename std::enable_if<std::is_convertible<
+ typename std::iterator_traits<in_iter>::iterator_category,
+ std::input_iterator_tag>::value>::type>
+ void assign(in_iter in_start, in_iter in_end) {
+ clear();
+ append(in_start, in_end);
+ }
+
+ void assign(std::initializer_list<T> IL) {
+ clear();
+ append(IL);
+ }
+
+ iterator erase(const_iterator CI) {
+ // Just cast away constness because this is a non-const member function.
+ iterator I = const_cast<iterator>(CI);
+
+ assert(I >= this->begin() && "Iterator to erase is out of bounds.");
+ assert(I < this->end() && "Erasing at past-the-end iterator.");
+
+ iterator N = I;
+ // Shift all elts down one.
+ std::move(I+1, this->end(), I);
+ // Drop the last elt.
+ this->pop_back();
+ return(N);
+ }
+
+ iterator erase(const_iterator CS, const_iterator CE) {
+ // Just cast away constness because this is a non-const member function.
+ iterator S = const_cast<iterator>(CS);
+ iterator E = const_cast<iterator>(CE);
+
+ assert(S >= this->begin() && "Range to erase is out of bounds.");
+ assert(S <= E && "Trying to erase invalid range.");
+ assert(E <= this->end() && "Trying to erase past the end.");
+
+ iterator N = S;
+ // Shift all elts down.
+ iterator I = std::move(E, this->end(), S);
+ // Drop the last elts.
+ this->destroy_range(I, this->end());
+ this->set_size(I - this->begin());
+ return(N);
+ }
+
+ iterator insert(iterator I, T &&Elt) {
+ if (I == this->end()) { // Important special case for empty vector.
+ this->push_back(::std::move(Elt));
+ return this->end()-1;
+ }
+
+ assert(I >= this->begin() && "Insertion iterator is out of bounds.");
+ assert(I <= this->end() && "Inserting past the end of the vector.");
+
+ if (this->size() >= this->capacity()) {
+ size_t EltNo = I-this->begin();
+ this->grow();
+ I = this->begin()+EltNo;
+ }
+
+ ::new ((void*) this->end()) T(::std::move(this->back()));
+ // Push everything else over.
+ std::move_backward(I, this->end()-1, this->end());
+ this->set_size(this->size() + 1);
+
+ // If we just moved the element we're inserting, be sure to update
+ // the reference.
+ T *EltPtr = &Elt;
+ if (I <= EltPtr && EltPtr < this->end())
+ ++EltPtr;
+
+ *I = ::std::move(*EltPtr);
+ return I;
+ }
+
+ iterator insert(iterator I, const T &Elt) {
+ if (I == this->end()) { // Important special case for empty vector.
+ this->push_back(Elt);
+ return this->end()-1;
+ }
+
+ assert(I >= this->begin() && "Insertion iterator is out of bounds.");
+ assert(I <= this->end() && "Inserting past the end of the vector.");
+
+ if (this->size() >= this->capacity()) {
+ size_t EltNo = I-this->begin();
+ this->grow();
+ I = this->begin()+EltNo;
+ }
+ ::new ((void*) this->end()) T(std::move(this->back()));
+ // Push everything else over.
+ std::move_backward(I, this->end()-1, this->end());
+ this->set_size(this->size() + 1);
+
+ // If we just moved the element we're inserting, be sure to update
+ // the reference.
+ const T *EltPtr = &Elt;
+ if (I <= EltPtr && EltPtr < this->end())
+ ++EltPtr;
+
+ *I = *EltPtr;
+ return I;
+ }
+
+ iterator insert(iterator I, size_type NumToInsert, const T &Elt) {
+ // Convert iterator to elt# to avoid invalidating iterator when we reserve()
+ size_t InsertElt = I - this->begin();
+
+ if (I == this->end()) { // Important special case for empty vector.
+ append(NumToInsert, Elt);
+ return this->begin()+InsertElt;
+ }
+
+ assert(I >= this->begin() && "Insertion iterator is out of bounds.");
+ assert(I <= this->end() && "Inserting past the end of the vector.");
+
+ // Ensure there is enough space.
+ reserve(this->size() + NumToInsert);
+
+ // Uninvalidate the iterator.
+ I = this->begin()+InsertElt;
+
+ // If there are more elements between the insertion point and the end of the
+ // range than there are being inserted, we can use a simple approach to
+ // insertion. Since we already reserved space, we know that this won't
+ // reallocate the vector.
+ if (size_t(this->end()-I) >= NumToInsert) {
+ T *OldEnd = this->end();
+ append(std::move_iterator<iterator>(this->end() - NumToInsert),
+ std::move_iterator<iterator>(this->end()));
+
+ // Copy the existing elements that get replaced.
+ std::move_backward(I, OldEnd-NumToInsert, OldEnd);
+
+ std::fill_n(I, NumToInsert, Elt);
+ return I;
+ }
+
+ // Otherwise, we're inserting more elements than exist already, and we're
+ // not inserting at the end.
+
+ // Move over the elements that we're about to overwrite.
+ T *OldEnd = this->end();
+ this->set_size(this->size() + NumToInsert);
+ size_t NumOverwritten = OldEnd-I;
+ this->uninitialized_move(I, OldEnd, this->end()-NumOverwritten);
+
+ // Replace the overwritten part.
+ std::fill_n(I, NumOverwritten, Elt);
+
+ // Insert the non-overwritten middle part.
+ std::uninitialized_fill_n(OldEnd, NumToInsert-NumOverwritten, Elt);
+ return I;
+ }
+
+ template <typename ItTy,
+ typename = typename std::enable_if<std::is_convertible<
+ typename std::iterator_traits<ItTy>::iterator_category,
+ std::input_iterator_tag>::value>::type>
+ iterator insert(iterator I, ItTy From, ItTy To) {
+ // Convert iterator to elt# to avoid invalidating iterator when we reserve()
+ size_t InsertElt = I - this->begin();
+
+ if (I == this->end()) { // Important special case for empty vector.
+ append(From, To);
+ return this->begin()+InsertElt;
+ }
+
+ assert(I >= this->begin() && "Insertion iterator is out of bounds.");
+ assert(I <= this->end() && "Inserting past the end of the vector.");
+
+ size_t NumToInsert = std::distance(From, To);
+
+ // Ensure there is enough space.
+ reserve(this->size() + NumToInsert);
+
+ // Uninvalidate the iterator.
+ I = this->begin()+InsertElt;
+
+ // If there are more elements between the insertion point and the end of the
+ // range than there are being inserted, we can use a simple approach to
+ // insertion. Since we already reserved space, we know that this won't
+ // reallocate the vector.
+ if (size_t(this->end()-I) >= NumToInsert) {
+ T *OldEnd = this->end();
+ append(std::move_iterator<iterator>(this->end() - NumToInsert),
+ std::move_iterator<iterator>(this->end()));
+
+ // Copy the existing elements that get replaced.
+ std::move_backward(I, OldEnd-NumToInsert, OldEnd);
+
+ std::copy(From, To, I);
+ return I;
+ }
+
+ // Otherwise, we're inserting more elements than exist already, and we're
+ // not inserting at the end.
+
+ // Move over the elements that we're about to overwrite.
+ T *OldEnd = this->end();
+ this->set_size(this->size() + NumToInsert);
+ size_t NumOverwritten = OldEnd-I;
+ this->uninitialized_move(I, OldEnd, this->end()-NumOverwritten);
+
+ // Replace the overwritten part.
+ for (T *J = I; NumOverwritten > 0; --NumOverwritten) {
+ *J = *From;
+ ++J; ++From;
+ }
+
+ // Insert the non-overwritten middle part.
+ this->uninitialized_copy(From, To, OldEnd);
+ return I;
+ }
+
+ void insert(iterator I, std::initializer_list<T> IL) {
+ insert(I, IL.begin(), IL.end());
+ }
+
+ template <typename... ArgTypes> reference emplace_back(ArgTypes &&... Args) {
+ if (LLVM_UNLIKELY(this->size() >= this->capacity()))
+ this->grow();
+ ::new ((void *)this->end()) T(std::forward<ArgTypes>(Args)...);
+ this->set_size(this->size() + 1);
+ return this->back();
+ }
+
+ SmallVectorImpl &operator=(const SmallVectorImpl &RHS);
+
+ SmallVectorImpl &operator=(SmallVectorImpl &&RHS);
+
+ bool operator==(const SmallVectorImpl &RHS) const {
+ if (this->size() != RHS.size()) return false;
+ return std::equal(this->begin(), this->end(), RHS.begin());
+ }
+ bool operator!=(const SmallVectorImpl &RHS) const {
+ return !(*this == RHS);
+ }
+
+ bool operator<(const SmallVectorImpl &RHS) const {
+ return std::lexicographical_compare(this->begin(), this->end(),
+ RHS.begin(), RHS.end());
+ }
+};
+
+template <typename T>
+void SmallVectorImpl<T>::swap(SmallVectorImpl<T> &RHS) {
+ if (this == &RHS) return;
+
+ // We can only avoid copying elements if neither vector is small.
+ if (!this->isSmall() && !RHS.isSmall()) {
+ std::swap(this->BeginX, RHS.BeginX);
+ std::swap(this->Size, RHS.Size);
+ std::swap(this->Capacity, RHS.Capacity);
+ return;
+ }
+ if (RHS.size() > this->capacity())
+ this->grow(RHS.size());
+ if (this->size() > RHS.capacity())
+ RHS.grow(this->size());
+
+ // Swap the shared elements.
+ size_t NumShared = this->size();
+ if (NumShared > RHS.size()) NumShared = RHS.size();
+ for (size_type i = 0; i != NumShared; ++i)
+ std::swap((*this)[i], RHS[i]);
+
+ // Copy over the extra elts.
+ if (this->size() > RHS.size()) {
+ size_t EltDiff = this->size() - RHS.size();
+ this->uninitialized_copy(this->begin()+NumShared, this->end(), RHS.end());
+ RHS.set_size(RHS.size() + EltDiff);
+ this->destroy_range(this->begin()+NumShared, this->end());
+ this->set_size(NumShared);
+ } else if (RHS.size() > this->size()) {
+ size_t EltDiff = RHS.size() - this->size();
+ this->uninitialized_copy(RHS.begin()+NumShared, RHS.end(), this->end());
+ this->set_size(this->size() + EltDiff);
+ this->destroy_range(RHS.begin()+NumShared, RHS.end());
+ RHS.set_size(NumShared);
+ }
+}
+
+template <typename T>
+SmallVectorImpl<T> &SmallVectorImpl<T>::
+ operator=(const SmallVectorImpl<T> &RHS) {
+ // Avoid self-assignment.
+ if (this == &RHS) return *this;
+
+ // If we already have sufficient space, assign the common elements, then
+ // destroy any excess.
+ size_t RHSSize = RHS.size();
+ size_t CurSize = this->size();
+ if (CurSize >= RHSSize) {
+ // Assign common elements.
+ iterator NewEnd;
+ if (RHSSize)
+ NewEnd = std::copy(RHS.begin(), RHS.begin()+RHSSize, this->begin());
+ else
+ NewEnd = this->begin();
+
+ // Destroy excess elements.
+ this->destroy_range(NewEnd, this->end());
+
+ // Trim.
+ this->set_size(RHSSize);
+ return *this;
+ }
+
+ // If we have to grow to have enough elements, destroy the current elements.
+ // This allows us to avoid copying them during the grow.
+ // FIXME: don't do this if they're efficiently moveable.
+ if (this->capacity() < RHSSize) {
+ // Destroy current elements.
+ this->destroy_range(this->begin(), this->end());
+ this->set_size(0);
+ CurSize = 0;
+ this->grow(RHSSize);
+ } else if (CurSize) {
+ // Otherwise, use assignment for the already-constructed elements.
+ std::copy(RHS.begin(), RHS.begin()+CurSize, this->begin());
+ }
+
+ // Copy construct the new elements in place.
+ this->uninitialized_copy(RHS.begin()+CurSize, RHS.end(),
+ this->begin()+CurSize);
+
+ // Set end.
+ this->set_size(RHSSize);
+ return *this;
+}
+
+template <typename T>
+SmallVectorImpl<T> &SmallVectorImpl<T>::operator=(SmallVectorImpl<T> &&RHS) {
+ // Avoid self-assignment.
+ if (this == &RHS) return *this;
+
+ // If the RHS isn't small, clear this vector and then steal its buffer.
+ if (!RHS.isSmall()) {
+ this->destroy_range(this->begin(), this->end());
+ if (!this->isSmall()) free(this->begin());
+ this->BeginX = RHS.BeginX;
+ this->Size = RHS.Size;
+ this->Capacity = RHS.Capacity;
+ RHS.resetToSmall();
+ return *this;
+ }
+
+ // If we already have sufficient space, assign the common elements, then
+ // destroy any excess.
+ size_t RHSSize = RHS.size();
+ size_t CurSize = this->size();
+ if (CurSize >= RHSSize) {
+ // Assign common elements.
+ iterator NewEnd = this->begin();
+ if (RHSSize)
+ NewEnd = std::move(RHS.begin(), RHS.end(), NewEnd);
+
+ // Destroy excess elements and trim the bounds.
+ this->destroy_range(NewEnd, this->end());
+ this->set_size(RHSSize);
+
+ // Clear the RHS.
+ RHS.clear();
+
+ return *this;
+ }
+
+ // If we have to grow to have enough elements, destroy the current elements.
+ // This allows us to avoid copying them during the grow.
+ // FIXME: this may not actually make any sense if we can efficiently move
+ // elements.
+ if (this->capacity() < RHSSize) {
+ // Destroy current elements.
+ this->destroy_range(this->begin(), this->end());
+ this->set_size(0);
+ CurSize = 0;
+ this->grow(RHSSize);
+ } else if (CurSize) {
+ // Otherwise, use assignment for the already-constructed elements.
+ std::move(RHS.begin(), RHS.begin()+CurSize, this->begin());
+ }
+
+ // Move-construct the new elements in place.
+ this->uninitialized_move(RHS.begin()+CurSize, RHS.end(),
+ this->begin()+CurSize);
+
+ // Set end.
+ this->set_size(RHSSize);
+
+ RHS.clear();
+ return *this;
+}
+
+/// Storage for the SmallVector elements. This is specialized for the N=0 case
+/// to avoid allocating unnecessary storage.
+template <typename T, unsigned N>
+struct SmallVectorStorage {
+ AlignedCharArrayUnion<T> InlineElts[N];
+};
+
+/// We need the storage to be properly aligned even for small-size of 0 so that
+/// the pointer math in \a SmallVectorTemplateCommon::getFirstEl() is
+/// well-defined.
+template <typename T> struct alignas(alignof(T)) SmallVectorStorage<T, 0> {};
+
+/// This is a 'vector' (really, a variable-sized array), optimized
+/// for the case when the array is small. It contains some number of elements
+/// in-place, which allows it to avoid heap allocation when the actual number of
+/// elements is below that threshold. This allows normal "small" cases to be
+/// fast without losing generality for large inputs.
+///
+/// Note that this does not attempt to be exception safe.
+///
+template <typename T, unsigned N>
+class SmallVector : public SmallVectorImpl<T>, SmallVectorStorage<T, N> {
+public:
+ SmallVector() : SmallVectorImpl<T>(N) {}
+
+ ~SmallVector() {
+ // Destroy the constructed elements in the vector.
+ this->destroy_range(this->begin(), this->end());
+ }
+
+ explicit SmallVector(size_t Size, const T &Value = T())
+ : SmallVectorImpl<T>(N) {
+ this->assign(Size, Value);
+ }
+
+ template <typename ItTy,
+ typename = typename std::enable_if<std::is_convertible<
+ typename std::iterator_traits<ItTy>::iterator_category,
+ std::input_iterator_tag>::value>::type>
+ SmallVector(ItTy S, ItTy E) : SmallVectorImpl<T>(N) {
+ this->append(S, E);
+ }
+
+ template <typename RangeTy>
+ explicit SmallVector(const iterator_range<RangeTy> &R)
+ : SmallVectorImpl<T>(N) {
+ this->append(R.begin(), R.end());
+ }
+
+ SmallVector(std::initializer_list<T> IL) : SmallVectorImpl<T>(N) {
+ this->assign(IL);
+ }
+
+ SmallVector(const SmallVector &RHS) : SmallVectorImpl<T>(N) {
+ if (!RHS.empty())
+ SmallVectorImpl<T>::operator=(RHS);
+ }
+
+ const SmallVector &operator=(const SmallVector &RHS) {
+ SmallVectorImpl<T>::operator=(RHS);
+ return *this;
+ }
+
+ SmallVector(SmallVector &&RHS) : SmallVectorImpl<T>(N) {
+ if (!RHS.empty())
+ SmallVectorImpl<T>::operator=(::std::move(RHS));
+ }
+
+ SmallVector(SmallVectorImpl<T> &&RHS) : SmallVectorImpl<T>(N) {
+ if (!RHS.empty())
+ SmallVectorImpl<T>::operator=(::std::move(RHS));
+ }
+
+ const SmallVector &operator=(SmallVector &&RHS) {
+ SmallVectorImpl<T>::operator=(::std::move(RHS));
+ return *this;
+ }
+
+ const SmallVector &operator=(SmallVectorImpl<T> &&RHS) {
+ SmallVectorImpl<T>::operator=(::std::move(RHS));
+ return *this;
+ }
+
+ const SmallVector &operator=(std::initializer_list<T> IL) {
+ this->assign(IL);
+ return *this;
+ }
+};
+
+template <typename T, unsigned N>
+inline size_t capacity_in_bytes(const SmallVector<T, N> &X) {
+ return X.capacity_in_bytes();
+}
+
+} // end namespace llvm
+
+namespace std {
+
+ /// Implement std::swap in terms of SmallVector swap.
+ template<typename T>
+ inline void
+ swap(llvm::SmallVectorImpl<T> &LHS, llvm::SmallVectorImpl<T> &RHS) {
+ LHS.swap(RHS);
+ }
+
+ /// Implement std::swap in terms of SmallVector swap.
+ template<typename T, unsigned N>
+ inline void
+ swap(llvm::SmallVector<T, N> &LHS, llvm::SmallVector<T, N> &RHS) {
+ LHS.swap(RHS);
+ }
+
+} // end namespace std
+
+#endif // LLVM_ADT_SMALLVECTOR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/StringExtras.h b/third_party/llvm-project/include/llvm/ADT/StringExtras.h
new file mode 100644
index 000000000..ef1a11e06
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/StringExtras.h
@@ -0,0 +1,401 @@
+//===- llvm/ADT/StringExtras.h - Useful string functions --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains some functions that are useful when dealing with strings.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_STRINGEXTRAS_H
+#define LLVM_ADT_STRINGEXTRAS_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstdlib>
+#include <cstring>
+#include <iterator>
+#include <string>
+#include <utility>
+
+namespace llvm {
+
+template<typename T> class SmallVectorImpl;
+class raw_ostream;
+
+/// hexdigit - Return the hexadecimal character for the
+/// given number \p X (which should be less than 16).
+inline char hexdigit(unsigned X, bool LowerCase = false) {
+ const char HexChar = LowerCase ? 'a' : 'A';
+ return X < 10 ? '0' + X : HexChar + X - 10;
+}
+
+/// Given an array of c-style strings terminated by a null pointer, construct
+/// a vector of StringRefs representing the same strings without the terminating
+/// null string.
+inline std::vector<StringRef> toStringRefArray(const char *const *Strings) {
+ std::vector<StringRef> Result;
+ while (*Strings)
+ Result.push_back(*Strings++);
+ return Result;
+}
+
+/// Construct a string ref from a boolean.
+inline StringRef toStringRef(bool B) { return StringRef(B ? "true" : "false"); }
+
+/// Construct a string ref from an array ref of unsigned chars.
+inline StringRef toStringRef(ArrayRef<uint8_t> Input) {
+ return StringRef(reinterpret_cast<const char *>(Input.begin()), Input.size());
+}
+
+/// Construct a string ref from an array ref of unsigned chars.
+inline ArrayRef<uint8_t> arrayRefFromStringRef(StringRef Input) {
+ return {Input.bytes_begin(), Input.bytes_end()};
+}
+
+/// Interpret the given character \p C as a hexadecimal digit and return its
+/// value.
+///
+/// If \p C is not a valid hex digit, -1U is returned.
+inline unsigned hexDigitValue(char C) {
+ if (C >= '0' && C <= '9') return C-'0';
+ if (C >= 'a' && C <= 'f') return C-'a'+10U;
+ if (C >= 'A' && C <= 'F') return C-'A'+10U;
+ return -1U;
+}
+
+/// Checks if character \p C is one of the 10 decimal digits.
+inline bool isDigit(char C) { return C >= '0' && C <= '9'; }
+
+/// Checks if character \p C is a hexadecimal numeric character.
+inline bool isHexDigit(char C) { return hexDigitValue(C) != -1U; }
+
+/// Checks if character \p C is a valid letter as classified by "C" locale.
+inline bool isAlpha(char C) {
+ return ('a' <= C && C <= 'z') || ('A' <= C && C <= 'Z');
+}
+
+/// Checks whether character \p C is either a decimal digit or an uppercase or
+/// lowercase letter as classified by "C" locale.
+inline bool isAlnum(char C) { return isAlpha(C) || isDigit(C); }
+
+/// Checks whether character \p C is valid ASCII (high bit is zero).
+inline bool isASCII(char C) { return static_cast<unsigned char>(C) <= 127; }
+
+/// Checks whether all characters in S are ASCII.
+inline bool isASCII(llvm::StringRef S) {
+ for (char C : S)
+ if (LLVM_UNLIKELY(!isASCII(C)))
+ return false;
+ return true;
+}
+
+/// Checks whether character \p C is printable.
+///
+/// Locale-independent version of the C standard library isprint whose results
+/// may differ on different platforms.
+inline bool isPrint(char C) {
+ unsigned char UC = static_cast<unsigned char>(C);
+ return (0x20 <= UC) && (UC <= 0x7E);
+}
+
+/// Returns the corresponding lowercase character if \p x is uppercase.
+inline char toLower(char x) {
+ if (x >= 'A' && x <= 'Z')
+ return x - 'A' + 'a';
+ return x;
+}
+
+/// Returns the corresponding uppercase character if \p x is lowercase.
+inline char toUpper(char x) {
+ if (x >= 'a' && x <= 'z')
+ return x - 'a' + 'A';
+ return x;
+}
+
+inline std::string utohexstr(uint64_t X, bool LowerCase = false) {
+ char Buffer[17];
+ char *BufPtr = std::end(Buffer);
+
+ if (X == 0) *--BufPtr = '0';
+
+ while (X) {
+ unsigned char Mod = static_cast<unsigned char>(X) & 15;
+ *--BufPtr = hexdigit(Mod, LowerCase);
+ X >>= 4;
+ }
+
+ return std::string(BufPtr, std::end(Buffer));
+}
+
+/// Convert buffer \p Input to its hexadecimal representation.
+/// The returned string is double the size of \p Input.
+inline std::string toHex(StringRef Input, bool LowerCase = false) {
+ static const char *const LUT = "0123456789ABCDEF";
+ const uint8_t Offset = LowerCase ? 32 : 0;
+ size_t Length = Input.size();
+
+ std::string Output;
+ Output.reserve(2 * Length);
+ for (size_t i = 0; i < Length; ++i) {
+ const unsigned char c = Input[i];
+ Output.push_back(LUT[c >> 4] | Offset);
+ Output.push_back(LUT[c & 15] | Offset);
+ }
+ return Output;
+}
+
+inline std::string toHex(ArrayRef<uint8_t> Input, bool LowerCase = false) {
+ return toHex(toStringRef(Input), LowerCase);
+}
+
+inline uint8_t hexFromNibbles(char MSB, char LSB) {
+ unsigned U1 = hexDigitValue(MSB);
+ unsigned U2 = hexDigitValue(LSB);
+ assert(U1 != -1U && U2 != -1U);
+
+ return static_cast<uint8_t>((U1 << 4) | U2);
+}
+
+/// Convert hexadecimal string \p Input to its binary representation.
+/// The return string is half the size of \p Input.
+inline std::string fromHex(StringRef Input) {
+ if (Input.empty())
+ return std::string();
+
+ std::string Output;
+ Output.reserve((Input.size() + 1) / 2);
+ if (Input.size() % 2 == 1) {
+ Output.push_back(hexFromNibbles('0', Input.front()));
+ Input = Input.drop_front();
+ }
+
+ assert(Input.size() % 2 == 0);
+ while (!Input.empty()) {
+ uint8_t Hex = hexFromNibbles(Input[0], Input[1]);
+ Output.push_back(Hex);
+ Input = Input.drop_front(2);
+ }
+ return Output;
+}
+
+/// Convert the string \p S to an integer of the specified type using
+/// the radix \p Base. If \p Base is 0, auto-detects the radix.
+/// Returns true if the number was successfully converted, false otherwise.
+template <typename N> bool to_integer(StringRef S, N &Num, unsigned Base = 0) {
+ return !S.getAsInteger(Base, Num);
+}
+
+namespace detail {
+template <typename N>
+inline bool to_float(const Twine &T, N &Num, N (*StrTo)(const char *, char **)) {
+ SmallString<32> Storage;
+ StringRef S = T.toNullTerminatedStringRef(Storage);
+ char *End;
+ N Temp = StrTo(S.data(), &End);
+ if (*End != '\0')
+ return false;
+ Num = Temp;
+ return true;
+}
+}
+
+inline bool to_float(const Twine &T, float &Num) {
+ return detail::to_float(T, Num, strtof);
+}
+
+inline bool to_float(const Twine &T, double &Num) {
+ return detail::to_float(T, Num, strtod);
+}
+
+inline bool to_float(const Twine &T, long double &Num) {
+ return detail::to_float(T, Num, strtold);
+}
+
+inline std::string utostr(uint64_t X, bool isNeg = false) {
+ char Buffer[21];
+ char *BufPtr = std::end(Buffer);
+
+ if (X == 0) *--BufPtr = '0'; // Handle special case...
+
+ while (X) {
+ *--BufPtr = '0' + char(X % 10);
+ X /= 10;
+ }
+
+ if (isNeg) *--BufPtr = '-'; // Add negative sign...
+ return std::string(BufPtr, std::end(Buffer));
+}
+
+inline std::string itostr(int64_t X) {
+ if (X < 0)
+ return utostr(static_cast<uint64_t>(-X), true);
+ else
+ return utostr(static_cast<uint64_t>(X));
+}
+
+/// StrInStrNoCase - Portable version of strcasestr. Locates the first
+/// occurrence of string 's1' in string 's2', ignoring case. Returns
+/// the offset of s2 in s1 or npos if s2 cannot be found.
+StringRef::size_type StrInStrNoCase(StringRef s1, StringRef s2);
+
+/// getToken - This function extracts one token from source, ignoring any
+/// leading characters that appear in the Delimiters string, and ending the
+/// token at any of the characters that appear in the Delimiters string. If
+/// there are no tokens in the source string, an empty string is returned.
+/// The function returns a pair containing the extracted token and the
+/// remaining tail string.
+std::pair<StringRef, StringRef> getToken(StringRef Source,
+ StringRef Delimiters = " \t\n\v\f\r");
+
+/// SplitString - Split up the specified string according to the specified
+/// delimiters, appending the result fragments to the output list.
+void SplitString(StringRef Source,
+ SmallVectorImpl<StringRef> &OutFragments,
+ StringRef Delimiters = " \t\n\v\f\r");
+
+/// Returns the English suffix for an ordinal integer (-st, -nd, -rd, -th).
+inline StringRef getOrdinalSuffix(unsigned Val) {
+ // It is critically important that we do this perfectly for
+ // user-written sequences with over 100 elements.
+ switch (Val % 100) {
+ case 11:
+ case 12:
+ case 13:
+ return "th";
+ default:
+ switch (Val % 10) {
+ case 1: return "st";
+ case 2: return "nd";
+ case 3: return "rd";
+ default: return "th";
+ }
+ }
+}
+
+/// Print each character of the specified string, escaping it if it is not
+/// printable or if it is an escape char.
+void printEscapedString(StringRef Name, raw_ostream &Out);
+
+/// Print each character of the specified string, escaping HTML special
+/// characters.
+void printHTMLEscaped(StringRef String, raw_ostream &Out);
+
+/// printLowerCase - Print each character as lowercase if it is uppercase.
+void printLowerCase(StringRef String, raw_ostream &Out);
+
+namespace detail {
+
+template <typename IteratorT>
+inline std::string join_impl(IteratorT Begin, IteratorT End,
+ StringRef Separator, std::input_iterator_tag) {
+ std::string S;
+ if (Begin == End)
+ return S;
+
+ S += (*Begin);
+ while (++Begin != End) {
+ S += Separator;
+ S += (*Begin);
+ }
+ return S;
+}
+
+template <typename IteratorT>
+inline std::string join_impl(IteratorT Begin, IteratorT End,
+ StringRef Separator, std::forward_iterator_tag) {
+ std::string S;
+ if (Begin == End)
+ return S;
+
+ size_t Len = (std::distance(Begin, End) - 1) * Separator.size();
+ for (IteratorT I = Begin; I != End; ++I)
+ Len += (*Begin).size();
+ S.reserve(Len);
+ S += (*Begin);
+ while (++Begin != End) {
+ S += Separator;
+ S += (*Begin);
+ }
+ return S;
+}
+
+template <typename Sep>
+inline void join_items_impl(std::string &Result, Sep Separator) {}
+
+template <typename Sep, typename Arg>
+inline void join_items_impl(std::string &Result, Sep Separator,
+ const Arg &Item) {
+ Result += Item;
+}
+
+template <typename Sep, typename Arg1, typename... Args>
+inline void join_items_impl(std::string &Result, Sep Separator, const Arg1 &A1,
+ Args &&... Items) {
+ Result += A1;
+ Result += Separator;
+ join_items_impl(Result, Separator, std::forward<Args>(Items)...);
+}
+
+inline size_t join_one_item_size(char) { return 1; }
+inline size_t join_one_item_size(const char *S) { return S ? ::strlen(S) : 0; }
+
+template <typename T> inline size_t join_one_item_size(const T &Str) {
+ return Str.size();
+}
+
+inline size_t join_items_size() { return 0; }
+
+template <typename A1> inline size_t join_items_size(const A1 &A) {
+ return join_one_item_size(A);
+}
+template <typename A1, typename... Args>
+inline size_t join_items_size(const A1 &A, Args &&... Items) {
+ return join_one_item_size(A) + join_items_size(std::forward<Args>(Items)...);
+}
+
+} // end namespace detail
+
+/// Joins the strings in the range [Begin, End), adding Separator between
+/// the elements.
+template <typename IteratorT>
+inline std::string join(IteratorT Begin, IteratorT End, StringRef Separator) {
+ using tag = typename std::iterator_traits<IteratorT>::iterator_category;
+ return detail::join_impl(Begin, End, Separator, tag());
+}
+
+/// Joins the strings in the range [R.begin(), R.end()), adding Separator
+/// between the elements.
+template <typename Range>
+inline std::string join(Range &&R, StringRef Separator) {
+ return join(R.begin(), R.end(), Separator);
+}
+
+/// Joins the strings in the parameter pack \p Items, adding \p Separator
+/// between the elements. All arguments must be implicitly convertible to
+/// std::string, or there should be an overload of std::string::operator+=()
+/// that accepts the argument explicitly.
+template <typename Sep, typename... Args>
+inline std::string join_items(Sep Separator, Args &&... Items) {
+ std::string Result;
+ if (sizeof...(Items) == 0)
+ return Result;
+
+ size_t NS = detail::join_one_item_size(Separator);
+ size_t NI = detail::join_items_size(std::forward<Args>(Items)...);
+ Result.reserve(NI + (sizeof...(Items) - 1) * NS + 1);
+ detail::join_items_impl(Result, Separator, std::forward<Args>(Items)...);
+ return Result;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STRINGEXTRAS_H
diff --git a/third_party/llvm-project/include/llvm/ADT/StringMap.h b/third_party/llvm-project/include/llvm/ADT/StringMap.h
new file mode 100644
index 000000000..108185bd0
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/StringMap.h
@@ -0,0 +1,593 @@
+//===- StringMap.h - String Hash table map interface ------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the StringMap class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_STRINGMAP_H
+#define LLVM_ADT_STRINGMAP_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/Allocator.h"
+#include "llvm/Support/PointerLikeTypeTraits.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+#include <cstdlib>
+#include <cstring>
+#include <initializer_list>
+#include <iterator>
+#include <utility>
+
+namespace llvm {
+
+template<typename ValueTy> class StringMapConstIterator;
+template<typename ValueTy> class StringMapIterator;
+template<typename ValueTy> class StringMapKeyIterator;
+
+/// StringMapEntryBase - Shared base class of StringMapEntry instances.
+class StringMapEntryBase {
+ size_t StrLen;
+
+public:
+ explicit StringMapEntryBase(size_t Len) : StrLen(Len) {}
+
+ size_t getKeyLength() const { return StrLen; }
+};
+
+/// StringMapImpl - This is the base class of StringMap that is shared among
+/// all of its instantiations.
+class StringMapImpl {
+protected:
+ // Array of NumBuckets pointers to entries, null pointers are holes.
+ // TheTable[NumBuckets] contains a sentinel value for easy iteration. Followed
+ // by an array of the actual hash values as unsigned integers.
+ StringMapEntryBase **TheTable = nullptr;
+ unsigned NumBuckets = 0;
+ unsigned NumItems = 0;
+ unsigned NumTombstones = 0;
+ unsigned ItemSize;
+
+protected:
+ explicit StringMapImpl(unsigned itemSize)
+ : ItemSize(itemSize) {}
+ StringMapImpl(StringMapImpl &&RHS)
+ : TheTable(RHS.TheTable), NumBuckets(RHS.NumBuckets),
+ NumItems(RHS.NumItems), NumTombstones(RHS.NumTombstones),
+ ItemSize(RHS.ItemSize) {
+ RHS.TheTable = nullptr;
+ RHS.NumBuckets = 0;
+ RHS.NumItems = 0;
+ RHS.NumTombstones = 0;
+ }
+
+ StringMapImpl(unsigned InitSize, unsigned ItemSize);
+ unsigned RehashTable(unsigned BucketNo = 0);
+
+ /// LookupBucketFor - Look up the bucket that the specified string should end
+ /// up in. If it already exists as a key in the map, the Item pointer for the
+ /// specified bucket will be non-null. Otherwise, it will be null. In either
+ /// case, the FullHashValue field of the bucket will be set to the hash value
+ /// of the string.
+ unsigned LookupBucketFor(StringRef Key);
+
+ /// FindKey - Look up the bucket that contains the specified key. If it exists
+ /// in the map, return the bucket number of the key. Otherwise return -1.
+ /// This does not modify the map.
+ int FindKey(StringRef Key) const;
+
+ /// RemoveKey - Remove the specified StringMapEntry from the table, but do not
+ /// delete it. This aborts if the value isn't in the table.
+ void RemoveKey(StringMapEntryBase *V);
+
+ /// RemoveKey - Remove the StringMapEntry for the specified key from the
+ /// table, returning it. If the key is not in the table, this returns null.
+ StringMapEntryBase *RemoveKey(StringRef Key);
+
+ /// Allocate the table with the specified number of buckets and otherwise
+ /// setup the map as empty.
+ void init(unsigned Size);
+
+public:
+ static StringMapEntryBase *getTombstoneVal() {
+ uintptr_t Val = static_cast<uintptr_t>(-1);
+ Val <<= PointerLikeTypeTraits<StringMapEntryBase *>::NumLowBitsAvailable;
+ return reinterpret_cast<StringMapEntryBase *>(Val);
+ }
+
+ unsigned getNumBuckets() const { return NumBuckets; }
+ unsigned getNumItems() const { return NumItems; }
+
+ bool empty() const { return NumItems == 0; }
+ unsigned size() const { return NumItems; }
+
+ void swap(StringMapImpl &Other) {
+ std::swap(TheTable, Other.TheTable);
+ std::swap(NumBuckets, Other.NumBuckets);
+ std::swap(NumItems, Other.NumItems);
+ std::swap(NumTombstones, Other.NumTombstones);
+ }
+};
+
+/// StringMapEntryStorage - Holds the value in a StringMapEntry.
+///
+/// Factored out into a separate base class to make it easier to specialize.
+/// This is primarily intended to support StringSet, which doesn't need a value
+/// stored at all.
+template<typename ValueTy>
+class StringMapEntryStorage : public StringMapEntryBase {
+public:
+ ValueTy second;
+
+ explicit StringMapEntryStorage(size_t strLen)
+ : StringMapEntryBase(strLen), second() {}
+ template <typename... InitTy>
+ StringMapEntryStorage(size_t strLen, InitTy &&... InitVals)
+ : StringMapEntryBase(strLen), second(std::forward<InitTy>(InitVals)...) {}
+ StringMapEntryStorage(StringMapEntryStorage &E) = delete;
+
+ const ValueTy &getValue() const { return second; }
+ ValueTy &getValue() { return second; }
+
+ void setValue(const ValueTy &V) { second = V; }
+};
+
+template<>
+class StringMapEntryStorage<NoneType> : public StringMapEntryBase {
+public:
+ explicit StringMapEntryStorage(size_t strLen, NoneType none = None)
+ : StringMapEntryBase(strLen) {}
+ StringMapEntryStorage(StringMapEntryStorage &E) = delete;
+
+ NoneType getValue() const { return None; }
+};
+
+/// StringMapEntry - This is used to represent one value that is inserted into
+/// a StringMap. It contains the Value itself and the key: the string length
+/// and data.
+template<typename ValueTy>
+class StringMapEntry final : public StringMapEntryStorage<ValueTy> {
+public:
+ using StringMapEntryStorage<ValueTy>::StringMapEntryStorage;
+
+ StringRef getKey() const {
+ return StringRef(getKeyData(), this->getKeyLength());
+ }
+
+ /// getKeyData - Return the start of the string data that is the key for this
+ /// value. The string data is always stored immediately after the
+ /// StringMapEntry object.
+ const char *getKeyData() const {return reinterpret_cast<const char*>(this+1);}
+
+ StringRef first() const {
+ return StringRef(getKeyData(), this->getKeyLength());
+ }
+
+ /// Create a StringMapEntry for the specified key construct the value using
+ /// \p InitiVals.
+ template <typename AllocatorTy, typename... InitTy>
+ static StringMapEntry *Create(StringRef Key, AllocatorTy &Allocator,
+ InitTy &&... InitVals) {
+ size_t KeyLength = Key.size();
+
+ // Allocate a new item with space for the string at the end and a null
+ // terminator.
+ size_t AllocSize = sizeof(StringMapEntry) + KeyLength + 1;
+ size_t Alignment = alignof(StringMapEntry);
+
+ StringMapEntry *NewItem =
+ static_cast<StringMapEntry*>(Allocator.Allocate(AllocSize,Alignment));
+ assert(NewItem && "Unhandled out-of-memory");
+
+ // Construct the value.
+ new (NewItem) StringMapEntry(KeyLength, std::forward<InitTy>(InitVals)...);
+
+ // Copy the string information.
+ char *StrBuffer = const_cast<char*>(NewItem->getKeyData());
+ if (KeyLength > 0)
+ memcpy(StrBuffer, Key.data(), KeyLength);
+ StrBuffer[KeyLength] = 0; // Null terminate for convenience of clients.
+ return NewItem;
+ }
+
+ /// Create - Create a StringMapEntry with normal malloc/free.
+ template <typename... InitType>
+ static StringMapEntry *Create(StringRef Key, InitType &&... InitVal) {
+ MallocAllocator A;
+ return Create(Key, A, std::forward<InitType>(InitVal)...);
+ }
+
+ static StringMapEntry *Create(StringRef Key) {
+ return Create(Key, ValueTy());
+ }
+
+ /// GetStringMapEntryFromKeyData - Given key data that is known to be embedded
+ /// into a StringMapEntry, return the StringMapEntry itself.
+ static StringMapEntry &GetStringMapEntryFromKeyData(const char *KeyData) {
+ char *Ptr = const_cast<char*>(KeyData) - sizeof(StringMapEntry<ValueTy>);
+ return *reinterpret_cast<StringMapEntry*>(Ptr);
+ }
+
+ /// Destroy - Destroy this StringMapEntry, releasing memory back to the
+ /// specified allocator.
+ template<typename AllocatorTy>
+ void Destroy(AllocatorTy &Allocator) {
+ // Free memory referenced by the item.
+ size_t AllocSize = sizeof(StringMapEntry) + this->getKeyLength() + 1;
+ this->~StringMapEntry();
+ Allocator.Deallocate(static_cast<void *>(this), AllocSize);
+ }
+
+ /// Destroy this object, releasing memory back to the malloc allocator.
+ void Destroy() {
+ MallocAllocator A;
+ Destroy(A);
+ }
+};
+
+/// StringMap - This is an unconventional map that is specialized for handling
+/// keys that are "strings", which are basically ranges of bytes. This does some
+/// funky memory allocation and hashing things to make it extremely efficient,
+/// storing the string data *after* the value in the map.
+template<typename ValueTy, typename AllocatorTy = MallocAllocator>
+class StringMap : public StringMapImpl {
+ AllocatorTy Allocator;
+
+public:
+ using MapEntryTy = StringMapEntry<ValueTy>;
+
+ StringMap() : StringMapImpl(static_cast<unsigned>(sizeof(MapEntryTy))) {}
+
+ explicit StringMap(unsigned InitialSize)
+ : StringMapImpl(InitialSize, static_cast<unsigned>(sizeof(MapEntryTy))) {}
+
+ explicit StringMap(AllocatorTy A)
+ : StringMapImpl(static_cast<unsigned>(sizeof(MapEntryTy))), Allocator(A) {}
+
+ StringMap(unsigned InitialSize, AllocatorTy A)
+ : StringMapImpl(InitialSize, static_cast<unsigned>(sizeof(MapEntryTy))),
+ Allocator(A) {}
+
+ StringMap(std::initializer_list<std::pair<StringRef, ValueTy>> List)
+ : StringMapImpl(List.size(), static_cast<unsigned>(sizeof(MapEntryTy))) {
+ for (const auto &P : List) {
+ insert(P);
+ }
+ }
+
+ StringMap(StringMap &&RHS)
+ : StringMapImpl(std::move(RHS)), Allocator(std::move(RHS.Allocator)) {}
+
+ StringMap(const StringMap &RHS) :
+ StringMapImpl(static_cast<unsigned>(sizeof(MapEntryTy))),
+ Allocator(RHS.Allocator) {
+ if (RHS.empty())
+ return;
+
+ // Allocate TheTable of the same size as RHS's TheTable, and set the
+ // sentinel appropriately (and NumBuckets).
+ init(RHS.NumBuckets);
+ unsigned *HashTable = (unsigned *)(TheTable + NumBuckets + 1),
+ *RHSHashTable = (unsigned *)(RHS.TheTable + NumBuckets + 1);
+
+ NumItems = RHS.NumItems;
+ NumTombstones = RHS.NumTombstones;
+ for (unsigned I = 0, E = NumBuckets; I != E; ++I) {
+ StringMapEntryBase *Bucket = RHS.TheTable[I];
+ if (!Bucket || Bucket == getTombstoneVal()) {
+ TheTable[I] = Bucket;
+ continue;
+ }
+
+ TheTable[I] = MapEntryTy::Create(
+ static_cast<MapEntryTy *>(Bucket)->getKey(), Allocator,
+ static_cast<MapEntryTy *>(Bucket)->getValue());
+ HashTable[I] = RHSHashTable[I];
+ }
+
+ // Note that here we've copied everything from the RHS into this object,
+ // tombstones included. We could, instead, have re-probed for each key to
+ // instantiate this new object without any tombstone buckets. The
+ // assumption here is that items are rarely deleted from most StringMaps,
+ // and so tombstones are rare, so the cost of re-probing for all inputs is
+ // not worthwhile.
+ }
+
+ StringMap &operator=(StringMap RHS) {
+ StringMapImpl::swap(RHS);
+ std::swap(Allocator, RHS.Allocator);
+ return *this;
+ }
+
+ ~StringMap() {
+ // Delete all the elements in the map, but don't reset the elements
+ // to default values. This is a copy of clear(), but avoids unnecessary
+ // work not required in the destructor.
+ if (!empty()) {
+ for (unsigned I = 0, E = NumBuckets; I != E; ++I) {
+ StringMapEntryBase *Bucket = TheTable[I];
+ if (Bucket && Bucket != getTombstoneVal()) {
+ static_cast<MapEntryTy*>(Bucket)->Destroy(Allocator);
+ }
+ }
+ }
+ free(TheTable);
+ }
+
+ AllocatorTy &getAllocator() { return Allocator; }
+ const AllocatorTy &getAllocator() const { return Allocator; }
+
+ using key_type = const char*;
+ using mapped_type = ValueTy;
+ using value_type = StringMapEntry<ValueTy>;
+ using size_type = size_t;
+
+ using const_iterator = StringMapConstIterator<ValueTy>;
+ using iterator = StringMapIterator<ValueTy>;
+
+ iterator begin() {
+ return iterator(TheTable, NumBuckets == 0);
+ }
+ iterator end() {
+ return iterator(TheTable+NumBuckets, true);
+ }
+ const_iterator begin() const {
+ return const_iterator(TheTable, NumBuckets == 0);
+ }
+ const_iterator end() const {
+ return const_iterator(TheTable+NumBuckets, true);
+ }
+
+ iterator_range<StringMapKeyIterator<ValueTy>> keys() const {
+ return make_range(StringMapKeyIterator<ValueTy>(begin()),
+ StringMapKeyIterator<ValueTy>(end()));
+ }
+
+ iterator find(StringRef Key) {
+ int Bucket = FindKey(Key);
+ if (Bucket == -1) return end();
+ return iterator(TheTable+Bucket, true);
+ }
+
+ const_iterator find(StringRef Key) const {
+ int Bucket = FindKey(Key);
+ if (Bucket == -1) return end();
+ return const_iterator(TheTable+Bucket, true);
+ }
+
+ /// lookup - Return the entry for the specified key, or a default
+ /// constructed value if no such entry exists.
+ ValueTy lookup(StringRef Key) const {
+ const_iterator it = find(Key);
+ if (it != end())
+ return it->second;
+ return ValueTy();
+ }
+
+ /// Lookup the ValueTy for the \p Key, or create a default constructed value
+ /// if the key is not in the map.
+ ValueTy &operator[](StringRef Key) { return try_emplace(Key).first->second; }
+
+ /// count - Return 1 if the element is in the map, 0 otherwise.
+ size_type count(StringRef Key) const {
+ return find(Key) == end() ? 0 : 1;
+ }
+
+ template <typename InputTy>
+ size_type count(const StringMapEntry<InputTy> &MapEntry) const {
+ return count(MapEntry.getKey());
+ }
+
+ /// insert - Insert the specified key/value pair into the map. If the key
+ /// already exists in the map, return false and ignore the request, otherwise
+ /// insert it and return true.
+ bool insert(MapEntryTy *KeyValue) {
+ unsigned BucketNo = LookupBucketFor(KeyValue->getKey());
+ StringMapEntryBase *&Bucket = TheTable[BucketNo];
+ if (Bucket && Bucket != getTombstoneVal())
+ return false; // Already exists in map.
+
+ if (Bucket == getTombstoneVal())
+ --NumTombstones;
+ Bucket = KeyValue;
+ ++NumItems;
+ assert(NumItems + NumTombstones <= NumBuckets);
+
+ RehashTable();
+ return true;
+ }
+
+ /// insert - Inserts the specified key/value pair into the map if the key
+ /// isn't already in the map. The bool component of the returned pair is true
+ /// if and only if the insertion takes place, and the iterator component of
+ /// the pair points to the element with key equivalent to the key of the pair.
+ std::pair<iterator, bool> insert(std::pair<StringRef, ValueTy> KV) {
+ return try_emplace(KV.first, std::move(KV.second));
+ }
+
+ /// Inserts an element or assigns to the current element if the key already
+ /// exists. The return type is the same as try_emplace.
+ template <typename V>
+ std::pair<iterator, bool> insert_or_assign(StringRef Key, V &&Val) {
+ auto Ret = try_emplace(Key, std::forward<V>(Val));
+ if (!Ret.second)
+ Ret.first->second = std::forward<V>(Val);
+ return Ret;
+ }
+
+ /// Emplace a new element for the specified key into the map if the key isn't
+ /// already in the map. The bool component of the returned pair is true
+ /// if and only if the insertion takes place, and the iterator component of
+ /// the pair points to the element with key equivalent to the key of the pair.
+ template <typename... ArgsTy>
+ std::pair<iterator, bool> try_emplace(StringRef Key, ArgsTy &&... Args) {
+ unsigned BucketNo = LookupBucketFor(Key);
+ StringMapEntryBase *&Bucket = TheTable[BucketNo];
+ if (Bucket && Bucket != getTombstoneVal())
+ return std::make_pair(iterator(TheTable + BucketNo, false),
+ false); // Already exists in map.
+
+ if (Bucket == getTombstoneVal())
+ --NumTombstones;
+ Bucket = MapEntryTy::Create(Key, Allocator, std::forward<ArgsTy>(Args)...);
+ ++NumItems;
+ assert(NumItems + NumTombstones <= NumBuckets);
+
+ BucketNo = RehashTable(BucketNo);
+ return std::make_pair(iterator(TheTable + BucketNo, false), true);
+ }
+
+ // clear - Empties out the StringMap
+ void clear() {
+ if (empty()) return;
+
+ // Zap all values, resetting the keys back to non-present (not tombstone),
+ // which is safe because we're removing all elements.
+ for (unsigned I = 0, E = NumBuckets; I != E; ++I) {
+ StringMapEntryBase *&Bucket = TheTable[I];
+ if (Bucket && Bucket != getTombstoneVal()) {
+ static_cast<MapEntryTy*>(Bucket)->Destroy(Allocator);
+ }
+ Bucket = nullptr;
+ }
+
+ NumItems = 0;
+ NumTombstones = 0;
+ }
+
+ /// remove - Remove the specified key/value pair from the map, but do not
+ /// erase it. This aborts if the key is not in the map.
+ void remove(MapEntryTy *KeyValue) {
+ RemoveKey(KeyValue);
+ }
+
+ void erase(iterator I) {
+ MapEntryTy &V = *I;
+ remove(&V);
+ V.Destroy(Allocator);
+ }
+
+ bool erase(StringRef Key) {
+ iterator I = find(Key);
+ if (I == end()) return false;
+ erase(I);
+ return true;
+ }
+};
+
+template <typename DerivedTy, typename ValueTy>
+class StringMapIterBase
+ : public iterator_facade_base<DerivedTy, std::forward_iterator_tag,
+ ValueTy> {
+protected:
+ StringMapEntryBase **Ptr = nullptr;
+
+public:
+ StringMapIterBase() = default;
+
+ explicit StringMapIterBase(StringMapEntryBase **Bucket,
+ bool NoAdvance = false)
+ : Ptr(Bucket) {
+ if (!NoAdvance) AdvancePastEmptyBuckets();
+ }
+
+ DerivedTy &operator=(const DerivedTy &Other) {
+ Ptr = Other.Ptr;
+ return static_cast<DerivedTy &>(*this);
+ }
+
+ bool operator==(const DerivedTy &RHS) const { return Ptr == RHS.Ptr; }
+
+ DerivedTy &operator++() { // Preincrement
+ ++Ptr;
+ AdvancePastEmptyBuckets();
+ return static_cast<DerivedTy &>(*this);
+ }
+
+ DerivedTy operator++(int) { // Post-increment
+ DerivedTy Tmp(Ptr);
+ ++*this;
+ return Tmp;
+ }
+
+private:
+ void AdvancePastEmptyBuckets() {
+ while (*Ptr == nullptr || *Ptr == StringMapImpl::getTombstoneVal())
+ ++Ptr;
+ }
+};
+
+template <typename ValueTy>
+class StringMapConstIterator
+ : public StringMapIterBase<StringMapConstIterator<ValueTy>,
+ const StringMapEntry<ValueTy>> {
+ using base = StringMapIterBase<StringMapConstIterator<ValueTy>,
+ const StringMapEntry<ValueTy>>;
+
+public:
+ StringMapConstIterator() = default;
+ explicit StringMapConstIterator(StringMapEntryBase **Bucket,
+ bool NoAdvance = false)
+ : base(Bucket, NoAdvance) {}
+
+ const StringMapEntry<ValueTy> &operator*() const {
+ return *static_cast<const StringMapEntry<ValueTy> *>(*this->Ptr);
+ }
+};
+
+template <typename ValueTy>
+class StringMapIterator : public StringMapIterBase<StringMapIterator<ValueTy>,
+ StringMapEntry<ValueTy>> {
+ using base =
+ StringMapIterBase<StringMapIterator<ValueTy>, StringMapEntry<ValueTy>>;
+
+public:
+ StringMapIterator() = default;
+ explicit StringMapIterator(StringMapEntryBase **Bucket,
+ bool NoAdvance = false)
+ : base(Bucket, NoAdvance) {}
+
+ StringMapEntry<ValueTy> &operator*() const {
+ return *static_cast<StringMapEntry<ValueTy> *>(*this->Ptr);
+ }
+
+ operator StringMapConstIterator<ValueTy>() const {
+ return StringMapConstIterator<ValueTy>(this->Ptr, true);
+ }
+};
+
+template <typename ValueTy>
+class StringMapKeyIterator
+ : public iterator_adaptor_base<StringMapKeyIterator<ValueTy>,
+ StringMapConstIterator<ValueTy>,
+ std::forward_iterator_tag, StringRef> {
+ using base = iterator_adaptor_base<StringMapKeyIterator<ValueTy>,
+ StringMapConstIterator<ValueTy>,
+ std::forward_iterator_tag, StringRef>;
+
+public:
+ StringMapKeyIterator() = default;
+ explicit StringMapKeyIterator(StringMapConstIterator<ValueTy> Iter)
+ : base(std::move(Iter)) {}
+
+ StringRef &operator*() {
+ Key = this->wrapped()->getKey();
+ return Key;
+ }
+
+private:
+ StringRef Key;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STRINGMAP_H
diff --git a/third_party/llvm-project/include/llvm/ADT/StringRef.h b/third_party/llvm-project/include/llvm/ADT/StringRef.h
new file mode 100644
index 000000000..e87a08f7e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/StringRef.h
@@ -0,0 +1,920 @@
+//===- StringRef.h - Constant String Reference Wrapper ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_STRINGREF_H
+#define LLVM_ADT_STRINGREF_H
+
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/Compiler.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstring>
+#include <limits>
+#include <string>
+#include <type_traits>
+#include <utility>
+
+// Declare the __builtin_strlen intrinsic for MSVC so it can be used in
+// constexpr context.
+#if defined(_MSC_VER)
+extern "C" size_t __builtin_strlen(const char *);
+#endif
+
+namespace llvm {
+
+ class APInt;
+ class hash_code;
+ template <typename T> class SmallVectorImpl;
+ class StringRef;
+
+ /// Helper functions for StringRef::getAsInteger.
+ bool getAsUnsignedInteger(StringRef Str, unsigned Radix,
+ unsigned long long &Result);
+
+ bool getAsSignedInteger(StringRef Str, unsigned Radix, long long &Result);
+
+ bool consumeUnsignedInteger(StringRef &Str, unsigned Radix,
+ unsigned long long &Result);
+ bool consumeSignedInteger(StringRef &Str, unsigned Radix, long long &Result);
+
+ /// StringRef - Represent a constant reference to a string, i.e. a character
+ /// array and a length, which need not be null terminated.
+ ///
+ /// This class does not own the string data, it is expected to be used in
+ /// situations where the character data resides in some other buffer, whose
+ /// lifetime extends past that of the StringRef. For this reason, it is not in
+ /// general safe to store a StringRef.
+ class StringRef {
+ public:
+ static const size_t npos = ~size_t(0);
+
+ using iterator = const char *;
+ using const_iterator = const char *;
+ using size_type = size_t;
+
+ private:
+ /// The start of the string, in an external buffer.
+ const char *Data = nullptr;
+
+ /// The length of the string.
+ size_t Length = 0;
+
+ // Workaround memcmp issue with null pointers (undefined behavior)
+ // by providing a specialized version
+ static int compareMemory(const char *Lhs, const char *Rhs, size_t Length) {
+ if (Length == 0) { return 0; }
+ return ::memcmp(Lhs,Rhs,Length);
+ }
+
+ // Constexpr version of std::strlen.
+ static constexpr size_t strLen(const char *Str) {
+#if __cplusplus > 201402L
+ return std::char_traits<char>::length(Str);
+#elif __has_builtin(__builtin_strlen) || defined(__GNUC__) || defined(_MSC_VER)
+ return __builtin_strlen(Str);
+#else
+ const char *Begin = Str;
+ while (*Str != '\0')
+ ++Str;
+ return Str - Begin;
+#endif
+ }
+
+ public:
+ /// @name Constructors
+ /// @{
+
+ /// Construct an empty string ref.
+ /*implicit*/ StringRef() = default;
+
+ /// Disable conversion from nullptr. This prevents things like
+ /// if (S == nullptr)
+ StringRef(std::nullptr_t) = delete;
+
+ /// Construct a string ref from a cstring.
+ /*implicit*/ constexpr StringRef(const char *Str)
+ : Data(Str), Length(Str ? strLen(Str) : 0) {}
+
+ /// Construct a string ref from a pointer and length.
+ /*implicit*/ constexpr StringRef(const char *data, size_t length)
+ : Data(data), Length(length) {}
+
+ /// Construct a string ref from an std::string.
+ /*implicit*/ StringRef(const std::string &Str)
+ : Data(Str.data()), Length(Str.length()) {}
+
+ static StringRef withNullAsEmpty(const char *data) {
+ return StringRef(data ? data : "");
+ }
+
+ /// @}
+ /// @name Iterators
+ /// @{
+
+ iterator begin() const { return Data; }
+
+ iterator end() const { return Data + Length; }
+
+ const unsigned char *bytes_begin() const {
+ return reinterpret_cast<const unsigned char *>(begin());
+ }
+ const unsigned char *bytes_end() const {
+ return reinterpret_cast<const unsigned char *>(end());
+ }
+ iterator_range<const unsigned char *> bytes() const {
+ return make_range(bytes_begin(), bytes_end());
+ }
+
+ /// @}
+ /// @name String Operations
+ /// @{
+
+ /// data - Get a pointer to the start of the string (which may not be null
+ /// terminated).
+ LLVM_NODISCARD
+ const char *data() const { return Data; }
+
+ /// empty - Check if the string is empty.
+ LLVM_NODISCARD
+ bool empty() const { return Length == 0; }
+
+ /// size - Get the string size.
+ LLVM_NODISCARD
+ size_t size() const { return Length; }
+
+ /// front - Get the first character in the string.
+ LLVM_NODISCARD
+ char front() const {
+ assert(!empty());
+ return Data[0];
+ }
+
+ /// back - Get the last character in the string.
+ LLVM_NODISCARD
+ char back() const {
+ assert(!empty());
+ return Data[Length-1];
+ }
+
+ // copy - Allocate copy in Allocator and return StringRef to it.
+ template <typename Allocator>
+ LLVM_NODISCARD StringRef copy(Allocator &A) const {
+ // Don't request a length 0 copy from the allocator.
+ if (empty())
+ return StringRef();
+ char *S = A.template Allocate<char>(Length);
+ std::copy(begin(), end(), S);
+ return StringRef(S, Length);
+ }
+
+ /// equals - Check for string equality, this is more efficient than
+ /// compare() when the relative ordering of inequal strings isn't needed.
+ LLVM_NODISCARD
+ bool equals(StringRef RHS) const {
+ return (Length == RHS.Length &&
+ compareMemory(Data, RHS.Data, RHS.Length) == 0);
+ }
+
+ /// equals_lower - Check for string equality, ignoring case.
+ LLVM_NODISCARD
+ bool equals_lower(StringRef RHS) const {
+ return Length == RHS.Length && compare_lower(RHS) == 0;
+ }
+
+ /// compare - Compare two strings; the result is -1, 0, or 1 if this string
+ /// is lexicographically less than, equal to, or greater than the \p RHS.
+ LLVM_NODISCARD
+ int compare(StringRef RHS) const {
+ // Check the prefix for a mismatch.
+ if (int Res = compareMemory(Data, RHS.Data, std::min(Length, RHS.Length)))
+ return Res < 0 ? -1 : 1;
+
+ // Otherwise the prefixes match, so we only need to check the lengths.
+ if (Length == RHS.Length)
+ return 0;
+ return Length < RHS.Length ? -1 : 1;
+ }
+
+ /// compare_lower - Compare two strings, ignoring case.
+ LLVM_NODISCARD
+ int compare_lower(StringRef RHS) const;
+
+ /// compare_numeric - Compare two strings, treating sequences of digits as
+ /// numbers.
+ LLVM_NODISCARD
+ int compare_numeric(StringRef RHS) const;
+
+ /// Determine the edit distance between this string and another
+ /// string.
+ ///
+ /// \param Other the string to compare this string against.
+ ///
+ /// \param AllowReplacements whether to allow character
+ /// replacements (change one character into another) as a single
+ /// operation, rather than as two operations (an insertion and a
+ /// removal).
+ ///
+ /// \param MaxEditDistance If non-zero, the maximum edit distance that
+ /// this routine is allowed to compute. If the edit distance will exceed
+ /// that maximum, returns \c MaxEditDistance+1.
+ ///
+ /// \returns the minimum number of character insertions, removals,
+ /// or (if \p AllowReplacements is \c true) replacements needed to
+ /// transform one of the given strings into the other. If zero,
+ /// the strings are identical.
+ LLVM_NODISCARD
+ unsigned edit_distance(StringRef Other, bool AllowReplacements = true,
+ unsigned MaxEditDistance = 0) const;
+
+ /// str - Get the contents as an std::string.
+ LLVM_NODISCARD
+ std::string str() const {
+ if (!Data) return std::string();
+ return std::string(Data, Length);
+ }
+
+ /// @}
+ /// @name Operator Overloads
+ /// @{
+
+ LLVM_NODISCARD
+ char operator[](size_t Index) const {
+ assert(Index < Length && "Invalid index!");
+ return Data[Index];
+ }
+
+ /// Disallow accidental assignment from a temporary std::string.
+ ///
+ /// The declaration here is extra complicated so that `stringRef = {}`
+ /// and `stringRef = "abc"` continue to select the move assignment operator.
+ template <typename T>
+ typename std::enable_if<std::is_same<T, std::string>::value,
+ StringRef>::type &
+ operator=(T &&Str) = delete;
+
+ /// @}
+ /// @name Type Conversions
+ /// @{
+
+ operator std::string() const {
+ return str();
+ }
+
+ /// @}
+ /// @name String Predicates
+ /// @{
+
+ /// Check if this string starts with the given \p Prefix.
+ LLVM_NODISCARD
+ bool startswith(StringRef Prefix) const {
+ return Length >= Prefix.Length &&
+ compareMemory(Data, Prefix.Data, Prefix.Length) == 0;
+ }
+
+ /// Check if this string starts with the given \p Prefix, ignoring case.
+ LLVM_NODISCARD
+ bool startswith_lower(StringRef Prefix) const;
+
+ /// Check if this string ends with the given \p Suffix.
+ LLVM_NODISCARD
+ bool endswith(StringRef Suffix) const {
+ return Length >= Suffix.Length &&
+ compareMemory(end() - Suffix.Length, Suffix.Data, Suffix.Length) == 0;
+ }
+
+ /// Check if this string ends with the given \p Suffix, ignoring case.
+ LLVM_NODISCARD
+ bool endswith_lower(StringRef Suffix) const;
+
+ /// @}
+ /// @name String Searching
+ /// @{
+
+ /// Search for the first character \p C in the string.
+ ///
+ /// \returns The index of the first occurrence of \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find(char C, size_t From = 0) const {
+ size_t FindBegin = std::min(From, Length);
+ if (FindBegin < Length) { // Avoid calling memchr with nullptr.
+ // Just forward to memchr, which is faster than a hand-rolled loop.
+ if (const void *P = ::memchr(Data + FindBegin, C, Length - FindBegin))
+ return static_cast<const char *>(P) - Data;
+ }
+ return npos;
+ }
+
+ /// Search for the first character \p C in the string, ignoring case.
+ ///
+ /// \returns The index of the first occurrence of \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find_lower(char C, size_t From = 0) const;
+
+ /// Search for the first character satisfying the predicate \p F
+ ///
+ /// \returns The index of the first character satisfying \p F starting from
+ /// \p From, or npos if not found.
+ LLVM_NODISCARD
+ size_t find_if(function_ref<bool(char)> F, size_t From = 0) const {
+ StringRef S = drop_front(From);
+ while (!S.empty()) {
+ if (F(S.front()))
+ return size() - S.size();
+ S = S.drop_front();
+ }
+ return npos;
+ }
+
+ /// Search for the first character not satisfying the predicate \p F
+ ///
+ /// \returns The index of the first character not satisfying \p F starting
+ /// from \p From, or npos if not found.
+ LLVM_NODISCARD
+ size_t find_if_not(function_ref<bool(char)> F, size_t From = 0) const {
+ return find_if([F](char c) { return !F(c); }, From);
+ }
+
+ /// Search for the first string \p Str in the string.
+ ///
+ /// \returns The index of the first occurrence of \p Str, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find(StringRef Str, size_t From = 0) const;
+
+ /// Search for the first string \p Str in the string, ignoring case.
+ ///
+ /// \returns The index of the first occurrence of \p Str, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find_lower(StringRef Str, size_t From = 0) const;
+
+ /// Search for the last character \p C in the string.
+ ///
+ /// \returns The index of the last occurrence of \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t rfind(char C, size_t From = npos) const {
+ From = std::min(From, Length);
+ size_t i = From;
+ while (i != 0) {
+ --i;
+ if (Data[i] == C)
+ return i;
+ }
+ return npos;
+ }
+
+ /// Search for the last character \p C in the string, ignoring case.
+ ///
+ /// \returns The index of the last occurrence of \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t rfind_lower(char C, size_t From = npos) const;
+
+ /// Search for the last string \p Str in the string.
+ ///
+ /// \returns The index of the last occurrence of \p Str, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t rfind(StringRef Str) const;
+
+ /// Search for the last string \p Str in the string, ignoring case.
+ ///
+ /// \returns The index of the last occurrence of \p Str, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t rfind_lower(StringRef Str) const;
+
+ /// Find the first character in the string that is \p C, or npos if not
+ /// found. Same as find.
+ LLVM_NODISCARD
+ size_t find_first_of(char C, size_t From = 0) const {
+ return find(C, From);
+ }
+
+ /// Find the first character in the string that is in \p Chars, or npos if
+ /// not found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ LLVM_NODISCARD
+ size_t find_first_of(StringRef Chars, size_t From = 0) const;
+
+ /// Find the first character in the string that is not \p C or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find_first_not_of(char C, size_t From = 0) const;
+
+ /// Find the first character in the string that is not in the string
+ /// \p Chars, or npos if not found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ LLVM_NODISCARD
+ size_t find_first_not_of(StringRef Chars, size_t From = 0) const;
+
+ /// Find the last character in the string that is \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find_last_of(char C, size_t From = npos) const {
+ return rfind(C, From);
+ }
+
+ /// Find the last character in the string that is in \p C, or npos if not
+ /// found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ LLVM_NODISCARD
+ size_t find_last_of(StringRef Chars, size_t From = npos) const;
+
+ /// Find the last character in the string that is not \p C, or npos if not
+ /// found.
+ LLVM_NODISCARD
+ size_t find_last_not_of(char C, size_t From = npos) const;
+
+ /// Find the last character in the string that is not in \p Chars, or
+ /// npos if not found.
+ ///
+ /// Complexity: O(size() + Chars.size())
+ LLVM_NODISCARD
+ size_t find_last_not_of(StringRef Chars, size_t From = npos) const;
+
+ /// Return true if the given string is a substring of *this, and false
+ /// otherwise.
+ LLVM_NODISCARD
+ bool contains(StringRef Other) const { return find(Other) != npos; }
+
+ /// Return true if the given character is contained in *this, and false
+ /// otherwise.
+ LLVM_NODISCARD
+ bool contains(char C) const { return find_first_of(C) != npos; }
+
+ /// Return true if the given string is a substring of *this, and false
+ /// otherwise.
+ LLVM_NODISCARD
+ bool contains_lower(StringRef Other) const {
+ return find_lower(Other) != npos;
+ }
+
+ /// Return true if the given character is contained in *this, and false
+ /// otherwise.
+ LLVM_NODISCARD
+ bool contains_lower(char C) const { return find_lower(C) != npos; }
+
+ /// @}
+ /// @name Helpful Algorithms
+ /// @{
+
+ /// Return the number of occurrences of \p C in the string.
+ LLVM_NODISCARD
+ size_t count(char C) const {
+ size_t Count = 0;
+ for (size_t i = 0, e = Length; i != e; ++i)
+ if (Data[i] == C)
+ ++Count;
+ return Count;
+ }
+
+ /// Return the number of non-overlapped occurrences of \p Str in
+ /// the string.
+ size_t count(StringRef Str) const;
+
+ /// Parse the current string as an integer of the specified radix. If
+ /// \p Radix is specified as zero, this does radix autosensing using
+ /// extended C rules: 0 is octal, 0x is hex, 0b is binary.
+ ///
+ /// If the string is invalid or if only a subset of the string is valid,
+ /// this returns true to signify the error. The string is considered
+ /// erroneous if empty or if it overflows T.
+ template <typename T>
+ typename std::enable_if<std::numeric_limits<T>::is_signed, bool>::type
+ getAsInteger(unsigned Radix, T &Result) const {
+ long long LLVal;
+ if (getAsSignedInteger(*this, Radix, LLVal) ||
+ static_cast<T>(LLVal) != LLVal)
+ return true;
+ Result = LLVal;
+ return false;
+ }
+
+ template <typename T>
+ typename std::enable_if<!std::numeric_limits<T>::is_signed, bool>::type
+ getAsInteger(unsigned Radix, T &Result) const {
+ unsigned long long ULLVal;
+ // The additional cast to unsigned long long is required to avoid the
+ // Visual C++ warning C4805: '!=' : unsafe mix of type 'bool' and type
+ // 'unsigned __int64' when instantiating getAsInteger with T = bool.
+ if (getAsUnsignedInteger(*this, Radix, ULLVal) ||
+ static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal)
+ return true;
+ Result = ULLVal;
+ return false;
+ }
+
+ /// Parse the current string as an integer of the specified radix. If
+ /// \p Radix is specified as zero, this does radix autosensing using
+ /// extended C rules: 0 is octal, 0x is hex, 0b is binary.
+ ///
+ /// If the string does not begin with a number of the specified radix,
+ /// this returns true to signify the error. The string is considered
+ /// erroneous if empty or if it overflows T.
+ /// The portion of the string representing the discovered numeric value
+ /// is removed from the beginning of the string.
+ template <typename T>
+ typename std::enable_if<std::numeric_limits<T>::is_signed, bool>::type
+ consumeInteger(unsigned Radix, T &Result) {
+ long long LLVal;
+ if (consumeSignedInteger(*this, Radix, LLVal) ||
+ static_cast<long long>(static_cast<T>(LLVal)) != LLVal)
+ return true;
+ Result = LLVal;
+ return false;
+ }
+
+ template <typename T>
+ typename std::enable_if<!std::numeric_limits<T>::is_signed, bool>::type
+ consumeInteger(unsigned Radix, T &Result) {
+ unsigned long long ULLVal;
+ if (consumeUnsignedInteger(*this, Radix, ULLVal) ||
+ static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal)
+ return true;
+ Result = ULLVal;
+ return false;
+ }
+
+ /// Parse the current string as an integer of the specified \p Radix, or of
+ /// an autosensed radix if the \p Radix given is 0. The current value in
+ /// \p Result is discarded, and the storage is changed to be wide enough to
+ /// store the parsed integer.
+ ///
+ /// \returns true if the string does not solely consist of a valid
+ /// non-empty number in the appropriate base.
+ ///
+ /// APInt::fromString is superficially similar but assumes the
+ /// string is well-formed in the given radix.
+ bool getAsInteger(unsigned Radix, APInt &Result) const;
+
+ /// Parse the current string as an IEEE double-precision floating
+ /// point value. The string must be a well-formed double.
+ ///
+ /// If \p AllowInexact is false, the function will fail if the string
+ /// cannot be represented exactly. Otherwise, the function only fails
+ /// in case of an overflow or underflow.
+ bool getAsDouble(double &Result, bool AllowInexact = true) const;
+
+ /// @}
+ /// @name String Operations
+ /// @{
+
+ // Convert the given ASCII string to lowercase.
+ LLVM_NODISCARD
+ std::string lower() const;
+
+ /// Convert the given ASCII string to uppercase.
+ LLVM_NODISCARD
+ std::string upper() const;
+
+ /// @}
+ /// @name Substring Operations
+ /// @{
+
+ /// Return a reference to the substring from [Start, Start + N).
+ ///
+ /// \param Start The index of the starting character in the substring; if
+ /// the index is npos or greater than the length of the string then the
+ /// empty substring will be returned.
+ ///
+ /// \param N The number of characters to included in the substring. If N
+ /// exceeds the number of characters remaining in the string, the string
+ /// suffix (starting with \p Start) will be returned.
+ LLVM_NODISCARD
+ StringRef substr(size_t Start, size_t N = npos) const {
+ Start = std::min(Start, Length);
+ return StringRef(Data + Start, std::min(N, Length - Start));
+ }
+
+ /// Return a StringRef equal to 'this' but with only the first \p N
+ /// elements remaining. If \p N is greater than the length of the
+ /// string, the entire string is returned.
+ LLVM_NODISCARD
+ StringRef take_front(size_t N = 1) const {
+ if (N >= size())
+ return *this;
+ return drop_back(size() - N);
+ }
+
+ /// Return a StringRef equal to 'this' but with only the last \p N
+ /// elements remaining. If \p N is greater than the length of the
+ /// string, the entire string is returned.
+ LLVM_NODISCARD
+ StringRef take_back(size_t N = 1) const {
+ if (N >= size())
+ return *this;
+ return drop_front(size() - N);
+ }
+
+ /// Return the longest prefix of 'this' such that every character
+ /// in the prefix satisfies the given predicate.
+ LLVM_NODISCARD
+ StringRef take_while(function_ref<bool(char)> F) const {
+ return substr(0, find_if_not(F));
+ }
+
+ /// Return the longest prefix of 'this' such that no character in
+ /// the prefix satisfies the given predicate.
+ LLVM_NODISCARD
+ StringRef take_until(function_ref<bool(char)> F) const {
+ return substr(0, find_if(F));
+ }
+
+ /// Return a StringRef equal to 'this' but with the first \p N elements
+ /// dropped.
+ LLVM_NODISCARD
+ StringRef drop_front(size_t N = 1) const {
+ assert(size() >= N && "Dropping more elements than exist");
+ return substr(N);
+ }
+
+ /// Return a StringRef equal to 'this' but with the last \p N elements
+ /// dropped.
+ LLVM_NODISCARD
+ StringRef drop_back(size_t N = 1) const {
+ assert(size() >= N && "Dropping more elements than exist");
+ return substr(0, size()-N);
+ }
+
+ /// Return a StringRef equal to 'this', but with all characters satisfying
+ /// the given predicate dropped from the beginning of the string.
+ LLVM_NODISCARD
+ StringRef drop_while(function_ref<bool(char)> F) const {
+ return substr(find_if_not(F));
+ }
+
+ /// Return a StringRef equal to 'this', but with all characters not
+ /// satisfying the given predicate dropped from the beginning of the string.
+ LLVM_NODISCARD
+ StringRef drop_until(function_ref<bool(char)> F) const {
+ return substr(find_if(F));
+ }
+
+ /// Returns true if this StringRef has the given prefix and removes that
+ /// prefix.
+ bool consume_front(StringRef Prefix) {
+ if (!startswith(Prefix))
+ return false;
+
+ *this = drop_front(Prefix.size());
+ return true;
+ }
+
+ /// Returns true if this StringRef has the given suffix and removes that
+ /// suffix.
+ bool consume_back(StringRef Suffix) {
+ if (!endswith(Suffix))
+ return false;
+
+ *this = drop_back(Suffix.size());
+ return true;
+ }
+
+ /// Return a reference to the substring from [Start, End).
+ ///
+ /// \param Start The index of the starting character in the substring; if
+ /// the index is npos or greater than the length of the string then the
+ /// empty substring will be returned.
+ ///
+ /// \param End The index following the last character to include in the
+ /// substring. If this is npos or exceeds the number of characters
+ /// remaining in the string, the string suffix (starting with \p Start)
+ /// will be returned. If this is less than \p Start, an empty string will
+ /// be returned.
+ LLVM_NODISCARD
+ StringRef slice(size_t Start, size_t End) const {
+ Start = std::min(Start, Length);
+ End = std::min(std::max(Start, End), Length);
+ return StringRef(Data + Start, End - Start);
+ }
+
+ /// Split into two substrings around the first occurrence of a separator
+ /// character.
+ ///
+ /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
+ /// such that (*this == LHS + Separator + RHS) is true and RHS is
+ /// maximal. If \p Separator is not in the string, then the result is a
+ /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
+ ///
+ /// \param Separator The character to split on.
+ /// \returns The split substrings.
+ LLVM_NODISCARD
+ std::pair<StringRef, StringRef> split(char Separator) const {
+ return split(StringRef(&Separator, 1));
+ }
+
+ /// Split into two substrings around the first occurrence of a separator
+ /// string.
+ ///
+ /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
+ /// such that (*this == LHS + Separator + RHS) is true and RHS is
+ /// maximal. If \p Separator is not in the string, then the result is a
+ /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
+ ///
+ /// \param Separator - The string to split on.
+ /// \return - The split substrings.
+ LLVM_NODISCARD
+ std::pair<StringRef, StringRef> split(StringRef Separator) const {
+ size_t Idx = find(Separator);
+ if (Idx == npos)
+ return std::make_pair(*this, StringRef());
+ return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos));
+ }
+
+ /// Split into two substrings around the last occurrence of a separator
+ /// string.
+ ///
+ /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
+ /// such that (*this == LHS + Separator + RHS) is true and RHS is
+ /// minimal. If \p Separator is not in the string, then the result is a
+ /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
+ ///
+ /// \param Separator - The string to split on.
+ /// \return - The split substrings.
+ LLVM_NODISCARD
+ std::pair<StringRef, StringRef> rsplit(StringRef Separator) const {
+ size_t Idx = rfind(Separator);
+ if (Idx == npos)
+ return std::make_pair(*this, StringRef());
+ return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos));
+ }
+
+ /// Split into substrings around the occurrences of a separator string.
+ ///
+ /// Each substring is stored in \p A. If \p MaxSplit is >= 0, at most
+ /// \p MaxSplit splits are done and consequently <= \p MaxSplit + 1
+ /// elements are added to A.
+ /// If \p KeepEmpty is false, empty strings are not added to \p A. They
+ /// still count when considering \p MaxSplit
+ /// An useful invariant is that
+ /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true
+ ///
+ /// \param A - Where to put the substrings.
+ /// \param Separator - The string to split on.
+ /// \param MaxSplit - The maximum number of times the string is split.
+ /// \param KeepEmpty - True if empty substring should be added.
+ void split(SmallVectorImpl<StringRef> &A,
+ StringRef Separator, int MaxSplit = -1,
+ bool KeepEmpty = true) const;
+
+ /// Split into substrings around the occurrences of a separator character.
+ ///
+ /// Each substring is stored in \p A. If \p MaxSplit is >= 0, at most
+ /// \p MaxSplit splits are done and consequently <= \p MaxSplit + 1
+ /// elements are added to A.
+ /// If \p KeepEmpty is false, empty strings are not added to \p A. They
+ /// still count when considering \p MaxSplit
+ /// An useful invariant is that
+ /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true
+ ///
+ /// \param A - Where to put the substrings.
+ /// \param Separator - The string to split on.
+ /// \param MaxSplit - The maximum number of times the string is split.
+ /// \param KeepEmpty - True if empty substring should be added.
+ void split(SmallVectorImpl<StringRef> &A, char Separator, int MaxSplit = -1,
+ bool KeepEmpty = true) const;
+
+ /// Split into two substrings around the last occurrence of a separator
+ /// character.
+ ///
+ /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
+ /// such that (*this == LHS + Separator + RHS) is true and RHS is
+ /// minimal. If \p Separator is not in the string, then the result is a
+ /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
+ ///
+ /// \param Separator - The character to split on.
+ /// \return - The split substrings.
+ LLVM_NODISCARD
+ std::pair<StringRef, StringRef> rsplit(char Separator) const {
+ return rsplit(StringRef(&Separator, 1));
+ }
+
+ /// Return string with consecutive \p Char characters starting from the
+ /// the left removed.
+ LLVM_NODISCARD
+ StringRef ltrim(char Char) const {
+ return drop_front(std::min(Length, find_first_not_of(Char)));
+ }
+
+ /// Return string with consecutive characters in \p Chars starting from
+ /// the left removed.
+ LLVM_NODISCARD
+ StringRef ltrim(StringRef Chars = " \t\n\v\f\r") const {
+ return drop_front(std::min(Length, find_first_not_of(Chars)));
+ }
+
+ /// Return string with consecutive \p Char characters starting from the
+ /// right removed.
+ LLVM_NODISCARD
+ StringRef rtrim(char Char) const {
+ return drop_back(Length - std::min(Length, find_last_not_of(Char) + 1));
+ }
+
+ /// Return string with consecutive characters in \p Chars starting from
+ /// the right removed.
+ LLVM_NODISCARD
+ StringRef rtrim(StringRef Chars = " \t\n\v\f\r") const {
+ return drop_back(Length - std::min(Length, find_last_not_of(Chars) + 1));
+ }
+
+ /// Return string with consecutive \p Char characters starting from the
+ /// left and right removed.
+ LLVM_NODISCARD
+ StringRef trim(char Char) const {
+ return ltrim(Char).rtrim(Char);
+ }
+
+ /// Return string with consecutive characters in \p Chars starting from
+ /// the left and right removed.
+ LLVM_NODISCARD
+ StringRef trim(StringRef Chars = " \t\n\v\f\r") const {
+ return ltrim(Chars).rtrim(Chars);
+ }
+
+ /// @}
+ };
+
+ /// A wrapper around a string literal that serves as a proxy for constructing
+ /// global tables of StringRefs with the length computed at compile time.
+ /// In order to avoid the invocation of a global constructor, StringLiteral
+ /// should *only* be used in a constexpr context, as such:
+ ///
+ /// constexpr StringLiteral S("test");
+ ///
+ class StringLiteral : public StringRef {
+ private:
+ constexpr StringLiteral(const char *Str, size_t N) : StringRef(Str, N) {
+ }
+
+ public:
+ template <size_t N>
+ constexpr StringLiteral(const char (&Str)[N])
+#if defined(__clang__) && __has_attribute(enable_if)
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wgcc-compat"
+ __attribute((enable_if(__builtin_strlen(Str) == N - 1,
+ "invalid string literal")))
+#pragma clang diagnostic pop
+#endif
+ : StringRef(Str, N - 1) {
+ }
+
+ // Explicit construction for strings like "foo\0bar".
+ template <size_t N>
+ static constexpr StringLiteral withInnerNUL(const char (&Str)[N]) {
+ return StringLiteral(Str, N - 1);
+ }
+ };
+
+ /// @name StringRef Comparison Operators
+ /// @{
+
+ inline bool operator==(StringRef LHS, StringRef RHS) {
+ return LHS.equals(RHS);
+ }
+
+ inline bool operator!=(StringRef LHS, StringRef RHS) { return !(LHS == RHS); }
+
+ inline bool operator<(StringRef LHS, StringRef RHS) {
+ return LHS.compare(RHS) == -1;
+ }
+
+ inline bool operator<=(StringRef LHS, StringRef RHS) {
+ return LHS.compare(RHS) != 1;
+ }
+
+ inline bool operator>(StringRef LHS, StringRef RHS) {
+ return LHS.compare(RHS) == 1;
+ }
+
+ inline bool operator>=(StringRef LHS, StringRef RHS) {
+ return LHS.compare(RHS) != -1;
+ }
+
+ inline std::string &operator+=(std::string &buffer, StringRef string) {
+ return buffer.append(string.data(), string.size());
+ }
+
+ /// @}
+
+ /// Compute a hash_code for a StringRef.
+ LLVM_NODISCARD
+ hash_code hash_value(StringRef S);
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STRINGREF_H
diff --git a/third_party/llvm-project/include/llvm/ADT/StringSet.h b/third_party/llvm-project/include/llvm/ADT/StringSet.h
new file mode 100644
index 000000000..60be09d3c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/StringSet.h
@@ -0,0 +1,58 @@
+//===- StringSet.h - The LLVM Compiler Driver -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// StringSet - A set-like wrapper for the StringMap.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_STRINGSET_H
+#define LLVM_ADT_STRINGSET_H
+
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Allocator.h"
+#include <cassert>
+#include <initializer_list>
+#include <utility>
+
+namespace llvm {
+
+ /// StringSet - A wrapper for StringMap that provides set-like functionality.
+ template <class AllocatorTy = MallocAllocator>
+ class StringSet : public StringMap<NoneType, AllocatorTy> {
+ using base = StringMap<NoneType, AllocatorTy>;
+
+ public:
+ StringSet() = default;
+ StringSet(std::initializer_list<StringRef> S) {
+ for (StringRef X : S)
+ insert(X);
+ }
+ explicit StringSet(AllocatorTy A) : base(A) {}
+
+ std::pair<typename base::iterator, bool> insert(StringRef Key) {
+ assert(!Key.empty());
+ return base::insert(std::make_pair(Key, None));
+ }
+
+ template <typename InputIt>
+ void insert(const InputIt &Begin, const InputIt &End) {
+ for (auto It = Begin; It != End; ++It)
+ base::insert(std::make_pair(*It, None));
+ }
+
+ template <typename ValueTy>
+ std::pair<typename base::iterator, bool>
+ insert(const StringMapEntry<ValueTy> &MapEntry) {
+ return insert(MapEntry.getKey());
+ }
+ };
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STRINGSET_H
diff --git a/third_party/llvm-project/include/llvm/ADT/StringSwitch.h b/third_party/llvm-project/include/llvm/ADT/StringSwitch.h
new file mode 100644
index 000000000..fea911f69
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/StringSwitch.h
@@ -0,0 +1,196 @@
+//===--- StringSwitch.h - Switch-on-literal-string Construct --------------===/
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//===----------------------------------------------------------------------===/
+//
+// This file implements the StringSwitch template, which mimics a switch()
+// statement whose cases are string literals.
+//
+//===----------------------------------------------------------------------===/
+#ifndef LLVM_ADT_STRINGSWITCH_H
+#define LLVM_ADT_STRINGSWITCH_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Compiler.h"
+#include <cassert>
+#include <cstring>
+
+namespace llvm {
+
+/// A switch()-like statement whose cases are string literals.
+///
+/// The StringSwitch class is a simple form of a switch() statement that
+/// determines whether the given string matches one of the given string
+/// literals. The template type parameter \p T is the type of the value that
+/// will be returned from the string-switch expression. For example,
+/// the following code switches on the name of a color in \c argv[i]:
+///
+/// \code
+/// Color color = StringSwitch<Color>(argv[i])
+/// .Case("red", Red)
+/// .Case("orange", Orange)
+/// .Case("yellow", Yellow)
+/// .Case("green", Green)
+/// .Case("blue", Blue)
+/// .Case("indigo", Indigo)
+/// .Cases("violet", "purple", Violet)
+/// .Default(UnknownColor);
+/// \endcode
+template<typename T, typename R = T>
+class StringSwitch {
+ /// The string we are matching.
+ const StringRef Str;
+
+ /// The pointer to the result of this switch statement, once known,
+ /// null before that.
+ Optional<T> Result;
+
+public:
+ explicit StringSwitch(StringRef S)
+ : Str(S), Result() { }
+
+ // StringSwitch is not copyable.
+ StringSwitch(const StringSwitch &) = delete;
+
+ // StringSwitch is not assignable due to 'Str' being 'const'.
+ void operator=(const StringSwitch &) = delete;
+ void operator=(StringSwitch &&other) = delete;
+
+ StringSwitch(StringSwitch &&other)
+ : Str(other.Str), Result(std::move(other.Result)) { }
+
+ ~StringSwitch() = default;
+
+ // Case-sensitive case matchers
+ StringSwitch &Case(StringLiteral S, T Value) {
+ if (!Result && Str == S) {
+ Result = std::move(Value);
+ }
+ return *this;
+ }
+
+ StringSwitch& EndsWith(StringLiteral S, T Value) {
+ if (!Result && Str.endswith(S)) {
+ Result = std::move(Value);
+ }
+ return *this;
+ }
+
+ StringSwitch& StartsWith(StringLiteral S, T Value) {
+ if (!Result && Str.startswith(S)) {
+ Result = std::move(Value);
+ }
+ return *this;
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, T Value) {
+ return Case(S0, Value).Case(S1, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ T Value) {
+ return Case(S0, Value).Cases(S1, S2, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, StringLiteral S5,
+ T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, S5, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, StringLiteral S5,
+ StringLiteral S6, T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, S5, S6, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, StringLiteral S5,
+ StringLiteral S6, StringLiteral S7, T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, S5, S6, S7, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, StringLiteral S5,
+ StringLiteral S6, StringLiteral S7, StringLiteral S8,
+ T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, S5, S6, S7, S8, Value);
+ }
+
+ StringSwitch &Cases(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, StringLiteral S5,
+ StringLiteral S6, StringLiteral S7, StringLiteral S8,
+ StringLiteral S9, T Value) {
+ return Case(S0, Value).Cases(S1, S2, S3, S4, S5, S6, S7, S8, S9, Value);
+ }
+
+ // Case-insensitive case matchers.
+ StringSwitch &CaseLower(StringLiteral S, T Value) {
+ if (!Result && Str.equals_lower(S))
+ Result = std::move(Value);
+
+ return *this;
+ }
+
+ StringSwitch &EndsWithLower(StringLiteral S, T Value) {
+ if (!Result && Str.endswith_lower(S))
+ Result = Value;
+
+ return *this;
+ }
+
+ StringSwitch &StartsWithLower(StringLiteral S, T Value) {
+ if (!Result && Str.startswith_lower(S))
+ Result = std::move(Value);
+
+ return *this;
+ }
+
+ StringSwitch &CasesLower(StringLiteral S0, StringLiteral S1, T Value) {
+ return CaseLower(S0, Value).CaseLower(S1, Value);
+ }
+
+ StringSwitch &CasesLower(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ T Value) {
+ return CaseLower(S0, Value).CasesLower(S1, S2, Value);
+ }
+
+ StringSwitch &CasesLower(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, T Value) {
+ return CaseLower(S0, Value).CasesLower(S1, S2, S3, Value);
+ }
+
+ StringSwitch &CasesLower(StringLiteral S0, StringLiteral S1, StringLiteral S2,
+ StringLiteral S3, StringLiteral S4, T Value) {
+ return CaseLower(S0, Value).CasesLower(S1, S2, S3, S4, Value);
+ }
+
+ LLVM_NODISCARD
+ R Default(T Value) {
+ if (Result)
+ return std::move(*Result);
+ return Value;
+ }
+
+ LLVM_NODISCARD
+ operator R() {
+ assert(Result && "Fell off the end of a string-switch");
+ return std::move(*Result);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_STRINGSWITCH_H
diff --git a/third_party/llvm-project/include/llvm/ADT/Triple.h b/third_party/llvm-project/include/llvm/ADT/Triple.h
new file mode 100644
index 000000000..edeb31efa
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/Triple.h
@@ -0,0 +1,881 @@
+//===-- llvm/ADT/Triple.h - Target triple helper class ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_TRIPLE_H
+#define LLVM_ADT_TRIPLE_H
+
+#include "llvm/ADT/Twine.h"
+
+// Some system headers or GCC predefined macros conflict with identifiers in
+// this file. Undefine them here.
+#undef NetBSD
+#undef mips
+#undef sparc
+
+namespace llvm {
+
+/// Triple - Helper class for working with autoconf configuration names. For
+/// historical reasons, we also call these 'triples' (they used to contain
+/// exactly three fields).
+///
+/// Configuration names are strings in the canonical form:
+/// ARCHITECTURE-VENDOR-OPERATING_SYSTEM
+/// or
+/// ARCHITECTURE-VENDOR-OPERATING_SYSTEM-ENVIRONMENT
+///
+/// This class is used for clients which want to support arbitrary
+/// configuration names, but also want to implement certain special
+/// behavior for particular configurations. This class isolates the mapping
+/// from the components of the configuration name to well known IDs.
+///
+/// At its core the Triple class is designed to be a wrapper for a triple
+/// string; the constructor does not change or normalize the triple string.
+/// Clients that need to handle the non-canonical triples that users often
+/// specify should use the normalize method.
+///
+/// See autoconf/config.guess for a glimpse into what configuration names
+/// look like in practice.
+class Triple {
+public:
+ enum ArchType {
+ UnknownArch,
+
+ arm, // ARM (little endian): arm, armv.*, xscale
+ armeb, // ARM (big endian): armeb
+ aarch64, // AArch64 (little endian): aarch64
+ aarch64_be, // AArch64 (big endian): aarch64_be
+ aarch64_32, // AArch64 (little endian) ILP32: aarch64_32
+ arc, // ARC: Synopsys ARC
+ avr, // AVR: Atmel AVR microcontroller
+ bpfel, // eBPF or extended BPF or 64-bit BPF (little endian)
+ bpfeb, // eBPF or extended BPF or 64-bit BPF (big endian)
+ hexagon, // Hexagon: hexagon
+ mips, // MIPS: mips, mipsallegrex, mipsr6
+ mipsel, // MIPSEL: mipsel, mipsallegrexe, mipsr6el
+ mips64, // MIPS64: mips64, mips64r6, mipsn32, mipsn32r6
+ mips64el, // MIPS64EL: mips64el, mips64r6el, mipsn32el, mipsn32r6el
+ msp430, // MSP430: msp430
+ ppc, // PPC: powerpc
+ ppc64, // PPC64: powerpc64, ppu
+ ppc64le, // PPC64LE: powerpc64le
+ r600, // R600: AMD GPUs HD2XXX - HD6XXX
+ amdgcn, // AMDGCN: AMD GCN GPUs
+ riscv32, // RISC-V (32-bit): riscv32
+ riscv64, // RISC-V (64-bit): riscv64
+ sparc, // Sparc: sparc
+ sparcv9, // Sparcv9: Sparcv9
+ sparcel, // Sparc: (endianness = little). NB: 'Sparcle' is a CPU variant
+ systemz, // SystemZ: s390x
+ tce, // TCE (http://tce.cs.tut.fi/): tce
+ tcele, // TCE little endian (http://tce.cs.tut.fi/): tcele
+ thumb, // Thumb (little endian): thumb, thumbv.*
+ thumbeb, // Thumb (big endian): thumbeb
+ x86, // X86: i[3-9]86
+ x86_64, // X86-64: amd64, x86_64
+ xcore, // XCore: xcore
+ nvptx, // NVPTX: 32-bit
+ nvptx64, // NVPTX: 64-bit
+ le32, // le32: generic little-endian 32-bit CPU (PNaCl)
+ le64, // le64: generic little-endian 64-bit CPU (PNaCl)
+ amdil, // AMDIL
+ amdil64, // AMDIL with 64-bit pointers
+ hsail, // AMD HSAIL
+ hsail64, // AMD HSAIL with 64-bit pointers
+ spir, // SPIR: standard portable IR for OpenCL 32-bit version
+ spir64, // SPIR: standard portable IR for OpenCL 64-bit version
+ kalimba, // Kalimba: generic kalimba
+ shave, // SHAVE: Movidius vector VLIW processors
+ lanai, // Lanai: Lanai 32-bit
+ wasm32, // WebAssembly with 32-bit pointers
+ wasm64, // WebAssembly with 64-bit pointers
+ renderscript32, // 32-bit RenderScript
+ renderscript64, // 64-bit RenderScript
+ LastArchType = renderscript64
+ };
+ enum SubArchType {
+ NoSubArch,
+
+ ARMSubArch_v8_5a,
+ ARMSubArch_v8_4a,
+ ARMSubArch_v8_3a,
+ ARMSubArch_v8_2a,
+ ARMSubArch_v8_1a,
+ ARMSubArch_v8,
+ ARMSubArch_v8r,
+ ARMSubArch_v8m_baseline,
+ ARMSubArch_v8m_mainline,
+ ARMSubArch_v8_1m_mainline,
+ ARMSubArch_v7,
+ ARMSubArch_v7em,
+ ARMSubArch_v7m,
+ ARMSubArch_v7s,
+ ARMSubArch_v7k,
+ ARMSubArch_v7ve,
+ ARMSubArch_v6,
+ ARMSubArch_v6m,
+ ARMSubArch_v6k,
+ ARMSubArch_v6t2,
+ ARMSubArch_v5,
+ ARMSubArch_v5te,
+ ARMSubArch_v4t,
+
+ KalimbaSubArch_v3,
+ KalimbaSubArch_v4,
+ KalimbaSubArch_v5,
+
+ MipsSubArch_r6
+ };
+ enum VendorType {
+ UnknownVendor,
+
+ Apple,
+ PC,
+ SCEI,
+ BGP,
+ BGQ,
+ Freescale,
+ IBM,
+ ImaginationTechnologies,
+ MipsTechnologies,
+ NVIDIA,
+ CSR,
+ Myriad,
+ AMD,
+ Mesa,
+ SUSE,
+ OpenEmbedded,
+ LastVendorType = OpenEmbedded
+ };
+ enum OSType {
+ UnknownOS,
+
+ Ananas,
+ CloudABI,
+ Darwin,
+ DragonFly,
+ FreeBSD,
+ Fuchsia,
+ IOS,
+ KFreeBSD,
+ Linux,
+ Lv2, // PS3
+ MacOSX,
+ NetBSD,
+ OpenBSD,
+ Solaris,
+ Win32,
+ Haiku,
+ Minix,
+ RTEMS,
+ NaCl, // Native Client
+ CNK, // BG/P Compute-Node Kernel
+ AIX,
+ CUDA, // NVIDIA CUDA
+ NVCL, // NVIDIA OpenCL
+ AMDHSA, // AMD HSA Runtime
+ PS4,
+ ELFIAMCU,
+ TvOS, // Apple tvOS
+ WatchOS, // Apple watchOS
+ Mesa3D,
+ Contiki,
+ AMDPAL, // AMD PAL Runtime
+ HermitCore, // HermitCore Unikernel/Multikernel
+ Hurd, // GNU/Hurd
+ WASI, // Experimental WebAssembly OS
+ Emscripten,
+ LastOSType = Emscripten
+ };
+ enum EnvironmentType {
+ UnknownEnvironment,
+
+ GNU,
+ GNUABIN32,
+ GNUABI64,
+ GNUEABI,
+ GNUEABIHF,
+ GNUX32,
+ CODE16,
+ EABI,
+ EABIHF,
+ ELFv1,
+ ELFv2,
+ Android,
+ Musl,
+ MuslEABI,
+ MuslEABIHF,
+
+ MSVC,
+ Itanium,
+ Cygnus,
+ CoreCLR,
+ Simulator, // Simulator variants of other systems, e.g., Apple's iOS
+ MacABI, // Mac Catalyst variant of Apple's iOS deployment target.
+ LastEnvironmentType = MacABI
+ };
+ enum ObjectFormatType {
+ UnknownObjectFormat,
+
+ COFF,
+ ELF,
+ MachO,
+ Wasm,
+ XCOFF,
+ };
+
+private:
+ std::string Data;
+
+ /// The parsed arch type.
+ ArchType Arch;
+
+ /// The parsed subarchitecture type.
+ SubArchType SubArch;
+
+ /// The parsed vendor type.
+ VendorType Vendor;
+
+ /// The parsed OS type.
+ OSType OS;
+
+ /// The parsed Environment type.
+ EnvironmentType Environment;
+
+ /// The object format type.
+ ObjectFormatType ObjectFormat;
+
+public:
+ /// @name Constructors
+ /// @{
+
+ /// Default constructor is the same as an empty string and leaves all
+ /// triple fields unknown.
+ Triple()
+ : Data(), Arch(), SubArch(), Vendor(), OS(), Environment(),
+ ObjectFormat() {}
+
+ explicit Triple(const Twine &Str);
+ Triple(const Twine &ArchStr, const Twine &VendorStr, const Twine &OSStr);
+ Triple(const Twine &ArchStr, const Twine &VendorStr, const Twine &OSStr,
+ const Twine &EnvironmentStr);
+
+ bool operator==(const Triple &Other) const {
+ return Arch == Other.Arch && SubArch == Other.SubArch &&
+ Vendor == Other.Vendor && OS == Other.OS &&
+ Environment == Other.Environment &&
+ ObjectFormat == Other.ObjectFormat;
+ }
+
+ bool operator!=(const Triple &Other) const {
+ return !(*this == Other);
+ }
+
+ /// @}
+ /// @name Normalization
+ /// @{
+
+ /// normalize - Turn an arbitrary machine specification into the canonical
+ /// triple form (or something sensible that the Triple class understands if
+ /// nothing better can reasonably be done). In particular, it handles the
+ /// common case in which otherwise valid components are in the wrong order.
+ static std::string normalize(StringRef Str);
+
+ /// Return the normalized form of this triple's string.
+ std::string normalize() const { return normalize(Data); }
+
+ /// @}
+ /// @name Typed Component Access
+ /// @{
+
+ /// getArch - Get the parsed architecture type of this triple.
+ ArchType getArch() const { return Arch; }
+
+ /// getSubArch - get the parsed subarchitecture type for this triple.
+ SubArchType getSubArch() const { return SubArch; }
+
+ /// getVendor - Get the parsed vendor type of this triple.
+ VendorType getVendor() const { return Vendor; }
+
+ /// getOS - Get the parsed operating system type of this triple.
+ OSType getOS() const { return OS; }
+
+ /// hasEnvironment - Does this triple have the optional environment
+ /// (fourth) component?
+ bool hasEnvironment() const {
+ return getEnvironmentName() != "";
+ }
+
+ /// getEnvironment - Get the parsed environment type of this triple.
+ EnvironmentType getEnvironment() const { return Environment; }
+
+ /// Parse the version number from the OS name component of the
+ /// triple, if present.
+ ///
+ /// For example, "fooos1.2.3" would return (1, 2, 3).
+ ///
+ /// If an entry is not defined, it will be returned as 0.
+ void getEnvironmentVersion(unsigned &Major, unsigned &Minor,
+ unsigned &Micro) const;
+
+ /// getFormat - Get the object format for this triple.
+ ObjectFormatType getObjectFormat() const { return ObjectFormat; }
+
+ /// getOSVersion - Parse the version number from the OS name component of the
+ /// triple, if present.
+ ///
+ /// For example, "fooos1.2.3" would return (1, 2, 3).
+ ///
+ /// If an entry is not defined, it will be returned as 0.
+ void getOSVersion(unsigned &Major, unsigned &Minor, unsigned &Micro) const;
+
+ /// getOSMajorVersion - Return just the major version number, this is
+ /// specialized because it is a common query.
+ unsigned getOSMajorVersion() const {
+ unsigned Maj, Min, Micro;
+ getOSVersion(Maj, Min, Micro);
+ return Maj;
+ }
+
+ /// getMacOSXVersion - Parse the version number as with getOSVersion and then
+ /// translate generic "darwin" versions to the corresponding OS X versions.
+ /// This may also be called with IOS triples but the OS X version number is
+ /// just set to a constant 10.4.0 in that case. Returns true if successful.
+ bool getMacOSXVersion(unsigned &Major, unsigned &Minor,
+ unsigned &Micro) const;
+
+ /// getiOSVersion - Parse the version number as with getOSVersion. This should
+ /// only be called with IOS or generic triples.
+ void getiOSVersion(unsigned &Major, unsigned &Minor,
+ unsigned &Micro) const;
+
+ /// getWatchOSVersion - Parse the version number as with getOSVersion. This
+ /// should only be called with WatchOS or generic triples.
+ void getWatchOSVersion(unsigned &Major, unsigned &Minor,
+ unsigned &Micro) const;
+
+ /// @}
+ /// @name Direct Component Access
+ /// @{
+
+ const std::string &str() const { return Data; }
+
+ const std::string &getTriple() const { return Data; }
+
+ /// getArchName - Get the architecture (first) component of the
+ /// triple.
+ StringRef getArchName() const;
+
+ /// getVendorName - Get the vendor (second) component of the triple.
+ StringRef getVendorName() const;
+
+ /// getOSName - Get the operating system (third) component of the
+ /// triple.
+ StringRef getOSName() const;
+
+ /// getEnvironmentName - Get the optional environment (fourth)
+ /// component of the triple, or "" if empty.
+ StringRef getEnvironmentName() const;
+
+ /// getOSAndEnvironmentName - Get the operating system and optional
+ /// environment components as a single string (separated by a '-'
+ /// if the environment component is present).
+ StringRef getOSAndEnvironmentName() const;
+
+ /// @}
+ /// @name Convenience Predicates
+ /// @{
+
+ /// Test whether the architecture is 64-bit
+ ///
+ /// Note that this tests for 64-bit pointer width, and nothing else. Note
+ /// that we intentionally expose only three predicates, 64-bit, 32-bit, and
+ /// 16-bit. The inner details of pointer width for particular architectures
+ /// is not summed up in the triple, and so only a coarse grained predicate
+ /// system is provided.
+ bool isArch64Bit() const;
+
+ /// Test whether the architecture is 32-bit
+ ///
+ /// Note that this tests for 32-bit pointer width, and nothing else.
+ bool isArch32Bit() const;
+
+ /// Test whether the architecture is 16-bit
+ ///
+ /// Note that this tests for 16-bit pointer width, and nothing else.
+ bool isArch16Bit() const;
+
+ /// isOSVersionLT - Helper function for doing comparisons against version
+ /// numbers included in the target triple.
+ bool isOSVersionLT(unsigned Major, unsigned Minor = 0,
+ unsigned Micro = 0) const {
+ unsigned LHS[3];
+ getOSVersion(LHS[0], LHS[1], LHS[2]);
+
+ if (LHS[0] != Major)
+ return LHS[0] < Major;
+ if (LHS[1] != Minor)
+ return LHS[1] < Minor;
+ if (LHS[2] != Micro)
+ return LHS[2] < Micro;
+
+ return false;
+ }
+
+ bool isOSVersionLT(const Triple &Other) const {
+ unsigned RHS[3];
+ Other.getOSVersion(RHS[0], RHS[1], RHS[2]);
+ return isOSVersionLT(RHS[0], RHS[1], RHS[2]);
+ }
+
+ /// isMacOSXVersionLT - Comparison function for checking OS X version
+ /// compatibility, which handles supporting skewed version numbering schemes
+ /// used by the "darwin" triples.
+ bool isMacOSXVersionLT(unsigned Major, unsigned Minor = 0,
+ unsigned Micro = 0) const {
+ assert(isMacOSX() && "Not an OS X triple!");
+
+ // If this is OS X, expect a sane version number.
+ if (getOS() == Triple::MacOSX)
+ return isOSVersionLT(Major, Minor, Micro);
+
+ // Otherwise, compare to the "Darwin" number.
+ assert(Major == 10 && "Unexpected major version");
+ return isOSVersionLT(Minor + 4, Micro, 0);
+ }
+
+ /// isMacOSX - Is this a Mac OS X triple. For legacy reasons, we support both
+ /// "darwin" and "osx" as OS X triples.
+ bool isMacOSX() const {
+ return getOS() == Triple::Darwin || getOS() == Triple::MacOSX;
+ }
+
+ /// Is this an iOS triple.
+ /// Note: This identifies tvOS as a variant of iOS. If that ever
+ /// changes, i.e., if the two operating systems diverge or their version
+ /// numbers get out of sync, that will need to be changed.
+ /// watchOS has completely different version numbers so it is not included.
+ bool isiOS() const {
+ return getOS() == Triple::IOS || isTvOS();
+ }
+
+ /// Is this an Apple tvOS triple.
+ bool isTvOS() const {
+ return getOS() == Triple::TvOS;
+ }
+
+ /// Is this an Apple watchOS triple.
+ bool isWatchOS() const {
+ return getOS() == Triple::WatchOS;
+ }
+
+ bool isWatchABI() const {
+ return getSubArch() == Triple::ARMSubArch_v7k;
+ }
+
+ /// isOSDarwin - Is this a "Darwin" OS (OS X, iOS, or watchOS).
+ bool isOSDarwin() const {
+ return isMacOSX() || isiOS() || isWatchOS();
+ }
+
+ bool isSimulatorEnvironment() const {
+ return getEnvironment() == Triple::Simulator;
+ }
+
+ bool isMacCatalystEnvironment() const {
+ return getEnvironment() == Triple::MacABI;
+ }
+
+ bool isOSNetBSD() const {
+ return getOS() == Triple::NetBSD;
+ }
+
+ bool isOSOpenBSD() const {
+ return getOS() == Triple::OpenBSD;
+ }
+
+ bool isOSFreeBSD() const {
+ return getOS() == Triple::FreeBSD;
+ }
+
+ bool isOSFuchsia() const {
+ return getOS() == Triple::Fuchsia;
+ }
+
+ bool isOSDragonFly() const { return getOS() == Triple::DragonFly; }
+
+ bool isOSSolaris() const {
+ return getOS() == Triple::Solaris;
+ }
+
+ bool isOSIAMCU() const {
+ return getOS() == Triple::ELFIAMCU;
+ }
+
+ bool isOSUnknown() const { return getOS() == Triple::UnknownOS; }
+
+ bool isGNUEnvironment() const {
+ EnvironmentType Env = getEnvironment();
+ return Env == Triple::GNU || Env == Triple::GNUABIN32 ||
+ Env == Triple::GNUABI64 || Env == Triple::GNUEABI ||
+ Env == Triple::GNUEABIHF || Env == Triple::GNUX32;
+ }
+
+ bool isOSContiki() const {
+ return getOS() == Triple::Contiki;
+ }
+
+ /// Tests whether the OS is Haiku.
+ bool isOSHaiku() const {
+ return getOS() == Triple::Haiku;
+ }
+
+ /// Tests whether the OS is Windows.
+ bool isOSWindows() const {
+ return getOS() == Triple::Win32;
+ }
+
+ /// Checks if the environment is MSVC.
+ bool isKnownWindowsMSVCEnvironment() const {
+ return isOSWindows() && getEnvironment() == Triple::MSVC;
+ }
+
+ /// Checks if the environment could be MSVC.
+ bool isWindowsMSVCEnvironment() const {
+ return isKnownWindowsMSVCEnvironment() ||
+ (isOSWindows() && getEnvironment() == Triple::UnknownEnvironment);
+ }
+
+ bool isWindowsCoreCLREnvironment() const {
+ return isOSWindows() && getEnvironment() == Triple::CoreCLR;
+ }
+
+ bool isWindowsItaniumEnvironment() const {
+ return isOSWindows() && getEnvironment() == Triple::Itanium;
+ }
+
+ bool isWindowsCygwinEnvironment() const {
+ return isOSWindows() && getEnvironment() == Triple::Cygnus;
+ }
+
+ bool isWindowsGNUEnvironment() const {
+ return isOSWindows() && getEnvironment() == Triple::GNU;
+ }
+
+ /// Tests for either Cygwin or MinGW OS
+ bool isOSCygMing() const {
+ return isWindowsCygwinEnvironment() || isWindowsGNUEnvironment();
+ }
+
+ /// Is this a "Windows" OS targeting a "MSVCRT.dll" environment.
+ bool isOSMSVCRT() const {
+ return isWindowsMSVCEnvironment() || isWindowsGNUEnvironment() ||
+ isWindowsItaniumEnvironment();
+ }
+
+ /// Tests whether the OS is NaCl (Native Client)
+ bool isOSNaCl() const {
+ return getOS() == Triple::NaCl;
+ }
+
+ /// Tests whether the OS is Linux.
+ bool isOSLinux() const {
+ return getOS() == Triple::Linux;
+ }
+
+ /// Tests whether the OS is kFreeBSD.
+ bool isOSKFreeBSD() const {
+ return getOS() == Triple::KFreeBSD;
+ }
+
+ /// Tests whether the OS is Hurd.
+ bool isOSHurd() const {
+ return getOS() == Triple::Hurd;
+ }
+
+ /// Tests whether the OS is WASI.
+ bool isOSWASI() const {
+ return getOS() == Triple::WASI;
+ }
+
+ /// Tests whether the OS is Emscripten.
+ bool isOSEmscripten() const {
+ return getOS() == Triple::Emscripten;
+ }
+
+ /// Tests whether the OS uses glibc.
+ bool isOSGlibc() const {
+ return (getOS() == Triple::Linux || getOS() == Triple::KFreeBSD ||
+ getOS() == Triple::Hurd) &&
+ !isAndroid();
+ }
+
+ /// Tests whether the OS is AIX.
+ bool isOSAIX() const {
+ return getOS() == Triple::AIX;
+ }
+
+ /// Tests whether the OS uses the ELF binary format.
+ bool isOSBinFormatELF() const {
+ return getObjectFormat() == Triple::ELF;
+ }
+
+ /// Tests whether the OS uses the COFF binary format.
+ bool isOSBinFormatCOFF() const {
+ return getObjectFormat() == Triple::COFF;
+ }
+
+ /// Tests whether the environment is MachO.
+ bool isOSBinFormatMachO() const {
+ return getObjectFormat() == Triple::MachO;
+ }
+
+ /// Tests whether the OS uses the Wasm binary format.
+ bool isOSBinFormatWasm() const {
+ return getObjectFormat() == Triple::Wasm;
+ }
+
+ /// Tests whether the OS uses the XCOFF binary format.
+ bool isOSBinFormatXCOFF() const {
+ return getObjectFormat() == Triple::XCOFF;
+ }
+
+ /// Tests whether the target is the PS4 CPU
+ bool isPS4CPU() const {
+ return getArch() == Triple::x86_64 &&
+ getVendor() == Triple::SCEI &&
+ getOS() == Triple::PS4;
+ }
+
+ /// Tests whether the target is the PS4 platform
+ bool isPS4() const {
+ return getVendor() == Triple::SCEI &&
+ getOS() == Triple::PS4;
+ }
+
+ /// Tests whether the target is Android
+ bool isAndroid() const { return getEnvironment() == Triple::Android; }
+
+ bool isAndroidVersionLT(unsigned Major) const {
+ assert(isAndroid() && "Not an Android triple!");
+
+ unsigned Env[3];
+ getEnvironmentVersion(Env[0], Env[1], Env[2]);
+
+ // 64-bit targets did not exist before API level 21 (Lollipop).
+ if (isArch64Bit() && Env[0] < 21)
+ Env[0] = 21;
+
+ return Env[0] < Major;
+ }
+
+ /// Tests whether the environment is musl-libc
+ bool isMusl() const {
+ return getEnvironment() == Triple::Musl ||
+ getEnvironment() == Triple::MuslEABI ||
+ getEnvironment() == Triple::MuslEABIHF;
+ }
+
+ /// Tests whether the target is SPIR (32- or 64-bit).
+ bool isSPIR() const {
+ return getArch() == Triple::spir || getArch() == Triple::spir64;
+ }
+
+ /// Tests whether the target is NVPTX (32- or 64-bit).
+ bool isNVPTX() const {
+ return getArch() == Triple::nvptx || getArch() == Triple::nvptx64;
+ }
+
+ /// Tests whether the target is Thumb (little and big endian).
+ bool isThumb() const {
+ return getArch() == Triple::thumb || getArch() == Triple::thumbeb;
+ }
+
+ /// Tests whether the target is ARM (little and big endian).
+ bool isARM() const {
+ return getArch() == Triple::arm || getArch() == Triple::armeb;
+ }
+
+ /// Tests whether the target is AArch64 (little and big endian).
+ bool isAArch64() const {
+ return getArch() == Triple::aarch64 || getArch() == Triple::aarch64_be;
+ }
+
+ /// Tests whether the target is MIPS 32-bit (little and big endian).
+ bool isMIPS32() const {
+ return getArch() == Triple::mips || getArch() == Triple::mipsel;
+ }
+
+ /// Tests whether the target is MIPS 64-bit (little and big endian).
+ bool isMIPS64() const {
+ return getArch() == Triple::mips64 || getArch() == Triple::mips64el;
+ }
+
+ /// Tests whether the target is MIPS (little and big endian, 32- or 64-bit).
+ bool isMIPS() const {
+ return isMIPS32() || isMIPS64();
+ }
+
+ /// Tests whether the target is 64-bit PowerPC (little and big endian).
+ bool isPPC64() const {
+ return getArch() == Triple::ppc64 || getArch() == Triple::ppc64le;
+ }
+
+ /// Tests whether the target is RISC-V (32- and 64-bit).
+ bool isRISCV() const {
+ return getArch() == Triple::riscv32 || getArch() == Triple::riscv64;
+ }
+
+ /// Tests whether the target supports comdat
+ bool supportsCOMDAT() const {
+ return !isOSBinFormatMachO();
+ }
+
+ /// Tests whether the target uses emulated TLS as default.
+ bool hasDefaultEmulatedTLS() const {
+ return isAndroid() || isOSOpenBSD() || isWindowsCygwinEnvironment();
+ }
+
+ /// @}
+ /// @name Mutators
+ /// @{
+
+ /// setArch - Set the architecture (first) component of the triple
+ /// to a known type.
+ void setArch(ArchType Kind);
+
+ /// setVendor - Set the vendor (second) component of the triple to a
+ /// known type.
+ void setVendor(VendorType Kind);
+
+ /// setOS - Set the operating system (third) component of the triple
+ /// to a known type.
+ void setOS(OSType Kind);
+
+ /// setEnvironment - Set the environment (fourth) component of the triple
+ /// to a known type.
+ void setEnvironment(EnvironmentType Kind);
+
+ /// setObjectFormat - Set the object file format
+ void setObjectFormat(ObjectFormatType Kind);
+
+ /// setTriple - Set all components to the new triple \p Str.
+ void setTriple(const Twine &Str);
+
+ /// setArchName - Set the architecture (first) component of the
+ /// triple by name.
+ void setArchName(StringRef Str);
+
+ /// setVendorName - Set the vendor (second) component of the triple
+ /// by name.
+ void setVendorName(StringRef Str);
+
+ /// setOSName - Set the operating system (third) component of the
+ /// triple by name.
+ void setOSName(StringRef Str);
+
+ /// setEnvironmentName - Set the optional environment (fourth)
+ /// component of the triple by name.
+ void setEnvironmentName(StringRef Str);
+
+ /// setOSAndEnvironmentName - Set the operating system and optional
+ /// environment components with a single string.
+ void setOSAndEnvironmentName(StringRef Str);
+
+ /// @}
+ /// @name Helpers to build variants of a particular triple.
+ /// @{
+
+ /// Form a triple with a 32-bit variant of the current architecture.
+ ///
+ /// This can be used to move across "families" of architectures where useful.
+ ///
+ /// \returns A new triple with a 32-bit architecture or an unknown
+ /// architecture if no such variant can be found.
+ llvm::Triple get32BitArchVariant() const;
+
+ /// Form a triple with a 64-bit variant of the current architecture.
+ ///
+ /// This can be used to move across "families" of architectures where useful.
+ ///
+ /// \returns A new triple with a 64-bit architecture or an unknown
+ /// architecture if no such variant can be found.
+ llvm::Triple get64BitArchVariant() const;
+
+ /// Form a triple with a big endian variant of the current architecture.
+ ///
+ /// This can be used to move across "families" of architectures where useful.
+ ///
+ /// \returns A new triple with a big endian architecture or an unknown
+ /// architecture if no such variant can be found.
+ llvm::Triple getBigEndianArchVariant() const;
+
+ /// Form a triple with a little endian variant of the current architecture.
+ ///
+ /// This can be used to move across "families" of architectures where useful.
+ ///
+ /// \returns A new triple with a little endian architecture or an unknown
+ /// architecture if no such variant can be found.
+ llvm::Triple getLittleEndianArchVariant() const;
+
+ /// Get the (LLVM) name of the minimum ARM CPU for the arch we are targeting.
+ ///
+ /// \param Arch the architecture name (e.g., "armv7s"). If it is an empty
+ /// string then the triple's arch name is used.
+ StringRef getARMCPUForArch(StringRef Arch = StringRef()) const;
+
+ /// Tests whether the target triple is little endian.
+ ///
+ /// \returns true if the triple is little endian, false otherwise.
+ bool isLittleEndian() const;
+
+ /// Test whether target triples are compatible.
+ bool isCompatibleWith(const Triple &Other) const;
+
+ /// Merge target triples.
+ std::string merge(const Triple &Other) const;
+
+ /// @}
+ /// @name Static helpers for IDs.
+ /// @{
+
+ /// getArchTypeName - Get the canonical name for the \p Kind architecture.
+ static StringRef getArchTypeName(ArchType Kind);
+
+ /// getArchTypePrefix - Get the "prefix" canonical name for the \p Kind
+ /// architecture. This is the prefix used by the architecture specific
+ /// builtins, and is suitable for passing to \see
+ /// Intrinsic::getIntrinsicForGCCBuiltin().
+ ///
+ /// \return - The architecture prefix, or 0 if none is defined.
+ static StringRef getArchTypePrefix(ArchType Kind);
+
+ /// getVendorTypeName - Get the canonical name for the \p Kind vendor.
+ static StringRef getVendorTypeName(VendorType Kind);
+
+ /// getOSTypeName - Get the canonical name for the \p Kind operating system.
+ static StringRef getOSTypeName(OSType Kind);
+
+ /// getEnvironmentTypeName - Get the canonical name for the \p Kind
+ /// environment.
+ static StringRef getEnvironmentTypeName(EnvironmentType Kind);
+
+ /// @}
+ /// @name Static helpers for converting alternate architecture names.
+ /// @{
+
+ /// getArchTypeForLLVMName - The canonical type for the given LLVM
+ /// architecture name (e.g., "x86").
+ static ArchType getArchTypeForLLVMName(StringRef Str);
+
+ /// @}
+};
+
+} // End llvm namespace
+
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/Twine.h b/third_party/llvm-project/include/llvm/ADT/Twine.h
new file mode 100644
index 000000000..2dc7486c9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/Twine.h
@@ -0,0 +1,544 @@
+//===- Twine.h - Fast Temporary String Concatenation ------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_TWINE_H
+#define LLVM_ADT_TWINE_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <cassert>
+#include <cstdint>
+#include <string>
+
+namespace llvm {
+
+ class formatv_object_base;
+ class raw_ostream;
+
+ /// Twine - A lightweight data structure for efficiently representing the
+ /// concatenation of temporary values as strings.
+ ///
+ /// A Twine is a kind of rope, it represents a concatenated string using a
+ /// binary-tree, where the string is the preorder of the nodes. Since the
+ /// Twine can be efficiently rendered into a buffer when its result is used,
+ /// it avoids the cost of generating temporary values for intermediate string
+ /// results -- particularly in cases when the Twine result is never
+ /// required. By explicitly tracking the type of leaf nodes, we can also avoid
+ /// the creation of temporary strings for conversions operations (such as
+ /// appending an integer to a string).
+ ///
+ /// A Twine is not intended for use directly and should not be stored, its
+ /// implementation relies on the ability to store pointers to temporary stack
+ /// objects which may be deallocated at the end of a statement. Twines should
+ /// only be used accepted as const references in arguments, when an API wishes
+ /// to accept possibly-concatenated strings.
+ ///
+ /// Twines support a special 'null' value, which always concatenates to form
+ /// itself, and renders as an empty string. This can be returned from APIs to
+ /// effectively nullify any concatenations performed on the result.
+ ///
+ /// \b Implementation
+ ///
+ /// Given the nature of a Twine, it is not possible for the Twine's
+ /// concatenation method to construct interior nodes; the result must be
+ /// represented inside the returned value. For this reason a Twine object
+ /// actually holds two values, the left- and right-hand sides of a
+ /// concatenation. We also have nullary Twine objects, which are effectively
+ /// sentinel values that represent empty strings.
+ ///
+ /// Thus, a Twine can effectively have zero, one, or two children. The \see
+ /// isNullary(), \see isUnary(), and \see isBinary() predicates exist for
+ /// testing the number of children.
+ ///
+ /// We maintain a number of invariants on Twine objects (FIXME: Why):
+ /// - Nullary twines are always represented with their Kind on the left-hand
+ /// side, and the Empty kind on the right-hand side.
+ /// - Unary twines are always represented with the value on the left-hand
+ /// side, and the Empty kind on the right-hand side.
+ /// - If a Twine has another Twine as a child, that child should always be
+ /// binary (otherwise it could have been folded into the parent).
+ ///
+ /// These invariants are check by \see isValid().
+ ///
+ /// \b Efficiency Considerations
+ ///
+ /// The Twine is designed to yield efficient and small code for common
+ /// situations. For this reason, the concat() method is inlined so that
+ /// concatenations of leaf nodes can be optimized into stores directly into a
+ /// single stack allocated object.
+ ///
+ /// In practice, not all compilers can be trusted to optimize concat() fully,
+ /// so we provide two additional methods (and accompanying operator+
+ /// overloads) to guarantee that particularly important cases (cstring plus
+ /// StringRef) codegen as desired.
+ class Twine {
+ /// NodeKind - Represent the type of an argument.
+ enum NodeKind : unsigned char {
+ /// An empty string; the result of concatenating anything with it is also
+ /// empty.
+ NullKind,
+
+ /// The empty string.
+ EmptyKind,
+
+ /// A pointer to a Twine instance.
+ TwineKind,
+
+ /// A pointer to a C string instance.
+ CStringKind,
+
+ /// A pointer to an std::string instance.
+ StdStringKind,
+
+ /// A pointer to a StringRef instance.
+ StringRefKind,
+
+ /// A pointer to a SmallString instance.
+ SmallStringKind,
+
+ /// A pointer to a formatv_object_base instance.
+ FormatvObjectKind,
+
+ /// A char value, to render as a character.
+ CharKind,
+
+ /// An unsigned int value, to render as an unsigned decimal integer.
+ DecUIKind,
+
+ /// An int value, to render as a signed decimal integer.
+ DecIKind,
+
+ /// A pointer to an unsigned long value, to render as an unsigned decimal
+ /// integer.
+ DecULKind,
+
+ /// A pointer to a long value, to render as a signed decimal integer.
+ DecLKind,
+
+ /// A pointer to an unsigned long long value, to render as an unsigned
+ /// decimal integer.
+ DecULLKind,
+
+ /// A pointer to a long long value, to render as a signed decimal integer.
+ DecLLKind,
+
+ /// A pointer to a uint64_t value, to render as an unsigned hexadecimal
+ /// integer.
+ UHexKind
+ };
+
+ union Child
+ {
+ const Twine *twine;
+ const char *cString;
+ const std::string *stdString;
+ const StringRef *stringRef;
+ const SmallVectorImpl<char> *smallString;
+ const formatv_object_base *formatvObject;
+ char character;
+ unsigned int decUI;
+ int decI;
+ const unsigned long *decUL;
+ const long *decL;
+ const unsigned long long *decULL;
+ const long long *decLL;
+ const uint64_t *uHex;
+ };
+
+ /// LHS - The prefix in the concatenation, which may be uninitialized for
+ /// Null or Empty kinds.
+ Child LHS = {0};
+
+ /// RHS - The suffix in the concatenation, which may be uninitialized for
+ /// Null or Empty kinds.
+ Child RHS = {0};
+
+ /// LHSKind - The NodeKind of the left hand side, \see getLHSKind().
+ NodeKind LHSKind = EmptyKind;
+
+ /// RHSKind - The NodeKind of the right hand side, \see getRHSKind().
+ NodeKind RHSKind = EmptyKind;
+
+ /// Construct a nullary twine; the kind must be NullKind or EmptyKind.
+ explicit Twine(NodeKind Kind) : LHSKind(Kind) {
+ assert(isNullary() && "Invalid kind!");
+ }
+
+ /// Construct a binary twine.
+ explicit Twine(const Twine &LHS, const Twine &RHS)
+ : LHSKind(TwineKind), RHSKind(TwineKind) {
+ this->LHS.twine = &LHS;
+ this->RHS.twine = &RHS;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct a twine from explicit values.
+ explicit Twine(Child LHS, NodeKind LHSKind, Child RHS, NodeKind RHSKind)
+ : LHS(LHS), RHS(RHS), LHSKind(LHSKind), RHSKind(RHSKind) {
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Check for the null twine.
+ bool isNull() const {
+ return getLHSKind() == NullKind;
+ }
+
+ /// Check for the empty twine.
+ bool isEmpty() const {
+ return getLHSKind() == EmptyKind;
+ }
+
+ /// Check if this is a nullary twine (null or empty).
+ bool isNullary() const {
+ return isNull() || isEmpty();
+ }
+
+ /// Check if this is a unary twine.
+ bool isUnary() const {
+ return getRHSKind() == EmptyKind && !isNullary();
+ }
+
+ /// Check if this is a binary twine.
+ bool isBinary() const {
+ return getLHSKind() != NullKind && getRHSKind() != EmptyKind;
+ }
+
+ /// Check if this is a valid twine (satisfying the invariants on
+ /// order and number of arguments).
+ bool isValid() const {
+ // Nullary twines always have Empty on the RHS.
+ if (isNullary() && getRHSKind() != EmptyKind)
+ return false;
+
+ // Null should never appear on the RHS.
+ if (getRHSKind() == NullKind)
+ return false;
+
+ // The RHS cannot be non-empty if the LHS is empty.
+ if (getRHSKind() != EmptyKind && getLHSKind() == EmptyKind)
+ return false;
+
+ // A twine child should always be binary.
+ if (getLHSKind() == TwineKind &&
+ !LHS.twine->isBinary())
+ return false;
+ if (getRHSKind() == TwineKind &&
+ !RHS.twine->isBinary())
+ return false;
+
+ return true;
+ }
+
+ /// Get the NodeKind of the left-hand side.
+ NodeKind getLHSKind() const { return LHSKind; }
+
+ /// Get the NodeKind of the right-hand side.
+ NodeKind getRHSKind() const { return RHSKind; }
+
+ /// Print one child from a twine.
+ void printOneChild(raw_ostream &OS, Child Ptr, NodeKind Kind) const;
+
+ /// Print the representation of one child from a twine.
+ void printOneChildRepr(raw_ostream &OS, Child Ptr,
+ NodeKind Kind) const;
+
+ public:
+ /// @name Constructors
+ /// @{
+
+ /// Construct from an empty string.
+ /*implicit*/ Twine() {
+ assert(isValid() && "Invalid twine!");
+ }
+
+ Twine(const Twine &) = default;
+
+ /// Construct from a C string.
+ ///
+ /// We take care here to optimize "" into the empty twine -- this will be
+ /// optimized out for string constants. This allows Twine arguments have
+ /// default "" values, without introducing unnecessary string constants.
+ /*implicit*/ Twine(const char *Str) {
+ if (Str[0] != '\0') {
+ LHS.cString = Str;
+ LHSKind = CStringKind;
+ } else
+ LHSKind = EmptyKind;
+
+ assert(isValid() && "Invalid twine!");
+ }
+ /// Delete the implicit conversion from nullptr as Twine(const char *)
+ /// cannot take nullptr.
+ /*implicit*/ Twine(std::nullptr_t) = delete;
+
+ /// Construct from an std::string.
+ /*implicit*/ Twine(const std::string &Str) : LHSKind(StdStringKind) {
+ LHS.stdString = &Str;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct from a StringRef.
+ /*implicit*/ Twine(const StringRef &Str) : LHSKind(StringRefKind) {
+ LHS.stringRef = &Str;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct from a SmallString.
+ /*implicit*/ Twine(const SmallVectorImpl<char> &Str)
+ : LHSKind(SmallStringKind) {
+ LHS.smallString = &Str;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct from a formatv_object_base.
+ /*implicit*/ Twine(const formatv_object_base &Fmt)
+ : LHSKind(FormatvObjectKind) {
+ LHS.formatvObject = &Fmt;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct from a char.
+ explicit Twine(char Val) : LHSKind(CharKind) {
+ LHS.character = Val;
+ }
+
+ /// Construct from a signed char.
+ explicit Twine(signed char Val) : LHSKind(CharKind) {
+ LHS.character = static_cast<char>(Val);
+ }
+
+ /// Construct from an unsigned char.
+ explicit Twine(unsigned char Val) : LHSKind(CharKind) {
+ LHS.character = static_cast<char>(Val);
+ }
+
+ /// Construct a twine to print \p Val as an unsigned decimal integer.
+ explicit Twine(unsigned Val) : LHSKind(DecUIKind) {
+ LHS.decUI = Val;
+ }
+
+ /// Construct a twine to print \p Val as a signed decimal integer.
+ explicit Twine(int Val) : LHSKind(DecIKind) {
+ LHS.decI = Val;
+ }
+
+ /// Construct a twine to print \p Val as an unsigned decimal integer.
+ explicit Twine(const unsigned long &Val) : LHSKind(DecULKind) {
+ LHS.decUL = &Val;
+ }
+
+ /// Construct a twine to print \p Val as a signed decimal integer.
+ explicit Twine(const long &Val) : LHSKind(DecLKind) {
+ LHS.decL = &Val;
+ }
+
+ /// Construct a twine to print \p Val as an unsigned decimal integer.
+ explicit Twine(const unsigned long long &Val) : LHSKind(DecULLKind) {
+ LHS.decULL = &Val;
+ }
+
+ /// Construct a twine to print \p Val as a signed decimal integer.
+ explicit Twine(const long long &Val) : LHSKind(DecLLKind) {
+ LHS.decLL = &Val;
+ }
+
+ // FIXME: Unfortunately, to make sure this is as efficient as possible we
+ // need extra binary constructors from particular types. We can't rely on
+ // the compiler to be smart enough to fold operator+()/concat() down to the
+ // right thing. Yet.
+
+ /// Construct as the concatenation of a C string and a StringRef.
+ /*implicit*/ Twine(const char *LHS, const StringRef &RHS)
+ : LHSKind(CStringKind), RHSKind(StringRefKind) {
+ this->LHS.cString = LHS;
+ this->RHS.stringRef = &RHS;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Construct as the concatenation of a StringRef and a C string.
+ /*implicit*/ Twine(const StringRef &LHS, const char *RHS)
+ : LHSKind(StringRefKind), RHSKind(CStringKind) {
+ this->LHS.stringRef = &LHS;
+ this->RHS.cString = RHS;
+ assert(isValid() && "Invalid twine!");
+ }
+
+ /// Since the intended use of twines is as temporary objects, assignments
+ /// when concatenating might cause undefined behavior or stack corruptions
+ Twine &operator=(const Twine &) = delete;
+
+ /// Create a 'null' string, which is an empty string that always
+ /// concatenates to form another empty string.
+ static Twine createNull() {
+ return Twine(NullKind);
+ }
+
+ /// @}
+ /// @name Numeric Conversions
+ /// @{
+
+ // Construct a twine to print \p Val as an unsigned hexadecimal integer.
+ static Twine utohexstr(const uint64_t &Val) {
+ Child LHS, RHS;
+ LHS.uHex = &Val;
+ RHS.twine = nullptr;
+ return Twine(LHS, UHexKind, RHS, EmptyKind);
+ }
+
+ /// @}
+ /// @name Predicate Operations
+ /// @{
+
+ /// Check if this twine is trivially empty; a false return value does not
+ /// necessarily mean the twine is empty.
+ bool isTriviallyEmpty() const {
+ return isNullary();
+ }
+
+ /// Return true if this twine can be dynamically accessed as a single
+ /// StringRef value with getSingleStringRef().
+ bool isSingleStringRef() const {
+ if (getRHSKind() != EmptyKind) return false;
+
+ switch (getLHSKind()) {
+ case EmptyKind:
+ case CStringKind:
+ case StdStringKind:
+ case StringRefKind:
+ case SmallStringKind:
+ return true;
+ default:
+ return false;
+ }
+ }
+
+ /// @}
+ /// @name String Operations
+ /// @{
+
+ Twine concat(const Twine &Suffix) const;
+
+ /// @}
+ /// @name Output & Conversion.
+ /// @{
+
+ /// Return the twine contents as a std::string.
+ std::string str() const;
+
+ /// Append the concatenated string into the given SmallString or SmallVector.
+ void toVector(SmallVectorImpl<char> &Out) const;
+
+ /// This returns the twine as a single StringRef. This method is only valid
+ /// if isSingleStringRef() is true.
+ StringRef getSingleStringRef() const {
+ assert(isSingleStringRef() &&"This cannot be had as a single stringref!");
+ switch (getLHSKind()) {
+ default: llvm_unreachable("Out of sync with isSingleStringRef");
+ case EmptyKind: return StringRef();
+ case CStringKind: return StringRef(LHS.cString);
+ case StdStringKind: return StringRef(*LHS.stdString);
+ case StringRefKind: return *LHS.stringRef;
+ case SmallStringKind:
+ return StringRef(LHS.smallString->data(), LHS.smallString->size());
+ }
+ }
+
+ /// This returns the twine as a single StringRef if it can be
+ /// represented as such. Otherwise the twine is written into the given
+ /// SmallVector and a StringRef to the SmallVector's data is returned.
+ StringRef toStringRef(SmallVectorImpl<char> &Out) const {
+ if (isSingleStringRef())
+ return getSingleStringRef();
+ toVector(Out);
+ return StringRef(Out.data(), Out.size());
+ }
+
+ /// This returns the twine as a single null terminated StringRef if it
+ /// can be represented as such. Otherwise the twine is written into the
+ /// given SmallVector and a StringRef to the SmallVector's data is returned.
+ ///
+ /// The returned StringRef's size does not include the null terminator.
+ StringRef toNullTerminatedStringRef(SmallVectorImpl<char> &Out) const;
+
+ /// Write the concatenated string represented by this twine to the
+ /// stream \p OS.
+ void print(raw_ostream &OS) const;
+
+ /// Dump the concatenated string represented by this twine to stderr.
+ void dump() const;
+
+ /// Write the representation of this twine to the stream \p OS.
+ void printRepr(raw_ostream &OS) const;
+
+ /// Dump the representation of this twine to stderr.
+ void dumpRepr() const;
+
+ /// @}
+ };
+
+ /// @name Twine Inline Implementations
+ /// @{
+
+ inline Twine Twine::concat(const Twine &Suffix) const {
+ // Concatenation with null is null.
+ if (isNull() || Suffix.isNull())
+ return Twine(NullKind);
+
+ // Concatenation with empty yields the other side.
+ if (isEmpty())
+ return Suffix;
+ if (Suffix.isEmpty())
+ return *this;
+
+ // Otherwise we need to create a new node, taking care to fold in unary
+ // twines.
+ Child NewLHS, NewRHS;
+ NewLHS.twine = this;
+ NewRHS.twine = &Suffix;
+ NodeKind NewLHSKind = TwineKind, NewRHSKind = TwineKind;
+ if (isUnary()) {
+ NewLHS = LHS;
+ NewLHSKind = getLHSKind();
+ }
+ if (Suffix.isUnary()) {
+ NewRHS = Suffix.LHS;
+ NewRHSKind = Suffix.getLHSKind();
+ }
+
+ return Twine(NewLHS, NewLHSKind, NewRHS, NewRHSKind);
+ }
+
+ inline Twine operator+(const Twine &LHS, const Twine &RHS) {
+ return LHS.concat(RHS);
+ }
+
+ /// Additional overload to guarantee simplified codegen; this is equivalent to
+ /// concat().
+
+ inline Twine operator+(const char *LHS, const StringRef &RHS) {
+ return Twine(LHS, RHS);
+ }
+
+ /// Additional overload to guarantee simplified codegen; this is equivalent to
+ /// concat().
+
+ inline Twine operator+(const StringRef &LHS, const char *RHS) {
+ return Twine(LHS, RHS);
+ }
+
+ inline raw_ostream &operator<<(raw_ostream &OS, const Twine &RHS) {
+ RHS.print(OS);
+ return OS;
+ }
+
+ /// @}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_TWINE_H
diff --git a/third_party/llvm-project/include/llvm/ADT/bit.h b/third_party/llvm-project/include/llvm/ADT/bit.h
new file mode 100644
index 000000000..a790d5ed2
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/bit.h
@@ -0,0 +1,58 @@
+//===-- llvm/ADT/bit.h - C++20 <bit> ----------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the C++20 <bit> header.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_BIT_H
+#define LLVM_ADT_BIT_H
+
+#include "llvm/Support/Compiler.h"
+#include <cstring>
+#include <type_traits>
+
+namespace llvm {
+
+// This implementation of bit_cast is different from the C++17 one in two ways:
+// - It isn't constexpr because that requires compiler support.
+// - It requires trivially-constructible To, to avoid UB in the implementation.
+template <typename To, typename From
+ , typename = typename std::enable_if<sizeof(To) == sizeof(From)>::type
+#if (__has_feature(is_trivially_constructible) && defined(_LIBCPP_VERSION)) || \
+ (defined(__GNUC__) && __GNUC__ >= 5)
+ , typename = typename std::is_trivially_constructible<To>::type
+#elif __has_feature(is_trivially_constructible)
+ , typename = typename std::enable_if<__is_trivially_constructible(To)>::type
+#else
+ // See comment below.
+#endif
+#if (__has_feature(is_trivially_copyable) && defined(_LIBCPP_VERSION)) || \
+ (defined(__GNUC__) && __GNUC__ >= 5)
+ , typename = typename std::enable_if<std::is_trivially_copyable<To>::value>::type
+ , typename = typename std::enable_if<std::is_trivially_copyable<From>::value>::type
+#elif __has_feature(is_trivially_copyable)
+ , typename = typename std::enable_if<__is_trivially_copyable(To)>::type
+ , typename = typename std::enable_if<__is_trivially_copyable(From)>::type
+#else
+// This case is GCC 4.x. clang with libc++ or libstdc++ never get here. Unlike
+// llvm/Support/type_traits.h's is_trivially_copyable we don't want to
+// provide a good-enough answer here: developers in that configuration will hit
+// compilation failures on the bots instead of locally. That's acceptable
+// because it's very few developers, and only until we move past C++11.
+#endif
+>
+inline To bit_cast(const From &from) noexcept {
+ To to;
+ std::memcpy(&to, &from, sizeof(To));
+ return to;
+}
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/edit_distance.h b/third_party/llvm-project/include/llvm/ADT/edit_distance.h
new file mode 100644
index 000000000..4f5134008
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/edit_distance.h
@@ -0,0 +1,102 @@
+//===-- llvm/ADT/edit_distance.h - Array edit distance function --- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines a Levenshtein distance function that works for any two
+// sequences, with each element of each sequence being analogous to a character
+// in a string.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_EDIT_DISTANCE_H
+#define LLVM_ADT_EDIT_DISTANCE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include <algorithm>
+#include <memory>
+
+namespace llvm {
+
+/// Determine the edit distance between two sequences.
+///
+/// \param FromArray the first sequence to compare.
+///
+/// \param ToArray the second sequence to compare.
+///
+/// \param AllowReplacements whether to allow element replacements (change one
+/// element into another) as a single operation, rather than as two operations
+/// (an insertion and a removal).
+///
+/// \param MaxEditDistance If non-zero, the maximum edit distance that this
+/// routine is allowed to compute. If the edit distance will exceed that
+/// maximum, returns \c MaxEditDistance+1.
+///
+/// \returns the minimum number of element insertions, removals, or (if
+/// \p AllowReplacements is \c true) replacements needed to transform one of
+/// the given sequences into the other. If zero, the sequences are identical.
+template<typename T>
+unsigned ComputeEditDistance(ArrayRef<T> FromArray, ArrayRef<T> ToArray,
+ bool AllowReplacements = true,
+ unsigned MaxEditDistance = 0) {
+ // The algorithm implemented below is the "classic"
+ // dynamic-programming algorithm for computing the Levenshtein
+ // distance, which is described here:
+ //
+ // http://en.wikipedia.org/wiki/Levenshtein_distance
+ //
+ // Although the algorithm is typically described using an m x n
+ // array, only one row plus one element are used at a time, so this
+ // implementation just keeps one vector for the row. To update one entry,
+ // only the entries to the left, top, and top-left are needed. The left
+ // entry is in Row[x-1], the top entry is what's in Row[x] from the last
+ // iteration, and the top-left entry is stored in Previous.
+ typename ArrayRef<T>::size_type m = FromArray.size();
+ typename ArrayRef<T>::size_type n = ToArray.size();
+
+ const unsigned SmallBufferSize = 64;
+ unsigned SmallBuffer[SmallBufferSize];
+ std::unique_ptr<unsigned[]> Allocated;
+ unsigned *Row = SmallBuffer;
+ if (n + 1 > SmallBufferSize) {
+ Row = new unsigned[n + 1];
+ Allocated.reset(Row);
+ }
+
+ for (unsigned i = 1; i <= n; ++i)
+ Row[i] = i;
+
+ for (typename ArrayRef<T>::size_type y = 1; y <= m; ++y) {
+ Row[0] = y;
+ unsigned BestThisRow = Row[0];
+
+ unsigned Previous = y - 1;
+ for (typename ArrayRef<T>::size_type x = 1; x <= n; ++x) {
+ int OldRow = Row[x];
+ if (AllowReplacements) {
+ Row[x] = std::min(
+ Previous + (FromArray[y-1] == ToArray[x-1] ? 0u : 1u),
+ std::min(Row[x-1], Row[x])+1);
+ }
+ else {
+ if (FromArray[y-1] == ToArray[x-1]) Row[x] = Previous;
+ else Row[x] = std::min(Row[x-1], Row[x]) + 1;
+ }
+ Previous = OldRow;
+ BestThisRow = std::min(BestThisRow, Row[x]);
+ }
+
+ if (MaxEditDistance && BestThisRow > MaxEditDistance)
+ return MaxEditDistance + 1;
+ }
+
+ unsigned Result = Row[n];
+ return Result;
+}
+
+} // End llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/fallible_iterator.h b/third_party/llvm-project/include/llvm/ADT/fallible_iterator.h
new file mode 100644
index 000000000..6501ad223
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/fallible_iterator.h
@@ -0,0 +1,243 @@
+//===--- fallible_iterator.h - Wrapper for fallible iterators ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_FALLIBLE_ITERATOR_H
+#define LLVM_ADT_FALLIBLE_ITERATOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/Error.h"
+
+#include <type_traits>
+
+namespace llvm {
+
+/// A wrapper class for fallible iterators.
+///
+/// The fallible_iterator template wraps an underlying iterator-like class
+/// whose increment and decrement operations are replaced with fallible versions
+/// like:
+///
+/// @code{.cpp}
+/// Error inc();
+/// Error dec();
+/// @endcode
+///
+/// It produces an interface that is (mostly) compatible with a traditional
+/// c++ iterator, including ++ and -- operators that do not fail.
+///
+/// Instances of the wrapper are constructed with an instance of the
+/// underlying iterator and (for non-end iterators) a reference to an Error
+/// instance. If the underlying increment/decrement operations fail, the Error
+/// is returned via this reference, and the resulting iterator value set to an
+/// end-of-range sentinel value. This enables the following loop idiom:
+///
+/// @code{.cpp}
+/// class Archive { // E.g. Potentially malformed on-disk archive
+/// public:
+/// fallible_iterator<ArchiveChildItr> children_begin(Error &Err);
+/// fallible_iterator<ArchiveChildItr> children_end();
+/// iterator_range<fallible_iterator<ArchiveChildItr>>
+/// children(Error &Err) {
+/// return make_range(children_begin(Err), children_end());
+/// //...
+/// };
+///
+/// void walk(Archive &A) {
+/// Error Err = Error::success();
+/// for (auto &C : A.children(Err)) {
+/// // Loop body only entered when increment succeeds.
+/// }
+/// if (Err) {
+/// // handle error.
+/// }
+/// }
+/// @endcode
+///
+/// The wrapper marks the referenced Error as unchecked after each increment
+/// and/or decrement operation, and clears the unchecked flag when a non-end
+/// value is compared against end (since, by the increment invariant, not being
+/// an end value proves that there was no error, and is equivalent to checking
+/// that the Error is success). This allows early exits from the loop body
+/// without requiring redundant error checks.
+template <typename Underlying> class fallible_iterator {
+private:
+ template <typename T>
+ using enable_if_struct_deref_supported = std::enable_if<
+ !std::is_void<decltype(std::declval<T>().operator->())>::value,
+ decltype(std::declval<T>().operator->())>;
+
+public:
+ /// Construct a fallible iterator that *cannot* be used as an end-of-range
+ /// value.
+ ///
+ /// A value created by this method can be dereferenced, incremented,
+ /// decremented and compared, providing the underlying type supports it.
+ ///
+ /// The error that is passed in will be initially marked as checked, so if the
+ /// iterator is not used at all the Error need not be checked.
+ static fallible_iterator itr(Underlying I, Error &Err) {
+ (void)!!Err;
+ return fallible_iterator(std::move(I), &Err);
+ }
+
+ /// Construct a fallible iteratro that can be used as an end-of-range value.
+ ///
+ /// A value created by this method can be dereferenced (if the underlying
+ /// value points at a valid value) and compared, but not incremented or
+ /// decremented.
+ static fallible_iterator end(Underlying I) {
+ return fallible_iterator(std::move(I), nullptr);
+ }
+
+ /// Forward dereference to the underlying iterator.
+ auto operator*() -> decltype(*std::declval<Underlying>()) { return *I; }
+
+ /// Forward const dereference to the underlying iterator.
+ auto operator*() const -> decltype(*std::declval<const Underlying>()) {
+ return *I;
+ }
+
+ /// Forward structure dereference to the underlying iterator (if the
+ /// underlying iterator supports it).
+ template <typename T = Underlying>
+ typename enable_if_struct_deref_supported<T>::type operator->() {
+ return I.operator->();
+ }
+
+ /// Forward const structure dereference to the underlying iterator (if the
+ /// underlying iterator supports it).
+ template <typename T = Underlying>
+ typename enable_if_struct_deref_supported<const T>::type operator->() const {
+ return I.operator->();
+ }
+
+ /// Increment the fallible iterator.
+ ///
+ /// If the underlying 'inc' operation fails, this will set the Error value
+ /// and update this iterator value to point to end-of-range.
+ ///
+ /// The Error value is marked as needing checking, regardless of whether the
+ /// 'inc' operation succeeds or fails.
+ fallible_iterator &operator++() {
+ assert(getErrPtr() && "Cannot increment end iterator");
+ if (auto Err = I.inc())
+ handleError(std::move(Err));
+ else
+ resetCheckedFlag();
+ return *this;
+ }
+
+ /// Decrement the fallible iterator.
+ ///
+ /// If the underlying 'dec' operation fails, this will set the Error value
+ /// and update this iterator value to point to end-of-range.
+ ///
+ /// The Error value is marked as needing checking, regardless of whether the
+ /// 'dec' operation succeeds or fails.
+ fallible_iterator &operator--() {
+ assert(getErrPtr() && "Cannot decrement end iterator");
+ if (auto Err = I.dec())
+ handleError(std::move(Err));
+ else
+ resetCheckedFlag();
+ return *this;
+ }
+
+ /// Compare fallible iterators for equality.
+ ///
+ /// Returns true if both LHS and RHS are end-of-range values, or if both are
+ /// non-end-of-range values whose underlying iterator values compare equal.
+ ///
+ /// If this is a comparison between an end-of-range iterator and a
+ /// non-end-of-range iterator, then the Error (referenced by the
+ /// non-end-of-range value) is marked as checked: Since all
+ /// increment/decrement operations result in an end-of-range value, comparing
+ /// false against end-of-range is equivalent to checking that the Error value
+ /// is success. This flag management enables early returns from loop bodies
+ /// without redundant Error checks.
+ friend bool operator==(const fallible_iterator &LHS,
+ const fallible_iterator &RHS) {
+ // If both iterators are in the end state they compare
+ // equal, regardless of whether either is valid.
+ if (LHS.isEnd() && RHS.isEnd())
+ return true;
+
+ assert(LHS.isValid() && RHS.isValid() &&
+ "Invalid iterators can only be compared against end");
+
+ bool Equal = LHS.I == RHS.I;
+
+ // If the iterators differ and this is a comparison against end then mark
+ // the Error as checked.
+ if (!Equal) {
+ if (LHS.isEnd())
+ (void)!!*RHS.getErrPtr();
+ else
+ (void)!!*LHS.getErrPtr();
+ }
+
+ return Equal;
+ }
+
+ /// Compare fallible iterators for inequality.
+ ///
+ /// See notes for operator==.
+ friend bool operator!=(const fallible_iterator &LHS,
+ const fallible_iterator &RHS) {
+ return !(LHS == RHS);
+ }
+
+private:
+ fallible_iterator(Underlying I, Error *Err)
+ : I(std::move(I)), ErrState(Err, false) {}
+
+ Error *getErrPtr() const { return ErrState.getPointer(); }
+
+ bool isEnd() const { return getErrPtr() == nullptr; }
+
+ bool isValid() const { return !ErrState.getInt(); }
+
+ void handleError(Error Err) {
+ *getErrPtr() = std::move(Err);
+ ErrState.setPointer(nullptr);
+ ErrState.setInt(true);
+ }
+
+ void resetCheckedFlag() {
+ *getErrPtr() = Error::success();
+ }
+
+ Underlying I;
+ mutable PointerIntPair<Error *, 1> ErrState;
+};
+
+/// Convenience wrapper to make a fallible_iterator value from an instance
+/// of an underlying iterator and an Error reference.
+template <typename Underlying>
+fallible_iterator<Underlying> make_fallible_itr(Underlying I, Error &Err) {
+ return fallible_iterator<Underlying>::itr(std::move(I), Err);
+}
+
+/// Convenience wrapper to make a fallible_iterator end value from an instance
+/// of an underlying iterator.
+template <typename Underlying>
+fallible_iterator<Underlying> make_fallible_end(Underlying E) {
+ return fallible_iterator<Underlying>::end(std::move(E));
+}
+
+template <typename Underlying>
+iterator_range<fallible_iterator<Underlying>>
+make_fallible_range(Underlying I, Underlying E, Error &Err) {
+ return make_range(make_fallible_itr(std::move(I), Err),
+ make_fallible_end(std::move(E)));
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_FALLIBLE_ITERATOR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ilist_base.h b/third_party/llvm-project/include/llvm/ADT/ilist_base.h
new file mode 100644
index 000000000..b8c098b95
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ilist_base.h
@@ -0,0 +1,92 @@
+//===- llvm/ADT/ilist_base.h - Intrusive List Base --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ILIST_BASE_H
+#define LLVM_ADT_ILIST_BASE_H
+
+#include "llvm/ADT/ilist_node_base.h"
+#include <cassert>
+
+namespace llvm {
+
+/// Implementations of list algorithms using ilist_node_base.
+template <bool EnableSentinelTracking> class ilist_base {
+public:
+ using node_base_type = ilist_node_base<EnableSentinelTracking>;
+
+ static void insertBeforeImpl(node_base_type &Next, node_base_type &N) {
+ node_base_type &Prev = *Next.getPrev();
+ N.setNext(&Next);
+ N.setPrev(&Prev);
+ Prev.setNext(&N);
+ Next.setPrev(&N);
+ }
+
+ static void removeImpl(node_base_type &N) {
+ node_base_type *Prev = N.getPrev();
+ node_base_type *Next = N.getNext();
+ Next->setPrev(Prev);
+ Prev->setNext(Next);
+
+ // Not strictly necessary, but helps catch a class of bugs.
+ N.setPrev(nullptr);
+ N.setNext(nullptr);
+ }
+
+ static void removeRangeImpl(node_base_type &First, node_base_type &Last) {
+ node_base_type *Prev = First.getPrev();
+ node_base_type *Final = Last.getPrev();
+ Last.setPrev(Prev);
+ Prev->setNext(&Last);
+
+ // Not strictly necessary, but helps catch a class of bugs.
+ First.setPrev(nullptr);
+ Final->setNext(nullptr);
+ }
+
+ static void transferBeforeImpl(node_base_type &Next, node_base_type &First,
+ node_base_type &Last) {
+ if (&Next == &Last || &First == &Last)
+ return;
+
+ // Position cannot be contained in the range to be transferred.
+ assert(&Next != &First &&
+ // Check for the most common mistake.
+ "Insertion point can't be one of the transferred nodes");
+
+ node_base_type &Final = *Last.getPrev();
+
+ // Detach from old list/position.
+ First.getPrev()->setNext(&Last);
+ Last.setPrev(First.getPrev());
+
+ // Splice [First, Final] into its new list/position.
+ node_base_type &Prev = *Next.getPrev();
+ Final.setNext(&Next);
+ First.setPrev(&Prev);
+ Prev.setNext(&First);
+ Next.setPrev(&Final);
+ }
+
+ template <class T> static void insertBefore(T &Next, T &N) {
+ insertBeforeImpl(Next, N);
+ }
+
+ template <class T> static void remove(T &N) { removeImpl(N); }
+ template <class T> static void removeRange(T &First, T &Last) {
+ removeRangeImpl(First, Last);
+ }
+
+ template <class T> static void transferBefore(T &Next, T &First, T &Last) {
+ transferBeforeImpl(Next, First, Last);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ILIST_BASE_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ilist_iterator.h b/third_party/llvm-project/include/llvm/ADT/ilist_iterator.h
new file mode 100644
index 000000000..cbe5cefa9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ilist_iterator.h
@@ -0,0 +1,198 @@
+//===- llvm/ADT/ilist_iterator.h - Intrusive List Iterator ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ILIST_ITERATOR_H
+#define LLVM_ADT_ILIST_ITERATOR_H
+
+#include "llvm/ADT/ilist_node.h"
+#include <cassert>
+#include <cstddef>
+#include <iterator>
+#include <type_traits>
+
+namespace llvm {
+
+namespace ilist_detail {
+
+/// Find const-correct node types.
+template <class OptionsT, bool IsConst> struct IteratorTraits;
+template <class OptionsT> struct IteratorTraits<OptionsT, false> {
+ using value_type = typename OptionsT::value_type;
+ using pointer = typename OptionsT::pointer;
+ using reference = typename OptionsT::reference;
+ using node_pointer = ilist_node_impl<OptionsT> *;
+ using node_reference = ilist_node_impl<OptionsT> &;
+};
+template <class OptionsT> struct IteratorTraits<OptionsT, true> {
+ using value_type = const typename OptionsT::value_type;
+ using pointer = typename OptionsT::const_pointer;
+ using reference = typename OptionsT::const_reference;
+ using node_pointer = const ilist_node_impl<OptionsT> *;
+ using node_reference = const ilist_node_impl<OptionsT> &;
+};
+
+template <bool IsReverse> struct IteratorHelper;
+template <> struct IteratorHelper<false> : ilist_detail::NodeAccess {
+ using Access = ilist_detail::NodeAccess;
+
+ template <class T> static void increment(T *&I) { I = Access::getNext(*I); }
+ template <class T> static void decrement(T *&I) { I = Access::getPrev(*I); }
+};
+template <> struct IteratorHelper<true> : ilist_detail::NodeAccess {
+ using Access = ilist_detail::NodeAccess;
+
+ template <class T> static void increment(T *&I) { I = Access::getPrev(*I); }
+ template <class T> static void decrement(T *&I) { I = Access::getNext(*I); }
+};
+
+} // end namespace ilist_detail
+
+/// Iterator for intrusive lists based on ilist_node.
+template <class OptionsT, bool IsReverse, bool IsConst>
+class ilist_iterator : ilist_detail::SpecificNodeAccess<OptionsT> {
+ friend ilist_iterator<OptionsT, IsReverse, !IsConst>;
+ friend ilist_iterator<OptionsT, !IsReverse, IsConst>;
+ friend ilist_iterator<OptionsT, !IsReverse, !IsConst>;
+
+ using Traits = ilist_detail::IteratorTraits<OptionsT, IsConst>;
+ using Access = ilist_detail::SpecificNodeAccess<OptionsT>;
+
+public:
+ using value_type = typename Traits::value_type;
+ using pointer = typename Traits::pointer;
+ using reference = typename Traits::reference;
+ using difference_type = ptrdiff_t;
+ using iterator_category = std::bidirectional_iterator_tag;
+ using const_pointer = typename OptionsT::const_pointer;
+ using const_reference = typename OptionsT::const_reference;
+
+private:
+ using node_pointer = typename Traits::node_pointer;
+ using node_reference = typename Traits::node_reference;
+
+ node_pointer NodePtr = nullptr;
+
+public:
+ /// Create from an ilist_node.
+ explicit ilist_iterator(node_reference N) : NodePtr(&N) {}
+
+ explicit ilist_iterator(pointer NP) : NodePtr(Access::getNodePtr(NP)) {}
+ explicit ilist_iterator(reference NR) : NodePtr(Access::getNodePtr(&NR)) {}
+ ilist_iterator() = default;
+
+ // This is templated so that we can allow constructing a const iterator from
+ // a nonconst iterator...
+ template <bool RHSIsConst>
+ ilist_iterator(
+ const ilist_iterator<OptionsT, IsReverse, RHSIsConst> &RHS,
+ typename std::enable_if<IsConst || !RHSIsConst, void *>::type = nullptr)
+ : NodePtr(RHS.NodePtr) {}
+
+ // This is templated so that we can allow assigning to a const iterator from
+ // a nonconst iterator...
+ template <bool RHSIsConst>
+ typename std::enable_if<IsConst || !RHSIsConst, ilist_iterator &>::type
+ operator=(const ilist_iterator<OptionsT, IsReverse, RHSIsConst> &RHS) {
+ NodePtr = RHS.NodePtr;
+ return *this;
+ }
+
+ /// Explicit conversion between forward/reverse iterators.
+ ///
+ /// Translate between forward and reverse iterators without changing range
+ /// boundaries. The resulting iterator will dereference (and have a handle)
+ /// to the previous node, which is somewhat unexpected; but converting the
+ /// two endpoints in a range will give the same range in reverse.
+ ///
+ /// This matches std::reverse_iterator conversions.
+ explicit ilist_iterator(
+ const ilist_iterator<OptionsT, !IsReverse, IsConst> &RHS)
+ : ilist_iterator(++RHS.getReverse()) {}
+
+ /// Get a reverse iterator to the same node.
+ ///
+ /// Gives a reverse iterator that will dereference (and have a handle) to the
+ /// same node. Converting the endpoint iterators in a range will give a
+ /// different range; for range operations, use the explicit conversions.
+ ilist_iterator<OptionsT, !IsReverse, IsConst> getReverse() const {
+ if (NodePtr)
+ return ilist_iterator<OptionsT, !IsReverse, IsConst>(*NodePtr);
+ return ilist_iterator<OptionsT, !IsReverse, IsConst>();
+ }
+
+ /// Const-cast.
+ ilist_iterator<OptionsT, IsReverse, false> getNonConst() const {
+ if (NodePtr)
+ return ilist_iterator<OptionsT, IsReverse, false>(
+ const_cast<typename ilist_iterator<OptionsT, IsReverse,
+ false>::node_reference>(*NodePtr));
+ return ilist_iterator<OptionsT, IsReverse, false>();
+ }
+
+ // Accessors...
+ reference operator*() const {
+ assert(!NodePtr->isKnownSentinel());
+ return *Access::getValuePtr(NodePtr);
+ }
+ pointer operator->() const { return &operator*(); }
+
+ // Comparison operators
+ friend bool operator==(const ilist_iterator &LHS, const ilist_iterator &RHS) {
+ return LHS.NodePtr == RHS.NodePtr;
+ }
+ friend bool operator!=(const ilist_iterator &LHS, const ilist_iterator &RHS) {
+ return LHS.NodePtr != RHS.NodePtr;
+ }
+
+ // Increment and decrement operators...
+ ilist_iterator &operator--() {
+ NodePtr = IsReverse ? NodePtr->getNext() : NodePtr->getPrev();
+ return *this;
+ }
+ ilist_iterator &operator++() {
+ NodePtr = IsReverse ? NodePtr->getPrev() : NodePtr->getNext();
+ return *this;
+ }
+ ilist_iterator operator--(int) {
+ ilist_iterator tmp = *this;
+ --*this;
+ return tmp;
+ }
+ ilist_iterator operator++(int) {
+ ilist_iterator tmp = *this;
+ ++*this;
+ return tmp;
+ }
+
+ /// Get the underlying ilist_node.
+ node_pointer getNodePtr() const { return static_cast<node_pointer>(NodePtr); }
+
+ /// Check for end. Only valid if ilist_sentinel_tracking<true>.
+ bool isEnd() const { return NodePtr ? NodePtr->isSentinel() : false; }
+};
+
+template <typename From> struct simplify_type;
+
+/// Allow ilist_iterators to convert into pointers to a node automatically when
+/// used by the dyn_cast, cast, isa mechanisms...
+///
+/// FIXME: remove this, since there is no implicit conversion to NodeTy.
+template <class OptionsT, bool IsConst>
+struct simplify_type<ilist_iterator<OptionsT, false, IsConst>> {
+ using iterator = ilist_iterator<OptionsT, false, IsConst>;
+ using SimpleType = typename iterator::pointer;
+
+ static SimpleType getSimplifiedValue(const iterator &Node) { return &*Node; }
+};
+template <class OptionsT, bool IsConst>
+struct simplify_type<const ilist_iterator<OptionsT, false, IsConst>>
+ : simplify_type<ilist_iterator<OptionsT, false, IsConst>> {};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ILIST_ITERATOR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ilist_node.h b/third_party/llvm-project/include/llvm/ADT/ilist_node.h
new file mode 100644
index 000000000..e040d9630
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ilist_node.h
@@ -0,0 +1,305 @@
+//===- llvm/ADT/ilist_node.h - Intrusive Linked List Helper -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the ilist_node class template, which is a convenient
+// base class for creating classes that can be used with ilists.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ILIST_NODE_H
+#define LLVM_ADT_ILIST_NODE_H
+
+#include "llvm/ADT/ilist_node_base.h"
+#include "llvm/ADT/ilist_node_options.h"
+
+namespace llvm {
+
+namespace ilist_detail {
+
+struct NodeAccess;
+
+} // end namespace ilist_detail
+
+template <class OptionsT, bool IsReverse, bool IsConst> class ilist_iterator;
+template <class OptionsT> class ilist_sentinel;
+
+/// Implementation for an ilist node.
+///
+/// Templated on an appropriate \a ilist_detail::node_options, usually computed
+/// by \a ilist_detail::compute_node_options.
+///
+/// This is a wrapper around \a ilist_node_base whose main purpose is to
+/// provide type safety: you can't insert nodes of \a ilist_node_impl into the
+/// wrong \a simple_ilist or \a iplist.
+template <class OptionsT> class ilist_node_impl : OptionsT::node_base_type {
+ using value_type = typename OptionsT::value_type;
+ using node_base_type = typename OptionsT::node_base_type;
+ using list_base_type = typename OptionsT::list_base_type;
+
+ friend typename OptionsT::list_base_type;
+ friend struct ilist_detail::NodeAccess;
+ friend class ilist_sentinel<OptionsT>;
+ friend class ilist_iterator<OptionsT, false, false>;
+ friend class ilist_iterator<OptionsT, false, true>;
+ friend class ilist_iterator<OptionsT, true, false>;
+ friend class ilist_iterator<OptionsT, true, true>;
+
+protected:
+ using self_iterator = ilist_iterator<OptionsT, false, false>;
+ using const_self_iterator = ilist_iterator<OptionsT, false, true>;
+ using reverse_self_iterator = ilist_iterator<OptionsT, true, false>;
+ using const_reverse_self_iterator = ilist_iterator<OptionsT, true, true>;
+
+ ilist_node_impl() = default;
+
+private:
+ ilist_node_impl *getPrev() {
+ return static_cast<ilist_node_impl *>(node_base_type::getPrev());
+ }
+
+ ilist_node_impl *getNext() {
+ return static_cast<ilist_node_impl *>(node_base_type::getNext());
+ }
+
+ const ilist_node_impl *getPrev() const {
+ return static_cast<ilist_node_impl *>(node_base_type::getPrev());
+ }
+
+ const ilist_node_impl *getNext() const {
+ return static_cast<ilist_node_impl *>(node_base_type::getNext());
+ }
+
+ void setPrev(ilist_node_impl *N) { node_base_type::setPrev(N); }
+ void setNext(ilist_node_impl *N) { node_base_type::setNext(N); }
+
+public:
+ self_iterator getIterator() { return self_iterator(*this); }
+ const_self_iterator getIterator() const { return const_self_iterator(*this); }
+
+ reverse_self_iterator getReverseIterator() {
+ return reverse_self_iterator(*this);
+ }
+
+ const_reverse_self_iterator getReverseIterator() const {
+ return const_reverse_self_iterator(*this);
+ }
+
+ // Under-approximation, but always available for assertions.
+ using node_base_type::isKnownSentinel;
+
+ /// Check whether this is the sentinel node.
+ ///
+ /// This requires sentinel tracking to be explicitly enabled. Use the
+ /// ilist_sentinel_tracking<true> option to get this API.
+ bool isSentinel() const {
+ static_assert(OptionsT::is_sentinel_tracking_explicit,
+ "Use ilist_sentinel_tracking<true> to enable isSentinel()");
+ return node_base_type::isSentinel();
+ }
+};
+
+/// An intrusive list node.
+///
+/// A base class to enable membership in intrusive lists, including \a
+/// simple_ilist, \a iplist, and \a ilist. The first template parameter is the
+/// \a value_type for the list.
+///
+/// An ilist node can be configured with compile-time options to change
+/// behaviour and/or add API.
+///
+/// By default, an \a ilist_node knows whether it is the list sentinel (an
+/// instance of \a ilist_sentinel) if and only if
+/// LLVM_ENABLE_ABI_BREAKING_CHECKS. The function \a isKnownSentinel() always
+/// returns \c false tracking is off. Sentinel tracking steals a bit from the
+/// "prev" link, which adds a mask operation when decrementing an iterator, but
+/// enables bug-finding assertions in \a ilist_iterator.
+///
+/// To turn sentinel tracking on all the time, pass in the
+/// ilist_sentinel_tracking<true> template parameter. This also enables the \a
+/// isSentinel() function. The same option must be passed to the intrusive
+/// list. (ilist_sentinel_tracking<false> turns sentinel tracking off all the
+/// time.)
+///
+/// A type can inherit from ilist_node multiple times by passing in different
+/// \a ilist_tag options. This allows a single instance to be inserted into
+/// multiple lists simultaneously, where each list is given the same tag.
+///
+/// \example
+/// struct A {};
+/// struct B {};
+/// struct N : ilist_node<N, ilist_tag<A>>, ilist_node<N, ilist_tag<B>> {};
+///
+/// void foo() {
+/// simple_ilist<N, ilist_tag<A>> ListA;
+/// simple_ilist<N, ilist_tag<B>> ListB;
+/// N N1;
+/// ListA.push_back(N1);
+/// ListB.push_back(N1);
+/// }
+/// \endexample
+///
+/// See \a is_valid_option for steps on adding a new option.
+template <class T, class... Options>
+class ilist_node
+ : public ilist_node_impl<
+ typename ilist_detail::compute_node_options<T, Options...>::type> {
+ static_assert(ilist_detail::check_options<Options...>::value,
+ "Unrecognized node option!");
+};
+
+namespace ilist_detail {
+
+/// An access class for ilist_node private API.
+///
+/// This gives access to the private parts of ilist nodes. Nodes for an ilist
+/// should friend this class if they inherit privately from ilist_node.
+///
+/// Using this class outside of the ilist implementation is unsupported.
+struct NodeAccess {
+protected:
+ template <class OptionsT>
+ static ilist_node_impl<OptionsT> *getNodePtr(typename OptionsT::pointer N) {
+ return N;
+ }
+
+ template <class OptionsT>
+ static const ilist_node_impl<OptionsT> *
+ getNodePtr(typename OptionsT::const_pointer N) {
+ return N;
+ }
+
+ template <class OptionsT>
+ static typename OptionsT::pointer getValuePtr(ilist_node_impl<OptionsT> *N) {
+ return static_cast<typename OptionsT::pointer>(N);
+ }
+
+ template <class OptionsT>
+ static typename OptionsT::const_pointer
+ getValuePtr(const ilist_node_impl<OptionsT> *N) {
+ return static_cast<typename OptionsT::const_pointer>(N);
+ }
+
+ template <class OptionsT>
+ static ilist_node_impl<OptionsT> *getPrev(ilist_node_impl<OptionsT> &N) {
+ return N.getPrev();
+ }
+
+ template <class OptionsT>
+ static ilist_node_impl<OptionsT> *getNext(ilist_node_impl<OptionsT> &N) {
+ return N.getNext();
+ }
+
+ template <class OptionsT>
+ static const ilist_node_impl<OptionsT> *
+ getPrev(const ilist_node_impl<OptionsT> &N) {
+ return N.getPrev();
+ }
+
+ template <class OptionsT>
+ static const ilist_node_impl<OptionsT> *
+ getNext(const ilist_node_impl<OptionsT> &N) {
+ return N.getNext();
+ }
+};
+
+template <class OptionsT> struct SpecificNodeAccess : NodeAccess {
+protected:
+ using pointer = typename OptionsT::pointer;
+ using const_pointer = typename OptionsT::const_pointer;
+ using node_type = ilist_node_impl<OptionsT>;
+
+ static node_type *getNodePtr(pointer N) {
+ return NodeAccess::getNodePtr<OptionsT>(N);
+ }
+
+ static const node_type *getNodePtr(const_pointer N) {
+ return NodeAccess::getNodePtr<OptionsT>(N);
+ }
+
+ static pointer getValuePtr(node_type *N) {
+ return NodeAccess::getValuePtr<OptionsT>(N);
+ }
+
+ static const_pointer getValuePtr(const node_type *N) {
+ return NodeAccess::getValuePtr<OptionsT>(N);
+ }
+};
+
+} // end namespace ilist_detail
+
+template <class OptionsT>
+class ilist_sentinel : public ilist_node_impl<OptionsT> {
+public:
+ ilist_sentinel() {
+ this->initializeSentinel();
+ reset();
+ }
+
+ void reset() {
+ this->setPrev(this);
+ this->setNext(this);
+ }
+
+ bool empty() const { return this == this->getPrev(); }
+};
+
+/// An ilist node that can access its parent list.
+///
+/// Requires \c NodeTy to have \a getParent() to find the parent node, and the
+/// \c ParentTy to have \a getSublistAccess() to get a reference to the list.
+template <typename NodeTy, typename ParentTy, class... Options>
+class ilist_node_with_parent : public ilist_node<NodeTy, Options...> {
+protected:
+ ilist_node_with_parent() = default;
+
+private:
+ /// Forward to NodeTy::getParent().
+ ///
+ /// Note: do not use the name "getParent()". We want a compile error
+ /// (instead of recursion) when the subclass fails to implement \a
+ /// getParent().
+ const ParentTy *getNodeParent() const {
+ return static_cast<const NodeTy *>(this)->getParent();
+ }
+
+public:
+ /// @name Adjacent Node Accessors
+ /// @{
+ /// Get the previous node, or \c nullptr for the list head.
+ NodeTy *getPrevNode() {
+ // Should be separated to a reused function, but then we couldn't use auto
+ // (and would need the type of the list).
+ const auto &List =
+ getNodeParent()->*(ParentTy::getSublistAccess((NodeTy *)nullptr));
+ return List.getPrevNode(*static_cast<NodeTy *>(this));
+ }
+
+ /// Get the previous node, or \c nullptr for the list head.
+ const NodeTy *getPrevNode() const {
+ return const_cast<ilist_node_with_parent *>(this)->getPrevNode();
+ }
+
+ /// Get the next node, or \c nullptr for the list tail.
+ NodeTy *getNextNode() {
+ // Should be separated to a reused function, but then we couldn't use auto
+ // (and would need the type of the list).
+ const auto &List =
+ getNodeParent()->*(ParentTy::getSublistAccess((NodeTy *)nullptr));
+ return List.getNextNode(*static_cast<NodeTy *>(this));
+ }
+
+ /// Get the next node, or \c nullptr for the list tail.
+ const NodeTy *getNextNode() const {
+ return const_cast<ilist_node_with_parent *>(this)->getNextNode();
+ }
+ /// @}
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ILIST_NODE_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ilist_node_base.h b/third_party/llvm-project/include/llvm/ADT/ilist_node_base.h
new file mode 100644
index 000000000..f6c518e6e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ilist_node_base.h
@@ -0,0 +1,52 @@
+//===- llvm/ADT/ilist_node_base.h - Intrusive List Node Base -----*- C++ -*-==//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ILIST_NODE_BASE_H
+#define LLVM_ADT_ILIST_NODE_BASE_H
+
+#include "llvm/ADT/PointerIntPair.h"
+
+namespace llvm {
+
+/// Base class for ilist nodes.
+///
+/// Optionally tracks whether this node is the sentinel.
+template <bool EnableSentinelTracking> class ilist_node_base;
+
+template <> class ilist_node_base<false> {
+ ilist_node_base *Prev = nullptr;
+ ilist_node_base *Next = nullptr;
+
+public:
+ void setPrev(ilist_node_base *Prev) { this->Prev = Prev; }
+ void setNext(ilist_node_base *Next) { this->Next = Next; }
+ ilist_node_base *getPrev() const { return Prev; }
+ ilist_node_base *getNext() const { return Next; }
+
+ bool isKnownSentinel() const { return false; }
+ void initializeSentinel() {}
+};
+
+template <> class ilist_node_base<true> {
+ PointerIntPair<ilist_node_base *, 1> PrevAndSentinel;
+ ilist_node_base *Next = nullptr;
+
+public:
+ void setPrev(ilist_node_base *Prev) { PrevAndSentinel.setPointer(Prev); }
+ void setNext(ilist_node_base *Next) { this->Next = Next; }
+ ilist_node_base *getPrev() const { return PrevAndSentinel.getPointer(); }
+ ilist_node_base *getNext() const { return Next; }
+
+ bool isSentinel() const { return PrevAndSentinel.getInt(); }
+ bool isKnownSentinel() const { return isSentinel(); }
+ void initializeSentinel() { PrevAndSentinel.setInt(true); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ILIST_NODE_BASE_H
diff --git a/third_party/llvm-project/include/llvm/ADT/ilist_node_options.h b/third_party/llvm-project/include/llvm/ADT/ilist_node_options.h
new file mode 100644
index 000000000..9b95cdbe0
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/ilist_node_options.h
@@ -0,0 +1,131 @@
+//===- llvm/ADT/ilist_node_options.h - ilist_node Options -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ILIST_NODE_OPTIONS_H
+#define LLVM_ADT_ILIST_NODE_OPTIONS_H
+
+#include "llvm/Config/abi-breaking.h"
+
+#include <type_traits>
+
+namespace llvm {
+
+template <bool EnableSentinelTracking> class ilist_node_base;
+template <bool EnableSentinelTracking> class ilist_base;
+
+/// Option to choose whether to track sentinels.
+///
+/// This option affects the ABI for the nodes. When not specified explicitly,
+/// the ABI depends on LLVM_ENABLE_ABI_BREAKING_CHECKS. Specify explicitly to
+/// enable \a ilist_node::isSentinel().
+template <bool EnableSentinelTracking> struct ilist_sentinel_tracking {};
+
+/// Option to specify a tag for the node type.
+///
+/// This option allows a single value type to be inserted in multiple lists
+/// simultaneously. See \a ilist_node for usage examples.
+template <class Tag> struct ilist_tag {};
+
+namespace ilist_detail {
+
+/// Helper trait for recording whether an option is specified explicitly.
+template <bool IsExplicit> struct explicitness {
+ static const bool is_explicit = IsExplicit;
+};
+typedef explicitness<true> is_explicit;
+typedef explicitness<false> is_implicit;
+
+/// Check whether an option is valid.
+///
+/// The steps for adding and enabling a new ilist option include:
+/// \li define the option, ilist_foo<Bar>, above;
+/// \li add new parameters for Bar to \a ilist_detail::node_options;
+/// \li add an extraction meta-function, ilist_detail::extract_foo;
+/// \li call extract_foo from \a ilist_detail::compute_node_options and pass it
+/// into \a ilist_detail::node_options; and
+/// \li specialize \c is_valid_option<ilist_foo<Bar>> to inherit from \c
+/// std::true_type to get static assertions passing in \a simple_ilist and \a
+/// ilist_node.
+template <class Option> struct is_valid_option : std::false_type {};
+
+/// Extract sentinel tracking option.
+///
+/// Look through \p Options for the \a ilist_sentinel_tracking option, with the
+/// default depending on LLVM_ENABLE_ABI_BREAKING_CHECKS.
+template <class... Options> struct extract_sentinel_tracking;
+template <bool EnableSentinelTracking, class... Options>
+struct extract_sentinel_tracking<
+ ilist_sentinel_tracking<EnableSentinelTracking>, Options...>
+ : std::integral_constant<bool, EnableSentinelTracking>, is_explicit {};
+template <class Option1, class... Options>
+struct extract_sentinel_tracking<Option1, Options...>
+ : extract_sentinel_tracking<Options...> {};
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+template <> struct extract_sentinel_tracking<> : std::true_type, is_implicit {};
+#else
+template <>
+struct extract_sentinel_tracking<> : std::false_type, is_implicit {};
+#endif
+template <bool EnableSentinelTracking>
+struct is_valid_option<ilist_sentinel_tracking<EnableSentinelTracking>>
+ : std::true_type {};
+
+/// Extract custom tag option.
+///
+/// Look through \p Options for the \a ilist_tag option, pulling out the
+/// custom tag type, using void as a default.
+template <class... Options> struct extract_tag;
+template <class Tag, class... Options>
+struct extract_tag<ilist_tag<Tag>, Options...> {
+ typedef Tag type;
+};
+template <class Option1, class... Options>
+struct extract_tag<Option1, Options...> : extract_tag<Options...> {};
+template <> struct extract_tag<> { typedef void type; };
+template <class Tag> struct is_valid_option<ilist_tag<Tag>> : std::true_type {};
+
+/// Check whether options are valid.
+///
+/// The conjunction of \a is_valid_option on each individual option.
+template <class... Options> struct check_options;
+template <> struct check_options<> : std::true_type {};
+template <class Option1, class... Options>
+struct check_options<Option1, Options...>
+ : std::integral_constant<bool, is_valid_option<Option1>::value &&
+ check_options<Options...>::value> {};
+
+/// Traits for options for \a ilist_node.
+///
+/// This is usually computed via \a compute_node_options.
+template <class T, bool EnableSentinelTracking, bool IsSentinelTrackingExplicit,
+ class TagT>
+struct node_options {
+ typedef T value_type;
+ typedef T *pointer;
+ typedef T &reference;
+ typedef const T *const_pointer;
+ typedef const T &const_reference;
+
+ static const bool enable_sentinel_tracking = EnableSentinelTracking;
+ static const bool is_sentinel_tracking_explicit = IsSentinelTrackingExplicit;
+ typedef TagT tag;
+ typedef ilist_node_base<enable_sentinel_tracking> node_base_type;
+ typedef ilist_base<enable_sentinel_tracking> list_base_type;
+};
+
+template <class T, class... Options> struct compute_node_options {
+ typedef node_options<T, extract_sentinel_tracking<Options...>::value,
+ extract_sentinel_tracking<Options...>::is_explicit,
+ typename extract_tag<Options...>::type>
+ type;
+};
+
+} // end namespace ilist_detail
+} // end namespace llvm
+
+#endif // LLVM_ADT_ILIST_NODE_OPTIONS_H
diff --git a/third_party/llvm-project/include/llvm/ADT/iterator.h b/third_party/llvm-project/include/llvm/ADT/iterator.h
new file mode 100644
index 000000000..467fd4c00
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/iterator.h
@@ -0,0 +1,366 @@
+//===- iterator.h - Utilities for using and defining iterators --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ITERATOR_H
+#define LLVM_ADT_ITERATOR_H
+
+#include "llvm/ADT/iterator_range.h"
+#include <algorithm>
+#include <cstddef>
+#include <iterator>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// CRTP base class which implements the entire standard iterator facade
+/// in terms of a minimal subset of the interface.
+///
+/// Use this when it is reasonable to implement most of the iterator
+/// functionality in terms of a core subset. If you need special behavior or
+/// there are performance implications for this, you may want to override the
+/// relevant members instead.
+///
+/// Note, one abstraction that this does *not* provide is implementing
+/// subtraction in terms of addition by negating the difference. Negation isn't
+/// always information preserving, and I can see very reasonable iterator
+/// designs where this doesn't work well. It doesn't really force much added
+/// boilerplate anyways.
+///
+/// Another abstraction that this doesn't provide is implementing increment in
+/// terms of addition of one. These aren't equivalent for all iterator
+/// categories, and respecting that adds a lot of complexity for little gain.
+///
+/// Classes wishing to use `iterator_facade_base` should implement the following
+/// methods:
+///
+/// Forward Iterators:
+/// (All of the following methods)
+/// - DerivedT &operator=(const DerivedT &R);
+/// - bool operator==(const DerivedT &R) const;
+/// - const T &operator*() const;
+/// - T &operator*();
+/// - DerivedT &operator++();
+///
+/// Bidirectional Iterators:
+/// (All methods of forward iterators, plus the following)
+/// - DerivedT &operator--();
+///
+/// Random-access Iterators:
+/// (All methods of bidirectional iterators excluding the following)
+/// - DerivedT &operator++();
+/// - DerivedT &operator--();
+/// (and plus the following)
+/// - bool operator<(const DerivedT &RHS) const;
+/// - DifferenceTypeT operator-(const DerivedT &R) const;
+/// - DerivedT &operator+=(DifferenceTypeT N);
+/// - DerivedT &operator-=(DifferenceTypeT N);
+///
+template <typename DerivedT, typename IteratorCategoryT, typename T,
+ typename DifferenceTypeT = std::ptrdiff_t, typename PointerT = T *,
+ typename ReferenceT = T &>
+class iterator_facade_base
+ : public std::iterator<IteratorCategoryT, T, DifferenceTypeT, PointerT,
+ ReferenceT> {
+protected:
+ enum {
+ IsRandomAccess = std::is_base_of<std::random_access_iterator_tag,
+ IteratorCategoryT>::value,
+ IsBidirectional = std::is_base_of<std::bidirectional_iterator_tag,
+ IteratorCategoryT>::value,
+ };
+
+ /// A proxy object for computing a reference via indirecting a copy of an
+ /// iterator. This is used in APIs which need to produce a reference via
+ /// indirection but for which the iterator object might be a temporary. The
+ /// proxy preserves the iterator internally and exposes the indirected
+ /// reference via a conversion operator.
+ class ReferenceProxy {
+ friend iterator_facade_base;
+
+ DerivedT I;
+
+ ReferenceProxy(DerivedT I) : I(std::move(I)) {}
+
+ public:
+ operator ReferenceT() const { return *I; }
+ };
+
+public:
+ DerivedT operator+(DifferenceTypeT n) const {
+ static_assert(std::is_base_of<iterator_facade_base, DerivedT>::value,
+ "Must pass the derived type to this template!");
+ static_assert(
+ IsRandomAccess,
+ "The '+' operator is only defined for random access iterators.");
+ DerivedT tmp = *static_cast<const DerivedT *>(this);
+ tmp += n;
+ return tmp;
+ }
+ friend DerivedT operator+(DifferenceTypeT n, const DerivedT &i) {
+ static_assert(
+ IsRandomAccess,
+ "The '+' operator is only defined for random access iterators.");
+ return i + n;
+ }
+ DerivedT operator-(DifferenceTypeT n) const {
+ static_assert(
+ IsRandomAccess,
+ "The '-' operator is only defined for random access iterators.");
+ DerivedT tmp = *static_cast<const DerivedT *>(this);
+ tmp -= n;
+ return tmp;
+ }
+
+ DerivedT &operator++() {
+ static_assert(std::is_base_of<iterator_facade_base, DerivedT>::value,
+ "Must pass the derived type to this template!");
+ return static_cast<DerivedT *>(this)->operator+=(1);
+ }
+ DerivedT operator++(int) {
+ DerivedT tmp = *static_cast<DerivedT *>(this);
+ ++*static_cast<DerivedT *>(this);
+ return tmp;
+ }
+ DerivedT &operator--() {
+ static_assert(
+ IsBidirectional,
+ "The decrement operator is only defined for bidirectional iterators.");
+ return static_cast<DerivedT *>(this)->operator-=(1);
+ }
+ DerivedT operator--(int) {
+ static_assert(
+ IsBidirectional,
+ "The decrement operator is only defined for bidirectional iterators.");
+ DerivedT tmp = *static_cast<DerivedT *>(this);
+ --*static_cast<DerivedT *>(this);
+ return tmp;
+ }
+
+ bool operator!=(const DerivedT &RHS) const {
+ return !static_cast<const DerivedT *>(this)->operator==(RHS);
+ }
+
+ bool operator>(const DerivedT &RHS) const {
+ static_assert(
+ IsRandomAccess,
+ "Relational operators are only defined for random access iterators.");
+ return !static_cast<const DerivedT *>(this)->operator<(RHS) &&
+ !static_cast<const DerivedT *>(this)->operator==(RHS);
+ }
+ bool operator<=(const DerivedT &RHS) const {
+ static_assert(
+ IsRandomAccess,
+ "Relational operators are only defined for random access iterators.");
+ return !static_cast<const DerivedT *>(this)->operator>(RHS);
+ }
+ bool operator>=(const DerivedT &RHS) const {
+ static_assert(
+ IsRandomAccess,
+ "Relational operators are only defined for random access iterators.");
+ return !static_cast<const DerivedT *>(this)->operator<(RHS);
+ }
+
+ PointerT operator->() { return &static_cast<DerivedT *>(this)->operator*(); }
+ PointerT operator->() const {
+ return &static_cast<const DerivedT *>(this)->operator*();
+ }
+ ReferenceProxy operator[](DifferenceTypeT n) {
+ static_assert(IsRandomAccess,
+ "Subscripting is only defined for random access iterators.");
+ return ReferenceProxy(static_cast<DerivedT *>(this)->operator+(n));
+ }
+ ReferenceProxy operator[](DifferenceTypeT n) const {
+ static_assert(IsRandomAccess,
+ "Subscripting is only defined for random access iterators.");
+ return ReferenceProxy(static_cast<const DerivedT *>(this)->operator+(n));
+ }
+};
+
+/// CRTP base class for adapting an iterator to a different type.
+///
+/// This class can be used through CRTP to adapt one iterator into another.
+/// Typically this is done through providing in the derived class a custom \c
+/// operator* implementation. Other methods can be overridden as well.
+template <
+ typename DerivedT, typename WrappedIteratorT,
+ typename IteratorCategoryT =
+ typename std::iterator_traits<WrappedIteratorT>::iterator_category,
+ typename T = typename std::iterator_traits<WrappedIteratorT>::value_type,
+ typename DifferenceTypeT =
+ typename std::iterator_traits<WrappedIteratorT>::difference_type,
+ typename PointerT = typename std::conditional<
+ std::is_same<T, typename std::iterator_traits<
+ WrappedIteratorT>::value_type>::value,
+ typename std::iterator_traits<WrappedIteratorT>::pointer, T *>::type,
+ typename ReferenceT = typename std::conditional<
+ std::is_same<T, typename std::iterator_traits<
+ WrappedIteratorT>::value_type>::value,
+ typename std::iterator_traits<WrappedIteratorT>::reference, T &>::type>
+class iterator_adaptor_base
+ : public iterator_facade_base<DerivedT, IteratorCategoryT, T,
+ DifferenceTypeT, PointerT, ReferenceT> {
+ using BaseT = typename iterator_adaptor_base::iterator_facade_base;
+
+protected:
+ WrappedIteratorT I;
+
+ iterator_adaptor_base() = default;
+
+ explicit iterator_adaptor_base(WrappedIteratorT u) : I(std::move(u)) {
+ static_assert(std::is_base_of<iterator_adaptor_base, DerivedT>::value,
+ "Must pass the derived type to this template!");
+ }
+
+ const WrappedIteratorT &wrapped() const { return I; }
+
+public:
+ using difference_type = DifferenceTypeT;
+
+ DerivedT &operator+=(difference_type n) {
+ static_assert(
+ BaseT::IsRandomAccess,
+ "The '+=' operator is only defined for random access iterators.");
+ I += n;
+ return *static_cast<DerivedT *>(this);
+ }
+ DerivedT &operator-=(difference_type n) {
+ static_assert(
+ BaseT::IsRandomAccess,
+ "The '-=' operator is only defined for random access iterators.");
+ I -= n;
+ return *static_cast<DerivedT *>(this);
+ }
+ using BaseT::operator-;
+ difference_type operator-(const DerivedT &RHS) const {
+ static_assert(
+ BaseT::IsRandomAccess,
+ "The '-' operator is only defined for random access iterators.");
+ return I - RHS.I;
+ }
+
+ // We have to explicitly provide ++ and -- rather than letting the facade
+ // forward to += because WrappedIteratorT might not support +=.
+ using BaseT::operator++;
+ DerivedT &operator++() {
+ ++I;
+ return *static_cast<DerivedT *>(this);
+ }
+ using BaseT::operator--;
+ DerivedT &operator--() {
+ static_assert(
+ BaseT::IsBidirectional,
+ "The decrement operator is only defined for bidirectional iterators.");
+ --I;
+ return *static_cast<DerivedT *>(this);
+ }
+
+ bool operator==(const DerivedT &RHS) const { return I == RHS.I; }
+ bool operator<(const DerivedT &RHS) const {
+ static_assert(
+ BaseT::IsRandomAccess,
+ "Relational operators are only defined for random access iterators.");
+ return I < RHS.I;
+ }
+
+ ReferenceT operator*() const { return *I; }
+};
+
+/// An iterator type that allows iterating over the pointees via some
+/// other iterator.
+///
+/// The typical usage of this is to expose a type that iterates over Ts, but
+/// which is implemented with some iterator over T*s:
+///
+/// \code
+/// using iterator = pointee_iterator<SmallVectorImpl<T *>::iterator>;
+/// \endcode
+template <typename WrappedIteratorT,
+ typename T = typename std::remove_reference<
+ decltype(**std::declval<WrappedIteratorT>())>::type>
+struct pointee_iterator
+ : iterator_adaptor_base<
+ pointee_iterator<WrappedIteratorT, T>, WrappedIteratorT,
+ typename std::iterator_traits<WrappedIteratorT>::iterator_category,
+ T> {
+ pointee_iterator() = default;
+ template <typename U>
+ pointee_iterator(U &&u)
+ : pointee_iterator::iterator_adaptor_base(std::forward<U &&>(u)) {}
+
+ T &operator*() const { return **this->I; }
+};
+
+template <typename RangeT, typename WrappedIteratorT =
+ decltype(std::begin(std::declval<RangeT>()))>
+iterator_range<pointee_iterator<WrappedIteratorT>>
+make_pointee_range(RangeT &&Range) {
+ using PointeeIteratorT = pointee_iterator<WrappedIteratorT>;
+ return make_range(PointeeIteratorT(std::begin(std::forward<RangeT>(Range))),
+ PointeeIteratorT(std::end(std::forward<RangeT>(Range))));
+}
+
+template <typename WrappedIteratorT,
+ typename T = decltype(&*std::declval<WrappedIteratorT>())>
+class pointer_iterator
+ : public iterator_adaptor_base<
+ pointer_iterator<WrappedIteratorT, T>, WrappedIteratorT,
+ typename std::iterator_traits<WrappedIteratorT>::iterator_category,
+ T> {
+ mutable T Ptr;
+
+public:
+ pointer_iterator() = default;
+
+ explicit pointer_iterator(WrappedIteratorT u)
+ : pointer_iterator::iterator_adaptor_base(std::move(u)) {}
+
+ T &operator*() { return Ptr = &*this->I; }
+ const T &operator*() const { return Ptr = &*this->I; }
+};
+
+template <typename RangeT, typename WrappedIteratorT =
+ decltype(std::begin(std::declval<RangeT>()))>
+iterator_range<pointer_iterator<WrappedIteratorT>>
+make_pointer_range(RangeT &&Range) {
+ using PointerIteratorT = pointer_iterator<WrappedIteratorT>;
+ return make_range(PointerIteratorT(std::begin(std::forward<RangeT>(Range))),
+ PointerIteratorT(std::end(std::forward<RangeT>(Range))));
+}
+
+// Wrapper iterator over iterator ItType, adding DataRef to the type of ItType,
+// to create NodeRef = std::pair<InnerTypeOfItType, DataRef>.
+template <typename ItType, typename NodeRef, typename DataRef>
+class WrappedPairNodeDataIterator
+ : public iterator_adaptor_base<
+ WrappedPairNodeDataIterator<ItType, NodeRef, DataRef>, ItType,
+ typename std::iterator_traits<ItType>::iterator_category, NodeRef,
+ std::ptrdiff_t, NodeRef *, NodeRef &> {
+ using BaseT = iterator_adaptor_base<
+ WrappedPairNodeDataIterator, ItType,
+ typename std::iterator_traits<ItType>::iterator_category, NodeRef,
+ std::ptrdiff_t, NodeRef *, NodeRef &>;
+
+ const DataRef DR;
+ mutable NodeRef NR;
+
+public:
+ WrappedPairNodeDataIterator(ItType Begin, const DataRef DR)
+ : BaseT(Begin), DR(DR) {
+ NR.first = DR;
+ }
+
+ NodeRef &operator*() const {
+ NR.second = *this->I;
+ return NR;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_ITERATOR_H
diff --git a/third_party/llvm-project/include/llvm/ADT/iterator_range.h b/third_party/llvm-project/include/llvm/ADT/iterator_range.h
new file mode 100644
index 000000000..aa8830943
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/iterator_range.h
@@ -0,0 +1,69 @@
+//===- iterator_range.h - A range adaptor for iterators ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+/// This provides a very simple, boring adaptor for a begin and end iterator
+/// into a range type. This should be used to build range views that work well
+/// with range based for loops and range based constructors.
+///
+/// Note that code here follows more standards-based coding conventions as it
+/// is mirroring proposed interfaces for standardization.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_ITERATOR_RANGE_H
+#define LLVM_ADT_ITERATOR_RANGE_H
+
+#include <iterator>
+#include <utility>
+
+namespace llvm {
+
+/// A range adaptor for a pair of iterators.
+///
+/// This just wraps two iterators into a range-compatible interface. Nothing
+/// fancy at all.
+template <typename IteratorT>
+class iterator_range {
+ IteratorT begin_iterator, end_iterator;
+
+public:
+ //TODO: Add SFINAE to test that the Container's iterators match the range's
+ // iterators.
+ template <typename Container>
+ iterator_range(Container &&c)
+ //TODO: Consider ADL/non-member begin/end calls.
+ : begin_iterator(c.begin()), end_iterator(c.end()) {}
+ iterator_range(IteratorT begin_iterator, IteratorT end_iterator)
+ : begin_iterator(std::move(begin_iterator)),
+ end_iterator(std::move(end_iterator)) {}
+
+ IteratorT begin() const { return begin_iterator; }
+ IteratorT end() const { return end_iterator; }
+ bool empty() const { return begin_iterator == end_iterator; }
+};
+
+/// Convenience function for iterating over sub-ranges.
+///
+/// This provides a bit of syntactic sugar to make using sub-ranges
+/// in for loops a bit easier. Analogous to std::make_pair().
+template <class T> iterator_range<T> make_range(T x, T y) {
+ return iterator_range<T>(std::move(x), std::move(y));
+}
+
+template <typename T> iterator_range<T> make_range(std::pair<T, T> p) {
+ return iterator_range<T>(std::move(p.first), std::move(p.second));
+}
+
+template <typename T>
+iterator_range<decltype(adl_begin(std::declval<T>()))> drop_begin(T &&t,
+ int n) {
+ return make_range(std::next(adl_begin(t), n), adl_end(t));
+}
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/ADT/simple_ilist.h b/third_party/llvm-project/include/llvm/ADT/simple_ilist.h
new file mode 100644
index 000000000..9257b47b9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/simple_ilist.h
@@ -0,0 +1,314 @@
+//===- llvm/ADT/simple_ilist.h - Simple Intrusive List ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_SIMPLE_ILIST_H
+#define LLVM_ADT_SIMPLE_ILIST_H
+
+#include "llvm/ADT/ilist_base.h"
+#include "llvm/ADT/ilist_iterator.h"
+#include "llvm/ADT/ilist_node.h"
+#include "llvm/ADT/ilist_node_options.h"
+#include "llvm/Support/Compiler.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <functional>
+#include <iterator>
+#include <utility>
+
+namespace llvm {
+
+/// A simple intrusive list implementation.
+///
+/// This is a simple intrusive list for a \c T that inherits from \c
+/// ilist_node<T>. The list never takes ownership of anything inserted in it.
+///
+/// Unlike \a iplist<T> and \a ilist<T>, \a simple_ilist<T> never allocates or
+/// deletes values, and has no callback traits.
+///
+/// The API for adding nodes include \a push_front(), \a push_back(), and \a
+/// insert(). These all take values by reference (not by pointer), except for
+/// the range version of \a insert().
+///
+/// There are three sets of API for discarding nodes from the list: \a
+/// remove(), which takes a reference to the node to remove, \a erase(), which
+/// takes an iterator or iterator range and returns the next one, and \a
+/// clear(), which empties out the container. All three are constant time
+/// operations. None of these deletes any nodes; in particular, if there is a
+/// single node in the list, then these have identical semantics:
+/// \li \c L.remove(L.front());
+/// \li \c L.erase(L.begin());
+/// \li \c L.clear();
+///
+/// As a convenience for callers, there are parallel APIs that take a \c
+/// Disposer (such as \c std::default_delete<T>): \a removeAndDispose(), \a
+/// eraseAndDispose(), and \a clearAndDispose(). These have different names
+/// because the extra semantic is otherwise non-obvious. They are equivalent
+/// to calling \a std::for_each() on the range to be discarded.
+///
+/// The currently available \p Options customize the nodes in the list. The
+/// same options must be specified in the \a ilist_node instantation for
+/// compatibility (although the order is irrelevant).
+/// \li Use \a ilist_tag to designate which ilist_node for a given \p T this
+/// list should use. This is useful if a type \p T is part of multiple,
+/// independent lists simultaneously.
+/// \li Use \a ilist_sentinel_tracking to always (or never) track whether a
+/// node is a sentinel. Specifying \c true enables the \a
+/// ilist_node::isSentinel() API. Unlike \a ilist_node::isKnownSentinel(),
+/// which is only appropriate for assertions, \a ilist_node::isSentinel() is
+/// appropriate for real logic.
+///
+/// Here are examples of \p Options usage:
+/// \li \c simple_ilist<T> gives the defaults. \li \c
+/// simple_ilist<T,ilist_sentinel_tracking<true>> enables the \a
+/// ilist_node::isSentinel() API.
+/// \li \c simple_ilist<T,ilist_tag<A>,ilist_sentinel_tracking<false>>
+/// specifies a tag of A and that tracking should be off (even when
+/// LLVM_ENABLE_ABI_BREAKING_CHECKS are enabled).
+/// \li \c simple_ilist<T,ilist_sentinel_tracking<false>,ilist_tag<A>> is
+/// equivalent to the last.
+///
+/// See \a is_valid_option for steps on adding a new option.
+template <typename T, class... Options>
+class simple_ilist
+ : ilist_detail::compute_node_options<T, Options...>::type::list_base_type,
+ ilist_detail::SpecificNodeAccess<
+ typename ilist_detail::compute_node_options<T, Options...>::type> {
+ static_assert(ilist_detail::check_options<Options...>::value,
+ "Unrecognized node option!");
+ using OptionsT =
+ typename ilist_detail::compute_node_options<T, Options...>::type;
+ using list_base_type = typename OptionsT::list_base_type;
+ ilist_sentinel<OptionsT> Sentinel;
+
+public:
+ using value_type = typename OptionsT::value_type;
+ using pointer = typename OptionsT::pointer;
+ using reference = typename OptionsT::reference;
+ using const_pointer = typename OptionsT::const_pointer;
+ using const_reference = typename OptionsT::const_reference;
+ using iterator = ilist_iterator<OptionsT, false, false>;
+ using const_iterator = ilist_iterator<OptionsT, false, true>;
+ using reverse_iterator = ilist_iterator<OptionsT, true, false>;
+ using const_reverse_iterator = ilist_iterator<OptionsT, true, true>;
+ using size_type = size_t;
+ using difference_type = ptrdiff_t;
+
+ simple_ilist() = default;
+ ~simple_ilist() = default;
+
+ // No copy constructors.
+ simple_ilist(const simple_ilist &) = delete;
+ simple_ilist &operator=(const simple_ilist &) = delete;
+
+ // Move constructors.
+ simple_ilist(simple_ilist &&X) { splice(end(), X); }
+ simple_ilist &operator=(simple_ilist &&X) {
+ clear();
+ splice(end(), X);
+ return *this;
+ }
+
+ iterator begin() { return ++iterator(Sentinel); }
+ const_iterator begin() const { return ++const_iterator(Sentinel); }
+ iterator end() { return iterator(Sentinel); }
+ const_iterator end() const { return const_iterator(Sentinel); }
+ reverse_iterator rbegin() { return ++reverse_iterator(Sentinel); }
+ const_reverse_iterator rbegin() const {
+ return ++const_reverse_iterator(Sentinel);
+ }
+ reverse_iterator rend() { return reverse_iterator(Sentinel); }
+ const_reverse_iterator rend() const {
+ return const_reverse_iterator(Sentinel);
+ }
+
+ /// Check if the list is empty in constant time.
+ LLVM_NODISCARD bool empty() const { return Sentinel.empty(); }
+
+ /// Calculate the size of the list in linear time.
+ LLVM_NODISCARD size_type size() const {
+ return std::distance(begin(), end());
+ }
+
+ reference front() { return *begin(); }
+ const_reference front() const { return *begin(); }
+ reference back() { return *rbegin(); }
+ const_reference back() const { return *rbegin(); }
+
+ /// Insert a node at the front; never copies.
+ void push_front(reference Node) { insert(begin(), Node); }
+
+ /// Insert a node at the back; never copies.
+ void push_back(reference Node) { insert(end(), Node); }
+
+ /// Remove the node at the front; never deletes.
+ void pop_front() { erase(begin()); }
+
+ /// Remove the node at the back; never deletes.
+ void pop_back() { erase(--end()); }
+
+ /// Swap with another list in place using std::swap.
+ void swap(simple_ilist &X) { std::swap(*this, X); }
+
+ /// Insert a node by reference; never copies.
+ iterator insert(iterator I, reference Node) {
+ list_base_type::insertBefore(*I.getNodePtr(), *this->getNodePtr(&Node));
+ return iterator(&Node);
+ }
+
+ /// Insert a range of nodes; never copies.
+ template <class Iterator>
+ void insert(iterator I, Iterator First, Iterator Last) {
+ for (; First != Last; ++First)
+ insert(I, *First);
+ }
+
+ /// Clone another list.
+ template <class Cloner, class Disposer>
+ void cloneFrom(const simple_ilist &L2, Cloner clone, Disposer dispose) {
+ clearAndDispose(dispose);
+ for (const_reference V : L2)
+ push_back(*clone(V));
+ }
+
+ /// Remove a node by reference; never deletes.
+ ///
+ /// \see \a erase() for removing by iterator.
+ /// \see \a removeAndDispose() if the node should be deleted.
+ void remove(reference N) { list_base_type::remove(*this->getNodePtr(&N)); }
+
+ /// Remove a node by reference and dispose of it.
+ template <class Disposer>
+ void removeAndDispose(reference N, Disposer dispose) {
+ remove(N);
+ dispose(&N);
+ }
+
+ /// Remove a node by iterator; never deletes.
+ ///
+ /// \see \a remove() for removing by reference.
+ /// \see \a eraseAndDispose() it the node should be deleted.
+ iterator erase(iterator I) {
+ assert(I != end() && "Cannot remove end of list!");
+ remove(*I++);
+ return I;
+ }
+
+ /// Remove a range of nodes; never deletes.
+ ///
+ /// \see \a eraseAndDispose() if the nodes should be deleted.
+ iterator erase(iterator First, iterator Last) {
+ list_base_type::removeRange(*First.getNodePtr(), *Last.getNodePtr());
+ return Last;
+ }
+
+ /// Remove a node by iterator and dispose of it.
+ template <class Disposer>
+ iterator eraseAndDispose(iterator I, Disposer dispose) {
+ auto Next = std::next(I);
+ erase(I);
+ dispose(&*I);
+ return Next;
+ }
+
+ /// Remove a range of nodes and dispose of them.
+ template <class Disposer>
+ iterator eraseAndDispose(iterator First, iterator Last, Disposer dispose) {
+ while (First != Last)
+ First = eraseAndDispose(First, dispose);
+ return Last;
+ }
+
+ /// Clear the list; never deletes.
+ ///
+ /// \see \a clearAndDispose() if the nodes should be deleted.
+ void clear() { Sentinel.reset(); }
+
+ /// Clear the list and dispose of the nodes.
+ template <class Disposer> void clearAndDispose(Disposer dispose) {
+ eraseAndDispose(begin(), end(), dispose);
+ }
+
+ /// Splice in another list.
+ void splice(iterator I, simple_ilist &L2) {
+ splice(I, L2, L2.begin(), L2.end());
+ }
+
+ /// Splice in a node from another list.
+ void splice(iterator I, simple_ilist &L2, iterator Node) {
+ splice(I, L2, Node, std::next(Node));
+ }
+
+ /// Splice in a range of nodes from another list.
+ void splice(iterator I, simple_ilist &, iterator First, iterator Last) {
+ list_base_type::transferBefore(*I.getNodePtr(), *First.getNodePtr(),
+ *Last.getNodePtr());
+ }
+
+ /// Merge in another list.
+ ///
+ /// \pre \c this and \p RHS are sorted.
+ ///@{
+ void merge(simple_ilist &RHS) { merge(RHS, std::less<T>()); }
+ template <class Compare> void merge(simple_ilist &RHS, Compare comp);
+ ///@}
+
+ /// Sort the list.
+ ///@{
+ void sort() { sort(std::less<T>()); }
+ template <class Compare> void sort(Compare comp);
+ ///@}
+};
+
+template <class T, class... Options>
+template <class Compare>
+void simple_ilist<T, Options...>::merge(simple_ilist &RHS, Compare comp) {
+ if (this == &RHS || RHS.empty())
+ return;
+ iterator LI = begin(), LE = end();
+ iterator RI = RHS.begin(), RE = RHS.end();
+ while (LI != LE) {
+ if (comp(*RI, *LI)) {
+ // Transfer a run of at least size 1 from RHS to LHS.
+ iterator RunStart = RI++;
+ RI = std::find_if(RI, RE, [&](reference RV) { return !comp(RV, *LI); });
+ splice(LI, RHS, RunStart, RI);
+ if (RI == RE)
+ return;
+ }
+ ++LI;
+ }
+ // Transfer the remaining RHS nodes once LHS is finished.
+ splice(LE, RHS, RI, RE);
+}
+
+template <class T, class... Options>
+template <class Compare>
+void simple_ilist<T, Options...>::sort(Compare comp) {
+ // Vacuously sorted.
+ if (empty() || std::next(begin()) == end())
+ return;
+
+ // Split the list in the middle.
+ iterator Center = begin(), End = begin();
+ while (End != end() && ++End != end()) {
+ ++Center;
+ ++End;
+ }
+ simple_ilist RHS;
+ RHS.splice(RHS.end(), *this, Center, end());
+
+ // Sort the sublists and merge back together.
+ sort(comp);
+ RHS.sort(comp);
+ merge(RHS, comp);
+}
+
+} // end namespace llvm
+
+#endif // LLVM_ADT_SIMPLE_ILIST_H
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/COFF.h b/third_party/llvm-project/include/llvm/BinaryFormat/COFF.h
new file mode 100644
index 000000000..0fe38a437
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/COFF.h
@@ -0,0 +1,729 @@
+//===-- llvm/BinaryFormat/COFF.h --------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains an definitions used in Windows COFF Files.
+//
+// Structures and enums defined within this file where created using
+// information from Microsoft's publicly available PE/COFF format document:
+//
+// Microsoft Portable Executable and Common Object File Format Specification
+// Revision 8.1 - February 15, 2008
+//
+// As of 5/2/2010, hosted by Microsoft at:
+// http://www.microsoft.com/whdc/system/platform/firmware/pecoff.mspx
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_COFF_H
+#define LLVM_BINARYFORMAT_COFF_H
+
+#include "llvm/Support/DataTypes.h"
+#include <cassert>
+#include <cstring>
+
+namespace llvm {
+namespace COFF {
+
+// The maximum number of sections that a COFF object can have (inclusive).
+const int32_t MaxNumberOfSections16 = 65279;
+
+// The PE signature bytes that follows the DOS stub header.
+static const char PEMagic[] = {'P', 'E', '\0', '\0'};
+
+static const char BigObjMagic[] = {
+ '\xc7', '\xa1', '\xba', '\xd1', '\xee', '\xba', '\xa9', '\x4b',
+ '\xaf', '\x20', '\xfa', '\xf6', '\x6a', '\xa4', '\xdc', '\xb8',
+};
+
+static const char ClGlObjMagic[] = {
+ '\x38', '\xfe', '\xb3', '\x0c', '\xa5', '\xd9', '\xab', '\x4d',
+ '\xac', '\x9b', '\xd6', '\xb6', '\x22', '\x26', '\x53', '\xc2',
+};
+
+// The signature bytes that start a .res file.
+static const char WinResMagic[] = {
+ '\x00', '\x00', '\x00', '\x00', '\x20', '\x00', '\x00', '\x00',
+ '\xff', '\xff', '\x00', '\x00', '\xff', '\xff', '\x00', '\x00',
+};
+
+// Sizes in bytes of various things in the COFF format.
+enum {
+ Header16Size = 20,
+ Header32Size = 56,
+ NameSize = 8,
+ Symbol16Size = 18,
+ Symbol32Size = 20,
+ SectionSize = 40,
+ RelocationSize = 10
+};
+
+struct header {
+ uint16_t Machine;
+ int32_t NumberOfSections;
+ uint32_t TimeDateStamp;
+ uint32_t PointerToSymbolTable;
+ uint32_t NumberOfSymbols;
+ uint16_t SizeOfOptionalHeader;
+ uint16_t Characteristics;
+};
+
+struct BigObjHeader {
+ enum : uint16_t { MinBigObjectVersion = 2 };
+
+ uint16_t Sig1; ///< Must be IMAGE_FILE_MACHINE_UNKNOWN (0).
+ uint16_t Sig2; ///< Must be 0xFFFF.
+ uint16_t Version;
+ uint16_t Machine;
+ uint32_t TimeDateStamp;
+ uint8_t UUID[16];
+ uint32_t unused1;
+ uint32_t unused2;
+ uint32_t unused3;
+ uint32_t unused4;
+ uint32_t NumberOfSections;
+ uint32_t PointerToSymbolTable;
+ uint32_t NumberOfSymbols;
+};
+
+enum MachineTypes : unsigned {
+ MT_Invalid = 0xffff,
+
+ IMAGE_FILE_MACHINE_UNKNOWN = 0x0,
+ IMAGE_FILE_MACHINE_AM33 = 0x1D3,
+ IMAGE_FILE_MACHINE_AMD64 = 0x8664,
+ IMAGE_FILE_MACHINE_ARM = 0x1C0,
+ IMAGE_FILE_MACHINE_ARMNT = 0x1C4,
+ IMAGE_FILE_MACHINE_ARM64 = 0xAA64,
+ IMAGE_FILE_MACHINE_EBC = 0xEBC,
+ IMAGE_FILE_MACHINE_I386 = 0x14C,
+ IMAGE_FILE_MACHINE_IA64 = 0x200,
+ IMAGE_FILE_MACHINE_M32R = 0x9041,
+ IMAGE_FILE_MACHINE_MIPS16 = 0x266,
+ IMAGE_FILE_MACHINE_MIPSFPU = 0x366,
+ IMAGE_FILE_MACHINE_MIPSFPU16 = 0x466,
+ IMAGE_FILE_MACHINE_POWERPC = 0x1F0,
+ IMAGE_FILE_MACHINE_POWERPCFP = 0x1F1,
+ IMAGE_FILE_MACHINE_R4000 = 0x166,
+ IMAGE_FILE_MACHINE_RISCV32 = 0x5032,
+ IMAGE_FILE_MACHINE_RISCV64 = 0x5064,
+ IMAGE_FILE_MACHINE_RISCV128 = 0x5128,
+ IMAGE_FILE_MACHINE_SH3 = 0x1A2,
+ IMAGE_FILE_MACHINE_SH3DSP = 0x1A3,
+ IMAGE_FILE_MACHINE_SH4 = 0x1A6,
+ IMAGE_FILE_MACHINE_SH5 = 0x1A8,
+ IMAGE_FILE_MACHINE_THUMB = 0x1C2,
+ IMAGE_FILE_MACHINE_WCEMIPSV2 = 0x169
+};
+
+enum Characteristics : unsigned {
+ C_Invalid = 0,
+
+ /// The file does not contain base relocations and must be loaded at its
+ /// preferred base. If this cannot be done, the loader will error.
+ IMAGE_FILE_RELOCS_STRIPPED = 0x0001,
+ /// The file is valid and can be run.
+ IMAGE_FILE_EXECUTABLE_IMAGE = 0x0002,
+ /// COFF line numbers have been stripped. This is deprecated and should be
+ /// 0.
+ IMAGE_FILE_LINE_NUMS_STRIPPED = 0x0004,
+ /// COFF symbol table entries for local symbols have been removed. This is
+ /// deprecated and should be 0.
+ IMAGE_FILE_LOCAL_SYMS_STRIPPED = 0x0008,
+ /// Aggressively trim working set. This is deprecated and must be 0.
+ IMAGE_FILE_AGGRESSIVE_WS_TRIM = 0x0010,
+ /// Image can handle > 2GiB addresses.
+ IMAGE_FILE_LARGE_ADDRESS_AWARE = 0x0020,
+ /// Little endian: the LSB precedes the MSB in memory. This is deprecated
+ /// and should be 0.
+ IMAGE_FILE_BYTES_REVERSED_LO = 0x0080,
+ /// Machine is based on a 32bit word architecture.
+ IMAGE_FILE_32BIT_MACHINE = 0x0100,
+ /// Debugging info has been removed.
+ IMAGE_FILE_DEBUG_STRIPPED = 0x0200,
+ /// If the image is on removable media, fully load it and copy it to swap.
+ IMAGE_FILE_REMOVABLE_RUN_FROM_SWAP = 0x0400,
+ /// If the image is on network media, fully load it and copy it to swap.
+ IMAGE_FILE_NET_RUN_FROM_SWAP = 0x0800,
+ /// The image file is a system file, not a user program.
+ IMAGE_FILE_SYSTEM = 0x1000,
+ /// The image file is a DLL.
+ IMAGE_FILE_DLL = 0x2000,
+ /// This file should only be run on a uniprocessor machine.
+ IMAGE_FILE_UP_SYSTEM_ONLY = 0x4000,
+ /// Big endian: the MSB precedes the LSB in memory. This is deprecated
+ /// and should be 0.
+ IMAGE_FILE_BYTES_REVERSED_HI = 0x8000
+};
+
+enum ResourceTypeID : unsigned {
+ RID_Cursor = 1,
+ RID_Bitmap = 2,
+ RID_Icon = 3,
+ RID_Menu = 4,
+ RID_Dialog = 5,
+ RID_String = 6,
+ RID_FontDir = 7,
+ RID_Font = 8,
+ RID_Accelerator = 9,
+ RID_RCData = 10,
+ RID_MessageTable = 11,
+ RID_Group_Cursor = 12,
+ RID_Group_Icon = 14,
+ RID_Version = 16,
+ RID_DLGInclude = 17,
+ RID_PlugPlay = 19,
+ RID_VXD = 20,
+ RID_AniCursor = 21,
+ RID_AniIcon = 22,
+ RID_HTML = 23,
+ RID_Manifest = 24,
+};
+
+struct symbol {
+ char Name[NameSize];
+ uint32_t Value;
+ int32_t SectionNumber;
+ uint16_t Type;
+ uint8_t StorageClass;
+ uint8_t NumberOfAuxSymbols;
+};
+
+enum SymbolSectionNumber : int32_t {
+ IMAGE_SYM_DEBUG = -2,
+ IMAGE_SYM_ABSOLUTE = -1,
+ IMAGE_SYM_UNDEFINED = 0
+};
+
+/// Storage class tells where and what the symbol represents
+enum SymbolStorageClass {
+ SSC_Invalid = 0xff,
+
+ IMAGE_SYM_CLASS_END_OF_FUNCTION = -1, ///< Physical end of function
+ IMAGE_SYM_CLASS_NULL = 0, ///< No symbol
+ IMAGE_SYM_CLASS_AUTOMATIC = 1, ///< Stack variable
+ IMAGE_SYM_CLASS_EXTERNAL = 2, ///< External symbol
+ IMAGE_SYM_CLASS_STATIC = 3, ///< Static
+ IMAGE_SYM_CLASS_REGISTER = 4, ///< Register variable
+ IMAGE_SYM_CLASS_EXTERNAL_DEF = 5, ///< External definition
+ IMAGE_SYM_CLASS_LABEL = 6, ///< Label
+ IMAGE_SYM_CLASS_UNDEFINED_LABEL = 7, ///< Undefined label
+ IMAGE_SYM_CLASS_MEMBER_OF_STRUCT = 8, ///< Member of structure
+ IMAGE_SYM_CLASS_ARGUMENT = 9, ///< Function argument
+ IMAGE_SYM_CLASS_STRUCT_TAG = 10, ///< Structure tag
+ IMAGE_SYM_CLASS_MEMBER_OF_UNION = 11, ///< Member of union
+ IMAGE_SYM_CLASS_UNION_TAG = 12, ///< Union tag
+ IMAGE_SYM_CLASS_TYPE_DEFINITION = 13, ///< Type definition
+ IMAGE_SYM_CLASS_UNDEFINED_STATIC = 14, ///< Undefined static
+ IMAGE_SYM_CLASS_ENUM_TAG = 15, ///< Enumeration tag
+ IMAGE_SYM_CLASS_MEMBER_OF_ENUM = 16, ///< Member of enumeration
+ IMAGE_SYM_CLASS_REGISTER_PARAM = 17, ///< Register parameter
+ IMAGE_SYM_CLASS_BIT_FIELD = 18, ///< Bit field
+ /// ".bb" or ".eb" - beginning or end of block
+ IMAGE_SYM_CLASS_BLOCK = 100,
+ /// ".bf" or ".ef" - beginning or end of function
+ IMAGE_SYM_CLASS_FUNCTION = 101,
+ IMAGE_SYM_CLASS_END_OF_STRUCT = 102, ///< End of structure
+ IMAGE_SYM_CLASS_FILE = 103, ///< File name
+ /// Line number, reformatted as symbol
+ IMAGE_SYM_CLASS_SECTION = 104,
+ IMAGE_SYM_CLASS_WEAK_EXTERNAL = 105, ///< Duplicate tag
+ /// External symbol in dmert public lib
+ IMAGE_SYM_CLASS_CLR_TOKEN = 107
+};
+
+enum SymbolBaseType : unsigned {
+ IMAGE_SYM_TYPE_NULL = 0, ///< No type information or unknown base type.
+ IMAGE_SYM_TYPE_VOID = 1, ///< Used with void pointers and functions.
+ IMAGE_SYM_TYPE_CHAR = 2, ///< A character (signed byte).
+ IMAGE_SYM_TYPE_SHORT = 3, ///< A 2-byte signed integer.
+ IMAGE_SYM_TYPE_INT = 4, ///< A natural integer type on the target.
+ IMAGE_SYM_TYPE_LONG = 5, ///< A 4-byte signed integer.
+ IMAGE_SYM_TYPE_FLOAT = 6, ///< A 4-byte floating-point number.
+ IMAGE_SYM_TYPE_DOUBLE = 7, ///< An 8-byte floating-point number.
+ IMAGE_SYM_TYPE_STRUCT = 8, ///< A structure.
+ IMAGE_SYM_TYPE_UNION = 9, ///< An union.
+ IMAGE_SYM_TYPE_ENUM = 10, ///< An enumerated type.
+ IMAGE_SYM_TYPE_MOE = 11, ///< A member of enumeration (a specific value).
+ IMAGE_SYM_TYPE_BYTE = 12, ///< A byte; unsigned 1-byte integer.
+ IMAGE_SYM_TYPE_WORD = 13, ///< A word; unsigned 2-byte integer.
+ IMAGE_SYM_TYPE_UINT = 14, ///< An unsigned integer of natural size.
+ IMAGE_SYM_TYPE_DWORD = 15 ///< An unsigned 4-byte integer.
+};
+
+enum SymbolComplexType : unsigned {
+ IMAGE_SYM_DTYPE_NULL = 0, ///< No complex type; simple scalar variable.
+ IMAGE_SYM_DTYPE_POINTER = 1, ///< A pointer to base type.
+ IMAGE_SYM_DTYPE_FUNCTION = 2, ///< A function that returns a base type.
+ IMAGE_SYM_DTYPE_ARRAY = 3, ///< An array of base type.
+
+ /// Type is formed as (base + (derived << SCT_COMPLEX_TYPE_SHIFT))
+ SCT_COMPLEX_TYPE_SHIFT = 4
+};
+
+enum AuxSymbolType { IMAGE_AUX_SYMBOL_TYPE_TOKEN_DEF = 1 };
+
+struct section {
+ char Name[NameSize];
+ uint32_t VirtualSize;
+ uint32_t VirtualAddress;
+ uint32_t SizeOfRawData;
+ uint32_t PointerToRawData;
+ uint32_t PointerToRelocations;
+ uint32_t PointerToLineNumbers;
+ uint16_t NumberOfRelocations;
+ uint16_t NumberOfLineNumbers;
+ uint32_t Characteristics;
+};
+
+enum SectionCharacteristics : uint32_t {
+ SC_Invalid = 0xffffffff,
+
+ IMAGE_SCN_TYPE_NOLOAD = 0x00000002,
+ IMAGE_SCN_TYPE_NO_PAD = 0x00000008,
+ IMAGE_SCN_CNT_CODE = 0x00000020,
+ IMAGE_SCN_CNT_INITIALIZED_DATA = 0x00000040,
+ IMAGE_SCN_CNT_UNINITIALIZED_DATA = 0x00000080,
+ IMAGE_SCN_LNK_OTHER = 0x00000100,
+ IMAGE_SCN_LNK_INFO = 0x00000200,
+ IMAGE_SCN_LNK_REMOVE = 0x00000800,
+ IMAGE_SCN_LNK_COMDAT = 0x00001000,
+ IMAGE_SCN_GPREL = 0x00008000,
+ IMAGE_SCN_MEM_PURGEABLE = 0x00020000,
+ IMAGE_SCN_MEM_16BIT = 0x00020000,
+ IMAGE_SCN_MEM_LOCKED = 0x00040000,
+ IMAGE_SCN_MEM_PRELOAD = 0x00080000,
+ IMAGE_SCN_ALIGN_1BYTES = 0x00100000,
+ IMAGE_SCN_ALIGN_2BYTES = 0x00200000,
+ IMAGE_SCN_ALIGN_4BYTES = 0x00300000,
+ IMAGE_SCN_ALIGN_8BYTES = 0x00400000,
+ IMAGE_SCN_ALIGN_16BYTES = 0x00500000,
+ IMAGE_SCN_ALIGN_32BYTES = 0x00600000,
+ IMAGE_SCN_ALIGN_64BYTES = 0x00700000,
+ IMAGE_SCN_ALIGN_128BYTES = 0x00800000,
+ IMAGE_SCN_ALIGN_256BYTES = 0x00900000,
+ IMAGE_SCN_ALIGN_512BYTES = 0x00A00000,
+ IMAGE_SCN_ALIGN_1024BYTES = 0x00B00000,
+ IMAGE_SCN_ALIGN_2048BYTES = 0x00C00000,
+ IMAGE_SCN_ALIGN_4096BYTES = 0x00D00000,
+ IMAGE_SCN_ALIGN_8192BYTES = 0x00E00000,
+ IMAGE_SCN_LNK_NRELOC_OVFL = 0x01000000,
+ IMAGE_SCN_MEM_DISCARDABLE = 0x02000000,
+ IMAGE_SCN_MEM_NOT_CACHED = 0x04000000,
+ IMAGE_SCN_MEM_NOT_PAGED = 0x08000000,
+ IMAGE_SCN_MEM_SHARED = 0x10000000,
+ IMAGE_SCN_MEM_EXECUTE = 0x20000000,
+ IMAGE_SCN_MEM_READ = 0x40000000,
+ IMAGE_SCN_MEM_WRITE = 0x80000000
+};
+
+struct relocation {
+ uint32_t VirtualAddress;
+ uint32_t SymbolTableIndex;
+ uint16_t Type;
+};
+
+enum RelocationTypeI386 : unsigned {
+ IMAGE_REL_I386_ABSOLUTE = 0x0000,
+ IMAGE_REL_I386_DIR16 = 0x0001,
+ IMAGE_REL_I386_REL16 = 0x0002,
+ IMAGE_REL_I386_DIR32 = 0x0006,
+ IMAGE_REL_I386_DIR32NB = 0x0007,
+ IMAGE_REL_I386_SEG12 = 0x0009,
+ IMAGE_REL_I386_SECTION = 0x000A,
+ IMAGE_REL_I386_SECREL = 0x000B,
+ IMAGE_REL_I386_TOKEN = 0x000C,
+ IMAGE_REL_I386_SECREL7 = 0x000D,
+ IMAGE_REL_I386_REL32 = 0x0014
+};
+
+enum RelocationTypeAMD64 : unsigned {
+ IMAGE_REL_AMD64_ABSOLUTE = 0x0000,
+ IMAGE_REL_AMD64_ADDR64 = 0x0001,
+ IMAGE_REL_AMD64_ADDR32 = 0x0002,
+ IMAGE_REL_AMD64_ADDR32NB = 0x0003,
+ IMAGE_REL_AMD64_REL32 = 0x0004,
+ IMAGE_REL_AMD64_REL32_1 = 0x0005,
+ IMAGE_REL_AMD64_REL32_2 = 0x0006,
+ IMAGE_REL_AMD64_REL32_3 = 0x0007,
+ IMAGE_REL_AMD64_REL32_4 = 0x0008,
+ IMAGE_REL_AMD64_REL32_5 = 0x0009,
+ IMAGE_REL_AMD64_SECTION = 0x000A,
+ IMAGE_REL_AMD64_SECREL = 0x000B,
+ IMAGE_REL_AMD64_SECREL7 = 0x000C,
+ IMAGE_REL_AMD64_TOKEN = 0x000D,
+ IMAGE_REL_AMD64_SREL32 = 0x000E,
+ IMAGE_REL_AMD64_PAIR = 0x000F,
+ IMAGE_REL_AMD64_SSPAN32 = 0x0010
+};
+
+enum RelocationTypesARM : unsigned {
+ IMAGE_REL_ARM_ABSOLUTE = 0x0000,
+ IMAGE_REL_ARM_ADDR32 = 0x0001,
+ IMAGE_REL_ARM_ADDR32NB = 0x0002,
+ IMAGE_REL_ARM_BRANCH24 = 0x0003,
+ IMAGE_REL_ARM_BRANCH11 = 0x0004,
+ IMAGE_REL_ARM_TOKEN = 0x0005,
+ IMAGE_REL_ARM_BLX24 = 0x0008,
+ IMAGE_REL_ARM_BLX11 = 0x0009,
+ IMAGE_REL_ARM_REL32 = 0x000A,
+ IMAGE_REL_ARM_SECTION = 0x000E,
+ IMAGE_REL_ARM_SECREL = 0x000F,
+ IMAGE_REL_ARM_MOV32A = 0x0010,
+ IMAGE_REL_ARM_MOV32T = 0x0011,
+ IMAGE_REL_ARM_BRANCH20T = 0x0012,
+ IMAGE_REL_ARM_BRANCH24T = 0x0014,
+ IMAGE_REL_ARM_BLX23T = 0x0015,
+ IMAGE_REL_ARM_PAIR = 0x0016,
+};
+
+enum RelocationTypesARM64 : unsigned {
+ IMAGE_REL_ARM64_ABSOLUTE = 0x0000,
+ IMAGE_REL_ARM64_ADDR32 = 0x0001,
+ IMAGE_REL_ARM64_ADDR32NB = 0x0002,
+ IMAGE_REL_ARM64_BRANCH26 = 0x0003,
+ IMAGE_REL_ARM64_PAGEBASE_REL21 = 0x0004,
+ IMAGE_REL_ARM64_REL21 = 0x0005,
+ IMAGE_REL_ARM64_PAGEOFFSET_12A = 0x0006,
+ IMAGE_REL_ARM64_PAGEOFFSET_12L = 0x0007,
+ IMAGE_REL_ARM64_SECREL = 0x0008,
+ IMAGE_REL_ARM64_SECREL_LOW12A = 0x0009,
+ IMAGE_REL_ARM64_SECREL_HIGH12A = 0x000A,
+ IMAGE_REL_ARM64_SECREL_LOW12L = 0x000B,
+ IMAGE_REL_ARM64_TOKEN = 0x000C,
+ IMAGE_REL_ARM64_SECTION = 0x000D,
+ IMAGE_REL_ARM64_ADDR64 = 0x000E,
+ IMAGE_REL_ARM64_BRANCH19 = 0x000F,
+ IMAGE_REL_ARM64_BRANCH14 = 0x0010,
+ IMAGE_REL_ARM64_REL32 = 0x0011,
+};
+
+enum COMDATType : uint8_t {
+ IMAGE_COMDAT_SELECT_NODUPLICATES = 1,
+ IMAGE_COMDAT_SELECT_ANY,
+ IMAGE_COMDAT_SELECT_SAME_SIZE,
+ IMAGE_COMDAT_SELECT_EXACT_MATCH,
+ IMAGE_COMDAT_SELECT_ASSOCIATIVE,
+ IMAGE_COMDAT_SELECT_LARGEST,
+ IMAGE_COMDAT_SELECT_NEWEST
+};
+
+// Auxiliary Symbol Formats
+struct AuxiliaryFunctionDefinition {
+ uint32_t TagIndex;
+ uint32_t TotalSize;
+ uint32_t PointerToLinenumber;
+ uint32_t PointerToNextFunction;
+ char unused[2];
+};
+
+struct AuxiliarybfAndefSymbol {
+ uint8_t unused1[4];
+ uint16_t Linenumber;
+ uint8_t unused2[6];
+ uint32_t PointerToNextFunction;
+ uint8_t unused3[2];
+};
+
+struct AuxiliaryWeakExternal {
+ uint32_t TagIndex;
+ uint32_t Characteristics;
+ uint8_t unused[10];
+};
+
+enum WeakExternalCharacteristics : unsigned {
+ IMAGE_WEAK_EXTERN_SEARCH_NOLIBRARY = 1,
+ IMAGE_WEAK_EXTERN_SEARCH_LIBRARY = 2,
+ IMAGE_WEAK_EXTERN_SEARCH_ALIAS = 3
+};
+
+struct AuxiliarySectionDefinition {
+ uint32_t Length;
+ uint16_t NumberOfRelocations;
+ uint16_t NumberOfLinenumbers;
+ uint32_t CheckSum;
+ uint32_t Number;
+ uint8_t Selection;
+ char unused;
+};
+
+struct AuxiliaryCLRToken {
+ uint8_t AuxType;
+ uint8_t unused1;
+ uint32_t SymbolTableIndex;
+ char unused2[12];
+};
+
+union Auxiliary {
+ AuxiliaryFunctionDefinition FunctionDefinition;
+ AuxiliarybfAndefSymbol bfAndefSymbol;
+ AuxiliaryWeakExternal WeakExternal;
+ AuxiliarySectionDefinition SectionDefinition;
+};
+
+/// The Import Directory Table.
+///
+/// There is a single array of these and one entry per imported DLL.
+struct ImportDirectoryTableEntry {
+ uint32_t ImportLookupTableRVA;
+ uint32_t TimeDateStamp;
+ uint32_t ForwarderChain;
+ uint32_t NameRVA;
+ uint32_t ImportAddressTableRVA;
+};
+
+/// The PE32 Import Lookup Table.
+///
+/// There is an array of these for each imported DLL. It represents either
+/// the ordinal to import from the target DLL, or a name to lookup and import
+/// from the target DLL.
+///
+/// This also happens to be the same format used by the Import Address Table
+/// when it is initially written out to the image.
+struct ImportLookupTableEntry32 {
+ uint32_t data;
+
+ /// Is this entry specified by ordinal, or name?
+ bool isOrdinal() const { return data & 0x80000000; }
+
+ /// Get the ordinal value of this entry. isOrdinal must be true.
+ uint16_t getOrdinal() const {
+ assert(isOrdinal() && "ILT entry is not an ordinal!");
+ return data & 0xFFFF;
+ }
+
+ /// Set the ordinal value and set isOrdinal to true.
+ void setOrdinal(uint16_t o) {
+ data = o;
+ data |= 0x80000000;
+ }
+
+ /// Get the Hint/Name entry RVA. isOrdinal must be false.
+ uint32_t getHintNameRVA() const {
+ assert(!isOrdinal() && "ILT entry is not a Hint/Name RVA!");
+ return data;
+ }
+
+ /// Set the Hint/Name entry RVA and set isOrdinal to false.
+ void setHintNameRVA(uint32_t rva) { data = rva; }
+};
+
+/// The DOS compatible header at the front of all PEs.
+struct DOSHeader {
+ uint16_t Magic;
+ uint16_t UsedBytesInTheLastPage;
+ uint16_t FileSizeInPages;
+ uint16_t NumberOfRelocationItems;
+ uint16_t HeaderSizeInParagraphs;
+ uint16_t MinimumExtraParagraphs;
+ uint16_t MaximumExtraParagraphs;
+ uint16_t InitialRelativeSS;
+ uint16_t InitialSP;
+ uint16_t Checksum;
+ uint16_t InitialIP;
+ uint16_t InitialRelativeCS;
+ uint16_t AddressOfRelocationTable;
+ uint16_t OverlayNumber;
+ uint16_t Reserved[4];
+ uint16_t OEMid;
+ uint16_t OEMinfo;
+ uint16_t Reserved2[10];
+ uint32_t AddressOfNewExeHeader;
+};
+
+struct PE32Header {
+ enum { PE32 = 0x10b, PE32_PLUS = 0x20b };
+
+ uint16_t Magic;
+ uint8_t MajorLinkerVersion;
+ uint8_t MinorLinkerVersion;
+ uint32_t SizeOfCode;
+ uint32_t SizeOfInitializedData;
+ uint32_t SizeOfUninitializedData;
+ uint32_t AddressOfEntryPoint; // RVA
+ uint32_t BaseOfCode; // RVA
+ uint32_t BaseOfData; // RVA
+ uint32_t ImageBase;
+ uint32_t SectionAlignment;
+ uint32_t FileAlignment;
+ uint16_t MajorOperatingSystemVersion;
+ uint16_t MinorOperatingSystemVersion;
+ uint16_t MajorImageVersion;
+ uint16_t MinorImageVersion;
+ uint16_t MajorSubsystemVersion;
+ uint16_t MinorSubsystemVersion;
+ uint32_t Win32VersionValue;
+ uint32_t SizeOfImage;
+ uint32_t SizeOfHeaders;
+ uint32_t CheckSum;
+ uint16_t Subsystem;
+ // FIXME: This should be DllCharacteristics to match the COFF spec.
+ uint16_t DLLCharacteristics;
+ uint32_t SizeOfStackReserve;
+ uint32_t SizeOfStackCommit;
+ uint32_t SizeOfHeapReserve;
+ uint32_t SizeOfHeapCommit;
+ uint32_t LoaderFlags;
+ // FIXME: This should be NumberOfRvaAndSizes to match the COFF spec.
+ uint32_t NumberOfRvaAndSize;
+};
+
+struct DataDirectory {
+ uint32_t RelativeVirtualAddress;
+ uint32_t Size;
+};
+
+enum DataDirectoryIndex : unsigned {
+ EXPORT_TABLE = 0,
+ IMPORT_TABLE,
+ RESOURCE_TABLE,
+ EXCEPTION_TABLE,
+ CERTIFICATE_TABLE,
+ BASE_RELOCATION_TABLE,
+ DEBUG_DIRECTORY,
+ ARCHITECTURE,
+ GLOBAL_PTR,
+ TLS_TABLE,
+ LOAD_CONFIG_TABLE,
+ BOUND_IMPORT,
+ IAT,
+ DELAY_IMPORT_DESCRIPTOR,
+ CLR_RUNTIME_HEADER,
+
+ NUM_DATA_DIRECTORIES
+};
+
+enum WindowsSubsystem : unsigned {
+ IMAGE_SUBSYSTEM_UNKNOWN = 0, ///< An unknown subsystem.
+ IMAGE_SUBSYSTEM_NATIVE = 1, ///< Device drivers and native Windows processes
+ IMAGE_SUBSYSTEM_WINDOWS_GUI = 2, ///< The Windows GUI subsystem.
+ IMAGE_SUBSYSTEM_WINDOWS_CUI = 3, ///< The Windows character subsystem.
+ IMAGE_SUBSYSTEM_OS2_CUI = 5, ///< The OS/2 character subsytem.
+ IMAGE_SUBSYSTEM_POSIX_CUI = 7, ///< The POSIX character subsystem.
+ IMAGE_SUBSYSTEM_NATIVE_WINDOWS = 8, ///< Native Windows 9x driver.
+ IMAGE_SUBSYSTEM_WINDOWS_CE_GUI = 9, ///< Windows CE.
+ IMAGE_SUBSYSTEM_EFI_APPLICATION = 10, ///< An EFI application.
+ IMAGE_SUBSYSTEM_EFI_BOOT_SERVICE_DRIVER = 11, ///< An EFI driver with boot
+ /// services.
+ IMAGE_SUBSYSTEM_EFI_RUNTIME_DRIVER = 12, ///< An EFI driver with run-time
+ /// services.
+ IMAGE_SUBSYSTEM_EFI_ROM = 13, ///< An EFI ROM image.
+ IMAGE_SUBSYSTEM_XBOX = 14, ///< XBOX.
+ IMAGE_SUBSYSTEM_WINDOWS_BOOT_APPLICATION = 16 ///< A BCD application.
+};
+
+enum DLLCharacteristics : unsigned {
+ /// ASLR with 64 bit address space.
+ IMAGE_DLL_CHARACTERISTICS_HIGH_ENTROPY_VA = 0x0020,
+ /// DLL can be relocated at load time.
+ IMAGE_DLL_CHARACTERISTICS_DYNAMIC_BASE = 0x0040,
+ /// Code integrity checks are enforced.
+ IMAGE_DLL_CHARACTERISTICS_FORCE_INTEGRITY = 0x0080,
+ ///< Image is NX compatible.
+ IMAGE_DLL_CHARACTERISTICS_NX_COMPAT = 0x0100,
+ /// Isolation aware, but do not isolate the image.
+ IMAGE_DLL_CHARACTERISTICS_NO_ISOLATION = 0x0200,
+ /// Does not use structured exception handling (SEH). No SEH handler may be
+ /// called in this image.
+ IMAGE_DLL_CHARACTERISTICS_NO_SEH = 0x0400,
+ /// Do not bind the image.
+ IMAGE_DLL_CHARACTERISTICS_NO_BIND = 0x0800,
+ ///< Image should execute in an AppContainer.
+ IMAGE_DLL_CHARACTERISTICS_APPCONTAINER = 0x1000,
+ ///< A WDM driver.
+ IMAGE_DLL_CHARACTERISTICS_WDM_DRIVER = 0x2000,
+ ///< Image supports Control Flow Guard.
+ IMAGE_DLL_CHARACTERISTICS_GUARD_CF = 0x4000,
+ /// Terminal Server aware.
+ IMAGE_DLL_CHARACTERISTICS_TERMINAL_SERVER_AWARE = 0x8000
+};
+
+enum DebugType : unsigned {
+ IMAGE_DEBUG_TYPE_UNKNOWN = 0,
+ IMAGE_DEBUG_TYPE_COFF = 1,
+ IMAGE_DEBUG_TYPE_CODEVIEW = 2,
+ IMAGE_DEBUG_TYPE_FPO = 3,
+ IMAGE_DEBUG_TYPE_MISC = 4,
+ IMAGE_DEBUG_TYPE_EXCEPTION = 5,
+ IMAGE_DEBUG_TYPE_FIXUP = 6,
+ IMAGE_DEBUG_TYPE_OMAP_TO_SRC = 7,
+ IMAGE_DEBUG_TYPE_OMAP_FROM_SRC = 8,
+ IMAGE_DEBUG_TYPE_BORLAND = 9,
+ IMAGE_DEBUG_TYPE_RESERVED10 = 10,
+ IMAGE_DEBUG_TYPE_CLSID = 11,
+ IMAGE_DEBUG_TYPE_VC_FEATURE = 12,
+ IMAGE_DEBUG_TYPE_POGO = 13,
+ IMAGE_DEBUG_TYPE_ILTCG = 14,
+ IMAGE_DEBUG_TYPE_MPX = 15,
+ IMAGE_DEBUG_TYPE_REPRO = 16,
+};
+
+enum BaseRelocationType : unsigned {
+ IMAGE_REL_BASED_ABSOLUTE = 0,
+ IMAGE_REL_BASED_HIGH = 1,
+ IMAGE_REL_BASED_LOW = 2,
+ IMAGE_REL_BASED_HIGHLOW = 3,
+ IMAGE_REL_BASED_HIGHADJ = 4,
+ IMAGE_REL_BASED_MIPS_JMPADDR = 5,
+ IMAGE_REL_BASED_ARM_MOV32A = 5,
+ IMAGE_REL_BASED_ARM_MOV32T = 7,
+ IMAGE_REL_BASED_MIPS_JMPADDR16 = 9,
+ IMAGE_REL_BASED_DIR64 = 10
+};
+
+enum ImportType : unsigned {
+ IMPORT_CODE = 0,
+ IMPORT_DATA = 1,
+ IMPORT_CONST = 2
+};
+
+enum ImportNameType : unsigned {
+ /// Import is by ordinal. This indicates that the value in the Ordinal/Hint
+ /// field of the import header is the import's ordinal. If this constant is
+ /// not specified, then the Ordinal/Hint field should always be interpreted
+ /// as the import's hint.
+ IMPORT_ORDINAL = 0,
+ /// The import name is identical to the public symbol name
+ IMPORT_NAME = 1,
+ /// The import name is the public symbol name, but skipping the leading ?,
+ /// @, or optionally _.
+ IMPORT_NAME_NOPREFIX = 2,
+ /// The import name is the public symbol name, but skipping the leading ?,
+ /// @, or optionally _, and truncating at the first @.
+ IMPORT_NAME_UNDECORATE = 3
+};
+
+struct ImportHeader {
+ uint16_t Sig1; ///< Must be IMAGE_FILE_MACHINE_UNKNOWN (0).
+ uint16_t Sig2; ///< Must be 0xFFFF.
+ uint16_t Version;
+ uint16_t Machine;
+ uint32_t TimeDateStamp;
+ uint32_t SizeOfData;
+ uint16_t OrdinalHint;
+ uint16_t TypeInfo;
+
+ ImportType getType() const { return static_cast<ImportType>(TypeInfo & 0x3); }
+
+ ImportNameType getNameType() const {
+ return static_cast<ImportNameType>((TypeInfo & 0x1C) >> 2);
+ }
+};
+
+enum CodeViewIdentifiers {
+ DEBUG_SECTION_MAGIC = 0x4,
+ DEBUG_HASHES_SECTION_MAGIC = 0x133C9C5
+};
+
+inline bool isReservedSectionNumber(int32_t SectionNumber) {
+ return SectionNumber <= 0;
+}
+
+} // End namespace COFF.
+} // End namespace llvm.
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.def b/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.def
new file mode 100644
index 000000000..34a7410f7
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.def
@@ -0,0 +1,964 @@
+//===- llvm/Support/Dwarf.def - Dwarf definitions ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Macros for running through Dwarf enumerators.
+//
+//===----------------------------------------------------------------------===//
+
+// TODO: Add other DW-based macros.
+#if !( \
+ defined HANDLE_DW_TAG || defined HANDLE_DW_AT || defined HANDLE_DW_FORM || \
+ defined HANDLE_DW_OP || defined HANDLE_DW_LANG || defined HANDLE_DW_ATE || \
+ defined HANDLE_DW_VIRTUALITY || defined HANDLE_DW_DEFAULTED || \
+ defined HANDLE_DW_CC || defined HANDLE_DW_LNS || defined HANDLE_DW_LNE || \
+ defined HANDLE_DW_LNCT || defined HANDLE_DW_MACRO || \
+ defined HANDLE_DW_RLE || defined HANDLE_DW_LLE || \
+ (defined HANDLE_DW_CFA && defined HANDLE_DW_CFA_PRED) || \
+ defined HANDLE_DW_APPLE_PROPERTY || defined HANDLE_DW_UT || \
+ defined HANDLE_DWARF_SECTION || defined HANDLE_DW_IDX || \
+ defined HANDLE_DW_END)
+#error "Missing macro definition of HANDLE_DW*"
+#endif
+
+#ifndef HANDLE_DW_TAG
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND)
+#endif
+
+// Note that DW_KIND is not a DWARF concept, but rather a way for us to
+// generate a list of tags that belong together.
