[PostEmscripten] Fix calcSegmentOffsets for large offsets (WebAssembl…

…y#6260) Specifically offsets larger than 2^32 which were being interpreted misinterpreted here as very large int64_t values.
dotnet · Jul 12, 2024 · fea72ee · fea72ee
1 parent f1ead41
commit fea72ee
Show file tree

Hide file tree

Showing 2 changed files with 35 additions and 14 deletions.
diff --git a/src/passes/PostEmscripten.cpp b/src/passes/PostEmscripten.cpp
@@ -95,12 +95,11 @@ static void calcSegmentOffsets(Module& wasm,
             return;
           }
         }
-        auto it = offsets.find(curr->segment);
-        if (it != offsets.end()) {
+        if (offsets.find(curr->segment) != offsets.end()) {
           Fatal() << "Cannot get offset of passive segment initialized "
                      "multiple times";
         }
-        offsets[curr->segment] = dest->value.getInteger();
+        offsets[curr->segment] = dest->value.getUnsigned();
       }
     } searcher(passiveOffsets);
     searcher.walkModule(&wasm);
@@ -317,7 +316,7 @@ struct PostEmscripten : public Pass {
         // The first operand is the function pointer index, which must be
         // constant if we are to optimize it statically.
         if (auto* index = curr->operands[0]->dynCast<Const>()) {
-          size_t indexValue = index->value.getInteger();
+          size_t indexValue = index->value.getUnsigned();
           if (indexValue >= flatTable.names.size()) {
             // UB can lead to indirect calls to invalid pointers.
             return;

diff --git a/test/lit/passes/post-emscripten.wast b/test/lit/passes/post-emscripten.wast
@@ -1,18 +1,27 @@
 ;; NOTE: Assertions have been generated by update_lit_checks.py --all-items and should not be edited.
-;; RUN: wasm-opt %s --post-emscripten -S -o - | filecheck %s
+;; RUN: wasm-opt %s --enable-bulk-memory --post-emscripten -S -o - | filecheck %s
 
 ;; Checks that the start/stop exports are removed and that the data they
 ;; refer to is either zero'd out, or the segment emptied.
 
+;; Two of the segments used here are active, so their offsets are trivial
+;; to dirive.  One segment is passive and its offset is derived from the
+;; memory.init instruction.
+
+;; Explictly use a data address that is larger then INT_MAX to verify
+;; that these offset are correctly interpreted as unsigned.
+
 (module
+ ;; CHECK:      (type $0 (func))
+
  ;; CHECK:      (global $em_asm_start i32 (i32.const 1000))
  (global $em_asm_start i32 (i32.const 1000))
  ;; CHECK:      (global $em_asm_stop i32 (i32.const 1011))
  (global $em_asm_stop i32 (i32.const 1011))
- ;; CHECK:      (global $em_js_start i32 (i32.const 2006))
- (global $em_js_start i32 (i32.const 2006))
- ;; CHECK:      (global $em_js_stop i32 (i32.const 2015))
- (global $em_js_stop i32 (i32.const 2015))
+ ;; CHECK:      (global $em_js_start i32 (i32.const -1294967290))
+ (global $em_js_start i32 (i32.const 3000000006))
+ ;; CHECK:      (global $em_js_stop i32 (i32.const -1294967281))
+ (global $em_js_stop i32 (i32.const 3000000015))
  ;; CHECK:      (global $em_lib_deps_start i32 (i32.const 3000))
  (global $em_lib_deps_start i32 (i32.const 3000))
  ;; CHECK:      (global $em_lib_deps_stop i32 (i32.const 3009))
@@ -21,13 +30,12 @@
  (global $foo_start i32 (i32.const 4000))
  ;; CHECK:      (global $foo_stop i32 (i32.const 4015))
  (global $foo_stop i32 (i32.const 4015))
- (memory 10 10)
- ;; CHECK:      (memory $0 10 10)
-
+ ;; CHECK:      (memory $mem 10 10)
+ (memory $mem 10 10)
  ;; CHECK:      (data $data1 (i32.const 1000) "")
  (data $data1 (i32.const 1000) "hello world")
- ;; CHECK:      (data $data2 (i32.const 2000) "hello \00\00\00\00\00\00\00\00\00 world")
- (data $data2 (i32.const 2000) "hello DELETE ME world")
+ ;; CHECK:      (data $data2 "hello \00\00\00\00\00\00\00\00\00 world")
+ (data $data2 "hello DELETE ME world")
  ;; CHECK:      (data $data3 (i32.const 3000) "")
  (data $data3 (i32.const 3000) "some deps")
  (export "__start_em_asm" (global $em_asm_start))
@@ -40,4 +48,18 @@
  (export "__start_foo" (global $foo_start))
  ;; CHECK:      (export "__stop_foo" (global $foo_stop))
  (export "__stop_foo" (global $foo_stop))
+ ;; CHECK:      (func $meminit
+ ;; CHECK-NEXT:  (memory.init $data2
+ ;; CHECK-NEXT:   (i32.const -1294967296)
+ ;; CHECK-NEXT:   (i32.const 0)
+ ;; CHECK-NEXT:   (i32.const 21)
+ ;; CHECK-NEXT:  )
+ ;; CHECK-NEXT: )
+ (func $meminit
+  (memory.init $mem $data2
+   (i32.const 3000000000)
+   (i32.const 0)
+   (i32.const 21)
+  )
+ )
 )