[Strings] Adopt new instruction binary encoding (#5714)

See WebAssembly/stringref#46.

This format is already adopted by V8: https://chromium-review.googlesource.com/c/v8/v8/+/3892695.
The text format is left unchanged (see #5607 for a discussion on the subject).

I have also added support for string.encode_lossy_utf8 and
string.encode_lossy_utf8 array (by allowing the replace policy for
Binaryen's string.encode_wtf8 instruction).

4 files changed, 125 insertions, 6 deletions

diff --git a/test/example/c-api-kitchen-sink.c b/test/example/c-api-kitchen-sink.c
index 2c6ec2d6c..cb26c372d 100644
--- a/test/example/c-api-kitchen-sink.c
+++ b/test/example/c-api-kitchen-sink.c
@@ -1174,7 +1174,7 @@ void test_core() {
                       0,
                       false),
     BinaryenStringNew(module,
-                      BinaryenStringNewReplace(),
+                      BinaryenStringNewLossyUTF8(),
                       makeInt32(module, 0),
                       makeInt32(module, 0),
                       0,
@@ -1209,7 +1209,7 @@ void test_core() {
                       makeInt32(module, 0),
                       false),
     BinaryenStringNew(module,
-                      BinaryenStringNewReplaceArray(),
+                      BinaryenStringNewLossyUTF8Array(),
                       BinaryenGlobalGet(module, "i8Array-global", i8Array),
                       0,
                       makeInt32(module, 0),
@@ -1261,6 +1261,12 @@ void test_core() {
       0),
     BinaryenStringEncode(
       module,
+      BinaryenStringEncodeLossyUTF8(),
+      BinaryenGlobalGet(module, "string-global", BinaryenTypeStringref()),
+      makeInt32(module, 0),
+      0),
+    BinaryenStringEncode(
+      module,
       BinaryenStringEncodeWTF8(),
       BinaryenGlobalGet(module, "string-global", BinaryenTypeStringref()),
       makeInt32(module, 0),
@@ -1279,6 +1285,12 @@ void test_core() {
       makeInt32(module, 0)),
     BinaryenStringEncode(
       module,
+      BinaryenStringEncodeLossyUTF8Array(),
+      BinaryenGlobalGet(module, "string-global", BinaryenTypeStringref()),
+      BinaryenGlobalGet(module, "i8Array-global", i8Array),
+      makeInt32(module, 0)),
+    BinaryenStringEncode(
+      module,
       BinaryenStringEncodeWTF8Array(),
       BinaryenGlobalGet(module, "string-global", BinaryenTypeStringref()),
       BinaryenGlobalGet(module, "i8Array-global", i8Array),
diff --git a/test/example/c-api-kitchen-sink.txt b/test/example/c-api-kitchen-sink.txt
index f9b688eaf..946245631 100644
--- a/test/example/c-api-kitchen-sink.txt
+++ b/test/example/c-api-kitchen-sink.txt
@@ -2434,6 +2434,12 @@ BinaryenFeatureAll: 126975
        )
       )
       (drop
+       (string.encode_wtf8 replace
+        (global.get $string-global)
+        (i32.const 0)
+       )
+      )
+      (drop
        (string.encode_wtf8 wtf8
         (global.get $string-global)
         (i32.const 0)
@@ -2453,6 +2459,13 @@ BinaryenFeatureAll: 126975
        )
       )
       (drop
+       (string.encode_wtf8_array replace
+        (global.get $string-global)
+        (global.get $i8Array-global)
+        (i32.const 0)
+       )
+      )
+      (drop
        (string.encode_wtf8_array wtf8
         (global.get $string-global)
         (global.get $i8Array-global)
diff --git a/test/lit/passes/simplify-locals-strings.wast b/test/lit/passes/simplify-locals-strings.wast
index 176d6fbb8..646eee6bd 100644
--- a/test/lit/passes/simplify-locals-strings.wast
+++ b/test/lit/passes/simplify-locals-strings.wast
@@ -320,7 +320,7 @@
   ;; CHECK-NEXT:   )
   ;; CHECK-NEXT:  )
   ;; CHECK-NEXT:  (drop
-  ;; CHECK-NEXT:   (string.encode_wtf16
+  ;; CHECK-NEXT:   (string.encode_wtf8 replace
   ;; CHECK-NEXT:    (local.get $ref)
   ;; CHECK-NEXT:    (i32.const 30)
   ;; CHECK-NEXT:   )
@@ -328,6 +328,20 @@
   ;; CHECK-NEXT:  (drop
   ;; CHECK-NEXT:   (local.get $temp)
   ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (local.set $temp
+  ;; CHECK-NEXT:   (i32.load
+  ;; CHECK-NEXT:    (i32.const 1)
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (string.encode_wtf16
+  ;; CHECK-NEXT:    (local.get $ref)
+  ;; CHECK-NEXT:    (i32.const 40)
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (local.get $temp)
+  ;; CHECK-NEXT:  )
   ;; CHECK-NEXT: )
   (func $no-load-past-encode (param $ref stringref)
     (local $temp i32)
@@ -366,7 +380,7 @@
       )
     )
     (drop
-      (string.encode_wtf16
+      (string.encode_wtf8 replace
         (local.get $ref)
         (i32.const 30)
       )
@@ -374,6 +388,20 @@
     (drop
       (local.get $temp)
     )
+    (local.set $temp
+      (i32.load
+        (i32.const 1)
+      )
+    )
+    (drop
+      (string.encode_wtf16
+        (local.get $ref)
+        (i32.const 40)
+      )
+    )
+    (drop
+      (local.get $temp)
+    )
   )
 