+#ifndef DW_KIND_NONE
+#define DW_KIND_NONE 0
+#endif
+
+#ifndef DW_KIND_TYPE
+#define DW_KIND_TYPE 1
+#endif
+
+#ifndef HANDLE_DW_AT
+#define HANDLE_DW_AT(ID, NAME, VERSION, VENDOR)
+#endif
+
+#ifndef HANDLE_DW_FORM
+#define HANDLE_DW_FORM(ID, NAME, VERSION, VENDOR)
+#endif
+
+#ifndef HANDLE_DW_OP
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR)
+#endif
+
+#ifndef HANDLE_DW_LANG
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR)
+#endif
+
+#ifndef HANDLE_DW_ATE
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR)
+#endif
+
+#ifndef HANDLE_DW_VIRTUALITY
+#define HANDLE_DW_VIRTUALITY(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_DEFAULTED
+#define HANDLE_DW_DEFAULTED(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_CC
+#define HANDLE_DW_CC(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_LNS
+#define HANDLE_DW_LNS(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_LNE
+#define HANDLE_DW_LNE(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_LNCT
+#define HANDLE_DW_LNCT(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_MACRO
+#define HANDLE_DW_MACRO(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_RLE
+#define HANDLE_DW_RLE(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_LLE
+#define HANDLE_DW_LLE(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_CFA
+#define HANDLE_DW_CFA(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_CFA_PRED
+#define HANDLE_DW_CFA_PRED(ID, NAME, PRED)
+#endif
+
+#ifndef HANDLE_DW_APPLE_PROPERTY
+#define HANDLE_DW_APPLE_PROPERTY(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_UT
+#define HANDLE_DW_UT(ID, NAME)
+#endif
+
+#ifndef HANDLE_DWARF_SECTION
+#define HANDLE_DWARF_SECTION(ENUM_NAME, ELF_NAME, CMDLINE_NAME)
+#endif
+
+#ifndef HANDLE_DW_IDX
+#define HANDLE_DW_IDX(ID, NAME)
+#endif
+
+#ifndef HANDLE_DW_END
+#define HANDLE_DW_END(ID, NAME)
+#endif
+
+HANDLE_DW_TAG(0x0000, null, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0001, array_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0002, class_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0003, entry_point, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0004, enumeration_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0005, formal_parameter, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0008, imported_declaration, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x000a, label, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x000b, lexical_block, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x000d, member, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x000f, pointer_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0010, reference_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0011, compile_unit, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0012, string_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0013, structure_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0015, subroutine_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0016, typedef, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0017, union_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0018, unspecified_parameters, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0019, variant, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001a, common_block, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001b, common_inclusion, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001c, inheritance, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001d, inlined_subroutine, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001e, module, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x001f, ptr_to_member_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0020, set_type, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0021, subrange_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0022, with_stmt, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0023, access_declaration, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0024, base_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0025, catch_block, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0026, const_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0027, constant, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0028, enumerator, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0029, file_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x002a, friend, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x002b, namelist, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x002c, namelist_item, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x002d, packed_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x002e, subprogram, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x002f, template_type_parameter, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0030, template_value_parameter, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0031, thrown_type, 2, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0032, try_block, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0033, variant_part, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0034, variable, 2, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0035, volatile_type, 2, DWARF, DW_KIND_TYPE)
+// New in DWARF v3:
+HANDLE_DW_TAG(0x0036, dwarf_procedure, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0037, restrict_type, 3, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0038, interface_type, 3, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0039, namespace, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x003a, imported_module, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x003b, unspecified_type, 3, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x003c, partial_unit, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x003d, imported_unit, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x003f, condition, 3, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0040, shared_type, 3, DWARF, DW_KIND_TYPE)
+// New in DWARF v4:
+HANDLE_DW_TAG(0x0041, type_unit, 4, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0042, rvalue_reference_type, 4, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0043, template_alias, 4, DWARF, DW_KIND_NONE)
+// New in DWARF v5:
+HANDLE_DW_TAG(0x0044, coarray_type, 5, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0045, generic_subrange, 5, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0046, dynamic_type, 5, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0047, atomic_type, 5, DWARF, DW_KIND_TYPE)
+HANDLE_DW_TAG(0x0048, call_site, 5, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x0049, call_site_parameter, 5, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x004a, skeleton_unit, 5, DWARF, DW_KIND_NONE)
+HANDLE_DW_TAG(0x004b, immutable_type, 5, DWARF, DW_KIND_TYPE)
+// Vendor extensions:
+HANDLE_DW_TAG(0x4081, MIPS_loop, 0, MIPS, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4101, format_label, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4102, function_template, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4103, class_template, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4106, GNU_template_template_param, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4107, GNU_template_parameter_pack, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4108, GNU_formal_parameter_pack, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4109, GNU_call_site, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x410a, GNU_call_site_parameter, 0, GNU, DW_KIND_NONE)
+HANDLE_DW_TAG(0x4200, APPLE_property, 0, APPLE, DW_KIND_NONE)
+HANDLE_DW_TAG(0xb000, BORLAND_property, 0, BORLAND, DW_KIND_NONE)
+HANDLE_DW_TAG(0xb001, BORLAND_Delphi_string, 0, BORLAND, DW_KIND_TYPE)
+HANDLE_DW_TAG(0xb002, BORLAND_Delphi_dynamic_array, 0, BORLAND, DW_KIND_TYPE)
+HANDLE_DW_TAG(0xb003, BORLAND_Delphi_set, 0, BORLAND, DW_KIND_TYPE)
+HANDLE_DW_TAG(0xb004, BORLAND_Delphi_variant, 0, BORLAND, DW_KIND_TYPE)
+
+// Attributes.
+HANDLE_DW_AT(0x01, sibling, 2, DWARF)
+HANDLE_DW_AT(0x02, location, 2, DWARF)
+HANDLE_DW_AT(0x03, name, 2, DWARF)
+HANDLE_DW_AT(0x09, ordering, 2, DWARF)
+HANDLE_DW_AT(0x0b, byte_size, 2, DWARF)
+HANDLE_DW_AT(0x0c, bit_offset, 2, DWARF)
+HANDLE_DW_AT(0x0d, bit_size, 2, DWARF)
+HANDLE_DW_AT(0x10, stmt_list, 2, DWARF)
+HANDLE_DW_AT(0x11, low_pc, 2, DWARF)
+HANDLE_DW_AT(0x12, high_pc, 2, DWARF)
+HANDLE_DW_AT(0x13, language, 2, DWARF)
+HANDLE_DW_AT(0x15, discr, 2, DWARF)
+HANDLE_DW_AT(0x16, discr_value, 2, DWARF)
+HANDLE_DW_AT(0x17, visibility, 2, DWARF)
+HANDLE_DW_AT(0x18, import, 2, DWARF)
+HANDLE_DW_AT(0x19, string_length, 2, DWARF)
+HANDLE_DW_AT(0x1a, common_reference, 2, DWARF)
+HANDLE_DW_AT(0x1b, comp_dir, 2, DWARF)
+HANDLE_DW_AT(0x1c, const_value, 2, DWARF)
+HANDLE_DW_AT(0x1d, containing_type, 2, DWARF)
+HANDLE_DW_AT(0x1e, default_value, 2, DWARF)
+HANDLE_DW_AT(0x20, inline, 2, DWARF)
+HANDLE_DW_AT(0x21, is_optional, 2, DWARF)
+HANDLE_DW_AT(0x22, lower_bound, 2, DWARF)
+HANDLE_DW_AT(0x25, producer, 2, DWARF)
+HANDLE_DW_AT(0x27, prototyped, 2, DWARF)
+HANDLE_DW_AT(0x2a, return_addr, 2, DWARF)
+HANDLE_DW_AT(0x2c, start_scope, 2, DWARF)
+HANDLE_DW_AT(0x2e, bit_stride, 2, DWARF)
+HANDLE_DW_AT(0x2f, upper_bound, 2, DWARF)
+HANDLE_DW_AT(0x31, abstract_origin, 2, DWARF)
+HANDLE_DW_AT(0x32, accessibility, 2, DWARF)
+HANDLE_DW_AT(0x33, address_class, 2, DWARF)
+HANDLE_DW_AT(0x34, artificial, 2, DWARF)
+HANDLE_DW_AT(0x35, base_types, 2, DWARF)
+HANDLE_DW_AT(0x36, calling_convention, 2, DWARF)
+HANDLE_DW_AT(0x37, count, 2, DWARF)
+HANDLE_DW_AT(0x38, data_member_location, 2, DWARF)
+HANDLE_DW_AT(0x39, decl_column, 2, DWARF)
+HANDLE_DW_AT(0x3a, decl_file, 2, DWARF)
+HANDLE_DW_AT(0x3b, decl_line, 2, DWARF)
+HANDLE_DW_AT(0x3c, declaration, 2, DWARF)
+HANDLE_DW_AT(0x3d, discr_list, 2, DWARF)
+HANDLE_DW_AT(0x3e, encoding, 2, DWARF)
+HANDLE_DW_AT(0x3f, external, 2, DWARF)
+HANDLE_DW_AT(0x40, frame_base, 2, DWARF)
+HANDLE_DW_AT(0x41, friend, 2, DWARF)
+HANDLE_DW_AT(0x42, identifier_case, 2, DWARF)
+HANDLE_DW_AT(0x43, macro_info, 2, DWARF)
+HANDLE_DW_AT(0x44, namelist_item, 2, DWARF)
+HANDLE_DW_AT(0x45, priority, 2, DWARF)
+HANDLE_DW_AT(0x46, segment, 2, DWARF)
+HANDLE_DW_AT(0x47, specification, 2, DWARF)
+HANDLE_DW_AT(0x48, static_link, 2, DWARF)
+HANDLE_DW_AT(0x49, type, 2, DWARF)
+HANDLE_DW_AT(0x4a, use_location, 2, DWARF)
+HANDLE_DW_AT(0x4b, variable_parameter, 2, DWARF)
+HANDLE_DW_AT(0x4c, virtuality, 2, DWARF)
+HANDLE_DW_AT(0x4d, vtable_elem_location, 2, DWARF)
+// New in DWARF v3:
+HANDLE_DW_AT(0x4e, allocated, 3, DWARF)
+HANDLE_DW_AT(0x4f, associated, 3, DWARF)
+HANDLE_DW_AT(0x50, data_location, 3, DWARF)
+HANDLE_DW_AT(0x51, byte_stride, 3, DWARF)
+HANDLE_DW_AT(0x52, entry_pc, 3, DWARF)
+HANDLE_DW_AT(0x53, use_UTF8, 3, DWARF)
+HANDLE_DW_AT(0x54, extension, 3, DWARF)
+HANDLE_DW_AT(0x55, ranges, 3, DWARF)
+HANDLE_DW_AT(0x56, trampoline, 3, DWARF)
+HANDLE_DW_AT(0x57, call_column, 3, DWARF)
+HANDLE_DW_AT(0x58, call_file, 3, DWARF)
+HANDLE_DW_AT(0x59, call_line, 3, DWARF)
+HANDLE_DW_AT(0x5a, description, 3, DWARF)
+HANDLE_DW_AT(0x5b, binary_scale, 3, DWARF)
+HANDLE_DW_AT(0x5c, decimal_scale, 3, DWARF)
+HANDLE_DW_AT(0x5d, small, 3, DWARF)
+HANDLE_DW_AT(0x5e, decimal_sign, 3, DWARF)
+HANDLE_DW_AT(0x5f, digit_count, 3, DWARF)
+HANDLE_DW_AT(0x60, picture_string, 3, DWARF)
+HANDLE_DW_AT(0x61, mutable, 3, DWARF)
+HANDLE_DW_AT(0x62, threads_scaled, 3, DWARF)
+HANDLE_DW_AT(0x63, explicit, 3, DWARF)
+HANDLE_DW_AT(0x64, object_pointer, 3, DWARF)
+HANDLE_DW_AT(0x65, endianity, 3, DWARF)
+HANDLE_DW_AT(0x66, elemental, 3, DWARF)
+HANDLE_DW_AT(0x67, pure, 3, DWARF)
+HANDLE_DW_AT(0x68, recursive, 3, DWARF)
+// New in DWARF v4:
+HANDLE_DW_AT(0x69, signature, 4, DWARF)
+HANDLE_DW_AT(0x6a, main_subprogram, 4, DWARF)
+HANDLE_DW_AT(0x6b, data_bit_offset, 4, DWARF)
+HANDLE_DW_AT(0x6c, const_expr, 4, DWARF)
+HANDLE_DW_AT(0x6d, enum_class, 4, DWARF)
+HANDLE_DW_AT(0x6e, linkage_name, 4, DWARF)
+// New in DWARF v5:
+HANDLE_DW_AT(0x6f, string_length_bit_size, 5, DWARF)
+HANDLE_DW_AT(0x70, string_length_byte_size, 5, DWARF)
+HANDLE_DW_AT(0x71, rank, 5, DWARF)
+HANDLE_DW_AT(0x72, str_offsets_base, 5, DWARF)
+HANDLE_DW_AT(0x73, addr_base, 5, DWARF)
+HANDLE_DW_AT(0x74, rnglists_base, 5, DWARF)
+HANDLE_DW_AT(0x75, dwo_id, 0, DWARF) ///< Retracted from DWARF v5.
+HANDLE_DW_AT(0x76, dwo_name, 5, DWARF)
+HANDLE_DW_AT(0x77, reference, 5, DWARF)
+HANDLE_DW_AT(0x78, rvalue_reference, 5, DWARF)
+HANDLE_DW_AT(0x79, macros, 5, DWARF)
+HANDLE_DW_AT(0x7a, call_all_calls, 5, DWARF)
+HANDLE_DW_AT(0x7b, call_all_source_calls, 5, DWARF)
+HANDLE_DW_AT(0x7c, call_all_tail_calls, 5, DWARF)
+HANDLE_DW_AT(0x7d, call_return_pc, 5, DWARF)
+HANDLE_DW_AT(0x7e, call_value, 5, DWARF)
+HANDLE_DW_AT(0x7f, call_origin, 5, DWARF)
+HANDLE_DW_AT(0x80, call_parameter, 5, DWARF)
+HANDLE_DW_AT(0x81, call_pc, 5, DWARF)
+HANDLE_DW_AT(0x82, call_tail_call, 5, DWARF)
+HANDLE_DW_AT(0x83, call_target, 5, DWARF)
+HANDLE_DW_AT(0x84, call_target_clobbered, 5, DWARF)
+HANDLE_DW_AT(0x85, call_data_location, 5, DWARF)
+HANDLE_DW_AT(0x86, call_data_value, 5, DWARF)
+HANDLE_DW_AT(0x87, noreturn, 5, DWARF)
+HANDLE_DW_AT(0x88, alignment, 5, DWARF)
+HANDLE_DW_AT(0x89, export_symbols, 5, DWARF)
+HANDLE_DW_AT(0x8a, deleted, 5, DWARF)
+HANDLE_DW_AT(0x8b, defaulted, 5, DWARF)
+HANDLE_DW_AT(0x8c, loclists_base, 5, DWARF)
+// Vendor extensions:
+HANDLE_DW_AT(0x2002, MIPS_loop_begin, 0, MIPS)
+HANDLE_DW_AT(0x2003, MIPS_tail_loop_begin, 0, MIPS)
+HANDLE_DW_AT(0x2004, MIPS_epilog_begin, 0, MIPS)
+HANDLE_DW_AT(0x2005, MIPS_loop_unroll_factor, 0, MIPS)
+HANDLE_DW_AT(0x2006, MIPS_software_pipeline_depth, 0, MIPS)
+HANDLE_DW_AT(0x2007, MIPS_linkage_name, 0, MIPS)
+HANDLE_DW_AT(0x2008, MIPS_stride, 0, MIPS)
+HANDLE_DW_AT(0x2009, MIPS_abstract_name, 0, MIPS)
+HANDLE_DW_AT(0x200a, MIPS_clone_origin, 0, MIPS)
+HANDLE_DW_AT(0x200b, MIPS_has_inlines, 0, MIPS)
+HANDLE_DW_AT(0x200c, MIPS_stride_byte, 0, MIPS)
+HANDLE_DW_AT(0x200d, MIPS_stride_elem, 0, MIPS)
+HANDLE_DW_AT(0x200e, MIPS_ptr_dopetype, 0, MIPS)
+HANDLE_DW_AT(0x200f, MIPS_allocatable_dopetype, 0, MIPS)
+HANDLE_DW_AT(0x2010, MIPS_assumed_shape_dopetype, 0, MIPS)
+// This one appears to have only been implemented by Open64 for
+// fortran and may conflict with other extensions.
+HANDLE_DW_AT(0x2011, MIPS_assumed_size, 0, MIPS)
+// GNU extensions
+HANDLE_DW_AT(0x2101, sf_names, 0, GNU)
+HANDLE_DW_AT(0x2102, src_info, 0, GNU)
+HANDLE_DW_AT(0x2103, mac_info, 0, GNU)
+HANDLE_DW_AT(0x2104, src_coords, 0, GNU)
+HANDLE_DW_AT(0x2105, body_begin, 0, GNU)
+HANDLE_DW_AT(0x2106, body_end, 0, GNU)
+HANDLE_DW_AT(0x2107, GNU_vector, 0, GNU)
+HANDLE_DW_AT(0x2110, GNU_template_name, 0, GNU)
+HANDLE_DW_AT(0x210f, GNU_odr_signature, 0, GNU)
+HANDLE_DW_AT(0x2111, GNU_call_site_value, 0, GNU)
+HANDLE_DW_AT (0x2112, GNU_call_site_data_value, 0, GNU)
+HANDLE_DW_AT (0x2113, GNU_call_site_target, 0, GNU)
+HANDLE_DW_AT (0x2114, GNU_call_site_target_clobbered, 0, GNU)
+HANDLE_DW_AT (0x2115, GNU_tail_call, 0, GNU)
+HANDLE_DW_AT (0x2116, GNU_all_tail_call_sites, 0, GNU)
+HANDLE_DW_AT(0x2117, GNU_all_call_sites, 0, GNU)
+HANDLE_DW_AT (0x2118, GNU_all_source_call_sites, 0, GNU)
+HANDLE_DW_AT(0x2119, GNU_macros, 0, GNU)
+// Extensions for Fission proposal.
+HANDLE_DW_AT(0x2130, GNU_dwo_name, 0, GNU)
+HANDLE_DW_AT(0x2131, GNU_dwo_id, 0, GNU)
+HANDLE_DW_AT(0x2132, GNU_ranges_base, 0, GNU)
+HANDLE_DW_AT(0x2133, GNU_addr_base, 0, GNU)
+HANDLE_DW_AT(0x2134, GNU_pubnames, 0, GNU)
+HANDLE_DW_AT(0x2135, GNU_pubtypes, 0, GNU)
+HANDLE_DW_AT(0x2136, GNU_discriminator, 0, GNU)
+// Borland extensions.
+HANDLE_DW_AT(0x3b11, BORLAND_property_read, 0, BORLAND)
+HANDLE_DW_AT(0x3b12, BORLAND_property_write, 0, BORLAND)
+HANDLE_DW_AT(0x3b13, BORLAND_property_implements, 0, BORLAND)
+HANDLE_DW_AT(0x3b14, BORLAND_property_index, 0, BORLAND)
+HANDLE_DW_AT(0x3b15, BORLAND_property_default, 0, BORLAND)
+HANDLE_DW_AT(0x3b20, BORLAND_Delphi_unit, 0, BORLAND)
+HANDLE_DW_AT(0x3b21, BORLAND_Delphi_class, 0, BORLAND)
+HANDLE_DW_AT(0x3b22, BORLAND_Delphi_record, 0, BORLAND)
+HANDLE_DW_AT(0x3b23, BORLAND_Delphi_metaclass, 0, BORLAND)
+HANDLE_DW_AT(0x3b24, BORLAND_Delphi_constructor, 0, BORLAND)
+HANDLE_DW_AT(0x3b25, BORLAND_Delphi_destructor, 0, BORLAND)
+HANDLE_DW_AT(0x3b26, BORLAND_Delphi_anonymous_method, 0, BORLAND)
+HANDLE_DW_AT(0x3b27, BORLAND_Delphi_interface, 0, BORLAND)
+HANDLE_DW_AT(0x3b28, BORLAND_Delphi_ABI, 0, BORLAND)
+HANDLE_DW_AT(0x3b29, BORLAND_Delphi_return, 0, BORLAND)
+HANDLE_DW_AT(0x3b30, BORLAND_Delphi_frameptr, 0, BORLAND)
+HANDLE_DW_AT(0x3b31, BORLAND_closure, 0, BORLAND)
+// LLVM project extensions.
+HANDLE_DW_AT(0x3e00, LLVM_include_path, 0, LLVM)
+HANDLE_DW_AT(0x3e01, LLVM_config_macros, 0, LLVM)
+HANDLE_DW_AT(0x3e02, LLVM_isysroot, 0, LLVM)
+HANDLE_DW_AT(0x3e03, LLVM_tag_offset, 0, LLVM)
+// Apple extensions.
+HANDLE_DW_AT(0x3fe1, APPLE_optimized, 0, APPLE)
+HANDLE_DW_AT(0x3fe2, APPLE_flags, 0, APPLE)
+HANDLE_DW_AT(0x3fe3, APPLE_isa, 0, APPLE)
+HANDLE_DW_AT(0x3fe4, APPLE_block, 0, APPLE)
+HANDLE_DW_AT(0x3fe5, APPLE_major_runtime_vers, 0, APPLE)
+HANDLE_DW_AT(0x3fe6, APPLE_runtime_class, 0, APPLE)
+HANDLE_DW_AT(0x3fe7, APPLE_omit_frame_ptr, 0, APPLE)
+HANDLE_DW_AT(0x3fe8, APPLE_property_name, 0, APPLE)
+HANDLE_DW_AT(0x3fe9, APPLE_property_getter, 0, APPLE)
+HANDLE_DW_AT(0x3fea, APPLE_property_setter, 0, APPLE)
+HANDLE_DW_AT(0x3feb, APPLE_property_attribute, 0, APPLE)
+HANDLE_DW_AT(0x3fec, APPLE_objc_complete_type, 0, APPLE)
+HANDLE_DW_AT(0x3fed, APPLE_property, 0, APPLE)
+
+// Attribute form encodings.
+HANDLE_DW_FORM(0x01, addr, 2, DWARF)
+HANDLE_DW_FORM(0x03, block2, 2, DWARF)
+HANDLE_DW_FORM(0x04, block4, 2, DWARF)
+HANDLE_DW_FORM(0x05, data2, 2, DWARF)
+HANDLE_DW_FORM(0x06, data4, 2, DWARF)
+HANDLE_DW_FORM(0x07, data8, 2, DWARF)
+HANDLE_DW_FORM(0x08, string, 2, DWARF)
+HANDLE_DW_FORM(0x09, block, 2, DWARF)
+HANDLE_DW_FORM(0x0a, block1, 2, DWARF)
+HANDLE_DW_FORM(0x0b, data1, 2, DWARF)
+HANDLE_DW_FORM(0x0c, flag, 2, DWARF)
+HANDLE_DW_FORM(0x0d, sdata, 2, DWARF)
+HANDLE_DW_FORM(0x0e, strp, 2, DWARF)
+HANDLE_DW_FORM(0x0f, udata, 2, DWARF)
+HANDLE_DW_FORM(0x10, ref_addr, 2, DWARF)
+HANDLE_DW_FORM(0x11, ref1, 2, DWARF)
+HANDLE_DW_FORM(0x12, ref2, 2, DWARF)
+HANDLE_DW_FORM(0x13, ref4, 2, DWARF)
+HANDLE_DW_FORM(0x14, ref8, 2, DWARF)
+HANDLE_DW_FORM(0x15, ref_udata, 2, DWARF)
+HANDLE_DW_FORM(0x16, indirect, 2, DWARF)
+// New in DWARF v4:
+HANDLE_DW_FORM(0x17, sec_offset, 4, DWARF)
+HANDLE_DW_FORM(0x18, exprloc, 4, DWARF)
+HANDLE_DW_FORM(0x19, flag_present, 4, DWARF)
+// This was defined out of sequence.
+HANDLE_DW_FORM(0x20, ref_sig8, 4, DWARF)
+// New in DWARF v5:
+HANDLE_DW_FORM(0x1a, strx, 5, DWARF)
+HANDLE_DW_FORM(0x1b, addrx, 5, DWARF)
+HANDLE_DW_FORM(0x1c, ref_sup4, 5, DWARF)
+HANDLE_DW_FORM(0x1d, strp_sup, 5, DWARF)
+HANDLE_DW_FORM(0x1e, data16, 5, DWARF)
+HANDLE_DW_FORM(0x1f, line_strp, 5, DWARF)
+HANDLE_DW_FORM(0x21, implicit_const, 5, DWARF)
+HANDLE_DW_FORM(0x22, loclistx, 5, DWARF)
+HANDLE_DW_FORM(0x23, rnglistx, 5, DWARF)
+HANDLE_DW_FORM(0x24, ref_sup8, 5, DWARF)
+HANDLE_DW_FORM(0x25, strx1, 5, DWARF)
+HANDLE_DW_FORM(0x26, strx2, 5, DWARF)
+HANDLE_DW_FORM(0x27, strx3, 5, DWARF)
+HANDLE_DW_FORM(0x28, strx4, 5, DWARF)
+HANDLE_DW_FORM(0x29, addrx1, 5, DWARF)
+HANDLE_DW_FORM(0x2a, addrx2, 5, DWARF)
+HANDLE_DW_FORM(0x2b, addrx3, 5, DWARF)
+HANDLE_DW_FORM(0x2c, addrx4, 5, DWARF)
+// Extensions for Fission proposal
+HANDLE_DW_FORM(0x1f01, GNU_addr_index, 0, GNU)
+HANDLE_DW_FORM(0x1f02, GNU_str_index, 0, GNU)
+// Alternate debug sections proposal (output of "dwz" tool).
+HANDLE_DW_FORM(0x1f20, GNU_ref_alt, 0, GNU)
+HANDLE_DW_FORM(0x1f21, GNU_strp_alt, 0, GNU)
+
+// DWARF Expression operators.
+HANDLE_DW_OP(0x03, addr, 2, DWARF)
+HANDLE_DW_OP(0x06, deref, 2, DWARF)
+HANDLE_DW_OP(0x08, const1u, 2, DWARF)
+HANDLE_DW_OP(0x09, const1s, 2, DWARF)
+HANDLE_DW_OP(0x0a, const2u, 2, DWARF)
+HANDLE_DW_OP(0x0b, const2s, 2, DWARF)
+HANDLE_DW_OP(0x0c, const4u, 2, DWARF)
+HANDLE_DW_OP(0x0d, const4s, 2, DWARF)
+HANDLE_DW_OP(0x0e, const8u, 2, DWARF)
+HANDLE_DW_OP(0x0f, const8s, 2, DWARF)
+HANDLE_DW_OP(0x10, constu, 2, DWARF)
+HANDLE_DW_OP(0x11, consts, 2, DWARF)
+HANDLE_DW_OP(0x12, dup, 2, DWARF)
+HANDLE_DW_OP(0x13, drop, 2, DWARF)
+HANDLE_DW_OP(0x14, over, 2, DWARF)
+HANDLE_DW_OP(0x15, pick, 2, DWARF)
+HANDLE_DW_OP(0x16, swap, 2, DWARF)
+HANDLE_DW_OP(0x17, rot, 2, DWARF)
+HANDLE_DW_OP(0x18, xderef, 2, DWARF)
+HANDLE_DW_OP(0x19, abs, 2, DWARF)
+HANDLE_DW_OP(0x1a, and, 2, DWARF)
+HANDLE_DW_OP(0x1b, div, 2, DWARF)
+HANDLE_DW_OP(0x1c, minus, 2, DWARF)
+HANDLE_DW_OP(0x1d, mod, 2, DWARF)
+HANDLE_DW_OP(0x1e, mul, 2, DWARF)
+HANDLE_DW_OP(0x1f, neg, 2, DWARF)
+HANDLE_DW_OP(0x20, not, 2, DWARF)
+HANDLE_DW_OP(0x21, or, 2, DWARF)
+HANDLE_DW_OP(0x22, plus, 2, DWARF)
+HANDLE_DW_OP(0x23, plus_uconst, 2, DWARF)
+HANDLE_DW_OP(0x24, shl, 2, DWARF)
+HANDLE_DW_OP(0x25, shr, 2, DWARF)
+HANDLE_DW_OP(0x26, shra, 2, DWARF)
+HANDLE_DW_OP(0x27, xor, 2, DWARF)
+HANDLE_DW_OP(0x28, bra, 2, DWARF)
+HANDLE_DW_OP(0x29, eq, 2, DWARF)
+HANDLE_DW_OP(0x2a, ge, 2, DWARF)
+HANDLE_DW_OP(0x2b, gt, 2, DWARF)
+HANDLE_DW_OP(0x2c, le, 2, DWARF)
+HANDLE_DW_OP(0x2d, lt, 2, DWARF)
+HANDLE_DW_OP(0x2e, ne, 2, DWARF)
+HANDLE_DW_OP(0x2f, skip, 2, DWARF)
+HANDLE_DW_OP(0x30, lit0, 2, DWARF)
+HANDLE_DW_OP(0x31, lit1, 2, DWARF)
+HANDLE_DW_OP(0x32, lit2, 2, DWARF)
+HANDLE_DW_OP(0x33, lit3, 2, DWARF)
+HANDLE_DW_OP(0x34, lit4, 2, DWARF)
+HANDLE_DW_OP(0x35, lit5, 2, DWARF)
+HANDLE_DW_OP(0x36, lit6, 2, DWARF)
+HANDLE_DW_OP(0x37, lit7, 2, DWARF)
+HANDLE_DW_OP(0x38, lit8, 2, DWARF)
+HANDLE_DW_OP(0x39, lit9, 2, DWARF)
+HANDLE_DW_OP(0x3a, lit10, 2, DWARF)
+HANDLE_DW_OP(0x3b, lit11, 2, DWARF)
+HANDLE_DW_OP(0x3c, lit12, 2, DWARF)
+HANDLE_DW_OP(0x3d, lit13, 2, DWARF)
+HANDLE_DW_OP(0x3e, lit14, 2, DWARF)
+HANDLE_DW_OP(0x3f, lit15, 2, DWARF)
+HANDLE_DW_OP(0x40, lit16, 2, DWARF)
+HANDLE_DW_OP(0x41, lit17, 2, DWARF)
+HANDLE_DW_OP(0x42, lit18, 2, DWARF)
+HANDLE_DW_OP(0x43, lit19, 2, DWARF)
+HANDLE_DW_OP(0x44, lit20, 2, DWARF)
+HANDLE_DW_OP(0x45, lit21, 2, DWARF)
+HANDLE_DW_OP(0x46, lit22, 2, DWARF)
+HANDLE_DW_OP(0x47, lit23, 2, DWARF)
+HANDLE_DW_OP(0x48, lit24, 2, DWARF)
+HANDLE_DW_OP(0x49, lit25, 2, DWARF)
+HANDLE_DW_OP(0x4a, lit26, 2, DWARF)
+HANDLE_DW_OP(0x4b, lit27, 2, DWARF)
+HANDLE_DW_OP(0x4c, lit28, 2, DWARF)
+HANDLE_DW_OP(0x4d, lit29, 2, DWARF)
+HANDLE_DW_OP(0x4e, lit30, 2, DWARF)
+HANDLE_DW_OP(0x4f, lit31, 2, DWARF)
+HANDLE_DW_OP(0x50, reg0, 2, DWARF)
+HANDLE_DW_OP(0x51, reg1, 2, DWARF)
+HANDLE_DW_OP(0x52, reg2, 2, DWARF)
+HANDLE_DW_OP(0x53, reg3, 2, DWARF)
+HANDLE_DW_OP(0x54, reg4, 2, DWARF)
+HANDLE_DW_OP(0x55, reg5, 2, DWARF)
+HANDLE_DW_OP(0x56, reg6, 2, DWARF)
+HANDLE_DW_OP(0x57, reg7, 2, DWARF)
+HANDLE_DW_OP(0x58, reg8, 2, DWARF)
+HANDLE_DW_OP(0x59, reg9, 2, DWARF)
+HANDLE_DW_OP(0x5a, reg10, 2, DWARF)
+HANDLE_DW_OP(0x5b, reg11, 2, DWARF)
+HANDLE_DW_OP(0x5c, reg12, 2, DWARF)
+HANDLE_DW_OP(0x5d, reg13, 2, DWARF)
+HANDLE_DW_OP(0x5e, reg14, 2, DWARF)
+HANDLE_DW_OP(0x5f, reg15, 2, DWARF)
+HANDLE_DW_OP(0x60, reg16, 2, DWARF)
+HANDLE_DW_OP(0x61, reg17, 2, DWARF)
+HANDLE_DW_OP(0x62, reg18, 2, DWARF)
+HANDLE_DW_OP(0x63, reg19, 2, DWARF)
+HANDLE_DW_OP(0x64, reg20, 2, DWARF)
+HANDLE_DW_OP(0x65, reg21, 2, DWARF)
+HANDLE_DW_OP(0x66, reg22, 2, DWARF)
+HANDLE_DW_OP(0x67, reg23, 2, DWARF)
+HANDLE_DW_OP(0x68, reg24, 2, DWARF)
+HANDLE_DW_OP(0x69, reg25, 2, DWARF)
+HANDLE_DW_OP(0x6a, reg26, 2, DWARF)
+HANDLE_DW_OP(0x6b, reg27, 2, DWARF)
+HANDLE_DW_OP(0x6c, reg28, 2, DWARF)
+HANDLE_DW_OP(0x6d, reg29, 2, DWARF)
+HANDLE_DW_OP(0x6e, reg30, 2, DWARF)
+HANDLE_DW_OP(0x6f, reg31, 2, DWARF)
+HANDLE_DW_OP(0x70, breg0, 2, DWARF)
+HANDLE_DW_OP(0x71, breg1, 2, DWARF)
+HANDLE_DW_OP(0x72, breg2, 2, DWARF)
+HANDLE_DW_OP(0x73, breg3, 2, DWARF)
+HANDLE_DW_OP(0x74, breg4, 2, DWARF)
+HANDLE_DW_OP(0x75, breg5, 2, DWARF)
+HANDLE_DW_OP(0x76, breg6, 2, DWARF)
+HANDLE_DW_OP(0x77, breg7, 2, DWARF)
+HANDLE_DW_OP(0x78, breg8, 2, DWARF)
+HANDLE_DW_OP(0x79, breg9, 2, DWARF)
+HANDLE_DW_OP(0x7a, breg10, 2, DWARF)
+HANDLE_DW_OP(0x7b, breg11, 2, DWARF)
+HANDLE_DW_OP(0x7c, breg12, 2, DWARF)
+HANDLE_DW_OP(0x7d, breg13, 2, DWARF)
+HANDLE_DW_OP(0x7e, breg14, 2, DWARF)
+HANDLE_DW_OP(0x7f, breg15, 2, DWARF)
+HANDLE_DW_OP(0x80, breg16, 2, DWARF)
+HANDLE_DW_OP(0x81, breg17, 2, DWARF)
+HANDLE_DW_OP(0x82, breg18, 2, DWARF)
+HANDLE_DW_OP(0x83, breg19, 2, DWARF)
+HANDLE_DW_OP(0x84, breg20, 2, DWARF)
+HANDLE_DW_OP(0x85, breg21, 2, DWARF)
+HANDLE_DW_OP(0x86, breg22, 2, DWARF)
+HANDLE_DW_OP(0x87, breg23, 2, DWARF)
+HANDLE_DW_OP(0x88, breg24, 2, DWARF)
+HANDLE_DW_OP(0x89, breg25, 2, DWARF)
+HANDLE_DW_OP(0x8a, breg26, 2, DWARF)
+HANDLE_DW_OP(0x8b, breg27, 2, DWARF)
+HANDLE_DW_OP(0x8c, breg28, 2, DWARF)
+HANDLE_DW_OP(0x8d, breg29, 2, DWARF)
+HANDLE_DW_OP(0x8e, breg30, 2, DWARF)
+HANDLE_DW_OP(0x8f, breg31, 2, DWARF)
+HANDLE_DW_OP(0x90, regx, 2, DWARF)
+HANDLE_DW_OP(0x91, fbreg, 2, DWARF)
+HANDLE_DW_OP(0x92, bregx, 2, DWARF)
+HANDLE_DW_OP(0x93, piece, 2, DWARF)
+HANDLE_DW_OP(0x94, deref_size, 2, DWARF)
+HANDLE_DW_OP(0x95, xderef_size, 2, DWARF)
+HANDLE_DW_OP(0x96, nop, 2, DWARF)
+// New in DWARF v3:
+HANDLE_DW_OP(0x97, push_object_address, 3, DWARF)
+HANDLE_DW_OP(0x98, call2, 3, DWARF)
+HANDLE_DW_OP(0x99, call4, 3, DWARF)
+HANDLE_DW_OP(0x9a, call_ref, 3, DWARF)
+HANDLE_DW_OP(0x9b, form_tls_address, 3, DWARF)
+HANDLE_DW_OP(0x9c, call_frame_cfa, 3, DWARF)
+HANDLE_DW_OP(0x9d, bit_piece, 3, DWARF)
+// New in DWARF v4:
+HANDLE_DW_OP(0x9e, implicit_value, 4, DWARF)
+HANDLE_DW_OP(0x9f, stack_value, 4, DWARF)
+// New in DWARF v5:
+HANDLE_DW_OP(0xa0, implicit_pointer, 5, DWARF)
+HANDLE_DW_OP(0xa1, addrx, 5, DWARF)
+HANDLE_DW_OP(0xa2, constx, 5, DWARF)
+HANDLE_DW_OP(0xa3, entry_value, 5, DWARF)
+HANDLE_DW_OP(0xa4, const_type, 5, DWARF)
+HANDLE_DW_OP(0xa5, regval_type, 5, DWARF)
+HANDLE_DW_OP(0xa6, deref_type, 5, DWARF)
+HANDLE_DW_OP(0xa7, xderef_type, 5, DWARF)
+HANDLE_DW_OP(0xa8, convert, 5, DWARF)
+HANDLE_DW_OP(0xa9, reinterpret, 5, DWARF)
+// Vendor extensions:
+// Extensions for GNU-style thread-local storage.
+HANDLE_DW_OP(0xe0, GNU_push_tls_address, 0, GNU)
+// The GNU entry value extension.
+HANDLE_DW_OP(0xf3, GNU_entry_value, 0, GNU)
+// Extensions for Fission proposal.
+HANDLE_DW_OP(0xfb, GNU_addr_index, 0, GNU)
+HANDLE_DW_OP(0xfc, GNU_const_index, 0, GNU)
+
+// DWARF languages.
+HANDLE_DW_LANG(0x0001, C89, 0, 2, DWARF)
+HANDLE_DW_LANG(0x0002, C, 0, 2, DWARF)
+HANDLE_DW_LANG(0x0003, Ada83, 1, 2, DWARF)
+HANDLE_DW_LANG(0x0004, C_plus_plus, 0, 2, DWARF)
+HANDLE_DW_LANG(0x0005, Cobol74, 1, 2, DWARF)
+HANDLE_DW_LANG(0x0006, Cobol85, 1, 2, DWARF)
+HANDLE_DW_LANG(0x0007, Fortran77, 1, 2, DWARF)
+HANDLE_DW_LANG(0x0008, Fortran90, 1, 2, DWARF)
+HANDLE_DW_LANG(0x0009, Pascal83, 1, 2, DWARF)
+HANDLE_DW_LANG(0x000a, Modula2, 1, 2, DWARF)
+// New in DWARF v3:
+HANDLE_DW_LANG(0x000b, Java, 0, 3, DWARF)
+HANDLE_DW_LANG(0x000c, C99, 0, 3, DWARF)
+HANDLE_DW_LANG(0x000d, Ada95, 1, 3, DWARF)
+HANDLE_DW_LANG(0x000e, Fortran95, 1, 3, DWARF)
+HANDLE_DW_LANG(0x000f, PLI, 1, 3, DWARF)
+HANDLE_DW_LANG(0x0010, ObjC, 0, 3, DWARF)
+HANDLE_DW_LANG(0x0011, ObjC_plus_plus, 0, 3, DWARF)
+HANDLE_DW_LANG(0x0012, UPC, 0, 3, DWARF)
+HANDLE_DW_LANG(0x0013, D, 0, 3, DWARF)
+// New in DWARF v4:
+HANDLE_DW_LANG(0x0014, Python, 0, 4, DWARF)
+// New in DWARF v5:
+HANDLE_DW_LANG(0x0015, OpenCL, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0016, Go, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0017, Modula3, 1, 5, DWARF)
+HANDLE_DW_LANG(0x0018, Haskell, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0019, C_plus_plus_03, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001a, C_plus_plus_11, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001b, OCaml, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001c, Rust, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001d, C11, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001e, Swift, 0, 5, DWARF)
+HANDLE_DW_LANG(0x001f, Julia, 1, 5, DWARF)
+HANDLE_DW_LANG(0x0020, Dylan, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0021, C_plus_plus_14, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0022, Fortran03, 1, 5, DWARF)
+HANDLE_DW_LANG(0x0023, Fortran08, 1, 5, DWARF)
+HANDLE_DW_LANG(0x0024, RenderScript, 0, 5, DWARF)
+HANDLE_DW_LANG(0x0025, BLISS, 0, 5, DWARF)
+// Vendor extensions:
+HANDLE_DW_LANG(0x8001, Mips_Assembler, None, 0, MIPS)
+HANDLE_DW_LANG(0x8e57, GOOGLE_RenderScript, 0, 0, GOOGLE)
+HANDLE_DW_LANG(0xb000, BORLAND_Delphi, 0, 0, BORLAND)
+
+// DWARF attribute type encodings.
+HANDLE_DW_ATE(0x01, address, 2, DWARF)
+HANDLE_DW_ATE(0x02, boolean, 2, DWARF)
+HANDLE_DW_ATE(0x03, complex_float, 2, DWARF)
+HANDLE_DW_ATE(0x04, float, 2, DWARF)
+HANDLE_DW_ATE(0x05, signed, 2, DWARF)
+HANDLE_DW_ATE(0x06, signed_char, 2, DWARF)
+HANDLE_DW_ATE(0x07, unsigned, 2, DWARF)
+HANDLE_DW_ATE(0x08, unsigned_char, 2, DWARF)
+// New in DWARF v3:
+HANDLE_DW_ATE(0x09, imaginary_float, 3, DWARF)
+HANDLE_DW_ATE(0x0a, packed_decimal, 3, DWARF)
+HANDLE_DW_ATE(0x0b, numeric_string, 3, DWARF)
+HANDLE_DW_ATE(0x0c, edited, 3, DWARF)
+HANDLE_DW_ATE(0x0d, signed_fixed, 3, DWARF)
+HANDLE_DW_ATE(0x0e, unsigned_fixed, 3, DWARF)
+HANDLE_DW_ATE(0x0f, decimal_float, 3, DWARF)
+// New in DWARF v4:
+HANDLE_DW_ATE(0x10, UTF, 4, DWARF)
+// New in DWARF v5:
+HANDLE_DW_ATE(0x11, UCS, 5, DWARF)
+HANDLE_DW_ATE(0x12, ASCII, 5, DWARF)
+
+// DWARF attribute endianity
+HANDLE_DW_END(0x00, default)
+HANDLE_DW_END(0x01, big)
+HANDLE_DW_END(0x02, little)
+
+// DWARF virtuality codes.
+HANDLE_DW_VIRTUALITY(0x00, none)
+HANDLE_DW_VIRTUALITY(0x01, virtual)
+HANDLE_DW_VIRTUALITY(0x02, pure_virtual)
+
+// DWARF v5 Defaulted Member Encodings.
+HANDLE_DW_DEFAULTED(0x00, no)
+HANDLE_DW_DEFAULTED(0x01, in_class)
+HANDLE_DW_DEFAULTED(0x02, out_of_class)
+
+// DWARF calling convention codes.
+HANDLE_DW_CC(0x01, normal)
+HANDLE_DW_CC(0x02, program)
+HANDLE_DW_CC(0x03, nocall)
+// New in DWARF v5:
+HANDLE_DW_CC(0x04, pass_by_reference)
+HANDLE_DW_CC(0x05, pass_by_value)
+// Vendor extensions:
+HANDLE_DW_CC(0x40, GNU_renesas_sh)
+HANDLE_DW_CC(0x41, GNU_borland_fastcall_i386)
+HANDLE_DW_CC(0xb0, BORLAND_safecall)
+HANDLE_DW_CC(0xb1, BORLAND_stdcall)
+HANDLE_DW_CC(0xb2, BORLAND_pascal)
+HANDLE_DW_CC(0xb3, BORLAND_msfastcall)
+HANDLE_DW_CC(0xb4, BORLAND_msreturn)
+HANDLE_DW_CC(0xb5, BORLAND_thiscall)
+HANDLE_DW_CC(0xb6, BORLAND_fastcall)
+HANDLE_DW_CC(0xc0, LLVM_vectorcall)
+HANDLE_DW_CC(0xc1, LLVM_Win64)
+HANDLE_DW_CC(0xc2, LLVM_X86_64SysV)
+HANDLE_DW_CC(0xc3, LLVM_AAPCS)
+HANDLE_DW_CC(0xc4, LLVM_AAPCS_VFP)
+HANDLE_DW_CC(0xc5, LLVM_IntelOclBicc)
+HANDLE_DW_CC(0xc6, LLVM_SpirFunction)
+HANDLE_DW_CC(0xc7, LLVM_OpenCLKernel)
+HANDLE_DW_CC(0xc8, LLVM_Swift)
+HANDLE_DW_CC(0xc9, LLVM_PreserveMost)
+HANDLE_DW_CC(0xca, LLVM_PreserveAll)
+HANDLE_DW_CC(0xcb, LLVM_X86RegCall)
+// From GCC source code (include/dwarf2.h): This DW_CC_ value is not currently
+// generated by any toolchain. It is used internally to GDB to indicate OpenCL C
+// functions that have been compiled with the IBM XL C for OpenCL compiler and use
+// a non-platform calling convention for passing OpenCL C vector types.
+HANDLE_DW_CC(0xff, GDB_IBM_OpenCL)
+
+// Line Number Extended Opcode Encodings
+HANDLE_DW_LNE(0x01, end_sequence)
+HANDLE_DW_LNE(0x02, set_address)
+HANDLE_DW_LNE(0x03, define_file)
+// New in DWARF v4:
+HANDLE_DW_LNE(0x04, set_discriminator)
+
+// Line Number Standard Opcode Encodings.
+HANDLE_DW_LNS(0x00, extended_op)
+HANDLE_DW_LNS(0x01, copy)
+HANDLE_DW_LNS(0x02, advance_pc)
+HANDLE_DW_LNS(0x03, advance_line)
+HANDLE_DW_LNS(0x04, set_file)
+HANDLE_DW_LNS(0x05, set_column)
+HANDLE_DW_LNS(0x06, negate_stmt)
+HANDLE_DW_LNS(0x07, set_basic_block)
+HANDLE_DW_LNS(0x08, const_add_pc)
+HANDLE_DW_LNS(0x09, fixed_advance_pc)
+// New in DWARF v3:
+HANDLE_DW_LNS(0x0a, set_prologue_end)
+HANDLE_DW_LNS(0x0b, set_epilogue_begin)
+HANDLE_DW_LNS(0x0c, set_isa)
+
+// DWARF v5 Line number header entry format.
+HANDLE_DW_LNCT(0x01, path)
+HANDLE_DW_LNCT(0x02, directory_index)
+HANDLE_DW_LNCT(0x03, timestamp)
+HANDLE_DW_LNCT(0x04, size)
+HANDLE_DW_LNCT(0x05, MD5)
+// A vendor extension until http://dwarfstd.org/ShowIssue.php?issue=180201.1 is
+// accepted and incorporated into the next DWARF standard.
+HANDLE_DW_LNCT(0x2001, LLVM_source)
+
+// DWARF v5 Macro information.
+HANDLE_DW_MACRO(0x01, define)
+HANDLE_DW_MACRO(0x02, undef)
+HANDLE_DW_MACRO(0x03, start_file)
+HANDLE_DW_MACRO(0x04, end_file)
+HANDLE_DW_MACRO(0x05, define_strp)
+HANDLE_DW_MACRO(0x06, undef_strp)
+HANDLE_DW_MACRO(0x07, import)
+HANDLE_DW_MACRO(0x08, define_sup)
+HANDLE_DW_MACRO(0x09, undef_sup)
+HANDLE_DW_MACRO(0x0a, import_sup)
+HANDLE_DW_MACRO(0x0b, define_strx)
+HANDLE_DW_MACRO(0x0c, undef_strx)
+
+// DWARF v5 Range List Entry encoding values.
+HANDLE_DW_RLE(0x00, end_of_list)
+HANDLE_DW_RLE(0x01, base_addressx)
+HANDLE_DW_RLE(0x02, startx_endx)
+HANDLE_DW_RLE(0x03, startx_length)
+HANDLE_DW_RLE(0x04, offset_pair)
+HANDLE_DW_RLE(0x05, base_address)
+HANDLE_DW_RLE(0x06, start_end)
+HANDLE_DW_RLE(0x07, start_length)
+
+// DWARF v5 Loc List Entry encoding values.
+HANDLE_DW_LLE(0x00, end_of_list)
+HANDLE_DW_LLE(0x01, base_addressx)
+HANDLE_DW_LLE(0x02, startx_endx)
+HANDLE_DW_LLE(0x03, startx_length)
+HANDLE_DW_LLE(0x04, offset_pair)
+HANDLE_DW_LLE(0x05, default_location)
+HANDLE_DW_LLE(0x06, base_address)
+HANDLE_DW_LLE(0x07, start_end)
+HANDLE_DW_LLE(0x08, start_length)
+
+// Call frame instruction encodings.
+HANDLE_DW_CFA(0x00, nop)
+HANDLE_DW_CFA(0x40, advance_loc)
+HANDLE_DW_CFA(0x80, offset)
+HANDLE_DW_CFA(0xc0, restore)
+HANDLE_DW_CFA(0x01, set_loc)
+HANDLE_DW_CFA(0x02, advance_loc1)
+HANDLE_DW_CFA(0x03, advance_loc2)
+HANDLE_DW_CFA(0x04, advance_loc4)
+HANDLE_DW_CFA(0x05, offset_extended)
+HANDLE_DW_CFA(0x06, restore_extended)
+HANDLE_DW_CFA(0x07, undefined)
+HANDLE_DW_CFA(0x08, same_value)
+HANDLE_DW_CFA(0x09, register)
+HANDLE_DW_CFA(0x0a, remember_state)
+HANDLE_DW_CFA(0x0b, restore_state)
+HANDLE_DW_CFA(0x0c, def_cfa)
+HANDLE_DW_CFA(0x0d, def_cfa_register)
+HANDLE_DW_CFA(0x0e, def_cfa_offset)
+// New in DWARF v3:
+HANDLE_DW_CFA(0x0f, def_cfa_expression)
+HANDLE_DW_CFA(0x10, expression)
+HANDLE_DW_CFA(0x11, offset_extended_sf)
+HANDLE_DW_CFA(0x12, def_cfa_sf)
+HANDLE_DW_CFA(0x13, def_cfa_offset_sf)
+HANDLE_DW_CFA(0x14, val_offset)
+HANDLE_DW_CFA(0x15, val_offset_sf)
+HANDLE_DW_CFA(0x16, val_expression)
+// Vendor extensions:
+HANDLE_DW_CFA_PRED(0x1d, MIPS_advance_loc8, SELECT_MIPS64)
+HANDLE_DW_CFA_PRED(0x2d, GNU_window_save, SELECT_SPARC)
+HANDLE_DW_CFA_PRED(0x2d, AARCH64_negate_ra_state, SELECT_AARCH64)
+HANDLE_DW_CFA_PRED(0x2e, GNU_args_size, SELECT_X86)
+
+// Apple Objective-C Property Attributes.
+// Keep this list in sync with clang's DeclSpec.h ObjCPropertyAttributeKind!
+HANDLE_DW_APPLE_PROPERTY(0x01, readonly)
+HANDLE_DW_APPLE_PROPERTY(0x02, getter)
+HANDLE_DW_APPLE_PROPERTY(0x04, assign)
+HANDLE_DW_APPLE_PROPERTY(0x08, readwrite)
+HANDLE_DW_APPLE_PROPERTY(0x10, retain)
+HANDLE_DW_APPLE_PROPERTY(0x20, copy)
+HANDLE_DW_APPLE_PROPERTY(0x40, nonatomic)
+HANDLE_DW_APPLE_PROPERTY(0x80, setter)
+HANDLE_DW_APPLE_PROPERTY(0x100, atomic)
+HANDLE_DW_APPLE_PROPERTY(0x200, weak)
+HANDLE_DW_APPLE_PROPERTY(0x400, strong)
+HANDLE_DW_APPLE_PROPERTY(0x800, unsafe_unretained)
+HANDLE_DW_APPLE_PROPERTY(0x1000, nullability)
+HANDLE_DW_APPLE_PROPERTY(0x2000, null_resettable)
+HANDLE_DW_APPLE_PROPERTY(0x4000, class)
+
+// DWARF v5 Unit Types.
+HANDLE_DW_UT(0x01, compile)
+HANDLE_DW_UT(0x02, type)
+HANDLE_DW_UT(0x03, partial)
+HANDLE_DW_UT(0x04, skeleton)
+HANDLE_DW_UT(0x05, split_compile)
+HANDLE_DW_UT(0x06, split_type)
+
+// DWARF section types. (enum name, ELF name, ELF DWO name, cmdline name)
+// Note that these IDs don't mean anything.
+// TODO: Add Mach-O and COFF names.
+// Official DWARF sections.
+HANDLE_DWARF_SECTION(DebugAbbrev, ".debug_abbrev", "debug-abbrev")
+HANDLE_DWARF_SECTION(DebugAddr, ".debug_addr", "debug-addr")
+HANDLE_DWARF_SECTION(DebugAranges, ".debug_aranges", "debug-aranges")
+HANDLE_DWARF_SECTION(DebugInfo, ".debug_info", "debug-info")
+HANDLE_DWARF_SECTION(DebugTypes, ".debug_types", "debug-types")
+HANDLE_DWARF_SECTION(DebugLine, ".debug_line", "debug-line")
+HANDLE_DWARF_SECTION(DebugLineStr, ".debug_line_str", "debug-line-str")
+HANDLE_DWARF_SECTION(DebugLoc, ".debug_loc", "debug-loc")
+HANDLE_DWARF_SECTION(DebugLoclists, ".debug_loclists", "debug-loclists")
+HANDLE_DWARF_SECTION(DebugFrame, ".debug_frame", "debug-frame")
+HANDLE_DWARF_SECTION(DebugMacro, ".debug_macro", "debug-macro")
+HANDLE_DWARF_SECTION(DebugNames, ".debug_names", "debug-names")
+HANDLE_DWARF_SECTION(DebugPubnames, ".debug_pubnames", "debug-pubnames")
+HANDLE_DWARF_SECTION(DebugPubtypes, ".debug_pubtypes", "debug-pubtypes")
+HANDLE_DWARF_SECTION(DebugGnuPubnames, ".debug_gnu_pubnames", "debug-gnu-pubnames")
+HANDLE_DWARF_SECTION(DebugGnuPubtypes, ".debug_gnu_pubtypes", "debug-gnu-pubtypes")
+HANDLE_DWARF_SECTION(DebugRanges, ".debug_ranges", "debug-ranges")
+HANDLE_DWARF_SECTION(DebugRnglists, ".debug_rnglists", "debug-rnglists")
+HANDLE_DWARF_SECTION(DebugStr, ".debug_str", "debug-str")
+HANDLE_DWARF_SECTION(DebugStrOffsets, ".debug_str_offsets", "debug-str-offsets")
+HANDLE_DWARF_SECTION(DebugCUIndex, ".debug_cu_index", "debug-cu-index")
+HANDLE_DWARF_SECTION(DebugTUIndex, ".debug_tu_index", "debug-tu-index")
+// Vendor extensions.
+HANDLE_DWARF_SECTION(AppleNames, ".apple_names", "apple-names")
+HANDLE_DWARF_SECTION(AppleTypes, ".apple_types", "apple-types")
+HANDLE_DWARF_SECTION(AppleNamespaces, ".apple_namespaces", "apple-namespaces")
+HANDLE_DWARF_SECTION(AppleObjC, ".apple_objc", "apple-objc")
+HANDLE_DWARF_SECTION(GdbIndex, ".gdb_index", "gdb-index")
+
+HANDLE_DW_IDX(0x01, compile_unit)
+HANDLE_DW_IDX(0x02, type_unit)
+HANDLE_DW_IDX(0x03, die_offset)
+HANDLE_DW_IDX(0x04, parent)
+HANDLE_DW_IDX(0x05, type_hash)
+
+
+#undef HANDLE_DW_TAG
+#undef HANDLE_DW_AT
+#undef HANDLE_DW_FORM
+#undef HANDLE_DW_OP
+#undef HANDLE_DW_LANG
+#undef HANDLE_DW_ATE
+#undef HANDLE_DW_VIRTUALITY
+#undef HANDLE_DW_DEFAULTED
+#undef HANDLE_DW_CC
+#undef HANDLE_DW_LNS
+#undef HANDLE_DW_LNE
+#undef HANDLE_DW_LNCT
+#undef HANDLE_DW_MACRO
+#undef HANDLE_DW_RLE
+#undef HANDLE_DW_LLE
+#undef HANDLE_DW_CFA
+#undef HANDLE_DW_CFA_PRED
+#undef HANDLE_DW_APPLE_PROPERTY
+#undef HANDLE_DW_UT
+#undef HANDLE_DWARF_SECTION
+#undef HANDLE_DW_IDX
+#undef HANDLE_DW_END
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.h b/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.h
new file mode 100644
index 000000000..93eaf3680
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/Dwarf.h
@@ -0,0 +1,677 @@
+//===-- llvm/BinaryFormat/Dwarf.h ---Dwarf Constants-------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+/// \file
+/// This file contains constants used for implementing Dwarf
+/// debug support.
+///
+/// For details on the Dwarf specfication see the latest DWARF Debugging
+/// Information Format standard document on http://www.dwarfstd.org. This
+/// file often includes support for non-released standard features.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_DWARF_H
+#define LLVM_BINARYFORMAT_DWARF_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatVariadicDetails.h"
+#include "llvm/ADT/Triple.h"
+
+namespace llvm {
+class StringRef;
+
+namespace dwarf {
+
+//===----------------------------------------------------------------------===//
+// DWARF constants as gleaned from the DWARF Debugging Information Format V.5
+// reference manual http://www.dwarfstd.org/.
+//
+
+// Do not mix the following two enumerations sets. DW_TAG_invalid changes the
+// enumeration base type.
+
+enum LLVMConstants : uint32_t {
+ // LLVM mock tags (see also llvm/BinaryFormat/Dwarf.def).
+ DW_TAG_invalid = ~0U, // Tag for invalid results.
+ DW_VIRTUALITY_invalid = ~0U, // Virtuality for invalid results.
+ DW_MACINFO_invalid = ~0U, // Macinfo type for invalid results.
+
+ // Special values for an initial length field.
+ DW_LENGTH_lo_reserved = 0xfffffff0, // Lower bound of the reserved range.
+ DW_LENGTH_DWARF64 = 0xffffffff, // Indicator of 64-bit DWARF format.
+ DW_LENGTH_hi_reserved = 0xffffffff, // Upper bound of the reserved range.
+
+ // Other constants.
+ DWARF_VERSION = 4, // Default dwarf version we output.
+ DW_PUBTYPES_VERSION = 2, // Section version number for .debug_pubtypes.
+ DW_PUBNAMES_VERSION = 2, // Section version number for .debug_pubnames.
+ DW_ARANGES_VERSION = 2, // Section version number for .debug_aranges.
+ // Identifiers we use to distinguish vendor extensions.
+ DWARF_VENDOR_DWARF = 0, // Defined in v2 or later of the DWARF standard.
+ DWARF_VENDOR_APPLE = 1,
+ DWARF_VENDOR_BORLAND = 2,
+ DWARF_VENDOR_GNU = 3,
+ DWARF_VENDOR_GOOGLE = 4,
+ DWARF_VENDOR_LLVM = 5,
+ DWARF_VENDOR_MIPS = 6
+};
+
+/// Constants that define the DWARF format as 32 or 64 bit.
+enum DwarfFormat : uint8_t { DWARF32, DWARF64 };
+
+/// Special ID values that distinguish a CIE from a FDE in DWARF CFI.
+/// Not inside an enum because a 64-bit value is needed.
+/// @{
+const uint32_t DW_CIE_ID = UINT32_MAX;
+const uint64_t DW64_CIE_ID = UINT64_MAX;
+/// @}
+
+/// Identifier of an invalid DIE offset in the .debug_info section.
+const uint32_t DW_INVALID_OFFSET = UINT32_MAX;
+
+enum Tag : uint16_t {
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) DW_TAG_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_TAG_lo_user = 0x4080,
+ DW_TAG_hi_user = 0xffff,
+ DW_TAG_user_base = 0x1000 ///< Recommended base for user tags.
+};
+
+inline bool isType(Tag T) {
+ switch (T) {
+ default:
+ return false;
+#define HANDLE_DW_TAG(ID, NAME, VERSION, VENDOR, KIND) \
+ case DW_TAG_##NAME: \
+ return (KIND == DW_KIND_TYPE);
+#include "llvm/BinaryFormat/Dwarf.def"
+ }
+}
+
+/// Attributes.
+enum Attribute : uint16_t {
+#define HANDLE_DW_AT(ID, NAME, VERSION, VENDOR) DW_AT_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_AT_lo_user = 0x2000,
+ DW_AT_hi_user = 0x3fff,
+};
+
+enum Form : uint16_t {
+#define HANDLE_DW_FORM(ID, NAME, VERSION, VENDOR) DW_FORM_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_FORM_lo_user = 0x1f00, ///< Not specified by DWARF.
+};
+
+enum LocationAtom {
+#define HANDLE_DW_OP(ID, NAME, VERSION, VENDOR) DW_OP_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_OP_lo_user = 0xe0,
+ DW_OP_hi_user = 0xff,
+ DW_OP_LLVM_fragment = 0x1000, ///< Only used in LLVM metadata.
+ DW_OP_LLVM_convert = 0x1001, ///< Only used in LLVM metadata.
+ DW_OP_LLVM_tag_offset = 0x1002, ///< Only used in LLVM metadata.
+ DW_OP_LLVM_entry_value = 0x1003, ///< Only used in LLVM metadata.
+};
+
+enum TypeKind : uint8_t {
+#define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR) DW_ATE_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_ATE_lo_user = 0x80,
+ DW_ATE_hi_user = 0xff
+};
+
+enum DecimalSignEncoding {
+ // Decimal sign attribute values
+ DW_DS_unsigned = 0x01,
+ DW_DS_leading_overpunch = 0x02,
+ DW_DS_trailing_overpunch = 0x03,
+ DW_DS_leading_separate = 0x04,
+ DW_DS_trailing_separate = 0x05
+};
+
+enum EndianityEncoding {
+ // Endianity attribute values
+#define HANDLE_DW_END(ID, NAME) DW_END_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_END_lo_user = 0x40,
+ DW_END_hi_user = 0xff
+};
+
+enum AccessAttribute {
+ // Accessibility codes
+ DW_ACCESS_public = 0x01,
+ DW_ACCESS_protected = 0x02,
+ DW_ACCESS_private = 0x03
+};
+
+enum VisibilityAttribute {
+ // Visibility codes
+ DW_VIS_local = 0x01,
+ DW_VIS_exported = 0x02,
+ DW_VIS_qualified = 0x03
+};
+
+enum VirtualityAttribute {
+#define HANDLE_DW_VIRTUALITY(ID, NAME) DW_VIRTUALITY_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_VIRTUALITY_max = 0x02
+};
+
+enum DefaultedMemberAttribute {
+#define HANDLE_DW_DEFAULTED(ID, NAME) DW_DEFAULTED_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_DEFAULTED_max = 0x02
+};
+
+enum SourceLanguage {
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR) \
+ DW_LANG_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_LANG_lo_user = 0x8000,
+ DW_LANG_hi_user = 0xffff
+};
+
+inline bool isCPlusPlus(SourceLanguage S) {
+ // Deliberately enumerate all the language options so we get a warning when
+ // new language options are added (-Wswitch) that'll hopefully help keep this
+ // switch up-to-date when new C++ versions are added.
+ switch (S) {
+ case DW_LANG_C_plus_plus:
+ case DW_LANG_C_plus_plus_03:
+ case DW_LANG_C_plus_plus_11:
+ case DW_LANG_C_plus_plus_14:
+ return true;
+ case DW_LANG_C89:
+ case DW_LANG_C:
+ case DW_LANG_Ada83:
+ case DW_LANG_Cobol74:
+ case DW_LANG_Cobol85:
+ case DW_LANG_Fortran77:
+ case DW_LANG_Fortran90:
+ case DW_LANG_Pascal83:
+ case DW_LANG_Modula2:
+ case DW_LANG_Java:
+ case DW_LANG_C99:
+ case DW_LANG_Ada95:
+ case DW_LANG_Fortran95:
+ case DW_LANG_PLI:
+ case DW_LANG_ObjC:
+ case DW_LANG_ObjC_plus_plus:
+ case DW_LANG_UPC:
+ case DW_LANG_D:
+ case DW_LANG_Python:
+ case DW_LANG_OpenCL:
+ case DW_LANG_Go:
+ case DW_LANG_Modula3:
+ case DW_LANG_Haskell:
+ case DW_LANG_OCaml:
+ case DW_LANG_Rust:
+ case DW_LANG_C11:
+ case DW_LANG_Swift:
+ case DW_LANG_Julia:
+ case DW_LANG_Dylan:
+ case DW_LANG_Fortran03:
+ case DW_LANG_Fortran08:
+ case DW_LANG_RenderScript:
+ case DW_LANG_BLISS:
+ case DW_LANG_Mips_Assembler:
+ case DW_LANG_GOOGLE_RenderScript:
+ case DW_LANG_BORLAND_Delphi:
+ case DW_LANG_lo_user:
+ case DW_LANG_hi_user:
+ return false;
+ }
+ llvm_unreachable("Invalid source language");
+}
+
+enum CaseSensitivity {
+ // Identifier case codes
+ DW_ID_case_sensitive = 0x00,
+ DW_ID_up_case = 0x01,
+ DW_ID_down_case = 0x02,
+ DW_ID_case_insensitive = 0x03
+};
+
+enum CallingConvention {
+// Calling convention codes
+#define HANDLE_DW_CC(ID, NAME) DW_CC_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_CC_lo_user = 0x40,
+ DW_CC_hi_user = 0xff
+};
+
+enum InlineAttribute {
+ // Inline codes
+ DW_INL_not_inlined = 0x00,
+ DW_INL_inlined = 0x01,
+ DW_INL_declared_not_inlined = 0x02,
+ DW_INL_declared_inlined = 0x03
+};
+
+enum ArrayDimensionOrdering {
+ // Array ordering
+ DW_ORD_row_major = 0x00,
+ DW_ORD_col_major = 0x01
+};
+
+enum DiscriminantList {
+ // Discriminant descriptor values
+ DW_DSC_label = 0x00,
+ DW_DSC_range = 0x01
+};
+
+/// Line Number Standard Opcode Encodings.
+enum LineNumberOps : uint8_t {
+#define HANDLE_DW_LNS(ID, NAME) DW_LNS_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+};
+
+/// Line Number Extended Opcode Encodings.
+enum LineNumberExtendedOps {
+#define HANDLE_DW_LNE(ID, NAME) DW_LNE_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_LNE_lo_user = 0x80,
+ DW_LNE_hi_user = 0xff
+};
+
+enum LineNumberEntryFormat {
+#define HANDLE_DW_LNCT(ID, NAME) DW_LNCT_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_LNCT_lo_user = 0x2000,
+ DW_LNCT_hi_user = 0x3fff,
+};
+
+enum MacinfoRecordType {
+ // Macinfo Type Encodings
+ DW_MACINFO_define = 0x01,
+ DW_MACINFO_undef = 0x02,
+ DW_MACINFO_start_file = 0x03,
+ DW_MACINFO_end_file = 0x04,
+ DW_MACINFO_vendor_ext = 0xff
+};
+
+/// DWARF v5 macro information entry type encodings.
+enum MacroEntryType {
+#define HANDLE_DW_MACRO(ID, NAME) DW_MACRO_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_MACRO_lo_user = 0xe0,
+ DW_MACRO_hi_user = 0xff
+};
+
+/// DWARF v5 range list entry encoding values.
+enum RnglistEntries {
+#define HANDLE_DW_RLE(ID, NAME) DW_RLE_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+};
+
+/// DWARF v5 loc list entry encoding values.
+enum LoclistEntries {
+#define HANDLE_DW_LLE(ID, NAME) DW_LLE_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+};
+
+/// Call frame instruction encodings.
+enum CallFrameInfo {
+#define HANDLE_DW_CFA(ID, NAME) DW_CFA_##NAME = ID,
+#define HANDLE_DW_CFA_PRED(ID, NAME, ARCH) DW_CFA_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_CFA_extended = 0x00,
+
+ DW_CFA_lo_user = 0x1c,
+ DW_CFA_hi_user = 0x3f
+};
+
+enum Constants {
+ // Children flag
+ DW_CHILDREN_no = 0x00,
+ DW_CHILDREN_yes = 0x01,
+
+ DW_EH_PE_absptr = 0x00,
+ DW_EH_PE_omit = 0xff,
+ DW_EH_PE_uleb128 = 0x01,
+ DW_EH_PE_udata2 = 0x02,
+ DW_EH_PE_udata4 = 0x03,
+ DW_EH_PE_udata8 = 0x04,
+ DW_EH_PE_sleb128 = 0x09,
+ DW_EH_PE_sdata2 = 0x0A,
+ DW_EH_PE_sdata4 = 0x0B,
+ DW_EH_PE_sdata8 = 0x0C,
+ DW_EH_PE_signed = 0x08,
+ DW_EH_PE_pcrel = 0x10,
+ DW_EH_PE_textrel = 0x20,
+ DW_EH_PE_datarel = 0x30,
+ DW_EH_PE_funcrel = 0x40,
+ DW_EH_PE_aligned = 0x50,
+ DW_EH_PE_indirect = 0x80
+};
+
+/// Constants for the DW_APPLE_PROPERTY_attributes attribute.
+/// Keep this list in sync with clang's DeclSpec.h ObjCPropertyAttributeKind!
+enum ApplePropertyAttributes {
+#define HANDLE_DW_APPLE_PROPERTY(ID, NAME) DW_APPLE_PROPERTY_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+};
+
+/// Constants for unit types in DWARF v5.
+enum UnitType : unsigned char {
+#define HANDLE_DW_UT(ID, NAME) DW_UT_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_UT_lo_user = 0x80,
+ DW_UT_hi_user = 0xff
+};
+
+enum Index {
+#define HANDLE_DW_IDX(ID, NAME) DW_IDX_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+ DW_IDX_lo_user = 0x2000,
+ DW_IDX_hi_user = 0x3fff
+};
+
+inline bool isUnitType(uint8_t UnitType) {
+ switch (UnitType) {
+ case DW_UT_compile:
+ case DW_UT_type:
+ case DW_UT_partial:
+ case DW_UT_skeleton:
+ case DW_UT_split_compile:
+ case DW_UT_split_type:
+ return true;
+ default:
+ return false;
+ }
+}
+
+inline bool isUnitType(dwarf::Tag T) {
+ switch (T) {
+ case DW_TAG_compile_unit:
+ case DW_TAG_type_unit:
+ case DW_TAG_partial_unit:
+ case DW_TAG_skeleton_unit:
+ return true;
+ default:
+ return false;
+ }
+}
+
+// Constants for the DWARF v5 Accelerator Table Proposal
+enum AcceleratorTable {
+ // Data layout descriptors.
+ DW_ATOM_null = 0u, /// Marker as the end of a list of atoms.
+ DW_ATOM_die_offset = 1u, // DIE offset in the debug_info section.
+ DW_ATOM_cu_offset = 2u, // Offset of the compile unit header that contains the
+ // item in question.
+ DW_ATOM_die_tag = 3u, // A tag entry.
+ DW_ATOM_type_flags = 4u, // Set of flags for a type.
+
+ DW_ATOM_type_type_flags = 5u, // Dsymutil type extension.
+ DW_ATOM_qual_name_hash = 6u, // Dsymutil qualified hash extension.
+
+ // DW_ATOM_type_flags values.
+
+ // Always set for C++, only set for ObjC if this is the @implementation for a
+ // class.
+ DW_FLAG_type_implementation = 2u,
+
+ // Hash functions.
+
+ // Daniel J. Bernstein hash.
+ DW_hash_function_djb = 0u
+};
+
+// Constants for the GNU pubnames/pubtypes extensions supporting gdb index.
+enum GDBIndexEntryKind {
+ GIEK_NONE,
+ GIEK_TYPE,
+ GIEK_VARIABLE,
+ GIEK_FUNCTION,
+ GIEK_OTHER,
+ GIEK_UNUSED5,
+ GIEK_UNUSED6,
+ GIEK_UNUSED7
+};
+
+enum GDBIndexEntryLinkage { GIEL_EXTERNAL, GIEL_STATIC };
+
+/// \defgroup DwarfConstantsDumping Dwarf constants dumping functions
+///
+/// All these functions map their argument's value back to the
+/// corresponding enumerator name or return an empty StringRef if the value
+/// isn't known.
+///
+/// @{
+StringRef TagString(unsigned Tag);
+StringRef ChildrenString(unsigned Children);
+StringRef AttributeString(unsigned Attribute);
+StringRef FormEncodingString(unsigned Encoding);
+StringRef OperationEncodingString(unsigned Encoding);
+StringRef AttributeEncodingString(unsigned Encoding);
+StringRef DecimalSignString(unsigned Sign);
+StringRef EndianityString(unsigned Endian);
+StringRef AccessibilityString(unsigned Access);
+StringRef DefaultedMemberString(unsigned DefaultedEncodings);
+StringRef VisibilityString(unsigned Visibility);
+StringRef VirtualityString(unsigned Virtuality);
+StringRef LanguageString(unsigned Language);
+StringRef CaseString(unsigned Case);
+StringRef ConventionString(unsigned Convention);
+StringRef InlineCodeString(unsigned Code);
+StringRef ArrayOrderString(unsigned Order);
+StringRef LNStandardString(unsigned Standard);
+StringRef LNExtendedString(unsigned Encoding);
+StringRef MacinfoString(unsigned Encoding);
+StringRef RangeListEncodingString(unsigned Encoding);
+StringRef LocListEncodingString(unsigned Encoding);
+StringRef CallFrameString(unsigned Encoding, Triple::ArchType Arch);
+StringRef ApplePropertyString(unsigned);
+StringRef UnitTypeString(unsigned);
+StringRef AtomTypeString(unsigned Atom);
+StringRef GDBIndexEntryKindString(GDBIndexEntryKind Kind);
+StringRef GDBIndexEntryLinkageString(GDBIndexEntryLinkage Linkage);
+StringRef IndexString(unsigned Idx);
+/// @}
+
+/// \defgroup DwarfConstantsParsing Dwarf constants parsing functions
+///
+/// These functions map their strings back to the corresponding enumeration
+/// value or return 0 if there is none, except for these exceptions:
+///
+/// \li \a getTag() returns \a DW_TAG_invalid on invalid input.
+/// \li \a getVirtuality() returns \a DW_VIRTUALITY_invalid on invalid input.
+/// \li \a getMacinfo() returns \a DW_MACINFO_invalid on invalid input.
+///
+/// @{
+unsigned getTag(StringRef TagString);
+unsigned getOperationEncoding(StringRef OperationEncodingString);
+unsigned getVirtuality(StringRef VirtualityString);
+unsigned getLanguage(StringRef LanguageString);
+unsigned getCallingConvention(StringRef LanguageString);
+unsigned getAttributeEncoding(StringRef EncodingString);
+unsigned getMacinfo(StringRef MacinfoString);
+/// @}
+
+/// \defgroup DwarfConstantsVersioning Dwarf version for constants
+///
+/// For constants defined by DWARF, returns the DWARF version when the constant
+/// was first defined. For vendor extensions, if there is a version-related
+/// policy for when to emit it, returns a version number for that policy.
+/// Otherwise returns 0.
+///
+/// @{
+unsigned TagVersion(Tag T);
+unsigned AttributeVersion(Attribute A);
+unsigned FormVersion(Form F);
+unsigned OperationVersion(LocationAtom O);
+unsigned AttributeEncodingVersion(TypeKind E);
+unsigned LanguageVersion(SourceLanguage L);
+/// @}
+
+/// \defgroup DwarfConstantsVendor Dwarf "vendor" for constants
+///
+/// These functions return an identifier describing "who" defined the constant,
+/// either the DWARF standard itself or the vendor who defined the extension.
+///
+/// @{
+unsigned TagVendor(Tag T);
+unsigned AttributeVendor(Attribute A);
+unsigned FormVendor(Form F);
+unsigned OperationVendor(LocationAtom O);
+unsigned AttributeEncodingVendor(TypeKind E);
+unsigned LanguageVendor(SourceLanguage L);
+/// @}
+
+Optional<unsigned> LanguageLowerBound(SourceLanguage L);
+
+/// A helper struct providing information about the byte size of DW_FORM
+/// values that vary in size depending on the DWARF version, address byte
+/// size, or DWARF32/DWARF64.
+struct FormParams {
+ uint16_t Version;
+ uint8_t AddrSize;
+ DwarfFormat Format;
+
+ /// The definition of the size of form DW_FORM_ref_addr depends on the
+ /// version. In DWARF v2 it's the size of an address; after that, it's the
+ /// size of a reference.
+ uint8_t getRefAddrByteSize() const {
+ if (Version == 2)
+ return AddrSize;
+ return getDwarfOffsetByteSize();
+ }
+
+ /// The size of a reference is determined by the DWARF 32/64-bit format.
+ uint8_t getDwarfOffsetByteSize() const {
+ switch (Format) {
+ case DwarfFormat::DWARF32:
+ return 4;
+ case DwarfFormat::DWARF64:
+ return 8;
+ }
+ llvm_unreachable("Invalid Format value");
+ }
+
+ explicit operator bool() const { return Version && AddrSize; }
+};
+
+/// Get the byte size of the unit length field depending on the DWARF format.
+inline uint8_t getUnitLengthFieldByteSize(DwarfFormat Format) {
+ switch (Format) {
+ case DwarfFormat::DWARF32:
+ return 4;
+ case DwarfFormat::DWARF64:
+ return 12;
+ }
+ llvm_unreachable("Invalid Format value");
+}
+
+/// Get the fixed byte size for a given form.
+///
+/// If the form has a fixed byte size, then an Optional with a value will be
+/// returned. If the form is always encoded using a variable length storage
+/// format (ULEB or SLEB numbers or blocks) then None will be returned.
+///
+/// \param Form DWARF form to get the fixed byte size for.
+/// \param Params DWARF parameters to help interpret forms.
+/// \returns Optional<uint8_t> value with the fixed byte size or None if
+/// \p Form doesn't have a fixed byte size.
+Optional<uint8_t> getFixedFormByteSize(dwarf::Form Form, FormParams Params);
+
+/// Tells whether the specified form is defined in the specified version,
+/// or is an extension if extensions are allowed.
+bool isValidFormForVersion(Form F, unsigned Version, bool ExtensionsOk = true);
+
+/// Returns the symbolic string representing Val when used as a value
+/// for attribute Attr.
+StringRef AttributeValueString(uint16_t Attr, unsigned Val);
+
+/// Returns the symbolic string representing Val when used as a value
+/// for atom Atom.
+StringRef AtomValueString(uint16_t Atom, unsigned Val);
+
+/// Describes an entry of the various gnu_pub* debug sections.
+///
+/// The gnu_pub* kind looks like:
+///
+/// 0-3 reserved
+/// 4-6 symbol kind
+/// 7 0 == global, 1 == static
+///
+/// A gdb_index descriptor includes the above kind, shifted 24 bits up with the
+/// offset of the cu within the debug_info section stored in those 24 bits.
+struct PubIndexEntryDescriptor {
+ GDBIndexEntryKind Kind;
+ GDBIndexEntryLinkage Linkage;
+ PubIndexEntryDescriptor(GDBIndexEntryKind Kind, GDBIndexEntryLinkage Linkage)
+ : Kind(Kind), Linkage(Linkage) {}
+ /* implicit */ PubIndexEntryDescriptor(GDBIndexEntryKind Kind)
+ : Kind(Kind), Linkage(GIEL_EXTERNAL) {}
+ explicit PubIndexEntryDescriptor(uint8_t Value)
+ : Kind(
+ static_cast<GDBIndexEntryKind>((Value & KIND_MASK) >> KIND_OFFSET)),
+ Linkage(static_cast<GDBIndexEntryLinkage>((Value & LINKAGE_MASK) >>
+ LINKAGE_OFFSET)) {}
+ uint8_t toBits() const {
+ return Kind << KIND_OFFSET | Linkage << LINKAGE_OFFSET;
+ }
+
+private:
+ enum {
+ KIND_OFFSET = 4,
+ KIND_MASK = 7 << KIND_OFFSET,
+ LINKAGE_OFFSET = 7,
+ LINKAGE_MASK = 1 << LINKAGE_OFFSET
+ };
+};
+
+template <typename Enum> struct EnumTraits : public std::false_type {};
+
+template <> struct EnumTraits<Attribute> : public std::true_type {
+ static constexpr char Type[3] = "AT";
+ static constexpr StringRef (*StringFn)(unsigned) = &AttributeString;
+};
+
+template <> struct EnumTraits<Form> : public std::true_type {
+ static constexpr char Type[5] = "FORM";
+ static constexpr StringRef (*StringFn)(unsigned) = &FormEncodingString;
+};
+
+template <> struct EnumTraits<Index> : public std::true_type {
+ static constexpr char Type[4] = "IDX";
+ static constexpr StringRef (*StringFn)(unsigned) = &IndexString;
+};
+
+template <> struct EnumTraits<Tag> : public std::true_type {
+ static constexpr char Type[4] = "TAG";
+ static constexpr StringRef (*StringFn)(unsigned) = &TagString;
+};
+} // End of namespace dwarf
+
+/// Dwarf constants format_provider
+///
+/// Specialization of the format_provider template for dwarf enums. Unlike the
+/// dumping functions above, these format unknown enumerator values as
+/// DW_TYPE_unknown_1234 (e.g. DW_TAG_unknown_ffff).
+template <typename Enum>
+struct format_provider<
+ Enum, typename std::enable_if<dwarf::EnumTraits<Enum>::value>::type> {
+ static void format(const Enum &E, raw_ostream &OS, StringRef Style) {
+ StringRef Str = dwarf::EnumTraits<Enum>::StringFn(E);
+ if (Str.empty()) {
+ OS << "DW_" << dwarf::EnumTraits<Enum>::Type << "_unknown_"
+ << llvm::format("%x", E);
+ } else
+ OS << Str;
+ }
+};
+} // End of namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/DynamicTags.def b/third_party/llvm-project/include/llvm/BinaryFormat/DynamicTags.def
new file mode 100644
index 000000000..aec408bd2
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/DynamicTags.def
@@ -0,0 +1,244 @@
+#ifndef DYNAMIC_TAG
+#error "DYNAMIC_TAG must be defined"
+#endif
+
+// Add separate macros for the architecture specific tags and the markers
+// such as DT_HIOS, etc. to allow using this file to in other contexts.
+// For example we can use it to generate a stringification switch statement.
+
+#ifndef AARCH64_DYNAMIC_TAG
+#define AARCH64_DYNAMIC_TAG(name, value) DYNAMIC_TAG(name, value)
+#define AARCH64_DYNAMIC_TAG_DEFINED
+#endif
+
+#ifndef HEXAGON_DYNAMIC_TAG
+#define HEXAGON_DYNAMIC_TAG(name, value) DYNAMIC_TAG(name, value)
+#define HEXAGON_DYNAMIC_TAG_DEFINED
+#endif
+
+#ifndef MIPS_DYNAMIC_TAG
+#define MIPS_DYNAMIC_TAG(name, value) DYNAMIC_TAG(name, value)
+#define MIPS_DYNAMIC_TAG_DEFINED
+#endif
+
+#ifndef PPC_DYNAMIC_TAG
+#define PPC_DYNAMIC_TAG(name, value) DYNAMIC_TAG(name, value)
+#define PPC_DYNAMIC_TAG_DEFINED
+#endif
+
+#ifndef PPC64_DYNAMIC_TAG
+#define PPC64_DYNAMIC_TAG(name, value) DYNAMIC_TAG(name, value)
+#define PPC64_DYNAMIC_TAG_DEFINED
+#endif
+
+#ifndef DYNAMIC_TAG_MARKER
+#define DYNAMIC_TAG_MARKER(name, value) DYNAMIC_TAG(name, value)
+#define DYNAMIC_TAG_MARKER_DEFINED
+#endif
+
+DYNAMIC_TAG(NULL, 0) // Marks end of dynamic array.
+DYNAMIC_TAG(NEEDED, 1) // String table offset of needed library.
+DYNAMIC_TAG(PLTRELSZ, 2) // Size of relocation entries in PLT.
+DYNAMIC_TAG(PLTGOT, 3) // Address associated with linkage table.
+DYNAMIC_TAG(HASH, 4) // Address of symbolic hash table.
+DYNAMIC_TAG(STRTAB, 5) // Address of dynamic string table.
+DYNAMIC_TAG(SYMTAB, 6) // Address of dynamic symbol table.
+DYNAMIC_TAG(RELA, 7) // Address of relocation table (Rela entries).
+DYNAMIC_TAG(RELASZ, 8) // Size of Rela relocation table.
+DYNAMIC_TAG(RELAENT, 9) // Size of a Rela relocation entry.
+DYNAMIC_TAG(STRSZ, 10) // Total size of the string table.
+DYNAMIC_TAG(SYMENT, 11) // Size of a symbol table entry.
+DYNAMIC_TAG(INIT, 12) // Address of initialization function.
+DYNAMIC_TAG(FINI, 13) // Address of termination function.
+DYNAMIC_TAG(SONAME, 14) // String table offset of a shared objects name.
+DYNAMIC_TAG(RPATH, 15) // String table offset of library search path.
+DYNAMIC_TAG(SYMBOLIC, 16) // Changes symbol resolution algorithm.
+DYNAMIC_TAG(REL, 17) // Address of relocation table (Rel entries).
+DYNAMIC_TAG(RELSZ, 18) // Size of Rel relocation table.
+DYNAMIC_TAG(RELENT, 19) // Size of a Rel relocation entry.
+DYNAMIC_TAG(PLTREL, 20) // Type of relocation entry used for linking.
+DYNAMIC_TAG(DEBUG, 21) // Reserved for debugger.
+DYNAMIC_TAG(TEXTREL, 22) // Relocations exist for non-writable segments.
+DYNAMIC_TAG(JMPREL, 23) // Address of relocations associated with PLT.
+DYNAMIC_TAG(BIND_NOW, 24) // Process all relocations before execution.
+DYNAMIC_TAG(INIT_ARRAY, 25) // Pointer to array of initialization functions.
+DYNAMIC_TAG(FINI_ARRAY, 26) // Pointer to array of termination functions.
+DYNAMIC_TAG(INIT_ARRAYSZ, 27) // Size of DT_INIT_ARRAY.
+DYNAMIC_TAG(FINI_ARRAYSZ, 28) // Size of DT_FINI_ARRAY.
+DYNAMIC_TAG(RUNPATH, 29) // String table offset of lib search path.
+DYNAMIC_TAG(FLAGS, 30) // Flags.
+DYNAMIC_TAG_MARKER(ENCODING, 32) // Values from here to DT_LOOS follow the rules
+ // for the interpretation of the d_un union.
+
+DYNAMIC_TAG(PREINIT_ARRAY, 32) // Pointer to array of preinit functions.
+DYNAMIC_TAG(PREINIT_ARRAYSZ, 33) // Size of the DT_PREINIT_ARRAY array.
+
+DYNAMIC_TAG(SYMTAB_SHNDX, 34) // Address of the SHT_SYMTAB_SHNDX section.
+
+// Experimental support for SHT_RELR sections. For details, see proposal
+// at https://groups.google.com/forum/#!topic/generic-abi/bX460iggiKg
+DYNAMIC_TAG(RELRSZ, 35) // Size of Relr relocation table.
+DYNAMIC_TAG(RELR, 36) // Address of relocation table (Relr entries).
+DYNAMIC_TAG(RELRENT, 37) // Size of a Relr relocation entry.
+
+DYNAMIC_TAG_MARKER(LOOS, 0x60000000) // Start of environment specific tags.
+DYNAMIC_TAG_MARKER(HIOS, 0x6FFFFFFF) // End of environment specific tags.
+DYNAMIC_TAG_MARKER(LOPROC, 0x70000000) // Start of processor specific tags.
+DYNAMIC_TAG_MARKER(HIPROC, 0x7FFFFFFF) // End of processor specific tags.
+
+// Android packed relocation section tags.
+// https://android.googlesource.com/platform/bionic/+/6f12bfece5dcc01325e0abba56a46b1bcf991c69/tools/relocation_packer/src/elf_file.cc#31
+DYNAMIC_TAG(ANDROID_REL, 0x6000000F)
+DYNAMIC_TAG(ANDROID_RELSZ, 0x60000010)
+DYNAMIC_TAG(ANDROID_RELA, 0x60000011)
+DYNAMIC_TAG(ANDROID_RELASZ, 0x60000012)
+
+// Android's experimental support for SHT_RELR sections.
+// https://android.googlesource.com/platform/bionic/+/b7feec74547f84559a1467aca02708ff61346d2a/libc/include/elf.h#253
+DYNAMIC_TAG(ANDROID_RELR, 0x6FFFE000) // Address of relocation table (Relr entries).
+DYNAMIC_TAG(ANDROID_RELRSZ, 0x6FFFE001) // Size of Relr relocation table.
+DYNAMIC_TAG(ANDROID_RELRENT, 0x6FFFE003) // Size of a Relr relocation entry.
+
+DYNAMIC_TAG(GNU_HASH, 0x6FFFFEF5) // Reference to the GNU hash table.
+DYNAMIC_TAG(TLSDESC_PLT, 0x6FFFFEF6) // Location of PLT entry for TLS
+ // descriptor resolver calls.
+DYNAMIC_TAG(TLSDESC_GOT, 0x6FFFFEF7) // Location of GOT entry used by TLS
+ // descriptor resolver PLT entry.
+DYNAMIC_TAG(RELACOUNT, 0x6FFFFFF9) // ELF32_Rela count.
+DYNAMIC_TAG(RELCOUNT, 0x6FFFFFFA) // ELF32_Rel count.
+
+DYNAMIC_TAG(FLAGS_1, 0X6FFFFFFB) // Flags_1.
+
+DYNAMIC_TAG(VERSYM, 0x6FFFFFF0) // The address of .gnu.version section.
+DYNAMIC_TAG(VERDEF, 0X6FFFFFFC) // The address of the version definition
+ // table.
+DYNAMIC_TAG(VERDEFNUM, 0X6FFFFFFD) // The number of entries in DT_VERDEF.
+DYNAMIC_TAG(VERNEED, 0X6FFFFFFE) // The address of the version dependency
+ // table.
+DYNAMIC_TAG(VERNEEDNUM, 0X6FFFFFFF) // The number of entries in DT_VERNEED.
+
+// AArch64 specific dynamic table entries
+AARCH64_DYNAMIC_TAG(AARCH64_BTI_PLT, 0x70000001)
+AARCH64_DYNAMIC_TAG(AARCH64_PAC_PLT, 0x70000003)
+
+// Hexagon specific dynamic table entries
+HEXAGON_DYNAMIC_TAG(HEXAGON_SYMSZ, 0x70000000)
+HEXAGON_DYNAMIC_TAG(HEXAGON_VER, 0x70000001)
+HEXAGON_DYNAMIC_TAG(HEXAGON_PLT, 0x70000002)
+
+// Mips specific dynamic table entry tags.
+
+MIPS_DYNAMIC_TAG(MIPS_RLD_VERSION, 0x70000001) // 32 bit version number for
+ // runtime linker interface.
+MIPS_DYNAMIC_TAG(MIPS_TIME_STAMP, 0x70000002) // Time stamp.
+MIPS_DYNAMIC_TAG(MIPS_ICHECKSUM, 0x70000003) // Checksum of external strings
+ // and common sizes.
+MIPS_DYNAMIC_TAG(MIPS_IVERSION, 0x70000004) // Index of version string
+ // in string table.
+MIPS_DYNAMIC_TAG(MIPS_FLAGS, 0x70000005) // 32 bits of flags.
+MIPS_DYNAMIC_TAG(MIPS_BASE_ADDRESS, 0x70000006) // Base address of the segment.
+MIPS_DYNAMIC_TAG(MIPS_MSYM, 0x70000007) // Address of .msym section.
+MIPS_DYNAMIC_TAG(MIPS_CONFLICT, 0x70000008) // Address of .conflict section.
+MIPS_DYNAMIC_TAG(MIPS_LIBLIST, 0x70000009) // Address of .liblist section.
+MIPS_DYNAMIC_TAG(MIPS_LOCAL_GOTNO, 0x7000000a) // Number of local global offset
+ // table entries.
+MIPS_DYNAMIC_TAG(MIPS_CONFLICTNO, 0x7000000b) // Number of entries
+ // in the .conflict section.
+MIPS_DYNAMIC_TAG(MIPS_LIBLISTNO, 0x70000010) // Number of entries
+ // in the .liblist section.
+MIPS_DYNAMIC_TAG(MIPS_SYMTABNO, 0x70000011) // Number of entries
+ // in the .dynsym section.
+MIPS_DYNAMIC_TAG(MIPS_UNREFEXTNO, 0x70000012) // Index of first external dynamic
+ // symbol not referenced locally.
+MIPS_DYNAMIC_TAG(MIPS_GOTSYM, 0x70000013) // Index of first dynamic symbol
+ // in global offset table.
+MIPS_DYNAMIC_TAG(MIPS_HIPAGENO, 0x70000014) // Number of page table entries
+ // in global offset table.
+MIPS_DYNAMIC_TAG(MIPS_RLD_MAP, 0x70000016) // Address of run time loader map
+ // used for debugging.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_CLASS, 0x70000017) // Delta C++ class definition.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_CLASS_NO, 0x70000018) // Number of entries
+ // in DT_MIPS_DELTA_CLASS.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_INSTANCE, 0x70000019) // Delta C++ class instances.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_INSTANCE_NO, 0x7000001A) // Number of entries
+ // in DT_MIPS_DELTA_INSTANCE.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_RELOC, 0x7000001B) // Delta relocations.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_RELOC_NO, 0x7000001C) // Number of entries
+ // in DT_MIPS_DELTA_RELOC.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_SYM, 0x7000001D) // Delta symbols that Delta
+ // relocations refer to.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_SYM_NO, 0x7000001E) // Number of entries
+ // in DT_MIPS_DELTA_SYM.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_CLASSSYM, 0x70000020) // Delta symbols that hold
+ // class declarations.
+MIPS_DYNAMIC_TAG(MIPS_DELTA_CLASSSYM_NO, 0x70000021) // Number of entries
+ // in DT_MIPS_DELTA_CLASSSYM.
+
+MIPS_DYNAMIC_TAG(MIPS_CXX_FLAGS, 0x70000022) // Flags indicating information
+ // about C++ flavor.
+MIPS_DYNAMIC_TAG(MIPS_PIXIE_INIT, 0x70000023) // Pixie information.
+MIPS_DYNAMIC_TAG(MIPS_SYMBOL_LIB, 0x70000024) // Address of .MIPS.symlib
+MIPS_DYNAMIC_TAG(MIPS_LOCALPAGE_GOTIDX, 0x70000025) // The GOT index of the first PTE
+ // for a segment
+MIPS_DYNAMIC_TAG(MIPS_LOCAL_GOTIDX, 0x70000026) // The GOT index of the first PTE
+ // for a local symbol
+MIPS_DYNAMIC_TAG(MIPS_HIDDEN_GOTIDX, 0x70000027) // The GOT index of the first PTE
+ // for a hidden symbol
+MIPS_DYNAMIC_TAG(MIPS_PROTECTED_GOTIDX, 0x70000028) // The GOT index of the first PTE
+ // for a protected symbol
+MIPS_DYNAMIC_TAG(MIPS_OPTIONS, 0x70000029) // Address of `.MIPS.options'.
+MIPS_DYNAMIC_TAG(MIPS_INTERFACE, 0x7000002A) // Address of `.interface'.
+MIPS_DYNAMIC_TAG(MIPS_DYNSTR_ALIGN, 0x7000002B) // Unknown.
+MIPS_DYNAMIC_TAG(MIPS_INTERFACE_SIZE, 0x7000002C) // Size of the .interface section.
+MIPS_DYNAMIC_TAG(MIPS_RLD_TEXT_RESOLVE_ADDR, 0x7000002D) // Size of rld_text_resolve
+ // function stored in the GOT.
+MIPS_DYNAMIC_TAG(MIPS_PERF_SUFFIX, 0x7000002E) // Default suffix of DSO to be added
+ // by rld on dlopen() calls.
+MIPS_DYNAMIC_TAG(MIPS_COMPACT_SIZE, 0x7000002F) // Size of compact relocation
+ // section (O32).
+MIPS_DYNAMIC_TAG(MIPS_GP_VALUE, 0x70000030) // GP value for auxiliary GOTs.
+MIPS_DYNAMIC_TAG(MIPS_AUX_DYNAMIC, 0x70000031) // Address of auxiliary .dynamic.
+MIPS_DYNAMIC_TAG(MIPS_PLTGOT, 0x70000032) // Address of the base of the PLTGOT.
+MIPS_DYNAMIC_TAG(MIPS_RWPLT, 0x70000034) // Points to the base
+ // of a writable PLT.
+MIPS_DYNAMIC_TAG(MIPS_RLD_MAP_REL, 0x70000035) // Relative offset of run time loader
+ // map, used for debugging.
+
+// PPC specific dynamic table entries.
+PPC_DYNAMIC_TAG(PPC_GOT, 0x70000000) // Uses Secure PLT ABI.
+PPC_DYNAMIC_TAG(PPC_OPT, 0x70000001) // Has TLS optimization.
+
+// PPC64 specific dynamic table entries.
+PPC64_DYNAMIC_TAG(PPC64_GLINK, 0x70000000) // Address of 32 bytes before the
+ // first glink lazy resolver stub.
+
+// Sun machine-independent extensions.
+DYNAMIC_TAG(AUXILIARY, 0x7FFFFFFD) // Shared object to load before self
+DYNAMIC_TAG(USED, 0x7FFFFFFE) // Same as DT_NEEDED
+DYNAMIC_TAG(FILTER, 0x7FFFFFFF) // Shared object to get values from
+
+
+#ifdef DYNAMIC_TAG_MARKER_DEFINED
+#undef DYNAMIC_TAG_MARKER
+#undef DYNAMIC_TAG_MARKER_DEFINED
+#endif
+#ifdef AARCH64_DYNAMIC_TAG_DEFINED
+#undef AARCH64_DYNAMIC_TAG
+#undef AARCH64_DYNAMIC_TAG_DEFINED
+#endif
+#ifdef MIPS_DYNAMIC_TAG_DEFINED
+#undef MIPS_DYNAMIC_TAG
+#undef MIPS_DYNAMIC_TAG_DEFINED
+#endif
+#ifdef HEXAGON_DYNAMIC_TAG_DEFINED
+#undef HEXAGON_DYNAMIC_TAG
+#undef HEXAGON_DYNAMIC_TAG_DEFINED
+#endif
+#ifdef PPC_DYNAMIC_TAG_DEFINED
+#undef PPC_DYNAMIC_TAG
+#undef PPC_DYNAMIC_TAG_DEFINED
+#endif
+#ifdef PPC64_DYNAMIC_TAG_DEFINED
+#undef PPC64_DYNAMIC_TAG
+#undef PPC64_DYNAMIC_TAG_DEFINED
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELF.h b/third_party/llvm-project/include/llvm/BinaryFormat/ELF.h
new file mode 100644
index 000000000..46edfb626
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELF.h
@@ -0,0 +1,1573 @@
+//===- llvm/BinaryFormat/ELF.h - ELF constants and structures ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This header contains common, non-processor-specific data structures and
+// constants for the ELF file format.
+//
+// The details of the ELF32 bits in this file are largely based on the Tool
+// Interface Standard (TIS) Executable and Linking Format (ELF) Specification
+// Version 1.2, May 1995. The ELF64 stuff is based on ELF-64 Object File Format
+// Version 1.5, Draft 2, May 1998 as well as OpenBSD header files.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_ELF_H
+#define LLVM_BINARYFORMAT_ELF_H
+
+#include <cstdint>
+#include <cstring>
+
+namespace llvm {
+namespace ELF {
+
+using Elf32_Addr = uint32_t; // Program address
+using Elf32_Off = uint32_t; // File offset
+using Elf32_Half = uint16_t;
+using Elf32_Word = uint32_t;
+using Elf32_Sword = int32_t;
+
+using Elf64_Addr = uint64_t;
+using Elf64_Off = uint64_t;
+using Elf64_Half = uint16_t;
+using Elf64_Word = uint32_t;
+using Elf64_Sword = int32_t;
+using Elf64_Xword = uint64_t;
+using Elf64_Sxword = int64_t;
+
+// Object file magic string.
+static const char ElfMagic[] = {0x7f, 'E', 'L', 'F', '\0'};
+
+// e_ident size and indices.
+enum {
+ EI_MAG0 = 0, // File identification index.
+ EI_MAG1 = 1, // File identification index.
+ EI_MAG2 = 2, // File identification index.
+ EI_MAG3 = 3, // File identification index.
+ EI_CLASS = 4, // File class.
+ EI_DATA = 5, // Data encoding.
+ EI_VERSION = 6, // File version.
+ EI_OSABI = 7, // OS/ABI identification.
+ EI_ABIVERSION = 8, // ABI version.
+ EI_PAD = 9, // Start of padding bytes.
+ EI_NIDENT = 16 // Number of bytes in e_ident.
+};
+
+struct Elf32_Ehdr {
+ unsigned char e_ident[EI_NIDENT]; // ELF Identification bytes
+ Elf32_Half e_type; // Type of file (see ET_* below)
+ Elf32_Half e_machine; // Required architecture for this file (see EM_*)
+ Elf32_Word e_version; // Must be equal to 1
+ Elf32_Addr e_entry; // Address to jump to in order to start program
+ Elf32_Off e_phoff; // Program header table's file offset, in bytes
+ Elf32_Off e_shoff; // Section header table's file offset, in bytes
+ Elf32_Word e_flags; // Processor-specific flags
+ Elf32_Half e_ehsize; // Size of ELF header, in bytes
+ Elf32_Half e_phentsize; // Size of an entry in the program header table
+ Elf32_Half e_phnum; // Number of entries in the program header table
+ Elf32_Half e_shentsize; // Size of an entry in the section header table
+ Elf32_Half e_shnum; // Number of entries in the section header table
+ Elf32_Half e_shstrndx; // Sect hdr table index of sect name string table
+
+ bool checkMagic() const {
+ return (memcmp(e_ident, ElfMagic, strlen(ElfMagic))) == 0;
+ }
+
+ unsigned char getFileClass() const { return e_ident[EI_CLASS]; }
+ unsigned char getDataEncoding() const { return e_ident[EI_DATA]; }
+};
+
+// 64-bit ELF header. Fields are the same as for ELF32, but with different
+// types (see above).
+struct Elf64_Ehdr {
+ unsigned char e_ident[EI_NIDENT];
+ Elf64_Half e_type;
+ Elf64_Half e_machine;
+ Elf64_Word e_version;
+ Elf64_Addr e_entry;
+ Elf64_Off e_phoff;
+ Elf64_Off e_shoff;
+ Elf64_Word e_flags;
+ Elf64_Half e_ehsize;
+ Elf64_Half e_phentsize;
+ Elf64_Half e_phnum;
+ Elf64_Half e_shentsize;
+ Elf64_Half e_shnum;
+ Elf64_Half e_shstrndx;
+
+ bool checkMagic() const {
+ return (memcmp(e_ident, ElfMagic, strlen(ElfMagic))) == 0;
+ }
+
+ unsigned char getFileClass() const { return e_ident[EI_CLASS]; }
+ unsigned char getDataEncoding() const { return e_ident[EI_DATA]; }
+};
+
+// File types
+enum {
+ ET_NONE = 0, // No file type
+ ET_REL = 1, // Relocatable file
+ ET_EXEC = 2, // Executable file
+ ET_DYN = 3, // Shared object file
+ ET_CORE = 4, // Core file
+ ET_LOPROC = 0xff00, // Beginning of processor-specific codes
+ ET_HIPROC = 0xffff // Processor-specific
+};
+
+// Versioning
+enum { EV_NONE = 0, EV_CURRENT = 1 };
+
+// Machine architectures
+// See current registered ELF machine architectures at:
+// http://www.uxsglobal.com/developers/gabi/latest/ch4.eheader.html
+enum {
+ EM_NONE = 0, // No machine
+ EM_M32 = 1, // AT&T WE 32100
+ EM_SPARC = 2, // SPARC
+ EM_386 = 3, // Intel 386
+ EM_68K = 4, // Motorola 68000
+ EM_88K = 5, // Motorola 88000
+ EM_IAMCU = 6, // Intel MCU
+ EM_860 = 7, // Intel 80860
+ EM_MIPS = 8, // MIPS R3000
+ EM_S370 = 9, // IBM System/370
+ EM_MIPS_RS3_LE = 10, // MIPS RS3000 Little-endian
+ EM_PARISC = 15, // Hewlett-Packard PA-RISC
+ EM_VPP500 = 17, // Fujitsu VPP500
+ EM_SPARC32PLUS = 18, // Enhanced instruction set SPARC
+ EM_960 = 19, // Intel 80960
+ EM_PPC = 20, // PowerPC
+ EM_PPC64 = 21, // PowerPC64
+ EM_S390 = 22, // IBM System/390
+ EM_SPU = 23, // IBM SPU/SPC
+ EM_V800 = 36, // NEC V800
+ EM_FR20 = 37, // Fujitsu FR20
+ EM_RH32 = 38, // TRW RH-32
+ EM_RCE = 39, // Motorola RCE
+ EM_ARM = 40, // ARM
+ EM_ALPHA = 41, // DEC Alpha
+ EM_SH = 42, // Hitachi SH
+ EM_SPARCV9 = 43, // SPARC V9
+ EM_TRICORE = 44, // Siemens TriCore
+ EM_ARC = 45, // Argonaut RISC Core
+ EM_H8_300 = 46, // Hitachi H8/300
+ EM_H8_300H = 47, // Hitachi H8/300H
+ EM_H8S = 48, // Hitachi H8S
+ EM_H8_500 = 49, // Hitachi H8/500
+ EM_IA_64 = 50, // Intel IA-64 processor architecture
+ EM_MIPS_X = 51, // Stanford MIPS-X
+ EM_COLDFIRE = 52, // Motorola ColdFire
+ EM_68HC12 = 53, // Motorola M68HC12
+ EM_MMA = 54, // Fujitsu MMA Multimedia Accelerator
+ EM_PCP = 55, // Siemens PCP
+ EM_NCPU = 56, // Sony nCPU embedded RISC processor
+ EM_NDR1 = 57, // Denso NDR1 microprocessor
+ EM_STARCORE = 58, // Motorola Star*Core processor
+ EM_ME16 = 59, // Toyota ME16 processor
+ EM_ST100 = 60, // STMicroelectronics ST100 processor
+ EM_TINYJ = 61, // Advanced Logic Corp. TinyJ embedded processor family
+ EM_X86_64 = 62, // AMD x86-64 architecture
+ EM_PDSP = 63, // Sony DSP Processor
+ EM_PDP10 = 64, // Digital Equipment Corp. PDP-10
+ EM_PDP11 = 65, // Digital Equipment Corp. PDP-11
+ EM_FX66 = 66, // Siemens FX66 microcontroller
+ EM_ST9PLUS = 67, // STMicroelectronics ST9+ 8/16 bit microcontroller
+ EM_ST7 = 68, // STMicroelectronics ST7 8-bit microcontroller
+ EM_68HC16 = 69, // Motorola MC68HC16 Microcontroller
+ EM_68HC11 = 70, // Motorola MC68HC11 Microcontroller
+ EM_68HC08 = 71, // Motorola MC68HC08 Microcontroller
+ EM_68HC05 = 72, // Motorola MC68HC05 Microcontroller
+ EM_SVX = 73, // Silicon Graphics SVx
+ EM_ST19 = 74, // STMicroelectronics ST19 8-bit microcontroller
+ EM_VAX = 75, // Digital VAX
+ EM_CRIS = 76, // Axis Communications 32-bit embedded processor
+ EM_JAVELIN = 77, // Infineon Technologies 32-bit embedded processor
+ EM_FIREPATH = 78, // Element 14 64-bit DSP Processor
+ EM_ZSP = 79, // LSI Logic 16-bit DSP Processor
+ EM_MMIX = 80, // Donald Knuth's educational 64-bit processor
+ EM_HUANY = 81, // Harvard University machine-independent object files
+ EM_PRISM = 82, // SiTera Prism
+ EM_AVR = 83, // Atmel AVR 8-bit microcontroller
+ EM_FR30 = 84, // Fujitsu FR30
+ EM_D10V = 85, // Mitsubishi D10V
+ EM_D30V = 86, // Mitsubishi D30V
+ EM_V850 = 87, // NEC v850
+ EM_M32R = 88, // Mitsubishi M32R
+ EM_MN10300 = 89, // Matsushita MN10300
+ EM_MN10200 = 90, // Matsushita MN10200
+ EM_PJ = 91, // picoJava
+ EM_OPENRISC = 92, // OpenRISC 32-bit embedded processor
+ EM_ARC_COMPACT = 93, // ARC International ARCompact processor (old
+ // spelling/synonym: EM_ARC_A5)
+ EM_XTENSA = 94, // Tensilica Xtensa Architecture
+ EM_VIDEOCORE = 95, // Alphamosaic VideoCore processor
+ EM_TMM_GPP = 96, // Thompson Multimedia General Purpose Processor
+ EM_NS32K = 97, // National Semiconductor 32000 series
+ EM_TPC = 98, // Tenor Network TPC processor
+ EM_SNP1K = 99, // Trebia SNP 1000 processor
+ EM_ST200 = 100, // STMicroelectronics (www.st.com) ST200
+ EM_IP2K = 101, // Ubicom IP2xxx microcontroller family
+ EM_MAX = 102, // MAX Processor
+ EM_CR = 103, // National Semiconductor CompactRISC microprocessor
+ EM_F2MC16 = 104, // Fujitsu F2MC16
+ EM_MSP430 = 105, // Texas Instruments embedded microcontroller msp430
+ EM_BLACKFIN = 106, // Analog Devices Blackfin (DSP) processor
+ EM_SE_C33 = 107, // S1C33 Family of Seiko Epson processors
+ EM_SEP = 108, // Sharp embedded microprocessor
+ EM_ARCA = 109, // Arca RISC Microprocessor
+ EM_UNICORE = 110, // Microprocessor series from PKU-Unity Ltd. and MPRC
+ // of Peking University
+ EM_EXCESS = 111, // eXcess: 16/32/64-bit configurable embedded CPU
+ EM_DXP = 112, // Icera Semiconductor Inc. Deep Execution Processor
+ EM_ALTERA_NIOS2 = 113, // Altera Nios II soft-core processor
+ EM_CRX = 114, // National Semiconductor CompactRISC CRX
+ EM_XGATE = 115, // Motorola XGATE embedded processor
+ EM_C166 = 116, // Infineon C16x/XC16x processor
+ EM_M16C = 117, // Renesas M16C series microprocessors
+ EM_DSPIC30F = 118, // Microchip Technology dsPIC30F Digital Signal
+ // Controller
+ EM_CE = 119, // Freescale Communication Engine RISC core
+ EM_M32C = 120, // Renesas M32C series microprocessors
+ EM_TSK3000 = 131, // Altium TSK3000 core
+ EM_RS08 = 132, // Freescale RS08 embedded processor
+ EM_SHARC = 133, // Analog Devices SHARC family of 32-bit DSP
+ // processors
+ EM_ECOG2 = 134, // Cyan Technology eCOG2 microprocessor
+ EM_SCORE7 = 135, // Sunplus S+core7 RISC processor
+ EM_DSP24 = 136, // New Japan Radio (NJR) 24-bit DSP Processor
+ EM_VIDEOCORE3 = 137, // Broadcom VideoCore III processor
+ EM_LATTICEMICO32 = 138, // RISC processor for Lattice FPGA architecture
+ EM_SE_C17 = 139, // Seiko Epson C17 family
+ EM_TI_C6000 = 140, // The Texas Instruments TMS320C6000 DSP family
+ EM_TI_C2000 = 141, // The Texas Instruments TMS320C2000 DSP family
+ EM_TI_C5500 = 142, // The Texas Instruments TMS320C55x DSP family
+ EM_MMDSP_PLUS = 160, // STMicroelectronics 64bit VLIW Data Signal Processor
+ EM_CYPRESS_M8C = 161, // Cypress M8C microprocessor
+ EM_R32C = 162, // Renesas R32C series microprocessors
+ EM_TRIMEDIA = 163, // NXP Semiconductors TriMedia architecture family
+ EM_HEXAGON = 164, // Qualcomm Hexagon processor
+ EM_8051 = 165, // Intel 8051 and variants
+ EM_STXP7X = 166, // STMicroelectronics STxP7x family of configurable
+ // and extensible RISC processors
+ EM_NDS32 = 167, // Andes Technology compact code size embedded RISC
+ // processor family
+ EM_ECOG1 = 168, // Cyan Technology eCOG1X family
+ EM_ECOG1X = 168, // Cyan Technology eCOG1X family
+ EM_MAXQ30 = 169, // Dallas Semiconductor MAXQ30 Core Micro-controllers
+ EM_XIMO16 = 170, // New Japan Radio (NJR) 16-bit DSP Processor
+ EM_MANIK = 171, // M2000 Reconfigurable RISC Microprocessor
+ EM_CRAYNV2 = 172, // Cray Inc. NV2 vector architecture
+ EM_RX = 173, // Renesas RX family
+ EM_METAG = 174, // Imagination Technologies META processor
+ // architecture
+ EM_MCST_ELBRUS = 175, // MCST Elbrus general purpose hardware architecture
+ EM_ECOG16 = 176, // Cyan Technology eCOG16 family
+ EM_CR16 = 177, // National Semiconductor CompactRISC CR16 16-bit
+ // microprocessor
+ EM_ETPU = 178, // Freescale Extended Time Processing Unit
+ EM_SLE9X = 179, // Infineon Technologies SLE9X core
+ EM_L10M = 180, // Intel L10M
+ EM_K10M = 181, // Intel K10M
+ EM_AARCH64 = 183, // ARM AArch64
+ EM_AVR32 = 185, // Atmel Corporation 32-bit microprocessor family
+ EM_STM8 = 186, // STMicroeletronics STM8 8-bit microcontroller
+ EM_TILE64 = 187, // Tilera TILE64 multicore architecture family
+ EM_TILEPRO = 188, // Tilera TILEPro multicore architecture family
+ EM_CUDA = 190, // NVIDIA CUDA architecture
+ EM_TILEGX = 191, // Tilera TILE-Gx multicore architecture family
+ EM_CLOUDSHIELD = 192, // CloudShield architecture family
+ EM_COREA_1ST = 193, // KIPO-KAIST Core-A 1st generation processor family
+ EM_COREA_2ND = 194, // KIPO-KAIST Core-A 2nd generation processor family
+ EM_ARC_COMPACT2 = 195, // Synopsys ARCompact V2
+ EM_OPEN8 = 196, // Open8 8-bit RISC soft processor core
+ EM_RL78 = 197, // Renesas RL78 family
+ EM_VIDEOCORE5 = 198, // Broadcom VideoCore V processor
+ EM_78KOR = 199, // Renesas 78KOR family
+ EM_56800EX = 200, // Freescale 56800EX Digital Signal Controller (DSC)
+ EM_BA1 = 201, // Beyond BA1 CPU architecture
+ EM_BA2 = 202, // Beyond BA2 CPU architecture
+ EM_XCORE = 203, // XMOS xCORE processor family
+ EM_MCHP_PIC = 204, // Microchip 8-bit PIC(r) family
+ EM_INTEL205 = 205, // Reserved by Intel
+ EM_INTEL206 = 206, // Reserved by Intel
+ EM_INTEL207 = 207, // Reserved by Intel
+ EM_INTEL208 = 208, // Reserved by Intel
+ EM_INTEL209 = 209, // Reserved by Intel
+ EM_KM32 = 210, // KM211 KM32 32-bit processor
+ EM_KMX32 = 211, // KM211 KMX32 32-bit processor
+ EM_KMX16 = 212, // KM211 KMX16 16-bit processor
+ EM_KMX8 = 213, // KM211 KMX8 8-bit processor
+ EM_KVARC = 214, // KM211 KVARC processor
+ EM_CDP = 215, // Paneve CDP architecture family
+ EM_COGE = 216, // Cognitive Smart Memory Processor
+ EM_COOL = 217, // iCelero CoolEngine
+ EM_NORC = 218, // Nanoradio Optimized RISC
+ EM_CSR_KALIMBA = 219, // CSR Kalimba architecture family
+ EM_AMDGPU = 224, // AMD GPU architecture
+ EM_RISCV = 243, // RISC-V
+ EM_LANAI = 244, // Lanai 32-bit processor
+ EM_BPF = 247, // Linux kernel bpf virtual machine
+};
+
+// Object file classes.
+enum {
+ ELFCLASSNONE = 0,
+ ELFCLASS32 = 1, // 32-bit object file
+ ELFCLASS64 = 2 // 64-bit object file
+};
+
+// Object file byte orderings.
+enum {
+ ELFDATANONE = 0, // Invalid data encoding.
+ ELFDATA2LSB = 1, // Little-endian object file
+ ELFDATA2MSB = 2 // Big-endian object file
+};
+
+// OS ABI identification.
+enum {
+ ELFOSABI_NONE = 0, // UNIX System V ABI
+ ELFOSABI_HPUX = 1, // HP-UX operating system
+ ELFOSABI_NETBSD = 2, // NetBSD
+ ELFOSABI_GNU = 3, // GNU/Linux
+ ELFOSABI_LINUX = 3, // Historical alias for ELFOSABI_GNU.
+ ELFOSABI_HURD = 4, // GNU/Hurd
+ ELFOSABI_SOLARIS = 6, // Solaris
+ ELFOSABI_AIX = 7, // AIX
+ ELFOSABI_IRIX = 8, // IRIX
+ ELFOSABI_FREEBSD = 9, // FreeBSD
+ ELFOSABI_TRU64 = 10, // TRU64 UNIX
+ ELFOSABI_MODESTO = 11, // Novell Modesto
+ ELFOSABI_OPENBSD = 12, // OpenBSD
+ ELFOSABI_OPENVMS = 13, // OpenVMS
+ ELFOSABI_NSK = 14, // Hewlett-Packard Non-Stop Kernel
+ ELFOSABI_AROS = 15, // AROS
+ ELFOSABI_FENIXOS = 16, // FenixOS
+ ELFOSABI_CLOUDABI = 17, // Nuxi CloudABI
+ ELFOSABI_FIRST_ARCH = 64, // First architecture-specific OS ABI
+ ELFOSABI_AMDGPU_HSA = 64, // AMD HSA runtime
+ ELFOSABI_AMDGPU_PAL = 65, // AMD PAL runtime
+ ELFOSABI_AMDGPU_MESA3D = 66, // AMD GCN GPUs (GFX6+) for MESA runtime
+ ELFOSABI_ARM = 97, // ARM
+ ELFOSABI_C6000_ELFABI = 64, // Bare-metal TMS320C6000
+ ELFOSABI_C6000_LINUX = 65, // Linux TMS320C6000
+ ELFOSABI_STANDALONE = 255, // Standalone (embedded) application
+ ELFOSABI_LAST_ARCH = 255 // Last Architecture-specific OS ABI
+};
+
+#define ELF_RELOC(name, value) name = value,
+
+// X86_64 relocations.
+enum {
+#include "ELFRelocs/x86_64.def"
+};
+
+// i386 relocations.
+enum {
+#include "ELFRelocs/i386.def"
+};
+
+// ELF Relocation types for PPC32
+enum {
+#include "ELFRelocs/PowerPC.def"
+};
+
+// Specific e_flags for PPC64
+enum {
+ // e_flags bits specifying ABI:
+ // 1 for original ABI using function descriptors,
+ // 2 for revised ABI without function descriptors,
+ // 0 for unspecified or not using any features affected by the differences.
+ EF_PPC64_ABI = 3
+};
+
+// Special values for the st_other field in the symbol table entry for PPC64.
+enum {
+ STO_PPC64_LOCAL_BIT = 5,
+ STO_PPC64_LOCAL_MASK = (7 << STO_PPC64_LOCAL_BIT)
+};
+static inline int64_t decodePPC64LocalEntryOffset(unsigned Other) {
+ unsigned Val = (Other & STO_PPC64_LOCAL_MASK) >> STO_PPC64_LOCAL_BIT;
+ return ((1 << Val) >> 2) << 2;
+}
+static inline unsigned encodePPC64LocalEntryOffset(int64_t Offset) {
+ unsigned Val =
+ (Offset >= 4 * 4 ? (Offset >= 8 * 4 ? (Offset >= 16 * 4 ? 6 : 5) : 4)
+ : (Offset >= 2 * 4 ? 3 : (Offset >= 1 * 4 ? 2 : 0)));
+ return Val << STO_PPC64_LOCAL_BIT;
+}
+
+// ELF Relocation types for PPC64
+enum {
+#include "ELFRelocs/PowerPC64.def"
+};
+
+// ELF Relocation types for AArch64
+enum {
+#include "ELFRelocs/AArch64.def"
+};
+
+// ARM Specific e_flags
+enum : unsigned {
+ EF_ARM_SOFT_FLOAT = 0x00000200U, // Legacy pre EABI_VER5
+ EF_ARM_ABI_FLOAT_SOFT = 0x00000200U, // EABI_VER5
+ EF_ARM_VFP_FLOAT = 0x00000400U, // Legacy pre EABI_VER5
+ EF_ARM_ABI_FLOAT_HARD = 0x00000400U, // EABI_VER5
+ EF_ARM_EABI_UNKNOWN = 0x00000000U,
+ EF_ARM_EABI_VER1 = 0x01000000U,
+ EF_ARM_EABI_VER2 = 0x02000000U,
+ EF_ARM_EABI_VER3 = 0x03000000U,
+ EF_ARM_EABI_VER4 = 0x04000000U,
+ EF_ARM_EABI_VER5 = 0x05000000U,
+ EF_ARM_EABIMASK = 0xFF000000U
+};
+
+// ELF Relocation types for ARM
+enum {
+#include "ELFRelocs/ARM.def"
+};
+
+// ARC Specific e_flags
+enum : unsigned {
+ EF_ARC_MACH_MSK = 0x000000ff,
+ EF_ARC_OSABI_MSK = 0x00000f00,
+ E_ARC_MACH_ARC600 = 0x00000002,
+ E_ARC_MACH_ARC601 = 0x00000004,
+ E_ARC_MACH_ARC700 = 0x00000003,
+ EF_ARC_CPU_ARCV2EM = 0x00000005,
+ EF_ARC_CPU_ARCV2HS = 0x00000006,
+ E_ARC_OSABI_ORIG = 0x00000000,
+ E_ARC_OSABI_V2 = 0x00000200,
+ E_ARC_OSABI_V3 = 0x00000300,
+ E_ARC_OSABI_V4 = 0x00000400,
+ EF_ARC_PIC = 0x00000100
+};
+
+// ELF Relocation types for ARC
+enum {
+#include "ELFRelocs/ARC.def"
+};
+
+// AVR specific e_flags
+enum : unsigned {
+ EF_AVR_ARCH_AVR1 = 1,
+ EF_AVR_ARCH_AVR2 = 2,
+ EF_AVR_ARCH_AVR25 = 25,
+ EF_AVR_ARCH_AVR3 = 3,
+ EF_AVR_ARCH_AVR31 = 31,
+ EF_AVR_ARCH_AVR35 = 35,
+ EF_AVR_ARCH_AVR4 = 4,
+ EF_AVR_ARCH_AVR5 = 5,
+ EF_AVR_ARCH_AVR51 = 51,
+ EF_AVR_ARCH_AVR6 = 6,
+ EF_AVR_ARCH_AVRTINY = 100,
+ EF_AVR_ARCH_XMEGA1 = 101,
+ EF_AVR_ARCH_XMEGA2 = 102,
+ EF_AVR_ARCH_XMEGA3 = 103,
+ EF_AVR_ARCH_XMEGA4 = 104,
+ EF_AVR_ARCH_XMEGA5 = 105,
+ EF_AVR_ARCH_XMEGA6 = 106,
+ EF_AVR_ARCH_XMEGA7 = 107
+};
+
+// ELF Relocation types for AVR
+enum {
+#include "ELFRelocs/AVR.def"
+};
+
+// Mips Specific e_flags
+enum : unsigned {
+ EF_MIPS_NOREORDER = 0x00000001, // Don't reorder instructions
+ EF_MIPS_PIC = 0x00000002, // Position independent code
+ EF_MIPS_CPIC = 0x00000004, // Call object with Position independent code
+ EF_MIPS_ABI2 = 0x00000020, // File uses N32 ABI
+ EF_MIPS_32BITMODE = 0x00000100, // Code compiled for a 64-bit machine
+ // in 32-bit mode
+ EF_MIPS_FP64 = 0x00000200, // Code compiled for a 32-bit machine
+ // but uses 64-bit FP registers
+ EF_MIPS_NAN2008 = 0x00000400, // Uses IEE 754-2008 NaN encoding
+
+ // ABI flags
+ EF_MIPS_ABI_O32 = 0x00001000, // This file follows the first MIPS 32 bit ABI
+ EF_MIPS_ABI_O64 = 0x00002000, // O32 ABI extended for 64-bit architecture.
+ EF_MIPS_ABI_EABI32 = 0x00003000, // EABI in 32 bit mode.
+ EF_MIPS_ABI_EABI64 = 0x00004000, // EABI in 64 bit mode.
+ EF_MIPS_ABI = 0x0000f000, // Mask for selecting EF_MIPS_ABI_ variant.
+
+ // MIPS machine variant
+ EF_MIPS_MACH_NONE = 0x00000000, // A standard MIPS implementation.
+ EF_MIPS_MACH_3900 = 0x00810000, // Toshiba R3900
+ EF_MIPS_MACH_4010 = 0x00820000, // LSI R4010
+ EF_MIPS_MACH_4100 = 0x00830000, // NEC VR4100
+ EF_MIPS_MACH_4650 = 0x00850000, // MIPS R4650
+ EF_MIPS_MACH_4120 = 0x00870000, // NEC VR4120
+ EF_MIPS_MACH_4111 = 0x00880000, // NEC VR4111/VR4181
+ EF_MIPS_MACH_SB1 = 0x008a0000, // Broadcom SB-1
+ EF_MIPS_MACH_OCTEON = 0x008b0000, // Cavium Networks Octeon
+ EF_MIPS_MACH_XLR = 0x008c0000, // RMI Xlr
+ EF_MIPS_MACH_OCTEON2 = 0x008d0000, // Cavium Networks Octeon2
+ EF_MIPS_MACH_OCTEON3 = 0x008e0000, // Cavium Networks Octeon3
+ EF_MIPS_MACH_5400 = 0x00910000, // NEC VR5400
+ EF_MIPS_MACH_5900 = 0x00920000, // MIPS R5900
+ EF_MIPS_MACH_5500 = 0x00980000, // NEC VR5500
+ EF_MIPS_MACH_9000 = 0x00990000, // Unknown
+ EF_MIPS_MACH_LS2E = 0x00a00000, // ST Microelectronics Loongson 2E
+ EF_MIPS_MACH_LS2F = 0x00a10000, // ST Microelectronics Loongson 2F
+ EF_MIPS_MACH_LS3A = 0x00a20000, // Loongson 3A
+ EF_MIPS_MACH = 0x00ff0000, // EF_MIPS_MACH_xxx selection mask
+
+ // ARCH_ASE
+ EF_MIPS_MICROMIPS = 0x02000000, // microMIPS
+ EF_MIPS_ARCH_ASE_M16 = 0x04000000, // Has Mips-16 ISA extensions
+ EF_MIPS_ARCH_ASE_MDMX = 0x08000000, // Has MDMX multimedia extensions
+ EF_MIPS_ARCH_ASE = 0x0f000000, // Mask for EF_MIPS_ARCH_ASE_xxx flags
+
+ // ARCH
+ EF_MIPS_ARCH_1 = 0x00000000, // MIPS1 instruction set
+ EF_MIPS_ARCH_2 = 0x10000000, // MIPS2 instruction set
+ EF_MIPS_ARCH_3 = 0x20000000, // MIPS3 instruction set
+ EF_MIPS_ARCH_4 = 0x30000000, // MIPS4 instruction set
+ EF_MIPS_ARCH_5 = 0x40000000, // MIPS5 instruction set
+ EF_MIPS_ARCH_32 = 0x50000000, // MIPS32 instruction set per linux not elf.h
+ EF_MIPS_ARCH_64 = 0x60000000, // MIPS64 instruction set per linux not elf.h
+ EF_MIPS_ARCH_32R2 = 0x70000000, // mips32r2, mips32r3, mips32r5
+ EF_MIPS_ARCH_64R2 = 0x80000000, // mips64r2, mips64r3, mips64r5
+ EF_MIPS_ARCH_32R6 = 0x90000000, // mips32r6
+ EF_MIPS_ARCH_64R6 = 0xa0000000, // mips64r6
+ EF_MIPS_ARCH = 0xf0000000 // Mask for applying EF_MIPS_ARCH_ variant
+};
+
+// ELF Relocation types for Mips
+enum {
+#include "ELFRelocs/Mips.def"
+};
+
+// Special values for the st_other field in the symbol table entry for MIPS.
+enum {
+ STO_MIPS_OPTIONAL = 0x04, // Symbol whose definition is optional
+ STO_MIPS_PLT = 0x08, // PLT entry related dynamic table record
+ STO_MIPS_PIC = 0x20, // PIC func in an object mixes PIC/non-PIC
+ STO_MIPS_MICROMIPS = 0x80, // MIPS Specific ISA for MicroMips
+ STO_MIPS_MIPS16 = 0xf0 // MIPS Specific ISA for Mips16
+};
+
+// .MIPS.options section descriptor kinds
+enum {
+ ODK_NULL = 0, // Undefined
+ ODK_REGINFO = 1, // Register usage information
+ ODK_EXCEPTIONS = 2, // Exception processing options
+ ODK_PAD = 3, // Section padding options
+ ODK_HWPATCH = 4, // Hardware patches applied
+ ODK_FILL = 5, // Linker fill value
+ ODK_TAGS = 6, // Space for tool identification
+ ODK_HWAND = 7, // Hardware AND patches applied
+ ODK_HWOR = 8, // Hardware OR patches applied
+ ODK_GP_GROUP = 9, // GP group to use for text/data sections
+ ODK_IDENT = 10, // ID information
+ ODK_PAGESIZE = 11 // Page size information
+};
+
+// Hexagon-specific e_flags
+enum {
+ // Object processor version flags, bits[11:0]
+ EF_HEXAGON_MACH_V2 = 0x00000001, // Hexagon V2
+ EF_HEXAGON_MACH_V3 = 0x00000002, // Hexagon V3
+ EF_HEXAGON_MACH_V4 = 0x00000003, // Hexagon V4
+ EF_HEXAGON_MACH_V5 = 0x00000004, // Hexagon V5
+ EF_HEXAGON_MACH_V55 = 0x00000005, // Hexagon V55
+ EF_HEXAGON_MACH_V60 = 0x00000060, // Hexagon V60
+ EF_HEXAGON_MACH_V62 = 0x00000062, // Hexagon V62
+ EF_HEXAGON_MACH_V65 = 0x00000065, // Hexagon V65
+ EF_HEXAGON_MACH_V66 = 0x00000066, // Hexagon V66
+
+ // Highest ISA version flags
+ EF_HEXAGON_ISA_MACH = 0x00000000, // Same as specified in bits[11:0]
+ // of e_flags
+ EF_HEXAGON_ISA_V2 = 0x00000010, // Hexagon V2 ISA
+ EF_HEXAGON_ISA_V3 = 0x00000020, // Hexagon V3 ISA
+ EF_HEXAGON_ISA_V4 = 0x00000030, // Hexagon V4 ISA
+ EF_HEXAGON_ISA_V5 = 0x00000040, // Hexagon V5 ISA
+ EF_HEXAGON_ISA_V55 = 0x00000050, // Hexagon V55 ISA
+ EF_HEXAGON_ISA_V60 = 0x00000060, // Hexagon V60 ISA
+ EF_HEXAGON_ISA_V62 = 0x00000062, // Hexagon V62 ISA
+ EF_HEXAGON_ISA_V65 = 0x00000065, // Hexagon V65 ISA
+ EF_HEXAGON_ISA_V66 = 0x00000066, // Hexagon V66 ISA
+};
+
+// Hexagon-specific section indexes for common small data
+enum {
+ SHN_HEXAGON_SCOMMON = 0xff00, // Other access sizes
+ SHN_HEXAGON_SCOMMON_1 = 0xff01, // Byte-sized access
+ SHN_HEXAGON_SCOMMON_2 = 0xff02, // Half-word-sized access
+ SHN_HEXAGON_SCOMMON_4 = 0xff03, // Word-sized access
+ SHN_HEXAGON_SCOMMON_8 = 0xff04 // Double-word-size access
+};
+
+// ELF Relocation types for Hexagon
+enum {
+#include "ELFRelocs/Hexagon.def"
+};
+
+// ELF Relocation type for Lanai.
+enum {
+#include "ELFRelocs/Lanai.def"
+};
+
+// RISCV Specific e_flags
+enum : unsigned {
+ EF_RISCV_RVC = 0x0001,
+ EF_RISCV_FLOAT_ABI = 0x0006,
+ EF_RISCV_FLOAT_ABI_SOFT = 0x0000,
+ EF_RISCV_FLOAT_ABI_SINGLE = 0x0002,
+ EF_RISCV_FLOAT_ABI_DOUBLE = 0x0004,
+ EF_RISCV_FLOAT_ABI_QUAD = 0x0006,
+ EF_RISCV_RVE = 0x0008
+};
+
+// ELF Relocation types for RISC-V
+enum {
+#include "ELFRelocs/RISCV.def"
+};
+
+// ELF Relocation types for S390/zSeries
+enum {
+#include "ELFRelocs/SystemZ.def"
+};
+
+// ELF Relocation type for Sparc.
+enum {
+#include "ELFRelocs/Sparc.def"
+};
+
+// AMDGPU specific e_flags.
+enum : unsigned {
+ // Processor selection mask for EF_AMDGPU_MACH_* values.
+ EF_AMDGPU_MACH = 0x0ff,
+
+ // Not specified processor.
+ EF_AMDGPU_MACH_NONE = 0x000,
+
+ // R600-based processors.
+
+ // Radeon HD 2000/3000 Series (R600).
+ EF_AMDGPU_MACH_R600_R600 = 0x001,
+ EF_AMDGPU_MACH_R600_R630 = 0x002,
+ EF_AMDGPU_MACH_R600_RS880 = 0x003,
+ EF_AMDGPU_MACH_R600_RV670 = 0x004,
+ // Radeon HD 4000 Series (R700).
+ EF_AMDGPU_MACH_R600_RV710 = 0x005,
+ EF_AMDGPU_MACH_R600_RV730 = 0x006,
+ EF_AMDGPU_MACH_R600_RV770 = 0x007,
+ // Radeon HD 5000 Series (Evergreen).
+ EF_AMDGPU_MACH_R600_CEDAR = 0x008,
+ EF_AMDGPU_MACH_R600_CYPRESS = 0x009,
+ EF_AMDGPU_MACH_R600_JUNIPER = 0x00a,
+ EF_AMDGPU_MACH_R600_REDWOOD = 0x00b,
+ EF_AMDGPU_MACH_R600_SUMO = 0x00c,
+ // Radeon HD 6000 Series (Northern Islands).
+ EF_AMDGPU_MACH_R600_BARTS = 0x00d,
+ EF_AMDGPU_MACH_R600_CAICOS = 0x00e,
+ EF_AMDGPU_MACH_R600_CAYMAN = 0x00f,
+ EF_AMDGPU_MACH_R600_TURKS = 0x010,
+
+ // Reserved for R600-based processors.
+ EF_AMDGPU_MACH_R600_RESERVED_FIRST = 0x011,
+ EF_AMDGPU_MACH_R600_RESERVED_LAST = 0x01f,
+
+ // First/last R600-based processors.
+ EF_AMDGPU_MACH_R600_FIRST = EF_AMDGPU_MACH_R600_R600,
+ EF_AMDGPU_MACH_R600_LAST = EF_AMDGPU_MACH_R600_TURKS,
+
+ // AMDGCN-based processors.
+
+ // AMDGCN GFX6.
+ EF_AMDGPU_MACH_AMDGCN_GFX600 = 0x020,
+ EF_AMDGPU_MACH_AMDGCN_GFX601 = 0x021,
+ // AMDGCN GFX7.
+ EF_AMDGPU_MACH_AMDGCN_GFX700 = 0x022,
+ EF_AMDGPU_MACH_AMDGCN_GFX701 = 0x023,
+ EF_AMDGPU_MACH_AMDGCN_GFX702 = 0x024,
+ EF_AMDGPU_MACH_AMDGCN_GFX703 = 0x025,
+ EF_AMDGPU_MACH_AMDGCN_GFX704 = 0x026,
+ // AMDGCN GFX8.
+ EF_AMDGPU_MACH_AMDGCN_GFX801 = 0x028,
+ EF_AMDGPU_MACH_AMDGCN_GFX802 = 0x029,
+ EF_AMDGPU_MACH_AMDGCN_GFX803 = 0x02a,
+ EF_AMDGPU_MACH_AMDGCN_GFX810 = 0x02b,
+ // AMDGCN GFX9.
+ EF_AMDGPU_MACH_AMDGCN_GFX900 = 0x02c,
+ EF_AMDGPU_MACH_AMDGCN_GFX902 = 0x02d,
+ EF_AMDGPU_MACH_AMDGCN_GFX904 = 0x02e,
+ EF_AMDGPU_MACH_AMDGCN_GFX906 = 0x02f,
+ EF_AMDGPU_MACH_AMDGCN_GFX908 = 0x030,
+ EF_AMDGPU_MACH_AMDGCN_GFX909 = 0x031,
+ // AMDGCN GFX10.
+ EF_AMDGPU_MACH_AMDGCN_GFX1010 = 0x033,
+ EF_AMDGPU_MACH_AMDGCN_GFX1011 = 0x034,
+ EF_AMDGPU_MACH_AMDGCN_GFX1012 = 0x035,
+
+ // Reserved for AMDGCN-based processors.
+ EF_AMDGPU_MACH_AMDGCN_RESERVED0 = 0x027,
+ EF_AMDGPU_MACH_AMDGCN_RESERVED1 = 0x032,
+
+ // First/last AMDGCN-based processors.
+ EF_AMDGPU_MACH_AMDGCN_FIRST = EF_AMDGPU_MACH_AMDGCN_GFX600,
+ EF_AMDGPU_MACH_AMDGCN_LAST = EF_AMDGPU_MACH_AMDGCN_GFX1012,
+
+ // Indicates if the "xnack" target feature is enabled for all code contained
+ // in the object.
+ EF_AMDGPU_XNACK = 0x100,
+ // Indicates if the "sram-ecc" target feature is enabled for all code
+ // contained in the object.
+ EF_AMDGPU_SRAM_ECC = 0x200,
+};
+
+// ELF Relocation types for AMDGPU
+enum {
+#include "ELFRelocs/AMDGPU.def"
+};
+
+// ELF Relocation types for BPF
+enum {
+#include "ELFRelocs/BPF.def"
+};
+
+// MSP430 specific e_flags
+enum : unsigned {
+ EF_MSP430_MACH_MSP430x11 = 11,
+ EF_MSP430_MACH_MSP430x11x1 = 110,
+ EF_MSP430_MACH_MSP430x12 = 12,
+ EF_MSP430_MACH_MSP430x13 = 13,
+ EF_MSP430_MACH_MSP430x14 = 14,
+ EF_MSP430_MACH_MSP430x15 = 15,
+ EF_MSP430_MACH_MSP430x16 = 16,
+ EF_MSP430_MACH_MSP430x20 = 20,
+ EF_MSP430_MACH_MSP430x22 = 22,
+ EF_MSP430_MACH_MSP430x23 = 23,
+ EF_MSP430_MACH_MSP430x24 = 24,
+ EF_MSP430_MACH_MSP430x26 = 26,
+ EF_MSP430_MACH_MSP430x31 = 31,
+ EF_MSP430_MACH_MSP430x32 = 32,
+ EF_MSP430_MACH_MSP430x33 = 33,
+ EF_MSP430_MACH_MSP430x41 = 41,
+ EF_MSP430_MACH_MSP430x42 = 42,
+ EF_MSP430_MACH_MSP430x43 = 43,
+ EF_MSP430_MACH_MSP430x44 = 44,
+ EF_MSP430_MACH_MSP430X = 45,
+ EF_MSP430_MACH_MSP430x46 = 46,
+ EF_MSP430_MACH_MSP430x47 = 47,
+ EF_MSP430_MACH_MSP430x54 = 54,
+};
+
+// ELF Relocation types for MSP430
+enum {
+#include "ELFRelocs/MSP430.def"
+};
+
+#undef ELF_RELOC
+
+// Section header.
+struct Elf32_Shdr {
+ Elf32_Word sh_name; // Section name (index into string table)
+ Elf32_Word sh_type; // Section type (SHT_*)
+ Elf32_Word sh_flags; // Section flags (SHF_*)
+ Elf32_Addr sh_addr; // Address where section is to be loaded
+ Elf32_Off sh_offset; // File offset of section data, in bytes
+ Elf32_Word sh_size; // Size of section, in bytes
+ Elf32_Word sh_link; // Section type-specific header table index link
+ Elf32_Word sh_info; // Section type-specific extra information
+ Elf32_Word sh_addralign; // Section address alignment
+ Elf32_Word sh_entsize; // Size of records contained within the section
+};
+
+// Section header for ELF64 - same fields as ELF32, different types.
+struct Elf64_Shdr {
+ Elf64_Word sh_name;
+ Elf64_Word sh_type;
+ Elf64_Xword sh_flags;
+ Elf64_Addr sh_addr;
+ Elf64_Off sh_offset;
+ Elf64_Xword sh_size;
+ Elf64_Word sh_link;
+ Elf64_Word sh_info;
+ Elf64_Xword sh_addralign;
+ Elf64_Xword sh_entsize;
+};
+
+// Special section indices.
+enum {
+ SHN_UNDEF = 0, // Undefined, missing, irrelevant, or meaningless
+ SHN_LORESERVE = 0xff00, // Lowest reserved index
+ SHN_LOPROC = 0xff00, // Lowest processor-specific index
+ SHN_HIPROC = 0xff1f, // Highest processor-specific index
+ SHN_LOOS = 0xff20, // Lowest operating system-specific index
+ SHN_HIOS = 0xff3f, // Highest operating system-specific index
+ SHN_ABS = 0xfff1, // Symbol has absolute value; does not need relocation
+ SHN_COMMON = 0xfff2, // FORTRAN COMMON or C external global variables
+ SHN_XINDEX = 0xffff, // Mark that the index is >= SHN_LORESERVE
+ SHN_HIRESERVE = 0xffff // Highest reserved index
+};
+
+// Section types.
+enum : unsigned {
+ SHT_NULL = 0, // No associated section (inactive entry).
+ SHT_PROGBITS = 1, // Program-defined contents.
+ SHT_SYMTAB = 2, // Symbol table.
+ SHT_STRTAB = 3, // String table.
+ SHT_RELA = 4, // Relocation entries; explicit addends.
+ SHT_HASH = 5, // Symbol hash table.
+ SHT_DYNAMIC = 6, // Information for dynamic linking.
+ SHT_NOTE = 7, // Information about the file.
+ SHT_NOBITS = 8, // Data occupies no space in the file.
+ SHT_REL = 9, // Relocation entries; no explicit addends.
+ SHT_SHLIB = 10, // Reserved.
+ SHT_DYNSYM = 11, // Symbol table.
+ SHT_INIT_ARRAY = 14, // Pointers to initialization functions.
+ SHT_FINI_ARRAY = 15, // Pointers to termination functions.
+ SHT_PREINIT_ARRAY = 16, // Pointers to pre-init functions.
+ SHT_GROUP = 17, // Section group.
+ SHT_SYMTAB_SHNDX = 18, // Indices for SHN_XINDEX entries.
+ // Experimental support for SHT_RELR sections. For details, see proposal
+ // at https://groups.google.com/forum/#!topic/generic-abi/bX460iggiKg
+ SHT_RELR = 19, // Relocation entries; only offsets.
+ SHT_LOOS = 0x60000000, // Lowest operating system-specific type.
+ // Android packed relocation section types.
+ // https://android.googlesource.com/platform/bionic/+/6f12bfece5dcc01325e0abba56a46b1bcf991c69/tools/relocation_packer/src/elf_file.cc#37
+ SHT_ANDROID_REL = 0x60000001,
+ SHT_ANDROID_RELA = 0x60000002,
+ SHT_LLVM_ODRTAB = 0x6fff4c00, // LLVM ODR table.
+ SHT_LLVM_LINKER_OPTIONS = 0x6fff4c01, // LLVM Linker Options.
+ SHT_LLVM_CALL_GRAPH_PROFILE = 0x6fff4c02, // LLVM Call Graph Profile.
+ SHT_LLVM_ADDRSIG = 0x6fff4c03, // List of address-significant symbols
+ // for safe ICF.
+ SHT_LLVM_DEPENDENT_LIBRARIES = 0x6fff4c04, // LLVM Dependent Library Specifiers.
+ SHT_LLVM_SYMPART = 0x6fff4c05, // Symbol partition specification.
+ SHT_LLVM_PART_EHDR = 0x6fff4c06, // ELF header for loadable partition.
+ SHT_LLVM_PART_PHDR = 0x6fff4c07, // Phdrs for loadable partition.
+ // Android's experimental support for SHT_RELR sections.
+ // https://android.googlesource.com/platform/bionic/+/b7feec74547f84559a1467aca02708ff61346d2a/libc/include/elf.h#512
+ SHT_ANDROID_RELR = 0x6fffff00, // Relocation entries; only offsets.
+ SHT_GNU_ATTRIBUTES = 0x6ffffff5, // Object attributes.
+ SHT_GNU_HASH = 0x6ffffff6, // GNU-style hash table.
+ SHT_GNU_verdef = 0x6ffffffd, // GNU version definitions.
+ SHT_GNU_verneed = 0x6ffffffe, // GNU version references.
+ SHT_GNU_versym = 0x6fffffff, // GNU symbol versions table.
+ SHT_HIOS = 0x6fffffff, // Highest operating system-specific type.
+ SHT_LOPROC = 0x70000000, // Lowest processor arch-specific type.
+ // Fixme: All this is duplicated in MCSectionELF. Why??
+ // Exception Index table
+ SHT_ARM_EXIDX = 0x70000001U,
+ // BPABI DLL dynamic linking pre-emption map
+ SHT_ARM_PREEMPTMAP = 0x70000002U,
+ // Object file compatibility attributes
+ SHT_ARM_ATTRIBUTES = 0x70000003U,
+ SHT_ARM_DEBUGOVERLAY = 0x70000004U,
+ SHT_ARM_OVERLAYSECTION = 0x70000005U,
+ SHT_HEX_ORDERED = 0x70000000, // Link editor is to sort the entries in
+ // this section based on their sizes
+ SHT_X86_64_UNWIND = 0x70000001, // Unwind information
+
+ SHT_MIPS_REGINFO = 0x70000006, // Register usage information
+ SHT_MIPS_OPTIONS = 0x7000000d, // General options
+ SHT_MIPS_DWARF = 0x7000001e, // DWARF debugging section.
+ SHT_MIPS_ABIFLAGS = 0x7000002a, // ABI information.
+
+ SHT_MSP430_ATTRIBUTES = 0x70000003U,
+
+ SHT_HIPROC = 0x7fffffff, // Highest processor arch-specific type.
+ SHT_LOUSER = 0x80000000, // Lowest type reserved for applications.
+ SHT_HIUSER = 0xffffffff // Highest type reserved for applications.
+};
+
+// Section flags.
+enum : unsigned {
+ // Section data should be writable during execution.
+ SHF_WRITE = 0x1,
+
+ // Section occupies memory during program execution.
+ SHF_ALLOC = 0x2,
+
+ // Section contains executable machine instructions.
+ SHF_EXECINSTR = 0x4,
+
+ // The data in this section may be merged.
+ SHF_MERGE = 0x10,
+
+ // The data in this section is null-terminated strings.
+ SHF_STRINGS = 0x20,
+
+ // A field in this section holds a section header table index.
+ SHF_INFO_LINK = 0x40U,
+
+ // Adds special ordering requirements for link editors.
+ SHF_LINK_ORDER = 0x80U,
+
+ // This section requires special OS-specific processing to avoid incorrect
+ // behavior.
+ SHF_OS_NONCONFORMING = 0x100U,
+
+ // This section is a member of a section group.
+ SHF_GROUP = 0x200U,
+
+ // This section holds Thread-Local Storage.
+ SHF_TLS = 0x400U,
+
+ // Identifies a section containing compressed data.
+ SHF_COMPRESSED = 0x800U,
+
+ // This section is excluded from the final executable or shared library.
+ SHF_EXCLUDE = 0x80000000U,
+
+ // Start of target-specific flags.
+
+ SHF_MASKOS = 0x0ff00000,
+
+ // Bits indicating processor-specific flags.
+ SHF_MASKPROC = 0xf0000000,
+
+ /// All sections with the "d" flag are grouped together by the linker to form
+ /// the data section and the dp register is set to the start of the section by
+ /// the boot code.
+ XCORE_SHF_DP_SECTION = 0x10000000,
+
+ /// All sections with the "c" flag are grouped together by the linker to form
+ /// the constant pool and the cp register is set to the start of the constant
+ /// pool by the boot code.
+ XCORE_SHF_CP_SECTION = 0x20000000,
+
+ // If an object file section does not have this flag set, then it may not hold
+ // more than 2GB and can be freely referred to in objects using smaller code
+ // models. Otherwise, only objects using larger code models can refer to them.
+ // For example, a medium code model object can refer to data in a section that
+ // sets this flag besides being able to refer to data in a section that does
+ // not set it; likewise, a small code model object can refer only to code in a
+ // section that does not set this flag.
+ SHF_X86_64_LARGE = 0x10000000,
+
+ // All sections with the GPREL flag are grouped into a global data area
+ // for faster accesses
+ SHF_HEX_GPREL = 0x10000000,
+
+ // Section contains text/data which may be replicated in other sections.
+ // Linker must retain only one copy.
+ SHF_MIPS_NODUPES = 0x01000000,
+
+ // Linker must generate implicit hidden weak names.
+ SHF_MIPS_NAMES = 0x02000000,
+
+ // Section data local to process.
+ SHF_MIPS_LOCAL = 0x04000000,
+
+ // Do not strip this section.
+ SHF_MIPS_NOSTRIP = 0x08000000,
+
+ // Section must be part of global data area.
+ SHF_MIPS_GPREL = 0x10000000,
+
+ // This section should be merged.
+ SHF_MIPS_MERGE = 0x20000000,
+
+ // Address size to be inferred from section entry size.
+ SHF_MIPS_ADDR = 0x40000000,
+
+ // Section data is string data by default.
+ SHF_MIPS_STRING = 0x80000000,
+
+ // Make code section unreadable when in execute-only mode
+ SHF_ARM_PURECODE = 0x20000000
+};
+
+// Section Group Flags
+enum : unsigned {
+ GRP_COMDAT = 0x1,
+ GRP_MASKOS = 0x0ff00000,
+ GRP_MASKPROC = 0xf0000000
+};
+
+// Symbol table entries for ELF32.
+struct Elf32_Sym {
+ Elf32_Word st_name; // Symbol name (index into string table)
+ Elf32_Addr st_value; // Value or address associated with the symbol
+ Elf32_Word st_size; // Size of the symbol
+ unsigned char st_info; // Symbol's type and binding attributes
+ unsigned char st_other; // Must be zero; reserved
+ Elf32_Half st_shndx; // Which section (header table index) it's defined in
+
+ // These accessors and mutators correspond to the ELF32_ST_BIND,
+ // ELF32_ST_TYPE, and ELF32_ST_INFO macros defined in the ELF specification:
+ unsigned char getBinding() const { return st_info >> 4; }
+ unsigned char getType() const { return st_info & 0x0f; }
+ void setBinding(unsigned char b) { setBindingAndType(b, getType()); }
+ void setType(unsigned char t) { setBindingAndType(getBinding(), t); }
+ void setBindingAndType(unsigned char b, unsigned char t) {
+ st_info = (b << 4) + (t & 0x0f);
+ }
+};
+
+// Symbol table entries for ELF64.
+struct Elf64_Sym {
+ Elf64_Word st_name; // Symbol name (index into string table)
+ unsigned char st_info; // Symbol's type and binding attributes
+ unsigned char st_other; // Must be zero; reserved
+ Elf64_Half st_shndx; // Which section (header tbl index) it's defined in
+ Elf64_Addr st_value; // Value or address associated with the symbol
+ Elf64_Xword st_size; // Size of the symbol
+
+ // These accessors and mutators are identical to those defined for ELF32
+ // symbol table entries.
+ unsigned char getBinding() const { return st_info >> 4; }
+ unsigned char getType() const { return st_info & 0x0f; }
+ void setBinding(unsigned char b) { setBindingAndType(b, getType()); }
+ void setType(unsigned char t) { setBindingAndType(getBinding(), t); }
+ void setBindingAndType(unsigned char b, unsigned char t) {
+ st_info = (b << 4) + (t & 0x0f);
+ }
+};
+
+// The size (in bytes) of symbol table entries.
+enum {
+ SYMENTRY_SIZE32 = 16, // 32-bit symbol entry size
+ SYMENTRY_SIZE64 = 24 // 64-bit symbol entry size.
+};
+
+// Symbol bindings.
+enum {
+ STB_LOCAL = 0, // Local symbol, not visible outside obj file containing def
+ STB_GLOBAL = 1, // Global symbol, visible to all object files being combined
+ STB_WEAK = 2, // Weak symbol, like global but lower-precedence
+ STB_GNU_UNIQUE = 10,
+ STB_LOOS = 10, // Lowest operating system-specific binding type
+ STB_HIOS = 12, // Highest operating system-specific binding type
+ STB_LOPROC = 13, // Lowest processor-specific binding type
+ STB_HIPROC = 15 // Highest processor-specific binding type
+};
+
+// Symbol types.
+enum {
+ STT_NOTYPE = 0, // Symbol's type is not specified
+ STT_OBJECT = 1, // Symbol is a data object (variable, array, etc.)
+ STT_FUNC = 2, // Symbol is executable code (function, etc.)
+ STT_SECTION = 3, // Symbol refers to a section
+ STT_FILE = 4, // Local, absolute symbol that refers to a file
+ STT_COMMON = 5, // An uninitialized common block
+ STT_TLS = 6, // Thread local data object
+ STT_GNU_IFUNC = 10, // GNU indirect function
+ STT_LOOS = 10, // Lowest operating system-specific symbol type
+ STT_HIOS = 12, // Highest operating system-specific symbol type
+ STT_LOPROC = 13, // Lowest processor-specific symbol type
+ STT_HIPROC = 15, // Highest processor-specific symbol type
+
+ // AMDGPU symbol types
+ STT_AMDGPU_HSA_KERNEL = 10
+};
+
+enum {
+ STV_DEFAULT = 0, // Visibility is specified by binding type
+ STV_INTERNAL = 1, // Defined by processor supplements
+ STV_HIDDEN = 2, // Not visible to other components
+ STV_PROTECTED = 3 // Visible in other components but not preemptable
+};
+
+// Symbol number.
+enum { STN_UNDEF = 0 };
+
+// Special relocation symbols used in the MIPS64 ELF relocation entries
+enum {
+ RSS_UNDEF = 0, // None
+ RSS_GP = 1, // Value of gp
+ RSS_GP0 = 2, // Value of gp used to create object being relocated
+ RSS_LOC = 3 // Address of location being relocated
+};
+
+// Relocation entry, without explicit addend.
+struct Elf32_Rel {
+ Elf32_Addr r_offset; // Location (file byte offset, or program virtual addr)
+ Elf32_Word r_info; // Symbol table index and type of relocation to apply
+
+ // These accessors and mutators correspond to the ELF32_R_SYM, ELF32_R_TYPE,
+ // and ELF32_R_INFO macros defined in the ELF specification:
+ Elf32_Word getSymbol() const { return (r_info >> 8); }
+ unsigned char getType() const { return (unsigned char)(r_info & 0x0ff); }
+ void setSymbol(Elf32_Word s) { setSymbolAndType(s, getType()); }
+ void setType(unsigned char t) { setSymbolAndType(getSymbol(), t); }
+ void setSymbolAndType(Elf32_Word s, unsigned char t) {
+ r_info = (s << 8) + t;
+ }
+};
+
+// Relocation entry with explicit addend.
+struct Elf32_Rela {
+ Elf32_Addr r_offset; // Location (file byte offset, or program virtual addr)
+ Elf32_Word r_info; // Symbol table index and type of relocation to apply
+ Elf32_Sword r_addend; // Compute value for relocatable field by adding this
+
+ // These accessors and mutators correspond to the ELF32_R_SYM, ELF32_R_TYPE,
+ // and ELF32_R_INFO macros defined in the ELF specification:
+ Elf32_Word getSymbol() const { return (r_info >> 8); }
+ unsigned char getType() const { return (unsigned char)(r_info & 0x0ff); }
+ void setSymbol(Elf32_Word s) { setSymbolAndType(s, getType()); }
+ void setType(unsigned char t) { setSymbolAndType(getSymbol(), t); }
+ void setSymbolAndType(Elf32_Word s, unsigned char t) {
+ r_info = (s << 8) + t;
+ }
+};
+
+// Relocation entry without explicit addend or info (relative relocations only).
+typedef Elf32_Word Elf32_Relr; // offset/bitmap for relative relocations
+
+// Relocation entry, without explicit addend.
+struct Elf64_Rel {
+ Elf64_Addr r_offset; // Location (file byte offset, or program virtual addr).
+ Elf64_Xword r_info; // Symbol table index and type of relocation to apply.
+
+ // These accessors and mutators correspond to the ELF64_R_SYM, ELF64_R_TYPE,
+ // and ELF64_R_INFO macros defined in the ELF specification:
+ Elf64_Word getSymbol() const { return (r_info >> 32); }
+ Elf64_Word getType() const { return (Elf64_Word)(r_info & 0xffffffffL); }
+ void setSymbol(Elf64_Word s) { setSymbolAndType(s, getType()); }
+ void setType(Elf64_Word t) { setSymbolAndType(getSymbol(), t); }
+ void setSymbolAndType(Elf64_Word s, Elf64_Word t) {
+ r_info = ((Elf64_Xword)s << 32) + (t & 0xffffffffL);
+ }
+};
+
+// Relocation entry with explicit addend.
+struct Elf64_Rela {
+ Elf64_Addr r_offset; // Location (file byte offset, or program virtual addr).
+ Elf64_Xword r_info; // Symbol table index and type of relocation to apply.
+ Elf64_Sxword r_addend; // Compute value for relocatable field by adding this.
+
+ // These accessors and mutators correspond to the ELF64_R_SYM, ELF64_R_TYPE,
+ // and ELF64_R_INFO macros defined in the ELF specification:
+ Elf64_Word getSymbol() const { return (r_info >> 32); }
+ Elf64_Word getType() const { return (Elf64_Word)(r_info & 0xffffffffL); }
+ void setSymbol(Elf64_Word s) { setSymbolAndType(s, getType()); }
+ void setType(Elf64_Word t) { setSymbolAndType(getSymbol(), t); }
+ void setSymbolAndType(Elf64_Word s, Elf64_Word t) {
+ r_info = ((Elf64_Xword)s << 32) + (t & 0xffffffffL);
+ }
+};
+
+// Relocation entry without explicit addend or info (relative relocations only).
+typedef Elf64_Xword Elf64_Relr; // offset/bitmap for relative relocations
+
+// Program header for ELF32.
+struct Elf32_Phdr {
+ Elf32_Word p_type; // Type of segment
+ Elf32_Off p_offset; // File offset where segment is located, in bytes
+ Elf32_Addr p_vaddr; // Virtual address of beginning of segment
+ Elf32_Addr p_paddr; // Physical address of beginning of segment (OS-specific)
+ Elf32_Word p_filesz; // Num. of bytes in file image of segment (may be zero)
+ Elf32_Word p_memsz; // Num. of bytes in mem image of segment (may be zero)
+ Elf32_Word p_flags; // Segment flags
+ Elf32_Word p_align; // Segment alignment constraint
+};
+
+// Program header for ELF64.
+struct Elf64_Phdr {
+ Elf64_Word p_type; // Type of segment
+ Elf64_Word p_flags; // Segment flags
+ Elf64_Off p_offset; // File offset where segment is located, in bytes
+ Elf64_Addr p_vaddr; // Virtual address of beginning of segment
+ Elf64_Addr p_paddr; // Physical addr of beginning of segment (OS-specific)
+ Elf64_Xword p_filesz; // Num. of bytes in file image of segment (may be zero)
+ Elf64_Xword p_memsz; // Num. of bytes in mem image of segment (may be zero)
+ Elf64_Xword p_align; // Segment alignment constraint
+};
+
+// Segment types.
+enum {
+ PT_NULL = 0, // Unused segment.
+ PT_LOAD = 1, // Loadable segment.
+ PT_DYNAMIC = 2, // Dynamic linking information.
+ PT_INTERP = 3, // Interpreter pathname.
+ PT_NOTE = 4, // Auxiliary information.
+ PT_SHLIB = 5, // Reserved.
+ PT_PHDR = 6, // The program header table itself.
+ PT_TLS = 7, // The thread-local storage template.
+ PT_LOOS = 0x60000000, // Lowest operating system-specific pt entry type.
+ PT_HIOS = 0x6fffffff, // Highest operating system-specific pt entry type.
+ PT_LOPROC = 0x70000000, // Lowest processor-specific program hdr entry type.
+ PT_HIPROC = 0x7fffffff, // Highest processor-specific program hdr entry type.
+
+ // x86-64 program header types.
+ // These all contain stack unwind tables.
+ PT_GNU_EH_FRAME = 0x6474e550,
+ PT_SUNW_EH_FRAME = 0x6474e550,
+ PT_SUNW_UNWIND = 0x6464e550,
+
+ PT_GNU_STACK = 0x6474e551, // Indicates stack executability.
+ PT_GNU_RELRO = 0x6474e552, // Read-only after relocation.
+
+ PT_OPENBSD_RANDOMIZE = 0x65a3dbe6, // Fill with random data.
+ PT_OPENBSD_WXNEEDED = 0x65a3dbe7, // Program does W^X violations.
+ PT_OPENBSD_BOOTDATA = 0x65a41be6, // Section for boot arguments.
+
+ // ARM program header types.
+ PT_ARM_ARCHEXT = 0x70000000, // Platform architecture compatibility info
+ // These all contain stack unwind tables.
+ PT_ARM_EXIDX = 0x70000001,
+ PT_ARM_UNWIND = 0x70000001,
+
+ // MIPS program header types.
+ PT_MIPS_REGINFO = 0x70000000, // Register usage information.
+ PT_MIPS_RTPROC = 0x70000001, // Runtime procedure table.
+ PT_MIPS_OPTIONS = 0x70000002, // Options segment.
+ PT_MIPS_ABIFLAGS = 0x70000003, // Abiflags segment.
+};
+
+// Segment flag bits.
+enum : unsigned {
+ PF_X = 1, // Execute
+ PF_W = 2, // Write
+ PF_R = 4, // Read
+ PF_MASKOS = 0x0ff00000, // Bits for operating system-specific semantics.
+ PF_MASKPROC = 0xf0000000 // Bits for processor-specific semantics.
+};
+
+// Dynamic table entry for ELF32.
+struct Elf32_Dyn {
+ Elf32_Sword d_tag; // Type of dynamic table entry.
+ union {
+ Elf32_Word d_val; // Integer value of entry.
+ Elf32_Addr d_ptr; // Pointer value of entry.
+ } d_un;
+};
+
+// Dynamic table entry for ELF64.
+struct Elf64_Dyn {
+ Elf64_Sxword d_tag; // Type of dynamic table entry.
+ union {
+ Elf64_Xword d_val; // Integer value of entry.
+ Elf64_Addr d_ptr; // Pointer value of entry.
+ } d_un;
+};
+
+// Dynamic table entry tags.
+enum {
+#define DYNAMIC_TAG(name, value) DT_##name = value,
+#include "DynamicTags.def"
+#undef DYNAMIC_TAG
+};
+
+// DT_FLAGS values.
+enum {
+ DF_ORIGIN = 0x01, // The object may reference $ORIGIN.
+ DF_SYMBOLIC = 0x02, // Search the shared lib before searching the exe.
+ DF_TEXTREL = 0x04, // Relocations may modify a non-writable segment.
+ DF_BIND_NOW = 0x08, // Process all relocations on load.
+ DF_STATIC_TLS = 0x10 // Reject attempts to load dynamically.
+};
+
+// State flags selectable in the `d_un.d_val' element of the DT_FLAGS_1 entry.
+enum {
+ DF_1_NOW = 0x00000001, // Set RTLD_NOW for this object.
+ DF_1_GLOBAL = 0x00000002, // Set RTLD_GLOBAL for this object.
+ DF_1_GROUP = 0x00000004, // Set RTLD_GROUP for this object.
+ DF_1_NODELETE = 0x00000008, // Set RTLD_NODELETE for this object.
+ DF_1_LOADFLTR = 0x00000010, // Trigger filtee loading at runtime.
+ DF_1_INITFIRST = 0x00000020, // Set RTLD_INITFIRST for this object.
+ DF_1_NOOPEN = 0x00000040, // Set RTLD_NOOPEN for this object.
+ DF_1_ORIGIN = 0x00000080, // $ORIGIN must be handled.
+ DF_1_DIRECT = 0x00000100, // Direct binding enabled.
+ DF_1_TRANS = 0x00000200,
+ DF_1_INTERPOSE = 0x00000400, // Object is used to interpose.
+ DF_1_NODEFLIB = 0x00000800, // Ignore default lib search path.
+ DF_1_NODUMP = 0x00001000, // Object can't be dldump'ed.
+ DF_1_CONFALT = 0x00002000, // Configuration alternative created.
+ DF_1_ENDFILTEE = 0x00004000, // Filtee terminates filters search.
+ DF_1_DISPRELDNE = 0x00008000, // Disp reloc applied at build time.
+ DF_1_DISPRELPND = 0x00010000, // Disp reloc applied at run-time.
+ DF_1_NODIRECT = 0x00020000, // Object has no-direct binding.
+ DF_1_IGNMULDEF = 0x00040000,
+ DF_1_NOKSYMS = 0x00080000,
+ DF_1_NOHDR = 0x00100000,
+ DF_1_EDITED = 0x00200000, // Object is modified after built.
+ DF_1_NORELOC = 0x00400000,
+ DF_1_SYMINTPOSE = 0x00800000, // Object has individual interposers.
+ DF_1_GLOBAUDIT = 0x01000000, // Global auditing required.
+ DF_1_SINGLETON = 0x02000000 // Singleton symbols are used.
+};
+
+// DT_MIPS_FLAGS values.
+enum {
+ RHF_NONE = 0x00000000, // No flags.
+ RHF_QUICKSTART = 0x00000001, // Uses shortcut pointers.
+ RHF_NOTPOT = 0x00000002, // Hash size is not a power of two.
+ RHS_NO_LIBRARY_REPLACEMENT = 0x00000004, // Ignore LD_LIBRARY_PATH.
+ RHF_NO_MOVE = 0x00000008, // DSO address may not be relocated.
+ RHF_SGI_ONLY = 0x00000010, // SGI specific features.
+ RHF_GUARANTEE_INIT = 0x00000020, // Guarantee that .init will finish
+ // executing before any non-init
+ // code in DSO is called.
+ RHF_DELTA_C_PLUS_PLUS = 0x00000040, // Contains Delta C++ code.
+ RHF_GUARANTEE_START_INIT = 0x00000080, // Guarantee that .init will start
+ // executing before any non-init
+ // code in DSO is called.
+ RHF_PIXIE = 0x00000100, // Generated by pixie.
+ RHF_DEFAULT_DELAY_LOAD = 0x00000200, // Delay-load DSO by default.
+ RHF_REQUICKSTART = 0x00000400, // Object may be requickstarted
+ RHF_REQUICKSTARTED = 0x00000800, // Object has been requickstarted
+ RHF_CORD = 0x00001000, // Generated by cord.
+ RHF_NO_UNRES_UNDEF = 0x00002000, // Object contains no unresolved
+ // undef symbols.
+ RHF_RLD_ORDER_SAFE = 0x00004000 // Symbol table is in a safe order.
+};
+
+// ElfXX_VerDef structure version (GNU versioning)
+enum { VER_DEF_NONE = 0, VER_DEF_CURRENT = 1 };
+
+// VerDef Flags (ElfXX_VerDef::vd_flags)
+enum { VER_FLG_BASE = 0x1, VER_FLG_WEAK = 0x2, VER_FLG_INFO = 0x4 };
+
+// Special constants for the version table. (SHT_GNU_versym/.gnu.version)
+enum {
+ VER_NDX_LOCAL = 0, // Unversioned local symbol
+ VER_NDX_GLOBAL = 1, // Unversioned global symbol
+ VERSYM_VERSION = 0x7fff, // Version Index mask
+ VERSYM_HIDDEN = 0x8000 // Hidden bit (non-default version)
+};
+
+// ElfXX_VerNeed structure version (GNU versioning)
+enum { VER_NEED_NONE = 0, VER_NEED_CURRENT = 1 };
+
+// SHT_NOTE section types
+enum {
+ NT_FREEBSD_THRMISC = 7,
+ NT_FREEBSD_PROCSTAT_PROC = 8,
+ NT_FREEBSD_PROCSTAT_FILES = 9,
+ NT_FREEBSD_PROCSTAT_VMMAP = 10,
+ NT_FREEBSD_PROCSTAT_GROUPS = 11,
+ NT_FREEBSD_PROCSTAT_UMASK = 12,
+ NT_FREEBSD_PROCSTAT_RLIMIT = 13,
+ NT_FREEBSD_PROCSTAT_OSREL = 14,
+ NT_FREEBSD_PROCSTAT_PSSTRINGS = 15,
+ NT_FREEBSD_PROCSTAT_AUXV = 16,
+};
+
+// Generic note types
+enum : unsigned {
+ NT_VERSION = 1,
+ NT_ARCH = 2,
+ NT_GNU_BUILD_ATTRIBUTE_OPEN = 0x100,
+ NT_GNU_BUILD_ATTRIBUTE_FUNC = 0x101,
+};
+
+// Core note types
+enum : unsigned {
+ NT_PRSTATUS = 1,
+ NT_FPREGSET = 2,
+ NT_PRPSINFO = 3,
+ NT_TASKSTRUCT = 4,
+ NT_AUXV = 6,
+ NT_PSTATUS = 10,
+ NT_FPREGS = 12,
+ NT_PSINFO = 13,
+ NT_LWPSTATUS = 16,
+ NT_LWPSINFO = 17,
+ NT_WIN32PSTATUS = 18,
+
+ NT_PPC_VMX = 0x100,
+ NT_PPC_VSX = 0x102,
+ NT_PPC_TAR = 0x103,
+ NT_PPC_PPR = 0x104,
+ NT_PPC_DSCR = 0x105,
+ NT_PPC_EBB = 0x106,
+ NT_PPC_PMU = 0x107,
+ NT_PPC_TM_CGPR = 0x108,
+ NT_PPC_TM_CFPR = 0x109,
+ NT_PPC_TM_CVMX = 0x10a,
+ NT_PPC_TM_CVSX = 0x10b,
+ NT_PPC_TM_SPR = 0x10c,
+ NT_PPC_TM_CTAR = 0x10d,
+ NT_PPC_TM_CPPR = 0x10e,
+ NT_PPC_TM_CDSCR = 0x10f,
+
+ NT_386_TLS = 0x200,
+ NT_386_IOPERM = 0x201,
+ NT_X86_XSTATE = 0x202,
+
+ NT_S390_HIGH_GPRS = 0x300,
+ NT_S390_TIMER = 0x301,
+ NT_S390_TODCMP = 0x302,
+ NT_S390_TODPREG = 0x303,
+ NT_S390_CTRS = 0x304,
+ NT_S390_PREFIX = 0x305,
+ NT_S390_LAST_BREAK = 0x306,
+ NT_S390_SYSTEM_CALL = 0x307,
+ NT_S390_TDB = 0x308,
+ NT_S390_VXRS_LOW = 0x309,
+ NT_S390_VXRS_HIGH = 0x30a,
+ NT_S390_GS_CB = 0x30b,
+ NT_S390_GS_BC = 0x30c,
+
+ NT_ARM_VFP = 0x400,
+ NT_ARM_TLS = 0x401,
+ NT_ARM_HW_BREAK = 0x402,
+ NT_ARM_HW_WATCH = 0x403,
+ NT_ARM_SVE = 0x405,
+ NT_ARM_PAC_MASK = 0x406,
+
+ NT_FILE = 0x46494c45,
+ NT_PRXFPREG = 0x46e62b7f,
+ NT_SIGINFO = 0x53494749,
+};
+
+// LLVM-specific notes.
+enum {
+ NT_LLVM_HWASAN_GLOBALS = 3,
+};
+
+// GNU note types
+enum {
+ NT_GNU_ABI_TAG = 1,
+ NT_GNU_HWCAP = 2,
+ NT_GNU_BUILD_ID = 3,
+ NT_GNU_GOLD_VERSION = 4,
+ NT_GNU_PROPERTY_TYPE_0 = 5,
+};
+
+// Property types used in GNU_PROPERTY_TYPE_0 notes.
+enum : unsigned {
+ GNU_PROPERTY_STACK_SIZE = 1,
+ GNU_PROPERTY_NO_COPY_ON_PROTECTED = 2,
+ GNU_PROPERTY_AARCH64_FEATURE_1_AND = 0xc0000000,
+ GNU_PROPERTY_X86_FEATURE_1_AND = 0xc0000002,
+ GNU_PROPERTY_X86_ISA_1_NEEDED = 0xc0008000,
+ GNU_PROPERTY_X86_FEATURE_2_NEEDED = 0xc0008001,
+ GNU_PROPERTY_X86_ISA_1_USED = 0xc0010000,
+ GNU_PROPERTY_X86_FEATURE_2_USED = 0xc0010001,
+};
+
+// aarch64 processor feature bits.
+enum : unsigned {
+ GNU_PROPERTY_AARCH64_FEATURE_1_BTI = 1 << 0,
+ GNU_PROPERTY_AARCH64_FEATURE_1_PAC = 1 << 1,
+};
+
+// x86 processor feature bits.
+enum : unsigned {
+ GNU_PROPERTY_X86_FEATURE_1_IBT = 1 << 0,
+ GNU_PROPERTY_X86_FEATURE_1_SHSTK = 1 << 1,
+
+ GNU_PROPERTY_X86_ISA_1_CMOV = 1 << 0,
+ GNU_PROPERTY_X86_ISA_1_SSE = 1 << 1,
+ GNU_PROPERTY_X86_ISA_1_SSE2 = 1 << 2,
+ GNU_PROPERTY_X86_ISA_1_SSE3 = 1 << 3,
+ GNU_PROPERTY_X86_ISA_1_SSSE3 = 1 << 4,
+ GNU_PROPERTY_X86_ISA_1_SSE4_1 = 1 << 5,
+ GNU_PROPERTY_X86_ISA_1_SSE4_2 = 1 << 6,
+ GNU_PROPERTY_X86_ISA_1_AVX = 1 << 7,
+ GNU_PROPERTY_X86_ISA_1_AVX2 = 1 << 8,
+ GNU_PROPERTY_X86_ISA_1_FMA = 1 << 9,
+ GNU_PROPERTY_X86_ISA_1_AVX512F = 1 << 10,
+ GNU_PROPERTY_X86_ISA_1_AVX512CD = 1 << 11,
+ GNU_PROPERTY_X86_ISA_1_AVX512ER = 1 << 12,
+ GNU_PROPERTY_X86_ISA_1_AVX512PF = 1 << 13,
+ GNU_PROPERTY_X86_ISA_1_AVX512VL = 1 << 14,
+ GNU_PROPERTY_X86_ISA_1_AVX512DQ = 1 << 15,
+ GNU_PROPERTY_X86_ISA_1_AVX512BW = 1 << 16,
+ GNU_PROPERTY_X86_ISA_1_AVX512_4FMAPS = 1 << 17,
+ GNU_PROPERTY_X86_ISA_1_AVX512_4VNNIW = 1 << 18,
+ GNU_PROPERTY_X86_ISA_1_AVX512_BITALG = 1 << 19,
+ GNU_PROPERTY_X86_ISA_1_AVX512_IFMA = 1 << 20,
+ GNU_PROPERTY_X86_ISA_1_AVX512_VBMI = 1 << 21,
+ GNU_PROPERTY_X86_ISA_1_AVX512_VBMI2 = 1 << 22,
+ GNU_PROPERTY_X86_ISA_1_AVX512_VNNI = 1 << 23,
+
+ GNU_PROPERTY_X86_FEATURE_2_X86 = 1 << 0,
+ GNU_PROPERTY_X86_FEATURE_2_X87 = 1 << 1,
+ GNU_PROPERTY_X86_FEATURE_2_MMX = 1 << 2,
+ GNU_PROPERTY_X86_FEATURE_2_XMM = 1 << 3,
+ GNU_PROPERTY_X86_FEATURE_2_YMM = 1 << 4,
+ GNU_PROPERTY_X86_FEATURE_2_ZMM = 1 << 5,
+ GNU_PROPERTY_X86_FEATURE_2_FXSR = 1 << 6,
+ GNU_PROPERTY_X86_FEATURE_2_XSAVE = 1 << 7,
+ GNU_PROPERTY_X86_FEATURE_2_XSAVEOPT = 1 << 8,
+ GNU_PROPERTY_X86_FEATURE_2_XSAVEC = 1 << 9,
+};
+
+// AMDGPU-specific section indices.
+enum {
+ SHN_AMDGPU_LDS = 0xff00, // Variable in LDS; symbol encoded like SHN_COMMON
+};
+
+// AMD specific notes. (Code Object V2)
+enum {
+ // Note types with values between 0 and 9 (inclusive) are reserved.
+ NT_AMD_AMDGPU_HSA_METADATA = 10,
+ NT_AMD_AMDGPU_ISA = 11,
+ NT_AMD_AMDGPU_PAL_METADATA = 12
+};
+
+// AMDGPU specific notes. (Code Object V3)
+enum {
+ // Note types with values between 0 and 31 (inclusive) are reserved.
+ NT_AMDGPU_METADATA = 32
+};
+
+enum {
+ GNU_ABI_TAG_LINUX = 0,
+ GNU_ABI_TAG_HURD = 1,
+ GNU_ABI_TAG_SOLARIS = 2,
+ GNU_ABI_TAG_FREEBSD = 3,
+ GNU_ABI_TAG_NETBSD = 4,
+ GNU_ABI_TAG_SYLLABLE = 5,
+ GNU_ABI_TAG_NACL = 6,
+};
+
+constexpr const char *ELF_NOTE_GNU = "GNU";
+
+// Android packed relocation group flags.
+enum {
+ RELOCATION_GROUPED_BY_INFO_FLAG = 1,
+ RELOCATION_GROUPED_BY_OFFSET_DELTA_FLAG = 2,
+ RELOCATION_GROUPED_BY_ADDEND_FLAG = 4,
+ RELOCATION_GROUP_HAS_ADDEND_FLAG = 8,
+};
+
+// Compressed section header for ELF32.
+struct Elf32_Chdr {
+ Elf32_Word ch_type;
+ Elf32_Word ch_size;
+ Elf32_Word ch_addralign;
+};
+
+// Compressed section header for ELF64.
+struct Elf64_Chdr {
+ Elf64_Word ch_type;
+ Elf64_Word ch_reserved;
+ Elf64_Xword ch_size;
+ Elf64_Xword ch_addralign;
+};
+
+// Node header for ELF32.
+struct Elf32_Nhdr {
+ Elf32_Word n_namesz;
+ Elf32_Word n_descsz;
+ Elf32_Word n_type;
+};
+
+// Node header for ELF64.
+struct Elf64_Nhdr {
+ Elf64_Word n_namesz;
+ Elf64_Word n_descsz;
+ Elf64_Word n_type;
+};
+
+// Legal values for ch_type field of compressed section header.
+enum {
+ ELFCOMPRESS_ZLIB = 1, // ZLIB/DEFLATE algorithm.
+ ELFCOMPRESS_LOOS = 0x60000000, // Start of OS-specific.
+ ELFCOMPRESS_HIOS = 0x6fffffff, // End of OS-specific.
+ ELFCOMPRESS_LOPROC = 0x70000000, // Start of processor-specific.
+ ELFCOMPRESS_HIPROC = 0x7fffffff // End of processor-specific.
+};
+
+} // end namespace ELF
+} // end namespace llvm
+
+#endif // LLVM_BINARYFORMAT_ELF_H
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AArch64.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AArch64.def
new file mode 100644
index 000000000..c8364133e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AArch64.def
@@ -0,0 +1,221 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// Based on ABI release 1.1-beta, dated 6 November 2013. NB: The cover page of
+// this document, IHI0056C_beta_aaelf64.pdf, on infocenter.arm.com, still
+// labels this as release 1.0.
+ELF_RELOC(R_AARCH64_NONE, 0)
+ELF_RELOC(R_AARCH64_ABS64, 0x101)
+ELF_RELOC(R_AARCH64_ABS32, 0x102)
+ELF_RELOC(R_AARCH64_ABS16, 0x103)
+ELF_RELOC(R_AARCH64_PREL64, 0x104)
+ELF_RELOC(R_AARCH64_PREL32, 0x105)
+ELF_RELOC(R_AARCH64_PREL16, 0x106)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G0, 0x107)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G0_NC, 0x108)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G1, 0x109)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G1_NC, 0x10a)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G2, 0x10b)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G2_NC, 0x10c)
+ELF_RELOC(R_AARCH64_MOVW_UABS_G3, 0x10d)
+ELF_RELOC(R_AARCH64_MOVW_SABS_G0, 0x10e)
+ELF_RELOC(R_AARCH64_MOVW_SABS_G1, 0x10f)
+ELF_RELOC(R_AARCH64_MOVW_SABS_G2, 0x110)
+ELF_RELOC(R_AARCH64_LD_PREL_LO19, 0x111)
+ELF_RELOC(R_AARCH64_ADR_PREL_LO21, 0x112)
+ELF_RELOC(R_AARCH64_ADR_PREL_PG_HI21, 0x113)
+ELF_RELOC(R_AARCH64_ADR_PREL_PG_HI21_NC, 0x114)
+ELF_RELOC(R_AARCH64_ADD_ABS_LO12_NC, 0x115)
+ELF_RELOC(R_AARCH64_LDST8_ABS_LO12_NC, 0x116)
+ELF_RELOC(R_AARCH64_TSTBR14, 0x117)
+ELF_RELOC(R_AARCH64_CONDBR19, 0x118)
+ELF_RELOC(R_AARCH64_JUMP26, 0x11a)
+ELF_RELOC(R_AARCH64_CALL26, 0x11b)
+ELF_RELOC(R_AARCH64_LDST16_ABS_LO12_NC, 0x11c)
+ELF_RELOC(R_AARCH64_LDST32_ABS_LO12_NC, 0x11d)
+ELF_RELOC(R_AARCH64_LDST64_ABS_LO12_NC, 0x11e)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G0, 0x11f)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G0_NC, 0x120)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G1, 0x121)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G1_NC, 0x122)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G2, 0x123)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G2_NC, 0x124)
+ELF_RELOC(R_AARCH64_MOVW_PREL_G3, 0x125)
+ELF_RELOC(R_AARCH64_LDST128_ABS_LO12_NC, 0x12b)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G0, 0x12c)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G0_NC, 0x12d)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G1, 0x12e)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G1_NC, 0x12f)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G2, 0x130)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G2_NC, 0x131)
+ELF_RELOC(R_AARCH64_MOVW_GOTOFF_G3, 0x132)
+ELF_RELOC(R_AARCH64_GOTREL64, 0x133)
+ELF_RELOC(R_AARCH64_GOTREL32, 0x134)
+ELF_RELOC(R_AARCH64_GOT_LD_PREL19, 0x135)
+ELF_RELOC(R_AARCH64_LD64_GOTOFF_LO15, 0x136)
+ELF_RELOC(R_AARCH64_ADR_GOT_PAGE, 0x137)
+ELF_RELOC(R_AARCH64_LD64_GOT_LO12_NC, 0x138)
+ELF_RELOC(R_AARCH64_LD64_GOTPAGE_LO15, 0x139)
+ELF_RELOC(R_AARCH64_TLSGD_ADR_PREL21, 0x200)
+ELF_RELOC(R_AARCH64_TLSGD_ADR_PAGE21, 0x201)
+ELF_RELOC(R_AARCH64_TLSGD_ADD_LO12_NC, 0x202)
+ELF_RELOC(R_AARCH64_TLSGD_MOVW_G1, 0x203)
+ELF_RELOC(R_AARCH64_TLSGD_MOVW_G0_NC, 0x204)
+ELF_RELOC(R_AARCH64_TLSLD_ADR_PREL21, 0x205)
+ELF_RELOC(R_AARCH64_TLSLD_ADR_PAGE21, 0x206)
+ELF_RELOC(R_AARCH64_TLSLD_ADD_LO12_NC, 0x207)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_G1, 0x208)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_G0_NC, 0x209)
+ELF_RELOC(R_AARCH64_TLSLD_LD_PREL19, 0x20a)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_DTPREL_G2, 0x20b)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_DTPREL_G1, 0x20c)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_DTPREL_G1_NC, 0x20d)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_DTPREL_G0, 0x20e)
+ELF_RELOC(R_AARCH64_TLSLD_MOVW_DTPREL_G0_NC, 0x20f)
+ELF_RELOC(R_AARCH64_TLSLD_ADD_DTPREL_HI12, 0x210)
+ELF_RELOC(R_AARCH64_TLSLD_ADD_DTPREL_LO12, 0x211)
+ELF_RELOC(R_AARCH64_TLSLD_ADD_DTPREL_LO12_NC, 0x212)
+ELF_RELOC(R_AARCH64_TLSLD_LDST8_DTPREL_LO12, 0x213)
+ELF_RELOC(R_AARCH64_TLSLD_LDST8_DTPREL_LO12_NC, 0x214)
+ELF_RELOC(R_AARCH64_TLSLD_LDST16_DTPREL_LO12, 0x215)
+ELF_RELOC(R_AARCH64_TLSLD_LDST16_DTPREL_LO12_NC, 0x216)
+ELF_RELOC(R_AARCH64_TLSLD_LDST32_DTPREL_LO12, 0x217)
+ELF_RELOC(R_AARCH64_TLSLD_LDST32_DTPREL_LO12_NC, 0x218)
+ELF_RELOC(R_AARCH64_TLSLD_LDST64_DTPREL_LO12, 0x219)
+ELF_RELOC(R_AARCH64_TLSLD_LDST64_DTPREL_LO12_NC, 0x21a)
+ELF_RELOC(R_AARCH64_TLSIE_MOVW_GOTTPREL_G1, 0x21b)
+ELF_RELOC(R_AARCH64_TLSIE_MOVW_GOTTPREL_G0_NC, 0x21c)
+ELF_RELOC(R_AARCH64_TLSIE_ADR_GOTTPREL_PAGE21, 0x21d)
+ELF_RELOC(R_AARCH64_TLSIE_LD64_GOTTPREL_LO12_NC, 0x21e)
+ELF_RELOC(R_AARCH64_TLSIE_LD_GOTTPREL_PREL19, 0x21f)
+ELF_RELOC(R_AARCH64_TLSLE_MOVW_TPREL_G2, 0x220)
+ELF_RELOC(R_AARCH64_TLSLE_MOVW_TPREL_G1, 0x221)
+ELF_RELOC(R_AARCH64_TLSLE_MOVW_TPREL_G1_NC, 0x222)
+ELF_RELOC(R_AARCH64_TLSLE_MOVW_TPREL_G0, 0x223)
+ELF_RELOC(R_AARCH64_TLSLE_MOVW_TPREL_G0_NC, 0x224)
+ELF_RELOC(R_AARCH64_TLSLE_ADD_TPREL_HI12, 0x225)
+ELF_RELOC(R_AARCH64_TLSLE_ADD_TPREL_LO12, 0x226)
+ELF_RELOC(R_AARCH64_TLSLE_ADD_TPREL_LO12_NC, 0x227)
+ELF_RELOC(R_AARCH64_TLSLE_LDST8_TPREL_LO12, 0x228)
+ELF_RELOC(R_AARCH64_TLSLE_LDST8_TPREL_LO12_NC, 0x229)
+ELF_RELOC(R_AARCH64_TLSLE_LDST16_TPREL_LO12, 0x22a)
+ELF_RELOC(R_AARCH64_TLSLE_LDST16_TPREL_LO12_NC, 0x22b)
+ELF_RELOC(R_AARCH64_TLSLE_LDST32_TPREL_LO12, 0x22c)
+ELF_RELOC(R_AARCH64_TLSLE_LDST32_TPREL_LO12_NC, 0x22d)
+ELF_RELOC(R_AARCH64_TLSLE_LDST64_TPREL_LO12, 0x22e)
+ELF_RELOC(R_AARCH64_TLSLE_LDST64_TPREL_LO12_NC, 0x22f)
+ELF_RELOC(R_AARCH64_TLSDESC_LD_PREL19, 0x230)
+ELF_RELOC(R_AARCH64_TLSDESC_ADR_PREL21, 0x231)
+ELF_RELOC(R_AARCH64_TLSDESC_ADR_PAGE21, 0x232)
+ELF_RELOC(R_AARCH64_TLSDESC_LD64_LO12, 0x233)
+ELF_RELOC(R_AARCH64_TLSDESC_ADD_LO12, 0x234)
+ELF_RELOC(R_AARCH64_TLSDESC_OFF_G1, 0x235)
+ELF_RELOC(R_AARCH64_TLSDESC_OFF_G0_NC, 0x236)
+ELF_RELOC(R_AARCH64_TLSDESC_LDR, 0x237)
+ELF_RELOC(R_AARCH64_TLSDESC_ADD, 0x238)
+ELF_RELOC(R_AARCH64_TLSDESC_CALL, 0x239)
+ELF_RELOC(R_AARCH64_TLSLE_LDST128_TPREL_LO12, 0x23a)
+ELF_RELOC(R_AARCH64_TLSLE_LDST128_TPREL_LO12_NC, 0x23b)
+ELF_RELOC(R_AARCH64_TLSLD_LDST128_DTPREL_LO12, 0x23c)
+ELF_RELOC(R_AARCH64_TLSLD_LDST128_DTPREL_LO12_NC, 0x23d)
+ELF_RELOC(R_AARCH64_COPY, 0x400)
+ELF_RELOC(R_AARCH64_GLOB_DAT, 0x401)
+ELF_RELOC(R_AARCH64_JUMP_SLOT, 0x402)
+ELF_RELOC(R_AARCH64_RELATIVE, 0x403)
+// 0x404 and 0x405 are now R_AARCH64_TLS_IMPDEF1 and R_AARCH64_TLS_IMPDEF2
+// We follow GNU and define TLS_IMPDEF1 as TLS_DTPMOD64 and TLS_IMPDEF2 as
+// TLS_DTPREL64
+ELF_RELOC(R_AARCH64_TLS_DTPMOD64, 0x404)
+ELF_RELOC(R_AARCH64_TLS_DTPREL64, 0x405)
+ELF_RELOC(R_AARCH64_TLS_TPREL64, 0x406)
+ELF_RELOC(R_AARCH64_TLSDESC, 0x407)
+ELF_RELOC(R_AARCH64_IRELATIVE, 0x408)
+
+// ELF_RELOC(R_AARCH64_P32_NONE, 0)
+ELF_RELOC(R_AARCH64_P32_ABS32, 0x001)
+ELF_RELOC(R_AARCH64_P32_ABS16, 0x002)
+ELF_RELOC(R_AARCH64_P32_PREL32, 0x003)
+ELF_RELOC(R_AARCH64_P32_PREL16, 0x004)
+ELF_RELOC(R_AARCH64_P32_MOVW_UABS_G0, 0x005)
+ELF_RELOC(R_AARCH64_P32_MOVW_UABS_G0_NC, 0x006)
+ELF_RELOC(R_AARCH64_P32_MOVW_UABS_G1, 0x007)
+ELF_RELOC(R_AARCH64_P32_MOVW_SABS_G0, 0x008)
+ELF_RELOC(R_AARCH64_P32_LD_PREL_LO19, 0x009)
+ELF_RELOC(R_AARCH64_P32_ADR_PREL_LO21, 0x00a)
+ELF_RELOC(R_AARCH64_P32_ADR_PREL_PG_HI21, 0x00b)
+ELF_RELOC(R_AARCH64_P32_ADD_ABS_LO12_NC, 0x00c)
+ELF_RELOC(R_AARCH64_P32_LDST8_ABS_LO12_NC, 0x00d)
+ELF_RELOC(R_AARCH64_P32_LDST16_ABS_LO12_NC, 0x00e)
+ELF_RELOC(R_AARCH64_P32_LDST32_ABS_LO12_NC, 0x00f)
+ELF_RELOC(R_AARCH64_P32_LDST64_ABS_LO12_NC, 0x010)
+ELF_RELOC(R_AARCH64_P32_LDST128_ABS_LO12_NC, 0x011)
+ELF_RELOC(R_AARCH64_P32_TSTBR14, 0x012)
+ELF_RELOC(R_AARCH64_P32_CONDBR19, 0x013)
+ELF_RELOC(R_AARCH64_P32_JUMP26, 0x014)
+ELF_RELOC(R_AARCH64_P32_CALL26, 0x015)
+ELF_RELOC(R_AARCH64_P32_MOVW_PREL_G0, 0x016)
+ELF_RELOC(R_AARCH64_P32_MOVW_PREL_G0_NC, 0x017)
+ELF_RELOC(R_AARCH64_P32_MOVW_PREL_G1, 0x018)
+ELF_RELOC(R_AARCH64_P32_GOT_LD_PREL19, 0x019)
+ELF_RELOC(R_AARCH64_P32_ADR_GOT_PAGE, 0x01a)
+ELF_RELOC(R_AARCH64_P32_LD32_GOT_LO12_NC, 0x01b)
+ELF_RELOC(R_AARCH64_P32_LD32_GOTPAGE_LO14, 0x01c)
+ELF_RELOC(R_AARCH64_P32_TLSGD_ADR_PREL21, 0x050)
+ELF_RELOC(R_AARCH64_P32_TLSGD_ADR_PAGE21, 0x051)
+ELF_RELOC(R_AARCH64_P32_TLSGD_ADD_LO12_NC, 0x052)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADR_PREL21, 0x053)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADR_PAGE21, 0x054)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADD_LO12_NC, 0x055)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LD_PREL19, 0x056)
+ELF_RELOC(R_AARCH64_P32_TLSLD_MOVW_DTPREL_G1, 0x057)
+ELF_RELOC(R_AARCH64_P32_TLSLD_MOVW_DTPREL_G0, 0x058)
+ELF_RELOC(R_AARCH64_P32_TLSLD_MOVW_DTPREL_G0_NC, 0x059)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADD_DTPREL_HI12, 0x05a)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADD_DTPREL_LO12, 0x05b)
+ELF_RELOC(R_AARCH64_P32_TLSLD_ADD_DTPREL_LO12_NC, 0x05c)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST8_DTPREL_LO12, 0x05d)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST8_DTPREL_LO12_NC, 0x05e)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST16_DTPREL_LO12, 0x05f)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST16_DTPREL_LO12_NC, 0x060)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST32_DTPREL_LO12, 0x061)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST32_DTPREL_LO12_NC, 0x062)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST64_DTPREL_LO12, 0x063)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST64_DTPREL_LO12_NC, 0x064)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST128_DTPREL_LO12, 0x065)
+ELF_RELOC(R_AARCH64_P32_TLSLD_LDST128_DTPREL_LO12_NC,0x066)
+ELF_RELOC(R_AARCH64_P32_TLSIE_ADR_GOTTPREL_PAGE21, 0x067)
+ELF_RELOC(R_AARCH64_P32_TLSIE_LD32_GOTTPREL_LO12_NC, 0x068)
+ELF_RELOC(R_AARCH64_P32_TLSIE_LD_GOTTPREL_PREL19, 0x069)
+ELF_RELOC(R_AARCH64_P32_TLSLE_MOVW_TPREL_G1, 0x06a)
+ELF_RELOC(R_AARCH64_P32_TLSLE_MOVW_TPREL_G0, 0x06b)
+ELF_RELOC(R_AARCH64_P32_TLSLE_MOVW_TPREL_G0_NC, 0x06c)
+ELF_RELOC(R_AARCH64_P32_TLSLE_ADD_TPREL_HI12, 0x06d)
+ELF_RELOC(R_AARCH64_P32_TLSLE_ADD_TPREL_LO12, 0x06e)
+ELF_RELOC(R_AARCH64_P32_TLSLE_ADD_TPREL_LO12_NC, 0x06f)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST8_TPREL_LO12, 0x070)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST8_TPREL_LO12_NC, 0x071)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST16_TPREL_LO12, 0x072)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST16_TPREL_LO12_NC, 0x073)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST32_TPREL_LO12, 0x074)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST32_TPREL_LO12_NC, 0x075)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST64_TPREL_LO12, 0x076)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST64_TPREL_LO12_NC, 0x077)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST128_TPREL_LO12, 0x078)
+ELF_RELOC(R_AARCH64_P32_TLSLE_LDST128_TPREL_LO12_NC, 0x079)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_LD_PREL19, 0x07a)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_ADR_PREL21, 0x07b)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_ADR_PAGE21, 0x07c)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_LD32_LO12, 0x07d)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_ADD_LO12, 0x07e)
+ELF_RELOC(R_AARCH64_P32_TLSDESC_CALL, 0x07f)
+ELF_RELOC(R_AARCH64_P32_COPY, 0x0b4)
+ELF_RELOC(R_AARCH64_P32_GLOB_DAT, 0x0b5)
+ELF_RELOC(R_AARCH64_P32_JUMP_SLOT, 0x0b6)
+ELF_RELOC(R_AARCH64_P32_RELATIVE, 0x0b7)
+ELF_RELOC(R_AARCH64_P32_TLS_DTPREL, 0x0b8)
+ELF_RELOC(R_AARCH64_P32_TLS_DTPMOD, 0x0b9)
+ELF_RELOC(R_AARCH64_P32_TLS_TPREL, 0x0ba)
+ELF_RELOC(R_AARCH64_P32_TLSDESC, 0x0bb)
+ELF_RELOC(R_AARCH64_P32_IRELATIVE, 0x0bc)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AMDGPU.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AMDGPU.def
new file mode 100644
index 000000000..00b19c416
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AMDGPU.def
@@ -0,0 +1,17 @@
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_AMDGPU_NONE, 0)
+ELF_RELOC(R_AMDGPU_ABS32_LO, 1)
+ELF_RELOC(R_AMDGPU_ABS32_HI, 2)
+ELF_RELOC(R_AMDGPU_ABS64, 3)
+ELF_RELOC(R_AMDGPU_REL32, 4)
+ELF_RELOC(R_AMDGPU_REL64, 5)
+ELF_RELOC(R_AMDGPU_ABS32, 6)
+ELF_RELOC(R_AMDGPU_GOTPCREL, 7)
+ELF_RELOC(R_AMDGPU_GOTPCREL32_LO, 8)
+ELF_RELOC(R_AMDGPU_GOTPCREL32_HI, 9)
+ELF_RELOC(R_AMDGPU_REL32_LO, 10)
+ELF_RELOC(R_AMDGPU_REL32_HI, 11)
+ELF_RELOC(R_AMDGPU_RELATIVE64, 13)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARC.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARC.def
new file mode 100644
index 000000000..5691fb345
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARC.def
@@ -0,0 +1,74 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_ARC_NONE, 0)
+ELF_RELOC(R_ARC_8, 1)
+ELF_RELOC(R_ARC_16, 2)
+ELF_RELOC(R_ARC_24, 3)
+ELF_RELOC(R_ARC_32, 4)
+ELF_RELOC(R_ARC_N8, 8)
+ELF_RELOC(R_ARC_N16, 9)
+ELF_RELOC(R_ARC_N24, 10)
+ELF_RELOC(R_ARC_N32, 11)
+ELF_RELOC(R_ARC_SDA, 12)
+ELF_RELOC(R_ARC_SECTOFF, 13)
+ELF_RELOC(R_ARC_S21H_PCREL, 14)
+ELF_RELOC(R_ARC_S21W_PCREL, 15)
+ELF_RELOC(R_ARC_S25H_PCREL, 16)
+ELF_RELOC(R_ARC_S25W_PCREL, 17)
+ELF_RELOC(R_ARC_SDA32, 18)
+ELF_RELOC(R_ARC_SDA_LDST, 19)
+ELF_RELOC(R_ARC_SDA_LDST1, 20)
+ELF_RELOC(R_ARC_SDA_LDST2, 21)
+ELF_RELOC(R_ARC_SDA16_LD, 22)
+ELF_RELOC(R_ARC_SDA16_LD1, 23)
+ELF_RELOC(R_ARC_SDA16_LD2, 24)
+ELF_RELOC(R_ARC_S13_PCREL, 25)
+ELF_RELOC(R_ARC_W, 26)
+ELF_RELOC(R_ARC_32_ME, 27)
+ELF_RELOC(R_ARC_32_ME_S, 105)
+ELF_RELOC(R_ARC_N32_ME, 28)
+ELF_RELOC(R_ARC_SECTOFF_ME, 29)
+ELF_RELOC(R_ARC_SDA32_ME, 30)
+ELF_RELOC(R_ARC_W_ME, 31)
+ELF_RELOC(R_AC_SECTOFF_U8, 35)
+ELF_RELOC(R_AC_SECTOFF_U8_1, 36)
+ELF_RELOC(R_AC_SECTOFF_U8_2, 37)
+ELF_RELOC(R_AC_SECTOFF_S9, 38)
+ELF_RELOC(R_AC_SECTOFF_S9_1, 39)
+ELF_RELOC(R_AC_SECTOFF_S9_2, 40)
+ELF_RELOC(R_ARC_SECTOFF_ME_1, 41)
+ELF_RELOC(R_ARC_SECTOFF_ME_2, 42)
+ELF_RELOC(R_ARC_SECTOFF_1, 43)
+ELF_RELOC(R_ARC_SECTOFF_2, 44)
+ELF_RELOC(R_ARC_SDA_12, 45)
+ELF_RELOC(R_ARC_SDA16_ST2, 48)
+ELF_RELOC(R_ARC_32_PCREL, 49)
+ELF_RELOC(R_ARC_PC32, 50)
+ELF_RELOC(R_ARC_GOT32, 59)
+ELF_RELOC(R_ARC_GOTPC32, 51)
+ELF_RELOC(R_ARC_PLT32, 52)
+ELF_RELOC(R_ARC_COPY, 53)
+ELF_RELOC(R_ARC_GLOB_DAT, 54)
+ELF_RELOC(R_ARC_JMP_SLOT, 55)
+ELF_RELOC(R_ARC_RELATIVE, 56)
+ELF_RELOC(R_ARC_GOTOFF, 57)
+ELF_RELOC(R_ARC_GOTPC, 58)
+ELF_RELOC(R_ARC_S21W_PCREL_PLT, 60)
+ELF_RELOC(R_ARC_S25H_PCREL_PLT, 61)
+ELF_RELOC(R_ARC_JLI_SECTOFF, 63)
+ELF_RELOC(R_ARC_TLS_DTPMOD, 66)
+ELF_RELOC(R_ARC_TLS_TPOFF, 68)
+ELF_RELOC(R_ARC_TLS_GD_GOT, 69)
+ELF_RELOC(R_ARC_TLS_GD_LD, 70)
+ELF_RELOC(R_ARC_TLS_GD_CALL, 71)
+ELF_RELOC(R_ARC_TLS_IE_GOT, 72)
+ELF_RELOC(R_ARC_TLS_DTPOFF, 67)
+ELF_RELOC(R_ARC_TLS_DTPOFF_S9, 73)
+ELF_RELOC(R_ARC_TLS_LE_S9, 74)
+ELF_RELOC(R_ARC_TLS_LE_32, 75)
+ELF_RELOC(R_ARC_S25W_PCREL_PLT, 76)
+ELF_RELOC(R_ARC_S21H_PCREL_PLT, 77)
+ELF_RELOC(R_ARC_NPS_CMEM16, 78)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARM.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARM.def
new file mode 100644
index 000000000..e0709fb81
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/ARM.def
@@ -0,0 +1,141 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// Meets 2.09 ABI Specs.
+ELF_RELOC(R_ARM_NONE, 0x00)
+ELF_RELOC(R_ARM_PC24, 0x01)
+ELF_RELOC(R_ARM_ABS32, 0x02)
+ELF_RELOC(R_ARM_REL32, 0x03)
+ELF_RELOC(R_ARM_LDR_PC_G0, 0x04)
+ELF_RELOC(R_ARM_ABS16, 0x05)
+ELF_RELOC(R_ARM_ABS12, 0x06)
+ELF_RELOC(R_ARM_THM_ABS5, 0x07)
+ELF_RELOC(R_ARM_ABS8, 0x08)
+ELF_RELOC(R_ARM_SBREL32, 0x09)
+ELF_RELOC(R_ARM_THM_CALL, 0x0a)
+ELF_RELOC(R_ARM_THM_PC8, 0x0b)
+ELF_RELOC(R_ARM_BREL_ADJ, 0x0c)
+ELF_RELOC(R_ARM_TLS_DESC, 0x0d)
+ELF_RELOC(R_ARM_THM_SWI8, 0x0e)
+ELF_RELOC(R_ARM_XPC25, 0x0f)
+ELF_RELOC(R_ARM_THM_XPC22, 0x10)
+ELF_RELOC(R_ARM_TLS_DTPMOD32, 0x11)
+ELF_RELOC(R_ARM_TLS_DTPOFF32, 0x12)
+ELF_RELOC(R_ARM_TLS_TPOFF32, 0x13)
+ELF_RELOC(R_ARM_COPY, 0x14)
+ELF_RELOC(R_ARM_GLOB_DAT, 0x15)
+ELF_RELOC(R_ARM_JUMP_SLOT, 0x16)
+ELF_RELOC(R_ARM_RELATIVE, 0x17)
+ELF_RELOC(R_ARM_GOTOFF32, 0x18)
+ELF_RELOC(R_ARM_BASE_PREL, 0x19)
+ELF_RELOC(R_ARM_GOT_BREL, 0x1a)
+ELF_RELOC(R_ARM_PLT32, 0x1b)
+ELF_RELOC(R_ARM_CALL, 0x1c)
+ELF_RELOC(R_ARM_JUMP24, 0x1d)
+ELF_RELOC(R_ARM_THM_JUMP24, 0x1e)
+ELF_RELOC(R_ARM_BASE_ABS, 0x1f)
+ELF_RELOC(R_ARM_ALU_PCREL_7_0, 0x20)
+ELF_RELOC(R_ARM_ALU_PCREL_15_8, 0x21)
+ELF_RELOC(R_ARM_ALU_PCREL_23_15, 0x22)
+ELF_RELOC(R_ARM_LDR_SBREL_11_0_NC, 0x23)
+ELF_RELOC(R_ARM_ALU_SBREL_19_12_NC, 0x24)
+ELF_RELOC(R_ARM_ALU_SBREL_27_20_CK, 0x25)
+ELF_RELOC(R_ARM_TARGET1, 0x26)
+ELF_RELOC(R_ARM_SBREL31, 0x27)
+ELF_RELOC(R_ARM_V4BX, 0x28)
+ELF_RELOC(R_ARM_TARGET2, 0x29)
+ELF_RELOC(R_ARM_PREL31, 0x2a)
+ELF_RELOC(R_ARM_MOVW_ABS_NC, 0x2b)
+ELF_RELOC(R_ARM_MOVT_ABS, 0x2c)
+ELF_RELOC(R_ARM_MOVW_PREL_NC, 0x2d)
+ELF_RELOC(R_ARM_MOVT_PREL, 0x2e)
+ELF_RELOC(R_ARM_THM_MOVW_ABS_NC, 0x2f)
+ELF_RELOC(R_ARM_THM_MOVT_ABS, 0x30)
+ELF_RELOC(R_ARM_THM_MOVW_PREL_NC, 0x31)
+ELF_RELOC(R_ARM_THM_MOVT_PREL, 0x32)
+ELF_RELOC(R_ARM_THM_JUMP19, 0x33)
+ELF_RELOC(R_ARM_THM_JUMP6, 0x34)
+ELF_RELOC(R_ARM_THM_ALU_PREL_11_0, 0x35)
+ELF_RELOC(R_ARM_THM_PC12, 0x36)
+ELF_RELOC(R_ARM_ABS32_NOI, 0x37)
+ELF_RELOC(R_ARM_REL32_NOI, 0x38)
+ELF_RELOC(R_ARM_ALU_PC_G0_NC, 0x39)
+ELF_RELOC(R_ARM_ALU_PC_G0, 0x3a)
+ELF_RELOC(R_ARM_ALU_PC_G1_NC, 0x3b)
+ELF_RELOC(R_ARM_ALU_PC_G1, 0x3c)
+ELF_RELOC(R_ARM_ALU_PC_G2, 0x3d)
+ELF_RELOC(R_ARM_LDR_PC_G1, 0x3e)
+ELF_RELOC(R_ARM_LDR_PC_G2, 0x3f)
+ELF_RELOC(R_ARM_LDRS_PC_G0, 0x40)
+ELF_RELOC(R_ARM_LDRS_PC_G1, 0x41)
+ELF_RELOC(R_ARM_LDRS_PC_G2, 0x42)
+ELF_RELOC(R_ARM_LDC_PC_G0, 0x43)
+ELF_RELOC(R_ARM_LDC_PC_G1, 0x44)
+ELF_RELOC(R_ARM_LDC_PC_G2, 0x45)
+ELF_RELOC(R_ARM_ALU_SB_G0_NC, 0x46)
+ELF_RELOC(R_ARM_ALU_SB_G0, 0x47)
+ELF_RELOC(R_ARM_ALU_SB_G1_NC, 0x48)
+ELF_RELOC(R_ARM_ALU_SB_G1, 0x49)
+ELF_RELOC(R_ARM_ALU_SB_G2, 0x4a)
+ELF_RELOC(R_ARM_LDR_SB_G0, 0x4b)
+ELF_RELOC(R_ARM_LDR_SB_G1, 0x4c)
+ELF_RELOC(R_ARM_LDR_SB_G2, 0x4d)
+ELF_RELOC(R_ARM_LDRS_SB_G0, 0x4e)
+ELF_RELOC(R_ARM_LDRS_SB_G1, 0x4f)
+ELF_RELOC(R_ARM_LDRS_SB_G2, 0x50)
+ELF_RELOC(R_ARM_LDC_SB_G0, 0x51)
+ELF_RELOC(R_ARM_LDC_SB_G1, 0x52)
+ELF_RELOC(R_ARM_LDC_SB_G2, 0x53)
+ELF_RELOC(R_ARM_MOVW_BREL_NC, 0x54)
+ELF_RELOC(R_ARM_MOVT_BREL, 0x55)
+ELF_RELOC(R_ARM_MOVW_BREL, 0x56)
+ELF_RELOC(R_ARM_THM_MOVW_BREL_NC, 0x57)
+ELF_RELOC(R_ARM_THM_MOVT_BREL, 0x58)
+ELF_RELOC(R_ARM_THM_MOVW_BREL, 0x59)
+ELF_RELOC(R_ARM_TLS_GOTDESC, 0x5a)
+ELF_RELOC(R_ARM_TLS_CALL, 0x5b)
+ELF_RELOC(R_ARM_TLS_DESCSEQ, 0x5c)
+ELF_RELOC(R_ARM_THM_TLS_CALL, 0x5d)
+ELF_RELOC(R_ARM_PLT32_ABS, 0x5e)
+ELF_RELOC(R_ARM_GOT_ABS, 0x5f)
+ELF_RELOC(R_ARM_GOT_PREL, 0x60)
+ELF_RELOC(R_ARM_GOT_BREL12, 0x61)
+ELF_RELOC(R_ARM_GOTOFF12, 0x62)
+ELF_RELOC(R_ARM_GOTRELAX, 0x63)
+ELF_RELOC(R_ARM_GNU_VTENTRY, 0x64)
+ELF_RELOC(R_ARM_GNU_VTINHERIT, 0x65)
+ELF_RELOC(R_ARM_THM_JUMP11, 0x66)
+ELF_RELOC(R_ARM_THM_JUMP8, 0x67)
+ELF_RELOC(R_ARM_TLS_GD32, 0x68)
+ELF_RELOC(R_ARM_TLS_LDM32, 0x69)
+ELF_RELOC(R_ARM_TLS_LDO32, 0x6a)
+ELF_RELOC(R_ARM_TLS_IE32, 0x6b)
+ELF_RELOC(R_ARM_TLS_LE32, 0x6c)
+ELF_RELOC(R_ARM_TLS_LDO12, 0x6d)
+ELF_RELOC(R_ARM_TLS_LE12, 0x6e)
+ELF_RELOC(R_ARM_TLS_IE12GP, 0x6f)
+ELF_RELOC(R_ARM_PRIVATE_0, 0x70)
+ELF_RELOC(R_ARM_PRIVATE_1, 0x71)
+ELF_RELOC(R_ARM_PRIVATE_2, 0x72)
+ELF_RELOC(R_ARM_PRIVATE_3, 0x73)
+ELF_RELOC(R_ARM_PRIVATE_4, 0x74)
+ELF_RELOC(R_ARM_PRIVATE_5, 0x75)
+ELF_RELOC(R_ARM_PRIVATE_6, 0x76)
+ELF_RELOC(R_ARM_PRIVATE_7, 0x77)
+ELF_RELOC(R_ARM_PRIVATE_8, 0x78)
+ELF_RELOC(R_ARM_PRIVATE_9, 0x79)
+ELF_RELOC(R_ARM_PRIVATE_10, 0x7a)
+ELF_RELOC(R_ARM_PRIVATE_11, 0x7b)
+ELF_RELOC(R_ARM_PRIVATE_12, 0x7c)
+ELF_RELOC(R_ARM_PRIVATE_13, 0x7d)
+ELF_RELOC(R_ARM_PRIVATE_14, 0x7e)
+ELF_RELOC(R_ARM_PRIVATE_15, 0x7f)
+ELF_RELOC(R_ARM_ME_TOO, 0x80)
+ELF_RELOC(R_ARM_THM_TLS_DESCSEQ16, 0x81)
+ELF_RELOC(R_ARM_THM_TLS_DESCSEQ32, 0x82)
+ELF_RELOC(R_ARM_THM_BF16, 0x88)
+ELF_RELOC(R_ARM_THM_BF12, 0x89)
+ELF_RELOC(R_ARM_THM_BF18, 0x8a)
+ELF_RELOC(R_ARM_IRELATIVE, 0xa0)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AVR.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AVR.def
new file mode 100644
index 000000000..696fc60b0
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/AVR.def
@@ -0,0 +1,41 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_AVR_NONE, 0)
+ELF_RELOC(R_AVR_32, 1)
+ELF_RELOC(R_AVR_7_PCREL, 2)
+ELF_RELOC(R_AVR_13_PCREL, 3)
+ELF_RELOC(R_AVR_16, 4)
+ELF_RELOC(R_AVR_16_PM, 5)
+ELF_RELOC(R_AVR_LO8_LDI, 6)
+ELF_RELOC(R_AVR_HI8_LDI, 7)
+ELF_RELOC(R_AVR_HH8_LDI, 8)
+ELF_RELOC(R_AVR_LO8_LDI_NEG, 9)
+ELF_RELOC(R_AVR_HI8_LDI_NEG, 10)
+ELF_RELOC(R_AVR_HH8_LDI_NEG, 11)
+ELF_RELOC(R_AVR_LO8_LDI_PM, 12)
+ELF_RELOC(R_AVR_HI8_LDI_PM, 13)
+ELF_RELOC(R_AVR_HH8_LDI_PM, 14)
+ELF_RELOC(R_AVR_LO8_LDI_PM_NEG, 15)
+ELF_RELOC(R_AVR_HI8_LDI_PM_NEG, 16)
+ELF_RELOC(R_AVR_HH8_LDI_PM_NEG, 17)
+ELF_RELOC(R_AVR_CALL, 18)
+ELF_RELOC(R_AVR_LDI, 19)
+ELF_RELOC(R_AVR_6, 20)
+ELF_RELOC(R_AVR_6_ADIW, 21)
+ELF_RELOC(R_AVR_MS8_LDI, 22)
+ELF_RELOC(R_AVR_MS8_LDI_NEG, 23)
+ELF_RELOC(R_AVR_LO8_LDI_GS, 24)
+ELF_RELOC(R_AVR_HI8_LDI_GS, 25)
+ELF_RELOC(R_AVR_8, 26)
+ELF_RELOC(R_AVR_8_LO8, 27)
+ELF_RELOC(R_AVR_8_HI8, 28)
+ELF_RELOC(R_AVR_8_HLO8, 29)
+ELF_RELOC(R_AVR_DIFF8, 30)
+ELF_RELOC(R_AVR_DIFF16, 31)
+ELF_RELOC(R_AVR_DIFF32, 32)
+ELF_RELOC(R_AVR_LDS_STS_16, 33)
+ELF_RELOC(R_AVR_PORT6, 34)
+ELF_RELOC(R_AVR_PORT5, 35)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/BPF.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/BPF.def
new file mode 100644
index 000000000..5dd7f70b6
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/BPF.def
@@ -0,0 +1,8 @@
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// No relocation
+ELF_RELOC(R_BPF_NONE, 0)
+ELF_RELOC(R_BPF_64_64, 1)
+ELF_RELOC(R_BPF_64_32, 10)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Hexagon.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Hexagon.def
new file mode 100644
index 000000000..5021e2b26
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Hexagon.def
@@ -0,0 +1,106 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// Release 5 ABI
+ELF_RELOC(R_HEX_NONE, 0)
+ELF_RELOC(R_HEX_B22_PCREL, 1)
+ELF_RELOC(R_HEX_B15_PCREL, 2)
+ELF_RELOC(R_HEX_B7_PCREL, 3)
+ELF_RELOC(R_HEX_LO16, 4)
+ELF_RELOC(R_HEX_HI16, 5)
+ELF_RELOC(R_HEX_32, 6)
+ELF_RELOC(R_HEX_16, 7)
+ELF_RELOC(R_HEX_8, 8)
+ELF_RELOC(R_HEX_GPREL16_0, 9)
+ELF_RELOC(R_HEX_GPREL16_1, 10)
+ELF_RELOC(R_HEX_GPREL16_2, 11)
+ELF_RELOC(R_HEX_GPREL16_3, 12)
+ELF_RELOC(R_HEX_HL16, 13)
+ELF_RELOC(R_HEX_B13_PCREL, 14)
+ELF_RELOC(R_HEX_B9_PCREL, 15)
+ELF_RELOC(R_HEX_B32_PCREL_X, 16)
+ELF_RELOC(R_HEX_32_6_X, 17)
+ELF_RELOC(R_HEX_B22_PCREL_X, 18)
+ELF_RELOC(R_HEX_B15_PCREL_X, 19)
+ELF_RELOC(R_HEX_B13_PCREL_X, 20)
+ELF_RELOC(R_HEX_B9_PCREL_X, 21)
+ELF_RELOC(R_HEX_B7_PCREL_X, 22)
+ELF_RELOC(R_HEX_16_X, 23)
+ELF_RELOC(R_HEX_12_X, 24)
+ELF_RELOC(R_HEX_11_X, 25)
+ELF_RELOC(R_HEX_10_X, 26)
+ELF_RELOC(R_HEX_9_X, 27)
+ELF_RELOC(R_HEX_8_X, 28)
+ELF_RELOC(R_HEX_7_X, 29)
+ELF_RELOC(R_HEX_6_X, 30)
+ELF_RELOC(R_HEX_32_PCREL, 31)
+ELF_RELOC(R_HEX_COPY, 32)
+ELF_RELOC(R_HEX_GLOB_DAT, 33)
+ELF_RELOC(R_HEX_JMP_SLOT, 34)
+ELF_RELOC(R_HEX_RELATIVE, 35)
+ELF_RELOC(R_HEX_PLT_B22_PCREL, 36)
+ELF_RELOC(R_HEX_GOTREL_LO16, 37)
+ELF_RELOC(R_HEX_GOTREL_HI16, 38)
+ELF_RELOC(R_HEX_GOTREL_32, 39)
+ELF_RELOC(R_HEX_GOT_LO16, 40)
+ELF_RELOC(R_HEX_GOT_HI16, 41)
+ELF_RELOC(R_HEX_GOT_32, 42)
+ELF_RELOC(R_HEX_GOT_16, 43)
+ELF_RELOC(R_HEX_DTPMOD_32, 44)
+ELF_RELOC(R_HEX_DTPREL_LO16, 45)
+ELF_RELOC(R_HEX_DTPREL_HI16, 46)
+ELF_RELOC(R_HEX_DTPREL_32, 47)
+ELF_RELOC(R_HEX_DTPREL_16, 48)
+ELF_RELOC(R_HEX_GD_PLT_B22_PCREL, 49)
+ELF_RELOC(R_HEX_GD_GOT_LO16, 50)
+ELF_RELOC(R_HEX_GD_GOT_HI16, 51)
+ELF_RELOC(R_HEX_GD_GOT_32, 52)
+ELF_RELOC(R_HEX_GD_GOT_16, 53)
+ELF_RELOC(R_HEX_IE_LO16, 54)
+ELF_RELOC(R_HEX_IE_HI16, 55)
+ELF_RELOC(R_HEX_IE_32, 56)
+ELF_RELOC(R_HEX_IE_GOT_LO16, 57)
+ELF_RELOC(R_HEX_IE_GOT_HI16, 58)
+ELF_RELOC(R_HEX_IE_GOT_32, 59)
+ELF_RELOC(R_HEX_IE_GOT_16, 60)
+ELF_RELOC(R_HEX_TPREL_LO16, 61)
+ELF_RELOC(R_HEX_TPREL_HI16, 62)
+ELF_RELOC(R_HEX_TPREL_32, 63)
+ELF_RELOC(R_HEX_TPREL_16, 64)
+ELF_RELOC(R_HEX_6_PCREL_X, 65)
+ELF_RELOC(R_HEX_GOTREL_32_6_X, 66)
+ELF_RELOC(R_HEX_GOTREL_16_X, 67)
+ELF_RELOC(R_HEX_GOTREL_11_X, 68)
+ELF_RELOC(R_HEX_GOT_32_6_X, 69)
+ELF_RELOC(R_HEX_GOT_16_X, 70)
+ELF_RELOC(R_HEX_GOT_11_X, 71)
+ELF_RELOC(R_HEX_DTPREL_32_6_X, 72)
+ELF_RELOC(R_HEX_DTPREL_16_X, 73)
+ELF_RELOC(R_HEX_DTPREL_11_X, 74)
+ELF_RELOC(R_HEX_GD_GOT_32_6_X, 75)
+ELF_RELOC(R_HEX_GD_GOT_16_X, 76)
+ELF_RELOC(R_HEX_GD_GOT_11_X, 77)
+ELF_RELOC(R_HEX_IE_32_6_X, 78)
+ELF_RELOC(R_HEX_IE_16_X, 79)
+ELF_RELOC(R_HEX_IE_GOT_32_6_X, 80)
+ELF_RELOC(R_HEX_IE_GOT_16_X, 81)
+ELF_RELOC(R_HEX_IE_GOT_11_X, 82)
+ELF_RELOC(R_HEX_TPREL_32_6_X, 83)
+ELF_RELOC(R_HEX_TPREL_16_X, 84)
+ELF_RELOC(R_HEX_TPREL_11_X, 85)
+ELF_RELOC(R_HEX_LD_PLT_B22_PCREL, 86)
+ELF_RELOC(R_HEX_LD_GOT_LO16, 87)
+ELF_RELOC(R_HEX_LD_GOT_HI16, 88)
+ELF_RELOC(R_HEX_LD_GOT_32, 89)
+ELF_RELOC(R_HEX_LD_GOT_16, 90)
+ELF_RELOC(R_HEX_LD_GOT_32_6_X, 91)
+ELF_RELOC(R_HEX_LD_GOT_16_X, 92)
+ELF_RELOC(R_HEX_LD_GOT_11_X, 93)
+ELF_RELOC(R_HEX_23_REG, 94)
+ELF_RELOC(R_HEX_GD_PLT_B22_PCREL_X, 95)
+ELF_RELOC(R_HEX_GD_PLT_B32_PCREL_X, 96)
+ELF_RELOC(R_HEX_LD_PLT_B22_PCREL_X, 97)
+ELF_RELOC(R_HEX_LD_PLT_B32_PCREL_X, 98)
+ELF_RELOC(R_HEX_27_REG, 99)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Lanai.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Lanai.def
new file mode 100644
index 000000000..77ecb0484
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Lanai.def
@@ -0,0 +1,19 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// No relocation
+ELF_RELOC(R_LANAI_NONE, 0)
+// 21-bit symbol relocation
+ELF_RELOC(R_LANAI_21, 1)
+// 21-bit symbol relocation with last two bits masked to 0
+ELF_RELOC(R_LANAI_21_F, 2)
+// 25-bit branch targets
+ELF_RELOC(R_LANAI_25, 3)
+// General 32-bit relocation
+ELF_RELOC(R_LANAI_32, 4)
+// Upper 16-bits of a symbolic relocation
+ELF_RELOC(R_LANAI_HI16, 5)
+// Lower 16-bits of a symbolic relocation
+ELF_RELOC(R_LANAI_LO16, 6)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/MSP430.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/MSP430.def
new file mode 100644
index 000000000..96990abf2
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/MSP430.def
@@ -0,0 +1,16 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_MSP430_NONE, 0)
+ELF_RELOC(R_MSP430_32, 1)
+ELF_RELOC(R_MSP430_10_PCREL, 2)
+ELF_RELOC(R_MSP430_16, 3)
+ELF_RELOC(R_MSP430_16_PCREL, 4)
+ELF_RELOC(R_MSP430_16_BYTE, 5)
+ELF_RELOC(R_MSP430_16_PCREL_BYTE, 6)
+ELF_RELOC(R_MSP430_2X_PCREL, 7)
+ELF_RELOC(R_MSP430_RL_PCREL, 8)
+ELF_RELOC(R_MSP430_8, 9)
+ELF_RELOC(R_MSP430_SYM_DIFF, 10)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Mips.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Mips.def
new file mode 100644
index 000000000..bc0088dff
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Mips.def
@@ -0,0 +1,117 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_MIPS_NONE, 0)
+ELF_RELOC(R_MIPS_16, 1)
+ELF_RELOC(R_MIPS_32, 2)
+ELF_RELOC(R_MIPS_REL32, 3)
+ELF_RELOC(R_MIPS_26, 4)
+ELF_RELOC(R_MIPS_HI16, 5)
+ELF_RELOC(R_MIPS_LO16, 6)
+ELF_RELOC(R_MIPS_GPREL16, 7)
+ELF_RELOC(R_MIPS_LITERAL, 8)
+ELF_RELOC(R_MIPS_GOT16, 9)
+ELF_RELOC(R_MIPS_PC16, 10)
+ELF_RELOC(R_MIPS_CALL16, 11)
+ELF_RELOC(R_MIPS_GPREL32, 12)
+ELF_RELOC(R_MIPS_UNUSED1, 13)
+ELF_RELOC(R_MIPS_UNUSED2, 14)
+ELF_RELOC(R_MIPS_UNUSED3, 15)
+ELF_RELOC(R_MIPS_SHIFT5, 16)
+ELF_RELOC(R_MIPS_SHIFT6, 17)
+ELF_RELOC(R_MIPS_64, 18)
+ELF_RELOC(R_MIPS_GOT_DISP, 19)
+ELF_RELOC(R_MIPS_GOT_PAGE, 20)
+ELF_RELOC(R_MIPS_GOT_OFST, 21)
+ELF_RELOC(R_MIPS_GOT_HI16, 22)
+ELF_RELOC(R_MIPS_GOT_LO16, 23)
+ELF_RELOC(R_MIPS_SUB, 24)
+ELF_RELOC(R_MIPS_INSERT_A, 25)
+ELF_RELOC(R_MIPS_INSERT_B, 26)
+ELF_RELOC(R_MIPS_DELETE, 27)
+ELF_RELOC(R_MIPS_HIGHER, 28)
+ELF_RELOC(R_MIPS_HIGHEST, 29)
+ELF_RELOC(R_MIPS_CALL_HI16, 30)
+ELF_RELOC(R_MIPS_CALL_LO16, 31)
+ELF_RELOC(R_MIPS_SCN_DISP, 32)
+ELF_RELOC(R_MIPS_REL16, 33)
+ELF_RELOC(R_MIPS_ADD_IMMEDIATE, 34)
+ELF_RELOC(R_MIPS_PJUMP, 35)
+ELF_RELOC(R_MIPS_RELGOT, 36)
+ELF_RELOC(R_MIPS_JALR, 37)
+ELF_RELOC(R_MIPS_TLS_DTPMOD32, 38)
+ELF_RELOC(R_MIPS_TLS_DTPREL32, 39)
+ELF_RELOC(R_MIPS_TLS_DTPMOD64, 40)
+ELF_RELOC(R_MIPS_TLS_DTPREL64, 41)
+ELF_RELOC(R_MIPS_TLS_GD, 42)
+ELF_RELOC(R_MIPS_TLS_LDM, 43)
+ELF_RELOC(R_MIPS_TLS_DTPREL_HI16, 44)
+ELF_RELOC(R_MIPS_TLS_DTPREL_LO16, 45)
+ELF_RELOC(R_MIPS_TLS_GOTTPREL, 46)
+ELF_RELOC(R_MIPS_TLS_TPREL32, 47)
+ELF_RELOC(R_MIPS_TLS_TPREL64, 48)
+ELF_RELOC(R_MIPS_TLS_TPREL_HI16, 49)
+ELF_RELOC(R_MIPS_TLS_TPREL_LO16, 50)
+ELF_RELOC(R_MIPS_GLOB_DAT, 51)
+ELF_RELOC(R_MIPS_PC21_S2, 60)
+ELF_RELOC(R_MIPS_PC26_S2, 61)
+ELF_RELOC(R_MIPS_PC18_S3, 62)
+ELF_RELOC(R_MIPS_PC19_S2, 63)
+ELF_RELOC(R_MIPS_PCHI16, 64)
+ELF_RELOC(R_MIPS_PCLO16, 65)
+ELF_RELOC(R_MIPS16_26, 100)
+ELF_RELOC(R_MIPS16_GPREL, 101)
+ELF_RELOC(R_MIPS16_GOT16, 102)
+ELF_RELOC(R_MIPS16_CALL16, 103)
+ELF_RELOC(R_MIPS16_HI16, 104)
+ELF_RELOC(R_MIPS16_LO16, 105)
+ELF_RELOC(R_MIPS16_TLS_GD, 106)
+ELF_RELOC(R_MIPS16_TLS_LDM, 107)
+ELF_RELOC(R_MIPS16_TLS_DTPREL_HI16, 108)
+ELF_RELOC(R_MIPS16_TLS_DTPREL_LO16, 109)
+ELF_RELOC(R_MIPS16_TLS_GOTTPREL, 110)
+ELF_RELOC(R_MIPS16_TLS_TPREL_HI16, 111)
+ELF_RELOC(R_MIPS16_TLS_TPREL_LO16, 112)
+ELF_RELOC(R_MIPS_COPY, 126)
+ELF_RELOC(R_MIPS_JUMP_SLOT, 127)
+ELF_RELOC(R_MICROMIPS_26_S1, 133)
+ELF_RELOC(R_MICROMIPS_HI16, 134)
+ELF_RELOC(R_MICROMIPS_LO16, 135)
+ELF_RELOC(R_MICROMIPS_GPREL16, 136)
+ELF_RELOC(R_MICROMIPS_LITERAL, 137)
+ELF_RELOC(R_MICROMIPS_GOT16, 138)
+ELF_RELOC(R_MICROMIPS_PC7_S1, 139)
+ELF_RELOC(R_MICROMIPS_PC10_S1, 140)
+ELF_RELOC(R_MICROMIPS_PC16_S1, 141)
+ELF_RELOC(R_MICROMIPS_CALL16, 142)
+ELF_RELOC(R_MICROMIPS_GOT_DISP, 145)
+ELF_RELOC(R_MICROMIPS_GOT_PAGE, 146)
+ELF_RELOC(R_MICROMIPS_GOT_OFST, 147)
+ELF_RELOC(R_MICROMIPS_GOT_HI16, 148)
+ELF_RELOC(R_MICROMIPS_GOT_LO16, 149)
+ELF_RELOC(R_MICROMIPS_SUB, 150)
+ELF_RELOC(R_MICROMIPS_HIGHER, 151)
+ELF_RELOC(R_MICROMIPS_HIGHEST, 152)
+ELF_RELOC(R_MICROMIPS_CALL_HI16, 153)
+ELF_RELOC(R_MICROMIPS_CALL_LO16, 154)
+ELF_RELOC(R_MICROMIPS_SCN_DISP, 155)
+ELF_RELOC(R_MICROMIPS_JALR, 156)
+ELF_RELOC(R_MICROMIPS_HI0_LO16, 157)
+ELF_RELOC(R_MICROMIPS_TLS_GD, 162)
+ELF_RELOC(R_MICROMIPS_TLS_LDM, 163)
+ELF_RELOC(R_MICROMIPS_TLS_DTPREL_HI16, 164)
+ELF_RELOC(R_MICROMIPS_TLS_DTPREL_LO16, 165)
+ELF_RELOC(R_MICROMIPS_TLS_GOTTPREL, 166)
+ELF_RELOC(R_MICROMIPS_TLS_TPREL_HI16, 169)
+ELF_RELOC(R_MICROMIPS_TLS_TPREL_LO16, 170)
+ELF_RELOC(R_MICROMIPS_GPREL7_S2, 172)
+ELF_RELOC(R_MICROMIPS_PC23_S2, 173)
+ELF_RELOC(R_MICROMIPS_PC21_S1, 174)
+ELF_RELOC(R_MICROMIPS_PC26_S1, 175)
+ELF_RELOC(R_MICROMIPS_PC18_S3, 176)
+ELF_RELOC(R_MICROMIPS_PC19_S2, 177)
+ELF_RELOC(R_MIPS_NUM, 218)
+ELF_RELOC(R_MIPS_PC32, 248)
+ELF_RELOC(R_MIPS_EH, 249)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC.def
new file mode 100644
index 000000000..28036889c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC.def
@@ -0,0 +1,156 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// glibc's PowerPC asm/sigcontext.h, when compiling for PPC64, has the
+// unfortunate behavior of including asm/elf.h, which defines R_PPC_NONE, etc.
+// to their corresponding integer values. As a result, we need to undef them
+// here before continuing.
+
+#undef R_PPC_NONE
+#undef R_PPC_ADDR32
+#undef R_PPC_ADDR24
+#undef R_PPC_ADDR16
+#undef R_PPC_ADDR16_LO
+#undef R_PPC_ADDR16_HI
+#undef R_PPC_ADDR16_HA
+#undef R_PPC_ADDR14
+#undef R_PPC_ADDR14_BRTAKEN
+#undef R_PPC_ADDR14_BRNTAKEN
+#undef R_PPC_REL24
+#undef R_PPC_REL14
+#undef R_PPC_REL14_BRTAKEN
+#undef R_PPC_REL14_BRNTAKEN
+#undef R_PPC_GOT16
+#undef R_PPC_GOT16_LO
+#undef R_PPC_GOT16_HI
+#undef R_PPC_GOT16_HA
+#undef R_PPC_PLTREL24
+#undef R_PPC_COPY
+#undef R_PPC_GLOB_DAT
+#undef R_PPC_JMP_SLOT
+#undef R_PPC_RELATIVE
+#undef R_PPC_LOCAL24PC
+#undef R_PPC_UADDR32
+#undef R_PPC_UADDR16
+#undef R_PPC_REL32
+#undef R_PPC_PLT32
+#undef R_PPC_PLTREL32
+#undef R_PPC_PLT16_LO
+#undef R_PPC_PLT16_HI
+#undef R_PPC_PLT16_HA
+#undef R_PPC_SDAREL16
+#undef R_PPC_SECTOFF
+#undef R_PPC_SECTOFF_LO
+#undef R_PPC_SECTOFF_HI
+#undef R_PPC_SECTOFF_HA
+#undef R_PPC_ADDR30
+#undef R_PPC_TLS
+#undef R_PPC_DTPMOD32
+#undef R_PPC_TPREL16
+#undef R_PPC_TPREL16_LO
+#undef R_PPC_TPREL16_HI
+#undef R_PPC_TPREL16_HA
+#undef R_PPC_TPREL32
+#undef R_PPC_DTPREL16
+#undef R_PPC_DTPREL16_LO
+#undef R_PPC_DTPREL16_HI
+#undef R_PPC_DTPREL16_HA
+#undef R_PPC_DTPREL32
+#undef R_PPC_GOT_TLSGD16
+#undef R_PPC_GOT_TLSGD16_LO
+#undef R_PPC_GOT_TLSGD16_HI
+#undef R_PPC_GOT_TLSGD16_HA
+#undef R_PPC_GOT_TLSLD16
+#undef R_PPC_GOT_TLSLD16_LO
+#undef R_PPC_GOT_TLSLD16_HI
+#undef R_PPC_GOT_TLSLD16_HA
+#undef R_PPC_GOT_TPREL16
+#undef R_PPC_GOT_TPREL16_LO
+#undef R_PPC_GOT_TPREL16_HI
+#undef R_PPC_GOT_TPREL16_HA
+#undef R_PPC_GOT_DTPREL16
+#undef R_PPC_GOT_DTPREL16_LO
+#undef R_PPC_GOT_DTPREL16_HI
+#undef R_PPC_GOT_DTPREL16_HA
+#undef R_PPC_TLSGD
+#undef R_PPC_TLSLD
+#undef R_PPC_REL16
+#undef R_PPC_REL16_LO
+#undef R_PPC_REL16_HI
+#undef R_PPC_REL16_HA
+
+ELF_RELOC(R_PPC_NONE, 0) /* No relocation. */
+ELF_RELOC(R_PPC_ADDR32, 1)
+ELF_RELOC(R_PPC_ADDR24, 2)
+ELF_RELOC(R_PPC_ADDR16, 3)
+ELF_RELOC(R_PPC_ADDR16_LO, 4)
+ELF_RELOC(R_PPC_ADDR16_HI, 5)
+ELF_RELOC(R_PPC_ADDR16_HA, 6)
+ELF_RELOC(R_PPC_ADDR14, 7)
+ELF_RELOC(R_PPC_ADDR14_BRTAKEN, 8)
+ELF_RELOC(R_PPC_ADDR14_BRNTAKEN, 9)
+ELF_RELOC(R_PPC_REL24, 10)
+ELF_RELOC(R_PPC_REL14, 11)
+ELF_RELOC(R_PPC_REL14_BRTAKEN, 12)
+ELF_RELOC(R_PPC_REL14_BRNTAKEN, 13)
+ELF_RELOC(R_PPC_GOT16, 14)
+ELF_RELOC(R_PPC_GOT16_LO, 15)
+ELF_RELOC(R_PPC_GOT16_HI, 16)
+ELF_RELOC(R_PPC_GOT16_HA, 17)
+ELF_RELOC(R_PPC_PLTREL24, 18)
+ELF_RELOC(R_PPC_COPY, 19)
+ELF_RELOC(R_PPC_GLOB_DAT, 20)
+ELF_RELOC(R_PPC_JMP_SLOT, 21)
+ELF_RELOC(R_PPC_RELATIVE, 22)
+ELF_RELOC(R_PPC_LOCAL24PC, 23)
+ELF_RELOC(R_PPC_UADDR32, 24)
+ELF_RELOC(R_PPC_UADDR16, 25)
+ELF_RELOC(R_PPC_REL32, 26)
+ELF_RELOC(R_PPC_PLT32, 27)
+ELF_RELOC(R_PPC_PLTREL32, 28)
+ELF_RELOC(R_PPC_PLT16_LO, 29)
+ELF_RELOC(R_PPC_PLT16_HI, 30)
+ELF_RELOC(R_PPC_PLT16_HA, 31)
+ELF_RELOC(R_PPC_SDAREL16, 32)
+ELF_RELOC(R_PPC_SECTOFF, 33)
+ELF_RELOC(R_PPC_SECTOFF_LO, 34)
+ELF_RELOC(R_PPC_SECTOFF_HI, 35)
+ELF_RELOC(R_PPC_SECTOFF_HA, 36)
+ELF_RELOC(R_PPC_ADDR30, 37)
+ELF_RELOC(R_PPC_TLS, 67)
+ELF_RELOC(R_PPC_DTPMOD32, 68)
+ELF_RELOC(R_PPC_TPREL16, 69)
+ELF_RELOC(R_PPC_TPREL16_LO, 70)
+ELF_RELOC(R_PPC_TPREL16_HI, 71)
+ELF_RELOC(R_PPC_TPREL16_HA, 72)
+ELF_RELOC(R_PPC_TPREL32, 73)
+ELF_RELOC(R_PPC_DTPREL16, 74)
+ELF_RELOC(R_PPC_DTPREL16_LO, 75)
+ELF_RELOC(R_PPC_DTPREL16_HI, 76)
+ELF_RELOC(R_PPC_DTPREL16_HA, 77)
+ELF_RELOC(R_PPC_DTPREL32, 78)
+ELF_RELOC(R_PPC_GOT_TLSGD16, 79)
+ELF_RELOC(R_PPC_GOT_TLSGD16_LO, 80)
+ELF_RELOC(R_PPC_GOT_TLSGD16_HI, 81)
+ELF_RELOC(R_PPC_GOT_TLSGD16_HA, 82)
+ELF_RELOC(R_PPC_GOT_TLSLD16, 83)
+ELF_RELOC(R_PPC_GOT_TLSLD16_LO, 84)
+ELF_RELOC(R_PPC_GOT_TLSLD16_HI, 85)
+ELF_RELOC(R_PPC_GOT_TLSLD16_HA, 86)
+ELF_RELOC(R_PPC_GOT_TPREL16, 87)
+ELF_RELOC(R_PPC_GOT_TPREL16_LO, 88)
+ELF_RELOC(R_PPC_GOT_TPREL16_HI, 89)
+ELF_RELOC(R_PPC_GOT_TPREL16_HA, 90)
+ELF_RELOC(R_PPC_GOT_DTPREL16, 91)
+ELF_RELOC(R_PPC_GOT_DTPREL16_LO, 92)
+ELF_RELOC(R_PPC_GOT_DTPREL16_HI, 93)
+ELF_RELOC(R_PPC_GOT_DTPREL16_HA, 94)
+ELF_RELOC(R_PPC_TLSGD, 95)
+ELF_RELOC(R_PPC_TLSLD, 96)
+ELF_RELOC(R_PPC_IRELATIVE, 248)
+ELF_RELOC(R_PPC_REL16, 249)
+ELF_RELOC(R_PPC_REL16_LO, 250)
+ELF_RELOC(R_PPC_REL16_HI, 251)
+ELF_RELOC(R_PPC_REL16_HA, 252)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC64.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC64.def
new file mode 100644
index 000000000..8c5b482f0
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/PowerPC64.def
@@ -0,0 +1,195 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// glibc's PowerPC asm/sigcontext.h, when compiling for PPC64, has the
+// unfortunate behavior of including asm/elf.h, which defines R_PPC_NONE, etc.
+// to their corresponding integer values. As a result, we need to undef them
+// here before continuing.
