[PATCH] D106013: [Verifier] Require same signature for intrinsic calls

Nikita Popov via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Jul 15 14:23:48 PDT 2021


nikic updated this revision to Diff 359134.
nikic edited the summary of this revision.
nikic added a comment.

Restore previous implementation.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D106013/new/

https://reviews.llvm.org/D106013

Files:
  llvm/lib/IR/Verifier.cpp
  llvm/test/Verifier/force-opaque-ptr.ll


Index: llvm/test/Verifier/force-opaque-ptr.ll
===================================================================
--- /dev/null
+++ llvm/test/Verifier/force-opaque-ptr.ll
@@ -0,0 +1,9 @@
+; RUN: not opt -passes=verify -force-opaque-pointers -S < %s 2>&1 | FileCheck %s
+
+declare i32 @llvm.umax.i32(i32, i32)
+
+define void @intrinsic_signature_mismatch() {
+; CHECK: Intrinsic called with incompatible signature
+  call i32 @llvm.umax.i32(i32 0)
+  ret void
+}
Index: llvm/lib/IR/Verifier.cpp
===================================================================
--- llvm/lib/IR/Verifier.cpp
+++ llvm/lib/IR/Verifier.cpp
@@ -3064,11 +3064,12 @@
   Assert(verifyAttributeCount(Attrs, Call.arg_size()),
          "Attribute after last parameter!", Call);
 
-  bool IsIntrinsic = Call.getCalledFunction() &&
-                     Call.getCalledFunction()->isIntrinsic();
-
   Function *Callee =
       dyn_cast<Function>(Call.getCalledOperand()->stripPointerCasts());
+  bool IsIntrinsic = Callee && Callee->isIntrinsic();
+  if (IsIntrinsic)
+    Assert(Callee->getValueType() == FTy,
+           "Intrinsic called with incompatible signature", Call);
 
   if (Attrs.hasFnAttribute(Attribute::Speculatable)) {
     // Don't allow speculatable on call sites, unless the underlying function


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D106013.359134.patch
Type: text/x-patch
Size: 1286 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210715/e79b72fb/attachment.bin>


More information about the llvm-commits mailing list