summaryrefslogtreecommitdiff
path: root/test/dump/loop.txt
Commit message (Collapse)AuthorAgeFilesLines
* Use stderr for all logging and error output (#1579)Sam Clegg2020-12-031-1/+3
| | | | I'm not sure why we were using stdout but the convention is normally to write all logging and error message to stderr.
* Update output from type to func type in binary writer and reader (#1570)Radu M2020-11-161-1/+1
| | | Signed-off-by: Radu M <root@radu.sh>
* Pass function body size in BinaryReader callback (#975)Ben Smith2018-12-121-1/+1
| | | This is useful for finding large functions with wasm-objdump.
* Remove `run-objdump.py`; replace with `RUN` cmds (#732)Ben Smith2018-01-241-1/+1
| | | | | | | | | | | | | | | | * Remove `run-objdump.py`; replace with `RUN` cmds This CL introduces two new TOOLs for tests: * `run-objdump-gen-wasm`: Run objdump on wasm files generated by `gen-wasm.py`. * `run-objdump-spec`: Run objdump on wasm files generated by `wast2json`. For now, this requires each test to specify the files manually: `%(temp_file)s.0.wasm %(temp_file)s.1.wasm` etc. `gen_wasm_py` has been added as a new RUN variable as well. * Call FixPythonExecutable after expanding variables
* Rename run args (#728)Ben Smith2018-01-211-1/+1
| | | | As suggested in https://github.com/WebAssembly/wabt/pull/725#discussion_r162516714.
* Rename wasmdump to wasm-objdump (#455)Sam Clegg2017-05-251-1/+1
| | | | This is more consistent with how llvm and binutils name such tools.
* Regenerate expected test output (#423)Sam Clegg2017-05-101-0/+1
| | | | | The expected test output was out of sync becasue whitespace at the start and end of stdout doesn't cause test failures (because stdout/stderr are stripped before comparison).
* Use 8 hex digits for wasm version rather than 6. (#391)pipcet2017-04-051-1/+1
|
* Add callback for end opcode and end of function (#358)Sam Clegg2017-03-161-0/+1
| | | | | Without this objdump doesn't show the final byte/opcode of each function.
* Make most enumerations into enum classes (#329)Ben Smith2017-03-011-3/+3
|
* Update wasm binary format label to 1 (#310)Derek Schuff2017-02-241-2/+2
| | | | | | | | | | | | * Update wasm binary format label to 1 Rebaseline tests, but no other changes. * Revert test/spec/globals.txt * Update testsuite * Update spec binary.wast test and testsuite repo
* Add function names and relocations to wasmdump disassembly (#285)Sam Clegg2017-01-231-1/+2
|
* wasmdump improvments (#192)Sam Clegg2016-10-311-0/+3
| | | | | | | | * wasmdump improvments - Split into three different passes: headers, details, and raw - Nest imports correctly - Show element count as part of section header rather than on its own line
* Fix types for binary 0xdBen Smith2016-10-261-4/+4
| | | | | | | * All types are unified (i32,i64,f32,f64,func,anyfunc,void) * Can't use array to get type name; use `wasm_get_type_name` instead * Encode types as vs7 (i.e. Signed LEB128 with max length 1 byte) * Change version to 0xd
* Remap the operator encoding for binary 0xdBen Smith2016-10-261-9/+9
|
* Update tests syntax to new flat syntax (#174)Kwadwo 'Que' Amankwa2016-10-201-3/+4
| | | | converted another batch
* Add wasmdump binary for inspecting wasm files (#161)Sam Clegg2016-10-121-4/+7
| | | | | | * Add wasmdump binary for inspecting wasm files Run this tool to as part of the 'dump' tests rather than using the -d flag to wast2wasm.
* Remove trailing whitespace from hexdumps (#154)Sam Clegg2016-10-071-29/+29
|
* Refactor all uses of s_opcode_ to use functionsBen Smith2016-10-061-5/+5
|
* append END opcode at end of functionBen Smith2016-09-301-3/+4
| | | | This fixes #132.
* Update 'dump' tests for binary_0xcSam Clegg2016-09-291-32/+29
|
* fix dump tests, update binary version to 0xcBen Smith2016-09-291-5/+5
|
* update dump testsBen Smith2016-04-281-36/+34
|
* remove end sectionBen Smith2016-03-111-6/+1
|
* sections have size and name as stringBen Smith2016-03-091-19/+39
|
* update v8, use function sig and body sectionsBen Smith2016-03-091-13/+14
| | | | v8 also has support for the names section, but I'll add that later.
* rebase master; remove need for local decl fixupBen Smith2016-03-071-1/+0
|
* Rebase the dumps on local decls changes.Ben L. Titzer2016-03-071-6/+8
|
* binary format has magic and version numberBen Smith2016-02-281-16/+19
|
* move shared binary format stuff to wasm-binary.hBen Smith2016-02-241-3/+3
|
* fix --dump-module flagBen Smith2015-12-111-0/+2
| | | | I broke this when I rewrote everything.
* don't write func name offset if func is not exportedBen Smith2015-12-031-9/+8
| | | | | This was originally done because it was difficult to do otherwise with the old binary generator.
* don't write memory section if not specifiedBen Smith2015-12-031-20/+16
| | | | | I originally did this to match the old binary generator, but it's not necessary.
* finished code for dump, doesn't actually write yetBen Smith2015-12-031-6/+2
|
* remove implicit branch to the top of loopBen Smith2015-11-051-7/+4
|
* fix gen after v8-native change to make stmt=exprBen Smith2015-11-051-5/+8
| | | | | | | | | | | The big changes: - there are no more statement ops, everything is an expression op - some of the opcodes are shuffled around - loops no longer implicitly branch to the top - loops have two break scopes, one is the continue scope, the deeper one is the exit scope. - the return statement was removed, so to return you need to break to a top-level block
* dedupe function signaturesBen Smith2015-10-231-1/+1
| | | | | | There is now one signature table, and all functions, function types, and imports reference it by index. This is in preparation for properly supporting call_indirect.
* update v8-native-prototype w/ new binary formatBen Smith2015-10-211-28/+25
| | | | | | | | | | | | | | | The new binary format has optional sections, which are marked with a byte. The names and segment data must come after all the sections. The other major change to the format is that function bodies are now inlined with the function header. This makes it much easier to append the assert_* functions to the module after the fact; we can place the function section last, and write the assertion functions directly into the buffer. The only tricky part is that the function count is now stored as leb128, which means that we may need to shift the function section down when the function count increases.
* change prefix for test file formatBen Smith2015-10-131-2/+3
| | | | | | | | | | | | | Use the comment characters for the s-expression format so the test files can easily be loaded into sexpr-wasm directly: ;;; KEY: VALUE (;; STDOUT ;;; some standard output ;;; STDOUT ;;)
* remove toplevel blockBen Smith2015-09-251-11/+8
| | | | | I thought this was needed, but it doesn't seem to be anymore (or maybe it never was).
* write correct expression count for toplevel blockBen Smith2015-09-101-1/+2
|
* wrap all functions in an OPCODE_BLOCKBen Smith2015-09-071-8/+10
| | | | The v8-native-prototype doesn't work properly without it.
* add section comments to verbose outputBen Smith2015-09-071-0/+3
|
* use memory max size in outputBen Smith2015-09-071-2/+2
|
* convenience function for dumping opcodeBen Smith2015-09-051-3/+3
|
* verbose dump outputBen Smith2015-09-051-1/+24
|
* write opcodes for block, loop, constBen Smith2015-09-051-0/+10
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-14 09:05:52 +0000