+
+#undef R_PPC64_NONE
+#undef R_PPC64_ADDR32
+#undef R_PPC64_ADDR24
+#undef R_PPC64_ADDR16
+#undef R_PPC64_ADDR16_LO
+#undef R_PPC64_ADDR16_HI
+#undef R_PPC64_ADDR16_HA
+#undef R_PPC64_ADDR14
+#undef R_PPC64_ADDR14_BRTAKEN
+#undef R_PPC64_ADDR14_BRNTAKEN
+#undef R_PPC64_REL24
+#undef R_PPC64_REL14
+#undef R_PPC64_REL14_BRTAKEN
+#undef R_PPC64_REL14_BRNTAKEN
+#undef R_PPC64_GOT16
+#undef R_PPC64_GOT16_LO
+#undef R_PPC64_GOT16_HI
+#undef R_PPC64_GOT16_HA
+#undef R_PPC64_GLOB_DAT
+#undef R_PPC64_JMP_SLOT
+#undef R_PPC64_RELATIVE
+#undef R_PPC64_REL32
+#undef R_PPC64_ADDR64
+#undef R_PPC64_ADDR16_HIGHER
+#undef R_PPC64_ADDR16_HIGHERA
+#undef R_PPC64_ADDR16_HIGHEST
+#undef R_PPC64_ADDR16_HIGHESTA
+#undef R_PPC64_REL64
+#undef R_PPC64_TOC16
+#undef R_PPC64_TOC16_LO
+#undef R_PPC64_TOC16_HI
+#undef R_PPC64_TOC16_HA
+#undef R_PPC64_TOC
+#undef R_PPC64_ADDR16_DS
+#undef R_PPC64_ADDR16_LO_DS
+#undef R_PPC64_GOT16_DS
+#undef R_PPC64_GOT16_LO_DS
+#undef R_PPC64_TOC16_DS
+#undef R_PPC64_TOC16_LO_DS
+#undef R_PPC64_TLS
+#undef R_PPC64_DTPMOD64
+#undef R_PPC64_TPREL16
+#undef R_PPC64_TPREL16_LO
+#undef R_PPC64_TPREL16_HI
+#undef R_PPC64_TPREL16_HA
+#undef R_PPC64_TPREL64
+#undef R_PPC64_DTPREL16
+#undef R_PPC64_DTPREL16_LO
+#undef R_PPC64_DTPREL16_HI
+#undef R_PPC64_DTPREL16_HA
+#undef R_PPC64_DTPREL64
+#undef R_PPC64_GOT_TLSGD16
+#undef R_PPC64_GOT_TLSGD16_LO
+#undef R_PPC64_GOT_TLSGD16_HI
+#undef R_PPC64_GOT_TLSGD16_HA
+#undef R_PPC64_GOT_TLSLD16
+#undef R_PPC64_GOT_TLSLD16_LO
+#undef R_PPC64_GOT_TLSLD16_HI
+#undef R_PPC64_GOT_TLSLD16_HA
+#undef R_PPC64_GOT_TPREL16_DS
+#undef R_PPC64_GOT_TPREL16_LO_DS
+#undef R_PPC64_GOT_TPREL16_HI
+#undef R_PPC64_GOT_TPREL16_HA
+#undef R_PPC64_GOT_DTPREL16_DS
+#undef R_PPC64_GOT_DTPREL16_LO_DS
+#undef R_PPC64_GOT_DTPREL16_HI
+#undef R_PPC64_GOT_DTPREL16_HA
+#undef R_PPC64_TPREL16_DS
+#undef R_PPC64_TPREL16_LO_DS
+#undef R_PPC64_TPREL16_HIGHER
+#undef R_PPC64_TPREL16_HIGHERA
+#undef R_PPC64_TPREL16_HIGHEST
+#undef R_PPC64_TPREL16_HIGHESTA
+#undef R_PPC64_DTPREL16_DS
+#undef R_PPC64_DTPREL16_LO_DS
+#undef R_PPC64_DTPREL16_HIGHER
+#undef R_PPC64_DTPREL16_HIGHERA
+#undef R_PPC64_DTPREL16_HIGHEST
+#undef R_PPC64_DTPREL16_HIGHESTA
+#undef R_PPC64_TLSGD
+#undef R_PPC64_TLSLD
+#undef R_PPC64_ADDR16_HIGH
+#undef R_PPC64_ADDR16_HIGHA
+#undef R_PPC64_TPREL16_HIGH
+#undef R_PPC64_TPREL16_HIGHA
+#undef R_PPC64_DTPREL16_HIGH
+#undef R_PPC64_DTPREL16_HIGHA
+#undef R_PPC64_IRELATIVE
+#undef R_PPC64_REL16
+#undef R_PPC64_REL16_LO
+#undef R_PPC64_REL16_HI
+#undef R_PPC64_REL16_HA
+
+ELF_RELOC(R_PPC64_NONE, 0)
+ELF_RELOC(R_PPC64_ADDR32, 1)
+ELF_RELOC(R_PPC64_ADDR24, 2)
+ELF_RELOC(R_PPC64_ADDR16, 3)
+ELF_RELOC(R_PPC64_ADDR16_LO, 4)
+ELF_RELOC(R_PPC64_ADDR16_HI, 5)
+ELF_RELOC(R_PPC64_ADDR16_HA, 6)
+ELF_RELOC(R_PPC64_ADDR14, 7)
+ELF_RELOC(R_PPC64_ADDR14_BRTAKEN, 8)
+ELF_RELOC(R_PPC64_ADDR14_BRNTAKEN, 9)
+ELF_RELOC(R_PPC64_REL24, 10)
+ELF_RELOC(R_PPC64_REL14, 11)
+ELF_RELOC(R_PPC64_REL14_BRTAKEN, 12)
+ELF_RELOC(R_PPC64_REL14_BRNTAKEN, 13)
+ELF_RELOC(R_PPC64_GOT16, 14)
+ELF_RELOC(R_PPC64_GOT16_LO, 15)
+ELF_RELOC(R_PPC64_GOT16_HI, 16)
+ELF_RELOC(R_PPC64_GOT16_HA, 17)
+ELF_RELOC(R_PPC64_GLOB_DAT, 20)
+ELF_RELOC(R_PPC64_JMP_SLOT, 21)
+ELF_RELOC(R_PPC64_RELATIVE, 22)
+ELF_RELOC(R_PPC64_REL32, 26)
+ELF_RELOC(R_PPC64_ADDR64, 38)
+ELF_RELOC(R_PPC64_ADDR16_HIGHER, 39)
+ELF_RELOC(R_PPC64_ADDR16_HIGHERA, 40)
+ELF_RELOC(R_PPC64_ADDR16_HIGHEST, 41)
+ELF_RELOC(R_PPC64_ADDR16_HIGHESTA, 42)
+ELF_RELOC(R_PPC64_REL64, 44)
+ELF_RELOC(R_PPC64_TOC16, 47)
+ELF_RELOC(R_PPC64_TOC16_LO, 48)
+ELF_RELOC(R_PPC64_TOC16_HI, 49)
+ELF_RELOC(R_PPC64_TOC16_HA, 50)
+ELF_RELOC(R_PPC64_TOC, 51)
+ELF_RELOC(R_PPC64_ADDR16_DS, 56)
+ELF_RELOC(R_PPC64_ADDR16_LO_DS, 57)
+ELF_RELOC(R_PPC64_GOT16_DS, 58)
+ELF_RELOC(R_PPC64_GOT16_LO_DS, 59)
+ELF_RELOC(R_PPC64_TOC16_DS, 63)
+ELF_RELOC(R_PPC64_TOC16_LO_DS, 64)
+ELF_RELOC(R_PPC64_TLS, 67)
+ELF_RELOC(R_PPC64_DTPMOD64, 68)
+ELF_RELOC(R_PPC64_TPREL16, 69)
+ELF_RELOC(R_PPC64_TPREL16_LO, 70)
+ELF_RELOC(R_PPC64_TPREL16_HI, 71)
+ELF_RELOC(R_PPC64_TPREL16_HA, 72)
+ELF_RELOC(R_PPC64_TPREL64, 73)
+ELF_RELOC(R_PPC64_DTPREL16, 74)
+ELF_RELOC(R_PPC64_DTPREL16_LO, 75)
+ELF_RELOC(R_PPC64_DTPREL16_HI, 76)
+ELF_RELOC(R_PPC64_DTPREL16_HA, 77)
+ELF_RELOC(R_PPC64_DTPREL64, 78)
+ELF_RELOC(R_PPC64_GOT_TLSGD16, 79)
+ELF_RELOC(R_PPC64_GOT_TLSGD16_LO, 80)
+ELF_RELOC(R_PPC64_GOT_TLSGD16_HI, 81)
+ELF_RELOC(R_PPC64_GOT_TLSGD16_HA, 82)
+ELF_RELOC(R_PPC64_GOT_TLSLD16, 83)
+ELF_RELOC(R_PPC64_GOT_TLSLD16_LO, 84)
+ELF_RELOC(R_PPC64_GOT_TLSLD16_HI, 85)
+ELF_RELOC(R_PPC64_GOT_TLSLD16_HA, 86)
+ELF_RELOC(R_PPC64_GOT_TPREL16_DS, 87)
+ELF_RELOC(R_PPC64_GOT_TPREL16_LO_DS, 88)
+ELF_RELOC(R_PPC64_GOT_TPREL16_HI, 89)
+ELF_RELOC(R_PPC64_GOT_TPREL16_HA, 90)
+ELF_RELOC(R_PPC64_GOT_DTPREL16_DS, 91)
+ELF_RELOC(R_PPC64_GOT_DTPREL16_LO_DS, 92)
+ELF_RELOC(R_PPC64_GOT_DTPREL16_HI, 93)
+ELF_RELOC(R_PPC64_GOT_DTPREL16_HA, 94)
+ELF_RELOC(R_PPC64_TPREL16_DS, 95)
+ELF_RELOC(R_PPC64_TPREL16_LO_DS, 96)
+ELF_RELOC(R_PPC64_TPREL16_HIGHER, 97)
+ELF_RELOC(R_PPC64_TPREL16_HIGHERA, 98)
+ELF_RELOC(R_PPC64_TPREL16_HIGHEST, 99)
+ELF_RELOC(R_PPC64_TPREL16_HIGHESTA, 100)
+ELF_RELOC(R_PPC64_DTPREL16_DS, 101)
+ELF_RELOC(R_PPC64_DTPREL16_LO_DS, 102)
+ELF_RELOC(R_PPC64_DTPREL16_HIGHER, 103)
+ELF_RELOC(R_PPC64_DTPREL16_HIGHERA, 104)
+ELF_RELOC(R_PPC64_DTPREL16_HIGHEST, 105)
+ELF_RELOC(R_PPC64_DTPREL16_HIGHESTA, 106)
+ELF_RELOC(R_PPC64_TLSGD, 107)
+ELF_RELOC(R_PPC64_TLSLD, 108)
+ELF_RELOC(R_PPC64_ADDR16_HIGH, 110)
+ELF_RELOC(R_PPC64_ADDR16_HIGHA, 111)
+ELF_RELOC(R_PPC64_TPREL16_HIGH, 112)
+ELF_RELOC(R_PPC64_TPREL16_HIGHA, 113)
+ELF_RELOC(R_PPC64_DTPREL16_HIGH, 114)
+ELF_RELOC(R_PPC64_DTPREL16_HIGHA, 115)
+ELF_RELOC(R_PPC64_IRELATIVE, 248)
+ELF_RELOC(R_PPC64_REL16, 249)
+ELF_RELOC(R_PPC64_REL16_LO, 250)
+ELF_RELOC(R_PPC64_REL16_HI, 251)
+ELF_RELOC(R_PPC64_REL16_HA, 252)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/RISCV.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/RISCV.def
new file mode 100644
index 000000000..5cc4c0ec3
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/RISCV.def
@@ -0,0 +1,59 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_RISCV_NONE, 0)
+ELF_RELOC(R_RISCV_32, 1)
+ELF_RELOC(R_RISCV_64, 2)
+ELF_RELOC(R_RISCV_RELATIVE, 3)
+ELF_RELOC(R_RISCV_COPY, 4)
+ELF_RELOC(R_RISCV_JUMP_SLOT, 5)
+ELF_RELOC(R_RISCV_TLS_DTPMOD32, 6)
+ELF_RELOC(R_RISCV_TLS_DTPMOD64, 7)
+ELF_RELOC(R_RISCV_TLS_DTPREL32, 8)
+ELF_RELOC(R_RISCV_TLS_DTPREL64, 9)
+ELF_RELOC(R_RISCV_TLS_TPREL32, 10)
+ELF_RELOC(R_RISCV_TLS_TPREL64, 11)
+ELF_RELOC(R_RISCV_BRANCH, 16)
+ELF_RELOC(R_RISCV_JAL, 17)
+ELF_RELOC(R_RISCV_CALL, 18)
+ELF_RELOC(R_RISCV_CALL_PLT, 19)
+ELF_RELOC(R_RISCV_GOT_HI20, 20)
+ELF_RELOC(R_RISCV_TLS_GOT_HI20, 21)
+ELF_RELOC(R_RISCV_TLS_GD_HI20, 22)
+ELF_RELOC(R_RISCV_PCREL_HI20, 23)
+ELF_RELOC(R_RISCV_PCREL_LO12_I, 24)
+ELF_RELOC(R_RISCV_PCREL_LO12_S, 25)
+ELF_RELOC(R_RISCV_HI20, 26)
+ELF_RELOC(R_RISCV_LO12_I, 27)
+ELF_RELOC(R_RISCV_LO12_S, 28)
+ELF_RELOC(R_RISCV_TPREL_HI20, 29)
+ELF_RELOC(R_RISCV_TPREL_LO12_I, 30)
+ELF_RELOC(R_RISCV_TPREL_LO12_S, 31)
+ELF_RELOC(R_RISCV_TPREL_ADD, 32)
+ELF_RELOC(R_RISCV_ADD8, 33)
+ELF_RELOC(R_RISCV_ADD16, 34)
+ELF_RELOC(R_RISCV_ADD32, 35)
+ELF_RELOC(R_RISCV_ADD64, 36)
+ELF_RELOC(R_RISCV_SUB8, 37)
+ELF_RELOC(R_RISCV_SUB16, 38)
+ELF_RELOC(R_RISCV_SUB32, 39)
+ELF_RELOC(R_RISCV_SUB64, 40)
+ELF_RELOC(R_RISCV_GNU_VTINHERIT, 41)
+ELF_RELOC(R_RISCV_GNU_VTENTRY, 42)
+ELF_RELOC(R_RISCV_ALIGN, 43)
+ELF_RELOC(R_RISCV_RVC_BRANCH, 44)
+ELF_RELOC(R_RISCV_RVC_JUMP, 45)
+ELF_RELOC(R_RISCV_RVC_LUI, 46)
+ELF_RELOC(R_RISCV_GPREL_I, 47)
+ELF_RELOC(R_RISCV_GPREL_S, 48)
+ELF_RELOC(R_RISCV_TPREL_I, 49)
+ELF_RELOC(R_RISCV_TPREL_S, 50)
+ELF_RELOC(R_RISCV_RELAX, 51)
+ELF_RELOC(R_RISCV_SUB6, 52)
+ELF_RELOC(R_RISCV_SET6, 53)
+ELF_RELOC(R_RISCV_SET8, 54)
+ELF_RELOC(R_RISCV_SET16, 55)
+ELF_RELOC(R_RISCV_SET32, 56)
+ELF_RELOC(R_RISCV_32_PCREL, 57)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Sparc.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Sparc.def
new file mode 100644
index 000000000..7e01a4a8a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/Sparc.def
@@ -0,0 +1,89 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_SPARC_NONE, 0)
+ELF_RELOC(R_SPARC_8, 1)
+ELF_RELOC(R_SPARC_16, 2)
+ELF_RELOC(R_SPARC_32, 3)
+ELF_RELOC(R_SPARC_DISP8, 4)
+ELF_RELOC(R_SPARC_DISP16, 5)
+ELF_RELOC(R_SPARC_DISP32, 6)
+ELF_RELOC(R_SPARC_WDISP30, 7)
+ELF_RELOC(R_SPARC_WDISP22, 8)
+ELF_RELOC(R_SPARC_HI22, 9)
+ELF_RELOC(R_SPARC_22, 10)
+ELF_RELOC(R_SPARC_13, 11)
+ELF_RELOC(R_SPARC_LO10, 12)
+ELF_RELOC(R_SPARC_GOT10, 13)
+ELF_RELOC(R_SPARC_GOT13, 14)
+ELF_RELOC(R_SPARC_GOT22, 15)
+ELF_RELOC(R_SPARC_PC10, 16)
+ELF_RELOC(R_SPARC_PC22, 17)
+ELF_RELOC(R_SPARC_WPLT30, 18)
+ELF_RELOC(R_SPARC_COPY, 19)
+ELF_RELOC(R_SPARC_GLOB_DAT, 20)
+ELF_RELOC(R_SPARC_JMP_SLOT, 21)
+ELF_RELOC(R_SPARC_RELATIVE, 22)
+ELF_RELOC(R_SPARC_UA32, 23)
+ELF_RELOC(R_SPARC_PLT32, 24)
+ELF_RELOC(R_SPARC_HIPLT22, 25)
+ELF_RELOC(R_SPARC_LOPLT10, 26)
+ELF_RELOC(R_SPARC_PCPLT32, 27)
+ELF_RELOC(R_SPARC_PCPLT22, 28)
+ELF_RELOC(R_SPARC_PCPLT10, 29)
+ELF_RELOC(R_SPARC_10, 30)
+ELF_RELOC(R_SPARC_11, 31)
+ELF_RELOC(R_SPARC_64, 32)
+ELF_RELOC(R_SPARC_OLO10, 33)
+ELF_RELOC(R_SPARC_HH22, 34)
+ELF_RELOC(R_SPARC_HM10, 35)
+ELF_RELOC(R_SPARC_LM22, 36)
+ELF_RELOC(R_SPARC_PC_HH22, 37)
+ELF_RELOC(R_SPARC_PC_HM10, 38)
+ELF_RELOC(R_SPARC_PC_LM22, 39)
+ELF_RELOC(R_SPARC_WDISP16, 40)
+ELF_RELOC(R_SPARC_WDISP19, 41)
+ELF_RELOC(R_SPARC_7, 43)
+ELF_RELOC(R_SPARC_5, 44)
+ELF_RELOC(R_SPARC_6, 45)
+ELF_RELOC(R_SPARC_DISP64, 46)
+ELF_RELOC(R_SPARC_PLT64, 47)
+ELF_RELOC(R_SPARC_HIX22, 48)
+ELF_RELOC(R_SPARC_LOX10, 49)
+ELF_RELOC(R_SPARC_H44, 50)
+ELF_RELOC(R_SPARC_M44, 51)
+ELF_RELOC(R_SPARC_L44, 52)
+ELF_RELOC(R_SPARC_REGISTER, 53)
+ELF_RELOC(R_SPARC_UA64, 54)
+ELF_RELOC(R_SPARC_UA16, 55)
+ELF_RELOC(R_SPARC_TLS_GD_HI22, 56)
+ELF_RELOC(R_SPARC_TLS_GD_LO10, 57)
+ELF_RELOC(R_SPARC_TLS_GD_ADD, 58)
+ELF_RELOC(R_SPARC_TLS_GD_CALL, 59)
+ELF_RELOC(R_SPARC_TLS_LDM_HI22, 60)
+ELF_RELOC(R_SPARC_TLS_LDM_LO10, 61)
+ELF_RELOC(R_SPARC_TLS_LDM_ADD, 62)
+ELF_RELOC(R_SPARC_TLS_LDM_CALL, 63)
+ELF_RELOC(R_SPARC_TLS_LDO_HIX22, 64)
+ELF_RELOC(R_SPARC_TLS_LDO_LOX10, 65)
+ELF_RELOC(R_SPARC_TLS_LDO_ADD, 66)
+ELF_RELOC(R_SPARC_TLS_IE_HI22, 67)
+ELF_RELOC(R_SPARC_TLS_IE_LO10, 68)
+ELF_RELOC(R_SPARC_TLS_IE_LD, 69)
+ELF_RELOC(R_SPARC_TLS_IE_LDX, 70)
+ELF_RELOC(R_SPARC_TLS_IE_ADD, 71)
+ELF_RELOC(R_SPARC_TLS_LE_HIX22, 72)
+ELF_RELOC(R_SPARC_TLS_LE_LOX10, 73)
+ELF_RELOC(R_SPARC_TLS_DTPMOD32, 74)
+ELF_RELOC(R_SPARC_TLS_DTPMOD64, 75)
+ELF_RELOC(R_SPARC_TLS_DTPOFF32, 76)
+ELF_RELOC(R_SPARC_TLS_DTPOFF64, 77)
+ELF_RELOC(R_SPARC_TLS_TPOFF32, 78)
+ELF_RELOC(R_SPARC_TLS_TPOFF64, 79)
+ELF_RELOC(R_SPARC_GOTDATA_HIX22, 80)
+ELF_RELOC(R_SPARC_GOTDATA_LOX10, 81)
+ELF_RELOC(R_SPARC_GOTDATA_OP_HIX22, 82)
+ELF_RELOC(R_SPARC_GOTDATA_OP_LOX10, 83)
+ELF_RELOC(R_SPARC_GOTDATA_OP, 84)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/SystemZ.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/SystemZ.def
new file mode 100644
index 000000000..d6c0b79d4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/SystemZ.def
@@ -0,0 +1,71 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_390_NONE, 0)
+ELF_RELOC(R_390_8, 1)
+ELF_RELOC(R_390_12, 2)
+ELF_RELOC(R_390_16, 3)
+ELF_RELOC(R_390_32, 4)
+ELF_RELOC(R_390_PC32, 5)
+ELF_RELOC(R_390_GOT12, 6)
+ELF_RELOC(R_390_GOT32, 7)
+ELF_RELOC(R_390_PLT32, 8)
+ELF_RELOC(R_390_COPY, 9)
+ELF_RELOC(R_390_GLOB_DAT, 10)
+ELF_RELOC(R_390_JMP_SLOT, 11)
+ELF_RELOC(R_390_RELATIVE, 12)
+ELF_RELOC(R_390_GOTOFF, 13)
+ELF_RELOC(R_390_GOTPC, 14)
+ELF_RELOC(R_390_GOT16, 15)
+ELF_RELOC(R_390_PC16, 16)
+ELF_RELOC(R_390_PC16DBL, 17)
+ELF_RELOC(R_390_PLT16DBL, 18)
+ELF_RELOC(R_390_PC32DBL, 19)
+ELF_RELOC(R_390_PLT32DBL, 20)
+ELF_RELOC(R_390_GOTPCDBL, 21)
+ELF_RELOC(R_390_64, 22)
+ELF_RELOC(R_390_PC64, 23)
+ELF_RELOC(R_390_GOT64, 24)
+ELF_RELOC(R_390_PLT64, 25)
+ELF_RELOC(R_390_GOTENT, 26)
+ELF_RELOC(R_390_GOTOFF16, 27)
+ELF_RELOC(R_390_GOTOFF64, 28)
+ELF_RELOC(R_390_GOTPLT12, 29)
+ELF_RELOC(R_390_GOTPLT16, 30)
+ELF_RELOC(R_390_GOTPLT32, 31)
+ELF_RELOC(R_390_GOTPLT64, 32)
+ELF_RELOC(R_390_GOTPLTENT, 33)
+ELF_RELOC(R_390_PLTOFF16, 34)
+ELF_RELOC(R_390_PLTOFF32, 35)
+ELF_RELOC(R_390_PLTOFF64, 36)
+ELF_RELOC(R_390_TLS_LOAD, 37)
+ELF_RELOC(R_390_TLS_GDCALL, 38)
+ELF_RELOC(R_390_TLS_LDCALL, 39)
+ELF_RELOC(R_390_TLS_GD32, 40)
+ELF_RELOC(R_390_TLS_GD64, 41)
+ELF_RELOC(R_390_TLS_GOTIE12, 42)
+ELF_RELOC(R_390_TLS_GOTIE32, 43)
+ELF_RELOC(R_390_TLS_GOTIE64, 44)
+ELF_RELOC(R_390_TLS_LDM32, 45)
+ELF_RELOC(R_390_TLS_LDM64, 46)
+ELF_RELOC(R_390_TLS_IE32, 47)
+ELF_RELOC(R_390_TLS_IE64, 48)
+ELF_RELOC(R_390_TLS_IEENT, 49)
+ELF_RELOC(R_390_TLS_LE32, 50)
+ELF_RELOC(R_390_TLS_LE64, 51)
+ELF_RELOC(R_390_TLS_LDO32, 52)
+ELF_RELOC(R_390_TLS_LDO64, 53)
+ELF_RELOC(R_390_TLS_DTPMOD, 54)
+ELF_RELOC(R_390_TLS_DTPOFF, 55)
+ELF_RELOC(R_390_TLS_TPOFF, 56)
+ELF_RELOC(R_390_20, 57)
+ELF_RELOC(R_390_GOT20, 58)
+ELF_RELOC(R_390_GOTPLT20, 59)
+ELF_RELOC(R_390_TLS_GOTIE20, 60)
+ELF_RELOC(R_390_IRELATIVE, 61)
+ELF_RELOC(R_390_PC12DBL, 62)
+ELF_RELOC(R_390_PLT12DBL, 63)
+ELF_RELOC(R_390_PC24DBL, 64)
+ELF_RELOC(R_390_PLT24DBL, 65)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/i386.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/i386.def
new file mode 100644
index 000000000..1d28cf595
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/i386.def
@@ -0,0 +1,47 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+// TODO: this is just a subset
+ELF_RELOC(R_386_NONE, 0)
+ELF_RELOC(R_386_32, 1)
+ELF_RELOC(R_386_PC32, 2)
+ELF_RELOC(R_386_GOT32, 3)
+ELF_RELOC(R_386_PLT32, 4)
+ELF_RELOC(R_386_COPY, 5)
+ELF_RELOC(R_386_GLOB_DAT, 6)
+ELF_RELOC(R_386_JUMP_SLOT, 7)
+ELF_RELOC(R_386_RELATIVE, 8)
+ELF_RELOC(R_386_GOTOFF, 9)
+ELF_RELOC(R_386_GOTPC, 10)
+ELF_RELOC(R_386_32PLT, 11)
+ELF_RELOC(R_386_TLS_TPOFF, 14)
+ELF_RELOC(R_386_TLS_IE, 15)
+ELF_RELOC(R_386_TLS_GOTIE, 16)
+ELF_RELOC(R_386_TLS_LE, 17)
+ELF_RELOC(R_386_TLS_GD, 18)
+ELF_RELOC(R_386_TLS_LDM, 19)
+ELF_RELOC(R_386_16, 20)
+ELF_RELOC(R_386_PC16, 21)
+ELF_RELOC(R_386_8, 22)
+ELF_RELOC(R_386_PC8, 23)
+ELF_RELOC(R_386_TLS_GD_32, 24)
+ELF_RELOC(R_386_TLS_GD_PUSH, 25)
+ELF_RELOC(R_386_TLS_GD_CALL, 26)
+ELF_RELOC(R_386_TLS_GD_POP, 27)
+ELF_RELOC(R_386_TLS_LDM_32, 28)
+ELF_RELOC(R_386_TLS_LDM_PUSH, 29)
+ELF_RELOC(R_386_TLS_LDM_CALL, 30)
+ELF_RELOC(R_386_TLS_LDM_POP, 31)
+ELF_RELOC(R_386_TLS_LDO_32, 32)
+ELF_RELOC(R_386_TLS_IE_32, 33)
+ELF_RELOC(R_386_TLS_LE_32, 34)
+ELF_RELOC(R_386_TLS_DTPMOD32, 35)
+ELF_RELOC(R_386_TLS_DTPOFF32, 36)
+ELF_RELOC(R_386_TLS_TPOFF32, 37)
+ELF_RELOC(R_386_TLS_GOTDESC, 39)
+ELF_RELOC(R_386_TLS_DESC_CALL, 40)
+ELF_RELOC(R_386_TLS_DESC, 41)
+ELF_RELOC(R_386_IRELATIVE, 42)
+ELF_RELOC(R_386_GOT32X, 43)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/x86_64.def b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/x86_64.def
new file mode 100644
index 000000000..18fdcf947
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/ELFRelocs/x86_64.def
@@ -0,0 +1,45 @@
+
+#ifndef ELF_RELOC
+#error "ELF_RELOC must be defined"
+#endif
+
+ELF_RELOC(R_X86_64_NONE, 0)
+ELF_RELOC(R_X86_64_64, 1)
+ELF_RELOC(R_X86_64_PC32, 2)
+ELF_RELOC(R_X86_64_GOT32, 3)
+ELF_RELOC(R_X86_64_PLT32, 4)
+ELF_RELOC(R_X86_64_COPY, 5)
+ELF_RELOC(R_X86_64_GLOB_DAT, 6)
+ELF_RELOC(R_X86_64_JUMP_SLOT, 7)
+ELF_RELOC(R_X86_64_RELATIVE, 8)
+ELF_RELOC(R_X86_64_GOTPCREL, 9)
+ELF_RELOC(R_X86_64_32, 10)
+ELF_RELOC(R_X86_64_32S, 11)
+ELF_RELOC(R_X86_64_16, 12)
+ELF_RELOC(R_X86_64_PC16, 13)
+ELF_RELOC(R_X86_64_8, 14)
+ELF_RELOC(R_X86_64_PC8, 15)
+ELF_RELOC(R_X86_64_DTPMOD64, 16)
+ELF_RELOC(R_X86_64_DTPOFF64, 17)
+ELF_RELOC(R_X86_64_TPOFF64, 18)
+ELF_RELOC(R_X86_64_TLSGD, 19)
+ELF_RELOC(R_X86_64_TLSLD, 20)
+ELF_RELOC(R_X86_64_DTPOFF32, 21)
+ELF_RELOC(R_X86_64_GOTTPOFF, 22)
+ELF_RELOC(R_X86_64_TPOFF32, 23)
+ELF_RELOC(R_X86_64_PC64, 24)
+ELF_RELOC(R_X86_64_GOTOFF64, 25)
+ELF_RELOC(R_X86_64_GOTPC32, 26)
+ELF_RELOC(R_X86_64_GOT64, 27)
+ELF_RELOC(R_X86_64_GOTPCREL64, 28)
+ELF_RELOC(R_X86_64_GOTPC64, 29)
+ELF_RELOC(R_X86_64_GOTPLT64, 30)
+ELF_RELOC(R_X86_64_PLTOFF64, 31)
+ELF_RELOC(R_X86_64_SIZE32, 32)
+ELF_RELOC(R_X86_64_SIZE64, 33)
+ELF_RELOC(R_X86_64_GOTPC32_TLSDESC, 34)
+ELF_RELOC(R_X86_64_TLSDESC_CALL, 35)
+ELF_RELOC(R_X86_64_TLSDESC, 36)
+ELF_RELOC(R_X86_64_IRELATIVE, 37)
+ELF_RELOC(R_X86_64_GOTPCRELX, 41)
+ELF_RELOC(R_X86_64_REX_GOTPCRELX, 42)
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/MachO.def b/third_party/llvm-project/include/llvm/BinaryFormat/MachO.def
new file mode 100644
index 000000000..76dcc58ba
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/MachO.def
@@ -0,0 +1,119 @@
+//,,,-- llvm/Support/MachO.def - The MachO file definitions -----*- C++ -*-,,,//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//,,,----------------------------------------------------------------------,,,//
+//
+// Definitions for MachO files
+//
+//,,,----------------------------------------------------------------------,,,//
+
+#ifdef HANDLE_LOAD_COMMAND
+
+HANDLE_LOAD_COMMAND(LC_SEGMENT, 0x00000001u, segment_command)
+HANDLE_LOAD_COMMAND(LC_SYMTAB, 0x00000002u, symtab_command)
+// LC_SYMSEG is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_SYMSEG, 0x00000003u, symseg_command)
+HANDLE_LOAD_COMMAND(LC_THREAD, 0x00000004u, thread_command)
+HANDLE_LOAD_COMMAND(LC_UNIXTHREAD, 0x00000005u, thread_command)
+// LC_LOADFVMLIB is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_LOADFVMLIB, 0x00000006u, fvmlib_command)
+// LC_IDFVMLIB is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_IDFVMLIB, 0x00000007u, fvmlib_command)
+// LC_IDENT is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_IDENT, 0x00000008u, ident_command)
+// LC_FVMFILE is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_FVMFILE, 0x00000009u, fvmfile_command)
+// LC_PREPAGE is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_PREPAGE, 0x0000000Au, load_command)
+HANDLE_LOAD_COMMAND(LC_DYSYMTAB, 0x0000000Bu, dysymtab_command)
+HANDLE_LOAD_COMMAND(LC_LOAD_DYLIB, 0x0000000Cu, dylib_command)
+HANDLE_LOAD_COMMAND(LC_ID_DYLIB, 0x0000000Du, dylib_command)
+HANDLE_LOAD_COMMAND(LC_LOAD_DYLINKER, 0x0000000Eu, dylinker_command)
+HANDLE_LOAD_COMMAND(LC_ID_DYLINKER, 0x0000000Fu, dylinker_command)
+// LC_PREBOUND_DYLIB is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_PREBOUND_DYLIB, 0x00000010u, prebound_dylib_command)
+HANDLE_LOAD_COMMAND(LC_ROUTINES, 0x00000011u, routines_command)
+HANDLE_LOAD_COMMAND(LC_SUB_FRAMEWORK, 0x00000012u, sub_framework_command)
+HANDLE_LOAD_COMMAND(LC_SUB_UMBRELLA, 0x00000013u, sub_umbrella_command)
+HANDLE_LOAD_COMMAND(LC_SUB_CLIENT, 0x00000014u, sub_client_command)
+HANDLE_LOAD_COMMAND(LC_SUB_LIBRARY, 0x00000015u, sub_library_command)
+// LC_TWOLEVEL_HINTS is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_TWOLEVEL_HINTS, 0x00000016u, twolevel_hints_command)
+// LC_PREBIND_CKSUM is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_PREBIND_CKSUM, 0x00000017u, prebind_cksum_command)
+// LC_LOAD_WEAK_DYLIB is obsolete and no longer supported.
+HANDLE_LOAD_COMMAND(LC_LOAD_WEAK_DYLIB, 0x80000018u, dylib_command)
+HANDLE_LOAD_COMMAND(LC_SEGMENT_64, 0x00000019u, segment_command_64)
+HANDLE_LOAD_COMMAND(LC_ROUTINES_64, 0x0000001Au, routines_command_64)
+HANDLE_LOAD_COMMAND(LC_UUID, 0x0000001Bu, uuid_command)
+HANDLE_LOAD_COMMAND(LC_RPATH, 0x8000001Cu, rpath_command)
+HANDLE_LOAD_COMMAND(LC_CODE_SIGNATURE, 0x0000001Du, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_SEGMENT_SPLIT_INFO, 0x0000001Eu, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_REEXPORT_DYLIB, 0x8000001Fu, dylib_command)
+HANDLE_LOAD_COMMAND(LC_LAZY_LOAD_DYLIB, 0x00000020u, dylib_command)
+HANDLE_LOAD_COMMAND(LC_ENCRYPTION_INFO, 0x00000021u, encryption_info_command)
+HANDLE_LOAD_COMMAND(LC_DYLD_INFO, 0x00000022u, dyld_info_command)
+HANDLE_LOAD_COMMAND(LC_DYLD_INFO_ONLY, 0x80000022u, dyld_info_command)
+HANDLE_LOAD_COMMAND(LC_LOAD_UPWARD_DYLIB, 0x80000023u, dylib_command)
+HANDLE_LOAD_COMMAND(LC_VERSION_MIN_MACOSX, 0x00000024u, version_min_command)
+HANDLE_LOAD_COMMAND(LC_VERSION_MIN_IPHONEOS, 0x00000025u, version_min_command)
+HANDLE_LOAD_COMMAND(LC_FUNCTION_STARTS, 0x00000026u, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_DYLD_ENVIRONMENT, 0x00000027u, dylinker_command)
+HANDLE_LOAD_COMMAND(LC_MAIN, 0x80000028u, entry_point_command)
+HANDLE_LOAD_COMMAND(LC_DATA_IN_CODE, 0x00000029u, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_SOURCE_VERSION, 0x0000002Au, source_version_command)
+HANDLE_LOAD_COMMAND(LC_DYLIB_CODE_SIGN_DRS, 0x0000002Bu, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_ENCRYPTION_INFO_64, 0x0000002Cu,
+ encryption_info_command_64)
+HANDLE_LOAD_COMMAND(LC_LINKER_OPTION, 0x0000002Du, linker_option_command)
+HANDLE_LOAD_COMMAND(LC_LINKER_OPTIMIZATION_HINT, 0x0000002Eu, linkedit_data_command)
+HANDLE_LOAD_COMMAND(LC_VERSION_MIN_TVOS, 0x0000002Fu, version_min_command)
+HANDLE_LOAD_COMMAND(LC_VERSION_MIN_WATCHOS, 0x00000030u, version_min_command)
+HANDLE_LOAD_COMMAND(LC_NOTE, 0x00000031u, note_command)
+HANDLE_LOAD_COMMAND(LC_BUILD_VERSION, 0x00000032u, build_version_command)
+
+#endif
+
+#ifdef LOAD_COMMAND_STRUCT
+
+LOAD_COMMAND_STRUCT(dyld_info_command)
+LOAD_COMMAND_STRUCT(dylib_command)
+LOAD_COMMAND_STRUCT(dylinker_command)
+LOAD_COMMAND_STRUCT(dysymtab_command)
+LOAD_COMMAND_STRUCT(encryption_info_command)
+LOAD_COMMAND_STRUCT(encryption_info_command_64)
+LOAD_COMMAND_STRUCT(entry_point_command)
+LOAD_COMMAND_STRUCT(fvmfile_command)
+LOAD_COMMAND_STRUCT(fvmlib_command)
+LOAD_COMMAND_STRUCT(ident_command)
+LOAD_COMMAND_STRUCT(linkedit_data_command)
+LOAD_COMMAND_STRUCT(linker_option_command)
+LOAD_COMMAND_STRUCT(load_command)
+LOAD_COMMAND_STRUCT(prebind_cksum_command)
+LOAD_COMMAND_STRUCT(prebound_dylib_command)
+LOAD_COMMAND_STRUCT(routines_command)
+LOAD_COMMAND_STRUCT(routines_command_64)
+LOAD_COMMAND_STRUCT(rpath_command)
+LOAD_COMMAND_STRUCT(segment_command)
+LOAD_COMMAND_STRUCT(segment_command_64)
+LOAD_COMMAND_STRUCT(source_version_command)
+LOAD_COMMAND_STRUCT(sub_client_command)
+LOAD_COMMAND_STRUCT(sub_framework_command)
+LOAD_COMMAND_STRUCT(sub_library_command)
+LOAD_COMMAND_STRUCT(sub_umbrella_command)
+LOAD_COMMAND_STRUCT(symseg_command)
+LOAD_COMMAND_STRUCT(symtab_command)
+LOAD_COMMAND_STRUCT(thread_command)
+LOAD_COMMAND_STRUCT(twolevel_hints_command)
+LOAD_COMMAND_STRUCT(uuid_command)
+LOAD_COMMAND_STRUCT(version_min_command)
+LOAD_COMMAND_STRUCT(note_command)
+LOAD_COMMAND_STRUCT(build_version_command)
+
+#endif
+
+#undef HANDLE_LOAD_COMMAND
+#undef LOAD_COMMAND_STRUCT
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/MachO.h b/third_party/llvm-project/include/llvm/BinaryFormat/MachO.h
new file mode 100644
index 000000000..fb50e549c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/MachO.h
@@ -0,0 +1,2006 @@
+//===-- llvm/BinaryFormat/MachO.h - The MachO file format -------*- C++/-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines manifest constants for the MachO object file format.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_MACHO_H
+#define LLVM_BINARYFORMAT_MACHO_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/Host.h"
+
+namespace llvm {
+namespace MachO {
+// Enums from <mach-o/loader.h>
+enum : uint32_t {
+ // Constants for the "magic" field in llvm::MachO::mach_header and
+ // llvm::MachO::mach_header_64
+ MH_MAGIC = 0xFEEDFACEu,
+ MH_CIGAM = 0xCEFAEDFEu,
+ MH_MAGIC_64 = 0xFEEDFACFu,
+ MH_CIGAM_64 = 0xCFFAEDFEu,
+ FAT_MAGIC = 0xCAFEBABEu,
+ FAT_CIGAM = 0xBEBAFECAu,
+ FAT_MAGIC_64 = 0xCAFEBABFu,
+ FAT_CIGAM_64 = 0xBFBAFECAu
+};
+
+enum HeaderFileType {
+ // Constants for the "filetype" field in llvm::MachO::mach_header and
+ // llvm::MachO::mach_header_64
+ MH_OBJECT = 0x1u,
+ MH_EXECUTE = 0x2u,
+ MH_FVMLIB = 0x3u,
+ MH_CORE = 0x4u,
+ MH_PRELOAD = 0x5u,
+ MH_DYLIB = 0x6u,
+ MH_DYLINKER = 0x7u,
+ MH_BUNDLE = 0x8u,
+ MH_DYLIB_STUB = 0x9u,
+ MH_DSYM = 0xAu,
+ MH_KEXT_BUNDLE = 0xBu
+};
+
+enum {
+ // Constant bits for the "flags" field in llvm::MachO::mach_header and
+ // llvm::MachO::mach_header_64
+ MH_NOUNDEFS = 0x00000001u,
+ MH_INCRLINK = 0x00000002u,
+ MH_DYLDLINK = 0x00000004u,
+ MH_BINDATLOAD = 0x00000008u,
+ MH_PREBOUND = 0x00000010u,
+ MH_SPLIT_SEGS = 0x00000020u,
+ MH_LAZY_INIT = 0x00000040u,
+ MH_TWOLEVEL = 0x00000080u,
+ MH_FORCE_FLAT = 0x00000100u,
+ MH_NOMULTIDEFS = 0x00000200u,
+ MH_NOFIXPREBINDING = 0x00000400u,
+ MH_PREBINDABLE = 0x00000800u,
+ MH_ALLMODSBOUND = 0x00001000u,
+ MH_SUBSECTIONS_VIA_SYMBOLS = 0x00002000u,
+ MH_CANONICAL = 0x00004000u,
+ MH_WEAK_DEFINES = 0x00008000u,
+ MH_BINDS_TO_WEAK = 0x00010000u,
+ MH_ALLOW_STACK_EXECUTION = 0x00020000u,
+ MH_ROOT_SAFE = 0x00040000u,
+ MH_SETUID_SAFE = 0x00080000u,
+ MH_NO_REEXPORTED_DYLIBS = 0x00100000u,
+ MH_PIE = 0x00200000u,
+ MH_DEAD_STRIPPABLE_DYLIB = 0x00400000u,
+ MH_HAS_TLV_DESCRIPTORS = 0x00800000u,
+ MH_NO_HEAP_EXECUTION = 0x01000000u,
+ MH_APP_EXTENSION_SAFE = 0x02000000u,
+ MH_NLIST_OUTOFSYNC_WITH_DYLDINFO = 0x04000000u
+};
+
+enum : uint32_t {
+ // Flags for the "cmd" field in llvm::MachO::load_command
+ LC_REQ_DYLD = 0x80000000u
+};
+
+#define HANDLE_LOAD_COMMAND(LCName, LCValue, LCStruct) LCName = LCValue,
+
+enum LoadCommandType : uint32_t {
+#include "llvm/BinaryFormat/MachO.def"
+};
+
+#undef HANDLE_LOAD_COMMAND
+
+enum : uint32_t {
+ // Constant bits for the "flags" field in llvm::MachO::segment_command
+ SG_HIGHVM = 0x1u,
+ SG_FVMLIB = 0x2u,
+ SG_NORELOC = 0x4u,
+ SG_PROTECTED_VERSION_1 = 0x8u,
+
+ // Constant masks for the "flags" field in llvm::MachO::section and
+ // llvm::MachO::section_64
+ SECTION_TYPE = 0x000000ffu, // SECTION_TYPE
+ SECTION_ATTRIBUTES = 0xffffff00u, // SECTION_ATTRIBUTES
+ SECTION_ATTRIBUTES_USR = 0xff000000u, // SECTION_ATTRIBUTES_USR
+ SECTION_ATTRIBUTES_SYS = 0x00ffff00u // SECTION_ATTRIBUTES_SYS
+};
+
+/// These are the section type and attributes fields. A MachO section can
+/// have only one Type, but can have any of the attributes specified.
+enum SectionType : uint32_t {
+ // Constant masks for the "flags[7:0]" field in llvm::MachO::section and
+ // llvm::MachO::section_64 (mask "flags" with SECTION_TYPE)
+
+ /// S_REGULAR - Regular section.
+ S_REGULAR = 0x00u,
+ /// S_ZEROFILL - Zero fill on demand section.
+ S_ZEROFILL = 0x01u,
+ /// S_CSTRING_LITERALS - Section with literal C strings.
+ S_CSTRING_LITERALS = 0x02u,
+ /// S_4BYTE_LITERALS - Section with 4 byte literals.
+ S_4BYTE_LITERALS = 0x03u,
+ /// S_8BYTE_LITERALS - Section with 8 byte literals.
+ S_8BYTE_LITERALS = 0x04u,
+ /// S_LITERAL_POINTERS - Section with pointers to literals.
+ S_LITERAL_POINTERS = 0x05u,
+ /// S_NON_LAZY_SYMBOL_POINTERS - Section with non-lazy symbol pointers.
+ S_NON_LAZY_SYMBOL_POINTERS = 0x06u,
+ /// S_LAZY_SYMBOL_POINTERS - Section with lazy symbol pointers.
+ S_LAZY_SYMBOL_POINTERS = 0x07u,
+ /// S_SYMBOL_STUBS - Section with symbol stubs, byte size of stub in
+ /// the Reserved2 field.
+ S_SYMBOL_STUBS = 0x08u,
+ /// S_MOD_INIT_FUNC_POINTERS - Section with only function pointers for
+ /// initialization.
+ S_MOD_INIT_FUNC_POINTERS = 0x09u,
+ /// S_MOD_TERM_FUNC_POINTERS - Section with only function pointers for
+ /// termination.
+ S_MOD_TERM_FUNC_POINTERS = 0x0au,
+ /// S_COALESCED - Section contains symbols that are to be coalesced.
+ S_COALESCED = 0x0bu,
+ /// S_GB_ZEROFILL - Zero fill on demand section (that can be larger than 4
+ /// gigabytes).
+ S_GB_ZEROFILL = 0x0cu,
+ /// S_INTERPOSING - Section with only pairs of function pointers for
+ /// interposing.
+ S_INTERPOSING = 0x0du,
+ /// S_16BYTE_LITERALS - Section with only 16 byte literals.
+ S_16BYTE_LITERALS = 0x0eu,
+ /// S_DTRACE_DOF - Section contains DTrace Object Format.
+ S_DTRACE_DOF = 0x0fu,
+ /// S_LAZY_DYLIB_SYMBOL_POINTERS - Section with lazy symbol pointers to
+ /// lazy loaded dylibs.
+ S_LAZY_DYLIB_SYMBOL_POINTERS = 0x10u,
+ /// S_THREAD_LOCAL_REGULAR - Thread local data section.
+ S_THREAD_LOCAL_REGULAR = 0x11u,
+ /// S_THREAD_LOCAL_ZEROFILL - Thread local zerofill section.
+ S_THREAD_LOCAL_ZEROFILL = 0x12u,
+ /// S_THREAD_LOCAL_VARIABLES - Section with thread local variable
+ /// structure data.
+ S_THREAD_LOCAL_VARIABLES = 0x13u,
+ /// S_THREAD_LOCAL_VARIABLE_POINTERS - Section with pointers to thread
+ /// local structures.
+ S_THREAD_LOCAL_VARIABLE_POINTERS = 0x14u,
+ /// S_THREAD_LOCAL_INIT_FUNCTION_POINTERS - Section with thread local
+ /// variable initialization pointers to functions.
+ S_THREAD_LOCAL_INIT_FUNCTION_POINTERS = 0x15u,
+
+ LAST_KNOWN_SECTION_TYPE = S_THREAD_LOCAL_INIT_FUNCTION_POINTERS
+};
+
+enum : uint32_t {
+ // Constant masks for the "flags[31:24]" field in llvm::MachO::section and
+ // llvm::MachO::section_64 (mask "flags" with SECTION_ATTRIBUTES_USR)
+
+ /// S_ATTR_PURE_INSTRUCTIONS - Section contains only true machine
+ /// instructions.
+ S_ATTR_PURE_INSTRUCTIONS = 0x80000000u,
+ /// S_ATTR_NO_TOC - Section contains coalesced symbols that are not to be
+ /// in a ranlib table of contents.
+ S_ATTR_NO_TOC = 0x40000000u,
+ /// S_ATTR_STRIP_STATIC_SYMS - Ok to strip static symbols in this section
+ /// in files with the MY_DYLDLINK flag.
+ S_ATTR_STRIP_STATIC_SYMS = 0x20000000u,
+ /// S_ATTR_NO_DEAD_STRIP - No dead stripping.
+ S_ATTR_NO_DEAD_STRIP = 0x10000000u,
+ /// S_ATTR_LIVE_SUPPORT - Blocks are live if they reference live blocks.
+ S_ATTR_LIVE_SUPPORT = 0x08000000u,
+ /// S_ATTR_SELF_MODIFYING_CODE - Used with i386 code stubs written on by
+ /// dyld.
+ S_ATTR_SELF_MODIFYING_CODE = 0x04000000u,
+ /// S_ATTR_DEBUG - A debug section.
+ S_ATTR_DEBUG = 0x02000000u,
+
+ // Constant masks for the "flags[23:8]" field in llvm::MachO::section and
+ // llvm::MachO::section_64 (mask "flags" with SECTION_ATTRIBUTES_SYS)
+
+ /// S_ATTR_SOME_INSTRUCTIONS - Section contains some machine instructions.
+ S_ATTR_SOME_INSTRUCTIONS = 0x00000400u,
+ /// S_ATTR_EXT_RELOC - Section has external relocation entries.
+ S_ATTR_EXT_RELOC = 0x00000200u,
+ /// S_ATTR_LOC_RELOC - Section has local relocation entries.
+ S_ATTR_LOC_RELOC = 0x00000100u,
+
+ // Constant masks for the value of an indirect symbol in an indirect
+ // symbol table
+ INDIRECT_SYMBOL_LOCAL = 0x80000000u,
+ INDIRECT_SYMBOL_ABS = 0x40000000u
+};
+
+enum DataRegionType {
+ // Constants for the "kind" field in a data_in_code_entry structure
+ DICE_KIND_DATA = 1u,
+ DICE_KIND_JUMP_TABLE8 = 2u,
+ DICE_KIND_JUMP_TABLE16 = 3u,
+ DICE_KIND_JUMP_TABLE32 = 4u,
+ DICE_KIND_ABS_JUMP_TABLE32 = 5u
+};
+
+enum RebaseType {
+ REBASE_TYPE_POINTER = 1u,
+ REBASE_TYPE_TEXT_ABSOLUTE32 = 2u,
+ REBASE_TYPE_TEXT_PCREL32 = 3u
+};
+
+enum { REBASE_OPCODE_MASK = 0xF0u, REBASE_IMMEDIATE_MASK = 0x0Fu };
+
+enum RebaseOpcode {
+ REBASE_OPCODE_DONE = 0x00u,
+ REBASE_OPCODE_SET_TYPE_IMM = 0x10u,
+ REBASE_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB = 0x20u,
+ REBASE_OPCODE_ADD_ADDR_ULEB = 0x30u,
+ REBASE_OPCODE_ADD_ADDR_IMM_SCALED = 0x40u,
+ REBASE_OPCODE_DO_REBASE_IMM_TIMES = 0x50u,
+ REBASE_OPCODE_DO_REBASE_ULEB_TIMES = 0x60u,
+ REBASE_OPCODE_DO_REBASE_ADD_ADDR_ULEB = 0x70u,
+ REBASE_OPCODE_DO_REBASE_ULEB_TIMES_SKIPPING_ULEB = 0x80u
+};
+
+enum BindType {
+ BIND_TYPE_POINTER = 1u,
+ BIND_TYPE_TEXT_ABSOLUTE32 = 2u,
+ BIND_TYPE_TEXT_PCREL32 = 3u
+};
+
+enum BindSpecialDylib {
+ BIND_SPECIAL_DYLIB_SELF = 0,
+ BIND_SPECIAL_DYLIB_MAIN_EXECUTABLE = -1,
+ BIND_SPECIAL_DYLIB_FLAT_LOOKUP = -2
+};
+
+enum {
+ BIND_SYMBOL_FLAGS_WEAK_IMPORT = 0x1u,
+ BIND_SYMBOL_FLAGS_NON_WEAK_DEFINITION = 0x8u,
+
+ BIND_OPCODE_MASK = 0xF0u,
+ BIND_IMMEDIATE_MASK = 0x0Fu
+};
+
+enum BindOpcode {
+ BIND_OPCODE_DONE = 0x00u,
+ BIND_OPCODE_SET_DYLIB_ORDINAL_IMM = 0x10u,
+ BIND_OPCODE_SET_DYLIB_ORDINAL_ULEB = 0x20u,
+ BIND_OPCODE_SET_DYLIB_SPECIAL_IMM = 0x30u,
+ BIND_OPCODE_SET_SYMBOL_TRAILING_FLAGS_IMM = 0x40u,
+ BIND_OPCODE_SET_TYPE_IMM = 0x50u,
+ BIND_OPCODE_SET_ADDEND_SLEB = 0x60u,
+ BIND_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB = 0x70u,
+ BIND_OPCODE_ADD_ADDR_ULEB = 0x80u,
+ BIND_OPCODE_DO_BIND = 0x90u,
+ BIND_OPCODE_DO_BIND_ADD_ADDR_ULEB = 0xA0u,
+ BIND_OPCODE_DO_BIND_ADD_ADDR_IMM_SCALED = 0xB0u,
+ BIND_OPCODE_DO_BIND_ULEB_TIMES_SKIPPING_ULEB = 0xC0u
+};
+
+enum {
+ EXPORT_SYMBOL_FLAGS_KIND_MASK = 0x03u,
+ EXPORT_SYMBOL_FLAGS_WEAK_DEFINITION = 0x04u,
+ EXPORT_SYMBOL_FLAGS_REEXPORT = 0x08u,
+ EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER = 0x10u
+};
+
+enum ExportSymbolKind {
+ EXPORT_SYMBOL_FLAGS_KIND_REGULAR = 0x00u,
+ EXPORT_SYMBOL_FLAGS_KIND_THREAD_LOCAL = 0x01u,
+ EXPORT_SYMBOL_FLAGS_KIND_ABSOLUTE = 0x02u
+};
+
+enum {
+ // Constant masks for the "n_type" field in llvm::MachO::nlist and
+ // llvm::MachO::nlist_64
+ N_STAB = 0xe0,
+ N_PEXT = 0x10,
+ N_TYPE = 0x0e,
+ N_EXT = 0x01
+};
+
+enum NListType : uint8_t {
+ // Constants for the "n_type & N_TYPE" llvm::MachO::nlist and
+ // llvm::MachO::nlist_64
+ N_UNDF = 0x0u,
+ N_ABS = 0x2u,
+ N_SECT = 0xeu,
+ N_PBUD = 0xcu,
+ N_INDR = 0xau
+};
+
+enum SectionOrdinal {
+ // Constants for the "n_sect" field in llvm::MachO::nlist and
+ // llvm::MachO::nlist_64
+ NO_SECT = 0u,
+ MAX_SECT = 0xffu
+};
+
+enum {
+ // Constant masks for the "n_desc" field in llvm::MachO::nlist and
+ // llvm::MachO::nlist_64
+ // The low 3 bits are the for the REFERENCE_TYPE.
+ REFERENCE_TYPE = 0x7,
+ REFERENCE_FLAG_UNDEFINED_NON_LAZY = 0,
+ REFERENCE_FLAG_UNDEFINED_LAZY = 1,
+ REFERENCE_FLAG_DEFINED = 2,
+ REFERENCE_FLAG_PRIVATE_DEFINED = 3,
+ REFERENCE_FLAG_PRIVATE_UNDEFINED_NON_LAZY = 4,
+ REFERENCE_FLAG_PRIVATE_UNDEFINED_LAZY = 5,
+ // Flag bits (some overlap with the library ordinal bits).
+ N_ARM_THUMB_DEF = 0x0008u,
+ REFERENCED_DYNAMICALLY = 0x0010u,
+ N_NO_DEAD_STRIP = 0x0020u,
+ N_WEAK_REF = 0x0040u,
+ N_WEAK_DEF = 0x0080u,
+ N_SYMBOL_RESOLVER = 0x0100u,
+ N_ALT_ENTRY = 0x0200u,
+ N_COLD_FUNC = 0x0400u,
+ // For undefined symbols coming from libraries, see GET_LIBRARY_ORDINAL()
+ // as these are in the top 8 bits.
+ SELF_LIBRARY_ORDINAL = 0x0,
+ MAX_LIBRARY_ORDINAL = 0xfd,
+ DYNAMIC_LOOKUP_ORDINAL = 0xfe,
+ EXECUTABLE_ORDINAL = 0xff
+};
+
+enum StabType {
+ // Constant values for the "n_type" field in llvm::MachO::nlist and
+ // llvm::MachO::nlist_64 when "(n_type & N_STAB) != 0"
+ N_GSYM = 0x20u,
+ N_FNAME = 0x22u,
+ N_FUN = 0x24u,
+ N_STSYM = 0x26u,
+ N_LCSYM = 0x28u,
+ N_BNSYM = 0x2Eu,
+ N_PC = 0x30u,
+ N_AST = 0x32u,
+ N_OPT = 0x3Cu,
+ N_RSYM = 0x40u,
+ N_SLINE = 0x44u,
+ N_ENSYM = 0x4Eu,
+ N_SSYM = 0x60u,
+ N_SO = 0x64u,
+ N_OSO = 0x66u,
+ N_LSYM = 0x80u,
+ N_BINCL = 0x82u,
+ N_SOL = 0x84u,
+ N_PARAMS = 0x86u,
+ N_VERSION = 0x88u,
+ N_OLEVEL = 0x8Au,
+ N_PSYM = 0xA0u,
+ N_EINCL = 0xA2u,
+ N_ENTRY = 0xA4u,
+ N_LBRAC = 0xC0u,
+ N_EXCL = 0xC2u,
+ N_RBRAC = 0xE0u,
+ N_BCOMM = 0xE2u,
+ N_ECOMM = 0xE4u,
+ N_ECOML = 0xE8u,
+ N_LENG = 0xFEu
+};
+
+enum : uint32_t {
+ // Constant values for the r_symbolnum field in an
+ // llvm::MachO::relocation_info structure when r_extern is 0.
+ R_ABS = 0,
+
+ // Constant bits for the r_address field in an
+ // llvm::MachO::relocation_info structure.
+ R_SCATTERED = 0x80000000
+};
+
+enum RelocationInfoType {
+ // Constant values for the r_type field in an
+ // llvm::MachO::relocation_info or llvm::MachO::scattered_relocation_info
+ // structure.
+ GENERIC_RELOC_VANILLA = 0,
+ GENERIC_RELOC_PAIR = 1,
+ GENERIC_RELOC_SECTDIFF = 2,
+ GENERIC_RELOC_PB_LA_PTR = 3,
+ GENERIC_RELOC_LOCAL_SECTDIFF = 4,
+ GENERIC_RELOC_TLV = 5,
+
+ // Constant values for the r_type field in a PowerPC architecture
+ // llvm::MachO::relocation_info or llvm::MachO::scattered_relocation_info
+ // structure.
+ PPC_RELOC_VANILLA = GENERIC_RELOC_VANILLA,
+ PPC_RELOC_PAIR = GENERIC_RELOC_PAIR,
+ PPC_RELOC_BR14 = 2,
+ PPC_RELOC_BR24 = 3,
+ PPC_RELOC_HI16 = 4,
+ PPC_RELOC_LO16 = 5,
+ PPC_RELOC_HA16 = 6,
+ PPC_RELOC_LO14 = 7,
+ PPC_RELOC_SECTDIFF = 8,
+ PPC_RELOC_PB_LA_PTR = 9,
+ PPC_RELOC_HI16_SECTDIFF = 10,
+ PPC_RELOC_LO16_SECTDIFF = 11,
+ PPC_RELOC_HA16_SECTDIFF = 12,
+ PPC_RELOC_JBSR = 13,
+ PPC_RELOC_LO14_SECTDIFF = 14,
+ PPC_RELOC_LOCAL_SECTDIFF = 15,
+
+ // Constant values for the r_type field in an ARM architecture
+ // llvm::MachO::relocation_info or llvm::MachO::scattered_relocation_info
+ // structure.
+ ARM_RELOC_VANILLA = GENERIC_RELOC_VANILLA,
+ ARM_RELOC_PAIR = GENERIC_RELOC_PAIR,
+ ARM_RELOC_SECTDIFF = GENERIC_RELOC_SECTDIFF,
+ ARM_RELOC_LOCAL_SECTDIFF = 3,
+ ARM_RELOC_PB_LA_PTR = 4,
+ ARM_RELOC_BR24 = 5,
+ ARM_THUMB_RELOC_BR22 = 6,
+ ARM_THUMB_32BIT_BRANCH = 7, // obsolete
+ ARM_RELOC_HALF = 8,
+ ARM_RELOC_HALF_SECTDIFF = 9,
+
+ // Constant values for the r_type field in an ARM64 architecture
+ // llvm::MachO::relocation_info or llvm::MachO::scattered_relocation_info
+ // structure.
+
+ // For pointers.
+ ARM64_RELOC_UNSIGNED = 0,
+ // Must be followed by an ARM64_RELOC_UNSIGNED
+ ARM64_RELOC_SUBTRACTOR = 1,
+ // A B/BL instruction with 26-bit displacement.
+ ARM64_RELOC_BRANCH26 = 2,
+ // PC-rel distance to page of target.
+ ARM64_RELOC_PAGE21 = 3,
+ // Offset within page, scaled by r_length.
+ ARM64_RELOC_PAGEOFF12 = 4,
+ // PC-rel distance to page of GOT slot.
+ ARM64_RELOC_GOT_LOAD_PAGE21 = 5,
+ // Offset within page of GOT slot, scaled by r_length.
+ ARM64_RELOC_GOT_LOAD_PAGEOFF12 = 6,
+ // For pointers to GOT slots.
+ ARM64_RELOC_POINTER_TO_GOT = 7,
+ // PC-rel distance to page of TLVP slot.
+ ARM64_RELOC_TLVP_LOAD_PAGE21 = 8,
+ // Offset within page of TLVP slot, scaled by r_length.
+ ARM64_RELOC_TLVP_LOAD_PAGEOFF12 = 9,
+ // Must be followed by ARM64_RELOC_PAGE21 or ARM64_RELOC_PAGEOFF12.
+ ARM64_RELOC_ADDEND = 10,
+
+ // Constant values for the r_type field in an x86_64 architecture
+ // llvm::MachO::relocation_info or llvm::MachO::scattered_relocation_info
+ // structure
+ X86_64_RELOC_UNSIGNED = 0,
+ X86_64_RELOC_SIGNED = 1,
+ X86_64_RELOC_BRANCH = 2,
+ X86_64_RELOC_GOT_LOAD = 3,
+ X86_64_RELOC_GOT = 4,
+ X86_64_RELOC_SUBTRACTOR = 5,
+ X86_64_RELOC_SIGNED_1 = 6,
+ X86_64_RELOC_SIGNED_2 = 7,
+ X86_64_RELOC_SIGNED_4 = 8,
+ X86_64_RELOC_TLV = 9
+};
+
+// Values for segment_command.initprot.
+// From <mach/vm_prot.h>
+enum { VM_PROT_READ = 0x1, VM_PROT_WRITE = 0x2, VM_PROT_EXECUTE = 0x4 };
+
+// Values for platform field in build_version_command.
+enum PlatformType {
+ PLATFORM_MACOS = 1,
+ PLATFORM_IOS = 2,
+ PLATFORM_TVOS = 3,
+ PLATFORM_WATCHOS = 4,
+ PLATFORM_BRIDGEOS = 5,
+ PLATFORM_MACCATALYST = 6,
+ PLATFORM_IOSSIMULATOR = 7,
+ PLATFORM_TVOSSIMULATOR = 8,
+ PLATFORM_WATCHOSSIMULATOR = 9
+};
+
+// Values for tools enum in build_tool_version.
+enum { TOOL_CLANG = 1, TOOL_SWIFT = 2, TOOL_LD = 3 };
+
+// Structs from <mach-o/loader.h>
+
+struct mach_header {
+ uint32_t magic;
+ uint32_t cputype;
+ uint32_t cpusubtype;
+ uint32_t filetype;
+ uint32_t ncmds;
+ uint32_t sizeofcmds;
+ uint32_t flags;
+};
+
+struct mach_header_64 {
+ uint32_t magic;
+ uint32_t cputype;
+ uint32_t cpusubtype;
+ uint32_t filetype;
+ uint32_t ncmds;
+ uint32_t sizeofcmds;
+ uint32_t flags;
+ uint32_t reserved;
+};
+
+struct load_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+};
+
+struct segment_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ char segname[16];
+ uint32_t vmaddr;
+ uint32_t vmsize;
+ uint32_t fileoff;
+ uint32_t filesize;
+ uint32_t maxprot;
+ uint32_t initprot;
+ uint32_t nsects;
+ uint32_t flags;
+};
+
+struct segment_command_64 {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ char segname[16];
+ uint64_t vmaddr;
+ uint64_t vmsize;
+ uint64_t fileoff;
+ uint64_t filesize;
+ uint32_t maxprot;
+ uint32_t initprot;
+ uint32_t nsects;
+ uint32_t flags;
+};
+
+struct section {
+ char sectname[16];
+ char segname[16];
+ uint32_t addr;
+ uint32_t size;
+ uint32_t offset;
+ uint32_t align;
+ uint32_t reloff;
+ uint32_t nreloc;
+ uint32_t flags;
+ uint32_t reserved1;
+ uint32_t reserved2;
+};
+
+struct section_64 {
+ char sectname[16];
+ char segname[16];
+ uint64_t addr;
+ uint64_t size;
+ uint32_t offset;
+ uint32_t align;
+ uint32_t reloff;
+ uint32_t nreloc;
+ uint32_t flags;
+ uint32_t reserved1;
+ uint32_t reserved2;
+ uint32_t reserved3;
+};
+
+inline bool isVirtualSection(uint8_t type) {
+ return (type == MachO::S_ZEROFILL || type == MachO::S_GB_ZEROFILL ||
+ type == MachO::S_THREAD_LOCAL_ZEROFILL);
+}
+
+struct fvmlib {
+ uint32_t name;
+ uint32_t minor_version;
+ uint32_t header_addr;
+};
+
+// The fvmlib_command is obsolete and no longer supported.
+struct fvmlib_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ struct fvmlib fvmlib;
+};
+
+struct dylib {
+ uint32_t name;
+ uint32_t timestamp;
+ uint32_t current_version;
+ uint32_t compatibility_version;
+};
+
+struct dylib_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ struct dylib dylib;
+};
+
+struct sub_framework_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t umbrella;
+};
+
+struct sub_client_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t client;
+};
+
+struct sub_umbrella_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t sub_umbrella;
+};
+
+struct sub_library_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t sub_library;
+};
+
+// The prebound_dylib_command is obsolete and no longer supported.
+struct prebound_dylib_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t name;
+ uint32_t nmodules;
+ uint32_t linked_modules;
+};
+
+struct dylinker_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t name;
+};
+
+struct thread_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+};
+
+struct routines_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t init_address;
+ uint32_t init_module;
+ uint32_t reserved1;
+ uint32_t reserved2;
+ uint32_t reserved3;
+ uint32_t reserved4;
+ uint32_t reserved5;
+ uint32_t reserved6;
+};
+
+struct routines_command_64 {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint64_t init_address;
+ uint64_t init_module;
+ uint64_t reserved1;
+ uint64_t reserved2;
+ uint64_t reserved3;
+ uint64_t reserved4;
+ uint64_t reserved5;
+ uint64_t reserved6;
+};
+
+struct symtab_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t symoff;
+ uint32_t nsyms;
+ uint32_t stroff;
+ uint32_t strsize;
+};
+
+struct dysymtab_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t ilocalsym;
+ uint32_t nlocalsym;
+ uint32_t iextdefsym;
+ uint32_t nextdefsym;
+ uint32_t iundefsym;
+ uint32_t nundefsym;
+ uint32_t tocoff;
+ uint32_t ntoc;
+ uint32_t modtaboff;
+ uint32_t nmodtab;
+ uint32_t extrefsymoff;
+ uint32_t nextrefsyms;
+ uint32_t indirectsymoff;
+ uint32_t nindirectsyms;
+ uint32_t extreloff;
+ uint32_t nextrel;
+ uint32_t locreloff;
+ uint32_t nlocrel;
+};
+
+struct dylib_table_of_contents {
+ uint32_t symbol_index;
+ uint32_t module_index;
+};
+
+struct dylib_module {
+ uint32_t module_name;
+ uint32_t iextdefsym;
+ uint32_t nextdefsym;
+ uint32_t irefsym;
+ uint32_t nrefsym;
+ uint32_t ilocalsym;
+ uint32_t nlocalsym;
+ uint32_t iextrel;
+ uint32_t nextrel;
+ uint32_t iinit_iterm;
+ uint32_t ninit_nterm;
+ uint32_t objc_module_info_addr;
+ uint32_t objc_module_info_size;
+};
+
+struct dylib_module_64 {
+ uint32_t module_name;
+ uint32_t iextdefsym;
+ uint32_t nextdefsym;
+ uint32_t irefsym;
+ uint32_t nrefsym;
+ uint32_t ilocalsym;
+ uint32_t nlocalsym;
+ uint32_t iextrel;
+ uint32_t nextrel;
+ uint32_t iinit_iterm;
+ uint32_t ninit_nterm;
+ uint32_t objc_module_info_size;
+ uint64_t objc_module_info_addr;
+};
+
+struct dylib_reference {
+ uint32_t isym : 24, flags : 8;
+};
+
+// The twolevel_hints_command is obsolete and no longer supported.
+struct twolevel_hints_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t offset;
+ uint32_t nhints;
+};
+
+// The twolevel_hints_command is obsolete and no longer supported.
+struct twolevel_hint {
+ uint32_t isub_image : 8, itoc : 24;
+};
+
+// The prebind_cksum_command is obsolete and no longer supported.
+struct prebind_cksum_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t cksum;
+};
+
+struct uuid_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint8_t uuid[16];
+};
+
+struct rpath_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t path;
+};
+
+struct linkedit_data_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t dataoff;
+ uint32_t datasize;
+};
+
+struct data_in_code_entry {
+ uint32_t offset;
+ uint16_t length;
+ uint16_t kind;
+};
+
+struct source_version_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint64_t version;
+};
+
+struct encryption_info_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t cryptoff;
+ uint32_t cryptsize;
+ uint32_t cryptid;
+};
+
+struct encryption_info_command_64 {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t cryptoff;
+ uint32_t cryptsize;
+ uint32_t cryptid;
+ uint32_t pad;
+};
+
+struct version_min_command {
+ uint32_t cmd; // LC_VERSION_MIN_MACOSX or
+ // LC_VERSION_MIN_IPHONEOS
+ uint32_t cmdsize; // sizeof(struct version_min_command)
+ uint32_t version; // X.Y.Z is encoded in nibbles xxxx.yy.zz
+ uint32_t sdk; // X.Y.Z is encoded in nibbles xxxx.yy.zz
+};
+
+struct note_command {
+ uint32_t cmd; // LC_NOTE
+ uint32_t cmdsize; // sizeof(struct note_command)
+ char data_owner[16]; // owner name for this LC_NOTE
+ uint64_t offset; // file offset of this data
+ uint64_t size; // length of data region
+};
+
+struct build_tool_version {
+ uint32_t tool; // enum for the tool
+ uint32_t version; // version of the tool
+};
+
+struct build_version_command {
+ uint32_t cmd; // LC_BUILD_VERSION
+ uint32_t cmdsize; // sizeof(struct build_version_command) +
+ // ntools * sizeof(struct build_tool_version)
+ uint32_t platform; // platform
+ uint32_t minos; // X.Y.Z is encoded in nibbles xxxx.yy.zz
+ uint32_t sdk; // X.Y.Z is encoded in nibbles xxxx.yy.zz
+ uint32_t ntools; // number of tool entries following this
+};
+
+struct dyld_info_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t rebase_off;
+ uint32_t rebase_size;
+ uint32_t bind_off;
+ uint32_t bind_size;
+ uint32_t weak_bind_off;
+ uint32_t weak_bind_size;
+ uint32_t lazy_bind_off;
+ uint32_t lazy_bind_size;
+ uint32_t export_off;
+ uint32_t export_size;
+};
+
+struct linker_option_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t count;
+};
+
+// The symseg_command is obsolete and no longer supported.
+struct symseg_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t offset;
+ uint32_t size;
+};
+
+// The ident_command is obsolete and no longer supported.
+struct ident_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+};
+
+// The fvmfile_command is obsolete and no longer supported.
+struct fvmfile_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint32_t name;
+ uint32_t header_addr;
+};
+
+struct tlv_descriptor_32 {
+ uint32_t thunk;
+ uint32_t key;
+ uint32_t offset;
+};
+
+struct tlv_descriptor_64 {
+ uint64_t thunk;
+ uint64_t key;
+ uint64_t offset;
+};
+
+struct tlv_descriptor {
+ uintptr_t thunk;
+ uintptr_t key;
+ uintptr_t offset;
+};
+
+struct entry_point_command {
+ uint32_t cmd;
+ uint32_t cmdsize;
+ uint64_t entryoff;
+ uint64_t stacksize;
+};
+
+// Structs from <mach-o/fat.h>
+struct fat_header {
+ uint32_t magic;
+ uint32_t nfat_arch;
+};
+
+struct fat_arch {
+ uint32_t cputype;
+ uint32_t cpusubtype;
+ uint32_t offset;
+ uint32_t size;
+ uint32_t align;
+};
+
+struct fat_arch_64 {
+ uint32_t cputype;
+ uint32_t cpusubtype;
+ uint64_t offset;
+ uint64_t size;
+ uint32_t align;
+ uint32_t reserved;
+};
+
+// Structs from <mach-o/reloc.h>
+struct relocation_info {
+ int32_t r_address;
+#if defined(BYTE_ORDER) && defined(BIG_ENDIAN) && (BYTE_ORDER == BIG_ENDIAN)
+ uint32_t r_type : 4, r_extern : 1, r_length : 2, r_pcrel : 1,
+ r_symbolnum : 24;
+#else
+ uint32_t r_symbolnum : 24, r_pcrel : 1, r_length : 2, r_extern : 1,
+ r_type : 4;
+#endif
+};
+
+struct scattered_relocation_info {
+#if defined(BYTE_ORDER) && defined(BIG_ENDIAN) && (BYTE_ORDER == BIG_ENDIAN)
+ uint32_t r_scattered : 1, r_pcrel : 1, r_length : 2, r_type : 4,
+ r_address : 24;
+#else
+ uint32_t r_address : 24, r_type : 4, r_length : 2, r_pcrel : 1,
+ r_scattered : 1;
+#endif
+ int32_t r_value;
+};
+
+// Structs NOT from <mach-o/reloc.h>, but that make LLVM's life easier
+struct any_relocation_info {
+ uint32_t r_word0, r_word1;
+};
+
+// Structs from <mach-o/nlist.h>
+struct nlist_base {
+ uint32_t n_strx;
+ uint8_t n_type;
+ uint8_t n_sect;
+ uint16_t n_desc;
+};
+
+struct nlist {
+ uint32_t n_strx;
+ uint8_t n_type;
+ uint8_t n_sect;
+ int16_t n_desc;
+ uint32_t n_value;
+};
+
+struct nlist_64 {
+ uint32_t n_strx;
+ uint8_t n_type;
+ uint8_t n_sect;
+ uint16_t n_desc;
+ uint64_t n_value;
+};
+
+// Byte order swapping functions for MachO structs
+
+inline void swapStruct(fat_header &mh) {
+ sys::swapByteOrder(mh.magic);
+ sys::swapByteOrder(mh.nfat_arch);
+}
+
+inline void swapStruct(fat_arch &mh) {
+ sys::swapByteOrder(mh.cputype);
+ sys::swapByteOrder(mh.cpusubtype);
+ sys::swapByteOrder(mh.offset);
+ sys::swapByteOrder(mh.size);
+ sys::swapByteOrder(mh.align);
+}
+
+inline void swapStruct(fat_arch_64 &mh) {
+ sys::swapByteOrder(mh.cputype);
+ sys::swapByteOrder(mh.cpusubtype);
+ sys::swapByteOrder(mh.offset);
+ sys::swapByteOrder(mh.size);
+ sys::swapByteOrder(mh.align);
+ sys::swapByteOrder(mh.reserved);
+}
+
+inline void swapStruct(mach_header &mh) {
+ sys::swapByteOrder(mh.magic);
+ sys::swapByteOrder(mh.cputype);
+ sys::swapByteOrder(mh.cpusubtype);
+ sys::swapByteOrder(mh.filetype);
+ sys::swapByteOrder(mh.ncmds);
+ sys::swapByteOrder(mh.sizeofcmds);
+ sys::swapByteOrder(mh.flags);
+}
+
+inline void swapStruct(mach_header_64 &H) {
+ sys::swapByteOrder(H.magic);
+ sys::swapByteOrder(H.cputype);
+ sys::swapByteOrder(H.cpusubtype);
+ sys::swapByteOrder(H.filetype);
+ sys::swapByteOrder(H.ncmds);
+ sys::swapByteOrder(H.sizeofcmds);
+ sys::swapByteOrder(H.flags);
+ sys::swapByteOrder(H.reserved);
+}
+
+inline void swapStruct(load_command &lc) {
+ sys::swapByteOrder(lc.cmd);
+ sys::swapByteOrder(lc.cmdsize);
+}
+
+inline void swapStruct(symtab_command &lc) {
+ sys::swapByteOrder(lc.cmd);
+ sys::swapByteOrder(lc.cmdsize);
+ sys::swapByteOrder(lc.symoff);
+ sys::swapByteOrder(lc.nsyms);
+ sys::swapByteOrder(lc.stroff);
+ sys::swapByteOrder(lc.strsize);
+}
+
+inline void swapStruct(segment_command_64 &seg) {
+ sys::swapByteOrder(seg.cmd);
+ sys::swapByteOrder(seg.cmdsize);
+ sys::swapByteOrder(seg.vmaddr);
+ sys::swapByteOrder(seg.vmsize);
+ sys::swapByteOrder(seg.fileoff);
+ sys::swapByteOrder(seg.filesize);
+ sys::swapByteOrder(seg.maxprot);
+ sys::swapByteOrder(seg.initprot);
+ sys::swapByteOrder(seg.nsects);
+ sys::swapByteOrder(seg.flags);
+}
+
+inline void swapStruct(segment_command &seg) {
+ sys::swapByteOrder(seg.cmd);
+ sys::swapByteOrder(seg.cmdsize);
+ sys::swapByteOrder(seg.vmaddr);
+ sys::swapByteOrder(seg.vmsize);
+ sys::swapByteOrder(seg.fileoff);
+ sys::swapByteOrder(seg.filesize);
+ sys::swapByteOrder(seg.maxprot);
+ sys::swapByteOrder(seg.initprot);
+ sys::swapByteOrder(seg.nsects);
+ sys::swapByteOrder(seg.flags);
+}
+
+inline void swapStruct(section_64 §) {
+ sys::swapByteOrder(sect.addr);
+ sys::swapByteOrder(sect.size);
+ sys::swapByteOrder(sect.offset);
+ sys::swapByteOrder(sect.align);
+ sys::swapByteOrder(sect.reloff);
+ sys::swapByteOrder(sect.nreloc);
+ sys::swapByteOrder(sect.flags);
+ sys::swapByteOrder(sect.reserved1);
+ sys::swapByteOrder(sect.reserved2);
+}
+
+inline void swapStruct(section §) {
+ sys::swapByteOrder(sect.addr);
+ sys::swapByteOrder(sect.size);
+ sys::swapByteOrder(sect.offset);
+ sys::swapByteOrder(sect.align);
+ sys::swapByteOrder(sect.reloff);
+ sys::swapByteOrder(sect.nreloc);
+ sys::swapByteOrder(sect.flags);
+ sys::swapByteOrder(sect.reserved1);
+ sys::swapByteOrder(sect.reserved2);
+}
+
+inline void swapStruct(dyld_info_command &info) {
+ sys::swapByteOrder(info.cmd);
+ sys::swapByteOrder(info.cmdsize);
+ sys::swapByteOrder(info.rebase_off);
+ sys::swapByteOrder(info.rebase_size);
+ sys::swapByteOrder(info.bind_off);
+ sys::swapByteOrder(info.bind_size);
+ sys::swapByteOrder(info.weak_bind_off);
+ sys::swapByteOrder(info.weak_bind_size);
+ sys::swapByteOrder(info.lazy_bind_off);
+ sys::swapByteOrder(info.lazy_bind_size);
+ sys::swapByteOrder(info.export_off);
+ sys::swapByteOrder(info.export_size);
+}
+
+inline void swapStruct(dylib_command &d) {
+ sys::swapByteOrder(d.cmd);
+ sys::swapByteOrder(d.cmdsize);
+ sys::swapByteOrder(d.dylib.name);
+ sys::swapByteOrder(d.dylib.timestamp);
+ sys::swapByteOrder(d.dylib.current_version);
+ sys::swapByteOrder(d.dylib.compatibility_version);
+}
+
+inline void swapStruct(sub_framework_command &s) {
+ sys::swapByteOrder(s.cmd);
+ sys::swapByteOrder(s.cmdsize);
+ sys::swapByteOrder(s.umbrella);
+}
+
+inline void swapStruct(sub_umbrella_command &s) {
+ sys::swapByteOrder(s.cmd);
+ sys::swapByteOrder(s.cmdsize);
+ sys::swapByteOrder(s.sub_umbrella);
+}
+
+inline void swapStruct(sub_library_command &s) {
+ sys::swapByteOrder(s.cmd);
+ sys::swapByteOrder(s.cmdsize);
+ sys::swapByteOrder(s.sub_library);
+}
+
+inline void swapStruct(sub_client_command &s) {
+ sys::swapByteOrder(s.cmd);
+ sys::swapByteOrder(s.cmdsize);
+ sys::swapByteOrder(s.client);
+}
+
+inline void swapStruct(routines_command &r) {
+ sys::swapByteOrder(r.cmd);
+ sys::swapByteOrder(r.cmdsize);
+ sys::swapByteOrder(r.init_address);
+ sys::swapByteOrder(r.init_module);
+ sys::swapByteOrder(r.reserved1);
+ sys::swapByteOrder(r.reserved2);
+ sys::swapByteOrder(r.reserved3);
+ sys::swapByteOrder(r.reserved4);
+ sys::swapByteOrder(r.reserved5);
+ sys::swapByteOrder(r.reserved6);
+}
+
+inline void swapStruct(routines_command_64 &r) {
+ sys::swapByteOrder(r.cmd);
+ sys::swapByteOrder(r.cmdsize);
+ sys::swapByteOrder(r.init_address);
+ sys::swapByteOrder(r.init_module);
+ sys::swapByteOrder(r.reserved1);
+ sys::swapByteOrder(r.reserved2);
+ sys::swapByteOrder(r.reserved3);
+ sys::swapByteOrder(r.reserved4);
+ sys::swapByteOrder(r.reserved5);
+ sys::swapByteOrder(r.reserved6);
+}
+
+inline void swapStruct(thread_command &t) {
+ sys::swapByteOrder(t.cmd);
+ sys::swapByteOrder(t.cmdsize);
+}
+
+inline void swapStruct(dylinker_command &d) {
+ sys::swapByteOrder(d.cmd);
+ sys::swapByteOrder(d.cmdsize);
+ sys::swapByteOrder(d.name);
+}
+
+inline void swapStruct(uuid_command &u) {
+ sys::swapByteOrder(u.cmd);
+ sys::swapByteOrder(u.cmdsize);
+}
+
+inline void swapStruct(rpath_command &r) {
+ sys::swapByteOrder(r.cmd);
+ sys::swapByteOrder(r.cmdsize);
+ sys::swapByteOrder(r.path);
+}
+
+inline void swapStruct(source_version_command &s) {
+ sys::swapByteOrder(s.cmd);
+ sys::swapByteOrder(s.cmdsize);
+ sys::swapByteOrder(s.version);
+}
+
+inline void swapStruct(entry_point_command &e) {
+ sys::swapByteOrder(e.cmd);
+ sys::swapByteOrder(e.cmdsize);
+ sys::swapByteOrder(e.entryoff);
+ sys::swapByteOrder(e.stacksize);
+}
+
+inline void swapStruct(encryption_info_command &e) {
+ sys::swapByteOrder(e.cmd);
+ sys::swapByteOrder(e.cmdsize);
+ sys::swapByteOrder(e.cryptoff);
+ sys::swapByteOrder(e.cryptsize);
+ sys::swapByteOrder(e.cryptid);
+}
+
+inline void swapStruct(encryption_info_command_64 &e) {
+ sys::swapByteOrder(e.cmd);
+ sys::swapByteOrder(e.cmdsize);
+ sys::swapByteOrder(e.cryptoff);
+ sys::swapByteOrder(e.cryptsize);
+ sys::swapByteOrder(e.cryptid);
+ sys::swapByteOrder(e.pad);
+}
+
+inline void swapStruct(dysymtab_command &dst) {
+ sys::swapByteOrder(dst.cmd);
+ sys::swapByteOrder(dst.cmdsize);
+ sys::swapByteOrder(dst.ilocalsym);
+ sys::swapByteOrder(dst.nlocalsym);
+ sys::swapByteOrder(dst.iextdefsym);
+ sys::swapByteOrder(dst.nextdefsym);
+ sys::swapByteOrder(dst.iundefsym);
+ sys::swapByteOrder(dst.nundefsym);
+ sys::swapByteOrder(dst.tocoff);
+ sys::swapByteOrder(dst.ntoc);
+ sys::swapByteOrder(dst.modtaboff);
+ sys::swapByteOrder(dst.nmodtab);
+ sys::swapByteOrder(dst.extrefsymoff);
+ sys::swapByteOrder(dst.nextrefsyms);
+ sys::swapByteOrder(dst.indirectsymoff);
+ sys::swapByteOrder(dst.nindirectsyms);
+ sys::swapByteOrder(dst.extreloff);
+ sys::swapByteOrder(dst.nextrel);
+ sys::swapByteOrder(dst.locreloff);
+ sys::swapByteOrder(dst.nlocrel);
+}
+
+inline void swapStruct(any_relocation_info &reloc) {
+ sys::swapByteOrder(reloc.r_word0);
+ sys::swapByteOrder(reloc.r_word1);
+}
+
+inline void swapStruct(nlist_base &S) {
+ sys::swapByteOrder(S.n_strx);
+ sys::swapByteOrder(S.n_desc);
+}
+
+inline void swapStruct(nlist &sym) {
+ sys::swapByteOrder(sym.n_strx);
+ sys::swapByteOrder(sym.n_desc);
+ sys::swapByteOrder(sym.n_value);
+}
+
+inline void swapStruct(nlist_64 &sym) {
+ sys::swapByteOrder(sym.n_strx);
+ sys::swapByteOrder(sym.n_desc);
+ sys::swapByteOrder(sym.n_value);
+}
+
+inline void swapStruct(linkedit_data_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.dataoff);
+ sys::swapByteOrder(C.datasize);
+}
+
+inline void swapStruct(linker_option_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.count);
+}
+
+inline void swapStruct(version_min_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.version);
+ sys::swapByteOrder(C.sdk);
+}
+
+inline void swapStruct(note_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.offset);
+ sys::swapByteOrder(C.size);
+}
+
+inline void swapStruct(build_version_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.platform);
+ sys::swapByteOrder(C.minos);
+ sys::swapByteOrder(C.sdk);
+ sys::swapByteOrder(C.ntools);
+}
+
+inline void swapStruct(build_tool_version &C) {
+ sys::swapByteOrder(C.tool);
+ sys::swapByteOrder(C.version);
+}
+
+inline void swapStruct(data_in_code_entry &C) {
+ sys::swapByteOrder(C.offset);
+ sys::swapByteOrder(C.length);
+ sys::swapByteOrder(C.kind);
+}
+
+inline void swapStruct(uint32_t &C) { sys::swapByteOrder(C); }
+
+// The prebind_cksum_command is obsolete and no longer supported.
+inline void swapStruct(prebind_cksum_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.cksum);
+}
+
+// The twolevel_hints_command is obsolete and no longer supported.
+inline void swapStruct(twolevel_hints_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.offset);
+ sys::swapByteOrder(C.nhints);
+}
+
+// The prebound_dylib_command is obsolete and no longer supported.
+inline void swapStruct(prebound_dylib_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.name);
+ sys::swapByteOrder(C.nmodules);
+ sys::swapByteOrder(C.linked_modules);
+}
+
+// The fvmfile_command is obsolete and no longer supported.
+inline void swapStruct(fvmfile_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.name);
+ sys::swapByteOrder(C.header_addr);
+}
+
+// The symseg_command is obsolete and no longer supported.
+inline void swapStruct(symseg_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ sys::swapByteOrder(C.offset);
+ sys::swapByteOrder(C.size);
+}
+
+// The ident_command is obsolete and no longer supported.
+inline void swapStruct(ident_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+}
+
+inline void swapStruct(fvmlib &C) {
+ sys::swapByteOrder(C.name);
+ sys::swapByteOrder(C.minor_version);
+ sys::swapByteOrder(C.header_addr);
+}
+
+// The fvmlib_command is obsolete and no longer supported.
+inline void swapStruct(fvmlib_command &C) {
+ sys::swapByteOrder(C.cmd);
+ sys::swapByteOrder(C.cmdsize);
+ swapStruct(C.fvmlib);
+}
+
+// Get/Set functions from <mach-o/nlist.h>
+
+inline uint16_t GET_LIBRARY_ORDINAL(uint16_t n_desc) {
+ return (((n_desc) >> 8u) & 0xffu);
+}
+
+inline void SET_LIBRARY_ORDINAL(uint16_t &n_desc, uint8_t ordinal) {
+ n_desc = (((n_desc)&0x00ff) | (((ordinal)&0xff) << 8));
+}
+
+inline uint8_t GET_COMM_ALIGN(uint16_t n_desc) {
+ return (n_desc >> 8u) & 0x0fu;
+}
+
+inline void SET_COMM_ALIGN(uint16_t &n_desc, uint8_t align) {
+ n_desc = ((n_desc & 0xf0ffu) | ((align & 0x0fu) << 8u));
+}
+
+// Enums from <mach/machine.h>
+enum : uint32_t {
+ // Capability bits used in the definition of cpu_type.
+ CPU_ARCH_MASK = 0xff000000, // Mask for architecture bits
+ CPU_ARCH_ABI64 = 0x01000000, // 64 bit ABI
+ CPU_ARCH_ABI64_32 = 0x02000000, // ILP32 ABI on 64-bit hardware
+};
+
+// Constants for the cputype field.
+enum CPUType {
+ CPU_TYPE_ANY = -1,
+ CPU_TYPE_X86 = 7,
+ CPU_TYPE_I386 = CPU_TYPE_X86,
+ CPU_TYPE_X86_64 = CPU_TYPE_X86 | CPU_ARCH_ABI64,
+ /* CPU_TYPE_MIPS = 8, */
+ CPU_TYPE_MC98000 = 10, // Old Motorola PowerPC
+ CPU_TYPE_ARM = 12,
+ CPU_TYPE_ARM64 = CPU_TYPE_ARM | CPU_ARCH_ABI64,
+ CPU_TYPE_ARM64_32 = CPU_TYPE_ARM | CPU_ARCH_ABI64_32,
+ CPU_TYPE_SPARC = 14,
+ CPU_TYPE_POWERPC = 18,
+ CPU_TYPE_POWERPC64 = CPU_TYPE_POWERPC | CPU_ARCH_ABI64
+};
+
+enum : uint32_t {
+ // Capability bits used in the definition of cpusubtype.
+ CPU_SUBTYPE_MASK = 0xff000000, // Mask for architecture bits
+ CPU_SUBTYPE_LIB64 = 0x80000000, // 64 bit libraries
+
+ // Special CPU subtype constants.
+ CPU_SUBTYPE_MULTIPLE = ~0u
+};
+
+// Constants for the cpusubtype field.
+enum CPUSubTypeX86 {
+ CPU_SUBTYPE_I386_ALL = 3,
+ CPU_SUBTYPE_386 = 3,
+ CPU_SUBTYPE_486 = 4,
+ CPU_SUBTYPE_486SX = 0x84,
+ CPU_SUBTYPE_586 = 5,
+ CPU_SUBTYPE_PENT = CPU_SUBTYPE_586,
+ CPU_SUBTYPE_PENTPRO = 0x16,
+ CPU_SUBTYPE_PENTII_M3 = 0x36,
+ CPU_SUBTYPE_PENTII_M5 = 0x56,
+ CPU_SUBTYPE_CELERON = 0x67,
+ CPU_SUBTYPE_CELERON_MOBILE = 0x77,
+ CPU_SUBTYPE_PENTIUM_3 = 0x08,
+ CPU_SUBTYPE_PENTIUM_3_M = 0x18,
+ CPU_SUBTYPE_PENTIUM_3_XEON = 0x28,
+ CPU_SUBTYPE_PENTIUM_M = 0x09,
+ CPU_SUBTYPE_PENTIUM_4 = 0x0a,
+ CPU_SUBTYPE_PENTIUM_4_M = 0x1a,
+ CPU_SUBTYPE_ITANIUM = 0x0b,
+ CPU_SUBTYPE_ITANIUM_2 = 0x1b,
+ CPU_SUBTYPE_XEON = 0x0c,
+ CPU_SUBTYPE_XEON_MP = 0x1c,
+
+ CPU_SUBTYPE_X86_ALL = 3,
+ CPU_SUBTYPE_X86_64_ALL = 3,
+ CPU_SUBTYPE_X86_ARCH1 = 4,
+ CPU_SUBTYPE_X86_64_H = 8
+};
+inline int CPU_SUBTYPE_INTEL(int Family, int Model) {
+ return Family | (Model << 4);
+}
+inline int CPU_SUBTYPE_INTEL_FAMILY(CPUSubTypeX86 ST) {
+ return ((int)ST) & 0x0f;
+}
+inline int CPU_SUBTYPE_INTEL_MODEL(CPUSubTypeX86 ST) { return ((int)ST) >> 4; }
+enum { CPU_SUBTYPE_INTEL_FAMILY_MAX = 15, CPU_SUBTYPE_INTEL_MODEL_ALL = 0 };
+
+enum CPUSubTypeARM {
+ CPU_SUBTYPE_ARM_ALL = 0,
+ CPU_SUBTYPE_ARM_V4T = 5,
+ CPU_SUBTYPE_ARM_V6 = 6,
+ CPU_SUBTYPE_ARM_V5 = 7,
+ CPU_SUBTYPE_ARM_V5TEJ = 7,
+ CPU_SUBTYPE_ARM_XSCALE = 8,
+ CPU_SUBTYPE_ARM_V7 = 9,
+ // unused ARM_V7F = 10,
+ CPU_SUBTYPE_ARM_V7S = 11,
+ CPU_SUBTYPE_ARM_V7K = 12,
+ CPU_SUBTYPE_ARM_V6M = 14,
+ CPU_SUBTYPE_ARM_V7M = 15,
+ CPU_SUBTYPE_ARM_V7EM = 16
+};
+
+enum CPUSubTypeARM64 {
+ CPU_SUBTYPE_ARM64_ALL = 0,
+ CPU_SUBTYPE_ARM64E = 2,
+};
+
+enum CPUSubTypeARM64_32 { CPU_SUBTYPE_ARM64_32_V8 = 1 };
+
+enum CPUSubTypeSPARC { CPU_SUBTYPE_SPARC_ALL = 0 };
+
+enum CPUSubTypePowerPC {
+ CPU_SUBTYPE_POWERPC_ALL = 0,
+ CPU_SUBTYPE_POWERPC_601 = 1,
+ CPU_SUBTYPE_POWERPC_602 = 2,
+ CPU_SUBTYPE_POWERPC_603 = 3,
+ CPU_SUBTYPE_POWERPC_603e = 4,
+ CPU_SUBTYPE_POWERPC_603ev = 5,
+ CPU_SUBTYPE_POWERPC_604 = 6,
+ CPU_SUBTYPE_POWERPC_604e = 7,
+ CPU_SUBTYPE_POWERPC_620 = 8,
+ CPU_SUBTYPE_POWERPC_750 = 9,
+ CPU_SUBTYPE_POWERPC_7400 = 10,
+ CPU_SUBTYPE_POWERPC_7450 = 11,
+ CPU_SUBTYPE_POWERPC_970 = 100,
+
+ CPU_SUBTYPE_MC980000_ALL = CPU_SUBTYPE_POWERPC_ALL,
+ CPU_SUBTYPE_MC98601 = CPU_SUBTYPE_POWERPC_601
+};
+
+struct x86_thread_state32_t {
+ uint32_t eax;
+ uint32_t ebx;
+ uint32_t ecx;
+ uint32_t edx;
+ uint32_t edi;
+ uint32_t esi;
+ uint32_t ebp;
+ uint32_t esp;
+ uint32_t ss;
+ uint32_t eflags;
+ uint32_t eip;
+ uint32_t cs;
+ uint32_t ds;
+ uint32_t es;
+ uint32_t fs;
+ uint32_t gs;
+};
+
+struct x86_thread_state64_t {
+ uint64_t rax;
+ uint64_t rbx;
+ uint64_t rcx;
+ uint64_t rdx;
+ uint64_t rdi;
+ uint64_t rsi;
+ uint64_t rbp;
+ uint64_t rsp;
+ uint64_t r8;
+ uint64_t r9;
+ uint64_t r10;
+ uint64_t r11;
+ uint64_t r12;
+ uint64_t r13;
+ uint64_t r14;
+ uint64_t r15;
+ uint64_t rip;
+ uint64_t rflags;
+ uint64_t cs;
+ uint64_t fs;
+ uint64_t gs;
+};
+
+enum x86_fp_control_precis {
+ x86_FP_PREC_24B = 0,
+ x86_FP_PREC_53B = 2,
+ x86_FP_PREC_64B = 3
+};
+
+enum x86_fp_control_rc {
+ x86_FP_RND_NEAR = 0,
+ x86_FP_RND_DOWN = 1,
+ x86_FP_RND_UP = 2,
+ x86_FP_CHOP = 3
+};
+
+struct fp_control_t {
+ unsigned short invalid : 1, denorm : 1, zdiv : 1, ovrfl : 1, undfl : 1,
+ precis : 1, : 2, pc : 2, rc : 2, : 1, : 3;
+};
+
+struct fp_status_t {
+ unsigned short invalid : 1, denorm : 1, zdiv : 1, ovrfl : 1, undfl : 1,
+ precis : 1, stkflt : 1, errsumm : 1, c0 : 1, c1 : 1, c2 : 1, tos : 3,
+ c3 : 1, busy : 1;
+};
+
+struct mmst_reg_t {
+ char mmst_reg[10];
+ char mmst_rsrv[6];
+};
+
+struct xmm_reg_t {
+ char xmm_reg[16];
+};
+
+struct x86_float_state64_t {
+ int32_t fpu_reserved[2];
+ fp_control_t fpu_fcw;
+ fp_status_t fpu_fsw;
+ uint8_t fpu_ftw;
+ uint8_t fpu_rsrv1;
+ uint16_t fpu_fop;
+ uint32_t fpu_ip;
+ uint16_t fpu_cs;
+ uint16_t fpu_rsrv2;
+ uint32_t fpu_dp;
+ uint16_t fpu_ds;
+ uint16_t fpu_rsrv3;
+ uint32_t fpu_mxcsr;
+ uint32_t fpu_mxcsrmask;
+ mmst_reg_t fpu_stmm0;
+ mmst_reg_t fpu_stmm1;
+ mmst_reg_t fpu_stmm2;
+ mmst_reg_t fpu_stmm3;
+ mmst_reg_t fpu_stmm4;
+ mmst_reg_t fpu_stmm5;
+ mmst_reg_t fpu_stmm6;
+ mmst_reg_t fpu_stmm7;
+ xmm_reg_t fpu_xmm0;
+ xmm_reg_t fpu_xmm1;
+ xmm_reg_t fpu_xmm2;
+ xmm_reg_t fpu_xmm3;
+ xmm_reg_t fpu_xmm4;
+ xmm_reg_t fpu_xmm5;
+ xmm_reg_t fpu_xmm6;
+ xmm_reg_t fpu_xmm7;
+ xmm_reg_t fpu_xmm8;
+ xmm_reg_t fpu_xmm9;
+ xmm_reg_t fpu_xmm10;
+ xmm_reg_t fpu_xmm11;
+ xmm_reg_t fpu_xmm12;
+ xmm_reg_t fpu_xmm13;
+ xmm_reg_t fpu_xmm14;
+ xmm_reg_t fpu_xmm15;
+ char fpu_rsrv4[6 * 16];
+ uint32_t fpu_reserved1;
+};
+
+struct x86_exception_state64_t {
+ uint16_t trapno;
+ uint16_t cpu;
+ uint32_t err;
+ uint64_t faultvaddr;
+};
+
+inline void swapStruct(x86_thread_state32_t &x) {
+ sys::swapByteOrder(x.eax);
+ sys::swapByteOrder(x.ebx);
+ sys::swapByteOrder(x.ecx);
+ sys::swapByteOrder(x.edx);
+ sys::swapByteOrder(x.edi);
+ sys::swapByteOrder(x.esi);
+ sys::swapByteOrder(x.ebp);
+ sys::swapByteOrder(x.esp);
+ sys::swapByteOrder(x.ss);
+ sys::swapByteOrder(x.eflags);
+ sys::swapByteOrder(x.eip);
+ sys::swapByteOrder(x.cs);
+ sys::swapByteOrder(x.ds);
+ sys::swapByteOrder(x.es);
+ sys::swapByteOrder(x.fs);
+ sys::swapByteOrder(x.gs);
+}
+
+inline void swapStruct(x86_thread_state64_t &x) {
+ sys::swapByteOrder(x.rax);
+ sys::swapByteOrder(x.rbx);
+ sys::swapByteOrder(x.rcx);
+ sys::swapByteOrder(x.rdx);
+ sys::swapByteOrder(x.rdi);
+ sys::swapByteOrder(x.rsi);
+ sys::swapByteOrder(x.rbp);
+ sys::swapByteOrder(x.rsp);
+ sys::swapByteOrder(x.r8);
+ sys::swapByteOrder(x.r9);
+ sys::swapByteOrder(x.r10);
+ sys::swapByteOrder(x.r11);
+ sys::swapByteOrder(x.r12);
+ sys::swapByteOrder(x.r13);
+ sys::swapByteOrder(x.r14);
+ sys::swapByteOrder(x.r15);
+ sys::swapByteOrder(x.rip);
+ sys::swapByteOrder(x.rflags);
+ sys::swapByteOrder(x.cs);
+ sys::swapByteOrder(x.fs);
+ sys::swapByteOrder(x.gs);
+}
+
+inline void swapStruct(x86_float_state64_t &x) {
+ sys::swapByteOrder(x.fpu_reserved[0]);
+ sys::swapByteOrder(x.fpu_reserved[1]);
+ // TODO swap: fp_control_t fpu_fcw;
+ // TODO swap: fp_status_t fpu_fsw;
+ sys::swapByteOrder(x.fpu_fop);
+ sys::swapByteOrder(x.fpu_ip);
+ sys::swapByteOrder(x.fpu_cs);
+ sys::swapByteOrder(x.fpu_rsrv2);
+ sys::swapByteOrder(x.fpu_dp);
+ sys::swapByteOrder(x.fpu_ds);
+ sys::swapByteOrder(x.fpu_rsrv3);
+ sys::swapByteOrder(x.fpu_mxcsr);
+ sys::swapByteOrder(x.fpu_mxcsrmask);
+ sys::swapByteOrder(x.fpu_reserved1);
+}
+
+inline void swapStruct(x86_exception_state64_t &x) {
+ sys::swapByteOrder(x.trapno);
+ sys::swapByteOrder(x.cpu);
+ sys::swapByteOrder(x.err);
+ sys::swapByteOrder(x.faultvaddr);
+}
+
+struct x86_state_hdr_t {
+ uint32_t flavor;
+ uint32_t count;
+};
+
+struct x86_thread_state_t {
+ x86_state_hdr_t tsh;
+ union {
+ x86_thread_state64_t ts64;
+ x86_thread_state32_t ts32;
+ } uts;
+};
+
+struct x86_float_state_t {
+ x86_state_hdr_t fsh;
+ union {
+ x86_float_state64_t fs64;
+ } ufs;
+};
+
+struct x86_exception_state_t {
+ x86_state_hdr_t esh;
+ union {
+ x86_exception_state64_t es64;
+ } ues;
+};
+
+inline void swapStruct(x86_state_hdr_t &x) {
+ sys::swapByteOrder(x.flavor);
+ sys::swapByteOrder(x.count);
+}
+
+enum X86ThreadFlavors {
+ x86_THREAD_STATE32 = 1,
+ x86_FLOAT_STATE32 = 2,
+ x86_EXCEPTION_STATE32 = 3,
+ x86_THREAD_STATE64 = 4,
+ x86_FLOAT_STATE64 = 5,
+ x86_EXCEPTION_STATE64 = 6,
+ x86_THREAD_STATE = 7,
+ x86_FLOAT_STATE = 8,
+ x86_EXCEPTION_STATE = 9,
+ x86_DEBUG_STATE32 = 10,
+ x86_DEBUG_STATE64 = 11,
+ x86_DEBUG_STATE = 12
+};
+
+inline void swapStruct(x86_thread_state_t &x) {
+ swapStruct(x.tsh);
+ if (x.tsh.flavor == x86_THREAD_STATE64)
+ swapStruct(x.uts.ts64);
+}
+
+inline void swapStruct(x86_float_state_t &x) {
+ swapStruct(x.fsh);
+ if (x.fsh.flavor == x86_FLOAT_STATE64)
+ swapStruct(x.ufs.fs64);
+}
+
+inline void swapStruct(x86_exception_state_t &x) {
+ swapStruct(x.esh);
+ if (x.esh.flavor == x86_EXCEPTION_STATE64)
+ swapStruct(x.ues.es64);
+}
+
+const uint32_t x86_THREAD_STATE32_COUNT =
+ sizeof(x86_thread_state32_t) / sizeof(uint32_t);
+
+const uint32_t x86_THREAD_STATE64_COUNT =
+ sizeof(x86_thread_state64_t) / sizeof(uint32_t);
+const uint32_t x86_FLOAT_STATE64_COUNT =
+ sizeof(x86_float_state64_t) / sizeof(uint32_t);
+const uint32_t x86_EXCEPTION_STATE64_COUNT =
+ sizeof(x86_exception_state64_t) / sizeof(uint32_t);
+
+const uint32_t x86_THREAD_STATE_COUNT =
+ sizeof(x86_thread_state_t) / sizeof(uint32_t);
+const uint32_t x86_FLOAT_STATE_COUNT =
+ sizeof(x86_float_state_t) / sizeof(uint32_t);
+const uint32_t x86_EXCEPTION_STATE_COUNT =
+ sizeof(x86_exception_state_t) / sizeof(uint32_t);
+
+struct arm_thread_state32_t {
+ uint32_t r[13];
+ uint32_t sp;
+ uint32_t lr;
+ uint32_t pc;
+ uint32_t cpsr;
+};
+
+inline void swapStruct(arm_thread_state32_t &x) {
+ for (int i = 0; i < 13; i++)
+ sys::swapByteOrder(x.r[i]);
+ sys::swapByteOrder(x.sp);
+ sys::swapByteOrder(x.lr);
+ sys::swapByteOrder(x.pc);
+ sys::swapByteOrder(x.cpsr);
+}
+
+struct arm_thread_state64_t {
+ uint64_t x[29];
+ uint64_t fp;
+ uint64_t lr;
+ uint64_t sp;
+ uint64_t pc;
+ uint32_t cpsr;
+ uint32_t pad;
+};
+
+inline void swapStruct(arm_thread_state64_t &x) {
+ for (int i = 0; i < 29; i++)
+ sys::swapByteOrder(x.x[i]);
+ sys::swapByteOrder(x.fp);
+ sys::swapByteOrder(x.lr);
+ sys::swapByteOrder(x.sp);
+ sys::swapByteOrder(x.pc);
+ sys::swapByteOrder(x.cpsr);
+}
+
+struct arm_state_hdr_t {
+ uint32_t flavor;
+ uint32_t count;
+};
+
+struct arm_thread_state_t {
+ arm_state_hdr_t tsh;
+ union {
+ arm_thread_state32_t ts32;
+ } uts;
+};
+
+inline void swapStruct(arm_state_hdr_t &x) {
+ sys::swapByteOrder(x.flavor);
+ sys::swapByteOrder(x.count);
+}
+
+enum ARMThreadFlavors {
+ ARM_THREAD_STATE = 1,
+ ARM_VFP_STATE = 2,
+ ARM_EXCEPTION_STATE = 3,
+ ARM_DEBUG_STATE = 4,
+ ARN_THREAD_STATE_NONE = 5,
+ ARM_THREAD_STATE64 = 6,
+ ARM_EXCEPTION_STATE64 = 7
+};
+
+inline void swapStruct(arm_thread_state_t &x) {
+ swapStruct(x.tsh);
+ if (x.tsh.flavor == ARM_THREAD_STATE)
+ swapStruct(x.uts.ts32);
+}
+
+const uint32_t ARM_THREAD_STATE_COUNT =
+ sizeof(arm_thread_state32_t) / sizeof(uint32_t);
+
+const uint32_t ARM_THREAD_STATE64_COUNT =
+ sizeof(arm_thread_state64_t) / sizeof(uint32_t);
+
+struct ppc_thread_state32_t {
+ uint32_t srr0;
+ uint32_t srr1;
+ uint32_t r0;
+ uint32_t r1;
+ uint32_t r2;
+ uint32_t r3;
+ uint32_t r4;
+ uint32_t r5;
+ uint32_t r6;
+ uint32_t r7;
+ uint32_t r8;
+ uint32_t r9;
+ uint32_t r10;
+ uint32_t r11;
+ uint32_t r12;
+ uint32_t r13;
+ uint32_t r14;
+ uint32_t r15;
+ uint32_t r16;
+ uint32_t r17;
+ uint32_t r18;
+ uint32_t r19;
+ uint32_t r20;
+ uint32_t r21;
+ uint32_t r22;
+ uint32_t r23;
+ uint32_t r24;
+ uint32_t r25;
+ uint32_t r26;
+ uint32_t r27;
+ uint32_t r28;
+ uint32_t r29;
+ uint32_t r30;
+ uint32_t r31;
+ uint32_t ct;
+ uint32_t xer;
+ uint32_t lr;
+ uint32_t ctr;
+ uint32_t mq;
+ uint32_t vrsave;
+};
+
+inline void swapStruct(ppc_thread_state32_t &x) {
+ sys::swapByteOrder(x.srr0);
+ sys::swapByteOrder(x.srr1);
+ sys::swapByteOrder(x.r0);
+ sys::swapByteOrder(x.r1);
+ sys::swapByteOrder(x.r2);
+ sys::swapByteOrder(x.r3);
+ sys::swapByteOrder(x.r4);
+ sys::swapByteOrder(x.r5);
+ sys::swapByteOrder(x.r6);
+ sys::swapByteOrder(x.r7);
+ sys::swapByteOrder(x.r8);
+ sys::swapByteOrder(x.r9);
+ sys::swapByteOrder(x.r10);
+ sys::swapByteOrder(x.r11);
+ sys::swapByteOrder(x.r12);
+ sys::swapByteOrder(x.r13);
+ sys::swapByteOrder(x.r14);
+ sys::swapByteOrder(x.r15);
+ sys::swapByteOrder(x.r16);
+ sys::swapByteOrder(x.r17);
+ sys::swapByteOrder(x.r18);
+ sys::swapByteOrder(x.r19);
+ sys::swapByteOrder(x.r20);
+ sys::swapByteOrder(x.r21);
+ sys::swapByteOrder(x.r22);
+ sys::swapByteOrder(x.r23);
+ sys::swapByteOrder(x.r24);
+ sys::swapByteOrder(x.r25);
+ sys::swapByteOrder(x.r26);
+ sys::swapByteOrder(x.r27);
+ sys::swapByteOrder(x.r28);
+ sys::swapByteOrder(x.r29);
+ sys::swapByteOrder(x.r30);
+ sys::swapByteOrder(x.r31);
+ sys::swapByteOrder(x.ct);
+ sys::swapByteOrder(x.xer);
+ sys::swapByteOrder(x.lr);
+ sys::swapByteOrder(x.ctr);
+ sys::swapByteOrder(x.mq);
+ sys::swapByteOrder(x.vrsave);
+}
+
+struct ppc_state_hdr_t {
+ uint32_t flavor;
+ uint32_t count;
+};
+
+struct ppc_thread_state_t {
+ ppc_state_hdr_t tsh;
+ union {
+ ppc_thread_state32_t ts32;
+ } uts;
+};
+
+inline void swapStruct(ppc_state_hdr_t &x) {
+ sys::swapByteOrder(x.flavor);
+ sys::swapByteOrder(x.count);
+}
+
+enum PPCThreadFlavors {
+ PPC_THREAD_STATE = 1,
+ PPC_FLOAT_STATE = 2,
+ PPC_EXCEPTION_STATE = 3,
+ PPC_VECTOR_STATE = 4,
+ PPC_THREAD_STATE64 = 5,
+ PPC_EXCEPTION_STATE64 = 6,
+ PPC_THREAD_STATE_NONE = 7
+};
+
+inline void swapStruct(ppc_thread_state_t &x) {
+ swapStruct(x.tsh);
+ if (x.tsh.flavor == PPC_THREAD_STATE)
+ swapStruct(x.uts.ts32);
+}
+
+const uint32_t PPC_THREAD_STATE_COUNT =
+ sizeof(ppc_thread_state32_t) / sizeof(uint32_t);
+
+// Define a union of all load command structs
+#define LOAD_COMMAND_STRUCT(LCStruct) LCStruct LCStruct##_data;
+
+LLVM_PACKED_START
+union alignas(4) macho_load_command {
+#include "llvm/BinaryFormat/MachO.def"
+};
+LLVM_PACKED_END
+
+} // end namespace MachO
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/Magic.h b/third_party/llvm-project/include/llvm/BinaryFormat/Magic.h
new file mode 100644
index 000000000..64c687262
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/Magic.h
@@ -0,0 +1,77 @@
+//===- llvm/BinaryFormat/Magic.h - File magic identification ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_MAGIC_H
+#define LLVM_BINARYFORMAT_MAGIC_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+
+#include <system_error>
+
+namespace llvm {
+/// file_magic - An "enum class" enumeration of file types based on magic (the
+/// first N bytes of the file).
+struct file_magic {
+ enum Impl {
+ unknown = 0, ///< Unrecognized file
+ bitcode, ///< Bitcode file
+ archive, ///< ar style archive file
+ elf, ///< ELF Unknown type
+ elf_relocatable, ///< ELF Relocatable object file
+ elf_executable, ///< ELF Executable image
+ elf_shared_object, ///< ELF dynamically linked shared lib
+ elf_core, ///< ELF core image
+ macho_object, ///< Mach-O Object file
+ macho_executable, ///< Mach-O Executable
+ macho_fixed_virtual_memory_shared_lib, ///< Mach-O Shared Lib, FVM
+ macho_core, ///< Mach-O Core File
+ macho_preload_executable, ///< Mach-O Preloaded Executable
+ macho_dynamically_linked_shared_lib, ///< Mach-O dynlinked shared lib
+ macho_dynamic_linker, ///< The Mach-O dynamic linker
+ macho_bundle, ///< Mach-O Bundle file
+ macho_dynamically_linked_shared_lib_stub, ///< Mach-O Shared lib stub
+ macho_dsym_companion, ///< Mach-O dSYM companion file
+ macho_kext_bundle, ///< Mach-O kext bundle file
+ macho_universal_binary, ///< Mach-O universal binary
+ minidump, ///< Windows minidump file
+ coff_cl_gl_object, ///< Microsoft cl.exe's intermediate code file
+ coff_object, ///< COFF object file
+ coff_import_library, ///< COFF import library
+ pecoff_executable, ///< PECOFF executable file
+ windows_resource, ///< Windows compiled resource file (.res)
+ xcoff_object_32, ///< 32-bit XCOFF object file
+ xcoff_object_64, ///< 64-bit XCOFF object file
+ wasm_object, ///< WebAssembly Object file
+ pdb, ///< Windows PDB debug info file
+ tapi_file, ///< Text-based Dynamic Library Stub file
+ };
+
+ bool is_object() const { return V != unknown; }
+
+ file_magic() = default;
+ file_magic(Impl V) : V(V) {}
+ operator Impl() const { return V; }
+
+private:
+ Impl V = unknown;
+};
+
+/// Identify the type of a binary file based on how magical it is.
+file_magic identify_magic(StringRef magic);
+
+/// Get and identify \a path's type based on its content.
+///
+/// @param path Input path.
+/// @param result Set to the type of file, or file_magic::unknown.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code identify_magic(const Twine &path, file_magic &result);
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/Wasm.h b/third_party/llvm-project/include/llvm/BinaryFormat/Wasm.h
new file mode 100644
index 000000000..f550d880f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/Wasm.h
@@ -0,0 +1,391 @@
+//===- Wasm.h - Wasm object file format -------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines manifest constants for the wasm object file format.
+// See: https://github.com/WebAssembly/design/blob/master/BinaryEncoding.md
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_BINARYFORMAT_WASM_H
+#define LLVM_BINARYFORMAT_WASM_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+
+namespace llvm {
+namespace wasm {
+
+// Object file magic string.
+const char WasmMagic[] = {'\0', 'a', 's', 'm'};
+// Wasm binary format version
+const uint32_t WasmVersion = 0x1;
+// Wasm linking metadata version
+const uint32_t WasmMetadataVersion = 0x2;
+// Wasm uses a 64k page size
+const uint32_t WasmPageSize = 65536;
+
+struct WasmObjectHeader {
+ StringRef Magic;
+ uint32_t Version;
+};
+
+struct WasmDylinkInfo {
+ uint32_t MemorySize; // Memory size in bytes
+ uint32_t MemoryAlignment; // P2 alignment of memory
+ uint32_t TableSize; // Table size in elements
+ uint32_t TableAlignment; // P2 alignment of table
+ std::vector<StringRef> Needed; // Shared library depenedencies
+};
+
+struct WasmProducerInfo {
+ std::vector<std::pair<std::string, std::string>> Languages;
+ std::vector<std::pair<std::string, std::string>> Tools;
+ std::vector<std::pair<std::string, std::string>> SDKs;
+};
+
+struct WasmFeatureEntry {
+ uint8_t Prefix;
+ std::string Name;
+};
+
+struct WasmExport {
+ StringRef Name;
+ uint8_t Kind;
+ uint32_t Index;
+};
+
+struct WasmLimits {
+ uint8_t Flags;
+ uint32_t Initial;
+ uint32_t Maximum;
+};
+
+struct WasmTable {
+ uint8_t ElemType;
+ WasmLimits Limits;
+};
+
+struct WasmInitExpr {
+ uint8_t Opcode;
+ union {
+ int32_t Int32;
+ int64_t Int64;
+ int32_t Float32;
+ int64_t Float64;
+ uint32_t Global;
+ } Value;
+};
+
+struct WasmGlobalType {
+ uint8_t Type;
+ bool Mutable;
+};
+
+struct WasmGlobal {
+ uint32_t Index;
+ WasmGlobalType Type;
+ WasmInitExpr InitExpr;
+ StringRef SymbolName; // from the "linking" section
+};
+
+struct WasmEventType {
+ // Kind of event. Currently only WASM_EVENT_ATTRIBUTE_EXCEPTION is possible.
+ uint32_t Attribute;
+ uint32_t SigIndex;
+};
+
+struct WasmEvent {
+ uint32_t Index;
+ WasmEventType Type;
+ StringRef SymbolName; // from the "linking" section
+};
+
+struct WasmImport {
+ StringRef Module;
+ StringRef Field;
+ uint8_t Kind;
+ union {
+ uint32_t SigIndex;
+ WasmGlobalType Global;
+ WasmTable Table;
+ WasmLimits Memory;
+ WasmEventType Event;
+ };
+};
+
+struct WasmLocalDecl {
+ uint8_t Type;
+ uint32_t Count;
+};
+
+struct WasmFunction {
+ uint32_t Index;
+ std::vector<WasmLocalDecl> Locals;
+ ArrayRef<uint8_t> Body;
+ uint32_t CodeSectionOffset;
+ uint32_t Size;
+ uint32_t CodeOffset; // start of Locals and Body
+ StringRef SymbolName; // from the "linking" section
+ StringRef DebugName; // from the "name" section
+ uint32_t Comdat; // from the "comdat info" section
+};
+
+struct WasmDataSegment {
+ uint32_t InitFlags;
+ uint32_t MemoryIndex; // present if InitFlags & WASM_SEGMENT_HAS_MEMINDEX
+ WasmInitExpr Offset; // present if InitFlags & WASM_SEGMENT_IS_PASSIVE == 0
+ ArrayRef<uint8_t> Content;
+ StringRef Name; // from the "segment info" section
+ uint32_t Alignment;
+ uint32_t LinkerFlags;
+ uint32_t Comdat; // from the "comdat info" section
+};
+
+struct WasmElemSegment {
+ uint32_t TableIndex;
+ WasmInitExpr Offset;
+ std::vector<uint32_t> Functions;
+};
+
+// Represents the location of a Wasm data symbol within a WasmDataSegment, as
+// the index of the segment, and the offset and size within the segment.
+struct WasmDataReference {
+ uint32_t Segment;
+ uint32_t Offset;
+ uint32_t Size;
+};
+
+struct WasmRelocation {
+ uint8_t Type; // The type of the relocation.
+ uint32_t Index; // Index into either symbol or type index space.
+ uint64_t Offset; // Offset from the start of the section.
+ int64_t Addend; // A value to add to the symbol.
+};
+
+struct WasmInitFunc {
+ uint32_t Priority;
+ uint32_t Symbol;
+};
+
+struct WasmSymbolInfo {
+ StringRef Name;
+ uint8_t Kind;
+ uint32_t Flags;
+ StringRef ImportModule; // For undefined symbols the module of the import
+ StringRef ImportName; // For undefined symbols the name of the import
+ union {
+ // For function or global symbols, the index in function or global index
+ // space.
+ uint32_t ElementIndex;
+ // For a data symbols, the address of the data relative to segment.
+ WasmDataReference DataRef;
+ };
+};
+
+struct WasmFunctionName {
+ uint32_t Index;
+ StringRef Name;
+};
+
+struct WasmLinkingData {
+ uint32_t Version;
+ std::vector<WasmInitFunc> InitFunctions;
+ std::vector<StringRef> Comdats;
+ std::vector<WasmSymbolInfo> SymbolTable;
+};
+
+enum : unsigned {
+ WASM_SEC_CUSTOM = 0, // Custom / User-defined section
+ WASM_SEC_TYPE = 1, // Function signature declarations
+ WASM_SEC_IMPORT = 2, // Import declarations
+ WASM_SEC_FUNCTION = 3, // Function declarations
+ WASM_SEC_TABLE = 4, // Indirect function table and other tables
+ WASM_SEC_MEMORY = 5, // Memory attributes
+ WASM_SEC_GLOBAL = 6, // Global declarations
+ WASM_SEC_EXPORT = 7, // Exports
+ WASM_SEC_START = 8, // Start function declaration
+ WASM_SEC_ELEM = 9, // Elements section
+ WASM_SEC_CODE = 10, // Function bodies (code)
+ WASM_SEC_DATA = 11, // Data segments
+ WASM_SEC_DATACOUNT = 12, // Data segment count
+ WASM_SEC_EVENT = 13 // Event declarations
+};
+
+// Type immediate encodings used in various contexts.
+enum : unsigned {
+ WASM_TYPE_I32 = 0x7F,
+ WASM_TYPE_I64 = 0x7E,
+ WASM_TYPE_F32 = 0x7D,
+ WASM_TYPE_F64 = 0x7C,
+ WASM_TYPE_V128 = 0x7B,
+ WASM_TYPE_FUNCREF = 0x70,
+ WASM_TYPE_EXNREF = 0x68,
+ WASM_TYPE_FUNC = 0x60,
+ WASM_TYPE_NORESULT = 0x40, // for blocks with no result values
+};
+
+// Kinds of externals (for imports and exports).
+enum : unsigned {
+ WASM_EXTERNAL_FUNCTION = 0x0,
+ WASM_EXTERNAL_TABLE = 0x1,
+ WASM_EXTERNAL_MEMORY = 0x2,
+ WASM_EXTERNAL_GLOBAL = 0x3,
+ WASM_EXTERNAL_EVENT = 0x4,
+};
+
+// Opcodes used in initializer expressions.
+enum : unsigned {
+ WASM_OPCODE_END = 0x0b,
+ WASM_OPCODE_CALL = 0x10,
+ WASM_OPCODE_LOCAL_GET = 0x20,
+ WASM_OPCODE_GLOBAL_GET = 0x23,
+ WASM_OPCODE_GLOBAL_SET = 0x24,
+ WASM_OPCODE_I32_STORE = 0x36,
+ WASM_OPCODE_I32_CONST = 0x41,
+ WASM_OPCODE_I64_CONST = 0x42,
+ WASM_OPCODE_F32_CONST = 0x43,
+ WASM_OPCODE_F64_CONST = 0x44,
+ WASM_OPCODE_I32_ADD = 0x6a,
+};
+
+// Opcodes used in synthetic functions.
+enum : unsigned {
+ WASM_OPCODE_IF = 0x04,
+ WASM_OPCODE_ELSE = 0x05,
+ WASM_OPCODE_DROP = 0x1a,
+ WASM_OPCODE_MISC_PREFIX = 0xfc,
+ WASM_OPCODE_MEMORY_INIT = 0x08,
+ WASM_OPCODE_DATA_DROP = 0x09,
+ WASM_OPCODE_ATOMICS_PREFIX = 0xfe,
+ WASM_OPCODE_ATOMIC_NOTIFY = 0x00,
+ WASM_OPCODE_I32_ATOMIC_WAIT = 0x01,
+ WASM_OPCODE_I32_ATOMIC_STORE = 0x17,
+ WASM_OPCODE_I32_RMW_CMPXCHG = 0x48,
+};
+
+enum : unsigned {
+ WASM_LIMITS_FLAG_HAS_MAX = 0x1,
+ WASM_LIMITS_FLAG_IS_SHARED = 0x2,
+};
+
+enum : unsigned {
+ WASM_SEGMENT_IS_PASSIVE = 0x01,
+ WASM_SEGMENT_HAS_MEMINDEX = 0x02,
+};
+
+// Feature policy prefixes used in the custom "target_features" section
+enum : uint8_t {
+ WASM_FEATURE_PREFIX_USED = '+',
+ WASM_FEATURE_PREFIX_REQUIRED = '=',
+ WASM_FEATURE_PREFIX_DISALLOWED = '-',
+};
+
+// Kind codes used in the custom "name" section
+enum : unsigned {
+ WASM_NAMES_FUNCTION = 0x1,
+ WASM_NAMES_LOCAL = 0x2,
+};
+
+// Kind codes used in the custom "linking" section
+enum : unsigned {
+ WASM_SEGMENT_INFO = 0x5,
+ WASM_INIT_FUNCS = 0x6,
+ WASM_COMDAT_INFO = 0x7,
+ WASM_SYMBOL_TABLE = 0x8,
+};
+
+// Kind codes used in the custom "linking" section in the WASM_COMDAT_INFO
+enum : unsigned {
+ WASM_COMDAT_DATA = 0x0,
+ WASM_COMDAT_FUNCTION = 0x1,
+};
+
+// Kind codes used in the custom "linking" section in the WASM_SYMBOL_TABLE
+enum WasmSymbolType : unsigned {
+ WASM_SYMBOL_TYPE_FUNCTION = 0x0,
+ WASM_SYMBOL_TYPE_DATA = 0x1,
+ WASM_SYMBOL_TYPE_GLOBAL = 0x2,
+ WASM_SYMBOL_TYPE_SECTION = 0x3,
+ WASM_SYMBOL_TYPE_EVENT = 0x4,
+};
+
+// Kinds of event attributes.
+enum WasmEventAttribute : unsigned {
+ WASM_EVENT_ATTRIBUTE_EXCEPTION = 0x0,
+};
+
+const unsigned WASM_SYMBOL_BINDING_MASK = 0x3;
+const unsigned WASM_SYMBOL_VISIBILITY_MASK = 0xc;
+
+const unsigned WASM_SYMBOL_BINDING_GLOBAL = 0x0;
+const unsigned WASM_SYMBOL_BINDING_WEAK = 0x1;
+const unsigned WASM_SYMBOL_BINDING_LOCAL = 0x2;
+const unsigned WASM_SYMBOL_VISIBILITY_DEFAULT = 0x0;
+const unsigned WASM_SYMBOL_VISIBILITY_HIDDEN = 0x4;
+const unsigned WASM_SYMBOL_UNDEFINED = 0x10;
+const unsigned WASM_SYMBOL_EXPORTED = 0x20;
+const unsigned WASM_SYMBOL_EXPLICIT_NAME = 0x40;
+const unsigned WASM_SYMBOL_NO_STRIP = 0x80;
+
+#define WASM_RELOC(name, value) name = value,
+
+enum : unsigned {
+#include "WasmRelocs.def"
+};
+
+#undef WASM_RELOC
+
+// Subset of types that a value can have
+enum class ValType {
+ I32 = WASM_TYPE_I32,
+ I64 = WASM_TYPE_I64,
+ F32 = WASM_TYPE_F32,
+ F64 = WASM_TYPE_F64,
+ V128 = WASM_TYPE_V128,
+ EXNREF = WASM_TYPE_EXNREF,
+};
+
+struct WasmSignature {
+ SmallVector<ValType, 1> Returns;
+ SmallVector<ValType, 4> Params;
+ // Support empty and tombstone instances, needed by DenseMap.
+ enum { Plain, Empty, Tombstone } State = Plain;
+
+ WasmSignature(SmallVector<ValType, 1> &&InReturns,
+ SmallVector<ValType, 4> &&InParams)
+ : Returns(InReturns), Params(InParams) {}
+ WasmSignature() = default;
+};
+
+// Useful comparison operators
+inline bool operator==(const WasmSignature &LHS, const WasmSignature &RHS) {
+ return LHS.State == RHS.State && LHS.Returns == RHS.Returns &&
+ LHS.Params == RHS.Params;
+}
+
+inline bool operator!=(const WasmSignature &LHS, const WasmSignature &RHS) {
+ return !(LHS == RHS);
+}
+
+inline bool operator==(const WasmGlobalType &LHS, const WasmGlobalType &RHS) {
+ return LHS.Type == RHS.Type && LHS.Mutable == RHS.Mutable;
+}
+
+inline bool operator!=(const WasmGlobalType &LHS, const WasmGlobalType &RHS) {
+ return !(LHS == RHS);
+}
+
+std::string toString(WasmSymbolType type);
+std::string relocTypetoString(uint32_t type);
+bool relocTypeHasAddend(uint32_t type);
+
+} // end namespace wasm
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/BinaryFormat/WasmRelocs.def b/third_party/llvm-project/include/llvm/BinaryFormat/WasmRelocs.def
new file mode 100644
index 000000000..00dacf72a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/BinaryFormat/WasmRelocs.def
@@ -0,0 +1,17 @@
+#ifndef WASM_RELOC
+#error "WASM_RELOC must be defined"
+#endif
+
+WASM_RELOC(R_WASM_FUNCTION_INDEX_LEB, 0)
+WASM_RELOC(R_WASM_TABLE_INDEX_SLEB, 1)
+WASM_RELOC(R_WASM_TABLE_INDEX_I32, 2)
+WASM_RELOC(R_WASM_MEMORY_ADDR_LEB, 3)
+WASM_RELOC(R_WASM_MEMORY_ADDR_SLEB, 4)
+WASM_RELOC(R_WASM_MEMORY_ADDR_I32, 5)
+WASM_RELOC(R_WASM_TYPE_INDEX_LEB, 6)
+WASM_RELOC(R_WASM_GLOBAL_INDEX_LEB, 7)
+WASM_RELOC(R_WASM_FUNCTION_OFFSET_I32, 8)
+WASM_RELOC(R_WASM_SECTION_OFFSET_I32, 9)
+WASM_RELOC(R_WASM_EVENT_INDEX_LEB, 10)
+WASM_RELOC(R_WASM_MEMORY_ADDR_REL_SLEB, 11)
+WASM_RELOC(R_WASM_TABLE_INDEX_REL_SLEB, 12)
diff --git a/third_party/llvm-project/include/llvm/Config/abi-breaking.h b/third_party/llvm-project/include/llvm/Config/abi-breaking.h
new file mode 100644
index 000000000..d6bffe75c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Config/abi-breaking.h
@@ -0,0 +1,3 @@
+
+// waka
+
diff --git a/third_party/llvm-project/include/llvm/Config/config.h b/third_party/llvm-project/include/llvm/Config/config.h
new file mode 100644
index 000000000..9cc7f7c31
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Config/config.h
@@ -0,0 +1,4 @@
+
+// waka waka
+
+#include <stdio.h>
diff --git a/third_party/llvm-project/include/llvm/Config/llvm-config.h b/third_party/llvm-project/include/llvm/Config/llvm-config.h
new file mode 100644
index 000000000..8a5702d4b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Config/llvm-config.h
@@ -0,0 +1,9 @@
+
+// This is all terrible
+
+#ifndef _WIN32
+#define LLVM_ON_UNIX
+#endif
+
+// Use simple std:: based atomics, no windows specifics
+#define LLVM_THREADING_USE_STD_CALL_ONCE 1
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DIContext.h b/third_party/llvm-project/include/llvm/DebugInfo/DIContext.h
new file mode 100644
index 000000000..fbebfe634
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DIContext.h
@@ -0,0 +1,302 @@
+//===- DIContext.h ----------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines DIContext, an abstract data structure that holds
+// debug information data.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DICONTEXT_H
+#define LLVM_DEBUGINFO_DICONTEXT_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cassert>
+#include <cstdint>
+#include <memory>
+#include <string>
+#include <tuple>
+#include <utility>
+
+namespace llvm {
+
+/// A format-neutral container for source line information.
+struct DILineInfo {
+ // DILineInfo contains "<invalid>" for function/filename it cannot fetch.
+ static constexpr const char *const BadString = "<invalid>";
+ // Use "??" instead of "<invalid>" to make our output closer to addr2line.
+ static constexpr const char *const Addr2LineBadString = "??";
+ std::string FileName;
+ std::string FunctionName;
+ Optional<StringRef> Source;
+ uint32_t Line = 0;
+ uint32_t Column = 0;
+ uint32_t StartLine = 0;
+
+ // DWARF-specific.
+ uint32_t Discriminator = 0;
+
+ DILineInfo() : FileName(BadString), FunctionName(BadString) {}
+
+ bool operator==(const DILineInfo &RHS) const {
+ return Line == RHS.Line && Column == RHS.Column &&
+ FileName == RHS.FileName && FunctionName == RHS.FunctionName &&
+ StartLine == RHS.StartLine && Discriminator == RHS.Discriminator;
+ }
+
+ bool operator!=(const DILineInfo &RHS) const {
+ return !(*this == RHS);
+ }
+
+ bool operator<(const DILineInfo &RHS) const {
+ return std::tie(FileName, FunctionName, Line, Column, StartLine,
+ Discriminator) <
+ std::tie(RHS.FileName, RHS.FunctionName, RHS.Line, RHS.Column,
+ RHS.StartLine, RHS.Discriminator);
+ }
+
+ explicit operator bool() const { return *this != DILineInfo(); }
+
+ void dump(raw_ostream &OS) {
+ OS << "Line info: ";
+ if (FileName != BadString)
+ OS << "file '" << FileName << "', ";
+ if (FunctionName != BadString)
+ OS << "function '" << FunctionName << "', ";
+ OS << "line " << Line << ", ";
+ OS << "column " << Column << ", ";
+ OS << "start line " << StartLine << '\n';
+ }
+};
+
+using DILineInfoTable = SmallVector<std::pair<uint64_t, DILineInfo>, 16>;
+
+/// A format-neutral container for inlined code description.
+class DIInliningInfo {
+ SmallVector<DILineInfo, 4> Frames;
+
+public:
+ DIInliningInfo() = default;
+
+ const DILineInfo & getFrame(unsigned Index) const {
+ assert(Index < Frames.size());
+ return Frames[Index];
+ }
+
+ DILineInfo *getMutableFrame(unsigned Index) {
+ assert(Index < Frames.size());
+ return &Frames[Index];
+ }
+
+ uint32_t getNumberOfFrames() const {
+ return Frames.size();
+ }
+
+ void addFrame(const DILineInfo &Frame) {
+ Frames.push_back(Frame);
+ }
+
+ void resize(unsigned i) {
+ Frames.resize(i);
+ }
+};
+
+/// Container for description of a global variable.
+struct DIGlobal {
+ std::string Name;
+ uint64_t Start = 0;
+ uint64_t Size = 0;
+
+ DIGlobal() : Name(DILineInfo::BadString) {}
+};
+
+struct DILocal {
+ std::string FunctionName;
+ std::string Name;
+ std::string DeclFile;
+ uint64_t DeclLine = 0;
+ Optional<int64_t> FrameOffset;
+ Optional<uint64_t> Size;
+ Optional<uint64_t> TagOffset;
+};
+
+/// A DINameKind is passed to name search methods to specify a
+/// preference regarding the type of name resolution the caller wants.
+enum class DINameKind { None, ShortName, LinkageName };
+
+/// Controls which fields of DILineInfo container should be filled
+/// with data.
+struct DILineInfoSpecifier {
+ enum class FileLineInfoKind { None, Default, AbsoluteFilePath };
+ using FunctionNameKind = DINameKind;
+
+ FileLineInfoKind FLIKind;
+ FunctionNameKind FNKind;
+
+ DILineInfoSpecifier(FileLineInfoKind FLIKind = FileLineInfoKind::Default,
+ FunctionNameKind FNKind = FunctionNameKind::None)
+ : FLIKind(FLIKind), FNKind(FNKind) {}
+};
+
+/// This is just a helper to programmatically construct DIDumpType.
+enum DIDumpTypeCounter {
+#define HANDLE_DWARF_SECTION(ENUM_NAME, ELF_NAME, CMDLINE_NAME) \
+ DIDT_ID_##ENUM_NAME,
+#include "llvm/BinaryFormat/Dwarf.def"
+#undef HANDLE_DWARF_SECTION
+ DIDT_ID_UUID,
+ DIDT_ID_Count
+};
+static_assert(DIDT_ID_Count <= 32, "section types overflow storage");
+
+/// Selects which debug sections get dumped.
+enum DIDumpType : unsigned {
+ DIDT_Null,
+ DIDT_All = ~0U,
+#define HANDLE_DWARF_SECTION(ENUM_NAME, ELF_NAME, CMDLINE_NAME) \
+ DIDT_##ENUM_NAME = 1U << DIDT_ID_##ENUM_NAME,
+#include "llvm/BinaryFormat/Dwarf.def"
+#undef HANDLE_DWARF_SECTION
+ DIDT_UUID = 1 << DIDT_ID_UUID,
+};
+
+/// Container for dump options that control which debug information will be
+/// dumped.
+struct DIDumpOptions {
+ unsigned DumpType = DIDT_All;
+ unsigned ChildRecurseDepth = -1U;
+ unsigned ParentRecurseDepth = -1U;
+ uint16_t Version = 0; // DWARF version to assume when extracting.
+ uint8_t AddrSize = 4; // Address byte size to assume when extracting.
+ bool ShowAddresses = true;
+ bool ShowChildren = false;
+ bool ShowParents = false;
+ bool ShowForm = false;
+ bool SummarizeTypes = false;
+ bool Verbose = false;
+ bool DisplayRawContents = false;
+
+ /// Return default option set for printing a single DIE without children.
+ static DIDumpOptions getForSingleDIE() {
+ DIDumpOptions Opts;
+ Opts.ChildRecurseDepth = 0;
+ Opts.ParentRecurseDepth = 0;
+ return Opts;
+ }
+
+ /// Return the options with RecurseDepth set to 0 unless explicitly required.
+ DIDumpOptions noImplicitRecursion() const {
+ DIDumpOptions Opts = *this;
+ if (ChildRecurseDepth == -1U && !ShowChildren)
+ Opts.ChildRecurseDepth = 0;
+ if (ParentRecurseDepth == -1U && !ShowParents)
+ Opts.ParentRecurseDepth = 0;
+ return Opts;
+ }
+};
+
+class DIContext {
+public:
+ enum DIContextKind {
+ CK_DWARF,
+ CK_PDB
+ };
+
+ DIContext(DIContextKind K) : Kind(K) {}
+ virtual ~DIContext() = default;
+
+ DIContextKind getKind() const { return Kind; }
+
+ virtual void dump(raw_ostream &OS, DIDumpOptions DumpOpts) = 0;
+
+ virtual bool verify(raw_ostream &OS, DIDumpOptions DumpOpts = {}) {
+ // No verifier? Just say things went well.
+ return true;
+ }
+
+ virtual DILineInfo getLineInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) = 0;
+ virtual DILineInfoTable getLineInfoForAddressRange(
+ object::SectionedAddress Address, uint64_t Size,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) = 0;
+ virtual DIInliningInfo getInliningInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) = 0;
+
+ virtual std::vector<DILocal>
+ getLocalsForAddress(object::SectionedAddress Address) = 0;
+
+private:
+ const DIContextKind Kind;
+};
+
+/// An inferface for inquiring the load address of a loaded object file
+/// to be used by the DIContext implementations when applying relocations
+/// on the fly.
+class LoadedObjectInfo {
+protected:
+ LoadedObjectInfo() = default;
+ LoadedObjectInfo(const LoadedObjectInfo &) = default;
+
+public:
+ virtual ~LoadedObjectInfo() = default;
+
+ /// Obtain the Load Address of a section by SectionRef.
+ ///
+ /// Calculate the address of the given section.
+ /// The section need not be present in the local address space. The addresses
+ /// need to be consistent with the addresses used to query the DIContext and
+ /// the output of this function should be deterministic, i.e. repeated calls
+ /// with the same Sec should give the same address.
+ virtual uint64_t getSectionLoadAddress(const object::SectionRef &Sec) const {
+ return 0;
+ }
+
+ /// If conveniently available, return the content of the given Section.
+ ///
+ /// When the section is available in the local address space, in relocated
+ /// (loaded) form, e.g. because it was relocated by a JIT for execution, this
+ /// function should provide the contents of said section in `Data`. If the
+ /// loaded section is not available, or the cost of retrieving it would be
+ /// prohibitive, this function should return false. In that case, relocations
+ /// will be read from the local (unrelocated) object file and applied on the
+ /// fly. Note that this method is used purely for optimzation purposes in the
+ /// common case of JITting in the local address space, so returning false
+ /// should always be correct.
+ virtual bool getLoadedSectionContents(const object::SectionRef &Sec,
+ StringRef &Data) const {
+ return false;
+ }
+
+ // FIXME: This is untested and unused anywhere in the LLVM project, it's
+ // used/needed by Julia (an external project). It should have some coverage
+ // (at least tests, but ideally example functionality).
+ /// Obtain a copy of this LoadedObjectInfo.
+ virtual std::unique_ptr<LoadedObjectInfo> clone() const = 0;
+};
+
+template <typename Derived, typename Base = LoadedObjectInfo>
+struct LoadedObjectInfoHelper : Base {
+protected:
+ LoadedObjectInfoHelper(const LoadedObjectInfoHelper &) = default;
+ LoadedObjectInfoHelper() = default;
+
+public:
+ template <typename... Ts>
+ LoadedObjectInfoHelper(Ts &&... Args) : Base(std::forward<Ts>(Args)...) {}
+
+ std::unique_ptr<llvm::LoadedObjectInfo> clone() const override {
+ return std::make_unique<Derived>(static_cast<const Derived &>(*this));
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DICONTEXT_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h
new file mode 100644
index 000000000..39ae53c4e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h
@@ -0,0 +1,183 @@
+//===- DWARFAbbreviationDeclaration.h ---------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFABBREVIATIONDECLARATION_H
+#define LLVM_DEBUGINFO_DWARFABBREVIATIONDECLARATION_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+
+namespace llvm {
+
+class DWARFFormValue;
+class DWARFUnit;
+class raw_ostream;
+
+class DWARFAbbreviationDeclaration {
+public:
+ struct AttributeSpec {
+ AttributeSpec(dwarf::Attribute A, dwarf::Form F, int64_t Value)
+ : Attr(A), Form(F), Value(Value) {
+ assert(isImplicitConst());
+ }
+ AttributeSpec(dwarf::Attribute A, dwarf::Form F, Optional<uint8_t> ByteSize)
+ : Attr(A), Form(F) {
+ assert(!isImplicitConst());
+ this->ByteSize.HasByteSize = ByteSize.hasValue();
+ if (this->ByteSize.HasByteSize)
+ this->ByteSize.ByteSize = *ByteSize;
+ }
+
+ dwarf::Attribute Attr;
+ dwarf::Form Form;
+
+ private:
+ /// The following field is used for ByteSize for non-implicit_const
+ /// attributes and as value for implicit_const ones, indicated by
+ /// Form == DW_FORM_implicit_const.
+ /// The following cases are distinguished:
+ /// * Form != DW_FORM_implicit_const and HasByteSize is true:
+ /// ByteSize contains the fixed size in bytes for the Form in this
+ /// object.
+ /// * Form != DW_FORM_implicit_const and HasByteSize is false:
+ /// byte size of Form either varies according to the DWARFUnit
+ /// that it is contained in or the value size varies and must be
+ /// decoded from the debug information in order to determine its size.
+ /// * Form == DW_FORM_implicit_const:
+ /// Value contains value for the implicit_const attribute.
+ struct ByteSizeStorage {
+ bool HasByteSize;
+ uint8_t ByteSize;
+ };
+ union {
+ ByteSizeStorage ByteSize;
+ int64_t Value;
+ };
+
+ public:
+ bool isImplicitConst() const {
+ return Form == dwarf::DW_FORM_implicit_const;
+ }
+
+ int64_t getImplicitConstValue() const {
+ assert(isImplicitConst());
+ return Value;
+ }
+
+ /// Get the fixed byte size of this Form if possible. This function might
+ /// use the DWARFUnit to calculate the size of the Form, like for
+ /// DW_AT_address and DW_AT_ref_addr, so this isn't just an accessor for
+ /// the ByteSize member.
+ Optional<int64_t> getByteSize(const DWARFUnit &U) const;
+ };
+ using AttributeSpecVector = SmallVector<AttributeSpec, 8>;
+
+ DWARFAbbreviationDeclaration();
+
+ uint32_t getCode() const { return Code; }
+ uint8_t getCodeByteSize() const { return CodeByteSize; }
+ dwarf::Tag getTag() const { return Tag; }
+ bool hasChildren() const { return HasChildren; }
+
+ using attr_iterator_range =
+ iterator_range<AttributeSpecVector::const_iterator>;
+
+ attr_iterator_range attributes() const {
+ return attr_iterator_range(AttributeSpecs.begin(), AttributeSpecs.end());
+ }
+
+ dwarf::Form getFormByIndex(uint32_t idx) const {
+ assert(idx < AttributeSpecs.size());
+ return AttributeSpecs[idx].Form;
+ }
+
+ size_t getNumAttributes() const {
+ return AttributeSpecs.size();
+ }
+
+ dwarf::Attribute getAttrByIndex(uint32_t idx) const {
+ assert(idx < AttributeSpecs.size());
+ return AttributeSpecs[idx].Attr;
+ }
+
+ /// Get the index of the specified attribute.
+ ///
+ /// Searches the this abbreviation declaration for the index of the specified
+ /// attribute.
+ ///
+ /// \param attr DWARF attribute to search for.
+ /// \returns Optional index of the attribute if found, None otherwise.
+ Optional<uint32_t> findAttributeIndex(dwarf::Attribute attr) const;
+
+ /// Extract a DWARF form value from a DIE specified by DIE offset.
+ ///
+ /// Extract an attribute value for a DWARFUnit given the DIE offset and the
+ /// attribute.
+ ///
+ /// \param DIEOffset the DIE offset that points to the ULEB128 abbreviation
+ /// code in the .debug_info data.
+ /// \param Attr DWARF attribute to search for.
+ /// \param U the DWARFUnit the contains the DIE.
+ /// \returns Optional DWARF form value if the attribute was extracted.
+ Optional<DWARFFormValue> getAttributeValue(const uint64_t DIEOffset,
+ const dwarf::Attribute Attr,
+ const DWARFUnit &U) const;
+
+ bool extract(DataExtractor Data, uint64_t* OffsetPtr);
+ void dump(raw_ostream &OS) const;
+
+ // Return an optional byte size of all attribute data in this abbreviation
+ // if a constant byte size can be calculated given a DWARFUnit. This allows
+ // DWARF parsing to be faster as many DWARF DIEs have a fixed byte size.
+ Optional<size_t> getFixedAttributesByteSize(const DWARFUnit &U) const;
+
+private:
+ void clear();
+
+ /// A helper structure that can quickly determine the size in bytes of an
+ /// abbreviation declaration.
+ struct FixedSizeInfo {
+ /// The fixed byte size for fixed size forms.
+ uint16_t NumBytes = 0;
+ /// Number of DW_FORM_address forms in this abbrevation declaration.
+ uint8_t NumAddrs = 0;
+ /// Number of DW_FORM_ref_addr forms in this abbrevation declaration.
+ uint8_t NumRefAddrs = 0;
+ /// Number of 4 byte in DWARF32 and 8 byte in DWARF64 forms.
+ uint8_t NumDwarfOffsets = 0;
+
+ FixedSizeInfo() = default;
+
+ /// Calculate the fixed size in bytes given a DWARFUnit.
+ ///
+ /// \param U the DWARFUnit to use when determing the byte size.
+ /// \returns the size in bytes for all attribute data in this abbreviation.
+ /// The returned size does not include bytes for the ULEB128 abbreviation
+ /// code
+ size_t getByteSize(const DWARFUnit &U) const;
+ };
+
+ uint32_t Code;
+ dwarf::Tag Tag;
+ uint8_t CodeByteSize;
+ bool HasChildren;
+ AttributeSpecVector AttributeSpecs;
+ /// If this abbreviation has a fixed byte size then FixedAttributeSize member
+ /// variable below will have a value.
+ Optional<FixedSizeInfo> FixedAttributeSize;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFABBREVIATIONDECLARATION_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h
new file mode 100644
index 000000000..c9042e593
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h
@@ -0,0 +1,599 @@
+//===- DWARFAcceleratorTable.h ----------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFACCELERATORTABLE_H
+#define LLVM_DEBUGINFO_DWARFACCELERATORTABLE_H
+
+#include "llvm/ADT/DenseSet.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+class raw_ostream;
+class ScopedPrinter;
+
+/// The accelerator tables are designed to allow efficient random access
+/// (using a symbol name as a key) into debug info by providing an index of the
+/// debug info DIEs. This class implements the common functionality of Apple and
+/// DWARF 5 accelerator tables.
+/// TODO: Generalize the rest of the AppleAcceleratorTable interface and move it
+/// to this class.
+class DWARFAcceleratorTable {
+protected:
+ DWARFDataExtractor AccelSection;
+ DataExtractor StringSection;
+
+public:
+ /// An abstract class representing a single entry in the accelerator tables.
+ class Entry {
+ protected:
+ SmallVector<DWARFFormValue, 3> Values;
+
+ Entry() = default;
+
+ // Make these protected so only (final) subclasses can be copied around.
+ Entry(const Entry &) = default;
+ Entry(Entry &&) = default;
+ Entry &operator=(const Entry &) = default;
+ Entry &operator=(Entry &&) = default;
+ ~Entry() = default;
+
+
+ public:
+ /// Returns the Offset of the Compilation Unit associated with this
+ /// Accelerator Entry or None if the Compilation Unit offset is not recorded
+ /// in this Accelerator Entry.
+ virtual Optional<uint64_t> getCUOffset() const = 0;
+
+ /// Returns the Tag of the Debug Info Entry associated with this
+ /// Accelerator Entry or None if the Tag is not recorded in this
+ /// Accelerator Entry.
+ virtual Optional<dwarf::Tag> getTag() const = 0;
+
+ /// Returns the raw values of fields in the Accelerator Entry. In general,
+ /// these can only be interpreted with the help of the metadata in the
+ /// owning Accelerator Table.
+ ArrayRef<DWARFFormValue> getValues() const { return Values; }
+ };
+
+ DWARFAcceleratorTable(const DWARFDataExtractor &AccelSection,
+ DataExtractor StringSection)
+ : AccelSection(AccelSection), StringSection(StringSection) {}
+ virtual ~DWARFAcceleratorTable();
+
+ virtual Error extract() = 0;
+ virtual void dump(raw_ostream &OS) const = 0;
+
+ DWARFAcceleratorTable(const DWARFAcceleratorTable &) = delete;
+ void operator=(const DWARFAcceleratorTable &) = delete;
+};
+
+/// This implements the Apple accelerator table format, a precursor of the
+/// DWARF 5 accelerator table format.
+class AppleAcceleratorTable : public DWARFAcceleratorTable {
+ struct Header {
+ uint32_t Magic;
+ uint16_t Version;
+ uint16_t HashFunction;
+ uint32_t BucketCount;
+ uint32_t HashCount;
+ uint32_t HeaderDataLength;
+
+ void dump(ScopedPrinter &W) const;
+ };
+
+ struct HeaderData {
+ using AtomType = uint16_t;
+ using Form = dwarf::Form;
+
+ uint64_t DIEOffsetBase;
+ SmallVector<std::pair<AtomType, Form>, 3> Atoms;
+
+ Optional<uint64_t> extractOffset(Optional<DWARFFormValue> Value) const;
+ };
+
+ struct Header Hdr;
+ struct HeaderData HdrData;
+ bool IsValid = false;
+
+ /// Returns true if we should continue scanning for entries or false if we've
+ /// reached the last (sentinel) entry of encountered a parsing error.
+ bool dumpName(ScopedPrinter &W, SmallVectorImpl<DWARFFormValue> &AtomForms,
+ uint64_t *DataOffset) const;
+
+public:
+ /// Apple-specific implementation of an Accelerator Entry.
+ class Entry final : public DWARFAcceleratorTable::Entry {
+ const HeaderData *HdrData = nullptr;
+
+ Entry(const HeaderData &Data);
+ Entry() = default;
+
+ void extract(const AppleAcceleratorTable &AccelTable, uint64_t *Offset);
+
+ public:
+ Optional<uint64_t> getCUOffset() const override;
+
+ /// Returns the Section Offset of the Debug Info Entry associated with this
+ /// Accelerator Entry or None if the DIE offset is not recorded in this
+ /// Accelerator Entry. The returned offset is relative to the start of the
+ /// Section containing the DIE.
+ Optional<uint64_t> getDIESectionOffset() const;
+
+ Optional<dwarf::Tag> getTag() const override;
+
+ /// Returns the value of the Atom in this Accelerator Entry, if the Entry
+ /// contains such Atom.
+ Optional<DWARFFormValue> lookup(HeaderData::AtomType Atom) const;
+
+ friend class AppleAcceleratorTable;
+ friend class ValueIterator;
+ };
+
+ class ValueIterator : public std::iterator<std::input_iterator_tag, Entry> {
+ const AppleAcceleratorTable *AccelTable = nullptr;
+ Entry Current; ///< The current entry.
+ uint64_t DataOffset = 0; ///< Offset into the section.
+ unsigned Data = 0; ///< Current data entry.
+ unsigned NumData = 0; ///< Number of data entries.
+
+ /// Advance the iterator.
+ void Next();
+ public:
+ /// Construct a new iterator for the entries at \p DataOffset.
+ ValueIterator(const AppleAcceleratorTable &AccelTable, uint64_t DataOffset);
+ /// End marker.
+ ValueIterator() = default;
+
+ const Entry &operator*() const { return Current; }
+ ValueIterator &operator++() { Next(); return *this; }
+ ValueIterator operator++(int) {
+ ValueIterator I = *this;
+ Next();
+ return I;
+ }
+ friend bool operator==(const ValueIterator &A, const ValueIterator &B) {
+ return A.NumData == B.NumData && A.DataOffset == B.DataOffset;
+ }
+ friend bool operator!=(const ValueIterator &A, const ValueIterator &B) {
+ return !(A == B);
+ }
+ };
+
+ AppleAcceleratorTable(const DWARFDataExtractor &AccelSection,
+ DataExtractor StringSection)
+ : DWARFAcceleratorTable(AccelSection, StringSection) {}
+
+ Error extract() override;
+ uint32_t getNumBuckets();
+ uint32_t getNumHashes();
+ uint32_t getSizeHdr();
+ uint32_t getHeaderDataLength();
+
+ /// Return the Atom description, which can be used to interpret the raw values
+ /// of the Accelerator Entries in this table.
+ ArrayRef<std::pair<HeaderData::AtomType, HeaderData::Form>> getAtomsDesc();
+ bool validateForms();
+
+ /// Return information related to the DWARF DIE we're looking for when
+ /// performing a lookup by name.
+ ///
+ /// \param HashDataOffset an offset into the hash data table
+ /// \returns <DieOffset, DieTag>
+ /// DieOffset is the offset into the .debug_info section for the DIE
+ /// related to the input hash data offset.
+ /// DieTag is the tag of the DIE
+ std::pair<uint64_t, dwarf::Tag> readAtoms(uint64_t *HashDataOffset);
+ void dump(raw_ostream &OS) const override;
+
+ /// Look up all entries in the accelerator table matching \c Key.
+ iterator_range<ValueIterator> equal_range(StringRef Key) const;
+};
+
+/// .debug_names section consists of one or more units. Each unit starts with a
+/// header, which is followed by a list of compilation units, local and foreign
+/// type units.
+///
+/// These may be followed by an (optional) hash lookup table, which consists of
+/// an array of buckets and hashes similar to the apple tables above. The only
+/// difference is that the hashes array is 1-based, and consequently an empty
+/// bucket is denoted by 0 and not UINT32_MAX.
+///
+/// Next is the name table, which consists of an array of names and array of
+/// entry offsets. This is different from the apple tables, which store names
+/// next to the actual entries.
+///
+/// The structure of the entries is described by an abbreviations table, which
+/// comes after the name table. Unlike the apple tables, which have a uniform
+/// entry structure described in the header, each .debug_names entry may have
+/// different index attributes (DW_IDX_???) attached to it.
+///
+/// The last segment consists of a list of entries, which is a 0-terminated list
+/// referenced by the name table and interpreted with the help of the
+/// abbreviation table.
+class DWARFDebugNames : public DWARFAcceleratorTable {
+ /// The fixed-size part of a DWARF v5 Name Index header
+ struct HeaderPOD {
+ uint32_t UnitLength;
+ uint16_t Version;
+ uint16_t Padding;
+ uint32_t CompUnitCount;
+ uint32_t LocalTypeUnitCount;
+ uint32_t ForeignTypeUnitCount;
+ uint32_t BucketCount;
+ uint32_t NameCount;
+ uint32_t AbbrevTableSize;
+ uint32_t AugmentationStringSize;
+ };
+
+public:
+ class NameIndex;
+ class NameIterator;
+ class ValueIterator;
+
+ /// DWARF v5 Name Index header.
+ struct Header : public HeaderPOD {
+ SmallString<8> AugmentationString;
+
+ Error extract(const DWARFDataExtractor &AS, uint64_t *Offset);
+ void dump(ScopedPrinter &W) const;
+ };
+
+ /// Index attribute and its encoding.
+ struct AttributeEncoding {
+ dwarf::Index Index;
+ dwarf::Form Form;
+
+ constexpr AttributeEncoding(dwarf::Index Index, dwarf::Form Form)
+ : Index(Index), Form(Form) {}
+
+ friend bool operator==(const AttributeEncoding &LHS,
+ const AttributeEncoding &RHS) {
+ return LHS.Index == RHS.Index && LHS.Form == RHS.Form;
+ }
+ };
+
+ /// Abbreviation describing the encoding of Name Index entries.
+ struct Abbrev {
+ uint32_t Code; ///< Abbreviation code
+ dwarf::Tag Tag; ///< Dwarf Tag of the described entity.
+ std::vector<AttributeEncoding> Attributes; ///< List of index attributes.
+
+ Abbrev(uint32_t Code, dwarf::Tag Tag,
+ std::vector<AttributeEncoding> Attributes)
+ : Code(Code), Tag(Tag), Attributes(std::move(Attributes)) {}
+
+ void dump(ScopedPrinter &W) const;
+ };
+
+ /// DWARF v5-specific implementation of an Accelerator Entry.
+ class Entry final : public DWARFAcceleratorTable::Entry {
+ const NameIndex *NameIdx;
+ const Abbrev *Abbr;
+
+ Entry(const NameIndex &NameIdx, const Abbrev &Abbr);
+
+ public:
+ Optional<uint64_t> getCUOffset() const override;
+ Optional<dwarf::Tag> getTag() const override { return tag(); }
+
+ /// Returns the Index into the Compilation Unit list of the owning Name
+ /// Index or None if this Accelerator Entry does not have an associated
+ /// Compilation Unit. It is up to the user to verify that the returned Index
+ /// is valid in the owning NameIndex (or use getCUOffset(), which will
+ /// handle that check itself). Note that entries in NameIndexes which index
+ /// just a single Compilation Unit are implicitly associated with that unit,
+ /// so this function will return 0 even without an explicit
+ /// DW_IDX_compile_unit attribute.
+ Optional<uint64_t> getCUIndex() const;
+
+ /// .debug_names-specific getter, which always succeeds (DWARF v5 index
+ /// entries always have a tag).
+ dwarf::Tag tag() const { return Abbr->Tag; }
+
+ /// Returns the Offset of the DIE within the containing CU or TU.
+ Optional<uint64_t> getDIEUnitOffset() const;
+
+ /// Return the Abbreviation that can be used to interpret the raw values of
+ /// this Accelerator Entry.
+ const Abbrev &getAbbrev() const { return *Abbr; }
+
+ /// Returns the value of the Index Attribute in this Accelerator Entry, if
+ /// the Entry contains such Attribute.
+ Optional<DWARFFormValue> lookup(dwarf::Index Index) const;
+
+ void dump(ScopedPrinter &W) const;
+
+ friend class NameIndex;
+ friend class ValueIterator;
+ };
+
+ /// Error returned by NameIndex::getEntry to report it has reached the end of
+ /// the entry list.
+ class SentinelError : public ErrorInfo<SentinelError> {
+ public:
+ static char ID;
+
+ void log(raw_ostream &OS) const override { OS << "Sentinel"; }
+ std::error_code convertToErrorCode() const override;
+ };
+
+private:
+ /// DenseMapInfo for struct Abbrev.
+ struct AbbrevMapInfo {
+ static Abbrev getEmptyKey();
+ static Abbrev getTombstoneKey();
+ static unsigned getHashValue(uint32_t Code) {
+ return DenseMapInfo<uint32_t>::getHashValue(Code);
+ }
+ static unsigned getHashValue(const Abbrev &Abbr) {
+ return getHashValue(Abbr.Code);
+ }
+ static bool isEqual(uint32_t LHS, const Abbrev &RHS) {
+ return LHS == RHS.Code;
+ }
+ static bool isEqual(const Abbrev &LHS, const Abbrev &RHS) {
+ return LHS.Code == RHS.Code;
+ }
+ };
+
+public:
+ /// A single entry in the Name Table (DWARF v5 sect. 6.1.1.4.6) of the Name
+ /// Index.
+ class NameTableEntry {
+ DataExtractor StrData;
+
+ uint32_t Index;
+ uint64_t StringOffset;
+ uint64_t EntryOffset;
+
+ public:
+ NameTableEntry(const DataExtractor &StrData, uint32_t Index,
+ uint64_t StringOffset, uint64_t EntryOffset)
+ : StrData(StrData), Index(Index), StringOffset(StringOffset),
+ EntryOffset(EntryOffset) {}
+
+ /// Return the index of this name in the parent Name Index.
+ uint32_t getIndex() const { return Index; }
+
+ /// Returns the offset of the name of the described entities.
+ uint64_t getStringOffset() const { return StringOffset; }
+
+ /// Return the string referenced by this name table entry or nullptr if the
+ /// string offset is not valid.
+ const char *getString() const {
+ uint64_t Off = StringOffset;
+ return StrData.getCStr(&Off);
+ }
+
+ /// Returns the offset of the first Entry in the list.
+ uint64_t getEntryOffset() const { return EntryOffset; }
+ };
+
+ /// Represents a single accelerator table within the DWARF v5 .debug_names
+ /// section.
+ class NameIndex {
+ DenseSet<Abbrev, AbbrevMapInfo> Abbrevs;
+ struct Header Hdr;
+ const DWARFDebugNames &Section;
+
+ // Base of the whole unit and of various important tables, as offsets from
+ // the start of the section.
+ uint64_t Base;
+ uint64_t CUsBase;
+ uint64_t BucketsBase;
+ uint64_t HashesBase;
+ uint64_t StringOffsetsBase;
+ uint64_t EntryOffsetsBase;
+ uint64_t EntriesBase;
+
+ void dumpCUs(ScopedPrinter &W) const;
+ void dumpLocalTUs(ScopedPrinter &W) const;
+ void dumpForeignTUs(ScopedPrinter &W) const;
+ void dumpAbbreviations(ScopedPrinter &W) const;
+ bool dumpEntry(ScopedPrinter &W, uint64_t *Offset) const;
+ void dumpName(ScopedPrinter &W, const NameTableEntry &NTE,
+ Optional<uint32_t> Hash) const;
+ void dumpBucket(ScopedPrinter &W, uint32_t Bucket) const;
+
+ Expected<AttributeEncoding> extractAttributeEncoding(uint64_t *Offset);
+
+ Expected<std::vector<AttributeEncoding>>
+ extractAttributeEncodings(uint64_t *Offset);
+
+ Expected<Abbrev> extractAbbrev(uint64_t *Offset);
+
+ public:
+ NameIndex(const DWARFDebugNames &Section, uint64_t Base)
+ : Section(Section), Base(Base) {}
+
+ /// Reads offset of compilation unit CU. CU is 0-based.
+ uint64_t getCUOffset(uint32_t CU) const;
+ uint32_t getCUCount() const { return Hdr.CompUnitCount; }
+
+ /// Reads offset of local type unit TU, TU is 0-based.
+ uint64_t getLocalTUOffset(uint32_t TU) const;
+ uint32_t getLocalTUCount() const { return Hdr.LocalTypeUnitCount; }
+
+ /// Reads signature of foreign type unit TU. TU is 0-based.
+ uint64_t getForeignTUSignature(uint32_t TU) const;
+ uint32_t getForeignTUCount() const { return Hdr.ForeignTypeUnitCount; }
+
+ /// Reads an entry in the Bucket Array for the given Bucket. The returned
+ /// value is a (1-based) index into the Names, StringOffsets and
+ /// EntryOffsets arrays. The input Bucket index is 0-based.
+ uint32_t getBucketArrayEntry(uint32_t Bucket) const;
+ uint32_t getBucketCount() const { return Hdr.BucketCount; }
+
+ /// Reads an entry in the Hash Array for the given Index. The input Index
+ /// is 1-based.
+ uint32_t getHashArrayEntry(uint32_t Index) const;
+
+ /// Reads an entry in the Name Table for the given Index. The Name Table
+ /// consists of two arrays -- String Offsets and Entry Offsets. The returned
+ /// offsets are relative to the starts of respective sections. Input Index
+ /// is 1-based.
+ NameTableEntry getNameTableEntry(uint32_t Index) const;
+
+ uint32_t getNameCount() const { return Hdr.NameCount; }
+
+ const DenseSet<Abbrev, AbbrevMapInfo> &getAbbrevs() const {
+ return Abbrevs;
+ }
+
+ Expected<Entry> getEntry(uint64_t *Offset) const;
+
+ /// Look up all entries in this Name Index matching \c Key.
+ iterator_range<ValueIterator> equal_range(StringRef Key) const;
+
+ NameIterator begin() const { return NameIterator(this, 1); }
+ NameIterator end() const { return NameIterator(this, getNameCount() + 1); }
+
+ Error extract();
+ uint64_t getUnitOffset() const { return Base; }
+ uint64_t getNextUnitOffset() const { return Base + 4 + Hdr.UnitLength; }
+ void dump(ScopedPrinter &W) const;
+
+ friend class DWARFDebugNames;
+ };
+
+ class ValueIterator : public std::iterator<std::input_iterator_tag, Entry> {
+
+ /// The Name Index we are currently iterating through. The implementation
+ /// relies on the fact that this can also be used as an iterator into the
+ /// "NameIndices" vector in the Accelerator section.
+ const NameIndex *CurrentIndex = nullptr;
+
+ /// Whether this is a local iterator (searches in CurrentIndex only) or not
+ /// (searches all name indices).
+ bool IsLocal;
+
+ Optional<Entry> CurrentEntry;
+ uint64_t DataOffset = 0; ///< Offset into the section.
+ std::string Key; ///< The Key we are searching for.
+ Optional<uint32_t> Hash; ///< Hash of Key, if it has been computed.
+
+ bool getEntryAtCurrentOffset();
+ Optional<uint64_t> findEntryOffsetInCurrentIndex();
+ bool findInCurrentIndex();
+ void searchFromStartOfCurrentIndex();
+ void next();
+
+ /// Set the iterator to the "end" state.
+ void setEnd() { *this = ValueIterator(); }
+
+ public:
+ /// Create a "begin" iterator for looping over all entries in the
+ /// accelerator table matching Key. The iterator will run through all Name
+ /// Indexes in the section in sequence.
+ ValueIterator(const DWARFDebugNames &AccelTable, StringRef Key);
+
+ /// Create a "begin" iterator for looping over all entries in a specific
+ /// Name Index. Other indices in the section will not be visited.
+ ValueIterator(const NameIndex &NI, StringRef Key);
+
+ /// End marker.
+ ValueIterator() = default;
+
+ const Entry &operator*() const { return *CurrentEntry; }
+ ValueIterator &operator++() {
+ next();
+ return *this;
+ }
+ ValueIterator operator++(int) {
+ ValueIterator I = *this;
+ next();
+ return I;
+ }
+
+ friend bool operator==(const ValueIterator &A, const ValueIterator &B) {
+ return A.CurrentIndex == B.CurrentIndex && A.DataOffset == B.DataOffset;
+ }
+ friend bool operator!=(const ValueIterator &A, const ValueIterator &B) {
+ return !(A == B);
+ }
+ };
+
+ class NameIterator {
+
+ /// The Name Index we are iterating through.
+ const NameIndex *CurrentIndex;
+
+ /// The current name in the Name Index.
+ uint32_t CurrentName;
+
+ void next() {
+ assert(CurrentName <= CurrentIndex->getNameCount());
+ ++CurrentName;
+ }
+
+ public:
+ using iterator_category = std::input_iterator_tag;
+ using value_type = NameTableEntry;
+ using difference_type = uint32_t;
+ using pointer = NameTableEntry *;
+ using reference = NameTableEntry; // We return entries by value.
+
+ /// Creates an iterator whose initial position is name CurrentName in
+ /// CurrentIndex.
+ NameIterator(const NameIndex *CurrentIndex, uint32_t CurrentName)
+ : CurrentIndex(CurrentIndex), CurrentName(CurrentName) {}
+
+ NameTableEntry operator*() const {
+ return CurrentIndex->getNameTableEntry(CurrentName);
+ }
+ NameIterator &operator++() {
+ next();
+ return *this;
+ }
+ NameIterator operator++(int) {
+ NameIterator I = *this;
+ next();
+ return I;
+ }
+
+ friend bool operator==(const NameIterator &A, const NameIterator &B) {
+ return A.CurrentIndex == B.CurrentIndex && A.CurrentName == B.CurrentName;
+ }
+ friend bool operator!=(const NameIterator &A, const NameIterator &B) {
+ return !(A == B);
+ }
+ };
+
+private:
+ SmallVector<NameIndex, 0> NameIndices;
+ DenseMap<uint64_t, const NameIndex *> CUToNameIndex;
+
+public:
+ DWARFDebugNames(const DWARFDataExtractor &AccelSection,
+ DataExtractor StringSection)
+ : DWARFAcceleratorTable(AccelSection, StringSection) {}
+
+ Error extract() override;
+ void dump(raw_ostream &OS) const override;
+
+ /// Look up all entries in the accelerator table matching \c Key.
+ iterator_range<ValueIterator> equal_range(StringRef Key) const;
+
+ using const_iterator = SmallVector<NameIndex, 0>::const_iterator;
+ const_iterator begin() const { return NameIndices.begin(); }
+ const_iterator end() const { return NameIndices.end(); }
+
+ /// Return the Name Index covering the compile unit at CUOffset, or nullptr if
+ /// there is no Name Index covering that unit.
+ const NameIndex *getCUNameIndex(uint64_t CUOffset);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFACCELERATORTABLE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAddressRange.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAddressRange.h
new file mode 100644
index 000000000..2d5f9f3c7
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAddressRange.h
@@ -0,0 +1,61 @@
+//===- DWARFAddressRange.h --------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFADDRESSRANGE_H
+#define LLVM_DEBUGINFO_DWARF_DWARFADDRESSRANGE_H
+
+#include "llvm/DebugInfo/DIContext.h"
+#include <cstdint>
+#include <tuple>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+struct DWARFAddressRange {
+ uint64_t LowPC;
+ uint64_t HighPC;
+ uint64_t SectionIndex;
+
+ DWARFAddressRange() = default;
+
+ /// Used for unit testing.
+ DWARFAddressRange(uint64_t LowPC, uint64_t HighPC, uint64_t SectionIndex = 0)
+ : LowPC(LowPC), HighPC(HighPC), SectionIndex(SectionIndex) {}
+
+ /// Returns true if LowPC is smaller or equal to HighPC. This accounts for
+ /// dead-stripped ranges.
+ bool valid() const { return LowPC <= HighPC; }
+
+ /// Returns true if [LowPC, HighPC) intersects with [RHS.LowPC, RHS.HighPC).
+ bool intersects(const DWARFAddressRange &RHS) const {
+ assert(valid() && RHS.valid());
+ // Empty ranges can't intersect.
+ if (LowPC == HighPC || RHS.LowPC == RHS.HighPC)
+ return false;
+ return LowPC < RHS.HighPC && RHS.LowPC < HighPC;
+ }
+
+ void dump(raw_ostream &OS, uint32_t AddressSize,
+ DIDumpOptions DumpOpts = {}) const;
+};
+
+static inline bool operator<(const DWARFAddressRange &LHS,
+ const DWARFAddressRange &RHS) {
+ return std::tie(LHS.LowPC, LHS.HighPC) < std::tie(RHS.LowPC, RHS.HighPC);
+}
+
+raw_ostream &operator<<(raw_ostream &OS, const DWARFAddressRange &R);
+
+/// DWARFAddressRangesVector - represents a set of absolute address ranges.
+using DWARFAddressRangesVector = std::vector<DWARFAddressRange>;
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFADDRESSRANGE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAttribute.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAttribute.h
new file mode 100644
index 000000000..dfc778346
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFAttribute.h
@@ -0,0 +1,49 @@
+//===- DWARFAttribute.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFATTRIBUTE_H
+#define LLVM_DEBUGINFO_DWARFATTRIBUTE_H
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include <cstdint>
+
+namespace llvm {
+
+//===----------------------------------------------------------------------===//
+/// Encapsulates a DWARF attribute value and all of the data required to
+/// describe the attribute value.
+///
+/// This class is designed to be used by clients that want to iterate across all
+/// attributes in a DWARFDie.
+struct DWARFAttribute {
+ /// The debug info/types offset for this attribute.
+ uint64_t Offset = 0;
+ /// The debug info/types section byte size of the data for this attribute.
+ uint32_t ByteSize = 0;
+ /// The attribute enumeration of this attribute.
+ dwarf::Attribute Attr = dwarf::Attribute(0);
+ /// The form and value for this attribute.
+ DWARFFormValue Value;
+
+ bool isValid() const {
+ return Offset != 0 && Attr != dwarf::Attribute(0);
+ }
+
+ explicit operator bool() const {
+ return isValid();
+ }
+
+ /// Identifies DWARF attributes that may contain a reference to a
+ /// DWARF expression.
+ static bool mayHaveLocationDescription(dwarf::Attribute Attr);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFATTRIBUTE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFCompileUnit.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFCompileUnit.h
new file mode 100644
index 000000000..16b9bfb5d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFCompileUnit.h
@@ -0,0 +1,38 @@
+//===- DWARFCompileUnit.h ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFCOMPILEUNIT_H
+#define LLVM_DEBUGINFO_DWARFCOMPILEUNIT_H
+
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+
+namespace llvm {
+
+class DWARFCompileUnit : public DWARFUnit {
+public:
+ DWARFCompileUnit(DWARFContext &Context, const DWARFSection &Section,
+ const DWARFUnitHeader &Header, const DWARFDebugAbbrev *DA,
+ const DWARFSection *RS, const DWARFSection *LocSection,
+ StringRef SS, const DWARFSection &SOS,
+ const DWARFSection *AOS, const DWARFSection &LS, bool LE,
+ bool IsDWO, const DWARFUnitVector &UnitVector)
+ : DWARFUnit(Context, Section, Header, DA, RS, LocSection, SS, SOS, AOS,
+ LS, LE, IsDWO, UnitVector) {}
+
+ /// VTable anchor.
+ ~DWARFCompileUnit() override;
+ /// Dump this compile unit to \p OS.
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts) override;
+ /// Enable LLVM-style RTTI.
+ static bool classof(const DWARFUnit *U) { return !U->isTypeUnit(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFCOMPILEUNIT_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFContext.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFContext.h
new file mode 100644
index 000000000..2dec107d1
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFContext.h
@@ -0,0 +1,382 @@
+//===- DWARFContext.h -------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===/
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
+#define LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
+
+#include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAranges.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugFrame.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLoc.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugMacro.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFGdbIndex.h"
+#include "llvm/DebugInfo/DWARF/DWARFObject.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/DebugInfo/DWARF/DWARFTypeUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Host.h"
+#include <cstdint>
+#include <deque>
+#include <map>
+#include <memory>
+
+namespace llvm {
+
+class MCRegisterInfo;
+class MemoryBuffer;
+class raw_ostream;
+
+/// Used as a return value for a error callback passed to DWARF context.
+/// Callback should return Halt if client application wants to stop
+/// object parsing, or should return Continue otherwise.
+enum class ErrorPolicy { Halt, Continue };
+
+/// DWARFContext
+/// This data structure is the top level entity that deals with dwarf debug
+/// information parsing. The actual data is supplied through DWARFObj.
+class DWARFContext : public DIContext {
+ DWARFUnitVector NormalUnits;
+ std::unique_ptr<DWARFUnitIndex> CUIndex;
+ std::unique_ptr<DWARFGdbIndex> GdbIndex;
+ std::unique_ptr<DWARFUnitIndex> TUIndex;
+ std::unique_ptr<DWARFDebugAbbrev> Abbrev;
+ std::unique_ptr<DWARFDebugLoc> Loc;
+ std::unique_ptr<DWARFDebugAranges> Aranges;
+ std::unique_ptr<DWARFDebugLine> Line;
+ std::unique_ptr<DWARFDebugFrame> DebugFrame;
+ std::unique_ptr<DWARFDebugFrame> EHFrame;
+ std::unique_ptr<DWARFDebugMacro> Macro;
+ std::unique_ptr<DWARFDebugNames> Names;
+ std::unique_ptr<AppleAcceleratorTable> AppleNames;
+ std::unique_ptr<AppleAcceleratorTable> AppleTypes;
+ std::unique_ptr<AppleAcceleratorTable> AppleNamespaces;
+ std::unique_ptr<AppleAcceleratorTable> AppleObjC;
+
+ DWARFUnitVector DWOUnits;
+ std::unique_ptr<DWARFDebugAbbrev> AbbrevDWO;
+
+ /// The maximum DWARF version of all units.
+ unsigned MaxVersion = 0;
+
+ struct DWOFile {
+ object::OwningBinary<object::ObjectFile> File;
+ std::unique_ptr<DWARFContext> Context;
+ };
+ StringMap<std::weak_ptr<DWOFile>> DWOFiles;
+ std::weak_ptr<DWOFile> DWP;
+ bool CheckedForDWP = false;
+ std::string DWPName;
+
+ std::unique_ptr<MCRegisterInfo> RegInfo;
+
+ /// Read compile units from the debug_info section (if necessary)
+ /// and type units from the debug_types sections (if necessary)
+ /// and store them in NormalUnits.
+ void parseNormalUnits();
+
+ /// Read compile units from the debug_info.dwo section (if necessary)
+ /// and type units from the debug_types.dwo section (if necessary)
+ /// and store them in DWOUnits.
+ /// If \p Lazy is true, set up to parse but don't actually parse them.
+ enum { EagerParse = false, LazyParse = true };
+ void parseDWOUnits(bool Lazy = false);
+
+ std::unique_ptr<const DWARFObject> DObj;
+
+public:
+ DWARFContext(std::unique_ptr<const DWARFObject> DObj,
+ std::string DWPName = "");
+ ~DWARFContext();
+
+ DWARFContext(DWARFContext &) = delete;
+ DWARFContext &operator=(DWARFContext &) = delete;
+
+ const DWARFObject &getDWARFObj() const { return *DObj; }
+
+ static bool classof(const DIContext *DICtx) {
+ return DICtx->getKind() == CK_DWARF;
+ }
+
+ /// Dump a textual representation to \p OS. If any \p DumpOffsets are present,
+ /// dump only the record at the specified offset.
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts,
+ std::array<Optional<uint64_t>, DIDT_ID_Count> DumpOffsets);
+
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts) override {
+ std::array<Optional<uint64_t>, DIDT_ID_Count> DumpOffsets;
+ dump(OS, DumpOpts, DumpOffsets);
+ }
+
+ bool verify(raw_ostream &OS, DIDumpOptions DumpOpts = {}) override;
+
+ using unit_iterator_range = DWARFUnitVector::iterator_range;
+
+ /// Get units from .debug_info in this context.
+ unit_iterator_range info_section_units() {
+ parseNormalUnits();
+ return unit_iterator_range(NormalUnits.begin(),
+ NormalUnits.begin() +
+ NormalUnits.getNumInfoUnits());
+ }
+
+ /// Get units from .debug_types in this context.
+ unit_iterator_range types_section_units() {
+ parseNormalUnits();
+ return unit_iterator_range(
+ NormalUnits.begin() + NormalUnits.getNumInfoUnits(), NormalUnits.end());
+ }
+
+ /// Get compile units in this context.
+ unit_iterator_range compile_units() { return info_section_units(); }
+
+ /// Get type units in this context.
+ unit_iterator_range type_units() { return types_section_units(); }
+
+ /// Get all normal compile/type units in this context.
+ unit_iterator_range normal_units() {
+ parseNormalUnits();
+ return unit_iterator_range(NormalUnits.begin(), NormalUnits.end());
+ }
+
+ /// Get units from .debug_info..dwo in the DWO context.
+ unit_iterator_range dwo_info_section_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin(),
+ DWOUnits.begin() + DWOUnits.getNumInfoUnits());
+ }
+
+ /// Get units from .debug_types.dwo in the DWO context.
+ unit_iterator_range dwo_types_section_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin() + DWOUnits.getNumInfoUnits(),
+ DWOUnits.end());
+ }
+
+ /// Get compile units in the DWO context.
+ unit_iterator_range dwo_compile_units() { return dwo_info_section_units(); }
+
+ /// Get type units in the DWO context.
+ unit_iterator_range dwo_type_units() { return dwo_types_section_units(); }
+
+ /// Get all units in the DWO context.
+ unit_iterator_range dwo_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin(), DWOUnits.end());
+ }
+
+ /// Get the number of compile units in this context.
+ unsigned getNumCompileUnits() {
+ parseNormalUnits();
+ return NormalUnits.getNumInfoUnits();
+ }
+
+ /// Get the number of type units in this context.
+ unsigned getNumTypeUnits() {
+ parseNormalUnits();
+ return NormalUnits.getNumTypesUnits();
+ }
+
+ /// Get the number of compile units in the DWO context.
+ unsigned getNumDWOCompileUnits() {
+ parseDWOUnits();
+ return DWOUnits.getNumInfoUnits();
+ }
+
+ /// Get the number of type units in the DWO context.
+ unsigned getNumDWOTypeUnits() {
+ parseDWOUnits();
+ return DWOUnits.getNumTypesUnits();
+ }
+
+ /// Get the unit at the specified index.
+ DWARFUnit *getUnitAtIndex(unsigned index) {
+ parseNormalUnits();
+ return NormalUnits[index].get();
+ }
+
+ /// Get the unit at the specified index for the DWO units.
+ DWARFUnit *getDWOUnitAtIndex(unsigned index) {
+ parseDWOUnits();
+ return DWOUnits[index].get();
+ }
+
+ DWARFCompileUnit *getDWOCompileUnitForHash(uint64_t Hash);
+
+ /// Return the compile unit that includes an offset (relative to .debug_info).
+ DWARFCompileUnit *getCompileUnitForOffset(uint64_t Offset);
+
+ /// Get a DIE given an exact offset.
+ DWARFDie getDIEForOffset(uint64_t Offset);
+
+ unsigned getMaxVersion() {
+ // Ensure info units have been parsed to discover MaxVersion
+ info_section_units();
+ return MaxVersion;
+ }
+
+ unsigned getMaxDWOVersion() {
+ // Ensure DWO info units have been parsed to discover MaxVersion
+ dwo_info_section_units();
+ return MaxVersion;
+ }
+
+ void setMaxVersionIfGreater(unsigned Version) {
+ if (Version > MaxVersion)
+ MaxVersion = Version;
+ }
+
+ const DWARFUnitIndex &getCUIndex();
+ DWARFGdbIndex &getGdbIndex();
+ const DWARFUnitIndex &getTUIndex();
+
+ /// Get a pointer to the parsed DebugAbbrev object.
+ const DWARFDebugAbbrev *getDebugAbbrev();
+
+ /// Get a pointer to the parsed DebugLoc object.
+ const DWARFDebugLoc *getDebugLoc();
+
+ /// Get a pointer to the parsed dwo abbreviations object.
+ const DWARFDebugAbbrev *getDebugAbbrevDWO();
+
+ /// Get a pointer to the parsed DebugAranges object.
+ const DWARFDebugAranges *getDebugAranges();
+
+ /// Get a pointer to the parsed frame information object.
+ const DWARFDebugFrame *getDebugFrame();
+
+ /// Get a pointer to the parsed eh frame information object.
+ const DWARFDebugFrame *getEHFrame();
+
+ /// Get a pointer to the parsed DebugMacro object.
+ const DWARFDebugMacro *getDebugMacro();
+
+ /// Get a reference to the parsed accelerator table object.
+ const DWARFDebugNames &getDebugNames();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleNames();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleTypes();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleNamespaces();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleObjC();
+
+ /// Get a pointer to a parsed line table corresponding to a compile unit.
+ /// Report any parsing issues as warnings on stderr.
+ const DWARFDebugLine::LineTable *getLineTableForUnit(DWARFUnit *U);
+
+ /// Get a pointer to a parsed line table corresponding to a compile unit.
+ /// Report any recoverable parsing problems using the callback.
+ Expected<const DWARFDebugLine::LineTable *>
+ getLineTableForUnit(DWARFUnit *U,
+ std::function<void(Error)> RecoverableErrorCallback);
+
+ DataExtractor getStringExtractor() const {
+ return DataExtractor(DObj->getStrSection(), false, 0);
+ }
+ DataExtractor getLineStringExtractor() const {
+ return DataExtractor(DObj->getLineStrSection(), false, 0);
+ }
+
+ /// Wraps the returned DIEs for a given address.
+ struct DIEsForAddress {
+ DWARFCompileUnit *CompileUnit = nullptr;
+ DWARFDie FunctionDIE;
+ DWARFDie BlockDIE;
+ explicit operator bool() const { return CompileUnit != nullptr; }
+ };
+
+ /// Get the compilation unit, the function DIE and lexical block DIE for the
+ /// given address where applicable.
+ /// TODO: change input parameter from "uint64_t Address"
+ /// into "SectionedAddress Address"
+ DIEsForAddress getDIEsForAddress(uint64_t Address);
+
+ DILineInfo getLineInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+ DILineInfoTable getLineInfoForAddressRange(
+ object::SectionedAddress Address, uint64_t Size,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+ DIInliningInfo getInliningInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+
+ std::vector<DILocal>
+ getLocalsForAddress(object::SectionedAddress Address) override;
+
+ bool isLittleEndian() const { return DObj->isLittleEndian(); }
+ static bool isSupportedVersion(unsigned version) {
+ return version == 2 || version == 3 || version == 4 || version == 5;
+ }
+
+ std::shared_ptr<DWARFContext> getDWOContext(StringRef AbsolutePath);
+
+ const MCRegisterInfo *getRegisterInfo() const { return RegInfo.get(); }
+
+ /// Function used to handle default error reporting policy. Prints a error
+ /// message and returns Continue, so DWARF context ignores the error.
+ static ErrorPolicy defaultErrorHandler(Error E);
+ static std::unique_ptr<DWARFContext>
+ create(const object::ObjectFile &Obj, const LoadedObjectInfo *L = nullptr,
+ function_ref<ErrorPolicy(Error)> HandleError = defaultErrorHandler,
+ std::string DWPName = "");
+
+ static std::unique_ptr<DWARFContext>
+ create(const StringMap<std::unique_ptr<MemoryBuffer>> &Sections,
+ uint8_t AddrSize, bool isLittleEndian = sys::IsLittleEndianHost);
+
+ /// Loads register info for the architecture of the provided object file.
+ /// Improves readability of dumped DWARF expressions. Requires the caller to
+ /// have initialized the relevant target descriptions.
+ Error loadRegisterInfo(const object::ObjectFile &Obj);
+
+ /// Get address size from CUs.
+ /// TODO: refactor compile_units() to make this const.
+ uint8_t getCUAddrSize();
+
+ /// Dump Error as warning message to stderr.
+ static void dumpWarning(Error Warning);
+
+ Triple::ArchType getArch() const {
+ return getDWARFObj().getFile()->getArch();
+ }
+
+private:
+ /// Return the compile unit which contains instruction with provided
+ /// address.
+ /// TODO: change input parameter from "uint64_t Address"
+ /// into "SectionedAddress Address"
+ DWARFCompileUnit *getCompileUnitForAddress(uint64_t Address);
+ void addLocalsForDie(DWARFCompileUnit *CU, DWARFDie Subprogram, DWARFDie Die,
+ std::vector<DILocal> &Result);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDataExtractor.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDataExtractor.h
new file mode 100644
index 000000000..980724c52
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDataExtractor.h
@@ -0,0 +1,64 @@
+//===- DWARFDataExtractor.h -------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDATAEXTRACTOR_H
+#define LLVM_DEBUGINFO_DWARFDATAEXTRACTOR_H
+
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/Support/DataExtractor.h"
+
+namespace llvm {
+class DWARFObject;
+
+/// A DataExtractor (typically for an in-memory copy of an object-file section)
+/// plus a relocation map for that section, if there is one.
+class DWARFDataExtractor : public DataExtractor {
+ const DWARFObject *Obj = nullptr;
+ const DWARFSection *Section = nullptr;
+
+public:
+ /// Constructor for the normal case of extracting data from a DWARF section.
+ /// The DWARFSection's lifetime must be at least as long as the extractor's.
+ DWARFDataExtractor(const DWARFObject &Obj, const DWARFSection &Section,
+ bool IsLittleEndian, uint8_t AddressSize)
+ : DataExtractor(Section.Data, IsLittleEndian, AddressSize), Obj(&Obj),
+ Section(&Section) {}
+
+ /// Constructor for cases when there are no relocations.
+ DWARFDataExtractor(StringRef Data, bool IsLittleEndian, uint8_t AddressSize)
+ : DataExtractor(Data, IsLittleEndian, AddressSize) {}
+
+ /// Extracts a value and applies a relocation to the result if
+ /// one exists for the given offset.
+ uint64_t getRelocatedValue(uint32_t Size, uint64_t *Off,
+ uint64_t *SectionIndex = nullptr,
+ Error *Err = nullptr) const;
+
+ /// Extracts an address-sized value and applies a relocation to the result if
+ /// one exists for the given offset.
+ uint64_t getRelocatedAddress(uint64_t *Off, uint64_t *SecIx = nullptr) const {
+ return getRelocatedValue(getAddressSize(), Off, SecIx);
+ }
+ uint64_t getRelocatedAddress(Cursor &C, uint64_t *SecIx = nullptr) const {
+ return getRelocatedValue(getAddressSize(), &getOffset(C), SecIx,
+ &getError(C));
+ }
+
+ /// Extracts a DWARF-encoded pointer in \p Offset using \p Encoding.
+ /// There is a DWARF encoding that uses a PC-relative adjustment.
+ /// For these values, \p AbsPosOffset is used to fix them, which should
+ /// reflect the absolute address of this pointer.
+ Optional<uint64_t> getEncodedPointer(uint64_t *Offset, uint8_t Encoding,
+ uint64_t AbsPosOffset = 0) const;
+
+ size_t size() const { return Section == nullptr ? 0 : Section->Data.size(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDATAEXTRACTOR_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h
new file mode 100644
index 000000000..1398e1625
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h
@@ -0,0 +1,87 @@
+//===- DWARFDebugAbbrev.h ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGABBREV_H
+#define LLVM_DEBUGINFO_DWARFDEBUGABBREV_H
+
+#include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+#include <map>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+class DWARFAbbreviationDeclarationSet {
+ uint64_t Offset;
+ /// Code of the first abbreviation, if all abbreviations in the set have
+ /// consecutive codes. UINT32_MAX otherwise.
+ uint32_t FirstAbbrCode;
+ std::vector<DWARFAbbreviationDeclaration> Decls;
+
+ using const_iterator =
+ std::vector<DWARFAbbreviationDeclaration>::const_iterator;
+
+public:
+ DWARFAbbreviationDeclarationSet();
+
+ uint64_t getOffset() const { return Offset; }
+ void dump(raw_ostream &OS) const;
+ bool extract(DataExtractor Data, uint64_t *OffsetPtr);
+
+ const DWARFAbbreviationDeclaration *
+ getAbbreviationDeclaration(uint32_t AbbrCode) const;
+
+ const_iterator begin() const {
+ return Decls.begin();
+ }
+
+ const_iterator end() const {
+ return Decls.end();
+ }
+
+private:
+ void clear();
+};
+
+class DWARFDebugAbbrev {
+ using DWARFAbbreviationDeclarationSetMap =
+ std::map<uint64_t, DWARFAbbreviationDeclarationSet>;
+
+ mutable DWARFAbbreviationDeclarationSetMap AbbrDeclSets;
+ mutable DWARFAbbreviationDeclarationSetMap::const_iterator PrevAbbrOffsetPos;
+ mutable Optional<DataExtractor> Data;
+
+public:
+ DWARFDebugAbbrev();
+
+ const DWARFAbbreviationDeclarationSet *
+ getAbbreviationDeclarationSet(uint64_t CUAbbrOffset) const;
+
+ void dump(raw_ostream &OS) const;
+ void parse() const;
+ void extract(DataExtractor Data);
+
+ DWARFAbbreviationDeclarationSetMap::const_iterator begin() const {
+ parse();
+ return AbbrDeclSets.begin();
+ }
+
+ DWARFAbbreviationDeclarationSetMap::const_iterator end() const {
+ return AbbrDeclSets.end();
+ }
+
+private:
+ void clear();
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGABBREV_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAddr.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAddr.h
new file mode 100644
index 000000000..4539b9c9d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAddr.h
@@ -0,0 +1,97 @@
+//===- DWARFDebugAddr.h -------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGADDR_H
+#define LLVM_DEBUGINFO_DWARFDEBUGADDR_H
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Error.h"
+#include <cstdint>
+#include <map>
+#include <vector>
+
+namespace llvm {
+
+class Error;
+class raw_ostream;
+
+/// A class representing an address table as specified in DWARF v5.
+/// The table consists of a header followed by an array of address values from
+/// .debug_addr section.
+class DWARFDebugAddrTable {
+public:
+ struct Header {
+ /// The total length of the entries for this table, not including the length
+ /// field itself.
+ uint32_t Length = 0;
+ /// The DWARF version number.
+ uint16_t Version = 5;
+ /// The size in bytes of an address on the target architecture. For
+ /// segmented addressing, this is the size of the offset portion of the
+ /// address.
+ uint8_t AddrSize;
+ /// The size in bytes of a segment selector on the target architecture.
+ /// If the target system uses a flat address space, this value is 0.
+ uint8_t SegSize = 0;
+ };
+
+private:
+ dwarf::DwarfFormat Format;
+ uint64_t HeaderOffset;
+ Header HeaderData;
+ uint32_t DataSize = 0;
+ std::vector<uint64_t> Addrs;
+
+public:
+ void clear();
+
+ /// Extract an entire table, including all addresses.
+ Error extract(DWARFDataExtractor Data, uint64_t *OffsetPtr,
+ uint16_t Version, uint8_t AddrSize,
+ std::function<void(Error)> WarnCallback);
+
+ uint64_t getHeaderOffset() const { return HeaderOffset; }
+ uint8_t getAddrSize() const { return HeaderData.AddrSize; }
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts = {}) const;
+
+ /// Return the address based on a given index.
+ Expected<uint64_t> getAddrEntry(uint32_t Index) const;
+
+ /// Return the size of the table header including the length
+ /// but not including the addresses.
+ uint8_t getHeaderSize() const {
+ switch (Format) {
+ case dwarf::DwarfFormat::DWARF32:
+ return 8; // 4 + 2 + 1 + 1
+ case dwarf::DwarfFormat::DWARF64:
+ return 16; // 12 + 2 + 1 + 1
+ }
+ llvm_unreachable("Invalid DWARF format (expected DWARF32 or DWARF64)");
+ }
+
+ /// Returns the length of this table, including the length field, or 0 if the
+ /// length has not been determined (e.g. because the table has not yet been
+ /// parsed, or there was a problem in parsing).
+ uint32_t getLength() const;
+
+ /// Verify that the given length is valid for this table.
+ bool hasValidLength() const { return getLength() != 0; }
+
+ /// Invalidate Length field to stop further processing.
+ void invalidateLength() { HeaderData.Length = 0; }
+
+ /// Returns the length of the array of addresses.
+ uint32_t getDataSize() const;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGADDR_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h
new file mode 100644
index 000000000..ebe4ad6e2
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h
@@ -0,0 +1,75 @@
+//===- DWARFDebugArangeSet.h ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGARANGESET_H
+#define LLVM_DEBUGINFO_DWARFDEBUGARANGESET_H
+
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+class DWARFDebugArangeSet {
+public:
+ struct Header {
+ /// The total length of the entries for that set, not including the length
+ /// field itself.
+ uint32_t Length;
+ /// The offset from the beginning of the .debug_info section of the
+ /// compilation unit entry referenced by the table.
+ uint32_t CuOffset;
+ /// The DWARF version number.
+ uint16_t Version;
+ /// The size in bytes of an address on the target architecture. For segmented
+ /// addressing, this is the size of the offset portion of the address.
+ uint8_t AddrSize;
+ /// The size in bytes of a segment descriptor on the target architecture.
+ /// If the target system uses a flat address space, this value is 0.
+ uint8_t SegSize;
+ };
+
+ struct Descriptor {
+ uint64_t Address;
+ uint64_t Length;
+
+ uint64_t getEndAddress() const { return Address + Length; }
+ void dump(raw_ostream &OS, uint32_t AddressSize) const;
+ };
+
+private:
+ using DescriptorColl = std::vector<Descriptor>;
+ using desc_iterator_range = iterator_range<DescriptorColl::const_iterator>;
+
+ uint64_t Offset;
+ Header HeaderData;
+ DescriptorColl ArangeDescriptors;
+
+public:
+ DWARFDebugArangeSet() { clear(); }
+
+ void clear();
+ bool extract(DataExtractor data, uint64_t *offset_ptr);
+ void dump(raw_ostream &OS) const;
+
+ uint32_t getCompileUnitDIEOffset() const { return HeaderData.CuOffset; }
+
+ const Header &getHeader() const { return HeaderData; }
+
+ desc_iterator_range descriptors() const {
+ return desc_iterator_range(ArangeDescriptors.begin(),
+ ArangeDescriptors.end());
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGARANGESET_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h
new file mode 100644
index 000000000..172f1d2c9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h
@@ -0,0 +1,84 @@
+//===- DWARFDebugAranges.h --------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGARANGES_H
+#define LLVM_DEBUGINFO_DWARFDEBUGARANGES_H
+
+#include "llvm/ADT/DenseSet.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+
+class DWARFContext;
+
+class DWARFDebugAranges {
+public:
+ void generate(DWARFContext *CTX);
+ uint32_t findAddress(uint64_t Address) const;
+
+private:
+ void clear();
+ void extract(DataExtractor DebugArangesData);
+
+ /// Call appendRange multiple times and then call construct.
+ void appendRange(uint64_t CUOffset, uint64_t LowPC, uint64_t HighPC);
+ void construct();
+
+ struct Range {
+ explicit Range(uint64_t LowPC = -1ULL, uint64_t HighPC = -1ULL,
+ uint32_t CUOffset = -1U)
+ : LowPC(LowPC), Length(HighPC - LowPC), CUOffset(CUOffset) {}
+
+ void setHighPC(uint64_t HighPC) {
+ if (HighPC == -1ULL || HighPC <= LowPC)
+ Length = 0;
+ else
+ Length = HighPC - LowPC;
+ }
+
+ uint64_t HighPC() const {
+ if (Length)
+ return LowPC + Length;
+ return -1ULL;
+ }
+
+ bool operator<(const Range &other) const {
+ return LowPC < other.LowPC;
+ }
+
+ uint64_t LowPC; /// Start of address range.
+ uint32_t Length; /// End of address range (not including this address).
+ uint32_t CUOffset; /// Offset of the compile unit or die.
+ };
+
+ struct RangeEndpoint {
+ uint64_t Address;
+ uint64_t CUOffset;
+ bool IsRangeStart;
+
+ RangeEndpoint(uint64_t Address, uint64_t CUOffset, bool IsRangeStart)
+ : Address(Address), CUOffset(CUOffset), IsRangeStart(IsRangeStart) {}
+
+ bool operator<(const RangeEndpoint &Other) const {
+ return Address < Other.Address;
+ }
+ };
+
+ using RangeColl = std::vector<Range>;
+ using RangeCollIterator = RangeColl::const_iterator;
+
+ std::vector<RangeEndpoint> Endpoints;
+ RangeColl Aranges;
+ DenseSet<uint64_t> ParsedCUOffsets;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGARANGES_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugFrame.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugFrame.h
new file mode 100644
index 000000000..c6539df0d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugFrame.h
@@ -0,0 +1,308 @@
+//===- DWARFDebugFrame.h - Parsing of .debug_frame --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFDEBUGFRAME_H
+#define LLVM_DEBUGINFO_DWARF_DWARFDEBUGFRAME_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
+#include "llvm/Support/Error.h"
+#include <memory>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+namespace dwarf {
+
+/// Represent a sequence of Call Frame Information instructions that, when read
+/// in order, construct a table mapping PC to frame state. This can also be
+/// referred to as "CFI rules" in DWARF literature to avoid confusion with
+/// computer programs in the broader sense, and in this context each instruction
+/// would be a rule to establish the mapping. Refer to pg. 172 in the DWARF5
+/// manual, "6.4.1 Structure of Call Frame Information".
+class CFIProgram {
+public:
+ typedef SmallVector<uint64_t, 2> Operands;
+
+ /// An instruction consists of a DWARF CFI opcode and an optional sequence of
+ /// operands. If it refers to an expression, then this expression has its own
+ /// sequence of operations and operands handled separately by DWARFExpression.
+ struct Instruction {
+ Instruction(uint8_t Opcode) : Opcode(Opcode) {}
+
+ uint8_t Opcode;
+ Operands Ops;
+ // Associated DWARF expression in case this instruction refers to one
+ Optional<DWARFExpression> Expression;
+ };
+
+ using InstrList = std::vector<Instruction>;
+ using iterator = InstrList::iterator;
+ using const_iterator = InstrList::const_iterator;
+
+ iterator begin() { return Instructions.begin(); }
+ const_iterator begin() const { return Instructions.begin(); }
+ iterator end() { return Instructions.end(); }
+ const_iterator end() const { return Instructions.end(); }
+
+ unsigned size() const { return (unsigned)Instructions.size(); }
+ bool empty() const { return Instructions.empty(); }
+
+ CFIProgram(uint64_t CodeAlignmentFactor, int64_t DataAlignmentFactor,
+ Triple::ArchType Arch)
+ : CodeAlignmentFactor(CodeAlignmentFactor),
+ DataAlignmentFactor(DataAlignmentFactor),
+ Arch(Arch) {}
+
+ /// Parse and store a sequence of CFI instructions from Data,
+ /// starting at *Offset and ending at EndOffset. *Offset is updated
+ /// to EndOffset upon successful parsing, or indicates the offset
+ /// where a problem occurred in case an error is returned.
+ Error parse(DWARFDataExtractor Data, uint64_t *Offset, uint64_t EndOffset);
+
+ void dump(raw_ostream &OS, const MCRegisterInfo *MRI, bool IsEH,
+ unsigned IndentLevel = 1) const;
+
+private:
+ std::vector<Instruction> Instructions;
+ const uint64_t CodeAlignmentFactor;
+ const int64_t DataAlignmentFactor;
+ Triple::ArchType Arch;
+
+ /// Convenience method to add a new instruction with the given opcode.
+ void addInstruction(uint8_t Opcode) {
+ Instructions.push_back(Instruction(Opcode));
+ }
+
+ /// Add a new single-operand instruction.
+ void addInstruction(uint8_t Opcode, uint64_t Operand1) {
+ Instructions.push_back(Instruction(Opcode));
+ Instructions.back().Ops.push_back(Operand1);
+ }
+
+ /// Add a new instruction that has two operands.
+ void addInstruction(uint8_t Opcode, uint64_t Operand1, uint64_t Operand2) {
+ Instructions.push_back(Instruction(Opcode));
+ Instructions.back().Ops.push_back(Operand1);
+ Instructions.back().Ops.push_back(Operand2);
+ }
+
+ /// Types of operands to CFI instructions
+ /// In DWARF, this type is implicitly tied to a CFI instruction opcode and
+ /// thus this type doesn't need to be explictly written to the file (this is
+ /// not a DWARF encoding). The relationship of instrs to operand types can
+ /// be obtained from getOperandTypes() and is only used to simplify
+ /// instruction printing.
+ enum OperandType {
+ OT_Unset,
+ OT_None,
+ OT_Address,
+ OT_Offset,
+ OT_FactoredCodeOffset,
+ OT_SignedFactDataOffset,
+ OT_UnsignedFactDataOffset,
+ OT_Register,
+ OT_Expression
+ };
+
+ /// Retrieve the array describing the types of operands according to the enum
+ /// above. This is indexed by opcode.
+ static ArrayRef<OperandType[2]> getOperandTypes();
+
+ /// Print \p Opcode's operand number \p OperandIdx which has value \p Operand.
+ void printOperand(raw_ostream &OS, const MCRegisterInfo *MRI, bool IsEH,
+ const Instruction &Instr, unsigned OperandIdx,
+ uint64_t Operand) const;
+};
+
+/// An entry in either debug_frame or eh_frame. This entry can be a CIE or an
+/// FDE.
+class FrameEntry {
+public:
+ enum FrameKind { FK_CIE, FK_FDE };
+
+ FrameEntry(FrameKind K, uint64_t Offset, uint64_t Length, uint64_t CodeAlign,
+ int64_t DataAlign, Triple::ArchType Arch)
+ : Kind(K), Offset(Offset), Length(Length),
+ CFIs(CodeAlign, DataAlign, Arch) {}
+
+ virtual ~FrameEntry() {}
+
+ FrameKind getKind() const { return Kind; }
+ uint64_t getOffset() const { return Offset; }
+ uint64_t getLength() const { return Length; }
+ const CFIProgram &cfis() const { return CFIs; }
+ CFIProgram &cfis() { return CFIs; }
+
+ /// Dump the instructions in this CFI fragment
+ virtual void dump(raw_ostream &OS, const MCRegisterInfo *MRI,
+ bool IsEH) const = 0;
+
+protected:
+ const FrameKind Kind;
+
+ /// Offset of this entry in the section.
+ const uint64_t Offset;
+
+ /// Entry length as specified in DWARF.
+ const uint64_t Length;
+
+ CFIProgram CFIs;
+};
+
+/// DWARF Common Information Entry (CIE)
+class CIE : public FrameEntry {
+public:
+ // CIEs (and FDEs) are simply container classes, so the only sensible way to
+ // create them is by providing the full parsed contents in the constructor.
+ CIE(uint64_t Offset, uint64_t Length, uint8_t Version,
+ SmallString<8> Augmentation, uint8_t AddressSize,
+ uint8_t SegmentDescriptorSize, uint64_t CodeAlignmentFactor,
+ int64_t DataAlignmentFactor, uint64_t ReturnAddressRegister,
+ SmallString<8> AugmentationData, uint32_t FDEPointerEncoding,
+ uint32_t LSDAPointerEncoding, Optional<uint64_t> Personality,
+ Optional<uint32_t> PersonalityEnc, Triple::ArchType Arch)
+ : FrameEntry(FK_CIE, Offset, Length, CodeAlignmentFactor,
+ DataAlignmentFactor, Arch),
+ Version(Version), Augmentation(std::move(Augmentation)),
+ AddressSize(AddressSize), SegmentDescriptorSize(SegmentDescriptorSize),
+ CodeAlignmentFactor(CodeAlignmentFactor),
+ DataAlignmentFactor(DataAlignmentFactor),
+ ReturnAddressRegister(ReturnAddressRegister),
+ AugmentationData(std::move(AugmentationData)),
+ FDEPointerEncoding(FDEPointerEncoding),
+ LSDAPointerEncoding(LSDAPointerEncoding), Personality(Personality),
+ PersonalityEnc(PersonalityEnc) {}
+
+ static bool classof(const FrameEntry *FE) { return FE->getKind() == FK_CIE; }
+
+ StringRef getAugmentationString() const { return Augmentation; }
+ uint64_t getCodeAlignmentFactor() const { return CodeAlignmentFactor; }
+ int64_t getDataAlignmentFactor() const { return DataAlignmentFactor; }
+ uint8_t getVersion() const { return Version; }
+ uint64_t getReturnAddressRegister() const { return ReturnAddressRegister; }
+ Optional<uint64_t> getPersonalityAddress() const { return Personality; }
+ Optional<uint32_t> getPersonalityEncoding() const { return PersonalityEnc; }
+
+ uint32_t getFDEPointerEncoding() const { return FDEPointerEncoding; }
+
+ uint32_t getLSDAPointerEncoding() const { return LSDAPointerEncoding; }
+
+ void dump(raw_ostream &OS, const MCRegisterInfo *MRI,
+ bool IsEH) const override;
+
+private:
+ /// The following fields are defined in section 6.4.1 of the DWARF standard v4
+ const uint8_t Version;
+ const SmallString<8> Augmentation;
+ const uint8_t AddressSize;
+ const uint8_t SegmentDescriptorSize;
+ const uint64_t CodeAlignmentFactor;
+ const int64_t DataAlignmentFactor;
+ const uint64_t ReturnAddressRegister;
+
+ // The following are used when the CIE represents an EH frame entry.
+ const SmallString<8> AugmentationData;
+ const uint32_t FDEPointerEncoding;
+ const uint32_t LSDAPointerEncoding;
+ const Optional<uint64_t> Personality;
+ const Optional<uint32_t> PersonalityEnc;
+};
+
+/// DWARF Frame Description Entry (FDE)
+class FDE : public FrameEntry {
+public:
+ // Each FDE has a CIE it's "linked to". Our FDE contains is constructed with
+ // an offset to the CIE (provided by parsing the FDE header). The CIE itself
+ // is obtained lazily once it's actually required.
+ FDE(uint64_t Offset, uint64_t Length, int64_t LinkedCIEOffset,
+ uint64_t InitialLocation, uint64_t AddressRange, CIE *Cie,
+ Optional<uint64_t> LSDAAddress, Triple::ArchType Arch)
+ : FrameEntry(FK_FDE, Offset, Length,
+ Cie ? Cie->getCodeAlignmentFactor() : 0,
+ Cie ? Cie->getDataAlignmentFactor() : 0,
+ Arch),
+ LinkedCIEOffset(LinkedCIEOffset), InitialLocation(InitialLocation),
+ AddressRange(AddressRange), LinkedCIE(Cie), LSDAAddress(LSDAAddress) {}
+
+ ~FDE() override = default;
+
+ const CIE *getLinkedCIE() const { return LinkedCIE; }
+ uint64_t getInitialLocation() const { return InitialLocation; }
+ uint64_t getAddressRange() const { return AddressRange; }
+ Optional<uint64_t> getLSDAAddress() const { return LSDAAddress; }
+
+ void dump(raw_ostream &OS, const MCRegisterInfo *MRI,
+ bool IsEH) const override;
+
+ static bool classof(const FrameEntry *FE) { return FE->getKind() == FK_FDE; }
+
+private:
+ /// The following fields are defined in section 6.4.1 of the DWARF standard v3
+ const uint64_t LinkedCIEOffset;
+ const uint64_t InitialLocation;
+ const uint64_t AddressRange;
+ const CIE *LinkedCIE;
+ const Optional<uint64_t> LSDAAddress;
+};
+
+} // end namespace dwarf
+
+/// A parsed .debug_frame or .eh_frame section
+class DWARFDebugFrame {
+ const Triple::ArchType Arch;
+ // True if this is parsing an eh_frame section.
+ const bool IsEH;
+ // Not zero for sane pointer values coming out of eh_frame
+ const uint64_t EHFrameAddress;
+
+ std::vector<std::unique_ptr<dwarf::FrameEntry>> Entries;
+ using iterator = pointee_iterator<decltype(Entries)::const_iterator>;
+
+ /// Return the entry at the given offset or nullptr.
+ dwarf::FrameEntry *getEntryAtOffset(uint64_t Offset) const;
+
+public:
+ // If IsEH is true, assume it is a .eh_frame section. Otherwise,
+ // it is a .debug_frame section. EHFrameAddress should be different
+ // than zero for correct parsing of .eh_frame addresses when they
+ // use a PC-relative encoding.
+ DWARFDebugFrame(Triple::ArchType Arch,
+ bool IsEH = false, uint64_t EHFrameAddress = 0);
+ ~DWARFDebugFrame();
+
+ /// Dump the section data into the given stream.
+ void dump(raw_ostream &OS, const MCRegisterInfo *MRI,
+ Optional<uint64_t> Offset) const;
+
+ /// Parse the section from raw data. \p Data is assumed to contain the whole
+ /// frame section contents to be parsed.
+ void parse(DWARFDataExtractor Data);
+
+ /// Return whether the section has any entries.
+ bool empty() const { return Entries.empty(); }
+
+ /// DWARF Frame entries accessors
+ iterator begin() const { return Entries.begin(); }
+ iterator end() const { return Entries.end(); }
+ iterator_range<iterator> entries() const {
+ return iterator_range<iterator>(Entries.begin(), Entries.end());
+ }
+
+ uint64_t getEHFrameAddress() const { return EHFrameAddress; }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFDEBUGFRAME_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h
new file mode 100644
index 000000000..ded960337
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h
@@ -0,0 +1,62 @@
+//===- DWARFDebugInfoEntry.h ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGINFOENTRY_H
+#define LLVM_DEBUGINFO_DWARFDEBUGINFOENTRY_H
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include <cstdint>
+
+namespace llvm {
+
+class DataExtractor;
+class DWARFUnit;
+
+/// DWARFDebugInfoEntry - A DIE with only the minimum required data.
+class DWARFDebugInfoEntry {
+ /// Offset within the .debug_info of the start of this entry.
+ uint64_t Offset = 0;
+
+ /// The integer depth of this DIE within the compile unit DIEs where the
+ /// compile/type unit DIE has a depth of zero.
+ uint32_t Depth = 0;
+
+ const DWARFAbbreviationDeclaration *AbbrevDecl = nullptr;
+
+public:
+ DWARFDebugInfoEntry() = default;
+
+ /// Extracts a debug info entry, which is a child of a given unit,
+ /// starting at a given offset. If DIE can't be extracted, returns false and
+ /// doesn't change OffsetPtr.
+ bool extractFast(const DWARFUnit &U, uint64_t *OffsetPtr);
+
+ /// High performance extraction should use this call.
+ bool extractFast(const DWARFUnit &U, uint64_t *OffsetPtr,
+ const DWARFDataExtractor &DebugInfoData, uint64_t UEndOffset,
+ uint32_t Depth);
+
+ uint64_t getOffset() const { return Offset; }
+ uint32_t getDepth() const { return Depth; }
+
+ dwarf::Tag getTag() const {
+ return AbbrevDecl ? AbbrevDecl->getTag() : dwarf::DW_TAG_null;
+ }
+
+ bool hasChildren() const { return AbbrevDecl && AbbrevDecl->hasChildren(); }
+
+ const DWARFAbbreviationDeclaration *getAbbreviationDeclarationPtr() const {
+ return AbbrevDecl;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGINFOENTRY_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLine.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLine.h
new file mode 100644
index 000000000..c2be8304a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLine.h
@@ -0,0 +1,392 @@
+//===- DWARFDebugLine.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGLINE_H
+#define LLVM_DEBUGINFO_DWARFDEBUGLINE_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/DebugInfo/DWARF/DWARFTypeUnit.h"
+#include "llvm/Support/MD5.h"
+#include "llvm/Support/Path.h"
+#include <cstdint>
+#include <map>
+#include <string>
+#include <vector>
+
+namespace llvm {
+
+class DWARFUnit;
+class raw_ostream;
+
+class DWARFDebugLine {
+public:
+ struct FileNameEntry {
+ FileNameEntry() = default;
+
+ DWARFFormValue Name;
+ uint64_t DirIdx = 0;
+ uint64_t ModTime = 0;
+ uint64_t Length = 0;
+ MD5::MD5Result Checksum;
+ DWARFFormValue Source;
+ };
+
+ /// Tracks which optional content types are present in a DWARF file name
+ /// entry format.
+ struct ContentTypeTracker {
+ ContentTypeTracker() = default;
+
+ /// Whether filename entries provide a modification timestamp.
+ bool HasModTime = false;
+ /// Whether filename entries provide a file size.
+ bool HasLength = false;
+ /// For v5, whether filename entries provide an MD5 checksum.
+ bool HasMD5 = false;
+ /// For v5, whether filename entries provide source text.
+ bool HasSource = false;
+
+ /// Update tracked content types with \p ContentType.
+ void trackContentType(dwarf::LineNumberEntryFormat ContentType);
+ };
+
+ struct Prologue {
+ Prologue();
+
+ /// The size in bytes of the statement information for this compilation unit
+ /// (not including the total_length field itself).
+ uint64_t TotalLength;
+ /// Version, address size (starting in v5), and DWARF32/64 format; these
+ /// parameters affect interpretation of forms (used in the directory and
+ /// file tables starting with v5).
+ dwarf::FormParams FormParams;
+ /// The number of bytes following the prologue_length field to the beginning
+ /// of the first byte of the statement program itself.
+ uint64_t PrologueLength;
+ /// In v5, size in bytes of a segment selector.
+ uint8_t SegSelectorSize;
+ /// The size in bytes of the smallest target machine instruction. Statement
+ /// program opcodes that alter the address register first multiply their
+ /// operands by this value.
+ uint8_t MinInstLength;
+ /// The maximum number of individual operations that may be encoded in an
+ /// instruction.
+ uint8_t MaxOpsPerInst;
+ /// The initial value of theis_stmtregister.
+ uint8_t DefaultIsStmt;
+ /// This parameter affects the meaning of the special opcodes. See below.
+ int8_t LineBase;
+ /// This parameter affects the meaning of the special opcodes. See below.
+ uint8_t LineRange;
+ /// The number assigned to the first special opcode.
+ uint8_t OpcodeBase;
+ /// This tracks which optional file format content types are present.
+ ContentTypeTracker ContentTypes;
+ std::vector<uint8_t> StandardOpcodeLengths;
+ std::vector<DWARFFormValue> IncludeDirectories;
+ std::vector<FileNameEntry> FileNames;
+
+ const dwarf::FormParams getFormParams() const { return FormParams; }
+ uint16_t getVersion() const { return FormParams.Version; }
+ uint8_t getAddressSize() const { return FormParams.AddrSize; }
+ bool isDWARF64() const { return FormParams.Format == dwarf::DWARF64; }
+
+ uint32_t sizeofTotalLength() const { return isDWARF64() ? 12 : 4; }
+
+ uint32_t sizeofPrologueLength() const { return isDWARF64() ? 8 : 4; }
+
+ bool totalLengthIsValid() const;
+
+ /// Length of the prologue in bytes.
+ uint32_t getLength() const {
+ return PrologueLength + sizeofTotalLength() + sizeof(getVersion()) +
+ sizeofPrologueLength();
+ }
+
+ /// Length of the line table data in bytes (not including the prologue).
+ uint32_t getStatementTableLength() const {
+ return TotalLength + sizeofTotalLength() - getLength();
+ }
+
+ int32_t getMaxLineIncrementForSpecialOpcode() const {
+ return LineBase + (int8_t)LineRange - 1;
+ }
+
+ /// Get DWARF-version aware access to the file name entry at the provided
+ /// index.
+ const llvm::DWARFDebugLine::FileNameEntry &
+ getFileNameEntry(uint64_t Index) const;
+
+ bool hasFileAtIndex(uint64_t FileIndex) const;
+
+ bool
+ getFileNameByIndex(uint64_t FileIndex, StringRef CompDir,
+ DILineInfoSpecifier::FileLineInfoKind Kind,
+ std::string &Result,
+ sys::path::Style Style = sys::path::Style::native) const;
+
+ void clear();
+ void dump(raw_ostream &OS, DIDumpOptions DumpOptions) const;
+ Error parse(const DWARFDataExtractor &DebugLineData, uint64_t *OffsetPtr,
+ const DWARFContext &Ctx, const DWARFUnit *U = nullptr);
+ };
+
+ /// Standard .debug_line state machine structure.
+ struct Row {
+ explicit Row(bool DefaultIsStmt = false);
+
+ /// Called after a row is appended to the matrix.
+ void postAppend();
+ void reset(bool DefaultIsStmt);
+ void dump(raw_ostream &OS) const;
+
+ static void dumpTableHeader(raw_ostream &OS);
+
+ static bool orderByAddress(const Row &LHS, const Row &RHS) {
+ return std::tie(LHS.Address.SectionIndex, LHS.Address.Address) <
+ std::tie(RHS.Address.SectionIndex, RHS.Address.Address);
+ }
+
+ /// The program-counter value corresponding to a machine instruction
+ /// generated by the compiler and section index pointing to the section
+ /// containg this PC. If relocation information is present then section
+ /// index is the index of the section which contains above address.
+ /// Otherwise this is object::SectionedAddress::Undef value.
+ object::SectionedAddress Address;
+ /// An unsigned integer indicating a source line number. Lines are numbered
+ /// beginning at 1. The compiler may emit the value 0 in cases where an
+ /// instruction cannot be attributed to any source line.
+ uint32_t Line;
+ /// An unsigned integer indicating a column number within a source line.
+ /// Columns are numbered beginning at 1. The value 0 is reserved to indicate
+ /// that a statement begins at the 'left edge' of the line.
+ uint16_t Column;
+ /// An unsigned integer indicating the identity of the source file
+ /// corresponding to a machine instruction.
+ uint16_t File;
+ /// An unsigned integer representing the DWARF path discriminator value
+ /// for this location.
+ uint32_t Discriminator;
+ /// An unsigned integer whose value encodes the applicable instruction set
+ /// architecture for the current instruction.
+ uint8_t Isa;
+ /// A boolean indicating that the current instruction is the beginning of a
+ /// statement.
+ uint8_t IsStmt : 1,
+ /// A boolean indicating that the current instruction is the
+ /// beginning of a basic block.
+ BasicBlock : 1,
+ /// A boolean indicating that the current address is that of the
+ /// first byte after the end of a sequence of target machine
+ /// instructions.
+ EndSequence : 1,
+ /// A boolean indicating that the current address is one (of possibly
+ /// many) where execution should be suspended for an entry breakpoint
+ /// of a function.
+ PrologueEnd : 1,
+ /// A boolean indicating that the current address is one (of possibly
+ /// many) where execution should be suspended for an exit breakpoint
+ /// of a function.
+ EpilogueBegin : 1;
+ };
+
+ /// Represents a series of contiguous machine instructions. Line table for
+ /// each compilation unit may consist of multiple sequences, which are not
+ /// guaranteed to be in the order of ascending instruction address.
+ struct Sequence {
+ Sequence();
+
+ /// Sequence describes instructions at address range [LowPC, HighPC)
+ /// and is described by line table rows [FirstRowIndex, LastRowIndex).
+ uint64_t LowPC;
+ uint64_t HighPC;
+ /// If relocation information is present then this is the index of the
+ /// section which contains above addresses. Otherwise this is
+ /// object::SectionedAddress::Undef value.
+ uint64_t SectionIndex;
+ unsigned FirstRowIndex;
+ unsigned LastRowIndex;
+ bool Empty;
+
+ void reset();
+
+ static bool orderByHighPC(const Sequence &LHS, const Sequence &RHS) {
+ return std::tie(LHS.SectionIndex, LHS.HighPC) <
+ std::tie(RHS.SectionIndex, RHS.HighPC);
+ }
+
+ bool isValid() const {
+ return !Empty && (LowPC < HighPC) && (FirstRowIndex < LastRowIndex);
+ }
+
+ bool containsPC(object::SectionedAddress PC) const {
+ return SectionIndex == PC.SectionIndex &&
+ (LowPC <= PC.Address && PC.Address < HighPC);
+ }
+ };
+
+ struct LineTable {
+ LineTable();
+
+ /// Represents an invalid row
+ const uint32_t UnknownRowIndex = UINT32_MAX;
+
+ void appendRow(const DWARFDebugLine::Row &R) { Rows.push_back(R); }
+
+ void appendSequence(const DWARFDebugLine::Sequence &S) {
+ Sequences.push_back(S);
+ }
+
+ /// Returns the index of the row with file/line info for a given address,
+ /// or UnknownRowIndex if there is no such row.
+ uint32_t lookupAddress(object::SectionedAddress Address) const;
+
+ bool lookupAddressRange(object::SectionedAddress Address, uint64_t Size,
+ std::vector<uint32_t> &Result) const;
+
+ bool hasFileAtIndex(uint64_t FileIndex) const {
+ return Prologue.hasFileAtIndex(FileIndex);
+ }
+
+ /// Extracts filename by its index in filename table in prologue.
+ /// In Dwarf 4, the files are 1-indexed and the current compilation file
+ /// name is not represented in the list. In DWARF v5, the files are
+ /// 0-indexed and the primary source file has the index 0.
+ /// Returns true on success.
+ bool getFileNameByIndex(uint64_t FileIndex, StringRef CompDir,
+ DILineInfoSpecifier::FileLineInfoKind Kind,
+ std::string &Result) const {
+ return Prologue.getFileNameByIndex(FileIndex, CompDir, Kind, Result);
+ }
+
+ /// Fills the Result argument with the file and line information
+ /// corresponding to Address. Returns true on success.
+ bool getFileLineInfoForAddress(object::SectionedAddress Address,
+ const char *CompDir,
+ DILineInfoSpecifier::FileLineInfoKind Kind,
+ DILineInfo &Result) const;
+
+ void dump(raw_ostream &OS, DIDumpOptions DumpOptions) const;
+ void clear();
+
+ /// Parse prologue and all rows.
+ Error parse(
+ DWARFDataExtractor &DebugLineData, uint64_t *OffsetPtr,
+ const DWARFContext &Ctx, const DWARFUnit *U,
+ std::function<void(Error)> RecoverableErrorCallback,
+ raw_ostream *OS = nullptr);
+
+ using RowVector = std::vector<Row>;
+ using RowIter = RowVector::const_iterator;
+ using SequenceVector = std::vector<Sequence>;
+ using SequenceIter = SequenceVector::const_iterator;
+
+ struct Prologue Prologue;
+ RowVector Rows;
+ SequenceVector Sequences;
+
+ private:
+ uint32_t findRowInSeq(const DWARFDebugLine::Sequence &Seq,
+ object::SectionedAddress Address) const;
+ Optional<StringRef>
+ getSourceByIndex(uint64_t FileIndex,
+ DILineInfoSpecifier::FileLineInfoKind Kind) const;
+
+ uint32_t lookupAddressImpl(object::SectionedAddress Address) const;
+
+ bool lookupAddressRangeImpl(object::SectionedAddress Address, uint64_t Size,
+ std::vector<uint32_t> &Result) const;
+ };
+
+ const LineTable *getLineTable(uint64_t Offset) const;
+ Expected<const LineTable *> getOrParseLineTable(
+ DWARFDataExtractor &DebugLineData, uint64_t Offset,
+ const DWARFContext &Ctx, const DWARFUnit *U,
+ std::function<void(Error)> RecoverableErrorCallback);
+
+ /// Helper to allow for parsing of an entire .debug_line section in sequence.
+ class SectionParser {
+ public:
+ using cu_range = DWARFUnitVector::iterator_range;
+ using tu_range = DWARFUnitVector::iterator_range;
+ using LineToUnitMap = std::map<uint64_t, DWARFUnit *>;
+
+ SectionParser(DWARFDataExtractor &Data, const DWARFContext &C, cu_range CUs,
+ tu_range TUs);
+
+ /// Get the next line table from the section. Report any issues via the
+ /// callbacks.
+ ///
+ /// \param RecoverableErrorCallback - any issues that don't prevent further
+ /// parsing of the table will be reported through this callback.
+ /// \param UnrecoverableErrorCallback - any issues that prevent further
+ /// parsing of the table will be reported through this callback.
+ /// \param OS - if not null, the parser will print information about the
+ /// table as it parses it.
+ LineTable
+ parseNext(
+ function_ref<void(Error)> RecoverableErrorCallback,
+ function_ref<void(Error)> UnrecoverableErrorCallback,
+ raw_ostream *OS = nullptr);
+
+ /// Skip the current line table and go to the following line table (if
+ /// present) immediately.
+ ///
+ /// \param ErrorCallback - report any prologue parsing issues via this
+ /// callback.
+ void skip(function_ref<void(Error)> ErrorCallback);
+
+ /// Indicates if the parser has parsed as much as possible.
+ ///
+ /// \note Certain problems with the line table structure might mean that
+ /// parsing stops before the end of the section is reached.
+ bool done() const { return Done; }
+
+ /// Get the offset the parser has reached.
+ uint64_t getOffset() const { return Offset; }
+
+ private:
+ DWARFUnit *prepareToParse(uint64_t Offset);
+ void moveToNextTable(uint64_t OldOffset, const Prologue &P);
+
+ LineToUnitMap LineToUnit;
+
+ DWARFDataExtractor &DebugLineData;
+ const DWARFContext &Context;
+ uint64_t Offset = 0;
+ bool Done = false;
+ };
+
+private:
+ struct ParsingState {
+ ParsingState(struct LineTable *LT);
+
+ void resetRowAndSequence();
+ void appendRowToMatrix();
+
+ /// Line table we're currently parsing.
+ struct LineTable *LineTable;
+ struct Row Row;
+ struct Sequence Sequence;
+ };
+
+ using LineTableMapTy = std::map<uint64_t, LineTable>;
+ using LineTableIter = LineTableMapTy::iterator;
+ using LineTableConstIter = LineTableMapTy::const_iterator;
+
+ LineTableMapTy LineTableMap;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGLINE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLoc.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLoc.h
new file mode 100644
index 000000000..25b771317
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugLoc.h
@@ -0,0 +1,121 @@
+//===- DWARFDebugLoc.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFDEBUGLOC_H
+#define LLVM_DEBUGINFO_DWARF_DWARFDEBUGLOC_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include <cstdint>
+
+namespace llvm {
+class DWARFUnit;
+class MCRegisterInfo;
+class raw_ostream;
+
+class DWARFDebugLoc {
+public:
+ /// A single location within a location list.
+ struct Entry {
+ /// The beginning address of the instruction range.
+ uint64_t Begin;
+ /// The ending address of the instruction range.
+ uint64_t End;
+ /// The location of the variable within the specified range.
+ SmallVector<uint8_t, 4> Loc;
+ };
+
+ /// A list of locations that contain one variable.
+ struct LocationList {
+ /// The beginning offset where this location list is stored in the debug_loc
+ /// section.
+ uint64_t Offset;
+ /// All the locations in which the variable is stored.
+ SmallVector<Entry, 2> Entries;
+ /// Dump this list on OS.
+ void dump(raw_ostream &OS, uint64_t BaseAddress, bool IsLittleEndian,
+ unsigned AddressSize, const MCRegisterInfo *MRI, DWARFUnit *U,
+ DIDumpOptions DumpOpts,
+ unsigned Indent) const;
+ };
+
+private:
+ using LocationLists = SmallVector<LocationList, 4>;
+
+ /// A list of all the variables in the debug_loc section, each one describing
+ /// the locations in which the variable is stored.
+ LocationLists Locations;
+
+ unsigned AddressSize;
+
+ bool IsLittleEndian;
+
+public:
+ /// Print the location lists found within the debug_loc section.
+ void dump(raw_ostream &OS, const MCRegisterInfo *RegInfo, DIDumpOptions DumpOpts,
+ Optional<uint64_t> Offset) const;
+
+ /// Parse the debug_loc section accessible via the 'data' parameter using the
+ /// address size also given in 'data' to interpret the address ranges.
+ void parse(const DWARFDataExtractor &data);
+
+ /// Return the location list at the given offset or nullptr.
+ LocationList const *getLocationListAtOffset(uint64_t Offset) const;
+
+ Expected<LocationList>
+ parseOneLocationList(const DWARFDataExtractor &Data, uint64_t *Offset);
+};
+
+class DWARFDebugLoclists {
+public:
+ // Unconstructible.
+ DWARFDebugLoclists() = delete;
+
+ struct Entry {
+ uint8_t Kind;
+ uint64_t Offset;
+ uint64_t Value0;
+ uint64_t Value1;
+ SmallVector<uint8_t, 4> Loc;
+ void dump(raw_ostream &OS, uint64_t &BaseAddr, bool IsLittleEndian,
+ unsigned AddressSize, const MCRegisterInfo *MRI, DWARFUnit *U,
+ DIDumpOptions DumpOpts, unsigned Indent, size_t MaxEncodingStringLength) const;
+ };
+
+ /// Call the user-provided callback for each entry (including the end-of-list
+ /// entry) in the location list starting at \p Offset. The callback can return
+ /// false to terminate the iteration early. Returns an error if it was unable
+ /// to parse the entire location list correctly. Upon successful termination
+ /// \p Offset will be updated point past the end of the list.
+ static Error visitLocationList(const DWARFDataExtractor &Data,
+ uint64_t *Offset, uint16_t Version,
+ llvm::function_ref<bool(const Entry &)> F);
+
+ /// Dump the location list at the given \p Offset. The function returns true
+ /// iff it has successfully reched the end of the list. This means that one
+ /// can attempt to parse another list after the current one (\p Offset will be
+ /// updated to point past the end of the current list).
+ static bool dumpLocationList(const DWARFDataExtractor &Data, uint64_t *Offset,
+ uint16_t Version, raw_ostream &OS,
+ uint64_t BaseAddr, const MCRegisterInfo *MRI,
+ DWARFUnit *U, DIDumpOptions DumpOpts,
+ unsigned Indent);
+
+ /// Dump all location lists within the given range.
+ static void dumpRange(const DWARFDataExtractor &Data, uint64_t StartOffset,
+ uint64_t Size, uint16_t Version, raw_ostream &OS,
+ uint64_t BaseAddr, const MCRegisterInfo *MRI,
+ DIDumpOptions DumpOpts);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFDEBUGLOC_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h
new file mode 100644
index 000000000..7880bcdf6
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h
@@ -0,0 +1,62 @@
+//===- DWARFDebugMacro.h ----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFDEBUGMACRO_H
+#define LLVM_DEBUGINFO_DWARF_DWARFDEBUGMACRO_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+
+namespace llvm {
+
+class raw_ostream;
+
+class DWARFDebugMacro {
+ /// A single macro entry within a macro list.
+ struct Entry {
+ /// The type of the macro entry.
+ uint32_t Type;
+ union {
+ /// The source line where the macro is defined.
+ uint64_t Line;
+ /// Vendor extension constant value.
+ uint64_t ExtConstant;
+ };
+
+ union {
+ /// The string (name, value) of the macro entry.
+ const char *MacroStr;
+ // An unsigned integer indicating the identity of the source file.
+ uint64_t File;
+ /// Vendor extension string.
+ const char *ExtStr;
+ };
+ };
+
+ using MacroList = SmallVector<Entry, 4>;
+
+ /// A list of all the macro entries in the debug_macinfo section.
+ std::vector<MacroList> MacroLists;
+
+public:
+ DWARFDebugMacro() = default;
+
+ /// Print the macro list found within the debug_macinfo section.
+ void dump(raw_ostream &OS) const;
+
+ /// Parse the debug_macinfo section accessible via the 'data' parameter.
+ void parse(DataExtractor data);
+
+ /// Return whether the section has any entries.
+ bool empty() const { return MacroLists.empty(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFDEBUGMACRO_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugPubTable.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugPubTable.h
new file mode 100644
index 000000000..ae57306b9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugPubTable.h
@@ -0,0 +1,80 @@
+//===- DWARFDebugPubTable.h -------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFDEBUGPUBTABLE_H
+#define LLVM_DEBUGINFO_DWARF_DWARFDEBUGPUBTABLE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFObject.h"
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+/// Represents structure for holding and parsing .debug_pub* tables.
+class DWARFDebugPubTable {
+public:
+ struct Entry {
+ /// Section offset from the beginning of the compilation unit.
+ uint64_t SecOffset;
+
+ /// An entry of the various gnu_pub* debug sections.
+ dwarf::PubIndexEntryDescriptor Descriptor;
+
+ /// The name of the object as given by the DW_AT_name attribute of the
+ /// referenced DIE.
+ StringRef Name;
+ };
+
+ /// Each table consists of sets of variable length entries. Each set describes
+ /// the names of global objects and functions, or global types, respectively,
+ /// whose definitions are represented by debugging information entries owned
+ /// by a single compilation unit.
+ struct Set {
+ /// The total length of the entries for that set, not including the length
+ /// field itself.
+ uint32_t Length;
+
+ /// This number is specific to the name lookup table and is independent of
+ /// the DWARF version number.
+ uint16_t Version;
+
+ /// The offset from the beginning of the .debug_info section of the
+ /// compilation unit header referenced by the set.
+ uint64_t Offset;
+
+ /// The size in bytes of the contents of the .debug_info section generated
+ /// to represent that compilation unit.
+ uint32_t Size;
+
+ std::vector<Entry> Entries;
+ };
+
+private:
+ std::vector<Set> Sets;
+
+ /// gnu styled tables contains additional information.
+ /// This flag determines whether or not section we parse is debug_gnu* table.
+ bool GnuStyle;
+
+public:
+ DWARFDebugPubTable(const DWARFObject &Obj, const DWARFSection &Sec,
+ bool LittleEndian, bool GnuStyle);
+
+ void dump(raw_ostream &OS) const;
+
+ ArrayRef<Set> getData() { return Sets; }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFDEBUGPUBTABLE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h
new file mode 100644
index 000000000..2f72c642a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h
@@ -0,0 +1,83 @@
+//===- DWARFDebugRangeList.h ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFDEBUGRANGELIST_H
+#define LLVM_DEBUGINFO_DWARF_DWARFDEBUGRANGELIST_H
+
+#include "llvm/DebugInfo/DWARF/DWARFAddressRange.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include <cassert>
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+
+class DWARFDebugRangeList {
+public:
+ struct RangeListEntry {
+ /// A beginning address offset. This address offset has the size of an
+ /// address and is relative to the applicable base address of the
+ /// compilation unit referencing this range list. It marks the beginning
+ /// of an address range.
+ uint64_t StartAddress;
+ /// An ending address offset. This address offset again has the size of
+ /// an address and is relative to the applicable base address of the
+ /// compilation unit referencing this range list. It marks the first
+ /// address past the end of the address range. The ending address must
+ /// be greater than or equal to the beginning address.
+ uint64_t EndAddress;
+ /// A section index this range belongs to.
+ uint64_t SectionIndex;
+
+ /// The end of any given range list is marked by an end of list entry,
+ /// which consists of a 0 for the beginning address offset
+ /// and a 0 for the ending address offset.
+ bool isEndOfListEntry() const {
+ return (StartAddress == 0) && (EndAddress == 0);
+ }
+
+ /// A base address selection entry consists of:
+ /// 1. The value of the largest representable address offset
+ /// (for example, 0xffffffff when the size of an address is 32 bits).
+ /// 2. An address, which defines the appropriate base address for
+ /// use in interpreting the beginning and ending address offsets of
+ /// subsequent entries of the location list.
+ bool isBaseAddressSelectionEntry(uint8_t AddressSize) const {
+ assert(AddressSize == 4 || AddressSize == 8);
+ if (AddressSize == 4)
+ return StartAddress == -1U;
+ return StartAddress == -1ULL;
+ }
+ };
+
+private:
+ /// Offset in .debug_ranges section.
+ uint64_t Offset;
+ uint8_t AddressSize;
+ std::vector<RangeListEntry> Entries;
+
+public:
+ DWARFDebugRangeList() { clear(); }
+
+ void clear();
+ void dump(raw_ostream &OS) const;
+ Error extract(const DWARFDataExtractor &data, uint64_t *offset_ptr);
+ const std::vector<RangeListEntry> &getEntries() { return Entries; }
+
+ /// getAbsoluteRanges - Returns absolute address ranges defined by this range
+ /// list. Has to be passed base address of the compile unit referencing this
+ /// range list.
+ DWARFAddressRangesVector
+ getAbsoluteRanges(llvm::Optional<object::SectionedAddress> BaseAddr) const;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFDEBUGRANGELIST_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRnglists.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRnglists.h
new file mode 100644
index 000000000..952c41e18
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDebugRnglists.h
@@ -0,0 +1,64 @@
+//===- DWARFDebugRnglists.h -------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDEBUGRNGLISTS_H
+#define LLVM_DEBUGINFO_DWARFDEBUGRNGLISTS_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRangeList.h"
+#include "llvm/DebugInfo/DWARF/DWARFListTable.h"
+#include <cstdint>
+#include <map>
+#include <vector>
+
+namespace llvm {
+
+class Error;
+class raw_ostream;
+class DWARFUnit;
+
+/// A class representing a single range list entry.
+struct RangeListEntry : public DWARFListEntryBase {
+ /// The values making up the range list entry. Most represent a range with
+ /// a start and end address or a start address and a length. Others are
+ /// single value base addresses or end-of-list with no values. The unneeded
+ /// values are semantically undefined, but initialized to 0.
+ uint64_t Value0;
+ uint64_t Value1;
+
+ Error extract(DWARFDataExtractor Data, uint64_t End, uint64_t *OffsetPtr);
+ void dump(raw_ostream &OS, uint8_t AddrSize, uint8_t MaxEncodingStringLength,
+ uint64_t &CurrentBase, DIDumpOptions DumpOpts,
+ llvm::function_ref<Optional<object::SectionedAddress>(uint32_t)>
+ LookupPooledAddress) const;
+ bool isSentinel() const { return EntryKind == dwarf::DW_RLE_end_of_list; }
+};
+
+/// A class representing a single rangelist.
+class DWARFDebugRnglist : public DWARFListType<RangeListEntry> {
+public:
+ /// Build a DWARFAddressRangesVector from a rangelist.
+ DWARFAddressRangesVector
+ getAbsoluteRanges(llvm::Optional<object::SectionedAddress> BaseAddr,
+ DWARFUnit &U) const;
+};
+
+class DWARFDebugRnglistTable : public DWARFListTableBase<DWARFDebugRnglist> {
+public:
+ DWARFDebugRnglistTable()
+ : DWARFListTableBase(/* SectionName = */ ".debug_rnglists",
+ /* HeaderString = */ "ranges:",
+ /* ListTypeString = */ "range") {}
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDEBUGRNGLISTS_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDie.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDie.h
new file mode 100644
index 000000000..f7f08b4a4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFDie.h
@@ -0,0 +1,468 @@
+//===- DWARFDie.h -----------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFDIE_H
+#define LLVM_DEBUGINFO_DWARFDIE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFAddressRange.h"
+#include "llvm/DebugInfo/DWARF/DWARFAttribute.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h"
+#include <cassert>
+#include <cstdint>
+#include <iterator>
+
+namespace llvm {
+
+class DWARFUnit;
+class raw_ostream;
+
+//===----------------------------------------------------------------------===//
+/// Utility class that carries the DWARF compile/type unit and the debug info
+/// entry in an object.
+///
+/// When accessing information from a debug info entry we always need to DWARF
+/// compile/type unit in order to extract the info correctly as some information
+/// is relative to the compile/type unit. Prior to this class the DWARFUnit and
+/// the DWARFDebugInfoEntry was passed around separately and there was the
+/// possibility for error if the wrong DWARFUnit was used to extract a unit
+/// relative offset. This class helps to ensure that this doesn't happen and
+/// also simplifies the attribute extraction calls by not having to specify the
+/// DWARFUnit for each call.
+class DWARFDie {
+ DWARFUnit *U = nullptr;
+ const DWARFDebugInfoEntry *Die = nullptr;
+
+public:
+ DWARFDie() = default;
+ DWARFDie(DWARFUnit *Unit, const DWARFDebugInfoEntry *D) : U(Unit), Die(D) {}
+
+ bool isValid() const { return U && Die; }
+ explicit operator bool() const { return isValid(); }
+ const DWARFDebugInfoEntry *getDebugInfoEntry() const { return Die; }
+ DWARFUnit *getDwarfUnit() const { return U; }
+
+ /// Get the abbreviation declaration for this DIE.
+ ///
+ /// \returns the abbreviation declaration or NULL for null tags.
+ const DWARFAbbreviationDeclaration *getAbbreviationDeclarationPtr() const {
+ assert(isValid() && "must check validity prior to calling");
+ return Die->getAbbreviationDeclarationPtr();
+ }
+
+ /// Get the absolute offset into the debug info or types section.
+ ///
+ /// \returns the DIE offset or -1U if invalid.
+ uint64_t getOffset() const {
+ assert(isValid() && "must check validity prior to calling");
+ return Die->getOffset();
+ }
+
+ dwarf::Tag getTag() const {
+ auto AbbrevDecl = getAbbreviationDeclarationPtr();
+ if (AbbrevDecl)
+ return AbbrevDecl->getTag();
+ return dwarf::DW_TAG_null;
+ }
+
+ bool hasChildren() const {
+ assert(isValid() && "must check validity prior to calling");
+ return Die->hasChildren();
+ }
+
+ /// Returns true for a valid DIE that terminates a sibling chain.
+ bool isNULL() const { return getAbbreviationDeclarationPtr() == nullptr; }
+
+ /// Returns true if DIE represents a subprogram (not inlined).
+ bool isSubprogramDIE() const;
+
+ /// Returns true if DIE represents a subprogram or an inlined subroutine.
+ bool isSubroutineDIE() const;
+
+ /// Get the parent of this DIE object.
+ ///
+ /// \returns a valid DWARFDie instance if this object has a parent or an
+ /// invalid DWARFDie instance if it doesn't.
+ DWARFDie getParent() const;
+
+ /// Get the sibling of this DIE object.
+ ///
+ /// \returns a valid DWARFDie instance if this object has a sibling or an
+ /// invalid DWARFDie instance if it doesn't.
+ DWARFDie getSibling() const;
+
+ /// Get the previous sibling of this DIE object.
+ ///
+ /// \returns a valid DWARFDie instance if this object has a sibling or an
+ /// invalid DWARFDie instance if it doesn't.
+ DWARFDie getPreviousSibling() const;
+
+ /// Get the first child of this DIE object.
+ ///
+ /// \returns a valid DWARFDie instance if this object has children or an
+ /// invalid DWARFDie instance if it doesn't.
+ DWARFDie getFirstChild() const;
+
+ /// Get the last child of this DIE object.
+ ///
+ /// \returns a valid null DWARFDie instance if this object has children or an
+ /// invalid DWARFDie instance if it doesn't.
+ DWARFDie getLastChild() const;
+
+ /// Dump the DIE and all of its attributes to the supplied stream.
+ ///
+ /// \param OS the stream to use for output.
+ /// \param indent the number of characters to indent each line that is output.
+ void dump(raw_ostream &OS, unsigned indent = 0,
+ DIDumpOptions DumpOpts = DIDumpOptions()) const;
+
+ /// Convenience zero-argument overload for debugging.
+ LLVM_DUMP_METHOD void dump() const;
+
+ /// Extract the specified attribute from this DIE.
+ ///
+ /// Extract an attribute value from this DIE only. This call doesn't look
+ /// for the attribute value in any DW_AT_specification or
+ /// DW_AT_abstract_origin referenced DIEs.
+ ///
+ /// \param Attr the attribute to extract.
+ /// \returns an optional DWARFFormValue that will have the form value if the
+ /// attribute was successfully extracted.
+ Optional<DWARFFormValue> find(dwarf::Attribute Attr) const;
+
+ /// Extract the first value of any attribute in Attrs from this DIE.
+ ///
+ /// Extract the first attribute that matches from this DIE only. This call
+ /// doesn't look for the attribute value in any DW_AT_specification or
+ /// DW_AT_abstract_origin referenced DIEs. The attributes will be searched
+ /// linearly in the order they are specified within Attrs.
+ ///
+ /// \param Attrs an array of DWARF attribute to look for.
+ /// \returns an optional that has a valid DWARFFormValue for the first
+ /// matching attribute in Attrs, or None if none of the attributes in Attrs
+ /// exist in this DIE.
+ Optional<DWARFFormValue> find(ArrayRef<dwarf::Attribute> Attrs) const;
+
+ /// Extract the first value of any attribute in Attrs from this DIE and
+ /// recurse into any DW_AT_specification or DW_AT_abstract_origin referenced
+ /// DIEs.
+ ///
+ /// \param Attrs an array of DWARF attribute to look for.
+ /// \returns an optional that has a valid DWARFFormValue for the first
+ /// matching attribute in Attrs, or None if none of the attributes in Attrs
+ /// exist in this DIE or in any DW_AT_specification or DW_AT_abstract_origin
+ /// DIEs.
+ Optional<DWARFFormValue>
+ findRecursively(ArrayRef<dwarf::Attribute> Attrs) const;
+
+ /// Extract the specified attribute from this DIE as the referenced DIE.
+ ///
+ /// Regardless of the reference type, return the correct DWARFDie instance if
+ /// the attribute exists. The returned DWARFDie object might be from another
+ /// DWARFUnit, but that is all encapsulated in the new DWARFDie object.
+ ///
+ /// Extract an attribute value from this DIE only. This call doesn't look
+ /// for the attribute value in any DW_AT_specification or
+ /// DW_AT_abstract_origin referenced DIEs.
+ ///
+ /// \param Attr the attribute to extract.
+ /// \returns a valid DWARFDie instance if the attribute exists, or an invalid
+ /// DWARFDie object if it doesn't.
+ DWARFDie getAttributeValueAsReferencedDie(dwarf::Attribute Attr) const;
+ DWARFDie getAttributeValueAsReferencedDie(const DWARFFormValue &V) const;
+
+ /// Extract the range base attribute from this DIE as absolute section offset.
+ ///
+ /// This is a utility function that checks for either the DW_AT_rnglists_base
+ /// or DW_AT_GNU_ranges_base attribute.
+ ///
+ /// \returns anm optional absolute section offset value for the attribute.
+ Optional<uint64_t> getRangesBaseAttribute() const;
+
+ /// Get the DW_AT_high_pc attribute value as an address.
+ ///
+ /// In DWARF version 4 and later the high PC can be encoded as an offset from
+ /// the DW_AT_low_pc. This function takes care of extracting the value as an
+ /// address or offset and adds it to the low PC if needed and returns the
+ /// value as an optional in case the DIE doesn't have a DW_AT_high_pc
+ /// attribute.
+ ///
+ /// \param LowPC the low PC that might be needed to calculate the high PC.
+ /// \returns an optional address value for the attribute.
+ Optional<uint64_t> getHighPC(uint64_t LowPC) const;
+
+ /// Retrieves DW_AT_low_pc and DW_AT_high_pc from CU.
+ /// Returns true if both attributes are present.
+ bool getLowAndHighPC(uint64_t &LowPC, uint64_t &HighPC,
+ uint64_t &SectionIndex) const;
+
+ /// Get the address ranges for this DIE.
+ ///
+ /// Get the hi/low PC range if both attributes are available or exrtracts the
+ /// non-contiguous address ranges from the DW_AT_ranges attribute.
+ ///
+ /// Extracts the range information from this DIE only. This call doesn't look
+ /// for the range in any DW_AT_specification or DW_AT_abstract_origin DIEs.
+ ///
+ /// \returns a address range vector that might be empty if no address range
+ /// information is available.
+ Expected<DWARFAddressRangesVector> getAddressRanges() const;
+
+ /// Get all address ranges for any DW_TAG_subprogram DIEs in this DIE or any
+ /// of its children.
+ ///
+ /// Get the hi/low PC range if both attributes are available or exrtracts the
+ /// non-contiguous address ranges from the DW_AT_ranges attribute for this DIE
+ /// and all children.
+ ///
+ /// \param Ranges the addres range vector to fill in.
+ void collectChildrenAddressRanges(DWARFAddressRangesVector &Ranges) const;
+
+ bool addressRangeContainsAddress(const uint64_t Address) const;
+
+ /// If a DIE represents a subprogram (or inlined subroutine), returns its
+ /// mangled name (or short name, if mangled is missing). This name may be
+ /// fetched from specification or abstract origin for this subprogram.
+ /// Returns null if no name is found.
+ const char *getSubroutineName(DINameKind Kind) const;
+
+ /// Return the DIE name resolving DW_AT_sepcification or DW_AT_abstract_origin
+ /// references if necessary. Returns null if no name is found.
+ const char *getName(DINameKind Kind) const;
+
+ /// Returns the declaration line (start line) for a DIE, assuming it specifies
+ /// a subprogram. This may be fetched from specification or abstract origin
+ /// for this subprogram by resolving DW_AT_sepcification or
+ /// DW_AT_abstract_origin references if necessary.
+ uint64_t getDeclLine() const;
+
+ /// Retrieves values of DW_AT_call_file, DW_AT_call_line and DW_AT_call_column
+ /// from DIE (or zeroes if they are missing). This function looks for
+ /// DW_AT_call attributes in this DIE only, it will not resolve the attribute
+ /// values in any DW_AT_specification or DW_AT_abstract_origin DIEs.
+ /// \param CallFile filled in with non-zero if successful, zero if there is no
+ /// DW_AT_call_file attribute in this DIE.
+ /// \param CallLine filled in with non-zero if successful, zero if there is no
+ /// DW_AT_call_line attribute in this DIE.
+ /// \param CallColumn filled in with non-zero if successful, zero if there is
+ /// no DW_AT_call_column attribute in this DIE.
+ /// \param CallDiscriminator filled in with non-zero if successful, zero if
+ /// there is no DW_AT_GNU_discriminator attribute in this DIE.
+ void getCallerFrame(uint32_t &CallFile, uint32_t &CallLine,
+ uint32_t &CallColumn, uint32_t &CallDiscriminator) const;
+
+ class attribute_iterator;
+
+ /// Get an iterator range to all attributes in the current DIE only.
+ ///
+ /// \returns an iterator range for the attributes of the current DIE.
+ iterator_range<attribute_iterator> attributes() const;
+
+ class iterator;
+
+ iterator begin() const;
+ iterator end() const;
+
+ std::reverse_iterator<iterator> rbegin() const;
+ std::reverse_iterator<iterator> rend() const;
+
+ iterator_range<iterator> children() const;
+};
+
+class DWARFDie::attribute_iterator
+ : public iterator_facade_base<attribute_iterator, std::forward_iterator_tag,
+ const DWARFAttribute> {
+ /// The DWARF DIE we are extracting attributes from.
+ DWARFDie Die;
+ /// The value vended to clients via the operator*() or operator->().
+ DWARFAttribute AttrValue;
+ /// The attribute index within the abbreviation declaration in Die.
+ uint32_t Index;
+
+ friend bool operator==(const attribute_iterator &LHS,
+ const attribute_iterator &RHS);
+
+ /// Update the attribute index and attempt to read the attribute value. If the
+ /// attribute is able to be read, update AttrValue and the Index member
+ /// variable. If the attribute value is not able to be read, an appropriate
+ /// error will be set if the Err member variable is non-NULL and the iterator
+ /// will be set to the end value so iteration stops.