   ;; CHECK:      (func $no-load-past-encode-gc (type $stringref_ref|$array|_ref|$array16|_=>_none) (param $ref stringref) (param $array (ref $array)) (param $array16 (ref $array16))
@@ -417,10 +445,26 @@
   ;; CHECK-NEXT:   )
   ;; CHECK-NEXT:  )
   ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (string.encode_wtf8_array replace
+  ;; CHECK-NEXT:    (local.get $ref)
+  ;; CHECK-NEXT:    (local.get $array)
+  ;; CHECK-NEXT:    (i32.const 30)
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (local.get $temp)
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (local.set $temp
+  ;; CHECK-NEXT:   (array.get_u $array
+  ;; CHECK-NEXT:    (local.get $array)
+  ;; CHECK-NEXT:    (i32.const 0)
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
   ;; CHECK-NEXT:   (string.encode_wtf16_array
   ;; CHECK-NEXT:    (local.get $ref)
   ;; CHECK-NEXT:    (local.get $array16)
-  ;; CHECK-NEXT:    (i32.const 30)
+  ;; CHECK-NEXT:    (i32.const 40)
   ;; CHECK-NEXT:   )
   ;; CHECK-NEXT:  )
   ;; CHECK-NEXT:  (drop
@@ -470,10 +514,26 @@
       )
     )
     (drop
+      (string.encode_wtf8_array replace
+        (local.get $ref)
+        (local.get $array)
+        (i32.const 30)
+      )
+    )
+    (drop
+      (local.get $temp)
+    )
+    (local.set $temp
+      (array.get $array
+        (local.get $array)
+        (i32.const 0)
+      )
+    )
+    (drop
       (string.encode_wtf16_array
         (local.get $ref)
         (local.get $array16)
-        (i32.const 30)
+        (i32.const 40)
       )
     )
     (drop
diff --git a/test/lit/strings.wast b/test/lit/strings.wast
index 1e38ff894..dcb1bffd2 100644
--- a/test/lit/strings.wast
+++ b/test/lit/strings.wast
@@ -184,6 +184,14 @@
   ;; CHECK-NEXT:   )
   ;; CHECK-NEXT:  )
   ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (i32.eqz
+  ;; CHECK-NEXT:    (string.encode_wtf8 replace
+  ;; CHECK-NEXT:     (local.get $ref)
+  ;; CHECK-NEXT:     (i32.const 10)
+  ;; CHECK-NEXT:    )
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
   ;; CHECK-NEXT:   (string.encode_wtf8 utf8
   ;; CHECK-NEXT:    (local.get $ref)
   ;; CHECK-NEXT:    (i32.const 20)
@@ -206,6 +214,14 @@
       )
     )
     (drop
+      (i32.eqz ;; validate the output is i32
+        (string.encode_wtf8 replace
+          (local.get $ref)
+          (i32.const 10)
+        )
+      )
+    )
+    (drop
       (string.encode_wtf8 utf8
         (local.get $ref)
         (i32.const 20)
@@ -546,6 +562,15 @@
   ;; CHECK-NEXT:   )
   ;; CHECK-NEXT:  )
   ;; CHECK-NEXT:  (drop
+  ;; CHECK-NEXT:   (i32.eqz
+  ;; CHECK-NEXT:    (string.encode_wtf8_array replace
+  ;; CHECK-NEXT:     (local.get $ref)
+  ;; CHECK-NEXT:     (local.get $array)
+  ;; CHECK-NEXT:     (i32.const 10)
+  ;; CHECK-NEXT:    )
+  ;; CHECK-NEXT:   )
+  ;; CHECK-NEXT:  )
+  ;; CHECK-NEXT:  (drop
   ;; CHECK-NEXT:   (string.encode_wtf8_array utf8
   ;; CHECK-NEXT:    (local.get $ref)
   ;; CHECK-NEXT:    (local.get $array)
@@ -571,6 +596,15 @@
       )
     )
     (drop
+      (i32.eqz ;; validate the output is i32
+        (string.encode_wtf8_array replace
+          (local.get $ref)
+          (local.get $array)
+          (i32.const 10)
+        )
+      )
+    )
+    (drop
       (string.encode_wtf8_array utf8
         (local.get $ref)
         (local.get $array)