+ void updateForIndex(const DWARFAbbreviationDeclaration &AbbrDecl, uint32_t I);
+
+public:
+ attribute_iterator() = delete;
+ explicit attribute_iterator(DWARFDie D, bool End);
+
+ attribute_iterator &operator++();
+ attribute_iterator &operator--();
+ explicit operator bool() const { return AttrValue.isValid(); }
+ const DWARFAttribute &operator*() const { return AttrValue; }
+};
+
+inline bool operator==(const DWARFDie::attribute_iterator &LHS,
+ const DWARFDie::attribute_iterator &RHS) {
+ return LHS.Index == RHS.Index;
+}
+
+inline bool operator!=(const DWARFDie::attribute_iterator &LHS,
+ const DWARFDie::attribute_iterator &RHS) {
+ return !(LHS == RHS);
+}
+
+inline bool operator==(const DWARFDie &LHS, const DWARFDie &RHS) {
+ return LHS.getDebugInfoEntry() == RHS.getDebugInfoEntry() &&
+ LHS.getDwarfUnit() == RHS.getDwarfUnit();
+}
+
+inline bool operator!=(const DWARFDie &LHS, const DWARFDie &RHS) {
+ return !(LHS == RHS);
+}
+
+inline bool operator<(const DWARFDie &LHS, const DWARFDie &RHS) {
+ return LHS.getOffset() < RHS.getOffset();
+}
+
+class DWARFDie::iterator
+ : public iterator_facade_base<iterator, std::bidirectional_iterator_tag,
+ const DWARFDie> {
+ DWARFDie Die;
+
+ friend std::reverse_iterator<llvm::DWARFDie::iterator>;
+ friend bool operator==(const DWARFDie::iterator &LHS,
+ const DWARFDie::iterator &RHS);
+
+public:
+ iterator() = default;
+
+ explicit iterator(DWARFDie D) : Die(D) {}
+
+ iterator &operator++() {
+ Die = Die.getSibling();
+ return *this;
+ }
+
+ iterator &operator--() {
+ Die = Die.getPreviousSibling();
+ return *this;
+ }
+
+ const DWARFDie &operator*() const { return Die; }
+};
+
+inline bool operator==(const DWARFDie::iterator &LHS,
+ const DWARFDie::iterator &RHS) {
+ return LHS.Die == RHS.Die;
+}
+
+inline bool operator!=(const DWARFDie::iterator &LHS,
+ const DWARFDie::iterator &RHS) {
+ return !(LHS == RHS);
+}
+
+// These inline functions must follow the DWARFDie::iterator definition above
+// as they use functions from that class.
+inline DWARFDie::iterator DWARFDie::begin() const {
+ return iterator(getFirstChild());
+}
+
+inline DWARFDie::iterator DWARFDie::end() const {
+ return iterator(getLastChild());
+}
+
+inline iterator_range<DWARFDie::iterator> DWARFDie::children() const {
+ return make_range(begin(), end());
+}
+
+} // end namespace llvm
+
+namespace std {
+
+template <>
+class reverse_iterator<llvm::DWARFDie::iterator>
+ : public llvm::iterator_facade_base<
+ reverse_iterator<llvm::DWARFDie::iterator>,
+ bidirectional_iterator_tag, const llvm::DWARFDie> {
+
+private:
+ llvm::DWARFDie Die;
+ bool AtEnd;
+
+public:
+ reverse_iterator(llvm::DWARFDie::iterator It)
+ : Die(It.Die), AtEnd(!It.Die.getPreviousSibling()) {
+ if (!AtEnd)
+ Die = Die.getPreviousSibling();
+ }
+
+ llvm::DWARFDie::iterator base() const {
+ return llvm::DWARFDie::iterator(AtEnd ? Die : Die.getSibling());
+ }
+
+ reverse_iterator<llvm::DWARFDie::iterator> &operator++() {
+ assert(!AtEnd && "Incrementing rend");
+ llvm::DWARFDie D = Die.getPreviousSibling();
+ if (D)
+ Die = D;
+ else
+ AtEnd = true;
+ return *this;
+ }
+
+ reverse_iterator<llvm::DWARFDie::iterator> &operator--() {
+ if (AtEnd) {
+ AtEnd = false;
+ return *this;
+ }
+ Die = Die.getSibling();
+ assert(!Die.isNULL() && "Decrementing rbegin");
+ return *this;
+ }
+
+ const llvm::DWARFDie &operator*() const {
+ assert(Die.isValid());
+ return Die;
+ }
+
+ // FIXME: We should be able to specify the equals operator as a friend, but
+ // that causes the compiler to think the operator overload is ambiguous
+ // with the friend declaration and the actual definition as candidates.
+ bool equals(const reverse_iterator<llvm::DWARFDie::iterator> &RHS) const {
+ return Die == RHS.Die && AtEnd == RHS.AtEnd;
+ }
+};
+
+} // namespace std
+
+namespace llvm {
+
+inline bool operator==(const std::reverse_iterator<DWARFDie::iterator> &LHS,
+ const std::reverse_iterator<DWARFDie::iterator> &RHS) {
+ return LHS.equals(RHS);
+}
+
+inline bool operator!=(const std::reverse_iterator<DWARFDie::iterator> &LHS,
+ const std::reverse_iterator<DWARFDie::iterator> &RHS) {
+ return !(LHS == RHS);
+}
+
+inline std::reverse_iterator<DWARFDie::iterator> DWARFDie::rbegin() const {
+ return llvm::make_reverse_iterator(end());
+}
+
+inline std::reverse_iterator<DWARFDie::iterator> DWARFDie::rend() const {
+ return llvm::make_reverse_iterator(begin());
+}
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFDIE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFExpression.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFExpression.h
new file mode 100644
index 000000000..456d9df95
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFExpression.h
@@ -0,0 +1,159 @@
+//===--- DWARFExpression.h - DWARF Expression handling ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFEXPRESSION_H
+#define LLVM_DEBUGINFO_DWARFEXPRESSION_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/DataExtractor.h"
+
+namespace llvm {
+class DWARFUnit;
+class MCRegisterInfo;
+class raw_ostream;
+
+class DWARFExpression {
+public:
+ class iterator;
+
+ /// This class represents an Operation in the Expression. Each operation can
+ /// have up to 2 oprerands.
+ ///
+ /// An Operation can be in Error state (check with isError()). This
+ /// means that it couldn't be decoded successfully and if it is the
+ /// case, all others fields contain undefined values.
+ class Operation {
+ public:
+ /// Size and signedness of expression operations' operands.
+ enum Encoding : uint8_t {
+ Size1 = 0,
+ Size2 = 1,
+ Size4 = 2,
+ Size8 = 3,
+ SizeLEB = 4,
+ SizeAddr = 5,
+ SizeRefAddr = 6,
+ SizeBlock = 7, ///< Preceding operand contains block size
+ BaseTypeRef = 8,
+ SignBit = 0x80,
+ SignedSize1 = SignBit | Size1,
+ SignedSize2 = SignBit | Size2,
+ SignedSize4 = SignBit | Size4,
+ SignedSize8 = SignBit | Size8,
+ SignedSizeLEB = SignBit | SizeLEB,
+ SizeNA = 0xFF ///< Unused operands get this encoding.
+ };
+
+ enum DwarfVersion : uint8_t {
+ DwarfNA, ///< Serves as a marker for unused entries
+ Dwarf2 = 2,
+ Dwarf3,
+ Dwarf4,
+ Dwarf5
+ };
+
+ /// Description of the encoding of one expression Op.
+ struct Description {
+ DwarfVersion Version; ///< Dwarf version where the Op was introduced.
+ Encoding Op[2]; ///< Encoding for Op operands, or SizeNA.
+
+ Description(DwarfVersion Version = DwarfNA, Encoding Op1 = SizeNA,
+ Encoding Op2 = SizeNA)
+ : Version(Version) {
+ Op[0] = Op1;
+ Op[1] = Op2;
+ }
+ };
+
+ private:
+ friend class DWARFExpression::iterator;
+ uint8_t Opcode; ///< The Op Opcode, DW_OP_<something>.
+ Description Desc;
+ bool Error;
+ uint64_t EndOffset;
+ uint64_t Operands[2];
+ uint64_t OperandEndOffsets[2];
+
+ public:
+ Description &getDescription() { return Desc; }
+ uint8_t getCode() { return Opcode; }
+ uint64_t getRawOperand(unsigned Idx) { return Operands[Idx]; }
+ uint64_t getOperandEndOffset(unsigned Idx) { return OperandEndOffsets[Idx]; }
+ uint64_t getEndOffset() { return EndOffset; }
+ bool extract(DataExtractor Data, uint16_t Version, uint8_t AddressSize,
+ uint64_t Offset);
+ bool isError() { return Error; }
+ bool print(raw_ostream &OS, const DWARFExpression *Expr,
+ const MCRegisterInfo *RegInfo, DWARFUnit *U, bool isEH);
+ bool verify(DWARFUnit *U);
+ };
+
+ /// An iterator to go through the expression operations.
+ class iterator
+ : public iterator_facade_base<iterator, std::forward_iterator_tag,
+ Operation> {
+ friend class DWARFExpression;
+ const DWARFExpression *Expr;
+ uint64_t Offset;
+ Operation Op;
+ iterator(const DWARFExpression *Expr, uint64_t Offset)
+ : Expr(Expr), Offset(Offset) {
+ Op.Error =
+ Offset >= Expr->Data.getData().size() ||
+ !Op.extract(Expr->Data, Expr->Version, Expr->AddressSize, Offset);
+ }
+
+ public:
+ class Operation &operator++() {
+ Offset = Op.isError() ? Expr->Data.getData().size() : Op.EndOffset;
+ Op.Error =
+ Offset >= Expr->Data.getData().size() ||
+ !Op.extract(Expr->Data, Expr->Version, Expr->AddressSize, Offset);
+ return Op;
+ }
+
+ class Operation &operator*() {
+ return Op;
+ }
+
+ // Comparison operators are provided out of line.
+ friend bool operator==(const iterator &, const iterator &);
+ };
+
+ DWARFExpression(DataExtractor Data, uint16_t Version, uint8_t AddressSize)
+ : Data(Data), Version(Version), AddressSize(AddressSize) {
+ assert(AddressSize == 8 || AddressSize == 4 || AddressSize == 2);
+ }
+
+ iterator begin() const { return iterator(this, 0); }
+ iterator end() const { return iterator(this, Data.getData().size()); }
+
+ void print(raw_ostream &OS, const MCRegisterInfo *RegInfo, DWARFUnit *U,
+ bool IsEH = false) const;
+
+ bool verify(DWARFUnit *U);
+
+private:
+ DataExtractor Data;
+ uint16_t Version;
+ uint8_t AddressSize;
+};
+
+inline bool operator==(const DWARFExpression::iterator &LHS,
+ const DWARFExpression::iterator &RHS) {
+ return LHS.Expr == RHS.Expr && LHS.Offset == RHS.Offset;
+}
+
+inline bool operator!=(const DWARFExpression::iterator &LHS,
+ const DWARFExpression::iterator &RHS) {
+ return !(LHS == RHS);
+}
+}
+#endif
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFFormValue.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFFormValue.h
new file mode 100644
index 000000000..6fec6fcb6
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFFormValue.h
@@ -0,0 +1,321 @@
+//===- DWARFFormValue.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFFORMVALUE_H
+#define LLVM_DEBUGINFO_DWARFFORMVALUE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include <cstdint>
+
+namespace llvm {
+
+class DWARFContext;
+class DWARFUnit;
+class raw_ostream;
+
+class DWARFFormValue {
+public:
+ enum FormClass {
+ FC_Unknown,
+ FC_Address,
+ FC_Block,
+ FC_Constant,
+ FC_String,
+ FC_Flag,
+ FC_Reference,
+ FC_Indirect,
+ FC_SectionOffset,
+ FC_Exprloc
+ };
+
+private:
+ struct ValueType {
+ ValueType() { uval = 0; }
+ ValueType(int64_t V) : sval(V) {}
+ ValueType(uint64_t V) : uval(V) {}
+ ValueType(const char *V) : cstr(V) {}
+
+ union {
+ uint64_t uval;
+ int64_t sval;
+ const char *cstr;
+ };
+ const uint8_t *data = nullptr;
+ uint64_t SectionIndex; /// Section index for reference forms.
+ };
+
+ dwarf::Form Form; /// Form for this value.
+ ValueType Value; /// Contains all data for the form.
+ const DWARFUnit *U = nullptr; /// Remember the DWARFUnit at extract time.
+ const DWARFContext *C = nullptr; /// Context for extract time.
+
+ DWARFFormValue(dwarf::Form F, ValueType V) : Form(F), Value(V) {}
+
+public:
+ DWARFFormValue(dwarf::Form F = dwarf::Form(0)) : Form(F) {}
+
+ static DWARFFormValue createFromSValue(dwarf::Form F, int64_t V);
+ static DWARFFormValue createFromUValue(dwarf::Form F, uint64_t V);
+ static DWARFFormValue createFromPValue(dwarf::Form F, const char *V);
+ static DWARFFormValue createFromBlockValue(dwarf::Form F,
+ ArrayRef<uint8_t> D);
+ static DWARFFormValue createFromUnit(dwarf::Form F, const DWARFUnit *Unit,
+ uint64_t *OffsetPtr);
+
+ dwarf::Form getForm() const { return Form; }
+ uint64_t getRawUValue() const { return Value.uval; }
+
+ bool isFormClass(FormClass FC) const;
+ const DWARFUnit *getUnit() const { return U; }
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts = DIDumpOptions()) const;
+ void dumpSectionedAddress(raw_ostream &OS, DIDumpOptions DumpOpts,
+ object::SectionedAddress SA) const;
+ static void dumpAddressSection(const DWARFObject &Obj, raw_ostream &OS,
+ DIDumpOptions DumpOpts, uint64_t SectionIndex);
+
+ /// Extracts a value in \p Data at offset \p *OffsetPtr. The information
+ /// in \p FormParams is needed to interpret some forms. The optional
+ /// \p Context and \p Unit allows extracting information if the form refers
+ /// to other sections (e.g., .debug_str).
+ bool extractValue(const DWARFDataExtractor &Data, uint64_t *OffsetPtr,
+ dwarf::FormParams FormParams,
+ const DWARFContext *Context = nullptr,
+ const DWARFUnit *Unit = nullptr);
+
+ bool extractValue(const DWARFDataExtractor &Data, uint64_t *OffsetPtr,
+ dwarf::FormParams FormParams, const DWARFUnit *U) {
+ return extractValue(Data, OffsetPtr, FormParams, nullptr, U);
+ }
+
+ bool isInlinedCStr() const {
+ return Value.data != nullptr && Value.data == (const uint8_t *)Value.cstr;
+ }
+
+ /// getAsFoo functions below return the extracted value as Foo if only
+ /// DWARFFormValue has form class is suitable for representing Foo.
+ Optional<uint64_t> getAsReference() const;
+ struct UnitOffset {
+ DWARFUnit *Unit;
+ uint64_t Offset;
+ };
+ Optional<UnitOffset> getAsRelativeReference() const;
+ Optional<uint64_t> getAsUnsignedConstant() const;
+ Optional<int64_t> getAsSignedConstant() const;
+ Optional<const char *> getAsCString() const;
+ Optional<uint64_t> getAsAddress() const;
+ Optional<object::SectionedAddress> getAsSectionedAddress() const;
+ Optional<uint64_t> getAsSectionOffset() const;
+ Optional<ArrayRef<uint8_t>> getAsBlock() const;
+ Optional<uint64_t> getAsCStringOffset() const;
+ Optional<uint64_t> getAsReferenceUVal() const;
+
+ /// Skip a form's value in \p DebugInfoData at the offset specified by
+ /// \p OffsetPtr.
+ ///
+ /// Skips the bytes for the current form and updates the offset.
+ ///
+ /// \param DebugInfoData The data where we want to skip the value.
+ /// \param OffsetPtr A reference to the offset that will be updated.
+ /// \param Params DWARF parameters to help interpret forms.
+ /// \returns true on success, false if the form was not skipped.
+ bool skipValue(DataExtractor DebugInfoData, uint64_t *OffsetPtr,
+ const dwarf::FormParams Params) const {
+ return DWARFFormValue::skipValue(Form, DebugInfoData, OffsetPtr, Params);
+ }
+
+ /// Skip a form's value in \p DebugInfoData at the offset specified by
+ /// \p OffsetPtr.
+ ///
+ /// Skips the bytes for the specified form and updates the offset.
+ ///
+ /// \param Form The DW_FORM enumeration that indicates the form to skip.
+ /// \param DebugInfoData The data where we want to skip the value.
+ /// \param OffsetPtr A reference to the offset that will be updated.
+ /// \param FormParams DWARF parameters to help interpret forms.
+ /// \returns true on success, false if the form was not skipped.
+ static bool skipValue(dwarf::Form Form, DataExtractor DebugInfoData,
+ uint64_t *OffsetPtr,
+ const dwarf::FormParams FormParams);
+
+private:
+ void dumpString(raw_ostream &OS) const;
+};
+
+namespace dwarf {
+
+/// Take an optional DWARFFormValue and try to extract a string value from it.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and was a string.
+inline Optional<const char *> toString(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsCString();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and try to extract a string value from it.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and was a string.
+inline StringRef toStringRef(const Optional<DWARFFormValue> &V,
+ StringRef Default = {}) {
+ if (V)
+ if (auto S = V->getAsCString())
+ return *S;
+ return Default;
+}
+
+/// Take an optional DWARFFormValue and extract a string value from it.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the string value or Default if the V doesn't have a value or the
+/// form value's encoding wasn't a string.
+inline const char *toString(const Optional<DWARFFormValue> &V,
+ const char *Default) {
+ return toString(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract an unsigned constant.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a unsigned constant form.
+inline Optional<uint64_t> toUnsigned(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsUnsignedConstant();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and extract a unsigned constant.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the extracted unsigned value or Default if the V doesn't have a
+/// value or the form value's encoding wasn't an unsigned constant form.
+inline uint64_t toUnsigned(const Optional<DWARFFormValue> &V,
+ uint64_t Default) {
+ return toUnsigned(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract an reference.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a reference form.
+inline Optional<uint64_t> toReference(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsReference();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and extract a reference.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the extracted reference value or Default if the V doesn't have a
+/// value or the form value's encoding wasn't a reference form.
+inline uint64_t toReference(const Optional<DWARFFormValue> &V,
+ uint64_t Default) {
+ return toReference(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract an signed constant.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a signed constant form.
+inline Optional<int64_t> toSigned(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsSignedConstant();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and extract a signed integer.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the extracted signed integer value or Default if the V doesn't
+/// have a value or the form value's encoding wasn't a signed integer form.
+inline int64_t toSigned(const Optional<DWARFFormValue> &V, int64_t Default) {
+ return toSigned(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract an address.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a address form.
+inline Optional<uint64_t> toAddress(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsAddress();
+ return None;
+}
+
+inline Optional<object::SectionedAddress>
+toSectionedAddress(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsSectionedAddress();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and extract a address.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the extracted address value or Default if the V doesn't have a
+/// value or the form value's encoding wasn't an address form.
+inline uint64_t toAddress(const Optional<DWARFFormValue> &V, uint64_t Default) {
+ return toAddress(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract an section offset.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a section offset form.
+inline Optional<uint64_t> toSectionOffset(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsSectionOffset();
+ return None;
+}
+
+/// Take an optional DWARFFormValue and extract a section offset.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \param Default the default value to return in case of failure.
+/// \returns the extracted section offset value or Default if the V doesn't
+/// have a value or the form value's encoding wasn't a section offset form.
+inline uint64_t toSectionOffset(const Optional<DWARFFormValue> &V,
+ uint64_t Default) {
+ return toSectionOffset(V).getValueOr(Default);
+}
+
+/// Take an optional DWARFFormValue and try to extract block data.
+///
+/// \param V and optional DWARFFormValue to attempt to extract the value from.
+/// \returns an optional value that contains a value if the form value
+/// was valid and has a block form.
+inline Optional<ArrayRef<uint8_t>> toBlock(const Optional<DWARFFormValue> &V) {
+ if (V)
+ return V->getAsBlock();
+ return None;
+}
+
+} // end namespace dwarf
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFFORMVALUE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h
new file mode 100644
index 000000000..38cd42ddb
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h
@@ -0,0 +1,82 @@
+//===- DWARFGdbIndex.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFGDBINDEX_H
+#define LLVM_DEBUGINFO_DWARF_DWARFGDBINDEX_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+class raw_ostream;
+
+class DWARFGdbIndex {
+ uint32_t Version;
+
+ uint32_t CuListOffset;
+ uint32_t TuListOffset;
+ uint32_t AddressAreaOffset;
+ uint32_t SymbolTableOffset;
+ uint32_t ConstantPoolOffset;
+
+ struct CompUnitEntry {
+ uint64_t Offset; /// Offset of a CU in the .debug_info section.
+ uint64_t Length; /// Length of that CU.
+ };
+ SmallVector<CompUnitEntry, 0> CuList;
+
+ struct TypeUnitEntry {
+ uint64_t Offset;
+ uint64_t TypeOffset;
+ uint64_t TypeSignature;
+ };
+ SmallVector<TypeUnitEntry, 0> TuList;
+
+ struct AddressEntry {
+ uint64_t LowAddress; /// The low address.
+ uint64_t HighAddress; /// The high address.
+ uint32_t CuIndex; /// The CU index.
+ };
+ SmallVector<AddressEntry, 0> AddressArea;
+
+ struct SymTableEntry {
+ uint32_t NameOffset; /// Offset of the symbol's name in the constant pool.
+ uint32_t VecOffset; /// Offset of the CU vector in the constant pool.
+ };
+ SmallVector<SymTableEntry, 0> SymbolTable;
+
+ /// Each value is CU index + attributes.
+ SmallVector<std::pair<uint32_t, SmallVector<uint32_t, 0>>, 0>
+ ConstantPoolVectors;
+
+ StringRef ConstantPoolStrings;
+ uint32_t StringPoolOffset;
+
+ void dumpCUList(raw_ostream &OS) const;
+ void dumpTUList(raw_ostream &OS) const;
+ void dumpAddressArea(raw_ostream &OS) const;
+ void dumpSymbolTable(raw_ostream &OS) const;
+ void dumpConstantPool(raw_ostream &OS) const;
+
+ bool parseImpl(DataExtractor Data);
+
+public:
+ void dump(raw_ostream &OS);
+ void parse(DataExtractor Data);
+
+ bool HasContent = false;
+ bool HasError = false;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFGDBINDEX_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFListTable.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFListTable.h
new file mode 100644
index 000000000..496fdb247
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFListTable.h
@@ -0,0 +1,290 @@
+//===- DWARFListTable.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARFLISTTABLE_H
+#define LLVM_DEBUGINFO_DWARFLISTTABLE_H
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFDataExtractor.h"
+#include "llvm/Support/Errc.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstdint>
+#include <map>
+#include <vector>
+
+namespace llvm {
+
+/// A base class for DWARF list entries, such as range or location list
+/// entries.
+struct DWARFListEntryBase {
+ /// The offset at which the entry is located in the section.
+ uint64_t Offset;
+ /// The DWARF encoding (DW_RLE_* or DW_LLE_*).
+ uint8_t EntryKind;
+ /// The index of the section this entry belongs to.
+ uint64_t SectionIndex;
+};
+
+/// A base class for lists of entries that are extracted from a particular
+/// section, such as range lists or location lists.
+template <typename ListEntryType> class DWARFListType {
+ using EntryType = ListEntryType;
+ using ListEntries = std::vector<EntryType>;
+
+protected:
+ ListEntries Entries;
+
+public:
+ const ListEntries &getEntries() const { return Entries; }
+ bool empty() const { return Entries.empty(); }
+ void clear() { Entries.clear(); }
+ Error extract(DWARFDataExtractor Data, uint64_t HeaderOffset, uint64_t End,
+ uint64_t *OffsetPtr, StringRef SectionName,
+ StringRef ListStringName);
+};
+
+/// A class representing the header of a list table such as the range list
+/// table in the .debug_rnglists section.
+class DWARFListTableHeader {
+ struct Header {
+ /// The total length of the entries for this table, not including the length
+ /// field itself.
+ uint64_t Length = 0;
+ /// The DWARF version number.
+ uint16_t Version;
+ /// The size in bytes of an address on the target architecture. For
+ /// segmented addressing, this is the size of the offset portion of the
+ /// address.
+ uint8_t AddrSize;
+ /// The size in bytes of a segment selector on the target architecture.
+ /// If the target system uses a flat address space, this value is 0.
+ uint8_t SegSize;
+ /// The number of offsets that follow the header before the range lists.
+ uint32_t OffsetEntryCount;
+ };
+
+ Header HeaderData;
+ /// The offset table, which contains offsets to the individual list entries.
+ /// It is used by forms such as DW_FORM_rnglistx.
+ /// FIXME: Generate the table and use the appropriate forms.
+ std::vector<uint64_t> Offsets;
+ /// The table's format, either DWARF32 or DWARF64.
+ dwarf::DwarfFormat Format;
+ /// The offset at which the header (and hence the table) is located within
+ /// its section.
+ uint64_t HeaderOffset;
+ /// The name of the section the list is located in.
+ StringRef SectionName;
+ /// A characterization of the list for dumping purposes, e.g. "range" or
+ /// "location".
+ StringRef ListTypeString;
+
+public:
+ DWARFListTableHeader(StringRef SectionName, StringRef ListTypeString)
+ : SectionName(SectionName), ListTypeString(ListTypeString) {}
+
+ void clear() {
+ HeaderData = {};
+ Offsets.clear();
+ }
+ uint64_t getHeaderOffset() const { return HeaderOffset; }
+ uint8_t getAddrSize() const { return HeaderData.AddrSize; }
+ uint64_t getLength() const { return HeaderData.Length; }
+ uint16_t getVersion() const { return HeaderData.Version; }
+ StringRef getSectionName() const { return SectionName; }
+ StringRef getListTypeString() const { return ListTypeString; }
+ dwarf::DwarfFormat getFormat() const { return Format; }
+
+ /// Return the size of the table header including the length but not including
+ /// the offsets.
+ static uint8_t getHeaderSize(dwarf::DwarfFormat Format) {
+ switch (Format) {
+ case dwarf::DwarfFormat::DWARF32:
+ return 12;
+ case dwarf::DwarfFormat::DWARF64:
+ return 20;
+ }
+ llvm_unreachable("Invalid DWARF format (expected DWARF32 or DWARF64");
+ }
+
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts = {}) const;
+ Optional<uint64_t> getOffsetEntry(uint32_t Index) const {
+ if (Index < Offsets.size())
+ return Offsets[Index];
+ return None;
+ }
+
+ /// Extract the table header and the array of offsets.
+ Error extract(DWARFDataExtractor Data, uint64_t *OffsetPtr);
+
+ /// Returns the length of the table, including the length field, or 0 if the
+ /// length has not been determined (e.g. because the table has not yet been
+ /// parsed, or there was a problem in parsing).
+ uint64_t length() const;
+};
+
+/// A class representing a table of lists as specified in the DWARF v5
+/// standard for location lists and range lists. The table consists of a header
+/// followed by an array of offsets into a DWARF section, followed by zero or
+/// more list entries. The list entries are kept in a map where the keys are
+/// the lists' section offsets.
+template <typename DWARFListType> class DWARFListTableBase {
+ DWARFListTableHeader Header;
+ /// A mapping between file offsets and lists. It is used to find a particular
+ /// list based on an offset (obtained from DW_AT_ranges, for example).
+ std::map<uint64_t, DWARFListType> ListMap;
+ /// This string is displayed as a heading before the list is dumped
+ /// (e.g. "ranges:").
+ StringRef HeaderString;
+
+protected:
+ DWARFListTableBase(StringRef SectionName, StringRef HeaderString,
+ StringRef ListTypeString)
+ : Header(SectionName, ListTypeString), HeaderString(HeaderString) {}
+
+public:
+ void clear() {
+ Header.clear();
+ ListMap.clear();
+ }
+ /// Extract the table header and the array of offsets.
+ Error extractHeaderAndOffsets(DWARFDataExtractor Data, uint64_t *OffsetPtr) {
+ return Header.extract(Data, OffsetPtr);
+ }
+ /// Extract an entire table, including all list entries.
+ Error extract(DWARFDataExtractor Data, uint64_t *OffsetPtr);
+ /// Look up a list based on a given offset. Extract it and enter it into the
+ /// list map if necessary.
+ Expected<DWARFListType> findList(DWARFDataExtractor Data, uint64_t Offset);
+
+ uint64_t getHeaderOffset() const { return Header.getHeaderOffset(); }
+ uint8_t getAddrSize() const { return Header.getAddrSize(); }
+ dwarf::DwarfFormat getFormat() const { return Header.getFormat(); }
+
+ void dump(raw_ostream &OS,
+ llvm::function_ref<Optional<object::SectionedAddress>(uint32_t)>
+ LookupPooledAddress,
+ DIDumpOptions DumpOpts = {}) const;
+
+ /// Return the contents of the offset entry designated by a given index.
+ Optional<uint64_t> getOffsetEntry(uint32_t Index) const {
+ return Header.getOffsetEntry(Index);
+ }
+ /// Return the size of the table header including the length but not including
+ /// the offsets. This is dependent on the table format, which is unambiguously
+ /// derived from parsing the table.
+ uint8_t getHeaderSize() const {
+ return DWARFListTableHeader::getHeaderSize(getFormat());
+ }
+
+ uint64_t length() { return Header.length(); }
+};
+
+template <typename DWARFListType>
+Error DWARFListTableBase<DWARFListType>::extract(DWARFDataExtractor Data,
+ uint64_t *OffsetPtr) {
+ clear();
+ if (Error E = extractHeaderAndOffsets(Data, OffsetPtr))
+ return E;
+
+ Data.setAddressSize(Header.getAddrSize());
+ uint64_t End = getHeaderOffset() + Header.length();
+ while (*OffsetPtr < End) {
+ DWARFListType CurrentList;
+ uint64_t Off = *OffsetPtr;
+ if (Error E = CurrentList.extract(Data, getHeaderOffset(), End, OffsetPtr,
+ Header.getSectionName(),
+ Header.getListTypeString()))
+ return E;
+ ListMap[Off] = CurrentList;
+ }
+
+ assert(*OffsetPtr == End &&
+ "mismatch between expected length of table and length "
+ "of extracted data");
+ return Error::success();
+}
+
+template <typename ListEntryType>
+Error DWARFListType<ListEntryType>::extract(DWARFDataExtractor Data,
+ uint64_t HeaderOffset, uint64_t End,
+ uint64_t *OffsetPtr,
+ StringRef SectionName,
+ StringRef ListTypeString) {
+ if (*OffsetPtr < HeaderOffset || *OffsetPtr >= End)
+ return createStringError(errc::invalid_argument,
+ "invalid %s list offset 0x%" PRIx64,
+ ListTypeString.data(), *OffsetPtr);
+ Entries.clear();
+ while (*OffsetPtr < End) {
+ ListEntryType Entry;
+ if (Error E = Entry.extract(Data, End, OffsetPtr))
+ return E;
+ Entries.push_back(Entry);
+ if (Entry.isSentinel())
+ return Error::success();
+ }
+ return createStringError(errc::illegal_byte_sequence,
+ "no end of list marker detected at end of %s table "
+ "starting at offset 0x%" PRIx64,
+ SectionName.data(), HeaderOffset);
+}
+
+template <typename DWARFListType>
+void DWARFListTableBase<DWARFListType>::dump(
+ raw_ostream &OS,
+ llvm::function_ref<Optional<object::SectionedAddress>(uint32_t)>
+ LookupPooledAddress,
+ DIDumpOptions DumpOpts) const {
+ Header.dump(OS, DumpOpts);
+ OS << HeaderString << "\n";
+
+ // Determine the length of the longest encoding string we have in the table,
+ // so we can align the output properly. We only need this in verbose mode.
+ size_t MaxEncodingStringLength = 0;
+ if (DumpOpts.Verbose) {
+ for (const auto &List : ListMap)
+ for (const auto &Entry : List.second.getEntries())
+ MaxEncodingStringLength =
+ std::max(MaxEncodingStringLength,
+ dwarf::RangeListEncodingString(Entry.EntryKind).size());
+ }
+
+ uint64_t CurrentBase = 0;
+ for (const auto &List : ListMap)
+ for (const auto &Entry : List.second.getEntries())
+ Entry.dump(OS, getAddrSize(), MaxEncodingStringLength, CurrentBase,
+ DumpOpts, LookupPooledAddress);
+}
+
+template <typename DWARFListType>
+Expected<DWARFListType>
+DWARFListTableBase<DWARFListType>::findList(DWARFDataExtractor Data,
+ uint64_t Offset) {
+ auto Entry = ListMap.find(Offset);
+ if (Entry != ListMap.end())
+ return Entry->second;
+
+ // Extract the list from the section and enter it into the list map.
+ DWARFListType List;
+ uint64_t End = getHeaderOffset() + Header.length();
+ uint64_t StartingOffset = Offset;
+ if (Error E =
+ List.extract(Data, getHeaderOffset(), End, &Offset,
+ Header.getSectionName(), Header.getListTypeString()))
+ return std::move(E);
+ ListMap[StartingOffset] = List;
+ return List;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARFLISTTABLE_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFObject.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFObject.h
new file mode 100644
index 000000000..88fe3f434
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFObject.h
@@ -0,0 +1,85 @@
+//===- DWARFObject.h --------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===-----------------------------------------------------------------------===/
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFOBJECT_H
+#define LLVM_DEBUGINFO_DWARF_DWARFOBJECT_H
+
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/Object/ObjectFile.h"
+
+namespace llvm {
+// This is responsible for low level access to the object file. It
+// knows how to find the required sections and compute relocated
+// values.
+// The default implementations of the get<Section> methods return dummy values.
+// This is to allow clients that only need some of those to implement just the
+// ones they need. We can't use unreachable for as many cases because the parser
+// implementation is eager and will call some of these methods even if the
+// result is not used.
+class DWARFObject {
+ DWARFSection Dummy;
+
+public:
+ virtual ~DWARFObject() = default;
+ virtual StringRef getFileName() const { llvm_unreachable("unimplemented"); }
+ virtual const object::ObjectFile *getFile() const { return nullptr; }
+ virtual ArrayRef<SectionName> getSectionNames() const { return {}; }
+ virtual bool isLittleEndian() const = 0;
+ virtual uint8_t getAddressSize() const { llvm_unreachable("unimplemented"); }
+ virtual void
+ forEachInfoSections(function_ref<void(const DWARFSection &)> F) const {}
+ virtual void
+ forEachTypesSections(function_ref<void(const DWARFSection &)> F) const {}
+ virtual StringRef getAbbrevSection() const { return ""; }
+ virtual const DWARFSection &getLocSection() const { return Dummy; }
+ virtual const DWARFSection &getLoclistsSection() const { return Dummy; }
+ virtual StringRef getArangesSection() const { return ""; }
+ virtual const DWARFSection &getFrameSection() const { return Dummy; }
+ virtual const DWARFSection &getEHFrameSection() const { return Dummy; }
+ virtual const DWARFSection &getLineSection() const { return Dummy; }
+ virtual StringRef getLineStrSection() const { return ""; }
+ virtual StringRef getStrSection() const { return ""; }
+ virtual const DWARFSection &getRangesSection() const { return Dummy; }
+ virtual const DWARFSection &getRnglistsSection() const { return Dummy; }
+ virtual StringRef getMacinfoSection() const { return ""; }
+ virtual const DWARFSection &getPubnamesSection() const { return Dummy; }
+ virtual const DWARFSection &getPubtypesSection() const { return Dummy; }
+ virtual const DWARFSection &getGnuPubnamesSection() const { return Dummy; }
+ virtual const DWARFSection &getGnuPubtypesSection() const { return Dummy; }
+ virtual const DWARFSection &getStrOffsetsSection() const { return Dummy; }
+ virtual void
+ forEachInfoDWOSections(function_ref<void(const DWARFSection &)> F) const {}
+ virtual void
+ forEachTypesDWOSections(function_ref<void(const DWARFSection &)> F) const {}
+ virtual StringRef getAbbrevDWOSection() const { return ""; }
+ virtual const DWARFSection &getLineDWOSection() const { return Dummy; }
+ virtual const DWARFSection &getLocDWOSection() const { return Dummy; }
+ virtual StringRef getStrDWOSection() const { return ""; }
+ virtual const DWARFSection &getStrOffsetsDWOSection() const {
+ return Dummy;
+ }
+ virtual const DWARFSection &getRangesDWOSection() const { return Dummy; }
+ virtual const DWARFSection &getRnglistsDWOSection() const { return Dummy; }
+ virtual const DWARFSection &getAddrSection() const { return Dummy; }
+ virtual const DWARFSection &getAppleNamesSection() const { return Dummy; }
+ virtual const DWARFSection &getAppleTypesSection() const { return Dummy; }
+ virtual const DWARFSection &getAppleNamespacesSection() const {
+ return Dummy;
+ }
+ virtual const DWARFSection &getNamesSection() const { return Dummy; }
+ virtual const DWARFSection &getAppleObjCSection() const { return Dummy; }
+ virtual StringRef getCUIndexSection() const { return ""; }
+ virtual StringRef getGdbIndexSection() const { return ""; }
+ virtual StringRef getTUIndexSection() const { return ""; }
+ virtual Optional<RelocAddrEntry> find(const DWARFSection &Sec,
+ uint64_t Pos) const = 0;
+};
+
+} // namespace llvm
+#endif
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h
new file mode 100644
index 000000000..3add71194
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h
@@ -0,0 +1,38 @@
+//===- DWARFRelocMap.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFRELOCMAP_H
+#define LLVM_DEBUGINFO_DWARF_DWARFRELOCMAP_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/Object/RelocationResolver.h"
+#include <cstdint>
+
+namespace llvm {
+
+/// RelocAddrEntry contains relocated value and section index.
+/// Section index is -1LL if relocation points to absolute symbol.
+struct RelocAddrEntry {
+ uint64_t SectionIndex;
+ object::RelocationRef Reloc;
+ uint64_t SymbolValue;
+ Optional<object::RelocationRef> Reloc2;
+ uint64_t SymbolValue2;
+ object::RelocationResolver Resolver;
+};
+
+/// In place of applying the relocations to the data we've read from disk we use
+/// a separate mapping table to the side and checking that at locations in the
+/// dwarf where we expect relocated values. This adds a bit of complexity to the
+/// dwarf parsing/extraction at the benefit of not allocating memory for the
+/// entire size of the debug info sections.
+using RelocAddrMap = DenseMap<uint64_t, RelocAddrEntry>;
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFRELOCMAP_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFSection.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFSection.h
new file mode 100644
index 000000000..054524d36
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFSection.h
@@ -0,0 +1,27 @@
+//===- DWARFSection.h -------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFSECTION_H
+#define LLVM_DEBUGINFO_DWARF_DWARFSECTION_H
+
+#include "llvm/ADT/StringRef.h"
+
+namespace llvm {
+
+struct DWARFSection {
+ StringRef Data;
+};
+
+struct SectionName {
+ StringRef Name;
+ bool IsNameUnique;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFSECTION_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFTypeUnit.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFTypeUnit.h
new file mode 100644
index 000000000..c95bdcbd8
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFTypeUnit.h
@@ -0,0 +1,46 @@
+//===- DWARFTypeUnit.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFTYPEUNIT_H
+#define LLVM_DEBUGINFO_DWARF_DWARFTYPEUNIT_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+
+namespace llvm {
+
+class DWARFContext;
+class DWARFDebugAbbrev;
+struct DWARFSection;
+class raw_ostream;
+
+class DWARFTypeUnit : public DWARFUnit {
+public:
+ DWARFTypeUnit(DWARFContext &Context, const DWARFSection &Section,
+ const DWARFUnitHeader &Header, const DWARFDebugAbbrev *DA,
+ const DWARFSection *RS, const DWARFSection *LocSection,
+ StringRef SS, const DWARFSection &SOS, const DWARFSection *AOS,
+ const DWARFSection &LS, bool LE, bool IsDWO,
+ const DWARFUnitVector &UnitVector)
+ : DWARFUnit(Context, Section, Header, DA, RS, LocSection, SS, SOS, AOS,
+ LS, LE, IsDWO, UnitVector) {}
+
+ uint64_t getTypeHash() const { return getHeader().getTypeHash(); }
+ uint64_t getTypeOffset() const { return getHeader().getTypeOffset(); }
+
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts = {}) override;
+ // Enable LLVM-style RTTI.
+ static bool classof(const DWARFUnit *U) { return U->isTypeUnit(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFTYPEUNIT_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnit.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnit.h
new file mode 100644
index 000000000..51de114a3
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnit.h
@@ -0,0 +1,527 @@
+//===- DWARFUnit.h ----------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFUNIT_H
+#define LLVM_DEBUGINFO_DWARF_DWARFUNIT_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugInfoEntry.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRangeList.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugRnglists.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
+#include "llvm/DebugInfo/DWARF/DWARFRelocMap.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/Support/DataExtractor.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <map>
+#include <memory>
+#include <utility>
+#include <vector>
+
+namespace llvm {
+
+class DWARFAbbreviationDeclarationSet;
+class DWARFContext;
+class DWARFDebugAbbrev;
+class DWARFUnit;
+
+/// Base class describing the header of any kind of "unit." Some information
+/// is specific to certain unit types. We separate this class out so we can
+/// parse the header before deciding what specific kind of unit to construct.
+class DWARFUnitHeader {
+ // Offset within section.
+ uint64_t Offset = 0;
+ // Version, address size, and DWARF format.
+ dwarf::FormParams FormParams;
+ uint64_t Length = 0;
+ uint64_t AbbrOffset = 0;
+
+ // For DWO units only.
+ const DWARFUnitIndex::Entry *IndexEntry = nullptr;
+
+ // For type units only.
+ uint64_t TypeHash = 0;
+ uint64_t TypeOffset = 0;
+
+ // For v5 split or skeleton compile units only.
+ Optional<uint64_t> DWOId;
+
+ // Unit type as parsed, or derived from the section kind.
+ uint8_t UnitType = 0;
+
+ // Size as parsed. uint8_t for compactness.
+ uint8_t Size = 0;
+
+public:
+ /// Parse a unit header from \p debug_info starting at \p offset_ptr.
+ bool extract(DWARFContext &Context, const DWARFDataExtractor &debug_info,
+ uint64_t *offset_ptr, DWARFSectionKind Kind = DW_SECT_INFO,
+ const DWARFUnitIndex *Index = nullptr,
+ const DWARFUnitIndex::Entry *Entry = nullptr);
+ uint64_t getOffset() const { return Offset; }
+ const dwarf::FormParams &getFormParams() const { return FormParams; }
+ uint16_t getVersion() const { return FormParams.Version; }
+ dwarf::DwarfFormat getFormat() const { return FormParams.Format; }
+ uint8_t getAddressByteSize() const { return FormParams.AddrSize; }
+ uint8_t getRefAddrByteSize() const { return FormParams.getRefAddrByteSize(); }
+ uint8_t getDwarfOffsetByteSize() const {
+ return FormParams.getDwarfOffsetByteSize();
+ }
+ uint64_t getLength() const { return Length; }
+ uint64_t getAbbrOffset() const { return AbbrOffset; }
+ Optional<uint64_t> getDWOId() const { return DWOId; }
+ void setDWOId(uint64_t Id) {
+ assert((!DWOId || *DWOId == Id) && "setting DWOId to a different value");
+ DWOId = Id;
+ }
+ const DWARFUnitIndex::Entry *getIndexEntry() const { return IndexEntry; }
+ uint64_t getTypeHash() const { return TypeHash; }
+ uint64_t getTypeOffset() const { return TypeOffset; }
+ uint8_t getUnitType() const { return UnitType; }
+ bool isTypeUnit() const {
+ return UnitType == dwarf::DW_UT_type || UnitType == dwarf::DW_UT_split_type;
+ }
+ uint8_t getSize() const { return Size; }
+ uint8_t getUnitLengthFieldByteSize() const {
+ return dwarf::getUnitLengthFieldByteSize(FormParams.Format);
+ }
+ uint64_t getNextUnitOffset() const {
+ return Offset + Length + getUnitLengthFieldByteSize();
+ }
+};
+
+const DWARFUnitIndex &getDWARFUnitIndex(DWARFContext &Context,
+ DWARFSectionKind Kind);
+
+/// Describe a collection of units. Intended to hold all units either from
+/// .debug_info and .debug_types, or from .debug_info.dwo and .debug_types.dwo.
+class DWARFUnitVector final : public SmallVector<std::unique_ptr<DWARFUnit>, 1> {
+ std::function<std::unique_ptr<DWARFUnit>(uint64_t, DWARFSectionKind,
+ const DWARFSection *,
+ const DWARFUnitIndex::Entry *)>
+ Parser;
+ int NumInfoUnits = -1;
+
+public:
+ using UnitVector = SmallVectorImpl<std::unique_ptr<DWARFUnit>>;
+ using iterator = typename UnitVector::iterator;
+ using iterator_range = llvm::iterator_range<typename UnitVector::iterator>;
+
+ DWARFUnit *getUnitForOffset(uint64_t Offset) const;
+ DWARFUnit *getUnitForIndexEntry(const DWARFUnitIndex::Entry &E);
+
+ /// Read units from a .debug_info or .debug_types section. Calls made
+ /// before finishedInfoUnits() are assumed to be for .debug_info sections,
+ /// calls after finishedInfoUnits() are for .debug_types sections. Caller
+ /// must not mix calls to addUnitsForSection and addUnitsForDWOSection.
+ void addUnitsForSection(DWARFContext &C, const DWARFSection &Section,
+ DWARFSectionKind SectionKind);
+ /// Read units from a .debug_info.dwo or .debug_types.dwo section. Calls
+ /// made before finishedInfoUnits() are assumed to be for .debug_info.dwo
+ /// sections, calls after finishedInfoUnits() are for .debug_types.dwo
+ /// sections. Caller must not mix calls to addUnitsForSection and
+ /// addUnitsForDWOSection.
+ void addUnitsForDWOSection(DWARFContext &C, const DWARFSection &DWOSection,
+ DWARFSectionKind SectionKind, bool Lazy = false);
+
+ /// Add an existing DWARFUnit to this UnitVector. This is used by the DWARF
+ /// verifier to process unit separately.
+ DWARFUnit *addUnit(std::unique_ptr<DWARFUnit> Unit);
+
+ /// Returns number of all units held by this instance.
+ unsigned getNumUnits() const { return size(); }
+ /// Returns number of units from all .debug_info[.dwo] sections.
+ unsigned getNumInfoUnits() const {
+ return NumInfoUnits == -1 ? size() : NumInfoUnits;
+ }
+ /// Returns number of units from all .debug_types[.dwo] sections.
+ unsigned getNumTypesUnits() const { return size() - NumInfoUnits; }
+ /// Indicate that parsing .debug_info[.dwo] is done, and remaining units
+ /// will be from .debug_types[.dwo].
+ void finishedInfoUnits() { NumInfoUnits = size(); }
+
+private:
+ void addUnitsImpl(DWARFContext &Context, const DWARFObject &Obj,
+ const DWARFSection &Section, const DWARFDebugAbbrev *DA,
+ const DWARFSection *RS, const DWARFSection *LocSection,
+ StringRef SS, const DWARFSection &SOS,
+ const DWARFSection *AOS, const DWARFSection &LS, bool LE,
+ bool IsDWO, bool Lazy, DWARFSectionKind SectionKind);
+};
+
+/// Represents base address of the CU.
+/// Represents a unit's contribution to the string offsets table.
+struct StrOffsetsContributionDescriptor {
+ uint64_t Base = 0;
+ /// The contribution size not including the header.
+ uint64_t Size = 0;
+ /// Format and version.
+ dwarf::FormParams FormParams = {0, 0, dwarf::DwarfFormat::DWARF32};
+
+ StrOffsetsContributionDescriptor(uint64_t Base, uint64_t Size,
+ uint8_t Version, dwarf::DwarfFormat Format)
+ : Base(Base), Size(Size), FormParams({Version, 0, Format}) {}
+ StrOffsetsContributionDescriptor() = default;
+
+ uint8_t getVersion() const { return FormParams.Version; }
+ dwarf::DwarfFormat getFormat() const { return FormParams.Format; }
+ uint8_t getDwarfOffsetByteSize() const {
+ return FormParams.getDwarfOffsetByteSize();
+ }
+ /// Determine whether a contribution to the string offsets table is
+ /// consistent with the relevant section size and that its length is
+ /// a multiple of the size of one of its entries.
+ Expected<StrOffsetsContributionDescriptor>
+ validateContributionSize(DWARFDataExtractor &DA);
+};
+
+class DWARFUnit {
+ DWARFContext &Context;
+ /// Section containing this DWARFUnit.
+ const DWARFSection &InfoSection;
+
+ DWARFUnitHeader Header;
+ const DWARFDebugAbbrev *Abbrev;
+ const DWARFSection *RangeSection;
+ uint64_t RangeSectionBase;
+ /// We either keep track of the location list section or its data, depending
+ /// on whether we are handling a split DWARF section or not.
+ union {
+ const DWARFSection *LocSection;
+ StringRef LocSectionData;
+ };
+ const DWARFSection &LineSection;
+ StringRef StringSection;
+ const DWARFSection &StringOffsetSection;
+ const DWARFSection *AddrOffsetSection;
+ uint32_t AddrOffsetSectionBase = 0;
+ bool isLittleEndian;
+ bool IsDWO;
+ const DWARFUnitVector &UnitVector;
+
+ /// Start, length, and DWARF format of the unit's contribution to the string
+ /// offsets table (DWARF v5).
+ Optional<StrOffsetsContributionDescriptor> StringOffsetsTableContribution;
+
+ /// A table of range lists (DWARF v5 and later).
+ Optional<DWARFDebugRnglistTable> RngListTable;
+
+ mutable const DWARFAbbreviationDeclarationSet *Abbrevs;
+ llvm::Optional<object::SectionedAddress> BaseAddr;
+ /// The compile unit debug information entry items.
+ std::vector<DWARFDebugInfoEntry> DieArray;
+
+ /// Map from range's start address to end address and corresponding DIE.
+ /// IntervalMap does not support range removal, as a result, we use the
+ /// std::map::upper_bound for address range lookup.
+ std::map<uint64_t, std::pair<uint64_t, DWARFDie>> AddrDieMap;
+
+ using die_iterator_range =
+ iterator_range<std::vector<DWARFDebugInfoEntry>::iterator>;
+
+ std::shared_ptr<DWARFUnit> DWO;
+
+ uint32_t getDIEIndex(const DWARFDebugInfoEntry *Die) {
+ auto First = DieArray.data();
+ assert(Die >= First && Die < First + DieArray.size());
+ return Die - First;
+ }
+
+protected:
+ const DWARFUnitHeader &getHeader() const { return Header; }
+
+ /// Size in bytes of the parsed unit header.
+ uint32_t getHeaderSize() const { return Header.getSize(); }
+
+ /// Find the unit's contribution to the string offsets table and determine its
+ /// length and form. The given offset is expected to be derived from the unit
+ /// DIE's DW_AT_str_offsets_base attribute.
+ Expected<Optional<StrOffsetsContributionDescriptor>>
+ determineStringOffsetsTableContribution(DWARFDataExtractor &DA);
+
+ /// Find the unit's contribution to the string offsets table and determine its
+ /// length and form. The given offset is expected to be 0 in a dwo file or,
+ /// in a dwp file, the start of the unit's contribution to the string offsets
+ /// table section (as determined by the index table).
+ Expected<Optional<StrOffsetsContributionDescriptor>>
+ determineStringOffsetsTableContributionDWO(DWARFDataExtractor &DA);
+
+public:
+ DWARFUnit(DWARFContext &Context, const DWARFSection &Section,
+ const DWARFUnitHeader &Header, const DWARFDebugAbbrev *DA,
+ const DWARFSection *RS, const DWARFSection *LocSection,
+ StringRef SS, const DWARFSection &SOS, const DWARFSection *AOS,
+ const DWARFSection &LS, bool LE, bool IsDWO,
+ const DWARFUnitVector &UnitVector);
+
+ virtual ~DWARFUnit();
+
+ bool isDWOUnit() const { return IsDWO; }
+ DWARFContext& getContext() const { return Context; }
+ const DWARFSection &getInfoSection() const { return InfoSection; }
+ const DWARFSection *getLocSection() const { return LocSection; }
+ StringRef getLocSectionData() const { return LocSectionData; }
+ uint64_t getOffset() const { return Header.getOffset(); }
+ const dwarf::FormParams &getFormParams() const {
+ return Header.getFormParams();
+ }
+ uint16_t getVersion() const { return Header.getVersion(); }
+ uint8_t getAddressByteSize() const { return Header.getAddressByteSize(); }
+ uint8_t getRefAddrByteSize() const { return Header.getRefAddrByteSize(); }
+ uint8_t getDwarfOffsetByteSize() const {
+ return Header.getDwarfOffsetByteSize();
+ }
+ uint64_t getLength() const { return Header.getLength(); }
+ uint8_t getUnitType() const { return Header.getUnitType(); }
+ bool isTypeUnit() const { return Header.isTypeUnit(); }
+ uint64_t getNextUnitOffset() const { return Header.getNextUnitOffset(); }
+ const DWARFSection &getLineSection() const { return LineSection; }
+ StringRef getStringSection() const { return StringSection; }
+ const DWARFSection &getStringOffsetSection() const {
+ return StringOffsetSection;
+ }
+
+ void setAddrOffsetSection(const DWARFSection *AOS, uint32_t Base) {
+ AddrOffsetSection = AOS;
+ AddrOffsetSectionBase = Base;
+ }
+
+ /// Recursively update address to Die map.
+ void updateAddressDieMap(DWARFDie Die);
+
+ void setRangesSection(const DWARFSection *RS, uint64_t Base) {
+ RangeSection = RS;
+ RangeSectionBase = Base;
+ }
+
+ Optional<object::SectionedAddress>
+ getAddrOffsetSectionItem(uint32_t Index) const;
+ Optional<uint64_t> getStringOffsetSectionItem(uint32_t Index) const;
+
+ DWARFDataExtractor getDebugInfoExtractor() const;
+
+ DataExtractor getStringExtractor() const {
+ return DataExtractor(StringSection, false, 0);
+ }
+
+ /// Extract the range list referenced by this compile unit from the
+ /// .debug_ranges section. If the extraction is unsuccessful, an error
+ /// is returned. Successful extraction requires that the compile unit
+ /// has already been extracted.
+ Error extractRangeList(uint64_t RangeListOffset,
+ DWARFDebugRangeList &RangeList) const;
+ void clear();
+
+ const Optional<StrOffsetsContributionDescriptor> &
+ getStringOffsetsTableContribution() const {
+ return StringOffsetsTableContribution;
+ }
+
+ uint8_t getDwarfStringOffsetsByteSize() const {
+ assert(StringOffsetsTableContribution);
+ return StringOffsetsTableContribution->getDwarfOffsetByteSize();
+ }
+
+ uint64_t getStringOffsetsBase() const {
+ assert(StringOffsetsTableContribution);
+ return StringOffsetsTableContribution->Base;
+ }
+
+ const DWARFAbbreviationDeclarationSet *getAbbreviations() const;
+
+ static bool isMatchingUnitTypeAndTag(uint8_t UnitType, dwarf::Tag Tag) {
+ switch (UnitType) {
+ case dwarf::DW_UT_compile:
+ return Tag == dwarf::DW_TAG_compile_unit;
+ case dwarf::DW_UT_type:
+ return Tag == dwarf::DW_TAG_type_unit;
+ case dwarf::DW_UT_partial:
+ return Tag == dwarf::DW_TAG_partial_unit;
+ case dwarf::DW_UT_skeleton:
+ return Tag == dwarf::DW_TAG_skeleton_unit;
+ case dwarf::DW_UT_split_compile:
+ case dwarf::DW_UT_split_type:
+ return dwarf::isUnitType(Tag);
+ }
+ return false;
+ }
+
+ /// Return the number of bytes for the header of a unit of
+ /// UnitType type.
+ ///
+ /// This function must be called with a valid unit type which in
+ /// DWARF5 is defined as one of the following six types.
+ static uint32_t getDWARF5HeaderSize(uint8_t UnitType) {
+ switch (UnitType) {
+ case dwarf::DW_UT_compile:
+ case dwarf::DW_UT_partial:
+ return 12;
+ case dwarf::DW_UT_skeleton:
+ case dwarf::DW_UT_split_compile:
+ return 20;
+ case dwarf::DW_UT_type:
+ case dwarf::DW_UT_split_type:
+ return 24;
+ }
+ llvm_unreachable("Invalid UnitType.");
+ }
+
+ llvm::Optional<object::SectionedAddress> getBaseAddress();
+
+ DWARFDie getUnitDIE(bool ExtractUnitDIEOnly = true) {
+ extractDIEsIfNeeded(ExtractUnitDIEOnly);
+ if (DieArray.empty())
+ return DWARFDie();
+ return DWARFDie(this, &DieArray[0]);
+ }
+
+ DWARFDie getNonSkeletonUnitDIE(bool ExtractUnitDIEOnly = true) {
+ parseDWO();
+ if (DWO)
+ return DWO->getUnitDIE(ExtractUnitDIEOnly);
+ return getUnitDIE(ExtractUnitDIEOnly);
+ }
+
+ const char *getCompilationDir();
+ Optional<uint64_t> getDWOId() {
+ extractDIEsIfNeeded(/*CUDieOnly*/ true);
+ return getHeader().getDWOId();
+ }
+ void setDWOId(uint64_t NewID) { Header.setDWOId(NewID); }
+
+ /// Return a vector of address ranges resulting from a (possibly encoded)
+ /// range list starting at a given offset in the appropriate ranges section.
+ Expected<DWARFAddressRangesVector> findRnglistFromOffset(uint64_t Offset);
+
+ /// Return a vector of address ranges retrieved from an encoded range
+ /// list whose offset is found via a table lookup given an index (DWARF v5
+ /// and later).
+ Expected<DWARFAddressRangesVector> findRnglistFromIndex(uint32_t Index);
+
+ /// Return a rangelist's offset based on an index. The index designates
+ /// an entry in the rangelist table's offset array and is supplied by
+ /// DW_FORM_rnglistx.
+ Optional<uint64_t> getRnglistOffset(uint32_t Index) {
+ if (RngListTable)
+ return RngListTable->getOffsetEntry(Index);
+ return None;
+ }
+
+ Expected<DWARFAddressRangesVector> collectAddressRanges();
+
+ /// Returns subprogram DIE with address range encompassing the provided
+ /// address. The pointer is alive as long as parsed compile unit DIEs are not
+ /// cleared.
+ DWARFDie getSubroutineForAddress(uint64_t Address);
+
+ /// getInlinedChainForAddress - fetches inlined chain for a given address.
+ /// Returns empty chain if there is no subprogram containing address. The
+ /// chain is valid as long as parsed compile unit DIEs are not cleared.
+ void getInlinedChainForAddress(uint64_t Address,
+ SmallVectorImpl<DWARFDie> &InlinedChain);
+
+ /// Return the DWARFUnitVector containing this unit.
+ const DWARFUnitVector &getUnitVector() const { return UnitVector; }
+
+ /// Returns the number of DIEs in the unit. Parses the unit
+ /// if necessary.
+ unsigned getNumDIEs() {
+ extractDIEsIfNeeded(false);
+ return DieArray.size();
+ }
+
+ /// Return the index of a DIE inside the unit's DIE vector.
+ ///
+ /// It is illegal to call this method with a DIE that hasn't be
+ /// created by this unit. In other word, it's illegal to call this
+ /// method on a DIE that isn't accessible by following
+ /// children/sibling links starting from this unit's getUnitDIE().
+ uint32_t getDIEIndex(const DWARFDie &D) {
+ return getDIEIndex(D.getDebugInfoEntry());
+ }
+
+ /// Return the DIE object at the given index.
+ DWARFDie getDIEAtIndex(unsigned Index) {
+ assert(Index < DieArray.size());
+ return DWARFDie(this, &DieArray[Index]);
+ }
+
+ DWARFDie getParent(const DWARFDebugInfoEntry *Die);
+ DWARFDie getSibling(const DWARFDebugInfoEntry *Die);
+ DWARFDie getPreviousSibling(const DWARFDebugInfoEntry *Die);
+ DWARFDie getFirstChild(const DWARFDebugInfoEntry *Die);
+ DWARFDie getLastChild(const DWARFDebugInfoEntry *Die);
+
+ /// Return the DIE object for a given offset inside the
+ /// unit's DIE vector.
+ ///
+ /// The unit needs to have its DIEs extracted for this method to work.
+ DWARFDie getDIEForOffset(uint64_t Offset) {
+ extractDIEsIfNeeded(false);
+ assert(!DieArray.empty());
+ auto It =
+ llvm::partition_point(DieArray, [=](const DWARFDebugInfoEntry &DIE) {
+ return DIE.getOffset() < Offset;
+ });
+ if (It != DieArray.end() && It->getOffset() == Offset)
+ return DWARFDie(this, &*It);
+ return DWARFDie();
+ }
+
+ uint32_t getLineTableOffset() const {
+ if (auto IndexEntry = Header.getIndexEntry())
+ if (const auto *Contrib = IndexEntry->getOffset(DW_SECT_LINE))
+ return Contrib->Offset;
+ return 0;
+ }
+
+ die_iterator_range dies() {
+ extractDIEsIfNeeded(false);
+ return die_iterator_range(DieArray.begin(), DieArray.end());
+ }
+
+ virtual void dump(raw_ostream &OS, DIDumpOptions DumpOpts) = 0;
+
+ Error tryExtractDIEsIfNeeded(bool CUDieOnly);
+
+private:
+ /// Size in bytes of the .debug_info data associated with this compile unit.
+ size_t getDebugInfoSize() const {
+ return Header.getLength() + Header.getUnitLengthFieldByteSize() -
+ getHeaderSize();
+ }
+
+ /// extractDIEsIfNeeded - Parses a compile unit and indexes its DIEs if it
+ /// hasn't already been done
+ void extractDIEsIfNeeded(bool CUDieOnly);
+
+ /// extractDIEsToVector - Appends all parsed DIEs to a vector.
+ void extractDIEsToVector(bool AppendCUDie, bool AppendNonCUDIEs,
+ std::vector<DWARFDebugInfoEntry> &DIEs) const;
+
+ /// clearDIEs - Clear parsed DIEs to keep memory usage low.
+ void clearDIEs(bool KeepCUDie);
+
+ /// parseDWO - Parses .dwo file for current compile unit. Returns true if
+ /// it was actually constructed.
+ bool parseDWO();
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFUNIT_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h
new file mode 100644
index 000000000..684103aac
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h
@@ -0,0 +1,105 @@
+//===- DWARFUnitIndex.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFUNITINDEX_H
+#define LLVM_DEBUGINFO_DWARF_DWARFUNITINDEX_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataExtractor.h"
+#include <cstdint>
+#include <memory>
+
+namespace llvm {
+
+class raw_ostream;
+
+enum DWARFSectionKind {
+ DW_SECT_INFO = 1,
+ DW_SECT_TYPES,
+ DW_SECT_ABBREV,
+ DW_SECT_LINE,
+ DW_SECT_LOC,
+ DW_SECT_STR_OFFSETS,
+ DW_SECT_MACINFO,
+ DW_SECT_MACRO,
+};
+
+class DWARFUnitIndex {
+ struct Header {
+ uint32_t Version;
+ uint32_t NumColumns;
+ uint32_t NumUnits;
+ uint32_t NumBuckets = 0;
+
+ bool parse(DataExtractor IndexData, uint64_t *OffsetPtr);
+ void dump(raw_ostream &OS) const;
+ };
+
+public:
+ class Entry {
+ public:
+ struct SectionContribution {
+ uint32_t Offset;
+ uint32_t Length;
+ };
+
+ private:
+ const DWARFUnitIndex *Index;
+ uint64_t Signature;
+ std::unique_ptr<SectionContribution[]> Contributions;
+ friend class DWARFUnitIndex;
+
+ public:
+ const SectionContribution *getOffset(DWARFSectionKind Sec) const;
+ const SectionContribution *getOffset() const;
+
+ const SectionContribution *getOffsets() const {
+ return Contributions.get();
+ }
+
+ uint64_t getSignature() const { return Signature; }
+ };
+
+private:
+ struct Header Header;
+
+ DWARFSectionKind InfoColumnKind;
+ int InfoColumn = -1;
+ std::unique_ptr<DWARFSectionKind[]> ColumnKinds;
+ std::unique_ptr<Entry[]> Rows;
+ mutable std::vector<Entry *> OffsetLookup;
+
+ static StringRef getColumnHeader(DWARFSectionKind DS);
+
+ bool parseImpl(DataExtractor IndexData);
+
+public:
+ DWARFUnitIndex(DWARFSectionKind InfoColumnKind)
+ : InfoColumnKind(InfoColumnKind) {}
+
+ explicit operator bool() const { return Header.NumBuckets; }
+
+ bool parse(DataExtractor IndexData);
+ void dump(raw_ostream &OS) const;
+
+ const Entry *getFromOffset(uint32_t Offset) const;
+ const Entry *getFromHash(uint64_t Offset) const;
+
+ ArrayRef<DWARFSectionKind> getColumnKinds() const {
+ return makeArrayRef(ColumnKinds.get(), Header.NumColumns);
+ }
+
+ ArrayRef<Entry> getRows() const {
+ return makeArrayRef(Rows.get(), Header.NumBuckets);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFUNITINDEX_H
diff --git a/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFVerifier.h b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
new file mode 100644
index 000000000..a4a3a11d4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
@@ -0,0 +1,336 @@
+//===- DWARFVerifier.h ----------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFVERIFIER_H
+#define LLVM_DEBUGINFO_DWARF_DWARFVERIFIER_H
+
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFAddressRange.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+
+#include <cstdint>
+#include <map>
+#include <set>
+
+namespace llvm {
+class raw_ostream;
+struct DWARFAttribute;
+class DWARFContext;
+class DWARFDie;
+class DWARFUnit;
+class DWARFCompileUnit;
+class DWARFDataExtractor;
+class DWARFDebugAbbrev;
+class DataExtractor;
+struct DWARFSection;
+
+/// A class that verifies DWARF debug information given a DWARF Context.
+class DWARFVerifier {
+public:
+ /// A class that keeps the address range information for a single DIE.
+ struct DieRangeInfo {
+ DWARFDie Die;
+
+ /// Sorted DWARFAddressRanges.
+ std::vector<DWARFAddressRange> Ranges;
+
+ /// Sorted DWARFAddressRangeInfo.
+ std::set<DieRangeInfo> Children;
+
+ DieRangeInfo() = default;
+ DieRangeInfo(DWARFDie Die) : Die(Die) {}
+
+ /// Used for unit testing.
+ DieRangeInfo(std::vector<DWARFAddressRange> Ranges)
+ : Ranges(std::move(Ranges)) {}
+
+ typedef std::vector<DWARFAddressRange>::const_iterator
+ address_range_iterator;
+ typedef std::set<DieRangeInfo>::const_iterator die_range_info_iterator;
+
+ /// Inserts the address range. If the range overlaps with an existing
+ /// range, the range is *not* added and an iterator to the overlapping
+ /// range is returned.
+ ///
+ /// This is used for finding overlapping ranges within the same DIE.
+ address_range_iterator insert(const DWARFAddressRange &R);
+
+ /// Finds an address range in the sorted vector of ranges.
+ address_range_iterator findRange(const DWARFAddressRange &R) const {
+ auto Begin = Ranges.begin();
+ auto End = Ranges.end();
+ auto Iter = std::upper_bound(Begin, End, R);
+ if (Iter != Begin)
+ --Iter;
+ return Iter;
+ }
+
+ /// Inserts the address range info. If any of its ranges overlaps with a
+ /// range in an existing range info, the range info is *not* added and an
+ /// iterator to the overlapping range info.
+ ///
+ /// This is used for finding overlapping children of the same DIE.
+ die_range_info_iterator insert(const DieRangeInfo &RI);
+
+ /// Return true if ranges in this object contains all ranges within RHS.
+ bool contains(const DieRangeInfo &RHS) const;
+
+ /// Return true if any range in this object intersects with any range in
+ /// RHS.
+ bool intersects(const DieRangeInfo &RHS) const;
+ };
+
+private:
+ raw_ostream &OS;
+ DWARFContext &DCtx;
+ DIDumpOptions DumpOpts;
+ /// A map that tracks all references (converted absolute references) so we
+ /// can verify each reference points to a valid DIE and not an offset that
+ /// lies between to valid DIEs.
+ std::map<uint64_t, std::set<uint64_t>> ReferenceToDIEOffsets;
+ uint32_t NumDebugLineErrors = 0;
+ // Used to relax some checks that do not currently work portably
+ bool IsObjectFile;
+ bool IsMachOObject;
+
+ raw_ostream &error() const;
+ raw_ostream &warn() const;
+ raw_ostream ¬e() const;
+ raw_ostream &dump(const DWARFDie &Die, unsigned indent = 0) const;
+
+ /// Verifies the abbreviations section.
+ ///
+ /// This function currently checks that:
+ /// --No abbreviation declaration has more than one attributes with the same
+ /// name.
+ ///
+ /// \param Abbrev Pointer to the abbreviations section we are verifying
+ /// Abbrev can be a pointer to either .debug_abbrev or debug_abbrev.dwo.
+ ///
+ /// \returns The number of errors that occurred during verification.
+ unsigned verifyAbbrevSection(const DWARFDebugAbbrev *Abbrev);
+
+ /// Verifies the header of a unit in a .debug_info or .debug_types section.
+ ///
+ /// This function currently checks for:
+ /// - Unit is in 32-bit DWARF format. The function can be modified to
+ /// support 64-bit format.
+ /// - The DWARF version is valid
+ /// - The unit type is valid (if unit is in version >=5)
+ /// - The unit doesn't extend beyond the containing section
+ /// - The address size is valid
+ /// - The offset in the .debug_abbrev section is valid
+ ///
+ /// \param DebugInfoData The section data
+ /// \param Offset A reference to the offset start of the unit. The offset will
+ /// be updated to point to the next unit in the section
+ /// \param UnitIndex The index of the unit to be verified
+ /// \param UnitType A reference to the type of the unit
+ /// \param isUnitDWARF64 A reference to a flag that shows whether the unit is
+ /// in 64-bit format.
+ ///
+ /// \returns true if the header is verified successfully, false otherwise.
+ bool verifyUnitHeader(const DWARFDataExtractor DebugInfoData,
+ uint64_t *Offset, unsigned UnitIndex, uint8_t &UnitType,
+ bool &isUnitDWARF64);
+
+ /// Verifies the header of a unit in a .debug_info or .debug_types section.
+ ///
+ /// This function currently verifies:
+ /// - The debug info attributes.
+ /// - The debug info form=s.
+ /// - The presence of a root DIE.
+ /// - That the root DIE is a unit DIE.
+ /// - If a unit type is provided, that the unit DIE matches the unit type.
+ /// - The DIE ranges.
+ /// - That call site entries are only nested within subprograms with a
+ /// DW_AT_call attribute.
+ ///
+ /// \param Unit The DWARF Unit to verify.
+ ///
+ /// \returns The number of errors that occurred during verification.
+ unsigned verifyUnitContents(DWARFUnit &Unit);
+
+ /// Verifies the unit headers and contents in a .debug_info or .debug_types
+ /// section.
+ ///
+ /// \param S The DWARF Section to verify.
+ /// \param SectionKind The object-file section kind that S comes from.
+ ///
+ /// \returns The number of errors that occurred during verification.
+ unsigned verifyUnitSection(const DWARFSection &S,
+ DWARFSectionKind SectionKind);
+
+ /// Verifies that a call site entry is nested within a subprogram with a
+ /// DW_AT_call attribute.
+ ///
+ /// \returns Number of errors that occurred during verification.
+ unsigned verifyDebugInfoCallSite(const DWARFDie &Die);
+
+ /// Verify that all Die ranges are valid.
+ ///
+ /// This function currently checks for:
+ /// - cases in which lowPC >= highPC
+ ///
+ /// \returns Number of errors that occurred during verification.
+ unsigned verifyDieRanges(const DWARFDie &Die, DieRangeInfo &ParentRI);
+
+ /// Verifies the attribute's DWARF attribute and its value.
+ ///
+ /// This function currently checks for:
+ /// - DW_AT_ranges values is a valid .debug_ranges offset
+ /// - DW_AT_stmt_list is a valid .debug_line offset
+ ///
+ /// \param Die The DWARF DIE that owns the attribute value
+ /// \param AttrValue The DWARF attribute value to check
+ ///
+ /// \returns NumErrors The number of errors occurred during verification of
+ /// attributes' values in a unit
+ unsigned verifyDebugInfoAttribute(const DWARFDie &Die,
+ DWARFAttribute &AttrValue);
+
+ /// Verifies the attribute's DWARF form.
+ ///
+ /// This function currently checks for:
+ /// - All DW_FORM_ref values that are CU relative have valid CU offsets
+ /// - All DW_FORM_ref_addr values have valid section offsets
+ /// - All DW_FORM_strp values have valid .debug_str offsets
+ ///
+ /// \param Die The DWARF DIE that owns the attribute value
+ /// \param AttrValue The DWARF attribute value to check
+ ///
+ /// \returns NumErrors The number of errors occurred during verification of
+ /// attributes' forms in a unit
+ unsigned verifyDebugInfoForm(const DWARFDie &Die, DWARFAttribute &AttrValue);
+
+ /// Verifies the all valid references that were found when iterating through
+ /// all of the DIE attributes.
+ ///
+ /// This function will verify that all references point to DIEs whose DIE
+ /// offset matches. This helps to ensure if a DWARF link phase moved things
+ /// around, that it doesn't create invalid references by failing to relocate
+ /// CU relative and absolute references.
+ ///
+ /// \returns NumErrors The number of errors occurred during verification of
+ /// references for the .debug_info and .debug_types sections
+ unsigned verifyDebugInfoReferences();
+
+ /// Verify the DW_AT_stmt_list encoding and value and ensure that no
+ /// compile units that have the same DW_AT_stmt_list value.
+ void verifyDebugLineStmtOffsets();
+
+ /// Verify that all of the rows in the line table are valid.
+ ///
+ /// This function currently checks for:
+ /// - addresses within a sequence that decrease in value
+ /// - invalid file indexes
+ void verifyDebugLineRows();
+
+ /// Verify that an Apple-style accelerator table is valid.
+ ///
+ /// This function currently checks that:
+ /// - The fixed part of the header fits in the section
+ /// - The size of the section is as large as what the header describes
+ /// - There is at least one atom
+ /// - The form for each atom is valid
+ /// - The tag for each DIE in the table is valid
+ /// - The buckets have a valid index, or they are empty
+ /// - Each hashdata offset is valid
+ /// - Each DIE is valid
+ ///
+ /// \param AccelSection pointer to the section containing the acceleration table
+ /// \param StrData pointer to the string section
+ /// \param SectionName the name of the table we're verifying
+ ///
+ /// \returns The number of errors occurred during verification
+ unsigned verifyAppleAccelTable(const DWARFSection *AccelSection,
+ DataExtractor *StrData,
+ const char *SectionName);
+
+ unsigned verifyDebugNamesCULists(const DWARFDebugNames &AccelTable);
+ unsigned verifyNameIndexBuckets(const DWARFDebugNames::NameIndex &NI,
+ const DataExtractor &StrData);
+ unsigned verifyNameIndexAbbrevs(const DWARFDebugNames::NameIndex &NI);
+ unsigned verifyNameIndexAttribute(const DWARFDebugNames::NameIndex &NI,
+ const DWARFDebugNames::Abbrev &Abbr,
+ DWARFDebugNames::AttributeEncoding AttrEnc);
+ unsigned verifyNameIndexEntries(const DWARFDebugNames::NameIndex &NI,
+ const DWARFDebugNames::NameTableEntry &NTE);
+ unsigned verifyNameIndexCompleteness(const DWARFDie &Die,
+ const DWARFDebugNames::NameIndex &NI);
+
+ /// Verify that the DWARF v5 accelerator table is valid.
+ ///
+ /// This function currently checks that:
+ /// - Headers individual Name Indices fit into the section and can be parsed.
+ /// - Abbreviation tables can be parsed and contain valid index attributes
+ /// with correct form encodings.
+ /// - The CU lists reference existing compile units.
+ /// - The buckets have a valid index, or they are empty.
+ /// - All names are reachable via the hash table (they have the correct hash,
+ /// and the hash is in the correct bucket).
+ /// - Information in the index entries is complete (all required entries are
+ /// present) and consistent with the debug_info section DIEs.
+ ///
+ /// \param AccelSection section containing the acceleration table
+ /// \param StrData string section
+ ///
+ /// \returns The number of errors occurred during verification
+ unsigned verifyDebugNames(const DWARFSection &AccelSection,
+ const DataExtractor &StrData);
+
+public:
+ DWARFVerifier(raw_ostream &S, DWARFContext &D,
+ DIDumpOptions DumpOpts = DIDumpOptions::getForSingleDIE());
+
+ /// Verify the information in any of the following sections, if available:
+ /// .debug_abbrev, debug_abbrev.dwo
+ ///
+ /// Any errors are reported to the stream that was this object was
+ /// constructed with.
+ ///
+ /// \returns true if .debug_abbrev and .debug_abbrev.dwo verify successfully,
+ /// false otherwise.
+ bool handleDebugAbbrev();
+
+ /// Verify the information in the .debug_info and .debug_types sections.
+ ///
+ /// Any errors are reported to the stream that this object was
+ /// constructed with.
+ ///
+ /// \returns true if all sections verify successfully, false otherwise.
+ bool handleDebugInfo();
+
+ /// Verify the information in the .debug_line section.
+ ///
+ /// Any errors are reported to the stream that was this object was
+ /// constructed with.
+ ///
+ /// \returns true if the .debug_line verifies successfully, false otherwise.
+ bool handleDebugLine();
+
+ /// Verify the information in accelerator tables, if they exist.
+ ///
+ /// Any errors are reported to the stream that was this object was
+ /// constructed with.
+ ///
+ /// \returns true if the existing Apple-style accelerator tables verify
+ /// successfully, false otherwise.
+ bool handleAccelTables();
+};
+
+static inline bool operator<(const DWARFVerifier::DieRangeInfo &LHS,
+ const DWARFVerifier::DieRangeInfo &RHS) {
+ return std::tie(LHS.Ranges, LHS.Die) < std::tie(RHS.Ranges, RHS.Die);
+}
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
diff --git a/third_party/llvm-project/include/llvm/LICENSE.TXT b/third_party/llvm-project/include/llvm/LICENSE.TXT
new file mode 100644
index 000000000..fa6ac5400
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/LICENSE.TXT
@@ -0,0 +1,279 @@
+==============================================================================
+The LLVM Project is under the Apache License v2.0 with LLVM Exceptions:
+==============================================================================
+
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
+
+ END OF TERMS AND CONDITIONS
+
+ APPENDIX: How to apply the Apache License to your work.
+
+ To apply the Apache License to your work, attach the following
+ boilerplate notice, with the fields enclosed by brackets "[]"
+ replaced with your own identifying information. (Don't include
+ the brackets!) The text should be enclosed in the appropriate
+ comment syntax for the file format. We also recommend that a
+ file or class name and description of purpose be included on the
+ same "printed page" as the copyright notice for easier
+ identification within third-party archives.
+
+ Copyright [yyyy] [name of copyright owner]
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
+
+
+---- LLVM Exceptions to the Apache 2.0 License ----
+
+As an exception, if, as a result of your compiling your source code, portions
+of this Software are embedded into an Object form of such source code, you
+may redistribute such embedded portions in such Object form without complying
+with the conditions of Sections 4(a), 4(b) and 4(d) of the License.
+
+In addition, if you combine or link compiled forms of this Software with
+software that is licensed under the GPLv2 ("Combined Software") and if a
+court of competent jurisdiction determines that the patent provision (Section
+3), the indemnity provision (Section 9) or other Section of the License
+conflicts with the conditions of the GPLv2, you may retroactively and
+prospectively choose to deem waived or otherwise exclude such Section(s) of
+the License, but only in their entirety and only with respect to the Combined
+Software.
+
+==============================================================================
+Software from third parties included in the LLVM Project:
+==============================================================================
+The LLVM Project contains third party software which is under different license
+terms. All such code will be identified clearly using at least one of two
+mechanisms:
+1) It will be in a separate directory tree with its own `LICENSE.txt` or
+ `LICENSE` file at the top containing the specific license and restrictions
+ which apply to that software, or
+2) It will contain specific license and restriction terms at the top of every
+ file.
+
+==============================================================================
+Legacy LLVM License (https://llvm.org/docs/DeveloperPolicy.html#legacy):
+==============================================================================
+University of Illinois/NCSA
+Open Source License
+
+Copyright (c) 2003-2019 University of Illinois at Urbana-Champaign.
+All rights reserved.
+
+Developed by:
+
+ LLVM Team
+
+ University of Illinois at Urbana-Champaign
+
+ http://llvm.org
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal with
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+
+ * Redistributions of source code must retain the above copyright notice,
+ this list of conditions and the following disclaimers.
+
+ * Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimers in the
+ documentation and/or other materials provided with the distribution.
+
+ * Neither the names of the LLVM Team, University of Illinois at
+ Urbana-Champaign, nor the names of its contributors may be used to
+ endorse or promote products derived from this Software without specific
+ prior written permission.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+CONTRIBUTORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS WITH THE
+SOFTWARE.
+
diff --git a/third_party/llvm-project/include/llvm/MC/LaneBitmask.h b/third_party/llvm-project/include/llvm/MC/LaneBitmask.h
new file mode 100644
index 000000000..d5f69287a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/LaneBitmask.h
@@ -0,0 +1,101 @@
+//===- llvm/MC/LaneBitmask.h ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// A common definition of LaneBitmask for use in TableGen and CodeGen.
+///
+/// A lane mask is a bitmask representing the covering of a register with
+/// sub-registers.
+///
+/// This is typically used to track liveness at sub-register granularity.
+/// Lane masks for sub-register indices are similar to register units for
+/// physical registers. The individual bits in a lane mask can't be assigned
+/// any specific meaning. They can be used to check if two sub-register
+/// indices overlap.
+///
+/// Iff the target has a register such that:
+///
+/// getSubReg(Reg, A) overlaps getSubReg(Reg, B)
+///
+/// then:
+///
+/// (getSubRegIndexLaneMask(A) & getSubRegIndexLaneMask(B)) != 0
+
+#ifndef LLVM_MC_LANEBITMASK_H
+#define LLVM_MC_LANEBITMASK_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/Printable.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+
+ struct LaneBitmask {
+ // When changing the underlying type, change the format string as well.
+ using Type = unsigned;
+ enum : unsigned { BitWidth = 8*sizeof(Type) };
+ constexpr static const char *const FormatStr = "%08X";
+
+ constexpr LaneBitmask() = default;
+ explicit constexpr LaneBitmask(Type V) : Mask(V) {}
+
+ constexpr bool operator== (LaneBitmask M) const { return Mask == M.Mask; }
+ constexpr bool operator!= (LaneBitmask M) const { return Mask != M.Mask; }
+ constexpr bool operator< (LaneBitmask M) const { return Mask < M.Mask; }
+ constexpr bool none() const { return Mask == 0; }
+ constexpr bool any() const { return Mask != 0; }
+ constexpr bool all() const { return ~Mask == 0; }
+
+ constexpr LaneBitmask operator~() const {
+ return LaneBitmask(~Mask);
+ }
+ constexpr LaneBitmask operator|(LaneBitmask M) const {
+ return LaneBitmask(Mask | M.Mask);
+ }
+ constexpr LaneBitmask operator&(LaneBitmask M) const {
+ return LaneBitmask(Mask & M.Mask);
+ }
+ LaneBitmask &operator|=(LaneBitmask M) {
+ Mask |= M.Mask;
+ return *this;
+ }
+ LaneBitmask &operator&=(LaneBitmask M) {
+ Mask &= M.Mask;
+ return *this;
+ }
+
+ constexpr Type getAsInteger() const { return Mask; }
+
+ unsigned getNumLanes() const {
+ return countPopulation(Mask);
+ }
+ unsigned getHighestLane() const {
+ return Log2_32(Mask);
+ }
+
+ static constexpr LaneBitmask getNone() { return LaneBitmask(0); }
+ static constexpr LaneBitmask getAll() { return ~LaneBitmask(0); }
+ static constexpr LaneBitmask getLane(unsigned Lane) {
+ return LaneBitmask(Type(1) << Lane);
+ }
+
+ private:
+ Type Mask = 0;
+ };
+
+ /// Create Printable object to print LaneBitmasks on a \ref raw_ostream.
+ inline Printable PrintLaneMask(LaneBitmask LaneMask) {
+ return Printable([LaneMask](raw_ostream &OS) {
+ OS << format(LaneBitmask::FormatStr, LaneMask.getAsInteger());
+ });
+ }
+
+} // end namespace llvm
+
+#endif // LLVM_MC_LANEBITMASK_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCExpr.h b/third_party/llvm-project/include/llvm/MC/MCExpr.h
new file mode 100644
index 000000000..eb2786501
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCExpr.h
@@ -0,0 +1,616 @@
+//===- MCExpr.h - Assembly Level Expressions --------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCEXPR_H
+#define LLVM_MC_MCEXPR_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/Support/SMLoc.h"
+#include <cstdint>
+
+namespace llvm {
+
+class MCAsmInfo;
+class MCAsmLayout;
+class MCAssembler;
+class MCContext;
+class MCFixup;
+class MCFragment;
+class MCSection;
+class MCStreamer;
+class MCSymbol;
+class MCValue;
+class raw_ostream;
+class StringRef;
+
+using SectionAddrMap = DenseMap<const MCSection *, uint64_t>;
+
+/// Base class for the full range of assembler expressions which are
+/// needed for parsing.
+class MCExpr {
+public:
+ enum ExprKind {
+ Binary, ///< Binary expressions.
+ Constant, ///< Constant expressions.
+ SymbolRef, ///< References to labels and assigned expressions.
+ Unary, ///< Unary expressions.
+ Target ///< Target specific expression.
+ };
+
+private:
+ ExprKind Kind;
+ SMLoc Loc;
+
+ bool evaluateAsAbsolute(int64_t &Res, const MCAssembler *Asm,
+ const MCAsmLayout *Layout,
+ const SectionAddrMap *Addrs, bool InSet) const;
+
+protected:
+ explicit MCExpr(ExprKind Kind, SMLoc Loc) : Kind(Kind), Loc(Loc) {}
+
+ bool evaluateAsRelocatableImpl(MCValue &Res, const MCAssembler *Asm,
+ const MCAsmLayout *Layout,
+ const MCFixup *Fixup,
+ const SectionAddrMap *Addrs, bool InSet) const;
+
+public:
+ MCExpr(const MCExpr &) = delete;
+ MCExpr &operator=(const MCExpr &) = delete;
+
+ /// \name Accessors
+ /// @{
+
+ ExprKind getKind() const { return Kind; }
+ SMLoc getLoc() const { return Loc; }
+
+ /// @}
+ /// \name Utility Methods
+ /// @{
+
+ void print(raw_ostream &OS, const MCAsmInfo *MAI,
+ bool InParens = false) const;
+ void dump() const;
+
+ /// @}
+ /// \name Expression Evaluation
+ /// @{
+
+ /// Try to evaluate the expression to an absolute value.
+ ///
+ /// \param Res - The absolute value, if evaluation succeeds.
+ /// \param Layout - The assembler layout object to use for evaluating symbol
+ /// values. If not given, then only non-symbolic expressions will be
+ /// evaluated.
+ /// \return - True on success.
+ bool evaluateAsAbsolute(int64_t &Res, const MCAsmLayout &Layout,
+ const SectionAddrMap &Addrs) const;
+ bool evaluateAsAbsolute(int64_t &Res) const;
+ bool evaluateAsAbsolute(int64_t &Res, const MCAssembler &Asm) const;
+ bool evaluateAsAbsolute(int64_t &Res, const MCAssembler *Asm) const;
+ bool evaluateAsAbsolute(int64_t &Res, const MCAsmLayout &Layout) const;
+
+ bool evaluateKnownAbsolute(int64_t &Res, const MCAsmLayout &Layout) const;
+
+ /// Try to evaluate the expression to a relocatable value, i.e. an
+ /// expression of the fixed form (a - b + constant).
+ ///
+ /// \param Res - The relocatable value, if evaluation succeeds.
+ /// \param Layout - The assembler layout object to use for evaluating values.
+ /// \param Fixup - The Fixup object if available.
+ /// \return - True on success.
+ bool evaluateAsRelocatable(MCValue &Res, const MCAsmLayout *Layout,
+ const MCFixup *Fixup) const;
+
+ /// Try to evaluate the expression to the form (a - b + constant) where
+ /// neither a nor b are variables.
+ ///
+ /// This is a more aggressive variant of evaluateAsRelocatable. The intended
+ /// use is for when relocations are not available, like the .size directive.
+ bool evaluateAsValue(MCValue &Res, const MCAsmLayout &Layout) const;
+
+ /// Find the "associated section" for this expression, which is
+ /// currently defined as the absolute section for constants, or
+ /// otherwise the section associated with the first defined symbol in the
+ /// expression.
+ MCFragment *findAssociatedFragment() const;
+
+ /// @}
+};
+
+inline raw_ostream &operator<<(raw_ostream &OS, const MCExpr &E) {
+ E.print(OS, nullptr);
+ return OS;
+}
+
+//// Represent a constant integer expression.
+class MCConstantExpr : public MCExpr {
+ int64_t Value;
+ bool PrintInHex = false;
+
+ explicit MCConstantExpr(int64_t Value)
+ : MCExpr(MCExpr::Constant, SMLoc()), Value(Value) {}
+
+ MCConstantExpr(int64_t Value, bool PrintInHex)
+ : MCExpr(MCExpr::Constant, SMLoc()), Value(Value),
+ PrintInHex(PrintInHex) {}
+
+public:
+ /// \name Construction
+ /// @{
+
+ static const MCConstantExpr *create(int64_t Value, MCContext &Ctx,
+ bool PrintInHex = false);
+
+ /// @}
+ /// \name Accessors
+ /// @{
+
+ int64_t getValue() const { return Value; }
+
+ bool useHexFormat() const { return PrintInHex; }
+
+ /// @}
+
+ static bool classof(const MCExpr *E) {
+ return E->getKind() == MCExpr::Constant;
+ }
+};
+
+/// Represent a reference to a symbol from inside an expression.
+///
+/// A symbol reference in an expression may be a use of a label, a use of an
+/// assembler variable (defined constant), or constitute an implicit definition
+/// of the symbol as external.
+class MCSymbolRefExpr : public MCExpr {
+public:
+ enum VariantKind : uint16_t {
+ VK_None,
+ VK_Invalid,
+
+ VK_GOT,
+ VK_GOTOFF,
+ VK_GOTREL,
+ VK_GOTPCREL,
+ VK_GOTTPOFF,
+ VK_INDNTPOFF,
+ VK_NTPOFF,
+ VK_GOTNTPOFF,
+ VK_PLT,
+ VK_TLSGD,
+ VK_TLSLD,
+ VK_TLSLDM,
+ VK_TPOFF,
+ VK_DTPOFF,
+ VK_TLSCALL, // symbol(tlscall)
+ VK_TLSDESC, // symbol(tlsdesc)
+ VK_TLVP, // Mach-O thread local variable relocations
+ VK_TLVPPAGE,
+ VK_TLVPPAGEOFF,
+ VK_PAGE,
+ VK_PAGEOFF,
+ VK_GOTPAGE,
+ VK_GOTPAGEOFF,
+ VK_SECREL,
+ VK_SIZE, // symbol@SIZE
+ VK_WEAKREF, // The link between the symbols in .weakref foo, bar
+
+ VK_X86_ABS8,
+
+ VK_ARM_NONE,
+ VK_ARM_GOT_PREL,
+ VK_ARM_TARGET1,
+ VK_ARM_TARGET2,
+ VK_ARM_PREL31,
+ VK_ARM_SBREL, // symbol(sbrel)
+ VK_ARM_TLSLDO, // symbol(tlsldo)
+ VK_ARM_TLSDESCSEQ,
+
+ VK_AVR_NONE,
+ VK_AVR_LO8,
+ VK_AVR_HI8,
+ VK_AVR_HLO8,
+ VK_AVR_DIFF8,
+ VK_AVR_DIFF16,
+ VK_AVR_DIFF32,
+
+ VK_PPC_LO, // symbol@l
+ VK_PPC_HI, // symbol@h
+ VK_PPC_HA, // symbol@ha
+ VK_PPC_HIGH, // symbol@high
+ VK_PPC_HIGHA, // symbol@higha
+ VK_PPC_HIGHER, // symbol@higher
+ VK_PPC_HIGHERA, // symbol@highera
+ VK_PPC_HIGHEST, // symbol@highest
+ VK_PPC_HIGHESTA, // symbol@highesta
+ VK_PPC_GOT_LO, // symbol@got@l
+ VK_PPC_GOT_HI, // symbol@got@h
+ VK_PPC_GOT_HA, // symbol@got@ha
+ VK_PPC_TOCBASE, // symbol@tocbase
+ VK_PPC_TOC, // symbol@toc
+ VK_PPC_TOC_LO, // symbol@toc@l
+ VK_PPC_TOC_HI, // symbol@toc@h
+ VK_PPC_TOC_HA, // symbol@toc@ha
+ VK_PPC_U, // symbol@u
+ VK_PPC_L, // symbol@l
+ VK_PPC_DTPMOD, // symbol@dtpmod
+ VK_PPC_TPREL_LO, // symbol@tprel@l
+ VK_PPC_TPREL_HI, // symbol@tprel@h
+ VK_PPC_TPREL_HA, // symbol@tprel@ha
+ VK_PPC_TPREL_HIGH, // symbol@tprel@high
+ VK_PPC_TPREL_HIGHA, // symbol@tprel@higha
+ VK_PPC_TPREL_HIGHER, // symbol@tprel@higher
+ VK_PPC_TPREL_HIGHERA, // symbol@tprel@highera
+ VK_PPC_TPREL_HIGHEST, // symbol@tprel@highest
+ VK_PPC_TPREL_HIGHESTA, // symbol@tprel@highesta
+ VK_PPC_DTPREL_LO, // symbol@dtprel@l
+ VK_PPC_DTPREL_HI, // symbol@dtprel@h
+ VK_PPC_DTPREL_HA, // symbol@dtprel@ha
+ VK_PPC_DTPREL_HIGH, // symbol@dtprel@high
+ VK_PPC_DTPREL_HIGHA, // symbol@dtprel@higha
+ VK_PPC_DTPREL_HIGHER, // symbol@dtprel@higher
+ VK_PPC_DTPREL_HIGHERA, // symbol@dtprel@highera
+ VK_PPC_DTPREL_HIGHEST, // symbol@dtprel@highest
+ VK_PPC_DTPREL_HIGHESTA,// symbol@dtprel@highesta
+ VK_PPC_GOT_TPREL, // symbol@got@tprel
+ VK_PPC_GOT_TPREL_LO, // symbol@got@tprel@l
+ VK_PPC_GOT_TPREL_HI, // symbol@got@tprel@h
+ VK_PPC_GOT_TPREL_HA, // symbol@got@tprel@ha
+ VK_PPC_GOT_DTPREL, // symbol@got@dtprel
+ VK_PPC_GOT_DTPREL_LO, // symbol@got@dtprel@l
+ VK_PPC_GOT_DTPREL_HI, // symbol@got@dtprel@h
+ VK_PPC_GOT_DTPREL_HA, // symbol@got@dtprel@ha
+ VK_PPC_TLS, // symbol@tls
+ VK_PPC_GOT_TLSGD, // symbol@got@tlsgd
+ VK_PPC_GOT_TLSGD_LO, // symbol@got@tlsgd@l
+ VK_PPC_GOT_TLSGD_HI, // symbol@got@tlsgd@h
+ VK_PPC_GOT_TLSGD_HA, // symbol@got@tlsgd@ha
+ VK_PPC_TLSGD, // symbol@tlsgd
+ VK_PPC_GOT_TLSLD, // symbol@got@tlsld
+ VK_PPC_GOT_TLSLD_LO, // symbol@got@tlsld@l
+ VK_PPC_GOT_TLSLD_HI, // symbol@got@tlsld@h
+ VK_PPC_GOT_TLSLD_HA, // symbol@got@tlsld@ha
+ VK_PPC_TLSLD, // symbol@tlsld
+ VK_PPC_LOCAL, // symbol@local
+
+ VK_COFF_IMGREL32, // symbol@imgrel (image-relative)
+
+ VK_Hexagon_PCREL,
+ VK_Hexagon_LO16,
+ VK_Hexagon_HI16,
+ VK_Hexagon_GPREL,
+ VK_Hexagon_GD_GOT,
+ VK_Hexagon_LD_GOT,
+ VK_Hexagon_GD_PLT,
+ VK_Hexagon_LD_PLT,
+ VK_Hexagon_IE,
+ VK_Hexagon_IE_GOT,
+
+ VK_WASM_TYPEINDEX, // Reference to a symbol's type (signature)
+ VK_WASM_MBREL, // Memory address relative to memory base
+ VK_WASM_TBREL, // Table index relative to table bare
+
+ VK_AMDGPU_GOTPCREL32_LO, // symbol@gotpcrel32@lo
+ VK_AMDGPU_GOTPCREL32_HI, // symbol@gotpcrel32@hi
+ VK_AMDGPU_REL32_LO, // symbol@rel32@lo
+ VK_AMDGPU_REL32_HI, // symbol@rel32@hi
+ VK_AMDGPU_REL64, // symbol@rel64
+ VK_AMDGPU_ABS32_LO, // symbol@abs32@lo
+ VK_AMDGPU_ABS32_HI, // symbol@abs32@hi
+
+ VK_TPREL,
+ VK_DTPREL
+ };
+
+private:
+ /// The symbol reference modifier.
+ const VariantKind Kind;
+
+ /// Specifies how the variant kind should be printed.
+ const unsigned UseParensForSymbolVariant : 1;
+
+ // FIXME: Remove this bit.
+ const unsigned HasSubsectionsViaSymbols : 1;
+
+ /// The symbol being referenced.
+ const MCSymbol *Symbol;
+
+ explicit MCSymbolRefExpr(const MCSymbol *Symbol, VariantKind Kind,
+ const MCAsmInfo *MAI, SMLoc Loc = SMLoc());
+
+public:
+ /// \name Construction
+ /// @{
+
+ static const MCSymbolRefExpr *create(const MCSymbol *Symbol, MCContext &Ctx) {
+ return MCSymbolRefExpr::create(Symbol, VK_None, Ctx);
+ }
+
+ static const MCSymbolRefExpr *create(const MCSymbol *Symbol, VariantKind Kind,
+ MCContext &Ctx, SMLoc Loc = SMLoc());
+ static const MCSymbolRefExpr *create(StringRef Name, VariantKind Kind,
+ MCContext &Ctx);
+
+ /// @}
+ /// \name Accessors
+ /// @{
+
+ const MCSymbol &getSymbol() const { return *Symbol; }
+
+ VariantKind getKind() const { return Kind; }
+
+ void printVariantKind(raw_ostream &OS) const;
+
+ bool hasSubsectionsViaSymbols() const { return HasSubsectionsViaSymbols; }
+
+ /// @}
+ /// \name Static Utility Functions
+ /// @{
+
+ static StringRef getVariantKindName(VariantKind Kind);
+
+ static VariantKind getVariantKindForName(StringRef Name);
+
+ /// @}
+
+ static bool classof(const MCExpr *E) {
+ return E->getKind() == MCExpr::SymbolRef;
+ }
+};
+
+/// Unary assembler expressions.
+class MCUnaryExpr : public MCExpr {
+public:
+ enum Opcode {
+ LNot, ///< Logical negation.
+ Minus, ///< Unary minus.
+ Not, ///< Bitwise negation.
+ Plus ///< Unary plus.
+ };
+
+private:
+ Opcode Op;
+ const MCExpr *Expr;
+
+ MCUnaryExpr(Opcode Op, const MCExpr *Expr, SMLoc Loc)
+ : MCExpr(MCExpr::Unary, Loc), Op(Op), Expr(Expr) {}
+
+public:
+ /// \name Construction
+ /// @{
+
+ static const MCUnaryExpr *create(Opcode Op, const MCExpr *Expr,
+ MCContext &Ctx, SMLoc Loc = SMLoc());
+
+ static const MCUnaryExpr *createLNot(const MCExpr *Expr, MCContext &Ctx, SMLoc Loc = SMLoc()) {
+ return create(LNot, Expr, Ctx, Loc);
+ }
+
+ static const MCUnaryExpr *createMinus(const MCExpr *Expr, MCContext &Ctx, SMLoc Loc = SMLoc()) {
+ return create(Minus, Expr, Ctx, Loc);
+ }
+
+ static const MCUnaryExpr *createNot(const MCExpr *Expr, MCContext &Ctx, SMLoc Loc = SMLoc()) {
+ return create(Not, Expr, Ctx, Loc);
+ }
+
+ static const MCUnaryExpr *createPlus(const MCExpr *Expr, MCContext &Ctx, SMLoc Loc = SMLoc()) {
+ return create(Plus, Expr, Ctx, Loc);
+ }
+
+ /// @}
+ /// \name Accessors
+ /// @{
+
+ /// Get the kind of this unary expression.
+ Opcode getOpcode() const { return Op; }
+
+ /// Get the child of this unary expression.
+ const MCExpr *getSubExpr() const { return Expr; }
+
+ /// @}
+
+ static bool classof(const MCExpr *E) {
+ return E->getKind() == MCExpr::Unary;
+ }
+};
+
+/// Binary assembler expressions.
+class MCBinaryExpr : public MCExpr {
+public:
+ enum Opcode {
+ Add, ///< Addition.
+ And, ///< Bitwise and.
+ Div, ///< Signed division.
+ EQ, ///< Equality comparison.
+ GT, ///< Signed greater than comparison (result is either 0 or some
+ ///< target-specific non-zero value)
+ GTE, ///< Signed greater than or equal comparison (result is either 0 or
+ ///< some target-specific non-zero value).
+ LAnd, ///< Logical and.
+ LOr, ///< Logical or.
+ LT, ///< Signed less than comparison (result is either 0 or
+ ///< some target-specific non-zero value).
+ LTE, ///< Signed less than or equal comparison (result is either 0 or
+ ///< some target-specific non-zero value).
+ Mod, ///< Signed remainder.
+ Mul, ///< Multiplication.
+ NE, ///< Inequality comparison.
+ Or, ///< Bitwise or.
+ Shl, ///< Shift left.
+ AShr, ///< Arithmetic shift right.
+ LShr, ///< Logical shift right.
+ Sub, ///< Subtraction.
+ Xor ///< Bitwise exclusive or.
+ };
+
+private:
+ Opcode Op;
+ const MCExpr *LHS, *RHS;
+
+ MCBinaryExpr(Opcode Op, const MCExpr *LHS, const MCExpr *RHS,
+ SMLoc Loc = SMLoc())
+ : MCExpr(MCExpr::Binary, Loc), Op(Op), LHS(LHS), RHS(RHS) {}
+
+public:
+ /// \name Construction
+ /// @{
+
+ static const MCBinaryExpr *create(Opcode Op, const MCExpr *LHS,
+ const MCExpr *RHS, MCContext &Ctx,
+ SMLoc Loc = SMLoc());
+
+ static const MCBinaryExpr *createAdd(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Add, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createAnd(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(And, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createDiv(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Div, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createEQ(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(EQ, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createGT(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(GT, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createGTE(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(GTE, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createLAnd(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(LAnd, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createLOr(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(LOr, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createLT(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(LT, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createLTE(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(LTE, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createMod(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Mod, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createMul(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Mul, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createNE(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(NE, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createOr(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Or, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createShl(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Shl, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createAShr(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(AShr, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createLShr(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(LShr, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createSub(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Sub, LHS, RHS, Ctx);
+ }
+
+ static const MCBinaryExpr *createXor(const MCExpr *LHS, const MCExpr *RHS,
+ MCContext &Ctx) {
+ return create(Xor, LHS, RHS, Ctx);
+ }
+
+ /// @}
+ /// \name Accessors
+ /// @{
+
+ /// Get the kind of this binary expression.
+ Opcode getOpcode() const { return Op; }
+
+ /// Get the left-hand side expression of the binary operator.
+ const MCExpr *getLHS() const { return LHS; }
+
+ /// Get the right-hand side expression of the binary operator.
+ const MCExpr *getRHS() const { return RHS; }
+
+ /// @}
+
+ static bool classof(const MCExpr *E) {
+ return E->getKind() == MCExpr::Binary;
+ }
+};
+
+/// This is an extension point for target-specific MCExpr subclasses to
+/// implement.
+///
+/// NOTE: All subclasses are required to have trivial destructors because
+/// MCExprs are bump pointer allocated and not destructed.
+class MCTargetExpr : public MCExpr {
+ virtual void anchor();
+
+protected:
+ MCTargetExpr() : MCExpr(Target, SMLoc()) {}
+ virtual ~MCTargetExpr() = default;
+
+public:
+ virtual void printImpl(raw_ostream &OS, const MCAsmInfo *MAI) const = 0;
+ virtual bool evaluateAsRelocatableImpl(MCValue &Res,
+ const MCAsmLayout *Layout,
+ const MCFixup *Fixup) const = 0;
+ // allow Target Expressions to be checked for equality
+ virtual bool isEqualTo(const MCExpr *x) const { return false; }
+ // This should be set when assigned expressions are not valid ".set"
+ // expressions, e.g. registers, and must be inlined.
+ virtual bool inlineAssignedExpr() const { return false; }
+ virtual void visitUsedExpr(MCStreamer& Streamer) const = 0;
+ virtual MCFragment *findAssociatedFragment() const = 0;
+
+ virtual void fixELFSymbolsInTLSFixups(MCAssembler &) const = 0;
+
+ static bool classof(const MCExpr *E) {
+ return E->getKind() == MCExpr::Target;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCEXPR_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCFixup.h b/third_party/llvm-project/include/llvm/MC/MCFixup.h
new file mode 100644
index 000000000..29e321e23
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCFixup.h
@@ -0,0 +1,202 @@
+//===-- llvm/MC/MCFixup.h - Instruction Relocation and Patching -*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCFIXUP_H
+#define LLVM_MC_MCFIXUP_H
+
+#include "llvm/MC/MCExpr.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/SMLoc.h"
+#include <cassert>
+
+namespace llvm {
+class MCExpr;
+
+/// Extensible enumeration to represent the type of a fixup.
+enum MCFixupKind {
+ FK_NONE = 0, ///< A no-op fixup.
+ FK_Data_1, ///< A one-byte fixup.
+ FK_Data_2, ///< A two-byte fixup.
+ FK_Data_4, ///< A four-byte fixup.
+ FK_Data_8, ///< A eight-byte fixup.
+ FK_Data_6b, ///< A six-bits fixup.
+ FK_PCRel_1, ///< A one-byte pc relative fixup.
+ FK_PCRel_2, ///< A two-byte pc relative fixup.
+ FK_PCRel_4, ///< A four-byte pc relative fixup.
+ FK_PCRel_8, ///< A eight-byte pc relative fixup.
+ FK_GPRel_1, ///< A one-byte gp relative fixup.
+ FK_GPRel_2, ///< A two-byte gp relative fixup.
+ FK_GPRel_4, ///< A four-byte gp relative fixup.
+ FK_GPRel_8, ///< A eight-byte gp relative fixup.
+ FK_DTPRel_4, ///< A four-byte dtp relative fixup.
+ FK_DTPRel_8, ///< A eight-byte dtp relative fixup.
+ FK_TPRel_4, ///< A four-byte tp relative fixup.
+ FK_TPRel_8, ///< A eight-byte tp relative fixup.
+ FK_SecRel_1, ///< A one-byte section relative fixup.
+ FK_SecRel_2, ///< A two-byte section relative fixup.
+ FK_SecRel_4, ///< A four-byte section relative fixup.
+ FK_SecRel_8, ///< A eight-byte section relative fixup.
+ FK_Data_Add_1, ///< A one-byte add fixup.
+ FK_Data_Add_2, ///< A two-byte add fixup.
+ FK_Data_Add_4, ///< A four-byte add fixup.
+ FK_Data_Add_8, ///< A eight-byte add fixup.
+ FK_Data_Add_6b, ///< A six-bits add fixup.
+ FK_Data_Sub_1, ///< A one-byte sub fixup.
+ FK_Data_Sub_2, ///< A two-byte sub fixup.
+ FK_Data_Sub_4, ///< A four-byte sub fixup.
+ FK_Data_Sub_8, ///< A eight-byte sub fixup.
+ FK_Data_Sub_6b, ///< A six-bits sub fixup.
+
+ FirstTargetFixupKind = 128,
+
+ // Limit range of target fixups, in case we want to pack more efficiently
+ // later.
+ MaxTargetFixupKind = (1 << 8)
+};
+
+/// Encode information on a single operation to perform on a byte
+/// sequence (e.g., an encoded instruction) which requires assemble- or run-
+/// time patching.
+///
+/// Fixups are used any time the target instruction encoder needs to represent
+/// some value in an instruction which is not yet concrete. The encoder will
+/// encode the instruction assuming the value is 0, and emit a fixup which
+/// communicates to the assembler backend how it should rewrite the encoded
+/// value.
+///
+/// During the process of relaxation, the assembler will apply fixups as
+/// symbolic values become concrete. When relaxation is complete, any remaining
+/// fixups become relocations in the object file (or errors, if the fixup cannot
+/// be encoded on the target).
+class MCFixup {
+ /// The value to put into the fixup location. The exact interpretation of the
+ /// expression is target dependent, usually it will be one of the operands to
+ /// an instruction or an assembler directive.
+ const MCExpr *Value = nullptr;
+
+ /// The byte index of start of the relocation inside the MCFragment.
+ uint32_t Offset = 0;
+
+ /// The target dependent kind of fixup item this is. The kind is used to
+ /// determine how the operand value should be encoded into the instruction.
+ MCFixupKind Kind = FK_NONE;
+
+ /// The source location which gave rise to the fixup, if any.
+ SMLoc Loc;
+public:
+ static MCFixup create(uint32_t Offset, const MCExpr *Value,
+ MCFixupKind Kind, SMLoc Loc = SMLoc()) {
+ assert(Kind < MaxTargetFixupKind && "Kind out of range!");
+ MCFixup FI;
+ FI.Value = Value;
+ FI.Offset = Offset;
+ FI.Kind = Kind;
+ FI.Loc = Loc;
+ return FI;
+ }
+
+ /// Return a fixup corresponding to the add half of a add/sub fixup pair for
+ /// the given Fixup.
+ static MCFixup createAddFor(const MCFixup &Fixup) {
+ MCFixup FI;
+ FI.Value = Fixup.getValue();
+ FI.Offset = Fixup.getOffset();
+ FI.Kind = getAddKindForKind(Fixup.getKind());
+ FI.Loc = Fixup.getLoc();
+ return FI;
+ }
+
+ /// Return a fixup corresponding to the sub half of a add/sub fixup pair for
+ /// the given Fixup.
+ static MCFixup createSubFor(const MCFixup &Fixup) {
+ MCFixup FI;
+ FI.Value = Fixup.getValue();
+ FI.Offset = Fixup.getOffset();
+ FI.Kind = getSubKindForKind(Fixup.getKind());
+ FI.Loc = Fixup.getLoc();
+ return FI;
+ }
+
+ MCFixupKind getKind() const { return Kind; }
+
+ unsigned getTargetKind() const { return Kind; }
+
+ uint32_t getOffset() const { return Offset; }
+ void setOffset(uint32_t Value) { Offset = Value; }
+
+ const MCExpr *getValue() const { return Value; }
+
+ /// Return the generic fixup kind for a value with the given size. It
+ /// is an error to pass an unsupported size.
+ static MCFixupKind getKindForSize(unsigned Size, bool IsPCRel) {
+ switch (Size) {
+ default: llvm_unreachable("Invalid generic fixup size!");
+ case 1:
+ return IsPCRel ? FK_PCRel_1 : FK_Data_1;
+ case 2:
+ return IsPCRel ? FK_PCRel_2 : FK_Data_2;
+ case 4:
+ return IsPCRel ? FK_PCRel_4 : FK_Data_4;
+ case 8:
+ return IsPCRel ? FK_PCRel_8 : FK_Data_8;
+ }
+ }
+
+ /// Return the generic fixup kind for a value with the given size in bits.
+ /// It is an error to pass an unsupported size.
+ static MCFixupKind getKindForSizeInBits(unsigned Size, bool IsPCRel) {
+ switch (Size) {
+ default:
+ llvm_unreachable("Invalid generic fixup size!");
+ case 6:
+ assert(!IsPCRel && "Invalid pc-relative fixup size!");
+ return FK_Data_6b;
+ case 8:
+ return IsPCRel ? FK_PCRel_1 : FK_Data_1;
+ case 16:
+ return IsPCRel ? FK_PCRel_2 : FK_Data_2;
+ case 32:
+ return IsPCRel ? FK_PCRel_4 : FK_Data_4;
+ case 64:
+ return IsPCRel ? FK_PCRel_8 : FK_Data_8;
+ }
+ }
+
+ /// Return the generic fixup kind for an addition with a given size. It
+ /// is an error to pass an unsupported size.
+ static MCFixupKind getAddKindForKind(MCFixupKind Kind) {
+ switch (Kind) {
+ default: llvm_unreachable("Unknown type to convert!");
+ case FK_Data_1: return FK_Data_Add_1;
+ case FK_Data_2: return FK_Data_Add_2;
+ case FK_Data_4: return FK_Data_Add_4;
+ case FK_Data_8: return FK_Data_Add_8;
+ case FK_Data_6b: return FK_Data_Add_6b;
+ }
+ }
+
+ /// Return the generic fixup kind for an subtraction with a given size. It
+ /// is an error to pass an unsupported size.
+ static MCFixupKind getSubKindForKind(MCFixupKind Kind) {
+ switch (Kind) {
+ default: llvm_unreachable("Unknown type to convert!");
+ case FK_Data_1: return FK_Data_Sub_1;
+ case FK_Data_2: return FK_Data_Sub_2;
+ case FK_Data_4: return FK_Data_Sub_4;
+ case FK_Data_8: return FK_Data_Sub_8;
+ case FK_Data_6b: return FK_Data_Sub_6b;
+ }
+ }
+
+ SMLoc getLoc() const { return Loc; }
+};
+
+} // End llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/MC/MCFragment.h b/third_party/llvm-project/include/llvm/MC/MCFragment.h
new file mode 100644
index 000000000..b0def566c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCFragment.h
@@ -0,0 +1,663 @@
+//===- MCFragment.h - Fragment type hierarchy -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCFRAGMENT_H
+#define LLVM_MC_MCFRAGMENT_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/ilist_node.h"
+#include "llvm/MC/MCFixup.h"
+#include "llvm/MC/MCInst.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/SMLoc.h"
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+class MCSection;
+class MCSubtargetInfo;
+class MCSymbol;
+
+class MCFragment : public ilist_node_with_parent<MCFragment, MCSection> {
+ friend class MCAsmLayout;
+
+public:
+ enum FragmentType : uint8_t {
+ FT_Align,
+ FT_Data,
+ FT_CompactEncodedInst,
+ FT_Fill,
+ FT_Relaxable,
+ FT_Org,
+ FT_Dwarf,
+ FT_DwarfFrame,
+ FT_LEB,
+ FT_Padding,
+ FT_SymbolId,
+ FT_CVInlineLines,
+ FT_CVDefRange,
+ FT_Dummy
+ };
+
+private:
+ FragmentType Kind;
+
+protected:
+ bool HasInstructions;
+
+private:
+ /// LayoutOrder - The layout order of this fragment.
+ unsigned LayoutOrder;
+
+ /// The data for the section this fragment is in.
+ MCSection *Parent;
+
+ /// Atom - The atom this fragment is in, as represented by its defining
+ /// symbol.
+ const MCSymbol *Atom;
+
+ /// \name Assembler Backend Data
+ /// @{
+ //
+ // FIXME: This could all be kept private to the assembler implementation.
+
+ /// Offset - The offset of this fragment in its section. This is ~0 until
+ /// initialized.
+ uint64_t Offset;
+
+ /// @}
+
+protected:
+ MCFragment(FragmentType Kind, bool HasInstructions,
+ MCSection *Parent = nullptr);
+
+ ~MCFragment();
+
+public:
+ MCFragment() = delete;
+ MCFragment(const MCFragment &) = delete;
+ MCFragment &operator=(const MCFragment &) = delete;
+
+ /// Destroys the current fragment.
+ ///
+ /// This must be used instead of delete as MCFragment is non-virtual.
+ /// This method will dispatch to the appropriate subclass.
+ void destroy();
+
+ FragmentType getKind() const { return Kind; }
+
+ MCSection *getParent() const { return Parent; }
+ void setParent(MCSection *Value) { Parent = Value; }
+
+ const MCSymbol *getAtom() const { return Atom; }
+ void setAtom(const MCSymbol *Value) { Atom = Value; }
+
+ unsigned getLayoutOrder() const { return LayoutOrder; }
+ void setLayoutOrder(unsigned Value) { LayoutOrder = Value; }
+
+ /// Does this fragment have instructions emitted into it? By default
+ /// this is false, but specific fragment types may set it to true.
+ bool hasInstructions() const { return HasInstructions; }
+
+ /// Return true if given frgment has FT_Dummy type.
+ bool isDummy() const { return Kind == FT_Dummy; }
+
+ void dump() const;
+};
+
+class MCDummyFragment : public MCFragment {
+public:
+ explicit MCDummyFragment(MCSection *Sec) : MCFragment(FT_Dummy, false, Sec) {}
+
+ static bool classof(const MCFragment *F) { return F->getKind() == FT_Dummy; }
+};
+
+/// Interface implemented by fragments that contain encoded instructions and/or
+/// data.
+///
+class MCEncodedFragment : public MCFragment {
+ /// Should this fragment be aligned to the end of a bundle?
+ bool AlignToBundleEnd = false;
+
+ uint8_t BundlePadding = 0;
+
+protected:
+ MCEncodedFragment(MCFragment::FragmentType FType, bool HasInstructions,
+ MCSection *Sec)
+ : MCFragment(FType, HasInstructions, Sec) {}
+
+ /// STI - The MCSubtargetInfo in effect when the instruction was encoded.
+ /// must be non-null for instructions.
+ const MCSubtargetInfo *STI = nullptr;
+
+public:
+ static bool classof(const MCFragment *F) {
+ MCFragment::FragmentType Kind = F->getKind();
+ switch (Kind) {
+ default:
+ return false;
+ case MCFragment::FT_Relaxable:
+ case MCFragment::FT_CompactEncodedInst:
+ case MCFragment::FT_Data:
+ case MCFragment::FT_Dwarf:
+ case MCFragment::FT_DwarfFrame:
+ return true;
+ }
+ }
+
+ /// Should this fragment be placed at the end of an aligned bundle?
+ bool alignToBundleEnd() const { return AlignToBundleEnd; }
+ void setAlignToBundleEnd(bool V) { AlignToBundleEnd = V; }
+
+ /// Get the padding size that must be inserted before this fragment.
+ /// Used for bundling. By default, no padding is inserted.
+ /// Note that padding size is restricted to 8 bits. This is an optimization
+ /// to reduce the amount of space used for each fragment. In practice, larger
+ /// padding should never be required.
+ uint8_t getBundlePadding() const { return BundlePadding; }
+
+ /// Set the padding size for this fragment. By default it's a no-op,
+ /// and only some fragments have a meaningful implementation.
+ void setBundlePadding(uint8_t N) { BundlePadding = N; }
+
+ /// Retrieve the MCSubTargetInfo in effect when the instruction was encoded.
+ /// Guaranteed to be non-null if hasInstructions() == true
+ const MCSubtargetInfo *getSubtargetInfo() const { return STI; }
+
+ /// Record that the fragment contains instructions with the MCSubtargetInfo in
+ /// effect when the instruction was encoded.
+ void setHasInstructions(const MCSubtargetInfo &STI) {
+ HasInstructions = true;
+ this->STI = &STI;
+ }
+};
+
+/// Interface implemented by fragments that contain encoded instructions and/or
+/// data.
+///
+template<unsigned ContentsSize>
+class MCEncodedFragmentWithContents : public MCEncodedFragment {
+ SmallVector<char, ContentsSize> Contents;
+
+protected:
+ MCEncodedFragmentWithContents(MCFragment::FragmentType FType,
+ bool HasInstructions,
+ MCSection *Sec)
+ : MCEncodedFragment(FType, HasInstructions, Sec) {}
+
+public:
+ SmallVectorImpl<char> &getContents() { return Contents; }
+ const SmallVectorImpl<char> &getContents() const { return Contents; }
+};
+
+/// Interface implemented by fragments that contain encoded instructions and/or
+/// data and also have fixups registered.
+///
+template<unsigned ContentsSize, unsigned FixupsSize>
+class MCEncodedFragmentWithFixups :
+ public MCEncodedFragmentWithContents<ContentsSize> {
+
+ /// Fixups - The list of fixups in this fragment.
+ SmallVector<MCFixup, FixupsSize> Fixups;
+
+protected:
+ MCEncodedFragmentWithFixups(MCFragment::FragmentType FType,
+ bool HasInstructions,
+ MCSection *Sec)
+ : MCEncodedFragmentWithContents<ContentsSize>(FType, HasInstructions,
+ Sec) {}
+
+public:
+
+ using const_fixup_iterator = SmallVectorImpl<MCFixup>::const_iterator;
+ using fixup_iterator = SmallVectorImpl<MCFixup>::iterator;
+
+ SmallVectorImpl<MCFixup> &getFixups() { return Fixups; }
+ const SmallVectorImpl<MCFixup> &getFixups() const { return Fixups; }
+
+ fixup_iterator fixup_begin() { return Fixups.begin(); }
+ const_fixup_iterator fixup_begin() const { return Fixups.begin(); }
+
+ fixup_iterator fixup_end() { return Fixups.end(); }
+ const_fixup_iterator fixup_end() const { return Fixups.end(); }
+
+ static bool classof(const MCFragment *F) {
+ MCFragment::FragmentType Kind = F->getKind();
+ return Kind == MCFragment::FT_Relaxable || Kind == MCFragment::FT_Data ||
+ Kind == MCFragment::FT_CVDefRange || Kind == MCFragment::FT_Dwarf ||
+ Kind == MCFragment::FT_DwarfFrame;
+ }
+};
+
+/// Fragment for data and encoded instructions.
+///
+class MCDataFragment : public MCEncodedFragmentWithFixups<32, 4> {
+public:
+ MCDataFragment(MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithFixups<32, 4>(FT_Data, false, Sec) {}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Data;
+ }
+};
+
+/// This is a compact (memory-size-wise) fragment for holding an encoded
+/// instruction (non-relaxable) that has no fixups registered. When applicable,
+/// it can be used instead of MCDataFragment and lead to lower memory
+/// consumption.
+///
+class MCCompactEncodedInstFragment : public MCEncodedFragmentWithContents<4> {
+public:
+ MCCompactEncodedInstFragment(MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithContents(FT_CompactEncodedInst, true, Sec) {
+ }
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_CompactEncodedInst;
+ }
+};
+
+/// A relaxable fragment holds on to its MCInst, since it may need to be
+/// relaxed during the assembler layout and relaxation stage.
+///
+class MCRelaxableFragment : public MCEncodedFragmentWithFixups<8, 1> {
+
+ /// Inst - The instruction this is a fragment for.
+ MCInst Inst;
+
+public:
+ MCRelaxableFragment(const MCInst &Inst, const MCSubtargetInfo &STI,
+ MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithFixups(FT_Relaxable, true, Sec),
+ Inst(Inst) { this->STI = &STI; }
+
+ const MCInst &getInst() const { return Inst; }
+ void setInst(const MCInst &Value) { Inst = Value; }
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Relaxable;
+ }
+};
+
+class MCAlignFragment : public MCFragment {
+ /// Alignment - The alignment to ensure, in bytes.
+ unsigned Alignment;
+
+ /// EmitNops - Flag to indicate that (optimal) NOPs should be emitted instead
+ /// of using the provided value. The exact interpretation of this flag is
+ /// target dependent.
+ bool EmitNops : 1;
+
+ /// Value - Value to use for filling padding bytes.
+ int64_t Value;
+
+ /// ValueSize - The size of the integer (in bytes) of \p Value.
+ unsigned ValueSize;
+
+ /// MaxBytesToEmit - The maximum number of bytes to emit; if the alignment
+ /// cannot be satisfied in this width then this fragment is ignored.
+ unsigned MaxBytesToEmit;
+
+public:
+ MCAlignFragment(unsigned Alignment, int64_t Value, unsigned ValueSize,
+ unsigned MaxBytesToEmit, MCSection *Sec = nullptr)
+ : MCFragment(FT_Align, false, Sec), Alignment(Alignment), EmitNops(false),
+ Value(Value), ValueSize(ValueSize), MaxBytesToEmit(MaxBytesToEmit) {}
+
+ /// \name Accessors
+ /// @{
+
+ unsigned getAlignment() const { return Alignment; }
+
+ int64_t getValue() const { return Value; }
+
+ unsigned getValueSize() const { return ValueSize; }
+
+ unsigned getMaxBytesToEmit() const { return MaxBytesToEmit; }
+
+ bool hasEmitNops() const { return EmitNops; }
+ void setEmitNops(bool Value) { EmitNops = Value; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Align;
+ }
+};
+
+/// Fragment for adding required padding.
+/// This fragment is always inserted before an instruction, and holds that
+/// instruction as context information (as well as a mask of kinds) for
+/// determining the padding size.
+///
+class MCPaddingFragment : public MCFragment {
+ /// A mask containing all the kinds relevant to this fragment. i.e. the i'th
+ /// bit will be set iff kind i is relevant to this fragment.
+ uint64_t PaddingPoliciesMask;
+ /// A boolean indicating if this fragment will actually hold padding. If its
+ /// value is false, then this fragment serves only as a placeholder,
+ /// containing data to assist other insertion point in their decision making.
+ bool IsInsertionPoint;
+
+ uint64_t Size;
+
+ struct MCInstInfo {
+ bool IsInitialized;
+ MCInst Inst;
+ /// A boolean indicating whether the instruction pointed by this fragment is
+ /// a fixed size instruction or a relaxable instruction held by a
+ /// MCRelaxableFragment.
+ bool IsImmutableSizedInst;
+ union {
+ /// If the instruction is a fixed size instruction, hold its size.
+ size_t InstSize;
+ /// Otherwise, hold a pointer to the MCRelaxableFragment holding it.
+ MCRelaxableFragment *InstFragment;
+ };
+ };
+ MCInstInfo InstInfo;
+
+public:
+ static const uint64_t PFK_None = UINT64_C(0);
+
+ enum MCPaddingFragmentKind {
+ // values 0-7 are reserved for future target independet values.
+
+ FirstTargetPerfNopFragmentKind = 8,
+
+ /// Limit range of target MCPerfNopFragment kinds to fit in uint64_t
+ MaxTargetPerfNopFragmentKind = 63
+ };
+
+ MCPaddingFragment(MCSection *Sec = nullptr)
+ : MCFragment(FT_Padding, false, Sec), PaddingPoliciesMask(PFK_None),
+ IsInsertionPoint(false), Size(UINT64_C(0)),
+ InstInfo({false, MCInst(), false, {0}}) {}
+
+ bool isInsertionPoint() const { return IsInsertionPoint; }
+ void setAsInsertionPoint() { IsInsertionPoint = true; }
+ uint64_t getPaddingPoliciesMask() const { return PaddingPoliciesMask; }
+ void setPaddingPoliciesMask(uint64_t Value) { PaddingPoliciesMask = Value; }
+ bool hasPaddingPolicy(uint64_t PolicyMask) const {
+ assert(isPowerOf2_64(PolicyMask) &&
+ "Policy mask must contain exactly one policy");
+ return (getPaddingPoliciesMask() & PolicyMask) != PFK_None;
+ }
+ const MCInst &getInst() const {
+ assert(isInstructionInitialized() && "Fragment has no instruction!");
+ return InstInfo.Inst;
+ }
+ size_t getInstSize() const {
+ assert(isInstructionInitialized() && "Fragment has no instruction!");
+ if (InstInfo.IsImmutableSizedInst)
+ return InstInfo.InstSize;
+ assert(InstInfo.InstFragment != nullptr &&
+ "Must have a valid InstFragment to retrieve InstSize from");
+ return InstInfo.InstFragment->getContents().size();
+ }
+ void setInstAndInstSize(const MCInst &Inst, size_t InstSize) {
+ InstInfo.IsInitialized = true;
+ InstInfo.IsImmutableSizedInst = true;
+ InstInfo.Inst = Inst;
+ InstInfo.InstSize = InstSize;
+ }
+ void setInstAndInstFragment(const MCInst &Inst,
+ MCRelaxableFragment *InstFragment) {
+ InstInfo.IsInitialized = true;
+ InstInfo.IsImmutableSizedInst = false;
+ InstInfo.Inst = Inst;
+ InstInfo.InstFragment = InstFragment;
+ }
+ uint64_t getSize() const { return Size; }
+ void setSize(uint64_t Value) { Size = Value; }
+ bool isInstructionInitialized() const { return InstInfo.IsInitialized; }
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Padding;
+ }
+};
+
+class MCFillFragment : public MCFragment {
+ /// Value to use for filling bytes.
+ uint64_t Value;
+ uint8_t ValueSize;
+ /// The number of bytes to insert.
+ const MCExpr &NumValues;
+
+ /// Source location of the directive that this fragment was created for.
+ SMLoc Loc;
+
+public:
+ MCFillFragment(uint64_t Value, uint8_t VSize, const MCExpr &NumValues,
+ SMLoc Loc, MCSection *Sec = nullptr)
+ : MCFragment(FT_Fill, false, Sec), Value(Value), ValueSize(VSize),
+ NumValues(NumValues), Loc(Loc) {}
+
+ uint64_t getValue() const { return Value; }
+ uint8_t getValueSize() const { return ValueSize; }
+ const MCExpr &getNumValues() const { return NumValues; }
+
+ SMLoc getLoc() const { return Loc; }
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Fill;
+ }
+};
+
+class MCOrgFragment : public MCFragment {
+ /// The offset this fragment should start at.
+ const MCExpr *Offset;
+
+ /// Value to use for filling bytes.
+ int8_t Value;
+
+ /// Source location of the directive that this fragment was created for.
+ SMLoc Loc;
+
+public:
+ MCOrgFragment(const MCExpr &Offset, int8_t Value, SMLoc Loc,
+ MCSection *Sec = nullptr)
+ : MCFragment(FT_Org, false, Sec), Offset(&Offset), Value(Value), Loc(Loc) {}
+
+ /// \name Accessors
+ /// @{
+
+ const MCExpr &getOffset() const { return *Offset; }
+
+ uint8_t getValue() const { return Value; }
+
+ SMLoc getLoc() const { return Loc; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Org;
+ }
+};
+
+class MCLEBFragment : public MCFragment {
+ /// Value - The value this fragment should contain.
+ const MCExpr *Value;
+
+ /// IsSigned - True if this is a sleb128, false if uleb128.
+ bool IsSigned;
+
+ SmallString<8> Contents;
+
+public:
+ MCLEBFragment(const MCExpr &Value_, bool IsSigned_, MCSection *Sec = nullptr)
+ : MCFragment(FT_LEB, false, Sec), Value(&Value_), IsSigned(IsSigned_) {
+ Contents.push_back(0);
+ }
+
+ /// \name Accessors
+ /// @{
+
+ const MCExpr &getValue() const { return *Value; }
+
+ bool isSigned() const { return IsSigned; }
+
+ SmallString<8> &getContents() { return Contents; }
+ const SmallString<8> &getContents() const { return Contents; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_LEB;
+ }
+};
+
+class MCDwarfLineAddrFragment : public MCEncodedFragmentWithFixups<8, 1> {
+ /// LineDelta - the value of the difference between the two line numbers
+ /// between two .loc dwarf directives.
+ int64_t LineDelta;
+
+ /// AddrDelta - The expression for the difference of the two symbols that
+ /// make up the address delta between two .loc dwarf directives.
+ const MCExpr *AddrDelta;
+
+public:
+ MCDwarfLineAddrFragment(int64_t LineDelta, const MCExpr &AddrDelta,
+ MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithFixups<8, 1>(FT_Dwarf, false, Sec),
+ LineDelta(LineDelta), AddrDelta(&AddrDelta) {}
+
+ /// \name Accessors
+ /// @{
+
+ int64_t getLineDelta() const { return LineDelta; }
+
+ const MCExpr &getAddrDelta() const { return *AddrDelta; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_Dwarf;
+ }
+};
+
+class MCDwarfCallFrameFragment : public MCEncodedFragmentWithFixups<8, 1> {
+ /// AddrDelta - The expression for the difference of the two symbols that
+ /// make up the address delta between two .cfi_* dwarf directives.
+ const MCExpr *AddrDelta;
+
+public:
+ MCDwarfCallFrameFragment(const MCExpr &AddrDelta, MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithFixups<8, 1>(FT_DwarfFrame, false, Sec),
+ AddrDelta(&AddrDelta) {}
+
+ /// \name Accessors
+ /// @{
+
+ const MCExpr &getAddrDelta() const { return *AddrDelta; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_DwarfFrame;
+ }
+};
+
+/// Represents a symbol table index fragment.
+class MCSymbolIdFragment : public MCFragment {
+ const MCSymbol *Sym;
+
+public:
+ MCSymbolIdFragment(const MCSymbol *Sym, MCSection *Sec = nullptr)
+ : MCFragment(FT_SymbolId, false, Sec), Sym(Sym) {}
+
+ /// \name Accessors
+ /// @{
+
+ const MCSymbol *getSymbol() { return Sym; }
+ const MCSymbol *getSymbol() const { return Sym; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_SymbolId;
+ }
+};
+
+/// Fragment representing the binary annotations produced by the
+/// .cv_inline_linetable directive.
+class MCCVInlineLineTableFragment : public MCFragment {
+ unsigned SiteFuncId;
+ unsigned StartFileId;
+ unsigned StartLineNum;
+ const MCSymbol *FnStartSym;
+ const MCSymbol *FnEndSym;
+ SmallString<8> Contents;
+
+ /// CodeViewContext has the real knowledge about this format, so let it access
+ /// our members.
+ friend class CodeViewContext;
+
+public:
+ MCCVInlineLineTableFragment(unsigned SiteFuncId, unsigned StartFileId,
+ unsigned StartLineNum, const MCSymbol *FnStartSym,
+ const MCSymbol *FnEndSym,
+ MCSection *Sec = nullptr)
+ : MCFragment(FT_CVInlineLines, false, Sec), SiteFuncId(SiteFuncId),
+ StartFileId(StartFileId), StartLineNum(StartLineNum),
+ FnStartSym(FnStartSym), FnEndSym(FnEndSym) {}
+
+ /// \name Accessors
+ /// @{
+
+ const MCSymbol *getFnStartSym() const { return FnStartSym; }
+ const MCSymbol *getFnEndSym() const { return FnEndSym; }
+
+ SmallString<8> &getContents() { return Contents; }
+ const SmallString<8> &getContents() const { return Contents; }
+
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_CVInlineLines;
+ }
+};
+
+/// Fragment representing the .cv_def_range directive.
+class MCCVDefRangeFragment : public MCEncodedFragmentWithFixups<32, 4> {
+ SmallVector<std::pair<const MCSymbol *, const MCSymbol *>, 2> Ranges;
+ SmallString<32> FixedSizePortion;
+
+ /// CodeViewContext has the real knowledge about this format, so let it access
+ /// our members.
+ friend class CodeViewContext;
+
+public:
+ MCCVDefRangeFragment(
+ ArrayRef<std::pair<const MCSymbol *, const MCSymbol *>> Ranges,
+ StringRef FixedSizePortion, MCSection *Sec = nullptr)
+ : MCEncodedFragmentWithFixups<32, 4>(FT_CVDefRange, false, Sec),
+ Ranges(Ranges.begin(), Ranges.end()),
+ FixedSizePortion(FixedSizePortion) {}
+
+ /// \name Accessors
+ /// @{
+ ArrayRef<std::pair<const MCSymbol *, const MCSymbol *>> getRanges() const {
+ return Ranges;
+ }
+
+ StringRef getFixedSizePortion() const { return FixedSizePortion; }
+ /// @}
+
+ static bool classof(const MCFragment *F) {
+ return F->getKind() == MCFragment::FT_CVDefRange;
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCFRAGMENT_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCInst.h b/third_party/llvm-project/include/llvm/MC/MCInst.h
new file mode 100644
index 000000000..8df8096bb
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCInst.h
@@ -0,0 +1,225 @@
+//===- llvm/MC/MCInst.h - MCInst class --------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains the declaration of the MCInst and MCOperand classes, which
+// is the basic representation used to represent low-level machine code
+// instructions.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCINST_H
+#define LLVM_MC_MCINST_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/SMLoc.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+
+namespace llvm {
+
+class MCExpr;
+class MCInst;
+class MCInstPrinter;
+class raw_ostream;
+
+/// Instances of this class represent operands of the MCInst class.
+/// This is a simple discriminated union.
+class MCOperand {
+ enum MachineOperandType : unsigned char {
+ kInvalid, ///< Uninitialized.
+ kRegister, ///< Register operand.
+ kImmediate, ///< Immediate operand.
+ kFPImmediate, ///< Floating-point immediate operand.
+ kExpr, ///< Relocatable immediate operand.
+ kInst ///< Sub-instruction operand.
+ };
+ MachineOperandType Kind = kInvalid;
+
+ union {
+ unsigned RegVal;
+ int64_t ImmVal;
+ double FPImmVal;
+ const MCExpr *ExprVal;
+ const MCInst *InstVal;
+ };
+
+public:
+ MCOperand() : FPImmVal(0.0) {}
+
+ bool isValid() const { return Kind != kInvalid; }
+ bool isReg() const { return Kind == kRegister; }
+ bool isImm() const { return Kind == kImmediate; }
+ bool isFPImm() const { return Kind == kFPImmediate; }
+ bool isExpr() const { return Kind == kExpr; }
+ bool isInst() const { return Kind == kInst; }
+
+ /// Returns the register number.
+ unsigned getReg() const {
+ assert(isReg() && "This is not a register operand!");
+ return RegVal;
+ }
+
+ /// Set the register number.
+ void setReg(unsigned Reg) {
+ assert(isReg() && "This is not a register operand!");
+ RegVal = Reg;
+ }
+
+ int64_t getImm() const {
+ assert(isImm() && "This is not an immediate");
+ return ImmVal;
+ }
+
+ void setImm(int64_t Val) {
+ assert(isImm() && "This is not an immediate");
+ ImmVal = Val;
+ }
+
+ double getFPImm() const {
+ assert(isFPImm() && "This is not an FP immediate");
+ return FPImmVal;
+ }
+
+ void setFPImm(double Val) {
+ assert(isFPImm() && "This is not an FP immediate");
+ FPImmVal = Val;
+ }
+
+ const MCExpr *getExpr() const {
+ assert(isExpr() && "This is not an expression");
+ return ExprVal;
+ }
+
+ void setExpr(const MCExpr *Val) {
+ assert(isExpr() && "This is not an expression");
+ ExprVal = Val;
+ }
+
+ const MCInst *getInst() const {
+ assert(isInst() && "This is not a sub-instruction");
+ return InstVal;
+ }
+
+ void setInst(const MCInst *Val) {
+ assert(isInst() && "This is not a sub-instruction");
+ InstVal = Val;
+ }
+
+ static MCOperand createReg(unsigned Reg) {
+ MCOperand Op;
+ Op.Kind = kRegister;
+ Op.RegVal = Reg;
+ return Op;
+ }
+
+ static MCOperand createImm(int64_t Val) {
+ MCOperand Op;
+ Op.Kind = kImmediate;
+ Op.ImmVal = Val;
+ return Op;
+ }
+
+ static MCOperand createFPImm(double Val) {
+ MCOperand Op;
+ Op.Kind = kFPImmediate;
+ Op.FPImmVal = Val;
+ return Op;
+ }
+
+ static MCOperand createExpr(const MCExpr *Val) {
+ MCOperand Op;
+ Op.Kind = kExpr;
+ Op.ExprVal = Val;
+ return Op;
+ }
+
+ static MCOperand createInst(const MCInst *Val) {
+ MCOperand Op;
+ Op.Kind = kInst;
+ Op.InstVal = Val;
+ return Op;
+ }
+
+ void print(raw_ostream &OS) const;
+ void dump() const;
+ bool isBareSymbolRef() const;
+ bool evaluateAsConstantImm(int64_t &Imm) const;
+};
+
+/// Instances of this class represent a single low-level machine
+/// instruction.
+class MCInst {
+ unsigned Opcode = 0;
+ SMLoc Loc;
+ SmallVector<MCOperand, 8> Operands;
+ // These flags could be used to pass some info from one target subcomponent
+ // to another, for example, from disassembler to asm printer. The values of
+ // the flags have any sense on target level only (e.g. prefixes on x86).
+ unsigned Flags = 0;
+
+public:
+ MCInst() = default;
+
+ void setOpcode(unsigned Op) { Opcode = Op; }
+ unsigned getOpcode() const { return Opcode; }
+
+ void setFlags(unsigned F) { Flags = F; }
+ unsigned getFlags() const { return Flags; }
+
+ void setLoc(SMLoc loc) { Loc = loc; }
+ SMLoc getLoc() const { return Loc; }
+
+ const MCOperand &getOperand(unsigned i) const { return Operands[i]; }
+ MCOperand &getOperand(unsigned i) { return Operands[i]; }
+ unsigned getNumOperands() const { return Operands.size(); }
+
+ void addOperand(const MCOperand &Op) { Operands.push_back(Op); }
+
+ using iterator = SmallVectorImpl<MCOperand>::iterator;
+ using const_iterator = SmallVectorImpl<MCOperand>::const_iterator;
+
+ void clear() { Operands.clear(); }
+ void erase(iterator I) { Operands.erase(I); }
+ void erase(iterator First, iterator Last) { Operands.erase(First, Last); }
+ size_t size() const { return Operands.size(); }
+ iterator begin() { return Operands.begin(); }
+ const_iterator begin() const { return Operands.begin(); }
+ iterator end() { return Operands.end(); }
+ const_iterator end() const { return Operands.end(); }
+
+ iterator insert(iterator I, const MCOperand &Op) {
+ return Operands.insert(I, Op);
+ }
+
+ void print(raw_ostream &OS) const;
+ void dump() const;
+
+ /// Dump the MCInst as prettily as possible using the additional MC
+ /// structures, if given. Operators are separated by the \p Separator
+ /// string.
+ void dump_pretty(raw_ostream &OS, const MCInstPrinter *Printer = nullptr,
+ StringRef Separator = " ") const;
+ void dump_pretty(raw_ostream &OS, StringRef Name,
+ StringRef Separator = " ") const;
+};
+
+inline raw_ostream& operator<<(raw_ostream &OS, const MCOperand &MO) {
+ MO.print(OS);
+ return OS;
+}
+
+inline raw_ostream& operator<<(raw_ostream &OS, const MCInst &MI) {
+ MI.print(OS);
+ return OS;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCINST_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCRegister.h b/third_party/llvm-project/include/llvm/MC/MCRegister.h
new file mode 100644
index 000000000..8372947a4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCRegister.h
@@ -0,0 +1,110 @@
+//===-- llvm/MC/Register.h --------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_REGISTER_H
+#define LLVM_MC_REGISTER_H
+
+#include "llvm/ADT/DenseMapInfo.h"
+#include <cassert>
+
+namespace llvm {
+
+/// An unsigned integer type large enough to represent all physical registers,
+/// but not necessarily virtual registers.
+using MCPhysReg = uint16_t;
+
+/// Wrapper class representing physical registers. Should be passed by value.
+class MCRegister {
+ unsigned Reg;
+
+public:
+ MCRegister(unsigned Val = 0): Reg(Val) {}
+
+ // Register numbers can represent physical registers, virtual registers, and
+ // sometimes stack slots. The unsigned values are divided into these ranges:
+ //
+ // 0 Not a register, can be used as a sentinel.
+ // [1;2^30) Physical registers assigned by TableGen.
+ // [2^30;2^31) Stack slots. (Rarely used.)
+ // [2^31;2^32) Virtual registers assigned by MachineRegisterInfo.
+ //
+ // Further sentinels can be allocated from the small negative integers.
+ // DenseMapInfo<unsigned> uses -1u and -2u.
+
+ /// This is the portion of the positive number space that is not a physical
+ /// register. StackSlot values do not exist in the MC layer, see
+ /// Register::isStackSlot() for the more information on them.
+ ///
+ /// Note that isVirtualRegister() and isPhysicalRegister() cannot handle stack
+ /// slots, so if a variable may contains a stack slot, always check
+ /// isStackSlot() first.
+ static bool isStackSlot(unsigned Reg) {
+ return int(Reg) >= (1 << 30);
+ }
+
+ /// Return true if the specified register number is in
+ /// the physical register namespace.
+ static bool isPhysicalRegister(unsigned Reg) {
+ assert(!isStackSlot(Reg) && "Not a register! Check isStackSlot() first.");
+ return int(Reg) > 0;
+ }
+
+ /// Return true if the specified register number is in the physical register
+ /// namespace.
+ bool isPhysical() const {
+ return isPhysicalRegister(Reg);
+ }
+
+ operator unsigned() const {
+ return Reg;
+ }
+
+ unsigned id() const {
+ return Reg;
+ }
+
+ bool isValid() const {
+ return Reg != 0;
+ }
+
+ /// Comparisons between register objects
+ bool operator==(const MCRegister &Other) const { return Reg == Other.Reg; }
+ bool operator!=(const MCRegister &Other) const { return Reg != Other.Reg; }
+
+ /// Comparisons against register constants. E.g.
+ /// * R == AArch64::WZR
+ /// * R == 0
+ /// * R == VirtRegMap::NO_PHYS_REG
+ bool operator==(unsigned Other) const { return Reg == Other; }
+ bool operator!=(unsigned Other) const { return Reg != Other; }
+ bool operator==(int Other) const { return Reg == unsigned(Other); }
+ bool operator!=(int Other) const { return Reg != unsigned(Other); }
+ // MSVC requires that we explicitly declare these two as well.
+ bool operator==(MCPhysReg Other) const { return Reg == unsigned(Other); }
+ bool operator!=(MCPhysReg Other) const { return Reg != unsigned(Other); }
+};
+
+// Provide DenseMapInfo for MCRegister
+template<> struct DenseMapInfo<MCRegister> {
+ static inline unsigned getEmptyKey() {
+ return DenseMapInfo<unsigned>::getEmptyKey();
+ }
+ static inline unsigned getTombstoneKey() {
+ return DenseMapInfo<unsigned>::getTombstoneKey();
+ }
+ static unsigned getHashValue(const MCRegister &Val) {
+ return DenseMapInfo<unsigned>::getHashValue(Val.id());
+ }
+ static bool isEqual(const MCRegister &LHS, const MCRegister &RHS) {
+ return DenseMapInfo<unsigned>::isEqual(LHS.id(), RHS.id());
+ }
+};
+
+}
+
+#endif // ifndef LLVM_MC_REGISTER_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCRegisterInfo.h b/third_party/llvm-project/include/llvm/MC/MCRegisterInfo.h
new file mode 100644
index 000000000..c7dc56ea5
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCRegisterInfo.h
@@ -0,0 +1,721 @@
+//===- MC/MCRegisterInfo.h - Target Register Description --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file describes an abstract interface used to get information about a
+// target machines register file. This information is used for a variety of
+// purposed, especially register allocation.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCREGISTERINFO_H
+#define LLVM_MC_MCREGISTERINFO_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/MC/LaneBitmask.h"
+#include "llvm/MC/MCRegister.h"
+#include <cassert>
+#include <cstdint>
+#include <utility>
+
+namespace llvm {
+
+/// MCRegisterClass - Base class of TargetRegisterClass.
+class MCRegisterClass {
+public:
+ using iterator = const MCPhysReg*;
+ using const_iterator = const MCPhysReg*;
+
+ const iterator RegsBegin;
+ const uint8_t *const RegSet;
+ const uint32_t NameIdx;
+ const uint16_t RegsSize;
+ const uint16_t RegSetSize;
+ const uint16_t ID;
+ const int8_t CopyCost;
+ const bool Allocatable;
+
+ /// getID() - Return the register class ID number.
+ ///
+ unsigned getID() const { return ID; }
+
+ /// begin/end - Return all of the registers in this class.
+ ///
+ iterator begin() const { return RegsBegin; }
+ iterator end() const { return RegsBegin + RegsSize; }
+
+ /// getNumRegs - Return the number of registers in this class.
+ ///
+ unsigned getNumRegs() const { return RegsSize; }
+
+ /// getRegister - Return the specified register in the class.
+ ///
+ unsigned getRegister(unsigned i) const {
+ assert(i < getNumRegs() && "Register number out of range!");
+ return RegsBegin[i];
+ }
+
+ /// contains - Return true if the specified register is included in this
+ /// register class. This does not include virtual registers.
+ bool contains(MCRegister Reg) const {
+ unsigned RegNo = unsigned(Reg);
+ unsigned InByte = RegNo % 8;
+ unsigned Byte = RegNo / 8;
+ if (Byte >= RegSetSize)
+ return false;
+ return (RegSet[Byte] & (1 << InByte)) != 0;
+ }
+
+ /// contains - Return true if both registers are in this class.
+ bool contains(MCRegister Reg1, MCRegister Reg2) const {
+ return contains(Reg1) && contains(Reg2);
+ }
+
+ /// getCopyCost - Return the cost of copying a value between two registers in
+ /// this class. A negative number means the register class is very expensive
+ /// to copy e.g. status flag register classes.
+ int getCopyCost() const { return CopyCost; }
+
+ /// isAllocatable - Return true if this register class may be used to create
+ /// virtual registers.
+ bool isAllocatable() const { return Allocatable; }
+};
+
+/// MCRegisterDesc - This record contains information about a particular
+/// register. The SubRegs field is a zero terminated array of registers that
+/// are sub-registers of the specific register, e.g. AL, AH are sub-registers
+/// of AX. The SuperRegs field is a zero terminated array of registers that are
+/// super-registers of the specific register, e.g. RAX, EAX, are
+/// super-registers of AX.
+///
+struct MCRegisterDesc {
+ uint32_t Name; // Printable name for the reg (for debugging)
+ uint32_t SubRegs; // Sub-register set, described above
+ uint32_t SuperRegs; // Super-register set, described above
+
+ // Offset into MCRI::SubRegIndices of a list of sub-register indices for each
+ // sub-register in SubRegs.
+ uint32_t SubRegIndices;
+
+ // RegUnits - Points to the list of register units. The low 4 bits holds the
+ // Scale, the high bits hold an offset into DiffLists. See MCRegUnitIterator.
+ uint32_t RegUnits;
+
+ /// Index into list with lane mask sequences. The sequence contains a lanemask
+ /// for every register unit.
+ uint16_t RegUnitLaneMasks;
+};
+
+/// MCRegisterInfo base class - We assume that the target defines a static
+/// array of MCRegisterDesc objects that represent all of the machine
+/// registers that the target has. As such, we simply have to track a pointer
+/// to this array so that we can turn register number into a register
+/// descriptor.
+///
+/// Note this class is designed to be a base class of TargetRegisterInfo, which
+/// is the interface used by codegen. However, specific targets *should never*
+/// specialize this class. MCRegisterInfo should only contain getters to access
+/// TableGen generated physical register data. It must not be extended with
+/// virtual methods.
+///
+class MCRegisterInfo {
+public:
+ using regclass_iterator = const MCRegisterClass *;
+
+ /// DwarfLLVMRegPair - Emitted by tablegen so Dwarf<->LLVM reg mappings can be
+ /// performed with a binary search.
+ struct DwarfLLVMRegPair {
+ unsigned FromReg;
+ unsigned ToReg;
+
+ bool operator<(DwarfLLVMRegPair RHS) const { return FromReg < RHS.FromReg; }
+ };
+
+ /// SubRegCoveredBits - Emitted by tablegen: bit range covered by a subreg
+ /// index, -1 in any being invalid.
+ struct SubRegCoveredBits {
+ uint16_t Offset;
+ uint16_t Size;
+ };
+
+private:
+ const MCRegisterDesc *Desc; // Pointer to the descriptor array
+ unsigned NumRegs; // Number of entries in the array
+ MCRegister RAReg; // Return address register
+ MCRegister PCReg; // Program counter register
+ const MCRegisterClass *Classes; // Pointer to the regclass array
+ unsigned NumClasses; // Number of entries in the array
+ unsigned NumRegUnits; // Number of regunits.
+ const MCPhysReg (*RegUnitRoots)[2]; // Pointer to regunit root table.
+ const MCPhysReg *DiffLists; // Pointer to the difflists array
+ const LaneBitmask *RegUnitMaskSequences; // Pointer to lane mask sequences
+ // for register units.
+ const char *RegStrings; // Pointer to the string table.
+ const char *RegClassStrings; // Pointer to the class strings.
+ const uint16_t *SubRegIndices; // Pointer to the subreg lookup
+ // array.
+ const SubRegCoveredBits *SubRegIdxRanges; // Pointer to the subreg covered
+ // bit ranges array.
+ unsigned NumSubRegIndices; // Number of subreg indices.
+ const uint16_t *RegEncodingTable; // Pointer to array of register
+ // encodings.
+
+ unsigned L2DwarfRegsSize;
+ unsigned EHL2DwarfRegsSize;
+ unsigned Dwarf2LRegsSize;
+ unsigned EHDwarf2LRegsSize;
+ const DwarfLLVMRegPair *L2DwarfRegs; // LLVM to Dwarf regs mapping
+ const DwarfLLVMRegPair *EHL2DwarfRegs; // LLVM to Dwarf regs mapping EH
+ const DwarfLLVMRegPair *Dwarf2LRegs; // Dwarf to LLVM regs mapping
+ const DwarfLLVMRegPair *EHDwarf2LRegs; // Dwarf to LLVM regs mapping EH
+ DenseMap<MCRegister, int> L2SEHRegs; // LLVM to SEH regs mapping
+ DenseMap<MCRegister, int> L2CVRegs; // LLVM to CV regs mapping
+
+public:
+ /// DiffListIterator - Base iterator class that can traverse the
+ /// differentially encoded register and regunit lists in DiffLists.
+ /// Don't use this class directly, use one of the specialized sub-classes
+ /// defined below.
+ class DiffListIterator {
+ uint16_t Val = 0;
+ const MCPhysReg *List = nullptr;
+
+ protected:
+ /// Create an invalid iterator. Call init() to point to something useful.
+ DiffListIterator() = default;
+
+ /// init - Point the iterator to InitVal, decoding subsequent values from
+ /// DiffList. The iterator will initially point to InitVal, sub-classes are
+ /// responsible for skipping the seed value if it is not part of the list.
+ void init(MCPhysReg InitVal, const MCPhysReg *DiffList) {
+ Val = InitVal;
+ List = DiffList;
+ }
+
+ /// advance - Move to the next list position, return the applied
+ /// differential. This function does not detect the end of the list, that
+ /// is the caller's responsibility (by checking for a 0 return value).
+ MCRegister advance() {
+ assert(isValid() && "Cannot move off the end of the list.");
+ MCPhysReg D = *List++;
+ Val += D;
+ return D;
+ }
+
+ public:
+ /// isValid - returns true if this iterator is not yet at the end.
+ bool isValid() const { return List; }
+
+ /// Dereference the iterator to get the value at the current position.
+ MCRegister operator*() const { return Val; }
+
+ /// Pre-increment to move to the next position.
+ void operator++() {
+ // The end of the list is encoded as a 0 differential.
+ if (!advance())
+ List = nullptr;
+ }
+ };
+
+ // These iterators are allowed to sub-class DiffListIterator and access
+ // internal list pointers.
+ friend class MCSubRegIterator;
+ friend class MCSubRegIndexIterator;
+ friend class MCSuperRegIterator;
+ friend class MCRegUnitIterator;
+ friend class MCRegUnitMaskIterator;
+ friend class MCRegUnitRootIterator;
+
+ /// Initialize MCRegisterInfo, called by TableGen
+ /// auto-generated routines. *DO NOT USE*.
+ void InitMCRegisterInfo(const MCRegisterDesc *D, unsigned NR, unsigned RA,
+ unsigned PC,
+ const MCRegisterClass *C, unsigned NC,
+ const MCPhysReg (*RURoots)[2],
+ unsigned NRU,
+ const MCPhysReg *DL,
+ const LaneBitmask *RUMS,
+ const char *Strings,
+ const char *ClassStrings,
+ const uint16_t *SubIndices,
+ unsigned NumIndices,
+ const SubRegCoveredBits *SubIdxRanges,
+ const uint16_t *RET) {
+ Desc = D;
+ NumRegs = NR;
+ RAReg = RA;
+ PCReg = PC;
+ Classes = C;
+ DiffLists = DL;
+ RegUnitMaskSequences = RUMS;
+ RegStrings = Strings;
+ RegClassStrings = ClassStrings;
+ NumClasses = NC;
+ RegUnitRoots = RURoots;
+ NumRegUnits = NRU;
+ SubRegIndices = SubIndices;
+ NumSubRegIndices = NumIndices;
+ SubRegIdxRanges = SubIdxRanges;
+ RegEncodingTable = RET;
+
+ // Initialize DWARF register mapping variables
+ EHL2DwarfRegs = nullptr;
+ EHL2DwarfRegsSize = 0;
+ L2DwarfRegs = nullptr;
+ L2DwarfRegsSize = 0;
+ EHDwarf2LRegs = nullptr;
+ EHDwarf2LRegsSize = 0;
+ Dwarf2LRegs = nullptr;
+ Dwarf2LRegsSize = 0;
+ }
+
+ /// Used to initialize LLVM register to Dwarf
+ /// register number mapping. Called by TableGen auto-generated routines.
+ /// *DO NOT USE*.
+ void mapLLVMRegsToDwarfRegs(const DwarfLLVMRegPair *Map, unsigned Size,
+ bool isEH) {
+ if (isEH) {
+ EHL2DwarfRegs = Map;
+ EHL2DwarfRegsSize = Size;
+ } else {
+ L2DwarfRegs = Map;
+ L2DwarfRegsSize = Size;
+ }
+ }
+
+ /// Used to initialize Dwarf register to LLVM
+ /// register number mapping. Called by TableGen auto-generated routines.
+ /// *DO NOT USE*.
+ void mapDwarfRegsToLLVMRegs(const DwarfLLVMRegPair *Map, unsigned Size,
+ bool isEH) {
+ if (isEH) {
+ EHDwarf2LRegs = Map;
+ EHDwarf2LRegsSize = Size;
+ } else {
+ Dwarf2LRegs = Map;
+ Dwarf2LRegsSize = Size;
+ }
+ }
+
+ /// mapLLVMRegToSEHReg - Used to initialize LLVM register to SEH register
+ /// number mapping. By default the SEH register number is just the same
+ /// as the LLVM register number.
+ /// FIXME: TableGen these numbers. Currently this requires target specific
+ /// initialization code.
+ void mapLLVMRegToSEHReg(MCRegister LLVMReg, int SEHReg) {
+ L2SEHRegs[LLVMReg] = SEHReg;
+ }
+
+ void mapLLVMRegToCVReg(MCRegister LLVMReg, int CVReg) {
+ L2CVRegs[LLVMReg] = CVReg;
+ }
+
+ /// This method should return the register where the return
+ /// address can be found.
+ MCRegister getRARegister() const {
+ return RAReg;
+ }
+
+ /// Return the register which is the program counter.
+ MCRegister getProgramCounter() const {
+ return PCReg;
+ }
+
+ const MCRegisterDesc &operator[](MCRegister RegNo) const {
+ assert(RegNo < NumRegs &&
+ "Attempting to access record for invalid register number!");
+ return Desc[RegNo];
+ }
+
+ /// Provide a get method, equivalent to [], but more useful with a
+ /// pointer to this object.
+ const MCRegisterDesc &get(MCRegister RegNo) const {
+ return operator[](RegNo);
+ }
+
+ /// Returns the physical register number of sub-register "Index"
+ /// for physical register RegNo. Return zero if the sub-register does not
+ /// exist.
+ MCRegister getSubReg(MCRegister Reg, unsigned Idx) const;
+
+ /// Return a super-register of the specified register
+ /// Reg so its sub-register of index SubIdx is Reg.
+ MCRegister getMatchingSuperReg(MCRegister Reg, unsigned SubIdx,
+ const MCRegisterClass *RC) const;
+
+ /// For a given register pair, return the sub-register index
+ /// if the second register is a sub-register of the first. Return zero
+ /// otherwise.
+ unsigned getSubRegIndex(MCRegister RegNo, MCRegister SubRegNo) const;
+
+ /// Get the size of the bit range covered by a sub-register index.
+ /// If the index isn't continuous, return the sum of the sizes of its parts.
+ /// If the index is used to access subregisters of different sizes, return -1.
+ unsigned getSubRegIdxSize(unsigned Idx) const;
+
+ /// Get the offset of the bit range covered by a sub-register index.
+ /// If an Offset doesn't make sense (the index isn't continuous, or is used to
+ /// access sub-registers at different offsets), return -1.
+ unsigned getSubRegIdxOffset(unsigned Idx) const;
+
+ /// Return the human-readable symbolic target-specific name for the
+ /// specified physical register.
+ const char *getName(MCRegister RegNo) const {
+ return RegStrings + get(RegNo).Name;
+ }
+
+ /// Return the number of registers this target has (useful for
+ /// sizing arrays holding per register information)
+ unsigned getNumRegs() const {
+ return NumRegs;
+ }
+
+ /// Return the number of sub-register indices
+ /// understood by the target. Index 0 is reserved for the no-op sub-register,
+ /// while 1 to getNumSubRegIndices() - 1 represent real sub-registers.
+ unsigned getNumSubRegIndices() const {
+ return NumSubRegIndices;
+ }
+
+ /// Return the number of (native) register units in the
+ /// target. Register units are numbered from 0 to getNumRegUnits() - 1. They
+ /// can be accessed through MCRegUnitIterator defined below.
+ unsigned getNumRegUnits() const {
+ return NumRegUnits;
+ }
+
+ /// Map a target register to an equivalent dwarf register
+ /// number. Returns -1 if there is no equivalent value. The second
+ /// parameter allows targets to use different numberings for EH info and
+ /// debugging info.
+ int getDwarfRegNum(MCRegister RegNum, bool isEH) const;
+
+ /// Map a dwarf register back to a target register. Returns None is there is
+ /// no mapping.
+ Optional<unsigned> getLLVMRegNum(unsigned RegNum, bool isEH) const;
+
+ /// Map a target EH register number to an equivalent DWARF register
+ /// number.
+ int getDwarfRegNumFromDwarfEHRegNum(unsigned RegNum) const;
+
+ /// Map a target register to an equivalent SEH register
+ /// number. Returns LLVM register number if there is no equivalent value.
+ int getSEHRegNum(MCRegister RegNum) const;
+
+ /// Map a target register to an equivalent CodeView register
+ /// number.
+ int getCodeViewRegNum(MCRegister RegNum) const;
+
+ regclass_iterator regclass_begin() const { return Classes; }
+ regclass_iterator regclass_end() const { return Classes+NumClasses; }
+ iterator_range<regclass_iterator> regclasses() const {
+ return make_range(regclass_begin(), regclass_end());
+ }
+
+ unsigned getNumRegClasses() const {
+ return (unsigned)(regclass_end()-regclass_begin());
+ }
+
+ /// Returns the register class associated with the enumeration
+ /// value. See class MCOperandInfo.
+ const MCRegisterClass& getRegClass(unsigned i) const {
+ assert(i < getNumRegClasses() && "Register Class ID out of range");
+ return Classes[i];
+ }
+
+ const char *getRegClassName(const MCRegisterClass *Class) const {
+ return RegClassStrings + Class->NameIdx;
+ }
+
+ /// Returns the encoding for RegNo
+ uint16_t getEncodingValue(MCRegister RegNo) const {
+ assert(RegNo < NumRegs &&
+ "Attempting to get encoding for invalid register number!");
+ return RegEncodingTable[RegNo];
+ }
+
+ /// Returns true if RegB is a sub-register of RegA.
+ bool isSubRegister(MCRegister RegA, MCRegister RegB) const {
+ return isSuperRegister(RegB, RegA);
+ }
+
+ /// Returns true if RegB is a super-register of RegA.
+ bool isSuperRegister(MCRegister RegA, MCRegister RegB) const;
+
+ /// Returns true if RegB is a sub-register of RegA or if RegB == RegA.
+ bool isSubRegisterEq(MCRegister RegA, MCRegister RegB) const {
+ return isSuperRegisterEq(RegB, RegA);
+ }
+
+ /// Returns true if RegB is a super-register of RegA or if
+ /// RegB == RegA.
+ bool isSuperRegisterEq(MCRegister RegA, MCRegister RegB) const {
+ return RegA == RegB || isSuperRegister(RegA, RegB);
+ }
+
+ /// Returns true if RegB is a super-register or sub-register of RegA
+ /// or if RegB == RegA.
+ bool isSuperOrSubRegisterEq(MCRegister RegA, MCRegister RegB) const {
+ return isSubRegisterEq(RegA, RegB) || isSuperRegister(RegA, RegB);
+ }
+};
+
+//===----------------------------------------------------------------------===//
+// Register List Iterators
+//===----------------------------------------------------------------------===//
+
+// MCRegisterInfo provides lists of super-registers, sub-registers, and
+// aliasing registers. Use these iterator classes to traverse the lists.
+
+/// MCSubRegIterator enumerates all sub-registers of Reg.
+/// If IncludeSelf is set, Reg itself is included in the list.
+class MCSubRegIterator : public MCRegisterInfo::DiffListIterator {
+public:
+ MCSubRegIterator(MCRegister Reg, const MCRegisterInfo *MCRI,
+ bool IncludeSelf = false) {
+ init(Reg, MCRI->DiffLists + MCRI->get(Reg).SubRegs);
+ // Initially, the iterator points to Reg itself.
+ if (!IncludeSelf)
+ ++*this;
+ }
+};
+
+/// Iterator that enumerates the sub-registers of a Reg and the associated
+/// sub-register indices.
+class MCSubRegIndexIterator {
+ MCSubRegIterator SRIter;
+ const uint16_t *SRIndex;
+
+public:
+ /// Constructs an iterator that traverses subregisters and their
+ /// associated subregister indices.
+ MCSubRegIndexIterator(MCRegister Reg, const MCRegisterInfo *MCRI)
+ : SRIter(Reg, MCRI) {
+ SRIndex = MCRI->SubRegIndices + MCRI->get(Reg).SubRegIndices;
+ }
+
+ /// Returns current sub-register.
+ MCRegister getSubReg() const {
+ return *SRIter;
+ }
+
+ /// Returns sub-register index of the current sub-register.
+ unsigned getSubRegIndex() const {
+ return *SRIndex;
+ }
+
+ /// Returns true if this iterator is not yet at the end.
+ bool isValid() const { return SRIter.isValid(); }
+
+ /// Moves to the next position.
+ void operator++() {
+ ++SRIter;
+ ++SRIndex;
+ }
+};
+
+/// MCSuperRegIterator enumerates all super-registers of Reg.
+/// If IncludeSelf is set, Reg itself is included in the list.
+class MCSuperRegIterator : public MCRegisterInfo::DiffListIterator {
+public:
+ MCSuperRegIterator() = default;
+
+ MCSuperRegIterator(MCRegister Reg, const MCRegisterInfo *MCRI,
+ bool IncludeSelf = false) {
+ init(Reg, MCRI->DiffLists + MCRI->get(Reg).SuperRegs);
+ // Initially, the iterator points to Reg itself.
+ if (!IncludeSelf)
+ ++*this;
+ }
+};
+
+// Definition for isSuperRegister. Put it down here since it needs the
+// iterator defined above in addition to the MCRegisterInfo class itself.
+inline bool MCRegisterInfo::isSuperRegister(MCRegister RegA, MCRegister RegB) const{
+ for (MCSuperRegIterator I(RegA, this); I.isValid(); ++I)
+ if (*I == RegB)
+ return true;
+ return false;
+}
+
+//===----------------------------------------------------------------------===//
+// Register Units
+//===----------------------------------------------------------------------===//
+
+// Register units are used to compute register aliasing. Every register has at
+// least one register unit, but it can have more. Two registers overlap if and
+// only if they have a common register unit.
+//
+// A target with a complicated sub-register structure will typically have many
+// fewer register units than actual registers. MCRI::getNumRegUnits() returns
+// the number of register units in the target.
+
+// MCRegUnitIterator enumerates a list of register units for Reg. The list is
+// in ascending numerical order.
+class MCRegUnitIterator : public MCRegisterInfo::DiffListIterator {
+public:
+ /// MCRegUnitIterator - Create an iterator that traverses the register units
+ /// in Reg.
+ MCRegUnitIterator() = default;
+
+ MCRegUnitIterator(MCRegister Reg, const MCRegisterInfo *MCRI) {
+ assert(Reg && "Null register has no regunits");
+ // Decode the RegUnits MCRegisterDesc field.
+ unsigned RU = MCRI->get(Reg).RegUnits;
+ unsigned Scale = RU & 15;
+ unsigned Offset = RU >> 4;
+
+ // Initialize the iterator to Reg * Scale, and the List pointer to
+ // DiffLists + Offset.
+ init(Reg * Scale, MCRI->DiffLists + Offset);
+
+ // That may not be a valid unit, we need to advance by one to get the real
+ // unit number. The first differential can be 0 which would normally
+ // terminate the list, but since we know every register has at least one
+ // unit, we can allow a 0 differential here.
+ advance();
+ }
+};
+
+/// MCRegUnitMaskIterator enumerates a list of register units and their
+/// associated lane masks for Reg. The register units are in ascending
+/// numerical order.
+class MCRegUnitMaskIterator {
+ MCRegUnitIterator RUIter;
+ const LaneBitmask *MaskListIter;
+
+public:
+ MCRegUnitMaskIterator() = default;
+
+ /// Constructs an iterator that traverses the register units and their
+ /// associated LaneMasks in Reg.
+ MCRegUnitMaskIterator(MCRegister Reg, const MCRegisterInfo *MCRI)
+ : RUIter(Reg, MCRI) {
+ uint16_t Idx = MCRI->get(Reg).RegUnitLaneMasks;
+ MaskListIter = &MCRI->RegUnitMaskSequences[Idx];
+ }
+
+ /// Returns a (RegUnit, LaneMask) pair.
+ std::pair<unsigned,LaneBitmask> operator*() const {
+ return std::make_pair(*RUIter, *MaskListIter);
+ }
+
+ /// Returns true if this iterator is not yet at the end.
+ bool isValid() const { return RUIter.isValid(); }
+
+ /// Moves to the next position.
+ void operator++() {
+ ++MaskListIter;
+ ++RUIter;
+ }
+};
+
+// Each register unit has one or two root registers. The complete set of
+// registers containing a register unit is the union of the roots and their
+// super-registers. All registers aliasing Unit can be visited like this:
+//
+// for (MCRegUnitRootIterator RI(Unit, MCRI); RI.isValid(); ++RI) {
+// for (MCSuperRegIterator SI(*RI, MCRI, true); SI.isValid(); ++SI)
+// visit(*SI);
+// }
+
+/// MCRegUnitRootIterator enumerates the root registers of a register unit.
+class MCRegUnitRootIterator {
+ uint16_t Reg0 = 0;
+ uint16_t Reg1 = 0;
+
+public:
+ MCRegUnitRootIterator() = default;
+
+ MCRegUnitRootIterator(unsigned RegUnit, const MCRegisterInfo *MCRI) {
+ assert(RegUnit < MCRI->getNumRegUnits() && "Invalid register unit");
+ Reg0 = MCRI->RegUnitRoots[RegUnit][0];
+ Reg1 = MCRI->RegUnitRoots[RegUnit][1];
+ }
+
+ /// Dereference to get the current root register.
+ unsigned operator*() const {
+ return Reg0;
+ }
+
+ /// Check if the iterator is at the end of the list.
+ bool isValid() const {
+ return Reg0;
+ }
+
+ /// Preincrement to move to the next root register.
+ void operator++() {
+ assert(isValid() && "Cannot move off the end of the list.");
+ Reg0 = Reg1;
+ Reg1 = 0;
+ }
+};
+
+/// MCRegAliasIterator enumerates all registers aliasing Reg. If IncludeSelf is
+/// set, Reg itself is included in the list. This iterator does not guarantee
+/// any ordering or that entries are unique.
+class MCRegAliasIterator {
+private:
+ MCRegister Reg;
+ const MCRegisterInfo *MCRI;
+ bool IncludeSelf;
+
+ MCRegUnitIterator RI;
+ MCRegUnitRootIterator RRI;
+ MCSuperRegIterator SI;
+
+public:
+ MCRegAliasIterator(MCRegister Reg, const MCRegisterInfo *MCRI,
+ bool IncludeSelf)
+ : Reg(Reg), MCRI(MCRI), IncludeSelf(IncludeSelf) {
+ // Initialize the iterators.
+ for (RI = MCRegUnitIterator(Reg, MCRI); RI.isValid(); ++RI) {
+ for (RRI = MCRegUnitRootIterator(*RI, MCRI); RRI.isValid(); ++RRI) {
+ for (SI = MCSuperRegIterator(*RRI, MCRI, true); SI.isValid(); ++SI) {
+ if (!(!IncludeSelf && Reg == *SI))
+ return;
+ }
+ }
+ }
+ }
+
+ bool isValid() const { return RI.isValid(); }
+
+ MCRegister operator*() const {
+ assert(SI.isValid() && "Cannot dereference an invalid iterator.");
+ return *SI;
+ }
+
+ void advance() {
+ // Assuming SI is valid.
+ ++SI;
+ if (SI.isValid()) return;
+
+ ++RRI;
+ if (RRI.isValid()) {
+ SI = MCSuperRegIterator(*RRI, MCRI, true);
+ return;
+ }
+
+ ++RI;
+ if (RI.isValid()) {
+ RRI = MCRegUnitRootIterator(*RI, MCRI);
+ SI = MCSuperRegIterator(*RRI, MCRI, true);
+ }
+ }
+
+ void operator++() {
+ assert(isValid() && "Cannot move off the end of the list.");
+ do advance();
+ while (!IncludeSelf && isValid() && *SI == Reg);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCREGISTERINFO_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCSymbol.h b/third_party/llvm-project/include/llvm/MC/MCSymbol.h
new file mode 100644
index 000000000..189484dea
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCSymbol.h
@@ -0,0 +1,441 @@
+//===- MCSymbol.h - Machine Code Symbols ------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains the declaration of the MCSymbol class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_MCSYMBOL_H
+#define LLVM_MC_MCSYMBOL_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/MC/MCFragment.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/MathExtras.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+
+namespace llvm {
+
+class MCAsmInfo;
+class MCContext;
+class MCExpr;
+class MCSection;
+class raw_ostream;
+
+/// MCSymbol - Instances of this class represent a symbol name in the MC file,
+/// and MCSymbols are created and uniqued by the MCContext class. MCSymbols
+/// should only be constructed with valid names for the object file.
+///
+/// If the symbol is defined/emitted into the current translation unit, the
+/// Section member is set to indicate what section it lives in. Otherwise, if
+/// it is a reference to an external entity, it has a null section.
+class MCSymbol {
+protected:
+ /// The kind of the symbol. If it is any value other than unset then this
+ /// class is actually one of the appropriate subclasses of MCSymbol.
+ enum SymbolKind {
+ SymbolKindUnset,
+ SymbolKindCOFF,
+ SymbolKindELF,
+ SymbolKindMachO,
+ SymbolKindWasm,
+ SymbolKindXCOFF,
+ };
+
+ /// A symbol can contain an Offset, or Value, or be Common, but never more
+ /// than one of these.
+ enum Contents : uint8_t {
+ SymContentsUnset,
+ SymContentsOffset,
+ SymContentsVariable,
+ SymContentsCommon,
+ SymContentsTargetCommon, // Index stores the section index
+ };
+
+ // Special sentinal value for the absolute pseudo fragment.
+ static MCFragment *AbsolutePseudoFragment;
+
+ /// If a symbol has a Fragment, the section is implied, so we only need
+ /// one pointer.
+ /// The special AbsolutePseudoFragment value is for absolute symbols.
+ /// If this is a variable symbol, this caches the variable value's fragment.
+ /// FIXME: We might be able to simplify this by having the asm streamer create
+ /// dummy fragments.
+ /// If this is a section, then it gives the symbol is defined in. This is null
+ /// for undefined symbols.
+ ///
+ /// If this is a fragment, then it gives the fragment this symbol's value is
+ /// relative to, if any.
+ ///
+ /// For the 'HasName' integer, this is true if this symbol is named.
+ /// A named symbol will have a pointer to the name allocated in the bytes
+ /// immediately prior to the MCSymbol.
+ mutable PointerIntPair<MCFragment *, 1> FragmentAndHasName;
+
+ /// IsTemporary - True if this is an assembler temporary label, which
+ /// typically does not survive in the .o file's symbol table. Usually
+ /// "Lfoo" or ".foo".
+ unsigned IsTemporary : 1;
+
+ /// True if this symbol can be redefined.
+ unsigned IsRedefinable : 1;
+
+ /// IsUsed - True if this symbol has been used.
+ mutable unsigned IsUsed : 1;
+
+ mutable unsigned IsRegistered : 1;
+
+ /// This symbol is visible outside this translation unit.
+ mutable unsigned IsExternal : 1;
+
+ /// This symbol is private extern.
+ mutable unsigned IsPrivateExtern : 1;
+
+ /// LLVM RTTI discriminator. This is actually a SymbolKind enumerator, but is
+ /// unsigned to avoid sign extension and achieve better bitpacking with MSVC.
+ unsigned Kind : 3;
+
+ /// True if we have created a relocation that uses this symbol.
+ mutable unsigned IsUsedInReloc : 1;
+
+ /// This is actually a Contents enumerator, but is unsigned to avoid sign
+ /// extension and achieve better bitpacking with MSVC.
+ unsigned SymbolContents : 3;
+
+ /// The alignment of the symbol, if it is 'common', or -1.
+ ///
+ /// The alignment is stored as log2(align) + 1. This allows all values from
+ /// 0 to 2^31 to be stored which is every power of 2 representable by an
+ /// unsigned.
+ enum : unsigned { NumCommonAlignmentBits = 5 };
+ unsigned CommonAlignLog2 : NumCommonAlignmentBits;
+
+ /// The Flags field is used by object file implementations to store
+ /// additional per symbol information which is not easily classified.
+ enum : unsigned { NumFlagsBits = 16 };
+ mutable uint32_t Flags : NumFlagsBits;
+
+ /// Index field, for use by the object file implementation.
+ mutable uint32_t Index = 0;
+
+ union {
+ /// The offset to apply to the fragment address to form this symbol's value.
+ uint64_t Offset;
+
+ /// The size of the symbol, if it is 'common'.
+ uint64_t CommonSize;
+
+ /// If non-null, the value for a variable symbol.
+ const MCExpr *Value;
+ };
+
+ // MCContext creates and uniques these.
+ friend class MCExpr;
+ friend class MCContext;
+
+ /// The name for a symbol.
+ /// MCSymbol contains a uint64_t so is probably aligned to 8. On a 32-bit
+ /// system, the name is a pointer so isn't going to satisfy the 8 byte
+ /// alignment of uint64_t. Account for that here.
+ using NameEntryStorageTy = union {
+ const StringMapEntry<bool> *NameEntry;
+ uint64_t AlignmentPadding;
+ };
+
+ MCSymbol(SymbolKind Kind, const StringMapEntry<bool> *Name, bool isTemporary)
+ : IsTemporary(isTemporary), IsRedefinable(false), IsUsed(false),
+ IsRegistered(false), IsExternal(false), IsPrivateExtern(false),
+ Kind(Kind), IsUsedInReloc(false), SymbolContents(SymContentsUnset),
+ CommonAlignLog2(0), Flags(0) {
+ Offset = 0;
+ FragmentAndHasName.setInt(!!Name);
+ if (Name)
+ getNameEntryPtr() = Name;
+ }
+
+ // Provide custom new/delete as we will only allocate space for a name
+ // if we need one.
+ void *operator new(size_t s, const StringMapEntry<bool> *Name,
+ MCContext &Ctx);
+
+private:
+ void operator delete(void *);
+ /// Placement delete - required by std, but never called.
+ void operator delete(void*, unsigned) {
+ llvm_unreachable("Constructor throws?");
+ }
+ /// Placement delete - required by std, but never called.
+ void operator delete(void*, unsigned, bool) {
+ llvm_unreachable("Constructor throws?");
+ }
+
+ MCSection *getSectionPtr() const {
+ if (MCFragment *F = getFragment()) {
+ assert(F != AbsolutePseudoFragment);
+ return F->getParent();
+ }
+ return nullptr;
+ }
+
+ /// Get a reference to the name field. Requires that we have a name
+ const StringMapEntry<bool> *&getNameEntryPtr() {
+ assert(FragmentAndHasName.getInt() && "Name is required");
+ NameEntryStorageTy *Name = reinterpret_cast<NameEntryStorageTy *>(this);
+ return (*(Name - 1)).NameEntry;
+ }
+ const StringMapEntry<bool> *&getNameEntryPtr() const {
+ return const_cast<MCSymbol*>(this)->getNameEntryPtr();
+ }
+
+public:
+ MCSymbol(const MCSymbol &) = delete;
+ MCSymbol &operator=(const MCSymbol &) = delete;
+
+ /// getName - Get the symbol name.
+ StringRef getName() const {
+ if (!FragmentAndHasName.getInt())
+ return StringRef();
+
+ return getNameEntryPtr()->first();
+ }
+
+ bool isRegistered() const { return IsRegistered; }
+ void setIsRegistered(bool Value) const { IsRegistered = Value; }
+
+ void setUsedInReloc() const { IsUsedInReloc = true; }
+ bool isUsedInReloc() const { return IsUsedInReloc; }
+
+ /// \name Accessors
+ /// @{
+
+ /// isTemporary - Check if this is an assembler temporary symbol.
+ bool isTemporary() const { return IsTemporary; }
+
+ /// isUsed - Check if this is used.
+ bool isUsed() const { return IsUsed; }
+
+ /// Check if this symbol is redefinable.
+ bool isRedefinable() const { return IsRedefinable; }
+ /// Mark this symbol as redefinable.
+ void setRedefinable(bool Value) { IsRedefinable = Value; }
+ /// Prepare this symbol to be redefined.
+ void redefineIfPossible() {
+ if (IsRedefinable) {
+ if (SymbolContents == SymContentsVariable) {
+ Value = nullptr;
+ SymbolContents = SymContentsUnset;
+ }
+ setUndefined();
+ IsRedefinable = false;
+ }
+ }
+
+ /// @}
+ /// \name Associated Sections
+ /// @{
+
+ /// isDefined - Check if this symbol is defined (i.e., it has an address).
+ ///
+ /// Defined symbols are either absolute or in some section.
+ bool isDefined() const { return !isUndefined(); }
+
+ /// isInSection - Check if this symbol is defined in some section (i.e., it
+ /// is defined but not absolute).
+ bool isInSection() const {
+ return isDefined() && !isAbsolute();
+ }
+
+ /// isUndefined - Check if this symbol undefined (i.e., implicitly defined).
+ bool isUndefined(bool SetUsed = true) const {
+ return getFragment(SetUsed) == nullptr;
+ }
+
+ /// isAbsolute - Check if this is an absolute symbol.
+ bool isAbsolute() const {
+ return getFragment() == AbsolutePseudoFragment;
+ }
+
+ /// Get the section associated with a defined, non-absolute symbol.
+ MCSection &getSection() const {
+ assert(isInSection() && "Invalid accessor!");
+ return *getSectionPtr();
+ }
+
+ /// Mark the symbol as defined in the fragment \p F.
+ void setFragment(MCFragment *F) const {
+ assert(!isVariable() && "Cannot set fragment of variable");
+ FragmentAndHasName.setPointer(F);
+ }
+
+ /// Mark the symbol as undefined.
+ void setUndefined() { FragmentAndHasName.setPointer(nullptr); }
+
+ bool isELF() const { return Kind == SymbolKindELF; }
+
+ bool isCOFF() const { return Kind == SymbolKindCOFF; }
+
+ bool isMachO() const { return Kind == SymbolKindMachO; }
+
+ bool isWasm() const { return Kind == SymbolKindWasm; }
+
+ bool isXCOFF() const { return Kind == SymbolKindXCOFF; }
+
+ /// @}
+ /// \name Variable Symbols
+ /// @{
+
+ /// isVariable - Check if this is a variable symbol.
+ bool isVariable() const {
+ return SymbolContents == SymContentsVariable;
+ }
+
+ /// getVariableValue - Get the value for variable symbols.
+ const MCExpr *getVariableValue(bool SetUsed = true) const {
+ assert(isVariable() && "Invalid accessor!");
+ IsUsed |= SetUsed;
+ return Value;
+ }
+
+ void setVariableValue(const MCExpr *Value);
+
+ /// @}
+
+ /// Get the (implementation defined) index.
+ uint32_t getIndex() const {
+ return Index;
+ }
+
+ /// Set the (implementation defined) index.
+ void setIndex(uint32_t Value) const {
+ Index = Value;
+ }
+
+ bool isUnset() const { return SymbolContents == SymContentsUnset; }
+
+ uint64_t getOffset() const {
+ assert((SymbolContents == SymContentsUnset ||
+ SymbolContents == SymContentsOffset) &&
+ "Cannot get offset for a common/variable symbol");
+ return Offset;
+ }
+ void setOffset(uint64_t Value) {
+ assert((SymbolContents == SymContentsUnset ||
+ SymbolContents == SymContentsOffset) &&
+ "Cannot set offset for a common/variable symbol");
+ Offset = Value;
+ SymbolContents = SymContentsOffset;
+ }
+
+ /// Return the size of a 'common' symbol.
+ uint64_t getCommonSize() const {
+ assert(isCommon() && "Not a 'common' symbol!");
+ return CommonSize;
+ }
+
+ /// Mark this symbol as being 'common'.
+ ///
+ /// \param Size - The size of the symbol.
+ /// \param Align - The alignment of the symbol.
+ /// \param Target - Is the symbol a target-specific common-like symbol.
+ void setCommon(uint64_t Size, unsigned Align, bool Target = false) {
+ assert(getOffset() == 0);
+ CommonSize = Size;
+ SymbolContents = Target ? SymContentsTargetCommon : SymContentsCommon;
+
+ assert((!Align || isPowerOf2_32(Align)) &&
+ "Alignment must be a power of 2");
+ unsigned Log2Align = Log2_32(Align) + 1;
+ assert(Log2Align < (1U << NumCommonAlignmentBits) &&
+ "Out of range alignment");
+ CommonAlignLog2 = Log2Align;
+ }
+
+ /// Return the alignment of a 'common' symbol.
+ unsigned getCommonAlignment() const {
+ assert(isCommon() && "Not a 'common' symbol!");
+ return CommonAlignLog2 ? (1U << (CommonAlignLog2 - 1)) : 0;
+ }
+
+ /// Declare this symbol as being 'common'.
+ ///
+ /// \param Size - The size of the symbol.
+ /// \param Align - The alignment of the symbol.
+ /// \param Target - Is the symbol a target-specific common-like symbol.
+ /// \return True if symbol was already declared as a different type
+ bool declareCommon(uint64_t Size, unsigned Align, bool Target = false) {
+ assert(isCommon() || getOffset() == 0);
+ if(isCommon()) {
+ if (CommonSize != Size || getCommonAlignment() != Align ||
+ isTargetCommon() != Target)
+ return true;
+ } else
+ setCommon(Size, Align, Target);
+ return false;
+ }
+
+ /// Is this a 'common' symbol.
+ bool isCommon() const {
+ return SymbolContents == SymContentsCommon ||
+ SymbolContents == SymContentsTargetCommon;
+ }
+
+ /// Is this a target-specific common-like symbol.
+ bool isTargetCommon() const {
+ return SymbolContents == SymContentsTargetCommon;
+ }
+
+ MCFragment *getFragment(bool SetUsed = true) const {
+ MCFragment *Fragment = FragmentAndHasName.getPointer();
+ if (Fragment || !isVariable())
+ return Fragment;
+ Fragment = getVariableValue(SetUsed)->findAssociatedFragment();
+ FragmentAndHasName.setPointer(Fragment);
+ return Fragment;
+ }
+
+ bool isExternal() const { return IsExternal; }
+ void setExternal(bool Value) const { IsExternal = Value; }
+
+ bool isPrivateExtern() const { return IsPrivateExtern; }
+ void setPrivateExtern(bool Value) { IsPrivateExtern = Value; }
+
+ /// print - Print the value to the stream \p OS.
+ void print(raw_ostream &OS, const MCAsmInfo *MAI) const;
+
+ /// dump - Print the value to stderr.
+ void dump() const;
+
+protected:
+ /// Get the (implementation defined) symbol flags.
+ uint32_t getFlags() const { return Flags; }
+
+ /// Set the (implementation defined) symbol flags.
+ void setFlags(uint32_t Value) const {
+ assert(Value < (1U << NumFlagsBits) && "Out of range flags");
+ Flags = Value;
+ }
+
+ /// Modify the flags via a mask
+ void modifyFlags(uint32_t Value, uint32_t Mask) const {
+ assert(Value < (1U << NumFlagsBits) && "Out of range flags");
+ Flags = (Flags & ~Mask) | Value;
+ }
+};
+
+inline raw_ostream &operator<<(raw_ostream &OS, const MCSymbol &Sym) {
+ Sym.print(OS, nullptr);
+ return OS;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCSYMBOL_H
diff --git a/third_party/llvm-project/include/llvm/MC/MCSymbolWasm.h b/third_party/llvm-project/include/llvm/MC/MCSymbolWasm.h
new file mode 100644
index 000000000..95beebe3f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/MCSymbolWasm.h
@@ -0,0 +1,110 @@
+//===- MCSymbolWasm.h - ----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+#ifndef LLVM_MC_MCSYMBOLWASM_H
+#define LLVM_MC_MCSYMBOLWASM_H
+
+#include "llvm/BinaryFormat/Wasm.h"
+#include "llvm/MC/MCSymbol.h"
+
+namespace llvm {
+
+class MCSymbolWasm : public MCSymbol {
+ wasm::WasmSymbolType Type = wasm::WASM_SYMBOL_TYPE_DATA;
+ bool IsWeak = false;
+ bool IsHidden = false;
+ bool IsComdat = false;
+ mutable bool IsUsedInGOT = false;
+ Optional<std::string> ImportModule;
+ Optional<std::string> ImportName;
+ wasm::WasmSignature *Signature = nullptr;
+ Optional<wasm::WasmGlobalType> GlobalType;
+ Optional<wasm::WasmEventType> EventType;
+
+ /// An expression describing how to calculate the size of a symbol. If a
+ /// symbol has no size this field will be NULL.
+ const MCExpr *SymbolSize = nullptr;
+
+public:
+ // Use a module name of "env" for now, for compatibility with existing tools.
+ // This is temporary, and may change, as the ABI is not yet stable.
+ MCSymbolWasm(const StringMapEntry<bool> *Name, bool isTemporary)
+ : MCSymbol(SymbolKindWasm, Name, isTemporary) {}
+ static bool classof(const MCSymbol *S) { return S->isWasm(); }
+
+ const MCExpr *getSize() const { return SymbolSize; }
+ void setSize(const MCExpr *SS) { SymbolSize = SS; }
+
+ bool isFunction() const { return Type == wasm::WASM_SYMBOL_TYPE_FUNCTION; }
+ bool isData() const { return Type == wasm::WASM_SYMBOL_TYPE_DATA; }
+ bool isGlobal() const { return Type == wasm::WASM_SYMBOL_TYPE_GLOBAL; }
+ bool isSection() const { return Type == wasm::WASM_SYMBOL_TYPE_SECTION; }
+ bool isEvent() const { return Type == wasm::WASM_SYMBOL_TYPE_EVENT; }
+ wasm::WasmSymbolType getType() const { return Type; }
+ void setType(wasm::WasmSymbolType type) { Type = type; }
+
+ bool isExported() const {
+ return getFlags() & wasm::WASM_SYMBOL_EXPORTED;
+ }
+ void setExported() const {
+ modifyFlags(wasm::WASM_SYMBOL_EXPORTED, wasm::WASM_SYMBOL_EXPORTED);
+ }
+
+ bool isNoStrip() const {
+ return getFlags() & wasm::WASM_SYMBOL_NO_STRIP;
+ }
+ void setNoStrip() const {
+ modifyFlags(wasm::WASM_SYMBOL_NO_STRIP, wasm::WASM_SYMBOL_NO_STRIP);
+ }
+
+ bool isWeak() const { return IsWeak; }
+ void setWeak(bool isWeak) { IsWeak = isWeak; }
+
+ bool isHidden() const { return IsHidden; }
+ void setHidden(bool isHidden) { IsHidden = isHidden; }
+
+ bool isComdat() const { return IsComdat; }
+ void setComdat(bool isComdat) { IsComdat = isComdat; }
+
+ const StringRef getImportModule() const {
+ if (ImportModule.hasValue()) {
+ return ImportModule.getValue();
+ }
+ return "env";
+ }
+ void setImportModule(StringRef Name) { ImportModule = Name; }
+
+ const StringRef getImportName() const {
+ if (ImportName.hasValue()) {
+ return ImportName.getValue();
+ }
+ return getName();
+ }
+ void setImportName(StringRef Name) { ImportName = Name; }
+
+ void setUsedInGOT() const { IsUsedInGOT = true; }
+ bool isUsedInGOT() const { return IsUsedInGOT; }
+
+ const wasm::WasmSignature *getSignature() const { return Signature; }
+ void setSignature(wasm::WasmSignature *Sig) { Signature = Sig; }
+
+ const wasm::WasmGlobalType &getGlobalType() const {
+ assert(GlobalType.hasValue());
+ return GlobalType.getValue();
+ }
+ void setGlobalType(wasm::WasmGlobalType GT) { GlobalType = GT; }
+
+ const wasm::WasmEventType &getEventType() const {
+ assert(EventType.hasValue());
+ return EventType.getValue();
+ }
+ void setEventType(wasm::WasmEventType ET) { EventType = ET; }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_MC_MCSYMBOLWASM_H
diff --git a/third_party/llvm-project/include/llvm/MC/SubtargetFeature.h b/third_party/llvm-project/include/llvm/MC/SubtargetFeature.h
new file mode 100644
index 000000000..defbc3c64
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/MC/SubtargetFeature.h
@@ -0,0 +1,236 @@
+//===- llvm/MC/SubtargetFeature.h - CPU characteristics ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+/// \file Defines and manages user or tool specified CPU characteristics.
+/// The intent is to be able to package specific features that should or should
+/// not be used on a specific target processor. A tool, such as llc, could, as
+/// as example, gather chip info from the command line, a long with features
+/// that should be used on that chip.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_MC_SUBTARGETFEATURE_H
+#define LLVM_MC_SUBTARGETFEATURE_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/MathExtras.h"
+#include <array>
+#include <bitset>
+#include <initializer_list>
+#include <string>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+class Triple;
+
+const unsigned MAX_SUBTARGET_WORDS = 3;
+const unsigned MAX_SUBTARGET_FEATURES = MAX_SUBTARGET_WORDS * 64;
+
+/// Container class for subtarget features.
+/// This is a constexpr reimplementation of a subset of std::bitset. It would be
+/// nice to use std::bitset directly, but it doesn't support constant
+/// initialization.
+class FeatureBitset {
+ static_assert((MAX_SUBTARGET_FEATURES % 64) == 0,
+ "Should be a multiple of 64!");
+ // This cannot be a std::array, operator[] is not constexpr until C++17.
+ uint64_t Bits[MAX_SUBTARGET_WORDS] = {};
+
+protected:
+ constexpr FeatureBitset(const std::array<uint64_t, MAX_SUBTARGET_WORDS> &B) {
+ for (unsigned I = 0; I != B.size(); ++I)
+ Bits[I] = B[I];
+ }
+
+public:
+ constexpr FeatureBitset() = default;
+ constexpr FeatureBitset(std::initializer_list<unsigned> Init) {
+ for (auto I : Init)
+ set(I);
+ }
+
+ FeatureBitset &set() {
+ std::fill(std::begin(Bits), std::end(Bits), -1ULL);
+ return *this;
+ }
+
+ constexpr FeatureBitset &set(unsigned I) {
+ // GCC <6.2 crashes if this is written in a single statement.
+ uint64_t NewBits = Bits[I / 64] | (uint64_t(1) << (I % 64));
+ Bits[I / 64] = NewBits;
+ return *this;
+ }
+
+ constexpr FeatureBitset &reset(unsigned I) {
+ // GCC <6.2 crashes if this is written in a single statement.
+ uint64_t NewBits = Bits[I / 64] & ~(uint64_t(1) << (I % 64));
+ Bits[I / 64] = NewBits;
+ return *this;
+ }
+
+ constexpr FeatureBitset &flip(unsigned I) {
+ // GCC <6.2 crashes if this is written in a single statement.
+ uint64_t NewBits = Bits[I / 64] ^ (uint64_t(1) << (I % 64));
+ Bits[I / 64] = NewBits;
+ return *this;
+ }
+
+ constexpr bool operator[](unsigned I) const {
+ uint64_t Mask = uint64_t(1) << (I % 64);
+ return (Bits[I / 64] & Mask) != 0;
+ }
+
+ constexpr bool test(unsigned I) const { return (*this)[I]; }
+
+ constexpr size_t size() const { return MAX_SUBTARGET_FEATURES; }
+
+ bool any() const {
+ return llvm::any_of(Bits, [](uint64_t I) { return I != 0; });
+ }
+ bool none() const { return !any(); }
+ size_t count() const {
+ size_t Count = 0;
+ for (auto B : Bits)
+ Count += countPopulation(B);
+ return Count;
+ }
+
+ constexpr FeatureBitset &operator^=(const FeatureBitset &RHS) {
+ for (unsigned I = 0, E = array_lengthof(Bits); I != E; ++I) {
+ Bits[I] ^= RHS.Bits[I];
+ }
+ return *this;
+ }
+ constexpr FeatureBitset operator^(const FeatureBitset &RHS) const {
+ FeatureBitset Result = *this;
+ Result ^= RHS;
+ return Result;
+ }
+
+ constexpr FeatureBitset &operator&=(const FeatureBitset &RHS) {
+ for (unsigned I = 0, E = array_lengthof(Bits); I != E; ++I) {
+ Bits[I] &= RHS.Bits[I];
+ }
+ return *this;
+ }
+ constexpr FeatureBitset operator&(const FeatureBitset &RHS) const {
+ FeatureBitset Result = *this;
+ Result &= RHS;
+ return Result;
+ }
+
+ constexpr FeatureBitset &operator|=(const FeatureBitset &RHS) {
+ for (unsigned I = 0, E = array_lengthof(Bits); I != E; ++I) {
+ Bits[I] |= RHS.Bits[I];
+ }
+ return *this;
+ }
+ constexpr FeatureBitset operator|(const FeatureBitset &RHS) const {
+ FeatureBitset Result = *this;
+ Result |= RHS;
+ return Result;
+ }
+
+ constexpr FeatureBitset operator~() const {
+ FeatureBitset Result = *this;
+ for (auto &B : Result.Bits)
+ B = ~B;
+ return Result;
+ }
+
+ bool operator==(const FeatureBitset &RHS) const {
+ return std::equal(std::begin(Bits), std::end(Bits), std::begin(RHS.Bits));
+ }
+
+ bool operator!=(const FeatureBitset &RHS) const { return !(*this == RHS); }
+
+ bool operator < (const FeatureBitset &Other) const {
+ for (unsigned I = 0, E = size(); I != E; ++I) {
+ bool LHS = test(I), RHS = Other.test(I);
+ if (LHS != RHS)
+ return LHS < RHS;
+ }
+ return false;
+ }
+};
+
+/// Class used to store the subtarget bits in the tables created by tablegen.
+class FeatureBitArray : public FeatureBitset {
+public:
+ constexpr FeatureBitArray(const std::array<uint64_t, MAX_SUBTARGET_WORDS> &B)
+ : FeatureBitset(B) {}
+
+ const FeatureBitset &getAsBitset() const { return *this; }
+};
+
+//===----------------------------------------------------------------------===//
+
+/// Manages the enabling and disabling of subtarget specific features.
+///
+/// Features are encoded as a string of the form
+/// "+attr1,+attr2,-attr3,...,+attrN"
+/// A comma separates each feature from the next (all lowercase.)
+/// Each of the remaining features is prefixed with + or - indicating whether
+/// that feature should be enabled or disabled contrary to the cpu
+/// specification.
+class SubtargetFeatures {
+ std::vector<std::string> Features; ///< Subtarget features as a vector
+
+public:
+ explicit SubtargetFeatures(StringRef Initial = "");
+
+ /// Returns features as a string.
+ std::string getString() const;
+
+ /// Adds Features.
+ void AddFeature(StringRef String, bool Enable = true);
+
+ /// Returns the vector of individual subtarget features.
+ const std::vector<std::string> &getFeatures() const { return Features; }
+
+ /// Prints feature string.
+ void print(raw_ostream &OS) const;
+
+ // Dumps feature info.
+ void dump() const;
+
+ /// Adds the default features for the specified target triple.
+ void getDefaultSubtargetFeatures(const Triple& Triple);
+
+ /// Determine if a feature has a flag; '+' or '-'
+ static bool hasFlag(StringRef Feature) {
+ assert(!Feature.empty() && "Empty string");
+ // Get first character
+ char Ch = Feature[0];
+ // Check if first character is '+' or '-' flag
+ return Ch == '+' || Ch =='-';
+ }
+
+ /// Return string stripped of flag.
+ static std::string StripFlag(StringRef Feature) {
+ return hasFlag(Feature) ? Feature.substr(1) : Feature;
+ }
+
+ /// Return true if enable flag; '+'.
+ static inline bool isEnabled(StringRef Feature) {
+ assert(!Feature.empty() && "Empty string");
+ // Get first character
+ char Ch = Feature[0];
+ // Check if first character is '+' for enabled
+ return Ch == '+';
+ }
+
+ /// Splits a string of comma separated items in to a vector of strings.
+ static void Split(std::vector<std::string> &V, StringRef S);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_MC_SUBTARGETFEATURE_H
diff --git a/third_party/llvm-project/include/llvm/Object/Archive.h b/third_party/llvm-project/include/llvm/Object/Archive.h
new file mode 100644
index 000000000..c3f36bdd9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Archive.h
@@ -0,0 +1,286 @@
+//===- Archive.h - ar archive file format -----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the ar archive file format class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_ARCHIVE_H
+#define LLVM_OBJECT_ARCHIVE_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/fallible_iterator.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Support/Chrono.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+#include <memory>
+#include <string>
+#include <vector>
+
+namespace llvm {
+namespace object {
+
+class Archive;
+
+class ArchiveMemberHeader {
+public:
+ friend class Archive;
+
+ ArchiveMemberHeader(Archive const *Parent, const char *RawHeaderPtr,
+ uint64_t Size, Error *Err);
+ // ArchiveMemberHeader() = default;
+
+ /// Get the name without looking up long names.
+ Expected<StringRef> getRawName() const;
+
+ /// Get the name looking up long names.
+ Expected<StringRef> getName(uint64_t Size) const;
+
+ Expected<uint64_t> getSize() const;
+
+ Expected<sys::fs::perms> getAccessMode() const;
+ Expected<sys::TimePoint<std::chrono::seconds>> getLastModified() const;
+
+ StringRef getRawLastModified() const {
+ return StringRef(ArMemHdr->LastModified,
+ sizeof(ArMemHdr->LastModified)).rtrim(' ');
+ }
+
+ Expected<unsigned> getUID() const;
+ Expected<unsigned> getGID() const;
+
+ // This returns the size of the private struct ArMemHdrType
+ uint64_t getSizeOf() const {
+ return sizeof(ArMemHdrType);
+ }
+
+private:
+ struct ArMemHdrType {
+ char Name[16];
+ char LastModified[12];
+ char UID[6];
+ char GID[6];
+ char AccessMode[8];
+ char Size[10]; ///< Size of data, not including header or padding.
+ char Terminator[2];
+ };
+ Archive const *Parent;
+ ArMemHdrType const *ArMemHdr;
+};
+
+class Archive : public Binary {
+ virtual void anchor();
+
+public:
+ class Child {
+ friend Archive;
+ friend ArchiveMemberHeader;
+
+ const Archive *Parent;
+ ArchiveMemberHeader Header;
+ /// Includes header but not padding byte.
+ StringRef Data;
+ /// Offset from Data to the start of the file.
+ uint16_t StartOfFile;
+
+ Expected<bool> isThinMember() const;
+
+ public:
+ Child(const Archive *Parent, const char *Start, Error *Err);
+ Child(const Archive *Parent, StringRef Data, uint16_t StartOfFile);
+
+ bool operator ==(const Child &other) const {
+ assert(!Parent || !other.Parent || Parent == other.Parent);
+ return Data.begin() == other.Data.begin();
+ }
+
+ const Archive *getParent() const { return Parent; }
+ Expected<Child> getNext() const;
+
+ Expected<StringRef> getName() const;
+ Expected<std::string> getFullName() const;
+ Expected<StringRef> getRawName() const { return Header.getRawName(); }
+
+ Expected<sys::TimePoint<std::chrono::seconds>> getLastModified() const {
+ return Header.getLastModified();
+ }
+
+ StringRef getRawLastModified() const {
+ return Header.getRawLastModified();
+ }
+
+ Expected<unsigned> getUID() const { return Header.getUID(); }
+ Expected<unsigned> getGID() const { return Header.getGID(); }
+
+ Expected<sys::fs::perms> getAccessMode() const {
+ return Header.getAccessMode();
+ }
+
+ /// \return the size of the archive member without the header or padding.
+ Expected<uint64_t> getSize() const;
+ /// \return the size in the archive header for this member.
+ Expected<uint64_t> getRawSize() const;
+
+ Expected<StringRef> getBuffer() const;
+ uint64_t getChildOffset() const;
+ uint64_t getDataOffset() const { return getChildOffset() + StartOfFile; }
+
+ Expected<MemoryBufferRef> getMemoryBufferRef() const;
+
+ Expected<std::unique_ptr<Binary>>
+ getAsBinary(LLVMContext *Context = nullptr) const;
+ };
+
+ class ChildFallibleIterator {
+ Child C;
+
+ public:
+ ChildFallibleIterator() : C(Child(nullptr, nullptr, nullptr)) {}
+ ChildFallibleIterator(const Child &C) : C(C) {}
+
+ const Child *operator->() const { return &C; }
+ const Child &operator*() const { return C; }
+
+ bool operator==(const ChildFallibleIterator &other) const {
+ // Ignore errors here: If an error occurred during increment then getNext
+ // will have been set to child_end(), and the following comparison should
+ // do the right thing.
+ return C == other.C;
+ }
+
+ bool operator!=(const ChildFallibleIterator &other) const {
+ return !(*this == other);
+ }
+
+ Error inc() {
+ auto NextChild = C.getNext();
+ if (!NextChild)
+ return NextChild.takeError();
+ C = std::move(*NextChild);
+ return Error::success();
+ }
+ };
+
+ using child_iterator = fallible_iterator<ChildFallibleIterator>;
+
+ class Symbol {
+ const Archive *Parent;
+ uint32_t SymbolIndex;
+ uint32_t StringIndex; // Extra index to the string.
+
+ public:
+ Symbol(const Archive *p, uint32_t symi, uint32_t stri)
+ : Parent(p)
+ , SymbolIndex(symi)
+ , StringIndex(stri) {}
+
+ bool operator ==(const Symbol &other) const {
+ return (Parent == other.Parent) && (SymbolIndex == other.SymbolIndex);
+ }
+
+ StringRef getName() const;
+ Expected<Child> getMember() const;
+ Symbol getNext() const;
+ };
+
+ class symbol_iterator {
+ Symbol symbol;
+
+ public:
+ symbol_iterator(const Symbol &s) : symbol(s) {}
+
+ const Symbol *operator->() const { return &symbol; }
+ const Symbol &operator*() const { return symbol; }
+
+ bool operator==(const symbol_iterator &other) const {
+ return symbol == other.symbol;
+ }
+
+ bool operator!=(const symbol_iterator &other) const {
+ return !(*this == other);
+ }
+
+ symbol_iterator& operator++() { // Preincrement
+ symbol = symbol.getNext();
+ return *this;
+ }
+ };
+
+ Archive(MemoryBufferRef Source, Error &Err);
+ static Expected<std::unique_ptr<Archive>> create(MemoryBufferRef Source);
+
+ /// Size field is 10 decimal digits long
+ static const uint64_t MaxMemberSize = 9999999999;
+
+ enum Kind {
+ K_GNU,
+ K_GNU64,
+ K_BSD,
+ K_DARWIN,
+ K_DARWIN64,
+ K_COFF
+ };
+
+ Kind kind() const { return (Kind)Format; }
+ bool isThin() const { return IsThin; }
+
+ child_iterator child_begin(Error &Err, bool SkipInternal = true) const;
+ child_iterator child_end() const;
+ iterator_range<child_iterator> children(Error &Err,
+ bool SkipInternal = true) const {
+ return make_range(child_begin(Err, SkipInternal), child_end());
+ }
+
+ symbol_iterator symbol_begin() const;
+ symbol_iterator symbol_end() const;
+ iterator_range<symbol_iterator> symbols() const {
+ return make_range(symbol_begin(), symbol_end());
+ }
+
+ // Cast methods.
+ static bool classof(Binary const *v) {
+ return v->isArchive();
+ }
+
+ // check if a symbol is in the archive
+ Expected<Optional<Child>> findSym(StringRef name) const;
+
+ bool isEmpty() const;
+ bool hasSymbolTable() const;
+ StringRef getSymbolTable() const { return SymbolTable; }
+ StringRef getStringTable() const { return StringTable; }
+ uint32_t getNumberOfSymbols() const;
+
+ std::vector<std::unique_ptr<MemoryBuffer>> takeThinBuffers() {
+ return std::move(ThinBuffers);
+ }
+
+private:
+ StringRef SymbolTable;
+ StringRef StringTable;
+
+ StringRef FirstRegularData;
+ uint16_t FirstRegularStartOfFile = -1;
+ void setFirstRegular(const Child &C);
+
+ unsigned Format : 3;
+ unsigned IsThin : 1;
+ mutable std::vector<std::unique_ptr<MemoryBuffer>> ThinBuffers;
+};
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_ARCHIVE_H
diff --git a/third_party/llvm-project/include/llvm/Object/Binary.h b/third_party/llvm-project/include/llvm/Object/Binary.h
new file mode 100644
index 000000000..aa5e718f5
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Binary.h
@@ -0,0 +1,237 @@
+//===- Binary.h - A generic binary file -------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the Binary class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_BINARY_H
+#define LLVM_OBJECT_BINARY_H
+
+#include "llvm-c/Types.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <algorithm>
+#include <memory>
+#include <utility>
+
+namespace llvm {
+
+class LLVMContext;
+class StringRef;
+
+namespace object {
+
+class Binary {
+private:
+ unsigned int TypeID;
+
+protected:
+ MemoryBufferRef Data;
+
+ Binary(unsigned int Type, MemoryBufferRef Source);
+
+ enum {
+ ID_Archive,
+ ID_MachOUniversalBinary,
+ ID_COFFImportFile,
+ ID_IR, // LLVM IR
+ ID_TapiUniversal, // Text-based Dynamic Library Stub file.
+ ID_TapiFile, // Text-based Dynamic Library Stub file.
+
+ ID_Minidump,
+
+ ID_WinRes, // Windows resource (.res) file.
+
+ // Object and children.
+ ID_StartObjects,
+ ID_COFF,
+
+ ID_XCOFF32, // AIX XCOFF 32-bit
+ ID_XCOFF64, // AIX XCOFF 64-bit
+
+ ID_ELF32L, // ELF 32-bit, little endian
+ ID_ELF32B, // ELF 32-bit, big endian
+ ID_ELF64L, // ELF 64-bit, little endian
+ ID_ELF64B, // ELF 64-bit, big endian
+
+ ID_MachO32L, // MachO 32-bit, little endian
+ ID_MachO32B, // MachO 32-bit, big endian
+ ID_MachO64L, // MachO 64-bit, little endian
+ ID_MachO64B, // MachO 64-bit, big endian
+
+ ID_Wasm,
+
+ ID_EndObjects
+ };
+
+ static inline unsigned int getELFType(bool isLE, bool is64Bits) {
+ if (isLE)
+ return is64Bits ? ID_ELF64L : ID_ELF32L;
+ else
+ return is64Bits ? ID_ELF64B : ID_ELF32B;
+ }
+
+ static unsigned int getMachOType(bool isLE, bool is64Bits) {
+ if (isLE)
+ return is64Bits ? ID_MachO64L : ID_MachO32L;
+ else
+ return is64Bits ? ID_MachO64B : ID_MachO32B;
+ }
+
+public:
+ Binary() = delete;
+ Binary(const Binary &other) = delete;
+ virtual ~Binary();
+
+ StringRef getData() const;
+ StringRef getFileName() const;
+ MemoryBufferRef getMemoryBufferRef() const;
+
+ // Cast methods.
+ unsigned int getType() const { return TypeID; }
+
+ // Convenience methods
+ bool isObject() const {
+ return TypeID > ID_StartObjects && TypeID < ID_EndObjects;
+ }
+
+ bool isSymbolic() const {
+ return isIR() || isObject() || isCOFFImportFile() || isTapiFile();
+ }
+
+ bool isArchive() const { return TypeID == ID_Archive; }
+
+ bool isMachOUniversalBinary() const {
+ return TypeID == ID_MachOUniversalBinary;
+ }
+
+ bool isTapiUniversal() const { return TypeID == ID_TapiUniversal; }
+
+ bool isELF() const {
+ return TypeID >= ID_ELF32L && TypeID <= ID_ELF64B;
+ }
+
+ bool isMachO() const {
+ return TypeID >= ID_MachO32L && TypeID <= ID_MachO64B;
+ }
+
+ bool isCOFF() const {
+ return TypeID == ID_COFF;
+ }
+
+ bool isXCOFF() const { return TypeID == ID_XCOFF32 || TypeID == ID_XCOFF64; }
+
+ bool isWasm() const { return TypeID == ID_Wasm; }
+
+ bool isCOFFImportFile() const {
+ return TypeID == ID_COFFImportFile;
+ }
+
+ bool isIR() const {
+ return TypeID == ID_IR;
+ }
+
+ bool isMinidump() const { return TypeID == ID_Minidump; }
+
+ bool isTapiFile() const { return TypeID == ID_TapiFile; }
+
+ bool isLittleEndian() const {
+ return !(TypeID == ID_ELF32B || TypeID == ID_ELF64B ||
+ TypeID == ID_MachO32B || TypeID == ID_MachO64B);
+ }
+
+ bool isWinRes() const { return TypeID == ID_WinRes; }
+
+ Triple::ObjectFormatType getTripleObjectFormat() const {
+ if (isCOFF())
+ return Triple::COFF;
+ if (isMachO())
+ return Triple::MachO;
+ if (isELF())
+ return Triple::ELF;
+ return Triple::UnknownObjectFormat;
+ }
+
+ static std::error_code checkOffset(MemoryBufferRef M, uintptr_t Addr,
+ const uint64_t Size) {
+ if (Addr + Size < Addr || Addr + Size < Size ||
+ Addr + Size > uintptr_t(M.getBufferEnd()) ||
+ Addr < uintptr_t(M.getBufferStart())) {
+ return object_error::unexpected_eof;
+ }
+ return std::error_code();
+ }
+};
+
+// Create wrappers for C Binding types (see CBindingWrapping.h).
+DEFINE_ISA_CONVERSION_FUNCTIONS(Binary, LLVMBinaryRef)
+
+/// Create a Binary from Source, autodetecting the file type.
+///
+/// @param Source The data to create the Binary from.
+Expected<std::unique_ptr<Binary>> createBinary(MemoryBufferRef Source,
+ LLVMContext *Context = nullptr);
+
+template <typename T> class OwningBinary {
+ std::unique_ptr<T> Bin;
+ std::unique_ptr<MemoryBuffer> Buf;
+
+public:
+ OwningBinary();
+ OwningBinary(std::unique_ptr<T> Bin, std::unique_ptr<MemoryBuffer> Buf);
+ OwningBinary(OwningBinary<T>&& Other);
+ OwningBinary<T> &operator=(OwningBinary<T> &&Other);
+
+ std::pair<std::unique_ptr<T>, std::unique_ptr<MemoryBuffer>> takeBinary();
+
+ T* getBinary();
+ const T* getBinary() const;
+};
+
+template <typename T>
+OwningBinary<T>::OwningBinary(std::unique_ptr<T> Bin,
+ std::unique_ptr<MemoryBuffer> Buf)
+ : Bin(std::move(Bin)), Buf(std::move(Buf)) {}
+
+template <typename T> OwningBinary<T>::OwningBinary() = default;
+
+template <typename T>
+OwningBinary<T>::OwningBinary(OwningBinary &&Other)
+ : Bin(std::move(Other.Bin)), Buf(std::move(Other.Buf)) {}
+
+template <typename T>
+OwningBinary<T> &OwningBinary<T>::operator=(OwningBinary &&Other) {
+ Bin = std::move(Other.Bin);
+ Buf = std::move(Other.Buf);
+ return *this;
+}
+
+template <typename T>
+std::pair<std::unique_ptr<T>, std::unique_ptr<MemoryBuffer>>
+OwningBinary<T>::takeBinary() {
+ return std::make_pair(std::move(Bin), std::move(Buf));
+}
+
+template <typename T> T* OwningBinary<T>::getBinary() {
+ return Bin.get();
+}
+
+template <typename T> const T* OwningBinary<T>::getBinary() const {
+ return Bin.get();
+}
+
+Expected<OwningBinary<Binary>> createBinary(StringRef Path);
+
+} // end namespace object
+
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_BINARY_H
diff --git a/third_party/llvm-project/include/llvm/Object/COFF.h b/third_party/llvm-project/include/llvm/Object/COFF.h
new file mode 100644
index 000000000..b91ee5887
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/COFF.h
@@ -0,0 +1,1269 @@
+//===- COFF.h - COFF object file implementation -----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the COFFObjectFile class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_COFF_H
+#define LLVM_OBJECT_COFF_H
+
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/COFF.h"
+#include "llvm/MC/SubtargetFeature.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/CVDebugRecord.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/BinaryByteStream.h"
+#include "llvm/Support/ConvertUTF.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <system_error>
+
+namespace llvm {
+
+template <typename T> class ArrayRef;
+
+namespace object {
+
+class BaseRelocRef;
+class DelayImportDirectoryEntryRef;
+class ExportDirectoryEntryRef;
+class ImportDirectoryEntryRef;
+class ImportedSymbolRef;
+class ResourceSectionRef;
+
+using import_directory_iterator = content_iterator<ImportDirectoryEntryRef>;
+using delay_import_directory_iterator =
+ content_iterator<DelayImportDirectoryEntryRef>;
+using export_directory_iterator = content_iterator<ExportDirectoryEntryRef>;
+using imported_symbol_iterator = content_iterator<ImportedSymbolRef>;
+using base_reloc_iterator = content_iterator<BaseRelocRef>;
+
+/// The DOS compatible header at the front of all PE/COFF executables.
+struct dos_header {
+ char Magic[2];
+ support::ulittle16_t UsedBytesInTheLastPage;
+ support::ulittle16_t FileSizeInPages;
+ support::ulittle16_t NumberOfRelocationItems;
+ support::ulittle16_t HeaderSizeInParagraphs;
+ support::ulittle16_t MinimumExtraParagraphs;
+ support::ulittle16_t MaximumExtraParagraphs;
+ support::ulittle16_t InitialRelativeSS;
+ support::ulittle16_t InitialSP;
+ support::ulittle16_t Checksum;
+ support::ulittle16_t InitialIP;
+ support::ulittle16_t InitialRelativeCS;
+ support::ulittle16_t AddressOfRelocationTable;
+ support::ulittle16_t OverlayNumber;
+ support::ulittle16_t Reserved[4];
+ support::ulittle16_t OEMid;
+ support::ulittle16_t OEMinfo;
+ support::ulittle16_t Reserved2[10];
+ support::ulittle32_t AddressOfNewExeHeader;
+};
+
+struct coff_file_header {
+ support::ulittle16_t Machine;
+ support::ulittle16_t NumberOfSections;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle32_t PointerToSymbolTable;
+ support::ulittle32_t NumberOfSymbols;
+ support::ulittle16_t SizeOfOptionalHeader;
+ support::ulittle16_t Characteristics;
+
+ bool isImportLibrary() const { return NumberOfSections == 0xffff; }
+};
+
+struct coff_bigobj_file_header {
+ support::ulittle16_t Sig1;
+ support::ulittle16_t Sig2;
+ support::ulittle16_t Version;
+ support::ulittle16_t Machine;
+ support::ulittle32_t TimeDateStamp;
+ uint8_t UUID[16];
+ support::ulittle32_t unused1;
+ support::ulittle32_t unused2;
+ support::ulittle32_t unused3;
+ support::ulittle32_t unused4;
+ support::ulittle32_t NumberOfSections;
+ support::ulittle32_t PointerToSymbolTable;
+ support::ulittle32_t NumberOfSymbols;
+};
+
+/// The 32-bit PE header that follows the COFF header.
+struct pe32_header {
+ support::ulittle16_t Magic;
+ uint8_t MajorLinkerVersion;
+ uint8_t MinorLinkerVersion;
+ support::ulittle32_t SizeOfCode;
+ support::ulittle32_t SizeOfInitializedData;
+ support::ulittle32_t SizeOfUninitializedData;
+ support::ulittle32_t AddressOfEntryPoint;
+ support::ulittle32_t BaseOfCode;
+ support::ulittle32_t BaseOfData;
+ support::ulittle32_t ImageBase;
+ support::ulittle32_t SectionAlignment;
+ support::ulittle32_t FileAlignment;
+ support::ulittle16_t MajorOperatingSystemVersion;
+ support::ulittle16_t MinorOperatingSystemVersion;
+ support::ulittle16_t MajorImageVersion;
+ support::ulittle16_t MinorImageVersion;
+ support::ulittle16_t MajorSubsystemVersion;
+ support::ulittle16_t MinorSubsystemVersion;
+ support::ulittle32_t Win32VersionValue;
+ support::ulittle32_t SizeOfImage;
+ support::ulittle32_t SizeOfHeaders;
+ support::ulittle32_t CheckSum;
+ support::ulittle16_t Subsystem;
+ // FIXME: This should be DllCharacteristics.
+ support::ulittle16_t DLLCharacteristics;
+ support::ulittle32_t SizeOfStackReserve;
+ support::ulittle32_t SizeOfStackCommit;
+ support::ulittle32_t SizeOfHeapReserve;
+ support::ulittle32_t SizeOfHeapCommit;
+ support::ulittle32_t LoaderFlags;
+ // FIXME: This should be NumberOfRvaAndSizes.
+ support::ulittle32_t NumberOfRvaAndSize;
+};
+
+/// The 64-bit PE header that follows the COFF header.
+struct pe32plus_header {
+ support::ulittle16_t Magic;
+ uint8_t MajorLinkerVersion;
+ uint8_t MinorLinkerVersion;
+ support::ulittle32_t SizeOfCode;
+ support::ulittle32_t SizeOfInitializedData;
+ support::ulittle32_t SizeOfUninitializedData;
+ support::ulittle32_t AddressOfEntryPoint;
+ support::ulittle32_t BaseOfCode;
+ support::ulittle64_t ImageBase;
+ support::ulittle32_t SectionAlignment;
+ support::ulittle32_t FileAlignment;
+ support::ulittle16_t MajorOperatingSystemVersion;
+ support::ulittle16_t MinorOperatingSystemVersion;
+ support::ulittle16_t MajorImageVersion;
+ support::ulittle16_t MinorImageVersion;
+ support::ulittle16_t MajorSubsystemVersion;
+ support::ulittle16_t MinorSubsystemVersion;
+ support::ulittle32_t Win32VersionValue;
+ support::ulittle32_t SizeOfImage;
+ support::ulittle32_t SizeOfHeaders;
+ support::ulittle32_t CheckSum;
+ support::ulittle16_t Subsystem;
+ support::ulittle16_t DLLCharacteristics;
+ support::ulittle64_t SizeOfStackReserve;
+ support::ulittle64_t SizeOfStackCommit;
+ support::ulittle64_t SizeOfHeapReserve;
+ support::ulittle64_t SizeOfHeapCommit;
+ support::ulittle32_t LoaderFlags;
+ support::ulittle32_t NumberOfRvaAndSize;
+};
+
+struct data_directory {
+ support::ulittle32_t RelativeVirtualAddress;
+ support::ulittle32_t Size;
+};
+
+struct debug_directory {
+ support::ulittle32_t Characteristics;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle16_t MajorVersion;
+ support::ulittle16_t MinorVersion;
+ support::ulittle32_t Type;
+ support::ulittle32_t SizeOfData;
+ support::ulittle32_t AddressOfRawData;
+ support::ulittle32_t PointerToRawData;
+};
+
+template <typename IntTy>
+struct import_lookup_table_entry {
+ IntTy Data;
+
+ bool isOrdinal() const { return Data < 0; }
+
+ uint16_t getOrdinal() const {
+ assert(isOrdinal() && "ILT entry is not an ordinal!");
+ return Data & 0xFFFF;
+ }
+
+ uint32_t getHintNameRVA() const {
+ assert(!isOrdinal() && "ILT entry is not a Hint/Name RVA!");
+ return Data & 0xFFFFFFFF;
+ }
+};
+
+using import_lookup_table_entry32 =
+ import_lookup_table_entry<support::little32_t>;
+using import_lookup_table_entry64 =
+ import_lookup_table_entry<support::little64_t>;
+
+struct delay_import_directory_table_entry {
+ // dumpbin reports this field as "Characteristics" instead of "Attributes".
+ support::ulittle32_t Attributes;
+ support::ulittle32_t Name;
+ support::ulittle32_t ModuleHandle;
+ support::ulittle32_t DelayImportAddressTable;
+ support::ulittle32_t DelayImportNameTable;
+ support::ulittle32_t BoundDelayImportTable;
+ support::ulittle32_t UnloadDelayImportTable;
+ support::ulittle32_t TimeStamp;
+};
+
+struct export_directory_table_entry {
+ support::ulittle32_t ExportFlags;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle16_t MajorVersion;
+ support::ulittle16_t MinorVersion;
+ support::ulittle32_t NameRVA;
+ support::ulittle32_t OrdinalBase;
+ support::ulittle32_t AddressTableEntries;
+ support::ulittle32_t NumberOfNamePointers;
+ support::ulittle32_t ExportAddressTableRVA;
+ support::ulittle32_t NamePointerRVA;
+ support::ulittle32_t OrdinalTableRVA;
+};
+
+union export_address_table_entry {
+ support::ulittle32_t ExportRVA;
+ support::ulittle32_t ForwarderRVA;
+};
+
+using export_name_pointer_table_entry = support::ulittle32_t;
+using export_ordinal_table_entry = support::ulittle16_t;
+
+struct StringTableOffset {
+ support::ulittle32_t Zeroes;
+ support::ulittle32_t Offset;
+};
+
+template <typename SectionNumberType>
+struct coff_symbol {
+ union {
+ char ShortName[COFF::NameSize];
+ StringTableOffset Offset;
+ } Name;
+
+ support::ulittle32_t Value;
+ SectionNumberType SectionNumber;
+
+ support::ulittle16_t Type;
+
+ uint8_t StorageClass;
+ uint8_t NumberOfAuxSymbols;
+};
+
+using coff_symbol16 = coff_symbol<support::ulittle16_t>;
+using coff_symbol32 = coff_symbol<support::ulittle32_t>;
+
+// Contains only common parts of coff_symbol16 and coff_symbol32.
+struct coff_symbol_generic {
+ union {
+ char ShortName[COFF::NameSize];
+ StringTableOffset Offset;
+ } Name;
+ support::ulittle32_t Value;
+};
+
+struct coff_aux_section_definition;
+struct coff_aux_weak_external;
+
+class COFFSymbolRef {
+public:
+ COFFSymbolRef() = default;
+ COFFSymbolRef(const coff_symbol16 *CS) : CS16(CS) {}
+ COFFSymbolRef(const coff_symbol32 *CS) : CS32(CS) {}
+
+ const void *getRawPtr() const {
+ return CS16 ? static_cast<const void *>(CS16) : CS32;
+ }
+
+ const coff_symbol_generic *getGeneric() const {
+ if (CS16)
+ return reinterpret_cast<const coff_symbol_generic *>(CS16);
+ return reinterpret_cast<const coff_symbol_generic *>(CS32);
+ }
+
+ friend bool operator<(COFFSymbolRef A, COFFSymbolRef B) {
+ return A.getRawPtr() < B.getRawPtr();
+ }
+
+ bool isBigObj() const {
+ if (CS16)
+ return false;
+ if (CS32)
+ return true;
+ llvm_unreachable("COFFSymbolRef points to nothing!");
+ }
+
+ const char *getShortName() const {
+ return CS16 ? CS16->Name.ShortName : CS32->Name.ShortName;
+ }
+
+ const StringTableOffset &getStringTableOffset() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ return CS16 ? CS16->Name.Offset : CS32->Name.Offset;
+ }
+
+ uint32_t getValue() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ return CS16 ? CS16->Value : CS32->Value;
+ }
+
+ int32_t getSectionNumber() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ if (CS16) {
+ // Reserved sections are returned as negative numbers.
+ if (CS16->SectionNumber <= COFF::MaxNumberOfSections16)
+ return CS16->SectionNumber;
+ return static_cast<int16_t>(CS16->SectionNumber);
+ }
+ return static_cast<int32_t>(CS32->SectionNumber);
+ }
+
+ uint16_t getType() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ return CS16 ? CS16->Type : CS32->Type;
+ }
+
+ uint8_t getStorageClass() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ return CS16 ? CS16->StorageClass : CS32->StorageClass;
+ }
+
+ uint8_t getNumberOfAuxSymbols() const {
+ assert(isSet() && "COFFSymbolRef points to nothing!");
+ return CS16 ? CS16->NumberOfAuxSymbols : CS32->NumberOfAuxSymbols;
+ }
+
+ uint8_t getBaseType() const { return getType() & 0x0F; }
+
+ uint8_t getComplexType() const {
+ return (getType() & 0xF0) >> COFF::SCT_COMPLEX_TYPE_SHIFT;
+ }
+
+ template <typename T> const T *getAux() const {
+ return CS16 ? reinterpret_cast<const T *>(CS16 + 1)
+ : reinterpret_cast<const T *>(CS32 + 1);
+ }
+
+ const coff_aux_section_definition *getSectionDefinition() const {
+ if (!getNumberOfAuxSymbols() ||
+ getStorageClass() != COFF::IMAGE_SYM_CLASS_STATIC)
+ return nullptr;
+ return getAux<coff_aux_section_definition>();
+ }
+
+ const coff_aux_weak_external *getWeakExternal() const {
+ if (!getNumberOfAuxSymbols() ||
+ getStorageClass() != COFF::IMAGE_SYM_CLASS_WEAK_EXTERNAL)
+ return nullptr;
+ return getAux<coff_aux_weak_external>();
+ }
+
+ bool isAbsolute() const {
+ return getSectionNumber() == -1;
+ }
+
+ bool isExternal() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_EXTERNAL;
+ }
+
+ bool isCommon() const {
+ return isExternal() && getSectionNumber() == COFF::IMAGE_SYM_UNDEFINED &&
+ getValue() != 0;
+ }
+
+ bool isUndefined() const {
+ return isExternal() && getSectionNumber() == COFF::IMAGE_SYM_UNDEFINED &&
+ getValue() == 0;
+ }
+
+ bool isWeakExternal() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_WEAK_EXTERNAL;
+ }
+
+ bool isFunctionDefinition() const {
+ return isExternal() && getBaseType() == COFF::IMAGE_SYM_TYPE_NULL &&
+ getComplexType() == COFF::IMAGE_SYM_DTYPE_FUNCTION &&
+ !COFF::isReservedSectionNumber(getSectionNumber());
+ }
+
+ bool isFunctionLineInfo() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_FUNCTION;
+ }
+
+ bool isAnyUndefined() const {
+ return isUndefined() || isWeakExternal();
+ }
+
+ bool isFileRecord() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_FILE;
+ }
+
+ bool isSection() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_SECTION;
+ }
+
+ bool isSectionDefinition() const {
+ // C++/CLI creates external ABS symbols for non-const appdomain globals.
+ // These are also followed by an auxiliary section definition.
+ bool isAppdomainGlobal =
+ getStorageClass() == COFF::IMAGE_SYM_CLASS_EXTERNAL &&
+ getSectionNumber() == COFF::IMAGE_SYM_ABSOLUTE;
+ bool isOrdinarySection = getStorageClass() == COFF::IMAGE_SYM_CLASS_STATIC;
+ if (!getNumberOfAuxSymbols())
+ return false;
+ return isAppdomainGlobal || isOrdinarySection;
+ }
+
+ bool isCLRToken() const {
+ return getStorageClass() == COFF::IMAGE_SYM_CLASS_CLR_TOKEN;
+ }
+
+private:
+ bool isSet() const { return CS16 || CS32; }
+
+ const coff_symbol16 *CS16 = nullptr;
+ const coff_symbol32 *CS32 = nullptr;
+};
+
+struct coff_section {
+ char Name[COFF::NameSize];
+ support::ulittle32_t VirtualSize;
+ support::ulittle32_t VirtualAddress;
+ support::ulittle32_t SizeOfRawData;
+ support::ulittle32_t PointerToRawData;
+ support::ulittle32_t PointerToRelocations;
+ support::ulittle32_t PointerToLinenumbers;
+ support::ulittle16_t NumberOfRelocations;
+ support::ulittle16_t NumberOfLinenumbers;
+ support::ulittle32_t Characteristics;
+
+ // Returns true if the actual number of relocations is stored in
+ // VirtualAddress field of the first relocation table entry.
+ bool hasExtendedRelocations() const {
+ return (Characteristics & COFF::IMAGE_SCN_LNK_NRELOC_OVFL) &&
+ NumberOfRelocations == UINT16_MAX;
+ }
+
+ uint32_t getAlignment() const {
+ // The IMAGE_SCN_TYPE_NO_PAD bit is a legacy way of getting to
+ // IMAGE_SCN_ALIGN_1BYTES.
+ if (Characteristics & COFF::IMAGE_SCN_TYPE_NO_PAD)
+ return 1;
+
+ // Bit [20:24] contains section alignment. 0 means use a default alignment
+ // of 16.
+ uint32_t Shift = (Characteristics >> 20) & 0xF;
+ if (Shift > 0)
+ return 1U << (Shift - 1);
+ return 16;
+ }
+};
+
+struct coff_relocation {
+ support::ulittle32_t VirtualAddress;
+ support::ulittle32_t SymbolTableIndex;
+ support::ulittle16_t Type;
+};
+
+struct coff_aux_function_definition {
+ support::ulittle32_t TagIndex;
+ support::ulittle32_t TotalSize;
+ support::ulittle32_t PointerToLinenumber;
+ support::ulittle32_t PointerToNextFunction;
+ char Unused1[2];
+};
+
+static_assert(sizeof(coff_aux_function_definition) == 18,
+ "auxiliary entry must be 18 bytes");
+
+struct coff_aux_bf_and_ef_symbol {
+ char Unused1[4];
+ support::ulittle16_t Linenumber;
+ char Unused2[6];
+ support::ulittle32_t PointerToNextFunction;
+ char Unused3[2];
+};
+
+static_assert(sizeof(coff_aux_bf_and_ef_symbol) == 18,
+ "auxiliary entry must be 18 bytes");
+
+struct coff_aux_weak_external {
+ support::ulittle32_t TagIndex;
+ support::ulittle32_t Characteristics;
+ char Unused1[10];
+};
+
+static_assert(sizeof(coff_aux_weak_external) == 18,
+ "auxiliary entry must be 18 bytes");
+
+struct coff_aux_section_definition {
+ support::ulittle32_t Length;
+ support::ulittle16_t NumberOfRelocations;
+ support::ulittle16_t NumberOfLinenumbers;
+ support::ulittle32_t CheckSum;
+ support::ulittle16_t NumberLowPart;
+ uint8_t Selection;
+ uint8_t Unused;
+ support::ulittle16_t NumberHighPart;
+ int32_t getNumber(bool IsBigObj) const {
+ uint32_t Number = static_cast<uint32_t>(NumberLowPart);
+ if (IsBigObj)
+ Number |= static_cast<uint32_t>(NumberHighPart) << 16;
+ return static_cast<int32_t>(Number);
+ }
+};
+
+static_assert(sizeof(coff_aux_section_definition) == 18,
+ "auxiliary entry must be 18 bytes");
+
+struct coff_aux_clr_token {
+ uint8_t AuxType;
+ uint8_t Reserved;
+ support::ulittle32_t SymbolTableIndex;
+ char MBZ[12];
+};
+
+static_assert(sizeof(coff_aux_clr_token) == 18,
+ "auxiliary entry must be 18 bytes");
+
+struct coff_import_header {
+ support::ulittle16_t Sig1;
+ support::ulittle16_t Sig2;
+ support::ulittle16_t Version;
+ support::ulittle16_t Machine;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle32_t SizeOfData;
+ support::ulittle16_t OrdinalHint;
+ support::ulittle16_t TypeInfo;
+
+ int getType() const { return TypeInfo & 0x3; }
+ int getNameType() const { return (TypeInfo >> 2) & 0x7; }
+};
+
+struct coff_import_directory_table_entry {
+ support::ulittle32_t ImportLookupTableRVA;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle32_t ForwarderChain;
+ support::ulittle32_t NameRVA;
+ support::ulittle32_t ImportAddressTableRVA;
+
+ bool isNull() const {
+ return ImportLookupTableRVA == 0 && TimeDateStamp == 0 &&
+ ForwarderChain == 0 && NameRVA == 0 && ImportAddressTableRVA == 0;
+ }
+};
+
+template <typename IntTy>
+struct coff_tls_directory {
+ IntTy StartAddressOfRawData;
+ IntTy EndAddressOfRawData;
+ IntTy AddressOfIndex;
+ IntTy AddressOfCallBacks;
+ support::ulittle32_t SizeOfZeroFill;
+ support::ulittle32_t Characteristics;
+
+ uint32_t getAlignment() const {
+ // Bit [20:24] contains section alignment.
+ uint32_t Shift = (Characteristics & 0x00F00000) >> 20;
+ if (Shift > 0)
+ return 1U << (Shift - 1);
+ return 0;
+ }
+};
+
+using coff_tls_directory32 = coff_tls_directory<support::little32_t>;
+using coff_tls_directory64 = coff_tls_directory<support::little64_t>;
+
+/// Bits in control flow guard flags as we understand them.
+enum class coff_guard_flags : uint32_t {
+ CFInstrumented = 0x00000100,
+ HasFidTable = 0x00000400,
+ ProtectDelayLoadIAT = 0x00001000,
+ DelayLoadIATSection = 0x00002000, // Delay load in separate section
+ HasLongJmpTable = 0x00010000,
+ FidTableHasFlags = 0x10000000, // Indicates that fid tables are 5 bytes
+};
+
+enum class frame_type : uint16_t { Fpo = 0, Trap = 1, Tss = 2, NonFpo = 3 };
+
+struct coff_load_config_code_integrity {
+ support::ulittle16_t Flags;
+ support::ulittle16_t Catalog;
+ support::ulittle32_t CatalogOffset;
+ support::ulittle32_t Reserved;
+};
+
+/// 32-bit load config (IMAGE_LOAD_CONFIG_DIRECTORY32)
+struct coff_load_configuration32 {
+ support::ulittle32_t Size;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle16_t MajorVersion;
+ support::ulittle16_t MinorVersion;
+ support::ulittle32_t GlobalFlagsClear;
+ support::ulittle32_t GlobalFlagsSet;
+ support::ulittle32_t CriticalSectionDefaultTimeout;
+ support::ulittle32_t DeCommitFreeBlockThreshold;
+ support::ulittle32_t DeCommitTotalFreeThreshold;
+ support::ulittle32_t LockPrefixTable;
+ support::ulittle32_t MaximumAllocationSize;
+ support::ulittle32_t VirtualMemoryThreshold;
+ support::ulittle32_t ProcessAffinityMask;
+ support::ulittle32_t ProcessHeapFlags;
+ support::ulittle16_t CSDVersion;
+ support::ulittle16_t DependentLoadFlags;
+ support::ulittle32_t EditList;
+ support::ulittle32_t SecurityCookie;
+ support::ulittle32_t SEHandlerTable;
+ support::ulittle32_t SEHandlerCount;
+
+ // Added in MSVC 2015 for /guard:cf.
+ support::ulittle32_t GuardCFCheckFunction;
+ support::ulittle32_t GuardCFCheckDispatch;
+ support::ulittle32_t GuardCFFunctionTable;
+ support::ulittle32_t GuardCFFunctionCount;
+ support::ulittle32_t GuardFlags; // coff_guard_flags
+
+ // Added in MSVC 2017
+ coff_load_config_code_integrity CodeIntegrity;
+ support::ulittle32_t GuardAddressTakenIatEntryTable;
+ support::ulittle32_t GuardAddressTakenIatEntryCount;
+ support::ulittle32_t GuardLongJumpTargetTable;
+ support::ulittle32_t GuardLongJumpTargetCount;
+ support::ulittle32_t DynamicValueRelocTable;
+ support::ulittle32_t CHPEMetadataPointer;
+ support::ulittle32_t GuardRFFailureRoutine;
+ support::ulittle32_t GuardRFFailureRoutineFunctionPointer;
+ support::ulittle32_t DynamicValueRelocTableOffset;
+ support::ulittle16_t DynamicValueRelocTableSection;
+ support::ulittle16_t Reserved2;
+ support::ulittle32_t GuardRFVerifyStackPointerFunctionPointer;
+ support::ulittle32_t HotPatchTableOffset;
+};
+
+/// 64-bit load config (IMAGE_LOAD_CONFIG_DIRECTORY64)
+struct coff_load_configuration64 {
+ support::ulittle32_t Size;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle16_t MajorVersion;
+ support::ulittle16_t MinorVersion;
+ support::ulittle32_t GlobalFlagsClear;
+ support::ulittle32_t GlobalFlagsSet;
+ support::ulittle32_t CriticalSectionDefaultTimeout;
+ support::ulittle64_t DeCommitFreeBlockThreshold;
+ support::ulittle64_t DeCommitTotalFreeThreshold;
+ support::ulittle64_t LockPrefixTable;
+ support::ulittle64_t MaximumAllocationSize;
+ support::ulittle64_t VirtualMemoryThreshold;
+ support::ulittle64_t ProcessAffinityMask;
+ support::ulittle32_t ProcessHeapFlags;
+ support::ulittle16_t CSDVersion;
+ support::ulittle16_t DependentLoadFlags;
+ support::ulittle64_t EditList;
+ support::ulittle64_t SecurityCookie;
+ support::ulittle64_t SEHandlerTable;
+ support::ulittle64_t SEHandlerCount;
+
+ // Added in MSVC 2015 for /guard:cf.
+ support::ulittle64_t GuardCFCheckFunction;
+ support::ulittle64_t GuardCFCheckDispatch;
+ support::ulittle64_t GuardCFFunctionTable;
+ support::ulittle64_t GuardCFFunctionCount;
+ support::ulittle32_t GuardFlags;
+
+ // Added in MSVC 2017
+ coff_load_config_code_integrity CodeIntegrity;
+ support::ulittle64_t GuardAddressTakenIatEntryTable;
+ support::ulittle64_t GuardAddressTakenIatEntryCount;
+ support::ulittle64_t GuardLongJumpTargetTable;
+ support::ulittle64_t GuardLongJumpTargetCount;
+ support::ulittle64_t DynamicValueRelocTable;
+ support::ulittle64_t CHPEMetadataPointer;
+ support::ulittle64_t GuardRFFailureRoutine;
+ support::ulittle64_t GuardRFFailureRoutineFunctionPointer;
+ support::ulittle32_t DynamicValueRelocTableOffset;
+ support::ulittle16_t DynamicValueRelocTableSection;
+ support::ulittle16_t Reserved2;
+ support::ulittle64_t GuardRFVerifyStackPointerFunctionPointer;
+ support::ulittle32_t HotPatchTableOffset;
+};
+
+struct coff_runtime_function_x64 {
+ support::ulittle32_t BeginAddress;
+ support::ulittle32_t EndAddress;
+ support::ulittle32_t UnwindInformation;
+};
+
+struct coff_base_reloc_block_header {
+ support::ulittle32_t PageRVA;
+ support::ulittle32_t BlockSize;
+};
+
+struct coff_base_reloc_block_entry {
+ support::ulittle16_t Data;
+
+ int getType() const { return Data >> 12; }
+ int getOffset() const { return Data & ((1 << 12) - 1); }
+};
+
+struct coff_resource_dir_entry {
+ union {
+ support::ulittle32_t NameOffset;
+ support::ulittle32_t ID;
+ uint32_t getNameOffset() const {
+ return maskTrailingOnes<uint32_t>(31) & NameOffset;
+ }
+ // Even though the PE/COFF spec doesn't mention this, the high bit of a name
+ // offset is set.
+ void setNameOffset(uint32_t Offset) { NameOffset = Offset | (1 << 31); }
+ } Identifier;
+ union {
+ support::ulittle32_t DataEntryOffset;
+ support::ulittle32_t SubdirOffset;
+
+ bool isSubDir() const { return SubdirOffset >> 31; }
+ uint32_t value() const {
+ return maskTrailingOnes<uint32_t>(31) & SubdirOffset;
+ }
+
+ } Offset;
+};
+
+struct coff_resource_data_entry {
+ support::ulittle32_t DataRVA;
+ support::ulittle32_t DataSize;
+ support::ulittle32_t Codepage;
+ support::ulittle32_t Reserved;
+};
+
+struct coff_resource_dir_table {
+ support::ulittle32_t Characteristics;
+ support::ulittle32_t TimeDateStamp;
+ support::ulittle16_t MajorVersion;
+ support::ulittle16_t MinorVersion;
+ support::ulittle16_t NumberOfNameEntries;
+ support::ulittle16_t NumberOfIDEntries;
+};
+
+struct debug_h_header {
+ support::ulittle32_t Magic;
+ support::ulittle16_t Version;
+ support::ulittle16_t HashAlgorithm;
+};
+
+class COFFObjectFile : public ObjectFile {
+private:
+ friend class ImportDirectoryEntryRef;
+ friend class ExportDirectoryEntryRef;
+ const coff_file_header *COFFHeader;
+ const coff_bigobj_file_header *COFFBigObjHeader;
+ const pe32_header *PE32Header;
+ const pe32plus_header *PE32PlusHeader;
+ const data_directory *DataDirectory;
+ const coff_section *SectionTable;
+ const coff_symbol16 *SymbolTable16;
+ const coff_symbol32 *SymbolTable32;
+ const char *StringTable;
+ uint32_t StringTableSize;
+ const coff_import_directory_table_entry *ImportDirectory;
+ const delay_import_directory_table_entry *DelayImportDirectory;
+ uint32_t NumberOfDelayImportDirectory;
+ const export_directory_table_entry *ExportDirectory;
+ const coff_base_reloc_block_header *BaseRelocHeader;
+ const coff_base_reloc_block_header *BaseRelocEnd;
+ const debug_directory *DebugDirectoryBegin;
+ const debug_directory *DebugDirectoryEnd;
+ // Either coff_load_configuration32 or coff_load_configuration64.
+ const void *LoadConfig = nullptr;
+
+ std::error_code getString(uint32_t offset, StringRef &Res) const;
+
+ template <typename coff_symbol_type>
+ const coff_symbol_type *toSymb(DataRefImpl Symb) const;
+ const coff_section *toSec(DataRefImpl Sec) const;
+ const coff_relocation *toRel(DataRefImpl Rel) const;
+
+ std::error_code initSymbolTablePtr();
+ std::error_code initImportTablePtr();
+ std::error_code initDelayImportTablePtr();
+ std::error_code initExportTablePtr();
+ std::error_code initBaseRelocPtr();
+ std::error_code initDebugDirectoryPtr();
+ std::error_code initLoadConfigPtr();
+
+public:
+ uintptr_t getSymbolTable() const {
+ if (SymbolTable16)
+ return reinterpret_cast<uintptr_t>(SymbolTable16);
+ if (SymbolTable32)
+ return reinterpret_cast<uintptr_t>(SymbolTable32);
+ return uintptr_t(0);
+ }
+
+ uint16_t getMachine() const {
+ if (COFFHeader)
+ return COFFHeader->Machine;
+ if (COFFBigObjHeader)
+ return COFFBigObjHeader->Machine;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint16_t getSizeOfOptionalHeader() const {
+ if (COFFHeader)
+ return COFFHeader->isImportLibrary() ? 0
+ : COFFHeader->SizeOfOptionalHeader;
+ // bigobj doesn't have this field.
+ if (COFFBigObjHeader)
+ return 0;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint16_t getCharacteristics() const {
+ if (COFFHeader)
+ return COFFHeader->isImportLibrary() ? 0 : COFFHeader->Characteristics;
+ // bigobj doesn't have characteristics to speak of,
+ // editbin will silently lie to you if you attempt to set any.
+ if (COFFBigObjHeader)
+ return 0;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint32_t getTimeDateStamp() const {
+ if (COFFHeader)
+ return COFFHeader->TimeDateStamp;
+ if (COFFBigObjHeader)
+ return COFFBigObjHeader->TimeDateStamp;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint32_t getNumberOfSections() const {
+ if (COFFHeader)
+ return COFFHeader->isImportLibrary() ? 0 : COFFHeader->NumberOfSections;
+ if (COFFBigObjHeader)
+ return COFFBigObjHeader->NumberOfSections;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint32_t getPointerToSymbolTable() const {
+ if (COFFHeader)
+ return COFFHeader->isImportLibrary() ? 0
+ : COFFHeader->PointerToSymbolTable;
+ if (COFFBigObjHeader)
+ return COFFBigObjHeader->PointerToSymbolTable;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint32_t getRawNumberOfSymbols() const {
+ if (COFFHeader)
+ return COFFHeader->isImportLibrary() ? 0 : COFFHeader->NumberOfSymbols;
+ if (COFFBigObjHeader)
+ return COFFBigObjHeader->NumberOfSymbols;
+ llvm_unreachable("no COFF header!");
+ }
+
+ uint32_t getNumberOfSymbols() const {
+ if (!SymbolTable16 && !SymbolTable32)
+ return 0;
+ return getRawNumberOfSymbols();
+ }
+
+ const coff_load_configuration32 *getLoadConfig32() const {
+ assert(!is64());
+ return reinterpret_cast<const coff_load_configuration32 *>(LoadConfig);
+ }
+
+ const coff_load_configuration64 *getLoadConfig64() const {
+ assert(is64());
+ return reinterpret_cast<const coff_load_configuration64 *>(LoadConfig);
+ }
+ StringRef getRelocationTypeName(uint16_t Type) const;
+
+protected:
+ void moveSymbolNext(DataRefImpl &Symb) const override;
+ Expected<StringRef> getSymbolName(DataRefImpl Symb) const override;
+ Expected<uint64_t> getSymbolAddress(DataRefImpl Symb) const override;
+ uint32_t getSymbolAlignment(DataRefImpl Symb) const override;
+ uint64_t getSymbolValueImpl(DataRefImpl Symb) const override;
+ uint64_t getCommonSymbolSizeImpl(DataRefImpl Symb) const override;
+ uint32_t getSymbolFlags(DataRefImpl Symb) const override;
+ Expected<SymbolRef::Type> getSymbolType(DataRefImpl Symb) const override;
+ Expected<section_iterator> getSymbolSection(DataRefImpl Symb) const override;
+ void moveSectionNext(DataRefImpl &Sec) const override;
+ Expected<StringRef> getSectionName(DataRefImpl Sec) const override;
+ uint64_t getSectionAddress(DataRefImpl Sec) const override;
+ uint64_t getSectionIndex(DataRefImpl Sec) const override;
+ uint64_t getSectionSize(DataRefImpl Sec) const override;
+ Expected<ArrayRef<uint8_t>>
+ getSectionContents(DataRefImpl Sec) const override;
+ uint64_t getSectionAlignment(DataRefImpl Sec) const override;
+ bool isSectionCompressed(DataRefImpl Sec) const override;
+ bool isSectionText(DataRefImpl Sec) const override;
+ bool isSectionData(DataRefImpl Sec) const override;
+ bool isSectionBSS(DataRefImpl Sec) const override;
+ bool isSectionVirtual(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_begin(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_end(DataRefImpl Sec) const override;
+
+ void moveRelocationNext(DataRefImpl &Rel) const override;
+ uint64_t getRelocationOffset(DataRefImpl Rel) const override;
+ symbol_iterator getRelocationSymbol(DataRefImpl Rel) const override;
+ uint64_t getRelocationType(DataRefImpl Rel) const override;
+ void getRelocationTypeName(DataRefImpl Rel,
+ SmallVectorImpl<char> &Result) const override;
+
+public:
+ COFFObjectFile(MemoryBufferRef Object, std::error_code &EC);
+
+ basic_symbol_iterator symbol_begin() const override;
+ basic_symbol_iterator symbol_end() const override;
+ section_iterator section_begin() const override;
+ section_iterator section_end() const override;
+
+ const coff_section *getCOFFSection(const SectionRef &Section) const;
+ COFFSymbolRef getCOFFSymbol(const DataRefImpl &Ref) const;
+ COFFSymbolRef getCOFFSymbol(const SymbolRef &Symbol) const;
+ const coff_relocation *getCOFFRelocation(const RelocationRef &Reloc) const;
+ unsigned getSectionID(SectionRef Sec) const;
+ unsigned getSymbolSectionID(SymbolRef Sym) const;
+
+ uint8_t getBytesInAddress() const override;
+ StringRef getFileFormatName() const override;
+ Triple::ArchType getArch() const override;
+ Expected<uint64_t> getStartAddress() const override;
+ SubtargetFeatures getFeatures() const override { return SubtargetFeatures(); }
+
+ import_directory_iterator import_directory_begin() const;
+ import_directory_iterator import_directory_end() const;
+ delay_import_directory_iterator delay_import_directory_begin() const;
+ delay_import_directory_iterator delay_import_directory_end() const;
+ export_directory_iterator export_directory_begin() const;
+ export_directory_iterator export_directory_end() const;
+ base_reloc_iterator base_reloc_begin() const;
+ base_reloc_iterator base_reloc_end() const;
+ const debug_directory *debug_directory_begin() const {
+ return DebugDirectoryBegin;
+ }
+ const debug_directory *debug_directory_end() const {
+ return DebugDirectoryEnd;
+ }
+
+ iterator_range<import_directory_iterator> import_directories() const;
+ iterator_range<delay_import_directory_iterator>
+ delay_import_directories() const;
+ iterator_range<export_directory_iterator> export_directories() const;
+ iterator_range<base_reloc_iterator> base_relocs() const;
+ iterator_range<const debug_directory *> debug_directories() const {
+ return make_range(debug_directory_begin(), debug_directory_end());
+ }
+
+ const dos_header *getDOSHeader() const {
+ if (!PE32Header && !PE32PlusHeader)
+ return nullptr;
+ return reinterpret_cast<const dos_header *>(base());
+ }
+
+ const coff_file_header *getCOFFHeader() const { return COFFHeader; }
+ const coff_bigobj_file_header *getCOFFBigObjHeader() const {
+ return COFFBigObjHeader;
+ }
+ const pe32_header *getPE32Header() const { return PE32Header; }
+ const pe32plus_header *getPE32PlusHeader() const { return PE32PlusHeader; }
+
+ std::error_code getDataDirectory(uint32_t index,
+ const data_directory *&Res) const;
+ std::error_code getSection(int32_t index, const coff_section *&Res) const;
+ std::error_code getSection(StringRef SectionName,
+ const coff_section *&Res) const;
+
+ template <typename coff_symbol_type>
+ std::error_code getSymbol(uint32_t Index,
+ const coff_symbol_type *&Res) const {
+ if (Index >= getNumberOfSymbols())
+ return object_error::parse_failed;
+
+ Res = reinterpret_cast<coff_symbol_type *>(getSymbolTable()) + Index;
+ return std::error_code();
+ }
+ Expected<COFFSymbolRef> getSymbol(uint32_t index) const {
+ if (SymbolTable16) {
+ const coff_symbol16 *Symb = nullptr;
+ if (std::error_code EC = getSymbol(index, Symb))
+ return errorCodeToError(EC);
+ return COFFSymbolRef(Symb);
+ }
+ if (SymbolTable32) {
+ const coff_symbol32 *Symb = nullptr;
+ if (std::error_code EC = getSymbol(index, Symb))
+ return errorCodeToError(EC);
+ return COFFSymbolRef(Symb);
+ }
+ return errorCodeToError(object_error::parse_failed);
+ }
+
+ template <typename T>
+ std::error_code getAuxSymbol(uint32_t index, const T *&Res) const {
+ Expected<COFFSymbolRef> S = getSymbol(index);
+ if (Error E = S.takeError())
+ return errorToErrorCode(std::move(E));
+ Res = reinterpret_cast<const T *>(S->getRawPtr());
+ return std::error_code();
+ }
+
+ std::error_code getSymbolName(COFFSymbolRef Symbol, StringRef &Res) const;
+ std::error_code getSymbolName(const coff_symbol_generic *Symbol,
+ StringRef &Res) const;
+
+ ArrayRef<uint8_t> getSymbolAuxData(COFFSymbolRef Symbol) const;
+
+ uint32_t getSymbolIndex(COFFSymbolRef Symbol) const;
+
+ size_t getSymbolTableEntrySize() const {
+ if (COFFHeader)
+ return sizeof(coff_symbol16);
+ if (COFFBigObjHeader)
+ return sizeof(coff_symbol32);
+ llvm_unreachable("null symbol table pointer!");
+ }
+
+ ArrayRef<coff_relocation> getRelocations(const coff_section *Sec) const;
+
+ Expected<StringRef> getSectionName(const coff_section *Sec) const;
+ uint64_t getSectionSize(const coff_section *Sec) const;
+ Error getSectionContents(const coff_section *Sec,
+ ArrayRef<uint8_t> &Res) const;
+
+ uint64_t getImageBase() const;
+ std::error_code getVaPtr(uint64_t VA, uintptr_t &Res) const;
+ std::error_code getRvaPtr(uint32_t Rva, uintptr_t &Res) const;
+
+ /// Given an RVA base and size, returns a valid array of bytes or an error
+ /// code if the RVA and size is not contained completely within a valid
+ /// section.
+ std::error_code getRvaAndSizeAsBytes(uint32_t RVA, uint32_t Size,
+ ArrayRef<uint8_t> &Contents) const;
+
+ std::error_code getHintName(uint32_t Rva, uint16_t &Hint,
+ StringRef &Name) const;
+
+ /// Get PDB information out of a codeview debug directory entry.
+ std::error_code getDebugPDBInfo(const debug_directory *DebugDir,
+ const codeview::DebugInfo *&Info,
+ StringRef &PDBFileName) const;
+
+ /// Get PDB information from an executable. If the information is not present,
+ /// Info will be set to nullptr and PDBFileName will be empty. An error is
+ /// returned only on corrupt object files. Convenience accessor that can be
+ /// used if the debug directory is not already handy.
+ std::error_code getDebugPDBInfo(const codeview::DebugInfo *&Info,
+ StringRef &PDBFileName) const;
+
+ bool isRelocatableObject() const override;
+ bool is64() const { return PE32PlusHeader; }
+
+ StringRef mapDebugSectionName(StringRef Name) const override;
+
+ static bool classof(const Binary *v) { return v->isCOFF(); }
+};
+
+// The iterator for the import directory table.
+class ImportDirectoryEntryRef {
+public:
+ ImportDirectoryEntryRef() = default;
+ ImportDirectoryEntryRef(const coff_import_directory_table_entry *Table,
+ uint32_t I, const COFFObjectFile *Owner)
+ : ImportTable(Table), Index(I), OwningObject(Owner) {}
+
+ bool operator==(const ImportDirectoryEntryRef &Other) const;
+ void moveNext();
+
+ imported_symbol_iterator imported_symbol_begin() const;
+ imported_symbol_iterator imported_symbol_end() const;
+ iterator_range<imported_symbol_iterator> imported_symbols() const;
+
+ imported_symbol_iterator lookup_table_begin() const;
+ imported_symbol_iterator lookup_table_end() const;
+ iterator_range<imported_symbol_iterator> lookup_table_symbols() const;
+
+ std::error_code getName(StringRef &Result) const;
+ std::error_code getImportLookupTableRVA(uint32_t &Result) const;
+ std::error_code getImportAddressTableRVA(uint32_t &Result) const;
+
+ std::error_code
+ getImportTableEntry(const coff_import_directory_table_entry *&Result) const;
+
+private:
+ const coff_import_directory_table_entry *ImportTable;
+ uint32_t Index;
+ const COFFObjectFile *OwningObject = nullptr;
+};
+
+class DelayImportDirectoryEntryRef {
+public:
+ DelayImportDirectoryEntryRef() = default;
+ DelayImportDirectoryEntryRef(const delay_import_directory_table_entry *T,
+ uint32_t I, const COFFObjectFile *Owner)
+ : Table(T), Index(I), OwningObject(Owner) {}
+
+ bool operator==(const DelayImportDirectoryEntryRef &Other) const;
+ void moveNext();
+
+ imported_symbol_iterator imported_symbol_begin() const;
+ imported_symbol_iterator imported_symbol_end() const;
+ iterator_range<imported_symbol_iterator> imported_symbols() const;
+
+ std::error_code getName(StringRef &Result) const;
+ std::error_code getDelayImportTable(
+ const delay_import_directory_table_entry *&Result) const;
+ std::error_code getImportAddress(int AddrIndex, uint64_t &Result) const;
+
+private:
+ const delay_import_directory_table_entry *Table;
+ uint32_t Index;
+ const COFFObjectFile *OwningObject = nullptr;
+};
+
+// The iterator for the export directory table entry.
+class ExportDirectoryEntryRef {
+public:
+ ExportDirectoryEntryRef() = default;
+ ExportDirectoryEntryRef(const export_directory_table_entry *Table, uint32_t I,
+ const COFFObjectFile *Owner)
+ : ExportTable(Table), Index(I), OwningObject(Owner) {}
+
+ bool operator==(const ExportDirectoryEntryRef &Other) const;
+ void moveNext();
+
+ std::error_code getDllName(StringRef &Result) const;
+ std::error_code getOrdinalBase(uint32_t &Result) const;
+ std::error_code getOrdinal(uint32_t &Result) const;
+ std::error_code getExportRVA(uint32_t &Result) const;
+ std::error_code getSymbolName(StringRef &Result) const;
+
+ std::error_code isForwarder(bool &Result) const;
+ std::error_code getForwardTo(StringRef &Result) const;
+
+private:
+ const export_directory_table_entry *ExportTable;
+ uint32_t Index;
+ const COFFObjectFile *OwningObject = nullptr;
+};
+
+class ImportedSymbolRef {
+public:
+ ImportedSymbolRef() = default;
+ ImportedSymbolRef(const import_lookup_table_entry32 *Entry, uint32_t I,
+ const COFFObjectFile *Owner)
+ : Entry32(Entry), Entry64(nullptr), Index(I), OwningObject(Owner) {}
+ ImportedSymbolRef(const import_lookup_table_entry64 *Entry, uint32_t I,
+ const COFFObjectFile *Owner)
+ : Entry32(nullptr), Entry64(Entry), Index(I), OwningObject(Owner) {}
+
+ bool operator==(const ImportedSymbolRef &Other) const;
+ void moveNext();
+
+ std::error_code getSymbolName(StringRef &Result) const;
+ std::error_code isOrdinal(bool &Result) const;
+ std::error_code getOrdinal(uint16_t &Result) const;
+ std::error_code getHintNameRVA(uint32_t &Result) const;
+
+private:
+ const import_lookup_table_entry32 *Entry32;
+ const import_lookup_table_entry64 *Entry64;
+ uint32_t Index;
+ const COFFObjectFile *OwningObject = nullptr;
+};
+
+class BaseRelocRef {
+public:
+ BaseRelocRef() = default;
+ BaseRelocRef(const coff_base_reloc_block_header *Header,
+ const COFFObjectFile *Owner)
+ : Header(Header), Index(0) {}
+
+ bool operator==(const BaseRelocRef &Other) const;
+ void moveNext();
+
+ std::error_code getType(uint8_t &Type) const;
+ std::error_code getRVA(uint32_t &Result) const;
+
+private:
+ const coff_base_reloc_block_header *Header;
+ uint32_t Index;
+};
+
+class ResourceSectionRef {
+public:
+ ResourceSectionRef() = default;
+ explicit ResourceSectionRef(StringRef Ref) : BBS(Ref, support::little) {}
+
+ Error load(const COFFObjectFile *O);
+ Error load(const COFFObjectFile *O, const SectionRef &S);
+
+ Expected<ArrayRef<UTF16>>
+ getEntryNameString(const coff_resource_dir_entry &Entry);
+ Expected<const coff_resource_dir_table &>
+ getEntrySubDir(const coff_resource_dir_entry &Entry);
+ Expected<const coff_resource_data_entry &>
+ getEntryData(const coff_resource_dir_entry &Entry);
+ Expected<const coff_resource_dir_table &> getBaseTable();
+ Expected<const coff_resource_dir_entry &>
+ getTableEntry(const coff_resource_dir_table &Table, uint32_t Index);
+
+ Expected<StringRef> getContents(const coff_resource_data_entry &Entry);
+
+private:
+ BinaryByteStream BBS;
+
+ SectionRef Section;
+ const COFFObjectFile *Obj;
+
+ std::vector<const coff_relocation *> Relocs;
+
+ Expected<const coff_resource_dir_table &> getTableAtOffset(uint32_t Offset);
+ Expected<const coff_resource_dir_entry &>
+ getTableEntryAtOffset(uint32_t Offset);
+ Expected<const coff_resource_data_entry &>
+ getDataEntryAtOffset(uint32_t Offset);
+ Expected<ArrayRef<UTF16>> getDirStringAtOffset(uint32_t Offset);
+};
+
+// Corresponds to `_FPO_DATA` structure in the PE/COFF spec.
+struct FpoData {
+ support::ulittle32_t Offset; // ulOffStart: Offset 1st byte of function code
+ support::ulittle32_t Size; // cbProcSize: # bytes in function
+ support::ulittle32_t NumLocals; // cdwLocals: # bytes in locals/4
+ support::ulittle16_t NumParams; // cdwParams: # bytes in params/4
+ support::ulittle16_t Attributes;
+
+ // cbProlog: # bytes in prolog
+ int getPrologSize() const { return Attributes & 0xF; }
+
+ // cbRegs: # regs saved
+ int getNumSavedRegs() const { return (Attributes >> 8) & 0x7; }
+
+ // fHasSEH: true if seh is func
+ bool hasSEH() const { return (Attributes >> 9) & 1; }
+
+ // fUseBP: true if EBP has been allocated
+ bool useBP() const { return (Attributes >> 10) & 1; }
+
+ // cbFrame: frame pointer
+ frame_type getFP() const { return static_cast<frame_type>(Attributes >> 14); }
+};
+
+} // end namespace object
+
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_COFF_H
diff --git a/third_party/llvm-project/include/llvm/Object/CVDebugRecord.h b/third_party/llvm-project/include/llvm/Object/CVDebugRecord.h
new file mode 100644
index 000000000..d41c7391f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/CVDebugRecord.h
@@ -0,0 +1,54 @@
+//===- CVDebugRecord.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_CVDEBUGRECORD_H
+#define LLVM_OBJECT_CVDEBUGRECORD_H
+
+#include "llvm/Support/Endian.h"
+
+namespace llvm {
+namespace OMF {
+struct Signature {
+ enum ID : uint32_t {
+ PDB70 = 0x53445352, // RSDS
+ PDB20 = 0x3031424e, // NB10
+ CV50 = 0x3131424e, // NB11
+ CV41 = 0x3930424e, // NB09
+ };
+
+ support::ulittle32_t CVSignature;
+ support::ulittle32_t Offset;
+};
+}
+
+namespace codeview {
+struct PDB70DebugInfo {
+ support::ulittle32_t CVSignature;
+ uint8_t Signature[16];
+ support::ulittle32_t Age;
+ // char PDBFileName[];
+};
+
+struct PDB20DebugInfo {
+ support::ulittle32_t CVSignature;
+ support::ulittle32_t Offset;
+ support::ulittle32_t Signature;
+ support::ulittle32_t Age;
+ // char PDBFileName[];
+};
+
+union DebugInfo {
+ struct OMF::Signature Signature;
+ struct PDB20DebugInfo PDB20;
+ struct PDB70DebugInfo PDB70;
+};
+}
+}
+
+#endif
+
diff --git a/third_party/llvm-project/include/llvm/Object/Decompressor.h b/third_party/llvm-project/include/llvm/Object/Decompressor.h
new file mode 100644
index 000000000..cc918481b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Decompressor.h
@@ -0,0 +1,66 @@
+//===-- Decompressor.h ------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===/
+
+#ifndef LLVM_OBJECT_DECOMPRESSOR_H
+#define LLVM_OBJECT_DECOMPRESSOR_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Object/ObjectFile.h"
+
+namespace llvm {
+namespace object {
+
+/// Decompressor helps to handle decompression of compressed sections.
+class Decompressor {
+public:
+ /// Create decompressor object.
+ /// @param Name Section name.
+ /// @param Data Section content.
+ /// @param IsLE Flag determines if Data is in little endian form.
+ /// @param Is64Bit Flag determines if object is 64 bit.
+ static Expected<Decompressor> create(StringRef Name, StringRef Data,
+ bool IsLE, bool Is64Bit);
+
+ /// Resize the buffer and uncompress section data into it.
+ /// @param Out Destination buffer.
+ template <class T> Error resizeAndDecompress(T &Out) {
+ Out.resize(DecompressedSize);
+ return decompress({Out.data(), (size_t)DecompressedSize});
+ }
+
+ /// Uncompress section data to raw buffer provided.
+ /// @param Buffer Destination buffer.
+ Error decompress(MutableArrayRef<char> Buffer);
+
+ /// Return memory buffer size required for decompression.
+ uint64_t getDecompressedSize() { return DecompressedSize; }
+
+ /// Return true if section is compressed, including gnu-styled case.
+ static bool isCompressed(const object::SectionRef &Section);
+
+ /// Return true if section is a ELF compressed one.
+ static bool isCompressedELFSection(uint64_t Flags, StringRef Name);
+
+ /// Return true if section name matches gnu style compressed one.
+ static bool isGnuStyle(StringRef Name);
+
+private:
+ Decompressor(StringRef Data);
+
+ Error consumeCompressedGnuHeader();
+ Error consumeCompressedZLibHeader(bool Is64Bit, bool IsLittleEndian);
+
+ StringRef SectionData;
+ uint64_t DecompressedSize;
+};
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_DECOMPRESSOR_H
diff --git a/third_party/llvm-project/include/llvm/Object/ELF.h b/third_party/llvm-project/include/llvm/Object/ELF.h
new file mode 100644
index 000000000..0d5482ae8
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/ELF.h
@@ -0,0 +1,723 @@
+//===- ELF.h - ELF object file implementation -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the ELFFile template class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_ELF_H
+#define LLVM_OBJECT_ELF_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/ELF.h"
+#include "llvm/Object/ELFTypes.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Error.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <limits>
+#include <utility>
+
+namespace llvm {
+namespace object {
+
+StringRef getELFRelocationTypeName(uint32_t Machine, uint32_t Type);
+uint32_t getELFRelativeRelocationType(uint32_t Machine);
+StringRef getELFSectionTypeName(uint32_t Machine, uint32_t Type);
+
+// Subclasses of ELFFile may need this for template instantiation
+inline std::pair<unsigned char, unsigned char>
+getElfArchType(StringRef Object) {
+ if (Object.size() < ELF::EI_NIDENT)
+ return std::make_pair((uint8_t)ELF::ELFCLASSNONE,
+ (uint8_t)ELF::ELFDATANONE);
+ return std::make_pair((uint8_t)Object[ELF::EI_CLASS],
+ (uint8_t)Object[ELF::EI_DATA]);
+}
+
+static inline Error createError(const Twine &Err) {
+ return make_error<StringError>(Err, object_error::parse_failed);
+}
+
+template <class ELFT> class ELFFile;
+
+template <class ELFT>
+std::string getSecIndexForError(const ELFFile<ELFT> *Obj,
+ const typename ELFT::Shdr *Sec) {
+ auto TableOrErr = Obj->sections();
+ if (TableOrErr)
+ return "[index " + std::to_string(Sec - &TableOrErr->front()) + "]";
+ // To make this helper be more convenient for error reporting purposes we
+ // drop the error. But really it should never be triggered. Before this point,
+ // our code should have called 'sections()' and reported a proper error on
+ // failure.
+ llvm::consumeError(TableOrErr.takeError());
+ return "[unknown index]";
+}
+
+static inline Error defaultWarningHandler(const Twine &Msg) {
+ return createError(Msg);
+}
+
+template <class ELFT>
+class ELFFile {
+public:
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ using uintX_t = typename ELFT::uint;
+ using Elf_Ehdr = typename ELFT::Ehdr;
+ using Elf_Shdr = typename ELFT::Shdr;
+ using Elf_Sym = typename ELFT::Sym;
+ using Elf_Dyn = typename ELFT::Dyn;
+ using Elf_Phdr = typename ELFT::Phdr;
+ using Elf_Rel = typename ELFT::Rel;
+ using Elf_Rela = typename ELFT::Rela;
+ using Elf_Relr = typename ELFT::Relr;
+ using Elf_Verdef = typename ELFT::Verdef;
+ using Elf_Verdaux = typename ELFT::Verdaux;
+ using Elf_Verneed = typename ELFT::Verneed;
+ using Elf_Vernaux = typename ELFT::Vernaux;
+ using Elf_Versym = typename ELFT::Versym;
+ using Elf_Hash = typename ELFT::Hash;
+ using Elf_GnuHash = typename ELFT::GnuHash;
+ using Elf_Nhdr = typename ELFT::Nhdr;
+ using Elf_Note = typename ELFT::Note;
+ using Elf_Note_Iterator = typename ELFT::NoteIterator;
+ using Elf_Dyn_Range = typename ELFT::DynRange;
+ using Elf_Shdr_Range = typename ELFT::ShdrRange;
+ using Elf_Sym_Range = typename ELFT::SymRange;
+ using Elf_Rel_Range = typename ELFT::RelRange;
+ using Elf_Rela_Range = typename ELFT::RelaRange;
+ using Elf_Relr_Range = typename ELFT::RelrRange;
+ using Elf_Phdr_Range = typename ELFT::PhdrRange;
+
+ // This is a callback that can be passed to a number of functions.
+ // It can be used to ignore non-critical errors (warnings), which is
+ // useful for dumpers, like llvm-readobj.
+ // It accepts a warning message string and returns a success
+ // when the warning should be ignored or an error otherwise.
+ using WarningHandler = llvm::function_ref<Error(const Twine &Msg)>;
+
+ const uint8_t *base() const { return Buf.bytes_begin(); }
+
+ size_t getBufSize() const { return Buf.size(); }
+
+private:
+ StringRef Buf;
+
+ ELFFile(StringRef Object);
+
+public:
+ const Elf_Ehdr *getHeader() const {
+ return reinterpret_cast<const Elf_Ehdr *>(base());
+ }
+
+ template <typename T>
+ Expected<const T *> getEntry(uint32_t Section, uint32_t Entry) const;
+ template <typename T>
+ Expected<const T *> getEntry(const Elf_Shdr *Section, uint32_t Entry) const;
+
+ Expected<StringRef>
+ getStringTable(const Elf_Shdr *Section,
+ WarningHandler WarnHandler = &defaultWarningHandler) const;
+ Expected<StringRef> getStringTableForSymtab(const Elf_Shdr &Section) const;
+ Expected<StringRef> getStringTableForSymtab(const Elf_Shdr &Section,
+ Elf_Shdr_Range Sections) const;
+
+ Expected<ArrayRef<Elf_Word>> getSHNDXTable(const Elf_Shdr &Section) const;
+ Expected<ArrayRef<Elf_Word>> getSHNDXTable(const Elf_Shdr &Section,
+ Elf_Shdr_Range Sections) const;
+
+ StringRef getRelocationTypeName(uint32_t Type) const;
+ void getRelocationTypeName(uint32_t Type,
+ SmallVectorImpl<char> &Result) const;
+ uint32_t getRelativeRelocationType() const;
+
+ std::string getDynamicTagAsString(unsigned Arch, uint64_t Type) const;
+ std::string getDynamicTagAsString(uint64_t Type) const;
+
+ /// Get the symbol for a given relocation.
+ Expected<const Elf_Sym *> getRelocationSymbol(const Elf_Rel *Rel,
+ const Elf_Shdr *SymTab) const;
+
+ static Expected<ELFFile> create(StringRef Object);
+
+ bool isLE() const {
+ return getHeader()->getDataEncoding() == ELF::ELFDATA2LSB;
+ }
+
+ bool isMipsELF64() const {
+ return getHeader()->e_machine == ELF::EM_MIPS &&
+ getHeader()->getFileClass() == ELF::ELFCLASS64;
+ }
+
+ bool isMips64EL() const { return isMipsELF64() && isLE(); }
+
+ Expected<Elf_Shdr_Range> sections() const;
+
+ Expected<Elf_Dyn_Range> dynamicEntries() const;
+
+ Expected<const uint8_t *> toMappedAddr(uint64_t VAddr) const;
+
+ Expected<Elf_Sym_Range> symbols(const Elf_Shdr *Sec) const {
+ if (!Sec)
+ return makeArrayRef<Elf_Sym>(nullptr, nullptr);
+ return getSectionContentsAsArray<Elf_Sym>(Sec);
+ }
+
+ Expected<Elf_Rela_Range> relas(const Elf_Shdr *Sec) const {
+ return getSectionContentsAsArray<Elf_Rela>(Sec);
+ }
+
+ Expected<Elf_Rel_Range> rels(const Elf_Shdr *Sec) const {
+ return getSectionContentsAsArray<Elf_Rel>(Sec);
+ }
+
+ Expected<Elf_Relr_Range> relrs(const Elf_Shdr *Sec) const {
+ return getSectionContentsAsArray<Elf_Relr>(Sec);
+ }
+
+ Expected<std::vector<Elf_Rela>> decode_relrs(Elf_Relr_Range relrs) const;
+
+ Expected<std::vector<Elf_Rela>> android_relas(const Elf_Shdr *Sec) const;
+
+ /// Iterate over program header table.
+ Expected<Elf_Phdr_Range> program_headers() const {
+ if (getHeader()->e_phnum && getHeader()->e_phentsize != sizeof(Elf_Phdr))
+ return createError("invalid e_phentsize: " +
+ Twine(getHeader()->e_phentsize));
+ if (getHeader()->e_phoff +
+ (getHeader()->e_phnum * getHeader()->e_phentsize) >
+ getBufSize())
+ return createError("program headers are longer than binary of size " +
+ Twine(getBufSize()) + ": e_phoff = 0x" +
+ Twine::utohexstr(getHeader()->e_phoff) +
+ ", e_phnum = " + Twine(getHeader()->e_phnum) +
+ ", e_phentsize = " + Twine(getHeader()->e_phentsize));
+ auto *Begin =
+ reinterpret_cast<const Elf_Phdr *>(base() + getHeader()->e_phoff);
+ return makeArrayRef(Begin, Begin + getHeader()->e_phnum);
+ }
+
+ /// Get an iterator over notes in a program header.
+ ///
+ /// The program header must be of type \c PT_NOTE.
+ ///
+ /// \param Phdr the program header to iterate over.
+ /// \param Err [out] an error to support fallible iteration, which should
+ /// be checked after iteration ends.
+ Elf_Note_Iterator notes_begin(const Elf_Phdr &Phdr, Error &Err) const {
+ assert(Phdr.p_type == ELF::PT_NOTE && "Phdr is not of type PT_NOTE");
+ ErrorAsOutParameter ErrAsOutParam(&Err);
+ if (Phdr.p_offset + Phdr.p_filesz > getBufSize()) {
+ Err = createError("PT_NOTE header has invalid offset (0x" +
+ Twine::utohexstr(Phdr.p_offset) + ") or size (0x" +
+ Twine::utohexstr(Phdr.p_filesz) + ")");
+ return Elf_Note_Iterator(Err);
+ }
+ return Elf_Note_Iterator(base() + Phdr.p_offset, Phdr.p_filesz, Err);
+ }
+
+ /// Get an iterator over notes in a section.
+ ///
+ /// The section must be of type \c SHT_NOTE.
+ ///
+ /// \param Shdr the section to iterate over.
+ /// \param Err [out] an error to support fallible iteration, which should
+ /// be checked after iteration ends.
+ Elf_Note_Iterator notes_begin(const Elf_Shdr &Shdr, Error &Err) const {
+ assert(Shdr.sh_type == ELF::SHT_NOTE && "Shdr is not of type SHT_NOTE");
+ ErrorAsOutParameter ErrAsOutParam(&Err);
+ if (Shdr.sh_offset + Shdr.sh_size > getBufSize()) {
+ Err = createError("SHT_NOTE section " + getSecIndexForError(this, &Shdr) +
+ " has invalid offset (0x" +
+ Twine::utohexstr(Shdr.sh_offset) + ") or size (0x" +
+ Twine::utohexstr(Shdr.sh_size) + ")");
+ return Elf_Note_Iterator(Err);
+ }
+ return Elf_Note_Iterator(base() + Shdr.sh_offset, Shdr.sh_size, Err);
+ }
+
+ /// Get the end iterator for notes.
+ Elf_Note_Iterator notes_end() const {
+ return Elf_Note_Iterator();
+ }
+
+ /// Get an iterator range over notes of a program header.
+ ///
+ /// The program header must be of type \c PT_NOTE.
+ ///
+ /// \param Phdr the program header to iterate over.
+ /// \param Err [out] an error to support fallible iteration, which should
+ /// be checked after iteration ends.
+ iterator_range<Elf_Note_Iterator> notes(const Elf_Phdr &Phdr,
+ Error &Err) const {
+ return make_range(notes_begin(Phdr, Err), notes_end());
+ }
+
+ /// Get an iterator range over notes of a section.
+ ///
+ /// The section must be of type \c SHT_NOTE.
+ ///
+ /// \param Shdr the section to iterate over.
+ /// \param Err [out] an error to support fallible iteration, which should
+ /// be checked after iteration ends.
+ iterator_range<Elf_Note_Iterator> notes(const Elf_Shdr &Shdr,
+ Error &Err) const {
+ return make_range(notes_begin(Shdr, Err), notes_end());
+ }
+
+ Expected<StringRef> getSectionStringTable(
+ Elf_Shdr_Range Sections,
+ WarningHandler WarnHandler = &defaultWarningHandler) const;
+ Expected<uint32_t> getSectionIndex(const Elf_Sym *Sym, Elf_Sym_Range Syms,
+ ArrayRef<Elf_Word> ShndxTable) const;
+ Expected<const Elf_Shdr *> getSection(const Elf_Sym *Sym,
+ const Elf_Shdr *SymTab,
+ ArrayRef<Elf_Word> ShndxTable) const;
+ Expected<const Elf_Shdr *> getSection(const Elf_Sym *Sym,
+ Elf_Sym_Range Symtab,
+ ArrayRef<Elf_Word> ShndxTable) const;
+ Expected<const Elf_Shdr *> getSection(uint32_t Index) const;
+
+ Expected<const Elf_Sym *> getSymbol(const Elf_Shdr *Sec,
+ uint32_t Index) const;
+
+ Expected<StringRef>
+ getSectionName(const Elf_Shdr *Section,
+ WarningHandler WarnHandler = &defaultWarningHandler) const;
+ Expected<StringRef> getSectionName(const Elf_Shdr *Section,
+ StringRef DotShstrtab) const;
+ template <typename T>
+ Expected<ArrayRef<T>> getSectionContentsAsArray(const Elf_Shdr *Sec) const;
+ Expected<ArrayRef<uint8_t>> getSectionContents(const Elf_Shdr *Sec) const;
+};
+
+using ELF32LEFile = ELFFile<ELF32LE>;
+using ELF64LEFile = ELFFile<ELF64LE>;
+using ELF32BEFile = ELFFile<ELF32BE>;
+using ELF64BEFile = ELFFile<ELF64BE>;
+
+template <class ELFT>
+inline Expected<const typename ELFT::Shdr *>
+getSection(typename ELFT::ShdrRange Sections, uint32_t Index) {
+ if (Index >= Sections.size())
+ return createError("invalid section index: " + Twine(Index));
+ return &Sections[Index];
+}
+
+template <class ELFT>
+inline Expected<uint32_t>
+getExtendedSymbolTableIndex(const typename ELFT::Sym *Sym,
+ const typename ELFT::Sym *FirstSym,
+ ArrayRef<typename ELFT::Word> ShndxTable) {
+ assert(Sym->st_shndx == ELF::SHN_XINDEX);
+ unsigned Index = Sym - FirstSym;
+ if (Index >= ShndxTable.size())
+ return createError(
+ "extended symbol index (" + Twine(Index) +
+ ") is past the end of the SHT_SYMTAB_SHNDX section of size " +
+ Twine(ShndxTable.size()));
+
+ // The size of the table was checked in getSHNDXTable.
+ return ShndxTable[Index];
+}
+
+template <class ELFT>
+Expected<uint32_t>
+ELFFile<ELFT>::getSectionIndex(const Elf_Sym *Sym, Elf_Sym_Range Syms,
+ ArrayRef<Elf_Word> ShndxTable) const {
+ uint32_t Index = Sym->st_shndx;
+ if (Index == ELF::SHN_XINDEX) {
+ auto ErrorOrIndex = getExtendedSymbolTableIndex<ELFT>(
+ Sym, Syms.begin(), ShndxTable);
+ if (!ErrorOrIndex)
+ return ErrorOrIndex.takeError();
+ return *ErrorOrIndex;
+ }
+ if (Index == ELF::SHN_UNDEF || Index >= ELF::SHN_LORESERVE)
+ return 0;
+ return Index;
+}
+
+template <class ELFT>
+Expected<const typename ELFT::Shdr *>
+ELFFile<ELFT>::getSection(const Elf_Sym *Sym, const Elf_Shdr *SymTab,
+ ArrayRef<Elf_Word> ShndxTable) const {
+ auto SymsOrErr = symbols(SymTab);
+ if (!SymsOrErr)
+ return SymsOrErr.takeError();
+ return getSection(Sym, *SymsOrErr, ShndxTable);
+}
+
+template <class ELFT>
+Expected<const typename ELFT::Shdr *>
+ELFFile<ELFT>::getSection(const Elf_Sym *Sym, Elf_Sym_Range Symbols,
+ ArrayRef<Elf_Word> ShndxTable) const {
+ auto IndexOrErr = getSectionIndex(Sym, Symbols, ShndxTable);
+ if (!IndexOrErr)
+ return IndexOrErr.takeError();
+ uint32_t Index = *IndexOrErr;
+ if (Index == 0)
+ return nullptr;
+ return getSection(Index);
+}
+
+template <class ELFT>
+Expected<const typename ELFT::Sym *>
+ELFFile<ELFT>::getSymbol(const Elf_Shdr *Sec, uint32_t Index) const {
+ auto SymsOrErr = symbols(Sec);
+ if (!SymsOrErr)
+ return SymsOrErr.takeError();
+
+ Elf_Sym_Range Symbols = *SymsOrErr;
+ if (Index >= Symbols.size())
+ return createError("unable to get symbol from section " +
+ getSecIndexForError(this, Sec) +
+ ": invalid symbol index (" + Twine(Index) + ")");
+ return &Symbols[Index];
+}
+
+template <class ELFT>
+template <typename T>
+Expected<ArrayRef<T>>
+ELFFile<ELFT>::getSectionContentsAsArray(const Elf_Shdr *Sec) const {
+ if (Sec->sh_entsize != sizeof(T) && sizeof(T) != 1)
+ return createError("section " + getSecIndexForError(this, Sec) +
+ " has an invalid sh_entsize: " + Twine(Sec->sh_entsize));
+
+ uintX_t Offset = Sec->sh_offset;
+ uintX_t Size = Sec->sh_size;
+
+ if (Size % sizeof(T))
+ return createError("section " + getSecIndexForError(this, Sec) +
+ " has an invalid sh_size (" + Twine(Size) +
+ ") which is not a multiple of its sh_entsize (" +
+ Twine(Sec->sh_entsize) + ")");
+ if ((std::numeric_limits<uintX_t>::max() - Offset < Size) ||
+ Offset + Size > Buf.size())
+ return createError("section " + getSecIndexForError(this, Sec) +
+ " has a sh_offset (0x" + Twine::utohexstr(Offset) +
+ ") + sh_size (0x" + Twine(Size) +
+ ") that cannot be represented");
+
+ if (Offset % alignof(T))
+ // TODO: this error is untested.
+ return createError("unaligned data");
+
+ const T *Start = reinterpret_cast<const T *>(base() + Offset);
+ return makeArrayRef(Start, Size / sizeof(T));
+}
+
+template <class ELFT>
+Expected<ArrayRef<uint8_t>>
+ELFFile<ELFT>::getSectionContents(const Elf_Shdr *Sec) const {
+ return getSectionContentsAsArray<uint8_t>(Sec);
+}
+
+template <class ELFT>
+StringRef ELFFile<ELFT>::getRelocationTypeName(uint32_t Type) const {
+ return getELFRelocationTypeName(getHeader()->e_machine, Type);
+}
+
+template <class ELFT>
+void ELFFile<ELFT>::getRelocationTypeName(uint32_t Type,
+ SmallVectorImpl<char> &Result) const {
+ if (!isMipsELF64()) {
+ StringRef Name = getRelocationTypeName(Type);
+ Result.append(Name.begin(), Name.end());
+ } else {
+ // The Mips N64 ABI allows up to three operations to be specified per
+ // relocation record. Unfortunately there's no easy way to test for the
+ // presence of N64 ELFs as they have no special flag that identifies them
+ // as being N64. We can safely assume at the moment that all Mips
+ // ELFCLASS64 ELFs are N64. New Mips64 ABIs should provide enough
+ // information to disambiguate between old vs new ABIs.
+ uint8_t Type1 = (Type >> 0) & 0xFF;
+ uint8_t Type2 = (Type >> 8) & 0xFF;
+ uint8_t Type3 = (Type >> 16) & 0xFF;
+
+ // Concat all three relocation type names.
+ StringRef Name = getRelocationTypeName(Type1);
+ Result.append(Name.begin(), Name.end());
+
+ Name = getRelocationTypeName(Type2);
+ Result.append(1, '/');
+ Result.append(Name.begin(), Name.end());
+
+ Name = getRelocationTypeName(Type3);
+ Result.append(1, '/');
+ Result.append(Name.begin(), Name.end());
+ }
+}
+
+template <class ELFT>
+uint32_t ELFFile<ELFT>::getRelativeRelocationType() const {
+ return getELFRelativeRelocationType(getHeader()->e_machine);
+}
+
+template <class ELFT>
+Expected<const typename ELFT::Sym *>
+ELFFile<ELFT>::getRelocationSymbol(const Elf_Rel *Rel,
+ const Elf_Shdr *SymTab) const {
+ uint32_t Index = Rel->getSymbol(isMips64EL());
+ if (Index == 0)
+ return nullptr;
+ return getEntry<Elf_Sym>(SymTab, Index);
+}
+
+template <class ELFT>
+Expected<StringRef>
+ELFFile<ELFT>::getSectionStringTable(Elf_Shdr_Range Sections,
+ WarningHandler WarnHandler) const {
+ uint32_t Index = getHeader()->e_shstrndx;
+ if (Index == ELF::SHN_XINDEX)
+ Index = Sections[0].sh_link;
+
+ if (!Index) // no section string table.
+ return "";
+ if (Index >= Sections.size())
+ return createError("section header string table index " + Twine(Index) +
+ " does not exist");
+ return getStringTable(&Sections[Index], WarnHandler);
+}
+
+template <class ELFT> ELFFile<ELFT>::ELFFile(StringRef Object) : Buf(Object) {}
+
+template <class ELFT>
+Expected<ELFFile<ELFT>> ELFFile<ELFT>::create(StringRef Object) {
+ if (sizeof(Elf_Ehdr) > Object.size())
+ return createError("invalid buffer: the size (" + Twine(Object.size()) +
+ ") is smaller than an ELF header (" +
+ Twine(sizeof(Elf_Ehdr)) + ")");
+ return ELFFile(Object);
+}
+
+template <class ELFT>
+Expected<typename ELFT::ShdrRange> ELFFile<ELFT>::sections() const {
+ const uintX_t SectionTableOffset = getHeader()->e_shoff;
+ if (SectionTableOffset == 0)
+ return ArrayRef<Elf_Shdr>();
+
+ if (getHeader()->e_shentsize != sizeof(Elf_Shdr))
+ return createError("invalid e_shentsize in ELF header: " +
+ Twine(getHeader()->e_shentsize));
+
+ const uint64_t FileSize = Buf.size();
+ if (SectionTableOffset + sizeof(Elf_Shdr) > FileSize ||
+ SectionTableOffset + (uintX_t)sizeof(Elf_Shdr) < SectionTableOffset)
+ return createError(
+ "section header table goes past the end of the file: e_shoff = 0x" +
+ Twine::utohexstr(SectionTableOffset));
+
+ // Invalid address alignment of section headers
+ if (SectionTableOffset & (alignof(Elf_Shdr) - 1))
+ // TODO: this error is untested.
+ return createError("invalid alignment of section headers");
+
+ const Elf_Shdr *First =
+ reinterpret_cast<const Elf_Shdr *>(base() + SectionTableOffset);
+
+ uintX_t NumSections = getHeader()->e_shnum;
+ if (NumSections == 0)
+ NumSections = First->sh_size;
+
+ if (NumSections > UINT64_MAX / sizeof(Elf_Shdr))
+ return createError("invalid number of sections specified in the NULL "
+ "section's sh_size field (" +
+ Twine(NumSections) + ")");
+
+ const uint64_t SectionTableSize = NumSections * sizeof(Elf_Shdr);
+ if (SectionTableOffset + SectionTableSize < SectionTableOffset)
+ return createError(
+ "invalid section header table offset (e_shoff = 0x" +
+ Twine::utohexstr(SectionTableOffset) +
+ ") or invalid number of sections specified in the first section "
+ "header's sh_size field (0x" +
+ Twine::utohexstr(NumSections) + ")");
+
+ // Section table goes past end of file!
+ if (SectionTableOffset + SectionTableSize > FileSize)
+ return createError("section table goes past the end of file");
+ return makeArrayRef(First, NumSections);
+}
+
+template <class ELFT>
+template <typename T>
+Expected<const T *> ELFFile<ELFT>::getEntry(uint32_t Section,
+ uint32_t Entry) const {
+ auto SecOrErr = getSection(Section);
+ if (!SecOrErr)
+ return SecOrErr.takeError();
+ return getEntry<T>(*SecOrErr, Entry);
+}
+
+template <class ELFT>
+template <typename T>
+Expected<const T *> ELFFile<ELFT>::getEntry(const Elf_Shdr *Section,
+ uint32_t Entry) const {
+ if (sizeof(T) != Section->sh_entsize)
+ return createError("section " + getSecIndexForError(this, Section) +
+ " has invalid sh_entsize: expected " + Twine(sizeof(T)) +
+ ", but got " + Twine(Section->sh_entsize));
+ size_t Pos = Section->sh_offset + Entry * sizeof(T);
+ if (Pos + sizeof(T) > Buf.size())
+ return createError("unable to access section " +
+ getSecIndexForError(this, Section) + " data at 0x" +
+ Twine::utohexstr(Pos) +
+ ": offset goes past the end of file");
+ return reinterpret_cast<const T *>(base() + Pos);
+}
+
+template <class ELFT>
+Expected<const typename ELFT::Shdr *>
+ELFFile<ELFT>::getSection(uint32_t Index) const {
+ auto TableOrErr = sections();
+ if (!TableOrErr)
+ return TableOrErr.takeError();
+ return object::getSection<ELFT>(*TableOrErr, Index);
+}
+
+template <class ELFT>
+Expected<StringRef>
+ELFFile<ELFT>::getStringTable(const Elf_Shdr *Section,
+ WarningHandler WarnHandler) const {
+ if (Section->sh_type != ELF::SHT_STRTAB)
+ if (Error E = WarnHandler("invalid sh_type for string table section " +
+ getSecIndexForError(this, Section) +
+ ": expected SHT_STRTAB, but got " +
+ object::getELFSectionTypeName(
+ getHeader()->e_machine, Section->sh_type)))
+ return std::move(E);
+
+ auto V = getSectionContentsAsArray<char>(Section);
+ if (!V)
+ return V.takeError();
+ ArrayRef<char> Data = *V;
+ if (Data.empty())
+ return createError("SHT_STRTAB string table section " +
+ getSecIndexForError(this, Section) + " is empty");
+ if (Data.back() != '\0')
+ return createError("SHT_STRTAB string table section " +
+ getSecIndexForError(this, Section) +
+ " is non-null terminated");
+ return StringRef(Data.begin(), Data.size());
+}
+
+template <class ELFT>
+Expected<ArrayRef<typename ELFT::Word>>
+ELFFile<ELFT>::getSHNDXTable(const Elf_Shdr &Section) const {
+ auto SectionsOrErr = sections();
+ if (!SectionsOrErr)
+ return SectionsOrErr.takeError();
+ return getSHNDXTable(Section, *SectionsOrErr);
+}
+
+template <class ELFT>
+Expected<ArrayRef<typename ELFT::Word>>
+ELFFile<ELFT>::getSHNDXTable(const Elf_Shdr &Section,
+ Elf_Shdr_Range Sections) const {
+ assert(Section.sh_type == ELF::SHT_SYMTAB_SHNDX);
+ auto VOrErr = getSectionContentsAsArray<Elf_Word>(&Section);
+ if (!VOrErr)
+ return VOrErr.takeError();
+ ArrayRef<Elf_Word> V = *VOrErr;
+ auto SymTableOrErr = object::getSection<ELFT>(Sections, Section.sh_link);
+ if (!SymTableOrErr)
+ return SymTableOrErr.takeError();
+ const Elf_Shdr &SymTable = **SymTableOrErr;
+ if (SymTable.sh_type != ELF::SHT_SYMTAB &&
+ SymTable.sh_type != ELF::SHT_DYNSYM)
+ return createError("SHT_SYMTAB_SHNDX section is linked with " +
+ object::getELFSectionTypeName(getHeader()->e_machine,
+ SymTable.sh_type) +
+ " section (expected SHT_SYMTAB/SHT_DYNSYM)");
+
+ uint64_t Syms = SymTable.sh_size / sizeof(Elf_Sym);
+ if (V.size() != Syms)
+ return createError("SHT_SYMTAB_SHNDX has " + Twine(V.size()) +
+ " entries, but the symbol table associated has " +
+ Twine(Syms));
+
+ return V;
+}
+
+template <class ELFT>
+Expected<StringRef>
+ELFFile<ELFT>::getStringTableForSymtab(const Elf_Shdr &Sec) const {
+ auto SectionsOrErr = sections();
+ if (!SectionsOrErr)
+ return SectionsOrErr.takeError();
+ return getStringTableForSymtab(Sec, *SectionsOrErr);
+}
+
+template <class ELFT>
+Expected<StringRef>
+ELFFile<ELFT>::getStringTableForSymtab(const Elf_Shdr &Sec,
+ Elf_Shdr_Range Sections) const {
+
+ if (Sec.sh_type != ELF::SHT_SYMTAB && Sec.sh_type != ELF::SHT_DYNSYM)
+ // TODO: this error is untested.
+ return createError(
+ "invalid sh_type for symbol table, expected SHT_SYMTAB or SHT_DYNSYM");
+ auto SectionOrErr = object::getSection<ELFT>(Sections, Sec.sh_link);
+ if (!SectionOrErr)
+ return SectionOrErr.takeError();
+ return getStringTable(*SectionOrErr);
+}
+
+template <class ELFT>
+Expected<StringRef>
+ELFFile<ELFT>::getSectionName(const Elf_Shdr *Section,
+ WarningHandler WarnHandler) const {
+ auto SectionsOrErr = sections();
+ if (!SectionsOrErr)
+ return SectionsOrErr.takeError();
+ auto Table = getSectionStringTable(*SectionsOrErr, WarnHandler);
+ if (!Table)
+ return Table.takeError();
+ return getSectionName(Section, *Table);
+}
+
+template <class ELFT>
+Expected<StringRef> ELFFile<ELFT>::getSectionName(const Elf_Shdr *Section,
+ StringRef DotShstrtab) const {
+ uint32_t Offset = Section->sh_name;
+ if (Offset == 0)
+ return StringRef();
+ if (Offset >= DotShstrtab.size())
+ return createError("a section " + getSecIndexForError(this, Section) +
+ " has an invalid sh_name (0x" +
+ Twine::utohexstr(Offset) +
+ ") offset which goes past the end of the "
+ "section name string table");
+ return StringRef(DotShstrtab.data() + Offset);
+}
+
+/// This function returns the hash value for a symbol in the .dynsym section
+/// Name of the API remains consistent as specified in the libelf
+/// REF : http://www.sco.com/developers/gabi/latest/ch5.dynamic.html#hash
+inline unsigned hashSysV(StringRef SymbolName) {
+ unsigned h = 0, g;
+ for (char C : SymbolName) {
+ h = (h << 4) + C;
+ g = h & 0xf0000000L;
+ if (g != 0)
+ h ^= g >> 24;
+ h &= ~g;
+ }
+ return h;
+}
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_ELF_H
diff --git a/third_party/llvm-project/include/llvm/Object/ELFObjectFile.h b/third_party/llvm-project/include/llvm/Object/ELFObjectFile.h
new file mode 100644
index 000000000..8a68e4947
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/ELFObjectFile.h
@@ -0,0 +1,1214 @@
+//===- ELFObjectFile.h - ELF object file implementation ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the ELFObjectFile template class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_ELFOBJECTFILE_H
+#define LLVM_OBJECT_ELFOBJECTFILE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/ELF.h"
+#include "llvm/MC/SubtargetFeature.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ELF.h"
+#include "llvm/Object/ELFTypes.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Object/SymbolicFile.h"
+#include "llvm/Support/ARMAttributeParser.h"
+#include "llvm/Support/ARMBuildAttributes.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <cassert>
+#include <cstdint>
+#include <system_error>
+
+namespace llvm {
+namespace object {
+
+constexpr int NumElfSymbolTypes = 16;
+extern const llvm::EnumEntry<unsigned> ElfSymbolTypes[NumElfSymbolTypes];
+
+class elf_symbol_iterator;
+
+class ELFObjectFileBase : public ObjectFile {
+ friend class ELFRelocationRef;
+ friend class ELFSectionRef;
+ friend class ELFSymbolRef;
+
+protected:
+ ELFObjectFileBase(unsigned int Type, MemoryBufferRef Source);
+
+ virtual uint64_t getSymbolSize(DataRefImpl Symb) const = 0;
+ virtual uint8_t getSymbolBinding(DataRefImpl Symb) const = 0;
+ virtual uint8_t getSymbolOther(DataRefImpl Symb) const = 0;
+ virtual uint8_t getSymbolELFType(DataRefImpl Symb) const = 0;
+
+ virtual uint32_t getSectionType(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionFlags(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionOffset(DataRefImpl Sec) const = 0;
+
+ virtual Expected<int64_t> getRelocationAddend(DataRefImpl Rel) const = 0;
+ virtual Error getBuildAttributes(ARMAttributeParser &Attributes) const = 0;
+
+public:
+ using elf_symbol_iterator_range = iterator_range<elf_symbol_iterator>;
+
+ virtual elf_symbol_iterator_range getDynamicSymbolIterators() const = 0;
+
+ /// Returns platform-specific object flags, if any.
+ virtual unsigned getPlatformFlags() const = 0;
+
+ elf_symbol_iterator_range symbols() const;
+
+ static bool classof(const Binary *v) { return v->isELF(); }
+
+ SubtargetFeatures getFeatures() const override;
+
+ SubtargetFeatures getMIPSFeatures() const;
+
+ SubtargetFeatures getARMFeatures() const;
+
+ SubtargetFeatures getRISCVFeatures() const;
+
+ void setARMSubArch(Triple &TheTriple) const override;
+
+ virtual uint16_t getEType() const = 0;
+
+ virtual uint16_t getEMachine() const = 0;
+
+ std::vector<std::pair<DataRefImpl, uint64_t>> getPltAddresses() const;
+};
+
+class ELFSectionRef : public SectionRef {
+public:
+ ELFSectionRef(const SectionRef &B) : SectionRef(B) {
+ assert(isa<ELFObjectFileBase>(SectionRef::getObject()));
+ }
+
+ const ELFObjectFileBase *getObject() const {
+ return cast<ELFObjectFileBase>(SectionRef::getObject());
+ }
+
+ uint32_t getType() const {
+ return getObject()->getSectionType(getRawDataRefImpl());
+ }
+
+ uint64_t getFlags() const {
+ return getObject()->getSectionFlags(getRawDataRefImpl());
+ }
+
+ uint64_t getOffset() const {
+ return getObject()->getSectionOffset(getRawDataRefImpl());
+ }
+};
+
+class elf_section_iterator : public section_iterator {
+public:
+ elf_section_iterator(const section_iterator &B) : section_iterator(B) {
+ assert(isa<ELFObjectFileBase>(B->getObject()));
+ }
+
+ const ELFSectionRef *operator->() const {
+ return static_cast<const ELFSectionRef *>(section_iterator::operator->());
+ }
+
+ const ELFSectionRef &operator*() const {
+ return static_cast<const ELFSectionRef &>(section_iterator::operator*());
+ }
+};
+
+class ELFSymbolRef : public SymbolRef {
+public:
+ ELFSymbolRef(const SymbolRef &B) : SymbolRef(B) {
+ assert(isa<ELFObjectFileBase>(SymbolRef::getObject()));
+ }
+
+ const ELFObjectFileBase *getObject() const {
+ return cast<ELFObjectFileBase>(BasicSymbolRef::getObject());
+ }
+
+ uint64_t getSize() const {
+ return getObject()->getSymbolSize(getRawDataRefImpl());
+ }
+
+ uint8_t getBinding() const {
+ return getObject()->getSymbolBinding(getRawDataRefImpl());
+ }
+
+ uint8_t getOther() const {
+ return getObject()->getSymbolOther(getRawDataRefImpl());
+ }
+
+ uint8_t getELFType() const {
+ return getObject()->getSymbolELFType(getRawDataRefImpl());
+ }
+
+ StringRef getELFTypeName() const {
+ uint8_t Type = getELFType();
+ for (auto &EE : ElfSymbolTypes) {
+ if (EE.Value == Type) {
+ return EE.AltName;
+ }
+ }
+ return "";
+ }
+};
+
+class elf_symbol_iterator : public symbol_iterator {
+public:
+ elf_symbol_iterator(const basic_symbol_iterator &B)
+ : symbol_iterator(SymbolRef(B->getRawDataRefImpl(),
+ cast<ELFObjectFileBase>(B->getObject()))) {}
+
+ const ELFSymbolRef *operator->() const {
+ return static_cast<const ELFSymbolRef *>(symbol_iterator::operator->());
+ }
+
+ const ELFSymbolRef &operator*() const {
+ return static_cast<const ELFSymbolRef &>(symbol_iterator::operator*());
+ }
+};
+
+class ELFRelocationRef : public RelocationRef {
+public:
+ ELFRelocationRef(const RelocationRef &B) : RelocationRef(B) {
+ assert(isa<ELFObjectFileBase>(RelocationRef::getObject()));
+ }
+
+ const ELFObjectFileBase *getObject() const {
+ return cast<ELFObjectFileBase>(RelocationRef::getObject());
+ }
+
+ Expected<int64_t> getAddend() const {
+ return getObject()->getRelocationAddend(getRawDataRefImpl());
+ }
+};
+
+class elf_relocation_iterator : public relocation_iterator {
+public:
+ elf_relocation_iterator(const relocation_iterator &B)
+ : relocation_iterator(RelocationRef(
+ B->getRawDataRefImpl(), cast<ELFObjectFileBase>(B->getObject()))) {}
+
+ const ELFRelocationRef *operator->() const {
+ return static_cast<const ELFRelocationRef *>(
+ relocation_iterator::operator->());
+ }
+
+ const ELFRelocationRef &operator*() const {
+ return static_cast<const ELFRelocationRef &>(
+ relocation_iterator::operator*());
+ }
+};
+
+inline ELFObjectFileBase::elf_symbol_iterator_range
+ELFObjectFileBase::symbols() const {
+ return elf_symbol_iterator_range(symbol_begin(), symbol_end());
+}
+
+template <class ELFT> class ELFObjectFile : public ELFObjectFileBase {
+ uint16_t getEMachine() const override;
+ uint16_t getEType() const override;
+ uint64_t getSymbolSize(DataRefImpl Sym) const override;
+
+public:
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+
+ using uintX_t = typename ELFT::uint;
+
+ using Elf_Sym = typename ELFT::Sym;
+ using Elf_Shdr = typename ELFT::Shdr;
+ using Elf_Ehdr = typename ELFT::Ehdr;
+ using Elf_Rel = typename ELFT::Rel;
+ using Elf_Rela = typename ELFT::Rela;
+ using Elf_Dyn = typename ELFT::Dyn;
+
+ SectionRef toSectionRef(const Elf_Shdr *Sec) const {
+ return SectionRef(toDRI(Sec), this);
+ }
+
+private:
+ ELFObjectFile(MemoryBufferRef Object, ELFFile<ELFT> EF,
+ const Elf_Shdr *DotDynSymSec, const Elf_Shdr *DotSymtabSec,
+ ArrayRef<Elf_Word> ShndxTable);
+
+protected:
+ ELFFile<ELFT> EF;
+
+ const Elf_Shdr *DotDynSymSec = nullptr; // Dynamic symbol table section.
+ const Elf_Shdr *DotSymtabSec = nullptr; // Symbol table section.
+ ArrayRef<Elf_Word> ShndxTable;
+
+ void moveSymbolNext(DataRefImpl &Symb) const override;
+ Expected<StringRef> getSymbolName(DataRefImpl Symb) const override;
+ Expected<uint64_t> getSymbolAddress(DataRefImpl Symb) const override;
+ uint64_t getSymbolValueImpl(DataRefImpl Symb) const override;
+ uint32_t getSymbolAlignment(DataRefImpl Symb) const override;
+ uint64_t getCommonSymbolSizeImpl(DataRefImpl Symb) const override;
+ uint32_t getSymbolFlags(DataRefImpl Symb) const override;
+ uint8_t getSymbolBinding(DataRefImpl Symb) const override;
+ uint8_t getSymbolOther(DataRefImpl Symb) const override;
+ uint8_t getSymbolELFType(DataRefImpl Symb) const override;
+ Expected<SymbolRef::Type> getSymbolType(DataRefImpl Symb) const override;
+ Expected<section_iterator> getSymbolSection(const Elf_Sym *Symb,
+ const Elf_Shdr *SymTab) const;
+ Expected<section_iterator> getSymbolSection(DataRefImpl Symb) const override;
+
+ void moveSectionNext(DataRefImpl &Sec) const override;
+ Expected<StringRef> getSectionName(DataRefImpl Sec) const override;
+ uint64_t getSectionAddress(DataRefImpl Sec) const override;
+ uint64_t getSectionIndex(DataRefImpl Sec) const override;
+ uint64_t getSectionSize(DataRefImpl Sec) const override;
+ Expected<ArrayRef<uint8_t>>
+ getSectionContents(DataRefImpl Sec) const override;
+ uint64_t getSectionAlignment(DataRefImpl Sec) const override;
+ bool isSectionCompressed(DataRefImpl Sec) const override;
+ bool isSectionText(DataRefImpl Sec) const override;
+ bool isSectionData(DataRefImpl Sec) const override;
+ bool isSectionBSS(DataRefImpl Sec) const override;
+ bool isSectionVirtual(DataRefImpl Sec) const override;
+ bool isBerkeleyText(DataRefImpl Sec) const override;
+ bool isBerkeleyData(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_begin(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_end(DataRefImpl Sec) const override;
+ std::vector<SectionRef> dynamic_relocation_sections() const override;
+ Expected<section_iterator>
+ getRelocatedSection(DataRefImpl Sec) const override;
+
+ void moveRelocationNext(DataRefImpl &Rel) const override;
+ uint64_t getRelocationOffset(DataRefImpl Rel) const override;
+ symbol_iterator getRelocationSymbol(DataRefImpl Rel) const override;
+ uint64_t getRelocationType(DataRefImpl Rel) const override;
+ void getRelocationTypeName(DataRefImpl Rel,
+ SmallVectorImpl<char> &Result) const override;
+
+ uint32_t getSectionType(DataRefImpl Sec) const override;
+ uint64_t getSectionFlags(DataRefImpl Sec) const override;
+ uint64_t getSectionOffset(DataRefImpl Sec) const override;
+ StringRef getRelocationTypeName(uint32_t Type) const;
+
+ /// Get the relocation section that contains \a Rel.
+ const Elf_Shdr *getRelSection(DataRefImpl Rel) const {
+ auto RelSecOrErr = EF.getSection(Rel.d.a);
+ if (!RelSecOrErr)
+ report_fatal_error(errorToErrorCode(RelSecOrErr.takeError()).message());
+ return *RelSecOrErr;
+ }
+
+ DataRefImpl toDRI(const Elf_Shdr *SymTable, unsigned SymbolNum) const {
+ DataRefImpl DRI;
+ if (!SymTable) {
+ DRI.d.a = 0;
+ DRI.d.b = 0;
+ return DRI;
+ }
+ assert(SymTable->sh_type == ELF::SHT_SYMTAB ||
+ SymTable->sh_type == ELF::SHT_DYNSYM);
+
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr) {
+ DRI.d.a = 0;
+ DRI.d.b = 0;
+ return DRI;
+ }
+ uintptr_t SHT = reinterpret_cast<uintptr_t>((*SectionsOrErr).begin());
+ unsigned SymTableIndex =
+ (reinterpret_cast<uintptr_t>(SymTable) - SHT) / sizeof(Elf_Shdr);
+
+ DRI.d.a = SymTableIndex;
+ DRI.d.b = SymbolNum;
+ return DRI;
+ }
+
+ const Elf_Shdr *toELFShdrIter(DataRefImpl Sec) const {
+ return reinterpret_cast<const Elf_Shdr *>(Sec.p);
+ }
+
+ DataRefImpl toDRI(const Elf_Shdr *Sec) const {
+ DataRefImpl DRI;
+ DRI.p = reinterpret_cast<uintptr_t>(Sec);
+ return DRI;
+ }
+
+ DataRefImpl toDRI(const Elf_Dyn *Dyn) const {
+ DataRefImpl DRI;
+ DRI.p = reinterpret_cast<uintptr_t>(Dyn);
+ return DRI;
+ }
+
+ bool isExportedToOtherDSO(const Elf_Sym *ESym) const {
+ unsigned char Binding = ESym->getBinding();
+ unsigned char Visibility = ESym->getVisibility();
+
+ // A symbol is exported if its binding is either GLOBAL or WEAK, and its
+ // visibility is either DEFAULT or PROTECTED. All other symbols are not
+ // exported.
+ return (
+ (Binding == ELF::STB_GLOBAL || Binding == ELF::STB_WEAK ||
+ Binding == ELF::STB_GNU_UNIQUE) &&
+ (Visibility == ELF::STV_DEFAULT || Visibility == ELF::STV_PROTECTED));
+ }
+
+ Error getBuildAttributes(ARMAttributeParser &Attributes) const override {
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return SectionsOrErr.takeError();
+
+ for (const Elf_Shdr &Sec : *SectionsOrErr) {
+ if (Sec.sh_type == ELF::SHT_ARM_ATTRIBUTES) {
+ auto ErrorOrContents = EF.getSectionContents(&Sec);
+ if (!ErrorOrContents)
+ return ErrorOrContents.takeError();
+
+ auto Contents = ErrorOrContents.get();
+ if (Contents[0] != ARMBuildAttrs::Format_Version || Contents.size() == 1)
+ return Error::success();
+
+ Attributes.Parse(Contents, ELFT::TargetEndianness == support::little);
+ break;
+ }
+ }
+ return Error::success();
+ }
+
+ // This flag is used for classof, to distinguish ELFObjectFile from
+ // its subclass. If more subclasses will be created, this flag will
+ // have to become an enum.
+ bool isDyldELFObject;
+
+public:
+ ELFObjectFile(ELFObjectFile<ELFT> &&Other);
+ static Expected<ELFObjectFile<ELFT>> create(MemoryBufferRef Object);
+
+ const Elf_Rel *getRel(DataRefImpl Rel) const;
+ const Elf_Rela *getRela(DataRefImpl Rela) const;
+
+ const Elf_Sym *getSymbol(DataRefImpl Sym) const {
+ auto Ret = EF.template getEntry<Elf_Sym>(Sym.d.a, Sym.d.b);
+ if (!Ret)
+ report_fatal_error(errorToErrorCode(Ret.takeError()).message());
+ return *Ret;
+ }
+
+ const Elf_Shdr *getSection(DataRefImpl Sec) const {
+ return reinterpret_cast<const Elf_Shdr *>(Sec.p);
+ }
+
+ basic_symbol_iterator symbol_begin() const override;
+ basic_symbol_iterator symbol_end() const override;
+
+ elf_symbol_iterator dynamic_symbol_begin() const;
+ elf_symbol_iterator dynamic_symbol_end() const;
+
+ section_iterator section_begin() const override;
+ section_iterator section_end() const override;
+
+ Expected<int64_t> getRelocationAddend(DataRefImpl Rel) const override;
+
+ uint8_t getBytesInAddress() const override;
+ StringRef getFileFormatName() const override;
+ Triple::ArchType getArch() const override;
+ Expected<uint64_t> getStartAddress() const override;
+
+ unsigned getPlatformFlags() const override { return EF.getHeader()->e_flags; }
+
+ const ELFFile<ELFT> *getELFFile() const { return &EF; }
+
+ bool isDyldType() const { return isDyldELFObject; }
+ static bool classof(const Binary *v) {
+ return v->getType() == getELFType(ELFT::TargetEndianness == support::little,
+ ELFT::Is64Bits);
+ }
+
+ elf_symbol_iterator_range getDynamicSymbolIterators() const override;
+
+ bool isRelocatableObject() const override;
+};
+
+using ELF32LEObjectFile = ELFObjectFile<ELF32LE>;
+using ELF64LEObjectFile = ELFObjectFile<ELF64LE>;
+using ELF32BEObjectFile = ELFObjectFile<ELF32BE>;
+using ELF64BEObjectFile = ELFObjectFile<ELF64BE>;
+
+template <class ELFT>
+void ELFObjectFile<ELFT>::moveSymbolNext(DataRefImpl &Sym) const {
+ ++Sym.d.b;
+}
+
+template <class ELFT>
+Expected<StringRef> ELFObjectFile<ELFT>::getSymbolName(DataRefImpl Sym) const {
+ const Elf_Sym *ESym = getSymbol(Sym);
+ auto SymTabOrErr = EF.getSection(Sym.d.a);
+ if (!SymTabOrErr)
+ return SymTabOrErr.takeError();
+ const Elf_Shdr *SymTableSec = *SymTabOrErr;
+ auto StrTabOrErr = EF.getSection(SymTableSec->sh_link);
+ if (!StrTabOrErr)
+ return StrTabOrErr.takeError();
+ const Elf_Shdr *StringTableSec = *StrTabOrErr;
+ auto SymStrTabOrErr = EF.getStringTable(StringTableSec);
+ if (!SymStrTabOrErr)
+ return SymStrTabOrErr.takeError();
+ Expected<StringRef> Name = ESym->getName(*SymStrTabOrErr);
+ if (Name && !Name->empty())
+ return Name;
+
+ // If the symbol name is empty use the section name.
+ if (ESym->getType() == ELF::STT_SECTION) {
+ if (Expected<section_iterator> SecOrErr = getSymbolSection(Sym)) {
+ consumeError(Name.takeError());
+ return (*SecOrErr)->getName();
+ }
+ }
+ return Name;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionFlags(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_flags;
+}
+
+template <class ELFT>
+uint32_t ELFObjectFile<ELFT>::getSectionType(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_type;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionOffset(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_offset;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSymbolValueImpl(DataRefImpl Symb) const {
+ const Elf_Sym *ESym = getSymbol(Symb);
+ uint64_t Ret = ESym->st_value;
+ if (ESym->st_shndx == ELF::SHN_ABS)
+ return Ret;
+
+ const Elf_Ehdr *Header = EF.getHeader();
+ // Clear the ARM/Thumb or microMIPS indicator flag.
+ if ((Header->e_machine == ELF::EM_ARM || Header->e_machine == ELF::EM_MIPS) &&
+ ESym->getType() == ELF::STT_FUNC)
+ Ret &= ~1;
+
+ return Ret;
+}
+
+template <class ELFT>
+Expected<uint64_t>
+ELFObjectFile<ELFT>::getSymbolAddress(DataRefImpl Symb) const {
+ uint64_t Result = getSymbolValue(Symb);
+ const Elf_Sym *ESym = getSymbol(Symb);
+ switch (ESym->st_shndx) {
+ case ELF::SHN_COMMON:
+ case ELF::SHN_UNDEF:
+ case ELF::SHN_ABS:
+ return Result;
+ }
+
+ const Elf_Ehdr *Header = EF.getHeader();
+ auto SymTabOrErr = EF.getSection(Symb.d.a);
+ if (!SymTabOrErr)
+ return SymTabOrErr.takeError();
+ const Elf_Shdr *SymTab = *SymTabOrErr;
+
+ if (Header->e_type == ELF::ET_REL) {
+ auto SectionOrErr = EF.getSection(ESym, SymTab, ShndxTable);
+ if (!SectionOrErr)
+ return SectionOrErr.takeError();
+ const Elf_Shdr *Section = *SectionOrErr;
+ if (Section)
+ Result += Section->sh_addr;
+ }
+
+ return Result;
+}
+
+template <class ELFT>
+uint32_t ELFObjectFile<ELFT>::getSymbolAlignment(DataRefImpl Symb) const {
+ const Elf_Sym *Sym = getSymbol(Symb);
+ if (Sym->st_shndx == ELF::SHN_COMMON)
+ return Sym->st_value;
+ return 0;
+}
+
+template <class ELFT>
+uint16_t ELFObjectFile<ELFT>::getEMachine() const {
+ return EF.getHeader()->e_machine;
+}
+
+template <class ELFT> uint16_t ELFObjectFile<ELFT>::getEType() const {
+ return EF.getHeader()->e_type;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSymbolSize(DataRefImpl Sym) const {
+ return getSymbol(Sym)->st_size;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getCommonSymbolSizeImpl(DataRefImpl Symb) const {
+ return getSymbol(Symb)->st_size;
+}
+
+template <class ELFT>
+uint8_t ELFObjectFile<ELFT>::getSymbolBinding(DataRefImpl Symb) const {
+ return getSymbol(Symb)->getBinding();
+}
+
+template <class ELFT>
+uint8_t ELFObjectFile<ELFT>::getSymbolOther(DataRefImpl Symb) const {
+ return getSymbol(Symb)->st_other;
+}
+
+template <class ELFT>
+uint8_t ELFObjectFile<ELFT>::getSymbolELFType(DataRefImpl Symb) const {
+ return getSymbol(Symb)->getType();
+}
+
+template <class ELFT>
+Expected<SymbolRef::Type>
+ELFObjectFile<ELFT>::getSymbolType(DataRefImpl Symb) const {
+ const Elf_Sym *ESym = getSymbol(Symb);
+
+ switch (ESym->getType()) {
+ case ELF::STT_NOTYPE:
+ return SymbolRef::ST_Unknown;
+ case ELF::STT_SECTION:
+ return SymbolRef::ST_Debug;
+ case ELF::STT_FILE:
+ return SymbolRef::ST_File;
+ case ELF::STT_FUNC:
+ return SymbolRef::ST_Function;
+ case ELF::STT_OBJECT:
+ case ELF::STT_COMMON:
+ case ELF::STT_TLS:
+ return SymbolRef::ST_Data;
+ default:
+ return SymbolRef::ST_Other;
+ }
+}
+
+template <class ELFT>
+uint32_t ELFObjectFile<ELFT>::getSymbolFlags(DataRefImpl Sym) const {
+ const Elf_Sym *ESym = getSymbol(Sym);
+
+ uint32_t Result = SymbolRef::SF_None;
+
+ if (ESym->getBinding() != ELF::STB_LOCAL)
+ Result |= SymbolRef::SF_Global;
+
+ if (ESym->getBinding() == ELF::STB_WEAK)
+ Result |= SymbolRef::SF_Weak;
+
+ if (ESym->st_shndx == ELF::SHN_ABS)
+ Result |= SymbolRef::SF_Absolute;
+
+ if (ESym->getType() == ELF::STT_FILE || ESym->getType() == ELF::STT_SECTION)
+ Result |= SymbolRef::SF_FormatSpecific;
+
+ auto DotSymtabSecSyms = EF.symbols(DotSymtabSec);
+ if (DotSymtabSecSyms && ESym == (*DotSymtabSecSyms).begin())
+ Result |= SymbolRef::SF_FormatSpecific;
+ auto DotDynSymSecSyms = EF.symbols(DotDynSymSec);
+ if (DotDynSymSecSyms && ESym == (*DotDynSymSecSyms).begin())
+ Result |= SymbolRef::SF_FormatSpecific;
+
+ if (EF.getHeader()->e_machine == ELF::EM_ARM) {
+ if (Expected<StringRef> NameOrErr = getSymbolName(Sym)) {
+ StringRef Name = *NameOrErr;
+ if (Name.startswith("$d") || Name.startswith("$t") ||
+ Name.startswith("$a"))
+ Result |= SymbolRef::SF_FormatSpecific;
+ } else {
+ // TODO: Actually report errors helpfully.
+ consumeError(NameOrErr.takeError());
+ }
+ if (ESym->getType() == ELF::STT_FUNC && (ESym->st_value & 1) == 1)
+ Result |= SymbolRef::SF_Thumb;
+ }
+
+ if (ESym->st_shndx == ELF::SHN_UNDEF)
+ Result |= SymbolRef::SF_Undefined;
+
+ if (ESym->getType() == ELF::STT_COMMON || ESym->st_shndx == ELF::SHN_COMMON)
+ Result |= SymbolRef::SF_Common;
+
+ if (isExportedToOtherDSO(ESym))
+ Result |= SymbolRef::SF_Exported;
+
+ if (ESym->getVisibility() == ELF::STV_HIDDEN)
+ Result |= SymbolRef::SF_Hidden;
+
+ return Result;
+}
+
+template <class ELFT>
+Expected<section_iterator>
+ELFObjectFile<ELFT>::getSymbolSection(const Elf_Sym *ESym,
+ const Elf_Shdr *SymTab) const {
+ auto ESecOrErr = EF.getSection(ESym, SymTab, ShndxTable);
+ if (!ESecOrErr)
+ return ESecOrErr.takeError();
+
+ const Elf_Shdr *ESec = *ESecOrErr;
+ if (!ESec)
+ return section_end();
+
+ DataRefImpl Sec;
+ Sec.p = reinterpret_cast<intptr_t>(ESec);
+ return section_iterator(SectionRef(Sec, this));
+}
+
+template <class ELFT>
+Expected<section_iterator>
+ELFObjectFile<ELFT>::getSymbolSection(DataRefImpl Symb) const {
+ const Elf_Sym *Sym = getSymbol(Symb);
+ auto SymTabOrErr = EF.getSection(Symb.d.a);
+ if (!SymTabOrErr)
+ return SymTabOrErr.takeError();
+ const Elf_Shdr *SymTab = *SymTabOrErr;
+ return getSymbolSection(Sym, SymTab);
+}
+
+template <class ELFT>
+void ELFObjectFile<ELFT>::moveSectionNext(DataRefImpl &Sec) const {
+ const Elf_Shdr *ESec = getSection(Sec);
+ Sec = toDRI(++ESec);
+}
+
+template <class ELFT>
+Expected<StringRef> ELFObjectFile<ELFT>::getSectionName(DataRefImpl Sec) const {
+ return EF.getSectionName(&*getSection(Sec));
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionAddress(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_addr;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionIndex(DataRefImpl Sec) const {
+ auto SectionsOrErr = EF.sections();
+ handleAllErrors(std::move(SectionsOrErr.takeError()),
+ [](const ErrorInfoBase &) {
+ llvm_unreachable("unable to get section index");
+ });
+ const Elf_Shdr *First = SectionsOrErr->begin();
+ return getSection(Sec) - First;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionSize(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_size;
+}
+
+template <class ELFT>
+Expected<ArrayRef<uint8_t>>
+ELFObjectFile<ELFT>::getSectionContents(DataRefImpl Sec) const {
+ const Elf_Shdr *EShdr = getSection(Sec);
+ if (EShdr->sh_type == ELF::SHT_NOBITS)
+ return makeArrayRef((const uint8_t *)base(), 0);
+ if (std::error_code EC =
+ checkOffset(getMemoryBufferRef(),
+ (uintptr_t)base() + EShdr->sh_offset, EShdr->sh_size))
+ return errorCodeToError(EC);
+ return makeArrayRef((const uint8_t *)base() + EShdr->sh_offset,
+ EShdr->sh_size);
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getSectionAlignment(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_addralign;
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isSectionCompressed(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_flags & ELF::SHF_COMPRESSED;
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isSectionText(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_flags & ELF::SHF_EXECINSTR;
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isSectionData(DataRefImpl Sec) const {
+ const Elf_Shdr *EShdr = getSection(Sec);
+ return EShdr->sh_type == ELF::SHT_PROGBITS &&
+ EShdr->sh_flags & ELF::SHF_ALLOC &&
+ !(EShdr->sh_flags & ELF::SHF_EXECINSTR);
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isSectionBSS(DataRefImpl Sec) const {
+ const Elf_Shdr *EShdr = getSection(Sec);
+ return EShdr->sh_flags & (ELF::SHF_ALLOC | ELF::SHF_WRITE) &&
+ EShdr->sh_type == ELF::SHT_NOBITS;
+}
+
+template <class ELFT>
+std::vector<SectionRef>
+ELFObjectFile<ELFT>::dynamic_relocation_sections() const {
+ std::vector<SectionRef> Res;
+ std::vector<uintptr_t> Offsets;
+
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return Res;
+
+ for (const Elf_Shdr &Sec : *SectionsOrErr) {
+ if (Sec.sh_type != ELF::SHT_DYNAMIC)
+ continue;
+ Elf_Dyn *Dynamic =
+ reinterpret_cast<Elf_Dyn *>((uintptr_t)base() + Sec.sh_offset);
+ for (; Dynamic->d_tag != ELF::DT_NULL; Dynamic++) {
+ if (Dynamic->d_tag == ELF::DT_REL || Dynamic->d_tag == ELF::DT_RELA ||
+ Dynamic->d_tag == ELF::DT_JMPREL) {
+ Offsets.push_back(Dynamic->d_un.d_val);
+ }
+ }
+ }
+ for (const Elf_Shdr &Sec : *SectionsOrErr) {
+ if (is_contained(Offsets, Sec.sh_addr))
+ Res.emplace_back(toDRI(&Sec), this);
+ }
+ return Res;
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isSectionVirtual(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_type == ELF::SHT_NOBITS;
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isBerkeleyText(DataRefImpl Sec) const {
+ return getSection(Sec)->sh_flags & ELF::SHF_ALLOC &&
+ (getSection(Sec)->sh_flags & ELF::SHF_EXECINSTR ||
+ !(getSection(Sec)->sh_flags & ELF::SHF_WRITE));
+}
+
+template <class ELFT>
+bool ELFObjectFile<ELFT>::isBerkeleyData(DataRefImpl Sec) const {
+ const Elf_Shdr *EShdr = getSection(Sec);
+ return !isBerkeleyText(Sec) && EShdr->sh_type != ELF::SHT_NOBITS &&
+ EShdr->sh_flags & ELF::SHF_ALLOC;
+}
+
+template <class ELFT>
+relocation_iterator
+ELFObjectFile<ELFT>::section_rel_begin(DataRefImpl Sec) const {
+ DataRefImpl RelData;
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return relocation_iterator(RelocationRef());
+ uintptr_t SHT = reinterpret_cast<uintptr_t>((*SectionsOrErr).begin());
+ RelData.d.a = (Sec.p - SHT) / EF.getHeader()->e_shentsize;
+ RelData.d.b = 0;
+ return relocation_iterator(RelocationRef(RelData, this));
+}
+
+template <class ELFT>
+relocation_iterator
+ELFObjectFile<ELFT>::section_rel_end(DataRefImpl Sec) const {
+ const Elf_Shdr *S = reinterpret_cast<const Elf_Shdr *>(Sec.p);
+ relocation_iterator Begin = section_rel_begin(Sec);
+ if (S->sh_type != ELF::SHT_RELA && S->sh_type != ELF::SHT_REL)
+ return Begin;
+ DataRefImpl RelData = Begin->getRawDataRefImpl();
+ const Elf_Shdr *RelSec = getRelSection(RelData);
+
+ // Error check sh_link here so that getRelocationSymbol can just use it.
+ auto SymSecOrErr = EF.getSection(RelSec->sh_link);
+ if (!SymSecOrErr)
+ report_fatal_error(errorToErrorCode(SymSecOrErr.takeError()).message());
+
+ RelData.d.b += S->sh_size / S->sh_entsize;
+ return relocation_iterator(RelocationRef(RelData, this));
+}
+
+template <class ELFT>
+Expected<section_iterator>
+ELFObjectFile<ELFT>::getRelocatedSection(DataRefImpl Sec) const {
+ if (EF.getHeader()->e_type != ELF::ET_REL)
+ return section_end();
+
+ const Elf_Shdr *EShdr = getSection(Sec);
+ uintX_t Type = EShdr->sh_type;
+ if (Type != ELF::SHT_REL && Type != ELF::SHT_RELA)
+ return section_end();
+
+ Expected<const Elf_Shdr *> SecOrErr = EF.getSection(EShdr->sh_info);
+ if (!SecOrErr)
+ return SecOrErr.takeError();
+ return section_iterator(SectionRef(toDRI(*SecOrErr), this));
+}
+
+// Relocations
+template <class ELFT>
+void ELFObjectFile<ELFT>::moveRelocationNext(DataRefImpl &Rel) const {
+ ++Rel.d.b;
+}
+
+template <class ELFT>
+symbol_iterator
+ELFObjectFile<ELFT>::getRelocationSymbol(DataRefImpl Rel) const {
+ uint32_t symbolIdx;
+ const Elf_Shdr *sec = getRelSection(Rel);
+ if (sec->sh_type == ELF::SHT_REL)
+ symbolIdx = getRel(Rel)->getSymbol(EF.isMips64EL());
+ else
+ symbolIdx = getRela(Rel)->getSymbol(EF.isMips64EL());
+ if (!symbolIdx)
+ return symbol_end();
+
+ // FIXME: error check symbolIdx
+ DataRefImpl SymbolData;
+ SymbolData.d.a = sec->sh_link;
+ SymbolData.d.b = symbolIdx;
+ return symbol_iterator(SymbolRef(SymbolData, this));
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getRelocationOffset(DataRefImpl Rel) const {
+ const Elf_Shdr *sec = getRelSection(Rel);
+ if (sec->sh_type == ELF::SHT_REL)
+ return getRel(Rel)->r_offset;
+
+ return getRela(Rel)->r_offset;
+}
+
+template <class ELFT>
+uint64_t ELFObjectFile<ELFT>::getRelocationType(DataRefImpl Rel) const {
+ const Elf_Shdr *sec = getRelSection(Rel);
+ if (sec->sh_type == ELF::SHT_REL)
+ return getRel(Rel)->getType(EF.isMips64EL());
+ else
+ return getRela(Rel)->getType(EF.isMips64EL());
+}
+
+template <class ELFT>
+StringRef ELFObjectFile<ELFT>::getRelocationTypeName(uint32_t Type) const {
+ return getELFRelocationTypeName(EF.getHeader()->e_machine, Type);
+}
+
+template <class ELFT>
+void ELFObjectFile<ELFT>::getRelocationTypeName(
+ DataRefImpl Rel, SmallVectorImpl<char> &Result) const {
+ uint32_t type = getRelocationType(Rel);
+ EF.getRelocationTypeName(type, Result);
+}
+
+template <class ELFT>
+Expected<int64_t>
+ELFObjectFile<ELFT>::getRelocationAddend(DataRefImpl Rel) const {
+ if (getRelSection(Rel)->sh_type != ELF::SHT_RELA)
+ return createError("Section is not SHT_RELA");
+ return (int64_t)getRela(Rel)->r_addend;
+}
+
+template <class ELFT>
+const typename ELFObjectFile<ELFT>::Elf_Rel *
+ELFObjectFile<ELFT>::getRel(DataRefImpl Rel) const {
+ assert(getRelSection(Rel)->sh_type == ELF::SHT_REL);
+ auto Ret = EF.template getEntry<Elf_Rel>(Rel.d.a, Rel.d.b);
+ if (!Ret)
+ report_fatal_error(errorToErrorCode(Ret.takeError()).message());
+ return *Ret;
+}
+
+template <class ELFT>
+const typename ELFObjectFile<ELFT>::Elf_Rela *
+ELFObjectFile<ELFT>::getRela(DataRefImpl Rela) const {
+ assert(getRelSection(Rela)->sh_type == ELF::SHT_RELA);
+ auto Ret = EF.template getEntry<Elf_Rela>(Rela.d.a, Rela.d.b);
+ if (!Ret)
+ report_fatal_error(errorToErrorCode(Ret.takeError()).message());
+ return *Ret;
+}
+
+template <class ELFT>
+Expected<ELFObjectFile<ELFT>>
+ELFObjectFile<ELFT>::create(MemoryBufferRef Object) {
+ auto EFOrErr = ELFFile<ELFT>::create(Object.getBuffer());
+ if (Error E = EFOrErr.takeError())
+ return std::move(E);
+ auto EF = std::move(*EFOrErr);
+
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return SectionsOrErr.takeError();
+
+ const Elf_Shdr *DotDynSymSec = nullptr;
+ const Elf_Shdr *DotSymtabSec = nullptr;
+ ArrayRef<Elf_Word> ShndxTable;
+ for (const Elf_Shdr &Sec : *SectionsOrErr) {
+ switch (Sec.sh_type) {
+ case ELF::SHT_DYNSYM: {
+ if (!DotDynSymSec)
+ DotDynSymSec = &Sec;
+ break;
+ }
+ case ELF::SHT_SYMTAB: {
+ if (!DotSymtabSec)
+ DotSymtabSec = &Sec;
+ break;
+ }
+ case ELF::SHT_SYMTAB_SHNDX: {
+ auto TableOrErr = EF.getSHNDXTable(Sec);
+ if (!TableOrErr)
+ return TableOrErr.takeError();
+ ShndxTable = *TableOrErr;
+ break;
+ }
+ }
+ }
+ return ELFObjectFile<ELFT>(Object, EF, DotDynSymSec, DotSymtabSec,
+ ShndxTable);
+}
+
+template <class ELFT>
+ELFObjectFile<ELFT>::ELFObjectFile(MemoryBufferRef Object, ELFFile<ELFT> EF,
+ const Elf_Shdr *DotDynSymSec,
+ const Elf_Shdr *DotSymtabSec,
+ ArrayRef<Elf_Word> ShndxTable)
+ : ELFObjectFileBase(
+ getELFType(ELFT::TargetEndianness == support::little, ELFT::Is64Bits),
+ Object),
+ EF(EF), DotDynSymSec(DotDynSymSec), DotSymtabSec(DotSymtabSec),
+ ShndxTable(ShndxTable) {}
+
+template <class ELFT>
+ELFObjectFile<ELFT>::ELFObjectFile(ELFObjectFile<ELFT> &&Other)
+ : ELFObjectFile(Other.Data, Other.EF, Other.DotDynSymSec,
+ Other.DotSymtabSec, Other.ShndxTable) {}
+
+template <class ELFT>
+basic_symbol_iterator ELFObjectFile<ELFT>::symbol_begin() const {
+ DataRefImpl Sym =
+ toDRI(DotSymtabSec,
+ DotSymtabSec && DotSymtabSec->sh_size >= sizeof(Elf_Sym) ? 1 : 0);
+ return basic_symbol_iterator(SymbolRef(Sym, this));
+}
+
+template <class ELFT>
+basic_symbol_iterator ELFObjectFile<ELFT>::symbol_end() const {
+ const Elf_Shdr *SymTab = DotSymtabSec;
+ if (!SymTab)
+ return symbol_begin();
+ DataRefImpl Sym = toDRI(SymTab, SymTab->sh_size / sizeof(Elf_Sym));
+ return basic_symbol_iterator(SymbolRef(Sym, this));
+}
+
+template <class ELFT>
+elf_symbol_iterator ELFObjectFile<ELFT>::dynamic_symbol_begin() const {
+ DataRefImpl Sym = toDRI(DotDynSymSec, 0);
+ return symbol_iterator(SymbolRef(Sym, this));
+}
+
+template <class ELFT>
+elf_symbol_iterator ELFObjectFile<ELFT>::dynamic_symbol_end() const {
+ const Elf_Shdr *SymTab = DotDynSymSec;
+ if (!SymTab)
+ return dynamic_symbol_begin();
+ DataRefImpl Sym = toDRI(SymTab, SymTab->sh_size / sizeof(Elf_Sym));
+ return basic_symbol_iterator(SymbolRef(Sym, this));
+}
+
+template <class ELFT>
+section_iterator ELFObjectFile<ELFT>::section_begin() const {
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return section_iterator(SectionRef());
+ return section_iterator(SectionRef(toDRI((*SectionsOrErr).begin()), this));
+}
+
+template <class ELFT>
+section_iterator ELFObjectFile<ELFT>::section_end() const {
+ auto SectionsOrErr = EF.sections();
+ if (!SectionsOrErr)
+ return section_iterator(SectionRef());
+ return section_iterator(SectionRef(toDRI((*SectionsOrErr).end()), this));
+}
+
+template <class ELFT>
+uint8_t ELFObjectFile<ELFT>::getBytesInAddress() const {
+ return ELFT::Is64Bits ? 8 : 4;
+}
+
+template <class ELFT>
+StringRef ELFObjectFile<ELFT>::getFileFormatName() const {
+ bool IsLittleEndian = ELFT::TargetEndianness == support::little;
+ switch (EF.getHeader()->e_ident[ELF::EI_CLASS]) {
+ case ELF::ELFCLASS32:
+ switch (EF.getHeader()->e_machine) {
+ case ELF::EM_386:
+ return "ELF32-i386";
+ case ELF::EM_IAMCU:
+ return "ELF32-iamcu";
+ case ELF::EM_X86_64:
+ return "ELF32-x86-64";
+ case ELF::EM_ARM:
+ return (IsLittleEndian ? "ELF32-arm-little" : "ELF32-arm-big");
+ case ELF::EM_AVR:
+ return "ELF32-avr";
+ case ELF::EM_HEXAGON:
+ return "ELF32-hexagon";
+ case ELF::EM_LANAI:
+ return "ELF32-lanai";
+ case ELF::EM_MIPS:
+ return "ELF32-mips";
+ case ELF::EM_MSP430:
+ return "ELF32-msp430";
+ case ELF::EM_PPC:
+ return "ELF32-ppc";
+ case ELF::EM_RISCV:
+ return "ELF32-riscv";
+ case ELF::EM_SPARC:
+ case ELF::EM_SPARC32PLUS:
+ return "ELF32-sparc";
+ case ELF::EM_AMDGPU:
+ return "ELF32-amdgpu";
+ default:
+ return "ELF32-unknown";
+ }
+ case ELF::ELFCLASS64:
+ switch (EF.getHeader()->e_machine) {
+ case ELF::EM_386:
+ return "ELF64-i386";
+ case ELF::EM_X86_64:
+ return "ELF64-x86-64";
+ case ELF::EM_AARCH64:
+ return (IsLittleEndian ? "ELF64-aarch64-little" : "ELF64-aarch64-big");
+ case ELF::EM_PPC64:
+ return "ELF64-ppc64";
+ case ELF::EM_RISCV:
+ return "ELF64-riscv";
+ case ELF::EM_S390:
+ return "ELF64-s390";
+ case ELF::EM_SPARCV9:
+ return "ELF64-sparc";
+ case ELF::EM_MIPS:
+ return "ELF64-mips";
+ case ELF::EM_AMDGPU:
+ return "ELF64-amdgpu";
+ case ELF::EM_BPF:
+ return "ELF64-BPF";
+ default:
+ return "ELF64-unknown";
+ }
+ default:
+ // FIXME: Proper error handling.
+ report_fatal_error("Invalid ELFCLASS!");
+ }
+}
+
+template <class ELFT> Triple::ArchType ELFObjectFile<ELFT>::getArch() const {
+ bool IsLittleEndian = ELFT::TargetEndianness == support::little;
+ switch (EF.getHeader()->e_machine) {
+ case ELF::EM_386:
+ case ELF::EM_IAMCU:
+ return Triple::x86;
+ case ELF::EM_X86_64:
+ return Triple::x86_64;
+ case ELF::EM_AARCH64:
+ return IsLittleEndian ? Triple::aarch64 : Triple::aarch64_be;
+ case ELF::EM_ARM:
+ return Triple::arm;
+ case ELF::EM_AVR:
+ return Triple::avr;
+ case ELF::EM_HEXAGON:
+ return Triple::hexagon;
+ case ELF::EM_LANAI:
+ return Triple::lanai;
+ case ELF::EM_MIPS:
+ switch (EF.getHeader()->e_ident[ELF::EI_CLASS]) {
+ case ELF::ELFCLASS32:
+ return IsLittleEndian ? Triple::mipsel : Triple::mips;
+ case ELF::ELFCLASS64:
+ return IsLittleEndian ? Triple::mips64el : Triple::mips64;
+ default:
+ report_fatal_error("Invalid ELFCLASS!");
+ }
+ case ELF::EM_MSP430:
+ return Triple::msp430;
+ case ELF::EM_PPC:
+ return Triple::ppc;
+ case ELF::EM_PPC64:
+ return IsLittleEndian ? Triple::ppc64le : Triple::ppc64;
+ case ELF::EM_RISCV:
+ switch (EF.getHeader()->e_ident[ELF::EI_CLASS]) {
+ case ELF::ELFCLASS32:
+ return Triple::riscv32;
+ case ELF::ELFCLASS64:
+ return Triple::riscv64;
+ default:
+ report_fatal_error("Invalid ELFCLASS!");
+ }
+ case ELF::EM_S390:
+ return Triple::systemz;
+
+ case ELF::EM_SPARC:
+ case ELF::EM_SPARC32PLUS:
+ return IsLittleEndian ? Triple::sparcel : Triple::sparc;
+ case ELF::EM_SPARCV9:
+ return Triple::sparcv9;
+
+ case ELF::EM_AMDGPU: {
+ if (!IsLittleEndian)
+ return Triple::UnknownArch;
+
+ unsigned MACH = EF.getHeader()->e_flags & ELF::EF_AMDGPU_MACH;
+ if (MACH >= ELF::EF_AMDGPU_MACH_R600_FIRST &&
+ MACH <= ELF::EF_AMDGPU_MACH_R600_LAST)
+ return Triple::r600;
+ if (MACH >= ELF::EF_AMDGPU_MACH_AMDGCN_FIRST &&
+ MACH <= ELF::EF_AMDGPU_MACH_AMDGCN_LAST)
+ return Triple::amdgcn;
+
+ return Triple::UnknownArch;
+ }
+
+ case ELF::EM_BPF:
+ return IsLittleEndian ? Triple::bpfel : Triple::bpfeb;
+
+ default:
+ return Triple::UnknownArch;
+ }
+}
+
+template <class ELFT>
+Expected<uint64_t> ELFObjectFile<ELFT>::getStartAddress() const {
+ return EF.getHeader()->e_entry;
+}
+
+template <class ELFT>
+ELFObjectFileBase::elf_symbol_iterator_range
+ELFObjectFile<ELFT>::getDynamicSymbolIterators() const {
+ return make_range(dynamic_symbol_begin(), dynamic_symbol_end());
+}
+
+template <class ELFT> bool ELFObjectFile<ELFT>::isRelocatableObject() const {
+ return EF.getHeader()->e_type == ELF::ET_REL;
+}
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_ELFOBJECTFILE_H
diff --git a/third_party/llvm-project/include/llvm/Object/ELFTypes.h b/third_party/llvm-project/include/llvm/Object/ELFTypes.h
new file mode 100644
index 000000000..7d1ade4d5
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/ELFTypes.h
@@ -0,0 +1,763 @@
+//===- ELFTypes.h - Endian specific types for ELF ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_ELFTYPES_H
+#define LLVM_OBJECT_ELFTYPES_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/ELF.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Error.h"
+#include <cassert>
+#include <cstdint>
+#include <cstring>
+#include <type_traits>
+
+namespace llvm {
+namespace object {
+
+using support::endianness;
+
+template <class ELFT> struct Elf_Ehdr_Impl;
+template <class ELFT> struct Elf_Shdr_Impl;
+template <class ELFT> struct Elf_Sym_Impl;
+template <class ELFT> struct Elf_Dyn_Impl;
+template <class ELFT> struct Elf_Phdr_Impl;
+template <class ELFT, bool isRela> struct Elf_Rel_Impl;
+template <class ELFT> struct Elf_Verdef_Impl;
+template <class ELFT> struct Elf_Verdaux_Impl;
+template <class ELFT> struct Elf_Verneed_Impl;
+template <class ELFT> struct Elf_Vernaux_Impl;
+template <class ELFT> struct Elf_Versym_Impl;
+template <class ELFT> struct Elf_Hash_Impl;
+template <class ELFT> struct Elf_GnuHash_Impl;
+template <class ELFT> struct Elf_Chdr_Impl;
+template <class ELFT> struct Elf_Nhdr_Impl;
+template <class ELFT> class Elf_Note_Impl;
+template <class ELFT> class Elf_Note_Iterator_Impl;
+template <class ELFT> struct Elf_CGProfile_Impl;
+
+template <endianness E, bool Is64> struct ELFType {
+private:
+ template <typename Ty>
+ using packed = support::detail::packed_endian_specific_integral<Ty, E, 1>;
+
+public:
+ static const endianness TargetEndianness = E;
+ static const bool Is64Bits = Is64;
+
+ using uint = typename std::conditional<Is64, uint64_t, uint32_t>::type;
+ using Ehdr = Elf_Ehdr_Impl<ELFType<E, Is64>>;
+ using Shdr = Elf_Shdr_Impl<ELFType<E, Is64>>;
+ using Sym = Elf_Sym_Impl<ELFType<E, Is64>>;
+ using Dyn = Elf_Dyn_Impl<ELFType<E, Is64>>;
+ using Phdr = Elf_Phdr_Impl<ELFType<E, Is64>>;
+ using Rel = Elf_Rel_Impl<ELFType<E, Is64>, false>;
+ using Rela = Elf_Rel_Impl<ELFType<E, Is64>, true>;
+ using Relr = packed<uint>;
+ using Verdef = Elf_Verdef_Impl<ELFType<E, Is64>>;
+ using Verdaux = Elf_Verdaux_Impl<ELFType<E, Is64>>;
+ using Verneed = Elf_Verneed_Impl<ELFType<E, Is64>>;
+ using Vernaux = Elf_Vernaux_Impl<ELFType<E, Is64>>;
+ using Versym = Elf_Versym_Impl<ELFType<E, Is64>>;
+ using Hash = Elf_Hash_Impl<ELFType<E, Is64>>;
+ using GnuHash = Elf_GnuHash_Impl<ELFType<E, Is64>>;
+ using Chdr = Elf_Chdr_Impl<ELFType<E, Is64>>;
+ using Nhdr = Elf_Nhdr_Impl<ELFType<E, Is64>>;
+ using Note = Elf_Note_Impl<ELFType<E, Is64>>;
+ using NoteIterator = Elf_Note_Iterator_Impl<ELFType<E, Is64>>;
+ using CGProfile = Elf_CGProfile_Impl<ELFType<E, Is64>>;
+ using DynRange = ArrayRef<Dyn>;
+ using ShdrRange = ArrayRef<Shdr>;
+ using SymRange = ArrayRef<Sym>;
+ using RelRange = ArrayRef<Rel>;
+ using RelaRange = ArrayRef<Rela>;
+ using RelrRange = ArrayRef<Relr>;
+ using PhdrRange = ArrayRef<Phdr>;
+
+ using Half = packed<uint16_t>;
+ using Word = packed<uint32_t>;
+ using Sword = packed<int32_t>;
+ using Xword = packed<uint64_t>;
+ using Sxword = packed<int64_t>;
+ using Addr = packed<uint>;
+ using Off = packed<uint>;
+};
+
+using ELF32LE = ELFType<support::little, false>;
+using ELF32BE = ELFType<support::big, false>;
+using ELF64LE = ELFType<support::little, true>;
+using ELF64BE = ELFType<support::big, true>;
+
+// Use an alignment of 2 for the typedefs since that is the worst case for
+// ELF files in archives.
+
+// I really don't like doing this, but the alternative is copypasta.
+#define LLVM_ELF_IMPORT_TYPES_ELFT(ELFT) \
+ using Elf_Addr = typename ELFT::Addr; \
+ using Elf_Off = typename ELFT::Off; \
+ using Elf_Half = typename ELFT::Half; \
+ using Elf_Word = typename ELFT::Word; \
+ using Elf_Sword = typename ELFT::Sword; \
+ using Elf_Xword = typename ELFT::Xword; \
+ using Elf_Sxword = typename ELFT::Sxword;
+
+#define LLVM_ELF_COMMA ,
+#define LLVM_ELF_IMPORT_TYPES(E, W) \
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFType<E LLVM_ELF_COMMA W>)
+
+// Section header.
+template <class ELFT> struct Elf_Shdr_Base;
+
+template <endianness TargetEndianness>
+struct Elf_Shdr_Base<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Word sh_name; // Section name (index into string table)
+ Elf_Word sh_type; // Section type (SHT_*)
+ Elf_Word sh_flags; // Section flags (SHF_*)
+ Elf_Addr sh_addr; // Address where section is to be loaded
+ Elf_Off sh_offset; // File offset of section data, in bytes
+ Elf_Word sh_size; // Size of section, in bytes
+ Elf_Word sh_link; // Section type-specific header table index link
+ Elf_Word sh_info; // Section type-specific extra information
+ Elf_Word sh_addralign; // Section address alignment
+ Elf_Word sh_entsize; // Size of records contained within the section
+};
+
+template <endianness TargetEndianness>
+struct Elf_Shdr_Base<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Word sh_name; // Section name (index into string table)
+ Elf_Word sh_type; // Section type (SHT_*)
+ Elf_Xword sh_flags; // Section flags (SHF_*)
+ Elf_Addr sh_addr; // Address where section is to be loaded
+ Elf_Off sh_offset; // File offset of section data, in bytes
+ Elf_Xword sh_size; // Size of section, in bytes
+ Elf_Word sh_link; // Section type-specific header table index link
+ Elf_Word sh_info; // Section type-specific extra information
+ Elf_Xword sh_addralign; // Section address alignment
+ Elf_Xword sh_entsize; // Size of records contained within the section
+};
+
+template <class ELFT>
+struct Elf_Shdr_Impl : Elf_Shdr_Base<ELFT> {
+ using Elf_Shdr_Base<ELFT>::sh_entsize;
+ using Elf_Shdr_Base<ELFT>::sh_size;
+
+ /// Get the number of entities this section contains if it has any.
+ unsigned getEntityCount() const {
+ if (sh_entsize == 0)
+ return 0;
+ return sh_size / sh_entsize;
+ }
+};
+
+template <class ELFT> struct Elf_Sym_Base;
+
+template <endianness TargetEndianness>
+struct Elf_Sym_Base<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Word st_name; // Symbol name (index into string table)
+ Elf_Addr st_value; // Value or address associated with the symbol
+ Elf_Word st_size; // Size of the symbol
+ unsigned char st_info; // Symbol's type and binding attributes
+ unsigned char st_other; // Must be zero; reserved
+ Elf_Half st_shndx; // Which section (header table index) it's defined in
+};
+
+template <endianness TargetEndianness>
+struct Elf_Sym_Base<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Word st_name; // Symbol name (index into string table)
+ unsigned char st_info; // Symbol's type and binding attributes
+ unsigned char st_other; // Must be zero; reserved
+ Elf_Half st_shndx; // Which section (header table index) it's defined in
+ Elf_Addr st_value; // Value or address associated with the symbol
+ Elf_Xword st_size; // Size of the symbol
+};
+
+template <class ELFT>
+struct Elf_Sym_Impl : Elf_Sym_Base<ELFT> {
+ using Elf_Sym_Base<ELFT>::st_info;
+ using Elf_Sym_Base<ELFT>::st_shndx;
+ using Elf_Sym_Base<ELFT>::st_other;
+ using Elf_Sym_Base<ELFT>::st_value;
+
+ // These accessors and mutators correspond to the ELF32_ST_BIND,
+ // ELF32_ST_TYPE, and ELF32_ST_INFO macros defined in the ELF specification:
+ unsigned char getBinding() const { return st_info >> 4; }
+ unsigned char getType() const { return st_info & 0x0f; }
+ uint64_t getValue() const { return st_value; }
+ void setBinding(unsigned char b) { setBindingAndType(b, getType()); }
+ void setType(unsigned char t) { setBindingAndType(getBinding(), t); }
+
+ void setBindingAndType(unsigned char b, unsigned char t) {
+ st_info = (b << 4) + (t & 0x0f);
+ }
+
+ /// Access to the STV_xxx flag stored in the first two bits of st_other.
+ /// STV_DEFAULT: 0
+ /// STV_INTERNAL: 1
+ /// STV_HIDDEN: 2
+ /// STV_PROTECTED: 3
+ unsigned char getVisibility() const { return st_other & 0x3; }
+ void setVisibility(unsigned char v) {
+ assert(v < 4 && "Invalid value for visibility");
+ st_other = (st_other & ~0x3) | v;
+ }
+
+ bool isAbsolute() const { return st_shndx == ELF::SHN_ABS; }
+
+ bool isCommon() const {
+ return getType() == ELF::STT_COMMON || st_shndx == ELF::SHN_COMMON;
+ }
+
+ bool isDefined() const { return !isUndefined(); }
+
+ bool isProcessorSpecific() const {
+ return st_shndx >= ELF::SHN_LOPROC && st_shndx <= ELF::SHN_HIPROC;
+ }
+
+ bool isOSSpecific() const {
+ return st_shndx >= ELF::SHN_LOOS && st_shndx <= ELF::SHN_HIOS;
+ }
+
+ bool isReserved() const {
+ // ELF::SHN_HIRESERVE is 0xffff so st_shndx <= ELF::SHN_HIRESERVE is always
+ // true and some compilers warn about it.
+ return st_shndx >= ELF::SHN_LORESERVE;
+ }
+
+ bool isUndefined() const { return st_shndx == ELF::SHN_UNDEF; }
+
+ bool isExternal() const {
+ return getBinding() != ELF::STB_LOCAL;
+ }
+
+ Expected<StringRef> getName(StringRef StrTab) const;
+};
+
+template <class ELFT>
+Expected<StringRef> Elf_Sym_Impl<ELFT>::getName(StringRef StrTab) const {
+ uint32_t Offset = this->st_name;
+ if (Offset >= StrTab.size())
+ return createStringError(object_error::parse_failed,
+ "st_name (0x%" PRIx32
+ ") is past the end of the string table"
+ " of size 0x%zx",
+ Offset, StrTab.size());
+ return StringRef(StrTab.data() + Offset);
+}
+
+/// Elf_Versym: This is the structure of entries in the SHT_GNU_versym section
+/// (.gnu.version). This structure is identical for ELF32 and ELF64.
+template <class ELFT>
+struct Elf_Versym_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Half vs_index; // Version index with flags (e.g. VERSYM_HIDDEN)
+};
+
+/// Elf_Verdef: This is the structure of entries in the SHT_GNU_verdef section
+/// (.gnu.version_d). This structure is identical for ELF32 and ELF64.
+template <class ELFT>
+struct Elf_Verdef_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ using Elf_Verdaux = Elf_Verdaux_Impl<ELFT>;
+ Elf_Half vd_version; // Version of this structure (e.g. VER_DEF_CURRENT)
+ Elf_Half vd_flags; // Bitwise flags (VER_DEF_*)
+ Elf_Half vd_ndx; // Version index, used in .gnu.version entries
+ Elf_Half vd_cnt; // Number of Verdaux entries
+ Elf_Word vd_hash; // Hash of name
+ Elf_Word vd_aux; // Offset to the first Verdaux entry (in bytes)
+ Elf_Word vd_next; // Offset to the next Verdef entry (in bytes)
+
+ /// Get the first Verdaux entry for this Verdef.
+ const Elf_Verdaux *getAux() const {
+ return reinterpret_cast<const Elf_Verdaux *>((const char *)this + vd_aux);
+ }
+};
+
+/// Elf_Verdaux: This is the structure of auxiliary data in the SHT_GNU_verdef
+/// section (.gnu.version_d). This structure is identical for ELF32 and ELF64.
+template <class ELFT>
+struct Elf_Verdaux_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word vda_name; // Version name (offset in string table)
+ Elf_Word vda_next; // Offset to next Verdaux entry (in bytes)
+};
+
+/// Elf_Verneed: This is the structure of entries in the SHT_GNU_verneed
+/// section (.gnu.version_r). This structure is identical for ELF32 and ELF64.
+template <class ELFT>
+struct Elf_Verneed_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Half vn_version; // Version of this structure (e.g. VER_NEED_CURRENT)
+ Elf_Half vn_cnt; // Number of associated Vernaux entries
+ Elf_Word vn_file; // Library name (string table offset)
+ Elf_Word vn_aux; // Offset to first Vernaux entry (in bytes)
+ Elf_Word vn_next; // Offset to next Verneed entry (in bytes)
+};
+
+/// Elf_Vernaux: This is the structure of auxiliary data in SHT_GNU_verneed
+/// section (.gnu.version_r). This structure is identical for ELF32 and ELF64.
+template <class ELFT>
+struct Elf_Vernaux_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word vna_hash; // Hash of dependency name
+ Elf_Half vna_flags; // Bitwise Flags (VER_FLAG_*)
+ Elf_Half vna_other; // Version index, used in .gnu.version entries
+ Elf_Word vna_name; // Dependency name
+ Elf_Word vna_next; // Offset to next Vernaux entry (in bytes)
+};
+
+/// Elf_Dyn_Base: This structure matches the form of entries in the dynamic
+/// table section (.dynamic) look like.
+template <class ELFT> struct Elf_Dyn_Base;
+
+template <endianness TargetEndianness>
+struct Elf_Dyn_Base<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Sword d_tag;
+ union {
+ Elf_Word d_val;
+ Elf_Addr d_ptr;
+ } d_un;
+};
+
+template <endianness TargetEndianness>
+struct Elf_Dyn_Base<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Sxword d_tag;
+ union {
+ Elf_Xword d_val;
+ Elf_Addr d_ptr;
+ } d_un;
+};
+
+/// Elf_Dyn_Impl: This inherits from Elf_Dyn_Base, adding getters.
+template <class ELFT>
+struct Elf_Dyn_Impl : Elf_Dyn_Base<ELFT> {
+ using Elf_Dyn_Base<ELFT>::d_tag;
+ using Elf_Dyn_Base<ELFT>::d_un;
+ using intX_t = typename std::conditional<ELFT::Is64Bits,
+ int64_t, int32_t>::type;
+ using uintX_t = typename std::conditional<ELFT::Is64Bits,
+ uint64_t, uint32_t>::type;
+ intX_t getTag() const { return d_tag; }
+ uintX_t getVal() const { return d_un.d_val; }
+ uintX_t getPtr() const { return d_un.d_ptr; }
+};
+
+template <endianness TargetEndianness>
+struct Elf_Rel_Impl<ELFType<TargetEndianness, false>, false> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ static const bool IsRela = false;
+ Elf_Addr r_offset; // Location (file byte offset, or program virtual addr)
+ Elf_Word r_info; // Symbol table index and type of relocation to apply
+
+ uint32_t getRInfo(bool isMips64EL) const {
+ assert(!isMips64EL);
+ return r_info;
+ }
+ void setRInfo(uint32_t R, bool IsMips64EL) {
+ assert(!IsMips64EL);
+ r_info = R;
+ }
+
+ // These accessors and mutators correspond to the ELF32_R_SYM, ELF32_R_TYPE,
+ // and ELF32_R_INFO macros defined in the ELF specification:
+ uint32_t getSymbol(bool isMips64EL) const {
+ return this->getRInfo(isMips64EL) >> 8;
+ }
+ unsigned char getType(bool isMips64EL) const {
+ return (unsigned char)(this->getRInfo(isMips64EL) & 0x0ff);
+ }
+ void setSymbol(uint32_t s, bool IsMips64EL) {
+ setSymbolAndType(s, getType(IsMips64EL), IsMips64EL);
+ }
+ void setType(unsigned char t, bool IsMips64EL) {
+ setSymbolAndType(getSymbol(IsMips64EL), t, IsMips64EL);
+ }
+ void setSymbolAndType(uint32_t s, unsigned char t, bool IsMips64EL) {
+ this->setRInfo((s << 8) + t, IsMips64EL);
+ }
+};
+
+template <endianness TargetEndianness>
+struct Elf_Rel_Impl<ELFType<TargetEndianness, false>, true>
+ : public Elf_Rel_Impl<ELFType<TargetEndianness, false>, false> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ static const bool IsRela = true;
+ Elf_Sword r_addend; // Compute value for relocatable field by adding this
+};
+
+template <endianness TargetEndianness>
+struct Elf_Rel_Impl<ELFType<TargetEndianness, true>, false> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ static const bool IsRela = false;
+ Elf_Addr r_offset; // Location (file byte offset, or program virtual addr)
+ Elf_Xword r_info; // Symbol table index and type of relocation to apply
+
+ uint64_t getRInfo(bool isMips64EL) const {
+ uint64_t t = r_info;
+ if (!isMips64EL)
+ return t;
+ // Mips64 little endian has a "special" encoding of r_info. Instead of one
+ // 64 bit little endian number, it is a little endian 32 bit number followed
+ // by a 32 bit big endian number.
+ return (t << 32) | ((t >> 8) & 0xff000000) | ((t >> 24) & 0x00ff0000) |
+ ((t >> 40) & 0x0000ff00) | ((t >> 56) & 0x000000ff);
+ }
+
+ void setRInfo(uint64_t R, bool IsMips64EL) {
+ if (IsMips64EL)
+ r_info = (R >> 32) | ((R & 0xff000000) << 8) | ((R & 0x00ff0000) << 24) |
+ ((R & 0x0000ff00) << 40) | ((R & 0x000000ff) << 56);
+ else
+ r_info = R;
+ }
+
+ // These accessors and mutators correspond to the ELF64_R_SYM, ELF64_R_TYPE,
+ // and ELF64_R_INFO macros defined in the ELF specification:
+ uint32_t getSymbol(bool isMips64EL) const {
+ return (uint32_t)(this->getRInfo(isMips64EL) >> 32);
+ }
+ uint32_t getType(bool isMips64EL) const {
+ return (uint32_t)(this->getRInfo(isMips64EL) & 0xffffffffL);
+ }
+ void setSymbol(uint32_t s, bool IsMips64EL) {
+ setSymbolAndType(s, getType(IsMips64EL), IsMips64EL);
+ }
+ void setType(uint32_t t, bool IsMips64EL) {
+ setSymbolAndType(getSymbol(IsMips64EL), t, IsMips64EL);
+ }
+ void setSymbolAndType(uint32_t s, uint32_t t, bool IsMips64EL) {
+ this->setRInfo(((uint64_t)s << 32) + (t & 0xffffffffL), IsMips64EL);
+ }
+};
+
+template <endianness TargetEndianness>
+struct Elf_Rel_Impl<ELFType<TargetEndianness, true>, true>
+ : public Elf_Rel_Impl<ELFType<TargetEndianness, true>, false> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ static const bool IsRela = true;
+ Elf_Sxword r_addend; // Compute value for relocatable field by adding this.
+};
+
+template <class ELFT>
+struct Elf_Ehdr_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ unsigned char e_ident[ELF::EI_NIDENT]; // ELF Identification bytes
+ Elf_Half e_type; // Type of file (see ET_*)
+ Elf_Half e_machine; // Required architecture for this file (see EM_*)
+ Elf_Word e_version; // Must be equal to 1
+ Elf_Addr e_entry; // Address to jump to in order to start program
+ Elf_Off e_phoff; // Program header table's file offset, in bytes
+ Elf_Off e_shoff; // Section header table's file offset, in bytes
+ Elf_Word e_flags; // Processor-specific flags
+ Elf_Half e_ehsize; // Size of ELF header, in bytes
+ Elf_Half e_phentsize; // Size of an entry in the program header table
+ Elf_Half e_phnum; // Number of entries in the program header table
+ Elf_Half e_shentsize; // Size of an entry in the section header table
+ Elf_Half e_shnum; // Number of entries in the section header table
+ Elf_Half e_shstrndx; // Section header table index of section name
+ // string table
+
+ bool checkMagic() const {
+ return (memcmp(e_ident, ELF::ElfMagic, strlen(ELF::ElfMagic))) == 0;
+ }
+
+ unsigned char getFileClass() const { return e_ident[ELF::EI_CLASS]; }
+ unsigned char getDataEncoding() const { return e_ident[ELF::EI_DATA]; }
+};
+
+template <endianness TargetEndianness>
+struct Elf_Phdr_Impl<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Word p_type; // Type of segment
+ Elf_Off p_offset; // FileOffset where segment is located, in bytes
+ Elf_Addr p_vaddr; // Virtual Address of beginning of segment
+ Elf_Addr p_paddr; // Physical address of beginning of segment (OS-specific)
+ Elf_Word p_filesz; // Num. of bytes in file image of segment (may be zero)
+ Elf_Word p_memsz; // Num. of bytes in mem image of segment (may be zero)
+ Elf_Word p_flags; // Segment flags
+ Elf_Word p_align; // Segment alignment constraint
+};
+
+template <endianness TargetEndianness>
+struct Elf_Phdr_Impl<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Word p_type; // Type of segment
+ Elf_Word p_flags; // Segment flags
+ Elf_Off p_offset; // FileOffset where segment is located, in bytes
+ Elf_Addr p_vaddr; // Virtual Address of beginning of segment
+ Elf_Addr p_paddr; // Physical address of beginning of segment (OS-specific)
+ Elf_Xword p_filesz; // Num. of bytes in file image of segment (may be zero)
+ Elf_Xword p_memsz; // Num. of bytes in mem image of segment (may be zero)
+ Elf_Xword p_align; // Segment alignment constraint
+};
+
+// ELFT needed for endianness.
+template <class ELFT>
+struct Elf_Hash_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word nbucket;
+ Elf_Word nchain;
+
+ ArrayRef<Elf_Word> buckets() const {
+ return ArrayRef<Elf_Word>(&nbucket + 2, &nbucket + 2 + nbucket);
+ }
+
+ ArrayRef<Elf_Word> chains() const {
+ return ArrayRef<Elf_Word>(&nbucket + 2 + nbucket,
+ &nbucket + 2 + nbucket + nchain);
+ }
+};
+
+// .gnu.hash section
+template <class ELFT>
+struct Elf_GnuHash_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word nbuckets;
+ Elf_Word symndx;
+ Elf_Word maskwords;
+ Elf_Word shift2;
+
+ ArrayRef<Elf_Off> filter() const {
+ return ArrayRef<Elf_Off>(reinterpret_cast<const Elf_Off *>(&shift2 + 1),
+ maskwords);
+ }
+
+ ArrayRef<Elf_Word> buckets() const {
+ return ArrayRef<Elf_Word>(
+ reinterpret_cast<const Elf_Word *>(filter().end()), nbuckets);
+ }
+
+ ArrayRef<Elf_Word> values(unsigned DynamicSymCount) const {
+ return ArrayRef<Elf_Word>(buckets().end(), DynamicSymCount - symndx);
+ }
+};
+
+// Compressed section headers.
+// http://www.sco.com/developers/gabi/latest/ch4.sheader.html#compression_header
+template <endianness TargetEndianness>
+struct Elf_Chdr_Impl<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Word ch_type;
+ Elf_Word ch_size;
+ Elf_Word ch_addralign;
+};
+
+template <endianness TargetEndianness>
+struct Elf_Chdr_Impl<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Word ch_type;
+ Elf_Word ch_reserved;
+ Elf_Xword ch_size;
+ Elf_Xword ch_addralign;
+};
+
+/// Note header
+template <class ELFT>
+struct Elf_Nhdr_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word n_namesz;
+ Elf_Word n_descsz;
+ Elf_Word n_type;
+
+ /// The alignment of the name and descriptor.
+ ///
+ /// Implementations differ from the specification here: in practice all
+ /// variants align both the name and descriptor to 4-bytes.
+ static const unsigned int Align = 4;
+
+ /// Get the size of the note, including name, descriptor, and padding.
+ size_t getSize() const {
+ return sizeof(*this) + alignTo<Align>(n_namesz) + alignTo<Align>(n_descsz);
+ }
+};
+
+/// An ELF note.
+///
+/// Wraps a note header, providing methods for accessing the name and
+/// descriptor safely.
+template <class ELFT>
+class Elf_Note_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+
+ const Elf_Nhdr_Impl<ELFT> &Nhdr;
+
+ template <class NoteIteratorELFT> friend class Elf_Note_Iterator_Impl;
+
+public:
+ Elf_Note_Impl(const Elf_Nhdr_Impl<ELFT> &Nhdr) : Nhdr(Nhdr) {}
+
+ /// Get the note's name, excluding the terminating null byte.
+ StringRef getName() const {
+ if (!Nhdr.n_namesz)
+ return StringRef();
+ return StringRef(reinterpret_cast<const char *>(&Nhdr) + sizeof(Nhdr),
+ Nhdr.n_namesz - 1);
+ }
+
+ /// Get the note's descriptor.
+ ArrayRef<uint8_t> getDesc() const {
+ if (!Nhdr.n_descsz)
+ return ArrayRef<uint8_t>();
+ return ArrayRef<uint8_t>(
+ reinterpret_cast<const uint8_t *>(&Nhdr) + sizeof(Nhdr) +
+ alignTo<Elf_Nhdr_Impl<ELFT>::Align>(Nhdr.n_namesz),
+ Nhdr.n_descsz);
+ }
+
+ /// Get the note's type.
+ Elf_Word getType() const { return Nhdr.n_type; }
+};
+
+template <class ELFT>
+class Elf_Note_Iterator_Impl
+ : std::iterator<std::forward_iterator_tag, Elf_Note_Impl<ELFT>> {
+ // Nhdr being a nullptr marks the end of iteration.
+ const Elf_Nhdr_Impl<ELFT> *Nhdr = nullptr;
+ size_t RemainingSize = 0u;
+ Error *Err = nullptr;
+
+ template <class ELFFileELFT> friend class ELFFile;
+
+ // Stop iteration and indicate an overflow.
+ void stopWithOverflowError() {
+ Nhdr = nullptr;
+ *Err = make_error<StringError>("ELF note overflows container",
+ object_error::parse_failed);
+ }
+
+ // Advance Nhdr by NoteSize bytes, starting from NhdrPos.
+ //
+ // Assumes NoteSize <= RemainingSize. Ensures Nhdr->getSize() <= RemainingSize
+ // upon returning. Handles stopping iteration when reaching the end of the
+ // container, either cleanly or with an overflow error.
+ void advanceNhdr(const uint8_t *NhdrPos, size_t NoteSize) {
+ RemainingSize -= NoteSize;
+ if (RemainingSize == 0u) {
+ // Ensure that if the iterator walks to the end, the error is checked
+ // afterwards.
+ *Err = Error::success();
+ Nhdr = nullptr;
+ } else if (sizeof(*Nhdr) > RemainingSize)
+ stopWithOverflowError();
+ else {
+ Nhdr = reinterpret_cast<const Elf_Nhdr_Impl<ELFT> *>(NhdrPos + NoteSize);
+ if (Nhdr->getSize() > RemainingSize)
+ stopWithOverflowError();
+ else
+ *Err = Error::success();
+ }
+ }
+
+ Elf_Note_Iterator_Impl() {}
+ explicit Elf_Note_Iterator_Impl(Error &Err) : Err(&Err) {}
+ Elf_Note_Iterator_Impl(const uint8_t *Start, size_t Size, Error &Err)
+ : RemainingSize(Size), Err(&Err) {
+ consumeError(std::move(Err));
+ assert(Start && "ELF note iterator starting at NULL");
+ advanceNhdr(Start, 0u);
+ }
+
+public:
+ Elf_Note_Iterator_Impl &operator++() {
+ assert(Nhdr && "incremented ELF note end iterator");
+ const uint8_t *NhdrPos = reinterpret_cast<const uint8_t *>(Nhdr);
+ size_t NoteSize = Nhdr->getSize();
+ advanceNhdr(NhdrPos, NoteSize);
+ return *this;
+ }
+ bool operator==(Elf_Note_Iterator_Impl Other) const {
+ if (!Nhdr && Other.Err)
+ (void)(bool)(*Other.Err);
+ if (!Other.Nhdr && Err)
+ (void)(bool)(*Err);
+ return Nhdr == Other.Nhdr;
+ }
+ bool operator!=(Elf_Note_Iterator_Impl Other) const {
+ return !(*this == Other);
+ }
+ Elf_Note_Impl<ELFT> operator*() const {
+ assert(Nhdr && "dereferenced ELF note end iterator");
+ return Elf_Note_Impl<ELFT>(*Nhdr);
+ }
+};
+
+template <class ELFT> struct Elf_CGProfile_Impl {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Word cgp_from;
+ Elf_Word cgp_to;
+ Elf_Xword cgp_weight;
+};
+
+// MIPS .reginfo section
+template <class ELFT>
+struct Elf_Mips_RegInfo;
+
+template <support::endianness TargetEndianness>
+struct Elf_Mips_RegInfo<ELFType<TargetEndianness, false>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, false)
+ Elf_Word ri_gprmask; // bit-mask of used general registers
+ Elf_Word ri_cprmask[4]; // bit-mask of used co-processor registers
+ Elf_Addr ri_gp_value; // gp register value
+};
+
+template <support::endianness TargetEndianness>
+struct Elf_Mips_RegInfo<ELFType<TargetEndianness, true>> {
+ LLVM_ELF_IMPORT_TYPES(TargetEndianness, true)
+ Elf_Word ri_gprmask; // bit-mask of used general registers
+ Elf_Word ri_pad; // unused padding field
+ Elf_Word ri_cprmask[4]; // bit-mask of used co-processor registers
+ Elf_Addr ri_gp_value; // gp register value
+};
+
+// .MIPS.options section
+template <class ELFT> struct Elf_Mips_Options {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ uint8_t kind; // Determines interpretation of variable part of descriptor
+ uint8_t size; // Byte size of descriptor, including this header
+ Elf_Half section; // Section header index of section affected,
+ // or 0 for global options
+ Elf_Word info; // Kind-specific information
+
+ Elf_Mips_RegInfo<ELFT> &getRegInfo() {
+ assert(kind == ELF::ODK_REGINFO);
+ return *reinterpret_cast<Elf_Mips_RegInfo<ELFT> *>(
+ (uint8_t *)this + sizeof(Elf_Mips_Options));
+ }
+ const Elf_Mips_RegInfo<ELFT> &getRegInfo() const {
+ return const_cast<Elf_Mips_Options *>(this)->getRegInfo();
+ }
+};
+
+// .MIPS.abiflags section content
+template <class ELFT> struct Elf_Mips_ABIFlags {
+ LLVM_ELF_IMPORT_TYPES_ELFT(ELFT)
+ Elf_Half version; // Version of the structure
+ uint8_t isa_level; // ISA level: 1-5, 32, and 64
+ uint8_t isa_rev; // ISA revision (0 for MIPS I - MIPS V)
+ uint8_t gpr_size; // General purpose registers size
+ uint8_t cpr1_size; // Co-processor 1 registers size
+ uint8_t cpr2_size; // Co-processor 2 registers size
+ uint8_t fp_abi; // Floating-point ABI flag
+ Elf_Word isa_ext; // Processor-specific extension
+ Elf_Word ases; // ASEs flags
+ Elf_Word flags1; // General flags
+ Elf_Word flags2; // General flags
+};
+
+} // end namespace object.
+} // end namespace llvm.
+
+#endif // LLVM_OBJECT_ELFTYPES_H
diff --git a/third_party/llvm-project/include/llvm/Object/Error.h b/third_party/llvm-project/include/llvm/Object/Error.h
new file mode 100644
index 000000000..b7bbf06fc
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Error.h
@@ -0,0 +1,92 @@
+//===- Error.h - system_error extensions for Object -------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This declares a new error_category for the Object library.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_ERROR_H
+#define LLVM_OBJECT_ERROR_H
+
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/Error.h"
+#include <system_error>
+
+namespace llvm {
+namespace object {
+
+class Binary;
+
+const std::error_category &object_category();
+
+enum class object_error {
+ // Error code 0 is absent. Use std::error_code() instead.
+ arch_not_found = 1,
+ invalid_file_type,
+ parse_failed,
+ unexpected_eof,
+ string_table_non_null_end,
+ invalid_section_index,
+ bitcode_section_not_found,
+ invalid_symbol_index,
+};
+
+inline std::error_code make_error_code(object_error e) {
+ return std::error_code(static_cast<int>(e), object_category());
+}
+
+/// Base class for all errors indicating malformed binary files.
+///
+/// Having a subclass for all malformed binary files allows archive-walking
+/// code to skip malformed files without having to understand every possible
+/// way that a binary file might be malformed.
+///
+/// Currently inherits from ECError for easy interoperability with
+/// std::error_code, but this will be removed in the future.
+class BinaryError : public ErrorInfo<BinaryError, ECError> {
+ virtual void anchor();
+public:
+ static char ID;
+ BinaryError() {
+ // Default to parse_failed, can be overridden with setErrorCode.
+ setErrorCode(make_error_code(object_error::parse_failed));
+ }
+};
+
+/// Generic binary error.
+///
+/// For errors that don't require their own specific sub-error (most errors)
+/// this class can be used to describe the error via a string message.
+class GenericBinaryError : public ErrorInfo<GenericBinaryError, BinaryError> {
+public:
+ static char ID;
+ GenericBinaryError(Twine Msg);
+ GenericBinaryError(Twine Msg, object_error ECOverride);
+ const std::string &getMessage() const { return Msg; }
+ void log(raw_ostream &OS) const override;
+private:
+ std::string Msg;
+};
+
+/// isNotObjectErrorInvalidFileType() is used when looping through the children
+/// of an archive after calling getAsBinary() on the child and it returns an
+/// llvm::Error. In the cases we want to loop through the children and ignore the
+/// non-objects in the archive this is used to test the error to see if an
+/// error() function needs to called on the llvm::Error.
+Error isNotObjectErrorInvalidFileType(llvm::Error Err);
+
+} // end namespace object.
+
+} // end namespace llvm.
+
+namespace std {
+template <>
+struct is_error_code_enum<llvm::object::object_error> : std::true_type {};
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Object/MachO.h b/third_party/llvm-project/include/llvm/Object/MachO.h
new file mode 100644
index 000000000..76be8049a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/MachO.h
@@ -0,0 +1,737 @@
+//===- MachO.h - MachO object file implementation ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the MachOObjectFile class, which implement the ObjectFile
+// interface for MachO files.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_MACHO_H
+#define LLVM_OBJECT_MACHO_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/MachO.h"
+#include "llvm/MC/SubtargetFeature.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Object/SymbolicFile.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstdint>
+#include <memory>
+#include <string>
+#include <system_error>
+
+namespace llvm {
+namespace object {
+
+/// DiceRef - This is a value type class that represents a single
+/// data in code entry in the table in a Mach-O object file.
+class DiceRef {
+ DataRefImpl DicePimpl;
+ const ObjectFile *OwningObject = nullptr;
+
+public:
+ DiceRef() = default;
+ DiceRef(DataRefImpl DiceP, const ObjectFile *Owner);
+
+ bool operator==(const DiceRef &Other) const;
+ bool operator<(const DiceRef &Other) const;
+
+ void moveNext();
+
+ std::error_code getOffset(uint32_t &Result) const;
+ std::error_code getLength(uint16_t &Result) const;
+ std::error_code getKind(uint16_t &Result) const;
+
+ DataRefImpl getRawDataRefImpl() const;
+ const ObjectFile *getObjectFile() const;
+};
+using dice_iterator = content_iterator<DiceRef>;
+
+/// ExportEntry encapsulates the current-state-of-the-walk used when doing a
+/// non-recursive walk of the trie data structure. This allows you to iterate
+/// across all exported symbols using:
+/// Error Err;
+/// for (const llvm::object::ExportEntry &AnExport : Obj->exports(&Err)) {
+/// }
+/// if (Err) { report error ...
+class ExportEntry {
+public:
+ ExportEntry(Error *Err, const MachOObjectFile *O, ArrayRef<uint8_t> Trie);
+
+ StringRef name() const;
+ uint64_t flags() const;
+ uint64_t address() const;
+ uint64_t other() const;
+ StringRef otherName() const;
+ uint32_t nodeOffset() const;
+
+ bool operator==(const ExportEntry &) const;
+
+ void moveNext();
+
+private:
+ friend class MachOObjectFile;
+
+ void moveToFirst();
+ void moveToEnd();
+ uint64_t readULEB128(const uint8_t *&p, const char **error);
+ void pushDownUntilBottom();
+ void pushNode(uint64_t Offset);
+
+ // Represents a node in the mach-o exports trie.
+ struct NodeState {
+ NodeState(const uint8_t *Ptr);
+
+ const uint8_t *Start;
+ const uint8_t *Current;
+ uint64_t Flags = 0;
+ uint64_t Address = 0;
+ uint64_t Other = 0;
+ const char *ImportName = nullptr;
+ unsigned ChildCount = 0;
+ unsigned NextChildIndex = 0;
+ unsigned ParentStringLength = 0;
+ bool IsExportNode = false;
+ };
+ using NodeList = SmallVector<NodeState, 16>;
+ using node_iterator = NodeList::const_iterator;
+
+ Error *E;
+ const MachOObjectFile *O;
+ ArrayRef<uint8_t> Trie;
+ SmallString<256> CumulativeString;
+ NodeList Stack;
+ bool Done = false;
+
+ iterator_range<node_iterator> nodes() const {
+ return make_range(Stack.begin(), Stack.end());
+ }
+};
+using export_iterator = content_iterator<ExportEntry>;
+
+// Segment info so SegIndex/SegOffset pairs in a Mach-O Bind or Rebase entry
+// can be checked and translated. Only the SegIndex/SegOffset pairs from
+// checked entries are to be used with the segmentName(), sectionName() and
+// address() methods below.
+class BindRebaseSegInfo {
+public:
+ BindRebaseSegInfo(const MachOObjectFile *Obj);
+
+ // Used to check a Mach-O Bind or Rebase entry for errors when iterating.
+ const char* checkSegAndOffsets(int32_t SegIndex, uint64_t SegOffset,
+ uint8_t PointerSize, uint32_t Count=1,
+ uint32_t Skip=0);
+ // Used with valid SegIndex/SegOffset values from checked entries.
+ StringRef segmentName(int32_t SegIndex);
+ StringRef sectionName(int32_t SegIndex, uint64_t SegOffset);
+ uint64_t address(uint32_t SegIndex, uint64_t SegOffset);
+
+private:
+ struct SectionInfo {
+ uint64_t Address;
+ uint64_t Size;
+ StringRef SectionName;
+ StringRef SegmentName;
+ uint64_t OffsetInSegment;
+ uint64_t SegmentStartAddress;
+ int32_t SegmentIndex;
+ };
+ const SectionInfo &findSection(int32_t SegIndex, uint64_t SegOffset);
+
+ SmallVector<SectionInfo, 32> Sections;
+ int32_t MaxSegIndex;
+};
+
+/// MachORebaseEntry encapsulates the current state in the decompression of
+/// rebasing opcodes. This allows you to iterate through the compressed table of
+/// rebasing using:
+/// Error Err;
+/// for (const llvm::object::MachORebaseEntry &Entry : Obj->rebaseTable(&Err)) {
+/// }
+/// if (Err) { report error ...
+class MachORebaseEntry {
+public:
+ MachORebaseEntry(Error *Err, const MachOObjectFile *O,
+ ArrayRef<uint8_t> opcodes, bool is64Bit);
+
+ int32_t segmentIndex() const;
+ uint64_t segmentOffset() const;
+ StringRef typeName() const;
+ StringRef segmentName() const;
+ StringRef sectionName() const;
+ uint64_t address() const;
+
+ bool operator==(const MachORebaseEntry &) const;
+
+ void moveNext();
+
+private:
+ friend class MachOObjectFile;
+
+ void moveToFirst();
+ void moveToEnd();
+ uint64_t readULEB128(const char **error);
+
+ Error *E;
+ const MachOObjectFile *O;
+ ArrayRef<uint8_t> Opcodes;
+ const uint8_t *Ptr;
+ uint64_t SegmentOffset = 0;
+ int32_t SegmentIndex = -1;
+ uint64_t RemainingLoopCount = 0;
+ uint64_t AdvanceAmount = 0;
+ uint8_t RebaseType = 0;
+ uint8_t PointerSize;
+ bool Done = false;
+};
+using rebase_iterator = content_iterator<MachORebaseEntry>;
+
+/// MachOBindEntry encapsulates the current state in the decompression of
+/// binding opcodes. This allows you to iterate through the compressed table of
+/// bindings using:
+/// Error Err;
+/// for (const llvm::object::MachOBindEntry &Entry : Obj->bindTable(&Err)) {
+/// }
+/// if (Err) { report error ...
+class MachOBindEntry {
+public:
+ enum class Kind { Regular, Lazy, Weak };
+
+ MachOBindEntry(Error *Err, const MachOObjectFile *O,
+ ArrayRef<uint8_t> Opcodes, bool is64Bit, MachOBindEntry::Kind);
+
+ int32_t segmentIndex() const;
+ uint64_t segmentOffset() const;
+ StringRef typeName() const;
+ StringRef symbolName() const;
+ uint32_t flags() const;
+ int64_t addend() const;
+ int ordinal() const;
+
+ StringRef segmentName() const;
+ StringRef sectionName() const;
+ uint64_t address() const;
+
+ bool operator==(const MachOBindEntry &) const;
+
+ void moveNext();
+
+private:
+ friend class MachOObjectFile;
+
+ void moveToFirst();
+ void moveToEnd();
+ uint64_t readULEB128(const char **error);
+ int64_t readSLEB128(const char **error);
+
+ Error *E;
+ const MachOObjectFile *O;
+ ArrayRef<uint8_t> Opcodes;
+ const uint8_t *Ptr;
+ uint64_t SegmentOffset = 0;
+ int32_t SegmentIndex = -1;
+ StringRef SymbolName;
+ bool LibraryOrdinalSet = false;
+ int Ordinal = 0;
+ uint32_t Flags = 0;
+ int64_t Addend = 0;
+ uint64_t RemainingLoopCount = 0;
+ uint64_t AdvanceAmount = 0;
+ uint8_t BindType = 0;
+ uint8_t PointerSize;
+ Kind TableKind;
+ bool Done = false;
+};
+using bind_iterator = content_iterator<MachOBindEntry>;
+
+class MachOObjectFile : public ObjectFile {
+public:
+ struct LoadCommandInfo {
+ const char *Ptr; // Where in memory the load command is.
+ MachO::load_command C; // The command itself.
+ };
+ using LoadCommandList = SmallVector<LoadCommandInfo, 4>;
+ using load_command_iterator = LoadCommandList::const_iterator;
+
+ static Expected<std::unique_ptr<MachOObjectFile>>
+ create(MemoryBufferRef Object, bool IsLittleEndian, bool Is64Bits,
+ uint32_t UniversalCputype = 0, uint32_t UniversalIndex = 0);
+
+ void moveSymbolNext(DataRefImpl &Symb) const override;
+
+ uint64_t getNValue(DataRefImpl Sym) const;
+ Expected<StringRef> getSymbolName(DataRefImpl Symb) const override;
+
+ // MachO specific.
+ Error checkSymbolTable() const;
+
+ std::error_code getIndirectName(DataRefImpl Symb, StringRef &Res) const;
+ unsigned getSectionType(SectionRef Sec) const;
+
+ Expected<uint64_t> getSymbolAddress(DataRefImpl Symb) const override;
+ uint32_t getSymbolAlignment(DataRefImpl Symb) const override;
+ uint64_t getCommonSymbolSizeImpl(DataRefImpl Symb) const override;
+ Expected<SymbolRef::Type> getSymbolType(DataRefImpl Symb) const override;
+ uint32_t getSymbolFlags(DataRefImpl Symb) const override;
+ Expected<section_iterator> getSymbolSection(DataRefImpl Symb) const override;
+ unsigned getSymbolSectionID(SymbolRef Symb) const;
+ unsigned getSectionID(SectionRef Sec) const;
+
+ void moveSectionNext(DataRefImpl &Sec) const override;
+ Expected<StringRef> getSectionName(DataRefImpl Sec) const override;
+ uint64_t getSectionAddress(DataRefImpl Sec) const override;
+ uint64_t getSectionIndex(DataRefImpl Sec) const override;
+ uint64_t getSectionSize(DataRefImpl Sec) const override;
+ ArrayRef<uint8_t> getSectionContents(uint32_t Offset, uint64_t Size) const;
+ Expected<ArrayRef<uint8_t>>
+ getSectionContents(DataRefImpl Sec) const override;
+ uint64_t getSectionAlignment(DataRefImpl Sec) const override;
+ Expected<SectionRef> getSection(unsigned SectionIndex) const;
+ Expected<SectionRef> getSection(StringRef SectionName) const;
+ bool isSectionCompressed(DataRefImpl Sec) const override;
+ bool isSectionText(DataRefImpl Sec) const override;
+ bool isSectionData(DataRefImpl Sec) const override;
+ bool isSectionBSS(DataRefImpl Sec) const override;
+ bool isSectionVirtual(DataRefImpl Sec) const override;
+ bool isSectionBitcode(DataRefImpl Sec) const override;
+
+ /// When dsymutil generates the companion file, it strips all unnecessary
+ /// sections (e.g. everything in the _TEXT segment) by omitting their body
+ /// and setting the offset in their corresponding load command to zero.
+ ///
+ /// While the load command itself is valid, reading the section corresponds
+ /// to reading the number of bytes specified in the load command, starting
+ /// from offset 0 (i.e. the Mach-O header at the beginning of the file).
+ bool isSectionStripped(DataRefImpl Sec) const override;
+
+ relocation_iterator section_rel_begin(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_end(DataRefImpl Sec) const override;
+
+ relocation_iterator extrel_begin() const;
+ relocation_iterator extrel_end() const;
+ iterator_range<relocation_iterator> external_relocations() const {
+ return make_range(extrel_begin(), extrel_end());
+ }
+
+ relocation_iterator locrel_begin() const;
+ relocation_iterator locrel_end() const;
+
+ void moveRelocationNext(DataRefImpl &Rel) const override;
+ uint64_t getRelocationOffset(DataRefImpl Rel) const override;
+ symbol_iterator getRelocationSymbol(DataRefImpl Rel) const override;
+ section_iterator getRelocationSection(DataRefImpl Rel) const;
+ uint64_t getRelocationType(DataRefImpl Rel) const override;
+ void getRelocationTypeName(DataRefImpl Rel,
+ SmallVectorImpl<char> &Result) const override;
+ uint8_t getRelocationLength(DataRefImpl Rel) const;
+
+ // MachO specific.
+ std::error_code getLibraryShortNameByIndex(unsigned Index, StringRef &) const;
+ uint32_t getLibraryCount() const;
+
+ section_iterator getRelocationRelocatedSection(relocation_iterator Rel) const;
+
+ // TODO: Would be useful to have an iterator based version
+ // of the load command interface too.
+
+ basic_symbol_iterator symbol_begin() const override;
+ basic_symbol_iterator symbol_end() const override;
+
+ // MachO specific.
+ symbol_iterator getSymbolByIndex(unsigned Index) const;
+ uint64_t getSymbolIndex(DataRefImpl Symb) const;
+
+ section_iterator section_begin() const override;
+ section_iterator section_end() const override;
+
+ uint8_t getBytesInAddress() const override;
+
+ StringRef getFileFormatName() const override;
+ Triple::ArchType getArch() const override;
+ SubtargetFeatures getFeatures() const override { return SubtargetFeatures(); }
+ Triple getArchTriple(const char **McpuDefault = nullptr) const;
+
+ relocation_iterator section_rel_begin(unsigned Index) const;
+ relocation_iterator section_rel_end(unsigned Index) const;
+
+ dice_iterator begin_dices() const;
+ dice_iterator end_dices() const;
+
+ load_command_iterator begin_load_commands() const;
+ load_command_iterator end_load_commands() const;
+ iterator_range<load_command_iterator> load_commands() const;
+
+ /// For use iterating over all exported symbols.
+ iterator_range<export_iterator> exports(Error &Err) const;
+
+ /// For use examining a trie not in a MachOObjectFile.
+ static iterator_range<export_iterator> exports(Error &Err,
+ ArrayRef<uint8_t> Trie,
+ const MachOObjectFile *O =
+ nullptr);
+
+ /// For use iterating over all rebase table entries.
+ iterator_range<rebase_iterator> rebaseTable(Error &Err);
+
+ /// For use examining rebase opcodes in a MachOObjectFile.
+ static iterator_range<rebase_iterator> rebaseTable(Error &Err,
+ MachOObjectFile *O,
+ ArrayRef<uint8_t> Opcodes,
+ bool is64);
+
+ /// For use iterating over all bind table entries.
+ iterator_range<bind_iterator> bindTable(Error &Err);
+
+ /// For use iterating over all lazy bind table entries.
+ iterator_range<bind_iterator> lazyBindTable(Error &Err);
+
+ /// For use iterating over all weak bind table entries.
+ iterator_range<bind_iterator> weakBindTable(Error &Err);
+
+ /// For use examining bind opcodes in a MachOObjectFile.
+ static iterator_range<bind_iterator> bindTable(Error &Err,
+ MachOObjectFile *O,
+ ArrayRef<uint8_t> Opcodes,
+ bool is64,
+ MachOBindEntry::Kind);
+
+ // Given a SegIndex, SegOffset, and PointerSize, verify a valid section exists
+ // that fully contains a pointer at that location. Multiple fixups in a bind
+ // (such as with the BIND_OPCODE_DO_BIND_ULEB_TIMES_SKIPPING_ULEB opcode) can
+ // be tested via the Count and Skip parameters.
+ //
+ // This is used by MachOBindEntry::moveNext() to validate a MachOBindEntry.
+ const char *BindEntryCheckSegAndOffsets(int32_t SegIndex, uint64_t SegOffset,
+ uint8_t PointerSize, uint32_t Count=1,
+ uint32_t Skip=0) const {
+ return BindRebaseSectionTable->checkSegAndOffsets(SegIndex, SegOffset,
+ PointerSize, Count, Skip);
+ }
+
+ // Given a SegIndex, SegOffset, and PointerSize, verify a valid section exists
+ // that fully contains a pointer at that location. Multiple fixups in a rebase
+ // (such as with the REBASE_OPCODE_DO_*_TIMES* opcodes) can be tested via the
+ // Count and Skip parameters.
+ //
+ // This is used by MachORebaseEntry::moveNext() to validate a MachORebaseEntry
+ const char *RebaseEntryCheckSegAndOffsets(int32_t SegIndex,
+ uint64_t SegOffset,
+ uint8_t PointerSize,
+ uint32_t Count=1,
+ uint32_t Skip=0) const {
+ return BindRebaseSectionTable->checkSegAndOffsets(SegIndex, SegOffset,
+ PointerSize, Count, Skip);
+ }
+
+ /// For use with the SegIndex of a checked Mach-O Bind or Rebase entry to
+ /// get the segment name.
+ StringRef BindRebaseSegmentName(int32_t SegIndex) const {
+ return BindRebaseSectionTable->segmentName(SegIndex);
+ }
+
+ /// For use with a SegIndex,SegOffset pair from a checked Mach-O Bind or
+ /// Rebase entry to get the section name.
+ StringRef BindRebaseSectionName(uint32_t SegIndex, uint64_t SegOffset) const {
+ return BindRebaseSectionTable->sectionName(SegIndex, SegOffset);
+ }
+
+ /// For use with a SegIndex,SegOffset pair from a checked Mach-O Bind or
+ /// Rebase entry to get the address.
+ uint64_t BindRebaseAddress(uint32_t SegIndex, uint64_t SegOffset) const {
+ return BindRebaseSectionTable->address(SegIndex, SegOffset);
+ }
+
+ // In a MachO file, sections have a segment name. This is used in the .o
+ // files. They have a single segment, but this field specifies which segment
+ // a section should be put in the final object.
+ StringRef getSectionFinalSegmentName(DataRefImpl Sec) const;
+
+ // Names are stored as 16 bytes. These returns the raw 16 bytes without
+ // interpreting them as a C string.
+ ArrayRef<char> getSectionRawName(DataRefImpl Sec) const;
+ ArrayRef<char> getSectionRawFinalSegmentName(DataRefImpl Sec) const;
+
+ // MachO specific Info about relocations.
+ bool isRelocationScattered(const MachO::any_relocation_info &RE) const;
+ unsigned getPlainRelocationSymbolNum(
+ const MachO::any_relocation_info &RE) const;
+ bool getPlainRelocationExternal(const MachO::any_relocation_info &RE) const;
+ bool getScatteredRelocationScattered(
+ const MachO::any_relocation_info &RE) const;
+ uint32_t getScatteredRelocationValue(
+ const MachO::any_relocation_info &RE) const;
+ uint32_t getScatteredRelocationType(
+ const MachO::any_relocation_info &RE) const;
+ unsigned getAnyRelocationAddress(const MachO::any_relocation_info &RE) const;
+ unsigned getAnyRelocationPCRel(const MachO::any_relocation_info &RE) const;
+ unsigned getAnyRelocationLength(const MachO::any_relocation_info &RE) const;
+ unsigned getAnyRelocationType(const MachO::any_relocation_info &RE) const;
+ SectionRef getAnyRelocationSection(const MachO::any_relocation_info &RE) const;
+
+ // MachO specific structures.
+ MachO::section getSection(DataRefImpl DRI) const;
+ MachO::section_64 getSection64(DataRefImpl DRI) const;
+ MachO::section getSection(const LoadCommandInfo &L, unsigned Index) const;
+ MachO::section_64 getSection64(const LoadCommandInfo &L,unsigned Index) const;
+ MachO::nlist getSymbolTableEntry(DataRefImpl DRI) const;
+ MachO::nlist_64 getSymbol64TableEntry(DataRefImpl DRI) const;
+
+ MachO::linkedit_data_command
+ getLinkeditDataLoadCommand(const LoadCommandInfo &L) const;
+ MachO::segment_command
+ getSegmentLoadCommand(const LoadCommandInfo &L) const;
+ MachO::segment_command_64
+ getSegment64LoadCommand(const LoadCommandInfo &L) const;
+ MachO::linker_option_command
+ getLinkerOptionLoadCommand(const LoadCommandInfo &L) const;
+ MachO::version_min_command
+ getVersionMinLoadCommand(const LoadCommandInfo &L) const;
+ MachO::note_command
+ getNoteLoadCommand(const LoadCommandInfo &L) const;
+ MachO::build_version_command
+ getBuildVersionLoadCommand(const LoadCommandInfo &L) const;
+ MachO::build_tool_version
+ getBuildToolVersion(unsigned index) const;
+ MachO::dylib_command
+ getDylibIDLoadCommand(const LoadCommandInfo &L) const;
+ MachO::dyld_info_command
+ getDyldInfoLoadCommand(const LoadCommandInfo &L) const;
+ MachO::dylinker_command
+ getDylinkerCommand(const LoadCommandInfo &L) const;
+ MachO::uuid_command
+ getUuidCommand(const LoadCommandInfo &L) const;
+ MachO::rpath_command
+ getRpathCommand(const LoadCommandInfo &L) const;
+ MachO::source_version_command
+ getSourceVersionCommand(const LoadCommandInfo &L) const;
+ MachO::entry_point_command
+ getEntryPointCommand(const LoadCommandInfo &L) const;
+ MachO::encryption_info_command
+ getEncryptionInfoCommand(const LoadCommandInfo &L) const;
+ MachO::encryption_info_command_64
+ getEncryptionInfoCommand64(const LoadCommandInfo &L) const;
+ MachO::sub_framework_command
+ getSubFrameworkCommand(const LoadCommandInfo &L) const;
+ MachO::sub_umbrella_command
+ getSubUmbrellaCommand(const LoadCommandInfo &L) const;
+ MachO::sub_library_command
+ getSubLibraryCommand(const LoadCommandInfo &L) const;
+ MachO::sub_client_command
+ getSubClientCommand(const LoadCommandInfo &L) const;
+ MachO::routines_command
+ getRoutinesCommand(const LoadCommandInfo &L) const;
+ MachO::routines_command_64
+ getRoutinesCommand64(const LoadCommandInfo &L) const;
+ MachO::thread_command
+ getThreadCommand(const LoadCommandInfo &L) const;
+
+ MachO::any_relocation_info getRelocation(DataRefImpl Rel) const;
+ MachO::data_in_code_entry getDice(DataRefImpl Rel) const;
+ const MachO::mach_header &getHeader() const;
+ const MachO::mach_header_64 &getHeader64() const;
+ uint32_t
+ getIndirectSymbolTableEntry(const MachO::dysymtab_command &DLC,
+ unsigned Index) const;
+ MachO::data_in_code_entry getDataInCodeTableEntry(uint32_t DataOffset,
+ unsigned Index) const;
+ MachO::symtab_command getSymtabLoadCommand() const;
+ MachO::dysymtab_command getDysymtabLoadCommand() const;
+ MachO::linkedit_data_command getDataInCodeLoadCommand() const;
+ MachO::linkedit_data_command getLinkOptHintsLoadCommand() const;
+ ArrayRef<uint8_t> getDyldInfoRebaseOpcodes() const;
+ ArrayRef<uint8_t> getDyldInfoBindOpcodes() const;
+ ArrayRef<uint8_t> getDyldInfoWeakBindOpcodes() const;
+ ArrayRef<uint8_t> getDyldInfoLazyBindOpcodes() const;
+ ArrayRef<uint8_t> getDyldInfoExportsTrie() const;
+ ArrayRef<uint8_t> getUuid() const;
+
+ StringRef getStringTableData() const;
+ bool is64Bit() const;
+ void ReadULEB128s(uint64_t Index, SmallVectorImpl<uint64_t> &Out) const;
+
+ static StringRef guessLibraryShortName(StringRef Name, bool &isFramework,
+ StringRef &Suffix);
+
+ static Triple::ArchType getArch(uint32_t CPUType);
+ static Triple getArchTriple(uint32_t CPUType, uint32_t CPUSubType,
+ const char **McpuDefault = nullptr,
+ const char **ArchFlag = nullptr);
+ static bool isValidArch(StringRef ArchFlag);
+ static ArrayRef<StringRef> getValidArchs();
+ static Triple getHostArch();
+
+ bool isRelocatableObject() const override;
+
+ StringRef mapDebugSectionName(StringRef Name) const override;
+
+ bool hasPageZeroSegment() const { return HasPageZeroSegment; }
+
+ static bool classof(const Binary *v) {
+ return v->isMachO();
+ }
+
+ static uint32_t
+ getVersionMinMajor(MachO::version_min_command &C, bool SDK) {
+ uint32_t VersionOrSDK = (SDK) ? C.sdk : C.version;
+ return (VersionOrSDK >> 16) & 0xffff;
+ }
+
+ static uint32_t
+ getVersionMinMinor(MachO::version_min_command &C, bool SDK) {
+ uint32_t VersionOrSDK = (SDK) ? C.sdk : C.version;
+ return (VersionOrSDK >> 8) & 0xff;
+ }
+
+ static uint32_t
+ getVersionMinUpdate(MachO::version_min_command &C, bool SDK) {
+ uint32_t VersionOrSDK = (SDK) ? C.sdk : C.version;
+ return VersionOrSDK & 0xff;
+ }
+
+ static std::string getBuildPlatform(uint32_t platform) {
+ switch (platform) {
+ case MachO::PLATFORM_MACOS: return "macos";
+ case MachO::PLATFORM_IOS: return "ios";
+ case MachO::PLATFORM_TVOS: return "tvos";
+ case MachO::PLATFORM_WATCHOS: return "watchos";
+ case MachO::PLATFORM_BRIDGEOS: return "bridgeos";
+ case MachO::PLATFORM_MACCATALYST: return "macCatalyst";
+ case MachO::PLATFORM_IOSSIMULATOR: return "iossimulator";
+ case MachO::PLATFORM_TVOSSIMULATOR: return "tvossimulator";
+ case MachO::PLATFORM_WATCHOSSIMULATOR: return "watchossimulator";
+ default:
+ std::string ret;
+ raw_string_ostream ss(ret);
+ ss << format_hex(platform, 8, true);
+ return ss.str();
+ }
+ }
+
+ static std::string getBuildTool(uint32_t tools) {
+ switch (tools) {
+ case MachO::TOOL_CLANG: return "clang";
+ case MachO::TOOL_SWIFT: return "swift";
+ case MachO::TOOL_LD: return "ld";
+ default:
+ std::string ret;
+ raw_string_ostream ss(ret);
+ ss << format_hex(tools, 8, true);
+ return ss.str();
+ }
+ }
+
+ static std::string getVersionString(uint32_t version) {
+ uint32_t major = (version >> 16) & 0xffff;
+ uint32_t minor = (version >> 8) & 0xff;
+ uint32_t update = version & 0xff;
+
+ SmallString<32> Version;
+ Version = utostr(major) + "." + utostr(minor);
+ if (update != 0)
+ Version += "." + utostr(update);
+ return Version.str();
+ }
+
+private:
+ MachOObjectFile(MemoryBufferRef Object, bool IsLittleEndian, bool Is64Bits,
+ Error &Err, uint32_t UniversalCputype = 0,
+ uint32_t UniversalIndex = 0);
+
+ uint64_t getSymbolValueImpl(DataRefImpl Symb) const override;
+
+ union {
+ MachO::mach_header_64 Header64;
+ MachO::mach_header Header;
+ };
+ using SectionList = SmallVector<const char*, 1>;
+ SectionList Sections;
+ using LibraryList = SmallVector<const char*, 1>;
+ LibraryList Libraries;
+ LoadCommandList LoadCommands;
+ using LibraryShortName = SmallVector<StringRef, 1>;
+ using BuildToolList = SmallVector<const char*, 1>;
+ BuildToolList BuildTools;
+ mutable LibraryShortName LibrariesShortNames;
+ std::unique_ptr<BindRebaseSegInfo> BindRebaseSectionTable;
+ const char *SymtabLoadCmd = nullptr;
+ const char *DysymtabLoadCmd = nullptr;
+ const char *DataInCodeLoadCmd = nullptr;
+ const char *LinkOptHintsLoadCmd = nullptr;
+ const char *DyldInfoLoadCmd = nullptr;
+ const char *UuidLoadCmd = nullptr;
+ bool HasPageZeroSegment = false;
+};
+
+/// DiceRef
+inline DiceRef::DiceRef(DataRefImpl DiceP, const ObjectFile *Owner)
+ : DicePimpl(DiceP) , OwningObject(Owner) {}
+
+inline bool DiceRef::operator==(const DiceRef &Other) const {
+ return DicePimpl == Other.DicePimpl;
+}
+
+inline bool DiceRef::operator<(const DiceRef &Other) const {
+ return DicePimpl < Other.DicePimpl;
+}
+
+inline void DiceRef::moveNext() {
+ const MachO::data_in_code_entry *P =
+ reinterpret_cast<const MachO::data_in_code_entry *>(DicePimpl.p);
+ DicePimpl.p = reinterpret_cast<uintptr_t>(P + 1);
+}
+
+// Since a Mach-O data in code reference, a DiceRef, can only be created when
+// the OwningObject ObjectFile is a MachOObjectFile a static_cast<> is used for
+// the methods that get the values of the fields of the reference.
+
+inline std::error_code DiceRef::getOffset(uint32_t &Result) const {
+ const MachOObjectFile *MachOOF =
+ static_cast<const MachOObjectFile *>(OwningObject);
+ MachO::data_in_code_entry Dice = MachOOF->getDice(DicePimpl);
+ Result = Dice.offset;
+ return std::error_code();
+}
+
+inline std::error_code DiceRef::getLength(uint16_t &Result) const {
+ const MachOObjectFile *MachOOF =
+ static_cast<const MachOObjectFile *>(OwningObject);
+ MachO::data_in_code_entry Dice = MachOOF->getDice(DicePimpl);
+ Result = Dice.length;
+ return std::error_code();
+}
+
+inline std::error_code DiceRef::getKind(uint16_t &Result) const {
+ const MachOObjectFile *MachOOF =
+ static_cast<const MachOObjectFile *>(OwningObject);
+ MachO::data_in_code_entry Dice = MachOOF->getDice(DicePimpl);
+ Result = Dice.kind;
+ return std::error_code();
+}
+
+inline DataRefImpl DiceRef::getRawDataRefImpl() const {
+ return DicePimpl;
+}
+
+inline const ObjectFile *DiceRef::getObjectFile() const {
+ return OwningObject;
+}
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_MACHO_H
diff --git a/third_party/llvm-project/include/llvm/Object/Minidump.h b/third_party/llvm-project/include/llvm/Object/Minidump.h
new file mode 100644
index 000000000..4429493af
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Minidump.h
@@ -0,0 +1,216 @@
+//===- Minidump.h - Minidump object file implementation ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_MINIDUMP_H
+#define LLVM_OBJECT_MINIDUMP_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/BinaryFormat/Minidump.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Support/Error.h"
+
+namespace llvm {
+namespace object {
+
+/// A class providing access to the contents of a minidump file.
+class MinidumpFile : public Binary {
+public:
+ /// Construct a new MinidumpFile object from the given memory buffer. Returns
+ /// an error if this file cannot be identified as a minidump file, or if its
+ /// contents are badly corrupted (i.e. we cannot read the stream directory).
+ static Expected<std::unique_ptr<MinidumpFile>> create(MemoryBufferRef Source);
+
+ static bool classof(const Binary *B) { return B->isMinidump(); }
+
+ /// Returns the contents of the minidump header.
+ const minidump::Header &header() const { return Header; }
+
+ /// Returns the list of streams (stream directory entries) in this file.
+ ArrayRef<minidump::Directory> streams() const { return Streams; }
+
+ /// Returns the raw contents of the stream given by the directory entry.
+ ArrayRef<uint8_t> getRawStream(const minidump::Directory &Stream) const {
+ return getData().slice(Stream.Location.RVA, Stream.Location.DataSize);
+ }
+
+ /// Returns the raw contents of the stream of the given type, or None if the
+ /// file does not contain a stream of this type.
+ Optional<ArrayRef<uint8_t>> getRawStream(minidump::StreamType Type) const;
+
+ /// Returns the raw contents of an object given by the LocationDescriptor. An
+ /// error is returned if the descriptor points outside of the minidump file.
+ Expected<ArrayRef<uint8_t>>
+ getRawData(minidump::LocationDescriptor Desc) const {
+ return getDataSlice(getData(), Desc.RVA, Desc.DataSize);
+ }
+
+ /// Returns the minidump string at the given offset. An error is returned if
+ /// we fail to parse the string, or the string is invalid UTF16.
+ Expected<std::string> getString(size_t Offset) const;
+
+ /// Returns the contents of the SystemInfo stream, cast to the appropriate
+ /// type. An error is returned if the file does not contain this stream, or
+ /// the stream is smaller than the size of the SystemInfo structure. The
+ /// internal consistency of the stream is not checked in any way.
+ Expected<const minidump::SystemInfo &> getSystemInfo() const {
+ return getStream<minidump::SystemInfo>(minidump::StreamType::SystemInfo);
+ }
+
+ /// Returns the module list embedded in the ModuleList stream. An error is
+ /// returned if the file does not contain this stream, or if the stream is
+ /// not large enough to contain the number of modules declared in the stream
+ /// header. The consistency of the Module entries themselves is not checked in
+ /// any way.
+ Expected<ArrayRef<minidump::Module>> getModuleList() const {
+ return getListStream<minidump::Module>(minidump::StreamType::ModuleList);
+ }
+
+ /// Returns the thread list embedded in the ThreadList stream. An error is
+ /// returned if the file does not contain this stream, or if the stream is
+ /// not large enough to contain the number of threads declared in the stream
+ /// header. The consistency of the Thread entries themselves is not checked in
+ /// any way.
+ Expected<ArrayRef<minidump::Thread>> getThreadList() const {
+ return getListStream<minidump::Thread>(minidump::StreamType::ThreadList);
+ }
+
+ /// Returns the contents of the Exception stream. An error is returned if the
+ /// file does not contain this stream, or the stream is smaller than the size
+ /// of the ExceptionStream structure. The internal consistency of the stream
+ /// is not checked in any way.
+ Expected<const minidump::ExceptionStream &> getExceptionStream() const {
+ return getStream<minidump::ExceptionStream>(
+ minidump::StreamType::Exception);
+ }
+
+ /// Returns the list of descriptors embedded in the MemoryList stream. The
+ /// descriptors provide the content of interesting regions of memory at the
+ /// time the minidump was taken. An error is returned if the file does not
+ /// contain this stream, or if the stream is not large enough to contain the
+ /// number of memory descriptors declared in the stream header. The
+ /// consistency of the MemoryDescriptor entries themselves is not checked in
+ /// any way.
+ Expected<ArrayRef<minidump::MemoryDescriptor>> getMemoryList() const {
+ return getListStream<minidump::MemoryDescriptor>(
+ minidump::StreamType::MemoryList);
+ }
+
+ class MemoryInfoIterator
+ : public iterator_facade_base<MemoryInfoIterator,
+ std::forward_iterator_tag,
+ minidump::MemoryInfo> {
+ public:
+ MemoryInfoIterator(ArrayRef<uint8_t> Storage, size_t Stride)
+ : Storage(Storage), Stride(Stride) {
+ assert(Storage.size() % Stride == 0);
+ }
+
+ bool operator==(const MemoryInfoIterator &R) const {
+ return Storage.size() == R.Storage.size();
+ }
+
+ const minidump::MemoryInfo &operator*() const {
+ assert(Storage.size() >= sizeof(minidump::MemoryInfo));
+ return *reinterpret_cast<const minidump::MemoryInfo *>(Storage.data());
+ }
+
+ MemoryInfoIterator &operator++() {
+ Storage = Storage.drop_front(Stride);
+ return *this;
+ }
+
+ private:
+ ArrayRef<uint8_t> Storage;
+ size_t Stride;
+ };
+
+ /// Returns the list of descriptors embedded in the MemoryInfoList stream. The
+ /// descriptors provide properties (e.g. permissions) of interesting regions
+ /// of memory at the time the minidump was taken. An error is returned if the
+ /// file does not contain this stream, or if the stream is not large enough to
+ /// contain the number of memory descriptors declared in the stream header.
+ /// The consistency of the MemoryInfoList entries themselves is not checked
+ /// in any way.
+ Expected<iterator_range<MemoryInfoIterator>> getMemoryInfoList() const;
+
+private:
+ static Error createError(StringRef Str) {
+ return make_error<GenericBinaryError>(Str, object_error::parse_failed);
+ }
+
+ static Error createEOFError() {
+ return make_error<GenericBinaryError>("Unexpected EOF",
+ object_error::unexpected_eof);
+ }
+
+ /// Return a slice of the given data array, with bounds checking.
+ static Expected<ArrayRef<uint8_t>> getDataSlice(ArrayRef<uint8_t> Data,
+ size_t Offset, size_t Size);
+
+ /// Return the slice of the given data array as an array of objects of the
+ /// given type. The function checks that the input array is large enough to
+ /// contain the correct number of objects of the given type.
+ template <typename T>
+ static Expected<ArrayRef<T>> getDataSliceAs(ArrayRef<uint8_t> Data,
+ size_t Offset, size_t Count);
+
+ MinidumpFile(MemoryBufferRef Source, const minidump::Header &Header,
+ ArrayRef<minidump::Directory> Streams,
+ DenseMap<minidump::StreamType, std::size_t> StreamMap)
+ : Binary(ID_Minidump, Source), Header(Header), Streams(Streams),
+ StreamMap(std::move(StreamMap)) {}
+
+ ArrayRef<uint8_t> getData() const {
+ return arrayRefFromStringRef(Data.getBuffer());
+ }
+
+ /// Return the stream of the given type, cast to the appropriate type. Checks
+ /// that the stream is large enough to hold an object of this type.
+ template <typename T>
+ Expected<const T &> getStream(minidump::StreamType Stream) const;
+
+ /// Return the contents of a stream which contains a list of fixed-size items,
+ /// prefixed by the list size.
+ template <typename T>
+ Expected<ArrayRef<T>> getListStream(minidump::StreamType Stream) const;
+
+ const minidump::Header &Header;
+ ArrayRef<minidump::Directory> Streams;
+ DenseMap<minidump::StreamType, std::size_t> StreamMap;
+};
+
+template <typename T>
+Expected<const T &> MinidumpFile::getStream(minidump::StreamType Type) const {
+ if (Optional<ArrayRef<uint8_t>> Stream = getRawStream(Type)) {
+ if (Stream->size() >= sizeof(T))
+ return *reinterpret_cast<const T *>(Stream->data());
+ return createEOFError();
+ }
+ return createError("No such stream");
+}
+
+template <typename T>
+Expected<ArrayRef<T>> MinidumpFile::getDataSliceAs(ArrayRef<uint8_t> Data,
+ size_t Offset,
+ size_t Count) {
+ // Check for overflow.
+ if (Count > std::numeric_limits<size_t>::max() / sizeof(T))
+ return createEOFError();
+ Expected<ArrayRef<uint8_t>> Slice =
+ getDataSlice(Data, Offset, sizeof(T) * Count);
+ if (!Slice)
+ return Slice.takeError();
+ return ArrayRef<T>(reinterpret_cast<const T *>(Slice->data()), Count);
+}
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_MINIDUMP_H
diff --git a/third_party/llvm-project/include/llvm/Object/ObjectFile.h b/third_party/llvm-project/include/llvm/Object/ObjectFile.h
new file mode 100644
index 000000000..d53c9b49c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/ObjectFile.h
@@ -0,0 +1,579 @@
+//===- ObjectFile.h - File format independent object file -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares a file format independent ObjectFile class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_OBJECTFILE_H
+#define LLVM_OBJECT_OBJECTFILE_H
+
+#include "llvm/ADT/DenseMapInfo.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/Magic.h"
+#include "llvm/MC/SubtargetFeature.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/Error.h"
+#include "llvm/Object/SymbolicFile.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <cassert>
+#include <cstdint>
+#include <memory>
+#include <system_error>
+
+namespace llvm {
+
+class ARMAttributeParser;
+
+namespace object {
+
+class COFFObjectFile;
+class MachOObjectFile;
+class ObjectFile;
+class SectionRef;
+class SymbolRef;
+class symbol_iterator;
+class WasmObjectFile;
+
+using section_iterator = content_iterator<SectionRef>;
+
+/// This is a value type class that represents a single relocation in the list
+/// of relocations in the object file.
+class RelocationRef {
+ DataRefImpl RelocationPimpl;
+ const ObjectFile *OwningObject = nullptr;
+
+public:
+ RelocationRef() = default;
+ RelocationRef(DataRefImpl RelocationP, const ObjectFile *Owner);
+
+ bool operator==(const RelocationRef &Other) const;
+
+ void moveNext();
+
+ uint64_t getOffset() const;
+ symbol_iterator getSymbol() const;
+ uint64_t getType() const;
+
+ /// Get a string that represents the type of this relocation.
+ ///
+ /// This is for display purposes only.
+ void getTypeName(SmallVectorImpl<char> &Result) const;
+
+ DataRefImpl getRawDataRefImpl() const;
+ const ObjectFile *getObject() const;
+};
+
+using relocation_iterator = content_iterator<RelocationRef>;
+
+/// This is a value type class that represents a single section in the list of
+/// sections in the object file.
+class SectionRef {
+ friend class SymbolRef;
+
+ DataRefImpl SectionPimpl;
+ const ObjectFile *OwningObject = nullptr;
+
+public:
+ SectionRef() = default;
+ SectionRef(DataRefImpl SectionP, const ObjectFile *Owner);
+
+ bool operator==(const SectionRef &Other) const;
+ bool operator!=(const SectionRef &Other) const;
+ bool operator<(const SectionRef &Other) const;
+
+ void moveNext();
+
+ Expected<StringRef> getName() const;
+ uint64_t getAddress() const;
+ uint64_t getIndex() const;
+ uint64_t getSize() const;
+ Expected<StringRef> getContents() const;
+
+ /// Get the alignment of this section as the actual value (not log 2).
+ uint64_t getAlignment() const;
+
+ bool isCompressed() const;
+ /// Whether this section contains instructions.
+ bool isText() const;
+ /// Whether this section contains data, not instructions.
+ bool isData() const;
+ /// Whether this section contains BSS uninitialized data.
+ bool isBSS() const;
+ bool isVirtual() const;
+ bool isBitcode() const;
+ bool isStripped() const;
+
+ /// Whether this section will be placed in the text segment, according to the
+ /// Berkeley size format. This is true if the section is allocatable, and
+ /// contains either code or readonly data.
+ bool isBerkeleyText() const;
+ /// Whether this section will be placed in the data segment, according to the
+ /// Berkeley size format. This is true if the section is allocatable and
+ /// contains data (e.g. PROGBITS), but is not text.
+ bool isBerkeleyData() const;
+
+ bool containsSymbol(SymbolRef S) const;
+
+ relocation_iterator relocation_begin() const;
+ relocation_iterator relocation_end() const;
+ iterator_range<relocation_iterator> relocations() const {
+ return make_range(relocation_begin(), relocation_end());
+ }
+ Expected<section_iterator> getRelocatedSection() const;
+
+ DataRefImpl getRawDataRefImpl() const;
+ const ObjectFile *getObject() const;
+};
+
+struct SectionedAddress {
+ const static uint64_t UndefSection = UINT64_MAX;
+
+ uint64_t Address = 0;
+ uint64_t SectionIndex = UndefSection;
+};
+
+inline bool operator<(const SectionedAddress &LHS,
+ const SectionedAddress &RHS) {
+ return std::tie(LHS.SectionIndex, LHS.Address) <
+ std::tie(RHS.SectionIndex, RHS.Address);
+}
+
+inline bool operator==(const SectionedAddress &LHS,
+ const SectionedAddress &RHS) {
+ return std::tie(LHS.SectionIndex, LHS.Address) ==
+ std::tie(RHS.SectionIndex, RHS.Address);
+}
+
+/// This is a value type class that represents a single symbol in the list of
+/// symbols in the object file.
+class SymbolRef : public BasicSymbolRef {
+ friend class SectionRef;
+
+public:
+ enum Type {
+ ST_Unknown, // Type not specified
+ ST_Data,
+ ST_Debug,
+ ST_File,
+ ST_Function,
+ ST_Other
+ };
+
+ SymbolRef() = default;
+ SymbolRef(DataRefImpl SymbolP, const ObjectFile *Owner);
+ SymbolRef(const BasicSymbolRef &B) : BasicSymbolRef(B) {
+ assert(isa<ObjectFile>(BasicSymbolRef::getObject()));
+ }
+
+ Expected<StringRef> getName() const;
+ /// Returns the symbol virtual address (i.e. address at which it will be
+ /// mapped).
+ Expected<uint64_t> getAddress() const;
+
+ /// Return the value of the symbol depending on the object this can be an
+ /// offset or a virtual address.
+ uint64_t getValue() const;
+
+ /// Get the alignment of this symbol as the actual value (not log 2).
+ uint32_t getAlignment() const;
+ uint64_t getCommonSize() const;
+ Expected<SymbolRef::Type> getType() const;
+
+ /// Get section this symbol is defined in reference to. Result is
+ /// end_sections() if it is undefined or is an absolute symbol.
+ Expected<section_iterator> getSection() const;
+
+ const ObjectFile *getObject() const;
+};
+
+class symbol_iterator : public basic_symbol_iterator {
+public:
+ symbol_iterator(SymbolRef Sym) : basic_symbol_iterator(Sym) {}
+ symbol_iterator(const basic_symbol_iterator &B)
+ : basic_symbol_iterator(SymbolRef(B->getRawDataRefImpl(),
+ cast<ObjectFile>(B->getObject()))) {}
+
+ const SymbolRef *operator->() const {
+ const BasicSymbolRef &P = basic_symbol_iterator::operator *();
+ return static_cast<const SymbolRef*>(&P);
+ }
+
+ const SymbolRef &operator*() const {
+ const BasicSymbolRef &P = basic_symbol_iterator::operator *();
+ return static_cast<const SymbolRef&>(P);
+ }
+};
+
+/// This class is the base class for all object file types. Concrete instances
+/// of this object are created by createObjectFile, which figures out which type
+/// to create.
+class ObjectFile : public SymbolicFile {
+ virtual void anchor();
+
+protected:
+ ObjectFile(unsigned int Type, MemoryBufferRef Source);
+
+ const uint8_t *base() const {
+ return reinterpret_cast<const uint8_t *>(Data.getBufferStart());
+ }
+
+ // These functions are for SymbolRef to call internally. The main goal of
+ // this is to allow SymbolRef::SymbolPimpl to point directly to the symbol
+ // entry in the memory mapped object file. SymbolPimpl cannot contain any
+ // virtual functions because then it could not point into the memory mapped
+ // file.
+ //
+ // Implementations assume that the DataRefImpl is valid and has not been
+ // modified externally. It's UB otherwise.
+ friend class SymbolRef;
+
+ virtual Expected<StringRef> getSymbolName(DataRefImpl Symb) const = 0;
+ Error printSymbolName(raw_ostream &OS,
+ DataRefImpl Symb) const override;
+ virtual Expected<uint64_t> getSymbolAddress(DataRefImpl Symb) const = 0;
+ virtual uint64_t getSymbolValueImpl(DataRefImpl Symb) const = 0;
+ virtual uint32_t getSymbolAlignment(DataRefImpl Symb) const;
+ virtual uint64_t getCommonSymbolSizeImpl(DataRefImpl Symb) const = 0;
+ virtual Expected<SymbolRef::Type> getSymbolType(DataRefImpl Symb) const = 0;
+ virtual Expected<section_iterator>
+ getSymbolSection(DataRefImpl Symb) const = 0;
+
+ // Same as above for SectionRef.
+ friend class SectionRef;
+
+ virtual void moveSectionNext(DataRefImpl &Sec) const = 0;
+ virtual Expected<StringRef> getSectionName(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionAddress(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionIndex(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionSize(DataRefImpl Sec) const = 0;
+ virtual Expected<ArrayRef<uint8_t>>
+ getSectionContents(DataRefImpl Sec) const = 0;
+ virtual uint64_t getSectionAlignment(DataRefImpl Sec) const = 0;
+ virtual bool isSectionCompressed(DataRefImpl Sec) const = 0;
+ virtual bool isSectionText(DataRefImpl Sec) const = 0;
+ virtual bool isSectionData(DataRefImpl Sec) const = 0;
+ virtual bool isSectionBSS(DataRefImpl Sec) const = 0;
+ // A section is 'virtual' if its contents aren't present in the object image.
+ virtual bool isSectionVirtual(DataRefImpl Sec) const = 0;
+ virtual bool isSectionBitcode(DataRefImpl Sec) const;
+ virtual bool isSectionStripped(DataRefImpl Sec) const;
+ virtual bool isBerkeleyText(DataRefImpl Sec) const;
+ virtual bool isBerkeleyData(DataRefImpl Sec) const;
+ virtual relocation_iterator section_rel_begin(DataRefImpl Sec) const = 0;
+ virtual relocation_iterator section_rel_end(DataRefImpl Sec) const = 0;
+ virtual Expected<section_iterator> getRelocatedSection(DataRefImpl Sec) const;
+
+ // Same as above for RelocationRef.
+ friend class RelocationRef;
+ virtual void moveRelocationNext(DataRefImpl &Rel) const = 0;
+ virtual uint64_t getRelocationOffset(DataRefImpl Rel) const = 0;
+ virtual symbol_iterator getRelocationSymbol(DataRefImpl Rel) const = 0;
+ virtual uint64_t getRelocationType(DataRefImpl Rel) const = 0;
+ virtual void getRelocationTypeName(DataRefImpl Rel,
+ SmallVectorImpl<char> &Result) const = 0;
+
+ uint64_t getSymbolValue(DataRefImpl Symb) const;
+
+public:
+ ObjectFile() = delete;
+ ObjectFile(const ObjectFile &other) = delete;
+
+ uint64_t getCommonSymbolSize(DataRefImpl Symb) const {
+ assert(getSymbolFlags(Symb) & SymbolRef::SF_Common);
+ return getCommonSymbolSizeImpl(Symb);
+ }
+
+ virtual std::vector<SectionRef> dynamic_relocation_sections() const {
+ return std::vector<SectionRef>();
+ }
+
+ using symbol_iterator_range = iterator_range<symbol_iterator>;
+ symbol_iterator_range symbols() const {
+ return symbol_iterator_range(symbol_begin(), symbol_end());
+ }
+
+ virtual section_iterator section_begin() const = 0;
+ virtual section_iterator section_end() const = 0;
+
+ using section_iterator_range = iterator_range<section_iterator>;
+ section_iterator_range sections() const {
+ return section_iterator_range(section_begin(), section_end());
+ }
+
+ /// The number of bytes used to represent an address in this object
+ /// file format.
+ virtual uint8_t getBytesInAddress() const = 0;
+
+ virtual StringRef getFileFormatName() const = 0;
+ virtual Triple::ArchType getArch() const = 0;
+ virtual SubtargetFeatures getFeatures() const = 0;
+ virtual void setARMSubArch(Triple &TheTriple) const { }
+ virtual Expected<uint64_t> getStartAddress() const {
+ // XXX BINARYEN
+ llvm_unreachable("getStartAddress");
+ //return errorCodeToError(object_error::parse_failed);
+ };
+
+ /// Create a triple from the data in this object file.
+ Triple makeTriple() const;
+
+ /// Maps a debug section name to a standard DWARF section name.
+ virtual StringRef mapDebugSectionName(StringRef Name) const { return Name; }
+
+ /// True if this is a relocatable object (.o/.obj).
+ virtual bool isRelocatableObject() const = 0;
+
+ /// @returns Pointer to ObjectFile subclass to handle this type of object.
+ /// @param ObjectPath The path to the object file. ObjectPath.isObject must
+ /// return true.
+ /// Create ObjectFile from path.
+ static Expected<OwningBinary<ObjectFile>>
+ createObjectFile(StringRef ObjectPath);
+
+ static Expected<std::unique_ptr<ObjectFile>>
+ createObjectFile(MemoryBufferRef Object, llvm::file_magic Type);
+ static Expected<std::unique_ptr<ObjectFile>>
+ createObjectFile(MemoryBufferRef Object) {
+ return createObjectFile(Object, llvm::file_magic::unknown);
+ }
+
+ static bool classof(const Binary *v) {
+ return v->isObject();
+ }
+
+ static Expected<std::unique_ptr<COFFObjectFile>>
+ createCOFFObjectFile(MemoryBufferRef Object);
+
+ static Expected<std::unique_ptr<ObjectFile>>
+ createXCOFFObjectFile(MemoryBufferRef Object, unsigned FileType);
+
+ static Expected<std::unique_ptr<ObjectFile>>
+ createELFObjectFile(MemoryBufferRef Object);
+
+ static Expected<std::unique_ptr<MachOObjectFile>>
+ createMachOObjectFile(MemoryBufferRef Object,
+ uint32_t UniversalCputype = 0,
+ uint32_t UniversalIndex = 0);
+
+ static Expected<std::unique_ptr<WasmObjectFile>>
+ createWasmObjectFile(MemoryBufferRef Object);
+};
+
+// Inline function definitions.
+inline SymbolRef::SymbolRef(DataRefImpl SymbolP, const ObjectFile *Owner)
+ : BasicSymbolRef(SymbolP, Owner) {}
+
+inline Expected<StringRef> SymbolRef::getName() const {
+ return getObject()->getSymbolName(getRawDataRefImpl());
+}
+
+inline Expected<uint64_t> SymbolRef::getAddress() const {
+ return getObject()->getSymbolAddress(getRawDataRefImpl());
+}
+
+inline uint64_t SymbolRef::getValue() const {
+ return getObject()->getSymbolValue(getRawDataRefImpl());
+}
+
+inline uint32_t SymbolRef::getAlignment() const {
+ return getObject()->getSymbolAlignment(getRawDataRefImpl());
+}
+
+inline uint64_t SymbolRef::getCommonSize() const {
+ return getObject()->getCommonSymbolSize(getRawDataRefImpl());
+}
+
+inline Expected<section_iterator> SymbolRef::getSection() const {
+ return getObject()->getSymbolSection(getRawDataRefImpl());
+}
+
+inline Expected<SymbolRef::Type> SymbolRef::getType() const {
+ return getObject()->getSymbolType(getRawDataRefImpl());
+}
+
+inline const ObjectFile *SymbolRef::getObject() const {
+ const SymbolicFile *O = BasicSymbolRef::getObject();
+ return cast<ObjectFile>(O);
+}
+
+/// SectionRef
+inline SectionRef::SectionRef(DataRefImpl SectionP,
+ const ObjectFile *Owner)
+ : SectionPimpl(SectionP)
+ , OwningObject(Owner) {}
+
+inline bool SectionRef::operator==(const SectionRef &Other) const {
+ return OwningObject == Other.OwningObject &&
+ SectionPimpl == Other.SectionPimpl;
+}
+
+inline bool SectionRef::operator!=(const SectionRef &Other) const {
+ return !(*this == Other);
+}
+
+inline bool SectionRef::operator<(const SectionRef &Other) const {
+ assert(OwningObject == Other.OwningObject);
+ return SectionPimpl < Other.SectionPimpl;
+}
+
+inline void SectionRef::moveNext() {
+ return OwningObject->moveSectionNext(SectionPimpl);
+}
+
+inline Expected<StringRef> SectionRef::getName() const {
+ return OwningObject->getSectionName(SectionPimpl);
+}
+
+inline uint64_t SectionRef::getAddress() const {
+ return OwningObject->getSectionAddress(SectionPimpl);
+}
+
+inline uint64_t SectionRef::getIndex() const {
+ return OwningObject->getSectionIndex(SectionPimpl);
+}
+
+inline uint64_t SectionRef::getSize() const {
+ return OwningObject->getSectionSize(SectionPimpl);
+}
+
+inline Expected<StringRef> SectionRef::getContents() const {
+ Expected<ArrayRef<uint8_t>> Res =
+ OwningObject->getSectionContents(SectionPimpl);
+ if (!Res)
+ return Res.takeError();
+ return StringRef(reinterpret_cast<const char *>(Res->data()), Res->size());
+}
+
+inline uint64_t SectionRef::getAlignment() const {
+ return OwningObject->getSectionAlignment(SectionPimpl);
+}
+
+inline bool SectionRef::isCompressed() const {
+ return OwningObject->isSectionCompressed(SectionPimpl);
+}
+
+inline bool SectionRef::isText() const {
+ return OwningObject->isSectionText(SectionPimpl);
+}
+
+inline bool SectionRef::isData() const {
+ return OwningObject->isSectionData(SectionPimpl);
+}
+
+inline bool SectionRef::isBSS() const {
+ return OwningObject->isSectionBSS(SectionPimpl);
+}
+
+inline bool SectionRef::isVirtual() const {
+ return OwningObject->isSectionVirtual(SectionPimpl);
+}
+
+inline bool SectionRef::isBitcode() const {
+ return OwningObject->isSectionBitcode(SectionPimpl);
+}
+
+inline bool SectionRef::isStripped() const {
+ return OwningObject->isSectionStripped(SectionPimpl);
+}
+
+inline bool SectionRef::isBerkeleyText() const {
+ return OwningObject->isBerkeleyText(SectionPimpl);
+}
+
+inline bool SectionRef::isBerkeleyData() const {
+ return OwningObject->isBerkeleyData(SectionPimpl);
+}
+
+inline relocation_iterator SectionRef::relocation_begin() const {
+ return OwningObject->section_rel_begin(SectionPimpl);
+}
+
+inline relocation_iterator SectionRef::relocation_end() const {
+ return OwningObject->section_rel_end(SectionPimpl);
+}
+
+inline Expected<section_iterator> SectionRef::getRelocatedSection() const {
+ return OwningObject->getRelocatedSection(SectionPimpl);
+}
+
+inline DataRefImpl SectionRef::getRawDataRefImpl() const {
+ return SectionPimpl;
+}
+
+inline const ObjectFile *SectionRef::getObject() const {
+ return OwningObject;
+}
+
+/// RelocationRef
+inline RelocationRef::RelocationRef(DataRefImpl RelocationP,
+ const ObjectFile *Owner)
+ : RelocationPimpl(RelocationP)
+ , OwningObject(Owner) {}
+
+inline bool RelocationRef::operator==(const RelocationRef &Other) const {
+ return RelocationPimpl == Other.RelocationPimpl;
+}
+
+inline void RelocationRef::moveNext() {
+ return OwningObject->moveRelocationNext(RelocationPimpl);
+}
+
+inline uint64_t RelocationRef::getOffset() const {
+ return OwningObject->getRelocationOffset(RelocationPimpl);
+}
+
+inline symbol_iterator RelocationRef::getSymbol() const {
+ return OwningObject->getRelocationSymbol(RelocationPimpl);
+}
+
+inline uint64_t RelocationRef::getType() const {
+ return OwningObject->getRelocationType(RelocationPimpl);
+}
+
+inline void RelocationRef::getTypeName(SmallVectorImpl<char> &Result) const {
+ return OwningObject->getRelocationTypeName(RelocationPimpl, Result);
+}
+
+inline DataRefImpl RelocationRef::getRawDataRefImpl() const {
+ return RelocationPimpl;
+}
+
+inline const ObjectFile *RelocationRef::getObject() const {
+ return OwningObject;
+}
+
+} // end namespace object
+
+template <> struct DenseMapInfo<object::SectionRef> {
+ static bool isEqual(const object::SectionRef &A,
+ const object::SectionRef &B) {
+ return A == B;
+ }
+ static object::SectionRef getEmptyKey() {
+ return object::SectionRef({}, nullptr);
+ }
+ static object::SectionRef getTombstoneKey() {
+ object::DataRefImpl TS;
+ TS.p = (uintptr_t)-1;
+ return object::SectionRef(TS, nullptr);
+ }
+ static unsigned getHashValue(const object::SectionRef &Sec) {
+ object::DataRefImpl Raw = Sec.getRawDataRefImpl();
+ return hash_combine(Raw.p, Raw.d.a, Raw.d.b);
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_OBJECTFILE_H
diff --git a/third_party/llvm-project/include/llvm/Object/RelocationResolver.h b/third_party/llvm-project/include/llvm/Object/RelocationResolver.h
new file mode 100644
index 000000000..1246dcc5e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/RelocationResolver.h
@@ -0,0 +1,42 @@
+//===- RelocVisitor.h - Visitor for object file relocations -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides a wrapper around all the different types of relocations
+// in different file formats, such that a client can handle them in a unified
+// manner by only implementing a minimal number of functions.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_RELOCVISITOR_H
+#define LLVM_OBJECT_RELOCVISITOR_H
+
+#include "llvm/ADT/Triple.h"
+#include "llvm/BinaryFormat/ELF.h"
+#include "llvm/BinaryFormat/MachO.h"
+#include "llvm/Object/COFF.h"
+#include "llvm/Object/ELFObjectFile.h"
+#include "llvm/Object/MachO.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Object/Wasm.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <cstdint>
+#include <system_error>
+
+namespace llvm {
+namespace object {
+
+using RelocationResolver = uint64_t (*)(RelocationRef R, uint64_t S, uint64_t A);
+
+std::pair<bool (*)(uint64_t), RelocationResolver>
+getRelocationResolver(const ObjectFile &Obj);
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_RELOCVISITOR_H
diff --git a/third_party/llvm-project/include/llvm/Object/SymbolicFile.h b/third_party/llvm-project/include/llvm/Object/SymbolicFile.h
new file mode 100644
index 000000000..1398fa134
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/SymbolicFile.h
@@ -0,0 +1,214 @@
+//===- SymbolicFile.h - Interface that only provides symbols ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the SymbolicFile interface.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_SYMBOLICFILE_H
+#define LLVM_OBJECT_SYMBOLICFILE_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/BinaryFormat/Magic.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <cinttypes>
+#include <cstdint>
+#include <cstring>
+#include <iterator>
+#include <memory>
+#include <system_error>
+
+namespace llvm {
+namespace object {
+
+union DataRefImpl {
+ // This entire union should probably be a
+ // char[max(8, sizeof(uintptr_t))] and require the impl to cast.
+ struct {
+ uint32_t a, b;
+ } d;
+ uintptr_t p;
+
+ DataRefImpl() { std::memset(this, 0, sizeof(DataRefImpl)); }
+};
+
+template <typename OStream>
+OStream& operator<<(OStream &OS, const DataRefImpl &D) {
+ OS << "(" << format("0x%08" PRIxPTR, D.p) << " (" << format("0x%08x", D.d.a)
+ << ", " << format("0x%08x", D.d.b) << "))";
+ return OS;
+}
+
+inline bool operator==(const DataRefImpl &a, const DataRefImpl &b) {
+ // Check bitwise identical. This is the only legal way to compare a union w/o
+ // knowing which member is in use.
+ return std::memcmp(&a, &b, sizeof(DataRefImpl)) == 0;
+}
+
+inline bool operator!=(const DataRefImpl &a, const DataRefImpl &b) {
+ return !operator==(a, b);
+}
+
+inline bool operator<(const DataRefImpl &a, const DataRefImpl &b) {
+ // Check bitwise identical. This is the only legal way to compare a union w/o
+ // knowing which member is in use.
+ return std::memcmp(&a, &b, sizeof(DataRefImpl)) < 0;
+}
+
+template <class content_type>
+class content_iterator
+ : public std::iterator<std::forward_iterator_tag, content_type> {
+ content_type Current;
+
+public:
+ content_iterator(content_type symb) : Current(std::move(symb)) {}
+
+ const content_type *operator->() const { return &Current; }
+
+ const content_type &operator*() const { return Current; }
+
+ bool operator==(const content_iterator &other) const {
+ return Current == other.Current;
+ }
+
+ bool operator!=(const content_iterator &other) const {
+ return !(*this == other);
+ }
+
+ content_iterator &operator++() { // preincrement
+ Current.moveNext();
+ return *this;
+ }
+};
+
+class SymbolicFile;
+
+/// This is a value type class that represents a single symbol in the list of
+/// symbols in the object file.
+class BasicSymbolRef {
+ DataRefImpl SymbolPimpl;
+ const SymbolicFile *OwningObject = nullptr;
+
+public:
+ enum Flags : unsigned {
+ SF_None = 0,
+ SF_Undefined = 1U << 0, // Symbol is defined in another object file
+ SF_Global = 1U << 1, // Global symbol
+ SF_Weak = 1U << 2, // Weak symbol
+ SF_Absolute = 1U << 3, // Absolute symbol
+ SF_Common = 1U << 4, // Symbol has common linkage
+ SF_Indirect = 1U << 5, // Symbol is an alias to another symbol
+ SF_Exported = 1U << 6, // Symbol is visible to other DSOs
+ SF_FormatSpecific = 1U << 7, // Specific to the object file format
+ // (e.g. section symbols)
+ SF_Thumb = 1U << 8, // Thumb symbol in a 32-bit ARM binary
+ SF_Hidden = 1U << 9, // Symbol has hidden visibility
+ SF_Const = 1U << 10, // Symbol value is constant
+ SF_Executable = 1U << 11, // Symbol points to an executable section
+ // (IR only)
+ };
+
+ BasicSymbolRef() = default;
+ BasicSymbolRef(DataRefImpl SymbolP, const SymbolicFile *Owner);
+
+ bool operator==(const BasicSymbolRef &Other) const;
+ bool operator<(const BasicSymbolRef &Other) const;
+
+ void moveNext();
+
+ Error printName(raw_ostream &OS) const;
+
+ /// Get symbol flags (bitwise OR of SymbolRef::Flags)
+ uint32_t getFlags() const;
+
+ DataRefImpl getRawDataRefImpl() const;
+ const SymbolicFile *getObject() const;
+};
+
+using basic_symbol_iterator = content_iterator<BasicSymbolRef>;
+
+class SymbolicFile : public Binary {
+public:
+ SymbolicFile(unsigned int Type, MemoryBufferRef Source);
+ ~SymbolicFile() override;
+
+ // virtual interface.
+ virtual void moveSymbolNext(DataRefImpl &Symb) const = 0;
+
+ virtual Error printSymbolName(raw_ostream &OS, DataRefImpl Symb) const = 0;
+
+ virtual uint32_t getSymbolFlags(DataRefImpl Symb) const = 0;
+
+ virtual basic_symbol_iterator symbol_begin() const = 0;
+
+ virtual basic_symbol_iterator symbol_end() const = 0;
+
+ // convenience wrappers.
+ using basic_symbol_iterator_range = iterator_range<basic_symbol_iterator>;
+ basic_symbol_iterator_range symbols() const {
+ return basic_symbol_iterator_range(symbol_begin(), symbol_end());
+ }
+
+ // construction aux.
+ static Expected<std::unique_ptr<SymbolicFile>>
+ createSymbolicFile(MemoryBufferRef Object, llvm::file_magic Type,
+ LLVMContext *Context);
+
+ static Expected<std::unique_ptr<SymbolicFile>>
+ createSymbolicFile(MemoryBufferRef Object) {
+ return createSymbolicFile(Object, llvm::file_magic::unknown, nullptr);
+ }
+ static Expected<OwningBinary<SymbolicFile>>
+ createSymbolicFile(StringRef ObjectPath);
+
+ static bool classof(const Binary *v) {
+ return v->isSymbolic();
+ }
+};
+
+inline BasicSymbolRef::BasicSymbolRef(DataRefImpl SymbolP,
+ const SymbolicFile *Owner)
+ : SymbolPimpl(SymbolP), OwningObject(Owner) {}
+
+inline bool BasicSymbolRef::operator==(const BasicSymbolRef &Other) const {
+ return SymbolPimpl == Other.SymbolPimpl;
+}
+
+inline bool BasicSymbolRef::operator<(const BasicSymbolRef &Other) const {
+ return SymbolPimpl < Other.SymbolPimpl;
+}
+
+inline void BasicSymbolRef::moveNext() {
+ return OwningObject->moveSymbolNext(SymbolPimpl);
+}
+
+inline Error BasicSymbolRef::printName(raw_ostream &OS) const {
+ return OwningObject->printSymbolName(OS, SymbolPimpl);
+}
+
+inline uint32_t BasicSymbolRef::getFlags() const {
+ return OwningObject->getSymbolFlags(SymbolPimpl);
+}
+
+inline DataRefImpl BasicSymbolRef::getRawDataRefImpl() const {
+ return SymbolPimpl;
+}
+
+inline const SymbolicFile *BasicSymbolRef::getObject() const {
+ return OwningObject;
+}
+
+} // end namespace object
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_SYMBOLICFILE_H
diff --git a/third_party/llvm-project/include/llvm/Object/Wasm.h b/third_party/llvm-project/include/llvm/Object/Wasm.h
new file mode 100644
index 000000000..e130ea32e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Object/Wasm.h
@@ -0,0 +1,356 @@
+//===- Wasm.h - Wasm object file implementation -----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the WasmObjectFile class, which implements the ObjectFile
+// interface for Wasm files.
+//
+// See: https://github.com/WebAssembly/design/blob/master/BinaryEncoding.md
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_WASM_H
+#define LLVM_OBJECT_WASM_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Wasm.h"
+#include "llvm/Config/llvm-config.h"
+#include "llvm/MC/MCSymbolWasm.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <cstddef>
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+namespace object {
+
+class WasmSymbol {
+public:
+ WasmSymbol(const wasm::WasmSymbolInfo &Info,
+ const wasm::WasmGlobalType *GlobalType,
+ const wasm::WasmEventType *EventType,
+ const wasm::WasmSignature *Signature)
+ : Info(Info), GlobalType(GlobalType), EventType(EventType),
+ Signature(Signature) {}
+
+ const wasm::WasmSymbolInfo &Info;
+ const wasm::WasmGlobalType *GlobalType;
+ const wasm::WasmEventType *EventType;
+ const wasm::WasmSignature *Signature;
+
+ bool isTypeFunction() const {
+ return Info.Kind == wasm::WASM_SYMBOL_TYPE_FUNCTION;
+ }
+
+ bool isTypeData() const { return Info.Kind == wasm::WASM_SYMBOL_TYPE_DATA; }
+
+ bool isTypeGlobal() const {
+ return Info.Kind == wasm::WASM_SYMBOL_TYPE_GLOBAL;
+ }
+
+ bool isTypeSection() const {
+ return Info.Kind == wasm::WASM_SYMBOL_TYPE_SECTION;
+ }
+
+ bool isTypeEvent() const { return Info.Kind == wasm::WASM_SYMBOL_TYPE_EVENT; }
+
+ bool isDefined() const { return !isUndefined(); }
+
+ bool isUndefined() const {
+ return (Info.Flags & wasm::WASM_SYMBOL_UNDEFINED) != 0;
+ }
+
+ bool isBindingWeak() const {
+ return getBinding() == wasm::WASM_SYMBOL_BINDING_WEAK;
+ }
+
+ bool isBindingGlobal() const {
+ return getBinding() == wasm::WASM_SYMBOL_BINDING_GLOBAL;
+ }
+
+ bool isBindingLocal() const {
+ return getBinding() == wasm::WASM_SYMBOL_BINDING_LOCAL;
+ }
+
+ unsigned getBinding() const {
+ return Info.Flags & wasm::WASM_SYMBOL_BINDING_MASK;
+ }
+
+ bool isHidden() const {
+ return getVisibility() == wasm::WASM_SYMBOL_VISIBILITY_HIDDEN;
+ }
+
+ unsigned getVisibility() const {
+ return Info.Flags & wasm::WASM_SYMBOL_VISIBILITY_MASK;
+ }
+
+ void print(raw_ostream &Out) const;
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+ LLVM_DUMP_METHOD void dump() const;
+#endif
+};
+
+struct WasmSection {
+ WasmSection() = default;
+
+ uint32_t Type = 0; // Section type (See below)
+ uint32_t Offset = 0; // Offset with in the file
+ StringRef Name; // Section name (User-defined sections only)
+ ArrayRef<uint8_t> Content; // Section content
+ std::vector<wasm::WasmRelocation> Relocations; // Relocations for this section
+};
+
+struct WasmSegment {
+ uint32_t SectionOffset;
+ wasm::WasmDataSegment Data;
+};
+
+class WasmObjectFile : public ObjectFile {
+
+public:
+ WasmObjectFile(MemoryBufferRef Object, Error &Err);
+
+ const wasm::WasmObjectHeader &getHeader() const;
+ const WasmSymbol &getWasmSymbol(const DataRefImpl &Symb) const;
+ const WasmSymbol &getWasmSymbol(const SymbolRef &Symbol) const;
+ const WasmSection &getWasmSection(const SectionRef &Section) const;
+ const wasm::WasmRelocation &getWasmRelocation(const RelocationRef &Ref) const;
+
+ static bool classof(const Binary *v) { return v->isWasm(); }
+
+ const wasm::WasmDylinkInfo &dylinkInfo() const { return DylinkInfo; }
+ const wasm::WasmProducerInfo &getProducerInfo() const { return ProducerInfo; }
+ ArrayRef<wasm::WasmFeatureEntry> getTargetFeatures() const {
+ return TargetFeatures;
+ }
+ ArrayRef<wasm::WasmSignature> types() const { return Signatures; }
+ ArrayRef<uint32_t> functionTypes() const { return FunctionTypes; }
+ ArrayRef<wasm::WasmImport> imports() const { return Imports; }
+ ArrayRef<wasm::WasmTable> tables() const { return Tables; }
+ ArrayRef<wasm::WasmLimits> memories() const { return Memories; }
+ ArrayRef<wasm::WasmGlobal> globals() const { return Globals; }
+ ArrayRef<wasm::WasmEvent> events() const { return Events; }
+ ArrayRef<wasm::WasmExport> exports() const { return Exports; }
+ ArrayRef<WasmSymbol> syms() const { return Symbols; }
+ const wasm::WasmLinkingData &linkingData() const { return LinkingData; }
+ uint32_t getNumberOfSymbols() const { return Symbols.size(); }
+ ArrayRef<wasm::WasmElemSegment> elements() const { return ElemSegments; }
+ ArrayRef<WasmSegment> dataSegments() const { return DataSegments; }
+ ArrayRef<wasm::WasmFunction> functions() const { return Functions; }
+ ArrayRef<wasm::WasmFunctionName> debugNames() const { return DebugNames; }
+ uint32_t startFunction() const { return StartFunction; }
+ uint32_t getNumImportedGlobals() const { return NumImportedGlobals; }
+ uint32_t getNumImportedFunctions() const { return NumImportedFunctions; }
+ uint32_t getNumImportedEvents() const { return NumImportedEvents; }
+ void moveSymbolNext(DataRefImpl &Symb) const override;
+
+ uint32_t getSymbolFlags(DataRefImpl Symb) const override;
+
+ basic_symbol_iterator symbol_begin() const override;
+
+ basic_symbol_iterator symbol_end() const override;
+ Expected<StringRef> getSymbolName(DataRefImpl Symb) const override;
+
+ Expected<uint64_t> getSymbolAddress(DataRefImpl Symb) const override;
+ uint64_t getWasmSymbolValue(const WasmSymbol &Sym) const;
+ uint64_t getSymbolValueImpl(DataRefImpl Symb) const override;
+ uint32_t getSymbolAlignment(DataRefImpl Symb) const override;
+ uint64_t getCommonSymbolSizeImpl(DataRefImpl Symb) const override;
+ Expected<SymbolRef::Type> getSymbolType(DataRefImpl Symb) const override;
+ Expected<section_iterator> getSymbolSection(DataRefImpl Symb) const override;
+
+ // Overrides from SectionRef.
+ void moveSectionNext(DataRefImpl &Sec) const override;
+ Expected<StringRef> getSectionName(DataRefImpl Sec) const override;
+ uint64_t getSectionAddress(DataRefImpl Sec) const override;
+ uint64_t getSectionIndex(DataRefImpl Sec) const override;
+ uint64_t getSectionSize(DataRefImpl Sec) const override;
+ Expected<ArrayRef<uint8_t>>
+ getSectionContents(DataRefImpl Sec) const override;
+ uint64_t getSectionAlignment(DataRefImpl Sec) const override;
+ bool isSectionCompressed(DataRefImpl Sec) const override;
+ bool isSectionText(DataRefImpl Sec) const override;
+ bool isSectionData(DataRefImpl Sec) const override;
+ bool isSectionBSS(DataRefImpl Sec) const override;
+ bool isSectionVirtual(DataRefImpl Sec) const override;
+ bool isSectionBitcode(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_begin(DataRefImpl Sec) const override;
+ relocation_iterator section_rel_end(DataRefImpl Sec) const override;
+
+ // Overrides from RelocationRef.
+ void moveRelocationNext(DataRefImpl &Rel) const override;
+ uint64_t getRelocationOffset(DataRefImpl Rel) const override;
+ symbol_iterator getRelocationSymbol(DataRefImpl Rel) const override;
+ uint64_t getRelocationType(DataRefImpl Rel) const override;
+ void getRelocationTypeName(DataRefImpl Rel,
+ SmallVectorImpl<char> &Result) const override;
+
+ section_iterator section_begin() const override;
+ section_iterator section_end() const override;
+ uint8_t getBytesInAddress() const override;
+ StringRef getFileFormatName() const override;
+ Triple::ArchType getArch() const override;
+ SubtargetFeatures getFeatures() const override;
+ bool isRelocatableObject() const override;
+ bool isSharedObject() const;
+
+ struct ReadContext {
+ const uint8_t *Start;
+ const uint8_t *Ptr;
+ const uint8_t *End;
+ };
+
+private:
+ bool isValidFunctionIndex(uint32_t Index) const;
+ bool isDefinedFunctionIndex(uint32_t Index) const;
+ bool isValidGlobalIndex(uint32_t Index) const;
+ bool isDefinedGlobalIndex(uint32_t Index) const;
+ bool isValidEventIndex(uint32_t Index) const;
+ bool isDefinedEventIndex(uint32_t Index) const;
+ bool isValidFunctionSymbol(uint32_t Index) const;
+ bool isValidGlobalSymbol(uint32_t Index) const;
+ bool isValidEventSymbol(uint32_t Index) const;
+ bool isValidDataSymbol(uint32_t Index) const;
+ bool isValidSectionSymbol(uint32_t Index) const;
+ wasm::WasmFunction &getDefinedFunction(uint32_t Index);
+ const wasm::WasmFunction &getDefinedFunction(uint32_t Index) const;
+ wasm::WasmGlobal &getDefinedGlobal(uint32_t Index);
+ wasm::WasmEvent &getDefinedEvent(uint32_t Index);
+
+ const WasmSection &getWasmSection(DataRefImpl Ref) const;
+ const wasm::WasmRelocation &getWasmRelocation(DataRefImpl Ref) const;
+
+ Error parseSection(WasmSection &Sec);
+ Error parseCustomSection(WasmSection &Sec, ReadContext &Ctx);
+
+ // Standard section types
+ Error parseTypeSection(ReadContext &Ctx);
+ Error parseImportSection(ReadContext &Ctx);
+ Error parseFunctionSection(ReadContext &Ctx);
+ Error parseTableSection(ReadContext &Ctx);
+ Error parseMemorySection(ReadContext &Ctx);
+ Error parseGlobalSection(ReadContext &Ctx);
+ Error parseEventSection(ReadContext &Ctx);
+ Error parseExportSection(ReadContext &Ctx);
+ Error parseStartSection(ReadContext &Ctx);
+ Error parseElemSection(ReadContext &Ctx);
+ Error parseCodeSection(ReadContext &Ctx);
+ Error parseDataSection(ReadContext &Ctx);
+ Error parseDataCountSection(ReadContext &Ctx);
+
+ // Custom section types
+ Error parseDylinkSection(ReadContext &Ctx);
+ Error parseNameSection(ReadContext &Ctx);
+ Error parseLinkingSection(ReadContext &Ctx);
+ Error parseLinkingSectionSymtab(ReadContext &Ctx);
+ Error parseLinkingSectionComdat(ReadContext &Ctx);
+ Error parseProducersSection(ReadContext &Ctx);
+ Error parseTargetFeaturesSection(ReadContext &Ctx);
+ Error parseRelocSection(StringRef Name, ReadContext &Ctx);
+
+ wasm::WasmObjectHeader Header;
+ std::vector<WasmSection> Sections;
+ wasm::WasmDylinkInfo DylinkInfo;
+ wasm::WasmProducerInfo ProducerInfo;
+ std::vector<wasm::WasmFeatureEntry> TargetFeatures;
+ std::vector<wasm::WasmSignature> Signatures;
+ std::vector<uint32_t> FunctionTypes;
+ std::vector<wasm::WasmTable> Tables;
+ std::vector<wasm::WasmLimits> Memories;
+ std::vector<wasm::WasmGlobal> Globals;
+ std::vector<wasm::WasmEvent> Events;
+ std::vector<wasm::WasmImport> Imports;
+ std::vector<wasm::WasmExport> Exports;
+ std::vector<wasm::WasmElemSegment> ElemSegments;
+ std::vector<WasmSegment> DataSegments;
+ llvm::Optional<size_t> DataCount;
+ std::vector<wasm::WasmFunction> Functions;
+ std::vector<WasmSymbol> Symbols;
+ std::vector<wasm::WasmFunctionName> DebugNames;
+ uint32_t StartFunction = -1;
+ bool HasLinkingSection = false;
+ bool HasDylinkSection = false;
+ wasm::WasmLinkingData LinkingData;
+ uint32_t NumImportedGlobals = 0;
+ uint32_t NumImportedFunctions = 0;
+ uint32_t NumImportedEvents = 0;
+ uint32_t CodeSection = 0;
+ uint32_t DataSection = 0;
+ uint32_t GlobalSection = 0;
+ uint32_t EventSection = 0;
+};
+
+class WasmSectionOrderChecker {
+public:
+ // We define orders for all core wasm sections and known custom sections.
+ enum : int {
+ // Sentinel, must be zero
+ WASM_SEC_ORDER_NONE = 0,
+
+ // Core sections
+ WASM_SEC_ORDER_TYPE,
+ WASM_SEC_ORDER_IMPORT,
+ WASM_SEC_ORDER_FUNCTION,
+ WASM_SEC_ORDER_TABLE,
+ WASM_SEC_ORDER_MEMORY,
+ WASM_SEC_ORDER_GLOBAL,
+ WASM_SEC_ORDER_EVENT,
+ WASM_SEC_ORDER_EXPORT,
+ WASM_SEC_ORDER_START,
+ WASM_SEC_ORDER_ELEM,
+ WASM_SEC_ORDER_DATACOUNT,
+ WASM_SEC_ORDER_CODE,
+ WASM_SEC_ORDER_DATA,
+
+ // Custom sections
+ // "dylink" should be the very first section in the module
+ WASM_SEC_ORDER_DYLINK,
+ // "linking" section requires DATA section in order to validate data symbols
+ WASM_SEC_ORDER_LINKING,
+ // Must come after "linking" section in order to validate reloc indexes.
+ WASM_SEC_ORDER_RELOC,
+ // "name" section must appear after DATA. Comes after "linking" to allow
+ // symbol table to set default function name.
+ WASM_SEC_ORDER_NAME,
+ // "producers" section must appear after "name" section.
+ WASM_SEC_ORDER_PRODUCERS,
+ // "target_features" section must appear after producers section
+ WASM_SEC_ORDER_TARGET_FEATURES,
+
+ // Must be last
+ WASM_NUM_SEC_ORDERS
+
+ };
+
+ // Sections that may or may not be present, but cannot be predecessors
+ static int DisallowedPredecessors[WASM_NUM_SEC_ORDERS][WASM_NUM_SEC_ORDERS];
+
+ bool isValidSectionOrder(unsigned ID, StringRef CustomSectionName = "");
+
+private:
+ bool Seen[WASM_NUM_SEC_ORDERS] = {}; // Sections that have been seen already
+
+ // Returns -1 for unknown sections.
+ int getSectionOrder(unsigned ID, StringRef CustomSectionName = "");
+};
+
+} // end namespace object
+
+inline raw_ostream &operator<<(raw_ostream &OS, const object::WasmSymbol &Sym) {
+ Sym.print(OS);
+ return OS;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_OBJECT_WASM_H
diff --git a/third_party/llvm-project/include/llvm/ObjectYAML/DWARFEmitter.h b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFEmitter.h
new file mode 100644
index 000000000..85bc81c60
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFEmitter.h
@@ -0,0 +1,50 @@
+//===--- DWARFEmitter.h - ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+/// Common declarations for yaml2obj
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECTYAML_DWARFEMITTER_H
+#define LLVM_OBJECTYAML_DWARFEMITTER_H
+
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Host.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <memory>
+
+namespace llvm {
+
+class raw_ostream;
+
+namespace DWARFYAML {
+
+struct Data;
+struct PubSection;
+
+void EmitDebugAbbrev(raw_ostream &OS, const Data &DI);
+void EmitDebugStr(raw_ostream &OS, const Data &DI);
+
+void EmitDebugAranges(raw_ostream &OS, const Data &DI);
+void EmitDebugRanges(raw_ostream &OS, const Data &DI); // XXX BINARYEN
+void EmitPubSection(raw_ostream &OS, const PubSection &Sect,
+ bool IsLittleEndian);
+void EmitDebugInfo(raw_ostream &OS, const Data &DI);
+void EmitDebugLine(raw_ostream &OS, const Data &DI);
+
+Expected<StringMap<std::unique_ptr<MemoryBuffer>>>
+EmitDebugSections(StringRef YAMLString, bool ApplyFixups = false,
+ bool IsLittleEndian = sys::IsLittleEndianHost);
+StringMap<std::unique_ptr<MemoryBuffer>>
+EmitDebugSections(llvm::DWARFYAML::Data &DI, bool ApplyFixups);
+
+} // end namespace DWARFYAML
+} // end namespace llvm
+
+#endif // LLVM_OBJECTYAML_DWARFEMITTER_H
diff --git a/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
new file mode 100644
index 000000000..8ab3de4f1
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
@@ -0,0 +1,320 @@
+//===- DWARFYAML.h - DWARF YAMLIO implementation ----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file declares classes for handling the YAML representation
+/// of DWARF Debug Info.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECTYAML_DWARFYAML_H
+#define LLVM_OBJECTYAML_DWARFYAML_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/YAMLTraits.h"
+#include <cstdint>
+#include <vector>
+
+namespace llvm {
+namespace DWARFYAML {
+
+struct InitialLength {
+ uint32_t TotalLength;
+ uint64_t TotalLength64;
+
+ bool isDWARF64() const { return TotalLength == UINT32_MAX; }
+
+ uint64_t getLength() const {
+ return isDWARF64() ? TotalLength64 : TotalLength;
+ }
+
+ void setLength(uint64_t Len) {
+ if (Len >= (uint64_t)UINT32_MAX) {
+ TotalLength64 = Len;
+ TotalLength = UINT32_MAX;
+ } else {
+ TotalLength = Len;
+ }
+ }
+};
+
+struct AttributeAbbrev {
+ llvm::dwarf::Attribute Attribute;
+ llvm::dwarf::Form Form;
+ llvm::yaml::Hex64 Value; // Some DWARF5 attributes have values
+};
+
+struct Abbrev {
+ llvm::yaml::Hex32 Code;
+ llvm::dwarf::Tag Tag;
+ llvm::dwarf::Constants Children;
+ std::vector<AttributeAbbrev> Attributes;
+};
+
+struct ARangeDescriptor {
+ llvm::yaml::Hex64 Address;
+ uint64_t Length;
+};
+
+struct ARange {
+ InitialLength Length;
+ uint16_t Version;
+ uint32_t CuOffset;
+ uint8_t AddrSize;
+ uint8_t SegSize;
+ std::vector<ARangeDescriptor> Descriptors;
+};
+
+// XXX BINARYEN <--
+struct Range {
+ uint64_t Start;
+ uint64_t End;
+ uint64_t SectionIndex; // XXX ?
+};
+// XXX BINARYEN -->
+
+struct PubEntry {
+ llvm::yaml::Hex32 DieOffset;
+ llvm::yaml::Hex8 Descriptor;
+ StringRef Name;
+};
+
+struct PubSection {
+ InitialLength Length;
+ uint16_t Version;
+ uint32_t UnitOffset;
+ uint32_t UnitSize;
+ bool IsGNUStyle = false;
+ std::vector<PubEntry> Entries;
+};
+
+struct FormValue {
+ llvm::yaml::Hex64 Value;
+ StringRef CStr;
+ std::vector<llvm::yaml::Hex8> BlockData;
+};
+
+struct Entry {
+ llvm::yaml::Hex32 AbbrCode;
+ std::vector<FormValue> Values;
+};
+
+struct Unit {
+ InitialLength Length;
+ uint16_t Version;
+ llvm::dwarf::UnitType Type; // Added in DWARF 5
+ uint32_t AbbrOffset;
+ uint8_t AddrSize;
+ std::vector<Entry> Entries;
+};
+
+struct File {
+ StringRef Name;
+ uint64_t DirIdx;
+ uint64_t ModTime;
+ uint64_t Length;
+};
+
+struct LineTableOpcode {
+ dwarf::LineNumberOps Opcode;
+ uint64_t ExtLen;
+ dwarf::LineNumberExtendedOps SubOpcode;
+ uint64_t Data;
+ int64_t SData;
+ File FileEntry;
+ std::vector<llvm::yaml::Hex8> UnknownOpcodeData;
+ std::vector<llvm::yaml::Hex64> StandardOpcodeData;
+};
+
+struct LineTable {
+ InitialLength Length;
+ uint16_t Version;
+ uint64_t PrologueLength;
+ uint8_t MinInstLength;
+ uint8_t MaxOpsPerInst;
+ uint8_t DefaultIsStmt;
+ uint8_t LineBase;
+ uint8_t LineRange;
+ uint8_t OpcodeBase;
+ std::vector<uint8_t> StandardOpcodeLengths;
+ std::vector<StringRef> IncludeDirs;
+ std::vector<File> Files;
+ std::vector<LineTableOpcode> Opcodes;
+};
+
+struct Data {
+ bool IsLittleEndian;
+ std::vector<Abbrev> AbbrevDecls;
+ std::vector<StringRef> DebugStrings;
+ std::vector<ARange> ARanges;
+ std::vector<Range> Ranges; // XXX BINARYEN
+ PubSection PubNames;
+ PubSection PubTypes;
+
+ PubSection GNUPubNames;
+ PubSection GNUPubTypes;
+
+ std::vector<Unit> CompileUnits;
+
+ std::vector<LineTable> DebugLines;
+
+ bool isEmpty() const;
+};
+
+} // end namespace DWARFYAML
+} // end namespace llvm
+
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::AttributeAbbrev)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::Abbrev)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::ARangeDescriptor)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::ARange)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::Range) // XXX BINARYEN
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::PubEntry)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::Unit)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::FormValue)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::Entry)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::File)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::LineTable)
+LLVM_YAML_IS_SEQUENCE_VECTOR(llvm::DWARFYAML::LineTableOpcode)
+
+namespace llvm {
+namespace yaml {
+
+template <> struct MappingTraits<DWARFYAML::Data> {
+ static void mapping(IO &IO, DWARFYAML::Data &DWARF);
+};
+
+template <> struct MappingTraits<DWARFYAML::Abbrev> {
+ static void mapping(IO &IO, DWARFYAML::Abbrev &Abbrev);
+};
+
+template <> struct MappingTraits<DWARFYAML::AttributeAbbrev> {
+ static void mapping(IO &IO, DWARFYAML::AttributeAbbrev &AttAbbrev);
+};
+
+template <> struct MappingTraits<DWARFYAML::ARangeDescriptor> {
+ static void mapping(IO &IO, DWARFYAML::ARangeDescriptor &Descriptor);
+};
+
+template <> struct MappingTraits<DWARFYAML::ARange> {
+ static void mapping(IO &IO, DWARFYAML::ARange &Range);
+};
+
+template <> struct MappingTraits<DWARFYAML::Range> { // XXX BINARYEN
+ static void mapping(IO &IO, DWARFYAML::Range &Range);
+};
+
+template <> struct MappingTraits<DWARFYAML::PubEntry> {
+ static void mapping(IO &IO, DWARFYAML::PubEntry &Entry);
+};
+
+template <> struct MappingTraits<DWARFYAML::PubSection> {
+ static void mapping(IO &IO, DWARFYAML::PubSection &Section);
+};
+
+template <> struct MappingTraits<DWARFYAML::Unit> {
+ static void mapping(IO &IO, DWARFYAML::Unit &Unit);
+};
+
+template <> struct MappingTraits<DWARFYAML::Entry> {
+ static void mapping(IO &IO, DWARFYAML::Entry &Entry);
+};
+
+template <> struct MappingTraits<DWARFYAML::FormValue> {
+ static void mapping(IO &IO, DWARFYAML::FormValue &FormValue);
+};
+
+template <> struct MappingTraits<DWARFYAML::File> {
+ static void mapping(IO &IO, DWARFYAML::File &File);
+};
+
+template <> struct MappingTraits<DWARFYAML::LineTableOpcode> {
+ static void mapping(IO &IO, DWARFYAML::LineTableOpcode &LineTableOpcode);
+};
+
+template <> struct MappingTraits<DWARFYAML::LineTable> {
+ static void mapping(IO &IO, DWARFYAML::LineTable &LineTable);
+};
+
+template <> struct MappingTraits<DWARFYAML::InitialLength> {
+ static void mapping(IO &IO, DWARFYAML::InitialLength &DWARF);
+};
+
+#define HANDLE_DW_TAG(unused, name, unused2, unused3, unused4) \
+ io.enumCase(value, "DW_TAG_" #name, dwarf::DW_TAG_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::Tag> {
+ static void enumeration(IO &io, dwarf::Tag &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex16>(value);
+ }
+};
+
+#define HANDLE_DW_LNS(unused, name) \
+ io.enumCase(value, "DW_LNS_" #name, dwarf::DW_LNS_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::LineNumberOps> {
+ static void enumeration(IO &io, dwarf::LineNumberOps &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex8>(value);
+ }
+};
+
+#define HANDLE_DW_LNE(unused, name) \
+ io.enumCase(value, "DW_LNE_" #name, dwarf::DW_LNE_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::LineNumberExtendedOps> {
+ static void enumeration(IO &io, dwarf::LineNumberExtendedOps &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex16>(value);
+ }
+};
+
+#define HANDLE_DW_AT(unused, name, unused2, unused3) \
+ io.enumCase(value, "DW_AT_" #name, dwarf::DW_AT_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::Attribute> {
+ static void enumeration(IO &io, dwarf::Attribute &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex16>(value);
+ }
+};
+
+#define HANDLE_DW_FORM(unused, name, unused2, unused3) \
+ io.enumCase(value, "DW_FORM_" #name, dwarf::DW_FORM_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::Form> {
+ static void enumeration(IO &io, dwarf::Form &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex16>(value);
+ }
+};
+
+#define HANDLE_DW_UT(unused, name) \
+ io.enumCase(value, "DW_UT_" #name, dwarf::DW_UT_##name);
+
+template <> struct ScalarEnumerationTraits<dwarf::UnitType> {
+ static void enumeration(IO &io, dwarf::UnitType &value) {
+#include "llvm/BinaryFormat/Dwarf.def"
+ io.enumFallback<Hex8>(value);
+ }
+};
+
+template <> struct ScalarEnumerationTraits<dwarf::Constants> {
+ static void enumeration(IO &io, dwarf::Constants &value) {
+ io.enumCase(value, "DW_CHILDREN_no", dwarf::DW_CHILDREN_no);
+ io.enumCase(value, "DW_CHILDREN_yes", dwarf::DW_CHILDREN_yes);
+ io.enumFallback<Hex16>(value);
+ }
+};
+
+} // end namespace yaml
+} // end namespace llvm
+
+#endif // LLVM_OBJECTYAML_DWARFYAML_H
diff --git a/third_party/llvm-project/include/llvm/ObjectYAML/ObjectYAML.h b/third_party/llvm-project/include/llvm/ObjectYAML/ObjectYAML.h
new file mode 100644
index 000000000..0015fd3dc
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ObjectYAML/ObjectYAML.h
@@ -0,0 +1,41 @@
+//===- ObjectYAML.h ---------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECTYAML_OBJECTYAML_H
+#define LLVM_OBJECTYAML_OBJECTYAML_H
+
+#include "llvm/ObjectYAML/COFFYAML.h"
+#include "llvm/ObjectYAML/ELFYAML.h"
+#include "llvm/ObjectYAML/MachOYAML.h"
+#include "llvm/ObjectYAML/MinidumpYAML.h"
+#include "llvm/ObjectYAML/WasmYAML.h"
+#include "llvm/Support/YAMLTraits.h"
+#include <memory>
+
+namespace llvm {
+namespace yaml {
+
+class IO;
+
+struct YamlObjectFile {
+ std::unique_ptr<ELFYAML::Object> Elf;
+ std::unique_ptr<COFFYAML::Object> Coff;
+ std::unique_ptr<MachOYAML::Object> MachO;
+ std::unique_ptr<MachOYAML::UniversalBinary> FatMachO;
+ std::unique_ptr<MinidumpYAML::Object> Minidump;
+ std::unique_ptr<WasmYAML::Object> Wasm;
+};
+
+template <> struct MappingTraits<YamlObjectFile> {
+ static void mapping(IO &IO, YamlObjectFile &ObjectFile);
+};
+
+} // end namespace yaml
+} // end namespace llvm
+
+#endif // LLVM_OBJECTYAML_OBJECTYAML_H
diff --git a/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.def b/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.def
new file mode 100644
index 000000000..7e8ba9122
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.def
@@ -0,0 +1,151 @@
+//===- AARCH64TargetParser.def - AARCH64 target parsing defines ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides defines to build up the AARCH64 target parser's logic.
+//
+//===----------------------------------------------------------------------===//
+
+// NOTE: NO INCLUDE GUARD DESIRED!
+
+#ifndef AARCH64_ARCH
+#define AARCH64_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, ARCH_BASE_EXT)
+#endif
+AARCH64_ARCH("invalid", INVALID, "", "",
+ ARMBuildAttrs::CPUArch::v8_A, FK_NONE, AArch64::AEK_NONE)
+AARCH64_ARCH("armv8-a", ARMV8A, "8-A", "v8", ARMBuildAttrs::CPUArch::v8_A,
+ FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRYPTO | AArch64::AEK_FP | AArch64::AEK_SIMD))
+AARCH64_ARCH("armv8.1-a", ARMV8_1A, "8.1-A", "v8.1a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRC | AArch64::AEK_CRYPTO | AArch64::AEK_FP |
+ AArch64::AEK_SIMD | AArch64::AEK_LSE | AArch64::AEK_RDM))
+AARCH64_ARCH("armv8.2-a", ARMV8_2A, "8.2-A", "v8.2a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRC | AArch64::AEK_CRYPTO | AArch64::AEK_FP |
+ AArch64::AEK_SIMD | AArch64::AEK_RAS | AArch64::AEK_LSE |
+ AArch64::AEK_RDM))
+AARCH64_ARCH("armv8.3-a", ARMV8_3A, "8.3-A", "v8.3a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRC | AArch64::AEK_CRYPTO | AArch64::AEK_FP |
+ AArch64::AEK_SIMD | AArch64::AEK_RAS | AArch64::AEK_LSE |
+ AArch64::AEK_RDM | AArch64::AEK_RCPC))
+AARCH64_ARCH("armv8.4-a", ARMV8_4A, "8.4-A", "v8.4a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRC | AArch64::AEK_CRYPTO | AArch64::AEK_FP |
+ AArch64::AEK_SIMD | AArch64::AEK_RAS | AArch64::AEK_LSE |
+ AArch64::AEK_RDM | AArch64::AEK_RCPC | AArch64::AEK_DOTPROD))
+AARCH64_ARCH("armv8.5-a", ARMV8_5A, "8.5-A", "v8.5a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (AArch64::AEK_CRC | AArch64::AEK_CRYPTO | AArch64::AEK_FP |
+ AArch64::AEK_SIMD | AArch64::AEK_RAS | AArch64::AEK_LSE |
+ AArch64::AEK_RDM | AArch64::AEK_RCPC | AArch64::AEK_DOTPROD))
+#undef AARCH64_ARCH
+
+#ifndef AARCH64_ARCH_EXT_NAME
+#define AARCH64_ARCH_EXT_NAME(NAME, ID, FEATURE, NEGFEATURE)
+#endif
+// FIXME: This would be nicer were it tablegen
+AARCH64_ARCH_EXT_NAME("invalid", AArch64::AEK_INVALID, nullptr, nullptr)
+AARCH64_ARCH_EXT_NAME("none", AArch64::AEK_NONE, nullptr, nullptr)
+AARCH64_ARCH_EXT_NAME("crc", AArch64::AEK_CRC, "+crc", "-crc")
+AARCH64_ARCH_EXT_NAME("lse", AArch64::AEK_LSE, "+lse", "-lse")
+AARCH64_ARCH_EXT_NAME("rdm", AArch64::AEK_RDM, "+rdm", "-rdm")
+AARCH64_ARCH_EXT_NAME("crypto", AArch64::AEK_CRYPTO, "+crypto","-crypto")
+AARCH64_ARCH_EXT_NAME("sm4", AArch64::AEK_SM4, "+sm4", "-sm4")
+AARCH64_ARCH_EXT_NAME("sha3", AArch64::AEK_SHA3, "+sha3", "-sha3")
+AARCH64_ARCH_EXT_NAME("sha2", AArch64::AEK_SHA2, "+sha2", "-sha2")
+AARCH64_ARCH_EXT_NAME("aes", AArch64::AEK_AES, "+aes", "-aes")
+AARCH64_ARCH_EXT_NAME("dotprod", AArch64::AEK_DOTPROD, "+dotprod","-dotprod")
+AARCH64_ARCH_EXT_NAME("fp", AArch64::AEK_FP, "+fp-armv8", "-fp-armv8")
+AARCH64_ARCH_EXT_NAME("simd", AArch64::AEK_SIMD, "+neon", "-neon")
+AARCH64_ARCH_EXT_NAME("fp16", AArch64::AEK_FP16, "+fullfp16", "-fullfp16")
+AARCH64_ARCH_EXT_NAME("fp16fml", AArch64::AEK_FP16FML, "+fp16fml", "-fp16fml")
+AARCH64_ARCH_EXT_NAME("profile", AArch64::AEK_PROFILE, "+spe", "-spe")
+AARCH64_ARCH_EXT_NAME("ras", AArch64::AEK_RAS, "+ras", "-ras")
+AARCH64_ARCH_EXT_NAME("sve", AArch64::AEK_SVE, "+sve", "-sve")
+AARCH64_ARCH_EXT_NAME("sve2", AArch64::AEK_SVE2, "+sve2", "-sve2")
+AARCH64_ARCH_EXT_NAME("sve2-aes", AArch64::AEK_SVE2AES, "+sve2-aes", "-sve2-aes")
+AARCH64_ARCH_EXT_NAME("sve2-sm4", AArch64::AEK_SVE2SM4, "+sve2-sm4", "-sve2-sm4")
+AARCH64_ARCH_EXT_NAME("sve2-sha3", AArch64::AEK_SVE2SHA3, "+sve2-sha3", "-sve2-sha3")
+AARCH64_ARCH_EXT_NAME("sve2-bitperm", AArch64::AEK_SVE2BITPERM, "+sve2-bitperm", "-sve2-bitperm")
+AARCH64_ARCH_EXT_NAME("rcpc", AArch64::AEK_RCPC, "+rcpc", "-rcpc")
+AARCH64_ARCH_EXT_NAME("rng", AArch64::AEK_RAND, "+rand", "-rand")
+AARCH64_ARCH_EXT_NAME("memtag", AArch64::AEK_MTE, "+mte", "-mte")
+AARCH64_ARCH_EXT_NAME("ssbs", AArch64::AEK_SSBS, "+ssbs", "-ssbs")
+AARCH64_ARCH_EXT_NAME("sb", AArch64::AEK_SB, "+sb", "-sb")
+AARCH64_ARCH_EXT_NAME("predres", AArch64::AEK_PREDRES, "+predres", "-predres")
+AARCH64_ARCH_EXT_NAME("tme", AArch64::AEK_TME, "+tme", "-tme")
+#undef AARCH64_ARCH_EXT_NAME
+
+#ifndef AARCH64_CPU_NAME
+#define AARCH64_CPU_NAME(NAME, ID, DEFAULT_FPU, IS_DEFAULT, DEFAULT_EXT)
+#endif
+AARCH64_CPU_NAME("cortex-a35", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("cortex-a53", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, true,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("cortex-a55", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_FP16 | AArch64::AEK_DOTPROD | AArch64::AEK_RCPC))
+AARCH64_CPU_NAME("cortex-a57", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("cortex-a65", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16 | AArch64::AEK_RAS |
+ AArch64::AEK_RCPC | AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("cortex-a65ae", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16 | AArch64::AEK_RAS |
+ AArch64::AEK_RCPC | AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("cortex-a72", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("cortex-a73", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("cortex-a75", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_FP16 | AArch64::AEK_DOTPROD | AArch64::AEK_RCPC))
+AARCH64_CPU_NAME("cortex-a76", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_FP16 | AArch64::AEK_DOTPROD | AArch64::AEK_RCPC |
+ AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("cortex-a76ae", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_FP16 | AArch64::AEK_DOTPROD | AArch64::AEK_RCPC |
+ AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("neoverse-e1", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16 | AArch64::AEK_RAS |
+ AArch64::AEK_RCPC | AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("neoverse-n1", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16 |
+ AArch64::AEK_PROFILE | AArch64::AEK_RAS | AArch64::AEK_RCPC |
+ AArch64::AEK_SSBS))
+AARCH64_CPU_NAME("cyclone", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_NONE))
+AARCH64_CPU_NAME("exynos-m3", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("exynos-m4", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16))
+AARCH64_CPU_NAME("exynos-m5", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD | AArch64::AEK_FP16))
+AARCH64_CPU_NAME("falkor", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC | AArch64::AEK_RDM))
+AARCH64_CPU_NAME("saphira", ARMV8_3A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_PROFILE))
+AARCH64_CPU_NAME("kryo", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC))
+AARCH64_CPU_NAME("thunderx2t99", ARMV8_1A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_NONE))
+AARCH64_CPU_NAME("thunderx", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC | AArch64::AEK_PROFILE))
+AARCH64_CPU_NAME("thunderxt88", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC | AArch64::AEK_PROFILE))
+AARCH64_CPU_NAME("thunderxt81", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC | AArch64::AEK_PROFILE))
+AARCH64_CPU_NAME("thunderxt83", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_CRC | AArch64::AEK_PROFILE))
+AARCH64_CPU_NAME("tsv110", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (AArch64::AEK_DOTPROD |
+ AArch64::AEK_FP16 | AArch64::AEK_FP16FML |
+ AArch64::AEK_PROFILE))
+// Invalid CPU
+AARCH64_CPU_NAME("invalid", INVALID, FK_INVALID, true, AArch64::AEK_INVALID)
+#undef AARCH64_CPU_NAME
diff --git a/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.h b/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.h
new file mode 100644
index 000000000..94f341c83
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/AArch64TargetParser.h
@@ -0,0 +1,129 @@
+//===-- AArch64TargetParser - Parser for AArch64 features -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a target parser to recognise AArch64 hardware features
+// such as FPU/CPU/ARCH and extension names.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_AARCH64TARGETPARSERCOMMON_H
+#define LLVM_SUPPORT_AARCH64TARGETPARSERCOMMON_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/Support/ARMTargetParser.h"
+#include <vector>
+
+// FIXME:This should be made into class design,to avoid dupplication.
+namespace llvm {
+namespace AArch64 {
+
+// Arch extension modifiers for CPUs.
+enum ArchExtKind : unsigned {
+ AEK_INVALID = 0,
+ AEK_NONE = 1,
+ AEK_CRC = 1 << 1,
+ AEK_CRYPTO = 1 << 2,
+ AEK_FP = 1 << 3,
+ AEK_SIMD = 1 << 4,
+ AEK_FP16 = 1 << 5,
+ AEK_PROFILE = 1 << 6,
+ AEK_RAS = 1 << 7,
+ AEK_LSE = 1 << 8,
+ AEK_SVE = 1 << 9,
+ AEK_DOTPROD = 1 << 10,
+ AEK_RCPC = 1 << 11,
+ AEK_RDM = 1 << 12,
+ AEK_SM4 = 1 << 13,
+ AEK_SHA3 = 1 << 14,
+ AEK_SHA2 = 1 << 15,
+ AEK_AES = 1 << 16,
+ AEK_FP16FML = 1 << 17,
+ AEK_RAND = 1 << 18,
+ AEK_MTE = 1 << 19,
+ AEK_SSBS = 1 << 20,
+ AEK_SB = 1 << 21,
+ AEK_PREDRES = 1 << 22,
+ AEK_SVE2 = 1 << 23,
+ AEK_SVE2AES = 1 << 24,
+ AEK_SVE2SM4 = 1 << 25,
+ AEK_SVE2SHA3 = 1 << 26,
+ AEK_SVE2BITPERM = 1 << 27,
+ AEK_TME = 1 << 28,
+};
+
+enum class ArchKind {
+#define AARCH64_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, ARCH_BASE_EXT) ID,
+#include "AArch64TargetParser.def"
+};
+
+const ARM::ArchNames<ArchKind> AArch64ARCHNames[] = {
+#define AARCH64_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, \
+ ARCH_BASE_EXT) \
+ {NAME, \
+ sizeof(NAME) - 1, \
+ CPU_ATTR, \
+ sizeof(CPU_ATTR) - 1, \
+ SUB_ARCH, \
+ sizeof(SUB_ARCH) - 1, \
+ ARM::FPUKind::ARCH_FPU, \
+ ARCH_BASE_EXT, \
+ AArch64::ArchKind::ID, \
+ ARCH_ATTR},
+#include "AArch64TargetParser.def"
+};
+
+const ARM::ExtName AArch64ARCHExtNames[] = {
+#define AARCH64_ARCH_EXT_NAME(NAME, ID, FEATURE, NEGFEATURE) \
+ {NAME, sizeof(NAME) - 1, ID, FEATURE, NEGFEATURE},
+#include "AArch64TargetParser.def"
+};
+
+const ARM::CpuNames<ArchKind> AArch64CPUNames[] = {
+#define AARCH64_CPU_NAME(NAME, ID, DEFAULT_FPU, IS_DEFAULT, DEFAULT_EXT) \
+ {NAME, sizeof(NAME) - 1, AArch64::ArchKind::ID, IS_DEFAULT, DEFAULT_EXT},
+#include "AArch64TargetParser.def"
+};
+
+const ArchKind ArchKinds[] = {
+#define AARCH64_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, ARCH_BASE_EXT) \
+ ArchKind::ID,
+#include "AArch64TargetParser.def"
+};
+
+// FIXME: These should be moved to TargetTuple once it exists
+bool getExtensionFeatures(unsigned Extensions,
+ std::vector<StringRef> &Features);
+bool getArchFeatures(ArchKind AK, std::vector<StringRef> &Features);
+
+StringRef getArchName(ArchKind AK);
+unsigned getArchAttr(ArchKind AK);
+StringRef getCPUAttr(ArchKind AK);
+StringRef getSubArch(ArchKind AK);
+StringRef getArchExtName(unsigned ArchExtKind);
+StringRef getArchExtFeature(StringRef ArchExt);
+
+// Information by Name
+unsigned getDefaultFPU(StringRef CPU, ArchKind AK);
+unsigned getDefaultExtensions(StringRef CPU, ArchKind AK);
+StringRef getDefaultCPU(StringRef Arch);
+ArchKind getCPUArchKind(StringRef CPU);
+
+// Parser
+ArchKind parseArch(StringRef Arch);
+ArchExtKind parseArchExt(StringRef ArchExt);
+ArchKind parseCPUArch(StringRef CPU);
+// Used by target parser tests
+void fillValidCPUArchList(SmallVectorImpl<StringRef> &Values);
+
+bool isX18ReservedByDefault(const Triple &TT);
+
+} // namespace AArch64
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/ARMAttributeParser.h b/third_party/llvm-project/include/llvm/Support/ARMAttributeParser.h
new file mode 100644
index 000000000..f6c39abb4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ARMAttributeParser.h
@@ -0,0 +1,141 @@
+//===--- ARMAttributeParser.h - ARM Attribute Information Printer ---------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ARMATTRIBUTEPARSER_H
+#define LLVM_SUPPORT_ARMATTRIBUTEPARSER_H
+
+#include "ARMBuildAttributes.h"
+#include "ScopedPrinter.h"
+
+#include <map>
+
+namespace llvm {
+class StringRef;
+
+class ARMAttributeParser {
+ ScopedPrinter *SW;
+
+ std::map<unsigned, unsigned> Attributes;
+
+ struct DisplayHandler {
+ ARMBuildAttrs::AttrType Attribute;
+ void (ARMAttributeParser::*Routine)(ARMBuildAttrs::AttrType,
+ const uint8_t *, uint32_t &);
+ };
+ static const DisplayHandler DisplayRoutines[];
+
+ uint64_t ParseInteger(const uint8_t *Data, uint32_t &Offset);
+ StringRef ParseString(const uint8_t *Data, uint32_t &Offset);
+
+ void IntegerAttribute(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void StringAttribute(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+
+ void PrintAttribute(unsigned Tag, unsigned Value, StringRef ValueDesc);
+
+ void CPU_arch(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void CPU_arch_profile(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ARM_ISA_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void THUMB_ISA_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void FP_arch(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void WMMX_arch(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void Advanced_SIMD_arch(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void MVE_arch(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void PCS_config(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_PCS_R9_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_PCS_RW_data(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_PCS_RO_data(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_PCS_GOT_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_PCS_wchar_t(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_rounding(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_denormal(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_exceptions(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_user_exceptions(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_number_model(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_align_needed(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_align_preserved(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_enum_size(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_HardFP_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_VFP_args(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_WMMX_args(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_optimization_goals(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_optimization_goals(ARMBuildAttrs::AttrType Tag,
+ const uint8_t *Data, uint32_t &Offset);
+ void compatibility(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void CPU_unaligned_access(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void FP_HP_extension(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void ABI_FP_16bit_format(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void MPextension_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void DIV_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void DSP_extension(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void T2EE_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void Virtualization_use(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+ void nodefaults(ARMBuildAttrs::AttrType Tag, const uint8_t *Data,
+ uint32_t &Offset);
+
+ void ParseAttributeList(const uint8_t *Data, uint32_t &Offset,
+ uint32_t Length);
+ void ParseIndexList(const uint8_t *Data, uint32_t &Offset,
+ SmallVectorImpl<uint8_t> &IndexList);
+ void ParseSubsection(const uint8_t *Data, uint32_t Length);
+public:
+ ARMAttributeParser(ScopedPrinter *SW) : SW(SW) {}
+
+ ARMAttributeParser() : SW(nullptr) { }
+
+ void Parse(ArrayRef<uint8_t> Section, bool isLittle);
+
+ bool hasAttribute(unsigned Tag) const {
+ return Attributes.count(Tag);
+ }
+
+ unsigned getAttributeValue(unsigned Tag) const {
+ return Attributes.find(Tag)->second;
+ }
+};
+
+}
+
+#endif
+
diff --git a/third_party/llvm-project/include/llvm/Support/ARMBuildAttributes.h b/third_party/llvm-project/include/llvm/Support/ARMBuildAttributes.h
new file mode 100644
index 000000000..90481eaa1
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ARMBuildAttributes.h
@@ -0,0 +1,253 @@
+//===-- ARMBuildAttributes.h - ARM Build Attributes -------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains enumerations and support routines for ARM build attributes
+// as defined in ARM ABI addenda document (ABI release 2.08).
+//
+// ELF for the ARM Architecture r2.09 - November 30, 2012
+//
+// http://infocenter.arm.com/help/topic/com.arm.doc.ihi0044e/IHI0044E_aaelf.pdf
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ARMBUILDATTRIBUTES_H
+#define LLVM_SUPPORT_ARMBUILDATTRIBUTES_H
+
+namespace llvm {
+class StringRef;
+
+namespace ARMBuildAttrs {
+
+enum SpecialAttr {
+ // This is for the .cpu asm attr. It translates into one or more
+ // AttrType (below) entries in the .ARM.attributes section in the ELF.
+ SEL_CPU
+};
+
+enum AttrType {
+ // Rest correspond to ELF/.ARM.attributes
+ File = 1,
+ CPU_raw_name = 4,
+ CPU_name = 5,
+ CPU_arch = 6,
+ CPU_arch_profile = 7,
+ ARM_ISA_use = 8,
+ THUMB_ISA_use = 9,
+ FP_arch = 10,
+ WMMX_arch = 11,
+ Advanced_SIMD_arch = 12,
+ PCS_config = 13,
+ ABI_PCS_R9_use = 14,
+ ABI_PCS_RW_data = 15,
+ ABI_PCS_RO_data = 16,
+ ABI_PCS_GOT_use = 17,
+ ABI_PCS_wchar_t = 18,
+ ABI_FP_rounding = 19,
+ ABI_FP_denormal = 20,
+ ABI_FP_exceptions = 21,
+ ABI_FP_user_exceptions = 22,
+ ABI_FP_number_model = 23,
+ ABI_align_needed = 24,
+ ABI_align_preserved = 25,
+ ABI_enum_size = 26,
+ ABI_HardFP_use = 27,
+ ABI_VFP_args = 28,
+ ABI_WMMX_args = 29,
+ ABI_optimization_goals = 30,
+ ABI_FP_optimization_goals = 31,
+ compatibility = 32,
+ CPU_unaligned_access = 34,
+ FP_HP_extension = 36,
+ ABI_FP_16bit_format = 38,
+ MPextension_use = 42, // recoded from 70 (ABI r2.08)
+ DIV_use = 44,
+ DSP_extension = 46,
+ MVE_arch = 48,
+ also_compatible_with = 65,
+ conformance = 67,
+ Virtualization_use = 68,
+
+ /// Legacy Tags
+ Section = 2, // deprecated (ABI r2.09)
+ Symbol = 3, // deprecated (ABI r2.09)
+ ABI_align8_needed = 24, // renamed to ABI_align_needed (ABI r2.09)
+ ABI_align8_preserved = 25, // renamed to ABI_align_preserved (ABI r2.09)
+ nodefaults = 64, // deprecated (ABI r2.09)
+ T2EE_use = 66, // deprecated (ABI r2.09)
+ MPextension_use_old = 70 // recoded to MPextension_use (ABI r2.08)
+};
+
+StringRef AttrTypeAsString(unsigned Attr, bool HasTagPrefix = true);
+StringRef AttrTypeAsString(AttrType Attr, bool HasTagPrefix = true);
+int AttrTypeFromString(StringRef Tag);
+
+// Magic numbers for .ARM.attributes
+enum AttrMagic {
+ Format_Version = 0x41
+};
+
+// Legal Values for CPU_arch, (=6), uleb128
+enum CPUArch {
+ Pre_v4 = 0,
+ v4 = 1, // e.g. SA110
+ v4T = 2, // e.g. ARM7TDMI
+ v5T = 3, // e.g. ARM9TDMI
+ v5TE = 4, // e.g. ARM946E_S
+ v5TEJ = 5, // e.g. ARM926EJ_S
+ v6 = 6, // e.g. ARM1136J_S
+ v6KZ = 7, // e.g. ARM1176JZ_S
+ v6T2 = 8, // e.g. ARM1156T2_S
+ v6K = 9, // e.g. ARM1176JZ_S
+ v7 = 10, // e.g. Cortex A8, Cortex M3
+ v6_M = 11, // e.g. Cortex M1
+ v6S_M = 12, // v6_M with the System extensions
+ v7E_M = 13, // v7_M with DSP extensions
+ v8_A = 14, // v8_A AArch32
+ v8_R = 15, // e.g. Cortex R52
+ v8_M_Base= 16, // v8_M_Base AArch32
+ v8_M_Main= 17, // v8_M_Main AArch32
+ v8_1_M_Main=21, // v8_1_M_Main AArch32
+};
+
+enum CPUArchProfile { // (=7), uleb128
+ Not_Applicable = 0, // pre v7, or cross-profile code
+ ApplicationProfile = (0x41), // 'A' (e.g. for Cortex A8)
+ RealTimeProfile = (0x52), // 'R' (e.g. for Cortex R4)
+ MicroControllerProfile = (0x4D), // 'M' (e.g. for Cortex M3)
+ SystemProfile = (0x53) // 'S' Application or real-time profile
+};
+
+// The following have a lot of common use cases
+enum {
+ Not_Allowed = 0,
+ Allowed = 1,
+
+ // Tag_ARM_ISA_use (=8), uleb128
+
+ // Tag_THUMB_ISA_use, (=9), uleb128
+ AllowThumb32 = 2, // 32-bit Thumb (implies 16-bit instructions)
+ AllowThumbDerived = 3, // Thumb allowed, derived from arch/profile
+
+ // Tag_FP_arch (=10), uleb128 (formerly Tag_VFP_arch = 10)
+ AllowFPv2 = 2, // v2 FP ISA permitted (implies use of the v1 FP ISA)
+ AllowFPv3A = 3, // v3 FP ISA permitted (implies use of the v2 FP ISA)
+ AllowFPv3B = 4, // v3 FP ISA permitted, but only D0-D15, S0-S31
+ AllowFPv4A = 5, // v4 FP ISA permitted (implies use of v3 FP ISA)
+ AllowFPv4B = 6, // v4 FP ISA was permitted, but only D0-D15, S0-S31
+ AllowFPARMv8A = 7, // Use of the ARM v8-A FP ISA was permitted
+ AllowFPARMv8B = 8, // Use of the ARM v8-A FP ISA was permitted, but only
+ // D0-D15, S0-S31
+
+ // Tag_WMMX_arch, (=11), uleb128
+ AllowWMMXv1 = 1, // The user permitted this entity to use WMMX v1
+ AllowWMMXv2 = 2, // The user permitted this entity to use WMMX v2
+
+ // Tag_Advanced_SIMD_arch, (=12), uleb128
+ AllowNeon = 1, // SIMDv1 was permitted
+ AllowNeon2 = 2, // SIMDv2 was permitted (Half-precision FP, MAC operations)
+ AllowNeonARMv8 = 3, // ARM v8-A SIMD was permitted
+ AllowNeonARMv8_1a = 4,// ARM v8.1-A SIMD was permitted (RDMA)
+
+ // Tag_MVE_arch, (=48), uleb128
+ AllowMVEInteger = 1, // integer-only MVE was permitted
+ AllowMVEIntegerAndFloat = 2, // both integer and floating point MVE were permitted
+
+ // Tag_ABI_PCS_R9_use, (=14), uleb128
+ R9IsGPR = 0, // R9 used as v6 (just another callee-saved register)
+ R9IsSB = 1, // R9 used as a global static base rgister
+ R9IsTLSPointer = 2, // R9 used as a thread local storage pointer
+ R9Reserved = 3, // R9 not used by code associated with attributed entity
+
+ // Tag_ABI_PCS_RW_data, (=15), uleb128
+ AddressRWPCRel = 1, // Address RW static data PC-relative
+ AddressRWSBRel = 2, // Address RW static data SB-relative
+ AddressRWNone = 3, // No RW static data permitted
+
+ // Tag_ABI_PCS_RO_data, (=14), uleb128
+ AddressROPCRel = 1, // Address RO static data PC-relative
+ AddressRONone = 2, // No RO static data permitted
+
+ // Tag_ABI_PCS_GOT_use, (=17), uleb128
+ AddressDirect = 1, // Address imported data directly
+ AddressGOT = 2, // Address imported data indirectly (via GOT)
+
+ // Tag_ABI_PCS_wchar_t, (=18), uleb128
+ WCharProhibited = 0, // wchar_t is not used
+ WCharWidth2Bytes = 2, // sizeof(wchar_t) == 2
+ WCharWidth4Bytes = 4, // sizeof(wchar_t) == 4
+
+ // Tag_ABI_align_needed, (=24), uleb128
+ Align8Byte = 1,
+ Align4Byte = 2,
+ AlignReserved = 3,
+
+ // Tag_ABI_align_needed, (=25), uleb128
+ AlignNotPreserved = 0,
+ AlignPreserve8Byte = 1,
+ AlignPreserveAll = 2,
+
+ // Tag_ABI_FP_denormal, (=20), uleb128
+ PositiveZero = 0,
+ IEEEDenormals = 1,
+ PreserveFPSign = 2, // sign when flushed-to-zero is preserved
+
+ // Tag_ABI_FP_number_model, (=23), uleb128
+ AllowIEEENormal = 1,
+ AllowRTABI = 2, // numbers, infinities, and one quiet NaN (see [RTABI])
+ AllowIEEE754 = 3, // this code to use all the IEEE 754-defined FP encodings
+
+ // Tag_ABI_enum_size, (=26), uleb128
+ EnumProhibited = 0, // The user prohibited the use of enums when building
+ // this entity.
+ EnumSmallest = 1, // Enum is smallest container big enough to hold all
+ // values.
+ Enum32Bit = 2, // Enum is at least 32 bits.
+ Enum32BitABI = 3, // Every enumeration visible across an ABI-complying
+ // interface contains a value needing 32 bits to encode
+ // it; other enums can be containerized.
+
+ // Tag_ABI_HardFP_use, (=27), uleb128
+ HardFPImplied = 0, // FP use should be implied by Tag_FP_arch
+ HardFPSinglePrecision = 1, // Single-precision only
+
+ // Tag_ABI_VFP_args, (=28), uleb128
+ BaseAAPCS = 0,
+ HardFPAAPCS = 1,
+ ToolChainFPPCS = 2,
+ CompatibleFPAAPCS = 3,
+
+ // Tag_FP_HP_extension, (=36), uleb128
+ AllowHPFP = 1, // Allow use of Half Precision FP
+
+ // Tag_FP_16bit_format, (=38), uleb128
+ FP16FormatIEEE = 1,
+ FP16VFP3 = 2,
+
+ // Tag_MPextension_use, (=42), uleb128
+ AllowMP = 1, // Allow use of MP extensions
+
+ // Tag_DIV_use, (=44), uleb128
+ // Note: AllowDIVExt must be emitted if and only if the permission to use
+ // hardware divide cannot be conveyed using AllowDIVIfExists or DisallowDIV
+ AllowDIVIfExists = 0, // Allow hardware divide if available in arch, or no
+ // info exists.
+ DisallowDIV = 1, // Hardware divide explicitly disallowed.
+ AllowDIVExt = 2, // Allow hardware divide as optional architecture
+ // extension above the base arch specified by
+ // Tag_CPU_arch and Tag_CPU_arch_profile.
+
+ // Tag_Virtualization_use, (=68), uleb128
+ AllowTZ = 1,
+ AllowVirtualization = 2,
+ AllowTZVirtualization = 3
+};
+
+} // namespace ARMBuildAttrs
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/ARMTargetParser.def b/third_party/llvm-project/include/llvm/Support/ARMTargetParser.def
new file mode 100644
index 000000000..7f03d9a13
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ARMTargetParser.def
@@ -0,0 +1,293 @@
+//===- ARMTargetParser.def - ARM target parsing defines ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides defines to build up the ARM target parser's logic.
+//
+//===----------------------------------------------------------------------===//
+
+// NOTE: NO INCLUDE GUARD DESIRED!
+
+#ifndef ARM_FPU
+#define ARM_FPU(NAME, KIND, VERSION, NEON_SUPPORT, RESTRICTION)
+#endif
+ARM_FPU("invalid", FK_INVALID, FPUVersion::NONE, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("none", FK_NONE, FPUVersion::NONE, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfp", FK_VFP, FPUVersion::VFPV2, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfpv2", FK_VFPV2, FPUVersion::VFPV2, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfpv3", FK_VFPV3, FPUVersion::VFPV3, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfpv3-fp16", FK_VFPV3_FP16, FPUVersion::VFPV3_FP16, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfpv3-d16", FK_VFPV3_D16, FPUVersion::VFPV3, NeonSupportLevel::None, FPURestriction::D16)
+ARM_FPU("vfpv3-d16-fp16", FK_VFPV3_D16_FP16, FPUVersion::VFPV3_FP16, NeonSupportLevel::None, FPURestriction::D16)
+ARM_FPU("vfpv3xd", FK_VFPV3XD, FPUVersion::VFPV3, NeonSupportLevel::None, FPURestriction::SP_D16)
+ARM_FPU("vfpv3xd-fp16", FK_VFPV3XD_FP16, FPUVersion::VFPV3_FP16, NeonSupportLevel::None, FPURestriction::SP_D16)
+ARM_FPU("vfpv4", FK_VFPV4, FPUVersion::VFPV4, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("vfpv4-d16", FK_VFPV4_D16, FPUVersion::VFPV4, NeonSupportLevel::None, FPURestriction::D16)
+ARM_FPU("fpv4-sp-d16", FK_FPV4_SP_D16, FPUVersion::VFPV4, NeonSupportLevel::None, FPURestriction::SP_D16)
+ARM_FPU("fpv5-d16", FK_FPV5_D16, FPUVersion::VFPV5, NeonSupportLevel::None, FPURestriction::D16)
+ARM_FPU("fpv5-sp-d16", FK_FPV5_SP_D16, FPUVersion::VFPV5, NeonSupportLevel::None, FPURestriction::SP_D16)
+ARM_FPU("fp-armv8", FK_FP_ARMV8, FPUVersion::VFPV5, NeonSupportLevel::None, FPURestriction::None)
+ARM_FPU("fp-armv8-fullfp16-d16", FK_FP_ARMV8_FULLFP16_D16, FPUVersion::VFPV5_FULLFP16, NeonSupportLevel::None, FPURestriction::D16)
+ARM_FPU("fp-armv8-fullfp16-sp-d16", FK_FP_ARMV8_FULLFP16_SP_D16, FPUVersion::VFPV5_FULLFP16, NeonSupportLevel::None, FPURestriction::SP_D16)
+ARM_FPU("neon", FK_NEON, FPUVersion::VFPV3, NeonSupportLevel::Neon, FPURestriction::None)
+ARM_FPU("neon-fp16", FK_NEON_FP16, FPUVersion::VFPV3_FP16, NeonSupportLevel::Neon, FPURestriction::None)
+ARM_FPU("neon-vfpv4", FK_NEON_VFPV4, FPUVersion::VFPV4, NeonSupportLevel::Neon, FPURestriction::None)
+ARM_FPU("neon-fp-armv8", FK_NEON_FP_ARMV8, FPUVersion::VFPV5, NeonSupportLevel::Neon, FPURestriction::None)
+ARM_FPU("crypto-neon-fp-armv8", FK_CRYPTO_NEON_FP_ARMV8, FPUVersion::VFPV5, NeonSupportLevel::Crypto,
+ FPURestriction::None)
+ARM_FPU("softvfp", FK_SOFTVFP, FPUVersion::NONE, NeonSupportLevel::None, FPURestriction::None)
+#undef ARM_FPU
+
+#ifndef ARM_ARCH
+#define ARM_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, ARCH_BASE_EXT)
+#endif
+ARM_ARCH("invalid", INVALID, "", "",
+ ARMBuildAttrs::CPUArch::Pre_v4, FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv2", ARMV2, "2", "v2", ARMBuildAttrs::CPUArch::Pre_v4,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv2a", ARMV2A, "2A", "v2a", ARMBuildAttrs::CPUArch::Pre_v4,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv3", ARMV3, "3", "v3", ARMBuildAttrs::CPUArch::Pre_v4,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv3m", ARMV3M, "3M", "v3m", ARMBuildAttrs::CPUArch::Pre_v4,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv4", ARMV4, "4", "v4", ARMBuildAttrs::CPUArch::v4,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv4t", ARMV4T, "4T", "v4t", ARMBuildAttrs::CPUArch::v4T,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv5t", ARMV5T, "5T", "v5", ARMBuildAttrs::CPUArch::v5T,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv5te", ARMV5TE, "5TE", "v5e", ARMBuildAttrs::CPUArch::v5TE,
+ FK_NONE, ARM::AEK_DSP)
+ARM_ARCH("armv5tej", ARMV5TEJ, "5TEJ", "v5e", ARMBuildAttrs::CPUArch::v5TEJ,
+ FK_NONE, ARM::AEK_DSP)
+ARM_ARCH("armv6", ARMV6, "6", "v6", ARMBuildAttrs::CPUArch::v6,
+ FK_VFPV2, ARM::AEK_DSP)
+ARM_ARCH("armv6k", ARMV6K, "6K", "v6k", ARMBuildAttrs::CPUArch::v6K,
+ FK_VFPV2, ARM::AEK_DSP)
+ARM_ARCH("armv6t2", ARMV6T2, "6T2", "v6t2", ARMBuildAttrs::CPUArch::v6T2,
+ FK_NONE, ARM::AEK_DSP)
+ARM_ARCH("armv6kz", ARMV6KZ, "6KZ", "v6kz", ARMBuildAttrs::CPUArch::v6KZ,
+ FK_VFPV2, (ARM::AEK_SEC | ARM::AEK_DSP))
+ARM_ARCH("armv6-m", ARMV6M, "6-M", "v6m", ARMBuildAttrs::CPUArch::v6_M,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv7-a", ARMV7A, "7-A", "v7", ARMBuildAttrs::CPUArch::v7,
+ FK_NEON, ARM::AEK_DSP)
+ARM_ARCH("armv7ve", ARMV7VE, "7VE", "v7ve", ARMBuildAttrs::CPUArch::v7,
+ FK_NEON, (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT |
+ ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP))
+ARM_ARCH("armv7-r", ARMV7R, "7-R", "v7r", ARMBuildAttrs::CPUArch::v7,
+ FK_NONE, (ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP))
+ARM_ARCH("armv7-m", ARMV7M, "7-M", "v7m", ARMBuildAttrs::CPUArch::v7,
+ FK_NONE, ARM::AEK_HWDIVTHUMB)
+ARM_ARCH("armv7e-m", ARMV7EM, "7E-M", "v7em", ARMBuildAttrs::CPUArch::v7E_M,
+ FK_NONE, (ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP))
+ARM_ARCH("armv8-a", ARMV8A, "8-A", "v8", ARMBuildAttrs::CPUArch::v8_A,
+ FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC))
+ARM_ARCH("armv8.1-a", ARMV8_1A, "8.1-A", "v8.1a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC))
+ARM_ARCH("armv8.2-a", ARMV8_2A, "8.2-A", "v8.2a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC | ARM::AEK_RAS))
+ARM_ARCH("armv8.3-a", ARMV8_3A, "8.3-A", "v8.3a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC | ARM::AEK_RAS))
+ARM_ARCH("armv8.4-a", ARMV8_4A, "8.4-A", "v8.4a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC | ARM::AEK_RAS |
+ ARM::AEK_DOTPROD))
+ARM_ARCH("armv8.5-a", ARMV8_5A, "8.5-A", "v8.5a",
+ ARMBuildAttrs::CPUArch::v8_A, FK_CRYPTO_NEON_FP_ARMV8,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB | ARM::AEK_DSP | ARM::AEK_CRC | ARM::AEK_RAS |
+ ARM::AEK_DOTPROD))
+ARM_ARCH("armv8-r", ARMV8R, "8-R", "v8r", ARMBuildAttrs::CPUArch::v8_R,
+ FK_NEON_FP_ARMV8,
+ (ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB |
+ ARM::AEK_DSP | ARM::AEK_CRC))
+ARM_ARCH("armv8-m.base", ARMV8MBaseline, "8-M.Baseline", "v8m.base",
+ ARMBuildAttrs::CPUArch::v8_M_Base, FK_NONE, ARM::AEK_HWDIVTHUMB)
+ARM_ARCH("armv8-m.main", ARMV8MMainline, "8-M.Mainline", "v8m.main",
+ ARMBuildAttrs::CPUArch::v8_M_Main, FK_FPV5_D16, ARM::AEK_HWDIVTHUMB)
+ARM_ARCH("armv8.1-m.main", ARMV8_1MMainline, "8.1-M.Mainline", "v8.1m.main",
+ ARMBuildAttrs::CPUArch::v8_1_M_Main, FK_FP_ARMV8_FULLFP16_SP_D16, ARM::AEK_HWDIVTHUMB | ARM::AEK_RAS | ARM::AEK_LOB)
+// Non-standard Arch names.
+ARM_ARCH("iwmmxt", IWMMXT, "iwmmxt", "", ARMBuildAttrs::CPUArch::v5TE,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("iwmmxt2", IWMMXT2, "iwmmxt2", "", ARMBuildAttrs::CPUArch::v5TE,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("xscale", XSCALE, "xscale", "v5e", ARMBuildAttrs::CPUArch::v5TE,
+ FK_NONE, ARM::AEK_NONE)
+ARM_ARCH("armv7s", ARMV7S, "7-S", "v7s", ARMBuildAttrs::CPUArch::v7,
+ FK_NEON_VFPV4, ARM::AEK_DSP)
+ARM_ARCH("armv7k", ARMV7K, "7-K", "v7k", ARMBuildAttrs::CPUArch::v7,
+ FK_NONE, ARM::AEK_DSP)
+#undef ARM_ARCH
+
+#ifndef ARM_ARCH_EXT_NAME
+#define ARM_ARCH_EXT_NAME(NAME, ID, FEATURE, NEGFEATURE)
+#endif
+// FIXME: This would be nicer were it tablegen
+ARM_ARCH_EXT_NAME("invalid", ARM::AEK_INVALID, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("none", ARM::AEK_NONE, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("crc", ARM::AEK_CRC, "+crc", "-crc")
+ARM_ARCH_EXT_NAME("crypto", ARM::AEK_CRYPTO, "+crypto","-crypto")
+ARM_ARCH_EXT_NAME("sha2", ARM::AEK_SHA2, "+sha2", "-sha2")
+ARM_ARCH_EXT_NAME("aes", ARM::AEK_AES, "+aes", "-aes")
+ARM_ARCH_EXT_NAME("dotprod", ARM::AEK_DOTPROD, "+dotprod","-dotprod")
+ARM_ARCH_EXT_NAME("dsp", ARM::AEK_DSP, "+dsp", "-dsp")
+ARM_ARCH_EXT_NAME("fp", ARM::AEK_FP, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("fp.dp", ARM::AEK_FP_DP, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("mve", (ARM::AEK_DSP | ARM::AEK_SIMD), "+mve", "-mve")
+ARM_ARCH_EXT_NAME("mve.fp", (ARM::AEK_DSP | ARM::AEK_SIMD | ARM::AEK_FP), "+mve.fp", "-mve.fp")
+ARM_ARCH_EXT_NAME("idiv", (ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB), nullptr, nullptr)
+ARM_ARCH_EXT_NAME("mp", ARM::AEK_MP, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("simd", ARM::AEK_SIMD, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("sec", ARM::AEK_SEC, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("virt", ARM::AEK_VIRT, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("fp16", ARM::AEK_FP16, "+fullfp16", "-fullfp16")
+ARM_ARCH_EXT_NAME("ras", ARM::AEK_RAS, "+ras", "-ras")
+ARM_ARCH_EXT_NAME("os", ARM::AEK_OS, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("iwmmxt", ARM::AEK_IWMMXT, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("iwmmxt2", ARM::AEK_IWMMXT2, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("maverick", ARM::AEK_MAVERICK, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("xscale", ARM::AEK_XSCALE, nullptr, nullptr)
+ARM_ARCH_EXT_NAME("fp16fml", ARM::AEK_FP16FML, "+fp16fml", "-fp16fml")
+ARM_ARCH_EXT_NAME("sb", ARM::AEK_SB, "+sb", "-sb")
+ARM_ARCH_EXT_NAME("lob", ARM::AEK_LOB, "+lob", "-lob")
+#undef ARM_ARCH_EXT_NAME
+
+#ifndef ARM_HW_DIV_NAME
+#define ARM_HW_DIV_NAME(NAME, ID)
+#endif
+ARM_HW_DIV_NAME("invalid", ARM::AEK_INVALID)
+ARM_HW_DIV_NAME("none", ARM::AEK_NONE)
+ARM_HW_DIV_NAME("thumb", ARM::AEK_HWDIVTHUMB)
+ARM_HW_DIV_NAME("arm", ARM::AEK_HWDIVARM)
+ARM_HW_DIV_NAME("arm,thumb", (ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB))
+#undef ARM_HW_DIV_NAME
+
+#ifndef ARM_CPU_NAME
+#define ARM_CPU_NAME(NAME, ID, DEFAULT_FPU, IS_DEFAULT, DEFAULT_EXT)
+#endif
+ARM_CPU_NAME("arm2", ARMV2, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm3", ARMV2A, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm6", ARMV3, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm7m", ARMV3M, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm8", ARMV4, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm810", ARMV4, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("strongarm", ARMV4, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("strongarm110", ARMV4, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("strongarm1100", ARMV4, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("strongarm1110", ARMV4, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm7tdmi", ARMV4T, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm7tdmi-s", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm710t", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm720t", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm9", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm9tdmi", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm920", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm920t", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm922t", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm9312", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm940t", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("ep9312", ARMV4T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm10tdmi", ARMV5T, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1020t", ARMV5T, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm9e", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm946e-s", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm966e-s", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm968e-s", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm10e", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1020e", ARMV5TE, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1022e", ARMV5TE, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm926ej-s", ARMV5TEJ, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1136j-s", ARMV6, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1136jf-s", ARMV6, FK_VFPV2, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1136jz-s", ARMV6, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("mpcore", ARMV6K, FK_VFPV2, true, ARM::AEK_NONE)
+ARM_CPU_NAME("mpcorenovfp", ARMV6K, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1176jz-s", ARMV6KZ, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1176jzf-s", ARMV6KZ, FK_VFPV2, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1156t2-s", ARMV6T2, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("arm1156t2f-s", ARMV6T2, FK_VFPV2, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m0", ARMV6M, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m0plus", ARMV6M, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m1", ARMV6M, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("sc000", ARMV6M, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-a5", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_SEC | ARM::AEK_MP))
+ARM_CPU_NAME("cortex-a7", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB))
+ARM_CPU_NAME("cortex-a8", ARMV7A, FK_NEON, false, ARM::AEK_SEC)
+ARM_CPU_NAME("cortex-a9", ARMV7A, FK_NEON_FP16, false, (ARM::AEK_SEC | ARM::AEK_MP))
+ARM_CPU_NAME("cortex-a12", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB))
+ARM_CPU_NAME("cortex-a15", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB))
+ARM_CPU_NAME("cortex-a17", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_SEC | ARM::AEK_MP | ARM::AEK_VIRT | ARM::AEK_HWDIVARM |
+ ARM::AEK_HWDIVTHUMB))
+ARM_CPU_NAME("krait", ARMV7A, FK_NEON_VFPV4, false,
+ (ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB))
+ARM_CPU_NAME("cortex-r4", ARMV7R, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-r4f", ARMV7R, FK_VFPV3_D16, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-r5", ARMV7R, FK_VFPV3_D16, false,
+ (ARM::AEK_MP | ARM::AEK_HWDIVARM))
+ARM_CPU_NAME("cortex-r7", ARMV7R, FK_VFPV3_D16_FP16, false,
+ (ARM::AEK_MP | ARM::AEK_HWDIVARM))
+ARM_CPU_NAME("cortex-r8", ARMV7R, FK_VFPV3_D16_FP16, false,
+ (ARM::AEK_MP | ARM::AEK_HWDIVARM))
+ARM_CPU_NAME("cortex-r52", ARMV8R, FK_NEON_FP_ARMV8, true, ARM::AEK_NONE)
+ARM_CPU_NAME("sc300", ARMV7M, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m3", ARMV7M, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m4", ARMV7EM, FK_FPV4_SP_D16, true, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m7", ARMV7EM, FK_FPV5_D16, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m23", ARMV8MBaseline, FK_NONE, false, ARM::AEK_NONE)
+ARM_CPU_NAME("cortex-m33", ARMV8MMainline, FK_FPV5_SP_D16, false, ARM::AEK_DSP)
+ARM_CPU_NAME("cortex-m35p", ARMV8MMainline, FK_FPV5_SP_D16, false, ARM::AEK_DSP)
+ARM_CPU_NAME("cortex-a32", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a35", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a53", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a55", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("cortex-a57", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a72", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a73", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("cortex-a75", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("cortex-a76", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("cortex-a76ae", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("neoverse-n1", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("cyclone", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("exynos-m3", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+ARM_CPU_NAME("exynos-m4", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("exynos-m5", ARMV8_2A, FK_CRYPTO_NEON_FP_ARMV8, false,
+ (ARM::AEK_FP16 | ARM::AEK_DOTPROD))
+ARM_CPU_NAME("kryo", ARMV8A, FK_CRYPTO_NEON_FP_ARMV8, false, ARM::AEK_CRC)
+// Non-standard Arch names.
+ARM_CPU_NAME("iwmmxt", IWMMXT, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("xscale", XSCALE, FK_NONE, true, ARM::AEK_NONE)
+ARM_CPU_NAME("swift", ARMV7S, FK_NEON_VFPV4, true,
+ (ARM::AEK_HWDIVARM | ARM::AEK_HWDIVTHUMB))
+// Invalid CPU
+ARM_CPU_NAME("invalid", INVALID, FK_INVALID, true, ARM::AEK_INVALID)
+#undef ARM_CPU_NAME
diff --git a/third_party/llvm-project/include/llvm/Support/ARMTargetParser.h b/third_party/llvm-project/include/llvm/Support/ARMTargetParser.h
new file mode 100644
index 000000000..02d4c9751
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ARMTargetParser.h
@@ -0,0 +1,267 @@
+//===-- ARMTargetParser - Parser for ARM target features --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a target parser to recognise ARM hardware features
+// such as FPU/CPU/ARCH/extensions and specific support such as HWDIV.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ARMTARGETPARSER_H
+#define LLVM_SUPPORT_ARMTARGETPARSER_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/Support/ARMBuildAttributes.h"
+#include <vector>
+
+namespace llvm {
+namespace ARM {
+
+// Arch extension modifiers for CPUs.
+// Note that this is not the same as the AArch64 list
+enum ArchExtKind : unsigned {
+ AEK_INVALID = 0,
+ AEK_NONE = 1,
+ AEK_CRC = 1 << 1,
+ AEK_CRYPTO = 1 << 2,
+ AEK_FP = 1 << 3,
+ AEK_HWDIVTHUMB = 1 << 4,
+ AEK_HWDIVARM = 1 << 5,
+ AEK_MP = 1 << 6,
+ AEK_SIMD = 1 << 7,
+ AEK_SEC = 1 << 8,
+ AEK_VIRT = 1 << 9,
+ AEK_DSP = 1 << 10,
+ AEK_FP16 = 1 << 11,
+ AEK_RAS = 1 << 12,
+ AEK_DOTPROD = 1 << 13,
+ AEK_SHA2 = 1 << 14,
+ AEK_AES = 1 << 15,
+ AEK_FP16FML = 1 << 16,
+ AEK_SB = 1 << 17,
+ AEK_FP_DP = 1 << 18,
+ AEK_LOB = 1 << 19,
+ // Unsupported extensions.
+ AEK_OS = 0x8000000,
+ AEK_IWMMXT = 0x10000000,
+ AEK_IWMMXT2 = 0x20000000,
+ AEK_MAVERICK = 0x40000000,
+ AEK_XSCALE = 0x80000000,
+};
+
+// List of Arch Extension names.
+// FIXME: TableGen this.
+struct ExtName {
+ const char *NameCStr;
+ size_t NameLength;
+ unsigned ID;
+ const char *Feature;
+ const char *NegFeature;
+
+ StringRef getName() const { return StringRef(NameCStr, NameLength); }
+};
+
+const ExtName ARCHExtNames[] = {
+#define ARM_ARCH_EXT_NAME(NAME, ID, FEATURE, NEGFEATURE) \
+ {NAME, sizeof(NAME) - 1, ID, FEATURE, NEGFEATURE},
+#include "ARMTargetParser.def"
+};
+
+// List of HWDiv names (use getHWDivSynonym) and which architectural
+// features they correspond to (use getHWDivFeatures).
+// FIXME: TableGen this.
+const struct {
+ const char *NameCStr;
+ size_t NameLength;
+ unsigned ID;
+
+ StringRef getName() const { return StringRef(NameCStr, NameLength); }
+} HWDivNames[] = {
+#define ARM_HW_DIV_NAME(NAME, ID) {NAME, sizeof(NAME) - 1, ID},
+#include "ARMTargetParser.def"
+};
+
+// Arch names.
+enum class ArchKind {
+#define ARM_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, ARCH_BASE_EXT) ID,
+#include "ARMTargetParser.def"
+};
+
+// List of CPU names and their arches.
+// The same CPU can have multiple arches and can be default on multiple arches.
+// When finding the Arch for a CPU, first-found prevails. Sort them accordingly.
+// When this becomes table-generated, we'd probably need two tables.
+// FIXME: TableGen this.
+template <typename T> struct CpuNames {
+ const char *NameCStr;
+ size_t NameLength;
+ T ArchID;
+ bool Default; // is $Name the default CPU for $ArchID ?
+ unsigned DefaultExtensions;
+
+ StringRef getName() const { return StringRef(NameCStr, NameLength); }
+};
+
+const CpuNames<ArchKind> CPUNames[] = {
+#define ARM_CPU_NAME(NAME, ID, DEFAULT_FPU, IS_DEFAULT, DEFAULT_EXT) \
+ {NAME, sizeof(NAME) - 1, ARM::ArchKind::ID, IS_DEFAULT, DEFAULT_EXT},
+#include "ARMTargetParser.def"
+};
+
+// FPU names.
+enum FPUKind {
+#define ARM_FPU(NAME, KIND, VERSION, NEON_SUPPORT, RESTRICTION) KIND,
+#include "ARMTargetParser.def"
+ FK_LAST
+};
+
+// FPU Version
+enum class FPUVersion {
+ NONE,
+ VFPV2,
+ VFPV3,
+ VFPV3_FP16,
+ VFPV4,
+ VFPV5,
+ VFPV5_FULLFP16,
+};
+
+// An FPU name restricts the FPU in one of three ways:
+enum class FPURestriction {
+ None = 0, ///< No restriction
+ D16, ///< Only 16 D registers
+ SP_D16 ///< Only single-precision instructions, with 16 D registers
+};
+
+// An FPU name implies one of three levels of Neon support:
+enum class NeonSupportLevel {
+ None = 0, ///< No Neon
+ Neon, ///< Neon
+ Crypto ///< Neon with Crypto
+};
+
+// ISA kinds.
+enum class ISAKind { INVALID = 0, ARM, THUMB, AARCH64 };
+
+// Endianness
+// FIXME: BE8 vs. BE32?
+enum class EndianKind { INVALID = 0, LITTLE, BIG };
+
+// v6/v7/v8 Profile
+enum class ProfileKind { INVALID = 0, A, R, M };
+
+// List of canonical FPU names (use getFPUSynonym) and which architectural
+// features they correspond to (use getFPUFeatures).
+// FIXME: TableGen this.
+// The entries must appear in the order listed in ARM::FPUKind for correct
+// indexing
+struct FPUName {
+ const char *NameCStr;
+ size_t NameLength;
+ FPUKind ID;
+ FPUVersion FPUVer;
+ NeonSupportLevel NeonSupport;
+ FPURestriction Restriction;
+
+ StringRef getName() const { return StringRef(NameCStr, NameLength); }
+};
+
+static const FPUName FPUNames[] = {
+#define ARM_FPU(NAME, KIND, VERSION, NEON_SUPPORT, RESTRICTION) \
+ {NAME, sizeof(NAME) - 1, KIND, VERSION, NEON_SUPPORT, RESTRICTION},
+#include "llvm/Support/ARMTargetParser.def"
+};
+
+// List of canonical arch names (use getArchSynonym).
+// This table also provides the build attribute fields for CPU arch
+// and Arch ID, according to the Addenda to the ARM ABI, chapters
+// 2.4 and 2.3.5.2 respectively.
+// FIXME: SubArch values were simplified to fit into the expectations
+// of the triples and are not conforming with their official names.
+// Check to see if the expectation should be changed.
+// FIXME: TableGen this.
+template <typename T> struct ArchNames {
+ const char *NameCStr;
+ size_t NameLength;
+ const char *CPUAttrCStr;
+ size_t CPUAttrLength;
+ const char *SubArchCStr;
+ size_t SubArchLength;
+ unsigned DefaultFPU;
+ unsigned ArchBaseExtensions;
+ T ID;
+ ARMBuildAttrs::CPUArch ArchAttr; // Arch ID in build attributes.
+
+ StringRef getName() const { return StringRef(NameCStr, NameLength); }
+
+ // CPU class in build attributes.
+ StringRef getCPUAttr() const { return StringRef(CPUAttrCStr, CPUAttrLength); }
+
+ // Sub-Arch name.
+ StringRef getSubArch() const { return StringRef(SubArchCStr, SubArchLength); }
+};
+
+static const ArchNames<ArchKind> ARCHNames[] = {
+#define ARM_ARCH(NAME, ID, CPU_ATTR, SUB_ARCH, ARCH_ATTR, ARCH_FPU, \
+ ARCH_BASE_EXT) \
+ {NAME, sizeof(NAME) - 1, \
+ CPU_ATTR, sizeof(CPU_ATTR) - 1, \
+ SUB_ARCH, sizeof(SUB_ARCH) - 1, \
+ ARCH_FPU, ARCH_BASE_EXT, \
+ ArchKind::ID, ARCH_ATTR},
+#include "llvm/Support/ARMTargetParser.def"
+};
+
+// Information by ID
+StringRef getFPUName(unsigned FPUKind);
+FPUVersion getFPUVersion(unsigned FPUKind);
+NeonSupportLevel getFPUNeonSupportLevel(unsigned FPUKind);
+FPURestriction getFPURestriction(unsigned FPUKind);
+
+// FIXME: These should be moved to TargetTuple once it exists
+bool getFPUFeatures(unsigned FPUKind, std::vector<StringRef> &Features);
+bool getHWDivFeatures(unsigned HWDivKind, std::vector<StringRef> &Features);
+bool getExtensionFeatures(unsigned Extensions,
+ std::vector<StringRef> &Features);
+
+StringRef getArchName(ArchKind AK);
+unsigned getArchAttr(ArchKind AK);
+StringRef getCPUAttr(ArchKind AK);
+StringRef getSubArch(ArchKind AK);
+StringRef getArchExtName(unsigned ArchExtKind);
+StringRef getArchExtFeature(StringRef ArchExt);
+bool appendArchExtFeatures(StringRef CPU, ARM::ArchKind AK, StringRef ArchExt,
+ std::vector<StringRef> &Features);
+StringRef getHWDivName(unsigned HWDivKind);
+
+// Information by Name
+unsigned getDefaultFPU(StringRef CPU, ArchKind AK);
+unsigned getDefaultExtensions(StringRef CPU, ArchKind AK);
+StringRef getDefaultCPU(StringRef Arch);
+StringRef getCanonicalArchName(StringRef Arch);
+StringRef getFPUSynonym(StringRef FPU);
+StringRef getArchSynonym(StringRef Arch);
+
+// Parser
+unsigned parseHWDiv(StringRef HWDiv);
+unsigned parseFPU(StringRef FPU);
+ArchKind parseArch(StringRef Arch);
+unsigned parseArchExt(StringRef ArchExt);
+ArchKind parseCPUArch(StringRef CPU);
+ISAKind parseArchISA(StringRef Arch);
+EndianKind parseArchEndian(StringRef Arch);
+ProfileKind parseArchProfile(StringRef Arch);
+unsigned parseArchVersion(StringRef Arch);
+
+void fillValidCPUArchList(SmallVectorImpl<StringRef> &Values);
+StringRef computeDefaultTargetABI(const Triple &TT, StringRef CPU);
+
+} // namespace ARM
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/AlignOf.h b/third_party/llvm-project/include/llvm/Support/AlignOf.h
new file mode 100644
index 000000000..eb42542b7
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/AlignOf.h
@@ -0,0 +1,55 @@
+//===--- AlignOf.h - Portable calculation of type alignment -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the AlignedCharArrayUnion class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ALIGNOF_H
+#define LLVM_SUPPORT_ALIGNOF_H
+
+#include "llvm/Support/Compiler.h"
+#include <cstddef>
+
+namespace llvm {
+
+namespace detail {
+
+template <typename T, typename... Ts> class AlignerImpl {
+ T t;
+ AlignerImpl<Ts...> rest;
+ AlignerImpl() = delete;
+};
+
+template <typename T> class AlignerImpl<T> {
+ T t;
+ AlignerImpl() = delete;
+};
+
+template <typename T, typename... Ts> union SizerImpl {
+ char arr[sizeof(T)];
+ SizerImpl<Ts...> rest;
+};
+
+template <typename T> union SizerImpl<T> { char arr[sizeof(T)]; };
+} // end namespace detail
+
+/// A suitably aligned and sized character array member which can hold elements
+/// of any type.
+///
+/// These types may be arrays, structs, or any other types. This exposes a
+/// `buffer` member which can be used as suitable storage for a placement new of
+/// any of these types.
+template <typename T, typename... Ts> struct AlignedCharArrayUnion {
+ alignas(::llvm::detail::AlignerImpl<T, Ts...>) char buffer[sizeof(
+ llvm::detail::SizerImpl<T, Ts...>)];
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_ALIGNOF_H
diff --git a/third_party/llvm-project/include/llvm/Support/Alignment.h b/third_party/llvm-project/include/llvm/Support/Alignment.h
new file mode 100644
index 000000000..72fad87dd
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Alignment.h
@@ -0,0 +1,403 @@
+//===-- llvm/Support/Alignment.h - Useful alignment functions ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains types to represent alignments.
+// They are instrumented to guarantee some invariants are preserved and prevent
+// invalid manipulations.
+//
+// - Align represents an alignment in bytes, it is always set and always a valid
+// power of two, its minimum value is 1 which means no alignment requirements.
+//
+// - MaybeAlign is an optional type, it may be undefined or set. When it's set
+// you can get the underlying Align type by using the getValue() method.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ALIGNMENT_H_
+#define LLVM_SUPPORT_ALIGNMENT_H_
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/MathExtras.h"
+#include <cassert>
+#include <limits>
+
+namespace llvm {
+
+#define ALIGN_CHECK_ISPOSITIVE(decl) \
+ assert(decl > 0 && (#decl " should be defined"))
+#define ALIGN_CHECK_ISSET(decl) \
+ assert(decl.hasValue() && (#decl " should be defined"))
+
+/// This struct is a compact representation of a valid (non-zero power of two)
+/// alignment.
+/// It is suitable for use as static global constants.
+struct Align {
+private:
+ uint8_t ShiftValue = 0; /// The log2 of the required alignment.
+ /// ShiftValue is less than 64 by construction.
+
+ friend struct MaybeAlign;
+ friend unsigned Log2(Align);
+ friend bool operator==(Align Lhs, Align Rhs);
+ friend bool operator!=(Align Lhs, Align Rhs);
+ friend bool operator<=(Align Lhs, Align Rhs);
+ friend bool operator>=(Align Lhs, Align Rhs);
+ friend bool operator<(Align Lhs, Align Rhs);
+ friend bool operator>(Align Lhs, Align Rhs);
+ friend unsigned encode(struct MaybeAlign A);
+ friend struct MaybeAlign decodeMaybeAlign(unsigned Value);
+
+ /// A trivial type to allow construction of constexpr Align.
+ /// This is currently needed to workaround a bug in GCC 5.3 which prevents
+ /// definition of constexpr assign operators.
+ /// https://stackoverflow.com/questions/46756288/explicitly-defaulted-function-cannot-be-declared-as-constexpr-because-the-implic
+ /// FIXME: Remove this, make all assign operators constexpr and introduce user
+ /// defined literals when we don't have to support GCC 5.3 anymore.
+ /// https://llvm.org/docs/GettingStarted.html#getting-a-modern-host-c-toolchain
+ struct LogValue {
+ uint8_t Log;
+ };
+
+public:
+ /// Default is byte-aligned.
+ constexpr Align() = default;
+ /// Do not perform checks in case of copy/move construct/assign, because the
+ /// checks have been performed when building `Other`.
+ constexpr Align(const Align &Other) = default;
+ constexpr Align(Align &&Other) = default;
+ Align &operator=(const Align &Other) = default;
+ Align &operator=(Align &&Other) = default;
+
+ explicit Align(uint64_t Value) {
+ assert(Value > 0 && "Value must not be 0");
+ assert(llvm::isPowerOf2_64(Value) && "Alignment is not a power of 2");
+ ShiftValue = Log2_64(Value);
+ assert(ShiftValue < 64 && "Broken invariant");
+ }
+
+ /// This is a hole in the type system and should not be abused.
+ /// Needed to interact with C for instance.
+ uint64_t value() const { return uint64_t(1) << ShiftValue; }
+
+ /// Returns a default constructed Align which corresponds to no alignment.
+ /// This is useful to test for unalignment as it conveys clear semantic.
+ /// `if (A != Align::None())`
+ /// would be better than
+ /// `if (A > Align(1))`
+ constexpr static const Align None() { return Align(); }
+
+ /// Allow constructions of constexpr Align.
+ template <size_t kValue> constexpr static LogValue Constant() {
+ return LogValue{static_cast<uint8_t>(CTLog2<kValue>())};
+ }
+
+ /// Allow constructions of constexpr Align from types.
+ /// Compile time equivalent to Align(alignof(T)).
+ template <typename T> constexpr static LogValue Of() {
+ return Constant<std::alignment_of<T>::value>();
+ }
+
+ /// Constexpr constructor from LogValue type.
+ constexpr Align(LogValue CA) : ShiftValue(CA.Log) {}
+};
+
+/// Treats the value 0 as a 1, so Align is always at least 1.
+inline Align assumeAligned(uint64_t Value) {
+ return Value ? Align(Value) : Align();
+}
+
+/// This struct is a compact representation of a valid (power of two) or
+/// undefined (0) alignment.
+struct MaybeAlign : public llvm::Optional<Align> {
+private:
+ using UP = llvm::Optional<Align>;
+
+public:
+ /// Default is undefined.
+ MaybeAlign() = default;
+ /// Do not perform checks in case of copy/move construct/assign, because the
+ /// checks have been performed when building `Other`.
+ MaybeAlign(const MaybeAlign &Other) = default;
+ MaybeAlign &operator=(const MaybeAlign &Other) = default;
+ MaybeAlign(MaybeAlign &&Other) = default;
+ MaybeAlign &operator=(MaybeAlign &&Other) = default;
+
+ /// Use llvm::Optional<Align> constructor.
+ using UP::UP;
+
+ explicit MaybeAlign(uint64_t Value) {
+ assert((Value == 0 || llvm::isPowerOf2_64(Value)) &&
+ "Alignment is neither 0 nor a power of 2");
+ if (Value)
+ emplace(Value);
+ }
+
+ /// For convenience, returns a valid alignment or 1 if undefined.
+ Align valueOrOne() const { return hasValue() ? getValue() : Align(); }
+};
+
+/// Checks that SizeInBytes is a multiple of the alignment.
+inline bool isAligned(Align Lhs, uint64_t SizeInBytes) {
+ return SizeInBytes % Lhs.value() == 0;
+}
+
+/// Checks that SizeInBytes is a multiple of the alignment.
+/// Returns false if the alignment is undefined.
+inline bool isAligned(MaybeAlign Lhs, uint64_t SizeInBytes) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return SizeInBytes % (*Lhs).value() == 0;
+}
+
+/// Checks that Addr is a multiple of the alignment.
+inline bool isAddrAligned(Align Lhs, const void *Addr) {
+ return isAligned(Lhs, reinterpret_cast<uintptr_t>(Addr));
+}
+
+/// Returns a multiple of A needed to store `Size` bytes.
+inline uint64_t alignTo(uint64_t Size, Align A) {
+ const uint64_t value = A.value();
+ // The following line is equivalent to `(Size + value - 1) / value * value`.
+
+ // The division followed by a multiplication can be thought of as a right
+ // shift followed by a left shift which zeros out the extra bits produced in
+ // the bump; `~(value - 1)` is a mask where all those bits being zeroed out
+ // are just zero.
+
+ // Most compilers can generate this code but the pattern may be missed when
+ // multiple functions gets inlined.
+ return (Size + value - 1) & ~(value - 1);
+}
+
+/// Returns a multiple of A needed to store `Size` bytes.
+/// Returns `Size` if current alignment is undefined.
+inline uint64_t alignTo(uint64_t Size, MaybeAlign A) {
+ return A ? alignTo(Size, A.getValue()) : Size;
+}
+
+/// Aligns `Addr` to `Alignment` bytes, rounding up.
+inline uintptr_t alignAddr(const void *Addr, Align Alignment) {
+ uintptr_t ArithAddr = reinterpret_cast<uintptr_t>(Addr);
+ assert(static_cast<uintptr_t>(ArithAddr + Alignment.value() - 1) >=
+ ArithAddr && "Overflow");
+ return alignTo(ArithAddr, Alignment);
+}
+
+/// Returns the offset to the next integer (mod 2**64) that is greater than
+/// or equal to \p Value and is a multiple of \p Align.
+inline uint64_t offsetToAlignment(uint64_t Value, Align Alignment) {
+ return alignTo(Value, Alignment) - Value;
+}
+
+/// Returns the necessary adjustment for aligning `Addr` to `Alignment`
+/// bytes, rounding up.
+inline uint64_t offsetToAlignedAddr(const void *Addr, Align Alignment) {
+ return offsetToAlignment(reinterpret_cast<uintptr_t>(Addr), Alignment);
+}
+
+/// Returns the log2 of the alignment.
+inline unsigned Log2(Align A) { return A.ShiftValue; }
+
+/// Returns the log2 of the alignment.
+/// \pre A must be defined.
+inline unsigned Log2(MaybeAlign A) {
+ ALIGN_CHECK_ISSET(A);
+ return Log2(A.getValue());
+}
+
+/// Returns the alignment that satisfies both alignments.
+/// Same semantic as MinAlign.
+inline Align commonAlignment(Align A, Align B) { return std::min(A, B); }
+
+/// Returns the alignment that satisfies both alignments.
+/// Same semantic as MinAlign.
+inline Align commonAlignment(Align A, uint64_t Offset) {
+ return Align(MinAlign(A.value(), Offset));
+}
+
+/// Returns the alignment that satisfies both alignments.
+/// Same semantic as MinAlign.
+inline MaybeAlign commonAlignment(MaybeAlign A, MaybeAlign B) {
+ return A && B ? commonAlignment(*A, *B) : A ? A : B;
+}
+
+/// Returns the alignment that satisfies both alignments.
+/// Same semantic as MinAlign.
+inline MaybeAlign commonAlignment(MaybeAlign A, uint64_t Offset) {
+ return MaybeAlign(MinAlign((*A).value(), Offset));
+}
+
+/// Returns a representation of the alignment that encodes undefined as 0.
+inline unsigned encode(MaybeAlign A) { return A ? A->ShiftValue + 1 : 0; }
+
+/// Dual operation of the encode function above.
+inline MaybeAlign decodeMaybeAlign(unsigned Value) {
+ if (Value == 0)
+ return MaybeAlign();
+ Align Out;
+ Out.ShiftValue = Value - 1;
+ return Out;
+}
+
+/// Returns a representation of the alignment, the encoded value is positive by
+/// definition.
+inline unsigned encode(Align A) { return encode(MaybeAlign(A)); }
+
+/// Comparisons between Align and scalars. Rhs must be positive.
+inline bool operator==(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() == Rhs;
+}
+inline bool operator!=(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() != Rhs;
+}
+inline bool operator<=(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() <= Rhs;
+}
+inline bool operator>=(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() >= Rhs;
+}
+inline bool operator<(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() < Rhs;
+}
+inline bool operator>(Align Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return Lhs.value() > Rhs;
+}
+
+/// Comparisons between MaybeAlign and scalars.
+inline bool operator==(MaybeAlign Lhs, uint64_t Rhs) {
+ return Lhs ? (*Lhs).value() == Rhs : Rhs == 0;
+}
+inline bool operator!=(MaybeAlign Lhs, uint64_t Rhs) {
+ return Lhs ? (*Lhs).value() != Rhs : Rhs != 0;
+}
+inline bool operator<=(MaybeAlign Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return (*Lhs).value() <= Rhs;
+}
+inline bool operator>=(MaybeAlign Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return (*Lhs).value() >= Rhs;
+}
+inline bool operator<(MaybeAlign Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return (*Lhs).value() < Rhs;
+}
+inline bool operator>(MaybeAlign Lhs, uint64_t Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ ALIGN_CHECK_ISPOSITIVE(Rhs);
+ return (*Lhs).value() > Rhs;
+}
+
+/// Comparisons operators between Align.
+inline bool operator==(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue == Rhs.ShiftValue;
+}
+inline bool operator!=(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue != Rhs.ShiftValue;
+}
+inline bool operator<=(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue <= Rhs.ShiftValue;
+}
+inline bool operator>=(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue >= Rhs.ShiftValue;
+}
+inline bool operator<(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue < Rhs.ShiftValue;
+}
+inline bool operator>(Align Lhs, Align Rhs) {
+ return Lhs.ShiftValue > Rhs.ShiftValue;
+}
+
+/// Comparisons operators between Align and MaybeAlign.
+inline bool operator==(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() == (*Rhs).value();
+}
+inline bool operator!=(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() != (*Rhs).value();
+}
+inline bool operator<=(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() <= (*Rhs).value();
+}
+inline bool operator>=(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() >= (*Rhs).value();
+}
+inline bool operator<(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() < (*Rhs).value();
+}
+inline bool operator>(Align Lhs, MaybeAlign Rhs) {
+ ALIGN_CHECK_ISSET(Rhs);
+ return Lhs.value() > (*Rhs).value();
+}
+
+/// Comparisons operators between MaybeAlign and Align.
+inline bool operator==(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() == Rhs.value();
+}
+inline bool operator!=(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() != Rhs.value();
+}
+inline bool operator<=(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() <= Rhs.value();
+}
+inline bool operator>=(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() >= Rhs.value();
+}
+inline bool operator<(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() < Rhs.value();
+}
+inline bool operator>(MaybeAlign Lhs, Align Rhs) {
+ ALIGN_CHECK_ISSET(Lhs);
+ return Lhs && (*Lhs).value() > Rhs.value();
+}
+
+inline Align operator/(Align Lhs, uint64_t Divisor) {
+ assert(llvm::isPowerOf2_64(Divisor) &&
+ "Divisor must be positive and a power of 2");
+ assert(Lhs != 1 && "Can't halve byte alignment");
+ return Align(Lhs.value() / Divisor);
+}
+
+inline MaybeAlign operator/(MaybeAlign Lhs, uint64_t Divisor) {
+ assert(llvm::isPowerOf2_64(Divisor) &&
+ "Divisor must be positive and a power of 2");
+ return Lhs ? Lhs.getValue() / Divisor : MaybeAlign();
+}
+
+inline Align max(MaybeAlign Lhs, Align Rhs) {
+ return Lhs && *Lhs > Rhs ? *Lhs : Rhs;
+}
+
+inline Align max(Align Lhs, MaybeAlign Rhs) {
+ return Rhs && *Rhs > Lhs ? *Rhs : Lhs;
+}
+
+#undef ALIGN_CHECK_ISPOSITIVE
+#undef ALIGN_CHECK_ISSET
+
+} // namespace llvm
+
+#endif // LLVM_SUPPORT_ALIGNMENT_H_
diff --git a/third_party/llvm-project/include/llvm/Support/Allocator.h b/third_party/llvm-project/include/llvm/Support/Allocator.h
new file mode 100644
index 000000000..106b90c35
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Allocator.h
@@ -0,0 +1,523 @@
+//===- Allocator.h - Simple memory allocation abstraction -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file defines the MallocAllocator and BumpPtrAllocator interfaces. Both
+/// of these conform to an LLVM "Allocator" concept which consists of an
+/// Allocate method accepting a size and alignment, and a Deallocate accepting
+/// a pointer and size. Further, the LLVM "Allocator" concept has overloads of
+/// Allocate and Deallocate for setting size and alignment based on the final
+/// type. These overloads are typically provided by a base class template \c
+/// AllocatorBase.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ALLOCATOR_H
+#define LLVM_SUPPORT_ALLOCATOR_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Support/Alignment.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/MemAlloc.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstdlib>
+#include <iterator>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// CRTP base class providing obvious overloads for the core \c
+/// Allocate() methods of LLVM-style allocators.
+///
+/// This base class both documents the full public interface exposed by all
+/// LLVM-style allocators, and redirects all of the overloads to a single core
+/// set of methods which the derived class must define.
+template <typename DerivedT> class AllocatorBase {
+public:
+ /// Allocate \a Size bytes of \a Alignment aligned memory. This method
+ /// must be implemented by \c DerivedT.
+ void *Allocate(size_t Size, size_t Alignment) {
+#ifdef __clang__
+ static_assert(static_cast<void *(AllocatorBase::*)(size_t, size_t)>(
+ &AllocatorBase::Allocate) !=
+ static_cast<void *(DerivedT::*)(size_t, size_t)>(
+ &DerivedT::Allocate),
+ "Class derives from AllocatorBase without implementing the "
+ "core Allocate(size_t, size_t) overload!");
+#endif
+ return static_cast<DerivedT *>(this)->Allocate(Size, Alignment);
+ }
+
+ /// Deallocate \a Ptr to \a Size bytes of memory allocated by this
+ /// allocator.
+ void Deallocate(const void *Ptr, size_t Size) {
+#ifdef __clang__
+ static_assert(static_cast<void (AllocatorBase::*)(const void *, size_t)>(
+ &AllocatorBase::Deallocate) !=
+ static_cast<void (DerivedT::*)(const void *, size_t)>(
+ &DerivedT::Deallocate),
+ "Class derives from AllocatorBase without implementing the "
+ "core Deallocate(void *) overload!");
+#endif
+ return static_cast<DerivedT *>(this)->Deallocate(Ptr, Size);
+ }
+
+ // The rest of these methods are helpers that redirect to one of the above
+ // core methods.
+
+ /// Allocate space for a sequence of objects without constructing them.
+ template <typename T> T *Allocate(size_t Num = 1) {
+ return static_cast<T *>(Allocate(Num * sizeof(T), alignof(T)));
+ }
+
+ /// Deallocate space for a sequence of objects without constructing them.
+ template <typename T>
+ typename std::enable_if<
+ !std::is_same<typename std::remove_cv<T>::type, void>::value, void>::type
+ Deallocate(T *Ptr, size_t Num = 1) {
+ Deallocate(static_cast<const void *>(Ptr), Num * sizeof(T));
+ }
+};
+
+class MallocAllocator : public AllocatorBase<MallocAllocator> {
+public:
+ void Reset() {}
+
+ LLVM_ATTRIBUTE_RETURNS_NONNULL void *Allocate(size_t Size,
+ size_t /*Alignment*/) {
+ return safe_malloc(Size);
+ }
+
+ // Pull in base class overloads.
+ using AllocatorBase<MallocAllocator>::Allocate;
+
+ void Deallocate(const void *Ptr, size_t /*Size*/) {
+ free(const_cast<void *>(Ptr));
+ }
+
+ // Pull in base class overloads.
+ using AllocatorBase<MallocAllocator>::Deallocate;
+
+ void PrintStats() const {}
+};
+
+namespace detail {
+
+// We call out to an external function to actually print the message as the
+// printing code uses Allocator.h in its implementation.
+void printBumpPtrAllocatorStats(unsigned NumSlabs, size_t BytesAllocated,
+ size_t TotalMemory);
+
+} // end namespace detail
+
+/// Allocate memory in an ever growing pool, as if by bump-pointer.
+///
+/// This isn't strictly a bump-pointer allocator as it uses backing slabs of
+/// memory rather than relying on a boundless contiguous heap. However, it has
+/// bump-pointer semantics in that it is a monotonically growing pool of memory
+/// where every allocation is found by merely allocating the next N bytes in
+/// the slab, or the next N bytes in the next slab.
+///
+/// Note that this also has a threshold for forcing allocations above a certain
+/// size into their own slab.
+///
+/// The BumpPtrAllocatorImpl template defaults to using a MallocAllocator
+/// object, which wraps malloc, to allocate memory, but it can be changed to
+/// use a custom allocator.
+template <typename AllocatorT = MallocAllocator, size_t SlabSize = 4096,
+ size_t SizeThreshold = SlabSize>
+class BumpPtrAllocatorImpl
+ : public AllocatorBase<
+ BumpPtrAllocatorImpl<AllocatorT, SlabSize, SizeThreshold>> {
+public:
+ static_assert(SizeThreshold <= SlabSize,
+ "The SizeThreshold must be at most the SlabSize to ensure "
+ "that objects larger than a slab go into their own memory "
+ "allocation.");
+
+ BumpPtrAllocatorImpl() = default;
+
+ template <typename T>
+ BumpPtrAllocatorImpl(T &&Allocator)
+ : Allocator(std::forward<T &&>(Allocator)) {}
+
+ // Manually implement a move constructor as we must clear the old allocator's
+ // slabs as a matter of correctness.
+ BumpPtrAllocatorImpl(BumpPtrAllocatorImpl &&Old)
+ : CurPtr(Old.CurPtr), End(Old.End), Slabs(std::move(Old.Slabs)),
+ CustomSizedSlabs(std::move(Old.CustomSizedSlabs)),
+ BytesAllocated(Old.BytesAllocated), RedZoneSize(Old.RedZoneSize),
+ Allocator(std::move(Old.Allocator)) {
+ Old.CurPtr = Old.End = nullptr;
+ Old.BytesAllocated = 0;
+ Old.Slabs.clear();
+ Old.CustomSizedSlabs.clear();
+ }
+
+ ~BumpPtrAllocatorImpl() {
+ DeallocateSlabs(Slabs.begin(), Slabs.end());
+ DeallocateCustomSizedSlabs();
+ }
+
+ BumpPtrAllocatorImpl &operator=(BumpPtrAllocatorImpl &&RHS) {
+ DeallocateSlabs(Slabs.begin(), Slabs.end());
+ DeallocateCustomSizedSlabs();
+
+ CurPtr = RHS.CurPtr;
+ End = RHS.End;
+ BytesAllocated = RHS.BytesAllocated;
+ RedZoneSize = RHS.RedZoneSize;
+ Slabs = std::move(RHS.Slabs);
+ CustomSizedSlabs = std::move(RHS.CustomSizedSlabs);
+ Allocator = std::move(RHS.Allocator);
+
+ RHS.CurPtr = RHS.End = nullptr;
+ RHS.BytesAllocated = 0;
+ RHS.Slabs.clear();
+ RHS.CustomSizedSlabs.clear();
+ return *this;
+ }
+
+ /// Deallocate all but the current slab and reset the current pointer
+ /// to the beginning of it, freeing all memory allocated so far.
+ void Reset() {
+ // Deallocate all but the first slab, and deallocate all custom-sized slabs.
+ DeallocateCustomSizedSlabs();
+ CustomSizedSlabs.clear();
+
+ if (Slabs.empty())
+ return;
+
+ // Reset the state.
+ BytesAllocated = 0;
+ CurPtr = (char *)Slabs.front();
+ End = CurPtr + SlabSize;
+
+ __asan_poison_memory_region(*Slabs.begin(), computeSlabSize(0));
+ DeallocateSlabs(std::next(Slabs.begin()), Slabs.end());
+ Slabs.erase(std::next(Slabs.begin()), Slabs.end());
+ }
+
+ /// Allocate space at the specified alignment.
+ LLVM_ATTRIBUTE_RETURNS_NONNULL LLVM_ATTRIBUTE_RETURNS_NOALIAS void *
+ Allocate(size_t Size, Align Alignment) {
+ // Keep track of how many bytes we've allocated.
+ BytesAllocated += Size;
+
+ size_t Adjustment = offsetToAlignedAddr(CurPtr, Alignment);
+ assert(Adjustment + Size >= Size && "Adjustment + Size must not overflow");
+
+ size_t SizeToAllocate = Size;
+#if LLVM_ADDRESS_SANITIZER_BUILD
+ // Add trailing bytes as a "red zone" under ASan.
+ SizeToAllocate += RedZoneSize;
+#endif
+
+ // Check if we have enough space.
+ if (Adjustment + SizeToAllocate <= size_t(End - CurPtr)) {
+ char *AlignedPtr = CurPtr + Adjustment;
+ CurPtr = AlignedPtr + SizeToAllocate;
+ // Update the allocation point of this memory block in MemorySanitizer.
+ // Without this, MemorySanitizer messages for values originated from here
+ // will point to the allocation of the entire slab.
+ __msan_allocated_memory(AlignedPtr, Size);
+ // Similarly, tell ASan about this space.
+ __asan_unpoison_memory_region(AlignedPtr, Size);
+ return AlignedPtr;
+ }
+
+ // If Size is really big, allocate a separate slab for it.
+ size_t PaddedSize = SizeToAllocate + Alignment.value() - 1;
+ if (PaddedSize > SizeThreshold) {
+ void *NewSlab = Allocator.Allocate(PaddedSize, 0);
+ // We own the new slab and don't want anyone reading anyting other than
+ // pieces returned from this method. So poison the whole slab.
+ __asan_poison_memory_region(NewSlab, PaddedSize);
+ CustomSizedSlabs.push_back(std::make_pair(NewSlab, PaddedSize));
+
+ uintptr_t AlignedAddr = alignAddr(NewSlab, Alignment);
+ assert(AlignedAddr + Size <= (uintptr_t)NewSlab + PaddedSize);
+ char *AlignedPtr = (char*)AlignedAddr;
+ __msan_allocated_memory(AlignedPtr, Size);
+ __asan_unpoison_memory_region(AlignedPtr, Size);
+ return AlignedPtr;
+ }
+
+ // Otherwise, start a new slab and try again.
+ StartNewSlab();
+ uintptr_t AlignedAddr = alignAddr(CurPtr, Alignment);
+ assert(AlignedAddr + SizeToAllocate <= (uintptr_t)End &&
+ "Unable to allocate memory!");
+ char *AlignedPtr = (char*)AlignedAddr;
+ CurPtr = AlignedPtr + SizeToAllocate;
+ __msan_allocated_memory(AlignedPtr, Size);
+ __asan_unpoison_memory_region(AlignedPtr, Size);
+ return AlignedPtr;
+ }
+
+ inline LLVM_ATTRIBUTE_RETURNS_NONNULL LLVM_ATTRIBUTE_RETURNS_NOALIAS void *
+ Allocate(size_t Size, size_t Alignment) {
+ assert(Alignment > 0 && "0-byte alignnment is not allowed. Use 1 instead.");
+ return Allocate(Size, Align(Alignment));
+ }
+
+ // Pull in base class overloads.
+ using AllocatorBase<BumpPtrAllocatorImpl>::Allocate;
+
+ // Bump pointer allocators are expected to never free their storage; and
+ // clients expect pointers to remain valid for non-dereferencing uses even
+ // after deallocation.
+ void Deallocate(const void *Ptr, size_t Size) {
+ __asan_poison_memory_region(Ptr, Size);
+ }
+
+ // Pull in base class overloads.
+ using AllocatorBase<BumpPtrAllocatorImpl>::Deallocate;
+
+ size_t GetNumSlabs() const { return Slabs.size() + CustomSizedSlabs.size(); }
+
+ /// \return An index uniquely and reproducibly identifying
+ /// an input pointer \p Ptr in the given allocator.
+ /// The returned value is negative iff the object is inside a custom-size
+ /// slab.
+ /// Returns an empty optional if the pointer is not found in the allocator.
+ llvm::Optional<int64_t> identifyObject(const void *Ptr) {
+ const char *P = static_cast<const char *>(Ptr);
+ int64_t InSlabIdx = 0;
+ for (size_t Idx = 0, E = Slabs.size(); Idx < E; Idx++) {
+ const char *S = static_cast<const char *>(Slabs[Idx]);
+ if (P >= S && P < S + computeSlabSize(Idx))
+ return InSlabIdx + static_cast<int64_t>(P - S);
+ InSlabIdx += static_cast<int64_t>(computeSlabSize(Idx));
+ }
+
+ // Use negative index to denote custom sized slabs.
+ int64_t InCustomSizedSlabIdx = -1;
+ for (size_t Idx = 0, E = CustomSizedSlabs.size(); Idx < E; Idx++) {
+ const char *S = static_cast<const char *>(CustomSizedSlabs[Idx].first);
+ size_t Size = CustomSizedSlabs[Idx].second;
+ if (P >= S && P < S + Size)
+ return InCustomSizedSlabIdx - static_cast<int64_t>(P - S);
+ InCustomSizedSlabIdx -= static_cast<int64_t>(Size);
+ }
+ return None;
+ }
+
+ /// A wrapper around identifyObject that additionally asserts that
+ /// the object is indeed within the allocator.
+ /// \return An index uniquely and reproducibly identifying
+ /// an input pointer \p Ptr in the given allocator.
+ int64_t identifyKnownObject(const void *Ptr) {
+ Optional<int64_t> Out = identifyObject(Ptr);
+ assert(Out && "Wrong allocator used");
+ return *Out;
+ }
+
+ /// A wrapper around identifyKnownObject. Accepts type information
+ /// about the object and produces a smaller identifier by relying on
+ /// the alignment information. Note that sub-classes may have different
+ /// alignment, so the most base class should be passed as template parameter
+ /// in order to obtain correct results. For that reason automatic template
+ /// parameter deduction is disabled.
+ /// \return An index uniquely and reproducibly identifying
+ /// an input pointer \p Ptr in the given allocator. This identifier is
+ /// different from the ones produced by identifyObject and
+ /// identifyAlignedObject.
+ template <typename T>
+ int64_t identifyKnownAlignedObject(const void *Ptr) {
+ int64_t Out = identifyKnownObject(Ptr);
+ assert(Out % alignof(T) == 0 && "Wrong alignment information");
+ return Out / alignof(T);
+ }
+
+ size_t getTotalMemory() const {
+ size_t TotalMemory = 0;
+ for (auto I = Slabs.begin(), E = Slabs.end(); I != E; ++I)
+ TotalMemory += computeSlabSize(std::distance(Slabs.begin(), I));
+ for (auto &PtrAndSize : CustomSizedSlabs)
+ TotalMemory += PtrAndSize.second;
+ return TotalMemory;
+ }
+
+ size_t getBytesAllocated() const { return BytesAllocated; }
+
+ void setRedZoneSize(size_t NewSize) {
+ RedZoneSize = NewSize;
+ }
+
+ void PrintStats() const {
+ detail::printBumpPtrAllocatorStats(Slabs.size(), BytesAllocated,
+ getTotalMemory());
+ }
+
+private:
+ /// The current pointer into the current slab.
+ ///
+ /// This points to the next free byte in the slab.
+ char *CurPtr = nullptr;
+
+ /// The end of the current slab.
+ char *End = nullptr;
+
+ /// The slabs allocated so far.
+ SmallVector<void *, 4> Slabs;
+
+ /// Custom-sized slabs allocated for too-large allocation requests.
+ SmallVector<std::pair<void *, size_t>, 0> CustomSizedSlabs;
+
+ /// How many bytes we've allocated.
+ ///
+ /// Used so that we can compute how much space was wasted.
+ size_t BytesAllocated = 0;
+
+ /// The number of bytes to put between allocations when running under
+ /// a sanitizer.
+ size_t RedZoneSize = 1;
+
+ /// The allocator instance we use to get slabs of memory.
+ AllocatorT Allocator;
+
+ static size_t computeSlabSize(unsigned SlabIdx) {
+ // Scale the actual allocated slab size based on the number of slabs
+ // allocated. Every 128 slabs allocated, we double the allocated size to
+ // reduce allocation frequency, but saturate at multiplying the slab size by
+ // 2^30.
+ return SlabSize * ((size_t)1 << std::min<size_t>(30, SlabIdx / 128));
+ }
+
+ /// Allocate a new slab and move the bump pointers over into the new
+ /// slab, modifying CurPtr and End.
+ void StartNewSlab() {
+ size_t AllocatedSlabSize = computeSlabSize(Slabs.size());
+
+ void *NewSlab = Allocator.Allocate(AllocatedSlabSize, 0);
+ // We own the new slab and don't want anyone reading anything other than
+ // pieces returned from this method. So poison the whole slab.
+ __asan_poison_memory_region(NewSlab, AllocatedSlabSize);
+
+ Slabs.push_back(NewSlab);
+ CurPtr = (char *)(NewSlab);
+ End = ((char *)NewSlab) + AllocatedSlabSize;
+ }
+
+ /// Deallocate a sequence of slabs.
+ void DeallocateSlabs(SmallVectorImpl<void *>::iterator I,
+ SmallVectorImpl<void *>::iterator E) {
+ for (; I != E; ++I) {
+ size_t AllocatedSlabSize =
+ computeSlabSize(std::distance(Slabs.begin(), I));
+ Allocator.Deallocate(*I, AllocatedSlabSize);
+ }
+ }
+
+ /// Deallocate all memory for custom sized slabs.
+ void DeallocateCustomSizedSlabs() {
+ for (auto &PtrAndSize : CustomSizedSlabs) {
+ void *Ptr = PtrAndSize.first;
+ size_t Size = PtrAndSize.second;
+ Allocator.Deallocate(Ptr, Size);
+ }
+ }
+
+ template <typename T> friend class SpecificBumpPtrAllocator;
+};
+
+/// The standard BumpPtrAllocator which just uses the default template
+/// parameters.
+typedef BumpPtrAllocatorImpl<> BumpPtrAllocator;
+
+/// A BumpPtrAllocator that allows only elements of a specific type to be
+/// allocated.
+///
+/// This allows calling the destructor in DestroyAll() and when the allocator is
+/// destroyed.
+template <typename T> class SpecificBumpPtrAllocator {
+ BumpPtrAllocator Allocator;
+
+public:
+ SpecificBumpPtrAllocator() {
+ // Because SpecificBumpPtrAllocator walks the memory to call destructors,
+ // it can't have red zones between allocations.
+ Allocator.setRedZoneSize(0);
+ }
+ SpecificBumpPtrAllocator(SpecificBumpPtrAllocator &&Old)
+ : Allocator(std::move(Old.Allocator)) {}
+ ~SpecificBumpPtrAllocator() { DestroyAll(); }
+
+ SpecificBumpPtrAllocator &operator=(SpecificBumpPtrAllocator &&RHS) {
+ Allocator = std::move(RHS.Allocator);
+ return *this;
+ }
+
+ /// Call the destructor of each allocated object and deallocate all but the
+ /// current slab and reset the current pointer to the beginning of it, freeing
+ /// all memory allocated so far.
+ void DestroyAll() {
+ auto DestroyElements = [](char *Begin, char *End) {
+ assert(Begin == (char *)alignAddr(Begin, Align::Of<T>()));
+ for (char *Ptr = Begin; Ptr + sizeof(T) <= End; Ptr += sizeof(T))
+ reinterpret_cast<T *>(Ptr)->~T();
+ };
+
+ for (auto I = Allocator.Slabs.begin(), E = Allocator.Slabs.end(); I != E;
+ ++I) {
+ size_t AllocatedSlabSize = BumpPtrAllocator::computeSlabSize(
+ std::distance(Allocator.Slabs.begin(), I));
+ char *Begin = (char *)alignAddr(*I, Align::Of<T>());
+ char *End = *I == Allocator.Slabs.back() ? Allocator.CurPtr
+ : (char *)*I + AllocatedSlabSize;
+
+ DestroyElements(Begin, End);
+ }
+
+ for (auto &PtrAndSize : Allocator.CustomSizedSlabs) {
+ void *Ptr = PtrAndSize.first;
+ size_t Size = PtrAndSize.second;
+ DestroyElements((char *)alignAddr(Ptr, Align::Of<T>()),
+ (char *)Ptr + Size);
+ }
+
+ Allocator.Reset();
+ }
+
+ /// Allocate space for an array of objects without constructing them.
+ T *Allocate(size_t num = 1) { return Allocator.Allocate<T>(num); }
+};
+
+} // end namespace llvm
+
+template <typename AllocatorT, size_t SlabSize, size_t SizeThreshold>
+void *operator new(size_t Size,
+ llvm::BumpPtrAllocatorImpl<AllocatorT, SlabSize,
+ SizeThreshold> &Allocator) {
+ struct S {
+ char c;
+ union {
+ double D;
+ long double LD;
+ long long L;
+ void *P;
+ } x;
+ };
+ return Allocator.Allocate(
+ Size, std::min((size_t)llvm::NextPowerOf2(Size), offsetof(S, x)));
+}
+
+template <typename AllocatorT, size_t SlabSize, size_t SizeThreshold>
+void operator delete(
+ void *, llvm::BumpPtrAllocatorImpl<AllocatorT, SlabSize, SizeThreshold> &) {
+}
+
+#endif // LLVM_SUPPORT_ALLOCATOR_H
diff --git a/third_party/llvm-project/include/llvm/Support/BinaryByteStream.h b/third_party/llvm-project/include/llvm/Support/BinaryByteStream.h
new file mode 100644
index 000000000..7acce9a03
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/BinaryByteStream.h
@@ -0,0 +1,273 @@
+//===- BinaryByteStream.h ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//===----------------------------------------------------------------------===//
+// A BinaryStream which stores data in a single continguous memory buffer.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_BINARYBYTESTREAM_H
+#define LLVM_SUPPORT_BINARYBYTESTREAM_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/BinaryStream.h"
+#include "llvm/Support/BinaryStreamError.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileOutputBuffer.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include <algorithm>
+#include <cstdint>
+#include <cstring>
+#include <memory>
+
+namespace llvm {
+
+/// An implementation of BinaryStream which holds its entire data set
+/// in a single contiguous buffer. BinaryByteStream guarantees that no read
+/// operation will ever incur a copy. Note that BinaryByteStream does not
+/// own the underlying buffer.
+class BinaryByteStream : public BinaryStream {
+public:
+ BinaryByteStream() = default;
+ BinaryByteStream(ArrayRef<uint8_t> Data, llvm::support::endianness Endian)
+ : Endian(Endian), Data(Data) {}
+ BinaryByteStream(StringRef Data, llvm::support::endianness Endian)
+ : Endian(Endian), Data(Data.bytes_begin(), Data.bytes_end()) {}
+
+ llvm::support::endianness getEndian() const override { return Endian; }
+
+ Error readBytes(uint32_t Offset, uint32_t Size,
+ ArrayRef<uint8_t> &Buffer) override {
+ if (auto EC = checkOffsetForRead(Offset, Size))
+ return EC;
+ Buffer = Data.slice(Offset, Size);
+ return Error::success();
+ }
+
+ Error readLongestContiguousChunk(uint32_t Offset,
+ ArrayRef<uint8_t> &Buffer) override {
+ if (auto EC = checkOffsetForRead(Offset, 1))
+ return EC;
+ Buffer = Data.slice(Offset);
+ return Error::success();
+ }
+
+ uint32_t getLength() override { return Data.size(); }
+
+ ArrayRef<uint8_t> data() const { return Data; }
+
+ StringRef str() const {
+ const char *CharData = reinterpret_cast<const char *>(Data.data());
+ return StringRef(CharData, Data.size());
+ }
+
+protected:
+ llvm::support::endianness Endian;
+ ArrayRef<uint8_t> Data;
+};
+
+/// An implementation of BinaryStream whose data is backed by an llvm
+/// MemoryBuffer object. MemoryBufferByteStream owns the MemoryBuffer in
+/// question. As with BinaryByteStream, reading from a MemoryBufferByteStream
+/// will never cause a copy.
+class MemoryBufferByteStream : public BinaryByteStream {
+public:
+ MemoryBufferByteStream(std::unique_ptr<MemoryBuffer> Buffer,
+ llvm::support::endianness Endian)
+ : BinaryByteStream(Buffer->getBuffer(), Endian),
+ MemBuffer(std::move(Buffer)) {}
+
+ std::unique_ptr<MemoryBuffer> MemBuffer;
+};
+
+/// An implementation of BinaryStream which holds its entire data set
+/// in a single contiguous buffer. As with BinaryByteStream, the mutable
+/// version also guarantees that no read operation will ever incur a copy,
+/// and similarly it does not own the underlying buffer.
+class MutableBinaryByteStream : public WritableBinaryStream {
+public:
+ MutableBinaryByteStream() = default;
+ MutableBinaryByteStream(MutableArrayRef<uint8_t> Data,
+ llvm::support::endianness Endian)
+ : Data(Data), ImmutableStream(Data, Endian) {}
+
+ llvm::support::endianness getEndian() const override {
+ return ImmutableStream.getEndian();
+ }
+
+ Error readBytes(uint32_t Offset, uint32_t Size,
+ ArrayRef<uint8_t> &Buffer) override {
+ return ImmutableStream.readBytes(Offset, Size, Buffer);
+ }
+
+ Error readLongestContiguousChunk(uint32_t Offset,
+ ArrayRef<uint8_t> &Buffer) override {
+ return ImmutableStream.readLongestContiguousChunk(Offset, Buffer);
+ }
+
+ uint32_t getLength() override { return ImmutableStream.getLength(); }
+
+ Error writeBytes(uint32_t Offset, ArrayRef<uint8_t> Buffer) override {
+ if (Buffer.empty())
+ return Error::success();
+
+ if (auto EC = checkOffsetForWrite(Offset, Buffer.size()))
+ return EC;
+
+ uint8_t *DataPtr = const_cast<uint8_t *>(Data.data());
+ ::memcpy(DataPtr + Offset, Buffer.data(), Buffer.size());
+ return Error::success();
+ }
+
+ Error commit() override { return Error::success(); }
+
+ MutableArrayRef<uint8_t> data() const { return Data; }
+
+private:
+ MutableArrayRef<uint8_t> Data;
+ BinaryByteStream ImmutableStream;
+};
+
+/// An implementation of WritableBinaryStream which can write at its end
+/// causing the underlying data to grow. This class owns the underlying data.
+class AppendingBinaryByteStream : public WritableBinaryStream {
+ std::vector<uint8_t> Data;
+ llvm::support::endianness Endian = llvm::support::little;
+
+public:
+ AppendingBinaryByteStream() = default;
+ AppendingBinaryByteStream(llvm::support::endianness Endian)
+ : Endian(Endian) {}
+
+ void clear() { Data.clear(); }
+
+ llvm::support::endianness getEndian() const override { return Endian; }
+
+ Error readBytes(uint32_t Offset, uint32_t Size,
+ ArrayRef<uint8_t> &Buffer) override {
+ if (auto EC = checkOffsetForWrite(Offset, Buffer.size()))
+ return EC;
+
+ Buffer = makeArrayRef(Data).slice(Offset, Size);
+ return Error::success();
+ }
+
+ void insert(uint32_t Offset, ArrayRef<uint8_t> Bytes) {
+ Data.insert(Data.begin() + Offset, Bytes.begin(), Bytes.end());
+ }
+
+ Error readLongestContiguousChunk(uint32_t Offset,
+ ArrayRef<uint8_t> &Buffer) override {
+ if (auto EC = checkOffsetForWrite(Offset, 1))
+ return EC;
+
+ Buffer = makeArrayRef(Data).slice(Offset);
+ return Error::success();
+ }
+
+ uint32_t getLength() override { return Data.size(); }
+
+ Error writeBytes(uint32_t Offset, ArrayRef<uint8_t> Buffer) override {
+ if (Buffer.empty())
+ return Error::success();
+
+ // This is well-defined for any case except where offset is strictly
+ // greater than the current length. If offset is equal to the current
+ // length, we can still grow. If offset is beyond the current length, we
+ // would have to decide how to deal with the intermediate uninitialized
+ // bytes. So we punt on that case for simplicity and just say it's an
+ // error.
+ if (Offset > getLength())
+ return make_error<BinaryStreamError>(stream_error_code::invalid_offset);
+
+ uint32_t RequiredSize = Offset + Buffer.size();
+ if (RequiredSize > Data.size())
+ Data.resize(RequiredSize);
+
+ ::memcpy(Data.data() + Offset, Buffer.data(), Buffer.size());
+ return Error::success();
+ }
+
+ Error commit() override { return Error::success(); }
+
+ /// Return the properties of this stream.
+ virtual BinaryStreamFlags getFlags() const override {
+ return BSF_Write | BSF_Append;
+ }
+
+ MutableArrayRef<uint8_t> data() { return Data; }
+};
+
+/// An implementation of WritableBinaryStream backed by an llvm
+/// FileOutputBuffer.
+class FileBufferByteStream : public WritableBinaryStream {
+private:
+ class StreamImpl : public MutableBinaryByteStream {
+ public:
+ StreamImpl(std::unique_ptr<FileOutputBuffer> Buffer,
+ llvm::support::endianness Endian)
+ : MutableBinaryByteStream(
+ MutableArrayRef<uint8_t>(Buffer->getBufferStart(),
+ Buffer->getBufferEnd()),
+ Endian),
+ FileBuffer(std::move(Buffer)) {}
+
+ Error commit() override {
+ if (FileBuffer->commit())
+ return make_error<BinaryStreamError>(
+ stream_error_code::filesystem_error);
+ return Error::success();
+ }
+
+ /// Returns a pointer to the start of the buffer.
+ uint8_t *getBufferStart() const { return FileBuffer->getBufferStart(); }
+
+ /// Returns a pointer to the end of the buffer.
+ uint8_t *getBufferEnd() const { return FileBuffer->getBufferEnd(); }
+
+ private:
+ std::unique_ptr<FileOutputBuffer> FileBuffer;
+ };
+
+public:
+ FileBufferByteStream(std::unique_ptr<FileOutputBuffer> Buffer,
+ llvm::support::endianness Endian)
+ : Impl(std::move(Buffer), Endian) {}
+
+ llvm::support::endianness getEndian() const override {
+ return Impl.getEndian();
+ }
+
+ Error readBytes(uint32_t Offset, uint32_t Size,
+ ArrayRef<uint8_t> &Buffer) override {
+ return Impl.readBytes(Offset, Size, Buffer);
+ }
+
+ Error readLongestContiguousChunk(uint32_t Offset,
+ ArrayRef<uint8_t> &Buffer) override {
+ return Impl.readLongestContiguousChunk(Offset, Buffer);
+ }
+
+ uint32_t getLength() override { return Impl.getLength(); }
+
+ Error writeBytes(uint32_t Offset, ArrayRef<uint8_t> Data) override {
+ return Impl.writeBytes(Offset, Data);
+ }
+
+ Error commit() override { return Impl.commit(); }
+
+ /// Returns a pointer to the start of the buffer.
+ uint8_t *getBufferStart() const { return Impl.getBufferStart(); }
+
+ /// Returns a pointer to the end of the buffer.
+ uint8_t *getBufferEnd() const { return Impl.getBufferEnd(); }
+
+private:
+ StreamImpl Impl;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_BYTESTREAM_H
diff --git a/third_party/llvm-project/include/llvm/Support/BinaryStream.h b/third_party/llvm-project/include/llvm/Support/BinaryStream.h
new file mode 100644
index 000000000..fcf439855
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/BinaryStream.h
@@ -0,0 +1,101 @@
+//===- BinaryStream.h - Base interface for a stream of data -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_BINARYSTREAM_H
+#define LLVM_SUPPORT_BINARYSTREAM_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/BitmaskEnum.h"
+#include "llvm/Support/BinaryStreamError.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Error.h"
+#include <cstdint>
+
+namespace llvm {
+
+enum BinaryStreamFlags {
+ BSF_None = 0,
+ BSF_Write = 1, // Stream supports writing.
+ BSF_Append = 2, // Writing can occur at offset == length.
+ LLVM_MARK_AS_BITMASK_ENUM(/* LargestValue = */ BSF_Append)
+};
+
+/// An interface for accessing data in a stream-like format, but which
+/// discourages copying. Instead of specifying a buffer in which to copy
+/// data on a read, the API returns an ArrayRef to data owned by the stream's
+/// implementation. Since implementations may not necessarily store data in a
+/// single contiguous buffer (or even in memory at all), in such cases a it may
+/// be necessary for an implementation to cache such a buffer so that it can
+/// return it.
+class BinaryStream {
+public:
+ virtual ~BinaryStream() = default;
+
+ virtual llvm::support::endianness getEndian() const = 0;
+
+ /// Given an offset into the stream and a number of bytes, attempt to
+ /// read the bytes and set the output ArrayRef to point to data owned by the
+ /// stream.
+ virtual Error readBytes(uint32_t Offset, uint32_t Size,
+ ArrayRef<uint8_t> &Buffer) = 0;
+
+ /// Given an offset into the stream, read as much as possible without
+ /// copying any data.
+ virtual Error readLongestContiguousChunk(uint32_t Offset,
+ ArrayRef<uint8_t> &Buffer) = 0;
+
+ /// Return the number of bytes of data in this stream.
+ virtual uint32_t getLength() = 0;
+
+ /// Return the properties of this stream.
+ virtual BinaryStreamFlags getFlags() const { return BSF_None; }
+
+protected:
+ Error checkOffsetForRead(uint32_t Offset, uint32_t DataSize) {
+ if (Offset > getLength())
+ return make_error<BinaryStreamError>(stream_error_code::invalid_offset);
+ if (getLength() < DataSize + Offset)
+ return make_error<BinaryStreamError>(stream_error_code::stream_too_short);
+ return Error::success();
+ }
+};
+
+/// A BinaryStream which can be read from as well as written to. Note
+/// that writing to a BinaryStream always necessitates copying from the input
+/// buffer to the stream's backing store. Streams are assumed to be buffered
+/// so that to be portable it is necessary to call commit() on the stream when
+/// all data has been written.
+class WritableBinaryStream : public BinaryStream {
+public:
+ ~WritableBinaryStream() override = default;
+
+ /// Attempt to write the given bytes into the stream at the desired
+ /// offset. This will always necessitate a copy. Cannot shrink or grow the
+ /// stream, only writes into existing allocated space.
+ virtual Error writeBytes(uint32_t Offset, ArrayRef<uint8_t> Data) = 0;
+
+ /// For buffered streams, commits changes to the backing store.
+ virtual Error commit() = 0;
+
+ /// Return the properties of this stream.
+ BinaryStreamFlags getFlags() const override { return BSF_Write; }
+
+protected:
+ Error checkOffsetForWrite(uint32_t Offset, uint32_t DataSize) {
+ if (!(getFlags() & BSF_Append))
+ return checkOffsetForRead(Offset, DataSize);
+
+ if (Offset > getLength())
+ return make_error<BinaryStreamError>(stream_error_code::invalid_offset);
+ return Error::success();
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_BINARYSTREAM_H
diff --git a/third_party/llvm-project/include/llvm/Support/BinaryStreamError.h b/third_party/llvm-project/include/llvm/Support/BinaryStreamError.h
new file mode 100644
index 000000000..cf6e034ff
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/BinaryStreamError.h
@@ -0,0 +1,47 @@
+//===- BinaryStreamError.h - Error extensions for Binary Streams *- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_BINARYSTREAMERROR_H
+#define LLVM_SUPPORT_BINARYSTREAMERROR_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Error.h"
+
+#include <string>
+
+namespace llvm {
+enum class stream_error_code {
+ unspecified,
+ stream_too_short,
+ invalid_array_size,
+ invalid_offset,
+ filesystem_error
+};
+
+/// Base class for errors originating when parsing raw PDB files
+class BinaryStreamError : public ErrorInfo<BinaryStreamError> {
+public:
+ static char ID;
+ explicit BinaryStreamError(stream_error_code C);
+ explicit BinaryStreamError(StringRef Context);
+ BinaryStreamError(stream_error_code C, StringRef Context);
+
+ void log(raw_ostream &OS) const override;
+ std::error_code convertToErrorCode() const override;
+
+ StringRef getErrorMessage() const;
+
+ stream_error_code getErrorCode() const { return Code; }
+
+private:
+ std::string ErrMsg;
+ stream_error_code Code;
+};
+} // namespace llvm
+
+#endif // LLVM_SUPPORT_BINARYSTREAMERROR_H
diff --git a/third_party/llvm-project/include/llvm/Support/CBindingWrapping.h b/third_party/llvm-project/include/llvm/Support/CBindingWrapping.h
new file mode 100644
index 000000000..46d6b4e3f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/CBindingWrapping.h
@@ -0,0 +1,46 @@
+//===- llvm/Support/CBindingWrapping.h - C Interface Wrapping ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the wrapping macros for the C interface.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_CBINDINGWRAPPING_H
+#define LLVM_SUPPORT_CBINDINGWRAPPING_H
+
+#include "llvm-c/Types.h"
+#include "llvm/Support/Casting.h"
+
+#define DEFINE_SIMPLE_CONVERSION_FUNCTIONS(ty, ref) \
+ inline ty *unwrap(ref P) { \
+ return reinterpret_cast<ty*>(P); \
+ } \
+ \
+ inline ref wrap(const ty *P) { \
+ return reinterpret_cast<ref>(const_cast<ty*>(P)); \
+ }
+
+#define DEFINE_ISA_CONVERSION_FUNCTIONS(ty, ref) \
+ DEFINE_SIMPLE_CONVERSION_FUNCTIONS(ty, ref) \
+ \
+ template<typename T> \
+ inline T *unwrap(ref P) { \
+ return cast<T>(unwrap(P)); \
+ }
+
+#define DEFINE_STDCXX_CONVERSION_FUNCTIONS(ty, ref) \
+ DEFINE_SIMPLE_CONVERSION_FUNCTIONS(ty, ref) \
+ \
+ template<typename T> \
+ inline T *unwrap(ref P) { \
+ T *Q = (T*)unwrap(P); \
+ assert(Q && "Invalid cast!"); \
+ return Q; \
+ }
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Casting.h b/third_party/llvm-project/include/llvm/Support/Casting.h
new file mode 100644
index 000000000..46bdedb04
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Casting.h
@@ -0,0 +1,408 @@
+//===- llvm/Support/Casting.h - Allow flexible, checked, casts --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the isa<X>(), cast<X>(), dyn_cast<X>(), cast_or_null<X>(),
+// and dyn_cast_or_null<X>() templates.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_CASTING_H
+#define LLVM_SUPPORT_CASTING_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/type_traits.h"
+#include <cassert>
+#include <memory>
+#include <type_traits>
+
+namespace llvm {
+
+//===----------------------------------------------------------------------===//
+// isa<x> Support Templates
+//===----------------------------------------------------------------------===//
+
+// Define a template that can be specialized by smart pointers to reflect the
+// fact that they are automatically dereferenced, and are not involved with the
+// template selection process... the default implementation is a noop.
+//
+template<typename From> struct simplify_type {
+ using SimpleType = From; // The real type this represents...
+
+ // An accessor to get the real value...
+ static SimpleType &getSimplifiedValue(From &Val) { return Val; }
+};
+
+template<typename From> struct simplify_type<const From> {
+ using NonConstSimpleType = typename simplify_type<From>::SimpleType;
+ using SimpleType =
+ typename add_const_past_pointer<NonConstSimpleType>::type;
+ using RetType =
+ typename add_lvalue_reference_if_not_pointer<SimpleType>::type;
+
+ static RetType getSimplifiedValue(const From& Val) {
+ return simplify_type<From>::getSimplifiedValue(const_cast<From&>(Val));
+ }
+};
+
+// The core of the implementation of isa<X> is here; To and From should be
+// the names of classes. This template can be specialized to customize the
+// implementation of isa<> without rewriting it from scratch.
+template <typename To, typename From, typename Enabler = void>
+struct isa_impl {
+ static inline bool doit(const From &Val) {
+ return To::classof(&Val);
+ }
+};
+
+/// Always allow upcasts, and perform no dynamic check for them.
+template <typename To, typename From>
+struct isa_impl<
+ To, From, typename std::enable_if<std::is_base_of<To, From>::value>::type> {
+ static inline bool doit(const From &) { return true; }
+};
+
+template <typename To, typename From> struct isa_impl_cl {
+ static inline bool doit(const From &Val) {
+ return isa_impl<To, From>::doit(Val);
+ }
+};
+
+template <typename To, typename From> struct isa_impl_cl<To, const From> {
+ static inline bool doit(const From &Val) {
+ return isa_impl<To, From>::doit(Val);
+ }
+};
+
+template <typename To, typename From>
+struct isa_impl_cl<To, const std::unique_ptr<From>> {
+ static inline bool doit(const std::unique_ptr<From> &Val) {
+ assert(Val && "isa<> used on a null pointer");
+ return isa_impl_cl<To, From>::doit(*Val);
+ }
+};
+
+template <typename To, typename From> struct isa_impl_cl<To, From*> {
+ static inline bool doit(const From *Val) {
+ assert(Val && "isa<> used on a null pointer");
+ return isa_impl<To, From>::doit(*Val);
+ }
+};
+
+template <typename To, typename From> struct isa_impl_cl<To, From*const> {
+ static inline bool doit(const From *Val) {
+ assert(Val && "isa<> used on a null pointer");
+ return isa_impl<To, From>::doit(*Val);
+ }
+};
+
+template <typename To, typename From> struct isa_impl_cl<To, const From*> {
+ static inline bool doit(const From *Val) {
+ assert(Val && "isa<> used on a null pointer");
+ return isa_impl<To, From>::doit(*Val);
+ }
+};
+
+template <typename To, typename From> struct isa_impl_cl<To, const From*const> {
+ static inline bool doit(const From *Val) {
+ assert(Val && "isa<> used on a null pointer");
+ return isa_impl<To, From>::doit(*Val);
+ }
+};
+
+template<typename To, typename From, typename SimpleFrom>
+struct isa_impl_wrap {
+ // When From != SimplifiedType, we can simplify the type some more by using
+ // the simplify_type template.
+ static bool doit(const From &Val) {
+ return isa_impl_wrap<To, SimpleFrom,
+ typename simplify_type<SimpleFrom>::SimpleType>::doit(
+ simplify_type<const From>::getSimplifiedValue(Val));
+ }
+};
+
+template<typename To, typename FromTy>
+struct isa_impl_wrap<To, FromTy, FromTy> {
+ // When From == SimpleType, we are as simple as we are going to get.
+ static bool doit(const FromTy &Val) {
+ return isa_impl_cl<To,FromTy>::doit(Val);
+ }
+};
+
+// isa<X> - Return true if the parameter to the template is an instance of the
+// template type argument. Used like this:
+//
+// if (isa<Type>(myVal)) { ... }
+//
+template <class X, class Y> LLVM_NODISCARD inline bool isa(const Y &Val) {
+ return isa_impl_wrap<X, const Y,
+ typename simplify_type<const Y>::SimpleType>::doit(Val);
+}
+
+// isa_and_nonnull<X> - Functionally identical to isa, except that a null value
+// is accepted.
+//
+template <class X, class Y>
+LLVM_NODISCARD inline bool isa_and_nonnull(const Y &Val) {
+ if (!Val)
+ return false;
+ return isa<X>(Val);
+}
+
+//===----------------------------------------------------------------------===//
+// cast<x> Support Templates
+//===----------------------------------------------------------------------===//
+
+template<class To, class From> struct cast_retty;
+
+// Calculate what type the 'cast' function should return, based on a requested
+// type of To and a source type of From.
+template<class To, class From> struct cast_retty_impl {
+ using ret_type = To &; // Normal case, return Ty&
+};
+template<class To, class From> struct cast_retty_impl<To, const From> {
+ using ret_type = const To &; // Normal case, return Ty&
+};
+
+template<class To, class From> struct cast_retty_impl<To, From*> {
+ using ret_type = To *; // Pointer arg case, return Ty*
+};
+
+template<class To, class From> struct cast_retty_impl<To, const From*> {
+ using ret_type = const To *; // Constant pointer arg case, return const Ty*
+};
+
+template<class To, class From> struct cast_retty_impl<To, const From*const> {
+ using ret_type = const To *; // Constant pointer arg case, return const Ty*
+};
+
+template <class To, class From>
+struct cast_retty_impl<To, std::unique_ptr<From>> {
+private:
+ using PointerType = typename cast_retty_impl<To, From *>::ret_type;
+ using ResultType = typename std::remove_pointer<PointerType>::type;
+
+public:
+ using ret_type = std::unique_ptr<ResultType>;
+};
+
+template<class To, class From, class SimpleFrom>
+struct cast_retty_wrap {
+ // When the simplified type and the from type are not the same, use the type
+ // simplifier to reduce the type, then reuse cast_retty_impl to get the
+ // resultant type.
+ using ret_type = typename cast_retty<To, SimpleFrom>::ret_type;
+};
+
+template<class To, class FromTy>
+struct cast_retty_wrap<To, FromTy, FromTy> {
+ // When the simplified type is equal to the from type, use it directly.
+ using ret_type = typename cast_retty_impl<To,FromTy>::ret_type;
+};
+
+template<class To, class From>
+struct cast_retty {
+ using ret_type = typename cast_retty_wrap<
+ To, From, typename simplify_type<From>::SimpleType>::ret_type;
+};
+
+// Ensure the non-simple values are converted using the simplify_type template
+// that may be specialized by smart pointers...
+//
+template<class To, class From, class SimpleFrom> struct cast_convert_val {
+ // This is not a simple type, use the template to simplify it...
+ static typename cast_retty<To, From>::ret_type doit(From &Val) {
+ return cast_convert_val<To, SimpleFrom,
+ typename simplify_type<SimpleFrom>::SimpleType>::doit(
+ simplify_type<From>::getSimplifiedValue(Val));
+ }
+};
+
+template<class To, class FromTy> struct cast_convert_val<To,FromTy,FromTy> {
+ // This _is_ a simple type, just cast it.
+ static typename cast_retty<To, FromTy>::ret_type doit(const FromTy &Val) {
+ typename cast_retty<To, FromTy>::ret_type Res2
+ = (typename cast_retty<To, FromTy>::ret_type)const_cast<FromTy&>(Val);
+ return Res2;
+ }
+};
+
+template <class X> struct is_simple_type {
+ static const bool value =
+ std::is_same<X, typename simplify_type<X>::SimpleType>::value;
+};
+
+// cast<X> - Return the argument parameter cast to the specified type. This
+// casting operator asserts that the type is correct, so it does not return null
+// on failure. It does not allow a null argument (use cast_or_null for that).
+// It is typically used like this:
+//
+// cast<Instruction>(myVal)->getParent()
+//
+template <class X, class Y>
+inline typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, const Y>::ret_type>::type
+cast(const Y &Val) {
+ assert(isa<X>(Val) && "cast<Ty>() argument of incompatible type!");
+ return cast_convert_val<
+ X, const Y, typename simplify_type<const Y>::SimpleType>::doit(Val);
+}
+
+template <class X, class Y>
+inline typename cast_retty<X, Y>::ret_type cast(Y &Val) {
+ assert(isa<X>(Val) && "cast<Ty>() argument of incompatible type!");
+ return cast_convert_val<X, Y,
+ typename simplify_type<Y>::SimpleType>::doit(Val);
+}
+
+template <class X, class Y>
+inline typename cast_retty<X, Y *>::ret_type cast(Y *Val) {
+ assert(isa<X>(Val) && "cast<Ty>() argument of incompatible type!");
+ return cast_convert_val<X, Y*,
+ typename simplify_type<Y*>::SimpleType>::doit(Val);
+}
+
+template <class X, class Y>
+inline typename cast_retty<X, std::unique_ptr<Y>>::ret_type
+cast(std::unique_ptr<Y> &&Val) {
+ assert(isa<X>(Val.get()) && "cast<Ty>() argument of incompatible type!");
+ using ret_type = typename cast_retty<X, std::unique_ptr<Y>>::ret_type;
+ return ret_type(
+ cast_convert_val<X, Y *, typename simplify_type<Y *>::SimpleType>::doit(
+ Val.release()));
+}
+
+// cast_or_null<X> - Functionally identical to cast, except that a null value is
+// accepted.
+//
+template <class X, class Y>
+LLVM_NODISCARD inline
+ typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, const Y>::ret_type>::type
+ cast_or_null(const Y &Val) {
+ if (!Val)
+ return nullptr;
+ assert(isa<X>(Val) && "cast_or_null<Ty>() argument of incompatible type!");
+ return cast<X>(Val);
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline
+ typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, Y>::ret_type>::type
+ cast_or_null(Y &Val) {
+ if (!Val)
+ return nullptr;
+ assert(isa<X>(Val) && "cast_or_null<Ty>() argument of incompatible type!");
+ return cast<X>(Val);
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline typename cast_retty<X, Y *>::ret_type
+cast_or_null(Y *Val) {
+ if (!Val) return nullptr;
+ assert(isa<X>(Val) && "cast_or_null<Ty>() argument of incompatible type!");
+ return cast<X>(Val);
+}
+
+template <class X, class Y>
+inline typename cast_retty<X, std::unique_ptr<Y>>::ret_type
+cast_or_null(std::unique_ptr<Y> &&Val) {
+ if (!Val)
+ return nullptr;
+ return cast<X>(std::move(Val));
+}
+
+// dyn_cast<X> - Return the argument parameter cast to the specified type. This
+// casting operator returns null if the argument is of the wrong type, so it can
+// be used to test for a type as well as cast if successful. This should be
+// used in the context of an if statement like this:
+//
+// if (const Instruction *I = dyn_cast<Instruction>(myVal)) { ... }
+//
+
+template <class X, class Y>
+LLVM_NODISCARD inline
+ typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, const Y>::ret_type>::type
+ dyn_cast(const Y &Val) {
+ return isa<X>(Val) ? cast<X>(Val) : nullptr;
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline typename cast_retty<X, Y>::ret_type dyn_cast(Y &Val) {
+ return isa<X>(Val) ? cast<X>(Val) : nullptr;
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline typename cast_retty<X, Y *>::ret_type dyn_cast(Y *Val) {
+ return isa<X>(Val) ? cast<X>(Val) : nullptr;
+}
+
+// dyn_cast_or_null<X> - Functionally identical to dyn_cast, except that a null
+// value is accepted.
+//
+template <class X, class Y>
+LLVM_NODISCARD inline
+ typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, const Y>::ret_type>::type
+ dyn_cast_or_null(const Y &Val) {
+ return (Val && isa<X>(Val)) ? cast<X>(Val) : nullptr;
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline
+ typename std::enable_if<!is_simple_type<Y>::value,
+ typename cast_retty<X, Y>::ret_type>::type
+ dyn_cast_or_null(Y &Val) {
+ return (Val && isa<X>(Val)) ? cast<X>(Val) : nullptr;
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline typename cast_retty<X, Y *>::ret_type
+dyn_cast_or_null(Y *Val) {
+ return (Val && isa<X>(Val)) ? cast<X>(Val) : nullptr;
+}
+
+// unique_dyn_cast<X> - Given a unique_ptr<Y>, try to return a unique_ptr<X>,
+// taking ownership of the input pointer iff isa<X>(Val) is true. If the
+// cast is successful, From refers to nullptr on exit and the casted value
+// is returned. If the cast is unsuccessful, the function returns nullptr
+// and From is unchanged.
+template <class X, class Y>
+LLVM_NODISCARD inline auto unique_dyn_cast(std::unique_ptr<Y> &Val)
+ -> decltype(cast<X>(Val)) {
+ if (!isa<X>(Val))
+ return nullptr;
+ return cast<X>(std::move(Val));
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline auto unique_dyn_cast(std::unique_ptr<Y> &&Val)
+ -> decltype(cast<X>(Val)) {
+ return unique_dyn_cast<X, Y>(Val);
+}
+
+// dyn_cast_or_null<X> - Functionally identical to unique_dyn_cast, except that
+// a null value is accepted.
+template <class X, class Y>
+LLVM_NODISCARD inline auto unique_dyn_cast_or_null(std::unique_ptr<Y> &Val)
+ -> decltype(cast<X>(Val)) {
+ if (!Val)
+ return nullptr;
+ return unique_dyn_cast<X, Y>(Val);
+}
+
+template <class X, class Y>
+LLVM_NODISCARD inline auto unique_dyn_cast_or_null(std::unique_ptr<Y> &&Val)
+ -> decltype(cast<X>(Val)) {
+ return unique_dyn_cast_or_null<X, Y>(Val);
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_CASTING_H
diff --git a/third_party/llvm-project/include/llvm/Support/Chrono.h b/third_party/llvm-project/include/llvm/Support/Chrono.h
new file mode 100644
index 000000000..334ab6083
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Chrono.h
@@ -0,0 +1,171 @@
+//===- llvm/Support/Chrono.h - Utilities for Timing Manipulation-*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_CHRONO_H
+#define LLVM_SUPPORT_CHRONO_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/FormatProviders.h"
+
+#include <chrono>
+#include <ctime>
+
+namespace llvm {
+
+class raw_ostream;
+
+namespace sys {
+
+/// A time point on the system clock. This is provided for two reasons:
+/// - to insulate us agains subtle differences in behavoir to differences in
+/// system clock precision (which is implementation-defined and differs between
+/// platforms).
+/// - to shorten the type name
+/// The default precision is nanoseconds. If need a specific precision specify
+/// it explicitly. If unsure, use the default. If you need a time point on a
+/// clock other than the system_clock, use std::chrono directly.
+template <typename D = std::chrono::nanoseconds>
+using TimePoint = std::chrono::time_point<std::chrono::system_clock, D>;
+
+/// Convert a TimePoint to std::time_t
+inline std::time_t toTimeT(TimePoint<> TP) {
+ using namespace std::chrono;
+ return system_clock::to_time_t(
+ time_point_cast<system_clock::time_point::duration>(TP));
+}
+
+/// Convert a std::time_t to a TimePoint
+inline TimePoint<std::chrono::seconds>
+toTimePoint(std::time_t T) {
+ using namespace std::chrono;
+ return time_point_cast<seconds>(system_clock::from_time_t(T));
+}
+
+/// Convert a std::time_t + nanoseconds to a TimePoint
+inline TimePoint<>
+toTimePoint(std::time_t T, uint32_t nsec) {
+ using namespace std::chrono;
+ return time_point_cast<nanoseconds>(system_clock::from_time_t(T))
+ + nanoseconds(nsec);
+}
+
+} // namespace sys
+
+raw_ostream &operator<<(raw_ostream &OS, sys::TimePoint<> TP);
+
+/// Format provider for TimePoint<>
+///
+/// The options string is a strftime format string, with extensions:
+/// - %L is millis: 000-999
+/// - %f is micros: 000000-999999
+/// - %N is nanos: 000000000 - 999999999
+///
+/// If no options are given, the default format is "%Y-%m-%d %H:%M:%S.%N".
+template <>
+struct format_provider<sys::TimePoint<>> {
+ static void format(const sys::TimePoint<> &TP, llvm::raw_ostream &OS,
+ StringRef Style);
+};
+
+/// Implementation of format_provider<T> for duration types.
+///
+/// The options string of a duration type has the grammar:
+///
+/// duration_options ::= [unit][show_unit [number_options]]
+/// unit ::= `h`|`m`|`s`|`ms|`us`|`ns`
+/// show_unit ::= `+` | `-`
+/// number_options ::= options string for a integral or floating point type
+///
+/// Examples
+/// =================================
+/// | options | Input | Output |
+/// =================================
+/// | "" | 1s | 1 s |
+/// | "ms" | 1s | 1000 ms |
+/// | "ms-" | 1s | 1000 |
+/// | "ms-n" | 1s | 1,000 |
+/// | "" | 1.0s | 1.00 s |
+/// =================================
+///
+/// If the unit of the duration type is not one of the units specified above,
+/// it is still possible to format it, provided you explicitly request a
+/// display unit or you request that the unit is not displayed.
+
+namespace detail {
+template <typename Period> struct unit { static const char value[]; };
+template <typename Period> const char unit<Period>::value[] = "";
+
+template <> struct unit<std::ratio<3600>> { static const char value[]; };
+template <> struct unit<std::ratio<60>> { static const char value[]; };
+template <> struct unit<std::ratio<1>> { static const char value[]; };
+template <> struct unit<std::milli> { static const char value[]; };
+template <> struct unit<std::micro> { static const char value[]; };
+template <> struct unit<std::nano> { static const char value[]; };
+} // namespace detail
+
+template <typename Rep, typename Period>
+struct format_provider<std::chrono::duration<Rep, Period>> {
+private:
+ typedef std::chrono::duration<Rep, Period> Dur;
+ typedef typename std::conditional<
+ std::chrono::treat_as_floating_point<Rep>::value, double, intmax_t>::type
+ InternalRep;
+
+ template <typename AsPeriod> static InternalRep getAs(const Dur &D) {
+ using namespace std::chrono;
+ return duration_cast<duration<InternalRep, AsPeriod>>(D).count();
+ }
+
+ static std::pair<InternalRep, StringRef> consumeUnit(StringRef &Style,
+ const Dur &D) {
+ using namespace std::chrono;
+ if (Style.consume_front("ns"))
+ return {getAs<std::nano>(D), "ns"};
+ if (Style.consume_front("us"))
+ return {getAs<std::micro>(D), "us"};
+ if (Style.consume_front("ms"))
+ return {getAs<std::milli>(D), "ms"};
+ if (Style.consume_front("s"))
+ return {getAs<std::ratio<1>>(D), "s"};
+ if (Style.consume_front("m"))
+ return {getAs<std::ratio<60>>(D), "m"};
+ if (Style.consume_front("h"))
+ return {getAs<std::ratio<3600>>(D), "h"};
+ return {D.count(), detail::unit<Period>::value};
+ }
+
+ static bool consumeShowUnit(StringRef &Style) {
+ if (Style.empty())
+ return true;
+ if (Style.consume_front("-"))
+ return false;
+ if (Style.consume_front("+"))
+ return true;
+ assert(0 && "Unrecognised duration format");
+ return true;
+ }
+
+public:
+ static void format(const Dur &D, llvm::raw_ostream &Stream, StringRef Style) {
+ InternalRep count;
+ StringRef unit;
+ std::tie(count, unit) = consumeUnit(Style, D);
+ bool show_unit = consumeShowUnit(Style);
+
+ format_provider<InternalRep>::format(count, Stream, Style);
+
+ if (show_unit) {
+ assert(!unit.empty());
+ Stream << " " << unit;
+ }
+ }
+};
+
+} // namespace llvm
+
+#endif // LLVM_SUPPORT_CHRONO_H
diff --git a/third_party/llvm-project/include/llvm/Support/CodeGen.h b/third_party/llvm-project/include/llvm/Support/CodeGen.h
new file mode 100644
index 000000000..a3f423e55
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/CodeGen.h
@@ -0,0 +1,67 @@
+//===-- llvm/Support/CodeGen.h - CodeGen Concepts ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file define some types which define code generation concepts. For
+// example, relocation model.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_CODEGEN_H
+#define LLVM_SUPPORT_CODEGEN_H
+
+namespace llvm {
+
+ // Relocation model types.
+ namespace Reloc {
+ // Cannot be named PIC due to collision with -DPIC
+ enum Model { Static, PIC_, DynamicNoPIC, ROPI, RWPI, ROPI_RWPI };
+ }
+
+ // Code model types.
+ namespace CodeModel {
+ // Sync changes with CodeGenCWrappers.h.
+ enum Model { Tiny, Small, Kernel, Medium, Large };
+ }
+
+ namespace PICLevel {
+ // This is used to map -fpic/-fPIC.
+ enum Level { NotPIC=0, SmallPIC=1, BigPIC=2 };
+ }
+
+ namespace PIELevel {
+ enum Level { Default=0, Small=1, Large=2 };
+ }
+
+ // TLS models.
+ namespace TLSModel {
+ enum Model {
+ GeneralDynamic,
+ LocalDynamic,
+ InitialExec,
+ LocalExec
+ };
+ }
+
+ // Code generation optimization level.
+ namespace CodeGenOpt {
+ enum Level {
+ None = 0, // -O0
+ Less = 1, // -O1
+ Default = 2, // -O2, -Os
+ Aggressive = 3 // -O3
+ };
+ }
+
+ // Specify effect of frame pointer elimination optimization.
+ namespace FramePointer {
+ enum FP {All, NonLeaf, None};
+ }
+
+} // end llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/CommandLine.h b/third_party/llvm-project/include/llvm/Support/CommandLine.h
new file mode 100644
index 000000000..4eb30f548
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/CommandLine.h
@@ -0,0 +1 @@
+// XXX BINARYEN - we don't need this, but stuff imports it
diff --git a/third_party/llvm-project/include/llvm/Support/Compiler.h b/third_party/llvm-project/include/llvm/Support/Compiler.h
new file mode 100644
index 000000000..cb7e57d4c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Compiler.h
@@ -0,0 +1,585 @@
+//===-- llvm/Support/Compiler.h - Compiler abstraction support --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines several macros, based on the current compiler. This allows
+// use of compiler-specific features in a way that remains portable. This header
+// can be included from either C or C++.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_COMPILER_H
+#define LLVM_SUPPORT_COMPILER_H
+
+#include "llvm/Config/llvm-config.h"
+
+#ifdef __cplusplus
+#include <new>
+#endif
+#include <stddef.h>
+
+#if defined(_MSC_VER)
+#include <sal.h>
+#endif
+
+#ifndef __has_feature
+# define __has_feature(x) 0
+#endif
+
+#ifndef __has_extension
+# define __has_extension(x) 0
+#endif
+
+#ifndef __has_attribute
+# define __has_attribute(x) 0
+#endif
+
+#ifndef __has_builtin
+# define __has_builtin(x) 0
+#endif
+
+// Only use __has_cpp_attribute in C++ mode. GCC defines __has_cpp_attribute in
+// C mode, but the :: in __has_cpp_attribute(scoped::attribute) is invalid.
+#ifndef LLVM_HAS_CPP_ATTRIBUTE
+#if defined(__cplusplus) && defined(__has_cpp_attribute)
+# define LLVM_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
+#else
+# define LLVM_HAS_CPP_ATTRIBUTE(x) 0
+#endif
+#endif
+
+/// \macro LLVM_GNUC_PREREQ
+/// Extend the default __GNUC_PREREQ even if glibc's features.h isn't
+/// available.
+#ifndef LLVM_GNUC_PREREQ
+# if defined(__GNUC__) && defined(__GNUC_MINOR__) && defined(__GNUC_PATCHLEVEL__)
+# define LLVM_GNUC_PREREQ(maj, min, patch) \
+ ((__GNUC__ << 20) + (__GNUC_MINOR__ << 10) + __GNUC_PATCHLEVEL__ >= \
+ ((maj) << 20) + ((min) << 10) + (patch))
+# elif defined(__GNUC__) && defined(__GNUC_MINOR__)
+# define LLVM_GNUC_PREREQ(maj, min, patch) \
+ ((__GNUC__ << 20) + (__GNUC_MINOR__ << 10) >= ((maj) << 20) + ((min) << 10))
+# else
+# define LLVM_GNUC_PREREQ(maj, min, patch) 0
+# endif
+#endif
+
+/// \macro LLVM_MSC_PREREQ
+/// Is the compiler MSVC of at least the specified version?
+/// The common \param version values to check for are:
+/// * 1910: VS2017, version 15.1 & 15.2
+/// * 1911: VS2017, version 15.3 & 15.4
+/// * 1912: VS2017, version 15.5
+/// * 1913: VS2017, version 15.6
+/// * 1914: VS2017, version 15.7
+/// * 1915: VS2017, version 15.8
+/// * 1916: VS2017, version 15.9
+/// * 1920: VS2019, version 16.0
+/// * 1921: VS2019, version 16.1
+#ifdef _MSC_VER
+#define LLVM_MSC_PREREQ(version) (_MSC_VER >= (version))
+
+// We require at least MSVC 2017.
+#if !LLVM_MSC_PREREQ(1910)
+#error LLVM requires at least MSVC 2017.
+#endif
+
+#else
+#define LLVM_MSC_PREREQ(version) 0
+#endif
+
+/// Does the compiler support ref-qualifiers for *this?
+///
+/// Sadly, this is separate from just rvalue reference support because GCC
+/// and MSVC implemented this later than everything else.
+#if __has_feature(cxx_rvalue_references) || LLVM_GNUC_PREREQ(4, 8, 1)
+#define LLVM_HAS_RVALUE_REFERENCE_THIS 1
+#else
+#define LLVM_HAS_RVALUE_REFERENCE_THIS 0
+#endif
+
+/// Expands to '&' if ref-qualifiers for *this are supported.
+///
+/// This can be used to provide lvalue/rvalue overrides of member functions.
+/// The rvalue override should be guarded by LLVM_HAS_RVALUE_REFERENCE_THIS
+#if LLVM_HAS_RVALUE_REFERENCE_THIS
+#define LLVM_LVALUE_FUNCTION &
+#else
+#define LLVM_LVALUE_FUNCTION
+#endif
+
+/// LLVM_LIBRARY_VISIBILITY - If a class marked with this attribute is linked
+/// into a shared library, then the class should be private to the library and
+/// not accessible from outside it. Can also be used to mark variables and
+/// functions, making them private to any shared library they are linked into.
+/// On PE/COFF targets, library visibility is the default, so this isn't needed.
+#if (__has_attribute(visibility) || LLVM_GNUC_PREREQ(4, 0, 0)) && \
+ !defined(__MINGW32__) && !defined(__CYGWIN__) && !defined(_WIN32)
+#define LLVM_LIBRARY_VISIBILITY __attribute__ ((visibility("hidden")))
+#else
+#define LLVM_LIBRARY_VISIBILITY
+#endif
+
+#if defined(__GNUC__)
+#define LLVM_PREFETCH(addr, rw, locality) __builtin_prefetch(addr, rw, locality)
+#else
+#define LLVM_PREFETCH(addr, rw, locality)
+#endif
+
+#if __has_attribute(used) || LLVM_GNUC_PREREQ(3, 1, 0)
+#define LLVM_ATTRIBUTE_USED __attribute__((__used__))
+#else
+#define LLVM_ATTRIBUTE_USED
+#endif
+
+/// LLVM_NODISCARD - Warn if a type or return value is discarded.
+
+// Use the 'nodiscard' attribute in C++17 or newer mode.
+#if __cplusplus > 201402L && LLVM_HAS_CPP_ATTRIBUTE(nodiscard)
+#define LLVM_NODISCARD [[nodiscard]]
+#elif LLVM_HAS_CPP_ATTRIBUTE(clang::warn_unused_result)
+#define LLVM_NODISCARD [[clang::warn_unused_result]]
+// Clang in C++14 mode claims that it has the 'nodiscard' attribute, but also
+// warns in the pedantic mode that 'nodiscard' is a C++17 extension (PR33518).
+// Use the 'nodiscard' attribute in C++14 mode only with GCC.
+// TODO: remove this workaround when PR33518 is resolved.
+#elif defined(__GNUC__) && LLVM_HAS_CPP_ATTRIBUTE(nodiscard)
+#define LLVM_NODISCARD [[nodiscard]]
+#else
+#define LLVM_NODISCARD
+#endif
+
+// Indicate that a non-static, non-const C++ member function reinitializes
+// the entire object to a known state, independent of the previous state of
+// the object.
+//
+// The clang-tidy check bugprone-use-after-move recognizes this attribute as a
+// marker that a moved-from object has left the indeterminate state and can be
+// reused.
+#if LLVM_HAS_CPP_ATTRIBUTE(clang::reinitializes)
+#define LLVM_ATTRIBUTE_REINITIALIZES [[clang::reinitializes]]
+#else
+#define LLVM_ATTRIBUTE_REINITIALIZES
+#endif
+
+// Some compilers warn about unused functions. When a function is sometimes
+// used or not depending on build settings (e.g. a function only called from
+// within "assert"), this attribute can be used to suppress such warnings.
+//
+// However, it shouldn't be used for unused *variables*, as those have a much
+// more portable solution:
+// (void)unused_var_name;
+// Prefer cast-to-void wherever it is sufficient.
+#if __has_attribute(unused) || LLVM_GNUC_PREREQ(3, 1, 0)
+#define LLVM_ATTRIBUTE_UNUSED __attribute__((__unused__))
+#else
+#define LLVM_ATTRIBUTE_UNUSED
+#endif
+
+// FIXME: Provide this for PE/COFF targets.
+#if (__has_attribute(weak) || LLVM_GNUC_PREREQ(4, 0, 0)) && \
+ (!defined(__MINGW32__) && !defined(__CYGWIN__) && !defined(_WIN32))
+#define LLVM_ATTRIBUTE_WEAK __attribute__((__weak__))
+#else
+#define LLVM_ATTRIBUTE_WEAK
+#endif
+
+// Prior to clang 3.2, clang did not accept any spelling of
+// __has_attribute(const), so assume it is supported.
+#if defined(__clang__) || defined(__GNUC__)
+// aka 'CONST' but following LLVM Conventions.
+#define LLVM_READNONE __attribute__((__const__))
+#else
+#define LLVM_READNONE
+#endif
+
+#if __has_attribute(pure) || defined(__GNUC__)
+// aka 'PURE' but following LLVM Conventions.
+#define LLVM_READONLY __attribute__((__pure__))
+#else
+#define LLVM_READONLY
+#endif
+
+#if __has_builtin(__builtin_expect) || LLVM_GNUC_PREREQ(4, 0, 0)
+#define LLVM_LIKELY(EXPR) __builtin_expect((bool)(EXPR), true)
+#define LLVM_UNLIKELY(EXPR) __builtin_expect((bool)(EXPR), false)
+#else
+#define LLVM_LIKELY(EXPR) (EXPR)
+#define LLVM_UNLIKELY(EXPR) (EXPR)
+#endif
+
+/// LLVM_ATTRIBUTE_NOINLINE - On compilers where we have a directive to do so,
+/// mark a method "not for inlining".
+#if __has_attribute(noinline) || LLVM_GNUC_PREREQ(3, 4, 0)
+#define LLVM_ATTRIBUTE_NOINLINE __attribute__((noinline))
+#elif defined(_MSC_VER)
+#define LLVM_ATTRIBUTE_NOINLINE __declspec(noinline)
+#else
+#define LLVM_ATTRIBUTE_NOINLINE
+#endif
+
+/// LLVM_ATTRIBUTE_ALWAYS_INLINE - On compilers where we have a directive to do
+/// so, mark a method "always inline" because it is performance sensitive. GCC
+/// 3.4 supported this but is buggy in various cases and produces unimplemented
+/// errors, just use it in GCC 4.0 and later.
+#if __has_attribute(always_inline) || LLVM_GNUC_PREREQ(4, 0, 0)
+#define LLVM_ATTRIBUTE_ALWAYS_INLINE __attribute__((always_inline))
+#elif defined(_MSC_VER)
+#define LLVM_ATTRIBUTE_ALWAYS_INLINE __forceinline
+#else
+#define LLVM_ATTRIBUTE_ALWAYS_INLINE
+#endif
+
+#ifdef __GNUC__
+#define LLVM_ATTRIBUTE_NORETURN __attribute__((noreturn))
+#elif defined(_MSC_VER)
+#define LLVM_ATTRIBUTE_NORETURN __declspec(noreturn)
+#else
+#define LLVM_ATTRIBUTE_NORETURN
+#endif
+
+#if __has_attribute(returns_nonnull) || LLVM_GNUC_PREREQ(4, 9, 0)
+#define LLVM_ATTRIBUTE_RETURNS_NONNULL __attribute__((returns_nonnull))
+#elif defined(_MSC_VER)
+#define LLVM_ATTRIBUTE_RETURNS_NONNULL _Ret_notnull_
+#else
+#define LLVM_ATTRIBUTE_RETURNS_NONNULL
+#endif
+
+/// \macro LLVM_ATTRIBUTE_RETURNS_NOALIAS Used to mark a function as returning a
+/// pointer that does not alias any other valid pointer.
+#ifdef __GNUC__
+#define LLVM_ATTRIBUTE_RETURNS_NOALIAS __attribute__((__malloc__))
+#elif defined(_MSC_VER)
+#define LLVM_ATTRIBUTE_RETURNS_NOALIAS __declspec(restrict)
+#else
+#define LLVM_ATTRIBUTE_RETURNS_NOALIAS
+#endif
+
+/// LLVM_FALLTHROUGH - Mark fallthrough cases in switch statements.
+#if __cplusplus > 201402L && LLVM_HAS_CPP_ATTRIBUTE(fallthrough)
+#define LLVM_FALLTHROUGH [[fallthrough]]
+#elif LLVM_HAS_CPP_ATTRIBUTE(gnu::fallthrough)
+#define LLVM_FALLTHROUGH [[gnu::fallthrough]]
+#elif __has_attribute(fallthrough)
+#define LLVM_FALLTHROUGH __attribute__((fallthrough))
+#elif LLVM_HAS_CPP_ATTRIBUTE(clang::fallthrough)
+#define LLVM_FALLTHROUGH [[clang::fallthrough]]
+#else
+#define LLVM_FALLTHROUGH
+#endif
+
+/// LLVM_REQUIRE_CONSTANT_INITIALIZATION - Apply this to globals to ensure that
+/// they are constant initialized.
+#if LLVM_HAS_CPP_ATTRIBUTE(clang::require_constant_initialization)
+#define LLVM_REQUIRE_CONSTANT_INITIALIZATION \
+ [[clang::require_constant_initialization]]
+#else
+#define LLVM_REQUIRE_CONSTANT_INITIALIZATION
+#endif
+
+/// LLVM_EXTENSION - Support compilers where we have a keyword to suppress
+/// pedantic diagnostics.
+#ifdef __GNUC__
+#define LLVM_EXTENSION __extension__
+#else
+#define LLVM_EXTENSION
+#endif
+
+// LLVM_ATTRIBUTE_DEPRECATED(decl, "message")
+#if __has_feature(attribute_deprecated_with_message)
+# define LLVM_ATTRIBUTE_DEPRECATED(decl, message) \
+ decl __attribute__((deprecated(message)))
+#elif defined(__GNUC__)
+# define LLVM_ATTRIBUTE_DEPRECATED(decl, message) \
+ decl __attribute__((deprecated))
+#elif defined(_MSC_VER)
+# define LLVM_ATTRIBUTE_DEPRECATED(decl, message) \
+ __declspec(deprecated(message)) decl
+#else
+# define LLVM_ATTRIBUTE_DEPRECATED(decl, message) \
+ decl
+#endif
+
+/// LLVM_BUILTIN_UNREACHABLE - On compilers which support it, expands
+/// to an expression which states that it is undefined behavior for the
+/// compiler to reach this point. Otherwise is not defined.
+#if __has_builtin(__builtin_unreachable) || LLVM_GNUC_PREREQ(4, 5, 0)
+# define LLVM_BUILTIN_UNREACHABLE __builtin_unreachable()
+#elif defined(_MSC_VER)
+# define LLVM_BUILTIN_UNREACHABLE __assume(false)
+#endif
+
+/// LLVM_BUILTIN_TRAP - On compilers which support it, expands to an expression
+/// which causes the program to exit abnormally.
+#if __has_builtin(__builtin_trap) || LLVM_GNUC_PREREQ(4, 3, 0)
+# define LLVM_BUILTIN_TRAP __builtin_trap()
+#elif defined(_MSC_VER)
+// The __debugbreak intrinsic is supported by MSVC, does not require forward
+// declarations involving platform-specific typedefs (unlike RaiseException),
+// results in a call to vectored exception handlers, and encodes to a short
+// instruction that still causes the trapping behavior we want.
+# define LLVM_BUILTIN_TRAP __debugbreak()
+#else
+# define LLVM_BUILTIN_TRAP *(volatile int*)0x11 = 0
+#endif
+
+/// LLVM_BUILTIN_DEBUGTRAP - On compilers which support it, expands to
+/// an expression which causes the program to break while running
+/// under a debugger.
+#if __has_builtin(__builtin_debugtrap)
+# define LLVM_BUILTIN_DEBUGTRAP __builtin_debugtrap()
+#elif defined(_MSC_VER)
+// The __debugbreak intrinsic is supported by MSVC and breaks while
+// running under the debugger, and also supports invoking a debugger
+// when the OS is configured appropriately.
+# define LLVM_BUILTIN_DEBUGTRAP __debugbreak()
+#else
+// Just continue execution when built with compilers that have no
+// support. This is a debugging aid and not intended to force the
+// program to abort if encountered.
+# define LLVM_BUILTIN_DEBUGTRAP
+#endif
+
+/// \macro LLVM_ASSUME_ALIGNED
+/// Returns a pointer with an assumed alignment.
+#if __has_builtin(__builtin_assume_aligned) || LLVM_GNUC_PREREQ(4, 7, 0)
+# define LLVM_ASSUME_ALIGNED(p, a) __builtin_assume_aligned(p, a)
+#elif defined(LLVM_BUILTIN_UNREACHABLE)
+// As of today, clang does not support __builtin_assume_aligned.
+# define LLVM_ASSUME_ALIGNED(p, a) \
+ (((uintptr_t(p) % (a)) == 0) ? (p) : (LLVM_BUILTIN_UNREACHABLE, (p)))
+#else
+# define LLVM_ASSUME_ALIGNED(p, a) (p)
+#endif
+
+/// \macro LLVM_PACKED
+/// Used to specify a packed structure.
+/// LLVM_PACKED(
+/// struct A {
+/// int i;
+/// int j;
+/// int k;
+/// long long l;
+/// });
+///
+/// LLVM_PACKED_START
+/// struct B {
+/// int i;
+/// int j;
+/// int k;
+/// long long l;
+/// };
+/// LLVM_PACKED_END
+#ifdef _MSC_VER
+# define LLVM_PACKED(d) __pragma(pack(push, 1)) d __pragma(pack(pop))
+# define LLVM_PACKED_START __pragma(pack(push, 1))
+# define LLVM_PACKED_END __pragma(pack(pop))
+#else
+# define LLVM_PACKED(d) d __attribute__((packed))
+# define LLVM_PACKED_START _Pragma("pack(push, 1)")
+# define LLVM_PACKED_END _Pragma("pack(pop)")
+#endif
+
+/// \macro LLVM_PTR_SIZE
+/// A constant integer equivalent to the value of sizeof(void*).
+/// Generally used in combination with alignas or when doing computation in the
+/// preprocessor.
+#ifdef __SIZEOF_POINTER__
+# define LLVM_PTR_SIZE __SIZEOF_POINTER__
+#elif defined(_WIN64)
+# define LLVM_PTR_SIZE 8
+#elif defined(_WIN32)
+# define LLVM_PTR_SIZE 4
+#elif defined(_MSC_VER)
+# error "could not determine LLVM_PTR_SIZE as a constant int for MSVC"
+#else
+# define LLVM_PTR_SIZE sizeof(void *)
+#endif
+
+/// \macro LLVM_MEMORY_SANITIZER_BUILD
+/// Whether LLVM itself is built with MemorySanitizer instrumentation.
+#if __has_feature(memory_sanitizer)
+# define LLVM_MEMORY_SANITIZER_BUILD 1
+# include <sanitizer/msan_interface.h>
+#else
+# define LLVM_MEMORY_SANITIZER_BUILD 0
+# define __msan_allocated_memory(p, size)
+# define __msan_unpoison(p, size)
+#endif
+
+/// \macro LLVM_ADDRESS_SANITIZER_BUILD
+/// Whether LLVM itself is built with AddressSanitizer instrumentation.
+#if __has_feature(address_sanitizer) || defined(__SANITIZE_ADDRESS__)
+# define LLVM_ADDRESS_SANITIZER_BUILD 1
+# include <sanitizer/asan_interface.h>
+#else
+# define LLVM_ADDRESS_SANITIZER_BUILD 0
+# define __asan_poison_memory_region(p, size)
+# define __asan_unpoison_memory_region(p, size)
+#endif
+
+/// \macro LLVM_THREAD_SANITIZER_BUILD
+/// Whether LLVM itself is built with ThreadSanitizer instrumentation.
+#if __has_feature(thread_sanitizer) || defined(__SANITIZE_THREAD__)
+# define LLVM_THREAD_SANITIZER_BUILD 1
+#else
+# define LLVM_THREAD_SANITIZER_BUILD 0
+#endif
+
+#if LLVM_THREAD_SANITIZER_BUILD
+// Thread Sanitizer is a tool that finds races in code.
+// See http://code.google.com/p/data-race-test/wiki/DynamicAnnotations .
+// tsan detects these exact functions by name.
+#ifdef __cplusplus
+extern "C" {
+#endif
+void AnnotateHappensAfter(const char *file, int line, const volatile void *cv);
+void AnnotateHappensBefore(const char *file, int line, const volatile void *cv);
+void AnnotateIgnoreWritesBegin(const char *file, int line);
+void AnnotateIgnoreWritesEnd(const char *file, int line);
+#ifdef __cplusplus
+}
+#endif
+
+// This marker is used to define a happens-before arc. The race detector will
+// infer an arc from the begin to the end when they share the same pointer
+// argument.
+# define TsanHappensBefore(cv) AnnotateHappensBefore(__FILE__, __LINE__, cv)
+
+// This marker defines the destination of a happens-before arc.
+# define TsanHappensAfter(cv) AnnotateHappensAfter(__FILE__, __LINE__, cv)
+
+// Ignore any races on writes between here and the next TsanIgnoreWritesEnd.
+# define TsanIgnoreWritesBegin() AnnotateIgnoreWritesBegin(__FILE__, __LINE__)
+
+// Resume checking for racy writes.
+# define TsanIgnoreWritesEnd() AnnotateIgnoreWritesEnd(__FILE__, __LINE__)
+#else
+# define TsanHappensBefore(cv)
+# define TsanHappensAfter(cv)
+# define TsanIgnoreWritesBegin()
+# define TsanIgnoreWritesEnd()
+#endif
+
+/// \macro LLVM_NO_SANITIZE
+/// Disable a particular sanitizer for a function.
+#if __has_attribute(no_sanitize)
+#define LLVM_NO_SANITIZE(KIND) __attribute__((no_sanitize(KIND)))
+#else
+#define LLVM_NO_SANITIZE(KIND)
+#endif
+
+/// Mark debug helper function definitions like dump() that should not be
+/// stripped from debug builds.
+/// Note that you should also surround dump() functions with
+/// `#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)` so they do always
+/// get stripped in release builds.
+// FIXME: Move this to a private config.h as it's not usable in public headers.
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+#define LLVM_DUMP_METHOD LLVM_ATTRIBUTE_NOINLINE LLVM_ATTRIBUTE_USED
+#else
+#define LLVM_DUMP_METHOD LLVM_ATTRIBUTE_NOINLINE
+#endif
+
+/// \macro LLVM_PRETTY_FUNCTION
+/// Gets a user-friendly looking function signature for the current scope
+/// using the best available method on each platform. The exact format of the
+/// resulting string is implementation specific and non-portable, so this should
+/// only be used, for example, for logging or diagnostics.
+#if defined(_MSC_VER)
+#define LLVM_PRETTY_FUNCTION __FUNCSIG__
+#elif defined(__GNUC__) || defined(__clang__)
+#define LLVM_PRETTY_FUNCTION __PRETTY_FUNCTION__
+#else
+#define LLVM_PRETTY_FUNCTION __func__
+#endif
+
+/// \macro LLVM_THREAD_LOCAL
+/// A thread-local storage specifier which can be used with globals,
+/// extern globals, and static globals.
+///
+/// This is essentially an extremely restricted analog to C++11's thread_local
+/// support, and uses that when available. However, it falls back on
+/// platform-specific or vendor-provided extensions when necessary. These
+/// extensions don't support many of the C++11 thread_local's features. You
+/// should only use this for PODs that you can statically initialize to
+/// some constant value. In almost all circumstances this is most appropriate
+/// for use with a pointer, integer, or small aggregation of pointers and
+/// integers.
+#if LLVM_ENABLE_THREADS
+#if __has_feature(cxx_thread_local)
+#define LLVM_THREAD_LOCAL thread_local
+#elif defined(_MSC_VER)
+// MSVC supports this with a __declspec.
+#define LLVM_THREAD_LOCAL __declspec(thread)
+#else
+// Clang, GCC, and other compatible compilers used __thread prior to C++11 and
+// we only need the restricted functionality that provides.
+#define LLVM_THREAD_LOCAL __thread
+#endif
+#else // !LLVM_ENABLE_THREADS
+// If threading is disabled entirely, this compiles to nothing and you get
+// a normal global variable.
+#define LLVM_THREAD_LOCAL
+#endif
+
+/// \macro LLVM_ENABLE_EXCEPTIONS
+/// Whether LLVM is built with exception support.
+#if __has_feature(cxx_exceptions)
+#define LLVM_ENABLE_EXCEPTIONS 1
+#elif defined(__GNUC__) && defined(__EXCEPTIONS)
+#define LLVM_ENABLE_EXCEPTIONS 1
+#elif defined(_MSC_VER) && defined(_CPPUNWIND)
+#define LLVM_ENABLE_EXCEPTIONS 1
+#endif
+
+#ifdef __cplusplus
+namespace llvm {
+
+/// Allocate a buffer of memory with the given size and alignment.
+///
+/// When the compiler supports aligned operator new, this will use it to to
+/// handle even over-aligned allocations.
+///
+/// However, this doesn't make any attempt to leverage the fancier techniques
+/// like posix_memalign due to portability. It is mostly intended to allow
+/// compatibility with platforms that, after aligned allocation was added, use
+/// reduced default alignment.
+inline void *allocate_buffer(size_t Size, size_t Alignment) {
+ return ::operator new(Size
+#ifdef __cpp_aligned_new
+ ,
+ std::align_val_t(Alignment)
+#endif
+ );
+}
+
+/// Deallocate a buffer of memory with the given size and alignment.
+///
+/// If supported, this will used the sized delete operator. Also if supported,
+/// this will pass the alignment to the delete operator.
+///
+/// The pointer must have been allocated with the corresponding new operator,
+/// most likely using the above helper.
+inline void deallocate_buffer(void *Ptr, size_t Size, size_t Alignment) {
+ ::operator delete(Ptr
+#ifdef __cpp_sized_deallocation
+ ,
+ Size
+#endif
+#ifdef __cpp_aligned_new
+ ,
+ std::align_val_t(Alignment)
+#endif
+ );
+}
+
+} // End namespace llvm
+
+#endif // __cplusplus
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/ConvertUTF.h b/third_party/llvm-project/include/llvm/Support/ConvertUTF.h
new file mode 100644
index 000000000..1add18533
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ConvertUTF.h
@@ -0,0 +1,306 @@
+/*===--- ConvertUTF.h - Universal Character Names conversions ---------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *==------------------------------------------------------------------------==*/
+/*
+ * Copyright 2001-2004 Unicode, Inc.
+ *
+ * Disclaimer
+ *
+ * This source code is provided as is by Unicode, Inc. No claims are
+ * made as to fitness for any particular purpose. No warranties of any
+ * kind are expressed or implied. The recipient agrees to determine
+ * applicability of information provided. If this file has been
+ * purchased on magnetic or optical media from Unicode, Inc., the
+ * sole remedy for any claim will be exchange of defective media
+ * within 90 days of receipt.
+ *
+ * Limitations on Rights to Redistribute This Code
+ *
+ * Unicode, Inc. hereby grants the right to freely use the information
+ * supplied in this file in the creation of products supporting the
+ * Unicode Standard, and to make copies of this file in any form
+ * for internal or external distribution as long as this notice
+ * remains attached.
+ */
+
+/* ---------------------------------------------------------------------
+
+ Conversions between UTF32, UTF-16, and UTF-8. Header file.
+
+ Several funtions are included here, forming a complete set of
+ conversions between the three formats. UTF-7 is not included
+ here, but is handled in a separate source file.
+
+ Each of these routines takes pointers to input buffers and output
+ buffers. The input buffers are const.
+
+ Each routine converts the text between *sourceStart and sourceEnd,
+ putting the result into the buffer between *targetStart and
+ targetEnd. Note: the end pointers are *after* the last item: e.g.
+ *(sourceEnd - 1) is the last item.
+
+ The return result indicates whether the conversion was successful,
+ and if not, whether the problem was in the source or target buffers.
+ (Only the first encountered problem is indicated.)
+
+ After the conversion, *sourceStart and *targetStart are both
+ updated to point to the end of last text successfully converted in
+ the respective buffers.
+
+ Input parameters:
+ sourceStart - pointer to a pointer to the source buffer.
+ The contents of this are modified on return so that
+ it points at the next thing to be converted.
+ targetStart - similarly, pointer to pointer to the target buffer.
+ sourceEnd, targetEnd - respectively pointers to the ends of the
+ two buffers, for overflow checking only.
+
+ These conversion functions take a ConversionFlags argument. When this
+ flag is set to strict, both irregular sequences and isolated surrogates
+ will cause an error. When the flag is set to lenient, both irregular
+ sequences and isolated surrogates are converted.
+
+ Whether the flag is strict or lenient, all illegal sequences will cause
+ an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
+ or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
+ must check for illegal sequences.
+
+ When the flag is set to lenient, characters over 0x10FFFF are converted
+ to the replacement character; otherwise (when the flag is set to strict)
+ they constitute an error.
+
+ Output parameters:
+ The value "sourceIllegal" is returned from some routines if the input
+ sequence is malformed. When "sourceIllegal" is returned, the source
+ value will point to the illegal value that caused the problem. E.g.,
+ in UTF-8 when a sequence is malformed, it points to the start of the
+ malformed sequence.
+
+ Author: Mark E. Davis, 1994.
+ Rev History: Rick McGowan, fixes & updates May 2001.
+ Fixes & updates, Sept 2001.
+
+------------------------------------------------------------------------ */
+
+#ifndef LLVM_SUPPORT_CONVERTUTF_H
+#define LLVM_SUPPORT_CONVERTUTF_H
+
+#include <cstddef>
+#include <string>
+#include <system_error>
+
+// Wrap everything in namespace llvm so that programs can link with llvm and
+// their own version of the unicode libraries.
+
+namespace llvm {
+
+/* ---------------------------------------------------------------------
+ The following 4 definitions are compiler-specific.
+ The C standard does not guarantee that wchar_t has at least
+ 16 bits, so wchar_t is no less portable than unsigned short!
+ All should be unsigned values to avoid sign extension during
+ bit mask & shift operations.
+------------------------------------------------------------------------ */
+
+typedef unsigned int UTF32; /* at least 32 bits */
+typedef unsigned short UTF16; /* at least 16 bits */
+typedef unsigned char UTF8; /* typically 8 bits */
+typedef unsigned char Boolean; /* 0 or 1 */
+
+/* Some fundamental constants */
+#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
+#define UNI_MAX_BMP (UTF32)0x0000FFFF
+#define UNI_MAX_UTF16 (UTF32)0x0010FFFF
+#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
+#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF
+
+#define UNI_MAX_UTF8_BYTES_PER_CODE_POINT 4
+
+#define UNI_UTF16_BYTE_ORDER_MARK_NATIVE 0xFEFF
+#define UNI_UTF16_BYTE_ORDER_MARK_SWAPPED 0xFFFE
+
+typedef enum {
+ conversionOK, /* conversion successful */
+ sourceExhausted, /* partial character in source, but hit end */
+ targetExhausted, /* insuff. room in target for conversion */
+ sourceIllegal /* source sequence is illegal/malformed */
+} ConversionResult;
+
+typedef enum {
+ strictConversion = 0,
+ lenientConversion
+} ConversionFlags;
+
+ConversionResult ConvertUTF8toUTF16 (
+ const UTF8** sourceStart, const UTF8* sourceEnd,
+ UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
+
+/**
+ * Convert a partial UTF8 sequence to UTF32. If the sequence ends in an
+ * incomplete code unit sequence, returns \c sourceExhausted.
+ */
+ConversionResult ConvertUTF8toUTF32Partial(
+ const UTF8** sourceStart, const UTF8* sourceEnd,
+ UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
+
+/**
+ * Convert a partial UTF8 sequence to UTF32. If the sequence ends in an
+ * incomplete code unit sequence, returns \c sourceIllegal.
+ */
+ConversionResult ConvertUTF8toUTF32(
+ const UTF8** sourceStart, const UTF8* sourceEnd,
+ UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF16toUTF8 (
+ const UTF16** sourceStart, const UTF16* sourceEnd,
+ UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF32toUTF8 (
+ const UTF32** sourceStart, const UTF32* sourceEnd,
+ UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF16toUTF32 (
+ const UTF16** sourceStart, const UTF16* sourceEnd,
+ UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF32toUTF16 (
+ const UTF32** sourceStart, const UTF32* sourceEnd,
+ UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
+
+Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd);
+
+Boolean isLegalUTF8String(const UTF8 **source, const UTF8 *sourceEnd);
+
+unsigned getNumBytesForUTF8(UTF8 firstByte);
+
+/*************************************************************************/
+/* Below are LLVM-specific wrappers of the functions above. */
+
+template <typename T> class ArrayRef;
+template <typename T> class SmallVectorImpl;
+class StringRef;
+
+/**
+ * Convert an UTF8 StringRef to UTF8, UTF16, or UTF32 depending on
+ * WideCharWidth. The converted data is written to ResultPtr, which needs to
+ * point to at least WideCharWidth * (Source.Size() + 1) bytes. On success,
+ * ResultPtr will point one after the end of the copied string. On failure,
+ * ResultPtr will not be changed, and ErrorPtr will be set to the location of
+ * the first character which could not be converted.
+ * \return true on success.
+ */
+bool ConvertUTF8toWide(unsigned WideCharWidth, llvm::StringRef Source,
+ char *&ResultPtr, const UTF8 *&ErrorPtr);
+
+/**
+* Converts a UTF-8 StringRef to a std::wstring.
+* \return true on success.
+*/
+bool ConvertUTF8toWide(llvm::StringRef Source, std::wstring &Result);
+
+/**
+* Converts a UTF-8 C-string to a std::wstring.
+* \return true on success.
+*/
+bool ConvertUTF8toWide(const char *Source, std::wstring &Result);
+
+/**
+* Converts a std::wstring to a UTF-8 encoded std::string.
+* \return true on success.
+*/
+bool convertWideToUTF8(const std::wstring &Source, std::string &Result);
+
+
+/**
+ * Convert an Unicode code point to UTF8 sequence.
+ *
+ * \param Source a Unicode code point.
+ * \param [in,out] ResultPtr pointer to the output buffer, needs to be at least
+ * \c UNI_MAX_UTF8_BYTES_PER_CODE_POINT bytes. On success \c ResultPtr is
+ * updated one past end of the converted sequence.
+ *
+ * \returns true on success.
+ */
+bool ConvertCodePointToUTF8(unsigned Source, char *&ResultPtr);
+
+/**
+ * Convert the first UTF8 sequence in the given source buffer to a UTF32
+ * code point.
+ *
+ * \param [in,out] source A pointer to the source buffer. If the conversion
+ * succeeds, this pointer will be updated to point to the byte just past the
+ * end of the converted sequence.
+ * \param sourceEnd A pointer just past the end of the source buffer.
+ * \param [out] target The converted code
+ * \param flags Whether the conversion is strict or lenient.
+ *
+ * \returns conversionOK on success
+ *
+ * \sa ConvertUTF8toUTF32
+ */
+inline ConversionResult convertUTF8Sequence(const UTF8 **source,
+ const UTF8 *sourceEnd,
+ UTF32 *target,
+ ConversionFlags flags) {
+ if (*source == sourceEnd)
+ return sourceExhausted;
+ unsigned size = getNumBytesForUTF8(**source);
+ if ((ptrdiff_t)size > sourceEnd - *source)
+ return sourceExhausted;
+ return ConvertUTF8toUTF32(source, *source + size, &target, target + 1, flags);
+}
+
+/**
+ * Returns true if a blob of text starts with a UTF-16 big or little endian byte
+ * order mark.
+ */
+bool hasUTF16ByteOrderMark(ArrayRef<char> SrcBytes);
+
+/**
+ * Converts a stream of raw bytes assumed to be UTF16 into a UTF8 std::string.
+ *
+ * \param [in] SrcBytes A buffer of what is assumed to be UTF-16 encoded text.
+ * \param [out] Out Converted UTF-8 is stored here on success.
+ * \returns true on success
+ */
+bool convertUTF16ToUTF8String(ArrayRef<char> SrcBytes, std::string &Out);
+
+/**
+* Converts a UTF16 string into a UTF8 std::string.
+*
+* \param [in] Src A buffer of UTF-16 encoded text.
+* \param [out] Out Converted UTF-8 is stored here on success.
+* \returns true on success
+*/
+bool convertUTF16ToUTF8String(ArrayRef<UTF16> Src, std::string &Out);
+
+/**
+ * Converts a UTF-8 string into a UTF-16 string with native endianness.
+ *
+ * \returns true on success
+ */
+bool convertUTF8ToUTF16String(StringRef SrcUTF8,
+ SmallVectorImpl<UTF16> &DstUTF16);
+
+#if defined(_WIN32)
+namespace sys {
+namespace windows {
+std::error_code UTF8ToUTF16(StringRef utf8, SmallVectorImpl<wchar_t> &utf16);
+/// Convert to UTF16 from the current code page used in the system
+std::error_code CurCPToUTF16(StringRef utf8, SmallVectorImpl<wchar_t> &utf16);
+std::error_code UTF16ToUTF8(const wchar_t *utf16, size_t utf16_len,
+ SmallVectorImpl<char> &utf8);
+/// Convert from UTF16 to the current code page used in the system
+std::error_code UTF16ToCurCP(const wchar_t *utf16, size_t utf16_len,
+ SmallVectorImpl<char> &utf8);
+} // namespace windows
+} // namespace sys
+#endif
+
+} /* end namespace llvm */
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/DJB.h b/third_party/llvm-project/include/llvm/Support/DJB.h
new file mode 100644
index 000000000..8a04a324a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/DJB.h
@@ -0,0 +1,32 @@
+//===-- llvm/Support/DJB.h ---DJB Hash --------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains support for the DJ Bernstein hash function.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_DJB_H
+#define LLVM_SUPPORT_DJB_H
+
+#include "llvm/ADT/StringRef.h"
+
+namespace llvm {
+
+/// The Bernstein hash function used by the DWARF accelerator tables.
+inline uint32_t djbHash(StringRef Buffer, uint32_t H = 5381) {
+ for (unsigned char C : Buffer.bytes())
+ H = (H << 5) + H + C;
+ return H;
+}
+
+/// Computes the Bernstein hash after folding the input according to the Dwarf 5
+/// standard case folding rules.
+uint32_t caseFoldingDjbHash(StringRef Buffer, uint32_t H = 5381);
+} // namespace llvm
+
+#endif // LLVM_SUPPORT_DJB_H
diff --git a/third_party/llvm-project/include/llvm/Support/DataExtractor.h b/third_party/llvm-project/include/llvm/Support/DataExtractor.h
new file mode 100644
index 000000000..f590a1e10
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/DataExtractor.h
@@ -0,0 +1,575 @@
+//===-- DataExtractor.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_DATAEXTRACTOR_H
+#define LLVM_SUPPORT_DATAEXTRACTOR_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/Error.h"
+
+namespace llvm {
+
+/// An auxiliary type to facilitate extraction of 3-byte entities.
+struct Uint24 {
+ uint8_t Bytes[3];
+ Uint24(uint8_t U) {
+ Bytes[0] = Bytes[1] = Bytes[2] = U;
+ }
+ Uint24(uint8_t U0, uint8_t U1, uint8_t U2) {
+ Bytes[0] = U0; Bytes[1] = U1; Bytes[2] = U2;
+ }
+ uint32_t getAsUint32(bool IsLittleEndian) const {
+ int LoIx = IsLittleEndian ? 0 : 2;
+ return Bytes[LoIx] + (Bytes[1] << 8) + (Bytes[2-LoIx] << 16);
+ }
+};
+
+using uint24_t = Uint24;
+static_assert(sizeof(uint24_t) == 3, "sizeof(uint24_t) != 3");
+
+/// Needed by swapByteOrder().
+inline uint24_t getSwappedBytes(uint24_t C) {
+ return uint24_t(C.Bytes[2], C.Bytes[1], C.Bytes[0]);
+}
+
+class DataExtractor {
+ StringRef Data;
+ uint8_t IsLittleEndian;
+ uint8_t AddressSize;
+public:
+ /// A class representing a position in a DataExtractor, as well as any error
+ /// encountered during extraction. It enables one to extract a sequence of
+ /// values without error-checking and then checking for errors in bulk at the
+ /// end. The class holds an Error object, so failing to check the result of
+ /// the parse will result in a runtime error. The error flag is sticky and
+ /// will cause all subsequent extraction functions to fail without even
+ /// attempting to parse and without updating the Cursor offset. After clearing
+ /// the error flag, one can again use the Cursor object for parsing.
+ class Cursor {
+ uint64_t Offset;
+ Error Err;
+
+ friend class DataExtractor;
+
+ public:
+ /// Construct a cursor for extraction from the given offset.
+ explicit Cursor(uint64_t Offset) : Offset(Offset), Err(Error::success()) {}
+
+ /// Checks whether the cursor is valid (i.e. no errors were encountered). In
+ /// case of errors, this does not clear the error flag -- one must call
+ /// takeError() instead.
+ explicit operator bool() { return !Err; }
+
+ /// Return the current position of this Cursor. In the error state this is
+ /// the position of the Cursor before the first error was encountered.
+ uint64_t tell() const { return Offset; }
+
+ /// Return error contained inside this Cursor, if any. Clears the internal
+ /// Cursor state.
+ Error takeError() { return std::move(Err); }
+ };
+
+ /// Construct with a buffer that is owned by the caller.
+ ///
+ /// This constructor allows us to use data that is owned by the
+ /// caller. The data must stay around as long as this object is
+ /// valid.
+ DataExtractor(StringRef Data, bool IsLittleEndian, uint8_t AddressSize)
+ : Data(Data), IsLittleEndian(IsLittleEndian), AddressSize(AddressSize) {}
+ DataExtractor(ArrayRef<uint8_t> Data, bool IsLittleEndian,
+ uint8_t AddressSize)
+ : Data(StringRef(reinterpret_cast<const char *>(Data.data()),
+ Data.size())),
+ IsLittleEndian(IsLittleEndian), AddressSize(AddressSize) {}
+
+ /// Get the data pointed to by this extractor.
+ StringRef getData() const { return Data; }
+ /// Get the endianness for this extractor.
+ bool isLittleEndian() const { return IsLittleEndian; }
+ /// Get the address size for this extractor.
+ uint8_t getAddressSize() const { return AddressSize; }
+ /// Set the address size for this extractor.
+ void setAddressSize(uint8_t Size) { AddressSize = Size; }
+
+ /// Extract a C string from \a *offset_ptr.
+ ///
+ /// Returns a pointer to a C String from the data at the offset
+ /// pointed to by \a offset_ptr. A variable length NULL terminated C
+ /// string will be extracted and the \a offset_ptr will be
+ /// updated with the offset of the byte that follows the NULL
+ /// terminator byte.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @return
+ /// A pointer to the C string value in the data. If the offset
+ /// pointed to by \a offset_ptr is out of bounds, or if the
+ /// offset plus the length of the C string is out of bounds,
+ /// NULL will be returned.
+ const char *getCStr(uint64_t *offset_ptr) const;
+
+ /// Extract a C string from \a *offset_ptr.
+ ///
+ /// Returns a StringRef for the C String from the data at the offset
+ /// pointed to by \a offset_ptr. A variable length NULL terminated C
+ /// string will be extracted and the \a offset_ptr will be
+ /// updated with the offset of the byte that follows the NULL
+ /// terminator byte.
+ ///
+ /// \param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// \return
+ /// A StringRef for the C string value in the data. If the offset
+ /// pointed to by \a offset_ptr is out of bounds, or if the
+ /// offset plus the length of the C string is out of bounds,
+ /// a default-initialized StringRef will be returned.
+ StringRef getCStrRef(uint64_t *offset_ptr) const;
+
+ /// Extract an unsigned integer of size \a byte_size from \a
+ /// *offset_ptr.
+ ///
+ /// Extract a single unsigned integer value and update the offset
+ /// pointed to by \a offset_ptr. The size of the extracted integer
+ /// is specified by the \a byte_size argument. \a byte_size should
+ /// have a value greater than or equal to one and less than or equal
+ /// to eight since the return value is 64 bits wide. Any
+ /// \a byte_size values less than 1 or greater than 8 will result in
+ /// nothing being extracted, and zero being returned.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in] byte_size
+ /// The size in byte of the integer to extract.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The unsigned integer value that was extracted, or zero on
+ /// failure.
+ uint64_t getUnsigned(uint64_t *offset_ptr, uint32_t byte_size,
+ Error *Err = nullptr) const;
+
+ /// Extract an unsigned integer of the given size from the location given by
+ /// the cursor. In case of an extraction error, or if the cursor is already in
+ /// an error state, zero is returned.
+ uint64_t getUnsigned(Cursor &C, uint32_t Size) const {
+ return getUnsigned(&C.Offset, Size, &C.Err);
+ }
+
+ /// Extract an signed integer of size \a byte_size from \a *offset_ptr.
+ ///
+ /// Extract a single signed integer value (sign extending if required)
+ /// and update the offset pointed to by \a offset_ptr. The size of
+ /// the extracted integer is specified by the \a byte_size argument.
+ /// \a byte_size should have a value greater than or equal to one
+ /// and less than or equal to eight since the return value is 64
+ /// bits wide. Any \a byte_size values less than 1 or greater than
+ /// 8 will result in nothing being extracted, and zero being returned.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in] size
+ /// The size in bytes of the integer to extract.
+ ///
+ /// @return
+ /// The sign extended signed integer value that was extracted,
+ /// or zero on failure.
+ int64_t getSigned(uint64_t *offset_ptr, uint32_t size) const;
+
+ //------------------------------------------------------------------
+ /// Extract an pointer from \a *offset_ptr.
+ ///
+ /// Extract a single pointer from the data and update the offset
+ /// pointed to by \a offset_ptr. The size of the extracted pointer
+ /// is \a getAddressSize(), so the address size has to be
+ /// set correctly prior to extracting any pointer values.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @return
+ /// The extracted pointer value as a 64 integer.
+ uint64_t getAddress(uint64_t *offset_ptr) const {
+ return getUnsigned(offset_ptr, AddressSize);
+ }
+
+ /// Extract a pointer-sized unsigned integer from the location given by the
+ /// cursor. In case of an extraction error, or if the cursor is already in
+ /// an error state, zero is returned.
+ uint64_t getAddress(Cursor &C) const { return getUnsigned(C, AddressSize); }
+
+ /// Extract a uint8_t value from \a *offset_ptr.
+ ///
+ /// Extract a single uint8_t from the binary data at the offset
+ /// pointed to by \a offset_ptr, and advance the offset on success.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The extracted uint8_t value.
+ uint8_t getU8(uint64_t *offset_ptr, Error *Err = nullptr) const;
+
+ /// Extract a single uint8_t value from the location given by the cursor. In
+ /// case of an extraction error, or if the cursor is already in an error
+ /// state, zero is returned.
+ uint8_t getU8(Cursor &C) const { return getU8(&C.Offset, &C.Err); }
+
+ /// Extract \a count uint8_t values from \a *offset_ptr.
+ ///
+ /// Extract \a count uint8_t values from the binary data at the
+ /// offset pointed to by \a offset_ptr, and advance the offset on
+ /// success. The extracted values are copied into \a dst.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[out] dst
+ /// A buffer to copy \a count uint8_t values into. \a dst must
+ /// be large enough to hold all requested data.
+ ///
+ /// @param[in] count
+ /// The number of uint8_t values to extract.
+ ///
+ /// @return
+ /// \a dst if all values were properly extracted and copied,
+ /// NULL otherise.
+ uint8_t *getU8(uint64_t *offset_ptr, uint8_t *dst, uint32_t count) const;
+
+ /// Extract \a Count uint8_t values from the location given by the cursor and
+ /// store them into the destination buffer. In case of an extraction error, or
+ /// if the cursor is already in an error state, a nullptr is returned and the
+ /// destination buffer is left unchanged.
+ uint8_t *getU8(Cursor &C, uint8_t *Dst, uint32_t Count) const;
+
+ /// Extract \a Count uint8_t values from the location given by the cursor and
+ /// store them into the destination vector. The vector is resized to fit the
+ /// extracted data. In case of an extraction error, or if the cursor is
+ /// already in an error state, the destination vector is left unchanged and
+ /// cursor is placed into an error state.
+ void getU8(Cursor &C, SmallVectorImpl<uint8_t> &Dst, uint32_t Count) const {
+ if (isValidOffsetForDataOfSize(C.Offset, Count))
+ Dst.resize(Count);
+
+ // This relies on the fact that getU8 will not attempt to write to the
+ // buffer if isValidOffsetForDataOfSize(C.Offset, Count) is false.
+ getU8(C, Dst.data(), Count);
+ }
+
+ //------------------------------------------------------------------
+ /// Extract a uint16_t value from \a *offset_ptr.
+ ///
+ /// Extract a single uint16_t from the binary data at the offset
+ /// pointed to by \a offset_ptr, and update the offset on success.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The extracted uint16_t value.
+ //------------------------------------------------------------------
+ uint16_t getU16(uint64_t *offset_ptr, Error *Err = nullptr) const;
+
+ /// Extract a single uint16_t value from the location given by the cursor. In
+ /// case of an extraction error, or if the cursor is already in an error
+ /// state, zero is returned.
+ uint16_t getU16(Cursor &C) const { return getU16(&C.Offset, &C.Err); }
+
+ /// Extract \a count uint16_t values from \a *offset_ptr.
+ ///
+ /// Extract \a count uint16_t values from the binary data at the
+ /// offset pointed to by \a offset_ptr, and advance the offset on
+ /// success. The extracted values are copied into \a dst.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[out] dst
+ /// A buffer to copy \a count uint16_t values into. \a dst must
+ /// be large enough to hold all requested data.
+ ///
+ /// @param[in] count
+ /// The number of uint16_t values to extract.
+ ///
+ /// @return
+ /// \a dst if all values were properly extracted and copied,
+ /// NULL otherise.
+ uint16_t *getU16(uint64_t *offset_ptr, uint16_t *dst, uint32_t count) const;
+
+ /// Extract a 24-bit unsigned value from \a *offset_ptr and return it
+ /// in a uint32_t.
+ ///
+ /// Extract 3 bytes from the binary data at the offset pointed to by
+ /// \a offset_ptr, construct a uint32_t from them and update the offset
+ /// on success.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the 3 bytes if the value is extracted correctly. If the offset
+ /// is out of bounds or there are not enough bytes to extract this value,
+ /// the offset will be left unmodified.
+ ///
+ /// @return
+ /// The extracted 24-bit value represented in a uint32_t.
+ uint32_t getU24(uint64_t *offset_ptr) const;
+
+ /// Extract a uint32_t value from \a *offset_ptr.
+ ///
+ /// Extract a single uint32_t from the binary data at the offset
+ /// pointed to by \a offset_ptr, and update the offset on success.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The extracted uint32_t value.
+ uint32_t getU32(uint64_t *offset_ptr, Error *Err = nullptr) const;
+
+ /// Extract a single uint32_t value from the location given by the cursor. In
+ /// case of an extraction error, or if the cursor is already in an error
+ /// state, zero is returned.
+ uint32_t getU32(Cursor &C) const { return getU32(&C.Offset, &C.Err); }
+
+ /// Extract \a count uint32_t values from \a *offset_ptr.
+ ///
+ /// Extract \a count uint32_t values from the binary data at the
+ /// offset pointed to by \a offset_ptr, and advance the offset on
+ /// success. The extracted values are copied into \a dst.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[out] dst
+ /// A buffer to copy \a count uint32_t values into. \a dst must
+ /// be large enough to hold all requested data.
+ ///
+ /// @param[in] count
+ /// The number of uint32_t values to extract.
+ ///
+ /// @return
+ /// \a dst if all values were properly extracted and copied,
+ /// NULL otherise.
+ uint32_t *getU32(uint64_t *offset_ptr, uint32_t *dst, uint32_t count) const;
+
+ /// Extract a uint64_t value from \a *offset_ptr.
+ ///
+ /// Extract a single uint64_t from the binary data at the offset
+ /// pointed to by \a offset_ptr, and update the offset on success.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The extracted uint64_t value.
+ uint64_t getU64(uint64_t *offset_ptr, Error *Err = nullptr) const;
+
+ /// Extract a single uint64_t value from the location given by the cursor. In
+ /// case of an extraction error, or if the cursor is already in an error
+ /// state, zero is returned.
+ uint64_t getU64(Cursor &C) const { return getU64(&C.Offset, &C.Err); }
+
+ /// Extract \a count uint64_t values from \a *offset_ptr.
+ ///
+ /// Extract \a count uint64_t values from the binary data at the
+ /// offset pointed to by \a offset_ptr, and advance the offset on
+ /// success. The extracted values are copied into \a dst.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[out] dst
+ /// A buffer to copy \a count uint64_t values into. \a dst must
+ /// be large enough to hold all requested data.
+ ///
+ /// @param[in] count
+ /// The number of uint64_t values to extract.
+ ///
+ /// @return
+ /// \a dst if all values were properly extracted and copied,
+ /// NULL otherise.
+ uint64_t *getU64(uint64_t *offset_ptr, uint64_t *dst, uint32_t count) const;
+
+ /// Extract a signed LEB128 value from \a *offset_ptr.
+ ///
+ /// Extracts an signed LEB128 number from this object's data
+ /// starting at the offset pointed to by \a offset_ptr. The offset
+ /// pointed to by \a offset_ptr will be updated with the offset of
+ /// the byte following the last extracted byte.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @return
+ /// The extracted signed integer value.
+ int64_t getSLEB128(uint64_t *offset_ptr) const;
+
+ /// Extract a unsigned LEB128 value from \a *offset_ptr.
+ ///
+ /// Extracts an unsigned LEB128 number from this object's data
+ /// starting at the offset pointed to by \a offset_ptr. The offset
+ /// pointed to by \a offset_ptr will be updated with the offset of
+ /// the byte following the last extracted byte.
+ ///
+ /// @param[in,out] offset_ptr
+ /// A pointer to an offset within the data that will be advanced
+ /// by the appropriate number of bytes if the value is extracted
+ /// correctly. If the offset is out of bounds or there are not
+ /// enough bytes to extract this value, the offset will be left
+ /// unmodified.
+ ///
+ /// @param[in,out] Err
+ /// A pointer to an Error object. Upon return the Error object is set to
+ /// indicate the result (success/failure) of the function. If the Error
+ /// object is already set when calling this function, no extraction is
+ /// performed.
+ ///
+ /// @return
+ /// The extracted unsigned integer value.
+ uint64_t getULEB128(uint64_t *offset_ptr, llvm::Error *Err = nullptr) const;
+
+ /// Extract an unsigned ULEB128 value from the location given by the cursor.
+ /// In case of an extraction error, or if the cursor is already in an error
+ /// state, zero is returned.
+ uint64_t getULEB128(Cursor &C) const { return getULEB128(&C.Offset, &C.Err); }
+
+ /// Advance the Cursor position by the given number of bytes. No-op if the
+ /// cursor is in an error state.
+ void skip(Cursor &C, uint64_t Length) const;
+
+ /// Return true iff the cursor is at the end of the buffer, regardless of the
+ /// error state of the cursor. The only way both eof and error states can be
+ /// true is if one attempts a read while the cursor is at the very end of the
+ /// data buffer.
+ bool eof(const Cursor &C) const { return Data.size() == C.Offset; }
+
+ /// Test the validity of \a offset.
+ ///
+ /// @return
+ /// \b true if \a offset is a valid offset into the data in this
+ /// object, \b false otherwise.
+ bool isValidOffset(uint64_t offset) const { return Data.size() > offset; }
+
+ /// Test the availability of \a length bytes of data from \a offset.
+ ///
+ /// @return
+ /// \b true if \a offset is a valid offset and there are \a
+ /// length bytes available at that offset, \b false otherwise.
+ bool isValidOffsetForDataOfSize(uint64_t offset, uint64_t length) const {
+ return offset + length >= offset && isValidOffset(offset + length - 1);
+ }
+
+ /// Test the availability of enough bytes of data for a pointer from
+ /// \a offset. The size of a pointer is \a getAddressSize().
+ ///
+ /// @return
+ /// \b true if \a offset is a valid offset and there are enough
+ /// bytes for a pointer available at that offset, \b false
+ /// otherwise.
+ bool isValidOffsetForAddress(uint64_t offset) const {
+ return isValidOffsetForDataOfSize(offset, AddressSize);
+ }
+
+protected:
+ // Make it possible for subclasses to access these fields without making them
+ // public.
+ static uint64_t &getOffset(Cursor &C) { return C.Offset; }
+ static Error &getError(Cursor &C) { return C.Err; }
+};
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/DataTypes.h b/third_party/llvm-project/include/llvm/Support/DataTypes.h
new file mode 100644
index 000000000..a3fcc8253
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/DataTypes.h
@@ -0,0 +1,16 @@
+//===-- llvm/Support/DataTypes.h - Define fixed size types ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Due to layering constraints (Support depends on llvm-c) this is a thin
+// wrapper around the implementation that lives in llvm-c, though most clients
+// can/should think of this as being provided by Support for simplicity (not
+// many clients are aware of their dependency on llvm-c).
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm-c/DataTypes.h"
diff --git a/third_party/llvm-project/include/llvm/Support/Debug.h b/third_party/llvm-project/include/llvm/Support/Debug.h
new file mode 100644
index 000000000..64b730951
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Debug.h
@@ -0,0 +1,126 @@
+//===- llvm/Support/Debug.h - Easy way to add debug output ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a handy way of adding debugging information to your
+// code, without it being enabled all of the time, and without having to add
+// command line options to enable it.
+//
+// In particular, just wrap your code with the LLVM_DEBUG() macro, and it will
+// be enabled automatically if you specify '-debug' on the command-line.
+// LLVM_DEBUG() requires the DEBUG_TYPE macro to be defined. Set it to "foo"
+// specify that your debug code belongs to class "foo". Be careful that you only
+// do this after including Debug.h and not around any #include of headers.
+// Headers should define and undef the macro acround the code that needs to use
+// the LLVM_DEBUG() macro. Then, on the command line, you can specify
+// '-debug-only=foo' to enable JUST the debug information for the foo class.
+//
+// When compiling without assertions, the -debug-* options and all code in
+// LLVM_DEBUG() statements disappears, so it does not affect the runtime of the
+// code.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_DEBUG_H
+#define LLVM_SUPPORT_DEBUG_H
+
+namespace llvm {
+
+class raw_ostream;
+
+#ifndef NDEBUG
+
+/// isCurrentDebugType - Return true if the specified string is the debug type
+/// specified on the command line, or if none was specified on the command line
+/// with the -debug-only=X option.
+///
+bool isCurrentDebugType(const char *Type);
+
+/// setCurrentDebugType - Set the current debug type, as if the -debug-only=X
+/// option were specified. Note that DebugFlag also needs to be set to true for
+/// debug output to be produced.
+///
+void setCurrentDebugType(const char *Type);
+
+/// setCurrentDebugTypes - Set the current debug type, as if the
+/// -debug-only=X,Y,Z option were specified. Note that DebugFlag
+/// also needs to be set to true for debug output to be produced.
+///
+void setCurrentDebugTypes(const char **Types, unsigned Count);
+
+/// DEBUG_WITH_TYPE macro - This macro should be used by passes to emit debug
+/// information. In the '-debug' option is specified on the commandline, and if
+/// this is a debug build, then the code specified as the option to the macro
+/// will be executed. Otherwise it will not be. Example:
+///
+/// DEBUG_WITH_TYPE("bitset", dbgs() << "Bitset contains: " << Bitset << "\n");
+///
+/// This will emit the debug information if -debug is present, and -debug-only
+/// is not specified, or is specified as "bitset".
+#define DEBUG_WITH_TYPE(TYPE, X) \
+ do { if (::llvm::DebugFlag && ::llvm::isCurrentDebugType(TYPE)) { X; } \
+ } while (false)
+
+#else
+#define isCurrentDebugType(X) (false)
+#define setCurrentDebugType(X)
+#define setCurrentDebugTypes(X, N)
+#define DEBUG_WITH_TYPE(TYPE, X) do { } while (false)
+#endif
+
+/// This boolean is set to true if the '-debug' command line option
+/// is specified. This should probably not be referenced directly, instead, use
+/// the DEBUG macro below.
+///
+extern bool DebugFlag;
+
+/// \name Verification flags.
+///
+/// These flags turns on/off that are expensive and are turned off by default,
+/// unless macro EXPENSIVE_CHECKS is defined. The flags allow selectively
+/// turning the checks on without need to recompile.
+/// \{
+
+/// Enables verification of dominator trees.
+///
+extern bool VerifyDomInfo;
+
+/// Enables verification of loop info.
+///
+extern bool VerifyLoopInfo;
+
+/// Enables verification of MemorySSA.
+///
+extern bool VerifyMemorySSA;
+
+///\}
+
+/// EnableDebugBuffering - This defaults to false. If true, the debug
+/// stream will install signal handlers to dump any buffered debug
+/// output. It allows clients to selectively allow the debug stream
+/// to install signal handlers if they are certain there will be no
+/// conflict.
+///
+extern bool EnableDebugBuffering;
+
+/// dbgs() - This returns a reference to a raw_ostream for debugging
+/// messages. If debugging is disabled it returns errs(). Use it
+/// like: dbgs() << "foo" << "bar";
+raw_ostream &dbgs();
+
+// DEBUG macro - This macro should be used by passes to emit debug information.
+// In the '-debug' option is specified on the commandline, and if this is a
+// debug build, then the code specified as the option to the macro will be
+// executed. Otherwise it will not be. Example:
+//
+// LLVM_DEBUG(dbgs() << "Bitset contains: " << Bitset << "\n");
+//
+#define LLVM_DEBUG(X) DEBUG_WITH_TYPE(DEBUG_TYPE, X)
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_DEBUG_H
diff --git a/third_party/llvm-project/include/llvm/Support/Endian.h b/third_party/llvm-project/include/llvm/Support/Endian.h
new file mode 100644
index 000000000..87aecedd3
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Endian.h
@@ -0,0 +1,429 @@
+//===- Endian.h - Utilities for IO with endian specific data ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares generic functions to read and write endian specific data.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ENDIAN_H
+#define LLVM_SUPPORT_ENDIAN_H
+
+#include "llvm/Support/AlignOf.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Host.h"
+#include "llvm/Support/SwapByteOrder.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstring>
+#include <type_traits>
+
+namespace llvm {
+namespace support {
+
+enum endianness {big, little, native};
+
+// These are named values for common alignments.
+enum {aligned = 0, unaligned = 1};
+
+namespace detail {
+
+/// ::value is either alignment, or alignof(T) if alignment is 0.
+template<class T, int alignment>
+struct PickAlignment {
+ enum { value = alignment == 0 ? alignof(T) : alignment };
+};
+
+} // end namespace detail
+
+namespace endian {
+
+constexpr endianness system_endianness() {
+ return sys::IsBigEndianHost ? big : little;
+}
+
+template <typename value_type>
+inline value_type byte_swap(value_type value, endianness endian) {
+ if ((endian != native) && (endian != system_endianness()))
+ sys::swapByteOrder(value);
+ return value;
+}
+
+/// Swap the bytes of value to match the given endianness.
+template<typename value_type, endianness endian>
+inline value_type byte_swap(value_type value) {
+ return byte_swap(value, endian);
+}
+
+/// Read a value of a particular endianness from memory.
+template <typename value_type, std::size_t alignment>
+inline value_type read(const void *memory, endianness endian) {
+ value_type ret;
+
+ memcpy(&ret,
+ LLVM_ASSUME_ALIGNED(
+ memory, (detail::PickAlignment<value_type, alignment>::value)),
+ sizeof(value_type));
+ return byte_swap<value_type>(ret, endian);
+}
+
+template<typename value_type,
+ endianness endian,
+ std::size_t alignment>
+inline value_type read(const void *memory) {
+ return read<value_type, alignment>(memory, endian);
+}
+
+/// Read a value of a particular endianness from a buffer, and increment the
+/// buffer past that value.
+template <typename value_type, std::size_t alignment, typename CharT>
+inline value_type readNext(const CharT *&memory, endianness endian) {
+ value_type ret = read<value_type, alignment>(memory, endian);
+ memory += sizeof(value_type);
+ return ret;
+}
+
+template<typename value_type, endianness endian, std::size_t alignment,
+ typename CharT>
+inline value_type readNext(const CharT *&memory) {
+ return readNext<value_type, alignment, CharT>(memory, endian);
+}
+
+/// Write a value to memory with a particular endianness.
+template <typename value_type, std::size_t alignment>
+inline void write(void *memory, value_type value, endianness endian) {
+ value = byte_swap<value_type>(value, endian);
+ memcpy(LLVM_ASSUME_ALIGNED(
+ memory, (detail::PickAlignment<value_type, alignment>::value)),
+ &value, sizeof(value_type));
+}
+
+template<typename value_type,
+ endianness endian,
+ std::size_t alignment>
+inline void write(void *memory, value_type value) {
+ write<value_type, alignment>(memory, value, endian);
+}
+
+template <typename value_type>
+using make_unsigned_t = typename std::make_unsigned<value_type>::type;
+
+/// Read a value of a particular endianness from memory, for a location
+/// that starts at the given bit offset within the first byte.
+template <typename value_type, endianness endian, std::size_t alignment>
+inline value_type readAtBitAlignment(const void *memory, uint64_t startBit) {
+ assert(startBit < 8);
+ if (startBit == 0)
+ return read<value_type, endian, alignment>(memory);
+ else {
+ // Read two values and compose the result from them.
+ value_type val[2];
+ memcpy(&val[0],
+ LLVM_ASSUME_ALIGNED(
+ memory, (detail::PickAlignment<value_type, alignment>::value)),
+ sizeof(value_type) * 2);
+ val[0] = byte_swap<value_type, endian>(val[0]);
+ val[1] = byte_swap<value_type, endian>(val[1]);
+
+ // Shift bits from the lower value into place.
+ make_unsigned_t<value_type> lowerVal = val[0] >> startBit;
+ // Mask off upper bits after right shift in case of signed type.
+ make_unsigned_t<value_type> numBitsFirstVal =
+ (sizeof(value_type) * 8) - startBit;
+ lowerVal &= ((make_unsigned_t<value_type>)1 << numBitsFirstVal) - 1;
+
+ // Get the bits from the upper value.
+ make_unsigned_t<value_type> upperVal =
+ val[1] & (((make_unsigned_t<value_type>)1 << startBit) - 1);
+ // Shift them in to place.
+ upperVal <<= numBitsFirstVal;
+
+ return lowerVal | upperVal;
+ }
+}
+
+/// Write a value to memory with a particular endianness, for a location
+/// that starts at the given bit offset within the first byte.
+template <typename value_type, endianness endian, std::size_t alignment>
+inline void writeAtBitAlignment(void *memory, value_type value,
+ uint64_t startBit) {
+ assert(startBit < 8);
+ if (startBit == 0)
+ write<value_type, endian, alignment>(memory, value);
+ else {
+ // Read two values and shift the result into them.
+ value_type val[2];
+ memcpy(&val[0],
+ LLVM_ASSUME_ALIGNED(
+ memory, (detail::PickAlignment<value_type, alignment>::value)),
+ sizeof(value_type) * 2);
+ val[0] = byte_swap<value_type, endian>(val[0]);
+ val[1] = byte_swap<value_type, endian>(val[1]);
+
+ // Mask off any existing bits in the upper part of the lower value that
+ // we want to replace.
+ val[0] &= ((make_unsigned_t<value_type>)1 << startBit) - 1;
+ make_unsigned_t<value_type> numBitsFirstVal =
+ (sizeof(value_type) * 8) - startBit;
+ make_unsigned_t<value_type> lowerVal = value;
+ if (startBit > 0) {
+ // Mask off the upper bits in the new value that are not going to go into
+ // the lower value. This avoids a left shift of a negative value, which
+ // is undefined behavior.
+ lowerVal &= (((make_unsigned_t<value_type>)1 << numBitsFirstVal) - 1);
+ // Now shift the new bits into place
+ lowerVal <<= startBit;
+ }
+ val[0] |= lowerVal;
+
+ // Mask off any existing bits in the lower part of the upper value that
+ // we want to replace.
+ val[1] &= ~(((make_unsigned_t<value_type>)1 << startBit) - 1);
+ // Next shift the bits that go into the upper value into position.
+ make_unsigned_t<value_type> upperVal = value >> numBitsFirstVal;
+ // Mask off upper bits after right shift in case of signed type.
+ upperVal &= ((make_unsigned_t<value_type>)1 << startBit) - 1;
+ val[1] |= upperVal;
+
+ // Finally, rewrite values.
+ val[0] = byte_swap<value_type, endian>(val[0]);
+ val[1] = byte_swap<value_type, endian>(val[1]);
+ memcpy(LLVM_ASSUME_ALIGNED(
+ memory, (detail::PickAlignment<value_type, alignment>::value)),
+ &val[0], sizeof(value_type) * 2);
+ }
+}
+
+} // end namespace endian
+
+namespace detail {
+
+template <typename ValueType, endianness Endian, std::size_t Alignment,
+ std::size_t ALIGN = PickAlignment<ValueType, Alignment>::value>
+struct packed_endian_specific_integral {
+ using value_type = ValueType;
+ static constexpr endianness endian = Endian;
+ static constexpr std::size_t alignment = Alignment;
+
+ packed_endian_specific_integral() = default;
+
+ explicit packed_endian_specific_integral(value_type val) { *this = val; }
+
+ operator value_type() const {
+ return endian::read<value_type, endian, alignment>(
+ (const void*)Value.buffer);
+ }
+
+ void operator=(value_type newValue) {
+ endian::write<value_type, endian, alignment>(
+ (void*)Value.buffer, newValue);
+ }
+
+ packed_endian_specific_integral &operator+=(value_type newValue) {
+ *this = *this + newValue;
+ return *this;
+ }
+
+ packed_endian_specific_integral &operator-=(value_type newValue) {
+ *this = *this - newValue;
+ return *this;
+ }
+
+ packed_endian_specific_integral &operator|=(value_type newValue) {
+ *this = *this | newValue;
+ return *this;
+ }
+
+ packed_endian_specific_integral &operator&=(value_type newValue) {
+ *this = *this & newValue;
+ return *this;
+ }
+
+private:
+ struct {
+ alignas(ALIGN) char buffer[sizeof(value_type)];
+ } Value;
+
+public:
+ struct ref {
+ explicit ref(void *Ptr) : Ptr(Ptr) {}
+
+ operator value_type() const {
+ return endian::read<value_type, endian, alignment>(Ptr);
+ }
+
+ void operator=(value_type NewValue) {
+ endian::write<value_type, endian, alignment>(Ptr, NewValue);
+ }
+
+ private:
+ void *Ptr;
+ };
+};
+
+} // end namespace detail
+
+using ulittle16_t =
+ detail::packed_endian_specific_integral<uint16_t, little, unaligned>;
+using ulittle32_t =
+ detail::packed_endian_specific_integral<uint32_t, little, unaligned>;
+using ulittle64_t =
+ detail::packed_endian_specific_integral<uint64_t, little, unaligned>;
+
+using little16_t =
+ detail::packed_endian_specific_integral<int16_t, little, unaligned>;
+using little32_t =
+ detail::packed_endian_specific_integral<int32_t, little, unaligned>;
+using little64_t =
+ detail::packed_endian_specific_integral<int64_t, little, unaligned>;
+
+using aligned_ulittle16_t =
+ detail::packed_endian_specific_integral<uint16_t, little, aligned>;
+using aligned_ulittle32_t =
+ detail::packed_endian_specific_integral<uint32_t, little, aligned>;
+using aligned_ulittle64_t =
+ detail::packed_endian_specific_integral<uint64_t, little, aligned>;
+
+using aligned_little16_t =
+ detail::packed_endian_specific_integral<int16_t, little, aligned>;
+using aligned_little32_t =
+ detail::packed_endian_specific_integral<int32_t, little, aligned>;
+using aligned_little64_t =
+ detail::packed_endian_specific_integral<int64_t, little, aligned>;
+
+using ubig16_t =
+ detail::packed_endian_specific_integral<uint16_t, big, unaligned>;
+using ubig32_t =
+ detail::packed_endian_specific_integral<uint32_t, big, unaligned>;
+using ubig64_t =
+ detail::packed_endian_specific_integral<uint64_t, big, unaligned>;
+
+using big16_t =
+ detail::packed_endian_specific_integral<int16_t, big, unaligned>;
+using big32_t =
+ detail::packed_endian_specific_integral<int32_t, big, unaligned>;
+using big64_t =
+ detail::packed_endian_specific_integral<int64_t, big, unaligned>;
+
+using aligned_ubig16_t =
+ detail::packed_endian_specific_integral<uint16_t, big, aligned>;
+using aligned_ubig32_t =
+ detail::packed_endian_specific_integral<uint32_t, big, aligned>;
+using aligned_ubig64_t =
+ detail::packed_endian_specific_integral<uint64_t, big, aligned>;
+
+using aligned_big16_t =
+ detail::packed_endian_specific_integral<int16_t, big, aligned>;
+using aligned_big32_t =
+ detail::packed_endian_specific_integral<int32_t, big, aligned>;
+using aligned_big64_t =
+ detail::packed_endian_specific_integral<int64_t, big, aligned>;
+
+using unaligned_uint16_t =
+ detail::packed_endian_specific_integral<uint16_t, native, unaligned>;
+using unaligned_uint32_t =
+ detail::packed_endian_specific_integral<uint32_t, native, unaligned>;
+using unaligned_uint64_t =
+ detail::packed_endian_specific_integral<uint64_t, native, unaligned>;
+
+using unaligned_int16_t =
+ detail::packed_endian_specific_integral<int16_t, native, unaligned>;
+using unaligned_int32_t =
+ detail::packed_endian_specific_integral<int32_t, native, unaligned>;
+using unaligned_int64_t =
+ detail::packed_endian_specific_integral<int64_t, native, unaligned>;
+
+template <typename T>
+using little_t = detail::packed_endian_specific_integral<T, little, unaligned>;
+template <typename T>
+using big_t = detail::packed_endian_specific_integral<T, big, unaligned>;
+
+template <typename T>
+using aligned_little_t =
+ detail::packed_endian_specific_integral<T, little, aligned>;
+template <typename T>
+using aligned_big_t = detail::packed_endian_specific_integral<T, big, aligned>;
+
+namespace endian {
+
+template <typename T> inline T read(const void *P, endianness E) {
+ return read<T, unaligned>(P, E);
+}
+
+template <typename T, endianness E> inline T read(const void *P) {
+ return *(const detail::packed_endian_specific_integral<T, E, unaligned> *)P;
+}
+
+inline uint16_t read16(const void *P, endianness E) {
+ return read<uint16_t>(P, E);
+}
+inline uint32_t read32(const void *P, endianness E) {
+ return read<uint32_t>(P, E);
+}
+inline uint64_t read64(const void *P, endianness E) {
+ return read<uint64_t>(P, E);
+}
+
+template <endianness E> inline uint16_t read16(const void *P) {
+ return read<uint16_t, E>(P);
+}
+template <endianness E> inline uint32_t read32(const void *P) {
+ return read<uint32_t, E>(P);
+}
+template <endianness E> inline uint64_t read64(const void *P) {
+ return read<uint64_t, E>(P);
+}
+
+inline uint16_t read16le(const void *P) { return read16<little>(P); }
+inline uint32_t read32le(const void *P) { return read32<little>(P); }
+inline uint64_t read64le(const void *P) { return read64<little>(P); }
+inline uint16_t read16be(const void *P) { return read16<big>(P); }
+inline uint32_t read32be(const void *P) { return read32<big>(P); }
+inline uint64_t read64be(const void *P) { return read64<big>(P); }
+
+template <typename T> inline void write(void *P, T V, endianness E) {
+ write<T, unaligned>(P, V, E);
+}
+
+template <typename T, endianness E> inline void write(void *P, T V) {
+ *(detail::packed_endian_specific_integral<T, E, unaligned> *)P = V;
+}
+
+inline void write16(void *P, uint16_t V, endianness E) {
+ write<uint16_t>(P, V, E);
+}
+inline void write32(void *P, uint32_t V, endianness E) {
+ write<uint32_t>(P, V, E);
+}
+inline void write64(void *P, uint64_t V, endianness E) {
+ write<uint64_t>(P, V, E);
+}
+
+template <endianness E> inline void write16(void *P, uint16_t V) {
+ write<uint16_t, E>(P, V);
+}
+template <endianness E> inline void write32(void *P, uint32_t V) {
+ write<uint32_t, E>(P, V);
+}
+template <endianness E> inline void write64(void *P, uint64_t V) {
+ write<uint64_t, E>(P, V);
+}
+
+inline void write16le(void *P, uint16_t V) { write16<little>(P, V); }
+inline void write32le(void *P, uint32_t V) { write32<little>(P, V); }
+inline void write64le(void *P, uint64_t V) { write64<little>(P, V); }
+inline void write16be(void *P, uint16_t V) { write16<big>(P, V); }
+inline void write32be(void *P, uint32_t V) { write32<big>(P, V); }
+inline void write64be(void *P, uint64_t V) { write64<big>(P, V); }
+
+} // end namespace endian
+
+} // end namespace support
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_ENDIAN_H
diff --git a/third_party/llvm-project/include/llvm/Support/Errc.h b/third_party/llvm-project/include/llvm/Support/Errc.h
new file mode 100644
index 000000000..9be8e5705
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Errc.h
@@ -0,0 +1,86 @@
+//===- llvm/Support/Errc.h - Defines the llvm::errc enum --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// While std::error_code works OK on all platforms we use, there are some
+// some problems with std::errc that can be avoided by using our own
+// enumeration:
+//
+// * std::errc is a namespace in some implementations. That meas that ADL
+// doesn't work and it is sometimes necessary to write std::make_error_code
+// or in templates:
+// using std::make_error_code;
+// make_error_code(...);
+//
+// with this enum it is safe to always just use make_error_code.
+//
+// * Some implementations define fewer names than others. This header has
+// the intersection of all the ones we support.
+//
+// * std::errc is just marked with is_error_condition_enum. This means that
+// common patters like AnErrorCode == errc::no_such_file_or_directory take
+// 4 virtual calls instead of two comparisons.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ERRC_H
+#define LLVM_SUPPORT_ERRC_H
+
+#include <system_error>
+
+namespace llvm {
+enum class errc {
+ argument_list_too_long = int(std::errc::argument_list_too_long),
+ argument_out_of_domain = int(std::errc::argument_out_of_domain),
+ bad_address = int(std::errc::bad_address),
+ bad_file_descriptor = int(std::errc::bad_file_descriptor),
+ broken_pipe = int(std::errc::broken_pipe),
+ device_or_resource_busy = int(std::errc::device_or_resource_busy),
+ directory_not_empty = int(std::errc::directory_not_empty),
+ executable_format_error = int(std::errc::executable_format_error),
+ file_exists = int(std::errc::file_exists),
+ file_too_large = int(std::errc::file_too_large),
+ filename_too_long = int(std::errc::filename_too_long),
+ function_not_supported = int(std::errc::function_not_supported),
+ illegal_byte_sequence = int(std::errc::illegal_byte_sequence),
+ inappropriate_io_control_operation =
+ int(std::errc::inappropriate_io_control_operation),
+ interrupted = int(std::errc::interrupted),
+ invalid_argument = int(std::errc::invalid_argument),
+ invalid_seek = int(std::errc::invalid_seek),
+ io_error = int(std::errc::io_error),
+ is_a_directory = int(std::errc::is_a_directory),
+ no_child_process = int(std::errc::no_child_process),
+ no_lock_available = int(std::errc::no_lock_available),
+ no_space_on_device = int(std::errc::no_space_on_device),
+ no_such_device_or_address = int(std::errc::no_such_device_or_address),
+ no_such_device = int(std::errc::no_such_device),
+ no_such_file_or_directory = int(std::errc::no_such_file_or_directory),
+ no_such_process = int(std::errc::no_such_process),
+ not_a_directory = int(std::errc::not_a_directory),
+ not_enough_memory = int(std::errc::not_enough_memory),
+ not_supported = int(std::errc::not_supported),
+ operation_not_permitted = int(std::errc::operation_not_permitted),
+ permission_denied = int(std::errc::permission_denied),
+ read_only_file_system = int(std::errc::read_only_file_system),
+ resource_deadlock_would_occur = int(std::errc::resource_deadlock_would_occur),
+ resource_unavailable_try_again =
+ int(std::errc::resource_unavailable_try_again),
+ result_out_of_range = int(std::errc::result_out_of_range),
+ too_many_files_open_in_system = int(std::errc::too_many_files_open_in_system),
+ too_many_files_open = int(std::errc::too_many_files_open),
+ too_many_links = int(std::errc::too_many_links)
+};
+
+inline std::error_code make_error_code(errc E) {
+ return std::error_code(static_cast<int>(E), std::generic_category());
+}
+}
+
+namespace std {
+template <> struct is_error_code_enum<llvm::errc> : std::true_type {};
+}
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Errno.h b/third_party/llvm-project/include/llvm/Support/Errno.h
new file mode 100644
index 000000000..aedb5fb29
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Errno.h
@@ -0,0 +1,46 @@
+//===- llvm/Support/Errno.h - Portable+convenient errno handling -*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares some portable and convenient functions to deal with errno.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ERRNO_H
+#define LLVM_SUPPORT_ERRNO_H
+
+#include <cerrno>
+#include <string>
+#include <type_traits>
+
+namespace llvm {
+namespace sys {
+
+/// Returns a string representation of the errno value, using whatever
+/// thread-safe variant of strerror() is available. Be sure to call this
+/// immediately after the function that set errno, or errno may have been
+/// overwritten by an intervening call.
+std::string StrError();
+
+/// Like the no-argument version above, but uses \p errnum instead of errno.
+std::string StrError(int errnum);
+
+template <typename FailT, typename Fun, typename... Args>
+inline auto RetryAfterSignal(const FailT &Fail, const Fun &F,
+ const Args &... As) -> decltype(F(As...)) {
+ decltype(F(As...)) Res;
+ do {
+ errno = 0;
+ Res = F(As...);
+ } while (Res == Fail && errno == EINTR);
+ return Res;
+}
+
+} // namespace sys
+} // namespace llvm
+
+#endif // LLVM_SYSTEM_ERRNO_H
diff --git a/third_party/llvm-project/include/llvm/Support/Error.h b/third_party/llvm-project/include/llvm/Support/Error.h
new file mode 100644
index 000000000..350877a21
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Error.h
@@ -0,0 +1,1351 @@
+//===- llvm/Support/Error.h - Recoverable error handling --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines an API used to report recoverable errors.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ERROR_H
+#define LLVM_SUPPORT_ERROR_H
+
+#include "llvm-c/Error.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Config/abi-breaking.h"
+#include "llvm/Support/AlignOf.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstdint>
+#include <cstdlib>
+#include <functional>
+#include <memory>
+#include <new>
+#include <string>
+#include <system_error>
+#include <type_traits>
+#include <utility>
+#include <vector>
+
+namespace llvm {
+
+class ErrorSuccess;
+
+/// Base class for error info classes. Do not extend this directly: Extend
+/// the ErrorInfo template subclass instead.
+class ErrorInfoBase {
+public:
+ virtual ~ErrorInfoBase() = default;
+
+ /// Print an error message to an output stream.
+ virtual void log(raw_ostream &OS) const = 0;
+
+ /// Return the error message as a string.
+ virtual std::string message() const {
+ std::string Msg;
+ raw_string_ostream OS(Msg);
+ log(OS);
+ return OS.str();
+ }
+
+ /// Convert this error to a std::error_code.
+ ///
+ /// This is a temporary crutch to enable interaction with code still
+ /// using std::error_code. It will be removed in the future.
+ virtual std::error_code convertToErrorCode() const = 0;
+
+ // Returns the class ID for this type.
+ static const void *classID() { return &ID; }
+
+ // Returns the class ID for the dynamic type of this ErrorInfoBase instance.
+ virtual const void *dynamicClassID() const = 0;
+
+ // Check whether this instance is a subclass of the class identified by
+ // ClassID.
+ virtual bool isA(const void *const ClassID) const {
+ return ClassID == classID();
+ }
+
+ // Check whether this instance is a subclass of ErrorInfoT.
+ template <typename ErrorInfoT> bool isA() const {
+ return isA(ErrorInfoT::classID());
+ }
+
+private:
+ virtual void anchor();
+
+ static char ID;
+};
+
+/// Lightweight error class with error context and mandatory checking.
+///
+/// Instances of this class wrap a ErrorInfoBase pointer. Failure states
+/// are represented by setting the pointer to a ErrorInfoBase subclass
+/// instance containing information describing the failure. Success is
+/// represented by a null pointer value.
+///
+/// Instances of Error also contains a 'Checked' flag, which must be set
+/// before the destructor is called, otherwise the destructor will trigger a
+/// runtime error. This enforces at runtime the requirement that all Error
+/// instances be checked or returned to the caller.
+///
+/// There are two ways to set the checked flag, depending on what state the
+/// Error instance is in. For Error instances indicating success, it
+/// is sufficient to invoke the boolean conversion operator. E.g.:
+///
+/// @code{.cpp}
+/// Error foo(<...>);
+///
+/// if (auto E = foo(<...>))
+/// return E; // <- Return E if it is in the error state.
+/// // We have verified that E was in the success state. It can now be safely
+/// // destroyed.
+/// @endcode
+///
+/// A success value *can not* be dropped. For example, just calling 'foo(<...>)'
+/// without testing the return value will raise a runtime error, even if foo
+/// returns success.
+///
+/// For Error instances representing failure, you must use either the
+/// handleErrors or handleAllErrors function with a typed handler. E.g.:
+///
+/// @code{.cpp}
+/// class MyErrorInfo : public ErrorInfo<MyErrorInfo> {
+/// // Custom error info.
+/// };
+///
+/// Error foo(<...>) { return make_error<MyErrorInfo>(...); }
+///
+/// auto E = foo(<...>); // <- foo returns failure with MyErrorInfo.
+/// auto NewE =
+/// handleErrors(E,
+/// [](const MyErrorInfo &M) {
+/// // Deal with the error.
+/// },
+/// [](std::unique_ptr<OtherError> M) -> Error {
+/// if (canHandle(*M)) {
+/// // handle error.
+/// return Error::success();
+/// }
+/// // Couldn't handle this error instance. Pass it up the stack.
+/// return Error(std::move(M));
+/// );
+/// // Note - we must check or return NewE in case any of the handlers
+/// // returned a new error.
+/// @endcode
+///
+/// The handleAllErrors function is identical to handleErrors, except
+/// that it has a void return type, and requires all errors to be handled and
+/// no new errors be returned. It prevents errors (assuming they can all be
+/// handled) from having to be bubbled all the way to the top-level.
+///
+/// *All* Error instances must be checked before destruction, even if
+/// they're moved-assigned or constructed from Success values that have already
+/// been checked. This enforces checking through all levels of the call stack.
+class LLVM_NODISCARD Error {
+ // Both ErrorList and FileError need to be able to yank ErrorInfoBase
+ // pointers out of this class to add to the error list.
+ friend class ErrorList;
+ friend class FileError;
+
+ // handleErrors needs to be able to set the Checked flag.
+ template <typename... HandlerTs>
+ friend Error handleErrors(Error E, HandlerTs &&... Handlers);
+
+ // Expected<T> needs to be able to steal the payload when constructed from an
+ // error.
+ template <typename T> friend class Expected;
+
+ // wrap needs to be able to steal the payload.
+ friend LLVMErrorRef wrap(Error);
+
+protected:
+ /// Create a success value. Prefer using 'Error::success()' for readability
+ Error() {
+ setPtr(nullptr);
+ setChecked(false);
+ }
+
+public:
+ /// Create a success value.
+ static ErrorSuccess success();
+
+ // Errors are not copy-constructable.
+ Error(const Error &Other) = delete;
+
+ /// Move-construct an error value. The newly constructed error is considered
+ /// unchecked, even if the source error had been checked. The original error
+ /// becomes a checked Success value, regardless of its original state.
+ Error(Error &&Other) {
+ setChecked(true);
+ *this = std::move(Other);
+ }
+
+ /// Create an error value. Prefer using the 'make_error' function, but
+ /// this constructor can be useful when "re-throwing" errors from handlers.
+ Error(std::unique_ptr<ErrorInfoBase> Payload) {
+ setPtr(Payload.release());
+ setChecked(false);
+ }
+
+ // Errors are not copy-assignable.
+ Error &operator=(const Error &Other) = delete;
+
+ /// Move-assign an error value. The current error must represent success, you
+ /// you cannot overwrite an unhandled error. The current error is then
+ /// considered unchecked. The source error becomes a checked success value,
+ /// regardless of its original state.
+ Error &operator=(Error &&Other) {
+ // Don't allow overwriting of unchecked values.
+ assertIsChecked();
+ setPtr(Other.getPtr());
+
+ // This Error is unchecked, even if the source error was checked.
+ setChecked(false);
+
+ // Null out Other's payload and set its checked bit.
+ Other.setPtr(nullptr);
+ Other.setChecked(true);
+
+ return *this;
+ }
+
+ /// Destroy a Error. Fails with a call to abort() if the error is
+ /// unchecked.
+ ~Error() {
+ assertIsChecked();
+ delete getPtr();
+ }
+
+ /// Bool conversion. Returns true if this Error is in a failure state,
+ /// and false if it is in an accept state. If the error is in a Success state
+ /// it will be considered checked.
+ explicit operator bool() {
+ setChecked(getPtr() == nullptr);
+ return getPtr() != nullptr;
+ }
+
+ /// Check whether one error is a subclass of another.
+ template <typename ErrT> bool isA() const {
+ return getPtr() && getPtr()->isA(ErrT::classID());
+ }
+
+ /// Returns the dynamic class id of this error, or null if this is a success
+ /// value.
+ const void* dynamicClassID() const {
+ if (!getPtr())
+ return nullptr;
+ return getPtr()->dynamicClassID();
+ }
+
+private:
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ // assertIsChecked() happens very frequently, but under normal circumstances
+ // is supposed to be a no-op. So we want it to be inlined, but having a bunch
+ // of debug prints can cause the function to be too large for inlining. So
+ // it's important that we define this function out of line so that it can't be
+ // inlined.
+ LLVM_ATTRIBUTE_NORETURN
+ void fatalUncheckedError() const;
+#endif
+
+ void assertIsChecked() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ if (LLVM_UNLIKELY(!getChecked() || getPtr()))
+ fatalUncheckedError();
+#endif
+ }
+
+ ErrorInfoBase *getPtr() const {
+ return reinterpret_cast<ErrorInfoBase*>(
+ reinterpret_cast<uintptr_t>(Payload) &
+ ~static_cast<uintptr_t>(0x1));
+ }
+
+ void setPtr(ErrorInfoBase *EI) {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ Payload = reinterpret_cast<ErrorInfoBase*>(
+ (reinterpret_cast<uintptr_t>(EI) &
+ ~static_cast<uintptr_t>(0x1)) |
+ (reinterpret_cast<uintptr_t>(Payload) & 0x1));
+#else
+ Payload = EI;
+#endif
+ }
+
+ bool getChecked() const {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ return (reinterpret_cast<uintptr_t>(Payload) & 0x1) == 0;
+#else
+ return true;
+#endif
+ }
+
+ void setChecked(bool V) {
+ Payload = reinterpret_cast<ErrorInfoBase*>(
+ (reinterpret_cast<uintptr_t>(Payload) &
+ ~static_cast<uintptr_t>(0x1)) |
+ (V ? 0 : 1));
+ }
+
+ std::unique_ptr<ErrorInfoBase> takePayload() {
+ std::unique_ptr<ErrorInfoBase> Tmp(getPtr());
+ setPtr(nullptr);
+ setChecked(true);
+ return Tmp;
+ }
+
+ friend raw_ostream &operator<<(raw_ostream &OS, const Error &E) {
+ if (auto P = E.getPtr())
+ P->log(OS);
+ else
+ OS << "success";
+ return OS;
+ }
+
+ ErrorInfoBase *Payload = nullptr;
+};
+
+/// Subclass of Error for the sole purpose of identifying the success path in
+/// the type system. This allows to catch invalid conversion to Expected<T> at
+/// compile time.
+class ErrorSuccess final : public Error {};
+
+inline ErrorSuccess Error::success() { return ErrorSuccess(); }
+
+/// Make a Error instance representing failure using the given error info
+/// type.
+template <typename ErrT, typename... ArgTs> Error make_error(ArgTs &&... Args) {
+ return Error(std::make_unique<ErrT>(std::forward<ArgTs>(Args)...));
+}
+
+/// Base class for user error types. Users should declare their error types
+/// like:
+///
+/// class MyError : public ErrorInfo<MyError> {
+/// ....
+/// };
+///
+/// This class provides an implementation of the ErrorInfoBase::kind
+/// method, which is used by the Error RTTI system.
+template <typename ThisErrT, typename ParentErrT = ErrorInfoBase>
+class ErrorInfo : public ParentErrT {
+public:
+ using ParentErrT::ParentErrT; // inherit constructors
+
+ static const void *classID() { return &ThisErrT::ID; }
+
+ const void *dynamicClassID() const override { return &ThisErrT::ID; }
+
+ bool isA(const void *const ClassID) const override {
+ return ClassID == classID() || ParentErrT::isA(ClassID);
+ }
+};
+
+/// Special ErrorInfo subclass representing a list of ErrorInfos.
+/// Instances of this class are constructed by joinError.
+class ErrorList final : public ErrorInfo<ErrorList> {
+ // handleErrors needs to be able to iterate the payload list of an
+ // ErrorList.
+ template <typename... HandlerTs>
+ friend Error handleErrors(Error E, HandlerTs &&... Handlers);
+
+ // joinErrors is implemented in terms of join.
+ friend Error joinErrors(Error, Error);
+
+public:
+ void log(raw_ostream &OS) const override {
+ OS << "Multiple errors:\n";
+ for (auto &ErrPayload : Payloads) {
+ ErrPayload->log(OS);
+ OS << "\n";
+ }
+ }
+
+ std::error_code convertToErrorCode() const override;
+
+ // Used by ErrorInfo::classID.
+ static char ID;
+
+private:
+ ErrorList(std::unique_ptr<ErrorInfoBase> Payload1,
+ std::unique_ptr<ErrorInfoBase> Payload2) {
+ assert(!Payload1->isA<ErrorList>() && !Payload2->isA<ErrorList>() &&
+ "ErrorList constructor payloads should be singleton errors");
+ Payloads.push_back(std::move(Payload1));
+ Payloads.push_back(std::move(Payload2));
+ }
+
+ static Error join(Error E1, Error E2) {
+ if (!E1)
+ return E2;
+ if (!E2)
+ return E1;
+ if (E1.isA<ErrorList>()) {
+ auto &E1List = static_cast<ErrorList &>(*E1.getPtr());
+ if (E2.isA<ErrorList>()) {
+ auto E2Payload = E2.takePayload();
+ auto &E2List = static_cast<ErrorList &>(*E2Payload);
+ for (auto &Payload : E2List.Payloads)
+ E1List.Payloads.push_back(std::move(Payload));
+ } else
+ E1List.Payloads.push_back(E2.takePayload());
+
+ return E1;
+ }
+ if (E2.isA<ErrorList>()) {
+ auto &E2List = static_cast<ErrorList &>(*E2.getPtr());
+ E2List.Payloads.insert(E2List.Payloads.begin(), E1.takePayload());
+ return E2;
+ }
+ return Error(std::unique_ptr<ErrorList>(
+ new ErrorList(E1.takePayload(), E2.takePayload())));
+ }
+
+ std::vector<std::unique_ptr<ErrorInfoBase>> Payloads;
+};
+
+/// Concatenate errors. The resulting Error is unchecked, and contains the
+/// ErrorInfo(s), if any, contained in E1, followed by the
+/// ErrorInfo(s), if any, contained in E2.
+inline Error joinErrors(Error E1, Error E2) {
+ return ErrorList::join(std::move(E1), std::move(E2));
+}
+
+/// Tagged union holding either a T or a Error.
+///
+/// This class parallels ErrorOr, but replaces error_code with Error. Since
+/// Error cannot be copied, this class replaces getError() with
+/// takeError(). It also adds an bool errorIsA<ErrT>() method for testing the
+/// error class type.
+template <class T> class LLVM_NODISCARD Expected {
+ template <class T1> friend class ExpectedAsOutParameter;
+ template <class OtherT> friend class Expected;
+
+ static const bool isRef = std::is_reference<T>::value;
+
+ using wrap = std::reference_wrapper<typename std::remove_reference<T>::type>;
+
+ using error_type = std::unique_ptr<ErrorInfoBase>;
+
+public:
+ using storage_type = typename std::conditional<isRef, wrap, T>::type;
+ using value_type = T;
+
+private:
+ using reference = typename std::remove_reference<T>::type &;
+ using const_reference = const typename std::remove_reference<T>::type &;
+ using pointer = typename std::remove_reference<T>::type *;
+ using const_pointer = const typename std::remove_reference<T>::type *;
+
+public:
+ /// Create an Expected<T> error value from the given Error.
+ Expected(Error Err)
+ : HasError(true)
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ // Expected is unchecked upon construction in Debug builds.
+ , Unchecked(true)
+#endif
+ {
+ assert(Err && "Cannot create Expected<T> from Error success value.");
+ new (getErrorStorage()) error_type(Err.takePayload());
+ }
+
+ /// Forbid to convert from Error::success() implicitly, this avoids having
+ /// Expected<T> foo() { return Error::success(); } which compiles otherwise
+ /// but triggers the assertion above.
+ Expected(ErrorSuccess) = delete;
+
+ /// Create an Expected<T> success value from the given OtherT value, which
+ /// must be convertible to T.
+ template <typename OtherT>
+ Expected(OtherT &&Val,
+ typename std::enable_if<std::is_convertible<OtherT, T>::value>::type
+ * = nullptr)
+ : HasError(false)
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ // Expected is unchecked upon construction in Debug builds.
+ , Unchecked(true)
+#endif
+ {
+ new (getStorage()) storage_type(std::forward<OtherT>(Val));
+ }
+
+ /// Move construct an Expected<T> value.
+ Expected(Expected &&Other) { moveConstruct(std::move(Other)); }
+
+ /// Move construct an Expected<T> value from an Expected<OtherT>, where OtherT
+ /// must be convertible to T.
+ template <class OtherT>
+ Expected(Expected<OtherT> &&Other,
+ typename std::enable_if<std::is_convertible<OtherT, T>::value>::type
+ * = nullptr) {
+ moveConstruct(std::move(Other));
+ }
+
+ /// Move construct an Expected<T> value from an Expected<OtherT>, where OtherT
+ /// isn't convertible to T.
+ template <class OtherT>
+ explicit Expected(
+ Expected<OtherT> &&Other,
+ typename std::enable_if<!std::is_convertible<OtherT, T>::value>::type * =
+ nullptr) {
+ moveConstruct(std::move(Other));
+ }
+
+ /// Move-assign from another Expected<T>.
+ Expected &operator=(Expected &&Other) {
+ moveAssign(std::move(Other));
+ return *this;
+ }
+
+ /// Destroy an Expected<T>.
+ ~Expected() {
+ assertIsChecked();
+ if (!HasError)
+ getStorage()->~storage_type();
+ else
+ getErrorStorage()->~error_type();
+ }
+
+ /// Return false if there is an error.
+ explicit operator bool() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ Unchecked = HasError;
+#endif
+ return !HasError;
+ }
+
+ /// Returns a reference to the stored T value.
+ reference get() {
+ assertIsChecked();
+ return *getStorage();
+ }
+
+ /// Returns a const reference to the stored T value.
+ const_reference get() const {
+ assertIsChecked();
+ return const_cast<Expected<T> *>(this)->get();
+ }
+
+ /// Check that this Expected<T> is an error of type ErrT.
+ template <typename ErrT> bool errorIsA() const {
+ return HasError && (*getErrorStorage())->template isA<ErrT>();
+ }
+
+ /// Take ownership of the stored error.
+ /// After calling this the Expected<T> is in an indeterminate state that can
+ /// only be safely destructed. No further calls (beside the destructor) should
+ /// be made on the Expected<T> value.
+ Error takeError() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ Unchecked = false;
+#endif
+ return HasError ? Error(std::move(*getErrorStorage())) : Error::success();
+ }
+
+ /// Returns a pointer to the stored T value.
+ pointer operator->() {
+ assertIsChecked();
+ return toPointer(getStorage());
+ }
+
+ /// Returns a const pointer to the stored T value.
+ const_pointer operator->() const {
+ assertIsChecked();
+ return toPointer(getStorage());
+ }
+
+ /// Returns a reference to the stored T value.
+ reference operator*() {
+ assertIsChecked();
+ return *getStorage();
+ }
+
+ /// Returns a const reference to the stored T value.
+ const_reference operator*() const {
+ assertIsChecked();
+ return *getStorage();
+ }
+
+private:
+ template <class T1>
+ static bool compareThisIfSameType(const T1 &a, const T1 &b) {
+ return &a == &b;
+ }
+
+ template <class T1, class T2>
+ static bool compareThisIfSameType(const T1 &a, const T2 &b) {
+ return false;
+ }
+
+ template <class OtherT> void moveConstruct(Expected<OtherT> &&Other) {
+ HasError = Other.HasError;
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ Unchecked = true;
+ Other.Unchecked = false;
+#endif
+
+ if (!HasError)
+ new (getStorage()) storage_type(std::move(*Other.getStorage()));
+ else
+ new (getErrorStorage()) error_type(std::move(*Other.getErrorStorage()));
+ }
+
+ template <class OtherT> void moveAssign(Expected<OtherT> &&Other) {
+ assertIsChecked();
+
+ if (compareThisIfSameType(*this, Other))
+ return;
+
+ this->~Expected();
+ new (this) Expected(std::move(Other));
+ }
+
+ pointer toPointer(pointer Val) { return Val; }
+
+ const_pointer toPointer(const_pointer Val) const { return Val; }
+
+ pointer toPointer(wrap *Val) { return &Val->get(); }
+
+ const_pointer toPointer(const wrap *Val) const { return &Val->get(); }
+
+ storage_type *getStorage() {
+ assert(!HasError && "Cannot get value when an error exists!");
+ return reinterpret_cast<storage_type *>(TStorage.buffer);
+ }
+
+ const storage_type *getStorage() const {
+ assert(!HasError && "Cannot get value when an error exists!");
+ return reinterpret_cast<const storage_type *>(TStorage.buffer);
+ }
+
+ error_type *getErrorStorage() {
+ assert(HasError && "Cannot get error when a value exists!");
+ return reinterpret_cast<error_type *>(ErrorStorage.buffer);
+ }
+
+ const error_type *getErrorStorage() const {
+ assert(HasError && "Cannot get error when a value exists!");
+ return reinterpret_cast<const error_type *>(ErrorStorage.buffer);
+ }
+
+ // Used by ExpectedAsOutParameter to reset the checked flag.
+ void setUnchecked() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ Unchecked = true;
+#endif
+ }
+
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ LLVM_ATTRIBUTE_NORETURN
+ LLVM_ATTRIBUTE_NOINLINE
+ void fatalUncheckedExpected() const {
+ dbgs() << "Expected<T> must be checked before access or destruction.\n";
+ if (HasError) {
+ dbgs() << "Unchecked Expected<T> contained error:\n";
+ (*getErrorStorage())->log(dbgs());
+ } else
+ dbgs() << "Expected<T> value was in success state. (Note: Expected<T> "
+ "values in success mode must still be checked prior to being "
+ "destroyed).\n";
+ abort();
+ }
+#endif
+
+ void assertIsChecked() {
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ if (LLVM_UNLIKELY(Unchecked))
+ fatalUncheckedExpected();
+#endif
+ }
+
+ union {
+ AlignedCharArrayUnion<storage_type> TStorage;
+ AlignedCharArrayUnion<error_type> ErrorStorage;
+ };
+ bool HasError : 1;
+#if LLVM_ENABLE_ABI_BREAKING_CHECKS
+ bool Unchecked : 1;
+#endif
+};
+
+/// Report a serious error, calling any installed error handler. See
+/// ErrorHandling.h.
+LLVM_ATTRIBUTE_NORETURN void report_fatal_error(Error Err,
+ bool gen_crash_diag = true);
+
+/// Report a fatal error if Err is a failure value.
+///
+/// This function can be used to wrap calls to fallible functions ONLY when it
+/// is known that the Error will always be a success value. E.g.
+///
+/// @code{.cpp}
+/// // foo only attempts the fallible operation if DoFallibleOperation is
+/// // true. If DoFallibleOperation is false then foo always returns
+/// // Error::success().
+/// Error foo(bool DoFallibleOperation);
+///
+/// cantFail(foo(false));
+/// @endcode
+inline void cantFail(Error Err, const char *Msg = nullptr) {
+ if (Err) {
+ if (!Msg)
+ Msg = "Failure value returned from cantFail wrapped call";
+#ifndef NDEBUG
+ std::string Str;
+ raw_string_ostream OS(Str);
+ OS << Msg << "\n" << Err;
+ Msg = OS.str().c_str();
+#endif
+ llvm_unreachable(Msg);
+ }
+}
+
+/// Report a fatal error if ValOrErr is a failure value, otherwise unwraps and
+/// returns the contained value.
+///
+/// This function can be used to wrap calls to fallible functions ONLY when it
+/// is known that the Error will always be a success value. E.g.
+///
+/// @code{.cpp}
+/// // foo only attempts the fallible operation if DoFallibleOperation is
+/// // true. If DoFallibleOperation is false then foo always returns an int.
+/// Expected<int> foo(bool DoFallibleOperation);
+///
+/// int X = cantFail(foo(false));
+/// @endcode
+template <typename T>
+T cantFail(Expected<T> ValOrErr, const char *Msg = nullptr) {
+ if (ValOrErr)
+ return std::move(*ValOrErr);
+ else {
+ if (!Msg)
+ Msg = "Failure value returned from cantFail wrapped call";
+#ifndef NDEBUG
+ std::string Str;
+ raw_string_ostream OS(Str);
+ auto E = ValOrErr.takeError();
+ OS << Msg << "\n" << E;
+ Msg = OS.str().c_str();
+#endif
+ llvm_unreachable(Msg);
+ }
+}
+
+/// Report a fatal error if ValOrErr is a failure value, otherwise unwraps and
+/// returns the contained reference.
+///
+/// This function can be used to wrap calls to fallible functions ONLY when it
+/// is known that the Error will always be a success value. E.g.
+///
+/// @code{.cpp}
+/// // foo only attempts the fallible operation if DoFallibleOperation is
+/// // true. If DoFallibleOperation is false then foo always returns a Bar&.
+/// Expected<Bar&> foo(bool DoFallibleOperation);
+///
+/// Bar &X = cantFail(foo(false));
+/// @endcode
+template <typename T>
+T& cantFail(Expected<T&> ValOrErr, const char *Msg = nullptr) {
+ if (ValOrErr)
+ return *ValOrErr;
+ else {
+ if (!Msg)
+ Msg = "Failure value returned from cantFail wrapped call";
+#ifndef NDEBUG
+ std::string Str;
+ raw_string_ostream OS(Str);
+ auto E = ValOrErr.takeError();
+ OS << Msg << "\n" << E;
+ Msg = OS.str().c_str();
+#endif
+ llvm_unreachable(Msg);
+ }
+}
+
+/// Helper for testing applicability of, and applying, handlers for
+/// ErrorInfo types.
+template <typename HandlerT>
+class ErrorHandlerTraits
+ : public ErrorHandlerTraits<decltype(
+ &std::remove_reference<HandlerT>::type::operator())> {};
+
+// Specialization functions of the form 'Error (const ErrT&)'.
+template <typename ErrT> class ErrorHandlerTraits<Error (&)(ErrT &)> {
+public:
+ static bool appliesTo(const ErrorInfoBase &E) {
+ return E.template isA<ErrT>();
+ }
+
+ template <typename HandlerT>
+ static Error apply(HandlerT &&H, std::unique_ptr<ErrorInfoBase> E) {
+ assert(appliesTo(*E) && "Applying incorrect handler");
+ return H(static_cast<ErrT &>(*E));
+ }
+};
+
+// Specialization functions of the form 'void (const ErrT&)'.
+template <typename ErrT> class ErrorHandlerTraits<void (&)(ErrT &)> {
+public:
+ static bool appliesTo(const ErrorInfoBase &E) {
+ return E.template isA<ErrT>();
+ }
+
+ template <typename HandlerT>
+ static Error apply(HandlerT &&H, std::unique_ptr<ErrorInfoBase> E) {
+ assert(appliesTo(*E) && "Applying incorrect handler");
+ H(static_cast<ErrT &>(*E));
+ return Error::success();
+ }
+};
+
+/// Specialization for functions of the form 'Error (std::unique_ptr<ErrT>)'.
+template <typename ErrT>
+class ErrorHandlerTraits<Error (&)(std::unique_ptr<ErrT>)> {
+public:
+ static bool appliesTo(const ErrorInfoBase &E) {
+ return E.template isA<ErrT>();
+ }
+
+ template <typename HandlerT>
+ static Error apply(HandlerT &&H, std::unique_ptr<ErrorInfoBase> E) {
+ assert(appliesTo(*E) && "Applying incorrect handler");
+ std::unique_ptr<ErrT> SubE(static_cast<ErrT *>(E.release()));
+ return H(std::move(SubE));
+ }
+};
+
+/// Specialization for functions of the form 'void (std::unique_ptr<ErrT>)'.
+template <typename ErrT>
+class ErrorHandlerTraits<void (&)(std::unique_ptr<ErrT>)> {
+public:
+ static bool appliesTo(const ErrorInfoBase &E) {
+ return E.template isA<ErrT>();
+ }
+
+ template <typename HandlerT>
+ static Error apply(HandlerT &&H, std::unique_ptr<ErrorInfoBase> E) {
+ assert(appliesTo(*E) && "Applying incorrect handler");
+ std::unique_ptr<ErrT> SubE(static_cast<ErrT *>(E.release()));
+ H(std::move(SubE));
+ return Error::success();
+ }
+};
+
+// Specialization for member functions of the form 'RetT (const ErrT&)'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(ErrT &)>
+ : public ErrorHandlerTraits<RetT (&)(ErrT &)> {};
+
+// Specialization for member functions of the form 'RetT (const ErrT&) const'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(ErrT &) const>
+ : public ErrorHandlerTraits<RetT (&)(ErrT &)> {};
+
+// Specialization for member functions of the form 'RetT (const ErrT&)'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(const ErrT &)>
+ : public ErrorHandlerTraits<RetT (&)(ErrT &)> {};
+
+// Specialization for member functions of the form 'RetT (const ErrT&) const'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(const ErrT &) const>
+ : public ErrorHandlerTraits<RetT (&)(ErrT &)> {};
+
+/// Specialization for member functions of the form
+/// 'RetT (std::unique_ptr<ErrT>)'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(std::unique_ptr<ErrT>)>
+ : public ErrorHandlerTraits<RetT (&)(std::unique_ptr<ErrT>)> {};
+
+/// Specialization for member functions of the form
+/// 'RetT (std::unique_ptr<ErrT>) const'.
+template <typename C, typename RetT, typename ErrT>
+class ErrorHandlerTraits<RetT (C::*)(std::unique_ptr<ErrT>) const>
+ : public ErrorHandlerTraits<RetT (&)(std::unique_ptr<ErrT>)> {};
+
+inline Error handleErrorImpl(std::unique_ptr<ErrorInfoBase> Payload) {
+ return Error(std::move(Payload));
+}
+
+template <typename HandlerT, typename... HandlerTs>
+Error handleErrorImpl(std::unique_ptr<ErrorInfoBase> Payload,
+ HandlerT &&Handler, HandlerTs &&... Handlers) {
+ if (ErrorHandlerTraits<HandlerT>::appliesTo(*Payload))
+ return ErrorHandlerTraits<HandlerT>::apply(std::forward<HandlerT>(Handler),
+ std::move(Payload));
+ return handleErrorImpl(std::move(Payload),
+ std::forward<HandlerTs>(Handlers)...);
+}
+
+/// Pass the ErrorInfo(s) contained in E to their respective handlers. Any
+/// unhandled errors (or Errors returned by handlers) are re-concatenated and
+/// returned.
+/// Because this function returns an error, its result must also be checked
+/// or returned. If you intend to handle all errors use handleAllErrors
+/// (which returns void, and will abort() on unhandled errors) instead.
+template <typename... HandlerTs>
+Error handleErrors(Error E, HandlerTs &&... Hs) {
+ if (!E)
+ return Error::success();
+
+ std::unique_ptr<ErrorInfoBase> Payload = E.takePayload();
+
+ if (Payload->isA<ErrorList>()) {
+ ErrorList &List = static_cast<ErrorList &>(*Payload);
+ Error R;
+ for (auto &P : List.Payloads)
+ R = ErrorList::join(
+ std::move(R),
+ handleErrorImpl(std::move(P), std::forward<HandlerTs>(Hs)...));
+ return R;
+ }
+
+ return handleErrorImpl(std::move(Payload), std::forward<HandlerTs>(Hs)...);
+}
+
+/// Behaves the same as handleErrors, except that by contract all errors
+/// *must* be handled by the given handlers (i.e. there must be no remaining
+/// errors after running the handlers, or llvm_unreachable is called).
+template <typename... HandlerTs>
+void handleAllErrors(Error E, HandlerTs &&... Handlers) {
+ cantFail(handleErrors(std::move(E), std::forward<HandlerTs>(Handlers)...));
+}
+
+/// Check that E is a non-error, then drop it.
+/// If E is an error, llvm_unreachable will be called.
+inline void handleAllErrors(Error E) {
+ cantFail(std::move(E));
+}
+
+/// Handle any errors (if present) in an Expected<T>, then try a recovery path.
+///
+/// If the incoming value is a success value it is returned unmodified. If it
+/// is a failure value then it the contained error is passed to handleErrors.
+/// If handleErrors is able to handle the error then the RecoveryPath functor
+/// is called to supply the final result. If handleErrors is not able to
+/// handle all errors then the unhandled errors are returned.
+///
+/// This utility enables the follow pattern:
+///
+/// @code{.cpp}
+/// enum FooStrategy { Aggressive, Conservative };
+/// Expected<Foo> foo(FooStrategy S);
+///
+/// auto ResultOrErr =
+/// handleExpected(
+/// foo(Aggressive),
+/// []() { return foo(Conservative); },
+/// [](AggressiveStrategyError&) {
+/// // Implicitly conusme this - we'll recover by using a conservative
+/// // strategy.
+/// });
+///
+/// @endcode
+template <typename T, typename RecoveryFtor, typename... HandlerTs>
+Expected<T> handleExpected(Expected<T> ValOrErr, RecoveryFtor &&RecoveryPath,
+ HandlerTs &&... Handlers) {
+ if (ValOrErr)
+ return ValOrErr;
+
+ if (auto Err = handleErrors(ValOrErr.takeError(),
+ std::forward<HandlerTs>(Handlers)...))
+ return std::move(Err);
+
+ return RecoveryPath();
+}
+
+/// Log all errors (if any) in E to OS. If there are any errors, ErrorBanner
+/// will be printed before the first one is logged. A newline will be printed
+/// after each error.
+///
+/// This function is compatible with the helpers from Support/WithColor.h. You
+/// can pass any of them as the OS. Please consider using them instead of
+/// including 'error: ' in the ErrorBanner.
+///
+/// This is useful in the base level of your program to allow clean termination
+/// (allowing clean deallocation of resources, etc.), while reporting error
+/// information to the user.
+void logAllUnhandledErrors(Error E, raw_ostream &OS, Twine ErrorBanner = {});
+
+/// Write all error messages (if any) in E to a string. The newline character
+/// is used to separate error messages.
+inline std::string toString(Error E) {
+ SmallVector<std::string, 2> Errors;
+ handleAllErrors(std::move(E), [&Errors](const ErrorInfoBase &EI) {
+ Errors.push_back(EI.message());
+ });
+ return join(Errors.begin(), Errors.end(), "\n");
+}
+
+/// Consume a Error without doing anything. This method should be used
+/// only where an error can be considered a reasonable and expected return
+/// value.
+///
+/// Uses of this method are potentially indicative of design problems: If it's
+/// legitimate to do nothing while processing an "error", the error-producer
+/// might be more clearly refactored to return an Optional<T>.
+inline void consumeError(Error Err) {
+ handleAllErrors(std::move(Err), [](const ErrorInfoBase &) {});
+}
+
+/// Convert an Expected to an Optional without doing anything. This method
+/// should be used only where an error can be considered a reasonable and
+/// expected return value.
+///
+/// Uses of this method are potentially indicative of problems: perhaps the
+/// error should be propagated further, or the error-producer should just
+/// return an Optional in the first place.
+template <typename T> Optional<T> expectedToOptional(Expected<T> &&E) {
+ if (E)
+ return std::move(*E);
+ consumeError(E.takeError());
+ return None;
+}
+
+/// Helper for converting an Error to a bool.
+///
+/// This method returns true if Err is in an error state, or false if it is
+/// in a success state. Puts Err in a checked state in both cases (unlike
+/// Error::operator bool(), which only does this for success states).
+inline bool errorToBool(Error Err) {
+ bool IsError = static_cast<bool>(Err);
+ if (IsError)
+ consumeError(std::move(Err));
+ return IsError;
+}
+
+/// Helper for Errors used as out-parameters.
+///
+/// This helper is for use with the Error-as-out-parameter idiom, where an error
+/// is passed to a function or method by reference, rather than being returned.
+/// In such cases it is helpful to set the checked bit on entry to the function
+/// so that the error can be written to (unchecked Errors abort on assignment)
+/// and clear the checked bit on exit so that clients cannot accidentally forget
+/// to check the result. This helper performs these actions automatically using
+/// RAII:
+///
+/// @code{.cpp}
+/// Result foo(Error &Err) {
+/// ErrorAsOutParameter ErrAsOutParam(&Err); // 'Checked' flag set
+/// // <body of foo>
+/// // <- 'Checked' flag auto-cleared when ErrAsOutParam is destructed.
+/// }
+/// @endcode
+///
+/// ErrorAsOutParameter takes an Error* rather than Error& so that it can be
+/// used with optional Errors (Error pointers that are allowed to be null). If
+/// ErrorAsOutParameter took an Error reference, an instance would have to be
+/// created inside every condition that verified that Error was non-null. By
+/// taking an Error pointer we can just create one instance at the top of the
+/// function.
+class ErrorAsOutParameter {
+public:
+ ErrorAsOutParameter(Error *Err) : Err(Err) {
+ // Raise the checked bit if Err is success.
+ if (Err)
+ (void)!!*Err;
+ }
+
+ ~ErrorAsOutParameter() {
+ // Clear the checked bit.
+ if (Err && !*Err)
+ *Err = Error::success();
+ }
+
+private:
+ Error *Err;
+};
+
+/// Helper for Expected<T>s used as out-parameters.
+///
+/// See ErrorAsOutParameter.
+template <typename T>
+class ExpectedAsOutParameter {
+public:
+ ExpectedAsOutParameter(Expected<T> *ValOrErr)
+ : ValOrErr(ValOrErr) {
+ if (ValOrErr)
+ (void)!!*ValOrErr;
+ }
+
+ ~ExpectedAsOutParameter() {
+ if (ValOrErr)
+ ValOrErr->setUnchecked();
+ }
+
+private:
+ Expected<T> *ValOrErr;
+};
+
+/// This class wraps a std::error_code in a Error.
+///
+/// This is useful if you're writing an interface that returns a Error
+/// (or Expected) and you want to call code that still returns
+/// std::error_codes.
+class ECError : public ErrorInfo<ECError> {
+ friend Error errorCodeToError(std::error_code);
+
+ virtual void anchor() override;
+
+public:
+ void setErrorCode(std::error_code EC) { this->EC = EC; }
+ std::error_code convertToErrorCode() const override { return EC; }
+ void log(raw_ostream &OS) const override { OS << EC.message(); }
+
+ // Used by ErrorInfo::classID.
+ static char ID;
+
+protected:
+ ECError() = default;
+ ECError(std::error_code EC) : EC(EC) {}
+
+ std::error_code EC;
+};
+
+/// The value returned by this function can be returned from convertToErrorCode
+/// for Error values where no sensible translation to std::error_code exists.
+/// It should only be used in this situation, and should never be used where a
+/// sensible conversion to std::error_code is available, as attempts to convert
+/// to/from this error will result in a fatal error. (i.e. it is a programmatic
+///error to try to convert such a value).
+std::error_code inconvertibleErrorCode();
+
+/// Helper for converting an std::error_code to a Error.
+Error errorCodeToError(std::error_code EC);
+
+/// Helper for converting an ECError to a std::error_code.
+///
+/// This method requires that Err be Error() or an ECError, otherwise it
+/// will trigger a call to abort().
+std::error_code errorToErrorCode(Error Err);
+
+/// Convert an ErrorOr<T> to an Expected<T>.
+template <typename T> Expected<T> errorOrToExpected(ErrorOr<T> &&EO) {
+ if (auto EC = EO.getError())
+ return errorCodeToError(EC);
+ return std::move(*EO);
+}
+
+/// Convert an Expected<T> to an ErrorOr<T>.
+template <typename T> ErrorOr<T> expectedToErrorOr(Expected<T> &&E) {
+ if (auto Err = E.takeError())
+ return errorToErrorCode(std::move(Err));
+ return std::move(*E);
+}
+
+/// This class wraps a string in an Error.
+///
+/// StringError is useful in cases where the client is not expected to be able
+/// to consume the specific error message programmatically (for example, if the
+/// error message is to be presented to the user).
+///
+/// StringError can also be used when additional information is to be printed
+/// along with a error_code message. Depending on the constructor called, this
+/// class can either display:
+/// 1. the error_code message (ECError behavior)
+/// 2. a string
+/// 3. the error_code message and a string
+///
+/// These behaviors are useful when subtyping is required; for example, when a
+/// specific library needs an explicit error type. In the example below,
+/// PDBError is derived from StringError:
+///
+/// @code{.cpp}
+/// Expected<int> foo() {
+/// return llvm::make_error<PDBError>(pdb_error_code::dia_failed_loading,
+/// "Additional information");
+/// }
+/// @endcode
+///
+class StringError : public ErrorInfo<StringError> {
+public:
+ static char ID;
+
+ // Prints EC + S and converts to EC
+ StringError(std::error_code EC, const Twine &S = Twine());
+
+ // Prints S and converts to EC
+ StringError(const Twine &S, std::error_code EC);
+
+ void log(raw_ostream &OS) const override;
+ std::error_code convertToErrorCode() const override;
+
+ const std::string &getMessage() const { return Msg; }
+
+private:
+ std::string Msg;
+ std::error_code EC;
+ const bool PrintMsgOnly = false;
+};
+
+/// Create formatted StringError object.
+template <typename... Ts>
+inline Error createStringError(std::error_code EC, char const *Fmt,
+ const Ts &... Vals) {
+ std::string Buffer;
+ raw_string_ostream Stream(Buffer);
+ Stream << format(Fmt, Vals...);
+ return make_error<StringError>(Stream.str(), EC);
+}
+
+Error createStringError(std::error_code EC, char const *Msg);
+
+inline Error createStringError(std::error_code EC, const Twine &S) {
+ return createStringError(EC, S.str().c_str());
+}
+
+template <typename... Ts>
+inline Error createStringError(std::errc EC, char const *Fmt,
+ const Ts &... Vals) {
+ return createStringError(std::make_error_code(EC), Fmt, Vals...);
+}
+
+/// This class wraps a filename and another Error.
+///
+/// In some cases, an error needs to live along a 'source' name, in order to
+/// show more detailed information to the user.
+class FileError final : public ErrorInfo<FileError> {
+
+ friend Error createFileError(const Twine &, Error);
+ friend Error createFileError(const Twine &, size_t, Error);
+
+public:
+ void log(raw_ostream &OS) const override {
+ assert(Err && !FileName.empty() && "Trying to log after takeError().");
+ OS << "'" << FileName << "': ";
+ if (Line.hasValue())
+ OS << "line " << Line.getValue() << ": ";
+ Err->log(OS);
+ }
+
+ Error takeError() { return Error(std::move(Err)); }
+
+ std::error_code convertToErrorCode() const override;
+
+ // Used by ErrorInfo::classID.
+ static char ID;
+
+private:
+ FileError(const Twine &F, Optional<size_t> LineNum,
+ std::unique_ptr<ErrorInfoBase> E) {
+ assert(E && "Cannot create FileError from Error success value.");
+ assert(!F.isTriviallyEmpty() &&
+ "The file name provided to FileError must not be empty.");
+ FileName = F.str();
+ Err = std::move(E);
+ Line = std::move(LineNum);
+ }
+
+ static Error build(const Twine &F, Optional<size_t> Line, Error E) {
+ return Error(
+ std::unique_ptr<FileError>(new FileError(F, Line, E.takePayload())));
+ }
+
+ std::string FileName;
+ Optional<size_t> Line;
+ std::unique_ptr<ErrorInfoBase> Err;
+};
+
+/// Concatenate a source file path and/or name with an Error. The resulting
+/// Error is unchecked.
+inline Error createFileError(const Twine &F, Error E) {
+ return FileError::build(F, Optional<size_t>(), std::move(E));
+}
+
+/// Concatenate a source file path and/or name with line number and an Error.
+/// The resulting Error is unchecked.
+inline Error createFileError(const Twine &F, size_t Line, Error E) {
+ return FileError::build(F, Optional<size_t>(Line), std::move(E));
+}
+
+/// Concatenate a source file path and/or name with a std::error_code
+/// to form an Error object.
+inline Error createFileError(const Twine &F, std::error_code EC) {
+ return createFileError(F, errorCodeToError(EC));
+}
+
+/// Concatenate a source file path and/or name with line number and
+/// std::error_code to form an Error object.
+inline Error createFileError(const Twine &F, size_t Line, std::error_code EC) {
+ return createFileError(F, Line, errorCodeToError(EC));
+}
+
+Error createFileError(const Twine &F, ErrorSuccess) = delete;
+
+/// Helper for check-and-exit error handling.
+///
+/// For tool use only. NOT FOR USE IN LIBRARY CODE.
+///
+class ExitOnError {
+public:
+ /// Create an error on exit helper.
+ ExitOnError(std::string Banner = "", int DefaultErrorExitCode = 1)
+ : Banner(std::move(Banner)),
+ GetExitCode([=](const Error &) { return DefaultErrorExitCode; }) {}
+
+ /// Set the banner string for any errors caught by operator().
+ void setBanner(std::string Banner) { this->Banner = std::move(Banner); }
+
+ /// Set the exit-code mapper function.
+ void setExitCodeMapper(std::function<int(const Error &)> GetExitCode) {
+ this->GetExitCode = std::move(GetExitCode);
+ }
+
+ /// Check Err. If it's in a failure state log the error(s) and exit.
+ void operator()(Error Err) const { checkError(std::move(Err)); }
+
+ /// Check E. If it's in a success state then return the contained value. If
+ /// it's in a failure state log the error(s) and exit.
+ template <typename T> T operator()(Expected<T> &&E) const {
+ checkError(E.takeError());
+ return std::move(*E);
+ }
+
+ /// Check E. If it's in a success state then return the contained reference. If
+ /// it's in a failure state log the error(s) and exit.
+ template <typename T> T& operator()(Expected<T&> &&E) const {
+ checkError(E.takeError());
+ return *E;
+ }
+
+private:
+ void checkError(Error Err) const {
+ if (Err) {
+ int ExitCode = GetExitCode(Err);
+ logAllUnhandledErrors(std::move(Err), errs(), Banner);
+ exit(ExitCode);
+ }
+ }
+
+ std::string Banner;
+ std::function<int(const Error &)> GetExitCode;
+};
+
+/// Conversion from Error to LLVMErrorRef for C error bindings.
+inline LLVMErrorRef wrap(Error Err) {
+ return reinterpret_cast<LLVMErrorRef>(Err.takePayload().release());
+}
+
+/// Conversion from LLVMErrorRef to Error for C error bindings.
+inline Error unwrap(LLVMErrorRef ErrRef) {
+ return Error(std::unique_ptr<ErrorInfoBase>(
+ reinterpret_cast<ErrorInfoBase *>(ErrRef)));
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_ERROR_H
diff --git a/third_party/llvm-project/include/llvm/Support/ErrorHandling.h b/third_party/llvm-project/include/llvm/Support/ErrorHandling.h
new file mode 100644
index 000000000..f75c2984a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ErrorHandling.h
@@ -0,0 +1,143 @@
+//===- llvm/Support/ErrorHandling.h - Fatal error handling ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines an API used to indicate fatal error conditions. Non-fatal
+// errors (most of them) should be handled through LLVMContext.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ERRORHANDLING_H
+#define LLVM_SUPPORT_ERRORHANDLING_H
+
+#include "llvm/Support/Compiler.h"
+#include <string>
+
+namespace llvm {
+class StringRef;
+ class Twine;
+
+ /// An error handler callback.
+ typedef void (*fatal_error_handler_t)(void *user_data,
+ const std::string& reason,
+ bool gen_crash_diag);
+
+ /// install_fatal_error_handler - Installs a new error handler to be used
+ /// whenever a serious (non-recoverable) error is encountered by LLVM.
+ ///
+ /// If no error handler is installed the default is to print the error message
+ /// to stderr, and call exit(1). If an error handler is installed then it is
+ /// the handler's responsibility to log the message, it will no longer be
+ /// printed to stderr. If the error handler returns, then exit(1) will be
+ /// called.
+ ///
+ /// It is dangerous to naively use an error handler which throws an exception.
+ /// Even though some applications desire to gracefully recover from arbitrary
+ /// faults, blindly throwing exceptions through unfamiliar code isn't a way to
+ /// achieve this.
+ ///
+ /// \param user_data - An argument which will be passed to the install error
+ /// handler.
+ void install_fatal_error_handler(fatal_error_handler_t handler,
+ void *user_data = nullptr);
+
+ /// Restores default error handling behaviour.
+ void remove_fatal_error_handler();
+
+ /// ScopedFatalErrorHandler - This is a simple helper class which just
+ /// calls install_fatal_error_handler in its constructor and
+ /// remove_fatal_error_handler in its destructor.
+ struct ScopedFatalErrorHandler {
+ explicit ScopedFatalErrorHandler(fatal_error_handler_t handler,
+ void *user_data = nullptr) {
+ install_fatal_error_handler(handler, user_data);
+ }
+
+ ~ScopedFatalErrorHandler() { remove_fatal_error_handler(); }
+ };
+
+/// Reports a serious error, calling any installed error handler. These
+/// functions are intended to be used for error conditions which are outside
+/// the control of the compiler (I/O errors, invalid user input, etc.)
+///
+/// If no error handler is installed the default is to print the message to
+/// standard error, followed by a newline.
+/// After the error handler is called this function will call exit(1), it
+/// does not return.
+LLVM_ATTRIBUTE_NORETURN void report_fatal_error(const char *reason,
+ bool gen_crash_diag = true);
+LLVM_ATTRIBUTE_NORETURN void report_fatal_error(const std::string &reason,
+ bool gen_crash_diag = true);
+LLVM_ATTRIBUTE_NORETURN void report_fatal_error(StringRef reason,
+ bool gen_crash_diag = true);
+LLVM_ATTRIBUTE_NORETURN void report_fatal_error(const Twine &reason,
+ bool gen_crash_diag = true);
+
+/// Installs a new bad alloc error handler that should be used whenever a
+/// bad alloc error, e.g. failing malloc/calloc, is encountered by LLVM.
+///
+/// The user can install a bad alloc handler, in order to define the behavior
+/// in case of failing allocations, e.g. throwing an exception. Note that this
+/// handler must not trigger any additional allocations itself.
+///
+/// If no error handler is installed the default is to print the error message
+/// to stderr, and call exit(1). If an error handler is installed then it is
+/// the handler's responsibility to log the message, it will no longer be
+/// printed to stderr. If the error handler returns, then exit(1) will be
+/// called.
+///
+///
+/// \param user_data - An argument which will be passed to the installed error
+/// handler.
+void install_bad_alloc_error_handler(fatal_error_handler_t handler,
+ void *user_data = nullptr);
+
+/// Restores default bad alloc error handling behavior.
+void remove_bad_alloc_error_handler();
+
+void install_out_of_memory_new_handler();
+
+/// Reports a bad alloc error, calling any user defined bad alloc
+/// error handler. In contrast to the generic 'report_fatal_error'
+/// functions, this function is expected to return, e.g. the user
+/// defined error handler throws an exception.
+///
+/// Note: When throwing an exception in the bad alloc handler, make sure that
+/// the following unwind succeeds, e.g. do not trigger additional allocations
+/// in the unwind chain.
+///
+/// If no error handler is installed (default), then a bad_alloc exception
+/// is thrown, if LLVM is compiled with exception support, otherwise an
+/// assertion is called.
+void report_bad_alloc_error(const char *Reason, bool GenCrashDiag = true);
+
+/// This function calls abort(), and prints the optional message to stderr.
+/// Use the llvm_unreachable macro (that adds location info), instead of
+/// calling this function directly.
+LLVM_ATTRIBUTE_NORETURN void
+llvm_unreachable_internal(const char *msg = nullptr, const char *file = nullptr,
+ unsigned line = 0);
+}
+
+/// Marks that the current location is not supposed to be reachable.
+/// In !NDEBUG builds, prints the message and location info to stderr.
+/// In NDEBUG builds, becomes an optimizer hint that the current location
+/// is not supposed to be reachable. On compilers that don't support
+/// such hints, prints a reduced message instead.
+///
+/// Use this instead of assert(0). It conveys intent more clearly and
+/// allows compilers to omit some unnecessary code.
+#ifndef NDEBUG
+#define llvm_unreachable(msg) \
+ ::llvm::llvm_unreachable_internal(msg, __FILE__, __LINE__)
+#elif defined(LLVM_BUILTIN_UNREACHABLE)
+#define llvm_unreachable(msg) LLVM_BUILTIN_UNREACHABLE
+#else
+#define llvm_unreachable(msg) ::llvm::llvm_unreachable_internal()
+#endif
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/ErrorOr.h b/third_party/llvm-project/include/llvm/Support/ErrorOr.h
new file mode 100644
index 000000000..8211f4d8a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ErrorOr.h
@@ -0,0 +1,278 @@
+//===- llvm/Support/ErrorOr.h - Error Smart Pointer -------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+///
+/// Provides ErrorOr<T> smart pointer.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_ERROROR_H
+#define LLVM_SUPPORT_ERROROR_H
+
+#include "llvm/Support/AlignOf.h"
+#include <cassert>
+#include <system_error>
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+/// Represents either an error or a value T.
+///
+/// ErrorOr<T> is a pointer-like class that represents the result of an
+/// operation. The result is either an error, or a value of type T. This is
+/// designed to emulate the usage of returning a pointer where nullptr indicates
+/// failure. However instead of just knowing that the operation failed, we also
+/// have an error_code and optional user data that describes why it failed.
+///
+/// It is used like the following.
+/// \code
+/// ErrorOr<Buffer> getBuffer();
+///
+/// auto buffer = getBuffer();
+/// if (error_code ec = buffer.getError())
+/// return ec;
+/// buffer->write("adena");
+/// \endcode
+///
+///
+/// Implicit conversion to bool returns true if there is a usable value. The
+/// unary * and -> operators provide pointer like access to the value. Accessing
+/// the value when there is an error has undefined behavior.
+///
+/// When T is a reference type the behavior is slightly different. The reference
+/// is held in a std::reference_wrapper<std::remove_reference<T>::type>, and
+/// there is special handling to make operator -> work as if T was not a
+/// reference.
+///
+/// T cannot be a rvalue reference.
+template<class T>
+class ErrorOr {
+ template <class OtherT> friend class ErrorOr;
+
+ static const bool isRef = std::is_reference<T>::value;
+
+ using wrap = std::reference_wrapper<typename std::remove_reference<T>::type>;
+
+public:
+ using storage_type = typename std::conditional<isRef, wrap, T>::type;
+
+private:
+ using reference = typename std::remove_reference<T>::type &;
+ using const_reference = const typename std::remove_reference<T>::type &;
+ using pointer = typename std::remove_reference<T>::type *;
+ using const_pointer = const typename std::remove_reference<T>::type *;
+
+public:
+ template <class E>
+ ErrorOr(E ErrorCode,
+ typename std::enable_if<std::is_error_code_enum<E>::value ||
+ std::is_error_condition_enum<E>::value,
+ void *>::type = nullptr)
+ : HasError(true) {
+ new (getErrorStorage()) std::error_code(make_error_code(ErrorCode));
+ }
+
+ ErrorOr(std::error_code EC) : HasError(true) {
+ new (getErrorStorage()) std::error_code(EC);
+ }
+
+ template <class OtherT>
+ ErrorOr(OtherT &&Val,
+ typename std::enable_if<std::is_convertible<OtherT, T>::value>::type
+ * = nullptr)
+ : HasError(false) {
+ new (getStorage()) storage_type(std::forward<OtherT>(Val));
+ }
+
+ ErrorOr(const ErrorOr &Other) {
+ copyConstruct(Other);
+ }
+
+ template <class OtherT>
+ ErrorOr(
+ const ErrorOr<OtherT> &Other,
+ typename std::enable_if<std::is_convertible<OtherT, T>::value>::type * =
+ nullptr) {
+ copyConstruct(Other);
+ }
+
+ template <class OtherT>
+ explicit ErrorOr(
+ const ErrorOr<OtherT> &Other,
+ typename std::enable_if<
+ !std::is_convertible<OtherT, const T &>::value>::type * = nullptr) {
+ copyConstruct(Other);
+ }
+
+ ErrorOr(ErrorOr &&Other) {
+ moveConstruct(std::move(Other));
+ }
+
+ template <class OtherT>
+ ErrorOr(
+ ErrorOr<OtherT> &&Other,
+ typename std::enable_if<std::is_convertible<OtherT, T>::value>::type * =
+ nullptr) {
+ moveConstruct(std::move(Other));
+ }
+
+ // This might eventually need SFINAE but it's more complex than is_convertible
+ // & I'm too lazy to write it right now.
+ template <class OtherT>
+ explicit ErrorOr(
+ ErrorOr<OtherT> &&Other,
+ typename std::enable_if<!std::is_convertible<OtherT, T>::value>::type * =
+ nullptr) {
+ moveConstruct(std::move(Other));
+ }
+
+ ErrorOr &operator=(const ErrorOr &Other) {
+ copyAssign(Other);
+ return *this;
+ }
+
+ ErrorOr &operator=(ErrorOr &&Other) {
+ moveAssign(std::move(Other));
+ return *this;
+ }
+
+ ~ErrorOr() {
+ if (!HasError)
+ getStorage()->~storage_type();
+ }
+
+ /// Return false if there is an error.
+ explicit operator bool() const {
+ return !HasError;
+ }
+
+ reference get() { return *getStorage(); }
+ const_reference get() const { return const_cast<ErrorOr<T> *>(this)->get(); }
+
+ std::error_code getError() const {
+ return HasError ? *getErrorStorage() : std::error_code();
+ }
+
+ pointer operator ->() {
+ return toPointer(getStorage());
+ }
+
+ const_pointer operator->() const { return toPointer(getStorage()); }
+
+ reference operator *() {
+ return *getStorage();
+ }
+
+ const_reference operator*() const { return *getStorage(); }
+
+private:
+ template <class OtherT>
+ void copyConstruct(const ErrorOr<OtherT> &Other) {
+ if (!Other.HasError) {
+ // Get the other value.
+ HasError = false;
+ new (getStorage()) storage_type(*Other.getStorage());
+ } else {
+ // Get other's error.
+ HasError = true;
+ new (getErrorStorage()) std::error_code(Other.getError());
+ }
+ }
+
+ template <class T1>
+ static bool compareThisIfSameType(const T1 &a, const T1 &b) {
+ return &a == &b;
+ }
+
+ template <class T1, class T2>
+ static bool compareThisIfSameType(const T1 &a, const T2 &b) {
+ return false;
+ }
+
+ template <class OtherT>
+ void copyAssign(const ErrorOr<OtherT> &Other) {
+ if (compareThisIfSameType(*this, Other))
+ return;
+
+ this->~ErrorOr();
+ new (this) ErrorOr(Other);
+ }
+
+ template <class OtherT>
+ void moveConstruct(ErrorOr<OtherT> &&Other) {
+ if (!Other.HasError) {
+ // Get the other value.
+ HasError = false;
+ new (getStorage()) storage_type(std::move(*Other.getStorage()));
+ } else {
+ // Get other's error.
+ HasError = true;
+ new (getErrorStorage()) std::error_code(Other.getError());
+ }
+ }
+
+ template <class OtherT>
+ void moveAssign(ErrorOr<OtherT> &&Other) {
+ if (compareThisIfSameType(*this, Other))
+ return;
+
+ this->~ErrorOr();
+ new (this) ErrorOr(std::move(Other));
+ }
+
+ pointer toPointer(pointer Val) {
+ return Val;
+ }
+
+ const_pointer toPointer(const_pointer Val) const { return Val; }
+
+ pointer toPointer(wrap *Val) {
+ return &Val->get();
+ }
+
+ const_pointer toPointer(const wrap *Val) const { return &Val->get(); }
+
+ storage_type *getStorage() {
+ assert(!HasError && "Cannot get value when an error exists!");
+ return reinterpret_cast<storage_type*>(TStorage.buffer);
+ }
+
+ const storage_type *getStorage() const {
+ assert(!HasError && "Cannot get value when an error exists!");
+ return reinterpret_cast<const storage_type*>(TStorage.buffer);
+ }
+
+ std::error_code *getErrorStorage() {
+ assert(HasError && "Cannot get error when a value exists!");
+ return reinterpret_cast<std::error_code *>(ErrorStorage.buffer);
+ }
+
+ const std::error_code *getErrorStorage() const {
+ return const_cast<ErrorOr<T> *>(this)->getErrorStorage();
+ }
+
+ union {
+ AlignedCharArrayUnion<storage_type> TStorage;
+ AlignedCharArrayUnion<std::error_code> ErrorStorage;
+ };
+ bool HasError : 1;
+};
+
+template <class T, class E>
+typename std::enable_if<std::is_error_code_enum<E>::value ||
+ std::is_error_condition_enum<E>::value,
+ bool>::type
+operator==(const ErrorOr<T> &Err, E Code) {
+ return Err.getError() == Code;
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_ERROROR_H
diff --git a/third_party/llvm-project/include/llvm/Support/FileOutputBuffer.h b/third_party/llvm-project/include/llvm/Support/FileOutputBuffer.h
new file mode 100644
index 000000000..bdc1425d4
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FileOutputBuffer.h
@@ -0,0 +1,88 @@
+//=== FileOutputBuffer.h - File Output Buffer -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Utility for creating a in-memory buffer that will be written to a file.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FILEOUTPUTBUFFER_H
+#define LLVM_SUPPORT_FILEOUTPUTBUFFER_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileSystem.h"
+
+namespace llvm {
+/// FileOutputBuffer - This interface provides simple way to create an in-memory
+/// buffer which will be written to a file. During the lifetime of these
+/// objects, the content or existence of the specified file is undefined. That
+/// is, creating an OutputBuffer for a file may immediately remove the file.
+/// If the FileOutputBuffer is committed, the target file's content will become
+/// the buffer content at the time of the commit. If the FileOutputBuffer is
+/// not committed, the file will be deleted in the FileOutputBuffer destructor.
+class FileOutputBuffer {
+public:
+ enum {
+ /// set the 'x' bit on the resulting file
+ F_executable = 1,
+
+ /// Don't use mmap and instead write an in-memory buffer to a file when this
+ /// buffer is closed.
+ F_no_mmap = 2,
+ };
+
+ /// Factory method to create an OutputBuffer object which manages a read/write
+ /// buffer of the specified size. When committed, the buffer will be written
+ /// to the file at the specified path.
+ ///
+ /// When F_modify is specified and \p FilePath refers to an existing on-disk
+ /// file \p Size may be set to -1, in which case the entire file is used.
+ /// Otherwise, the file shrinks or grows as necessary based on the value of
+ /// \p Size. It is an error to specify F_modify and Size=-1 if \p FilePath
+ /// does not exist.
+ static Expected<std::unique_ptr<FileOutputBuffer>>
+ create(StringRef FilePath, size_t Size, unsigned Flags = 0);
+
+ /// Returns a pointer to the start of the buffer.
+ virtual uint8_t *getBufferStart() const = 0;
+
+ /// Returns a pointer to the end of the buffer.
+ virtual uint8_t *getBufferEnd() const = 0;
+
+ /// Returns size of the buffer.
+ virtual size_t getBufferSize() const = 0;
+
+ /// Returns path where file will show up if buffer is committed.
+ StringRef getPath() const { return FinalPath; }
+
+ /// Flushes the content of the buffer to its file and deallocates the
+ /// buffer. If commit() is not called before this object's destructor
+ /// is called, the file is deleted in the destructor. The optional parameter
+ /// is used if it turns out you want the file size to be smaller than
+ /// initially requested.
+ virtual Error commit() = 0;
+
+ /// If this object was previously committed, the destructor just deletes
+ /// this object. If this object was not committed, the destructor
+ /// deallocates the buffer and the target file is never written.
+ virtual ~FileOutputBuffer() {}
+
+ /// This removes the temporary file (unless it already was committed)
+ /// but keeps the memory mapping alive.
+ virtual void discard() {}
+
+protected:
+ FileOutputBuffer(StringRef Path) : FinalPath(Path) {}
+
+ std::string FinalPath;
+};
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FileSystem.h b/third_party/llvm-project/include/llvm/Support/FileSystem.h
new file mode 100644
index 000000000..a29a9d787
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FileSystem.h
@@ -0,0 +1,1444 @@
+//===- llvm/Support/FileSystem.h - File System OS Concept -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the llvm::sys::fs namespace. It is designed after
+// TR2/boost filesystem (v3), but modified to remove exception handling and the
+// path class.
+//
+// All functions return an error_code and their actual work via the last out
+// argument. The out argument is defined if and only if errc::success is
+// returned. A function may return any error code in the generic or system
+// category. However, they shall be equivalent to any error conditions listed
+// in each functions respective documentation if the condition applies. [ note:
+// this does not guarantee that error_code will be in the set of explicitly
+// listed codes, but it does guarantee that if any of the explicitly listed
+// errors occur, the correct error_code will be used ]. All functions may
+// return errc::not_enough_memory if there is not enough memory to complete the
+// operation.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FILESYSTEM_H
+#define LLVM_SUPPORT_FILESYSTEM_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Config/llvm-config.h"
+#include "llvm/Support/Chrono.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/MD5.h"
+#include <cassert>
+#include <cstdint>
+#include <ctime>
+#include <memory>
+#include <stack>
+#include <string>
+#include <system_error>
+#include <tuple>
+#include <vector>
+
+#ifdef HAVE_SYS_STAT_H
+#include <sys/stat.h>
+#endif
+
+namespace llvm {
+namespace sys {
+namespace fs {
+
+#if defined(_WIN32)
+// A Win32 HANDLE is a typedef of void*
+using file_t = void *;
+#else
+using file_t = int;
+#endif
+
+extern const file_t kInvalidFile;
+
+/// An enumeration for the file system's view of the type.
+enum class file_type {
+ status_error,
+ file_not_found,
+ regular_file,
+ directory_file,
+ symlink_file,
+ block_file,
+ character_file,
+ fifo_file,
+ socket_file,
+ type_unknown
+};
+
+/// space_info - Self explanatory.
+struct space_info {
+ uint64_t capacity;
+ uint64_t free;
+ uint64_t available;
+};
+
+enum perms {
+ no_perms = 0,
+ owner_read = 0400,
+ owner_write = 0200,
+ owner_exe = 0100,
+ owner_all = owner_read | owner_write | owner_exe,
+ group_read = 040,
+ group_write = 020,
+ group_exe = 010,
+ group_all = group_read | group_write | group_exe,
+ others_read = 04,
+ others_write = 02,
+ others_exe = 01,
+ others_all = others_read | others_write | others_exe,
+ all_read = owner_read | group_read | others_read,
+ all_write = owner_write | group_write | others_write,
+ all_exe = owner_exe | group_exe | others_exe,
+ all_all = owner_all | group_all | others_all,
+ set_uid_on_exe = 04000,
+ set_gid_on_exe = 02000,
+ sticky_bit = 01000,
+ all_perms = all_all | set_uid_on_exe | set_gid_on_exe | sticky_bit,
+ perms_not_known = 0xFFFF
+};
+
+// Helper functions so that you can use & and | to manipulate perms bits:
+inline perms operator|(perms l, perms r) {
+ return static_cast<perms>(static_cast<unsigned short>(l) |
+ static_cast<unsigned short>(r));
+}
+inline perms operator&(perms l, perms r) {
+ return static_cast<perms>(static_cast<unsigned short>(l) &
+ static_cast<unsigned short>(r));
+}
+inline perms &operator|=(perms &l, perms r) {
+ l = l | r;
+ return l;
+}
+inline perms &operator&=(perms &l, perms r) {
+ l = l & r;
+ return l;
+}
+inline perms operator~(perms x) {
+ // Avoid UB by explicitly truncating the (unsigned) ~ result.
+ return static_cast<perms>(
+ static_cast<unsigned short>(~static_cast<unsigned short>(x)));
+}
+
+class UniqueID {
+ uint64_t Device;
+ uint64_t File;
+
+public:
+ UniqueID() = default;
+ UniqueID(uint64_t Device, uint64_t File) : Device(Device), File(File) {}
+
+ bool operator==(const UniqueID &Other) const {
+ return Device == Other.Device && File == Other.File;
+ }
+ bool operator!=(const UniqueID &Other) const { return !(*this == Other); }
+ bool operator<(const UniqueID &Other) const {
+ return std::tie(Device, File) < std::tie(Other.Device, Other.File);
+ }
+
+ uint64_t getDevice() const { return Device; }
+ uint64_t getFile() const { return File; }
+};
+
+/// Represents the result of a call to directory_iterator::status(). This is a
+/// subset of the information returned by a regular sys::fs::status() call, and
+/// represents the information provided by Windows FileFirstFile/FindNextFile.
+class basic_file_status {
+protected:
+ #if defined(LLVM_ON_UNIX)
+ time_t fs_st_atime = 0;
+ time_t fs_st_mtime = 0;
+ uint32_t fs_st_atime_nsec = 0;
+ uint32_t fs_st_mtime_nsec = 0;
+ uid_t fs_st_uid = 0;
+ gid_t fs_st_gid = 0;
+ off_t fs_st_size = 0;
+ #elif defined (_WIN32)
+ uint32_t LastAccessedTimeHigh = 0;
+ uint32_t LastAccessedTimeLow = 0;
+ uint32_t LastWriteTimeHigh = 0;
+ uint32_t LastWriteTimeLow = 0;
+ uint32_t FileSizeHigh = 0;
+ uint32_t FileSizeLow = 0;
+ #endif
+ file_type Type = file_type::status_error;
+ perms Perms = perms_not_known;
+
+public:
+ basic_file_status() = default;
+
+ explicit basic_file_status(file_type Type) : Type(Type) {}
+
+ #if defined(LLVM_ON_UNIX)
+ basic_file_status(file_type Type, perms Perms, time_t ATime,
+ uint32_t ATimeNSec, time_t MTime, uint32_t MTimeNSec,
+ uid_t UID, gid_t GID, off_t Size)
+ : fs_st_atime(ATime), fs_st_mtime(MTime),
+ fs_st_atime_nsec(ATimeNSec), fs_st_mtime_nsec(MTimeNSec),
+ fs_st_uid(UID), fs_st_gid(GID),
+ fs_st_size(Size), Type(Type), Perms(Perms) {}
+#elif defined(_WIN32)
+ basic_file_status(file_type Type, perms Perms, uint32_t LastAccessTimeHigh,
+ uint32_t LastAccessTimeLow, uint32_t LastWriteTimeHigh,
+ uint32_t LastWriteTimeLow, uint32_t FileSizeHigh,
+ uint32_t FileSizeLow)
+ : LastAccessedTimeHigh(LastAccessTimeHigh),
+ LastAccessedTimeLow(LastAccessTimeLow),
+ LastWriteTimeHigh(LastWriteTimeHigh),
+ LastWriteTimeLow(LastWriteTimeLow), FileSizeHigh(FileSizeHigh),
+ FileSizeLow(FileSizeLow), Type(Type), Perms(Perms) {}
+ #endif
+
+ // getters
+ file_type type() const { return Type; }
+ perms permissions() const { return Perms; }
+
+ /// The file access time as reported from the underlying file system.
+ ///
+ /// Also see comments on \c getLastModificationTime() related to the precision
+ /// of the returned value.
+ TimePoint<> getLastAccessedTime() const;
+
+ /// The file modification time as reported from the underlying file system.
+ ///
+ /// The returned value allows for nanosecond precision but the actual
+ /// resolution is an implementation detail of the underlying file system.
+ /// There is no guarantee for what kind of resolution you can expect, the
+ /// resolution can differ across platforms and even across mountpoints on the
+ /// same machine.
+ TimePoint<> getLastModificationTime() const;
+
+ #if defined(LLVM_ON_UNIX)
+ uint32_t getUser() const { return fs_st_uid; }
+ uint32_t getGroup() const { return fs_st_gid; }
+ uint64_t getSize() const { return fs_st_size; }
+ #elif defined (_WIN32)
+ uint32_t getUser() const {
+ return 9999; // Not applicable to Windows, so...
+ }
+
+ uint32_t getGroup() const {
+ return 9999; // Not applicable to Windows, so...
+ }
+
+ uint64_t getSize() const {
+ return (uint64_t(FileSizeHigh) << 32) + FileSizeLow;
+ }
+ #endif
+
+ // setters
+ void type(file_type v) { Type = v; }
+ void permissions(perms p) { Perms = p; }
+};
+
+/// Represents the result of a call to sys::fs::status().
+class file_status : public basic_file_status {
+ friend bool equivalent(file_status A, file_status B);
+
+ #if defined(LLVM_ON_UNIX)
+ dev_t fs_st_dev = 0;
+ nlink_t fs_st_nlinks = 0;
+ ino_t fs_st_ino = 0;
+ #elif defined (_WIN32)
+ uint32_t NumLinks = 0;
+ uint32_t VolumeSerialNumber = 0;
+ uint32_t FileIndexHigh = 0;
+ uint32_t FileIndexLow = 0;
+ #endif
+
+public:
+ file_status() = default;
+
+ explicit file_status(file_type Type) : basic_file_status(Type) {}
+
+ #if defined(LLVM_ON_UNIX)
+ file_status(file_type Type, perms Perms, dev_t Dev, nlink_t Links, ino_t Ino,
+ time_t ATime, uint32_t ATimeNSec,
+ time_t MTime, uint32_t MTimeNSec,
+ uid_t UID, gid_t GID, off_t Size)
+ : basic_file_status(Type, Perms, ATime, ATimeNSec, MTime, MTimeNSec,
+ UID, GID, Size),
+ fs_st_dev(Dev), fs_st_nlinks(Links), fs_st_ino(Ino) {}
+ #elif defined(_WIN32)
+ file_status(file_type Type, perms Perms, uint32_t LinkCount,
+ uint32_t LastAccessTimeHigh, uint32_t LastAccessTimeLow,
+ uint32_t LastWriteTimeHigh, uint32_t LastWriteTimeLow,
+ uint32_t VolumeSerialNumber, uint32_t FileSizeHigh,
+ uint32_t FileSizeLow, uint32_t FileIndexHigh,
+ uint32_t FileIndexLow)
+ : basic_file_status(Type, Perms, LastAccessTimeHigh, LastAccessTimeLow,
+ LastWriteTimeHigh, LastWriteTimeLow, FileSizeHigh,
+ FileSizeLow),
+ NumLinks(LinkCount), VolumeSerialNumber(VolumeSerialNumber),
+ FileIndexHigh(FileIndexHigh), FileIndexLow(FileIndexLow) {}
+ #endif
+
+ UniqueID getUniqueID() const;
+ uint32_t getLinkCount() const;
+};
+
+/// @}
+/// @name Physical Operators
+/// @{
+
+/// Make \a path an absolute path.
+///
+/// Makes \a path absolute using the \a current_directory if it is not already.
+/// An empty \a path will result in the \a current_directory.
+///
+/// /absolute/path => /absolute/path
+/// relative/../path => <current-directory>/relative/../path
+///
+/// @param path A path that is modified to be an absolute path.
+void make_absolute(const Twine ¤t_directory, SmallVectorImpl<char> &path);
+
+/// Make \a path an absolute path.
+///
+/// Makes \a path absolute using the current directory if it is not already. An
+/// empty \a path will result in the current directory.
+///
+/// /absolute/path => /absolute/path
+/// relative/../path => <current-directory>/relative/../path
+///
+/// @param path A path that is modified to be an absolute path.
+/// @returns errc::success if \a path has been made absolute, otherwise a
+/// platform-specific error_code.
+std::error_code make_absolute(SmallVectorImpl<char> &path);
+
+/// Create all the non-existent directories in path.
+///
+/// @param path Directories to create.
+/// @returns errc::success if is_directory(path), otherwise a platform
+/// specific error_code. If IgnoreExisting is false, also returns
+/// error if the directory already existed.
+std::error_code create_directories(const Twine &path,
+ bool IgnoreExisting = true,
+ perms Perms = owner_all | group_all);
+
+/// Create the directory in path.
+///
+/// @param path Directory to create.
+/// @returns errc::success if is_directory(path), otherwise a platform
+/// specific error_code. If IgnoreExisting is false, also returns
+/// error if the directory already existed.
+std::error_code create_directory(const Twine &path, bool IgnoreExisting = true,
+ perms Perms = owner_all | group_all);
+
+/// Create a link from \a from to \a to.
+///
+/// The link may be a soft or a hard link, depending on the platform. The caller
+/// may not assume which one. Currently on windows it creates a hard link since
+/// soft links require extra privileges. On unix, it creates a soft link since
+/// hard links don't work on SMB file systems.
+///
+/// @param to The path to hard link to.
+/// @param from The path to hard link from. This is created.
+/// @returns errc::success if the link was created, otherwise a platform
+/// specific error_code.
+std::error_code create_link(const Twine &to, const Twine &from);
+
+/// Create a hard link from \a from to \a to, or return an error.
+///
+/// @param to The path to hard link to.
+/// @param from The path to hard link from. This is created.
+/// @returns errc::success if the link was created, otherwise a platform
+/// specific error_code.
+std::error_code create_hard_link(const Twine &to, const Twine &from);
+
+/// Collapse all . and .. patterns, resolve all symlinks, and optionally
+/// expand ~ expressions to the user's home directory.
+///
+/// @param path The path to resolve.
+/// @param output The location to store the resolved path.
+/// @param expand_tilde If true, resolves ~ expressions to the user's home
+/// directory.
+std::error_code real_path(const Twine &path, SmallVectorImpl<char> &output,
+ bool expand_tilde = false);
+
+/// Expands ~ expressions to the user's home directory. On Unix ~user
+/// directories are resolved as well.
+///
+/// @param path The path to resolve.
+void expand_tilde(const Twine &path, SmallVectorImpl<char> &output);
+
+/// Get the current path.
+///
+/// @param result Holds the current path on return.
+/// @returns errc::success if the current path has been stored in result,
+/// otherwise a platform-specific error_code.
+std::error_code current_path(SmallVectorImpl<char> &result);
+
+/// Set the current path.
+///
+/// @param path The path to set.
+/// @returns errc::success if the current path was successfully set,
+/// otherwise a platform-specific error_code.
+std::error_code set_current_path(const Twine &path);
+
+/// Remove path. Equivalent to POSIX remove().
+///
+/// @param path Input path.
+/// @returns errc::success if path has been removed or didn't exist, otherwise a
+/// platform-specific error code. If IgnoreNonExisting is false, also
+/// returns error if the file didn't exist.
+std::error_code remove(const Twine &path, bool IgnoreNonExisting = true);
+
+/// Recursively delete a directory.
+///
+/// @param path Input path.
+/// @returns errc::success if path has been removed or didn't exist, otherwise a
+/// platform-specific error code.
+std::error_code remove_directories(const Twine &path, bool IgnoreErrors = true);
+
+/// Rename \a from to \a to.
+///
+/// Files are renamed as if by POSIX rename(), except that on Windows there may
+/// be a short interval of time during which the destination file does not
+/// exist.
+///
+/// @param from The path to rename from.
+/// @param to The path to rename to. This is created.
+std::error_code rename(const Twine &from, const Twine &to);
+
+/// Copy the contents of \a From to \a To.
+///
+/// @param From The path to copy from.
+/// @param To The path to copy to. This is created.
+std::error_code copy_file(const Twine &From, const Twine &To);
+
+/// Copy the contents of \a From to \a To.
+///
+/// @param From The path to copy from.
+/// @param ToFD The open file descriptor of the destination file.
+std::error_code copy_file(const Twine &From, int ToFD);
+
+/// Resize path to size. File is resized as if by POSIX truncate().
+///
+/// @param FD Input file descriptor.
+/// @param Size Size to resize to.
+/// @returns errc::success if \a path has been resized to \a size, otherwise a
+/// platform-specific error_code.
+std::error_code resize_file(int FD, uint64_t Size);
+
+/// Compute an MD5 hash of a file's contents.
+///
+/// @param FD Input file descriptor.
+/// @returns An MD5Result with the hash computed, if successful, otherwise a
+/// std::error_code.
+ErrorOr<MD5::MD5Result> md5_contents(int FD);
+
+/// Version of compute_md5 that doesn't require an open file descriptor.
+ErrorOr<MD5::MD5Result> md5_contents(const Twine &Path);
+
+/// @}
+/// @name Physical Observers
+/// @{
+
+/// Does file exist?
+///
+/// @param status A basic_file_status previously returned from stat.
+/// @returns True if the file represented by status exists, false if it does
+/// not.
+bool exists(const basic_file_status &status);
+
+enum class AccessMode { Exist, Write, Execute };
+
+/// Can the file be accessed?
+///
+/// @param Path Input path.
+/// @returns errc::success if the path can be accessed, otherwise a
+/// platform-specific error_code.
+std::error_code access(const Twine &Path, AccessMode Mode);
+
+/// Does file exist?
+///
+/// @param Path Input path.
+/// @returns True if it exists, false otherwise.
+inline bool exists(const Twine &Path) {
+ return !access(Path, AccessMode::Exist);
+}
+
+/// Can we execute this file?
+///
+/// @param Path Input path.
+/// @returns True if we can execute it, false otherwise.
+bool can_execute(const Twine &Path);
+
+/// Can we write this file?
+///
+/// @param Path Input path.
+/// @returns True if we can write to it, false otherwise.
+inline bool can_write(const Twine &Path) {
+ return !access(Path, AccessMode::Write);
+}
+
+/// Do file_status's represent the same thing?
+///
+/// @param A Input file_status.
+/// @param B Input file_status.
+///
+/// assert(status_known(A) || status_known(B));
+///
+/// @returns True if A and B both represent the same file system entity, false
+/// otherwise.
+bool equivalent(file_status A, file_status B);
+
+/// Do paths represent the same thing?
+///
+/// assert(status_known(A) || status_known(B));
+///
+/// @param A Input path A.
+/// @param B Input path B.
+/// @param result Set to true if stat(A) and stat(B) have the same device and
+/// inode (or equivalent).
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code equivalent(const Twine &A, const Twine &B, bool &result);
+
+/// Simpler version of equivalent for clients that don't need to
+/// differentiate between an error and false.
+inline bool equivalent(const Twine &A, const Twine &B) {
+ bool result;
+ return !equivalent(A, B, result) && result;
+}
+
+/// Is the file mounted on a local filesystem?
+///
+/// @param path Input path.
+/// @param result Set to true if \a path is on fixed media such as a hard disk,
+/// false if it is not.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform specific error_code.
+std::error_code is_local(const Twine &path, bool &result);
+
+/// Version of is_local accepting an open file descriptor.
+std::error_code is_local(int FD, bool &result);
+
+/// Simpler version of is_local for clients that don't need to
+/// differentiate between an error and false.
+inline bool is_local(const Twine &Path) {
+ bool Result;
+ return !is_local(Path, Result) && Result;
+}
+
+/// Simpler version of is_local accepting an open file descriptor for
+/// clients that don't need to differentiate between an error and false.
+inline bool is_local(int FD) {
+ bool Result;
+ return !is_local(FD, Result) && Result;
+}
+
+/// Does status represent a directory?
+///
+/// @param Path The path to get the type of.
+/// @param Follow For symbolic links, indicates whether to return the file type
+/// of the link itself, or of the target.
+/// @returns A value from the file_type enumeration indicating the type of file.
+file_type get_file_type(const Twine &Path, bool Follow = true);
+
+/// Does status represent a directory?
+///
+/// @param status A basic_file_status previously returned from status.
+/// @returns status.type() == file_type::directory_file.
+bool is_directory(const basic_file_status &status);
+
+/// Is path a directory?
+///
+/// @param path Input path.
+/// @param result Set to true if \a path is a directory (after following
+/// symlinks, false if it is not. Undefined otherwise.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code is_directory(const Twine &path, bool &result);
+
+/// Simpler version of is_directory for clients that don't need to
+/// differentiate between an error and false.
+inline bool is_directory(const Twine &Path) {
+ bool Result;
+ return !is_directory(Path, Result) && Result;
+}
+
+/// Does status represent a regular file?
+///
+/// @param status A basic_file_status previously returned from status.
+/// @returns status_known(status) && status.type() == file_type::regular_file.
+bool is_regular_file(const basic_file_status &status);
+
+/// Is path a regular file?
+///
+/// @param path Input path.
+/// @param result Set to true if \a path is a regular file (after following
+/// symlinks), false if it is not. Undefined otherwise.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code is_regular_file(const Twine &path, bool &result);
+
+/// Simpler version of is_regular_file for clients that don't need to
+/// differentiate between an error and false.
+inline bool is_regular_file(const Twine &Path) {
+ bool Result;
+ if (is_regular_file(Path, Result))
+ return false;
+ return Result;
+}
+
+/// Does status represent a symlink file?
+///
+/// @param status A basic_file_status previously returned from status.
+/// @returns status_known(status) && status.type() == file_type::symlink_file.
+bool is_symlink_file(const basic_file_status &status);
+
+/// Is path a symlink file?
+///
+/// @param path Input path.
+/// @param result Set to true if \a path is a symlink file, false if it is not.
+/// Undefined otherwise.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code is_symlink_file(const Twine &path, bool &result);
+
+/// Simpler version of is_symlink_file for clients that don't need to
+/// differentiate between an error and false.
+inline bool is_symlink_file(const Twine &Path) {
+ bool Result;
+ if (is_symlink_file(Path, Result))
+ return false;
+ return Result;
+}
+
+/// Does this status represent something that exists but is not a
+/// directory or regular file?
+///
+/// @param status A basic_file_status previously returned from status.
+/// @returns exists(s) && !is_regular_file(s) && !is_directory(s)
+bool is_other(const basic_file_status &status);
+
+/// Is path something that exists but is not a directory,
+/// regular file, or symlink?
+///
+/// @param path Input path.
+/// @param result Set to true if \a path exists, but is not a directory, regular
+/// file, or a symlink, false if it does not. Undefined otherwise.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code is_other(const Twine &path, bool &result);
+
+/// Get file status as if by POSIX stat().
+///
+/// @param path Input path.
+/// @param result Set to the file status.
+/// @param follow When true, follows symlinks. Otherwise, the symlink itself is
+/// statted.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code status(const Twine &path, file_status &result,
+ bool follow = true);
+
+/// A version for when a file descriptor is already available.
+std::error_code status(int FD, file_status &Result);
+
+#ifdef _WIN32
+/// A version for when a file descriptor is already available.
+std::error_code status(file_t FD, file_status &Result);
+#endif
+
+/// Get file creation mode mask of the process.
+///
+/// @returns Mask reported by umask(2)
+/// @note There is no umask on Windows. This function returns 0 always
+/// on Windows. This function does not return an error_code because
+/// umask(2) never fails. It is not thread safe.
+unsigned getUmask();
+
+/// Set file permissions.
+///
+/// @param Path File to set permissions on.
+/// @param Permissions New file permissions.
+/// @returns errc::success if the permissions were successfully set, otherwise
+/// a platform-specific error_code.
+/// @note On Windows, all permissions except *_write are ignored. Using any of
+/// owner_write, group_write, or all_write will make the file writable.
+/// Otherwise, the file will be marked as read-only.
+std::error_code setPermissions(const Twine &Path, perms Permissions);
+
+/// Vesion of setPermissions accepting a file descriptor.
+/// TODO Delete the path based overload once we implement the FD based overload
+/// on Windows.
+std::error_code setPermissions(int FD, perms Permissions);
+
+/// Get file permissions.
+///
+/// @param Path File to get permissions from.
+/// @returns the permissions if they were successfully retrieved, otherwise a
+/// platform-specific error_code.
+/// @note On Windows, if the file does not have the FILE_ATTRIBUTE_READONLY
+/// attribute, all_all will be returned. Otherwise, all_read | all_exe
+/// will be returned.
+ErrorOr<perms> getPermissions(const Twine &Path);
+
+/// Get file size.
+///
+/// @param Path Input path.
+/// @param Result Set to the size of the file in \a Path.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+inline std::error_code file_size(const Twine &Path, uint64_t &Result) {
+ file_status Status;
+ std::error_code EC = status(Path, Status);
+ if (EC)
+ return EC;
+ Result = Status.getSize();
+ return std::error_code();
+}
+
+/// Set the file modification and access time.
+///
+/// @returns errc::success if the file times were successfully set, otherwise a
+/// platform-specific error_code or errc::function_not_supported on
+/// platforms where the functionality isn't available.
+std::error_code setLastAccessAndModificationTime(int FD, TimePoint<> AccessTime,
+ TimePoint<> ModificationTime);
+
+/// Simpler version that sets both file modification and access time to the same
+/// time.
+inline std::error_code setLastAccessAndModificationTime(int FD,
+ TimePoint<> Time) {
+ return setLastAccessAndModificationTime(FD, Time, Time);
+}
+
+/// Is status available?
+///
+/// @param s Input file status.
+/// @returns True if status() != status_error.
+bool status_known(const basic_file_status &s);
+
+/// Is status available?
+///
+/// @param path Input path.
+/// @param result Set to true if status() != status_error.
+/// @returns errc::success if result has been successfully set, otherwise a
+/// platform-specific error_code.
+std::error_code status_known(const Twine &path, bool &result);
+
+enum CreationDisposition : unsigned {
+ /// CD_CreateAlways - When opening a file:
+ /// * If it already exists, truncate it.
+ /// * If it does not already exist, create a new file.
+ CD_CreateAlways = 0,
+
+ /// CD_CreateNew - When opening a file:
+ /// * If it already exists, fail.
+ /// * If it does not already exist, create a new file.
+ CD_CreateNew = 1,
+
+ /// CD_OpenExisting - When opening a file:
+ /// * If it already exists, open the file with the offset set to 0.
+ /// * If it does not already exist, fail.
+ CD_OpenExisting = 2,
+
+ /// CD_OpenAlways - When opening a file:
+ /// * If it already exists, open the file with the offset set to 0.
+ /// * If it does not already exist, create a new file.
+ CD_OpenAlways = 3,
+};
+
+enum FileAccess : unsigned {
+ FA_Read = 1,
+ FA_Write = 2,
+};
+
+enum OpenFlags : unsigned {
+ OF_None = 0,
+ F_None = 0, // For compatibility
+
+ /// The file should be opened in text mode on platforms that make this
+ /// distinction.
+ OF_Text = 1,
+ F_Text = 1, // For compatibility
+
+ /// The file should be opened in append mode.
+ OF_Append = 2,
+ F_Append = 2, // For compatibility
+
+ /// Delete the file on close. Only makes a difference on windows.
+ OF_Delete = 4,
+
+ /// When a child process is launched, this file should remain open in the
+ /// child process.
+ OF_ChildInherit = 8,
+
+ /// Force files Atime to be updated on access. Only makes a difference on windows.
+ OF_UpdateAtime = 16,
+};
+
+/// Create a potentially unique file name but does not create it.
+///
+/// Generates a unique path suitable for a temporary file but does not
+/// open or create the file. The name is based on \a Model with '%'
+/// replaced by a random char in [0-9a-f]. If \a MakeAbsolute is true
+/// then the system's temp directory is prepended first. If \a MakeAbsolute
+/// is false the current directory will be used instead.
+///
+/// This function does not check if the file exists. If you want to be sure
+/// that the file does not yet exist, you should use use enough '%' characters
+/// in your model to ensure this. Each '%' gives 4-bits of entropy so you can
+/// use 32 of them to get 128 bits of entropy.
+///
+/// Example: clang-%%-%%-%%-%%-%%.s => clang-a0-b1-c2-d3-e4.s
+///
+/// @param Model Name to base unique path off of.
+/// @param ResultPath Set to the file's path.
+/// @param MakeAbsolute Whether to use the system temp directory.
+void createUniquePath(const Twine &Model, SmallVectorImpl<char> &ResultPath,
+ bool MakeAbsolute);
+
+/// Create a uniquely named file.
+///
+/// Generates a unique path suitable for a temporary file and then opens it as a
+/// file. The name is based on \a Model with '%' replaced by a random char in
+/// [0-9a-f]. If \a Model is not an absolute path, the temporary file will be
+/// created in the current directory.
+///
+/// Example: clang-%%-%%-%%-%%-%%.s => clang-a0-b1-c2-d3-e4.s
+///
+/// This is an atomic operation. Either the file is created and opened, or the
+/// file system is left untouched.
+///
+/// The intended use is for files that are to be kept, possibly after
+/// renaming them. For example, when running 'clang -c foo.o', the file can
+/// be first created as foo-abc123.o and then renamed.
+///
+/// @param Model Name to base unique path off of.
+/// @param ResultFD Set to the opened file's file descriptor.
+/// @param ResultPath Set to the opened file's absolute path.
+/// @returns errc::success if Result{FD,Path} have been successfully set,
+/// otherwise a platform-specific error_code.
+std::error_code createUniqueFile(const Twine &Model, int &ResultFD,
+ SmallVectorImpl<char> &ResultPath,
+ unsigned Mode = all_read | all_write);
+
+/// Simpler version for clients that don't want an open file. An empty
+/// file will still be created.
+std::error_code createUniqueFile(const Twine &Model,
+ SmallVectorImpl<char> &ResultPath,
+ unsigned Mode = all_read | all_write);
+
+/// Represents a temporary file.
+///
+/// The temporary file must be eventually discarded or given a final name and
+/// kept.
+///
+/// The destructor doesn't implicitly discard because there is no way to
+/// properly handle errors in a destructor.
+class TempFile {
+ bool Done = false;
+ TempFile(StringRef Name, int FD);
+
+public:
+ /// This creates a temporary file with createUniqueFile and schedules it for
+ /// deletion with sys::RemoveFileOnSignal.
+ static Expected<TempFile> create(const Twine &Model,
+ unsigned Mode = all_read | all_write);
+ TempFile(TempFile &&Other);
+ TempFile &operator=(TempFile &&Other);
+
+ // Name of the temporary file.
+ std::string TmpName;
+
+ // The open file descriptor.
+ int FD = -1;
+
+ // Keep this with the given name.
+ Error keep(const Twine &Name);
+
+ // Keep this with the temporary name.
+ Error keep();
+
+ // Delete the file.
+ Error discard();
+
+ // This checks that keep or delete was called.
+ ~TempFile();
+};
+
+/// Create a file in the system temporary directory.
+///
+/// The filename is of the form prefix-random_chars.suffix. Since the directory
+/// is not know to the caller, Prefix and Suffix cannot have path separators.
+/// The files are created with mode 0600.
+///
+/// This should be used for things like a temporary .s that is removed after
+/// running the assembler.
+std::error_code createTemporaryFile(const Twine &Prefix, StringRef Suffix,
+ int &ResultFD,
+ SmallVectorImpl<char> &ResultPath);
+
+/// Simpler version for clients that don't want an open file. An empty
+/// file will still be created.
+std::error_code createTemporaryFile(const Twine &Prefix, StringRef Suffix,
+ SmallVectorImpl<char> &ResultPath);
+
+std::error_code createUniqueDirectory(const Twine &Prefix,
+ SmallVectorImpl<char> &ResultPath);
+
+/// Get a unique name, not currently exisiting in the filesystem. Subject
+/// to race conditions, prefer to use createUniqueFile instead.
+///
+/// Similar to createUniqueFile, but instead of creating a file only
+/// checks if it exists. This function is subject to race conditions, if you
+/// want to use the returned name to actually create a file, use
+/// createUniqueFile instead.
+std::error_code getPotentiallyUniqueFileName(const Twine &Model,
+ SmallVectorImpl<char> &ResultPath);
+
+/// Get a unique temporary file name, not currently exisiting in the
+/// filesystem. Subject to race conditions, prefer to use createTemporaryFile
+/// instead.
+///
+/// Similar to createTemporaryFile, but instead of creating a file only
+/// checks if it exists. This function is subject to race conditions, if you
+/// want to use the returned name to actually create a file, use
+/// createTemporaryFile instead.
+std::error_code
+getPotentiallyUniqueTempFileName(const Twine &Prefix, StringRef Suffix,
+ SmallVectorImpl<char> &ResultPath);
+
+inline OpenFlags operator|(OpenFlags A, OpenFlags B) {
+ return OpenFlags(unsigned(A) | unsigned(B));
+}
+
+inline OpenFlags &operator|=(OpenFlags &A, OpenFlags B) {
+ A = A | B;
+ return A;
+}
+
+inline FileAccess operator|(FileAccess A, FileAccess B) {
+ return FileAccess(unsigned(A) | unsigned(B));
+}
+
+inline FileAccess &operator|=(FileAccess &A, FileAccess B) {
+ A = A | B;
+ return A;
+}
+
+/// @brief Opens a file with the specified creation disposition, access mode,
+/// and flags and returns a file descriptor.
+///
+/// The caller is responsible for closing the file descriptor once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param ResultFD If the file could be opened successfully, its descriptor
+/// is stored in this location. Otherwise, this is set to -1.
+/// @param Disp Value specifying the existing-file behavior.
+/// @param Access Value specifying whether to open the file in read, write, or
+/// read-write mode.
+/// @param Flags Additional flags.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns errc::success if \a Name has been opened, otherwise a
+/// platform-specific error_code.
+std::error_code openFile(const Twine &Name, int &ResultFD,
+ CreationDisposition Disp, FileAccess Access,
+ OpenFlags Flags, unsigned Mode = 0666);
+
+/// @brief Opens a file with the specified creation disposition, access mode,
+/// and flags and returns a platform-specific file object.
+///
+/// The caller is responsible for closing the file object once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param Disp Value specifying the existing-file behavior.
+/// @param Access Value specifying whether to open the file in read, write, or
+/// read-write mode.
+/// @param Flags Additional flags.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns errc::success if \a Name has been opened, otherwise a
+/// platform-specific error_code.
+Expected<file_t> openNativeFile(const Twine &Name, CreationDisposition Disp,
+ FileAccess Access, OpenFlags Flags,
+ unsigned Mode = 0666);
+
+/// Converts from a Posix file descriptor number to a native file handle.
+/// On Windows, this retreives the underlying handle. On non-Windows, this is a
+/// no-op.
+file_t convertFDToNativeFile(int FD);
+
+#ifndef _WIN32
+inline file_t convertFDToNativeFile(int FD) { return FD; }
+#endif
+
+/// Return an open handle to standard in. On Unix, this is typically FD 0.
+/// Returns kInvalidFile when the stream is closed.
+file_t getStdinHandle();
+
+/// Return an open handle to standard out. On Unix, this is typically FD 1.
+/// Returns kInvalidFile when the stream is closed.
+file_t getStdoutHandle();
+
+/// Return an open handle to standard error. On Unix, this is typically FD 2.
+/// Returns kInvalidFile when the stream is closed.
+file_t getStderrHandle();
+
+/// Reads \p Buf.size() bytes from \p FileHandle into \p Buf. Returns the number
+/// of bytes actually read. On Unix, this is equivalent to `return ::read(FD,
+/// Buf.data(), Buf.size())`, with error reporting. Returns 0 when reaching EOF.
+///
+/// @param FileHandle File to read from.
+/// @param Buf Buffer to read into.
+/// @returns The number of bytes read, or error.
+Expected<size_t> readNativeFile(file_t FileHandle, MutableArrayRef<char> Buf);
+
+/// Reads \p Buf.size() bytes from \p FileHandle at offset \p Offset into \p
+/// Buf. If 'pread' is available, this will use that, otherwise it will use
+/// 'lseek'. Returns the number of bytes actually read. Returns 0 when reaching
+/// EOF.
+///
+/// @param FileHandle File to read from.
+/// @param Buf Buffer to read into.
+/// @param Offset Offset into the file at which the read should occur.
+/// @returns The number of bytes read, or error.
+Expected<size_t> readNativeFileSlice(file_t FileHandle,
+ MutableArrayRef<char> Buf,
+ uint64_t Offset);
+
+/// @brief Opens the file with the given name in a write-only or read-write
+/// mode, returning its open file descriptor. If the file does not exist, it
+/// is created.
+///
+/// The caller is responsible for closing the file descriptor once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param ResultFD If the file could be opened successfully, its descriptor
+/// is stored in this location. Otherwise, this is set to -1.
+/// @param Flags Additional flags used to determine whether the file should be
+/// opened in, for example, read-write or in write-only mode.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns errc::success if \a Name has been opened, otherwise a
+/// platform-specific error_code.
+inline std::error_code
+openFileForWrite(const Twine &Name, int &ResultFD,
+ CreationDisposition Disp = CD_CreateAlways,
+ OpenFlags Flags = OF_None, unsigned Mode = 0666) {
+ return openFile(Name, ResultFD, Disp, FA_Write, Flags, Mode);
+}
+
+/// @brief Opens the file with the given name in a write-only or read-write
+/// mode, returning its open file descriptor. If the file does not exist, it
+/// is created.
+///
+/// The caller is responsible for closing the freeing the file once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param Flags Additional flags used to determine whether the file should be
+/// opened in, for example, read-write or in write-only mode.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns a platform-specific file descriptor if \a Name has been opened,
+/// otherwise an error object.
+inline Expected<file_t> openNativeFileForWrite(const Twine &Name,
+ CreationDisposition Disp,
+ OpenFlags Flags,
+ unsigned Mode = 0666) {
+ return openNativeFile(Name, Disp, FA_Write, Flags, Mode);
+}
+
+/// @brief Opens the file with the given name in a write-only or read-write
+/// mode, returning its open file descriptor. If the file does not exist, it
+/// is created.
+///
+/// The caller is responsible for closing the file descriptor once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param ResultFD If the file could be opened successfully, its descriptor
+/// is stored in this location. Otherwise, this is set to -1.
+/// @param Flags Additional flags used to determine whether the file should be
+/// opened in, for example, read-write or in write-only mode.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns errc::success if \a Name has been opened, otherwise a
+/// platform-specific error_code.
+inline std::error_code openFileForReadWrite(const Twine &Name, int &ResultFD,
+ CreationDisposition Disp,
+ OpenFlags Flags,
+ unsigned Mode = 0666) {
+ return openFile(Name, ResultFD, Disp, FA_Write | FA_Read, Flags, Mode);
+}
+
+/// @brief Opens the file with the given name in a write-only or read-write
+/// mode, returning its open file descriptor. If the file does not exist, it
+/// is created.
+///
+/// The caller is responsible for closing the freeing the file once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param Flags Additional flags used to determine whether the file should be
+/// opened in, for example, read-write or in write-only mode.
+/// @param Mode The access permissions of the file, represented in octal.
+/// @returns a platform-specific file descriptor if \a Name has been opened,
+/// otherwise an error object.
+inline Expected<file_t> openNativeFileForReadWrite(const Twine &Name,
+ CreationDisposition Disp,
+ OpenFlags Flags,
+ unsigned Mode = 0666) {
+ return openNativeFile(Name, Disp, FA_Write | FA_Read, Flags, Mode);
+}
+
+/// @brief Opens the file with the given name in a read-only mode, returning
+/// its open file descriptor.
+///
+/// The caller is responsible for closing the file descriptor once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param ResultFD If the file could be opened successfully, its descriptor
+/// is stored in this location. Otherwise, this is set to -1.
+/// @param RealPath If nonnull, extra work is done to determine the real path
+/// of the opened file, and that path is stored in this
+/// location.
+/// @returns errc::success if \a Name has been opened, otherwise a
+/// platform-specific error_code.
+std::error_code openFileForRead(const Twine &Name, int &ResultFD,
+ OpenFlags Flags = OF_None,
+ SmallVectorImpl<char> *RealPath = nullptr);
+
+/// @brief Opens the file with the given name in a read-only mode, returning
+/// its open file descriptor.
+///
+/// The caller is responsible for closing the freeing the file once they are
+/// finished with it.
+///
+/// @param Name The path of the file to open, relative or absolute.
+/// @param RealPath If nonnull, extra work is done to determine the real path
+/// of the opened file, and that path is stored in this
+/// location.
+/// @returns a platform-specific file descriptor if \a Name has been opened,
+/// otherwise an error object.
+Expected<file_t>
+openNativeFileForRead(const Twine &Name, OpenFlags Flags = OF_None,
+ SmallVectorImpl<char> *RealPath = nullptr);
+
+/// @brief Close the file object. This should be used instead of ::close for
+/// portability. On error, the caller should assume the file is closed, as is
+/// the case for Process::SafelyCloseFileDescriptor
+///
+/// @param F On input, this is the file to close. On output, the file is
+/// set to kInvalidFile.
+///
+/// @returns An error code if closing the file failed. Typically, an error here
+/// means that the filesystem may have failed to perform some buffered writes.
+std::error_code closeFile(file_t &F);
+
+std::error_code getUniqueID(const Twine Path, UniqueID &Result);
+
+/// Get disk space usage information.
+///
+/// Note: Users must be careful about "Time Of Check, Time Of Use" kind of bug.
+/// Note: Windows reports results according to the quota allocated to the user.
+///
+/// @param Path Input path.
+/// @returns a space_info structure filled with the capacity, free, and
+/// available space on the device \a Path is on. A platform specific error_code
+/// is returned on error.
+ErrorOr<space_info> disk_space(const Twine &Path);
+
+/// This class represents a memory mapped file. It is based on
+/// boost::iostreams::mapped_file.
+class mapped_file_region {
+public:
+ enum mapmode {
+ readonly, ///< May only access map via const_data as read only.
+ readwrite, ///< May access map via data and modify it. Written to path.
+ priv ///< May modify via data, but changes are lost on destruction.
+ };
+
+private:
+ /// Platform-specific mapping state.
+ size_t Size;
+ void *Mapping;
+#ifdef _WIN32
+ sys::fs::file_t FileHandle;
+#endif
+ mapmode Mode;
+
+ std::error_code init(sys::fs::file_t FD, uint64_t Offset, mapmode Mode);
+
+public:
+ mapped_file_region() = delete;
+ mapped_file_region(mapped_file_region&) = delete;
+ mapped_file_region &operator =(mapped_file_region&) = delete;
+
+ /// \param fd An open file descriptor to map. Does not take ownership of fd.
+ mapped_file_region(sys::fs::file_t fd, mapmode mode, size_t length, uint64_t offset,
+ std::error_code &ec);
+
+ ~mapped_file_region();
+
+ size_t size() const;
+ char *data() const;
+
+ /// Get a const view of the data. Modifying this memory has undefined
+ /// behavior.
+ const char *const_data() const;
+
+ /// \returns The minimum alignment offset must be.
+ static int alignment();
+};
+
+/// Return the path to the main executable, given the value of argv[0] from
+/// program startup and the address of main itself. In extremis, this function
+/// may fail and return an empty path.
+std::string getMainExecutable(const char *argv0, void *MainExecAddr);
+
+/// @}
+/// @name Iterators
+/// @{
+
+/// directory_entry - A single entry in a directory.
+class directory_entry {
+ // FIXME: different platforms make different information available "for free"
+ // when traversing a directory. The design of this class wraps most of the
+ // information in basic_file_status, so on platforms where we can't populate
+ // that whole structure, callers end up paying for a stat().
+ // std::filesystem::directory_entry may be a better model.
+ std::string Path;
+ file_type Type = file_type::type_unknown; // Most platforms can provide this.
+ bool FollowSymlinks = true; // Affects the behavior of status().
+ basic_file_status Status; // If available.
+
+public:
+ explicit directory_entry(const Twine &Path, bool FollowSymlinks = true,
+ file_type Type = file_type::type_unknown,
+ basic_file_status Status = basic_file_status())
+ : Path(Path.str()), Type(Type), FollowSymlinks(FollowSymlinks),
+ Status(Status) {}
+
+ directory_entry() = default;
+
+ void replace_filename(const Twine &Filename, file_type Type,
+ basic_file_status Status = basic_file_status());
+
+ const std::string &path() const { return Path; }
+ // Get basic information about entry file (a subset of fs::status()).
+ // On most platforms this is a stat() call.
+ // On windows the information was already retrieved from the directory.
+ ErrorOr<basic_file_status> status() const;
+ // Get the type of this file.
+ // On most platforms (Linux/Mac/Windows/BSD), this was already retrieved.
+ // On some platforms (e.g. Solaris) this is a stat() call.
+ file_type type() const {
+ if (Type != file_type::type_unknown)
+ return Type;
+ auto S = status();
+ return S ? S->type() : file_type::type_unknown;
+ }
+
+ bool operator==(const directory_entry& RHS) const { return Path == RHS.Path; }
+ bool operator!=(const directory_entry& RHS) const { return !(*this == RHS); }
+ bool operator< (const directory_entry& RHS) const;
+ bool operator<=(const directory_entry& RHS) const;
+ bool operator> (const directory_entry& RHS) const;
+ bool operator>=(const directory_entry& RHS) const;
+};
+
+namespace detail {
+
+ struct DirIterState;
+
+ std::error_code directory_iterator_construct(DirIterState &, StringRef, bool);
+ std::error_code directory_iterator_increment(DirIterState &);
+ std::error_code directory_iterator_destruct(DirIterState &);
+
+ /// Keeps state for the directory_iterator.
+ struct DirIterState {
+ ~DirIterState() {
+ directory_iterator_destruct(*this);
+ }
+
+ intptr_t IterationHandle = 0;
+ directory_entry CurrentEntry;
+ };
+
+} // end namespace detail
+
+/// directory_iterator - Iterates through the entries in path. There is no
+/// operator++ because we need an error_code. If it's really needed we can make
+/// it call report_fatal_error on error.
+class directory_iterator {
+ std::shared_ptr<detail::DirIterState> State;
+ bool FollowSymlinks = true;
+
+public:
+ explicit directory_iterator(const Twine &path, std::error_code &ec,
+ bool follow_symlinks = true)
+ : FollowSymlinks(follow_symlinks) {
+ State = std::make_shared<detail::DirIterState>();
+ SmallString<128> path_storage;
+ ec = detail::directory_iterator_construct(
+ *State, path.toStringRef(path_storage), FollowSymlinks);
+ }
+
+ explicit directory_iterator(const directory_entry &de, std::error_code &ec,
+ bool follow_symlinks = true)
+ : FollowSymlinks(follow_symlinks) {
+ State = std::make_shared<detail::DirIterState>();
+ ec = detail::directory_iterator_construct(
+ *State, de.path(), FollowSymlinks);
+ }
+
+ /// Construct end iterator.
+ directory_iterator() = default;
+
+ // No operator++ because we need error_code.
+ directory_iterator &increment(std::error_code &ec) {
+ ec = directory_iterator_increment(*State);
+ return *this;
+ }
+
+ const directory_entry &operator*() const { return State->CurrentEntry; }
+ const directory_entry *operator->() const { return &State->CurrentEntry; }
+
+ bool operator==(const directory_iterator &RHS) const {
+ if (State == RHS.State)
+ return true;
+ if (!RHS.State)
+ return State->CurrentEntry == directory_entry();
+ if (!State)
+ return RHS.State->CurrentEntry == directory_entry();
+ return State->CurrentEntry == RHS.State->CurrentEntry;
+ }
+
+ bool operator!=(const directory_iterator &RHS) const {
+ return !(*this == RHS);
+ }
+};
+
+namespace detail {
+
+ /// Keeps state for the recursive_directory_iterator.
+ struct RecDirIterState {
+ std::stack<directory_iterator, std::vector<directory_iterator>> Stack;
+ uint16_t Level = 0;
+ bool HasNoPushRequest = false;
+ };
+
+} // end namespace detail
+
+/// recursive_directory_iterator - Same as directory_iterator except for it
+/// recurses down into child directories.
+class recursive_directory_iterator {
+ std::shared_ptr<detail::RecDirIterState> State;
+ bool Follow;
+
+public:
+ recursive_directory_iterator() = default;
+ explicit recursive_directory_iterator(const Twine &path, std::error_code &ec,
+ bool follow_symlinks = true)
+ : State(std::make_shared<detail::RecDirIterState>()),
+ Follow(follow_symlinks) {
+ State->Stack.push(directory_iterator(path, ec, Follow));
+ if (State->Stack.top() == directory_iterator())
+ State.reset();
+ }
+
+ // No operator++ because we need error_code.
+ recursive_directory_iterator &increment(std::error_code &ec) {
+ const directory_iterator end_itr = {};
+
+ if (State->HasNoPushRequest)
+ State->HasNoPushRequest = false;
+ else {
+ file_type type = State->Stack.top()->type();
+ if (type == file_type::symlink_file && Follow) {
+ // Resolve the symlink: is it a directory to recurse into?
+ ErrorOr<basic_file_status> status = State->Stack.top()->status();
+ if (status)
+ type = status->type();
+ // Otherwise broken symlink, and we'll continue.
+ }
+ if (type == file_type::directory_file) {
+ State->Stack.push(directory_iterator(*State->Stack.top(), ec, Follow));
+ if (State->Stack.top() != end_itr) {
+ ++State->Level;
+ return *this;
+ }
+ State->Stack.pop();
+ }
+ }
+
+ while (!State->Stack.empty()
+ && State->Stack.top().increment(ec) == end_itr) {
+ State->Stack.pop();
+ --State->Level;
+ }
+
+ // Check if we are done. If so, create an end iterator.
+ if (State->Stack.empty())
+ State.reset();
+
+ return *this;
+ }
+
+ const directory_entry &operator*() const { return *State->Stack.top(); }
+ const directory_entry *operator->() const { return &*State->Stack.top(); }
+
+ // observers
+ /// Gets the current level. Starting path is at level 0.
+ int level() const { return State->Level; }
+
+ /// Returns true if no_push has been called for this directory_entry.
+ bool no_push_request() const { return State->HasNoPushRequest; }
+
+ // modifiers
+ /// Goes up one level if Level > 0.
+ void pop() {
+ assert(State && "Cannot pop an end iterator!");
+ assert(State->Level > 0 && "Cannot pop an iterator with level < 1");
+
+ const directory_iterator end_itr = {};
+ std::error_code ec;
+ do {
+ if (ec)
+ report_fatal_error("Error incrementing directory iterator.");
+ State->Stack.pop();
+ --State->Level;
+ } while (!State->Stack.empty()
+ && State->Stack.top().increment(ec) == end_itr);
+
+ // Check if we are done. If so, create an end iterator.
+ if (State->Stack.empty())
+ State.reset();
+ }
+
+ /// Does not go down into the current directory_entry.
+ void no_push() { State->HasNoPushRequest = true; }
+
+ bool operator==(const recursive_directory_iterator &RHS) const {
+ return State == RHS.State;
+ }
+
+ bool operator!=(const recursive_directory_iterator &RHS) const {
+ return !(*this == RHS);
+ }
+};
+
+/// @}
+
+} // end namespace fs
+} // end namespace sys
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_FILESYSTEM_H
diff --git a/third_party/llvm-project/include/llvm/Support/Format.h b/third_party/llvm-project/include/llvm/Support/Format.h
new file mode 100644
index 000000000..9dd7b401b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Format.h
@@ -0,0 +1,257 @@
+//===- Format.h - Efficient printf-style formatting for streams -*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the format() function, which can be used with other
+// LLVM subsystems to provide printf-style formatting. This gives all the power
+// and risk of printf. This can be used like this (with raw_ostreams as an
+// example):
+//
+// OS << "mynumber: " << format("%4.5f", 1234.412) << '\n';
+//
+// Or if you prefer:
+//
+// OS << format("mynumber: %4.5f\n", 1234.412);
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMAT_H
+#define LLVM_SUPPORT_FORMAT_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataTypes.h"
+#include <cassert>
+#include <cstdio>
+#include <tuple>
+#include <utility>
+
+namespace llvm {
+
+/// This is a helper class used for handling formatted output. It is the
+/// abstract base class of a templated derived class.
+class format_object_base {
+protected:
+ const char *Fmt;
+ ~format_object_base() = default; // Disallow polymorphic deletion.
+ format_object_base(const format_object_base &) = default;
+ virtual void home(); // Out of line virtual method.
+
+ /// Call snprintf() for this object, on the given buffer and size.
+ virtual int snprint(char *Buffer, unsigned BufferSize) const = 0;
+
+public:
+ format_object_base(const char *fmt) : Fmt(fmt) {}
+
+ /// Format the object into the specified buffer. On success, this returns
+ /// the length of the formatted string. If the buffer is too small, this
+ /// returns a length to retry with, which will be larger than BufferSize.
+ unsigned print(char *Buffer, unsigned BufferSize) const {
+ assert(BufferSize && "Invalid buffer size!");
+
+ // Print the string, leaving room for the terminating null.
+ int N = snprint(Buffer, BufferSize);
+
+ // VC++ and old GlibC return negative on overflow, just double the size.
+ if (N < 0)
+ return BufferSize * 2;
+
+ // Other implementations yield number of bytes needed, not including the
+ // final '\0'.
+ if (unsigned(N) >= BufferSize)
+ return N + 1;
+
+ // Otherwise N is the length of output (not including the final '\0').
+ return N;
+ }
+};
+
+/// These are templated helper classes used by the format function that
+/// capture the object to be formatted and the format string. When actually
+/// printed, this synthesizes the string into a temporary buffer provided and
+/// returns whether or not it is big enough.
+
+// Helper to validate that format() parameters are scalars or pointers.
+template <typename... Args> struct validate_format_parameters;
+template <typename Arg, typename... Args>
+struct validate_format_parameters<Arg, Args...> {
+ static_assert(std::is_scalar<Arg>::value,
+ "format can't be used with non fundamental / non pointer type");
+ validate_format_parameters() { validate_format_parameters<Args...>(); }
+};
+template <> struct validate_format_parameters<> {};
+
+template <typename... Ts>
+class format_object final : public format_object_base {
+ std::tuple<Ts...> Vals;
+
+ template <std::size_t... Is>
+ int snprint_tuple(char *Buffer, unsigned BufferSize,
+ std::index_sequence<Is...>) const {
+#ifdef _MSC_VER
+ return _snprintf(Buffer, BufferSize, Fmt, std::get<Is>(Vals)...);
+#else
+ return snprintf(Buffer, BufferSize, Fmt, std::get<Is>(Vals)...);
+#endif
+ }
+
+public:
+ format_object(const char *fmt, const Ts &... vals)
+ : format_object_base(fmt), Vals(vals...) {
+ validate_format_parameters<Ts...>();
+ }
+
+ int snprint(char *Buffer, unsigned BufferSize) const override {
+ return snprint_tuple(Buffer, BufferSize, std::index_sequence_for<Ts...>());
+ }
+};
+
+/// These are helper functions used to produce formatted output. They use
+/// template type deduction to construct the appropriate instance of the
+/// format_object class to simplify their construction.
+///
+/// This is typically used like:
+/// \code
+/// OS << format("%0.4f", myfloat) << '\n';
+/// \endcode
+
+template <typename... Ts>
+inline format_object<Ts...> format(const char *Fmt, const Ts &... Vals) {
+ return format_object<Ts...>(Fmt, Vals...);
+}
+
+/// This is a helper class for left_justify, right_justify, and center_justify.
+class FormattedString {
+public:
+ enum Justification { JustifyNone, JustifyLeft, JustifyRight, JustifyCenter };
+ FormattedString(StringRef S, unsigned W, Justification J)
+ : Str(S), Width(W), Justify(J) {}
+
+private:
+ StringRef Str;
+ unsigned Width;
+ Justification Justify;
+ friend class raw_ostream;
+};
+
+/// left_justify - append spaces after string so total output is
+/// \p Width characters. If \p Str is larger that \p Width, full string
+/// is written with no padding.
+inline FormattedString left_justify(StringRef Str, unsigned Width) {
+ return FormattedString(Str, Width, FormattedString::JustifyLeft);
+}
+
+/// right_justify - add spaces before string so total output is
+/// \p Width characters. If \p Str is larger that \p Width, full string
+/// is written with no padding.
+inline FormattedString right_justify(StringRef Str, unsigned Width) {
+ return FormattedString(Str, Width, FormattedString::JustifyRight);
+}
+
+/// center_justify - add spaces before and after string so total output is
+/// \p Width characters. If \p Str is larger that \p Width, full string
+/// is written with no padding.
+inline FormattedString center_justify(StringRef Str, unsigned Width) {
+ return FormattedString(Str, Width, FormattedString::JustifyCenter);
+}
+
+/// This is a helper class used for format_hex() and format_decimal().
+class FormattedNumber {
+ uint64_t HexValue;
+ int64_t DecValue;
+ unsigned Width;
+ bool Hex;
+ bool Upper;
+ bool HexPrefix;
+ friend class raw_ostream;
+
+public:
+ FormattedNumber(uint64_t HV, int64_t DV, unsigned W, bool H, bool U,
+ bool Prefix)
+ : HexValue(HV), DecValue(DV), Width(W), Hex(H), Upper(U),
+ HexPrefix(Prefix) {}
+};
+
+/// format_hex - Output \p N as a fixed width hexadecimal. If number will not
+/// fit in width, full number is still printed. Examples:
+/// OS << format_hex(255, 4) => 0xff
+/// OS << format_hex(255, 4, true) => 0xFF
+/// OS << format_hex(255, 6) => 0x00ff
+/// OS << format_hex(255, 2) => 0xff
+inline FormattedNumber format_hex(uint64_t N, unsigned Width,
+ bool Upper = false) {
+ assert(Width <= 18 && "hex width must be <= 18");
+ return FormattedNumber(N, 0, Width, true, Upper, true);
+}
+
+/// format_hex_no_prefix - Output \p N as a fixed width hexadecimal. Does not
+/// prepend '0x' to the outputted string. If number will not fit in width,
+/// full number is still printed. Examples:
+/// OS << format_hex_no_prefix(255, 2) => ff
+/// OS << format_hex_no_prefix(255, 2, true) => FF
+/// OS << format_hex_no_prefix(255, 4) => 00ff
+/// OS << format_hex_no_prefix(255, 1) => ff
+inline FormattedNumber format_hex_no_prefix(uint64_t N, unsigned Width,
+ bool Upper = false) {
+ assert(Width <= 16 && "hex width must be <= 16");
+ return FormattedNumber(N, 0, Width, true, Upper, false);
+}
+
+/// format_decimal - Output \p N as a right justified, fixed-width decimal. If
+/// number will not fit in width, full number is still printed. Examples:
+/// OS << format_decimal(0, 5) => " 0"
+/// OS << format_decimal(255, 5) => " 255"
+/// OS << format_decimal(-1, 3) => " -1"
+/// OS << format_decimal(12345, 3) => "12345"
+inline FormattedNumber format_decimal(int64_t N, unsigned Width) {
+ return FormattedNumber(0, N, Width, false, false, false);
+}
+
+class FormattedBytes {
+ ArrayRef<uint8_t> Bytes;
+
+ // If not None, display offsets for each line relative to starting value.
+ Optional<uint64_t> FirstByteOffset;
+ uint32_t IndentLevel; // Number of characters to indent each line.
+ uint32_t NumPerLine; // Number of bytes to show per line.
+ uint8_t ByteGroupSize; // How many hex bytes are grouped without spaces
+ bool Upper; // Show offset and hex bytes as upper case.
+ bool ASCII; // Show the ASCII bytes for the hex bytes to the right.
+ friend class raw_ostream;
+
+public:
+ FormattedBytes(ArrayRef<uint8_t> B, uint32_t IL, Optional<uint64_t> O,
+ uint32_t NPL, uint8_t BGS, bool U, bool A)
+ : Bytes(B), FirstByteOffset(O), IndentLevel(IL), NumPerLine(NPL),
+ ByteGroupSize(BGS), Upper(U), ASCII(A) {
+
+ if (ByteGroupSize > NumPerLine)
+ ByteGroupSize = NumPerLine;
+ }
+};
+
+inline FormattedBytes
+format_bytes(ArrayRef<uint8_t> Bytes, Optional<uint64_t> FirstByteOffset = None,
+ uint32_t NumPerLine = 16, uint8_t ByteGroupSize = 4,
+ uint32_t IndentLevel = 0, bool Upper = false) {
+ return FormattedBytes(Bytes, IndentLevel, FirstByteOffset, NumPerLine,
+ ByteGroupSize, Upper, false);
+}
+
+inline FormattedBytes
+format_bytes_with_ascii(ArrayRef<uint8_t> Bytes,
+ Optional<uint64_t> FirstByteOffset = None,
+ uint32_t NumPerLine = 16, uint8_t ByteGroupSize = 4,
+ uint32_t IndentLevel = 0, bool Upper = false) {
+ return FormattedBytes(Bytes, IndentLevel, FirstByteOffset, NumPerLine,
+ ByteGroupSize, Upper, true);
+}
+
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FormatAdapters.h b/third_party/llvm-project/include/llvm/Support/FormatAdapters.h
new file mode 100644
index 000000000..a0e8cc439
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormatAdapters.h
@@ -0,0 +1,108 @@
+//===- FormatAdapters.h - Formatters for common LLVM types -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATADAPTERS_H
+#define LLVM_SUPPORT_FORMATADAPTERS_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FormatCommon.h"
+#include "llvm/Support/FormatVariadicDetails.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+template <typename T> class FormatAdapter : public detail::format_adapter {
+protected:
+ explicit FormatAdapter(T &&Item) : Item(std::forward<T>(Item)) {}
+
+ T Item;
+};
+
+namespace detail {
+template <typename T> class AlignAdapter final : public FormatAdapter<T> {
+ AlignStyle Where;
+ size_t Amount;
+ char Fill;
+
+public:
+ AlignAdapter(T &&Item, AlignStyle Where, size_t Amount, char Fill)
+ : FormatAdapter<T>(std::forward<T>(Item)), Where(Where), Amount(Amount),
+ Fill(Fill) {}
+
+ void format(llvm::raw_ostream &Stream, StringRef Style) {
+ auto Adapter = detail::build_format_adapter(std::forward<T>(this->Item));
+ FmtAlign(Adapter, Where, Amount, Fill).format(Stream, Style);
+ }
+};
+
+template <typename T> class PadAdapter final : public FormatAdapter<T> {
+ size_t Left;
+ size_t Right;
+
+public:
+ PadAdapter(T &&Item, size_t Left, size_t Right)
+ : FormatAdapter<T>(std::forward<T>(Item)), Left(Left), Right(Right) {}
+
+ void format(llvm::raw_ostream &Stream, StringRef Style) {
+ auto Adapter = detail::build_format_adapter(std::forward<T>(this->Item));
+ Stream.indent(Left);
+ Adapter.format(Stream, Style);
+ Stream.indent(Right);
+ }
+};
+
+template <typename T> class RepeatAdapter final : public FormatAdapter<T> {
+ size_t Count;
+
+public:
+ RepeatAdapter(T &&Item, size_t Count)
+ : FormatAdapter<T>(std::forward<T>(Item)), Count(Count) {}
+
+ void format(llvm::raw_ostream &Stream, StringRef Style) {
+ auto Adapter = detail::build_format_adapter(std::forward<T>(this->Item));
+ for (size_t I = 0; I < Count; ++I) {
+ Adapter.format(Stream, Style);
+ }
+ }
+};
+
+class ErrorAdapter : public FormatAdapter<Error> {
+public:
+ ErrorAdapter(Error &&Item) : FormatAdapter(std::move(Item)) {}
+ ErrorAdapter(ErrorAdapter &&) = default;
+ ~ErrorAdapter() { consumeError(std::move(Item)); }
+ void format(llvm::raw_ostream &Stream, StringRef Style) { Stream << Item; }
+};
+}
+
+template <typename T>
+detail::AlignAdapter<T> fmt_align(T &&Item, AlignStyle Where, size_t Amount,
+ char Fill = ' ') {
+ return detail::AlignAdapter<T>(std::forward<T>(Item), Where, Amount, Fill);
+}
+
+template <typename T>
+detail::PadAdapter<T> fmt_pad(T &&Item, size_t Left, size_t Right) {
+ return detail::PadAdapter<T>(std::forward<T>(Item), Left, Right);
+}
+
+template <typename T>
+detail::RepeatAdapter<T> fmt_repeat(T &&Item, size_t Count) {
+ return detail::RepeatAdapter<T>(std::forward<T>(Item), Count);
+}
+
+// llvm::Error values must be consumed before being destroyed.
+// Wrapping an error in fmt_consume explicitly indicates that the formatv_object
+// should take ownership and consume it.
+inline detail::ErrorAdapter fmt_consume(Error &&Item) {
+ return detail::ErrorAdapter(std::move(Item));
+}
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FormatCommon.h b/third_party/llvm-project/include/llvm/Support/FormatCommon.h
new file mode 100644
index 000000000..3c119d125
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormatCommon.h
@@ -0,0 +1,76 @@
+//===- FormatCommon.h - Formatters for common LLVM types --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATCOMMON_H
+#define LLVM_SUPPORT_FORMATCOMMON_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/Support/FormatVariadicDetails.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+enum class AlignStyle { Left, Center, Right };
+
+struct FmtAlign {
+ detail::format_adapter &Adapter;
+ AlignStyle Where;
+ size_t Amount;
+ char Fill;
+
+ FmtAlign(detail::format_adapter &Adapter, AlignStyle Where, size_t Amount,
+ char Fill = ' ')
+ : Adapter(Adapter), Where(Where), Amount(Amount), Fill(Fill) {}
+
+ void format(raw_ostream &S, StringRef Options) {
+ // If we don't need to align, we can format straight into the underlying
+ // stream. Otherwise we have to go through an intermediate stream first
+ // in order to calculate how long the output is so we can align it.
+ // TODO: Make the format method return the number of bytes written, that
+ // way we can also skip the intermediate stream for left-aligned output.
+ if (Amount == 0) {
+ Adapter.format(S, Options);
+ return;
+ }
+ SmallString<64> Item;
+ raw_svector_ostream Stream(Item);
+
+ Adapter.format(Stream, Options);
+ if (Amount <= Item.size()) {
+ S << Item;
+ return;
+ }
+
+ size_t PadAmount = Amount - Item.size();
+ switch (Where) {
+ case AlignStyle::Left:
+ S << Item;
+ fill(S, PadAmount);
+ break;
+ case AlignStyle::Center: {
+ size_t X = PadAmount / 2;
+ fill(S, X);
+ S << Item;
+ fill(S, PadAmount - X);
+ break;
+ }
+ default:
+ fill(S, PadAmount);
+ S << Item;
+ break;
+ }
+ }
+
+private:
+ void fill(llvm::raw_ostream &S, uint32_t Count) {
+ for (uint32_t I = 0; I < Count; ++I)
+ S << Fill;
+ }
+};
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FormatProviders.h b/third_party/llvm-project/include/llvm/Support/FormatProviders.h
new file mode 100644
index 000000000..629a48457
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormatProviders.h
@@ -0,0 +1,422 @@
+//===- FormatProviders.h - Formatters for common LLVM types -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements format providers for many common LLVM types, for example
+// allowing precision and width specifiers for scalar and string types.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATPROVIDERS_H
+#define LLVM_SUPPORT_FORMATPROVIDERS_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/StringSwitch.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/FormatVariadicDetails.h"
+#include "llvm/Support/NativeFormatting.h"
+
+#include <type_traits>
+#include <vector>
+
+namespace llvm {
+namespace detail {
+template <typename T>
+struct use_integral_formatter
+ : public std::integral_constant<
+ bool, is_one_of<T, uint8_t, int16_t, uint16_t, int32_t, uint32_t,
+ int64_t, uint64_t, int, unsigned, long, unsigned long,
+ long long, unsigned long long>::value> {};
+
+template <typename T>
+struct use_char_formatter
+ : public std::integral_constant<bool, std::is_same<T, char>::value> {};
+
+template <typename T>
+struct is_cstring
+ : public std::integral_constant<bool,
+ is_one_of<T, char *, const char *>::value> {
+};
+
+template <typename T>
+struct use_string_formatter
+ : public std::integral_constant<bool,
+ std::is_convertible<T, llvm::StringRef>::value> {};
+
+template <typename T>
+struct use_pointer_formatter
+ : public std::integral_constant<bool, std::is_pointer<T>::value &&
+ !is_cstring<T>::value> {};
+
+template <typename T>
+struct use_double_formatter
+ : public std::integral_constant<bool, std::is_floating_point<T>::value> {};
+
+class HelperFunctions {
+protected:
+ static Optional<size_t> parseNumericPrecision(StringRef Str) {
+ size_t Prec;
+ Optional<size_t> Result;
+ if (Str.empty())
+ Result = None;
+ else if (Str.getAsInteger(10, Prec)) {
+ assert(false && "Invalid precision specifier");
+ Result = None;
+ } else {
+ assert(Prec < 100 && "Precision out of range");
+ Result = std::min<size_t>(99u, Prec);
+ }
+ return Result;
+ }
+
+ static bool consumeHexStyle(StringRef &Str, HexPrintStyle &Style) {
+ if (!Str.startswith_lower("x"))
+ return false;
+
+ if (Str.consume_front("x-"))
+ Style = HexPrintStyle::Lower;
+ else if (Str.consume_front("X-"))
+ Style = HexPrintStyle::Upper;
+ else if (Str.consume_front("x+") || Str.consume_front("x"))
+ Style = HexPrintStyle::PrefixLower;
+ else if (Str.consume_front("X+") || Str.consume_front("X"))
+ Style = HexPrintStyle::PrefixUpper;
+ return true;
+ }
+
+ static size_t consumeNumHexDigits(StringRef &Str, HexPrintStyle Style,
+ size_t Default) {
+ Str.consumeInteger(10, Default);
+ if (isPrefixedHexStyle(Style))
+ Default += 2;
+ return Default;
+ }
+};
+}
+
+/// Implementation of format_provider<T> for integral arithmetic types.
+///
+/// The options string of an integral type has the grammar:
+///
+/// integer_options :: [style][digits]
+/// style :: <see table below>
+/// digits :: <non-negative integer> 0-99
+///
+/// ==========================================================================
+/// | style | Meaning | Example | Digits Meaning |
+/// --------------------------------------------------------------------------
+/// | | | Input | Output | |
+/// ==========================================================================
+/// | x- | Hex no prefix, lower | 42 | 2a | Minimum # digits |
+/// | X- | Hex no prefix, upper | 42 | 2A | Minimum # digits |
+/// | x+ / x | Hex + prefix, lower | 42 | 0x2a | Minimum # digits |
+/// | X+ / X | Hex + prefix, upper | 42 | 0x2A | Minimum # digits |
+/// | N / n | Digit grouped number | 123456 | 123,456 | Ignored |
+/// | D / d | Integer | 100000 | 100000 | Ignored |
+/// | (empty) | Same as D / d | | | |
+/// ==========================================================================
+///
+
+template <typename T>
+struct format_provider<
+ T, typename std::enable_if<detail::use_integral_formatter<T>::value>::type>
+ : public detail::HelperFunctions {
+private:
+public:
+ static void format(const T &V, llvm::raw_ostream &Stream, StringRef Style) {
+ HexPrintStyle HS;
+ size_t Digits = 0;
+ if (consumeHexStyle(Style, HS)) {
+ Digits = consumeNumHexDigits(Style, HS, 0);
+ write_hex(Stream, V, HS, Digits);
+ return;
+ }
+
+ IntegerStyle IS = IntegerStyle::Integer;
+ if (Style.consume_front("N") || Style.consume_front("n"))
+ IS = IntegerStyle::Number;
+ else if (Style.consume_front("D") || Style.consume_front("d"))
+ IS = IntegerStyle::Integer;
+
+ Style.consumeInteger(10, Digits);
+ assert(Style.empty() && "Invalid integral format style!");
+ write_integer(Stream, V, Digits, IS);
+ }
+};
+
+/// Implementation of format_provider<T> for integral pointer types.
+///
+/// The options string of a pointer type has the grammar:
+///
+/// pointer_options :: [style][precision]
+/// style :: <see table below>
+/// digits :: <non-negative integer> 0-sizeof(void*)
+///
+/// ==========================================================================
+/// | S | Meaning | Example |
+/// --------------------------------------------------------------------------
+/// | | | Input | Output |
+/// ==========================================================================
+/// | x- | Hex no prefix, lower | 0xDEADBEEF | deadbeef |
+/// | X- | Hex no prefix, upper | 0xDEADBEEF | DEADBEEF |
+/// | x+ / x | Hex + prefix, lower | 0xDEADBEEF | 0xdeadbeef |
+/// | X+ / X | Hex + prefix, upper | 0xDEADBEEF | 0xDEADBEEF |
+/// | (empty) | Same as X+ / X | | |
+/// ==========================================================================
+///
+/// The default precision is the number of nibbles in a machine word, and in all
+/// cases indicates the minimum number of nibbles to print.
+template <typename T>
+struct format_provider<
+ T, typename std::enable_if<detail::use_pointer_formatter<T>::value>::type>
+ : public detail::HelperFunctions {
+private:
+public:
+ static void format(const T &V, llvm::raw_ostream &Stream, StringRef Style) {
+ HexPrintStyle HS = HexPrintStyle::PrefixUpper;
+ consumeHexStyle(Style, HS);
+ size_t Digits = consumeNumHexDigits(Style, HS, sizeof(void *) * 2);
+ write_hex(Stream, reinterpret_cast<std::uintptr_t>(V), HS, Digits);
+ }
+};
+
+/// Implementation of format_provider<T> for c-style strings and string
+/// objects such as std::string and llvm::StringRef.
+///
+/// The options string of a string type has the grammar:
+///
+/// string_options :: [length]
+///
+/// where `length` is an optional integer specifying the maximum number of
+/// characters in the string to print. If `length` is omitted, the string is
+/// printed up to the null terminator.
+
+template <typename T>
+struct format_provider<
+ T, typename std::enable_if<detail::use_string_formatter<T>::value>::type> {
+ static void format(const T &V, llvm::raw_ostream &Stream, StringRef Style) {
+ size_t N = StringRef::npos;
+ if (!Style.empty() && Style.getAsInteger(10, N)) {
+ assert(false && "Style is not a valid integer");
+ }
+ llvm::StringRef S = V;
+ Stream << S.substr(0, N);
+ }
+};
+
+/// Implementation of format_provider<T> for llvm::Twine.
+///
+/// This follows the same rules as the string formatter.
+
+template <> struct format_provider<Twine> {
+ static void format(const Twine &V, llvm::raw_ostream &Stream,
+ StringRef Style) {
+ format_provider<std::string>::format(V.str(), Stream, Style);
+ }
+};
+
+/// Implementation of format_provider<T> for characters.
+///
+/// The options string of a character type has the grammar:
+///
+/// char_options :: (empty) | [integer_options]
+///
+/// If `char_options` is empty, the character is displayed as an ASCII
+/// character. Otherwise, it is treated as an integer options string.
+///
+template <typename T>
+struct format_provider<
+ T, typename std::enable_if<detail::use_char_formatter<T>::value>::type> {
+ static void format(const char &V, llvm::raw_ostream &Stream,
+ StringRef Style) {
+ if (Style.empty())
+ Stream << V;
+ else {
+ int X = static_cast<int>(V);
+ format_provider<int>::format(X, Stream, Style);
+ }
+ }
+};
+
+/// Implementation of format_provider<T> for type `bool`
+///
+/// The options string of a boolean type has the grammar:
+///
+/// bool_options :: "" | "Y" | "y" | "D" | "d" | "T" | "t"
+///
+/// ==================================
+/// | C | Meaning |
+/// ==================================
+/// | Y | YES / NO |
+/// | y | yes / no |
+/// | D / d | Integer 0 or 1 |
+/// | T | TRUE / FALSE |
+/// | t | true / false |
+/// | (empty) | Equivalent to 't' |
+/// ==================================
+template <> struct format_provider<bool> {
+ static void format(const bool &B, llvm::raw_ostream &Stream,
+ StringRef Style) {
+ Stream << StringSwitch<const char *>(Style)
+ .Case("Y", B ? "YES" : "NO")
+ .Case("y", B ? "yes" : "no")
+ .CaseLower("D", B ? "1" : "0")
+ .Case("T", B ? "TRUE" : "FALSE")
+ .Cases("t", "", B ? "true" : "false")
+ .Default(B ? "1" : "0");
+ }
+};
+
+/// Implementation of format_provider<T> for floating point types.
+///
+/// The options string of a floating point type has the format:
+///
+/// float_options :: [style][precision]
+/// style :: <see table below>
+/// precision :: <non-negative integer> 0-99
+///
+/// =====================================================
+/// | style | Meaning | Example |
+/// -----------------------------------------------------
+/// | | | Input | Output |
+/// =====================================================
+/// | P / p | Percentage | 0.05 | 5.00% |
+/// | F / f | Fixed point | 1.0 | 1.00 |
+/// | E | Exponential with E | 100000 | 1.0E+05 |
+/// | e | Exponential with e | 100000 | 1.0e+05 |
+/// | (empty) | Same as F / f | | |
+/// =====================================================
+///
+/// The default precision is 6 for exponential (E / e) and 2 for everything
+/// else.
+
+template <typename T>
+struct format_provider<
+ T, typename std::enable_if<detail::use_double_formatter<T>::value>::type>
+ : public detail::HelperFunctions {
+ static void format(const T &V, llvm::raw_ostream &Stream, StringRef Style) {
+ FloatStyle S;
+ if (Style.consume_front("P") || Style.consume_front("p"))
+ S = FloatStyle::Percent;
+ else if (Style.consume_front("F") || Style.consume_front("f"))
+ S = FloatStyle::Fixed;
+ else if (Style.consume_front("E"))
+ S = FloatStyle::ExponentUpper;
+ else if (Style.consume_front("e"))
+ S = FloatStyle::Exponent;
+ else
+ S = FloatStyle::Fixed;
+
+ Optional<size_t> Precision = parseNumericPrecision(Style);
+ if (!Precision.hasValue())
+ Precision = getDefaultPrecision(S);
+
+ write_double(Stream, static_cast<double>(V), S, Precision);
+ }
+};
+
+namespace detail {
+template <typename IterT>
+using IterValue = typename std::iterator_traits<IterT>::value_type;
+
+template <typename IterT>
+struct range_item_has_provider
+ : public std::integral_constant<
+ bool, !uses_missing_provider<IterValue<IterT>>::value> {};
+}
+
+/// Implementation of format_provider<T> for ranges.
+///
+/// This will print an arbitrary range as a delimited sequence of items.
+///
+/// The options string of a range type has the grammar:
+///
+/// range_style ::= [separator] [element_style]
+/// separator ::= "$" delimeted_expr
+/// element_style ::= "@" delimeted_expr
+/// delimeted_expr ::= "[" expr "]" | "(" expr ")" | "<" expr ">"
+/// expr ::= <any string not containing delimeter>
+///
+/// where the separator expression is the string to insert between consecutive
+/// items in the range and the argument expression is the Style specification to
+/// be used when formatting the underlying type. The default separator if
+/// unspecified is ' ' (space). The syntax of the argument expression follows
+/// whatever grammar is dictated by the format provider or format adapter used
+/// to format the value type.
+///
+/// Note that attempting to format an `iterator_range<T>` where no format
+/// provider can be found for T will result in a compile error.
+///
+
+template <typename IterT> class format_provider<llvm::iterator_range<IterT>> {
+ using value = typename std::iterator_traits<IterT>::value_type;
+ using reference = typename std::iterator_traits<IterT>::reference;
+
+ static StringRef consumeOneOption(StringRef &Style, char Indicator,
+ StringRef Default) {
+ if (Style.empty())
+ return Default;
+ if (Style.front() != Indicator)
+ return Default;
+ Style = Style.drop_front();
+ if (Style.empty()) {
+ assert(false && "Invalid range style");
+ return Default;
+ }
+
+ for (const char *D : {"[]", "<>", "()"}) {
+ if (Style.front() != D[0])
+ continue;
+ size_t End = Style.find_first_of(D[1]);
+ if (End == StringRef::npos) {
+ assert(false && "Missing range option end delimeter!");
+ return Default;
+ }
+ StringRef Result = Style.slice(1, End);
+ Style = Style.drop_front(End + 1);
+ return Result;
+ }
+ assert(false && "Invalid range style!");
+ return Default;
+ }
+
+ static std::pair<StringRef, StringRef> parseOptions(StringRef Style) {
+ StringRef Sep = consumeOneOption(Style, '$', ", ");
+ StringRef Args = consumeOneOption(Style, '@', "");
+ assert(Style.empty() && "Unexpected text in range option string!");
+ return std::make_pair(Sep, Args);
+ }
+
+public:
+ static_assert(detail::range_item_has_provider<IterT>::value,
+ "Range value_type does not have a format provider!");
+ static void format(const llvm::iterator_range<IterT> &V,
+ llvm::raw_ostream &Stream, StringRef Style) {
+ StringRef Sep;
+ StringRef ArgStyle;
+ std::tie(Sep, ArgStyle) = parseOptions(Style);
+ auto Begin = V.begin();
+ auto End = V.end();
+ if (Begin != End) {
+ auto Adapter =
+ detail::build_format_adapter(std::forward<reference>(*Begin));
+ Adapter.format(Stream, ArgStyle);
+ ++Begin;
+ }
+ while (Begin != End) {
+ Stream << Sep;
+ auto Adapter =
+ detail::build_format_adapter(std::forward<reference>(*Begin));
+ Adapter.format(Stream, ArgStyle);
+ ++Begin;
+ }
+ }
+};
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FormatVariadic.h b/third_party/llvm-project/include/llvm/Support/FormatVariadic.h
new file mode 100644
index 000000000..86a9d30cc
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormatVariadic.h
@@ -0,0 +1,264 @@
+//===- FormatVariadic.h - Efficient type-safe string formatting --*- C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the formatv() function which can be used with other LLVM
+// subsystems to provide printf-like formatting, but with improved safety and
+// flexibility. The result of `formatv` is an object which can be streamed to
+// a raw_ostream or converted to a std::string or llvm::SmallString.
+//
+// // Convert to std::string.
+// std::string S = formatv("{0} {1}", 1234.412, "test").str();
+//
+// // Convert to llvm::SmallString
+// SmallString<8> S = formatv("{0} {1}", 1234.412, "test").sstr<8>();
+//
+// // Stream to an existing raw_ostream.
+// OS << formatv("{0} {1}", 1234.412, "test");
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATVARIADIC_H
+#define LLVM_SUPPORT_FORMATVARIADIC_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/FormatCommon.h"
+#include "llvm/Support/FormatProviders.h"
+#include "llvm/Support/FormatVariadicDetails.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstddef>
+#include <string>
+#include <tuple>
+#include <utility>
+#include <vector>
+
+namespace llvm {
+
+enum class ReplacementType { Empty, Format, Literal };
+
+struct ReplacementItem {
+ ReplacementItem() = default;
+ explicit ReplacementItem(StringRef Literal)
+ : Type(ReplacementType::Literal), Spec(Literal) {}
+ ReplacementItem(StringRef Spec, size_t Index, size_t Align, AlignStyle Where,
+ char Pad, StringRef Options)
+ : Type(ReplacementType::Format), Spec(Spec), Index(Index), Align(Align),
+ Where(Where), Pad(Pad), Options(Options) {}
+
+ ReplacementType Type = ReplacementType::Empty;
+ StringRef Spec;
+ size_t Index = 0;
+ size_t Align = 0;
+ AlignStyle Where = AlignStyle::Right;
+ char Pad = 0;
+ StringRef Options;
+};
+
+class formatv_object_base {
+protected:
+ // The parameters are stored in a std::tuple, which does not provide runtime
+ // indexing capabilities. In order to enable runtime indexing, we use this
+ // structure to put the parameters into a std::vector. Since the parameters
+ // are not all the same type, we use some type-erasure by wrapping the
+ // parameters in a template class that derives from a non-template superclass.
+ // Essentially, we are converting a std::tuple<Derived<Ts...>> to a
+ // std::vector<Base*>.
+ struct create_adapters {
+ template <typename... Ts>
+ std::vector<detail::format_adapter *> operator()(Ts &... Items) {
+ return std::vector<detail::format_adapter *>{&Items...};
+ }
+ };
+
+ StringRef Fmt;
+ std::vector<detail::format_adapter *> Adapters;
+ std::vector<ReplacementItem> Replacements;
+
+ static bool consumeFieldLayout(StringRef &Spec, AlignStyle &Where,
+ size_t &Align, char &Pad);
+
+ static std::pair<ReplacementItem, StringRef>
+ splitLiteralAndReplacement(StringRef Fmt);
+
+public:
+ formatv_object_base(StringRef Fmt, std::size_t ParamCount)
+ : Fmt(Fmt), Replacements(parseFormatString(Fmt)) {
+ Adapters.reserve(ParamCount);
+ }
+
+ formatv_object_base(formatv_object_base const &rhs) = delete;
+
+ formatv_object_base(formatv_object_base &&rhs)
+ : Fmt(std::move(rhs.Fmt)),
+ Adapters(), // Adapters are initialized by formatv_object
+ Replacements(std::move(rhs.Replacements)) {
+ Adapters.reserve(rhs.Adapters.size());
+ };
+
+ void format(raw_ostream &S) const {
+ for (auto &R : Replacements) {
+ if (R.Type == ReplacementType::Empty)
+ continue;
+ if (R.Type == ReplacementType::Literal) {
+ S << R.Spec;
+ continue;
+ }
+ if (R.Index >= Adapters.size()) {
+ S << R.Spec;
+ continue;
+ }
+
+ auto W = Adapters[R.Index];
+
+ FmtAlign Align(*W, R.Where, R.Align, R.Pad);
+ Align.format(S, R.Options);
+ }
+ }
+ static std::vector<ReplacementItem> parseFormatString(StringRef Fmt);
+
+ static Optional<ReplacementItem> parseReplacementItem(StringRef Spec);
+
+ std::string str() const {
+ std::string Result;
+ raw_string_ostream Stream(Result);
+ Stream << *this;
+ Stream.flush();
+ return Result;
+ }
+
+ template <unsigned N> SmallString<N> sstr() const {
+ SmallString<N> Result;
+ raw_svector_ostream Stream(Result);
+ Stream << *this;
+ return Result;
+ }
+
+ template <unsigned N> operator SmallString<N>() const { return sstr<N>(); }
+
+ operator std::string() const { return str(); }
+};
+
+template <typename Tuple> class formatv_object : public formatv_object_base {
+ // Storage for the parameter adapters. Since the base class erases the type
+ // of the parameters, we have to own the storage for the parameters here, and
+ // have the base class store type-erased pointers into this tuple.
+ Tuple Parameters;
+
+public:
+ formatv_object(StringRef Fmt, Tuple &&Params)
+ : formatv_object_base(Fmt, std::tuple_size<Tuple>::value),
+ Parameters(std::move(Params)) {
+ Adapters = apply_tuple(create_adapters(), Parameters);
+ }
+
+ formatv_object(formatv_object const &rhs) = delete;
+
+ formatv_object(formatv_object &&rhs)
+ : formatv_object_base(std::move(rhs)),
+ Parameters(std::move(rhs.Parameters)) {
+ Adapters = apply_tuple(create_adapters(), Parameters);
+ }
+};
+
+// Format text given a format string and replacement parameters.
+//
+// ===General Description===
+//
+// Formats textual output. `Fmt` is a string consisting of one or more
+// replacement sequences with the following grammar:
+//
+// rep_field ::= "{" [index] ["," layout] [":" format] "}"
+// index ::= <non-negative integer>
+// layout ::= [[[char]loc]width]
+// format ::= <any string not containing "{" or "}">
+// char ::= <any character except "{" or "}">
+// loc ::= "-" | "=" | "+"
+// width ::= <positive integer>
+//
+// index - A non-negative integer specifying the index of the item in the
+// parameter pack to print. Any other value is invalid.
+// layout - A string controlling how the field is laid out within the available
+// space.
+// format - A type-dependent string used to provide additional options to
+// the formatting operation. Refer to the documentation of the
+// various individual format providers for per-type options.
+// char - The padding character. Defaults to ' ' (space). Only valid if
+// `loc` is also specified.
+// loc - Where to print the formatted text within the field. Only valid if
+// `width` is also specified.
+// '-' : The field is left aligned within the available space.
+// '=' : The field is centered within the available space.
+// '+' : The field is right aligned within the available space (this
+// is the default).
+// width - The width of the field within which to print the formatted text.
+// If this is less than the required length then the `char` and `loc`
+// fields are ignored, and the field is printed with no leading or
+// trailing padding. If this is greater than the required length,
+// then the text is output according to the value of `loc`, and padded
+// as appropriate on the left and/or right by `char`.
+//
+// ===Special Characters===
+//
+// The characters '{' and '}' are reserved and cannot appear anywhere within a
+// replacement sequence. Outside of a replacement sequence, in order to print
+// a literal '{' or '}' it must be doubled -- "{{" to print a literal '{' and
+// "}}" to print a literal '}'.
+//
+// ===Parameter Indexing===
+// `index` specifies the index of the parameter in the parameter pack to format
+// into the output. Note that it is possible to refer to the same parameter
+// index multiple times in a given format string. This makes it possible to
+// output the same value multiple times without passing it multiple times to the
+// function. For example:
+//
+// formatv("{0} {1} {0}", "a", "bb")
+//
+// would yield the string "abba". This can be convenient when it is expensive
+// to compute the value of the parameter, and you would otherwise have had to
+// save it to a temporary.
+//
+// ===Formatter Search===
+//
+// For a given parameter of type T, the following steps are executed in order
+// until a match is found:
+//
+// 1. If the parameter is of class type, and inherits from format_adapter,
+// Then format() is invoked on it to produce the formatted output. The
+// implementation should write the formatted text into `Stream`.
+// 2. If there is a suitable template specialization of format_provider<>
+// for type T containing a method whose signature is:
+// void format(const T &Obj, raw_ostream &Stream, StringRef Options)
+// Then this method is invoked as described in Step 1.
+// 3. If an appropriate operator<< for raw_ostream exists, it will be used.
+// For this to work, (raw_ostream& << const T&) must return raw_ostream&.
+//
+// If a match cannot be found through either of the above methods, a compiler
+// error is generated.
+//
+// ===Invalid Format String Handling===
+//
+// In the case of a format string which does not match the grammar described
+// above, the output is undefined. With asserts enabled, LLVM will trigger an
+// assertion. Otherwise, it will try to do something reasonable, but in general
+// the details of what that is are undefined.
+//
+template <typename... Ts>
+inline auto formatv(const char *Fmt, Ts &&... Vals) -> formatv_object<decltype(
+ std::make_tuple(detail::build_format_adapter(std::forward<Ts>(Vals))...))> {
+ using ParamTuple = decltype(
+ std::make_tuple(detail::build_format_adapter(std::forward<Ts>(Vals))...));
+ return formatv_object<ParamTuple>(
+ Fmt,
+ std::make_tuple(detail::build_format_adapter(std::forward<Ts>(Vals))...));
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_FORMATVARIADIC_H
diff --git a/third_party/llvm-project/include/llvm/Support/FormatVariadicDetails.h b/third_party/llvm-project/include/llvm/Support/FormatVariadicDetails.h
new file mode 100644
index 000000000..e3c185134
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormatVariadicDetails.h
@@ -0,0 +1,164 @@
+//===- FormatVariadicDetails.h - Helpers for FormatVariadic.h ----*- C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATVARIADIC_DETAILS_H
+#define LLVM_SUPPORT_FORMATVARIADIC_DETAILS_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/raw_ostream.h"
+
+#include <type_traits>
+
+namespace llvm {
+template <typename T, typename Enable = void> struct format_provider {};
+class Error;
+
+namespace detail {
+class format_adapter {
+ virtual void anchor();
+
+protected:
+ virtual ~format_adapter() {}
+
+public:
+ virtual void format(raw_ostream &S, StringRef Options) = 0;
+};
+
+template <typename T> class provider_format_adapter : public format_adapter {
+ T Item;
+
+public:
+ explicit provider_format_adapter(T &&Item) : Item(std::forward<T>(Item)) {}
+
+ void format(llvm::raw_ostream &S, StringRef Options) override {
+ format_provider<typename std::decay<T>::type>::format(Item, S, Options);
+ }
+};
+
+template <typename T>
+class stream_operator_format_adapter : public format_adapter {
+ T Item;
+
+public:
+ explicit stream_operator_format_adapter(T &&Item)
+ : Item(std::forward<T>(Item)) {}
+
+ void format(llvm::raw_ostream &S, StringRef Options) override { S << Item; }
+};
+
+template <typename T> class missing_format_adapter;
+
+// Test if format_provider<T> is defined on T and contains a member function
+// with the signature:
+// static void format(const T&, raw_stream &, StringRef);
+//
+template <class T> class has_FormatProvider {
+public:
+ using Decayed = typename std::decay<T>::type;
+ typedef void (*Signature_format)(const Decayed &, llvm::raw_ostream &,
+ StringRef);
+
+ template <typename U>
+ static char test(SameType<Signature_format, &U::format> *);
+
+ template <typename U> static double test(...);
+
+ static bool const value =
+ (sizeof(test<llvm::format_provider<Decayed>>(nullptr)) == 1);
+};
+
+// Test if raw_ostream& << T -> raw_ostream& is findable via ADL.
+template <class T> class has_StreamOperator {
+public:
+ using ConstRefT = const typename std::decay<T>::type &;
+
+ template <typename U>
+ static char test(typename std::enable_if<
+ std::is_same<decltype(std::declval<llvm::raw_ostream &>()
+ << std::declval<U>()),
+ llvm::raw_ostream &>::value,
+ int *>::type);
+
+ template <typename U> static double test(...);
+
+ static bool const value = (sizeof(test<ConstRefT>(nullptr)) == 1);
+};
+
+// Simple template that decides whether a type T should use the member-function
+// based format() invocation.
+template <typename T>
+struct uses_format_member
+ : public std::integral_constant<
+ bool,
+ std::is_base_of<format_adapter,
+ typename std::remove_reference<T>::type>::value> {};
+
+// Simple template that decides whether a type T should use the format_provider
+// based format() invocation. The member function takes priority, so this test
+// will only be true if there is not ALSO a format member.
+template <typename T>
+struct uses_format_provider
+ : public std::integral_constant<
+ bool, !uses_format_member<T>::value && has_FormatProvider<T>::value> {
+};
+
+// Simple template that decides whether a type T should use the operator<<
+// based format() invocation. This takes last priority.
+template <typename T>
+struct uses_stream_operator
+ : public std::integral_constant<bool, !uses_format_member<T>::value &&
+ !uses_format_provider<T>::value &&
+ has_StreamOperator<T>::value> {};
+
+// Simple template that decides whether a type T has neither a member-function
+// nor format_provider based implementation that it can use. Mostly used so
+// that the compiler spits out a nice diagnostic when a type with no format
+// implementation can be located.
+template <typename T>
+struct uses_missing_provider
+ : public std::integral_constant<bool, !uses_format_member<T>::value &&
+ !uses_format_provider<T>::value &&
+ !uses_stream_operator<T>::value> {
+};
+
+template <typename T>
+typename std::enable_if<uses_format_member<T>::value, T>::type
+build_format_adapter(T &&Item) {
+ return std::forward<T>(Item);
+}
+
+template <typename T>
+typename std::enable_if<uses_format_provider<T>::value,
+ provider_format_adapter<T>>::type
+build_format_adapter(T &&Item) {
+ return provider_format_adapter<T>(std::forward<T>(Item));
+}
+
+template <typename T>
+typename std::enable_if<uses_stream_operator<T>::value,
+ stream_operator_format_adapter<T>>::type
+build_format_adapter(T &&Item) {
+ // If the caller passed an Error by value, then stream_operator_format_adapter
+ // would be responsible for consuming it.
+ // Make the caller opt into this by calling fmt_consume().
+ static_assert(
+ !std::is_same<llvm::Error, typename std::remove_cv<T>::type>::value,
+ "llvm::Error-by-value must be wrapped in fmt_consume() for formatv");
+ return stream_operator_format_adapter<T>(std::forward<T>(Item));
+}
+
+template <typename T>
+typename std::enable_if<uses_missing_provider<T>::value,
+ missing_format_adapter<T>>::type
+build_format_adapter(T &&Item) {
+ return missing_format_adapter<T>();
+}
+}
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/FormattedStream.h b/third_party/llvm-project/include/llvm/Support/FormattedStream.h
new file mode 100644
index 000000000..b49c8d865
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/FormattedStream.h
@@ -0,0 +1,161 @@
+//===-- llvm/Support/FormattedStream.h - Formatted streams ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains raw_ostream implementations for streams to do
+// things like pretty-print comments.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_FORMATTEDSTREAM_H
+#define LLVM_SUPPORT_FORMATTEDSTREAM_H
+
+#include "llvm/Support/raw_ostream.h"
+#include <utility>
+
+namespace llvm {
+
+/// formatted_raw_ostream - A raw_ostream that wraps another one and keeps track
+/// of line and column position, allowing padding out to specific column
+/// boundaries and querying the number of lines written to the stream.
+///
+class formatted_raw_ostream : public raw_ostream {
+ /// TheStream - The real stream we output to. We set it to be
+ /// unbuffered, since we're already doing our own buffering.
+ ///
+ raw_ostream *TheStream;
+
+ /// Position - The current output column and line of the data that's
+ /// been flushed and the portion of the buffer that's been
+ /// scanned. The line and column scheme is zero-based.
+ ///
+ std::pair<unsigned, unsigned> Position;
+
+ /// Scanned - This points to one past the last character in the
+ /// buffer we've scanned.
+ ///
+ const char *Scanned;
+
+ void write_impl(const char *Ptr, size_t Size) override;
+
+ /// current_pos - Return the current position within the stream,
+ /// not counting the bytes currently in the buffer.
+ uint64_t current_pos() const override {
+ // Our current position in the stream is all the contents which have been
+ // written to the underlying stream (*not* the current position of the
+ // underlying stream).
+ return TheStream->tell();
+ }
+
+ /// ComputePosition - Examine the given output buffer and figure out the new
+ /// position after output.
+ ///
+ void ComputePosition(const char *Ptr, size_t size);
+
+ void setStream(raw_ostream &Stream) {
+ releaseStream();
+
+ TheStream = &Stream;
+
+ // This formatted_raw_ostream inherits from raw_ostream, so it'll do its
+ // own buffering, and it doesn't need or want TheStream to do another
+ // layer of buffering underneath. Resize the buffer to what TheStream
+ // had been using, and tell TheStream not to do its own buffering.
+ if (size_t BufferSize = TheStream->GetBufferSize())
+ SetBufferSize(BufferSize);
+ else
+ SetUnbuffered();
+ TheStream->SetUnbuffered();
+
+ Scanned = nullptr;
+ }
+
+public:
+ /// formatted_raw_ostream - Open the specified file for
+ /// writing. If an error occurs, information about the error is
+ /// put into ErrorInfo, and the stream should be immediately
+ /// destroyed; the string will be empty if no error occurred.
+ ///
+ /// As a side effect, the given Stream is set to be Unbuffered.
+ /// This is because formatted_raw_ostream does its own buffering,
+ /// so it doesn't want another layer of buffering to be happening
+ /// underneath it.
+ ///
+ formatted_raw_ostream(raw_ostream &Stream)
+ : TheStream(nullptr), Position(0, 0) {
+ setStream(Stream);
+ }
+ explicit formatted_raw_ostream() : TheStream(nullptr), Position(0, 0) {
+ Scanned = nullptr;
+ }
+
+ ~formatted_raw_ostream() override {
+ flush();
+ releaseStream();
+ }
+
+ /// PadToColumn - Align the output to some column number. If the current
+ /// column is already equal to or more than NewCol, PadToColumn inserts one
+ /// space.
+ ///
+ /// \param NewCol - The column to move to.
+ formatted_raw_ostream &PadToColumn(unsigned NewCol);
+
+ /// getColumn - Return the column number
+ unsigned getColumn() { return Position.first; }
+
+ /// getLine - Return the line number
+ unsigned getLine() { return Position.second; }
+
+ raw_ostream &resetColor() override {
+ TheStream->resetColor();
+ return *this;
+ }
+
+ raw_ostream &reverseColor() override {
+ TheStream->reverseColor();
+ return *this;
+ }
+
+ raw_ostream &changeColor(enum Colors Color, bool Bold, bool BG) override {
+ TheStream->changeColor(Color, Bold, BG);
+ return *this;
+ }
+
+ bool is_displayed() const override {
+ return TheStream->is_displayed();
+ }
+
+private:
+ void releaseStream() {
+ // Transfer the buffer settings from this raw_ostream back to the underlying
+ // stream.
+ if (!TheStream)
+ return;
+ if (size_t BufferSize = GetBufferSize())
+ TheStream->SetBufferSize(BufferSize);
+ else
+ TheStream->SetUnbuffered();
+ }
+};
+
+/// fouts() - This returns a reference to a formatted_raw_ostream for
+/// standard output. Use it like: fouts() << "foo" << "bar";
+formatted_raw_ostream &fouts();
+
+/// ferrs() - This returns a reference to a formatted_raw_ostream for
+/// standard error. Use it like: ferrs() << "foo" << "bar";
+formatted_raw_ostream &ferrs();
+
+/// fdbgs() - This returns a reference to a formatted_raw_ostream for
+/// debug output. Use it like: fdbgs() << "foo" << "bar";
+formatted_raw_ostream &fdbgs();
+
+} // end llvm namespace
+
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Host.h b/third_party/llvm-project/include/llvm/Support/Host.h
new file mode 100644
index 000000000..44f543c36
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Host.h
@@ -0,0 +1,70 @@
+//===- llvm/Support/Host.h - Host machine characteristics --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Methods for querying the nature of the host machine.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_HOST_H
+#define LLVM_SUPPORT_HOST_H
+
+#include "llvm/ADT/StringMap.h"
+
+#include <string>
+
+namespace llvm {
+namespace sys {
+
+ /// getDefaultTargetTriple() - Return the default target triple the compiler
+ /// has been configured to produce code for.
+ ///
+ /// The target triple is a string in the format of:
+ /// CPU_TYPE-VENDOR-OPERATING_SYSTEM
+ /// or
+ /// CPU_TYPE-VENDOR-KERNEL-OPERATING_SYSTEM
+ std::string getDefaultTargetTriple();
+
+ /// getProcessTriple() - Return an appropriate target triple for generating
+ /// code to be loaded into the current process, e.g. when using the JIT.
+ std::string getProcessTriple();
+
+ /// getHostCPUName - Get the LLVM name for the host CPU. The particular format
+ /// of the name is target dependent, and suitable for passing as -mcpu to the
+ /// target which matches the host.
+ ///
+ /// \return - The host CPU name, or empty if the CPU could not be determined.
+ StringRef getHostCPUName();
+
+ /// getHostCPUFeatures - Get the LLVM names for the host CPU features.
+ /// The particular format of the names are target dependent, and suitable for
+ /// passing as -mattr to the target which matches the host.
+ ///
+ /// \param Features - A string mapping feature names to either
+ /// true (if enabled) or false (if disabled). This routine makes no guarantees
+ /// about exactly which features may appear in this map, except that they are
+ /// all valid LLVM feature names.
+ ///
+ /// \return - True on success.
+ bool getHostCPUFeatures(StringMap<bool> &Features);
+
+ /// Get the number of physical cores (as opposed to logical cores returned
+ /// from thread::hardware_concurrency(), which includes hyperthreads).
+ /// Returns -1 if unknown for the current host system.
+ int getHostNumPhysicalCores();
+
+ namespace detail {
+ /// Helper functions to extract HostCPUName from /proc/cpuinfo on linux.
+ StringRef getHostCPUNameForPowerPC(StringRef ProcCpuinfoContent);
+ StringRef getHostCPUNameForARM(StringRef ProcCpuinfoContent);
+ StringRef getHostCPUNameForS390x(StringRef ProcCpuinfoContent);
+ StringRef getHostCPUNameForBPF();
+ }
+}
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/LEB128.h b/third_party/llvm-project/include/llvm/Support/LEB128.h
new file mode 100644
index 000000000..a02b83ca9
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/LEB128.h
@@ -0,0 +1,198 @@
+//===- llvm/Support/LEB128.h - [SU]LEB128 utility functions -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares some utility functions for encoding SLEB128 and
+// ULEB128 values.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_LEB128_H
+#define LLVM_SUPPORT_LEB128_H
+
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+
+/// Utility function to encode a SLEB128 value to an output stream. Returns
+/// the length in bytes of the encoded value.
+inline unsigned encodeSLEB128(int64_t Value, raw_ostream &OS,
+ unsigned PadTo = 0) {
+ bool More;
+ unsigned Count = 0;
+ do {
+ uint8_t Byte = Value & 0x7f;
+ // NOTE: this assumes that this signed shift is an arithmetic right shift.
+ Value >>= 7;
+ More = !((((Value == 0 ) && ((Byte & 0x40) == 0)) ||
+ ((Value == -1) && ((Byte & 0x40) != 0))));
+ Count++;
+ if (More || Count < PadTo)
+ Byte |= 0x80; // Mark this byte to show that more bytes will follow.
+ OS << char(Byte);
+ } while (More);
+
+ // Pad with 0x80 and emit a terminating byte at the end.
+ if (Count < PadTo) {
+ uint8_t PadValue = Value < 0 ? 0x7f : 0x00;
+ for (; Count < PadTo - 1; ++Count)
+ OS << char(PadValue | 0x80);
+ OS << char(PadValue);
+ Count++;
+ }
+ return Count;
+}
+
+/// Utility function to encode a SLEB128 value to a buffer. Returns
+/// the length in bytes of the encoded value.
+inline unsigned encodeSLEB128(int64_t Value, uint8_t *p, unsigned PadTo = 0) {
+ uint8_t *orig_p = p;
+ unsigned Count = 0;
+ bool More;
+ do {
+ uint8_t Byte = Value & 0x7f;
+ // NOTE: this assumes that this signed shift is an arithmetic right shift.
+ Value >>= 7;
+ More = !((((Value == 0 ) && ((Byte & 0x40) == 0)) ||
+ ((Value == -1) && ((Byte & 0x40) != 0))));
+ Count++;
+ if (More || Count < PadTo)
+ Byte |= 0x80; // Mark this byte to show that more bytes will follow.
+ *p++ = Byte;
+ } while (More);
+
+ // Pad with 0x80 and emit a terminating byte at the end.
+ if (Count < PadTo) {
+ uint8_t PadValue = Value < 0 ? 0x7f : 0x00;
+ for (; Count < PadTo - 1; ++Count)
+ *p++ = (PadValue | 0x80);
+ *p++ = PadValue;
+ }
+ return (unsigned)(p - orig_p);
+}
+
+/// Utility function to encode a ULEB128 value to an output stream. Returns
+/// the length in bytes of the encoded value.
+inline unsigned encodeULEB128(uint64_t Value, raw_ostream &OS,
+ unsigned PadTo = 0) {
+ unsigned Count = 0;
+ do {
+ uint8_t Byte = Value & 0x7f;
+ Value >>= 7;
+ Count++;
+ if (Value != 0 || Count < PadTo)
+ Byte |= 0x80; // Mark this byte to show that more bytes will follow.
+ OS << char(Byte);
+ } while (Value != 0);
+
+ // Pad with 0x80 and emit a null byte at the end.
+ if (Count < PadTo) {
+ for (; Count < PadTo - 1; ++Count)
+ OS << '\x80';
+ OS << '\x00';
+ Count++;
+ }
+ return Count;
+}
+
+/// Utility function to encode a ULEB128 value to a buffer. Returns
+/// the length in bytes of the encoded value.
+inline unsigned encodeULEB128(uint64_t Value, uint8_t *p,
+ unsigned PadTo = 0) {
+ uint8_t *orig_p = p;
+ unsigned Count = 0;
+ do {
+ uint8_t Byte = Value & 0x7f;
+ Value >>= 7;
+ Count++;
+ if (Value != 0 || Count < PadTo)
+ Byte |= 0x80; // Mark this byte to show that more bytes will follow.
+ *p++ = Byte;
+ } while (Value != 0);
+
+ // Pad with 0x80 and emit a null byte at the end.
+ if (Count < PadTo) {
+ for (; Count < PadTo - 1; ++Count)
+ *p++ = '\x80';
+ *p++ = '\x00';
+ }
+
+ return (unsigned)(p - orig_p);
+}
+
+/// Utility function to decode a ULEB128 value.
+inline uint64_t decodeULEB128(const uint8_t *p, unsigned *n = nullptr,
+ const uint8_t *end = nullptr,
+ const char **error = nullptr) {
+ const uint8_t *orig_p = p;
+ uint64_t Value = 0;
+ unsigned Shift = 0;
+ if (error)
+ *error = nullptr;
+ do {
+ if (end && p == end) {
+ if (error)
+ *error = "malformed uleb128, extends past end";
+ if (n)
+ *n = (unsigned)(p - orig_p);
+ return 0;
+ }
+ uint64_t Slice = *p & 0x7f;
+ if (Shift >= 64 || Slice << Shift >> Shift != Slice) {
+ if (error)
+ *error = "uleb128 too big for uint64";
+ if (n)
+ *n = (unsigned)(p - orig_p);
+ return 0;
+ }
+ Value += uint64_t(*p & 0x7f) << Shift;
+ Shift += 7;
+ } while (*p++ >= 128);
+ if (n)
+ *n = (unsigned)(p - orig_p);
+ return Value;
+}
+
+/// Utility function to decode a SLEB128 value.
+inline int64_t decodeSLEB128(const uint8_t *p, unsigned *n = nullptr,
+ const uint8_t *end = nullptr,
+ const char **error = nullptr) {
+ const uint8_t *orig_p = p;
+ int64_t Value = 0;
+ unsigned Shift = 0;
+ uint8_t Byte;
+ if (error)
+ *error = nullptr;
+ do {
+ if (end && p == end) {
+ if (error)
+ *error = "malformed sleb128, extends past end";
+ if (n)
+ *n = (unsigned)(p - orig_p);
+ return 0;
+ }
+ Byte = *p++;
+ Value |= (uint64_t(Byte & 0x7f) << Shift);
+ Shift += 7;
+ } while (Byte >= 128);
+ // Sign extend negative numbers if needed.
+ if (Shift < 64 && (Byte & 0x40))
+ Value |= (-1ULL) << Shift;
+ if (n)
+ *n = (unsigned)(p - orig_p);
+ return Value;
+}
+
+/// Utility function to get the size of the ULEB128-encoded value.
+extern unsigned getULEB128Size(uint64_t Value);
+
+/// Utility function to get the size of the SLEB128-encoded value.
+extern unsigned getSLEB128Size(int64_t Value);
+
+} // namespace llvm
+
+#endif // LLVM_SYSTEM_LEB128_H
diff --git a/third_party/llvm-project/include/llvm/Support/LICENSE.TXT b/third_party/llvm-project/include/llvm/Support/LICENSE.TXT
new file mode 100644
index 000000000..3479b3fd7
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/LICENSE.TXT
@@ -0,0 +1,6 @@
+LLVM System Interface Library
+-------------------------------------------------------------------------------
+The LLVM System Interface Library is licensed under the Illinois Open Source
+License and has the following additional copyright:
+
+Copyright (C) 2004 eXtensible Systems, Inc.
diff --git a/third_party/llvm-project/include/llvm/Support/LineIterator.h b/third_party/llvm-project/include/llvm/Support/LineIterator.h
new file mode 100644
index 000000000..2a1e47bfe
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/LineIterator.h
@@ -0,0 +1,87 @@
+//===- LineIterator.h - Iterator to read a text buffer's lines --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_LINEITERATOR_H
+#define LLVM_SUPPORT_LINEITERATOR_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataTypes.h"
+#include <iterator>
+
+namespace llvm {
+
+class MemoryBuffer;
+
+/// A forward iterator which reads text lines from a buffer.
+///
+/// This class provides a forward iterator interface for reading one line at
+/// a time from a buffer. When default constructed the iterator will be the
+/// "end" iterator.
+///
+/// The iterator is aware of what line number it is currently processing. It
+/// strips blank lines by default, and comment lines given a comment-starting
+/// character.
+///
+/// Note that this iterator requires the buffer to be nul terminated.
+class line_iterator
+ : public std::iterator<std::forward_iterator_tag, StringRef> {
+ const MemoryBuffer *Buffer = nullptr;
+ char CommentMarker = '\0';
+ bool SkipBlanks = true;
+
+ unsigned LineNumber = 1;
+ StringRef CurrentLine;
+
+public:
+ /// Default construct an "end" iterator.
+ line_iterator() = default;
+
+ /// Construct a new iterator around some memory buffer.
+ explicit line_iterator(const MemoryBuffer &Buffer, bool SkipBlanks = true,
+ char CommentMarker = '\0');
+
+ /// Return true if we've reached EOF or are an "end" iterator.
+ bool is_at_eof() const { return !Buffer; }
+
+ /// Return true if we're an "end" iterator or have reached EOF.
+ bool is_at_end() const { return is_at_eof(); }
+
+ /// Return the current line number. May return any number at EOF.
+ int64_t line_number() const { return LineNumber; }
+
+ /// Advance to the next (non-empty, non-comment) line.
+ line_iterator &operator++() {
+ advance();
+ return *this;
+ }
+ line_iterator operator++(int) {
+ line_iterator tmp(*this);
+ advance();
+ return tmp;
+ }
+
+ /// Get the current line as a \c StringRef.
+ StringRef operator*() const { return CurrentLine; }
+ const StringRef *operator->() const { return &CurrentLine; }
+
+ friend bool operator==(const line_iterator &LHS, const line_iterator &RHS) {
+ return LHS.Buffer == RHS.Buffer &&
+ LHS.CurrentLine.begin() == RHS.CurrentLine.begin();
+ }
+
+ friend bool operator!=(const line_iterator &LHS, const line_iterator &RHS) {
+ return !(LHS == RHS);
+ }
+
+private:
+ /// Advance the iterator to the next line.
+ void advance();
+};
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Locale.h b/third_party/llvm-project/include/llvm/Support/Locale.h
new file mode 100644
index 000000000..f7a2c036e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Locale.h
@@ -0,0 +1,17 @@
+#ifndef LLVM_SUPPORT_LOCALE_H
+#define LLVM_SUPPORT_LOCALE_H
+
+namespace llvm {
+class StringRef;
+
+namespace sys {
+namespace locale {
+
+int columnWidth(StringRef s);
+bool isPrint(int c);
+
+}
+}
+}
+
+#endif // LLVM_SUPPORT_LOCALE_H
diff --git a/third_party/llvm-project/include/llvm/Support/MD5.h b/third_party/llvm-project/include/llvm/Support/MD5.h
new file mode 100644
index 000000000..bb2bdbf1b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/MD5.h
@@ -0,0 +1,122 @@
+/* -*- C++ -*-
+ * This code is derived from (original license follows):
+ *
+ * This is an OpenSSL-compatible implementation of the RSA Data Security, Inc.
+ * MD5 Message-Digest Algorithm (RFC 1321).
+ *
+ * Homepage:
+ * http://openwall.info/wiki/people/solar/software/public-domain-source-code/md5
+ *
+ * Author:
+ * Alexander Peslyak, better known as Solar Designer <solar at openwall.com>
+ *
+ * This software was written by Alexander Peslyak in 2001. No copyright is
+ * claimed, and the software is hereby placed in the public domain.
+ * In case this attempt to disclaim copyright and place the software in the
+ * public domain is deemed null and void, then the software is
+ * Copyright (c) 2001 Alexander Peslyak and it is hereby released to the
+ * general public under the following terms:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * There's ABSOLUTELY NO WARRANTY, express or implied.
+ *
+ * See md5.c for more information.
+ */
+
+#ifndef LLVM_SUPPORT_MD5_H
+#define LLVM_SUPPORT_MD5_H
+
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Endian.h"
+#include <array>
+#include <cstdint>
+
+namespace llvm {
+
+template <typename T> class ArrayRef;
+
+class MD5 {
+ // Any 32-bit or wider unsigned integer data type will do.
+ typedef uint32_t MD5_u32plus;
+
+ MD5_u32plus a = 0x67452301;
+ MD5_u32plus b = 0xefcdab89;
+ MD5_u32plus c = 0x98badcfe;
+ MD5_u32plus d = 0x10325476;
+ MD5_u32plus hi = 0;
+ MD5_u32plus lo = 0;
+ uint8_t buffer[64];
+ MD5_u32plus block[16];
+
+public:
+ struct MD5Result {
+ std::array<uint8_t, 16> Bytes;
+
+ operator std::array<uint8_t, 16>() const { return Bytes; }
+
+ const uint8_t &operator[](size_t I) const { return Bytes[I]; }
+ uint8_t &operator[](size_t I) { return Bytes[I]; }
+
+ SmallString<32> digest() const;
+
+ uint64_t low() const {
+ // Our MD5 implementation returns the result in little endian, so the low
+ // word is first.
+ using namespace support;
+ return endian::read<uint64_t, little, unaligned>(Bytes.data());
+ }
+
+ uint64_t high() const {
+ using namespace support;
+ return endian::read<uint64_t, little, unaligned>(Bytes.data() + 8);
+ }
+ std::pair<uint64_t, uint64_t> words() const {
+ using namespace support;
+ return std::make_pair(high(), low());
+ }
+ };
+
+ MD5();
+
+ /// Updates the hash for the byte stream provided.
+ void update(ArrayRef<uint8_t> Data);
+
+ /// Updates the hash for the StringRef provided.
+ void update(StringRef Str);
+
+ /// Finishes off the hash and puts the result in result.
+ void final(MD5Result &Result);
+
+ /// Translates the bytes in \p Res to a hex string that is
+ /// deposited into \p Str. The result will be of length 32.
+ static void stringifyResult(MD5Result &Result, SmallString<32> &Str);
+
+ /// Computes the hash for a given bytes.
+ static std::array<uint8_t, 16> hash(ArrayRef<uint8_t> Data);
+
+private:
+ const uint8_t *body(ArrayRef<uint8_t> Data);
+};
+
+inline bool operator==(const MD5::MD5Result &LHS, const MD5::MD5Result &RHS) {
+ return LHS.Bytes == RHS.Bytes;
+}
+
+/// Helper to compute and return lower 64 bits of the given string's MD5 hash.
+inline uint64_t MD5Hash(StringRef Str) {
+ using namespace support;
+
+ MD5 Hash;
+ Hash.update(Str);
+ MD5::MD5Result Result;
+ Hash.final(Result);
+ // Return the least significant word.
+ return Result.low();
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_MD5_H
diff --git a/third_party/llvm-project/include/llvm/Support/ManagedStatic.h b/third_party/llvm-project/include/llvm/Support/ManagedStatic.h
new file mode 100644
index 000000000..4eb30f548
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ManagedStatic.h
@@ -0,0 +1 @@
+// XXX BINARYEN - we don't need this, but stuff imports it
diff --git a/third_party/llvm-project/include/llvm/Support/MathExtras.h b/third_party/llvm-project/include/llvm/Support/MathExtras.h
new file mode 100644
index 000000000..004a6f5f6
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/MathExtras.h
@@ -0,0 +1,951 @@
+//===-- llvm/Support/MathExtras.h - Useful math functions -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains some functions that are useful for math stuff.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_MATHEXTRAS_H
+#define LLVM_SUPPORT_MATHEXTRAS_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/SwapByteOrder.h"
+#include <algorithm>
+#include <cassert>
+#include <climits>
+#include <cstring>
+#include <limits>
+#include <type_traits>
+
+#ifdef __ANDROID_NDK__
+#include <android/api-level.h>
+#endif
+
+#ifdef _MSC_VER
+// Declare these intrinsics manually rather including intrin.h. It's very
+// expensive, and MathExtras.h is popular.
+// #include <intrin.h>
+extern "C" {
+unsigned char _BitScanForward(unsigned long *_Index, unsigned long _Mask);
+unsigned char _BitScanForward64(unsigned long *_Index, unsigned __int64 _Mask);
+unsigned char _BitScanReverse(unsigned long *_Index, unsigned long _Mask);
+unsigned char _BitScanReverse64(unsigned long *_Index, unsigned __int64 _Mask);
+}
+#endif
+
+namespace llvm {
+
+/// The behavior an operation has on an input of 0.
+enum ZeroBehavior {
+ /// The returned value is undefined.
+ ZB_Undefined,
+ /// The returned value is numeric_limits<T>::max()
+ ZB_Max,
+ /// The returned value is numeric_limits<T>::digits
+ ZB_Width
+};
+
+/// Mathematical constants.
+namespace numbers {
+// TODO: Track C++20 std::numbers.
+// TODO: Favor using the hexadecimal FP constants (requires C++17).
+constexpr double e = 2.7182818284590452354, // (0x1.5bf0a8b145749P+1) https://oeis.org/A001113
+ egamma = .57721566490153286061, // (0x1.2788cfc6fb619P-1) https://oeis.org/A001620
+ ln2 = .69314718055994530942, // (0x1.62e42fefa39efP-1) https://oeis.org/A002162
+ ln10 = 2.3025850929940456840, // (0x1.24bb1bbb55516P+1) https://oeis.org/A002392
+ log2e = 1.4426950408889634074, // (0x1.71547652b82feP+0)
+ log10e = .43429448190325182765, // (0x1.bcb7b1526e50eP-2)
+ pi = 3.1415926535897932385, // (0x1.921fb54442d18P+1) https://oeis.org/A000796
+ inv_pi = .31830988618379067154, // (0x1.45f306bc9c883P-2) https://oeis.org/A049541
+ sqrtpi = 1.7724538509055160273, // (0x1.c5bf891b4ef6bP+0) https://oeis.org/A002161
+ inv_sqrtpi = .56418958354775628695, // (0x1.20dd750429b6dP-1) https://oeis.org/A087197
+ sqrt2 = 1.4142135623730950488, // (0x1.6a09e667f3bcdP+0) https://oeis.org/A00219
+ inv_sqrt2 = .70710678118654752440, // (0x1.6a09e667f3bcdP-1)
+ sqrt3 = 1.7320508075688772935, // (0x1.bb67ae8584caaP+0) https://oeis.org/A002194
+ inv_sqrt3 = .57735026918962576451, // (0x1.279a74590331cP-1)
+ phi = 1.6180339887498948482; // (0x1.9e3779b97f4a8P+0) https://oeis.org/A001622
+constexpr float ef = 2.71828183F, // (0x1.5bf0a8P+1) https://oeis.org/A001113
+ egammaf = .577215665F, // (0x1.2788d0P-1) https://oeis.org/A001620
+ ln2f = .693147181F, // (0x1.62e430P-1) https://oeis.org/A002162
+ ln10f = 2.30258509F, // (0x1.26bb1cP+1) https://oeis.org/A002392
+ log2ef = 1.44269504F, // (0x1.715476P+0)
+ log10ef = .434294482F, // (0x1.bcb7b2P-2)
+ pif = 3.14159265F, // (0x1.921fb6P+1) https://oeis.org/A000796
+ inv_pif = .318309886F, // (0x1.45f306P-2) https://oeis.org/A049541
+ sqrtpif = 1.77245385F, // (0x1.c5bf8aP+0) https://oeis.org/A002161
+ inv_sqrtpif = .564189584F, // (0x1.20dd76P-1) https://oeis.org/A087197
+ sqrt2f = 1.41421356F, // (0x1.6a09e6P+0) https://oeis.org/A002193
+ inv_sqrt2f = .707106781F, // (0x1.6a09e6P-1)
+ sqrt3f = 1.73205081F, // (0x1.bb67aeP+0) https://oeis.org/A002194
+ inv_sqrt3f = .577350269F, // (0x1.279a74P-1)
+ phif = 1.61803399F; // (0x1.9e377aP+0) https://oeis.org/A001622
+} // namespace numbers
+
+namespace detail {
+template <typename T, std::size_t SizeOfT> struct TrailingZerosCounter {
+ static unsigned count(T Val, ZeroBehavior) {
+ if (!Val)
+ return std::numeric_limits<T>::digits;
+ if (Val & 0x1)
+ return 0;
+
+ // Bisection method.
+ unsigned ZeroBits = 0;
+ T Shift = std::numeric_limits<T>::digits >> 1;
+ T Mask = std::numeric_limits<T>::max() >> Shift;
+ while (Shift) {
+ if ((Val & Mask) == 0) {
+ Val >>= Shift;
+ ZeroBits |= Shift;
+ }
+ Shift >>= 1;
+ Mask >>= Shift;
+ }
+ return ZeroBits;
+ }
+};
+
+#if defined(__GNUC__) || defined(_MSC_VER)
+template <typename T> struct TrailingZerosCounter<T, 4> {
+ static unsigned count(T Val, ZeroBehavior ZB) {
+ if (ZB != ZB_Undefined && Val == 0)
+ return 32;
+
+#if __has_builtin(__builtin_ctz) || defined(__GNUC__)
+ return __builtin_ctz(Val);
+#elif defined(_MSC_VER)
+ unsigned long Index;
+ _BitScanForward(&Index, Val);
+ return Index;
+#endif
+ }
+};
+
+#if !defined(_MSC_VER) || defined(_M_X64)
+template <typename T> struct TrailingZerosCounter<T, 8> {
+ static unsigned count(T Val, ZeroBehavior ZB) {
+ if (ZB != ZB_Undefined && Val == 0)
+ return 64;
+
+#if __has_builtin(__builtin_ctzll) || defined(__GNUC__)
+ return __builtin_ctzll(Val);
+#elif defined(_MSC_VER)
+ unsigned long Index;
+ _BitScanForward64(&Index, Val);
+ return Index;
+#endif
+ }
+};
+#endif
+#endif
+} // namespace detail
+
+/// Count number of 0's from the least significant bit to the most
+/// stopping at the first 1.
+///
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of 0. Only ZB_Width and ZB_Undefined are
+/// valid arguments.
+template <typename T>
+unsigned countTrailingZeros(T Val, ZeroBehavior ZB = ZB_Width) {
+ static_assert(std::numeric_limits<T>::is_integer &&
+ !std::numeric_limits<T>::is_signed,
+ "Only unsigned integral types are allowed.");
+ return llvm::detail::TrailingZerosCounter<T, sizeof(T)>::count(Val, ZB);
+}
+
+namespace detail {
+template <typename T, std::size_t SizeOfT> struct LeadingZerosCounter {
+ static unsigned count(T Val, ZeroBehavior) {
+ if (!Val)
+ return std::numeric_limits<T>::digits;
+
+ // Bisection method.
+ unsigned ZeroBits = 0;
+ for (T Shift = std::numeric_limits<T>::digits >> 1; Shift; Shift >>= 1) {
+ T Tmp = Val >> Shift;
+ if (Tmp)
+ Val = Tmp;
+ else
+ ZeroBits |= Shift;
+ }
+ return ZeroBits;
+ }
+};
+
+#if defined(__GNUC__) || defined(_MSC_VER)
+template <typename T> struct LeadingZerosCounter<T, 4> {
+ static unsigned count(T Val, ZeroBehavior ZB) {
+ if (ZB != ZB_Undefined && Val == 0)
+ return 32;
+
+#if __has_builtin(__builtin_clz) || defined(__GNUC__)
+ return __builtin_clz(Val);
+#elif defined(_MSC_VER)
+ unsigned long Index;
+ _BitScanReverse(&Index, Val);
+ return Index ^ 31;
+#endif
+ }
+};
+
+#if !defined(_MSC_VER) || defined(_M_X64)
+template <typename T> struct LeadingZerosCounter<T, 8> {
+ static unsigned count(T Val, ZeroBehavior ZB) {
+ if (ZB != ZB_Undefined && Val == 0)
+ return 64;
+
+#if __has_builtin(__builtin_clzll) || defined(__GNUC__)
+ return __builtin_clzll(Val);
+#elif defined(_MSC_VER)
+ unsigned long Index;
+ _BitScanReverse64(&Index, Val);
+ return Index ^ 63;
+#endif
+ }
+};
+#endif
+#endif
+} // namespace detail
+
+/// Count number of 0's from the most significant bit to the least
+/// stopping at the first 1.
+///
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of 0. Only ZB_Width and ZB_Undefined are
+/// valid arguments.
+template <typename T>
+unsigned countLeadingZeros(T Val, ZeroBehavior ZB = ZB_Width) {
+ static_assert(std::numeric_limits<T>::is_integer &&
+ !std::numeric_limits<T>::is_signed,
+ "Only unsigned integral types are allowed.");
+ return llvm::detail::LeadingZerosCounter<T, sizeof(T)>::count(Val, ZB);
+}
+
+/// Get the index of the first set bit starting from the least
+/// significant bit.
+///
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of 0. Only ZB_Max and ZB_Undefined are
+/// valid arguments.
+template <typename T> T findFirstSet(T Val, ZeroBehavior ZB = ZB_Max) {
+ if (ZB == ZB_Max && Val == 0)
+ return std::numeric_limits<T>::max();
+
+ return countTrailingZeros(Val, ZB_Undefined);
+}
+
+/// Create a bitmask with the N right-most bits set to 1, and all other
+/// bits set to 0. Only unsigned types are allowed.
+template <typename T> T maskTrailingOnes(unsigned N) {
+ static_assert(std::is_unsigned<T>::value, "Invalid type!");
+ const unsigned Bits = CHAR_BIT * sizeof(T);
+ assert(N <= Bits && "Invalid bit index");
+ return N == 0 ? 0 : (T(-1) >> (Bits - N));
+}
+
+/// Create a bitmask with the N left-most bits set to 1, and all other
+/// bits set to 0. Only unsigned types are allowed.
+template <typename T> T maskLeadingOnes(unsigned N) {
+ return ~maskTrailingOnes<T>(CHAR_BIT * sizeof(T) - N);
+}
+
+/// Create a bitmask with the N right-most bits set to 0, and all other
+/// bits set to 1. Only unsigned types are allowed.
+template <typename T> T maskTrailingZeros(unsigned N) {
+ return maskLeadingOnes<T>(CHAR_BIT * sizeof(T) - N);
+}
+
+/// Create a bitmask with the N left-most bits set to 0, and all other
+/// bits set to 1. Only unsigned types are allowed.
+template <typename T> T maskLeadingZeros(unsigned N) {
+ return maskTrailingOnes<T>(CHAR_BIT * sizeof(T) - N);
+}
+
+/// Get the index of the last set bit starting from the least
+/// significant bit.
+///
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of 0. Only ZB_Max and ZB_Undefined are
+/// valid arguments.
+template <typename T> T findLastSet(T Val, ZeroBehavior ZB = ZB_Max) {
+ if (ZB == ZB_Max && Val == 0)
+ return std::numeric_limits<T>::max();
+
+ // Use ^ instead of - because both gcc and llvm can remove the associated ^
+ // in the __builtin_clz intrinsic on x86.
+ return countLeadingZeros(Val, ZB_Undefined) ^
+ (std::numeric_limits<T>::digits - 1);
+}
+
+/// Macro compressed bit reversal table for 256 bits.
+///
+/// http://graphics.stanford.edu/~seander/bithacks.html#BitReverseTable
+static const unsigned char BitReverseTable256[256] = {
+#define R2(n) n, n + 2 * 64, n + 1 * 64, n + 3 * 64
+#define R4(n) R2(n), R2(n + 2 * 16), R2(n + 1 * 16), R2(n + 3 * 16)
+#define R6(n) R4(n), R4(n + 2 * 4), R4(n + 1 * 4), R4(n + 3 * 4)
+ R6(0), R6(2), R6(1), R6(3)
+#undef R2
+#undef R4
+#undef R6
+};
+
+/// Reverse the bits in \p Val.
+template <typename T>
+T reverseBits(T Val) {
+ unsigned char in[sizeof(Val)];
+ unsigned char out[sizeof(Val)];
+ std::memcpy(in, &Val, sizeof(Val));
+ for (unsigned i = 0; i < sizeof(Val); ++i)
+ out[(sizeof(Val) - i) - 1] = BitReverseTable256[in[i]];
+ std::memcpy(&Val, out, sizeof(Val));
+ return Val;
+}
+
+// NOTE: The following support functions use the _32/_64 extensions instead of
+// type overloading so that signed and unsigned integers can be used without
+// ambiguity.
+
+/// Return the high 32 bits of a 64 bit value.
+constexpr inline uint32_t Hi_32(uint64_t Value) {
+ return static_cast<uint32_t>(Value >> 32);
+}
+
+/// Return the low 32 bits of a 64 bit value.
+constexpr inline uint32_t Lo_32(uint64_t Value) {
+ return static_cast<uint32_t>(Value);
+}
+
+/// Make a 64-bit integer from a high / low pair of 32-bit integers.
+constexpr inline uint64_t Make_64(uint32_t High, uint32_t Low) {
+ return ((uint64_t)High << 32) | (uint64_t)Low;
+}
+
+/// Checks if an integer fits into the given bit width.
+template <unsigned N> constexpr inline bool isInt(int64_t x) {
+ return N >= 64 || (-(INT64_C(1)<<(N-1)) <= x && x < (INT64_C(1)<<(N-1)));
+}
+// Template specializations to get better code for common cases.
+template <> constexpr inline bool isInt<8>(int64_t x) {
+ return static_cast<int8_t>(x) == x;
+}
+template <> constexpr inline bool isInt<16>(int64_t x) {
+ return static_cast<int16_t>(x) == x;
+}
+template <> constexpr inline bool isInt<32>(int64_t x) {
+ return static_cast<int32_t>(x) == x;
+}
+
+/// Checks if a signed integer is an N bit number shifted left by S.
+template <unsigned N, unsigned S>
+constexpr inline bool isShiftedInt(int64_t x) {
+ static_assert(
+ N > 0, "isShiftedInt<0> doesn't make sense (refers to a 0-bit number.");
+ static_assert(N + S <= 64, "isShiftedInt<N, S> with N + S > 64 is too wide.");
+ return isInt<N + S>(x) && (x % (UINT64_C(1) << S) == 0);
+}
+
+/// Checks if an unsigned integer fits into the given bit width.
+///
+/// This is written as two functions rather than as simply
+///
+/// return N >= 64 || X < (UINT64_C(1) << N);
+///
+/// to keep MSVC from (incorrectly) warning on isUInt<64> that we're shifting
+/// left too many places.
+template <unsigned N>
+constexpr inline typename std::enable_if<(N < 64), bool>::type
+isUInt(uint64_t X) {
+ static_assert(N > 0, "isUInt<0> doesn't make sense");
+ return X < (UINT64_C(1) << (N));
+}
+template <unsigned N>
+constexpr inline typename std::enable_if<N >= 64, bool>::type
+isUInt(uint64_t X) {
+ return true;
+}
+
+// Template specializations to get better code for common cases.
+template <> constexpr inline bool isUInt<8>(uint64_t x) {
+ return static_cast<uint8_t>(x) == x;
+}
+template <> constexpr inline bool isUInt<16>(uint64_t x) {
+ return static_cast<uint16_t>(x) == x;
+}
+template <> constexpr inline bool isUInt<32>(uint64_t x) {
+ return static_cast<uint32_t>(x) == x;
+}
+
+/// Checks if a unsigned integer is an N bit number shifted left by S.
+template <unsigned N, unsigned S>
+constexpr inline bool isShiftedUInt(uint64_t x) {
+ static_assert(
+ N > 0, "isShiftedUInt<0> doesn't make sense (refers to a 0-bit number)");
+ static_assert(N + S <= 64,
+ "isShiftedUInt<N, S> with N + S > 64 is too wide.");
+ // Per the two static_asserts above, S must be strictly less than 64. So
+ // 1 << S is not undefined behavior.
+ return isUInt<N + S>(x) && (x % (UINT64_C(1) << S) == 0);
+}
+
+/// Gets the maximum value for a N-bit unsigned integer.
+inline uint64_t maxUIntN(uint64_t N) {
+ assert(N > 0 && N <= 64 && "integer width out of range");
+
+ // uint64_t(1) << 64 is undefined behavior, so we can't do
+ // (uint64_t(1) << N) - 1
+ // without checking first that N != 64. But this works and doesn't have a
+ // branch.
+ return UINT64_MAX >> (64 - N);
+}
+
+/// Gets the minimum value for a N-bit signed integer.
+inline int64_t minIntN(int64_t N) {
+ assert(N > 0 && N <= 64 && "integer width out of range");
+
+ return -(UINT64_C(1)<<(N-1));
+}
+
+/// Gets the maximum value for a N-bit signed integer.
+inline int64_t maxIntN(int64_t N) {
+ assert(N > 0 && N <= 64 && "integer width out of range");
+
+ // This relies on two's complement wraparound when N == 64, so we convert to
+ // int64_t only at the very end to avoid UB.
+ return (UINT64_C(1) << (N - 1)) - 1;
+}
+
+/// Checks if an unsigned integer fits into the given (dynamic) bit width.
+inline bool isUIntN(unsigned N, uint64_t x) {
+ return N >= 64 || x <= maxUIntN(N);
+}
+
+/// Checks if an signed integer fits into the given (dynamic) bit width.
+inline bool isIntN(unsigned N, int64_t x) {
+ return N >= 64 || (minIntN(N) <= x && x <= maxIntN(N));
+}
+
+/// Return true if the argument is a non-empty sequence of ones starting at the
+/// least significant bit with the remainder zero (32 bit version).
+/// Ex. isMask_32(0x0000FFFFU) == true.
+constexpr inline bool isMask_32(uint32_t Value) {
+ return Value && ((Value + 1) & Value) == 0;
+}
+
+/// Return true if the argument is a non-empty sequence of ones starting at the
+/// least significant bit with the remainder zero (64 bit version).
+constexpr inline bool isMask_64(uint64_t Value) {
+ return Value && ((Value + 1) & Value) == 0;
+}
+
+/// Return true if the argument contains a non-empty sequence of ones with the
+/// remainder zero (32 bit version.) Ex. isShiftedMask_32(0x0000FF00U) == true.
+constexpr inline bool isShiftedMask_32(uint32_t Value) {
+ return Value && isMask_32((Value - 1) | Value);
+}
+
+/// Return true if the argument contains a non-empty sequence of ones with the
+/// remainder zero (64 bit version.)
+constexpr inline bool isShiftedMask_64(uint64_t Value) {
+ return Value && isMask_64((Value - 1) | Value);
+}
+
+/// Return true if the argument is a power of two > 0.
+/// Ex. isPowerOf2_32(0x00100000U) == true (32 bit edition.)
+constexpr inline bool isPowerOf2_32(uint32_t Value) {
+ return Value && !(Value & (Value - 1));
+}
+
+/// Return true if the argument is a power of two > 0 (64 bit edition.)
+constexpr inline bool isPowerOf2_64(uint64_t Value) {
+ return Value && !(Value & (Value - 1));
+}
+
+/// Return a byte-swapped representation of the 16-bit argument.
+inline uint16_t ByteSwap_16(uint16_t Value) {
+ return sys::SwapByteOrder_16(Value);
+}
+
+/// Return a byte-swapped representation of the 32-bit argument.
+inline uint32_t ByteSwap_32(uint32_t Value) {
+ return sys::SwapByteOrder_32(Value);
+}
+
+/// Return a byte-swapped representation of the 64-bit argument.
+inline uint64_t ByteSwap_64(uint64_t Value) {
+ return sys::SwapByteOrder_64(Value);
+}
+
+/// Count the number of ones from the most significant bit to the first
+/// zero bit.
+///
+/// Ex. countLeadingOnes(0xFF0FFF00) == 8.
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of all ones. Only ZB_Width and
+/// ZB_Undefined are valid arguments.
+template <typename T>
+unsigned countLeadingOnes(T Value, ZeroBehavior ZB = ZB_Width) {
+ static_assert(std::numeric_limits<T>::is_integer &&
+ !std::numeric_limits<T>::is_signed,
+ "Only unsigned integral types are allowed.");
+ return countLeadingZeros<T>(~Value, ZB);
+}
+
+/// Count the number of ones from the least significant bit to the first
+/// zero bit.
+///
+/// Ex. countTrailingOnes(0x00FF00FF) == 8.
+/// Only unsigned integral types are allowed.
+///
+/// \param ZB the behavior on an input of all ones. Only ZB_Width and
+/// ZB_Undefined are valid arguments.
+template <typename T>
+unsigned countTrailingOnes(T Value, ZeroBehavior ZB = ZB_Width) {
+ static_assert(std::numeric_limits<T>::is_integer &&
+ !std::numeric_limits<T>::is_signed,
+ "Only unsigned integral types are allowed.");
+ return countTrailingZeros<T>(~Value, ZB);
+}
+
+namespace detail {
+template <typename T, std::size_t SizeOfT> struct PopulationCounter {
+ static unsigned count(T Value) {
+ // Generic version, forward to 32 bits.
+ static_assert(SizeOfT <= 4, "Not implemented!");
+#if defined(__GNUC__)
+ return __builtin_popcount(Value);
+#else
+ uint32_t v = Value;
+ v = v - ((v >> 1) & 0x55555555);
+ v = (v & 0x33333333) + ((v >> 2) & 0x33333333);
+ return ((v + (v >> 4) & 0xF0F0F0F) * 0x1010101) >> 24;
+#endif
+ }
+};
+
+template <typename T> struct PopulationCounter<T, 8> {
+ static unsigned count(T Value) {
+#if defined(__GNUC__)
+ return __builtin_popcountll(Value);
+#else
+ uint64_t v = Value;
+ v = v - ((v >> 1) & 0x5555555555555555ULL);
+ v = (v & 0x3333333333333333ULL) + ((v >> 2) & 0x3333333333333333ULL);
+ v = (v + (v >> 4)) & 0x0F0F0F0F0F0F0F0FULL;
+ return unsigned((uint64_t)(v * 0x0101010101010101ULL) >> 56);
+#endif
+ }
+};
+} // namespace detail
+
+/// Count the number of set bits in a value.
+/// Ex. countPopulation(0xF000F000) = 8
+/// Returns 0 if the word is zero.
+template <typename T>
+inline unsigned countPopulation(T Value) {
+ static_assert(std::numeric_limits<T>::is_integer &&
+ !std::numeric_limits<T>::is_signed,
+ "Only unsigned integral types are allowed.");
+ return detail::PopulationCounter<T, sizeof(T)>::count(Value);
+}
+
+/// Compile time Log2.
+/// Valid only for positive powers of two.
+template <size_t kValue> constexpr inline size_t CTLog2() {
+ static_assert(kValue > 0 && llvm::isPowerOf2_64(kValue),
+ "Value is not a valid power of 2");
+ return 1 + CTLog2<kValue / 2>();
+}
+
+template <> constexpr inline size_t CTLog2<1>() { return 0; }
+
+/// Return the log base 2 of the specified value.
+inline double Log2(double Value) {
+#if defined(__ANDROID_API__) && __ANDROID_API__ < 18
+ return __builtin_log(Value) / __builtin_log(2.0);
+#else
+ return log2(Value);
+#endif
+}
+
+/// Return the floor log base 2 of the specified value, -1 if the value is zero.
+/// (32 bit edition.)
+/// Ex. Log2_32(32) == 5, Log2_32(1) == 0, Log2_32(0) == -1, Log2_32(6) == 2
+inline unsigned Log2_32(uint32_t Value) {
+ return 31 - countLeadingZeros(Value);
+}
+
+/// Return the floor log base 2 of the specified value, -1 if the value is zero.
+/// (64 bit edition.)
+inline unsigned Log2_64(uint64_t Value) {
+ return 63 - countLeadingZeros(Value);
+}
+
+/// Return the ceil log base 2 of the specified value, 32 if the value is zero.
+/// (32 bit edition).
+/// Ex. Log2_32_Ceil(32) == 5, Log2_32_Ceil(1) == 0, Log2_32_Ceil(6) == 3
+inline unsigned Log2_32_Ceil(uint32_t Value) {
+ return 32 - countLeadingZeros(Value - 1);
+}
+
+/// Return the ceil log base 2 of the specified value, 64 if the value is zero.
+/// (64 bit edition.)
+inline unsigned Log2_64_Ceil(uint64_t Value) {
+ return 64 - countLeadingZeros(Value - 1);
+}
+
+/// Return the greatest common divisor of the values using Euclid's algorithm.
+template <typename T>
+inline T greatestCommonDivisor(T A, T B) {
+ while (B) {
+ T Tmp = B;
+ B = A % B;
+ A = Tmp;
+ }
+ return A;
+}
+
+inline uint64_t GreatestCommonDivisor64(uint64_t A, uint64_t B) {
+ return greatestCommonDivisor<uint64_t>(A, B);
+}
+
+/// This function takes a 64-bit integer and returns the bit equivalent double.
+inline double BitsToDouble(uint64_t Bits) {
+ double D;
+ static_assert(sizeof(uint64_t) == sizeof(double), "Unexpected type sizes");
+ memcpy(&D, &Bits, sizeof(Bits));
+ return D;
+}
+
+/// This function takes a 32-bit integer and returns the bit equivalent float.
+inline float BitsToFloat(uint32_t Bits) {
+ float F;
+ static_assert(sizeof(uint32_t) == sizeof(float), "Unexpected type sizes");
+ memcpy(&F, &Bits, sizeof(Bits));
+ return F;
+}
+
+/// This function takes a double and returns the bit equivalent 64-bit integer.
+/// Note that copying doubles around changes the bits of NaNs on some hosts,
+/// notably x86, so this routine cannot be used if these bits are needed.
+inline uint64_t DoubleToBits(double Double) {
+ uint64_t Bits;
+ static_assert(sizeof(uint64_t) == sizeof(double), "Unexpected type sizes");
+ memcpy(&Bits, &Double, sizeof(Double));
+ return Bits;
+}
+
+/// This function takes a float and returns the bit equivalent 32-bit integer.
+/// Note that copying floats around changes the bits of NaNs on some hosts,
+/// notably x86, so this routine cannot be used if these bits are needed.
+inline uint32_t FloatToBits(float Float) {
+ uint32_t Bits;
+ static_assert(sizeof(uint32_t) == sizeof(float), "Unexpected type sizes");
+ memcpy(&Bits, &Float, sizeof(Float));
+ return Bits;
+}
+
+/// A and B are either alignments or offsets. Return the minimum alignment that
+/// may be assumed after adding the two together.
+constexpr inline uint64_t MinAlign(uint64_t A, uint64_t B) {
+ // The largest power of 2 that divides both A and B.
+ //
+ // Replace "-Value" by "1+~Value" in the following commented code to avoid
+ // MSVC warning C4146
+ // return (A | B) & -(A | B);
+ return (A | B) & (1 + ~(A | B));
+}
+
+/// Returns the next power of two (in 64-bits) that is strictly greater than A.
+/// Returns zero on overflow.
+inline uint64_t NextPowerOf2(uint64_t A) {
+ A |= (A >> 1);
+ A |= (A >> 2);
+ A |= (A >> 4);
+ A |= (A >> 8);
+ A |= (A >> 16);
+ A |= (A >> 32);
+ return A + 1;
+}
+
+/// Returns the power of two which is less than or equal to the given value.
+/// Essentially, it is a floor operation across the domain of powers of two.
+inline uint64_t PowerOf2Floor(uint64_t A) {
+ if (!A) return 0;
+ return 1ull << (63 - countLeadingZeros(A, ZB_Undefined));
+}
+
+/// Returns the power of two which is greater than or equal to the given value.
+/// Essentially, it is a ceil operation across the domain of powers of two.
+inline uint64_t PowerOf2Ceil(uint64_t A) {
+ if (!A)
+ return 0;
+ return NextPowerOf2(A - 1);
+}
+
+/// Returns the next integer (mod 2**64) that is greater than or equal to
+/// \p Value and is a multiple of \p Align. \p Align must be non-zero.
+///
+/// If non-zero \p Skew is specified, the return value will be a minimal
+/// integer that is greater than or equal to \p Value and equal to
+/// \p Align * N + \p Skew for some integer N. If \p Skew is larger than
+/// \p Align, its value is adjusted to '\p Skew mod \p Align'.
+///
+/// Examples:
+/// \code
+/// alignTo(5, 8) = 8
+/// alignTo(17, 8) = 24
+/// alignTo(~0LL, 8) = 0
+/// alignTo(321, 255) = 510
+///
+/// alignTo(5, 8, 7) = 7
+/// alignTo(17, 8, 1) = 17
+/// alignTo(~0LL, 8, 3) = 3
+/// alignTo(321, 255, 42) = 552
+/// \endcode
+inline uint64_t alignTo(uint64_t Value, uint64_t Align, uint64_t Skew = 0) {
+ assert(Align != 0u && "Align can't be 0.");
+ Skew %= Align;
+ return (Value + Align - 1 - Skew) / Align * Align + Skew;
+}
+
+/// Returns the next integer (mod 2**64) that is greater than or equal to
+/// \p Value and is a multiple of \c Align. \c Align must be non-zero.
+template <uint64_t Align> constexpr inline uint64_t alignTo(uint64_t Value) {
+ static_assert(Align != 0u, "Align must be non-zero");
+ return (Value + Align - 1) / Align * Align;
+}
+
+/// Returns the integer ceil(Numerator / Denominator).
+inline uint64_t divideCeil(uint64_t Numerator, uint64_t Denominator) {
+ return alignTo(Numerator, Denominator) / Denominator;
+}
+
+/// Returns the largest uint64_t less than or equal to \p Value and is
+/// \p Skew mod \p Align. \p Align must be non-zero
+inline uint64_t alignDown(uint64_t Value, uint64_t Align, uint64_t Skew = 0) {
+ assert(Align != 0u && "Align can't be 0.");
+ Skew %= Align;
+ return (Value - Skew) / Align * Align + Skew;
+}
+
+/// Sign-extend the number in the bottom B bits of X to a 32-bit integer.
+/// Requires 0 < B <= 32.
+template <unsigned B> constexpr inline int32_t SignExtend32(uint32_t X) {
+ static_assert(B > 0, "Bit width can't be 0.");
+ static_assert(B <= 32, "Bit width out of range.");
+ return int32_t(X << (32 - B)) >> (32 - B);
+}
+
+/// Sign-extend the number in the bottom B bits of X to a 32-bit integer.
+/// Requires 0 < B < 32.
+inline int32_t SignExtend32(uint32_t X, unsigned B) {
+ assert(B > 0 && "Bit width can't be 0.");
+ assert(B <= 32 && "Bit width out of range.");
+ return int32_t(X << (32 - B)) >> (32 - B);
+}
+
+/// Sign-extend the number in the bottom B bits of X to a 64-bit integer.
+/// Requires 0 < B < 64.
+template <unsigned B> constexpr inline int64_t SignExtend64(uint64_t x) {
+ static_assert(B > 0, "Bit width can't be 0.");
+ static_assert(B <= 64, "Bit width out of range.");
+ return int64_t(x << (64 - B)) >> (64 - B);
+}
+
+/// Sign-extend the number in the bottom B bits of X to a 64-bit integer.
+/// Requires 0 < B < 64.
+inline int64_t SignExtend64(uint64_t X, unsigned B) {
+ assert(B > 0 && "Bit width can't be 0.");
+ assert(B <= 64 && "Bit width out of range.");
+ return int64_t(X << (64 - B)) >> (64 - B);
+}
+
+/// Subtract two unsigned integers, X and Y, of type T and return the absolute
+/// value of the result.
+template <typename T>
+typename std::enable_if<std::is_unsigned<T>::value, T>::type
+AbsoluteDifference(T X, T Y) {
+ return std::max(X, Y) - std::min(X, Y);
+}
+
+/// Add two unsigned integers, X and Y, of type T. Clamp the result to the
+/// maximum representable value of T on overflow. ResultOverflowed indicates if
+/// the result is larger than the maximum representable value of type T.
+template <typename T>
+typename std::enable_if<std::is_unsigned<T>::value, T>::type
+SaturatingAdd(T X, T Y, bool *ResultOverflowed = nullptr) {
+ bool Dummy;
+ bool &Overflowed = ResultOverflowed ? *ResultOverflowed : Dummy;
+ // Hacker's Delight, p. 29
+ T Z = X + Y;
+ Overflowed = (Z < X || Z < Y);
+ if (Overflowed)
+ return std::numeric_limits<T>::max();
+ else
+ return Z;
+}
+
+/// Multiply two unsigned integers, X and Y, of type T. Clamp the result to the
+/// maximum representable value of T on overflow. ResultOverflowed indicates if
+/// the result is larger than the maximum representable value of type T.
+template <typename T>
+typename std::enable_if<std::is_unsigned<T>::value, T>::type
+SaturatingMultiply(T X, T Y, bool *ResultOverflowed = nullptr) {
+ bool Dummy;
+ bool &Overflowed = ResultOverflowed ? *ResultOverflowed : Dummy;
+
+ // Hacker's Delight, p. 30 has a different algorithm, but we don't use that
+ // because it fails for uint16_t (where multiplication can have undefined
+ // behavior due to promotion to int), and requires a division in addition
+ // to the multiplication.
+
+ Overflowed = false;
+
+ // Log2(Z) would be either Log2Z or Log2Z + 1.
+ // Special case: if X or Y is 0, Log2_64 gives -1, and Log2Z
+ // will necessarily be less than Log2Max as desired.
+ int Log2Z = Log2_64(X) + Log2_64(Y);
+ const T Max = std::numeric_limits<T>::max();
+ int Log2Max = Log2_64(Max);
+ if (Log2Z < Log2Max) {
+ return X * Y;
+ }
+ if (Log2Z > Log2Max) {
+ Overflowed = true;
+ return Max;
+ }
+
+ // We're going to use the top bit, and maybe overflow one
+ // bit past it. Multiply all but the bottom bit then add
+ // that on at the end.
+ T Z = (X >> 1) * Y;
+ if (Z & ~(Max >> 1)) {
+ Overflowed = true;
+ return Max;
+ }
+ Z <<= 1;
+ if (X & 1)
+ return SaturatingAdd(Z, Y, ResultOverflowed);
+
+ return Z;
+}
+
+/// Multiply two unsigned integers, X and Y, and add the unsigned integer, A to
+/// the product. Clamp the result to the maximum representable value of T on
+/// overflow. ResultOverflowed indicates if the result is larger than the
+/// maximum representable value of type T.
+template <typename T>
+typename std::enable_if<std::is_unsigned<T>::value, T>::type
+SaturatingMultiplyAdd(T X, T Y, T A, bool *ResultOverflowed = nullptr) {
+ bool Dummy;
+ bool &Overflowed = ResultOverflowed ? *ResultOverflowed : Dummy;
+
+ T Product = SaturatingMultiply(X, Y, &Overflowed);
+ if (Overflowed)
+ return Product;
+
+ return SaturatingAdd(A, Product, &Overflowed);
+}
+
+/// Use this rather than HUGE_VALF; the latter causes warnings on MSVC.
+extern const float huge_valf;
+
+
+/// Add two signed integers, computing the two's complement truncated result,
+/// returning true if overflow occured.
+template <typename T>
+typename std::enable_if<std::is_signed<T>::value, T>::type
+AddOverflow(T X, T Y, T &Result) {
+#if __has_builtin(__builtin_add_overflow)
+ return __builtin_add_overflow(X, Y, &Result);
+#else
+ // Perform the unsigned addition.
+ using U = typename std::make_unsigned<T>::type;
+ const U UX = static_cast<U>(X);
+ const U UY = static_cast<U>(Y);
+ const U UResult = UX + UY;
+
+ // Convert to signed.
+ Result = static_cast<T>(UResult);
+
+ // Adding two positive numbers should result in a positive number.
+ if (X > 0 && Y > 0)
+ return Result <= 0;
+ // Adding two negatives should result in a negative number.
+ if (X < 0 && Y < 0)
+ return Result >= 0;
+ return false;
+#endif
+}
+
+/// Subtract two signed integers, computing the two's complement truncated
+/// result, returning true if an overflow ocurred.
+template <typename T>
+typename std::enable_if<std::is_signed<T>::value, T>::type
+SubOverflow(T X, T Y, T &Result) {
+#if __has_builtin(__builtin_sub_overflow)
+ return __builtin_sub_overflow(X, Y, &Result);
+#else
+ // Perform the unsigned addition.
+ using U = typename std::make_unsigned<T>::type;
+ const U UX = static_cast<U>(X);
+ const U UY = static_cast<U>(Y);
+ const U UResult = UX - UY;
+
+ // Convert to signed.
+ Result = static_cast<T>(UResult);
+
+ // Subtracting a positive number from a negative results in a negative number.
+ if (X <= 0 && Y > 0)
+ return Result >= 0;
+ // Subtracting a negative number from a positive results in a positive number.
+ if (X >= 0 && Y < 0)
+ return Result <= 0;
+ return false;
+#endif
+}
+
+
+/// Multiply two signed integers, computing the two's complement truncated
+/// result, returning true if an overflow ocurred.
+template <typename T>
+typename std::enable_if<std::is_signed<T>::value, T>::type
+MulOverflow(T X, T Y, T &Result) {
+ // Perform the unsigned multiplication on absolute values.
+ using U = typename std::make_unsigned<T>::type;
+ const U UX = X < 0 ? (0 - static_cast<U>(X)) : static_cast<U>(X);
+ const U UY = Y < 0 ? (0 - static_cast<U>(Y)) : static_cast<U>(Y);
+ const U UResult = UX * UY;
+
+ // Convert to signed.
+ const bool IsNegative = (X < 0) ^ (Y < 0);
+ Result = IsNegative ? (0 - UResult) : UResult;
+
+ // If any of the args was 0, result is 0 and no overflow occurs.
+ if (UX == 0 || UY == 0)
+ return false;
+
+ // UX and UY are in [1, 2^n], where n is the number of digits.
+ // Check how the max allowed absolute value (2^n for negative, 2^(n-1) for
+ // positive) divided by an argument compares to the other.
+ if (IsNegative)
+ return UX > (static_cast<U>(std::numeric_limits<T>::max()) + U(1)) / UY;
+ else
+ return UX > (static_cast<U>(std::numeric_limits<T>::max())) / UY;
+}
+
+} // End llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/MemAlloc.h b/third_party/llvm-project/include/llvm/Support/MemAlloc.h
new file mode 100644
index 000000000..0e5869141
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/MemAlloc.h
@@ -0,0 +1,66 @@
+//===- MemAlloc.h - Memory allocation functions -----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file defines counterparts of C library allocation functions defined in
+/// the namespace 'std'. The new allocation functions crash on allocation
+/// failure instead of returning null pointer.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_MEMALLOC_H
+#define LLVM_SUPPORT_MEMALLOC_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <cstdlib>
+
+namespace llvm {
+
+LLVM_ATTRIBUTE_RETURNS_NONNULL inline void *safe_malloc(size_t Sz) {
+ void *Result = std::malloc(Sz);
+ if (Result == nullptr) {
+ // It is implementation-defined whether allocation occurs if the space
+ // requested is zero (ISO/IEC 9899:2018 7.22.3). Retry, requesting
+ // non-zero, if the space requested was zero.
+ if (Sz == 0)
+ return safe_malloc(1);
+ report_bad_alloc_error("Allocation failed");
+ }
+ return Result;
+}
+
+LLVM_ATTRIBUTE_RETURNS_NONNULL inline void *safe_calloc(size_t Count,
+ size_t Sz) {
+ void *Result = std::calloc(Count, Sz);
+ if (Result == nullptr) {
+ // It is implementation-defined whether allocation occurs if the space
+ // requested is zero (ISO/IEC 9899:2018 7.22.3). Retry, requesting
+ // non-zero, if the space requested was zero.
+ if (Count == 0 || Sz == 0)
+ return safe_malloc(1);
+ report_bad_alloc_error("Allocation failed");
+ }
+ return Result;
+}
+
+LLVM_ATTRIBUTE_RETURNS_NONNULL inline void *safe_realloc(void *Ptr, size_t Sz) {
+ void *Result = std::realloc(Ptr, Sz);
+ if (Result == nullptr) {
+ // It is implementation-defined whether allocation occurs if the space
+ // requested is zero (ISO/IEC 9899:2018 7.22.3). Retry, requesting
+ // non-zero, if the space requested was zero.
+ if (Sz == 0)
+ return safe_malloc(1);
+ report_bad_alloc_error("Allocation failed");
+ }
+ return Result;
+}
+
+}
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/MemoryBuffer.h b/third_party/llvm-project/include/llvm/Support/MemoryBuffer.h
new file mode 100644
index 000000000..b5196cd84
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/MemoryBuffer.h
@@ -0,0 +1,286 @@
+//===--- MemoryBuffer.h - Memory Buffer Interface ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the MemoryBuffer interface.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_MEMORYBUFFER_H
+#define LLVM_SUPPORT_MEMORYBUFFER_H
+
+#include "llvm-c/Types.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/CBindingWrapping.h"
+#include "llvm/Support/ErrorOr.h"
+#include "llvm/Support/FileSystem.h"
+#include <cstddef>
+#include <cstdint>
+#include <memory>
+
+namespace llvm {
+
+class MemoryBufferRef;
+
+/// This interface provides simple read-only access to a block of memory, and
+/// provides simple methods for reading files and standard input into a memory
+/// buffer. In addition to basic access to the characters in the file, this
+/// interface guarantees you can read one character past the end of the file,
+/// and that this character will read as '\0'.
+///
+/// The '\0' guarantee is needed to support an optimization -- it's intended to
+/// be more efficient for clients which are reading all the data to stop
+/// reading when they encounter a '\0' than to continually check the file
+/// position to see if it has reached the end of the file.
+class MemoryBuffer {
+ const char *BufferStart; // Start of the buffer.
+ const char *BufferEnd; // End of the buffer.
+
+protected:
+ MemoryBuffer() = default;
+
+ void init(const char *BufStart, const char *BufEnd,
+ bool RequiresNullTerminator);
+
+ static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
+ sys::fs::mapped_file_region::readonly;
+
+public:
+ MemoryBuffer(const MemoryBuffer &) = delete;
+ MemoryBuffer &operator=(const MemoryBuffer &) = delete;
+ virtual ~MemoryBuffer();
+
+ const char *getBufferStart() const { return BufferStart; }
+ const char *getBufferEnd() const { return BufferEnd; }
+ size_t getBufferSize() const { return BufferEnd-BufferStart; }
+
+ StringRef getBuffer() const {
+ return StringRef(BufferStart, getBufferSize());
+ }
+
+ /// Return an identifier for this buffer, typically the filename it was read
+ /// from.
+ virtual StringRef getBufferIdentifier() const { return "Unknown buffer"; }
+
+ /// Open the specified file as a MemoryBuffer, returning a new MemoryBuffer
+ /// if successful, otherwise returning null. If FileSize is specified, this
+ /// means that the client knows that the file exists and that it has the
+ /// specified size.
+ ///
+ /// \param IsVolatile Set to true to indicate that the contents of the file
+ /// can change outside the user's control, e.g. when libclang tries to parse
+ /// while the user is editing/updating the file or if the file is on an NFS.
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getFile(const Twine &Filename, int64_t FileSize = -1,
+ bool RequiresNullTerminator = true, bool IsVolatile = false);
+
+ /// Read all of the specified file into a MemoryBuffer as a stream
+ /// (i.e. until EOF reached). This is useful for special files that
+ /// look like a regular file but have 0 size (e.g. /proc/cpuinfo on Linux).
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getFileAsStream(const Twine &Filename);
+
+ /// Given an already-open file descriptor, map some slice of it into a
+ /// MemoryBuffer. The slice is specified by an \p Offset and \p MapSize.
+ /// Since this is in the middle of a file, the buffer is not null terminated.
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getOpenFileSlice(sys::fs::file_t FD, const Twine &Filename, uint64_t MapSize,
+ int64_t Offset, bool IsVolatile = false);
+
+ /// Given an already-open file descriptor, read the file and return a
+ /// MemoryBuffer.
+ ///
+ /// \param IsVolatile Set to true to indicate that the contents of the file
+ /// can change outside the user's control, e.g. when libclang tries to parse
+ /// while the user is editing/updating the file or if the file is on an NFS.
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getOpenFile(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
+ bool RequiresNullTerminator = true, bool IsVolatile = false);
+
+ /// Open the specified memory range as a MemoryBuffer. Note that InputData
+ /// must be null terminated if RequiresNullTerminator is true.
+ static std::unique_ptr<MemoryBuffer>
+ getMemBuffer(StringRef InputData, StringRef BufferName = "",
+ bool RequiresNullTerminator = true);
+
+ static std::unique_ptr<MemoryBuffer>
+ getMemBuffer(MemoryBufferRef Ref, bool RequiresNullTerminator = true);
+
+ /// Open the specified memory range as a MemoryBuffer, copying the contents
+ /// and taking ownership of it. InputData does not have to be null terminated.
+ static std::unique_ptr<MemoryBuffer>
+ getMemBufferCopy(StringRef InputData, const Twine &BufferName = "");
+
+ /// Read all of stdin into a file buffer, and return it.
+ static ErrorOr<std::unique_ptr<MemoryBuffer>> getSTDIN();
+
+ /// Open the specified file as a MemoryBuffer, or open stdin if the Filename
+ /// is "-".
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getFileOrSTDIN(const Twine &Filename, int64_t FileSize = -1,
+ bool RequiresNullTerminator = true);
+
+ /// Map a subrange of the specified file as a MemoryBuffer.
+ static ErrorOr<std::unique_ptr<MemoryBuffer>>
+ getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset,
+ bool IsVolatile = false);
+
+ //===--------------------------------------------------------------------===//
+ // Provided for performance analysis.
+ //===--------------------------------------------------------------------===//
+
+ /// The kind of memory backing used to support the MemoryBuffer.
+ enum BufferKind {
+ MemoryBuffer_Malloc,
+ MemoryBuffer_MMap
+ };
+
+ /// Return information on the memory mechanism used to support the
+ /// MemoryBuffer.
+ virtual BufferKind getBufferKind() const = 0;
+
+ MemoryBufferRef getMemBufferRef() const;
+};
+
+/// This class is an extension of MemoryBuffer, which allows copy-on-write
+/// access to the underlying contents. It only supports creation methods that
+/// are guaranteed to produce a writable buffer. For example, mapping a file
+/// read-only is not supported.
+class WritableMemoryBuffer : public MemoryBuffer {
+protected:
+ WritableMemoryBuffer() = default;
+
+ static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
+ sys::fs::mapped_file_region::priv;
+
+public:
+ using MemoryBuffer::getBuffer;
+ using MemoryBuffer::getBufferEnd;
+ using MemoryBuffer::getBufferStart;
+
+ // const_cast is well-defined here, because the underlying buffer is
+ // guaranteed to have been initialized with a mutable buffer.
+ char *getBufferStart() {
+ return const_cast<char *>(MemoryBuffer::getBufferStart());
+ }
+ char *getBufferEnd() {
+ return const_cast<char *>(MemoryBuffer::getBufferEnd());
+ }
+ MutableArrayRef<char> getBuffer() {
+ return {getBufferStart(), getBufferEnd()};
+ }
+
+ static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+ getFile(const Twine &Filename, int64_t FileSize = -1,
+ bool IsVolatile = false);
+
+ /// Map a subrange of the specified file as a WritableMemoryBuffer.
+ static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
+ getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset,
+ bool IsVolatile = false);
+
+ /// Allocate a new MemoryBuffer of the specified size that is not initialized.
+ /// Note that the caller should initialize the memory allocated by this
+ /// method. The memory is owned by the MemoryBuffer object.
+ static std::unique_ptr<WritableMemoryBuffer>
+ getNewUninitMemBuffer(size_t Size, const Twine &BufferName = "");
+
+ /// Allocate a new zero-initialized MemoryBuffer of the specified size. Note
+ /// that the caller need not initialize the memory allocated by this method.
+ /// The memory is owned by the MemoryBuffer object.
+ static std::unique_ptr<WritableMemoryBuffer>
+ getNewMemBuffer(size_t Size, const Twine &BufferName = "");
+
+private:
+ // Hide these base class factory function so one can't write
+ // WritableMemoryBuffer::getXXX()
+ // and be surprised that he got a read-only Buffer.
+ using MemoryBuffer::getFileAsStream;
+ using MemoryBuffer::getFileOrSTDIN;
+ using MemoryBuffer::getMemBuffer;
+ using MemoryBuffer::getMemBufferCopy;
+ using MemoryBuffer::getOpenFile;
+ using MemoryBuffer::getOpenFileSlice;
+ using MemoryBuffer::getSTDIN;
+};
+
+/// This class is an extension of MemoryBuffer, which allows write access to
+/// the underlying contents and committing those changes to the original source.
+/// It only supports creation methods that are guaranteed to produce a writable
+/// buffer. For example, mapping a file read-only is not supported.
+class WriteThroughMemoryBuffer : public MemoryBuffer {
+protected:
+ WriteThroughMemoryBuffer() = default;
+
+ static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
+ sys::fs::mapped_file_region::readwrite;
+
+public:
+ using MemoryBuffer::getBuffer;
+ using MemoryBuffer::getBufferEnd;
+ using MemoryBuffer::getBufferStart;
+
+ // const_cast is well-defined here, because the underlying buffer is
+ // guaranteed to have been initialized with a mutable buffer.
+ char *getBufferStart() {
+ return const_cast<char *>(MemoryBuffer::getBufferStart());
+ }
+ char *getBufferEnd() {
+ return const_cast<char *>(MemoryBuffer::getBufferEnd());
+ }
+ MutableArrayRef<char> getBuffer() {
+ return {getBufferStart(), getBufferEnd()};
+ }
+
+ static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
+ getFile(const Twine &Filename, int64_t FileSize = -1);
+
+ /// Map a subrange of the specified file as a ReadWriteMemoryBuffer.
+ static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
+ getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset);
+
+private:
+ // Hide these base class factory function so one can't write
+ // WritableMemoryBuffer::getXXX()
+ // and be surprised that he got a read-only Buffer.
+ using MemoryBuffer::getFileAsStream;
+ using MemoryBuffer::getFileOrSTDIN;
+ using MemoryBuffer::getMemBuffer;
+ using MemoryBuffer::getMemBufferCopy;
+ using MemoryBuffer::getOpenFile;
+ using MemoryBuffer::getOpenFileSlice;
+ using MemoryBuffer::getSTDIN;
+};
+
+class MemoryBufferRef {
+ StringRef Buffer;
+ StringRef Identifier;
+
+public:
+ MemoryBufferRef() = default;
+ MemoryBufferRef(const MemoryBuffer& Buffer)
+ : Buffer(Buffer.getBuffer()), Identifier(Buffer.getBufferIdentifier()) {}
+ MemoryBufferRef(StringRef Buffer, StringRef Identifier)
+ : Buffer(Buffer), Identifier(Identifier) {}
+
+ StringRef getBuffer() const { return Buffer; }
+
+ StringRef getBufferIdentifier() const { return Identifier; }
+
+ const char *getBufferStart() const { return Buffer.begin(); }
+ const char *getBufferEnd() const { return Buffer.end(); }
+ size_t getBufferSize() const { return Buffer.size(); }
+};
+
+// Create wrappers for C Binding types (see CBindingWrapping.h).
+DEFINE_SIMPLE_CONVERSION_FUNCTIONS(MemoryBuffer, LLVMMemoryBufferRef)
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_MEMORYBUFFER_H
diff --git a/third_party/llvm-project/include/llvm/Support/NativeFormatting.h b/third_party/llvm-project/include/llvm/Support/NativeFormatting.h
new file mode 100644
index 000000000..825a44c77
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/NativeFormatting.h
@@ -0,0 +1,48 @@
+//===- NativeFormatting.h - Low level formatting helpers ---------*- C++-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_NATIVE_FORMATTING_H
+#define LLVM_SUPPORT_NATIVE_FORMATTING_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/Support/raw_ostream.h"
+
+#include <cstdint>
+
+namespace llvm {
+enum class FloatStyle { Exponent, ExponentUpper, Fixed, Percent };
+enum class IntegerStyle {
+ Integer,
+ Number,
+};
+enum class HexPrintStyle { Upper, Lower, PrefixUpper, PrefixLower };
+
+size_t getDefaultPrecision(FloatStyle Style);
+
+bool isPrefixedHexStyle(HexPrintStyle S);
+
+void write_integer(raw_ostream &S, unsigned int N, size_t MinDigits,
+ IntegerStyle Style);
+void write_integer(raw_ostream &S, int N, size_t MinDigits, IntegerStyle Style);
+void write_integer(raw_ostream &S, unsigned long N, size_t MinDigits,
+ IntegerStyle Style);
+void write_integer(raw_ostream &S, long N, size_t MinDigits,
+ IntegerStyle Style);
+void write_integer(raw_ostream &S, unsigned long long N, size_t MinDigits,
+ IntegerStyle Style);
+void write_integer(raw_ostream &S, long long N, size_t MinDigits,
+ IntegerStyle Style);
+
+void write_hex(raw_ostream &S, uint64_t N, HexPrintStyle Style,
+ Optional<size_t> Width = None);
+void write_double(raw_ostream &S, double D, FloatStyle Style,
+ Optional<size_t> Precision = None);
+}
+
+#endif
+
diff --git a/third_party/llvm-project/include/llvm/Support/Path.h b/third_party/llvm-project/include/llvm/Support/Path.h
new file mode 100644
index 000000000..488f17427
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Path.h
@@ -0,0 +1,464 @@
+//===- llvm/Support/Path.h - Path Operating System Concept ------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the llvm::sys::path namespace. It is designed after
+// TR2/boost filesystem (v3), but modified to remove exception handling and the
+// path class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_PATH_H
+#define LLVM_SUPPORT_PATH_H
+
+#include "llvm/ADT/Twine.h"
+#include "llvm/ADT/iterator.h"
+#include "llvm/Support/DataTypes.h"
+#include <iterator>
+#include <system_error>
+
+namespace llvm {
+namespace sys {
+namespace path {
+
+enum class Style { windows, posix, native };
+
+/// @name Lexical Component Iterator
+/// @{
+
+/// Path iterator.
+///
+/// This is an input iterator that iterates over the individual components in
+/// \a path. The traversal order is as follows:
+/// * The root-name element, if present.
+/// * The root-directory element, if present.
+/// * Each successive filename element, if present.
+/// * Dot, if one or more trailing non-root slash characters are present.
+/// Traversing backwards is possible with \a reverse_iterator
+///
+/// Iteration examples. Each component is separated by ',':
+/// @code
+/// / => /
+/// /foo => /,foo
+/// foo/ => foo,.
+/// /foo/bar => /,foo,bar
+/// ../ => ..,.
+/// C:\foo\bar => C:,/,foo,bar
+/// @endcode
+class const_iterator
+ : public iterator_facade_base<const_iterator, std::input_iterator_tag,
+ const StringRef> {
+ StringRef Path; ///< The entire path.
+ StringRef Component; ///< The current component. Not necessarily in Path.
+ size_t Position = 0; ///< The iterators current position within Path.
+ Style S = Style::native; ///< The path style to use.
+
+ // An end iterator has Position = Path.size() + 1.
+ friend const_iterator begin(StringRef path, Style style);
+ friend const_iterator end(StringRef path);
+
+public:
+ reference operator*() const { return Component; }
+ const_iterator &operator++(); // preincrement
+ bool operator==(const const_iterator &RHS) const;
+
+ /// Difference in bytes between this and RHS.
+ ptrdiff_t operator-(const const_iterator &RHS) const;
+};
+
+/// Reverse path iterator.
+///
+/// This is an input iterator that iterates over the individual components in
+/// \a path in reverse order. The traversal order is exactly reversed from that
+/// of \a const_iterator
+class reverse_iterator
+ : public iterator_facade_base<reverse_iterator, std::input_iterator_tag,
+ const StringRef> {
+ StringRef Path; ///< The entire path.
+ StringRef Component; ///< The current component. Not necessarily in Path.
+ size_t Position = 0; ///< The iterators current position within Path.
+ Style S = Style::native; ///< The path style to use.
+
+ friend reverse_iterator rbegin(StringRef path, Style style);
+ friend reverse_iterator rend(StringRef path);
+
+public:
+ reference operator*() const { return Component; }
+ reverse_iterator &operator++(); // preincrement
+ bool operator==(const reverse_iterator &RHS) const;
+
+ /// Difference in bytes between this and RHS.
+ ptrdiff_t operator-(const reverse_iterator &RHS) const;
+};
+
+/// Get begin iterator over \a path.
+/// @param path Input path.
+/// @returns Iterator initialized with the first component of \a path.
+const_iterator begin(StringRef path, Style style = Style::native);
+
+/// Get end iterator over \a path.
+/// @param path Input path.
+/// @returns Iterator initialized to the end of \a path.
+const_iterator end(StringRef path);
+
+/// Get reverse begin iterator over \a path.
+/// @param path Input path.
+/// @returns Iterator initialized with the first reverse component of \a path.
+reverse_iterator rbegin(StringRef path, Style style = Style::native);
+
+/// Get reverse end iterator over \a path.
+/// @param path Input path.
+/// @returns Iterator initialized to the reverse end of \a path.
+reverse_iterator rend(StringRef path);
+
+/// @}
+/// @name Lexical Modifiers
+/// @{
+
+/// Remove the last component from \a path unless it is the root dir.
+///
+/// Similar to the POSIX "dirname" utility.
+///
+/// @code
+/// directory/filename.cpp => directory/
+/// directory/ => directory
+/// filename.cpp => <empty>
+/// / => /
+/// @endcode
+///
+/// @param path A path that is modified to not have a file component.
+void remove_filename(SmallVectorImpl<char> &path, Style style = Style::native);
+
+/// Replace the file extension of \a path with \a extension.
+///
+/// @code
+/// ./filename.cpp => ./filename.extension
+/// ./filename => ./filename.extension
+/// ./ => ./.extension
+/// @endcode
+///
+/// @param path A path that has its extension replaced with \a extension.
+/// @param extension The extension to be added. It may be empty. It may also
+/// optionally start with a '.', if it does not, one will be
+/// prepended.
+void replace_extension(SmallVectorImpl<char> &path, const Twine &extension,
+ Style style = Style::native);
+
+/// Replace matching path prefix with another path.
+///
+/// @code
+/// /foo, /old, /new => /foo
+/// /old/foo, /old, /new => /new/foo
+/// /foo, <empty>, /new => /new/foo
+/// /old/foo, /old, <empty> => /foo
+/// @endcode
+///
+/// @param Path If \a Path starts with \a OldPrefix modify to instead
+/// start with \a NewPrefix.
+/// @param OldPrefix The path prefix to strip from \a Path.
+/// @param NewPrefix The path prefix to replace \a NewPrefix with.
+void replace_path_prefix(SmallVectorImpl<char> &Path,
+ const StringRef &OldPrefix, const StringRef &NewPrefix,
+ Style style = Style::native);
+
+/// Append to path.
+///
+/// @code
+/// /foo + bar/f => /foo/bar/f
+/// /foo/ + bar/f => /foo/bar/f
+/// foo + bar/f => foo/bar/f
+/// @endcode
+///
+/// @param path Set to \a path + \a component.
+/// @param a The component to be appended to \a path.
+void append(SmallVectorImpl<char> &path, const Twine &a,
+ const Twine &b = "",
+ const Twine &c = "",
+ const Twine &d = "");
+
+void append(SmallVectorImpl<char> &path, Style style, const Twine &a,
+ const Twine &b = "", const Twine &c = "", const Twine &d = "");
+
+/// Append to path.
+///
+/// @code
+/// /foo + [bar,f] => /foo/bar/f
+/// /foo/ + [bar,f] => /foo/bar/f
+/// foo + [bar,f] => foo/bar/f
+/// @endcode
+///
+/// @param path Set to \a path + [\a begin, \a end).
+/// @param begin Start of components to append.
+/// @param end One past the end of components to append.
+void append(SmallVectorImpl<char> &path, const_iterator begin,
+ const_iterator end, Style style = Style::native);
+
+/// @}
+/// @name Transforms (or some other better name)
+/// @{
+
+/// Convert path to the native form. This is used to give paths to users and
+/// operating system calls in the platform's normal way. For example, on Windows
+/// all '/' are converted to '\'.
+///
+/// @param path A path that is transformed to native format.
+/// @param result Holds the result of the transformation.
+void native(const Twine &path, SmallVectorImpl<char> &result,
+ Style style = Style::native);
+
+/// Convert path to the native form in place. This is used to give paths to
+/// users and operating system calls in the platform's normal way. For example,
+/// on Windows all '/' are converted to '\'.
+///
+/// @param path A path that is transformed to native format.
+void native(SmallVectorImpl<char> &path, Style style = Style::native);
+
+/// Replaces backslashes with slashes if Windows.
+///
+/// @param path processed path
+/// @result The result of replacing backslashes with forward slashes if Windows.
+/// On Unix, this function is a no-op because backslashes are valid path
+/// chracters.
+std::string convert_to_slash(StringRef path, Style style = Style::native);
+
+/// @}
+/// @name Lexical Observers
+/// @{
+
+/// Get root name.
+///
+/// @code
+/// //net/hello => //net
+/// c:/hello => c: (on Windows, on other platforms nothing)
+/// /hello => <empty>
+/// @endcode
+///
+/// @param path Input path.
+/// @result The root name of \a path if it has one, otherwise "".
+StringRef root_name(StringRef path, Style style = Style::native);
+
+/// Get root directory.
+///
+/// @code
+/// /goo/hello => /
+/// c:/hello => /
+/// d/file.txt => <empty>
+/// @endcode
+///
+/// @param path Input path.
+/// @result The root directory of \a path if it has one, otherwise
+/// "".
+StringRef root_directory(StringRef path, Style style = Style::native);
+
+/// Get root path.
+///
+/// Equivalent to root_name + root_directory.
+///
+/// @param path Input path.
+/// @result The root path of \a path if it has one, otherwise "".
+StringRef root_path(StringRef path, Style style = Style::native);
+
+/// Get relative path.
+///
+/// @code
+/// C:\hello\world => hello\world
+/// foo/bar => foo/bar
+/// /foo/bar => foo/bar
+/// @endcode
+///
+/// @param path Input path.
+/// @result The path starting after root_path if one exists, otherwise "".
+StringRef relative_path(StringRef path, Style style = Style::native);
+
+/// Get parent path.
+///
+/// @code
+/// / => <empty>
+/// /foo => /
+/// foo/../bar => foo/..
+/// @endcode
+///
+/// @param path Input path.
+/// @result The parent path of \a path if one exists, otherwise "".
+StringRef parent_path(StringRef path, Style style = Style::native);
+
+/// Get filename.
+///
+/// @code
+/// /foo.txt => foo.txt
+/// . => .
+/// .. => ..
+/// / => /
+/// @endcode
+///
+/// @param path Input path.
+/// @result The filename part of \a path. This is defined as the last component
+/// of \a path. Similar to the POSIX "basename" utility.
+StringRef filename(StringRef path, Style style = Style::native);
+
+/// Get stem.
+///
+/// If filename contains a dot but not solely one or two dots, result is the
+/// substring of filename ending at (but not including) the last dot. Otherwise
+/// it is filename.
+///
+/// @code
+/// /foo/bar.txt => bar
+/// /foo/bar => bar
+/// /foo/.txt => <empty>
+/// /foo/. => .
+/// /foo/.. => ..
+/// @endcode
+///
+/// @param path Input path.
+/// @result The stem of \a path.
+StringRef stem(StringRef path, Style style = Style::native);
+
+/// Get extension.
+///
+/// If filename contains a dot but not solely one or two dots, result is the
+/// substring of filename starting at (and including) the last dot, and ending
+/// at the end of \a path. Otherwise "".
+///
+/// @code
+/// /foo/bar.txt => .txt
+/// /foo/bar => <empty>
+/// /foo/.txt => .txt
+/// @endcode
+///
+/// @param path Input path.
+/// @result The extension of \a path.
+StringRef extension(StringRef path, Style style = Style::native);
+
+/// Check whether the given char is a path separator on the host OS.
+///
+/// @param value a character
+/// @result true if \a value is a path separator character on the host OS
+bool is_separator(char value, Style style = Style::native);
+
+/// Return the preferred separator for this platform.
+///
+/// @result StringRef of the preferred separator, null-terminated.
+StringRef get_separator(Style style = Style::native);
+
+/// Get the typical temporary directory for the system, e.g.,
+/// "/var/tmp" or "C:/TEMP"
+///
+/// @param erasedOnReboot Whether to favor a path that is erased on reboot
+/// rather than one that potentially persists longer. This parameter will be
+/// ignored if the user or system has set the typical environment variable
+/// (e.g., TEMP on Windows, TMPDIR on *nix) to specify a temporary directory.
+///
+/// @param result Holds the resulting path name.
+void system_temp_directory(bool erasedOnReboot, SmallVectorImpl<char> &result);
+
+/// Get the user's home directory.
+///
+/// @param result Holds the resulting path name.
+/// @result True if a home directory is set, false otherwise.
+bool home_directory(SmallVectorImpl<char> &result);
+
+/// Has root name?
+///
+/// root_name != ""
+///
+/// @param path Input path.
+/// @result True if the path has a root name, false otherwise.
+bool has_root_name(const Twine &path, Style style = Style::native);
+
+/// Has root directory?
+///
+/// root_directory != ""
+///
+/// @param path Input path.
+/// @result True if the path has a root directory, false otherwise.
+bool has_root_directory(const Twine &path, Style style = Style::native);
+
+/// Has root path?
+///
+/// root_path != ""
+///
+/// @param path Input path.
+/// @result True if the path has a root path, false otherwise.
+bool has_root_path(const Twine &path, Style style = Style::native);
+
+/// Has relative path?
+///
+/// relative_path != ""
+///
+/// @param path Input path.
+/// @result True if the path has a relative path, false otherwise.
+bool has_relative_path(const Twine &path, Style style = Style::native);
+
+/// Has parent path?
+///
+/// parent_path != ""
+///
+/// @param path Input path.
+/// @result True if the path has a parent path, false otherwise.
+bool has_parent_path(const Twine &path, Style style = Style::native);
+
+/// Has filename?
+///
+/// filename != ""
+///
+/// @param path Input path.
+/// @result True if the path has a filename, false otherwise.
+bool has_filename(const Twine &path, Style style = Style::native);
+
+/// Has stem?
+///
+/// stem != ""
+///
+/// @param path Input path.
+/// @result True if the path has a stem, false otherwise.
+bool has_stem(const Twine &path, Style style = Style::native);
+
+/// Has extension?
+///
+/// extension != ""
+///
+/// @param path Input path.
+/// @result True if the path has a extension, false otherwise.
+bool has_extension(const Twine &path, Style style = Style::native);
+
+/// Is path absolute?
+///
+/// @param path Input path.
+/// @result True if the path is absolute, false if it is not.
+bool is_absolute(const Twine &path, Style style = Style::native);
+
+/// Is path relative?
+///
+/// @param path Input path.
+/// @result True if the path is relative, false if it is not.
+bool is_relative(const Twine &path, Style style = Style::native);
+
+/// Remove redundant leading "./" pieces and consecutive separators.
+///
+/// @param path Input path.
+/// @result The cleaned-up \a path.
+StringRef remove_leading_dotslash(StringRef path, Style style = Style::native);
+
+/// In-place remove any './' and optionally '../' components from a path.
+///
+/// @param path processed path
+/// @param remove_dot_dot specify if '../' (except for leading "../") should be
+/// removed
+/// @result True if path was changed
+bool remove_dots(SmallVectorImpl<char> &path, bool remove_dot_dot = false,
+ Style style = Style::native);
+
+#if defined(_WIN32)
+std::error_code widenPath(const Twine &Path8, SmallVectorImpl<wchar_t> &Path16);
+#endif
+
+} // end namespace path
+} // end namespace sys
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/PointerLikeTypeTraits.h b/third_party/llvm-project/include/llvm/Support/PointerLikeTypeTraits.h
new file mode 100644
index 000000000..1e7e5b53c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/PointerLikeTypeTraits.h
@@ -0,0 +1,149 @@
+//===- llvm/Support/PointerLikeTypeTraits.h - Pointer Traits ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the PointerLikeTypeTraits class. This allows data
+// structures to reason about pointers and other things that are pointer sized.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_POINTERLIKETYPETRAITS_H
+#define LLVM_SUPPORT_POINTERLIKETYPETRAITS_H
+
+#include "llvm/Support/DataTypes.h"
+#include <assert.h>
+#include <type_traits>
+
+namespace llvm {
+
+/// A traits type that is used to handle pointer types and things that are just
+/// wrappers for pointers as a uniform entity.
+template <typename T> struct PointerLikeTypeTraits;
+
+namespace detail {
+/// A tiny meta function to compute the log2 of a compile time constant.
+template <size_t N>
+struct ConstantLog2
+ : std::integral_constant<size_t, ConstantLog2<N / 2>::value + 1> {};
+template <> struct ConstantLog2<1> : std::integral_constant<size_t, 0> {};
+
+// Provide a trait to check if T is pointer-like.
+template <typename T, typename U = void> struct HasPointerLikeTypeTraits {
+ static const bool value = false;
+};
+
+// sizeof(T) is valid only for a complete T.
+template <typename T> struct HasPointerLikeTypeTraits<
+ T, decltype((sizeof(PointerLikeTypeTraits<T>) + sizeof(T)), void())> {
+ static const bool value = true;
+};
+
+template <typename T> struct IsPointerLike {
+ static const bool value = HasPointerLikeTypeTraits<T>::value;
+};
+
+template <typename T> struct IsPointerLike<T *> {
+ static const bool value = true;
+};
+} // namespace detail
+
+// Provide PointerLikeTypeTraits for non-cvr pointers.
+template <typename T> struct PointerLikeTypeTraits<T *> {
+ static inline void *getAsVoidPointer(T *P) { return P; }
+ static inline T *getFromVoidPointer(void *P) { return static_cast<T *>(P); }
+
+ enum { NumLowBitsAvailable = detail::ConstantLog2<alignof(T)>::value };
+};
+
+template <> struct PointerLikeTypeTraits<void *> {
+ static inline void *getAsVoidPointer(void *P) { return P; }
+ static inline void *getFromVoidPointer(void *P) { return P; }
+
+ /// Note, we assume here that void* is related to raw malloc'ed memory and
+ /// that malloc returns objects at least 4-byte aligned. However, this may be
+ /// wrong, or pointers may be from something other than malloc. In this case,
+ /// you should specify a real typed pointer or avoid this template.
+ ///
+ /// All clients should use assertions to do a run-time check to ensure that
+ /// this is actually true.
+ enum { NumLowBitsAvailable = 2 };
+};
+
+// Provide PointerLikeTypeTraits for const things.
+template <typename T> struct PointerLikeTypeTraits<const T> {
+ typedef PointerLikeTypeTraits<T> NonConst;
+
+ static inline const void *getAsVoidPointer(const T P) {
+ return NonConst::getAsVoidPointer(P);
+ }
+ static inline const T getFromVoidPointer(const void *P) {
+ return NonConst::getFromVoidPointer(const_cast<void *>(P));
+ }
+ enum { NumLowBitsAvailable = NonConst::NumLowBitsAvailable };
+};
+
+// Provide PointerLikeTypeTraits for const pointers.
+template <typename T> struct PointerLikeTypeTraits<const T *> {
+ typedef PointerLikeTypeTraits<T *> NonConst;
+
+ static inline const void *getAsVoidPointer(const T *P) {
+ return NonConst::getAsVoidPointer(const_cast<T *>(P));
+ }
+ static inline const T *getFromVoidPointer(const void *P) {
+ return NonConst::getFromVoidPointer(const_cast<void *>(P));
+ }
+ enum { NumLowBitsAvailable = NonConst::NumLowBitsAvailable };
+};
+
+// Provide PointerLikeTypeTraits for uintptr_t.
+template <> struct PointerLikeTypeTraits<uintptr_t> {
+ static inline void *getAsVoidPointer(uintptr_t P) {
+ return reinterpret_cast<void *>(P);
+ }
+ static inline uintptr_t getFromVoidPointer(void *P) {
+ return reinterpret_cast<uintptr_t>(P);
+ }
+ // No bits are available!
+ enum { NumLowBitsAvailable = 0 };
+};
+
+/// Provide suitable custom traits struct for function pointers.
+///
+/// Function pointers can't be directly given these traits as functions can't
+/// have their alignment computed with `alignof` and we need different casting.
+///
+/// To rely on higher alignment for a specialized use, you can provide a
+/// customized form of this template explicitly with higher alignment, and
+/// potentially use alignment attributes on functions to satisfy that.
+template <int Alignment, typename FunctionPointerT>
+struct FunctionPointerLikeTypeTraits {
+ enum { NumLowBitsAvailable = detail::ConstantLog2<Alignment>::value };
+ static inline void *getAsVoidPointer(FunctionPointerT P) {
+ assert((reinterpret_cast<uintptr_t>(P) &
+ ~((uintptr_t)-1 << NumLowBitsAvailable)) == 0 &&
+ "Alignment not satisfied for an actual function pointer!");
+ return reinterpret_cast<void *>(P);
+ }
+ static inline FunctionPointerT getFromVoidPointer(void *P) {
+ return reinterpret_cast<FunctionPointerT>(P);
+ }
+};
+
+/// Provide a default specialization for function pointers that assumes 4-byte
+/// alignment.
+///
+/// We assume here that functions used with this are always at least 4-byte
+/// aligned. This means that, for example, thumb functions won't work or systems
+/// with weird unaligned function pointers won't work. But all practical systems
+/// we support satisfy this requirement.
+template <typename ReturnT, typename... ParamTs>
+struct PointerLikeTypeTraits<ReturnT (*)(ParamTs...)>
+ : FunctionPointerLikeTypeTraits<4, ReturnT (*)(ParamTs...)> {};
+
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Printable.h b/third_party/llvm-project/include/llvm/Support/Printable.h
new file mode 100644
index 000000000..0f8670d04
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Printable.h
@@ -0,0 +1,51 @@
+//===--- Printable.h - Print function helpers -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the Printable struct.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_PRINTABLE_H
+#define LLVM_SUPPORT_PRINTABLE_H
+
+#include <functional>
+
+namespace llvm {
+
+class raw_ostream;
+
+/// Simple wrapper around std::function<void(raw_ostream&)>.
+/// This class is useful to construct print helpers for raw_ostream.
+///
+/// Example:
+/// Printable PrintRegister(unsigned Register) {
+/// return Printable([Register](raw_ostream &OS) {
+/// OS << getRegisterName(Register);
+/// }
+/// }
+/// ... OS << PrintRegister(Register); ...
+///
+/// Implementation note: Ideally this would just be a typedef, but doing so
+/// leads to operator << being ambiguous as function has matching constructors
+/// in some STL versions. I have seen the problem on gcc 4.6 libstdc++ and
+/// microsoft STL.
+class Printable {
+public:
+ std::function<void(raw_ostream &OS)> Print;
+ Printable(std::function<void(raw_ostream &OS)> Print)
+ : Print(std::move(Print)) {}
+};
+
+inline raw_ostream &operator<<(raw_ostream &OS, const Printable &P) {
+ P.Print(OS);
+ return OS;
+}
+
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Process.h b/third_party/llvm-project/include/llvm/Support/Process.h
new file mode 100644
index 000000000..67e379125
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Process.h
@@ -0,0 +1,209 @@
+//===- llvm/Support/Process.h -----------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// Provides a library for accessing information about this process and other
+/// processes on the operating system. Also provides means of spawning
+/// subprocess for commands. The design of this library is modeled after the
+/// proposed design of the Boost.Process library, and is design specifically to
+/// follow the style of standard libraries and potentially become a proposal
+/// for a standard library.
+///
+/// This file declares the llvm::sys::Process class which contains a collection
+/// of legacy static interfaces for extracting various information about the
+/// current process. The goal is to migrate users of this API over to the new
+/// interfaces.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_PROCESS_H
+#define LLVM_SUPPORT_PROCESS_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/Support/Allocator.h"
+#include "llvm/Support/Chrono.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/Error.h"
+#include <system_error>
+
+namespace llvm {
+template <typename T> class ArrayRef;
+class StringRef;
+
+namespace sys {
+
+
+/// A collection of legacy interfaces for querying information about the
+/// current executing process.
+class Process {
+public:
+ /// Get the process's page size.
+ /// This may fail if the underlying syscall returns an error. In most cases,
+ /// page size information is used for optimization, and this error can be
+ /// safely discarded by calling consumeError, and an estimated page size
+ /// substituted instead.
+ static Expected<unsigned> getPageSize();
+
+ /// Get the process's estimated page size.
+ /// This function always succeeds, but if the underlying syscall to determine
+ /// the page size fails then this will silently return an estimated page size.
+ /// The estimated page size is guaranteed to be a power of 2.
+ static unsigned getPageSizeEstimate() {
+ if (auto PageSize = getPageSize())
+ return *PageSize;
+ else {
+ consumeError(PageSize.takeError());
+ return 4096;
+ }
+ }
+
+ /// Return process memory usage.
+ /// This static function will return the total amount of memory allocated
+ /// by the process. This only counts the memory allocated via the malloc,
+ /// calloc and realloc functions and includes any "free" holes in the
+ /// allocated space.
+ static size_t GetMallocUsage();
+
+ /// This static function will set \p user_time to the amount of CPU time
+ /// spent in user (non-kernel) mode and \p sys_time to the amount of CPU
+ /// time spent in system (kernel) mode. If the operating system does not
+ /// support collection of these metrics, a zero duration will be for both
+ /// values.
+ /// \param elapsed Returns the system_clock::now() giving current time
+ /// \param user_time Returns the current amount of user time for the process
+ /// \param sys_time Returns the current amount of system time for the process
+ static void GetTimeUsage(TimePoint<> &elapsed,
+ std::chrono::nanoseconds &user_time,
+ std::chrono::nanoseconds &sys_time);
+
+ /// This function makes the necessary calls to the operating system to
+ /// prevent core files or any other kind of large memory dumps that can
+ /// occur when a program fails.
+ /// Prevent core file generation.
+ static void PreventCoreFiles();
+
+ /// true if PreventCoreFiles has been called, false otherwise.
+ static bool AreCoreFilesPrevented();
+
+ // This function returns the environment variable \arg name's value as a UTF-8
+ // string. \arg Name is assumed to be in UTF-8 encoding too.
+ static Optional<std::string> GetEnv(StringRef name);
+
+ /// This function searches for an existing file in the list of directories
+ /// in a PATH like environment variable, and returns the first file found,
+ /// according to the order of the entries in the PATH like environment
+ /// variable. If an ignore list is specified, then any folder which is in
+ /// the PATH like environment variable but is also in IgnoreList is not
+ /// considered.
+ static Optional<std::string> FindInEnvPath(StringRef EnvName,
+ StringRef FileName,
+ ArrayRef<std::string> IgnoreList);
+
+ static Optional<std::string> FindInEnvPath(StringRef EnvName,
+ StringRef FileName);
+
+ // This functions ensures that the standard file descriptors (input, output,
+ // and error) are properly mapped to a file descriptor before we use any of
+ // them. This should only be called by standalone programs, library
+ // components should not call this.
+ static std::error_code FixupStandardFileDescriptors();
+
+ // This function safely closes a file descriptor. It is not safe to retry
+ // close(2) when it returns with errno equivalent to EINTR; this is because
+ // *nixen cannot agree if the file descriptor is, in fact, closed when this
+ // occurs.
+ //
+ // N.B. Some operating systems, due to thread cancellation, cannot properly
+ // guarantee that it will or will not be closed one way or the other!
+ static std::error_code SafelyCloseFileDescriptor(int FD);
+
+ /// This function determines if the standard input is connected directly
+ /// to a user's input (keyboard probably), rather than coming from a file
+ /// or pipe.
+ static bool StandardInIsUserInput();
+
+ /// This function determines if the standard output is connected to a
+ /// "tty" or "console" window. That is, the output would be displayed to
+ /// the user rather than being put on a pipe or stored in a file.
+ static bool StandardOutIsDisplayed();
+
+ /// This function determines if the standard error is connected to a
+ /// "tty" or "console" window. That is, the output would be displayed to
+ /// the user rather than being put on a pipe or stored in a file.
+ static bool StandardErrIsDisplayed();
+
+ /// This function determines if the given file descriptor is connected to
+ /// a "tty" or "console" window. That is, the output would be displayed to
+ /// the user rather than being put on a pipe or stored in a file.
+ static bool FileDescriptorIsDisplayed(int fd);
+
+ /// This function determines if the given file descriptor is displayd and
+ /// supports colors.
+ static bool FileDescriptorHasColors(int fd);
+
+ /// This function determines the number of columns in the window
+ /// if standard output is connected to a "tty" or "console"
+ /// window. If standard output is not connected to a tty or
+ /// console, or if the number of columns cannot be determined,
+ /// this routine returns zero.
+ static unsigned StandardOutColumns();
+
+ /// This function determines the number of columns in the window
+ /// if standard error is connected to a "tty" or "console"
+ /// window. If standard error is not connected to a tty or
+ /// console, or if the number of columns cannot be determined,
+ /// this routine returns zero.
+ static unsigned StandardErrColumns();
+
+ /// This function determines whether the terminal connected to standard
+ /// output supports colors. If standard output is not connected to a
+ /// terminal, this function returns false.
+ static bool StandardOutHasColors();
+
+ /// This function determines whether the terminal connected to standard
+ /// error supports colors. If standard error is not connected to a
+ /// terminal, this function returns false.
+ static bool StandardErrHasColors();
+
+ /// Enables or disables whether ANSI escape sequences are used to output
+ /// colors. This only has an effect on Windows.
+ /// Note: Setting this option is not thread-safe and should only be done
+ /// during initialization.
+ static void UseANSIEscapeCodes(bool enable);
+
+ /// Whether changing colors requires the output to be flushed.
+ /// This is needed on systems that don't support escape sequences for
+ /// changing colors.
+ static bool ColorNeedsFlush();
+
+ /// This function returns the colorcode escape sequences.
+ /// If ColorNeedsFlush() is true then this function will change the colors
+ /// and return an empty escape sequence. In that case it is the
+ /// responsibility of the client to flush the output stream prior to
+ /// calling this function.
+ static const char *OutputColor(char c, bool bold, bool bg);
+
+ /// Same as OutputColor, but only enables the bold attribute.
+ static const char *OutputBold(bool bg);
+
+ /// This function returns the escape sequence to reverse forground and
+ /// background colors.
+ static const char *OutputReverse();
+
+ /// Resets the terminals colors, or returns an escape sequence to do so.
+ static const char *ResetColor();
+
+ /// Get the result of a process wide random number generator. The
+ /// generator will be automatically seeded in non-deterministic fashion.
+ static unsigned GetRandomNumber();
+};
+
+}
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Program.h b/third_party/llvm-project/include/llvm/Support/Program.h
new file mode 100644
index 000000000..6b2315c5d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Program.h
@@ -0,0 +1,208 @@
+//===- llvm/Support/Program.h ------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the llvm::sys::Program class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_PROGRAM_H
+#define LLVM_SUPPORT_PROGRAM_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Config/llvm-config.h"
+#include "llvm/Support/ErrorOr.h"
+#include <system_error>
+
+namespace llvm {
+namespace sys {
+
+ /// This is the OS-specific separator for PATH like environment variables:
+ // a colon on Unix or a semicolon on Windows.
+#if defined(LLVM_ON_UNIX)
+ const char EnvPathSeparator = ':';
+#elif defined (_WIN32)
+ const char EnvPathSeparator = ';';
+#endif
+
+#if defined(_WIN32)
+ typedef unsigned long procid_t; // Must match the type of DWORD on Windows.
+ typedef void *process_t; // Must match the type of HANDLE on Windows.
+#else
+ typedef pid_t procid_t;
+ typedef procid_t process_t;
+#endif
+
+ /// This struct encapsulates information about a process.
+ struct ProcessInfo {
+ enum : procid_t { InvalidPid = 0 };
+
+ procid_t Pid; /// The process identifier.
+ process_t Process; /// Platform-dependent process object.
+
+ /// The return code, set after execution.
+ int ReturnCode;
+
+ ProcessInfo();
+ };
+
+ /// Find the first executable file \p Name in \p Paths.
+ ///
+ /// This does not perform hashing as a shell would but instead stats each PATH
+ /// entry individually so should generally be avoided. Core LLVM library
+ /// functions and options should instead require fully specified paths.
+ ///
+ /// \param Name name of the executable to find. If it contains any system
+ /// slashes, it will be returned as is.
+ /// \param Paths optional list of paths to search for \p Name. If empty it
+ /// will use the system PATH environment instead.
+ ///
+ /// \returns The fully qualified path to the first \p Name in \p Paths if it
+ /// exists. \p Name if \p Name has slashes in it. Otherwise an error.
+ ErrorOr<std::string>
+ findProgramByName(StringRef Name, ArrayRef<StringRef> Paths = {});
+
+ // These functions change the specified standard stream (stdin or stdout) to
+ // binary mode. They return errc::success if the specified stream
+ // was changed. Otherwise a platform dependent error is returned.
+ std::error_code ChangeStdinToBinary();
+ std::error_code ChangeStdoutToBinary();
+
+ /// This function executes the program using the arguments provided. The
+ /// invoked program will inherit the stdin, stdout, and stderr file
+ /// descriptors, the environment and other configuration settings of the
+ /// invoking program.
+ /// This function waits for the program to finish, so should be avoided in
+ /// library functions that aren't expected to block. Consider using
+ /// ExecuteNoWait() instead.
+ /// \returns an integer result code indicating the status of the program.
+ /// A zero or positive value indicates the result code of the program.
+ /// -1 indicates failure to execute
+ /// -2 indicates a crash during execution or timeout
+ int ExecuteAndWait(
+ StringRef Program, ///< Path of the program to be executed. It is
+ ///< presumed this is the result of the findProgramByName method.
+ ArrayRef<StringRef> Args, ///< An array of strings that are passed to the
+ ///< program. The first element should be the name of the program.
+ ///< The array should **not** be terminated by an empty StringRef.
+ Optional<ArrayRef<StringRef>> Env = None, ///< An optional vector of
+ ///< strings to use for the program's environment. If not provided, the
+ ///< current program's environment will be used. If specified, the
+ ///< vector should **not** be terminated by an empty StringRef.
+ ArrayRef<Optional<StringRef>> Redirects = {}, ///<
+ ///< An array of optional paths. Should have a size of zero or three.
+ ///< If the array is empty, no redirections are performed.
+ ///< Otherwise, the inferior process's stdin(0), stdout(1), and stderr(2)
+ ///< will be redirected to the corresponding paths, if the optional path
+ ///< is present (not \c llvm::None).
+ ///< When an empty path is passed in, the corresponding file descriptor
+ ///< will be disconnected (ie, /dev/null'd) in a portable way.
+ unsigned SecondsToWait = 0, ///< If non-zero, this specifies the amount
+ ///< of time to wait for the child process to exit. If the time
+ ///< expires, the child is killed and this call returns. If zero,
+ ///< this function will wait until the child finishes or forever if
+ ///< it doesn't.
+ unsigned MemoryLimit = 0, ///< If non-zero, this specifies max. amount
+ ///< of memory can be allocated by process. If memory usage will be
+ ///< higher limit, the child is killed and this call returns. If zero
+ ///< - no memory limit.
+ std::string *ErrMsg = nullptr, ///< If non-zero, provides a pointer to a
+ ///< string instance in which error messages will be returned. If the
+ ///< string is non-empty upon return an error occurred while invoking the
+ ///< program.
+ bool *ExecutionFailed = nullptr);
+
+ /// Similar to ExecuteAndWait, but returns immediately.
+ /// @returns The \see ProcessInfo of the newly launced process.
+ /// \note On Microsoft Windows systems, users will need to either call
+ /// \see Wait until the process finished execution or win32 CloseHandle() API
+ /// on ProcessInfo.ProcessHandle to avoid memory leaks.
+ ProcessInfo ExecuteNoWait(StringRef Program, ArrayRef<StringRef> Args,
+ Optional<ArrayRef<StringRef>> Env,
+ ArrayRef<Optional<StringRef>> Redirects = {},
+ unsigned MemoryLimit = 0,
+ std::string *ErrMsg = nullptr,
+ bool *ExecutionFailed = nullptr);
+
+ /// Return true if the given arguments fit within system-specific
+ /// argument length limits.
+ bool commandLineFitsWithinSystemLimits(StringRef Program,
+ ArrayRef<StringRef> Args);
+
+ /// Return true if the given arguments fit within system-specific
+ /// argument length limits.
+ bool commandLineFitsWithinSystemLimits(StringRef Program,
+ ArrayRef<const char *> Args);
+
+ /// File encoding options when writing contents that a non-UTF8 tool will
+ /// read (on Windows systems). For UNIX, we always use UTF-8.
+ enum WindowsEncodingMethod {
+ /// UTF-8 is the LLVM native encoding, being the same as "do not perform
+ /// encoding conversion".
+ WEM_UTF8,
+ WEM_CurrentCodePage,
+ WEM_UTF16
+ };
+
+ /// Saves the UTF8-encoded \p contents string into the file \p FileName
+ /// using a specific encoding.
+ ///
+ /// This write file function adds the possibility to choose which encoding
+ /// to use when writing a text file. On Windows, this is important when
+ /// writing files with internationalization support with an encoding that is
+ /// different from the one used in LLVM (UTF-8). We use this when writing
+ /// response files, since GCC tools on MinGW only understand legacy code
+ /// pages, and VisualStudio tools only understand UTF-16.
+ /// For UNIX, using different encodings is silently ignored, since all tools
+ /// work well with UTF-8.
+ /// This function assumes that you only use UTF-8 *text* data and will convert
+ /// it to your desired encoding before writing to the file.
+ ///
+ /// FIXME: We use EM_CurrentCodePage to write response files for GNU tools in
+ /// a MinGW/MinGW-w64 environment, which has serious flaws but currently is
+ /// our best shot to make gcc/ld understand international characters. This
+ /// should be changed as soon as binutils fix this to support UTF16 on mingw.
+ ///
+ /// \returns non-zero error_code if failed
+ std::error_code
+ writeFileWithEncoding(StringRef FileName, StringRef Contents,
+ WindowsEncodingMethod Encoding = WEM_UTF8);
+
+ /// This function waits for the process specified by \p PI to finish.
+ /// \returns A \see ProcessInfo struct with Pid set to:
+ /// \li The process id of the child process if the child process has changed
+ /// state.
+ /// \li 0 if the child process has not changed state.
+ /// \note Users of this function should always check the ReturnCode member of
+ /// the \see ProcessInfo returned from this function.
+ ProcessInfo Wait(
+ const ProcessInfo &PI, ///< The child process that should be waited on.
+ unsigned SecondsToWait, ///< If non-zero, this specifies the amount of
+ ///< time to wait for the child process to exit. If the time expires, the
+ ///< child is killed and this function returns. If zero, this function
+ ///< will perform a non-blocking wait on the child process.
+ bool WaitUntilTerminates, ///< If true, ignores \p SecondsToWait and waits
+ ///< until child has terminated.
+ std::string *ErrMsg = nullptr ///< If non-zero, provides a pointer to a
+ ///< string instance in which error messages will be returned. If the
+ ///< string is non-empty upon return an error occurred while invoking the
+ ///< program.
+ );
+
+#if defined(_WIN32)
+ /// Given a list of command line arguments, quote and escape them as necessary
+ /// to build a single flat command line appropriate for calling CreateProcess
+ /// on
+ /// Windows.
+ std::string flattenWindowsCommandLine(ArrayRef<StringRef> Args);
+#endif
+ }
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Regex.h b/third_party/llvm-project/include/llvm/Support/Regex.h
new file mode 100644
index 000000000..b2620ab4c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Regex.h
@@ -0,0 +1,109 @@
+//===-- Regex.h - Regular Expression matcher implementation -*- C++ -*-----===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a POSIX regular expression matcher. Both Basic and
+// Extended POSIX regular expressions (ERE) are supported. EREs were extended
+// to support backreferences in matches.
+// This implementation also supports matching strings with embedded NUL chars.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_REGEX_H
+#define LLVM_SUPPORT_REGEX_H
+
+#include <string>
+
+struct llvm_regex;
+
+namespace llvm {
+ class StringRef;
+ template<typename T> class SmallVectorImpl;
+
+ class Regex {
+ public:
+ enum {
+ NoFlags=0,
+ /// Compile for matching that ignores upper/lower case distinctions.
+ IgnoreCase=1,
+ /// Compile for newline-sensitive matching. With this flag '[^' bracket
+ /// expressions and '.' never match newline. A ^ anchor matches the
+ /// null string after any newline in the string in addition to its normal
+ /// function, and the $ anchor matches the null string before any
+ /// newline in the string in addition to its normal function.
+ Newline=2,
+ /// By default, the POSIX extended regular expression (ERE) syntax is
+ /// assumed. Pass this flag to turn on basic regular expressions (BRE)
+ /// instead.
+ BasicRegex=4
+ };
+
+ Regex();
+ /// Compiles the given regular expression \p Regex.
+ ///
+ /// \param Regex - referenced string is no longer needed after this
+ /// constructor does finish. Only its compiled form is kept stored.
+ Regex(StringRef Regex, unsigned Flags = NoFlags);
+ Regex(const Regex &) = delete;
+ Regex &operator=(Regex regex) {
+ std::swap(preg, regex.preg);
+ std::swap(error, regex.error);
+ return *this;
+ }
+ Regex(Regex &®ex);
+ ~Regex();
+
+ /// isValid - returns the error encountered during regex compilation, if
+ /// any.
+ bool isValid(std::string &Error) const;
+ bool isValid() const { return !error; }
+
+ /// getNumMatches - In a valid regex, return the number of parenthesized
+ /// matches it contains. The number filled in by match will include this
+ /// many entries plus one for the whole regex (as element 0).
+ unsigned getNumMatches() const;
+
+ /// matches - Match the regex against a given \p String.
+ ///
+ /// \param Matches - If given, on a successful match this will be filled in
+ /// with references to the matched group expressions (inside \p String),
+ /// the first group is always the entire pattern.
+ ///
+ /// \param Error - If non-null, any errors in the matching will be recorded
+ /// as a non-empty string. If there is no error, it will be an empty string.
+ ///
+ /// This returns true on a successful match.
+ bool match(StringRef String, SmallVectorImpl<StringRef> *Matches = nullptr,
+ std::string *Error = nullptr) const;
+
+ /// sub - Return the result of replacing the first match of the regex in
+ /// \p String with the \p Repl string. Backreferences like "\0" in the
+ /// replacement string are replaced with the appropriate match substring.
+ ///
+ /// Note that the replacement string has backslash escaping performed on
+ /// it. Invalid backreferences are ignored (replaced by empty strings).
+ ///
+ /// \param Error If non-null, any errors in the substitution (invalid
+ /// backreferences, trailing backslashes) will be recorded as a non-empty
+ /// string. If there is no error, it will be an empty string.
+ std::string sub(StringRef Repl, StringRef String,
+ std::string *Error = nullptr) const;
+
+ /// If this function returns true, ^Str$ is an extended regular
+ /// expression that matches Str and only Str.
+ static bool isLiteralERE(StringRef Str);
+
+ /// Turn String into a regex by escaping its special characters.
+ static std::string escape(StringRef String);
+
+ private:
+ struct llvm_regex *preg;
+ int error;
+ };
+}
+
+#endif // LLVM_SUPPORT_REGEX_H
diff --git a/third_party/llvm-project/include/llvm/Support/ReverseIteration.h b/third_party/llvm-project/include/llvm/Support/ReverseIteration.h
new file mode 100644
index 000000000..5e0238d81
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ReverseIteration.h
@@ -0,0 +1,19 @@
+#ifndef LLVM_SUPPORT_REVERSEITERATION_H
+#define LLVM_SUPPORT_REVERSEITERATION_H
+
+#include "llvm/Config/abi-breaking.h"
+#include "llvm/Support/PointerLikeTypeTraits.h"
+
+namespace llvm {
+
+template<class T = void *>
+bool shouldReverseIterate() {
+#if LLVM_ENABLE_REVERSE_ITERATION
+ return detail::IsPointerLike<T>::value;
+#else
+ return false;
+#endif
+}
+
+}
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/SMLoc.h b/third_party/llvm-project/include/llvm/Support/SMLoc.h
new file mode 100644
index 000000000..d8607034e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/SMLoc.h
@@ -0,0 +1,64 @@
+//===- SMLoc.h - Source location for use with diagnostics -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the SMLoc class. This class encapsulates a location in
+// source code for use in diagnostics.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_SMLOC_H
+#define LLVM_SUPPORT_SMLOC_H
+
+#include "llvm/ADT/None.h"
+#include <cassert>
+
+namespace llvm {
+
+/// Represents a location in source code.
+class SMLoc {
+ const char *Ptr = nullptr;
+
+public:
+ SMLoc() = default;
+
+ bool isValid() const { return Ptr != nullptr; }
+
+ bool operator==(const SMLoc &RHS) const { return RHS.Ptr == Ptr; }
+ bool operator!=(const SMLoc &RHS) const { return RHS.Ptr != Ptr; }
+
+ const char *getPointer() const { return Ptr; }
+
+ static SMLoc getFromPointer(const char *Ptr) {
+ SMLoc L;
+ L.Ptr = Ptr;
+ return L;
+ }
+};
+
+/// Represents a range in source code.
+///
+/// SMRange is implemented using a half-open range, as is the convention in C++.
+/// In the string "abc", the range [1,3) represents the substring "bc", and the
+/// range [2,2) represents an empty range between the characters "b" and "c".
+class SMRange {
+public:
+ SMLoc Start, End;
+
+ SMRange() = default;
+ SMRange(NoneType) {}
+ SMRange(SMLoc St, SMLoc En) : Start(St), End(En) {
+ assert(Start.isValid() == End.isValid() &&
+ "Start and End should either both be valid or both be invalid!");
+ }
+
+ bool isValid() const { return Start.isValid(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_SMLOC_H
diff --git a/third_party/llvm-project/include/llvm/Support/ScopedPrinter.h b/third_party/llvm-project/include/llvm/Support/ScopedPrinter.h
new file mode 100644
index 000000000..88daedc87
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/ScopedPrinter.h
@@ -0,0 +1,388 @@
+//===-- ScopedPrinter.h ----------------------------------------*- C++ -*--===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_SCOPEDPRINTER_H
+#define LLVM_SUPPORT_SCOPEDPRINTER_H
+
+#include "llvm/ADT/APSInt.h"
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/DataTypes.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+
+namespace llvm {
+
+template <typename T> struct EnumEntry {
+ StringRef Name;
+ // While Name suffices in most of the cases, in certain cases
+ // GNU style and LLVM style of ELFDumper do not
+ // display same string for same enum. The AltName if initialized appropriately
+ // will hold the string that GNU style emits.
+ // Example:
+ // "EM_X86_64" string on LLVM style for Elf_Ehdr->e_machine corresponds to
+ // "Advanced Micro Devices X86-64" on GNU style
+ StringRef AltName;
+ T Value;
+ EnumEntry(StringRef N, StringRef A, T V) : Name(N), AltName(A), Value(V) {}
+ EnumEntry(StringRef N, T V) : Name(N), AltName(N), Value(V) {}
+};
+
+struct HexNumber {
+ // To avoid sign-extension we have to explicitly cast to the appropriate
+ // unsigned type. The overloads are here so that every type that is implicitly
+ // convertible to an integer (including enums and endian helpers) can be used
+ // without requiring type traits or call-site changes.
+ HexNumber(char Value) : Value(static_cast<unsigned char>(Value)) {}
+ HexNumber(signed char Value) : Value(static_cast<unsigned char>(Value)) {}
+ HexNumber(signed short Value) : Value(static_cast<unsigned short>(Value)) {}
+ HexNumber(signed int Value) : Value(static_cast<unsigned int>(Value)) {}
+ HexNumber(signed long Value) : Value(static_cast<unsigned long>(Value)) {}
+ HexNumber(signed long long Value)
+ : Value(static_cast<unsigned long long>(Value)) {}
+ HexNumber(unsigned char Value) : Value(Value) {}
+ HexNumber(unsigned short Value) : Value(Value) {}
+ HexNumber(unsigned int Value) : Value(Value) {}
+ HexNumber(unsigned long Value) : Value(Value) {}
+ HexNumber(unsigned long long Value) : Value(Value) {}
+ uint64_t Value;
+};
+
+raw_ostream &operator<<(raw_ostream &OS, const HexNumber &Value);
+const std::string to_hexString(uint64_t Value, bool UpperCase = true);
+
+template <class T> const std::string to_string(const T &Value) {
+ std::string number;
+ llvm::raw_string_ostream stream(number);
+ stream << Value;
+ return stream.str();
+}
+
+class ScopedPrinter {
+public:
+ ScopedPrinter(raw_ostream &OS) : OS(OS), IndentLevel(0) {}
+
+ void flush() { OS.flush(); }
+
+ void indent(int Levels = 1) { IndentLevel += Levels; }
+
+ void unindent(int Levels = 1) {
+ IndentLevel = std::max(0, IndentLevel - Levels);
+ }
+
+ void resetIndent() { IndentLevel = 0; }
+
+ int getIndentLevel() { return IndentLevel; }
+
+ void setPrefix(StringRef P) { Prefix = P; }
+
+ void printIndent() {
+ OS << Prefix;
+ for (int i = 0; i < IndentLevel; ++i)
+ OS << " ";
+ }
+
+ template <typename T> HexNumber hex(T Value) { return HexNumber(Value); }
+
+ template <typename T, typename TEnum>
+ void printEnum(StringRef Label, T Value,
+ ArrayRef<EnumEntry<TEnum>> EnumValues) {
+ StringRef Name;
+ bool Found = false;
+ for (const auto &EnumItem : EnumValues) {
+ if (EnumItem.Value == Value) {
+ Name = EnumItem.Name;
+ Found = true;
+ break;
+ }
+ }
+
+ if (Found) {
+ startLine() << Label << ": " << Name << " (" << hex(Value) << ")\n";
+ } else {
+ startLine() << Label << ": " << hex(Value) << "\n";
+ }
+ }
+
+ template <typename T, typename TFlag>
+ void printFlags(StringRef Label, T Value, ArrayRef<EnumEntry<TFlag>> Flags,
+ TFlag EnumMask1 = {}, TFlag EnumMask2 = {},
+ TFlag EnumMask3 = {}) {
+ typedef EnumEntry<TFlag> FlagEntry;
+ typedef SmallVector<FlagEntry, 10> FlagVector;
+ FlagVector SetFlags;
+
+ for (const auto &Flag : Flags) {
+ if (Flag.Value == 0)
+ continue;
+
+ TFlag EnumMask{};
+ if (Flag.Value & EnumMask1)
+ EnumMask = EnumMask1;
+ else if (Flag.Value & EnumMask2)
+ EnumMask = EnumMask2;
+ else if (Flag.Value & EnumMask3)
+ EnumMask = EnumMask3;
+ bool IsEnum = (Flag.Value & EnumMask) != 0;
+ if ((!IsEnum && (Value & Flag.Value) == Flag.Value) ||
+ (IsEnum && (Value & EnumMask) == Flag.Value)) {
+ SetFlags.push_back(Flag);
+ }
+ }
+
+ llvm::sort(SetFlags, &flagName<TFlag>);
+
+ startLine() << Label << " [ (" << hex(Value) << ")\n";
+ for (const auto &Flag : SetFlags) {
+ startLine() << " " << Flag.Name << " (" << hex(Flag.Value) << ")\n";
+ }
+ startLine() << "]\n";
+ }
+
+ template <typename T> void printFlags(StringRef Label, T Value) {
+ startLine() << Label << " [ (" << hex(Value) << ")\n";
+ uint64_t Flag = 1;
+ uint64_t Curr = Value;
+ while (Curr > 0) {
+ if (Curr & 1)
+ startLine() << " " << hex(Flag) << "\n";
+ Curr >>= 1;
+ Flag <<= 1;
+ }
+ startLine() << "]\n";
+ }
+
+ void printNumber(StringRef Label, uint64_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, uint32_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, uint16_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, uint8_t Value) {
+ startLine() << Label << ": " << unsigned(Value) << "\n";
+ }
+
+ void printNumber(StringRef Label, int64_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, int32_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, int16_t Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printNumber(StringRef Label, int8_t Value) {
+ startLine() << Label << ": " << int(Value) << "\n";
+ }
+
+ void printNumber(StringRef Label, const APSInt &Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printBoolean(StringRef Label, bool Value) {
+ startLine() << Label << ": " << (Value ? "Yes" : "No") << '\n';
+ }
+
+ template <typename... T> void printVersion(StringRef Label, T... Version) {
+ startLine() << Label << ": ";
+ printVersionInternal(Version...);
+ getOStream() << "\n";
+ }
+
+ template <typename T> void printList(StringRef Label, const T &List) {
+ startLine() << Label << ": [";
+ bool Comma = false;
+ for (const auto &Item : List) {
+ if (Comma)
+ OS << ", ";
+ OS << Item;
+ Comma = true;
+ }
+ OS << "]\n";
+ }
+
+ template <typename T, typename U>
+ void printList(StringRef Label, const T &List, const U &Printer) {
+ startLine() << Label << ": [";
+ bool Comma = false;
+ for (const auto &Item : List) {
+ if (Comma)
+ OS << ", ";
+ Printer(OS, Item);
+ Comma = true;
+ }
+ OS << "]\n";
+ }
+
+ template <typename T> void printHexList(StringRef Label, const T &List) {
+ startLine() << Label << ": [";
+ bool Comma = false;
+ for (const auto &Item : List) {
+ if (Comma)
+ OS << ", ";
+ OS << hex(Item);
+ Comma = true;
+ }
+ OS << "]\n";
+ }
+
+ template <typename T> void printHex(StringRef Label, T Value) {
+ startLine() << Label << ": " << hex(Value) << "\n";
+ }
+
+ template <typename T> void printHex(StringRef Label, StringRef Str, T Value) {
+ startLine() << Label << ": " << Str << " (" << hex(Value) << ")\n";
+ }
+
+ template <typename T>
+ void printSymbolOffset(StringRef Label, StringRef Symbol, T Value) {
+ startLine() << Label << ": " << Symbol << '+' << hex(Value) << '\n';
+ }
+
+ void printString(StringRef Value) { startLine() << Value << "\n"; }
+
+ void printString(StringRef Label, StringRef Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ void printString(StringRef Label, const std::string &Value) {
+ printString(Label, StringRef(Value));
+ }
+
+ void printString(StringRef Label, const char* Value) {
+ printString(Label, StringRef(Value));
+ }
+
+ template <typename T>
+ void printNumber(StringRef Label, StringRef Str, T Value) {
+ startLine() << Label << ": " << Str << " (" << Value << ")\n";
+ }
+
+ void printBinary(StringRef Label, StringRef Str, ArrayRef<uint8_t> Value) {
+ printBinaryImpl(Label, Str, Value, false);
+ }
+
+ void printBinary(StringRef Label, StringRef Str, ArrayRef<char> Value) {
+ auto V = makeArrayRef(reinterpret_cast<const uint8_t *>(Value.data()),
+ Value.size());
+ printBinaryImpl(Label, Str, V, false);
+ }
+
+ void printBinary(StringRef Label, ArrayRef<uint8_t> Value) {
+ printBinaryImpl(Label, StringRef(), Value, false);
+ }
+
+ void printBinary(StringRef Label, ArrayRef<char> Value) {
+ auto V = makeArrayRef(reinterpret_cast<const uint8_t *>(Value.data()),
+ Value.size());
+ printBinaryImpl(Label, StringRef(), V, false);
+ }
+
+ void printBinary(StringRef Label, StringRef Value) {
+ auto V = makeArrayRef(reinterpret_cast<const uint8_t *>(Value.data()),
+ Value.size());
+ printBinaryImpl(Label, StringRef(), V, false);
+ }
+
+ void printBinaryBlock(StringRef Label, ArrayRef<uint8_t> Value,
+ uint32_t StartOffset) {
+ printBinaryImpl(Label, StringRef(), Value, true, StartOffset);
+ }
+
+ void printBinaryBlock(StringRef Label, ArrayRef<uint8_t> Value) {
+ printBinaryImpl(Label, StringRef(), Value, true);
+ }
+
+ void printBinaryBlock(StringRef Label, StringRef Value) {
+ auto V = makeArrayRef(reinterpret_cast<const uint8_t *>(Value.data()),
+ Value.size());
+ printBinaryImpl(Label, StringRef(), V, true);
+ }
+
+ template <typename T> void printObject(StringRef Label, const T &Value) {
+ startLine() << Label << ": " << Value << "\n";
+ }
+
+ raw_ostream &startLine() {
+ printIndent();
+ return OS;
+ }
+
+ raw_ostream &getOStream() { return OS; }
+
+private:
+ template <typename T> void printVersionInternal(T Value) {
+ getOStream() << Value;
+ }
+
+ template <typename S, typename T, typename... TArgs>
+ void printVersionInternal(S Value, T Value2, TArgs... Args) {
+ getOStream() << Value << ".";
+ printVersionInternal(Value2, Args...);
+ }
+
+ template <typename T>
+ static bool flagName(const EnumEntry<T> &lhs, const EnumEntry<T> &rhs) {
+ return lhs.Name < rhs.Name;
+ }
+
+ void printBinaryImpl(StringRef Label, StringRef Str, ArrayRef<uint8_t> Value,
+ bool Block, uint32_t StartOffset = 0);
+
+ raw_ostream &OS;
+ int IndentLevel;
+ StringRef Prefix;
+};
+
+template <>
+inline void
+ScopedPrinter::printHex<support::ulittle16_t>(StringRef Label,
+ support::ulittle16_t Value) {
+ startLine() << Label << ": " << hex(Value) << "\n";
+}
+
+template<char Open, char Close>
+struct DelimitedScope {
+ explicit DelimitedScope(ScopedPrinter &W) : W(W) {
+ W.startLine() << Open << '\n';
+ W.indent();
+ }
+
+ DelimitedScope(ScopedPrinter &W, StringRef N) : W(W) {
+ W.startLine() << N;
+ if (!N.empty())
+ W.getOStream() << ' ';
+ W.getOStream() << Open << '\n';
+ W.indent();
+ }
+
+ ~DelimitedScope() {
+ W.unindent();
+ W.startLine() << Close << '\n';
+ }
+
+ ScopedPrinter &W;
+};
+
+using DictScope = DelimitedScope<'{', '}'>;
+using ListScope = DelimitedScope<'[', ']'>;
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/Signals.h b/third_party/llvm-project/include/llvm/Support/Signals.h
new file mode 100644
index 000000000..a6b215a24
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Signals.h
@@ -0,0 +1,90 @@
+//===- llvm/Support/Signals.h - Signal Handling support ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines some helpful functions for dealing with the possibility of
+// unix signals occurring while your program is running.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_SIGNALS_H
+#define LLVM_SUPPORT_SIGNALS_H
+
+#include <string>
+
+namespace llvm {
+class StringRef;
+class raw_ostream;
+
+namespace sys {
+
+ /// This function runs all the registered interrupt handlers, including the
+ /// removal of files registered by RemoveFileOnSignal.
+ void RunInterruptHandlers();
+
+ /// This function registers signal handlers to ensure that if a signal gets
+ /// delivered that the named file is removed.
+ /// Remove a file if a fatal signal occurs.
+ bool RemoveFileOnSignal(StringRef Filename, std::string* ErrMsg = nullptr);
+
+ /// This function removes a file from the list of files to be removed on
+ /// signal delivery.
+ void DontRemoveFileOnSignal(StringRef Filename);
+
+ /// When an error signal (such as SIGABRT or SIGSEGV) is delivered to the
+ /// process, print a stack trace and then exit.
+ /// Print a stack trace if a fatal signal occurs.
+ /// \param Argv0 the current binary name, used to find the symbolizer
+ /// relative to the current binary before searching $PATH; can be
+ /// StringRef(), in which case we will only search $PATH.
+ /// \param DisableCrashReporting if \c true, disable the normal crash
+ /// reporting mechanisms on the underlying operating system.
+ void PrintStackTraceOnErrorSignal(StringRef Argv0,
+ bool DisableCrashReporting = false);
+
+ /// Disable all system dialog boxes that appear when the process crashes.
+ void DisableSystemDialogsOnCrash();
+
+ /// Print the stack trace using the given \c raw_ostream object.
+ void PrintStackTrace(raw_ostream &OS);
+
+ // Run all registered signal handlers.
+ void RunSignalHandlers();
+
+ using SignalHandlerCallback = void (*)(void *);
+
+ /// Add a function to be called when an abort/kill signal is delivered to the
+ /// process. The handler can have a cookie passed to it to identify what
+ /// instance of the handler it is.
+ void AddSignalHandler(SignalHandlerCallback FnPtr, void *Cookie);
+
+ /// This function registers a function to be called when the user "interrupts"
+ /// the program (typically by pressing ctrl-c). When the user interrupts the
+ /// program, the specified interrupt function is called instead of the program
+ /// being killed, and the interrupt function automatically disabled.
+ ///
+ /// Note that interrupt functions are not allowed to call any non-reentrant
+ /// functions. An null interrupt function pointer disables the current
+ /// installed function. Note also that the handler may be executed on a
+ /// different thread on some platforms.
+ void SetInterruptFunction(void (*IF)());
+
+ /// Registers a function to be called when an "info" signal is delivered to
+ /// the process.
+ ///
+ /// On POSIX systems, this will be SIGUSR1; on systems that have it, SIGINFO
+ /// will also be used (typically ctrl-t).
+ ///
+ /// Note that signal handlers are not allowed to call any non-reentrant
+ /// functions. An null function pointer disables the current installed
+ /// function. Note also that the handler may be executed on a different
+ /// thread on some platforms.
+ void SetInfoSignalFunction(void (*Handler)());
+} // End sys namespace
+} // End llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/SmallVectorMemoryBuffer.h b/third_party/llvm-project/include/llvm/Support/SmallVectorMemoryBuffer.h
new file mode 100644
index 000000000..b63b58e3a
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/SmallVectorMemoryBuffer.h
@@ -0,0 +1,65 @@
+//===- SmallVectorMemoryBuffer.h --------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares a wrapper class to hold the memory into which an
+// object will be generated.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_EXECUTIONENGINE_OBJECTMEMORYBUFFER_H
+#define LLVM_EXECUTIONENGINE_OBJECTMEMORYBUFFER_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+
+/// SmallVector-backed MemoryBuffer instance.
+///
+/// This class enables efficient construction of MemoryBuffers from SmallVector
+/// instances. This is useful for MCJIT and Orc, where object files are streamed
+/// into SmallVectors, then inspected using ObjectFile (which takes a
+/// MemoryBuffer).
+class SmallVectorMemoryBuffer : public MemoryBuffer {
+public:
+ /// Construct an SmallVectorMemoryBuffer from the given SmallVector
+ /// r-value.
+ ///
+ /// FIXME: It'd be nice for this to be a non-templated constructor taking a
+ /// SmallVectorImpl here instead of a templated one taking a SmallVector<N>,
+ /// but SmallVector's move-construction/assignment currently only take
+ /// SmallVectors. If/when that is fixed we can simplify this constructor and
+ /// the following one.
+ SmallVectorMemoryBuffer(SmallVectorImpl<char> &&SV)
+ : SV(std::move(SV)), BufferName("<in-memory object>") {
+ init(this->SV.begin(), this->SV.end(), false);
+ }
+
+ /// Construct a named SmallVectorMemoryBuffer from the given
+ /// SmallVector r-value and StringRef.
+ SmallVectorMemoryBuffer(SmallVectorImpl<char> &&SV, StringRef Name)
+ : SV(std::move(SV)), BufferName(Name) {
+ init(this->SV.begin(), this->SV.end(), false);
+ }
+
+ // Key function.
+ ~SmallVectorMemoryBuffer() override;
+
+ StringRef getBufferIdentifier() const override { return BufferName; }
+
+ BufferKind getBufferKind() const override { return MemoryBuffer_Malloc; }
+
+private:
+ SmallVector<char, 0> SV;
+ std::string BufferName;
+};
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/SourceMgr.h b/third_party/llvm-project/include/llvm/Support/SourceMgr.h
new file mode 100644
index 000000000..aa6026c23
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/SourceMgr.h
@@ -0,0 +1,310 @@
+//===- SourceMgr.h - Manager for Source Buffers & Diagnostics ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the SMDiagnostic and SourceMgr classes. This
+// provides a simple substrate for diagnostics, #include handling, and other low
+// level things for simple parsers.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_SOURCEMGR_H
+#define LLVM_SUPPORT_SOURCEMGR_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/PointerUnion.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/SMLoc.h"
+#include <algorithm>
+#include <cassert>
+#include <memory>
+#include <string>
+#include <utility>
+#include <vector>
+
+namespace llvm {
+
+class raw_ostream;
+class SMDiagnostic;
+class SMFixIt;
+
+/// This owns the files read by a parser, handles include stacks,
+/// and handles diagnostic wrangling.
+class SourceMgr {
+public:
+ enum DiagKind {
+ DK_Error,
+ DK_Warning,
+ DK_Remark,
+ DK_Note,
+ };
+
+ /// Clients that want to handle their own diagnostics in a custom way can
+ /// register a function pointer+context as a diagnostic handler.
+ /// It gets called each time PrintMessage is invoked.
+ using DiagHandlerTy = void (*)(const SMDiagnostic &, void *Context);
+
+private:
+ struct SrcBuffer {
+ /// The memory buffer for the file.
+ std::unique_ptr<MemoryBuffer> Buffer;
+
+ /// Helper type for OffsetCache below: since we're storing many offsets
+ /// into relatively small files (often smaller than 2^8 or 2^16 bytes),
+ /// we select the offset vector element type dynamically based on the
+ /// size of Buffer.
+ using VariableSizeOffsets = PointerUnion4<std::vector<uint8_t> *,
+ std::vector<uint16_t> *,
+ std::vector<uint32_t> *,
+ std::vector<uint64_t> *>;
+
+ /// Vector of offsets into Buffer at which there are line-endings
+ /// (lazily populated). Once populated, the '\n' that marks the end of
+ /// line number N from [1..] is at Buffer[OffsetCache[N-1]]. Since
+ /// these offsets are in sorted (ascending) order, they can be
+ /// binary-searched for the first one after any given offset (eg. an
+ /// offset corresponding to a particular SMLoc).
+ mutable VariableSizeOffsets OffsetCache;
+
+ /// Populate \c OffsetCache and look up a given \p Ptr in it, assuming
+ /// it points somewhere into \c Buffer. The static type parameter \p T
+ /// must be an unsigned integer type from uint{8,16,32,64}_t large
+ /// enough to store offsets inside \c Buffer.
+ template<typename T>
+ unsigned getLineNumber(const char *Ptr) const;
+
+ /// This is the location of the parent include, or null if at the top level.
+ SMLoc IncludeLoc;
+
+ SrcBuffer() = default;
+ SrcBuffer(SrcBuffer &&);
+ SrcBuffer(const SrcBuffer &) = delete;
+ SrcBuffer &operator=(const SrcBuffer &) = delete;
+ ~SrcBuffer();
+ };
+
+ /// This is all of the buffers that we are reading from.
+ std::vector<SrcBuffer> Buffers;
+
+ // This is the list of directories we should search for include files in.
+ std::vector<std::string> IncludeDirectories;
+
+ DiagHandlerTy DiagHandler = nullptr;
+ void *DiagContext = nullptr;
+
+ bool isValidBufferID(unsigned i) const { return i && i <= Buffers.size(); }
+
+public:
+ SourceMgr() = default;
+ SourceMgr(const SourceMgr &) = delete;
+ SourceMgr &operator=(const SourceMgr &) = delete;
+ SourceMgr(SourceMgr &&) = default;
+ SourceMgr &operator=(SourceMgr &&) = default;
+ ~SourceMgr() = default;
+
+ void setIncludeDirs(const std::vector<std::string> &Dirs) {
+ IncludeDirectories = Dirs;
+ }
+
+ /// Specify a diagnostic handler to be invoked every time PrintMessage is
+ /// called. \p Ctx is passed into the handler when it is invoked.
+ void setDiagHandler(DiagHandlerTy DH, void *Ctx = nullptr) {
+ DiagHandler = DH;
+ DiagContext = Ctx;
+ }
+
+ DiagHandlerTy getDiagHandler() const { return DiagHandler; }
+ void *getDiagContext() const { return DiagContext; }
+
+ const SrcBuffer &getBufferInfo(unsigned i) const {
+ assert(isValidBufferID(i));
+ return Buffers[i - 1];
+ }
+
+ const MemoryBuffer *getMemoryBuffer(unsigned i) const {
+ assert(isValidBufferID(i));
+ return Buffers[i - 1].Buffer.get();
+ }
+
+ unsigned getNumBuffers() const {
+ return Buffers.size();
+ }
+
+ unsigned getMainFileID() const {
+ assert(getNumBuffers());
+ return 1;
+ }
+
+ SMLoc getParentIncludeLoc(unsigned i) const {
+ assert(isValidBufferID(i));
+ return Buffers[i - 1].IncludeLoc;
+ }
+
+ /// Add a new source buffer to this source manager. This takes ownership of
+ /// the memory buffer.
+ unsigned AddNewSourceBuffer(std::unique_ptr<MemoryBuffer> F,
+ SMLoc IncludeLoc) {
+ SrcBuffer NB;
+ NB.Buffer = std::move(F);
+ NB.IncludeLoc = IncludeLoc;
+ Buffers.push_back(std::move(NB));
+ return Buffers.size();
+ }
+
+ /// Search for a file with the specified name in the current directory or in
+ /// one of the IncludeDirs.
+ ///
+ /// If no file is found, this returns 0, otherwise it returns the buffer ID
+ /// of the stacked file. The full path to the included file can be found in
+ /// \p IncludedFile.
+ unsigned AddIncludeFile(const std::string &Filename, SMLoc IncludeLoc,
+ std::string &IncludedFile);
+
+ /// Return the ID of the buffer containing the specified location.
+ ///
+ /// 0 is returned if the buffer is not found.
+ unsigned FindBufferContainingLoc(SMLoc Loc) const;
+
+ /// Find the line number for the specified location in the specified file.
+ /// This is not a fast method.
+ unsigned FindLineNumber(SMLoc Loc, unsigned BufferID = 0) const {
+ return getLineAndColumn(Loc, BufferID).first;
+ }
+
+ /// Find the line and column number for the specified location in the
+ /// specified file. This is not a fast method.
+ std::pair<unsigned, unsigned> getLineAndColumn(SMLoc Loc,
+ unsigned BufferID = 0) const;
+
+ /// Emit a message about the specified location with the specified string.
+ ///
+ /// \param ShowColors Display colored messages if output is a terminal and
+ /// the default error handler is used.
+ void PrintMessage(raw_ostream &OS, SMLoc Loc, DiagKind Kind,
+ const Twine &Msg,
+ ArrayRef<SMRange> Ranges = None,
+ ArrayRef<SMFixIt> FixIts = None,
+ bool ShowColors = true) const;
+
+ /// Emits a diagnostic to llvm::errs().
+ void PrintMessage(SMLoc Loc, DiagKind Kind, const Twine &Msg,
+ ArrayRef<SMRange> Ranges = None,
+ ArrayRef<SMFixIt> FixIts = None,
+ bool ShowColors = true) const;
+
+ /// Emits a manually-constructed diagnostic to the given output stream.
+ ///
+ /// \param ShowColors Display colored messages if output is a terminal and
+ /// the default error handler is used.
+ void PrintMessage(raw_ostream &OS, const SMDiagnostic &Diagnostic,
+ bool ShowColors = true) const;
+
+ /// Return an SMDiagnostic at the specified location with the specified
+ /// string.
+ ///
+ /// \param Msg If non-null, the kind of message (e.g., "error") which is
+ /// prefixed to the message.
+ SMDiagnostic GetMessage(SMLoc Loc, DiagKind Kind, const Twine &Msg,
+ ArrayRef<SMRange> Ranges = None,
+ ArrayRef<SMFixIt> FixIts = None) const;
+
+ /// Prints the names of included files and the line of the file they were
+ /// included from. A diagnostic handler can use this before printing its
+ /// custom formatted message.
+ ///
+ /// \param IncludeLoc The location of the include.
+ /// \param OS the raw_ostream to print on.
+ void PrintIncludeStack(SMLoc IncludeLoc, raw_ostream &OS) const;
+};
+
+/// Represents a single fixit, a replacement of one range of text with another.
+class SMFixIt {
+ SMRange Range;
+
+ std::string Text;
+
+public:
+ // FIXME: Twine.str() is not very efficient.
+ SMFixIt(SMLoc Loc, const Twine &Insertion)
+ : Range(Loc, Loc), Text(Insertion.str()) {
+ assert(Loc.isValid());
+ }
+
+ // FIXME: Twine.str() is not very efficient.
+ SMFixIt(SMRange R, const Twine &Replacement)
+ : Range(R), Text(Replacement.str()) {
+ assert(R.isValid());
+ }
+
+ StringRef getText() const { return Text; }
+ SMRange getRange() const { return Range; }
+
+ bool operator<(const SMFixIt &Other) const {
+ if (Range.Start.getPointer() != Other.Range.Start.getPointer())
+ return Range.Start.getPointer() < Other.Range.Start.getPointer();
+ if (Range.End.getPointer() != Other.Range.End.getPointer())
+ return Range.End.getPointer() < Other.Range.End.getPointer();
+ return Text < Other.Text;
+ }
+};
+
+/// Instances of this class encapsulate one diagnostic report, allowing
+/// printing to a raw_ostream as a caret diagnostic.
+class SMDiagnostic {
+ const SourceMgr *SM = nullptr;
+ SMLoc Loc;
+ std::string Filename;
+ int LineNo = 0;
+ int ColumnNo = 0;
+ SourceMgr::DiagKind Kind = SourceMgr::DK_Error;
+ std::string Message, LineContents;
+ std::vector<std::pair<unsigned, unsigned>> Ranges;
+ SmallVector<SMFixIt, 4> FixIts;
+
+public:
+ // Null diagnostic.
+ SMDiagnostic() = default;
+ // Diagnostic with no location (e.g. file not found, command line arg error).
+ SMDiagnostic(StringRef filename, SourceMgr::DiagKind Knd, StringRef Msg)
+ : Filename(filename), LineNo(-1), ColumnNo(-1), Kind(Knd), Message(Msg) {}
+
+ // Diagnostic with a location.
+ SMDiagnostic(const SourceMgr &sm, SMLoc L, StringRef FN,
+ int Line, int Col, SourceMgr::DiagKind Kind,
+ StringRef Msg, StringRef LineStr,
+ ArrayRef<std::pair<unsigned,unsigned>> Ranges,
+ ArrayRef<SMFixIt> FixIts = None);
+
+ const SourceMgr *getSourceMgr() const { return SM; }
+ SMLoc getLoc() const { return Loc; }
+ StringRef getFilename() const { return Filename; }
+ int getLineNo() const { return LineNo; }
+ int getColumnNo() const { return ColumnNo; }
+ SourceMgr::DiagKind getKind() const { return Kind; }
+ StringRef getMessage() const { return Message; }
+ StringRef getLineContents() const { return LineContents; }
+ ArrayRef<std::pair<unsigned, unsigned>> getRanges() const { return Ranges; }
+
+ void addFixIt(const SMFixIt &Hint) {
+ FixIts.push_back(Hint);
+ }
+
+ ArrayRef<SMFixIt> getFixIts() const {
+ return FixIts;
+ }
+
+ void print(const char *ProgName, raw_ostream &S, bool ShowColors = true,
+ bool ShowKindLabel = true) const;
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_SOURCEMGR_H
diff --git a/third_party/llvm-project/include/llvm/Support/SwapByteOrder.h b/third_party/llvm-project/include/llvm/Support/SwapByteOrder.h
new file mode 100644
index 000000000..6cec87006
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/SwapByteOrder.h
@@ -0,0 +1,160 @@
+//===- SwapByteOrder.h - Generic and optimized byte swaps -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares generic and optimized functions to swap the byte order of
+// an integral type.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_SWAPBYTEORDER_H
+#define LLVM_SUPPORT_SWAPBYTEORDER_H
+
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/DataTypes.h"
+#include <cstddef>
+#include <type_traits>
+#if defined(_MSC_VER) && !defined(_DEBUG)
+#include <stdlib.h>
+#endif
+
+#if defined(__linux__) || defined(__GNU__) || defined(__HAIKU__)
+#include <endian.h>
+#elif defined(_AIX)
+#include <sys/machine.h>
+#elif defined(__sun)
+/* Solaris provides _BIG_ENDIAN/_LITTLE_ENDIAN selector in sys/types.h */
+#include <sys/types.h>
+#define BIG_ENDIAN 4321
+#define LITTLE_ENDIAN 1234
+#if defined(_BIG_ENDIAN)
+#define BYTE_ORDER BIG_ENDIAN
+#else
+#define BYTE_ORDER LITTLE_ENDIAN
+#endif
+#else
+#if !defined(BYTE_ORDER) && !defined(_WIN32)
+#include <machine/endian.h>
+#endif
+#endif
+
+namespace llvm {
+namespace sys {
+
+#if defined(BYTE_ORDER) && defined(BIG_ENDIAN) && BYTE_ORDER == BIG_ENDIAN
+constexpr bool IsBigEndianHost = true;
+#else
+constexpr bool IsBigEndianHost = false;
+#endif
+
+static const bool IsLittleEndianHost = !IsBigEndianHost;
+
+/// SwapByteOrder_16 - This function returns a byte-swapped representation of
+/// the 16-bit argument.
+inline uint16_t SwapByteOrder_16(uint16_t value) {
+#if defined(_MSC_VER) && !defined(_DEBUG)
+ // The DLL version of the runtime lacks these functions (bug!?), but in a
+ // release build they're replaced with BSWAP instructions anyway.
+ return _byteswap_ushort(value);
+#else
+ uint16_t Hi = value << 8;
+ uint16_t Lo = value >> 8;
+ return Hi | Lo;
+#endif
+}
+
+/// This function returns a byte-swapped representation of the 32-bit argument.
+inline uint32_t SwapByteOrder_32(uint32_t value) {
+#if defined(__llvm__) || (defined(__GNUC__) && !defined(__ICC))
+ return __builtin_bswap32(value);
+#elif defined(_MSC_VER) && !defined(_DEBUG)
+ return _byteswap_ulong(value);
+#else
+ uint32_t Byte0 = value & 0x000000FF;
+ uint32_t Byte1 = value & 0x0000FF00;
+ uint32_t Byte2 = value & 0x00FF0000;
+ uint32_t Byte3 = value & 0xFF000000;
+ return (Byte0 << 24) | (Byte1 << 8) | (Byte2 >> 8) | (Byte3 >> 24);
+#endif
+}
+
+/// This function returns a byte-swapped representation of the 64-bit argument.
+inline uint64_t SwapByteOrder_64(uint64_t value) {
+#if defined(__llvm__) || (defined(__GNUC__) && !defined(__ICC))
+ return __builtin_bswap64(value);
+#elif defined(_MSC_VER) && !defined(_DEBUG)
+ return _byteswap_uint64(value);
+#else
+ uint64_t Hi = SwapByteOrder_32(uint32_t(value));
+ uint32_t Lo = SwapByteOrder_32(uint32_t(value >> 32));
+ return (Hi << 32) | Lo;
+#endif
+}
+
+inline unsigned char getSwappedBytes(unsigned char C) { return C; }
+inline signed char getSwappedBytes(signed char C) { return C; }
+inline char getSwappedBytes(char C) { return C; }
+
+inline unsigned short getSwappedBytes(unsigned short C) { return SwapByteOrder_16(C); }
+inline signed short getSwappedBytes( signed short C) { return SwapByteOrder_16(C); }
+
+inline unsigned int getSwappedBytes(unsigned int C) { return SwapByteOrder_32(C); }
+inline signed int getSwappedBytes( signed int C) { return SwapByteOrder_32(C); }
+
+#if __LONG_MAX__ == __INT_MAX__
+inline unsigned long getSwappedBytes(unsigned long C) { return SwapByteOrder_32(C); }
+inline signed long getSwappedBytes( signed long C) { return SwapByteOrder_32(C); }
+#elif __LONG_MAX__ == __LONG_LONG_MAX__
+inline unsigned long getSwappedBytes(unsigned long C) { return SwapByteOrder_64(C); }
+inline signed long getSwappedBytes( signed long C) { return SwapByteOrder_64(C); }
+#else
+#error "Unknown long size!"
+#endif
+
+inline unsigned long long getSwappedBytes(unsigned long long C) {
+ return SwapByteOrder_64(C);
+}
+inline signed long long getSwappedBytes(signed long long C) {
+ return SwapByteOrder_64(C);
+}
+
+inline float getSwappedBytes(float C) {
+ union {
+ uint32_t i;
+ float f;
+ } in, out;
+ in.f = C;
+ out.i = SwapByteOrder_32(in.i);
+ return out.f;
+}
+
+inline double getSwappedBytes(double C) {
+ union {
+ uint64_t i;
+ double d;
+ } in, out;
+ in.d = C;
+ out.i = SwapByteOrder_64(in.i);
+ return out.d;
+}
+
+template <typename T>
+inline typename std::enable_if<std::is_enum<T>::value, T>::type
+getSwappedBytes(T C) {
+ return static_cast<T>(
+ getSwappedBytes(static_cast<typename std::underlying_type<T>::type>(C)));
+}
+
+template<typename T>
+inline void swapByteOrder(T &Value) {
+ Value = getSwappedBytes(Value);
+}
+
+} // end namespace sys
+} // end namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/TargetParser.h b/third_party/llvm-project/include/llvm/Support/TargetParser.h
new file mode 100644
index 000000000..a7e1a752d
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/TargetParser.h
@@ -0,0 +1,174 @@
+//===-- TargetParser - Parser for target features ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements a target parser to recognise hardware features such as
+// FPU/CPU/ARCH names as well as specific support such as HDIV, etc.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_TARGETPARSER_H
+#define LLVM_SUPPORT_TARGETPARSER_H
+
+// FIXME: vector is used because that's what clang uses for subtarget feature
+// lists, but SmallVector would probably be better
+#include "llvm/ADT/Triple.h"
+#include "llvm/Support/ARMTargetParser.h"
+#include "llvm/Support/AArch64TargetParser.h"
+#include <vector>
+
+namespace llvm {
+class StringRef;
+
+// Target specific information in their own namespaces.
+// (ARM/AArch64 are declared in ARM/AArch64TargetParser.h)
+// These should be generated from TableGen because the information is already
+// there, and there is where new information about targets will be added.
+// FIXME: To TableGen this we need to make some table generated files available
+// even if the back-end is not compiled with LLVM, plus we need to create a new
+// back-end to TableGen to create these clean tables.
+namespace X86 {
+
+// This should be kept in sync with libcc/compiler-rt as its included by clang
+// as a proxy for what's in libgcc/compiler-rt.
+enum ProcessorVendors : unsigned {
+ VENDOR_DUMMY,
+#define X86_VENDOR(ENUM, STRING) \
+ ENUM,
+#include "llvm/Support/X86TargetParser.def"
+ VENDOR_OTHER
+};
+
+// This should be kept in sync with libcc/compiler-rt as its included by clang
+// as a proxy for what's in libgcc/compiler-rt.
+enum ProcessorTypes : unsigned {
+ CPU_TYPE_DUMMY,
+#define X86_CPU_TYPE(ARCHNAME, ENUM) \
+ ENUM,
+#include "llvm/Support/X86TargetParser.def"
+ CPU_TYPE_MAX
+};
+
+// This should be kept in sync with libcc/compiler-rt as its included by clang
+// as a proxy for what's in libgcc/compiler-rt.
+enum ProcessorSubtypes : unsigned {
+ CPU_SUBTYPE_DUMMY,
+#define X86_CPU_SUBTYPE(ARCHNAME, ENUM) \
+ ENUM,
+#include "llvm/Support/X86TargetParser.def"
+ CPU_SUBTYPE_MAX
+};
+
+// This should be kept in sync with libcc/compiler-rt as it should be used
+// by clang as a proxy for what's in libgcc/compiler-rt.
+enum ProcessorFeatures {
+#define X86_FEATURE(VAL, ENUM) \
+ ENUM = VAL,
+#include "llvm/Support/X86TargetParser.def"
+
+};
+
+} // namespace X86
+
+namespace AMDGPU {
+
+/// GPU kinds supported by the AMDGPU target.
+enum GPUKind : uint32_t {
+ // Not specified processor.
+ GK_NONE = 0,
+
+ // R600-based processors.
+ GK_R600 = 1,
+ GK_R630 = 2,
+ GK_RS880 = 3,
+ GK_RV670 = 4,
+ GK_RV710 = 5,
+ GK_RV730 = 6,
+ GK_RV770 = 7,
+ GK_CEDAR = 8,
+ GK_CYPRESS = 9,
+ GK_JUNIPER = 10,
+ GK_REDWOOD = 11,
+ GK_SUMO = 12,
+ GK_BARTS = 13,
+ GK_CAICOS = 14,
+ GK_CAYMAN = 15,
+ GK_TURKS = 16,
+
+ GK_R600_FIRST = GK_R600,
+ GK_R600_LAST = GK_TURKS,
+
+ // AMDGCN-based processors.
+ GK_GFX600 = 32,
+ GK_GFX601 = 33,
+
+ GK_GFX700 = 40,
+ GK_GFX701 = 41,
+ GK_GFX702 = 42,
+ GK_GFX703 = 43,
+ GK_GFX704 = 44,
+
+ GK_GFX801 = 50,
+ GK_GFX802 = 51,
+ GK_GFX803 = 52,
+ GK_GFX810 = 53,
+
+ GK_GFX900 = 60,
+ GK_GFX902 = 61,
+ GK_GFX904 = 62,
+ GK_GFX906 = 63,
+ GK_GFX908 = 64,
+ GK_GFX909 = 65,
+
+ GK_GFX1010 = 71,
+ GK_GFX1011 = 72,
+ GK_GFX1012 = 73,
+
+ GK_AMDGCN_FIRST = GK_GFX600,
+ GK_AMDGCN_LAST = GK_GFX1012,
+};
+
+/// Instruction set architecture version.
+struct IsaVersion {
+ unsigned Major;
+ unsigned Minor;
+ unsigned Stepping;
+};
+
+// This isn't comprehensive for now, just things that are needed from the
+// frontend driver.
+enum ArchFeatureKind : uint32_t {
+ FEATURE_NONE = 0,
+
+ // These features only exist for r600, and are implied true for amdgcn.
+ FEATURE_FMA = 1 << 1,
+ FEATURE_LDEXP = 1 << 2,
+ FEATURE_FP64 = 1 << 3,
+
+ // Common features.
+ FEATURE_FAST_FMA_F32 = 1 << 4,
+ FEATURE_FAST_DENORMAL_F32 = 1 << 5
+};
+
+StringRef getArchNameAMDGCN(GPUKind AK);
+StringRef getArchNameR600(GPUKind AK);
+StringRef getCanonicalArchName(StringRef Arch);
+GPUKind parseArchAMDGCN(StringRef CPU);
+GPUKind parseArchR600(StringRef CPU);
+unsigned getArchAttrAMDGCN(GPUKind AK);
+unsigned getArchAttrR600(GPUKind AK);
+
+void fillValidArchListAMDGCN(SmallVectorImpl<StringRef> &Values);
+void fillValidArchListR600(SmallVectorImpl<StringRef> &Values);
+
+IsaVersion getIsaVersion(StringRef GPU);
+
+} // namespace AMDGPU
+
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/TargetRegistry.h b/third_party/llvm-project/include/llvm/Support/TargetRegistry.h
new file mode 100644
index 000000000..d91eabae8
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/TargetRegistry.h
@@ -0,0 +1,1209 @@
+//===- Support/TargetRegistry.h - Target Registration -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file exposes the TargetRegistry interface, which tools can use to access
+// the appropriate target specific classes (TargetMachine, AsmPrinter, etc.)
+// which have been registered.
+//
+// Target specific class implementations should register themselves using the
+// appropriate TargetRegistry interfaces.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_TARGETREGISTRY_H
+#define LLVM_SUPPORT_TARGETREGISTRY_H
+
+#include "llvm-c/DisassemblerTypes.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/Support/CodeGen.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/FormattedStream.h"
+#include <algorithm>
+#include <cassert>
+#include <cstddef>
+#include <iterator>
+#include <memory>
+#include <string>
+
+namespace llvm {
+
+class AsmPrinter;
+class MCAsmBackend;
+class MCAsmInfo;
+class MCAsmParser;
+class MCCodeEmitter;
+class MCContext;
+class MCDisassembler;
+class MCInstPrinter;
+class MCInstrAnalysis;
+class MCInstrInfo;
+class MCObjectWriter;
+class MCRegisterInfo;
+class MCRelocationInfo;
+class MCStreamer;
+class MCSubtargetInfo;
+class MCSymbolizer;
+class MCTargetAsmParser;
+class MCTargetOptions;
+class MCTargetStreamer;
+class raw_ostream;
+class raw_pwrite_stream;
+class TargetMachine;
+class TargetOptions;
+
+MCStreamer *createNullStreamer(MCContext &Ctx);
+// Takes ownership of \p TAB and \p CE.
+
+/// Create a machine code streamer which will print out assembly for the native
+/// target, suitable for compiling with a native assembler.
+///
+/// \param InstPrint - If given, the instruction printer to use. If not given
+/// the MCInst representation will be printed. This method takes ownership of
+/// InstPrint.
+///
+/// \param CE - If given, a code emitter to use to show the instruction
+/// encoding inline with the assembly. This method takes ownership of \p CE.
+///
+/// \param TAB - If given, a target asm backend to use to show the fixup
+/// information in conjunction with encoding information. This method takes
+/// ownership of \p TAB.
+///
+/// \param ShowInst - Whether to show the MCInst representation inline with
+/// the assembly.
+MCStreamer *
+createAsmStreamer(MCContext &Ctx, std::unique_ptr<formatted_raw_ostream> OS,
+ bool isVerboseAsm, bool useDwarfDirectory,
+ MCInstPrinter *InstPrint, std::unique_ptr<MCCodeEmitter> &&CE,
+ std::unique_ptr<MCAsmBackend> &&TAB, bool ShowInst);
+
+MCStreamer *createELFStreamer(MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&CE,
+ bool RelaxAll);
+MCStreamer *createMachOStreamer(MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&CE,
+ bool RelaxAll, bool DWARFMustBeAtTheEnd,
+ bool LabelSections = false);
+MCStreamer *createWasmStreamer(MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&CE,
+ bool RelaxAll);
+MCStreamer *createXCOFFStreamer(MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&CE,
+ bool RelaxAll);
+
+MCRelocationInfo *createMCRelocationInfo(const Triple &TT, MCContext &Ctx);
+
+MCSymbolizer *createMCSymbolizer(const Triple &TT, LLVMOpInfoCallback GetOpInfo,
+ LLVMSymbolLookupCallback SymbolLookUp,
+ void *DisInfo, MCContext *Ctx,
+ std::unique_ptr<MCRelocationInfo> &&RelInfo);
+
+/// Target - Wrapper for Target specific information.
+///
+/// For registration purposes, this is a POD type so that targets can be
+/// registered without the use of static constructors.
+///
+/// Targets should implement a single global instance of this class (which
+/// will be zero initialized), and pass that instance to the TargetRegistry as
+/// part of their initialization.
+class Target {
+public:
+ friend struct TargetRegistry;
+
+ using ArchMatchFnTy = bool (*)(Triple::ArchType Arch);
+
+ using MCAsmInfoCtorFnTy = MCAsmInfo *(*)(const MCRegisterInfo &MRI,
+ const Triple &TT,
+ const MCTargetOptions &Options);
+ using MCInstrInfoCtorFnTy = MCInstrInfo *(*)();
+ using MCInstrAnalysisCtorFnTy = MCInstrAnalysis *(*)(const MCInstrInfo *Info);
+ using MCRegInfoCtorFnTy = MCRegisterInfo *(*)(const Triple &TT);
+ using MCSubtargetInfoCtorFnTy = MCSubtargetInfo *(*)(const Triple &TT,
+ StringRef CPU,
+ StringRef Features);
+ using TargetMachineCtorTy = TargetMachine
+ *(*)(const Target &T, const Triple &TT, StringRef CPU, StringRef Features,
+ const TargetOptions &Options, Optional<Reloc::Model> RM,
+ Optional<CodeModel::Model> CM, CodeGenOpt::Level OL, bool JIT);
+ // If it weren't for layering issues (this header is in llvm/Support, but
+ // depends on MC?) this should take the Streamer by value rather than rvalue
+ // reference.
+ using AsmPrinterCtorTy = AsmPrinter *(*)(
+ TargetMachine &TM, std::unique_ptr<MCStreamer> &&Streamer);
+ using MCAsmBackendCtorTy = MCAsmBackend *(*)(const Target &T,
+ const MCSubtargetInfo &STI,
+ const MCRegisterInfo &MRI,
+ const MCTargetOptions &Options);
+ using MCAsmParserCtorTy = MCTargetAsmParser *(*)(
+ const MCSubtargetInfo &STI, MCAsmParser &P, const MCInstrInfo &MII,
+ const MCTargetOptions &Options);
+ using MCDisassemblerCtorTy = MCDisassembler *(*)(const Target &T,
+ const MCSubtargetInfo &STI,
+ MCContext &Ctx);
+ using MCInstPrinterCtorTy = MCInstPrinter *(*)(const Triple &T,
+ unsigned SyntaxVariant,
+ const MCAsmInfo &MAI,
+ const MCInstrInfo &MII,
+ const MCRegisterInfo &MRI);
+ using MCCodeEmitterCtorTy = MCCodeEmitter *(*)(const MCInstrInfo &II,
+ const MCRegisterInfo &MRI,
+ MCContext &Ctx);
+ using ELFStreamerCtorTy =
+ MCStreamer *(*)(const Triple &T, MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+ using MachOStreamerCtorTy =
+ MCStreamer *(*)(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+ bool DWARFMustBeAtTheEnd);
+ using COFFStreamerCtorTy =
+ MCStreamer *(*)(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+ bool IncrementalLinkerCompatible);
+ using WasmStreamerCtorTy =
+ MCStreamer *(*)(const Triple &T, MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+ using NullTargetStreamerCtorTy = MCTargetStreamer *(*)(MCStreamer &S);
+ using AsmTargetStreamerCtorTy = MCTargetStreamer *(*)(
+ MCStreamer &S, formatted_raw_ostream &OS, MCInstPrinter *InstPrint,
+ bool IsVerboseAsm);
+ using ObjectTargetStreamerCtorTy = MCTargetStreamer *(*)(
+ MCStreamer &S, const MCSubtargetInfo &STI);
+ using MCRelocationInfoCtorTy = MCRelocationInfo *(*)(const Triple &TT,
+ MCContext &Ctx);
+ using MCSymbolizerCtorTy = MCSymbolizer *(*)(
+ const Triple &TT, LLVMOpInfoCallback GetOpInfo,
+ LLVMSymbolLookupCallback SymbolLookUp, void *DisInfo, MCContext *Ctx,
+ std::unique_ptr<MCRelocationInfo> &&RelInfo);
+
+private:
+ /// Next - The next registered target in the linked list, maintained by the
+ /// TargetRegistry.
+ Target *Next;
+
+ /// The target function for checking if an architecture is supported.
+ ArchMatchFnTy ArchMatchFn;
+
+ /// Name - The target name.
+ const char *Name;
+
+ /// ShortDesc - A short description of the target.
+ const char *ShortDesc;
+
+ /// BackendName - The name of the backend implementation. This must match the
+ /// name of the 'def X : Target ...' in TableGen.
+ const char *BackendName;
+
+ /// HasJIT - Whether this target supports the JIT.
+ bool HasJIT;
+
+ /// MCAsmInfoCtorFn - Constructor function for this target's MCAsmInfo, if
+ /// registered.
+ MCAsmInfoCtorFnTy MCAsmInfoCtorFn;
+
+ /// MCInstrInfoCtorFn - Constructor function for this target's MCInstrInfo,
+ /// if registered.
+ MCInstrInfoCtorFnTy MCInstrInfoCtorFn;
+
+ /// MCInstrAnalysisCtorFn - Constructor function for this target's
+ /// MCInstrAnalysis, if registered.
+ MCInstrAnalysisCtorFnTy MCInstrAnalysisCtorFn;
+
+ /// MCRegInfoCtorFn - Constructor function for this target's MCRegisterInfo,
+ /// if registered.
+ MCRegInfoCtorFnTy MCRegInfoCtorFn;
+
+ /// MCSubtargetInfoCtorFn - Constructor function for this target's
+ /// MCSubtargetInfo, if registered.
+ MCSubtargetInfoCtorFnTy MCSubtargetInfoCtorFn;
+
+ /// TargetMachineCtorFn - Construction function for this target's
+ /// TargetMachine, if registered.
+ TargetMachineCtorTy TargetMachineCtorFn;
+
+ /// MCAsmBackendCtorFn - Construction function for this target's
+ /// MCAsmBackend, if registered.
+ MCAsmBackendCtorTy MCAsmBackendCtorFn;
+
+ /// MCAsmParserCtorFn - Construction function for this target's
+ /// MCTargetAsmParser, if registered.
+ MCAsmParserCtorTy MCAsmParserCtorFn;
+
+ /// AsmPrinterCtorFn - Construction function for this target's AsmPrinter,
+ /// if registered.
+ AsmPrinterCtorTy AsmPrinterCtorFn;
+
+ /// MCDisassemblerCtorFn - Construction function for this target's
+ /// MCDisassembler, if registered.
+ MCDisassemblerCtorTy MCDisassemblerCtorFn;
+
+ /// MCInstPrinterCtorFn - Construction function for this target's
+ /// MCInstPrinter, if registered.
+ MCInstPrinterCtorTy MCInstPrinterCtorFn;
+
+ /// MCCodeEmitterCtorFn - Construction function for this target's
+ /// CodeEmitter, if registered.
+ MCCodeEmitterCtorTy MCCodeEmitterCtorFn;
+
+ // Construction functions for the various object formats, if registered.
+ COFFStreamerCtorTy COFFStreamerCtorFn = nullptr;
+ MachOStreamerCtorTy MachOStreamerCtorFn = nullptr;
+ ELFStreamerCtorTy ELFStreamerCtorFn = nullptr;
+ WasmStreamerCtorTy WasmStreamerCtorFn = nullptr;
+
+ /// Construction function for this target's null TargetStreamer, if
+ /// registered (default = nullptr).
+ NullTargetStreamerCtorTy NullTargetStreamerCtorFn = nullptr;
+
+ /// Construction function for this target's asm TargetStreamer, if
+ /// registered (default = nullptr).
+ AsmTargetStreamerCtorTy AsmTargetStreamerCtorFn = nullptr;
+
+ /// Construction function for this target's obj TargetStreamer, if
+ /// registered (default = nullptr).
+ ObjectTargetStreamerCtorTy ObjectTargetStreamerCtorFn = nullptr;
+
+ /// MCRelocationInfoCtorFn - Construction function for this target's
+ /// MCRelocationInfo, if registered (default = llvm::createMCRelocationInfo)
+ MCRelocationInfoCtorTy MCRelocationInfoCtorFn = nullptr;
+
+ /// MCSymbolizerCtorFn - Construction function for this target's
+ /// MCSymbolizer, if registered (default = llvm::createMCSymbolizer)
+ MCSymbolizerCtorTy MCSymbolizerCtorFn = nullptr;
+
+public:
+ Target() = default;
+
+ /// @name Target Information
+ /// @{
+
+ // getNext - Return the next registered target.
+ const Target *getNext() const { return Next; }
+
+ /// getName - Get the target name.
+ const char *getName() const { return Name; }
+
+ /// getShortDescription - Get a short description of the target.
+ const char *getShortDescription() const { return ShortDesc; }
+
+ /// getBackendName - Get the backend name.
+ const char *getBackendName() const { return BackendName; }
+
+ /// @}
+ /// @name Feature Predicates
+ /// @{
+
+ /// hasJIT - Check if this targets supports the just-in-time compilation.
+ bool hasJIT() const { return HasJIT; }
+
+ /// hasTargetMachine - Check if this target supports code generation.
+ bool hasTargetMachine() const { return TargetMachineCtorFn != nullptr; }
+
+ /// hasMCAsmBackend - Check if this target supports .o generation.
+ bool hasMCAsmBackend() const { return MCAsmBackendCtorFn != nullptr; }
+
+ /// hasMCAsmParser - Check if this target supports assembly parsing.
+ bool hasMCAsmParser() const { return MCAsmParserCtorFn != nullptr; }
+
+ /// @}
+ /// @name Feature Constructors
+ /// @{
+
+ /// createMCAsmInfo - Create a MCAsmInfo implementation for the specified
+ /// target triple.
+ ///
+ /// \param TheTriple This argument is used to determine the target machine
+ /// feature set; it should always be provided. Generally this should be
+ /// either the target triple from the module, or the target triple of the
+ /// host if that does not exist.
+ MCAsmInfo *createMCAsmInfo(const MCRegisterInfo &MRI, StringRef TheTriple,
+ const MCTargetOptions &Options) const {
+ if (!MCAsmInfoCtorFn)
+ return nullptr;
+ return MCAsmInfoCtorFn(MRI, Triple(TheTriple), Options);
+ }
+
+ /// createMCInstrInfo - Create a MCInstrInfo implementation.
+ ///
+ MCInstrInfo *createMCInstrInfo() const {
+ if (!MCInstrInfoCtorFn)
+ return nullptr;
+ return MCInstrInfoCtorFn();
+ }
+
+ /// createMCInstrAnalysis - Create a MCInstrAnalysis implementation.
+ ///
+ MCInstrAnalysis *createMCInstrAnalysis(const MCInstrInfo *Info) const {
+ if (!MCInstrAnalysisCtorFn)
+ return nullptr;
+ return MCInstrAnalysisCtorFn(Info);
+ }
+
+ /// createMCRegInfo - Create a MCRegisterInfo implementation.
+ ///
+ MCRegisterInfo *createMCRegInfo(StringRef TT) const {
+ if (!MCRegInfoCtorFn)
+ return nullptr;
+ return MCRegInfoCtorFn(Triple(TT));
+ }
+
+ /// createMCSubtargetInfo - Create a MCSubtargetInfo implementation.
+ ///
+ /// \param TheTriple This argument is used to determine the target machine
+ /// feature set; it should always be provided. Generally this should be
+ /// either the target triple from the module, or the target triple of the
+ /// host if that does not exist.
+ /// \param CPU This specifies the name of the target CPU.
+ /// \param Features This specifies the string representation of the
+ /// additional target features.
+ MCSubtargetInfo *createMCSubtargetInfo(StringRef TheTriple, StringRef CPU,
+ StringRef Features) const {
+ if (!MCSubtargetInfoCtorFn)
+ return nullptr;
+ return MCSubtargetInfoCtorFn(Triple(TheTriple), CPU, Features);
+ }
+
+ /// createTargetMachine - Create a target specific machine implementation
+ /// for the specified \p Triple.
+ ///
+ /// \param TT This argument is used to determine the target machine
+ /// feature set; it should always be provided. Generally this should be
+ /// either the target triple from the module, or the target triple of the
+ /// host if that does not exist.
+ TargetMachine *createTargetMachine(StringRef TT, StringRef CPU,
+ StringRef Features,
+ const TargetOptions &Options,
+ Optional<Reloc::Model> RM,
+ Optional<CodeModel::Model> CM = None,
+ CodeGenOpt::Level OL = CodeGenOpt::Default,
+ bool JIT = false) const {
+ if (!TargetMachineCtorFn)
+ return nullptr;
+ return TargetMachineCtorFn(*this, Triple(TT), CPU, Features, Options, RM,
+ CM, OL, JIT);
+ }
+
+ /// createMCAsmBackend - Create a target specific assembly parser.
+ MCAsmBackend *createMCAsmBackend(const MCSubtargetInfo &STI,
+ const MCRegisterInfo &MRI,
+ const MCTargetOptions &Options) const {
+ if (!MCAsmBackendCtorFn)
+ return nullptr;
+ return MCAsmBackendCtorFn(*this, STI, MRI, Options);
+ }
+
+ /// createMCAsmParser - Create a target specific assembly parser.
+ ///
+ /// \param Parser The target independent parser implementation to use for
+ /// parsing and lexing.
+ MCTargetAsmParser *createMCAsmParser(const MCSubtargetInfo &STI,
+ MCAsmParser &Parser,
+ const MCInstrInfo &MII,
+ const MCTargetOptions &Options) const {
+ if (!MCAsmParserCtorFn)
+ return nullptr;
+ return MCAsmParserCtorFn(STI, Parser, MII, Options);
+ }
+
+ /// createAsmPrinter - Create a target specific assembly printer pass. This
+ /// takes ownership of the MCStreamer object.
+ AsmPrinter *createAsmPrinter(TargetMachine &TM,
+ std::unique_ptr<MCStreamer> &&Streamer) const {
+ if (!AsmPrinterCtorFn)
+ return nullptr;
+ return AsmPrinterCtorFn(TM, std::move(Streamer));
+ }
+
+ MCDisassembler *createMCDisassembler(const MCSubtargetInfo &STI,
+ MCContext &Ctx) const {
+ if (!MCDisassemblerCtorFn)
+ return nullptr;
+ return MCDisassemblerCtorFn(*this, STI, Ctx);
+ }
+
+ MCInstPrinter *createMCInstPrinter(const Triple &T, unsigned SyntaxVariant,
+ const MCAsmInfo &MAI,
+ const MCInstrInfo &MII,
+ const MCRegisterInfo &MRI) const {
+ if (!MCInstPrinterCtorFn)
+ return nullptr;
+ return MCInstPrinterCtorFn(T, SyntaxVariant, MAI, MII, MRI);
+ }
+
+ /// createMCCodeEmitter - Create a target specific code emitter.
+ MCCodeEmitter *createMCCodeEmitter(const MCInstrInfo &II,
+ const MCRegisterInfo &MRI,
+ MCContext &Ctx) const {
+ if (!MCCodeEmitterCtorFn)
+ return nullptr;
+ return MCCodeEmitterCtorFn(II, MRI, Ctx);
+ }
+
+ /// Create a target specific MCStreamer.
+ ///
+ /// \param T The target triple.
+ /// \param Ctx The target context.
+ /// \param TAB The target assembler backend object. Takes ownership.
+ /// \param OW The stream object.
+ /// \param Emitter The target independent assembler object.Takes ownership.
+ /// \param RelaxAll Relax all fixups?
+ MCStreamer *createMCObjectStreamer(const Triple &T, MCContext &Ctx,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ std::unique_ptr<MCObjectWriter> &&OW,
+ std::unique_ptr<MCCodeEmitter> &&Emitter,
+ const MCSubtargetInfo &STI, bool RelaxAll,
+ bool IncrementalLinkerCompatible,
+ bool DWARFMustBeAtTheEnd) const {
+ MCStreamer *S = nullptr;
+ switch (T.getObjectFormat()) {
+ case Triple::UnknownObjectFormat:
+ llvm_unreachable("Unknown object format");
+ case Triple::COFF:
+ assert(T.isOSWindows() && "only Windows COFF is supported");
+ S = COFFStreamerCtorFn(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll,
+ IncrementalLinkerCompatible);
+ break;
+ case Triple::MachO:
+ if (MachOStreamerCtorFn)
+ S = MachOStreamerCtorFn(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll,
+ DWARFMustBeAtTheEnd);
+ else
+ S = createMachOStreamer(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll,
+ DWARFMustBeAtTheEnd);
+ break;
+ case Triple::ELF:
+ if (ELFStreamerCtorFn)
+ S = ELFStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll);
+ else
+ S = createELFStreamer(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll);
+ break;
+ case Triple::Wasm:
+ if (WasmStreamerCtorFn)
+ S = WasmStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll);
+ else
+ S = createWasmStreamer(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll);
+ break;
+ case Triple::XCOFF:
+ S = createXCOFFStreamer(Ctx, std::move(TAB), std::move(OW),
+ std::move(Emitter), RelaxAll);
+ break;
+ }
+ if (ObjectTargetStreamerCtorFn)
+ ObjectTargetStreamerCtorFn(*S, STI);
+ return S;
+ }
+
+ MCStreamer *createAsmStreamer(MCContext &Ctx,
+ std::unique_ptr<formatted_raw_ostream> OS,
+ bool IsVerboseAsm, bool UseDwarfDirectory,
+ MCInstPrinter *InstPrint,
+ std::unique_ptr<MCCodeEmitter> &&CE,
+ std::unique_ptr<MCAsmBackend> &&TAB,
+ bool ShowInst) const {
+ formatted_raw_ostream &OSRef = *OS;
+ MCStreamer *S = llvm::createAsmStreamer(
+ Ctx, std::move(OS), IsVerboseAsm, UseDwarfDirectory, InstPrint,
+ std::move(CE), std::move(TAB), ShowInst);
+ createAsmTargetStreamer(*S, OSRef, InstPrint, IsVerboseAsm);
+ return S;
+ }
+
+ MCTargetStreamer *createAsmTargetStreamer(MCStreamer &S,
+ formatted_raw_ostream &OS,
+ MCInstPrinter *InstPrint,
+ bool IsVerboseAsm) const {
+ if (AsmTargetStreamerCtorFn)
+ return AsmTargetStreamerCtorFn(S, OS, InstPrint, IsVerboseAsm);
+ return nullptr;
+ }
+
+ MCStreamer *createNullStreamer(MCContext &Ctx) const {
+ MCStreamer *S = llvm::createNullStreamer(Ctx);
+ createNullTargetStreamer(*S);
+ return S;
+ }
+
+ MCTargetStreamer *createNullTargetStreamer(MCStreamer &S) const {
+ if (NullTargetStreamerCtorFn)
+ return NullTargetStreamerCtorFn(S);
+ return nullptr;
+ }
+
+ /// createMCRelocationInfo - Create a target specific MCRelocationInfo.
+ ///
+ /// \param TT The target triple.
+ /// \param Ctx The target context.
+ MCRelocationInfo *createMCRelocationInfo(StringRef TT, MCContext &Ctx) const {
+ MCRelocationInfoCtorTy Fn = MCRelocationInfoCtorFn
+ ? MCRelocationInfoCtorFn
+ : llvm::createMCRelocationInfo;
+ return Fn(Triple(TT), Ctx);
+ }
+
+ /// createMCSymbolizer - Create a target specific MCSymbolizer.
+ ///
+ /// \param TT The target triple.
+ /// \param GetOpInfo The function to get the symbolic information for
+ /// operands.
+ /// \param SymbolLookUp The function to lookup a symbol name.
+ /// \param DisInfo The pointer to the block of symbolic information for above
+ /// call
+ /// back.
+ /// \param Ctx The target context.
+ /// \param RelInfo The relocation information for this target. Takes
+ /// ownership.
+ MCSymbolizer *
+ createMCSymbolizer(StringRef TT, LLVMOpInfoCallback GetOpInfo,
+ LLVMSymbolLookupCallback SymbolLookUp, void *DisInfo,
+ MCContext *Ctx,
+ std::unique_ptr<MCRelocationInfo> &&RelInfo) const {
+ MCSymbolizerCtorTy Fn =
+ MCSymbolizerCtorFn ? MCSymbolizerCtorFn : llvm::createMCSymbolizer;
+ return Fn(Triple(TT), GetOpInfo, SymbolLookUp, DisInfo, Ctx,
+ std::move(RelInfo));
+ }
+
+ /// @}
+};
+
+/// TargetRegistry - Generic interface to target specific features.
+struct TargetRegistry {
+ // FIXME: Make this a namespace, probably just move all the Register*
+ // functions into Target (currently they all just set members on the Target
+ // anyway, and Target friends this class so those functions can...
+ // function).
+ TargetRegistry() = delete;
+
+ class iterator
+ : public std::iterator<std::forward_iterator_tag, Target, ptrdiff_t> {
+ friend struct TargetRegistry;
+
+ const Target *Current = nullptr;
+
+ explicit iterator(Target *T) : Current(T) {}
+
+ public:
+ iterator() = default;
+
+ bool operator==(const iterator &x) const { return Current == x.Current; }
+ bool operator!=(const iterator &x) const { return !operator==(x); }
+
+ // Iterator traversal: forward iteration only
+ iterator &operator++() { // Preincrement
+ assert(Current && "Cannot increment end iterator!");
+ Current = Current->getNext();
+ return *this;
+ }
+ iterator operator++(int) { // Postincrement
+ iterator tmp = *this;
+ ++*this;
+ return tmp;
+ }
+
+ const Target &operator*() const {
+ assert(Current && "Cannot dereference end iterator!");
+ return *Current;
+ }
+
+ const Target *operator->() const { return &operator*(); }
+ };
+
+ /// printRegisteredTargetsForVersion - Print the registered targets
+ /// appropriately for inclusion in a tool's version output.
+ static void printRegisteredTargetsForVersion(raw_ostream &OS);
+
+ /// @name Registry Access
+ /// @{
+
+ static iterator_range<iterator> targets();
+
+ /// lookupTarget - Lookup a target based on a target triple.
+ ///
+ /// \param Triple - The triple to use for finding a target.
+ /// \param Error - On failure, an error string describing why no target was
+ /// found.
+ static const Target *lookupTarget(const std::string &Triple,
+ std::string &Error);
+
+ /// lookupTarget - Lookup a target based on an architecture name
+ /// and a target triple. If the architecture name is non-empty,
+ /// then the lookup is done by architecture. Otherwise, the target
+ /// triple is used.
+ ///
+ /// \param ArchName - The architecture to use for finding a target.
+ /// \param TheTriple - The triple to use for finding a target. The
+ /// triple is updated with canonical architecture name if a lookup
+ /// by architecture is done.
+ /// \param Error - On failure, an error string describing why no target was
+ /// found.
+ static const Target *lookupTarget(const std::string &ArchName,
+ Triple &TheTriple, std::string &Error);
+
+ /// @}
+ /// @name Target Registration
+ /// @{
+
+ /// RegisterTarget - Register the given target. Attempts to register a
+ /// target which has already been registered will be ignored.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Name - The target name. This should be a static string.
+ /// @param ShortDesc - A short target description. This should be a static
+ /// string.
+ /// @param BackendName - The name of the backend. This should be a static
+ /// string that is the same for all targets that share a backend
+ /// implementation and must match the name used in the 'def X : Target ...' in
+ /// TableGen.
+ /// @param ArchMatchFn - The arch match checking function for this target.
+ /// @param HasJIT - Whether the target supports JIT code
+ /// generation.
+ static void RegisterTarget(Target &T, const char *Name, const char *ShortDesc,
+ const char *BackendName,
+ Target::ArchMatchFnTy ArchMatchFn,
+ bool HasJIT = false);
+
+ /// RegisterMCAsmInfo - Register a MCAsmInfo implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct a MCAsmInfo for the target.
+ static void RegisterMCAsmInfo(Target &T, Target::MCAsmInfoCtorFnTy Fn) {
+ T.MCAsmInfoCtorFn = Fn;
+ }
+
+ /// RegisterMCInstrInfo - Register a MCInstrInfo implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct a MCInstrInfo for the target.
+ static void RegisterMCInstrInfo(Target &T, Target::MCInstrInfoCtorFnTy Fn) {
+ T.MCInstrInfoCtorFn = Fn;
+ }
+
+ /// RegisterMCInstrAnalysis - Register a MCInstrAnalysis implementation for
+ /// the given target.
+ static void RegisterMCInstrAnalysis(Target &T,
+ Target::MCInstrAnalysisCtorFnTy Fn) {
+ T.MCInstrAnalysisCtorFn = Fn;
+ }
+
+ /// RegisterMCRegInfo - Register a MCRegisterInfo implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct a MCRegisterInfo for the target.
+ static void RegisterMCRegInfo(Target &T, Target::MCRegInfoCtorFnTy Fn) {
+ T.MCRegInfoCtorFn = Fn;
+ }
+
+ /// RegisterMCSubtargetInfo - Register a MCSubtargetInfo implementation for
+ /// the given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct a MCSubtargetInfo for the target.
+ static void RegisterMCSubtargetInfo(Target &T,
+ Target::MCSubtargetInfoCtorFnTy Fn) {
+ T.MCSubtargetInfoCtorFn = Fn;
+ }
+
+ /// RegisterTargetMachine - Register a TargetMachine implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct a TargetMachine for the target.
+ static void RegisterTargetMachine(Target &T, Target::TargetMachineCtorTy Fn) {
+ T.TargetMachineCtorFn = Fn;
+ }
+
+ /// RegisterMCAsmBackend - Register a MCAsmBackend implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an AsmBackend for the target.
+ static void RegisterMCAsmBackend(Target &T, Target::MCAsmBackendCtorTy Fn) {
+ T.MCAsmBackendCtorFn = Fn;
+ }
+
+ /// RegisterMCAsmParser - Register a MCTargetAsmParser implementation for
+ /// the given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCTargetAsmParser for the target.
+ static void RegisterMCAsmParser(Target &T, Target::MCAsmParserCtorTy Fn) {
+ T.MCAsmParserCtorFn = Fn;
+ }
+
+ /// RegisterAsmPrinter - Register an AsmPrinter implementation for the given
+ /// target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an AsmPrinter for the target.
+ static void RegisterAsmPrinter(Target &T, Target::AsmPrinterCtorTy Fn) {
+ T.AsmPrinterCtorFn = Fn;
+ }
+
+ /// RegisterMCDisassembler - Register a MCDisassembler implementation for
+ /// the given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCDisassembler for the target.
+ static void RegisterMCDisassembler(Target &T,
+ Target::MCDisassemblerCtorTy Fn) {
+ T.MCDisassemblerCtorFn = Fn;
+ }
+
+ /// RegisterMCInstPrinter - Register a MCInstPrinter implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCInstPrinter for the target.
+ static void RegisterMCInstPrinter(Target &T, Target::MCInstPrinterCtorTy Fn) {
+ T.MCInstPrinterCtorFn = Fn;
+ }
+
+ /// RegisterMCCodeEmitter - Register a MCCodeEmitter implementation for the
+ /// given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCCodeEmitter for the target.
+ static void RegisterMCCodeEmitter(Target &T, Target::MCCodeEmitterCtorTy Fn) {
+ T.MCCodeEmitterCtorFn = Fn;
+ }
+
+ static void RegisterCOFFStreamer(Target &T, Target::COFFStreamerCtorTy Fn) {
+ T.COFFStreamerCtorFn = Fn;
+ }
+
+ static void RegisterMachOStreamer(Target &T, Target::MachOStreamerCtorTy Fn) {
+ T.MachOStreamerCtorFn = Fn;
+ }
+
+ static void RegisterELFStreamer(Target &T, Target::ELFStreamerCtorTy Fn) {
+ T.ELFStreamerCtorFn = Fn;
+ }
+
+ static void RegisterWasmStreamer(Target &T, Target::WasmStreamerCtorTy Fn) {
+ T.WasmStreamerCtorFn = Fn;
+ }
+
+ static void RegisterNullTargetStreamer(Target &T,
+ Target::NullTargetStreamerCtorTy Fn) {
+ T.NullTargetStreamerCtorFn = Fn;
+ }
+
+ static void RegisterAsmTargetStreamer(Target &T,
+ Target::AsmTargetStreamerCtorTy Fn) {
+ T.AsmTargetStreamerCtorFn = Fn;
+ }
+
+ static void
+ RegisterObjectTargetStreamer(Target &T,
+ Target::ObjectTargetStreamerCtorTy Fn) {
+ T.ObjectTargetStreamerCtorFn = Fn;
+ }
+
+ /// RegisterMCRelocationInfo - Register an MCRelocationInfo
+ /// implementation for the given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCRelocationInfo for the target.
+ static void RegisterMCRelocationInfo(Target &T,
+ Target::MCRelocationInfoCtorTy Fn) {
+ T.MCRelocationInfoCtorFn = Fn;
+ }
+
+ /// RegisterMCSymbolizer - Register an MCSymbolizer
+ /// implementation for the given target.
+ ///
+ /// Clients are responsible for ensuring that registration doesn't occur
+ /// while another thread is attempting to access the registry. Typically
+ /// this is done by initializing all targets at program startup.
+ ///
+ /// @param T - The target being registered.
+ /// @param Fn - A function to construct an MCSymbolizer for the target.
+ static void RegisterMCSymbolizer(Target &T, Target::MCSymbolizerCtorTy Fn) {
+ T.MCSymbolizerCtorFn = Fn;
+ }
+
+ /// @}
+};
+
+//===--------------------------------------------------------------------===//
+
+/// RegisterTarget - Helper template for registering a target, for use in the
+/// target's initialization function. Usage:
+///
+///
+/// Target &getTheFooTarget() { // The global target instance.
+/// static Target TheFooTarget;
+/// return TheFooTarget;
+/// }
+/// extern "C" void LLVMInitializeFooTargetInfo() {
+/// RegisterTarget<Triple::foo> X(getTheFooTarget(), "foo", "Foo
+/// description", "Foo" /* Backend Name */);
+/// }
+template <Triple::ArchType TargetArchType = Triple::UnknownArch,
+ bool HasJIT = false>
+struct RegisterTarget {
+ RegisterTarget(Target &T, const char *Name, const char *Desc,
+ const char *BackendName) {
+ TargetRegistry::RegisterTarget(T, Name, Desc, BackendName, &getArchMatch,
+ HasJIT);
+ }
+
+ static bool getArchMatch(Triple::ArchType Arch) {
+ return Arch == TargetArchType;
+ }
+};
+
+/// RegisterMCAsmInfo - Helper template for registering a target assembly info
+/// implementation. This invokes the static "Create" method on the class to
+/// actually do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCAsmInfo<FooMCAsmInfo> X(TheFooTarget);
+/// }
+template <class MCAsmInfoImpl> struct RegisterMCAsmInfo {
+ RegisterMCAsmInfo(Target &T) {
+ TargetRegistry::RegisterMCAsmInfo(T, &Allocator);
+ }
+
+private:
+ static MCAsmInfo *Allocator(const MCRegisterInfo & /*MRI*/, const Triple &TT,
+ const MCTargetOptions &Options) {
+ return new MCAsmInfoImpl(TT, Options);
+ }
+};
+
+/// RegisterMCAsmInfoFn - Helper template for registering a target assembly info
+/// implementation. This invokes the specified function to do the
+/// construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCAsmInfoFn X(TheFooTarget, TheFunction);
+/// }
+struct RegisterMCAsmInfoFn {
+ RegisterMCAsmInfoFn(Target &T, Target::MCAsmInfoCtorFnTy Fn) {
+ TargetRegistry::RegisterMCAsmInfo(T, Fn);
+ }
+};
+
+/// RegisterMCInstrInfo - Helper template for registering a target instruction
+/// info implementation. This invokes the static "Create" method on the class
+/// to actually do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCInstrInfo<FooMCInstrInfo> X(TheFooTarget);
+/// }
+template <class MCInstrInfoImpl> struct RegisterMCInstrInfo {
+ RegisterMCInstrInfo(Target &T) {
+ TargetRegistry::RegisterMCInstrInfo(T, &Allocator);
+ }
+
+private:
+ static MCInstrInfo *Allocator() { return new MCInstrInfoImpl(); }
+};
+
+/// RegisterMCInstrInfoFn - Helper template for registering a target
+/// instruction info implementation. This invokes the specified function to
+/// do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCInstrInfoFn X(TheFooTarget, TheFunction);
+/// }
+struct RegisterMCInstrInfoFn {
+ RegisterMCInstrInfoFn(Target &T, Target::MCInstrInfoCtorFnTy Fn) {
+ TargetRegistry::RegisterMCInstrInfo(T, Fn);
+ }
+};
+
+/// RegisterMCInstrAnalysis - Helper template for registering a target
+/// instruction analyzer implementation. This invokes the static "Create"
+/// method on the class to actually do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCInstrAnalysis<FooMCInstrAnalysis> X(TheFooTarget);
+/// }
+template <class MCInstrAnalysisImpl> struct RegisterMCInstrAnalysis {
+ RegisterMCInstrAnalysis(Target &T) {
+ TargetRegistry::RegisterMCInstrAnalysis(T, &Allocator);
+ }
+
+private:
+ static MCInstrAnalysis *Allocator(const MCInstrInfo *Info) {
+ return new MCInstrAnalysisImpl(Info);
+ }
+};
+
+/// RegisterMCInstrAnalysisFn - Helper template for registering a target
+/// instruction analyzer implementation. This invokes the specified function
+/// to do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCInstrAnalysisFn X(TheFooTarget, TheFunction);
+/// }
+struct RegisterMCInstrAnalysisFn {
+ RegisterMCInstrAnalysisFn(Target &T, Target::MCInstrAnalysisCtorFnTy Fn) {
+ TargetRegistry::RegisterMCInstrAnalysis(T, Fn);
+ }
+};
+
+/// RegisterMCRegInfo - Helper template for registering a target register info
+/// implementation. This invokes the static "Create" method on the class to
+/// actually do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCRegInfo<FooMCRegInfo> X(TheFooTarget);
+/// }
+template <class MCRegisterInfoImpl> struct RegisterMCRegInfo {
+ RegisterMCRegInfo(Target &T) {
+ TargetRegistry::RegisterMCRegInfo(T, &Allocator);
+ }
+
+private:
+ static MCRegisterInfo *Allocator(const Triple & /*TT*/) {
+ return new MCRegisterInfoImpl();
+ }
+};
+
+/// RegisterMCRegInfoFn - Helper template for registering a target register
+/// info implementation. This invokes the specified function to do the
+/// construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCRegInfoFn X(TheFooTarget, TheFunction);
+/// }
+struct RegisterMCRegInfoFn {
+ RegisterMCRegInfoFn(Target &T, Target::MCRegInfoCtorFnTy Fn) {
+ TargetRegistry::RegisterMCRegInfo(T, Fn);
+ }
+};
+
+/// RegisterMCSubtargetInfo - Helper template for registering a target
+/// subtarget info implementation. This invokes the static "Create" method
+/// on the class to actually do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCSubtargetInfo<FooMCSubtargetInfo> X(TheFooTarget);
+/// }
+template <class MCSubtargetInfoImpl> struct RegisterMCSubtargetInfo {
+ RegisterMCSubtargetInfo(Target &T) {
+ TargetRegistry::RegisterMCSubtargetInfo(T, &Allocator);
+ }
+
+private:
+ static MCSubtargetInfo *Allocator(const Triple & /*TT*/, StringRef /*CPU*/,
+ StringRef /*FS*/) {
+ return new MCSubtargetInfoImpl();
+ }
+};
+
+/// RegisterMCSubtargetInfoFn - Helper template for registering a target
+/// subtarget info implementation. This invokes the specified function to
+/// do the construction. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterMCSubtargetInfoFn X(TheFooTarget, TheFunction);
+/// }
+struct RegisterMCSubtargetInfoFn {
+ RegisterMCSubtargetInfoFn(Target &T, Target::MCSubtargetInfoCtorFnTy Fn) {
+ TargetRegistry::RegisterMCSubtargetInfo(T, Fn);
+ }
+};
+
+/// RegisterTargetMachine - Helper template for registering a target machine
+/// implementation, for use in the target machine initialization
+/// function. Usage:
+///
+/// extern "C" void LLVMInitializeFooTarget() {
+/// extern Target TheFooTarget;
+/// RegisterTargetMachine<FooTargetMachine> X(TheFooTarget);
+/// }
+template <class TargetMachineImpl> struct RegisterTargetMachine {
+ RegisterTargetMachine(Target &T) {
+ TargetRegistry::RegisterTargetMachine(T, &Allocator);
+ }
+
+private:
+ static TargetMachine *
+ Allocator(const Target &T, const Triple &TT, StringRef CPU, StringRef FS,
+ const TargetOptions &Options, Optional<Reloc::Model> RM,
+ Optional<CodeModel::Model> CM, CodeGenOpt::Level OL, bool JIT) {
+ return new TargetMachineImpl(T, TT, CPU, FS, Options, RM, CM, OL, JIT);
+ }
+};
+
+/// RegisterMCAsmBackend - Helper template for registering a target specific
+/// assembler backend. Usage:
+///
+/// extern "C" void LLVMInitializeFooMCAsmBackend() {
+/// extern Target TheFooTarget;
+/// RegisterMCAsmBackend<FooAsmLexer> X(TheFooTarget);
+/// }
+template <class MCAsmBackendImpl> struct RegisterMCAsmBackend {
+ RegisterMCAsmBackend(Target &T) {
+ TargetRegistry::RegisterMCAsmBackend(T, &Allocator);
+ }
+
+private:
+ static MCAsmBackend *Allocator(const Target &T, const MCSubtargetInfo &STI,
+ const MCRegisterInfo &MRI,
+ const MCTargetOptions &Options) {
+ return new MCAsmBackendImpl(T, STI, MRI);
+ }
+};
+
+/// RegisterMCAsmParser - Helper template for registering a target specific
+/// assembly parser, for use in the target machine initialization
+/// function. Usage:
+///
+/// extern "C" void LLVMInitializeFooMCAsmParser() {
+/// extern Target TheFooTarget;
+/// RegisterMCAsmParser<FooAsmParser> X(TheFooTarget);
+/// }
+template <class MCAsmParserImpl> struct RegisterMCAsmParser {
+ RegisterMCAsmParser(Target &T) {
+ TargetRegistry::RegisterMCAsmParser(T, &Allocator);
+ }
+
+private:
+ static MCTargetAsmParser *Allocator(const MCSubtargetInfo &STI,
+ MCAsmParser &P, const MCInstrInfo &MII,
+ const MCTargetOptions &Options) {
+ return new MCAsmParserImpl(STI, P, MII, Options);
+ }
+};
+
+/// RegisterAsmPrinter - Helper template for registering a target specific
+/// assembly printer, for use in the target machine initialization
+/// function. Usage:
+///
+/// extern "C" void LLVMInitializeFooAsmPrinter() {
+/// extern Target TheFooTarget;
+/// RegisterAsmPrinter<FooAsmPrinter> X(TheFooTarget);
+/// }
+template <class AsmPrinterImpl> struct RegisterAsmPrinter {
+ RegisterAsmPrinter(Target &T) {
+ TargetRegistry::RegisterAsmPrinter(T, &Allocator);
+ }
+
+private:
+ static AsmPrinter *Allocator(TargetMachine &TM,
+ std::unique_ptr<MCStreamer> &&Streamer) {
+ return new AsmPrinterImpl(TM, std::move(Streamer));
+ }
+};
+
+/// RegisterMCCodeEmitter - Helper template for registering a target specific
+/// machine code emitter, for use in the target initialization
+/// function. Usage:
+///
+/// extern "C" void LLVMInitializeFooMCCodeEmitter() {
+/// extern Target TheFooTarget;
+/// RegisterMCCodeEmitter<FooCodeEmitter> X(TheFooTarget);
+/// }
+template <class MCCodeEmitterImpl> struct RegisterMCCodeEmitter {
+ RegisterMCCodeEmitter(Target &T) {
+ TargetRegistry::RegisterMCCodeEmitter(T, &Allocator);
+ }
+
+private:
+ static MCCodeEmitter *Allocator(const MCInstrInfo & /*II*/,
+ const MCRegisterInfo & /*MRI*/,
+ MCContext & /*Ctx*/) {
+ return new MCCodeEmitterImpl();
+ }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_TARGETREGISTRY_H
diff --git a/third_party/llvm-project/include/llvm/Support/Threading.h b/third_party/llvm-project/include/llvm/Support/Threading.h
new file mode 100644
index 000000000..e69de29bb
diff --git a/third_party/llvm-project/include/llvm/Support/TypeSize.h b/third_party/llvm-project/include/llvm/Support/TypeSize.h
new file mode 100644
index 000000000..711679cdc
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/TypeSize.h
@@ -0,0 +1,201 @@
+//===- TypeSize.h - Wrapper around type sizes -------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides a struct that can be used to query the size of IR types
+// which may be scalable vectors. It provides convenience operators so that
+// it can be used in much the same way as a single scalar value.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_TYPESIZE_H
+#define LLVM_SUPPORT_TYPESIZE_H
+
+#include <cassert>
+#include <tuple>
+
+namespace llvm {
+
+class ElementCount {
+public:
+ unsigned Min; // Minimum number of vector elements.
+ bool Scalable; // If true, NumElements is a multiple of 'Min' determined
+ // at runtime rather than compile time.
+
+ ElementCount(unsigned Min, bool Scalable)
+ : Min(Min), Scalable(Scalable) {}
+
+ ElementCount operator*(unsigned RHS) {
+ return { Min * RHS, Scalable };
+ }
+ ElementCount operator/(unsigned RHS) {
+ return { Min / RHS, Scalable };
+ }
+
+ bool operator==(const ElementCount& RHS) const {
+ return Min == RHS.Min && Scalable == RHS.Scalable;
+ }
+ bool operator!=(const ElementCount& RHS) const {
+ return !(*this == RHS);
+ }
+};
+
+// This class is used to represent the size of types. If the type is of fixed
+// size, it will represent the exact size. If the type is a scalable vector,
+// it will represent the known minimum size.
+class TypeSize {
+ uint64_t MinSize; // The known minimum size.
+ bool IsScalable; // If true, then the runtime size is an integer multiple
+ // of MinSize.
+
+public:
+ constexpr TypeSize(uint64_t MinSize, bool Scalable)
+ : MinSize(MinSize), IsScalable(Scalable) {}
+
+ static constexpr TypeSize Fixed(uint64_t Size) {
+ return TypeSize(Size, /*IsScalable=*/false);
+ }
+
+ static constexpr TypeSize Scalable(uint64_t MinSize) {
+ return TypeSize(MinSize, /*IsScalable=*/true);
+ }
+
+ // Scalable vector types with the same minimum size as a fixed size type are
+ // not guaranteed to be the same size at runtime, so they are never
+ // considered to be equal.
+ friend bool operator==(const TypeSize &LHS, const TypeSize &RHS) {
+ return std::tie(LHS.MinSize, LHS.IsScalable) ==
+ std::tie(RHS.MinSize, RHS.IsScalable);
+ }
+
+ friend bool operator!=(const TypeSize &LHS, const TypeSize &RHS) {
+ return !(LHS == RHS);
+ }
+
+ // For many cases, size ordering between scalable and fixed size types cannot
+ // be determined at compile time, so such comparisons aren't allowed.
+ //
+ // e.g. <vscale x 2 x i16> could be bigger than <4 x i32> with a runtime
+ // vscale >= 5, equal sized with a vscale of 4, and smaller with
+ // a vscale <= 3.
+ //
+ // If the scalable flags match, just perform the requested comparison
+ // between the minimum sizes.
+ friend bool operator<(const TypeSize &LHS, const TypeSize &RHS) {
+ assert(LHS.IsScalable == RHS.IsScalable &&
+ "Ordering comparison of scalable and fixed types");
+
+ return LHS.MinSize < RHS.MinSize;
+ }
+
+ friend bool operator>(const TypeSize &LHS, const TypeSize &RHS) {
+ return RHS < LHS;
+ }
+
+ friend bool operator<=(const TypeSize &LHS, const TypeSize &RHS) {
+ return !(RHS < LHS);
+ }
+
+ friend bool operator>=(const TypeSize &LHS, const TypeSize& RHS) {
+ return !(LHS < RHS);
+ }
+
+ // Convenience operators to obtain relative sizes independently of
+ // the scalable flag.
+ TypeSize operator*(unsigned RHS) const {
+ return { MinSize * RHS, IsScalable };
+ }
+
+ friend TypeSize operator*(const unsigned LHS, const TypeSize &RHS) {
+ return { LHS * RHS.MinSize, RHS.IsScalable };
+ }
+
+ TypeSize operator/(unsigned RHS) const {
+ return { MinSize / RHS, IsScalable };
+ }
+
+ // Return the minimum size with the assumption that the size is exact.
+ // Use in places where a scalable size doesn't make sense (e.g. non-vector
+ // types, or vectors in backends which don't support scalable vectors).
+ uint64_t getFixedSize() const {
+ assert(!IsScalable && "Request for a fixed size on a scalable object");
+ return MinSize;
+ }
+
+ // Return the known minimum size. Use in places where the scalable property
+ // doesn't matter (e.g. determining alignment) or in conjunction with the
+ // isScalable method below.
+ uint64_t getKnownMinSize() const {
+ return MinSize;
+ }
+
+ // Return whether or not the size is scalable.
+ bool isScalable() const {
+ return IsScalable;
+ }
+
+ // Casts to a uint64_t if this is a fixed-width size.
+ //
+ // NOTE: This interface is obsolete and will be removed in a future version
+ // of LLVM in favour of calling getFixedSize() directly.
+ operator uint64_t() const {
+ return getFixedSize();
+ }
+
+ // Additional convenience operators needed to avoid ambiguous parses.
+ // TODO: Make uint64_t the default operator?
+ TypeSize operator*(uint64_t RHS) const {
+ return { MinSize * RHS, IsScalable };
+ }
+
+ TypeSize operator*(int RHS) const {
+ return { MinSize * RHS, IsScalable };
+ }
+
+ TypeSize operator*(int64_t RHS) const {
+ return { MinSize * RHS, IsScalable };
+ }
+
+ friend TypeSize operator*(const uint64_t LHS, const TypeSize &RHS) {
+ return { LHS * RHS.MinSize, RHS.IsScalable };
+ }
+
+ friend TypeSize operator*(const int LHS, const TypeSize &RHS) {
+ return { LHS * RHS.MinSize, RHS.IsScalable };
+ }
+
+ friend TypeSize operator*(const int64_t LHS, const TypeSize &RHS) {
+ return { LHS * RHS.MinSize, RHS.IsScalable };
+ }
+
+ TypeSize operator/(uint64_t RHS) const {
+ return { MinSize / RHS, IsScalable };
+ }
+
+ TypeSize operator/(int RHS) const {
+ return { MinSize / RHS, IsScalable };
+ }
+
+ TypeSize operator/(int64_t RHS) const {
+ return { MinSize / RHS, IsScalable };
+ }
+};
+
+/// Returns a TypeSize with a known minimum size that is the next integer
+/// (mod 2**64) that is greater than or equal to \p Value and is a multiple
+/// of \p Align. \p Align must be non-zero.
+///
+/// Similar to the alignTo functions in MathExtras.h
+inline TypeSize alignTo(TypeSize Size, uint64_t Align) {
+ assert(Align != 0u && "Align must be non-zero");
+ return {(Size.getKnownMinSize() + Align - 1) / Align * Align,
+ Size.isScalable()};
+}
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_TypeSize_H
diff --git a/third_party/llvm-project/include/llvm/Support/Unicode.h b/third_party/llvm-project/include/llvm/Support/Unicode.h
new file mode 100644
index 000000000..ca17bba2f
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/Unicode.h
@@ -0,0 +1,70 @@
+//===- llvm/Support/Unicode.h - Unicode character properties -*- C++ -*-=====//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines functions that allow querying certain properties of Unicode
+// characters.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_UNICODE_H
+#define LLVM_SUPPORT_UNICODE_H
+
+namespace llvm {
+class StringRef;
+
+namespace sys {
+namespace unicode {
+
+enum ColumnWidthErrors {
+ ErrorInvalidUTF8 = -2,
+ ErrorNonPrintableCharacter = -1
+};
+
+/// Determines if a character is likely to be displayed correctly on the
+/// terminal. Exact implementation would have to depend on the specific
+/// terminal, so we define the semantic that should be suitable for generic case
+/// of a terminal capable to output Unicode characters.
+///
+/// All characters from the Unicode code point range are considered printable
+/// except for:
+/// * C0 and C1 control character ranges;
+/// * default ignorable code points as per 5.21 of
+/// http://www.unicode.org/versions/Unicode6.2.0/UnicodeStandard-6.2.pdf
+/// except for U+00AD SOFT HYPHEN, as it's actually displayed on most
+/// terminals;
+/// * format characters (category = Cf);
+/// * surrogates (category = Cs);
+/// * unassigned characters (category = Cn).
+/// \return true if the character is considered printable.
+bool isPrintable(int UCS);
+
+/// Gets the number of positions the UTF8-encoded \p Text is likely to occupy
+/// when output on a terminal ("character width"). This depends on the
+/// implementation of the terminal, and there's no standard definition of
+/// character width.
+///
+/// The implementation defines it in a way that is expected to be compatible
+/// with a generic Unicode-capable terminal.
+///
+/// \return Character width:
+/// * ErrorNonPrintableCharacter (-1) if \p Text contains non-printable
+/// characters (as identified by isPrintable);
+/// * 0 for each non-spacing and enclosing combining mark;
+/// * 2 for each CJK character excluding halfwidth forms;
+/// * 1 for each of the remaining characters.
+int columnWidthUTF8(StringRef Text);
+
+/// Fold input unicode character according the Simple unicode case folding
+/// rules.
+int foldCharSimple(int C);
+
+} // namespace unicode
+} // namespace sys
+} // namespace llvm
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/UnicodeCharRanges.h b/third_party/llvm-project/include/llvm/Support/UnicodeCharRanges.h
new file mode 100644
index 000000000..73d3603b7
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/UnicodeCharRanges.h
@@ -0,0 +1,103 @@
+//===--- UnicodeCharRanges.h - Types and functions for character ranges ---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+#ifndef LLVM_SUPPORT_UNICODECHARRANGES_H
+#define LLVM_SUPPORT_UNICODECHARRANGES_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
+#include <algorithm>
+
+#define DEBUG_TYPE "unicode"
+
+namespace llvm {
+namespace sys {
+
+/// Represents a closed range of Unicode code points [Lower, Upper].
+struct UnicodeCharRange {
+ uint32_t Lower;
+ uint32_t Upper;
+};
+
+inline bool operator<(uint32_t Value, UnicodeCharRange Range) {
+ return Value < Range.Lower;
+}
+inline bool operator<(UnicodeCharRange Range, uint32_t Value) {
+ return Range.Upper < Value;
+}
+
+/// Holds a reference to an ordered array of UnicodeCharRange and allows
+/// to quickly check if a code point is contained in the set represented by this
+/// array.
+class UnicodeCharSet {
+public:
+ typedef ArrayRef<UnicodeCharRange> CharRanges;
+
+ /// Constructs a UnicodeCharSet instance from an array of
+ /// UnicodeCharRanges.
+ ///
+ /// Array pointed by \p Ranges should have the lifetime at least as long as
+ /// the UnicodeCharSet instance, and should not change. Array is validated by
+ /// the constructor, so it makes sense to create as few UnicodeCharSet
+ /// instances per each array of ranges, as possible.
+#ifdef NDEBUG
+
+ // FIXME: This could use constexpr + static_assert. This way we
+ // may get rid of NDEBUG in this header. Unfortunately there are some
+ // problems to get this working with MSVC 2013. Change this when
+ // the support for MSVC 2013 is dropped.
+ constexpr UnicodeCharSet(CharRanges Ranges) : Ranges(Ranges) {}
+#else
+ UnicodeCharSet(CharRanges Ranges) : Ranges(Ranges) {
+ assert(rangesAreValid());
+ }
+#endif
+
+ /// Returns true if the character set contains the Unicode code point
+ /// \p C.
+ bool contains(uint32_t C) const {
+ return std::binary_search(Ranges.begin(), Ranges.end(), C);
+ }
+
+private:
+ /// Returns true if each of the ranges is a proper closed range
+ /// [min, max], and if the ranges themselves are ordered and non-overlapping.
+ bool rangesAreValid() const {
+ uint32_t Prev = 0;
+ for (CharRanges::const_iterator I = Ranges.begin(), E = Ranges.end();
+ I != E; ++I) {
+ if (I != Ranges.begin() && Prev >= I->Lower) {
+ LLVM_DEBUG(dbgs() << "Upper bound 0x");
+ LLVM_DEBUG(dbgs().write_hex(Prev));
+ LLVM_DEBUG(dbgs() << " should be less than succeeding lower bound 0x");
+ LLVM_DEBUG(dbgs().write_hex(I->Lower) << "\n");
+ return false;
+ }
+ if (I->Upper < I->Lower) {
+ LLVM_DEBUG(dbgs() << "Upper bound 0x");
+ LLVM_DEBUG(dbgs().write_hex(I->Lower));
+ LLVM_DEBUG(dbgs() << " should not be less than lower bound 0x");
+ LLVM_DEBUG(dbgs().write_hex(I->Upper) << "\n");
+ return false;
+ }
+ Prev = I->Upper;
+ }
+
+ return true;
+ }
+
+ const CharRanges Ranges;
+};
+
+} // namespace sys
+} // namespace llvm
+
+#undef DEBUG_TYPE // "unicode"
+
+#endif // LLVM_SUPPORT_UNICODECHARRANGES_H
diff --git a/third_party/llvm-project/include/llvm/Support/WindowsError.h b/third_party/llvm-project/include/llvm/Support/WindowsError.h
new file mode 100644
index 000000000..195405224
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/WindowsError.h
@@ -0,0 +1,18 @@
+//===-- WindowsError.h - Support for mapping windows errors to posix-------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_WINDOWSERROR_H
+#define LLVM_SUPPORT_WINDOWSERROR_H
+
+#include <system_error>
+
+namespace llvm {
+std::error_code mapWindowsError(unsigned EV);
+}
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/WithColor.h b/third_party/llvm-project/include/llvm/Support/WithColor.h
new file mode 100644
index 000000000..2814f7e0e
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/WithColor.h
@@ -0,0 +1,118 @@
+//===- WithColor.h ----------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_WITHCOLOR_H
+#define LLVM_SUPPORT_WITHCOLOR_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+
+#if 0 // XXX BINARYEN
+extern cl::OptionCategory ColorCategory;
+#endif
+
+class raw_ostream;
+
+// Symbolic names for various syntax elements.
+enum class HighlightColor {
+ Address,
+ String,
+ Tag,
+ Attribute,
+ Enumerator,
+ Macro,
+ Error,
+ Warning,
+ Note,
+ Remark
+};
+
+/// An RAII object that temporarily switches an output stream to a specific
+/// color.
+class WithColor {
+ raw_ostream &OS;
+ bool DisableColors;
+
+public:
+ /// To be used like this: WithColor(OS, HighlightColor::String) << "text";
+ /// @param OS The output stream
+ /// @param S Symbolic name for syntax element to color
+ /// @param DisableColors Whether to ignore color changes regardless of -color
+ /// and support in OS
+ WithColor(raw_ostream &OS, HighlightColor S, bool DisableColors = false);
+ /// To be used like this: WithColor(OS, raw_ostream::Black) << "text";
+ /// @param OS The output stream
+ /// @param Color ANSI color to use, the special SAVEDCOLOR can be used to
+ /// change only the bold attribute, and keep colors untouched
+ /// @param Bold Bold/brighter text, default false
+ /// @param BG If true, change the background, default: change foreground
+ /// @param DisableColors Whether to ignore color changes regardless of -color
+ /// and support in OS
+ WithColor(raw_ostream &OS,
+ raw_ostream::Colors Color = raw_ostream::SAVEDCOLOR,
+ bool Bold = false, bool BG = false, bool DisableColors = false)
+ : OS(OS), DisableColors(DisableColors) {
+ changeColor(Color, Bold, BG);
+ }
+ ~WithColor();
+
+ raw_ostream &get() { return OS; }
+ operator raw_ostream &() { return OS; }
+ template <typename T> WithColor &operator<<(T &O) {
+ OS << O;
+ return *this;
+ }
+ template <typename T> WithColor &operator<<(const T &O) {
+ OS << O;
+ return *this;
+ }
+
+ /// Convenience method for printing "error: " to stderr.
+ static raw_ostream &error();
+ /// Convenience method for printing "warning: " to stderr.
+ static raw_ostream &warning();
+ /// Convenience method for printing "note: " to stderr.
+ static raw_ostream ¬e();
+ /// Convenience method for printing "remark: " to stderr.
+ static raw_ostream &remark();
+
+ /// Convenience method for printing "error: " to the given stream.
+ static raw_ostream &error(raw_ostream &OS, StringRef Prefix = "",
+ bool DisableColors = false);
+ /// Convenience method for printing "warning: " to the given stream.
+ static raw_ostream &warning(raw_ostream &OS, StringRef Prefix = "",
+ bool DisableColors = false);
+ /// Convenience method for printing "note: " to the given stream.
+ static raw_ostream ¬e(raw_ostream &OS, StringRef Prefix = "",
+ bool DisableColors = false);
+ /// Convenience method for printing "remark: " to the given stream.
+ static raw_ostream &remark(raw_ostream &OS, StringRef Prefix = "",
+ bool DisableColors = false);
+
+ /// Determine whether colors are displayed.
+ bool colorsEnabled();
+
+ /// Change the color of text that will be output from this point forward.
+ /// @param Color ANSI color to use, the special SAVEDCOLOR can be used to
+ /// change only the bold attribute, and keep colors untouched
+ /// @param Bold Bold/brighter text, default false
+ /// @param BG If true, change the background, default: change foreground
+ WithColor &changeColor(raw_ostream::Colors Color, bool Bold = false,
+ bool BG = false);
+
+ /// Reset the colors to terminal defaults. Call this when you are done
+ /// outputting colored text, or before program exit.
+ WithColor &resetColor();
+};
+
+} // end namespace llvm
+
+#endif // LLVM_LIB_DEBUGINFO_WITHCOLOR_H
diff --git a/third_party/llvm-project/include/llvm/Support/X86TargetParser.def b/third_party/llvm-project/include/llvm/Support/X86TargetParser.def
new file mode 100644
index 000000000..4ebf2d79c
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/X86TargetParser.def
@@ -0,0 +1,173 @@
+//===- X86TargetParser.def - X86 target parsing defines ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides defines to build up the X86 target parser's logic.
+//
+//===----------------------------------------------------------------------===//
+
+// NOTE: NO INCLUDE GUARD DESIRED!
+
+#ifndef X86_VENDOR
+#define X86_VENDOR(ENUM, STR)
+#endif
+X86_VENDOR(VENDOR_INTEL, "intel")
+X86_VENDOR(VENDOR_AMD, "amd")
+#undef X86_VENDOR
+
+// This macro is used to implement CPU types that have an alias. As of now
+// there is only ever one alias.
+#ifndef X86_CPU_TYPE_COMPAT_WITH_ALIAS
+#define X86_CPU_TYPE_COMPAT_WITH_ALIAS(ARCHNAME, ENUM, STR, ALIAS) X86_CPU_TYPE_COMPAT(ARCHNAME, ENUM, STR)
+#endif
+
+// This macro is used for cpu types present in compiler-rt/libgcc.
+#ifndef X86_CPU_TYPE_COMPAT
+#define X86_CPU_TYPE_COMPAT(ARCHNAME, ENUM, STR) X86_CPU_TYPE(ARCHNAME, ENUM)
+#endif
+
+#ifndef X86_CPU_TYPE
+#define X86_CPU_TYPE(ARCHNAME, ENUM)
+#endif
+// The first part of this list must match what is implemented in libgcc and
+// compilert-rt. Clang uses this to know how to implement __builtin_cpu_is.
+X86_CPU_TYPE_COMPAT_WITH_ALIAS("bonnell", INTEL_BONNELL, "bonnell", "atom")
+X86_CPU_TYPE_COMPAT ("core2", INTEL_CORE2, "core2")
+X86_CPU_TYPE_COMPAT ("nehalem", INTEL_COREI7, "corei7")
+X86_CPU_TYPE_COMPAT_WITH_ALIAS("amdfam10", AMDFAM10H, "amdfam10h", "amdfam10")
+X86_CPU_TYPE_COMPAT_WITH_ALIAS("bdver1", AMDFAM15H, "amdfam15h", "amdfam15")
+X86_CPU_TYPE_COMPAT_WITH_ALIAS("silvermont", INTEL_SILVERMONT, "silvermont", "slm")
+X86_CPU_TYPE_COMPAT ("knl", INTEL_KNL, "knl")
+X86_CPU_TYPE_COMPAT ("btver1", AMD_BTVER1, "btver1")
+X86_CPU_TYPE_COMPAT ("btver2", AMD_BTVER2, "btver2")
+X86_CPU_TYPE_COMPAT ("znver1", AMDFAM17H, "amdfam17h")
+X86_CPU_TYPE_COMPAT ("knm", INTEL_KNM, "knm")
+X86_CPU_TYPE_COMPAT ("goldmont", INTEL_GOLDMONT, "goldmont")
+X86_CPU_TYPE_COMPAT ("goldmont-plus", INTEL_GOLDMONT_PLUS, "goldmont-plus")
+X86_CPU_TYPE_COMPAT ("tremont", INTEL_TREMONT, "tremont")
+// Entries below this are not in libgcc/compiler-rt.
+X86_CPU_TYPE ("i386", INTEL_i386)
+X86_CPU_TYPE ("i486", INTEL_i486)
+X86_CPU_TYPE ("pentium", INTEL_PENTIUM)
+X86_CPU_TYPE ("pentium-mmx", INTEL_PENTIUM_MMX)
+X86_CPU_TYPE ("pentiumpro", INTEL_PENTIUM_PRO)
+X86_CPU_TYPE ("pentium2", INTEL_PENTIUM_II)
+X86_CPU_TYPE ("pentium3", INTEL_PENTIUM_III)
+X86_CPU_TYPE ("pentium4", INTEL_PENTIUM_IV)
+X86_CPU_TYPE ("pentium-m", INTEL_PENTIUM_M)
+X86_CPU_TYPE ("yonah", INTEL_CORE_DUO)
+X86_CPU_TYPE ("nocona", INTEL_NOCONA)
+X86_CPU_TYPE ("prescott", INTEL_PRESCOTT)
+X86_CPU_TYPE ("i486", AMD_i486)
+X86_CPU_TYPE ("pentium", AMDPENTIUM)
+X86_CPU_TYPE ("athlon", AMD_ATHLON)
+X86_CPU_TYPE ("athlon-xp", AMD_ATHLON_XP)
+X86_CPU_TYPE ("k8", AMD_K8)
+X86_CPU_TYPE ("k8-sse3", AMD_K8SSE3)
+#undef X86_CPU_TYPE_COMPAT_WITH_ALIAS
+#undef X86_CPU_TYPE_COMPAT
+#undef X86_CPU_TYPE
+
+// This macro is used for cpu subtypes present in compiler-rt/libgcc.
+#ifndef X86_CPU_SUBTYPE_COMPAT
+#define X86_CPU_SUBTYPE_COMPAT(ARCHNAME, ENUM, STR) X86_CPU_SUBTYPE(ARCHNAME, ENUM)
+#endif
+
+#ifndef X86_CPU_SUBTYPE
+#define X86_CPU_SUBTYPE(ARCHNAME, ENUM)
+#endif
+
+// The first part of this list must match what is implemented in libgcc and
+// compilert-rt. Clang uses this to know how to implement __builtin_cpu_is.
+X86_CPU_SUBTYPE_COMPAT("nehalem", INTEL_COREI7_NEHALEM, "nehalem")
+X86_CPU_SUBTYPE_COMPAT("westmere", INTEL_COREI7_WESTMERE, "westmere")
+X86_CPU_SUBTYPE_COMPAT("sandybridge", INTEL_COREI7_SANDYBRIDGE, "sandybridge")
+X86_CPU_SUBTYPE_COMPAT("amdfam10", AMDFAM10H_BARCELONA, "barcelona")
+X86_CPU_SUBTYPE_COMPAT("amdfam10", AMDFAM10H_SHANGHAI, "shanghai")
+X86_CPU_SUBTYPE_COMPAT("amdfam10", AMDFAM10H_ISTANBUL, "istanbul")
+X86_CPU_SUBTYPE_COMPAT("bdver1", AMDFAM15H_BDVER1, "bdver1")
+X86_CPU_SUBTYPE_COMPAT("bdver2", AMDFAM15H_BDVER2, "bdver2")
+X86_CPU_SUBTYPE_COMPAT("bdver3", AMDFAM15H_BDVER3, "bdver3")
+X86_CPU_SUBTYPE_COMPAT("bdver4", AMDFAM15H_BDVER4, "bdver4")
+X86_CPU_SUBTYPE_COMPAT("znver1", AMDFAM17H_ZNVER1, "znver1")
+X86_CPU_SUBTYPE_COMPAT("ivybridge", INTEL_COREI7_IVYBRIDGE, "ivybridge")
+X86_CPU_SUBTYPE_COMPAT("haswell", INTEL_COREI7_HASWELL, "haswell")
+X86_CPU_SUBTYPE_COMPAT("broadwell", INTEL_COREI7_BROADWELL, "broadwell")
+X86_CPU_SUBTYPE_COMPAT("skylake", INTEL_COREI7_SKYLAKE, "skylake")
+X86_CPU_SUBTYPE_COMPAT("skylake-avx512", INTEL_COREI7_SKYLAKE_AVX512, "skylake-avx512")
+X86_CPU_SUBTYPE_COMPAT("cannonlake", INTEL_COREI7_CANNONLAKE, "cannonlake")
+X86_CPU_SUBTYPE_COMPAT("icelake-client", INTEL_COREI7_ICELAKE_CLIENT, "icelake-client")
+X86_CPU_SUBTYPE_COMPAT("icelake-server", INTEL_COREI7_ICELAKE_SERVER, "icelake-server")
+X86_CPU_SUBTYPE_COMPAT("znver2", AMDFAM17H_ZNVER2, "znver2")
+X86_CPU_SUBTYPE_COMPAT("cascadelake", INTEL_COREI7_CASCADELAKE, "cascadelake")
+// Entries below this are not in libgcc/compiler-rt.
+X86_CPU_SUBTYPE ("core2", INTEL_CORE2_65)
+X86_CPU_SUBTYPE ("penryn", INTEL_CORE2_45)
+X86_CPU_SUBTYPE ("k6", AMDPENTIUM_K6)
+X86_CPU_SUBTYPE ("k6-2", AMDPENTIUM_K62)
+X86_CPU_SUBTYPE ("k6-3", AMDPENTIUM_K63)
+X86_CPU_SUBTYPE ("geode", AMDPENTIUM_GEODE)
+X86_CPU_SUBTYPE ("cooperlake", INTEL_COREI7_COOPERLAKE)
+X86_CPU_SUBTYPE ("tigerlake", INTEL_COREI7_TIGERLAKE)
+#undef X86_CPU_SUBTYPE_COMPAT
+#undef X86_CPU_SUBTYPE
+
+
+// This macro is used for cpu types present in compiler-rt/libgcc.
+#ifndef X86_FEATURE_COMPAT
+#define X86_FEATURE_COMPAT(VAL, ENUM, STR) X86_FEATURE(VAL, ENUM)
+#endif
+
+#ifndef X86_FEATURE
+#define X86_FEATURE(VAL, ENUM)
+#endif
+X86_FEATURE_COMPAT( 0, FEATURE_CMOV, "cmov")
+X86_FEATURE_COMPAT( 1, FEATURE_MMX, "mmx")
+X86_FEATURE_COMPAT( 2, FEATURE_POPCNT, "popcnt")
+X86_FEATURE_COMPAT( 3, FEATURE_SSE, "sse")
+X86_FEATURE_COMPAT( 4, FEATURE_SSE2, "sse2")
+X86_FEATURE_COMPAT( 5, FEATURE_SSE3, "sse3")
+X86_FEATURE_COMPAT( 6, FEATURE_SSSE3, "ssse3")
+X86_FEATURE_COMPAT( 7, FEATURE_SSE4_1, "sse4.1")
+X86_FEATURE_COMPAT( 8, FEATURE_SSE4_2, "sse4.2")
+X86_FEATURE_COMPAT( 9, FEATURE_AVX, "avx")
+X86_FEATURE_COMPAT(10, FEATURE_AVX2, "avx2")
+X86_FEATURE_COMPAT(11, FEATURE_SSE4_A, "sse4a")
+X86_FEATURE_COMPAT(12, FEATURE_FMA4, "fma4")
+X86_FEATURE_COMPAT(13, FEATURE_XOP, "xop")
+X86_FEATURE_COMPAT(14, FEATURE_FMA, "fma")
+X86_FEATURE_COMPAT(15, FEATURE_AVX512F, "avx512f")
+X86_FEATURE_COMPAT(16, FEATURE_BMI, "bmi")
+X86_FEATURE_COMPAT(17, FEATURE_BMI2, "bmi2")
+X86_FEATURE_COMPAT(18, FEATURE_AES, "aes")
+X86_FEATURE_COMPAT(19, FEATURE_PCLMUL, "pclmul")
+X86_FEATURE_COMPAT(20, FEATURE_AVX512VL, "avx512vl")
+X86_FEATURE_COMPAT(21, FEATURE_AVX512BW, "avx512bw")
+X86_FEATURE_COMPAT(22, FEATURE_AVX512DQ, "avx512dq")
+X86_FEATURE_COMPAT(23, FEATURE_AVX512CD, "avx512cd")
+X86_FEATURE_COMPAT(24, FEATURE_AVX512ER, "avx512er")
+X86_FEATURE_COMPAT(25, FEATURE_AVX512PF, "avx512pf")
+X86_FEATURE_COMPAT(26, FEATURE_AVX512VBMI, "avx512vbmi")
+X86_FEATURE_COMPAT(27, FEATURE_AVX512IFMA, "avx512ifma")
+X86_FEATURE_COMPAT(28, FEATURE_AVX5124VNNIW, "avx5124vnniw")
+X86_FEATURE_COMPAT(29, FEATURE_AVX5124FMAPS, "avx5124fmaps")
+X86_FEATURE_COMPAT(30, FEATURE_AVX512VPOPCNTDQ, "avx512vpopcntdq")
+X86_FEATURE_COMPAT(31, FEATURE_AVX512VBMI2, "avx512vbmi2")
+X86_FEATURE_COMPAT(32, FEATURE_GFNI, "gfni")
+X86_FEATURE_COMPAT(33, FEATURE_VPCLMULQDQ, "vpclmulqdq")
+X86_FEATURE_COMPAT(34, FEATURE_AVX512VNNI, "avx512vnni")
+X86_FEATURE_COMPAT(35, FEATURE_AVX512BITALG, "avx512bitalg")
+X86_FEATURE_COMPAT(36, FEATURE_AVX512BF16, "avx512bf16")
+// Features below here are not in libgcc/compiler-rt.
+X86_FEATURE (64, FEATURE_MOVBE)
+X86_FEATURE (65, FEATURE_ADX)
+X86_FEATURE (66, FEATURE_EM64T)
+X86_FEATURE (67, FEATURE_CLFLUSHOPT)
+X86_FEATURE (68, FEATURE_SHA)
+X86_FEATURE (69, FEATURE_AVX512VP2INTERSECT)
+#undef X86_FEATURE_COMPAT
+#undef X86_FEATURE
diff --git a/third_party/llvm-project/include/llvm/Support/YAMLParser.h b/third_party/llvm-project/include/llvm/Support/YAMLParser.h
new file mode 100644
index 000000000..3570119a3
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/YAMLParser.h
@@ -0,0 +1,619 @@
+//===- YAMLParser.h - Simple YAML parser ------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This is a YAML 1.2 parser.
+//
+// See http://www.yaml.org/spec/1.2/spec.html for the full standard.
+//
+// This currently does not implement the following:
+// * Multi-line literal folding.
+// * Tag resolution.
+// * UTF-16.
+// * BOMs anywhere other than the first Unicode scalar value in the file.
+//
+// The most important class here is Stream. This represents a YAML stream with
+// 0, 1, or many documents.
+//
+// SourceMgr sm;
+// StringRef input = getInput();
+// yaml::Stream stream(input, sm);
+//
+// for (yaml::document_iterator di = stream.begin(), de = stream.end();
+// di != de; ++di) {
+// yaml::Node *n = di->getRoot();
+// if (n) {
+// // Do something with n...
+// } else
+// break;
+// }
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_YAMLPARSER_H
+#define LLVM_SUPPORT_YAMLPARSER_H
+
+#include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Allocator.h"
+#include "llvm/Support/SMLoc.h"
+#include <cassert>
+#include <cstddef>
+#include <iterator>
+#include <map>
+#include <memory>
+#include <string>
+#include <system_error>
+
+namespace llvm {
+
+class MemoryBufferRef;
+class SourceMgr;
+class raw_ostream;
+class Twine;
+
+namespace yaml {
+
+class Document;
+class document_iterator;
+class Node;
+class Scanner;
+struct Token;
+
+/// Dump all the tokens in this stream to OS.
+/// \returns true if there was an error, false otherwise.
+bool dumpTokens(StringRef Input, raw_ostream &);
+
+/// Scans all tokens in input without outputting anything. This is used
+/// for benchmarking the tokenizer.
+/// \returns true if there was an error, false otherwise.
+bool scanTokens(StringRef Input);
+
+/// Escape \a Input for a double quoted scalar; if \p EscapePrintable
+/// is true, all UTF8 sequences will be escaped, if \p EscapePrintable is
+/// false, those UTF8 sequences encoding printable unicode scalars will not be
+/// escaped, but emitted verbatim.
+std::string escape(StringRef Input, bool EscapePrintable = true);
+
+/// This class represents a YAML stream potentially containing multiple
+/// documents.
+class Stream {
+public:
+ /// This keeps a reference to the string referenced by \p Input.
+ Stream(StringRef Input, SourceMgr &, bool ShowColors = true,
+ std::error_code *EC = nullptr);
+
+ Stream(MemoryBufferRef InputBuffer, SourceMgr &, bool ShowColors = true,
+ std::error_code *EC = nullptr);
+ ~Stream();
+
+ document_iterator begin();
+ document_iterator end();
+ void skip();
+ bool failed();
+
+ bool validate() {
+ skip();
+ return !failed();
+ }
+
+ void printError(Node *N, const Twine &Msg);
+
+private:
+ friend class Document;
+
+ std::unique_ptr<Scanner> scanner;
+ std::unique_ptr<Document> CurrentDoc;
+};
+
+/// Abstract base class for all Nodes.
+class Node {
+ virtual void anchor();
+
+public:
+ enum NodeKind {
+ NK_Null,
+ NK_Scalar,
+ NK_BlockScalar,
+ NK_KeyValue,
+ NK_Mapping,
+ NK_Sequence,
+ NK_Alias
+ };
+
+ Node(unsigned int Type, std::unique_ptr<Document> &, StringRef Anchor,
+ StringRef Tag);
+
+ // It's not safe to copy YAML nodes; the document is streamed and the position
+ // is part of the state.
+ Node(const Node &) = delete;
+ void operator=(const Node &) = delete;
+
+ void *operator new(size_t Size, BumpPtrAllocator &Alloc,
+ size_t Alignment = 16) noexcept {
+ return Alloc.Allocate(Size, Alignment);
+ }
+
+ void operator delete(void *Ptr, BumpPtrAllocator &Alloc,
+ size_t Size) noexcept {
+ Alloc.Deallocate(Ptr, Size);
+ }
+
+ void operator delete(void *) noexcept = delete;
+
+ /// Get the value of the anchor attached to this node. If it does not
+ /// have one, getAnchor().size() will be 0.
+ StringRef getAnchor() const { return Anchor; }
+
+ /// Get the tag as it was written in the document. This does not
+ /// perform tag resolution.
+ StringRef getRawTag() const { return Tag; }
+
+ /// Get the verbatium tag for a given Node. This performs tag resoluton
+ /// and substitution.
+ std::string getVerbatimTag() const;
+
+ SMRange getSourceRange() const { return SourceRange; }
+ void setSourceRange(SMRange SR) { SourceRange = SR; }
+
+ // These functions forward to Document and Scanner.
+ Token &peekNext();
+ Token getNext();
+ Node *parseBlockNode();
+ BumpPtrAllocator &getAllocator();
+ void setError(const Twine &Message, Token &Location) const;
+ bool failed() const;
+
+ virtual void skip() {}
+
+ unsigned int getType() const { return TypeID; }
+
+protected:
+ std::unique_ptr<Document> &Doc;
+ SMRange SourceRange;
+
+ ~Node() = default;
+
+private:
+ unsigned int TypeID;
+ StringRef Anchor;
+ /// The tag as typed in the document.
+ StringRef Tag;
+};
+
+/// A null value.
+///
+/// Example:
+/// !!null null
+class NullNode final : public Node {
+ void anchor() override;
+
+public:
+ NullNode(std::unique_ptr<Document> &D)
+ : Node(NK_Null, D, StringRef(), StringRef()) {}
+
+ static bool classof(const Node *N) { return N->getType() == NK_Null; }
+};
+
+/// A scalar node is an opaque datum that can be presented as a
+/// series of zero or more Unicode scalar values.
+///
+/// Example:
+/// Adena
+class ScalarNode final : public Node {
+ void anchor() override;
+
+public:
+ ScalarNode(std::unique_ptr<Document> &D, StringRef Anchor, StringRef Tag,
+ StringRef Val)
+ : Node(NK_Scalar, D, Anchor, Tag), Value(Val) {
+ SMLoc Start = SMLoc::getFromPointer(Val.begin());
+ SMLoc End = SMLoc::getFromPointer(Val.end());
+ SourceRange = SMRange(Start, End);
+ }
+
+ // Return Value without any escaping or folding or other fun YAML stuff. This
+ // is the exact bytes that are contained in the file (after conversion to
+ // utf8).
+ StringRef getRawValue() const { return Value; }
+
+ /// Gets the value of this node as a StringRef.
+ ///
+ /// \param Storage is used to store the content of the returned StringRef iff
+ /// it requires any modification from how it appeared in the source.
+ /// This happens with escaped characters and multi-line literals.
+ StringRef getValue(SmallVectorImpl<char> &Storage) const;
+
+ static bool classof(const Node *N) {
+ return N->getType() == NK_Scalar;
+ }
+
+private:
+ StringRef Value;
+
+ StringRef unescapeDoubleQuoted(StringRef UnquotedValue,
+ StringRef::size_type Start,
+ SmallVectorImpl<char> &Storage) const;
+};
+
+/// A block scalar node is an opaque datum that can be presented as a
+/// series of zero or more Unicode scalar values.
+///
+/// Example:
+/// |
+/// Hello
+/// World
+class BlockScalarNode final : public Node {
+ void anchor() override;
+
+public:
+ BlockScalarNode(std::unique_ptr<Document> &D, StringRef Anchor, StringRef Tag,
+ StringRef Value, StringRef RawVal)
+ : Node(NK_BlockScalar, D, Anchor, Tag), Value(Value) {
+ SMLoc Start = SMLoc::getFromPointer(RawVal.begin());
+ SMLoc End = SMLoc::getFromPointer(RawVal.end());
+ SourceRange = SMRange(Start, End);
+ }
+
+ /// Gets the value of this node as a StringRef.
+ StringRef getValue() const { return Value; }
+
+ static bool classof(const Node *N) {
+ return N->getType() == NK_BlockScalar;
+ }
+
+private:
+ StringRef Value;
+};
+
+/// A key and value pair. While not technically a Node under the YAML
+/// representation graph, it is easier to treat them this way.
+///
+/// TODO: Consider making this not a child of Node.
+///
+/// Example:
+/// Section: .text
+class KeyValueNode final : public Node {
+ void anchor() override;
+
+public:
+ KeyValueNode(std::unique_ptr<Document> &D)
+ : Node(NK_KeyValue, D, StringRef(), StringRef()) {}
+
+ /// Parse and return the key.
+ ///
+ /// This may be called multiple times.
+ ///
+ /// \returns The key, or nullptr if failed() == true.
+ Node *getKey();
+
+ /// Parse and return the value.
+ ///
+ /// This may be called multiple times.
+ ///
+ /// \returns The value, or nullptr if failed() == true.
+ Node *getValue();
+
+ void skip() override {
+ if (Node *Key = getKey()) {
+ Key->skip();
+ if (Node *Val = getValue())
+ Val->skip();
+ }
+ }
+
+ static bool classof(const Node *N) {
+ return N->getType() == NK_KeyValue;
+ }
+
+private:
+ Node *Key = nullptr;
+ Node *Value = nullptr;
+};
+
+/// This is an iterator abstraction over YAML collections shared by both
+/// sequences and maps.
+///
+/// BaseT must have a ValueT* member named CurrentEntry and a member function
+/// increment() which must set CurrentEntry to 0 to create an end iterator.
+template <class BaseT, class ValueT>
+class basic_collection_iterator
+ : public std::iterator<std::input_iterator_tag, ValueT> {
+public:
+ basic_collection_iterator() = default;
+ basic_collection_iterator(BaseT *B) : Base(B) {}
+
+ ValueT *operator->() const {
+ assert(Base && Base->CurrentEntry && "Attempted to access end iterator!");
+ return Base->CurrentEntry;
+ }
+
+ ValueT &operator*() const {
+ assert(Base && Base->CurrentEntry &&
+ "Attempted to dereference end iterator!");
+ return *Base->CurrentEntry;
+ }
+
+ operator ValueT *() const {
+ assert(Base && Base->CurrentEntry && "Attempted to access end iterator!");
+ return Base->CurrentEntry;
+ }
+
+ /// Note on EqualityComparable:
+ ///
+ /// The iterator is not re-entrant,
+ /// it is meant to be used for parsing YAML on-demand
+ /// Once iteration started - it can point only to one entry at a time
+ /// hence Base.CurrentEntry and Other.Base.CurrentEntry are equal
+ /// iff Base and Other.Base are equal.
+ bool operator==(const basic_collection_iterator &Other) const {
+ if (Base && (Base == Other.Base)) {
+ assert((Base->CurrentEntry == Other.Base->CurrentEntry)
+ && "Equal Bases expected to point to equal Entries");
+ }
+
+ return Base == Other.Base;
+ }
+
+ bool operator!=(const basic_collection_iterator &Other) const {
+ return !(Base == Other.Base);
+ }
+
+ basic_collection_iterator &operator++() {
+ assert(Base && "Attempted to advance iterator past end!");
+ Base->increment();
+ // Create an end iterator.
+ if (!Base->CurrentEntry)
+ Base = nullptr;
+ return *this;
+ }
+
+private:
+ BaseT *Base = nullptr;
+};
+
+// The following two templates are used for both MappingNode and Sequence Node.
+template <class CollectionType>
+typename CollectionType::iterator begin(CollectionType &C) {
+ assert(C.IsAtBeginning && "You may only iterate over a collection once!");
+ C.IsAtBeginning = false;
+ typename CollectionType::iterator ret(&C);
+ ++ret;
+ return ret;
+}
+
+template <class CollectionType> void skip(CollectionType &C) {
+ // TODO: support skipping from the middle of a parsed collection ;/
+ assert((C.IsAtBeginning || C.IsAtEnd) && "Cannot skip mid parse!");
+ if (C.IsAtBeginning)
+ for (typename CollectionType::iterator i = begin(C), e = C.end(); i != e;
+ ++i)
+ i->skip();
+}
+
+/// Represents a YAML map created from either a block map for a flow map.
+///
+/// This parses the YAML stream as increment() is called.
+///
+/// Example:
+/// Name: _main
+/// Scope: Global
+class MappingNode final : public Node {
+ void anchor() override;
+
+public:
+ enum MappingType {
+ MT_Block,
+ MT_Flow,
+ MT_Inline ///< An inline mapping node is used for "[key: value]".
+ };
+
+ MappingNode(std::unique_ptr<Document> &D, StringRef Anchor, StringRef Tag,
+ MappingType MT)
+ : Node(NK_Mapping, D, Anchor, Tag), Type(MT) {}
+
+ friend class basic_collection_iterator<MappingNode, KeyValueNode>;
+
+ using iterator = basic_collection_iterator<MappingNode, KeyValueNode>;
+
+ template <class T> friend typename T::iterator yaml::begin(T &);
+ template <class T> friend void yaml::skip(T &);
+
+ iterator begin() { return yaml::begin(*this); }
+
+ iterator end() { return iterator(); }
+
+ void skip() override { yaml::skip(*this); }
+
+ static bool classof(const Node *N) {
+ return N->getType() == NK_Mapping;
+ }
+
+private:
+ MappingType Type;
+ bool IsAtBeginning = true;
+ bool IsAtEnd = false;
+ KeyValueNode *CurrentEntry = nullptr;
+
+ void increment();
+};
+
+/// Represents a YAML sequence created from either a block sequence for a
+/// flow sequence.
+///
+/// This parses the YAML stream as increment() is called.
+///
+/// Example:
+/// - Hello
+/// - World
+class SequenceNode final : public Node {
+ void anchor() override;
+
+public:
+ enum SequenceType {
+ ST_Block,
+ ST_Flow,
+ // Use for:
+ //
+ // key:
+ // - val1
+ // - val2
+ //
+ // As a BlockMappingEntry and BlockEnd are not created in this case.
+ ST_Indentless
+ };
+
+ SequenceNode(std::unique_ptr<Document> &D, StringRef Anchor, StringRef Tag,
+ SequenceType ST)
+ : Node(NK_Sequence, D, Anchor, Tag), SeqType(ST) {}
+
+ friend class basic_collection_iterator<SequenceNode, Node>;
+
+ using iterator = basic_collection_iterator<SequenceNode, Node>;
+
+ template <class T> friend typename T::iterator yaml::begin(T &);
+ template <class T> friend void yaml::skip(T &);
+
+ void increment();
+
+ iterator begin() { return yaml::begin(*this); }
+
+ iterator end() { return iterator(); }
+
+ void skip() override { yaml::skip(*this); }
+
+ static bool classof(const Node *N) {
+ return N->getType() == NK_Sequence;
+ }
+
+private:
+ SequenceType SeqType;
+ bool IsAtBeginning = true;
+ bool IsAtEnd = false;
+ bool WasPreviousTokenFlowEntry = true; // Start with an imaginary ','.
+ Node *CurrentEntry = nullptr;
+};
+
+/// Represents an alias to a Node with an anchor.
+///
+/// Example:
+/// *AnchorName
+class AliasNode final : public Node {
+ void anchor() override;
+
+public:
+ AliasNode(std::unique_ptr<Document> &D, StringRef Val)
+ : Node(NK_Alias, D, StringRef(), StringRef()), Name(Val) {}
+
+ StringRef getName() const { return Name; }
+ Node *getTarget();
+
+ static bool classof(const Node *N) { return N->getType() == NK_Alias; }
+
+private:
+ StringRef Name;
+};
+
+/// A YAML Stream is a sequence of Documents. A document contains a root
+/// node.
+class Document {
+public:
+ Document(Stream &ParentStream);
+
+ /// Root for parsing a node. Returns a single node.
+ Node *parseBlockNode();
+
+ /// Finish parsing the current document and return true if there are
+ /// more. Return false otherwise.
+ bool skip();
+
+ /// Parse and return the root level node.
+ Node *getRoot() {
+ if (Root)
+ return Root;
+ return Root = parseBlockNode();
+ }
+
+ const std::map<StringRef, StringRef> &getTagMap() const { return TagMap; }
+
+private:
+ friend class Node;
+ friend class document_iterator;
+
+ /// Stream to read tokens from.
+ Stream &stream;
+
+ /// Used to allocate nodes to. All are destroyed without calling their
+ /// destructor when the document is destroyed.
+ BumpPtrAllocator NodeAllocator;
+
+ /// The root node. Used to support skipping a partially parsed
+ /// document.
+ Node *Root;
+
+ /// Maps tag prefixes to their expansion.
+ std::map<StringRef, StringRef> TagMap;
+
+ Token &peekNext();
+ Token getNext();
+ void setError(const Twine &Message, Token &Location) const;
+ bool failed() const;
+
+ /// Parse %BLAH directives and return true if any were encountered.
+ bool parseDirectives();
+
+ /// Parse %YAML
+ void parseYAMLDirective();
+
+ /// Parse %TAG
+ void parseTAGDirective();
+
+ /// Consume the next token and error if it is not \a TK.
+ bool expectToken(int TK);
+};
+
+/// Iterator abstraction for Documents over a Stream.
+class document_iterator {
+public:
+ document_iterator() = default;
+ document_iterator(std::unique_ptr<Document> &D) : Doc(&D) {}
+
+ bool operator==(const document_iterator &Other) const {
+ if (isAtEnd() || Other.isAtEnd())
+ return isAtEnd() && Other.isAtEnd();
+
+ return Doc == Other.Doc;
+ }
+ bool operator!=(const document_iterator &Other) const {
+ return !(*this == Other);
+ }
+
+ document_iterator operator++() {
+ assert(Doc && "incrementing iterator past the end.");
+ if (!(*Doc)->skip()) {
+ Doc->reset(nullptr);
+ } else {
+ Stream &S = (*Doc)->stream;
+ Doc->reset(new Document(S));
+ }
+ return *this;
+ }
+
+ Document &operator*() { return *Doc->get(); }
+
+ std::unique_ptr<Document> &operator->() { return *Doc; }
+
+private:
+ bool isAtEnd() const { return !Doc || !*Doc; }
+
+ std::unique_ptr<Document> *Doc = nullptr;
+};
+
+} // end namespace yaml
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_YAMLPARSER_H
diff --git a/third_party/llvm-project/include/llvm/Support/YAMLTraits.h b/third_party/llvm-project/include/llvm/Support/YAMLTraits.h
new file mode 100644
index 000000000..8642069ad
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/YAMLTraits.h
@@ -0,0 +1,2043 @@
+//===- llvm/Support/YAMLTraits.h --------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_YAMLTRAITS_H
+#define LLVM_SUPPORT_YAMLTRAITS_H
+
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Support/AlignOf.h"
+#include "llvm/Support/Allocator.h"
+#include "llvm/Support/Endian.h"
+#include "llvm/Support/Regex.h"
+#include "llvm/Support/SourceMgr.h"
+#include "llvm/Support/YAMLParser.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cassert>
+#include <cctype>
+#include <cstddef>
+#include <cstdint>
+#include <iterator>
+#include <map>
+#include <memory>
+#include <new>
+#include <string>
+#include <system_error>
+#include <type_traits>
+#include <vector>
+
+namespace llvm {
+namespace yaml {
+
+enum class NodeKind : uint8_t {
+ Scalar,
+ Map,
+ Sequence,
+};
+
+struct EmptyContext {};
+
+/// This class should be specialized by any type that needs to be converted
+/// to/from a YAML mapping. For example:
+///
+/// struct MappingTraits<MyStruct> {
+/// static void mapping(IO &io, MyStruct &s) {
+/// io.mapRequired("name", s.name);
+/// io.mapRequired("size", s.size);
+/// io.mapOptional("age", s.age);
+/// }
+/// };
+template<class T>
+struct MappingTraits {
+ // Must provide:
+ // static void mapping(IO &io, T &fields);
+ // Optionally may provide:
+ // static StringRef validate(IO &io, T &fields);
+ //
+ // The optional flow flag will cause generated YAML to use a flow mapping
+ // (e.g. { a: 0, b: 1 }):
+ // static const bool flow = true;
+};
+
+/// This class is similar to MappingTraits<T> but allows you to pass in
+/// additional context for each map operation. For example:
+///
+/// struct MappingContextTraits<MyStruct, MyContext> {
+/// static void mapping(IO &io, MyStruct &s, MyContext &c) {
+/// io.mapRequired("name", s.name);
+/// io.mapRequired("size", s.size);
+/// io.mapOptional("age", s.age);
+/// ++c.TimesMapped;
+/// }
+/// };
+template <class T, class Context> struct MappingContextTraits {
+ // Must provide:
+ // static void mapping(IO &io, T &fields, Context &Ctx);
+ // Optionally may provide:
+ // static StringRef validate(IO &io, T &fields, Context &Ctx);
+ //
+ // The optional flow flag will cause generated YAML to use a flow mapping
+ // (e.g. { a: 0, b: 1 }):
+ // static const bool flow = true;
+};
+
+/// This class should be specialized by any integral type that converts
+/// to/from a YAML scalar where there is a one-to-one mapping between
+/// in-memory values and a string in YAML. For example:
+///
+/// struct ScalarEnumerationTraits<Colors> {
+/// static void enumeration(IO &io, Colors &value) {
+/// io.enumCase(value, "red", cRed);
+/// io.enumCase(value, "blue", cBlue);
+/// io.enumCase(value, "green", cGreen);
+/// }
+/// };
+template <typename T, typename Enable = void> struct ScalarEnumerationTraits {
+ // Must provide:
+ // static void enumeration(IO &io, T &value);
+};
+
+/// This class should be specialized by any integer type that is a union
+/// of bit values and the YAML representation is a flow sequence of
+/// strings. For example:
+///
+/// struct ScalarBitSetTraits<MyFlags> {
+/// static void bitset(IO &io, MyFlags &value) {
+/// io.bitSetCase(value, "big", flagBig);
+/// io.bitSetCase(value, "flat", flagFlat);
+/// io.bitSetCase(value, "round", flagRound);
+/// }
+/// };
+template <typename T, typename Enable = void> struct ScalarBitSetTraits {
+ // Must provide:
+ // static void bitset(IO &io, T &value);
+};
+
+/// Describe which type of quotes should be used when quoting is necessary.
+/// Some non-printable characters need to be double-quoted, while some others
+/// are fine with simple-quoting, and some don't need any quoting.
+enum class QuotingType { None, Single, Double };
+
+/// This class should be specialized by type that requires custom conversion
+/// to/from a yaml scalar. For example:
+///
+/// template<>
+/// struct ScalarTraits<MyType> {
+/// static void output(const MyType &val, void*, llvm::raw_ostream &out) {
+/// // stream out custom formatting
+/// out << llvm::format("%x", val);
+/// }
+/// static StringRef input(StringRef scalar, void*, MyType &value) {
+/// // parse scalar and set `value`
+/// // return empty string on success, or error string
+/// return StringRef();
+/// }
+/// static QuotingType mustQuote(StringRef) { return QuotingType::Single; }
+/// };
+template <typename T, typename Enable = void> struct ScalarTraits {
+ // Must provide:
+ //
+ // Function to write the value as a string:
+ // static void output(const T &value, void *ctxt, llvm::raw_ostream &out);
+ //
+ // Function to convert a string to a value. Returns the empty
+ // StringRef on success or an error string if string is malformed:
+ // static StringRef input(StringRef scalar, void *ctxt, T &value);
+ //
+ // Function to determine if the value should be quoted.
+ // static QuotingType mustQuote(StringRef);
+};
+
+/// This class should be specialized by type that requires custom conversion
+/// to/from a YAML literal block scalar. For example:
+///
+/// template <>
+/// struct BlockScalarTraits<MyType> {
+/// static void output(const MyType &Value, void*, llvm::raw_ostream &Out)
+/// {
+/// // stream out custom formatting
+/// Out << Value;
+/// }
+/// static StringRef input(StringRef Scalar, void*, MyType &Value) {
+/// // parse scalar and set `value`
+/// // return empty string on success, or error string
+/// return StringRef();
+/// }
+/// };
+template <typename T>
+struct BlockScalarTraits {
+ // Must provide:
+ //
+ // Function to write the value as a string:
+ // static void output(const T &Value, void *ctx, llvm::raw_ostream &Out);
+ //
+ // Function to convert a string to a value. Returns the empty
+ // StringRef on success or an error string if string is malformed:
+ // static StringRef input(StringRef Scalar, void *ctxt, T &Value);
+ //
+ // Optional:
+ // static StringRef inputTag(T &Val, std::string Tag)
+ // static void outputTag(const T &Val, raw_ostream &Out)
+};
+
+/// This class should be specialized by type that requires custom conversion
+/// to/from a YAML scalar with optional tags. For example:
+///
+/// template <>
+/// struct TaggedScalarTraits<MyType> {
+/// static void output(const MyType &Value, void*, llvm::raw_ostream
+/// &ScalarOut, llvm::raw_ostream &TagOut)
+/// {
+/// // stream out custom formatting including optional Tag
+/// Out << Value;
+/// }
+/// static StringRef input(StringRef Scalar, StringRef Tag, void*, MyType
+/// &Value) {
+/// // parse scalar and set `value`
+/// // return empty string on success, or error string
+/// return StringRef();
+/// }
+/// static QuotingType mustQuote(const MyType &Value, StringRef) {
+/// return QuotingType::Single;
+/// }
+/// };
+template <typename T> struct TaggedScalarTraits {
+ // Must provide:
+ //
+ // Function to write the value and tag as strings:
+ // static void output(const T &Value, void *ctx, llvm::raw_ostream &ScalarOut,
+ // llvm::raw_ostream &TagOut);
+ //
+ // Function to convert a string to a value. Returns the empty
+ // StringRef on success or an error string if string is malformed:
+ // static StringRef input(StringRef Scalar, StringRef Tag, void *ctxt, T
+ // &Value);
+ //
+ // Function to determine if the value should be quoted.
+ // static QuotingType mustQuote(const T &Value, StringRef Scalar);
+};
+
+/// This class should be specialized by any type that needs to be converted
+/// to/from a YAML sequence. For example:
+///
+/// template<>
+/// struct SequenceTraits<MyContainer> {
+/// static size_t size(IO &io, MyContainer &seq) {
+/// return seq.size();
+/// }
+/// static MyType& element(IO &, MyContainer &seq, size_t index) {
+/// if ( index >= seq.size() )
+/// seq.resize(index+1);
+/// return seq[index];
+/// }
+/// };
+template<typename T, typename EnableIf = void>
+struct SequenceTraits {
+ // Must provide:
+ // static size_t size(IO &io, T &seq);
+ // static T::value_type& element(IO &io, T &seq, size_t index);
+ //
+ // The following is option and will cause generated YAML to use
+ // a flow sequence (e.g. [a,b,c]).
+ // static const bool flow = true;
+};
+
+/// This class should be specialized by any type for which vectors of that
+/// type need to be converted to/from a YAML sequence.
+template<typename T, typename EnableIf = void>
+struct SequenceElementTraits {
+ // Must provide:
+ // static const bool flow;
+};
+
+/// This class should be specialized by any type that needs to be converted
+/// to/from a list of YAML documents.
+template<typename T>
+struct DocumentListTraits {
+ // Must provide:
+ // static size_t size(IO &io, T &seq);
+ // static T::value_type& element(IO &io, T &seq, size_t index);
+};
+
+/// This class should be specialized by any type that needs to be converted
+/// to/from a YAML mapping in the case where the names of the keys are not known
+/// in advance, e.g. a string map.
+template <typename T>
+struct CustomMappingTraits {
+ // static void inputOne(IO &io, StringRef key, T &elem);
+ // static void output(IO &io, T &elem);
+};
+
+/// This class should be specialized by any type that can be represented as
+/// a scalar, map, or sequence, decided dynamically. For example:
+///
+/// typedef std::unique_ptr<MyBase> MyPoly;
+///
+/// template<>
+/// struct PolymorphicTraits<MyPoly> {
+/// static NodeKind getKind(const MyPoly &poly) {
+/// return poly->getKind();
+/// }
+/// static MyScalar& getAsScalar(MyPoly &poly) {
+/// if (!poly || !isa<MyScalar>(poly))
+/// poly.reset(new MyScalar());
+/// return *cast<MyScalar>(poly.get());
+/// }
+/// // ...
+/// };
+template <typename T> struct PolymorphicTraits {
+ // Must provide:
+ // static NodeKind getKind(const T &poly);
+ // static scalar_type &getAsScalar(T &poly);
+ // static map_type &getAsMap(T &poly);
+ // static sequence_type &getAsSequence(T &poly);
+};
+
+// Only used for better diagnostics of missing traits
+template <typename T>
+struct MissingTrait;
+
+// Test if ScalarEnumerationTraits<T> is defined on type T.
+template <class T>
+struct has_ScalarEnumerationTraits
+{
+ using Signature_enumeration = void (*)(class IO&, T&);
+
+ template <typename U>
+ static char test(SameType<Signature_enumeration, &U::enumeration>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<ScalarEnumerationTraits<T>>(nullptr)) == 1);
+};
+
+// Test if ScalarBitSetTraits<T> is defined on type T.
+template <class T>
+struct has_ScalarBitSetTraits
+{
+ using Signature_bitset = void (*)(class IO&, T&);
+
+ template <typename U>
+ static char test(SameType<Signature_bitset, &U::bitset>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value = (sizeof(test<ScalarBitSetTraits<T>>(nullptr)) == 1);
+};
+
+// Test if ScalarTraits<T> is defined on type T.
+template <class T>
+struct has_ScalarTraits
+{
+ using Signature_input = StringRef (*)(StringRef, void*, T&);
+ using Signature_output = void (*)(const T&, void*, raw_ostream&);
+ using Signature_mustQuote = QuotingType (*)(StringRef);
+
+ template <typename U>
+ static char test(SameType<Signature_input, &U::input> *,
+ SameType<Signature_output, &U::output> *,
+ SameType<Signature_mustQuote, &U::mustQuote> *);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<ScalarTraits<T>>(nullptr, nullptr, nullptr)) == 1);
+};
+
+// Test if BlockScalarTraits<T> is defined on type T.
+template <class T>
+struct has_BlockScalarTraits
+{
+ using Signature_input = StringRef (*)(StringRef, void *, T &);
+ using Signature_output = void (*)(const T &, void *, raw_ostream &);
+
+ template <typename U>
+ static char test(SameType<Signature_input, &U::input> *,
+ SameType<Signature_output, &U::output> *);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<BlockScalarTraits<T>>(nullptr, nullptr)) == 1);
+};
+
+// Test if TaggedScalarTraits<T> is defined on type T.
+template <class T> struct has_TaggedScalarTraits {
+ using Signature_input = StringRef (*)(StringRef, StringRef, void *, T &);
+ using Signature_output = void (*)(const T &, void *, raw_ostream &,
+ raw_ostream &);
+ using Signature_mustQuote = QuotingType (*)(const T &, StringRef);
+
+ template <typename U>
+ static char test(SameType<Signature_input, &U::input> *,
+ SameType<Signature_output, &U::output> *,
+ SameType<Signature_mustQuote, &U::mustQuote> *);
+
+ template <typename U> static double test(...);
+
+ static bool const value =
+ (sizeof(test<TaggedScalarTraits<T>>(nullptr, nullptr, nullptr)) == 1);
+};
+
+// Test if MappingContextTraits<T> is defined on type T.
+template <class T, class Context> struct has_MappingTraits {
+ using Signature_mapping = void (*)(class IO &, T &, Context &);
+
+ template <typename U>
+ static char test(SameType<Signature_mapping, &U::mapping>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<MappingContextTraits<T, Context>>(nullptr)) == 1);
+};
+
+// Test if MappingTraits<T> is defined on type T.
+template <class T> struct has_MappingTraits<T, EmptyContext> {
+ using Signature_mapping = void (*)(class IO &, T &);
+
+ template <typename U>
+ static char test(SameType<Signature_mapping, &U::mapping> *);
+
+ template <typename U> static double test(...);
+
+ static bool const value = (sizeof(test<MappingTraits<T>>(nullptr)) == 1);
+};
+
+// Test if MappingContextTraits<T>::validate() is defined on type T.
+template <class T, class Context> struct has_MappingValidateTraits {
+ using Signature_validate = StringRef (*)(class IO &, T &, Context &);
+
+ template <typename U>
+ static char test(SameType<Signature_validate, &U::validate>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<MappingContextTraits<T, Context>>(nullptr)) == 1);
+};
+
+// Test if MappingTraits<T>::validate() is defined on type T.
+template <class T> struct has_MappingValidateTraits<T, EmptyContext> {
+ using Signature_validate = StringRef (*)(class IO &, T &);
+
+ template <typename U>
+ static char test(SameType<Signature_validate, &U::validate> *);
+
+ template <typename U> static double test(...);
+
+ static bool const value = (sizeof(test<MappingTraits<T>>(nullptr)) == 1);
+};
+
+// Test if SequenceTraits<T> is defined on type T.
+template <class T>
+struct has_SequenceMethodTraits
+{
+ using Signature_size = size_t (*)(class IO&, T&);
+
+ template <typename U>
+ static char test(SameType<Signature_size, &U::size>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value = (sizeof(test<SequenceTraits<T>>(nullptr)) == 1);
+};
+
+// Test if CustomMappingTraits<T> is defined on type T.
+template <class T>
+struct has_CustomMappingTraits
+{
+ using Signature_input = void (*)(IO &io, StringRef key, T &v);
+
+ template <typename U>
+ static char test(SameType<Signature_input, &U::inputOne>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value =
+ (sizeof(test<CustomMappingTraits<T>>(nullptr)) == 1);
+};
+
+// has_FlowTraits<int> will cause an error with some compilers because
+// it subclasses int. Using this wrapper only instantiates the
+// real has_FlowTraits only if the template type is a class.
+template <typename T, bool Enabled = std::is_class<T>::value>
+class has_FlowTraits
+{
+public:
+ static const bool value = false;
+};
+
+// Some older gcc compilers don't support straight forward tests
+// for members, so test for ambiguity cause by the base and derived
+// classes both defining the member.
+template <class T>
+struct has_FlowTraits<T, true>
+{
+ struct Fallback { bool flow; };
+ struct Derived : T, Fallback { };
+
+ template<typename C>
+ static char (&f(SameType<bool Fallback::*, &C::flow>*))[1];
+
+ template<typename C>
+ static char (&f(...))[2];
+
+ static bool const value = sizeof(f<Derived>(nullptr)) == 2;
+};
+
+// Test if SequenceTraits<T> is defined on type T
+template<typename T>
+struct has_SequenceTraits : public std::integral_constant<bool,
+ has_SequenceMethodTraits<T>::value > { };
+
+// Test if DocumentListTraits<T> is defined on type T
+template <class T>
+struct has_DocumentListTraits
+{
+ using Signature_size = size_t (*)(class IO &, T &);
+
+ template <typename U>
+ static char test(SameType<Signature_size, &U::size>*);
+
+ template <typename U>
+ static double test(...);
+
+ static bool const value = (sizeof(test<DocumentListTraits<T>>(nullptr))==1);
+};
+
+template <class T> struct has_PolymorphicTraits {
+ using Signature_getKind = NodeKind (*)(const T &);
+
+ template <typename U>
+ static char test(SameType<Signature_getKind, &U::getKind> *);
+
+ template <typename U> static double test(...);
+
+ static bool const value = (sizeof(test<PolymorphicTraits<T>>(nullptr)) == 1);
+};
+
+inline bool isNumeric(StringRef S) {
+ const static auto skipDigits = [](StringRef Input) {
+ return Input.drop_front(
+ std::min(Input.find_first_not_of("0123456789"), Input.size()));
+ };
+
+ // Make S.front() and S.drop_front().front() (if S.front() is [+-]) calls
+ // safe.
+ if (S.empty() || S.equals("+") || S.equals("-"))
+ return false;
+
+ if (S.equals(".nan") || S.equals(".NaN") || S.equals(".NAN"))
+ return true;
+
+ // Infinity and decimal numbers can be prefixed with sign.
+ StringRef Tail = (S.front() == '-' || S.front() == '+') ? S.drop_front() : S;
+
+ // Check for infinity first, because checking for hex and oct numbers is more
+ // expensive.
+ if (Tail.equals(".inf") || Tail.equals(".Inf") || Tail.equals(".INF"))
+ return true;
+
+ // Section 10.3.2 Tag Resolution
+ // YAML 1.2 Specification prohibits Base 8 and Base 16 numbers prefixed with
+ // [-+], so S should be used instead of Tail.
+ if (S.startswith("0o"))
+ return S.size() > 2 &&
+ S.drop_front(2).find_first_not_of("01234567") == StringRef::npos;
+
+ if (S.startswith("0x"))
+ return S.size() > 2 && S.drop_front(2).find_first_not_of(
+ "0123456789abcdefABCDEF") == StringRef::npos;
+
+ // Parse float: [-+]? (\. [0-9]+ | [0-9]+ (\. [0-9]* )?) ([eE] [-+]? [0-9]+)?
+ S = Tail;
+
+ // Handle cases when the number starts with '.' and hence needs at least one
+ // digit after dot (as opposed by number which has digits before the dot), but
+ // doesn't have one.
+ if (S.startswith(".") &&
+ (S.equals(".") ||
+ (S.size() > 1 && std::strchr("0123456789", S[1]) == nullptr)))
+ return false;
+
+ if (S.startswith("E") || S.startswith("e"))
+ return false;
+
+ enum ParseState {
+ Default,
+ FoundDot,
+ FoundExponent,
+ };
+ ParseState State = Default;
+
+ S = skipDigits(S);
+
+ // Accept decimal integer.
+ if (S.empty())
+ return true;
+
+ if (S.front() == '.') {
+ State = FoundDot;
+ S = S.drop_front();
+ } else if (S.front() == 'e' || S.front() == 'E') {
+ State = FoundExponent;
+ S = S.drop_front();
+ } else {
+ return false;
+ }
+
+ if (State == FoundDot) {
+ S = skipDigits(S);
+ if (S.empty())
+ return true;
+
+ if (S.front() == 'e' || S.front() == 'E') {
+ State = FoundExponent;
+ S = S.drop_front();
+ } else {
+ return false;
+ }
+ }
+
+ assert(State == FoundExponent && "Should have found exponent at this point.");
+ if (S.empty())
+ return false;
+
+ if (S.front() == '+' || S.front() == '-') {
+ S = S.drop_front();
+ if (S.empty())
+ return false;
+ }
+
+ return skipDigits(S).empty();
+}
+
+inline bool isNull(StringRef S) {
+ return S.equals("null") || S.equals("Null") || S.equals("NULL") ||
+ S.equals("~");
+}
+
+inline bool isBool(StringRef S) {
+ return S.equals("true") || S.equals("True") || S.equals("TRUE") ||
+ S.equals("false") || S.equals("False") || S.equals("FALSE");
+}
+
+// 5.1. Character Set
+// The allowed character range explicitly excludes the C0 control block #x0-#x1F
+// (except for TAB #x9, LF #xA, and CR #xD which are allowed), DEL #x7F, the C1
+// control block #x80-#x9F (except for NEL #x85 which is allowed), the surrogate
+// block #xD800-#xDFFF, #xFFFE, and #xFFFF.
+inline QuotingType needsQuotes(StringRef S) {
+ if (S.empty())
+ return QuotingType::Single;
+ if (isspace(static_cast<unsigned char>(S.front())) ||
+ isspace(static_cast<unsigned char>(S.back())))
+ return QuotingType::Single;
+ if (isNull(S))
+ return QuotingType::Single;
+ if (isBool(S))
+ return QuotingType::Single;
+ if (isNumeric(S))
+ return QuotingType::Single;
+
+ // 7.3.3 Plain Style
+ // Plain scalars must not begin with most indicators, as this would cause
+ // ambiguity with other YAML constructs.
+ static constexpr char Indicators[] = R"(-?:\,[]{}#&*!|>'"%@`)";
+ if (S.find_first_of(Indicators) == 0)
+ return QuotingType::Single;
+
+ QuotingType MaxQuotingNeeded = QuotingType::None;
+ for (unsigned char C : S) {
+ // Alphanum is safe.
+ if (isAlnum(C))
+ continue;
+
+ switch (C) {
+ // Safe scalar characters.
+ case '_':
+ case '-':
+ case '^':
+ case '.':
+ case ',':
+ case ' ':
+ // TAB (0x9) is allowed in unquoted strings.
+ case 0x9:
+ continue;
+ // LF(0xA) and CR(0xD) may delimit values and so require at least single
+ // quotes.
+ case 0xA:
+ case 0xD:
+ MaxQuotingNeeded = QuotingType::Single;
+ continue;
+ // DEL (0x7F) are excluded from the allowed character range.
+ case 0x7F:
+ return QuotingType::Double;
+ // Forward slash is allowed to be unquoted, but we quote it anyway. We have
+ // many tests that use FileCheck against YAML output, and this output often
+ // contains paths. If we quote backslashes but not forward slashes then
+ // paths will come out either quoted or unquoted depending on which platform
+ // the test is run on, making FileCheck comparisons difficult.
+ case '/':
+ default: {
+ // C0 control block (0x0 - 0x1F) is excluded from the allowed character
+ // range.
+ if (C <= 0x1F)
+ return QuotingType::Double;
+
+ // Always double quote UTF-8.
+ if ((C & 0x80) != 0)
+ return QuotingType::Double;
+
+ // The character is not safe, at least simple quoting needed.
+ MaxQuotingNeeded = QuotingType::Single;
+ }
+ }
+ }
+
+ return MaxQuotingNeeded;
+}
+
+template <typename T, typename Context>
+struct missingTraits
+ : public std::integral_constant<bool,
+ !has_ScalarEnumerationTraits<T>::value &&
+ !has_ScalarBitSetTraits<T>::value &&
+ !has_ScalarTraits<T>::value &&
+ !has_BlockScalarTraits<T>::value &&
+ !has_TaggedScalarTraits<T>::value &&
+ !has_MappingTraits<T, Context>::value &&
+ !has_SequenceTraits<T>::value &&
+ !has_CustomMappingTraits<T>::value &&
+ !has_DocumentListTraits<T>::value &&
+ !has_PolymorphicTraits<T>::value> {};
+
+template <typename T, typename Context>
+struct validatedMappingTraits
+ : public std::integral_constant<
+ bool, has_MappingTraits<T, Context>::value &&
+ has_MappingValidateTraits<T, Context>::value> {};
+
+template <typename T, typename Context>
+struct unvalidatedMappingTraits
+ : public std::integral_constant<
+ bool, has_MappingTraits<T, Context>::value &&
+ !has_MappingValidateTraits<T, Context>::value> {};
+
+// Base class for Input and Output.
+class IO {
+public:
+ IO(void *Ctxt = nullptr);
+ virtual ~IO();
+
+ virtual bool outputting() const = 0;
+
+ virtual unsigned beginSequence() = 0;
+ virtual bool preflightElement(unsigned, void *&) = 0;
+ virtual void postflightElement(void*) = 0;
+ virtual void endSequence() = 0;
+ virtual bool canElideEmptySequence() = 0;
+
+ virtual unsigned beginFlowSequence() = 0;
+ virtual bool preflightFlowElement(unsigned, void *&) = 0;
+ virtual void postflightFlowElement(void*) = 0;
+ virtual void endFlowSequence() = 0;
+
+ virtual bool mapTag(StringRef Tag, bool Default=false) = 0;
+ virtual void beginMapping() = 0;
+ virtual void endMapping() = 0;
+ virtual bool preflightKey(const char*, bool, bool, bool &, void *&) = 0;
+ virtual void postflightKey(void*) = 0;
+ virtual std::vector<StringRef> keys() = 0;
+
+ virtual void beginFlowMapping() = 0;
+ virtual void endFlowMapping() = 0;
+
+ virtual void beginEnumScalar() = 0;
+ virtual bool matchEnumScalar(const char*, bool) = 0;
+ virtual bool matchEnumFallback() = 0;
+ virtual void endEnumScalar() = 0;
+
+ virtual bool beginBitSetScalar(bool &) = 0;
+ virtual bool bitSetMatch(const char*, bool) = 0;
+ virtual void endBitSetScalar() = 0;
+
+ virtual void scalarString(StringRef &, QuotingType) = 0;
+ virtual void blockScalarString(StringRef &) = 0;
+ virtual void scalarTag(std::string &) = 0;
+
+ virtual NodeKind getNodeKind() = 0;
+
+ virtual void setError(const Twine &) = 0;
+
+ template <typename T>
+ void enumCase(T &Val, const char* Str, const T ConstVal) {
+ if ( matchEnumScalar(Str, outputting() && Val == ConstVal) ) {
+ Val = ConstVal;
+ }
+ }
+
+ // allow anonymous enum values to be used with LLVM_YAML_STRONG_TYPEDEF
+ template <typename T>
+ void enumCase(T &Val, const char* Str, const uint32_t ConstVal) {
+ if ( matchEnumScalar(Str, outputting() && Val == static_cast<T>(ConstVal)) ) {
+ Val = ConstVal;
+ }
+ }
+
+ template <typename FBT, typename T>
+ void enumFallback(T &Val) {
+ if (matchEnumFallback()) {
+ EmptyContext Context;
+ // FIXME: Force integral conversion to allow strong typedefs to convert.
+ FBT Res = static_cast<typename FBT::BaseType>(Val);
+ yamlize(*this, Res, true, Context);
+ Val = static_cast<T>(static_cast<typename FBT::BaseType>(Res));
+ }
+ }
+
+ template <typename T>
+ void bitSetCase(T &Val, const char* Str, const T ConstVal) {
+ if ( bitSetMatch(Str, outputting() && (Val & ConstVal) == ConstVal) ) {
+ Val = static_cast<T>(Val | ConstVal);
+ }
+ }
+
+ // allow anonymous enum values to be used with LLVM_YAML_STRONG_TYPEDEF
+ template <typename T>
+ void bitSetCase(T &Val, const char* Str, const uint32_t ConstVal) {
+ if ( bitSetMatch(Str, outputting() && (Val & ConstVal) == ConstVal) ) {
+ Val = static_cast<T>(Val | ConstVal);
+ }
+ }
+
+ template <typename T>
+ void maskedBitSetCase(T &Val, const char *Str, T ConstVal, T Mask) {
+ if (bitSetMatch(Str, outputting() && (Val & Mask) == ConstVal))
+ Val = Val | ConstVal;
+ }
+
+ template <typename T>
+ void maskedBitSetCase(T &Val, const char *Str, uint32_t ConstVal,
+ uint32_t Mask) {
+ if (bitSetMatch(Str, outputting() && (Val & Mask) == ConstVal))
+ Val = Val | ConstVal;
+ }
+
+ void *getContext() const;
+ void setContext(void *);
+
+ template <typename T> void mapRequired(const char *Key, T &Val) {
+ EmptyContext Ctx;
+ this->processKey(Key, Val, true, Ctx);
+ }
+
+ template <typename T, typename Context>
+ void mapRequired(const char *Key, T &Val, Context &Ctx) {
+ this->processKey(Key, Val, true, Ctx);
+ }
+
+ template <typename T> void mapOptional(const char *Key, T &Val) {
+ EmptyContext Ctx;
+ mapOptionalWithContext(Key, Val, Ctx);
+ }
+
+ template <typename T, typename DefaultT>
+ void mapOptional(const char *Key, T &Val, const DefaultT &Default) {
+ EmptyContext Ctx;
+ mapOptionalWithContext(Key, Val, Default, Ctx);
+ }
+
+ template <typename T, typename Context>
+ typename std::enable_if<has_SequenceTraits<T>::value, void>::type
+ mapOptionalWithContext(const char *Key, T &Val, Context &Ctx) {
+ // omit key/value instead of outputting empty sequence
+ if (this->canElideEmptySequence() && !(Val.begin() != Val.end()))
+ return;
+ this->processKey(Key, Val, false, Ctx);
+ }
+
+ template <typename T, typename Context>
+ void mapOptionalWithContext(const char *Key, Optional<T> &Val, Context &Ctx) {
+ this->processKeyWithDefault(Key, Val, Optional<T>(), /*Required=*/false,
+ Ctx);
+ }
+
+ template <typename T, typename Context>
+ typename std::enable_if<!has_SequenceTraits<T>::value, void>::type
+ mapOptionalWithContext(const char *Key, T &Val, Context &Ctx) {
+ this->processKey(Key, Val, false, Ctx);
+ }
+
+ template <typename T, typename Context, typename DefaultT>
+ void mapOptionalWithContext(const char *Key, T &Val, const DefaultT &Default,
+ Context &Ctx) {
+ static_assert(std::is_convertible<DefaultT, T>::value,
+ "Default type must be implicitly convertible to value type!");
+ this->processKeyWithDefault(Key, Val, static_cast<const T &>(Default),
+ false, Ctx);
+ }
+
+private:
+ template <typename T, typename Context>
+ void processKeyWithDefault(const char *Key, Optional<T> &Val,
+ const Optional<T> &DefaultValue, bool Required,
+ Context &Ctx) {
+ assert(DefaultValue.hasValue() == false &&
+ "Optional<T> shouldn't have a value!");
+ void *SaveInfo;
+ bool UseDefault = true;
+ const bool sameAsDefault = outputting() && !Val.hasValue();
+ if (!outputting() && !Val.hasValue())
+ Val = T();
+ if (Val.hasValue() &&
+ this->preflightKey(Key, Required, sameAsDefault, UseDefault,
+ SaveInfo)) {
+ yamlize(*this, Val.getValue(), Required, Ctx);
+ this->postflightKey(SaveInfo);
+ } else {
+ if (UseDefault)
+ Val = DefaultValue;
+ }
+ }
+
+ template <typename T, typename Context>
+ void processKeyWithDefault(const char *Key, T &Val, const T &DefaultValue,
+ bool Required, Context &Ctx) {
+ void *SaveInfo;
+ bool UseDefault;
+ const bool sameAsDefault = outputting() && Val == DefaultValue;
+ if ( this->preflightKey(Key, Required, sameAsDefault, UseDefault,
+ SaveInfo) ) {
+ yamlize(*this, Val, Required, Ctx);
+ this->postflightKey(SaveInfo);
+ }
+ else {
+ if ( UseDefault )
+ Val = DefaultValue;
+ }
+ }
+
+ template <typename T, typename Context>
+ void processKey(const char *Key, T &Val, bool Required, Context &Ctx) {
+ void *SaveInfo;
+ bool UseDefault;
+ if ( this->preflightKey(Key, Required, false, UseDefault, SaveInfo) ) {
+ yamlize(*this, Val, Required, Ctx);
+ this->postflightKey(SaveInfo);
+ }
+ }
+
+private:
+ void *Ctxt;
+};
+
+namespace detail {
+
+template <typename T, typename Context>
+void doMapping(IO &io, T &Val, Context &Ctx) {
+ MappingContextTraits<T, Context>::mapping(io, Val, Ctx);
+}
+
+template <typename T> void doMapping(IO &io, T &Val, EmptyContext &Ctx) {
+ MappingTraits<T>::mapping(io, Val);
+}
+
+} // end namespace detail
+
+template <typename T>
+typename std::enable_if<has_ScalarEnumerationTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ io.beginEnumScalar();
+ ScalarEnumerationTraits<T>::enumeration(io, Val);
+ io.endEnumScalar();
+}
+
+template <typename T>
+typename std::enable_if<has_ScalarBitSetTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ bool DoClear;
+ if ( io.beginBitSetScalar(DoClear) ) {
+ if ( DoClear )
+ Val = T();
+ ScalarBitSetTraits<T>::bitset(io, Val);
+ io.endBitSetScalar();
+ }
+}
+
+template <typename T>
+typename std::enable_if<has_ScalarTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ if ( io.outputting() ) {
+ std::string Storage;
+ raw_string_ostream Buffer(Storage);
+ ScalarTraits<T>::output(Val, io.getContext(), Buffer);
+ StringRef Str = Buffer.str();
+ io.scalarString(Str, ScalarTraits<T>::mustQuote(Str));
+ }
+ else {
+ StringRef Str;
+ io.scalarString(Str, ScalarTraits<T>::mustQuote(Str));
+ StringRef Result = ScalarTraits<T>::input(Str, io.getContext(), Val);
+ if ( !Result.empty() ) {
+ io.setError(Twine(Result));
+ }
+ }
+}
+
+template <typename T>
+typename std::enable_if<has_BlockScalarTraits<T>::value, void>::type
+yamlize(IO &YamlIO, T &Val, bool, EmptyContext &Ctx) {
+ if (YamlIO.outputting()) {
+ std::string Storage;
+ raw_string_ostream Buffer(Storage);
+ BlockScalarTraits<T>::output(Val, YamlIO.getContext(), Buffer);
+ StringRef Str = Buffer.str();
+ YamlIO.blockScalarString(Str);
+ } else {
+ StringRef Str;
+ YamlIO.blockScalarString(Str);
+ StringRef Result =
+ BlockScalarTraits<T>::input(Str, YamlIO.getContext(), Val);
+ if (!Result.empty())
+ YamlIO.setError(Twine(Result));
+ }
+}
+
+template <typename T>
+typename std::enable_if<has_TaggedScalarTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ if (io.outputting()) {
+ std::string ScalarStorage, TagStorage;
+ raw_string_ostream ScalarBuffer(ScalarStorage), TagBuffer(TagStorage);
+ TaggedScalarTraits<T>::output(Val, io.getContext(), ScalarBuffer,
+ TagBuffer);
+ io.scalarTag(TagBuffer.str());
+ StringRef ScalarStr = ScalarBuffer.str();
+ io.scalarString(ScalarStr,
+ TaggedScalarTraits<T>::mustQuote(Val, ScalarStr));
+ } else {
+ std::string Tag;
+ io.scalarTag(Tag);
+ StringRef Str;
+ io.scalarString(Str, QuotingType::None);
+ StringRef Result =
+ TaggedScalarTraits<T>::input(Str, Tag, io.getContext(), Val);
+ if (!Result.empty()) {
+ io.setError(Twine(Result));
+ }
+ }
+}
+
+template <typename T, typename Context>
+typename std::enable_if<validatedMappingTraits<T, Context>::value, void>::type
+yamlize(IO &io, T &Val, bool, Context &Ctx) {
+ if (has_FlowTraits<MappingTraits<T>>::value)
+ io.beginFlowMapping();
+ else
+ io.beginMapping();
+ if (io.outputting()) {
+ StringRef Err = MappingTraits<T>::validate(io, Val);
+ if (!Err.empty()) {
+ errs() << Err << "\n";
+ assert(Err.empty() && "invalid struct trying to be written as yaml");
+ }
+ }
+ detail::doMapping(io, Val, Ctx);
+ if (!io.outputting()) {
+ StringRef Err = MappingTraits<T>::validate(io, Val);
+ if (!Err.empty())
+ io.setError(Err);
+ }
+ if (has_FlowTraits<MappingTraits<T>>::value)
+ io.endFlowMapping();
+ else
+ io.endMapping();
+}
+
+template <typename T, typename Context>
+typename std::enable_if<unvalidatedMappingTraits<T, Context>::value, void>::type
+yamlize(IO &io, T &Val, bool, Context &Ctx) {
+ if (has_FlowTraits<MappingTraits<T>>::value) {
+ io.beginFlowMapping();
+ detail::doMapping(io, Val, Ctx);
+ io.endFlowMapping();
+ } else {
+ io.beginMapping();
+ detail::doMapping(io, Val, Ctx);
+ io.endMapping();
+ }
+}
+
+template <typename T>
+typename std::enable_if<has_CustomMappingTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ if ( io.outputting() ) {
+ io.beginMapping();
+ CustomMappingTraits<T>::output(io, Val);
+ io.endMapping();
+ } else {
+ io.beginMapping();
+ for (StringRef key : io.keys())
+ CustomMappingTraits<T>::inputOne(io, key, Val);
+ io.endMapping();
+ }
+}
+
+template <typename T>
+typename std::enable_if<has_PolymorphicTraits<T>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ switch (io.outputting() ? PolymorphicTraits<T>::getKind(Val)
+ : io.getNodeKind()) {
+ case NodeKind::Scalar:
+ return yamlize(io, PolymorphicTraits<T>::getAsScalar(Val), true, Ctx);
+ case NodeKind::Map:
+ return yamlize(io, PolymorphicTraits<T>::getAsMap(Val), true, Ctx);
+ case NodeKind::Sequence:
+ return yamlize(io, PolymorphicTraits<T>::getAsSequence(Val), true, Ctx);
+ }
+}
+
+template <typename T>
+typename std::enable_if<missingTraits<T, EmptyContext>::value, void>::type
+yamlize(IO &io, T &Val, bool, EmptyContext &Ctx) {
+ char missing_yaml_trait_for_type[sizeof(MissingTrait<T>)];
+}
+
+template <typename T, typename Context>
+typename std::enable_if<has_SequenceTraits<T>::value, void>::type
+yamlize(IO &io, T &Seq, bool, Context &Ctx) {
+ if ( has_FlowTraits< SequenceTraits<T>>::value ) {
+ unsigned incnt = io.beginFlowSequence();
+ unsigned count = io.outputting() ? SequenceTraits<T>::size(io, Seq) : incnt;
+ for(unsigned i=0; i < count; ++i) {
+ void *SaveInfo;
+ if ( io.preflightFlowElement(i, SaveInfo) ) {
+ yamlize(io, SequenceTraits<T>::element(io, Seq, i), true, Ctx);
+ io.postflightFlowElement(SaveInfo);
+ }
+ }
+ io.endFlowSequence();
+ }
+ else {
+ unsigned incnt = io.beginSequence();
+ unsigned count = io.outputting() ? SequenceTraits<T>::size(io, Seq) : incnt;
+ for(unsigned i=0; i < count; ++i) {
+ void *SaveInfo;
+ if ( io.preflightElement(i, SaveInfo) ) {
+ yamlize(io, SequenceTraits<T>::element(io, Seq, i), true, Ctx);
+ io.postflightElement(SaveInfo);
+ }
+ }
+ io.endSequence();
+ }
+}
+
+template<>
+struct ScalarTraits<bool> {
+ static void output(const bool &, void* , raw_ostream &);
+ static StringRef input(StringRef, void *, bool &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<StringRef> {
+ static void output(const StringRef &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, StringRef &);
+ static QuotingType mustQuote(StringRef S) { return needsQuotes(S); }
+};
+
+template<>
+struct ScalarTraits<std::string> {
+ static void output(const std::string &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, std::string &);
+ static QuotingType mustQuote(StringRef S) { return needsQuotes(S); }
+};
+
+template<>
+struct ScalarTraits<uint8_t> {
+ static void output(const uint8_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, uint8_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<uint16_t> {
+ static void output(const uint16_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, uint16_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<uint32_t> {
+ static void output(const uint32_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, uint32_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<uint64_t> {
+ static void output(const uint64_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, uint64_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<int8_t> {
+ static void output(const int8_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, int8_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<int16_t> {
+ static void output(const int16_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, int16_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<int32_t> {
+ static void output(const int32_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, int32_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<int64_t> {
+ static void output(const int64_t &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, int64_t &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<float> {
+ static void output(const float &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, float &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<double> {
+ static void output(const double &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, double &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+// For endian types, we use existing scalar Traits class for the underlying
+// type. This way endian aware types are supported whenever the traits are
+// defined for the underlying type.
+template <typename value_type, support::endianness endian, size_t alignment>
+struct ScalarTraits<
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>,
+ typename std::enable_if<has_ScalarTraits<value_type>::value>::type> {
+ using endian_type =
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>;
+
+ static void output(const endian_type &E, void *Ctx, raw_ostream &Stream) {
+ ScalarTraits<value_type>::output(static_cast<value_type>(E), Ctx, Stream);
+ }
+
+ static StringRef input(StringRef Str, void *Ctx, endian_type &E) {
+ value_type V;
+ auto R = ScalarTraits<value_type>::input(Str, Ctx, V);
+ E = static_cast<endian_type>(V);
+ return R;
+ }
+
+ static QuotingType mustQuote(StringRef Str) {
+ return ScalarTraits<value_type>::mustQuote(Str);
+ }
+};
+
+template <typename value_type, support::endianness endian, size_t alignment>
+struct ScalarEnumerationTraits<
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>,
+ typename std::enable_if<
+ has_ScalarEnumerationTraits<value_type>::value>::type> {
+ using endian_type =
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>;
+
+ static void enumeration(IO &io, endian_type &E) {
+ value_type V = E;
+ ScalarEnumerationTraits<value_type>::enumeration(io, V);
+ E = V;
+ }
+};
+
+template <typename value_type, support::endianness endian, size_t alignment>
+struct ScalarBitSetTraits<
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>,
+ typename std::enable_if<has_ScalarBitSetTraits<value_type>::value>::type> {
+ using endian_type =
+ support::detail::packed_endian_specific_integral<value_type, endian,
+ alignment>;
+ static void bitset(IO &io, endian_type &E) {
+ value_type V = E;
+ ScalarBitSetTraits<value_type>::bitset(io, V);
+ E = V;
+ }
+};
+
+// Utility for use within MappingTraits<>::mapping() method
+// to [de]normalize an object for use with YAML conversion.
+template <typename TNorm, typename TFinal>
+struct MappingNormalization {
+ MappingNormalization(IO &i_o, TFinal &Obj)
+ : io(i_o), BufPtr(nullptr), Result(Obj) {
+ if ( io.outputting() ) {
+ BufPtr = new (&Buffer) TNorm(io, Obj);
+ }
+ else {
+ BufPtr = new (&Buffer) TNorm(io);
+ }
+ }
+
+ ~MappingNormalization() {
+ if ( ! io.outputting() ) {
+ Result = BufPtr->denormalize(io);
+ }
+ BufPtr->~TNorm();
+ }
+
+ TNorm* operator->() { return BufPtr; }
+
+private:
+ using Storage = AlignedCharArrayUnion<TNorm>;
+
+ Storage Buffer;
+ IO &io;
+ TNorm *BufPtr;
+ TFinal &Result;
+};
+
+// Utility for use within MappingTraits<>::mapping() method
+// to [de]normalize an object for use with YAML conversion.
+template <typename TNorm, typename TFinal>
+struct MappingNormalizationHeap {
+ MappingNormalizationHeap(IO &i_o, TFinal &Obj, BumpPtrAllocator *allocator)
+ : io(i_o), Result(Obj) {
+ if ( io.outputting() ) {
+ BufPtr = new (&Buffer) TNorm(io, Obj);
+ }
+ else if (allocator) {
+ BufPtr = allocator->Allocate<TNorm>();
+ new (BufPtr) TNorm(io);
+ } else {
+ BufPtr = new TNorm(io);
+ }
+ }
+
+ ~MappingNormalizationHeap() {
+ if ( io.outputting() ) {
+ BufPtr->~TNorm();
+ }
+ else {
+ Result = BufPtr->denormalize(io);
+ }
+ }
+
+ TNorm* operator->() { return BufPtr; }
+
+private:
+ using Storage = AlignedCharArrayUnion<TNorm>;
+
+ Storage Buffer;
+ IO &io;
+ TNorm *BufPtr = nullptr;
+ TFinal &Result;
+};
+
+///
+/// The Input class is used to parse a yaml document into in-memory structs
+/// and vectors.
+///
+/// It works by using YAMLParser to do a syntax parse of the entire yaml
+/// document, then the Input class builds a graph of HNodes which wraps
+/// each yaml Node. The extra layer is buffering. The low level yaml
+/// parser only lets you look at each node once. The buffering layer lets
+/// you search and interate multiple times. This is necessary because
+/// the mapRequired() method calls may not be in the same order
+/// as the keys in the document.
+///
+class Input : public IO {
+public:
+ // Construct a yaml Input object from a StringRef and optional
+ // user-data. The DiagHandler can be specified to provide
+ // alternative error reporting.
+ Input(StringRef InputContent,
+ void *Ctxt = nullptr,
+ SourceMgr::DiagHandlerTy DiagHandler = nullptr,
+ void *DiagHandlerCtxt = nullptr);
+ Input(MemoryBufferRef Input,
+ void *Ctxt = nullptr,
+ SourceMgr::DiagHandlerTy DiagHandler = nullptr,
+ void *DiagHandlerCtxt = nullptr);
+ ~Input() override;
+
+ // Check if there was an syntax or semantic error during parsing.
+ std::error_code error();
+
+private:
+ bool outputting() const override;
+ bool mapTag(StringRef, bool) override;
+ void beginMapping() override;
+ void endMapping() override;
+ bool preflightKey(const char *, bool, bool, bool &, void *&) override;
+ void postflightKey(void *) override;
+ std::vector<StringRef> keys() override;
+ void beginFlowMapping() override;
+ void endFlowMapping() override;
+ unsigned beginSequence() override;
+ void endSequence() override;
+ bool preflightElement(unsigned index, void *&) override;
+ void postflightElement(void *) override;
+ unsigned beginFlowSequence() override;
+ bool preflightFlowElement(unsigned , void *&) override;
+ void postflightFlowElement(void *) override;
+ void endFlowSequence() override;
+ void beginEnumScalar() override;
+ bool matchEnumScalar(const char*, bool) override;
+ bool matchEnumFallback() override;
+ void endEnumScalar() override;
+ bool beginBitSetScalar(bool &) override;
+ bool bitSetMatch(const char *, bool ) override;
+ void endBitSetScalar() override;
+ void scalarString(StringRef &, QuotingType) override;
+ void blockScalarString(StringRef &) override;
+ void scalarTag(std::string &) override;
+ NodeKind getNodeKind() override;
+ void setError(const Twine &message) override;
+ bool canElideEmptySequence() override;
+
+ class HNode {
+ virtual void anchor();
+
+ public:
+ HNode(Node *n) : _node(n) { }
+ virtual ~HNode() = default;
+
+ static bool classof(const HNode *) { return true; }
+
+ Node *_node;
+ };
+
+ class EmptyHNode : public HNode {
+ void anchor() override;
+
+ public:
+ EmptyHNode(Node *n) : HNode(n) { }
+
+ static bool classof(const HNode *n) { return NullNode::classof(n->_node); }
+
+ static bool classof(const EmptyHNode *) { return true; }
+ };
+
+ class ScalarHNode : public HNode {
+ void anchor() override;
+
+ public:
+ ScalarHNode(Node *n, StringRef s) : HNode(n), _value(s) { }
+
+ StringRef value() const { return _value; }
+
+ static bool classof(const HNode *n) {
+ return ScalarNode::classof(n->_node) ||
+ BlockScalarNode::classof(n->_node);
+ }
+
+ static bool classof(const ScalarHNode *) { return true; }
+
+ protected:
+ StringRef _value;
+ };
+
+ class MapHNode : public HNode {
+ void anchor() override;
+
+ public:
+ MapHNode(Node *n) : HNode(n) { }
+
+ static bool classof(const HNode *n) {
+ return MappingNode::classof(n->_node);
+ }
+
+ static bool classof(const MapHNode *) { return true; }
+
+ using NameToNode = StringMap<std::unique_ptr<HNode>>;
+
+ NameToNode Mapping;
+ SmallVector<std::string, 6> ValidKeys;
+ };
+
+ class SequenceHNode : public HNode {
+ void anchor() override;
+
+ public:
+ SequenceHNode(Node *n) : HNode(n) { }
+
+ static bool classof(const HNode *n) {
+ return SequenceNode::classof(n->_node);
+ }
+
+ static bool classof(const SequenceHNode *) { return true; }
+
+ std::vector<std::unique_ptr<HNode>> Entries;
+ };
+
+ std::unique_ptr<Input::HNode> createHNodes(Node *node);
+ void setError(HNode *hnode, const Twine &message);
+ void setError(Node *node, const Twine &message);
+
+public:
+ // These are only used by operator>>. They could be private
+ // if those templated things could be made friends.
+ bool setCurrentDocument();
+ bool nextDocument();
+
+ /// Returns the current node that's being parsed by the YAML Parser.
+ const Node *getCurrentNode() const;
+
+private:
+ SourceMgr SrcMgr; // must be before Strm
+ std::unique_ptr<llvm::yaml::Stream> Strm;
+ std::unique_ptr<HNode> TopNode;
+ std::error_code EC;
+ BumpPtrAllocator StringAllocator;
+ document_iterator DocIterator;
+ std::vector<bool> BitValuesUsed;
+ HNode *CurrentNode = nullptr;
+ bool ScalarMatchFound = false;
+};
+
+///
+/// The Output class is used to generate a yaml document from in-memory structs
+/// and vectors.
+///
+class Output : public IO {
+public:
+ Output(raw_ostream &, void *Ctxt = nullptr, int WrapColumn = 70);
+ ~Output() override;
+
+ /// Set whether or not to output optional values which are equal
+ /// to the default value. By default, when outputting if you attempt
+ /// to write a value that is equal to the default, the value gets ignored.
+ /// Sometimes, it is useful to be able to see these in the resulting YAML
+ /// anyway.
+ void setWriteDefaultValues(bool Write) { WriteDefaultValues = Write; }
+
+ bool outputting() const override;
+ bool mapTag(StringRef, bool) override;
+ void beginMapping() override;
+ void endMapping() override;
+ bool preflightKey(const char *key, bool, bool, bool &, void *&) override;
+ void postflightKey(void *) override;
+ std::vector<StringRef> keys() override;
+ void beginFlowMapping() override;
+ void endFlowMapping() override;
+ unsigned beginSequence() override;
+ void endSequence() override;
+ bool preflightElement(unsigned, void *&) override;
+ void postflightElement(void *) override;
+ unsigned beginFlowSequence() override;
+ bool preflightFlowElement(unsigned, void *&) override;
+ void postflightFlowElement(void *) override;
+ void endFlowSequence() override;
+ void beginEnumScalar() override;
+ bool matchEnumScalar(const char*, bool) override;
+ bool matchEnumFallback() override;
+ void endEnumScalar() override;
+ bool beginBitSetScalar(bool &) override;
+ bool bitSetMatch(const char *, bool ) override;
+ void endBitSetScalar() override;
+ void scalarString(StringRef &, QuotingType) override;
+ void blockScalarString(StringRef &) override;
+ void scalarTag(std::string &) override;
+ NodeKind getNodeKind() override;
+ void setError(const Twine &message) override;
+ bool canElideEmptySequence() override;
+
+ // These are only used by operator<<. They could be private
+ // if that templated operator could be made a friend.
+ void beginDocuments();
+ bool preflightDocument(unsigned);
+ void postflightDocument();
+ void endDocuments();
+
+private:
+ void output(StringRef s);
+ void outputUpToEndOfLine(StringRef s);
+ void newLineCheck();
+ void outputNewLine();
+ void paddedKey(StringRef key);
+ void flowKey(StringRef Key);
+
+ enum InState {
+ inSeqFirstElement,
+ inSeqOtherElement,
+ inFlowSeqFirstElement,
+ inFlowSeqOtherElement,
+ inMapFirstKey,
+ inMapOtherKey,
+ inFlowMapFirstKey,
+ inFlowMapOtherKey
+ };
+
+ static bool inSeqAnyElement(InState State);
+ static bool inFlowSeqAnyElement(InState State);
+ static bool inMapAnyKey(InState State);
+ static bool inFlowMapAnyKey(InState State);
+
+ raw_ostream &Out;
+ int WrapColumn;
+ SmallVector<InState, 8> StateStack;
+ int Column = 0;
+ int ColumnAtFlowStart = 0;
+ int ColumnAtMapFlowStart = 0;
+ bool NeedBitValueComma = false;
+ bool NeedFlowSequenceComma = false;
+ bool EnumerationMatchFound = false;
+ bool WriteDefaultValues = false;
+ StringRef Padding;
+ StringRef PaddingBeforeContainer;
+};
+
+/// YAML I/O does conversion based on types. But often native data types
+/// are just a typedef of built in intergral types (e.g. int). But the C++
+/// type matching system sees through the typedef and all the typedefed types
+/// look like a built in type. This will cause the generic YAML I/O conversion
+/// to be used. To provide better control over the YAML conversion, you can
+/// use this macro instead of typedef. It will create a class with one field
+/// and automatic conversion operators to and from the base type.
+/// Based on BOOST_STRONG_TYPEDEF
+#define LLVM_YAML_STRONG_TYPEDEF(_base, _type) \
+ struct _type { \
+ _type() = default; \
+ _type(const _base v) : value(v) {} \
+ _type(const _type &v) = default; \
+ _type &operator=(const _type &rhs) = default; \
+ _type &operator=(const _base &rhs) { value = rhs; return *this; } \
+ operator const _base & () const { return value; } \
+ bool operator==(const _type &rhs) const { return value == rhs.value; } \
+ bool operator==(const _base &rhs) const { return value == rhs; } \
+ bool operator<(const _type &rhs) const { return value < rhs.value; } \
+ _base value; \
+ using BaseType = _base; \
+ };
+
+///
+/// Use these types instead of uintXX_t in any mapping to have
+/// its yaml output formatted as hexadecimal.
+///
+LLVM_YAML_STRONG_TYPEDEF(uint8_t, Hex8)
+LLVM_YAML_STRONG_TYPEDEF(uint16_t, Hex16)
+LLVM_YAML_STRONG_TYPEDEF(uint32_t, Hex32)
+LLVM_YAML_STRONG_TYPEDEF(uint64_t, Hex64)
+
+template<>
+struct ScalarTraits<Hex8> {
+ static void output(const Hex8 &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, Hex8 &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<Hex16> {
+ static void output(const Hex16 &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, Hex16 &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<Hex32> {
+ static void output(const Hex32 &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, Hex32 &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+template<>
+struct ScalarTraits<Hex64> {
+ static void output(const Hex64 &, void *, raw_ostream &);
+ static StringRef input(StringRef, void *, Hex64 &);
+ static QuotingType mustQuote(StringRef) { return QuotingType::None; }
+};
+
+// Define non-member operator>> so that Input can stream in a document list.
+template <typename T>
+inline
+typename std::enable_if<has_DocumentListTraits<T>::value, Input &>::type
+operator>>(Input &yin, T &docList) {
+ int i = 0;
+ EmptyContext Ctx;
+ while ( yin.setCurrentDocument() ) {
+ yamlize(yin, DocumentListTraits<T>::element(yin, docList, i), true, Ctx);
+ if ( yin.error() )
+ return yin;
+ yin.nextDocument();
+ ++i;
+ }
+ return yin;
+}
+
+// Define non-member operator>> so that Input can stream in a map as a document.
+template <typename T>
+inline typename std::enable_if<has_MappingTraits<T, EmptyContext>::value,
+ Input &>::type
+operator>>(Input &yin, T &docMap) {
+ EmptyContext Ctx;
+ yin.setCurrentDocument();
+ yamlize(yin, docMap, true, Ctx);
+ return yin;
+}
+
+// Define non-member operator>> so that Input can stream in a sequence as
+// a document.
+template <typename T>
+inline
+typename std::enable_if<has_SequenceTraits<T>::value, Input &>::type
+operator>>(Input &yin, T &docSeq) {
+ EmptyContext Ctx;
+ if (yin.setCurrentDocument())
+ yamlize(yin, docSeq, true, Ctx);
+ return yin;
+}
+
+// Define non-member operator>> so that Input can stream in a block scalar.
+template <typename T>
+inline
+typename std::enable_if<has_BlockScalarTraits<T>::value, Input &>::type
+operator>>(Input &In, T &Val) {
+ EmptyContext Ctx;
+ if (In.setCurrentDocument())
+ yamlize(In, Val, true, Ctx);
+ return In;
+}
+
+// Define non-member operator>> so that Input can stream in a string map.
+template <typename T>
+inline
+typename std::enable_if<has_CustomMappingTraits<T>::value, Input &>::type
+operator>>(Input &In, T &Val) {
+ EmptyContext Ctx;
+ if (In.setCurrentDocument())
+ yamlize(In, Val, true, Ctx);
+ return In;
+}
+
+// Define non-member operator>> so that Input can stream in a polymorphic type.
+template <typename T>
+inline typename std::enable_if<has_PolymorphicTraits<T>::value, Input &>::type
+operator>>(Input &In, T &Val) {
+ EmptyContext Ctx;
+ if (In.setCurrentDocument())
+ yamlize(In, Val, true, Ctx);
+ return In;
+}
+
+// Provide better error message about types missing a trait specialization
+template <typename T>
+inline typename std::enable_if<missingTraits<T, EmptyContext>::value,
+ Input &>::type
+operator>>(Input &yin, T &docSeq) {
+ char missing_yaml_trait_for_type[sizeof(MissingTrait<T>)];
+ return yin;
+}
+
+// Define non-member operator<< so that Output can stream out document list.
+template <typename T>
+inline
+typename std::enable_if<has_DocumentListTraits<T>::value, Output &>::type
+operator<<(Output &yout, T &docList) {
+ EmptyContext Ctx;
+ yout.beginDocuments();
+ const size_t count = DocumentListTraits<T>::size(yout, docList);
+ for(size_t i=0; i < count; ++i) {
+ if ( yout.preflightDocument(i) ) {
+ yamlize(yout, DocumentListTraits<T>::element(yout, docList, i), true,
+ Ctx);
+ yout.postflightDocument();
+ }
+ }
+ yout.endDocuments();
+ return yout;
+}
+
+// Define non-member operator<< so that Output can stream out a map.
+template <typename T>
+inline typename std::enable_if<has_MappingTraits<T, EmptyContext>::value,
+ Output &>::type
+operator<<(Output &yout, T &map) {
+ EmptyContext Ctx;
+ yout.beginDocuments();
+ if ( yout.preflightDocument(0) ) {
+ yamlize(yout, map, true, Ctx);
+ yout.postflightDocument();
+ }
+ yout.endDocuments();
+ return yout;
+}
+
+// Define non-member operator<< so that Output can stream out a sequence.
+template <typename T>
+inline
+typename std::enable_if<has_SequenceTraits<T>::value, Output &>::type
+operator<<(Output &yout, T &seq) {
+ EmptyContext Ctx;
+ yout.beginDocuments();
+ if ( yout.preflightDocument(0) ) {
+ yamlize(yout, seq, true, Ctx);
+ yout.postflightDocument();
+ }
+ yout.endDocuments();
+ return yout;
+}
+
+// Define non-member operator<< so that Output can stream out a block scalar.
+template <typename T>
+inline
+typename std::enable_if<has_BlockScalarTraits<T>::value, Output &>::type
+operator<<(Output &Out, T &Val) {
+ EmptyContext Ctx;
+ Out.beginDocuments();
+ if (Out.preflightDocument(0)) {
+ yamlize(Out, Val, true, Ctx);
+ Out.postflightDocument();
+ }
+ Out.endDocuments();
+ return Out;
+}
+
+// Define non-member operator<< so that Output can stream out a string map.
+template <typename T>
+inline
+typename std::enable_if<has_CustomMappingTraits<T>::value, Output &>::type
+operator<<(Output &Out, T &Val) {
+ EmptyContext Ctx;
+ Out.beginDocuments();
+ if (Out.preflightDocument(0)) {
+ yamlize(Out, Val, true, Ctx);
+ Out.postflightDocument();
+ }
+ Out.endDocuments();
+ return Out;
+}
+
+// Define non-member operator<< so that Output can stream out a polymorphic
+// type.
+template <typename T>
+inline typename std::enable_if<has_PolymorphicTraits<T>::value, Output &>::type
+operator<<(Output &Out, T &Val) {
+ EmptyContext Ctx;
+ Out.beginDocuments();
+ if (Out.preflightDocument(0)) {
+ // FIXME: The parser does not support explicit documents terminated with a
+ // plain scalar; the end-marker is included as part of the scalar token.
+ assert(PolymorphicTraits<T>::getKind(Val) != NodeKind::Scalar && "plain scalar documents are not supported");
+ yamlize(Out, Val, true, Ctx);
+ Out.postflightDocument();
+ }
+ Out.endDocuments();
+ return Out;
+}
+
+// Provide better error message about types missing a trait specialization
+template <typename T>
+inline typename std::enable_if<missingTraits<T, EmptyContext>::value,
+ Output &>::type
+operator<<(Output &yout, T &seq) {
+ char missing_yaml_trait_for_type[sizeof(MissingTrait<T>)];
+ return yout;
+}
+
+template <bool B> struct IsFlowSequenceBase {};
+template <> struct IsFlowSequenceBase<true> { static const bool flow = true; };
+
+template <typename T, bool Flow>
+struct SequenceTraitsImpl : IsFlowSequenceBase<Flow> {
+private:
+ using type = typename T::value_type;
+
+public:
+ static size_t size(IO &io, T &seq) { return seq.size(); }
+
+ static type &element(IO &io, T &seq, size_t index) {
+ if (index >= seq.size())
+ seq.resize(index + 1);
+ return seq[index];
+ }
+};
+
+// Simple helper to check an expression can be used as a bool-valued template
+// argument.
+template <bool> struct CheckIsBool { static const bool value = true; };
+
+// If T has SequenceElementTraits, then vector<T> and SmallVector<T, N> have
+// SequenceTraits that do the obvious thing.
+template <typename T>
+struct SequenceTraits<std::vector<T>,
+ typename std::enable_if<CheckIsBool<
+ SequenceElementTraits<T>::flow>::value>::type>
+ : SequenceTraitsImpl<std::vector<T>, SequenceElementTraits<T>::flow> {};
+template <typename T, unsigned N>
+struct SequenceTraits<SmallVector<T, N>,
+ typename std::enable_if<CheckIsBool<
+ SequenceElementTraits<T>::flow>::value>::type>
+ : SequenceTraitsImpl<SmallVector<T, N>, SequenceElementTraits<T>::flow> {};
+template <typename T>
+struct SequenceTraits<SmallVectorImpl<T>,
+ typename std::enable_if<CheckIsBool<
+ SequenceElementTraits<T>::flow>::value>::type>
+ : SequenceTraitsImpl<SmallVectorImpl<T>, SequenceElementTraits<T>::flow> {};
+
+// Sequences of fundamental types use flow formatting.
+template <typename T>
+struct SequenceElementTraits<
+ T, typename std::enable_if<std::is_fundamental<T>::value>::type> {
+ static const bool flow = true;
+};
+
+// Sequences of strings use block formatting.
+template<> struct SequenceElementTraits<std::string> {
+ static const bool flow = false;
+};
+template<> struct SequenceElementTraits<StringRef> {
+ static const bool flow = false;
+};
+template<> struct SequenceElementTraits<std::pair<std::string, std::string>> {
+ static const bool flow = false;
+};
+
+/// Implementation of CustomMappingTraits for std::map<std::string, T>.
+template <typename T> struct StdMapStringCustomMappingTraitsImpl {
+ using map_type = std::map<std::string, T>;
+
+ static void inputOne(IO &io, StringRef key, map_type &v) {
+ io.mapRequired(key.str().c_str(), v[key]);
+ }
+
+ static void output(IO &io, map_type &v) {
+ for (auto &p : v)
+ io.mapRequired(p.first.c_str(), p.second);
+ }
+};
+
+} // end namespace yaml
+} // end namespace llvm
+
+#define LLVM_YAML_IS_SEQUENCE_VECTOR_IMPL(TYPE, FLOW) \
+ namespace llvm { \
+ namespace yaml { \
+ static_assert( \
+ !std::is_fundamental<TYPE>::value && \
+ !std::is_same<TYPE, std::string>::value && \
+ !std::is_same<TYPE, llvm::StringRef>::value, \
+ "only use LLVM_YAML_IS_SEQUENCE_VECTOR for types you control"); \
+ template <> struct SequenceElementTraits<TYPE> { \
+ static const bool flow = FLOW; \
+ }; \
+ } \
+ }
+
+/// Utility for declaring that a std::vector of a particular type
+/// should be considered a YAML sequence.
+#define LLVM_YAML_IS_SEQUENCE_VECTOR(type) \
+ LLVM_YAML_IS_SEQUENCE_VECTOR_IMPL(type, false)
+
+/// Utility for declaring that a std::vector of a particular type
+/// should be considered a YAML flow sequence.
+#define LLVM_YAML_IS_FLOW_SEQUENCE_VECTOR(type) \
+ LLVM_YAML_IS_SEQUENCE_VECTOR_IMPL(type, true)
+
+#define LLVM_YAML_DECLARE_MAPPING_TRAITS(Type) \
+ namespace llvm { \
+ namespace yaml { \
+ template <> struct MappingTraits<Type> { \
+ static void mapping(IO &IO, Type &Obj); \
+ }; \
+ } \
+ }
+
+#define LLVM_YAML_DECLARE_ENUM_TRAITS(Type) \
+ namespace llvm { \
+ namespace yaml { \
+ template <> struct ScalarEnumerationTraits<Type> { \
+ static void enumeration(IO &io, Type &Value); \
+ }; \
+ } \
+ }
+
+#define LLVM_YAML_DECLARE_BITSET_TRAITS(Type) \
+ namespace llvm { \
+ namespace yaml { \
+ template <> struct ScalarBitSetTraits<Type> { \
+ static void bitset(IO &IO, Type &Options); \
+ }; \
+ } \
+ }
+
+#define LLVM_YAML_DECLARE_SCALAR_TRAITS(Type, MustQuote) \
+ namespace llvm { \
+ namespace yaml { \
+ template <> struct ScalarTraits<Type> { \
+ static void output(const Type &Value, void *ctx, raw_ostream &Out); \
+ static StringRef input(StringRef Scalar, void *ctxt, Type &Value); \
+ static QuotingType mustQuote(StringRef) { return MustQuote; } \
+ }; \
+ } \
+ }
+
+/// Utility for declaring that a std::vector of a particular type
+/// should be considered a YAML document list.
+#define LLVM_YAML_IS_DOCUMENT_LIST_VECTOR(_type) \
+ namespace llvm { \
+ namespace yaml { \
+ template <unsigned N> \
+ struct DocumentListTraits<SmallVector<_type, N>> \
+ : public SequenceTraitsImpl<SmallVector<_type, N>, false> {}; \
+ template <> \
+ struct DocumentListTraits<std::vector<_type>> \
+ : public SequenceTraitsImpl<std::vector<_type>, false> {}; \
+ } \
+ }
+
+/// Utility for declaring that std::map<std::string, _type> should be considered
+/// a YAML map.
+#define LLVM_YAML_IS_STRING_MAP(_type) \
+ namespace llvm { \
+ namespace yaml { \
+ template <> \
+ struct CustomMappingTraits<std::map<std::string, _type>> \
+ : public StdMapStringCustomMappingTraitsImpl<_type> {}; \
+ } \
+ }
+
+LLVM_YAML_IS_FLOW_SEQUENCE_VECTOR(llvm::yaml::Hex64)
+LLVM_YAML_IS_FLOW_SEQUENCE_VECTOR(llvm::yaml::Hex32)
+LLVM_YAML_IS_FLOW_SEQUENCE_VECTOR(llvm::yaml::Hex16)
+LLVM_YAML_IS_FLOW_SEQUENCE_VECTOR(llvm::yaml::Hex8)
+
+#endif // LLVM_SUPPORT_YAMLTRAITS_H
diff --git a/third_party/llvm-project/include/llvm/Support/circular_raw_ostream.h b/third_party/llvm-project/include/llvm/Support/circular_raw_ostream.h
new file mode 100644
index 000000000..a72acd4fe
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/circular_raw_ostream.h
@@ -0,0 +1,159 @@
+//===-- llvm/Support/circular_raw_ostream.h - Buffered streams --*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains raw_ostream implementations for streams to do circular
+// buffering of their output.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_CIRCULAR_RAW_OSTREAM_H
+#define LLVM_SUPPORT_CIRCULAR_RAW_OSTREAM_H
+
+#include "llvm/Support/raw_ostream.h"
+
+namespace llvm {
+ /// circular_raw_ostream - A raw_ostream which *can* save its data
+ /// to a circular buffer, or can pass it through directly to an
+ /// underlying stream if specified with a buffer of zero.
+ ///
+ class circular_raw_ostream : public raw_ostream {
+ public:
+ /// TAKE_OWNERSHIP - Tell this stream that it owns the underlying
+ /// stream and is responsible for cleanup, memory management
+ /// issues, etc.
+ ///
+ static const bool TAKE_OWNERSHIP = true;
+
+ /// REFERENCE_ONLY - Tell this stream it should not manage the
+ /// held stream.
+ ///
+ static const bool REFERENCE_ONLY = false;
+
+ private:
+ /// TheStream - The real stream we output to. We set it to be
+ /// unbuffered, since we're already doing our own buffering.
+ ///
+ raw_ostream *TheStream;
+
+ /// OwnsStream - Are we responsible for managing the underlying
+ /// stream?
+ ///
+ bool OwnsStream;
+
+ /// BufferSize - The size of the buffer in bytes.
+ ///
+ size_t BufferSize;
+
+ /// BufferArray - The actual buffer storage.
+ ///
+ char *BufferArray;
+
+ /// Cur - Pointer to the current output point in BufferArray.
+ ///
+ char *Cur;
+
+ /// Filled - Indicate whether the buffer has been completely
+ /// filled. This helps avoid garbage output.
+ ///
+ bool Filled;
+
+ /// Banner - A pointer to a banner to print before dumping the
+ /// log.
+ ///
+ const char *Banner;
+
+ /// flushBuffer - Dump the contents of the buffer to Stream.
+ ///
+ void flushBuffer() {
+ if (Filled)
+ // Write the older portion of the buffer.
+ TheStream->write(Cur, BufferArray + BufferSize - Cur);
+ // Write the newer portion of the buffer.
+ TheStream->write(BufferArray, Cur - BufferArray);
+ Cur = BufferArray;
+ Filled = false;
+ }
+
+ void write_impl(const char *Ptr, size_t Size) override;
+
+ /// current_pos - Return the current position within the stream,
+ /// not counting the bytes currently in the buffer.
+ ///
+ uint64_t current_pos() const override {
+ // This has the same effect as calling TheStream.current_pos(),
+ // but that interface is private.
+ return TheStream->tell() - TheStream->GetNumBytesInBuffer();
+ }
+
+ public:
+ /// circular_raw_ostream - Construct an optionally
+ /// circular-buffered stream, handing it an underlying stream to
+ /// do the "real" output.
+ ///
+ /// As a side effect, if BuffSize is nonzero, the given Stream is
+ /// set to be Unbuffered. This is because circular_raw_ostream
+ /// does its own buffering, so it doesn't want another layer of
+ /// buffering to be happening underneath it.
+ ///
+ /// "Owns" tells the circular_raw_ostream whether it is
+ /// responsible for managing the held stream, doing memory
+ /// management of it, etc.
+ ///
+ circular_raw_ostream(raw_ostream &Stream, const char *Header,
+ size_t BuffSize = 0, bool Owns = REFERENCE_ONLY)
+ : raw_ostream(/*unbuffered*/ true), TheStream(nullptr),
+ OwnsStream(Owns), BufferSize(BuffSize), BufferArray(nullptr),
+ Filled(false), Banner(Header) {
+ if (BufferSize != 0)
+ BufferArray = new char[BufferSize];
+ Cur = BufferArray;
+ setStream(Stream, Owns);
+ }
+
+ ~circular_raw_ostream() override {
+ flush();
+ flushBufferWithBanner();
+ releaseStream();
+ delete[] BufferArray;
+ }
+
+ bool is_displayed() const override {
+ return TheStream->is_displayed();
+ }
+
+ /// setStream - Tell the circular_raw_ostream to output a
+ /// different stream. "Owns" tells circular_raw_ostream whether
+ /// it should take responsibility for managing the underlying
+ /// stream.
+ ///
+ void setStream(raw_ostream &Stream, bool Owns = REFERENCE_ONLY) {
+ releaseStream();
+ TheStream = &Stream;
+ OwnsStream = Owns;
+ }
+
+ /// flushBufferWithBanner - Force output of the buffer along with
+ /// a small header.
+ ///
+ void flushBufferWithBanner();
+
+ private:
+ /// releaseStream - Delete the held stream if needed. Otherwise,
+ /// transfer the buffer settings from this circular_raw_ostream
+ /// back to the underlying stream.
+ ///
+ void releaseStream() {
+ if (!TheStream)
+ return;
+ if (OwnsStream)
+ delete TheStream;
+ }
+ };
+} // end llvm namespace
+
+#endif
diff --git a/third_party/llvm-project/include/llvm/Support/raw_ostream.h b/third_party/llvm-project/include/llvm/Support/raw_ostream.h
new file mode 100644
index 000000000..705f1790b
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/raw_ostream.h
@@ -0,0 +1,587 @@
+//===--- raw_ostream.h - Raw output stream ----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the raw_ostream class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_RAW_OSTREAM_H
+#define LLVM_SUPPORT_RAW_OSTREAM_H
+
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringRef.h"
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstring>
+#include <string>
+#include <system_error>
+
+namespace llvm {
+
+class formatv_object_base;
+class format_object_base;
+class FormattedString;
+class FormattedNumber;
+class FormattedBytes;
+
+namespace sys {
+namespace fs {
+enum FileAccess : unsigned;
+enum OpenFlags : unsigned;
+enum CreationDisposition : unsigned;
+} // end namespace fs
+} // end namespace sys
+
+/// This class implements an extremely fast bulk output stream that can *only*
+/// output to a stream. It does not support seeking, reopening, rewinding, line
+/// buffered disciplines etc. It is a simple buffer that outputs
+/// a chunk at a time.
+class raw_ostream {
+private:
+ /// The buffer is handled in such a way that the buffer is
+ /// uninitialized, unbuffered, or out of space when OutBufCur >=
+ /// OutBufEnd. Thus a single comparison suffices to determine if we
+ /// need to take the slow path to write a single character.
+ ///
+ /// The buffer is in one of three states:
+ /// 1. Unbuffered (BufferMode == Unbuffered)
+ /// 1. Uninitialized (BufferMode != Unbuffered && OutBufStart == 0).
+ /// 2. Buffered (BufferMode != Unbuffered && OutBufStart != 0 &&
+ /// OutBufEnd - OutBufStart >= 1).
+ ///
+ /// If buffered, then the raw_ostream owns the buffer if (BufferMode ==
+ /// InternalBuffer); otherwise the buffer has been set via SetBuffer and is
+ /// managed by the subclass.
+ ///
+ /// If a subclass installs an external buffer using SetBuffer then it can wait
+ /// for a \see write_impl() call to handle the data which has been put into
+ /// this buffer.
+ char *OutBufStart, *OutBufEnd, *OutBufCur;
+
+ enum class BufferKind {
+ Unbuffered = 0,
+ InternalBuffer,
+ ExternalBuffer
+ } BufferMode;
+
+public:
+ // color order matches ANSI escape sequence, don't change
+ enum class Colors {
+ BLACK = 0,
+ RED,
+ GREEN,
+ YELLOW,
+ BLUE,
+ MAGENTA,
+ CYAN,
+ WHITE,
+ SAVEDCOLOR,
+ RESET,
+ };
+
+ static const Colors BLACK = Colors::BLACK;
+ static const Colors RED = Colors::RED;
+ static const Colors GREEN = Colors::GREEN;
+ static const Colors YELLOW = Colors::YELLOW;
+ static const Colors BLUE = Colors::BLUE;
+ static const Colors MAGENTA = Colors::MAGENTA;
+ static const Colors CYAN = Colors::CYAN;
+ static const Colors WHITE = Colors::WHITE;
+ static const Colors SAVEDCOLOR = Colors::SAVEDCOLOR;
+ static const Colors RESET = Colors::RESET;
+
+ explicit raw_ostream(bool unbuffered = false)
+ : BufferMode(unbuffered ? BufferKind::Unbuffered
+ : BufferKind::InternalBuffer) {
+ // Start out ready to flush.
+ OutBufStart = OutBufEnd = OutBufCur = nullptr;
+ }
+
+ raw_ostream(const raw_ostream &) = delete;
+ void operator=(const raw_ostream &) = delete;
+
+ virtual ~raw_ostream();
+
+ /// tell - Return the current offset with the file.
+ uint64_t tell() const { return current_pos() + GetNumBytesInBuffer(); }
+
+ //===--------------------------------------------------------------------===//
+ // Configuration Interface
+ //===--------------------------------------------------------------------===//
+
+ /// Set the stream to be buffered, with an automatically determined buffer
+ /// size.
+ void SetBuffered();
+
+ /// Set the stream to be buffered, using the specified buffer size.
+ void SetBufferSize(size_t Size) {
+ flush();
+ SetBufferAndMode(new char[Size], Size, BufferKind::InternalBuffer);
+ }
+
+ size_t GetBufferSize() const {
+ // If we're supposed to be buffered but haven't actually gotten around
+ // to allocating the buffer yet, return the value that would be used.
+ if (BufferMode != BufferKind::Unbuffered && OutBufStart == nullptr)
+ return preferred_buffer_size();
+
+ // Otherwise just return the size of the allocated buffer.
+ return OutBufEnd - OutBufStart;
+ }
+
+ /// Set the stream to be unbuffered. When unbuffered, the stream will flush
+ /// after every write. This routine will also flush the buffer immediately
+ /// when the stream is being set to unbuffered.
+ void SetUnbuffered() {
+ flush();
+ SetBufferAndMode(nullptr, 0, BufferKind::Unbuffered);
+ }
+
+ size_t GetNumBytesInBuffer() const {
+ return OutBufCur - OutBufStart;
+ }
+
+ //===--------------------------------------------------------------------===//
+ // Data Output Interface
+ //===--------------------------------------------------------------------===//
+
+ void flush() {
+ if (OutBufCur != OutBufStart)
+ flush_nonempty();
+ }
+
+ raw_ostream &operator<<(char C) {
+ if (OutBufCur >= OutBufEnd)
+ return write(C);
+ *OutBufCur++ = C;
+ return *this;
+ }
+
+ raw_ostream &operator<<(unsigned char C) {
+ if (OutBufCur >= OutBufEnd)
+ return write(C);
+ *OutBufCur++ = C;
+ return *this;
+ }
+
+ raw_ostream &operator<<(signed char C) {
+ if (OutBufCur >= OutBufEnd)
+ return write(C);
+ *OutBufCur++ = C;
+ return *this;
+ }
+
+ raw_ostream &operator<<(StringRef Str) {
+ // Inline fast path, particularly for strings with a known length.
+ size_t Size = Str.size();
+
+ // Make sure we can use the fast path.
+ if (Size > (size_t)(OutBufEnd - OutBufCur))
+ return write(Str.data(), Size);
+
+ if (Size) {
+ memcpy(OutBufCur, Str.data(), Size);
+ OutBufCur += Size;
+ }
+ return *this;
+ }
+
+ raw_ostream &operator<<(const char *Str) {
+ // Inline fast path, particularly for constant strings where a sufficiently
+ // smart compiler will simplify strlen.
+
+ return this->operator<<(StringRef(Str));
+ }
+
+ raw_ostream &operator<<(const std::string &Str) {
+ // Avoid the fast path, it would only increase code size for a marginal win.
+ return write(Str.data(), Str.length());
+ }
+
+ raw_ostream &operator<<(const SmallVectorImpl<char> &Str) {
+ return write(Str.data(), Str.size());
+ }
+
+ raw_ostream &operator<<(unsigned long N);
+ raw_ostream &operator<<(long N);
+ raw_ostream &operator<<(unsigned long long N);
+ raw_ostream &operator<<(long long N);
+ raw_ostream &operator<<(const void *P);
+
+ raw_ostream &operator<<(unsigned int N) {
+ return this->operator<<(static_cast<unsigned long>(N));
+ }
+
+ raw_ostream &operator<<(int N) {
+ return this->operator<<(static_cast<long>(N));
+ }
+
+ raw_ostream &operator<<(double N);
+
+ /// Output \p N in hexadecimal, without any prefix or padding.
+ raw_ostream &write_hex(unsigned long long N);
+
+ // Change the foreground color of text.
+ raw_ostream &operator<<(Colors C);
+
+ /// Output a formatted UUID with dash separators.
+ using uuid_t = uint8_t[16];
+ raw_ostream &write_uuid(const uuid_t UUID);
+
+ /// Output \p Str, turning '\\', '\t', '\n', '"', and anything that doesn't
+ /// satisfy llvm::isPrint into an escape sequence.
+ raw_ostream &write_escaped(StringRef Str, bool UseHexEscapes = false);
+
+ raw_ostream &write(unsigned char C);
+ raw_ostream &write(const char *Ptr, size_t Size);
+
+ // Formatted output, see the format() function in Support/Format.h.
+ raw_ostream &operator<<(const format_object_base &Fmt);
+
+ // Formatted output, see the leftJustify() function in Support/Format.h.
+ raw_ostream &operator<<(const FormattedString &);
+
+ // Formatted output, see the formatHex() function in Support/Format.h.
+ raw_ostream &operator<<(const FormattedNumber &);
+
+ // Formatted output, see the formatv() function in Support/FormatVariadic.h.
+ raw_ostream &operator<<(const formatv_object_base &);
+
+ // Formatted output, see the format_bytes() function in Support/Format.h.
+ raw_ostream &operator<<(const FormattedBytes &);
+
+ /// indent - Insert 'NumSpaces' spaces.
+ raw_ostream &indent(unsigned NumSpaces);
+
+ /// write_zeros - Insert 'NumZeros' nulls.
+ raw_ostream &write_zeros(unsigned NumZeros);
+
+ /// Changes the foreground color of text that will be output from this point
+ /// forward.
+ /// @param Color ANSI color to use, the special SAVEDCOLOR can be used to
+ /// change only the bold attribute, and keep colors untouched
+ /// @param Bold bold/brighter text, default false
+ /// @param BG if true change the background, default: change foreground
+ /// @returns itself so it can be used within << invocations
+ virtual raw_ostream &changeColor(enum Colors Color,
+ bool Bold = false,
+ bool BG = false) {
+ (void)Color;
+ (void)Bold;
+ (void)BG;
+ return *this;
+ }
+
+ /// Resets the colors to terminal defaults. Call this when you are done
+ /// outputting colored text, or before program exit.
+ virtual raw_ostream &resetColor() { return *this; }
+
+ /// Reverses the foreground and background colors.
+ virtual raw_ostream &reverseColor() { return *this; }
+
+ /// This function determines if this stream is connected to a "tty" or
+ /// "console" window. That is, the output would be displayed to the user
+ /// rather than being put on a pipe or stored in a file.
+ virtual bool is_displayed() const { return false; }
+
+ /// This function determines if this stream is displayed and supports colors.
+ virtual bool has_colors() const { return is_displayed(); }
+
+ // Enable or disable colors. Once disable_colors() is called,
+ // changeColor() has no effect until enable_colors() is called.
+ virtual void enable_colors(bool /*enable*/) {}
+
+ //===--------------------------------------------------------------------===//
+ // Subclass Interface
+ //===--------------------------------------------------------------------===//
+
+private:
+ /// The is the piece of the class that is implemented by subclasses. This
+ /// writes the \p Size bytes starting at
+ /// \p Ptr to the underlying stream.
+ ///
+ /// This function is guaranteed to only be called at a point at which it is
+ /// safe for the subclass to install a new buffer via SetBuffer.
+ ///
+ /// \param Ptr The start of the data to be written. For buffered streams this
+ /// is guaranteed to be the start of the buffer.
+ ///
+ /// \param Size The number of bytes to be written.
+ ///
+ /// \invariant { Size > 0 }
+ virtual void write_impl(const char *Ptr, size_t Size) = 0;
+
+ /// Return the current position within the stream, not counting the bytes
+ /// currently in the buffer.
+ virtual uint64_t current_pos() const = 0;
+
+protected:
+ /// Use the provided buffer as the raw_ostream buffer. This is intended for
+ /// use only by subclasses which can arrange for the output to go directly
+ /// into the desired output buffer, instead of being copied on each flush.
+ void SetBuffer(char *BufferStart, size_t Size) {
+ SetBufferAndMode(BufferStart, Size, BufferKind::ExternalBuffer);
+ }
+
+ /// Return an efficient buffer size for the underlying output mechanism.
+ virtual size_t preferred_buffer_size() const;
+
+ /// Return the beginning of the current stream buffer, or 0 if the stream is
+ /// unbuffered.
+ const char *getBufferStart() const { return OutBufStart; }
+
+ //===--------------------------------------------------------------------===//
+ // Private Interface
+ //===--------------------------------------------------------------------===//
+private:
+ /// Install the given buffer and mode.
+ void SetBufferAndMode(char *BufferStart, size_t Size, BufferKind Mode);
+
+ /// Flush the current buffer, which is known to be non-empty. This outputs the
+ /// currently buffered data and resets the buffer to empty.
+ void flush_nonempty();
+
+ /// Copy data into the buffer. Size must not be greater than the number of
+ /// unused bytes in the buffer.
+ void copy_to_buffer(const char *Ptr, size_t Size);
+
+ virtual void anchor();
+};
+
+/// An abstract base class for streams implementations that also support a
+/// pwrite operation. This is useful for code that can mostly stream out data,
+/// but needs to patch in a header that needs to know the output size.
+class raw_pwrite_stream : public raw_ostream {
+ virtual void pwrite_impl(const char *Ptr, size_t Size, uint64_t Offset) = 0;
+ void anchor() override;
+
+public:
+ explicit raw_pwrite_stream(bool Unbuffered = false)
+ : raw_ostream(Unbuffered) {}
+ void pwrite(const char *Ptr, size_t Size, uint64_t Offset) {
+#ifndef NDEBUG
+ uint64_t Pos = tell();
+ // /dev/null always reports a pos of 0, so we cannot perform this check
+ // in that case.
+ if (Pos)
+ assert(Size + Offset <= Pos && "We don't support extending the stream");
+#endif
+ pwrite_impl(Ptr, Size, Offset);
+ }
+};
+
+//===----------------------------------------------------------------------===//
+// File Output Streams
+//===----------------------------------------------------------------------===//
+
+/// A raw_ostream that writes to a file descriptor.
+///
+class raw_fd_ostream : public raw_pwrite_stream {
+ int FD;
+ bool ShouldClose;
+ bool SupportsSeeking = false;
+ bool ColorEnabled = true;
+
+#ifdef _WIN32
+ /// True if this fd refers to a Windows console device. Mintty and other
+ /// terminal emulators are TTYs, but they are not consoles.
+ bool IsWindowsConsole = false;
+#endif
+
+ std::error_code EC;
+
+ uint64_t pos = 0;
+
+ /// See raw_ostream::write_impl.
+ void write_impl(const char *Ptr, size_t Size) override;
+
+ void pwrite_impl(const char *Ptr, size_t Size, uint64_t Offset) override;
+
+ /// Return the current position within the stream, not counting the bytes
+ /// currently in the buffer.
+ uint64_t current_pos() const override { return pos; }
+
+ /// Determine an efficient buffer size.
+ size_t preferred_buffer_size() const override;
+
+ /// Set the flag indicating that an output error has been encountered.
+ void error_detected(std::error_code EC) { this->EC = EC; }
+
+ void anchor() override;
+
+public:
+ /// Open the specified file for writing. If an error occurs, information
+ /// about the error is put into EC, and the stream should be immediately
+ /// destroyed;
+ /// \p Flags allows optional flags to control how the file will be opened.
+ ///
+ /// As a special case, if Filename is "-", then the stream will use
+ /// STDOUT_FILENO instead of opening a file. This will not close the stdout
+ /// descriptor.
+ raw_fd_ostream(StringRef Filename, std::error_code &EC);
+ raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::CreationDisposition Disp);
+ raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::FileAccess Access);
+ raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::OpenFlags Flags);
+ raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::CreationDisposition Disp, sys::fs::FileAccess Access,
+ sys::fs::OpenFlags Flags);
+
+ /// FD is the file descriptor that this writes to. If ShouldClose is true,
+ /// this closes the file when the stream is destroyed. If FD is for stdout or
+ /// stderr, it will not be closed.
+ raw_fd_ostream(int fd, bool shouldClose, bool unbuffered=false);
+
+ ~raw_fd_ostream() override;
+
+ /// Manually flush the stream and close the file. Note that this does not call
+ /// fsync.
+ void close();
+
+ bool supportsSeeking() { return SupportsSeeking; }
+
+ /// Flushes the stream and repositions the underlying file descriptor position
+ /// to the offset specified from the beginning of the file.
+ uint64_t seek(uint64_t off);
+
+ raw_ostream &changeColor(enum Colors colors, bool bold=false,
+ bool bg=false) override;
+ raw_ostream &resetColor() override;
+
+ raw_ostream &reverseColor() override;
+
+ bool is_displayed() const override;
+
+ bool has_colors() const override;
+
+ void enable_colors(bool enable) override { ColorEnabled = enable; }
+
+ std::error_code error() const { return EC; }
+
+ /// Return the value of the flag in this raw_fd_ostream indicating whether an
+ /// output error has been encountered.
+ /// This doesn't implicitly flush any pending output. Also, it doesn't
+ /// guarantee to detect all errors unless the stream has been closed.
+ bool has_error() const { return bool(EC); }
+
+ /// Set the flag read by has_error() to false. If the error flag is set at the
+ /// time when this raw_ostream's destructor is called, report_fatal_error is
+ /// called to report the error. Use clear_error() after handling the error to
+ /// avoid this behavior.
+ ///
+ /// "Errors should never pass silently.
+ /// Unless explicitly silenced."
+ /// - from The Zen of Python, by Tim Peters
+ ///
+ void clear_error() { EC = std::error_code(); }
+};
+
+/// This returns a reference to a raw_ostream for standard output. Use it like:
+/// outs() << "foo" << "bar";
+raw_ostream &outs();
+
+/// This returns a reference to a raw_ostream for standard error. Use it like:
+/// errs() << "foo" << "bar";
+raw_ostream &errs();
+
+/// This returns a reference to a raw_ostream which simply discards output.
+raw_ostream &nulls();
+
+//===----------------------------------------------------------------------===//
+// Output Stream Adaptors
+//===----------------------------------------------------------------------===//
+
+/// A raw_ostream that writes to an std::string. This is a simple adaptor
+/// class. This class does not encounter output errors.
+class raw_string_ostream : public raw_ostream {
+ std::string &OS;
+
+ /// See raw_ostream::write_impl.
+ void write_impl(const char *Ptr, size_t Size) override;
+
+ /// Return the current position within the stream, not counting the bytes
+ /// currently in the buffer.
+ uint64_t current_pos() const override { return OS.size(); }
+
+public:
+ explicit raw_string_ostream(std::string &O) : OS(O) {}
+ ~raw_string_ostream() override;
+
+ /// Flushes the stream contents to the target string and returns the string's
+ /// reference.
+ std::string& str() {
+ flush();
+ return OS;
+ }
+};
+
+/// A raw_ostream that writes to an SmallVector or SmallString. This is a
+/// simple adaptor class. This class does not encounter output errors.
+/// raw_svector_ostream operates without a buffer, delegating all memory
+/// management to the SmallString. Thus the SmallString is always up-to-date,
+/// may be used directly and there is no need to call flush().
+class raw_svector_ostream : public raw_pwrite_stream {
+ SmallVectorImpl<char> &OS;
+
+ /// See raw_ostream::write_impl.
+ void write_impl(const char *Ptr, size_t Size) override;
+
+ void pwrite_impl(const char *Ptr, size_t Size, uint64_t Offset) override;
+
+ /// Return the current position within the stream.
+ uint64_t current_pos() const override;
+
+public:
+ /// Construct a new raw_svector_ostream.
+ ///
+ /// \param O The vector to write to; this should generally have at least 128
+ /// bytes free to avoid any extraneous memory overhead.
+ explicit raw_svector_ostream(SmallVectorImpl<char> &O) : OS(O) {
+ SetUnbuffered();
+ }
+
+ ~raw_svector_ostream() override = default;
+
+ void flush() = delete;
+
+ /// Return a StringRef for the vector contents.
+ StringRef str() { return StringRef(OS.data(), OS.size()); }
+};
+
+/// A raw_ostream that discards all output.
+class raw_null_ostream : public raw_pwrite_stream {
+ /// See raw_ostream::write_impl.
+ void write_impl(const char *Ptr, size_t size) override;
+ void pwrite_impl(const char *Ptr, size_t Size, uint64_t Offset) override;
+
+ /// Return the current position within the stream, not counting the bytes
+ /// currently in the buffer.
+ uint64_t current_pos() const override;
+
+public:
+ explicit raw_null_ostream() = default;
+ ~raw_null_ostream() override;
+};
+
+class buffer_ostream : public raw_svector_ostream {
+ raw_ostream &OS;
+ SmallVector<char, 0> Buffer;
+
+ virtual void anchor() override;
+
+public:
+ buffer_ostream(raw_ostream &OS) : raw_svector_ostream(Buffer), OS(OS) {}
+ ~buffer_ostream() override { OS << str(); }
+};
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_RAW_OSTREAM_H
diff --git a/third_party/llvm-project/include/llvm/Support/type_traits.h b/third_party/llvm-project/include/llvm/Support/type_traits.h
new file mode 100644
index 000000000..b7d48e8e1
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/Support/type_traits.h
@@ -0,0 +1,192 @@
+//===- llvm/Support/type_traits.h - Simplfied type traits -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides useful additions to the standard type_traits library.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_TYPE_TRAITS_H
+#define LLVM_SUPPORT_TYPE_TRAITS_H
+
+#include "llvm/Support/Compiler.h"
+#include <type_traits>
+#include <utility>
+
+namespace llvm {
+
+
+/// Metafunction that determines whether the given type is either an
+/// integral type or an enumeration type, including enum classes.
+///
+/// Note that this accepts potentially more integral types than is_integral
+/// because it is based on being implicitly convertible to an integral type.
+/// Also note that enum classes aren't implicitly convertible to integral types,
+/// the value may therefore need to be explicitly converted before being used.
+template <typename T> class is_integral_or_enum {
+ using UnderlyingT = typename std::remove_reference<T>::type;
+
+public:
+ static const bool value =
+ !std::is_class<UnderlyingT>::value && // Filter conversion operators.
+ !std::is_pointer<UnderlyingT>::value &&
+ !std::is_floating_point<UnderlyingT>::value &&
+ (std::is_enum<UnderlyingT>::value ||
+ std::is_convertible<UnderlyingT, unsigned long long>::value);
+};
+
+/// If T is a pointer, just return it. If it is not, return T&.
+template<typename T, typename Enable = void>
+struct add_lvalue_reference_if_not_pointer { using type = T &; };
+
+template <typename T>
+struct add_lvalue_reference_if_not_pointer<
+ T, typename std::enable_if<std::is_pointer<T>::value>::type> {
+ using type = T;
+};
+
+/// If T is a pointer to X, return a pointer to const X. If it is not,
+/// return const T.
+template<typename T, typename Enable = void>
+struct add_const_past_pointer { using type = const T; };
+
+template <typename T>
+struct add_const_past_pointer<
+ T, typename std::enable_if<std::is_pointer<T>::value>::type> {
+ using type = const typename std::remove_pointer<T>::type *;
+};
+
+template <typename T, typename Enable = void>
+struct const_pointer_or_const_ref {
+ using type = const T &;
+};
+template <typename T>
+struct const_pointer_or_const_ref<
+ T, typename std::enable_if<std::is_pointer<T>::value>::type> {
+ using type = typename add_const_past_pointer<T>::type;
+};
+
+namespace detail {
+/// Internal utility to detect trivial copy construction.
+template<typename T> union copy_construction_triviality_helper {
+ T t;
+ copy_construction_triviality_helper() = default;
+ copy_construction_triviality_helper(const copy_construction_triviality_helper&) = default;
+ ~copy_construction_triviality_helper() = default;
+};
+/// Internal utility to detect trivial move construction.
+template<typename T> union move_construction_triviality_helper {
+ T t;
+ move_construction_triviality_helper() = default;
+ move_construction_triviality_helper(move_construction_triviality_helper&&) = default;
+ ~move_construction_triviality_helper() = default;
+};
+
+template<class T>
+union trivial_helper {
+ T t;
+};
+
+} // end namespace detail
+
+/// An implementation of `std::is_trivially_copy_constructible` since we have
+/// users with STLs that don't yet include it.
+template <typename T>
+struct is_trivially_copy_constructible
+ : std::is_copy_constructible<
+ ::llvm::detail::copy_construction_triviality_helper<T>> {};
+template <typename T>
+struct is_trivially_copy_constructible<T &> : std::true_type {};
+template <typename T>
+struct is_trivially_copy_constructible<T &&> : std::false_type {};
+
+/// An implementation of `std::is_trivially_move_constructible` since we have
+/// users with STLs that don't yet include it.
+template <typename T>
+struct is_trivially_move_constructible
+ : std::is_move_constructible<
+ ::llvm::detail::move_construction_triviality_helper<T>> {};
+template <typename T>
+struct is_trivially_move_constructible<T &> : std::true_type {};
+template <typename T>
+struct is_trivially_move_constructible<T &&> : std::true_type {};
+
+
+template <typename T>
+struct is_copy_assignable {
+ template<class F>
+ static auto get(F*) -> decltype(std::declval<F &>() = std::declval<const F &>(), std::true_type{});
+ static std::false_type get(...);
+ static constexpr bool value = decltype(get((T*)nullptr))::value;
+};
+
+template <typename T>
+struct is_move_assignable {
+ template<class F>
+ static auto get(F*) -> decltype(std::declval<F &>() = std::declval<F &&>(), std::true_type{});
+ static std::false_type get(...);
+ static constexpr bool value = decltype(get((T*)nullptr))::value;
+};
+
+
+// An implementation of `std::is_trivially_copyable` since STL version
+// is not equally supported by all compilers, especially GCC 4.9.
+// Uniform implementation of this trait is important for ABI compatibility
+// as it has an impact on SmallVector's ABI (among others).
+template <typename T>
+class is_trivially_copyable {
+
+ // copy constructors
+ static constexpr bool has_trivial_copy_constructor =
+ std::is_copy_constructible<detail::trivial_helper<T>>::value;
+ static constexpr bool has_deleted_copy_constructor =
+ !std::is_copy_constructible<T>::value;
+
+ // move constructors
+ static constexpr bool has_trivial_move_constructor =
+ std::is_move_constructible<detail::trivial_helper<T>>::value;
+ static constexpr bool has_deleted_move_constructor =
+ !std::is_move_constructible<T>::value;
+
+ // copy assign
+ static constexpr bool has_trivial_copy_assign =
+ is_copy_assignable<detail::trivial_helper<T>>::value;
+ static constexpr bool has_deleted_copy_assign =
+ !is_copy_assignable<T>::value;
+
+ // move assign
+ static constexpr bool has_trivial_move_assign =
+ is_move_assignable<detail::trivial_helper<T>>::value;
+ static constexpr bool has_deleted_move_assign =
+ !is_move_assignable<T>::value;
+
+ // destructor
+ static constexpr bool has_trivial_destructor =
+ std::is_destructible<detail::trivial_helper<T>>::value;
+
+ public:
+
+ static constexpr bool value =
+ has_trivial_destructor &&
+ (has_deleted_move_assign || has_trivial_move_assign) &&
+ (has_deleted_move_constructor || has_trivial_move_constructor) &&
+ (has_deleted_copy_assign || has_trivial_copy_assign) &&
+ (has_deleted_copy_constructor || has_trivial_copy_constructor);
+
+#ifdef HAVE_STD_IS_TRIVIALLY_COPYABLE
+ static_assert(value == std::is_trivially_copyable<T>::value,
+ "inconsistent behavior between llvm:: and std:: implementation of is_trivially_copyable");
+#endif
+};
+template <typename T>
+class is_trivially_copyable<T*> : public std::true_type {
+};
+
+
+} // end namespace llvm
+
+#endif // LLVM_SUPPORT_TYPE_TRAITS_H
diff --git a/third_party/llvm-project/include/llvm/include/llvm/DebugInfo/DWARFContext.h b/third_party/llvm-project/include/llvm/include/llvm/DebugInfo/DWARFContext.h
new file mode 100644
index 000000000..2dec107d1
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/include/llvm/DebugInfo/DWARFContext.h
@@ -0,0 +1,382 @@
+//===- DWARFContext.h -------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===/
+
+#ifndef LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
+#define LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
+
+#include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/iterator_range.h"
+#include "llvm/DebugInfo/DIContext.h"
+#include "llvm/DebugInfo/DWARF/DWARFAcceleratorTable.h"
+#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugAranges.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugFrame.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugLoc.h"
+#include "llvm/DebugInfo/DWARF/DWARFDebugMacro.h"
+#include "llvm/DebugInfo/DWARF/DWARFDie.h"
+#include "llvm/DebugInfo/DWARF/DWARFGdbIndex.h"
+#include "llvm/DebugInfo/DWARF/DWARFObject.h"
+#include "llvm/DebugInfo/DWARF/DWARFSection.h"
+#include "llvm/DebugInfo/DWARF/DWARFTypeUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnit.h"
+#include "llvm/DebugInfo/DWARF/DWARFUnitIndex.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/Host.h"
+#include <cstdint>
+#include <deque>
+#include <map>
+#include <memory>
+
+namespace llvm {
+
+class MCRegisterInfo;
+class MemoryBuffer;
+class raw_ostream;
+
+/// Used as a return value for a error callback passed to DWARF context.
+/// Callback should return Halt if client application wants to stop
+/// object parsing, or should return Continue otherwise.
+enum class ErrorPolicy { Halt, Continue };
+
+/// DWARFContext
+/// This data structure is the top level entity that deals with dwarf debug
+/// information parsing. The actual data is supplied through DWARFObj.
+class DWARFContext : public DIContext {
+ DWARFUnitVector NormalUnits;
+ std::unique_ptr<DWARFUnitIndex> CUIndex;
+ std::unique_ptr<DWARFGdbIndex> GdbIndex;
+ std::unique_ptr<DWARFUnitIndex> TUIndex;
+ std::unique_ptr<DWARFDebugAbbrev> Abbrev;
+ std::unique_ptr<DWARFDebugLoc> Loc;
+ std::unique_ptr<DWARFDebugAranges> Aranges;
+ std::unique_ptr<DWARFDebugLine> Line;
+ std::unique_ptr<DWARFDebugFrame> DebugFrame;
+ std::unique_ptr<DWARFDebugFrame> EHFrame;
+ std::unique_ptr<DWARFDebugMacro> Macro;
+ std::unique_ptr<DWARFDebugNames> Names;
+ std::unique_ptr<AppleAcceleratorTable> AppleNames;
+ std::unique_ptr<AppleAcceleratorTable> AppleTypes;
+ std::unique_ptr<AppleAcceleratorTable> AppleNamespaces;
+ std::unique_ptr<AppleAcceleratorTable> AppleObjC;
+
+ DWARFUnitVector DWOUnits;
+ std::unique_ptr<DWARFDebugAbbrev> AbbrevDWO;
+
+ /// The maximum DWARF version of all units.
+ unsigned MaxVersion = 0;
+
+ struct DWOFile {
+ object::OwningBinary<object::ObjectFile> File;
+ std::unique_ptr<DWARFContext> Context;
+ };
+ StringMap<std::weak_ptr<DWOFile>> DWOFiles;
+ std::weak_ptr<DWOFile> DWP;
+ bool CheckedForDWP = false;
+ std::string DWPName;
+
+ std::unique_ptr<MCRegisterInfo> RegInfo;
+
+ /// Read compile units from the debug_info section (if necessary)
+ /// and type units from the debug_types sections (if necessary)
+ /// and store them in NormalUnits.
+ void parseNormalUnits();
+
+ /// Read compile units from the debug_info.dwo section (if necessary)
+ /// and type units from the debug_types.dwo section (if necessary)
+ /// and store them in DWOUnits.
+ /// If \p Lazy is true, set up to parse but don't actually parse them.
+ enum { EagerParse = false, LazyParse = true };
+ void parseDWOUnits(bool Lazy = false);
+
+ std::unique_ptr<const DWARFObject> DObj;
+
+public:
+ DWARFContext(std::unique_ptr<const DWARFObject> DObj,
+ std::string DWPName = "");
+ ~DWARFContext();
+
+ DWARFContext(DWARFContext &) = delete;
+ DWARFContext &operator=(DWARFContext &) = delete;
+
+ const DWARFObject &getDWARFObj() const { return *DObj; }
+
+ static bool classof(const DIContext *DICtx) {
+ return DICtx->getKind() == CK_DWARF;
+ }
+
+ /// Dump a textual representation to \p OS. If any \p DumpOffsets are present,
+ /// dump only the record at the specified offset.
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts,
+ std::array<Optional<uint64_t>, DIDT_ID_Count> DumpOffsets);
+
+ void dump(raw_ostream &OS, DIDumpOptions DumpOpts) override {
+ std::array<Optional<uint64_t>, DIDT_ID_Count> DumpOffsets;
+ dump(OS, DumpOpts, DumpOffsets);
+ }
+
+ bool verify(raw_ostream &OS, DIDumpOptions DumpOpts = {}) override;
+
+ using unit_iterator_range = DWARFUnitVector::iterator_range;
+
+ /// Get units from .debug_info in this context.
+ unit_iterator_range info_section_units() {
+ parseNormalUnits();
+ return unit_iterator_range(NormalUnits.begin(),
+ NormalUnits.begin() +
+ NormalUnits.getNumInfoUnits());
+ }
+
+ /// Get units from .debug_types in this context.
+ unit_iterator_range types_section_units() {
+ parseNormalUnits();
+ return unit_iterator_range(
+ NormalUnits.begin() + NormalUnits.getNumInfoUnits(), NormalUnits.end());
+ }
+
+ /// Get compile units in this context.
+ unit_iterator_range compile_units() { return info_section_units(); }
+
+ /// Get type units in this context.
+ unit_iterator_range type_units() { return types_section_units(); }
+
+ /// Get all normal compile/type units in this context.
+ unit_iterator_range normal_units() {
+ parseNormalUnits();
+ return unit_iterator_range(NormalUnits.begin(), NormalUnits.end());
+ }
+
+ /// Get units from .debug_info..dwo in the DWO context.
+ unit_iterator_range dwo_info_section_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin(),
+ DWOUnits.begin() + DWOUnits.getNumInfoUnits());
+ }
+
+ /// Get units from .debug_types.dwo in the DWO context.
+ unit_iterator_range dwo_types_section_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin() + DWOUnits.getNumInfoUnits(),
+ DWOUnits.end());
+ }
+
+ /// Get compile units in the DWO context.
+ unit_iterator_range dwo_compile_units() { return dwo_info_section_units(); }
+
+ /// Get type units in the DWO context.
+ unit_iterator_range dwo_type_units() { return dwo_types_section_units(); }
+
+ /// Get all units in the DWO context.
+ unit_iterator_range dwo_units() {
+ parseDWOUnits();
+ return unit_iterator_range(DWOUnits.begin(), DWOUnits.end());
+ }
+
+ /// Get the number of compile units in this context.
+ unsigned getNumCompileUnits() {
+ parseNormalUnits();
+ return NormalUnits.getNumInfoUnits();
+ }
+
+ /// Get the number of type units in this context.
+ unsigned getNumTypeUnits() {
+ parseNormalUnits();
+ return NormalUnits.getNumTypesUnits();
+ }
+
+ /// Get the number of compile units in the DWO context.
+ unsigned getNumDWOCompileUnits() {
+ parseDWOUnits();
+ return DWOUnits.getNumInfoUnits();
+ }
+
+ /// Get the number of type units in the DWO context.
+ unsigned getNumDWOTypeUnits() {
+ parseDWOUnits();
+ return DWOUnits.getNumTypesUnits();
+ }
+
+ /// Get the unit at the specified index.
+ DWARFUnit *getUnitAtIndex(unsigned index) {
+ parseNormalUnits();
+ return NormalUnits[index].get();
+ }
+
+ /// Get the unit at the specified index for the DWO units.
+ DWARFUnit *getDWOUnitAtIndex(unsigned index) {
+ parseDWOUnits();
+ return DWOUnits[index].get();
+ }
+
+ DWARFCompileUnit *getDWOCompileUnitForHash(uint64_t Hash);
+
+ /// Return the compile unit that includes an offset (relative to .debug_info).
+ DWARFCompileUnit *getCompileUnitForOffset(uint64_t Offset);
+
+ /// Get a DIE given an exact offset.
+ DWARFDie getDIEForOffset(uint64_t Offset);
+
+ unsigned getMaxVersion() {
+ // Ensure info units have been parsed to discover MaxVersion
+ info_section_units();
+ return MaxVersion;
+ }
+
+ unsigned getMaxDWOVersion() {
+ // Ensure DWO info units have been parsed to discover MaxVersion
+ dwo_info_section_units();
+ return MaxVersion;
+ }
+
+ void setMaxVersionIfGreater(unsigned Version) {
+ if (Version > MaxVersion)
+ MaxVersion = Version;
+ }
+
+ const DWARFUnitIndex &getCUIndex();
+ DWARFGdbIndex &getGdbIndex();
+ const DWARFUnitIndex &getTUIndex();
+
+ /// Get a pointer to the parsed DebugAbbrev object.
+ const DWARFDebugAbbrev *getDebugAbbrev();
+
+ /// Get a pointer to the parsed DebugLoc object.
+ const DWARFDebugLoc *getDebugLoc();
+
+ /// Get a pointer to the parsed dwo abbreviations object.
+ const DWARFDebugAbbrev *getDebugAbbrevDWO();
+
+ /// Get a pointer to the parsed DebugAranges object.
+ const DWARFDebugAranges *getDebugAranges();
+
+ /// Get a pointer to the parsed frame information object.
+ const DWARFDebugFrame *getDebugFrame();
+
+ /// Get a pointer to the parsed eh frame information object.
+ const DWARFDebugFrame *getEHFrame();
+
+ /// Get a pointer to the parsed DebugMacro object.
+ const DWARFDebugMacro *getDebugMacro();
+
+ /// Get a reference to the parsed accelerator table object.
+ const DWARFDebugNames &getDebugNames();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleNames();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleTypes();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleNamespaces();
+
+ /// Get a reference to the parsed accelerator table object.
+ const AppleAcceleratorTable &getAppleObjC();
+
+ /// Get a pointer to a parsed line table corresponding to a compile unit.
+ /// Report any parsing issues as warnings on stderr.
+ const DWARFDebugLine::LineTable *getLineTableForUnit(DWARFUnit *U);
+
+ /// Get a pointer to a parsed line table corresponding to a compile unit.
+ /// Report any recoverable parsing problems using the callback.
+ Expected<const DWARFDebugLine::LineTable *>
+ getLineTableForUnit(DWARFUnit *U,
+ std::function<void(Error)> RecoverableErrorCallback);
+
+ DataExtractor getStringExtractor() const {
+ return DataExtractor(DObj->getStrSection(), false, 0);
+ }
+ DataExtractor getLineStringExtractor() const {
+ return DataExtractor(DObj->getLineStrSection(), false, 0);
+ }
+
+ /// Wraps the returned DIEs for a given address.
+ struct DIEsForAddress {
+ DWARFCompileUnit *CompileUnit = nullptr;
+ DWARFDie FunctionDIE;
+ DWARFDie BlockDIE;
+ explicit operator bool() const { return CompileUnit != nullptr; }
+ };
+
+ /// Get the compilation unit, the function DIE and lexical block DIE for the
+ /// given address where applicable.
+ /// TODO: change input parameter from "uint64_t Address"
+ /// into "SectionedAddress Address"
+ DIEsForAddress getDIEsForAddress(uint64_t Address);
+
+ DILineInfo getLineInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+ DILineInfoTable getLineInfoForAddressRange(
+ object::SectionedAddress Address, uint64_t Size,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+ DIInliningInfo getInliningInfoForAddress(
+ object::SectionedAddress Address,
+ DILineInfoSpecifier Specifier = DILineInfoSpecifier()) override;
+
+ std::vector<DILocal>
+ getLocalsForAddress(object::SectionedAddress Address) override;
+
+ bool isLittleEndian() const { return DObj->isLittleEndian(); }
+ static bool isSupportedVersion(unsigned version) {
+ return version == 2 || version == 3 || version == 4 || version == 5;
+ }
+
+ std::shared_ptr<DWARFContext> getDWOContext(StringRef AbsolutePath);
+
+ const MCRegisterInfo *getRegisterInfo() const { return RegInfo.get(); }
+
+ /// Function used to handle default error reporting policy. Prints a error
+ /// message and returns Continue, so DWARF context ignores the error.
+ static ErrorPolicy defaultErrorHandler(Error E);
+ static std::unique_ptr<DWARFContext>
+ create(const object::ObjectFile &Obj, const LoadedObjectInfo *L = nullptr,
+ function_ref<ErrorPolicy(Error)> HandleError = defaultErrorHandler,
+ std::string DWPName = "");
+
+ static std::unique_ptr<DWARFContext>
+ create(const StringMap<std::unique_ptr<MemoryBuffer>> &Sections,
+ uint8_t AddrSize, bool isLittleEndian = sys::IsLittleEndianHost);
+
+ /// Loads register info for the architecture of the provided object file.
+ /// Improves readability of dumped DWARF expressions. Requires the caller to
+ /// have initialized the relevant target descriptions.
+ Error loadRegisterInfo(const object::ObjectFile &Obj);
+
+ /// Get address size from CUs.
+ /// TODO: refactor compile_units() to make this const.
+ uint8_t getCUAddrSize();
+
+ /// Dump Error as warning message to stderr.
+ static void dumpWarning(Error Warning);
+
+ Triple::ArchType getArch() const {
+ return getDWARFObj().getFile()->getArch();
+ }
+
+private:
+ /// Return the compile unit which contains instruction with provided
+ /// address.
+ /// TODO: change input parameter from "uint64_t Address"
+ /// into "SectionedAddress Address"
+ DWARFCompileUnit *getCompileUnitForAddress(uint64_t Address);
+ void addLocalsForDie(DWARFCompileUnit *CU, DWARFDie Subprogram, DWARFDie Die,
+ std::vector<DILocal> &Result);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_DEBUGINFO_DWARF_DWARFCONTEXT_H
diff --git a/third_party/llvm-project/include/llvm/readme.txt b/third_party/llvm-project/include/llvm/readme.txt
new file mode 100644
index 000000000..9549438f6
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/readme.txt
@@ -0,0 +1,15 @@
+This folder contains files from LLVM. See LICENSE.TXT for their license.
+
+These files were synced from
+
+commit 6c86d6efaf129c42d37121f1e7e9a7adffb54c1a (origin/master, master)
+Author: Craig Topper <craig.topper@intel.com>
+Date: Mon Nov 11 16:30:42 2019 -0800
+
+ [X86] Remove some else branches after checking for !useSoftFloat() that set operations to Expand.
+
+ If we're using soft floats, then these operations shoudl be
+ softened during type legalization. They'll never get to
+ LegalizeVectorOps or LegalizeDAG so they don't need to be
+ Expanded there.
+
diff --git a/third_party/llvm-project/obj2yaml_Error.cpp b/third_party/llvm-project/obj2yaml_Error.cpp
new file mode 100644
index 000000000..d11d47b5b
--- /dev/null
+++ b/third_party/llvm-project/obj2yaml_Error.cpp
@@ -0,0 +1,61 @@
+//===- Error.cpp - system_error extensions for obj2yaml ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "Error.h"
+#include "llvm/Support/ErrorHandling.h"
+
+using namespace llvm;
+
+namespace {
+// FIXME: This class is only here to support the transition to llvm::Error. It
+// will be removed once this transition is complete. Clients should prefer to
+// deal with the Error value directly, rather than converting to error_code.
+class _obj2yaml_error_category : public std::error_category {
+public:
+ const char *name() const noexcept override;
+ std::string message(int ev) const override;
+};
+} // namespace
+
+const char *_obj2yaml_error_category::name() const noexcept {
+ return "obj2yaml";
+}
+
+std::string _obj2yaml_error_category::message(int ev) const {
+ switch (static_cast<obj2yaml_error>(ev)) {
+ case obj2yaml_error::success:
+ return "Success";
+ case obj2yaml_error::file_not_found:
+ return "No such file.";
+ case obj2yaml_error::unrecognized_file_format:
+ return "Unrecognized file type.";
+ case obj2yaml_error::unsupported_obj_file_format:
+ return "Unsupported object file format.";
+ case obj2yaml_error::not_implemented:
+ return "Feature not yet implemented.";
+ }
+ llvm_unreachable("An enumerator of obj2yaml_error does not have a message "
+ "defined.");
+}
+
+namespace llvm {
+
+const std::error_category &obj2yaml_category() {
+ static _obj2yaml_error_category o;
+ return o;
+}
+
+char Obj2YamlError::ID = 0;
+
+void Obj2YamlError::log(raw_ostream &OS) const { OS << ErrMsg; }
+
+std::error_code Obj2YamlError::convertToErrorCode() const {
+ return std::error_code(static_cast<int>(Code), obj2yaml_category());
+}
+
+} // namespace llvm
diff --git a/third_party/llvm-project/raw_ostream.cpp b/third_party/llvm-project/raw_ostream.cpp
new file mode 100644
index 000000000..63e0fcbf1
--- /dev/null
+++ b/third_party/llvm-project/raw_ostream.cpp
@@ -0,0 +1,712 @@
+//===--- raw_ostream.cpp - Implement the raw_ostream classes --------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This implements support for bulk buffered stream output.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/raw_ostream.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/Config/config.h"
+#include "llvm/Support/Compiler.h"
+#include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/Format.h"
+#include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/MathExtras.h"
+#include "llvm/Support/NativeFormatting.h"
+#include "llvm/Support/Process.h"
+#include "llvm/Support/Program.h"
+#include <algorithm>
+#include <cctype>
+#include <cerrno>
+#include <cstdio>
+#include <iterator>
+#include <sys/stat.h>
+#include <system_error>
+
+// XXX BINARYEn
+#include <iostream>
+
+// <fcntl.h> may provide O_BINARY.
+#if defined(HAVE_FCNTL_H)
+# include <fcntl.h>
+#endif
+
+#if defined(HAVE_UNISTD_H)
+# include <unistd.h>
+#endif
+
+#if defined(__CYGWIN__)
+#include <io.h>
+#endif
+
+#if defined(_MSC_VER)
+#include <io.h>
+#ifndef STDIN_FILENO
+# define STDIN_FILENO 0
+#endif
+#ifndef STDOUT_FILENO
+# define STDOUT_FILENO 1
+#endif
+#ifndef STDERR_FILENO
+# define STDERR_FILENO 2
+#endif
+#endif
+
+#if 0 // XXX BINARYEN def _WIN32
+#include "llvm/Support/ConvertUTF.h"
+#include "Windows/WindowsSupport.h"
+#endif
+
+using namespace llvm;
+
+const raw_ostream::Colors raw_ostream::BLACK;
+const raw_ostream::Colors raw_ostream::RED;
+const raw_ostream::Colors raw_ostream::GREEN;
+const raw_ostream::Colors raw_ostream::YELLOW;
+const raw_ostream::Colors raw_ostream::BLUE;
+const raw_ostream::Colors raw_ostream::MAGENTA;
+const raw_ostream::Colors raw_ostream::CYAN;
+const raw_ostream::Colors raw_ostream::WHITE;
+const raw_ostream::Colors raw_ostream::SAVEDCOLOR;
+const raw_ostream::Colors raw_ostream::RESET;
+
+raw_ostream::~raw_ostream() {
+ // raw_ostream's subclasses should take care to flush the buffer
+ // in their destructors.
+ assert(OutBufCur == OutBufStart &&
+ "raw_ostream destructor called with non-empty buffer!");
+
+ if (BufferMode == BufferKind::InternalBuffer)
+ delete [] OutBufStart;
+}
+
+size_t raw_ostream::preferred_buffer_size() const {
+ // BUFSIZ is intended to be a reasonable default.
+ return BUFSIZ;
+}
+
+void raw_ostream::SetBuffered() {
+ // Ask the subclass to determine an appropriate buffer size.
+ if (size_t Size = preferred_buffer_size())
+ SetBufferSize(Size);
+ else
+ // It may return 0, meaning this stream should be unbuffered.
+ SetUnbuffered();
+}
+
+void raw_ostream::SetBufferAndMode(char *BufferStart, size_t Size,
+ BufferKind Mode) {
+ assert(((Mode == BufferKind::Unbuffered && !BufferStart && Size == 0) ||
+ (Mode != BufferKind::Unbuffered && BufferStart && Size != 0)) &&
+ "stream must be unbuffered or have at least one byte");
+ // Make sure the current buffer is free of content (we can't flush here; the
+ // child buffer management logic will be in write_impl).
+ assert(GetNumBytesInBuffer() == 0 && "Current buffer is non-empty!");
+
+ if (BufferMode == BufferKind::InternalBuffer)
+ delete [] OutBufStart;
+ OutBufStart = BufferStart;
+ OutBufEnd = OutBufStart+Size;
+ OutBufCur = OutBufStart;
+ BufferMode = Mode;
+
+ assert(OutBufStart <= OutBufEnd && "Invalid size!");
+}
+
+raw_ostream &raw_ostream::operator<<(unsigned long N) {
+ write_integer(*this, static_cast<uint64_t>(N), 0, IntegerStyle::Integer);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(long N) {
+ write_integer(*this, static_cast<int64_t>(N), 0, IntegerStyle::Integer);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(unsigned long long N) {
+ write_integer(*this, static_cast<uint64_t>(N), 0, IntegerStyle::Integer);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(long long N) {
+ write_integer(*this, static_cast<int64_t>(N), 0, IntegerStyle::Integer);
+ return *this;
+}
+
+raw_ostream &raw_ostream::write_hex(unsigned long long N) {
+ llvm::write_hex(*this, N, HexPrintStyle::Lower);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(Colors C) {
+ if (C == Colors::RESET)
+ resetColor();
+ else
+ changeColor(C);
+ return *this;
+}
+
+raw_ostream &raw_ostream::write_uuid(const uuid_t UUID) {
+ for (int Idx = 0; Idx < 16; ++Idx) {
+ *this << format("%02" PRIX32, UUID[Idx]);
+ if (Idx == 3 || Idx == 5 || Idx == 7 || Idx == 9)
+ *this << "-";
+ }
+ return *this;
+}
+
+
+raw_ostream &raw_ostream::write_escaped(StringRef Str,
+ bool UseHexEscapes) {
+ for (unsigned char c : Str) {
+ switch (c) {
+ case '\\':
+ *this << '\\' << '\\';
+ break;
+ case '\t':
+ *this << '\\' << 't';
+ break;
+ case '\n':
+ *this << '\\' << 'n';
+ break;
+ case '"':
+ *this << '\\' << '"';
+ break;
+ default:
+ if (isPrint(c)) {
+ *this << c;
+ break;
+ }
+
+ // Write out the escaped representation.
+ if (UseHexEscapes) {
+ *this << '\\' << 'x';
+ *this << hexdigit((c >> 4 & 0xF));
+ *this << hexdigit((c >> 0) & 0xF);
+ } else {
+ // Always use a full 3-character octal escape.
+ *this << '\\';
+ *this << char('0' + ((c >> 6) & 7));
+ *this << char('0' + ((c >> 3) & 7));
+ *this << char('0' + ((c >> 0) & 7));
+ }
+ }
+ }
+
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(const void *P) {
+ llvm::write_hex(*this, (uintptr_t)P, HexPrintStyle::PrefixLower);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(double N) {
+ llvm::write_double(*this, N, FloatStyle::Exponent);
+ return *this;
+}
+
+void raw_ostream::flush_nonempty() {
+ assert(OutBufCur > OutBufStart && "Invalid call to flush_nonempty.");
+ size_t Length = OutBufCur - OutBufStart;
+ OutBufCur = OutBufStart;
+ write_impl(OutBufStart, Length);
+}
+
+raw_ostream &raw_ostream::write(unsigned char C) {
+ // Group exceptional cases into a single branch.
+ if (LLVM_UNLIKELY(OutBufCur >= OutBufEnd)) {
+ if (LLVM_UNLIKELY(!OutBufStart)) {
+ if (BufferMode == BufferKind::Unbuffered) {
+ write_impl(reinterpret_cast<char*>(&C), 1);
+ return *this;
+ }
+ // Set up a buffer and start over.
+ SetBuffered();
+ return write(C);
+ }
+
+ flush_nonempty();
+ }
+
+ *OutBufCur++ = C;
+ return *this;
+}
+
+raw_ostream &raw_ostream::write(const char *Ptr, size_t Size) {
+ // Group exceptional cases into a single branch.
+ if (LLVM_UNLIKELY(size_t(OutBufEnd - OutBufCur) < Size)) {
+ if (LLVM_UNLIKELY(!OutBufStart)) {
+ if (BufferMode == BufferKind::Unbuffered) {
+ write_impl(Ptr, Size);
+ return *this;
+ }
+ // Set up a buffer and start over.
+ SetBuffered();
+ return write(Ptr, Size);
+ }
+
+ size_t NumBytes = OutBufEnd - OutBufCur;
+
+ // If the buffer is empty at this point we have a string that is larger
+ // than the buffer. Directly write the chunk that is a multiple of the
+ // preferred buffer size and put the remainder in the buffer.
+ if (LLVM_UNLIKELY(OutBufCur == OutBufStart)) {
+ assert(NumBytes != 0 && "undefined behavior");
+ size_t BytesToWrite = Size - (Size % NumBytes);
+ write_impl(Ptr, BytesToWrite);
+ size_t BytesRemaining = Size - BytesToWrite;
+ if (BytesRemaining > size_t(OutBufEnd - OutBufCur)) {
+ // Too much left over to copy into our buffer.
+ return write(Ptr + BytesToWrite, BytesRemaining);
+ }
+ copy_to_buffer(Ptr + BytesToWrite, BytesRemaining);
+ return *this;
+ }
+
+ // We don't have enough space in the buffer to fit the string in. Insert as
+ // much as possible, flush and start over with the remainder.
+ copy_to_buffer(Ptr, NumBytes);
+ flush_nonempty();
+ return write(Ptr + NumBytes, Size - NumBytes);
+ }
+
+ copy_to_buffer(Ptr, Size);
+
+ return *this;
+}
+
+void raw_ostream::copy_to_buffer(const char *Ptr, size_t Size) {
+ assert(Size <= size_t(OutBufEnd - OutBufCur) && "Buffer overrun!");
+
+ // Handle short strings specially, memcpy isn't very good at very short
+ // strings.
+ switch (Size) {
+ case 4: OutBufCur[3] = Ptr[3]; LLVM_FALLTHROUGH;
+ case 3: OutBufCur[2] = Ptr[2]; LLVM_FALLTHROUGH;
+ case 2: OutBufCur[1] = Ptr[1]; LLVM_FALLTHROUGH;
+ case 1: OutBufCur[0] = Ptr[0]; LLVM_FALLTHROUGH;
+ case 0: break;
+ default:
+ memcpy(OutBufCur, Ptr, Size);
+ break;
+ }
+
+ OutBufCur += Size;
+}
+
+// Formatted output.
+raw_ostream &raw_ostream::operator<<(const format_object_base &Fmt) {
+ // If we have more than a few bytes left in our output buffer, try
+ // formatting directly onto its end.
+ size_t NextBufferSize = 127;
+ size_t BufferBytesLeft = OutBufEnd - OutBufCur;
+ if (BufferBytesLeft > 3) {
+ size_t BytesUsed = Fmt.print(OutBufCur, BufferBytesLeft);
+
+ // Common case is that we have plenty of space.
+ if (BytesUsed <= BufferBytesLeft) {
+ OutBufCur += BytesUsed;
+ return *this;
+ }
+
+ // Otherwise, we overflowed and the return value tells us the size to try
+ // again with.
+ NextBufferSize = BytesUsed;
+ }
+
+ // If we got here, we didn't have enough space in the output buffer for the
+ // string. Try printing into a SmallVector that is resized to have enough
+ // space. Iterate until we win.
+ SmallVector<char, 128> V;
+
+ while (true) {
+ V.resize(NextBufferSize);
+
+ // Try formatting into the SmallVector.
+ size_t BytesUsed = Fmt.print(V.data(), NextBufferSize);
+
+ // If BytesUsed fit into the vector, we win.
+ if (BytesUsed <= NextBufferSize)
+ return write(V.data(), BytesUsed);
+
+ // Otherwise, try again with a new size.
+ assert(BytesUsed > NextBufferSize && "Didn't grow buffer!?");
+ NextBufferSize = BytesUsed;
+ }
+}
+
+raw_ostream &raw_ostream::operator<<(const formatv_object_base &Obj) {
+ SmallString<128> S;
+ Obj.format(*this);
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(const FormattedString &FS) {
+ if (FS.Str.size() >= FS.Width || FS.Justify == FormattedString::JustifyNone) {
+ this->operator<<(FS.Str);
+ return *this;
+ }
+ const size_t Difference = FS.Width - FS.Str.size();
+ switch (FS.Justify) {
+ case FormattedString::JustifyLeft:
+ this->operator<<(FS.Str);
+ this->indent(Difference);
+ break;
+ case FormattedString::JustifyRight:
+ this->indent(Difference);
+ this->operator<<(FS.Str);
+ break;
+ case FormattedString::JustifyCenter: {
+ int PadAmount = Difference / 2;
+ this->indent(PadAmount);
+ this->operator<<(FS.Str);
+ this->indent(Difference - PadAmount);
+ break;
+ }
+ default:
+ llvm_unreachable("Bad Justification");
+ }
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(const FormattedNumber &FN) {
+ if (FN.Hex) {
+ HexPrintStyle Style;
+ if (FN.Upper && FN.HexPrefix)
+ Style = HexPrintStyle::PrefixUpper;
+ else if (FN.Upper && !FN.HexPrefix)
+ Style = HexPrintStyle::Upper;
+ else if (!FN.Upper && FN.HexPrefix)
+ Style = HexPrintStyle::PrefixLower;
+ else
+ Style = HexPrintStyle::Lower;
+ llvm::write_hex(*this, FN.HexValue, Style, FN.Width);
+ } else {
+ llvm::SmallString<16> Buffer;
+ llvm::raw_svector_ostream Stream(Buffer);
+ llvm::write_integer(Stream, FN.DecValue, 0, IntegerStyle::Integer);
+ if (Buffer.size() < FN.Width)
+ indent(FN.Width - Buffer.size());
+ (*this) << Buffer;
+ }
+ return *this;
+}
+
+raw_ostream &raw_ostream::operator<<(const FormattedBytes &FB) {
+ if (FB.Bytes.empty())
+ return *this;
+
+ size_t LineIndex = 0;
+ auto Bytes = FB.Bytes;
+ const size_t Size = Bytes.size();
+ HexPrintStyle HPS = FB.Upper ? HexPrintStyle::Upper : HexPrintStyle::Lower;
+ uint64_t OffsetWidth = 0;
+ if (FB.FirstByteOffset.hasValue()) {
+ // Figure out how many nibbles are needed to print the largest offset
+ // represented by this data set, so that we can align the offset field
+ // to the right width.
+ size_t Lines = Size / FB.NumPerLine;
+ uint64_t MaxOffset = *FB.FirstByteOffset + Lines * FB.NumPerLine;
+ unsigned Power = 0;
+ if (MaxOffset > 0)
+ Power = llvm::Log2_64_Ceil(MaxOffset);
+ OffsetWidth = std::max<uint64_t>(4, llvm::alignTo(Power, 4) / 4);
+ }
+
+ // The width of a block of data including all spaces for group separators.
+ unsigned NumByteGroups =
+ alignTo(FB.NumPerLine, FB.ByteGroupSize) / FB.ByteGroupSize;
+ unsigned BlockCharWidth = FB.NumPerLine * 2 + NumByteGroups - 1;
+
+ while (!Bytes.empty()) {
+ indent(FB.IndentLevel);
+
+ if (FB.FirstByteOffset.hasValue()) {
+ uint64_t Offset = FB.FirstByteOffset.getValue();
+ llvm::write_hex(*this, Offset + LineIndex, HPS, OffsetWidth);
+ *this << ": ";
+ }
+
+ auto Line = Bytes.take_front(FB.NumPerLine);
+
+ size_t CharsPrinted = 0;
+ // Print the hex bytes for this line in groups
+ for (size_t I = 0; I < Line.size(); ++I, CharsPrinted += 2) {
+ if (I && (I % FB.ByteGroupSize) == 0) {
+ ++CharsPrinted;
+ *this << " ";
+ }
+ llvm::write_hex(*this, Line[I], HPS, 2);
+ }
+
+ if (FB.ASCII) {
+ // Print any spaces needed for any bytes that we didn't print on this
+ // line so that the ASCII bytes are correctly aligned.
+ assert(BlockCharWidth >= CharsPrinted);
+ indent(BlockCharWidth - CharsPrinted + 2);
+ *this << "|";
+
+ // Print the ASCII char values for each byte on this line
+ for (uint8_t Byte : Line) {
+ if (isPrint(Byte))
+ *this << static_cast<char>(Byte);
+ else
+ *this << '.';
+ }
+ *this << '|';
+ }
+
+ Bytes = Bytes.drop_front(Line.size());
+ LineIndex += Line.size();
+ if (LineIndex < Size)
+ *this << '\n';
+ }
+ return *this;
+}
+
+template <char C>
+static raw_ostream &write_padding(raw_ostream &OS, unsigned NumChars) {
+ static const char Chars[] = {C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C,
+ C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C,
+ C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C,
+ C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C,
+ C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C};
+
+ // Usually the indentation is small, handle it with a fastpath.
+ if (NumChars < array_lengthof(Chars))
+ return OS.write(Chars, NumChars);
+
+ while (NumChars) {
+ unsigned NumToWrite = std::min(NumChars,
+ (unsigned)array_lengthof(Chars)-1);
+ OS.write(Chars, NumToWrite);
+ NumChars -= NumToWrite;
+ }
+ return OS;
+}
+
+/// indent - Insert 'NumSpaces' spaces.
+raw_ostream &raw_ostream::indent(unsigned NumSpaces) {
+ return write_padding<' '>(*this, NumSpaces);
+}
+
+/// write_zeros - Insert 'NumZeros' nulls.
+raw_ostream &raw_ostream::write_zeros(unsigned NumZeros) {
+ return write_padding<'\0'>(*this, NumZeros);
+}
+
+void raw_ostream::anchor() {}
+
+//===----------------------------------------------------------------------===//
+// Formatted Output
+//===----------------------------------------------------------------------===//
+
+// Out of line virtual method.
+void format_object_base::home() {
+}
+
+//===----------------------------------------------------------------------===//
+// raw_fd_ostream
+//===----------------------------------------------------------------------===//
+
+static int getFD(StringRef Filename, std::error_code &EC,
+ sys::fs::CreationDisposition Disp, sys::fs::FileAccess Access,
+ sys::fs::OpenFlags Flags) {
+ // XXX BINARYEN - we only ever use IO from LLVM to log to stdout
+ return ::fileno(stdout);
+}
+
+raw_fd_ostream::raw_fd_ostream(StringRef Filename, std::error_code &EC)
+ : raw_fd_ostream(Filename, EC, sys::fs::CD_CreateAlways, sys::fs::FA_Write,
+ sys::fs::OF_None) {}
+
+raw_fd_ostream::raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::CreationDisposition Disp)
+ : raw_fd_ostream(Filename, EC, Disp, sys::fs::FA_Write, sys::fs::OF_None) {}
+
+raw_fd_ostream::raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::FileAccess Access)
+ : raw_fd_ostream(Filename, EC, sys::fs::CD_CreateAlways, Access,
+ sys::fs::OF_None) {}
+
+raw_fd_ostream::raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::OpenFlags Flags)
+ : raw_fd_ostream(Filename, EC, sys::fs::CD_CreateAlways, sys::fs::FA_Write,
+ Flags) {}
+
+raw_fd_ostream::raw_fd_ostream(StringRef Filename, std::error_code &EC,
+ sys::fs::CreationDisposition Disp,
+ sys::fs::FileAccess Access,
+ sys::fs::OpenFlags Flags)
+ : raw_fd_ostream(getFD(Filename, EC, Disp, Access, Flags), true) {}
+
+/// FD is the file descriptor that this writes to. If ShouldClose is true, this
+/// closes the file when the stream is destroyed.
+raw_fd_ostream::raw_fd_ostream(int fd, bool shouldClose, bool unbuffered)
+ : raw_pwrite_stream(unbuffered), FD(fd), ShouldClose(shouldClose) {
+ // XXX BINARYEN: do nothing here
+}
+
+raw_fd_ostream::~raw_fd_ostream() {
+ // XXX BINARYEN: do nothing here
+}
+
+void raw_fd_ostream::write_impl(const char *Ptr, size_t Size) {
+ // XXX BINARYEN: just log it out
+ for (size_t i = 0; i < Size; i++) {
+ std::cout << Ptr[i];
+ }
+}
+
+void raw_fd_ostream::close() {
+ assert(ShouldClose);
+ ShouldClose = false;
+ flush();
+ llvm_unreachable("close"); // XXX BINARYEN
+#if 0
+ if (auto EC = sys::Process::SafelyCloseFileDescriptor(FD))
+ error_detected(EC);
+#endif
+ FD = -1;
+}
+
+uint64_t raw_fd_ostream::seek(uint64_t off) {
+ llvm_unreachable("seek");
+}
+
+void raw_fd_ostream::pwrite_impl(const char *Ptr, size_t Size,
+ uint64_t Offset) {
+ uint64_t Pos = tell();
+ seek(Offset);
+ write(Ptr, Size);
+ seek(Pos);
+}
+
+size_t raw_fd_ostream::preferred_buffer_size() const {
+ return 0; // XXX BINARYEN
+}
+
+raw_ostream &raw_fd_ostream::changeColor(enum Colors colors, bool bold,
+ bool bg) {
+ if (!ColorEnabled)
+ return *this;
+
+ llvm_unreachable("color"); // XXX BINARYEN
+}
+
+raw_ostream &raw_fd_ostream::resetColor() {
+ if (!ColorEnabled)
+ return *this;
+
+ llvm_unreachable("color"); // XXX BINARYEN
+}
+
+raw_ostream &raw_fd_ostream::reverseColor() {
+ if (!ColorEnabled)
+ return *this;
+
+ llvm_unreachable("color"); // XXX BINARYEN
+}
+
+bool raw_fd_ostream::is_displayed() const {
+ llvm_unreachable("is_displayed"); // XXX BINARYEN
+}
+
+bool raw_fd_ostream::has_colors() const {
+ llvm_unreachable("is_displayed"); // XXX BINARYEN
+}
+
+void raw_fd_ostream::anchor() {}
+
+//===----------------------------------------------------------------------===//
+// outs(), errs(), nulls()
+//===----------------------------------------------------------------------===//
+
+/// outs() - This returns a reference to a raw_ostream for standard output.
+/// Use it like: outs() << "foo" << "bar";
+raw_ostream &llvm::outs() {
+ // Set buffer settings to model stdout behavior.
+ std::error_code EC;
+ static raw_fd_ostream S("-", EC, sys::fs::OF_None);
+ assert(!EC);
+ return S;
+}
+
+/// errs() - This returns a reference to a raw_ostream for standard error.
+/// Use it like: errs() << "foo" << "bar";
+raw_ostream &llvm::errs() {
+ // Set standard error to be unbuffered by default.
+ const int fd = 2; // XXX BINARYEN: stderr, but it doesn't matter anyhow
+ static raw_fd_ostream S(fd, false, true);
+ return S;
+}
+
+/// nulls() - This returns a reference to a raw_ostream which discards output.
+raw_ostream &llvm::nulls() {
+ static raw_null_ostream S;
+ return S;
+}
+
+//===----------------------------------------------------------------------===//
+// raw_string_ostream
+//===----------------------------------------------------------------------===//
+
+raw_string_ostream::~raw_string_ostream() {
+ flush();
+}
+
+void raw_string_ostream::write_impl(const char *Ptr, size_t Size) {
+ OS.append(Ptr, Size);
+}
+
+//===----------------------------------------------------------------------===//
+// raw_svector_ostream
+//===----------------------------------------------------------------------===//
+
+uint64_t raw_svector_ostream::current_pos() const { return OS.size(); }
+
+void raw_svector_ostream::write_impl(const char *Ptr, size_t Size) {
+ OS.append(Ptr, Ptr + Size);
+}
+
+void raw_svector_ostream::pwrite_impl(const char *Ptr, size_t Size,
+ uint64_t Offset) {
+ memcpy(OS.data() + Offset, Ptr, Size);
+}
+
+//===----------------------------------------------------------------------===//
+// raw_null_ostream
+//===----------------------------------------------------------------------===//
+
+raw_null_ostream::~raw_null_ostream() {
+#ifndef NDEBUG
+ // ~raw_ostream asserts that the buffer is empty. This isn't necessary
+ // with raw_null_ostream, but it's better to have raw_null_ostream follow
+ // the rules than to change the rules just for raw_null_ostream.
+ flush();
+#endif
+}
+
+void raw_null_ostream::write_impl(const char *Ptr, size_t Size) {
+}
+
+uint64_t raw_null_ostream::current_pos() const {
+ return 0;
+}
+
+void raw_null_ostream::pwrite_impl(const char *Ptr, size_t Size,
+ uint64_t Offset) {}
+
+void raw_pwrite_stream::anchor() {}
+
+void buffer_ostream::anchor() {}
diff --git a/third_party/llvm-project/readme.txt b/third_party/llvm-project/readme.txt
new file mode 100644
index 000000000..a1202b857
--- /dev/null
+++ b/third_party/llvm-project/readme.txt
@@ -0,0 +1,19 @@
+This folder contains files from LLVM. See LICENSE.TXT for their license.
+
+These files were synced from
+
+commit 6c86d6efaf129c42d37121f1e7e9a7adffb54c1a (origin/master, master)
+Author: Craig Topper <craig.topper@intel.com>
+Date: Mon Nov 11 16:30:42 2019 -0800
+
+ [X86] Remove some else branches after checking for !useSoftFloat() that set operations to Expand.
+
+ If we're using soft floats, then these operations shoudl be
+ softened during type legalization. They'll never get to
+ LegalizeVectorOps or LegalizeDAG so they don't need to be
+ Expanded there.
+
+Local changes in those files are marked with
+
+ // XXX BINARYEN
+
--
cgit v1.2.3