[MemCpy] Check for alias in performMemCpyToMemSetOptzn, instead of the identity of...

Summary:
This fixes pr29105. The reason is that lifetime marks creates new
aliasing pointers the original ones, but before this patch aliases
were not checked in performMemCpyToMemSetOptzn.

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D23846

llvm-svn: 279769

diff --git a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
index e77a893..27c37cd 100644 (file)
--- a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
+++ b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
@@ -1108,8 +1108,10 @@ bool MemCpyOptPass::processMemSetMemCpyDependence(MemCpyInst *MemCpy,
 /// The \p MemCpy must have a Constant length.
 bool MemCpyOptPass::performMemCpyToMemSetOptzn(MemCpyInst *MemCpy,
                                                MemSetInst *MemSet) {
+  AliasAnalysis &AA = LookupAliasAnalysis();
+
   // This only makes sense on memcpy(..., memset(...), ...).
-  if (MemSet->getRawDest() != MemCpy->getRawSource())
+  if (!AA.isMustAlias(MemSet->getRawDest(), MemCpy->getRawSource()))
     return false;
 
   ConstantInt *CopySize = cast<ConstantInt>(MemCpy->getLength());

diff --git a/llvm/test/Transforms/MemCpyOpt/pr29105.ll b/llvm/test/Transforms/MemCpyOpt/pr29105.ll
new file mode 100644 (file)
index 0000000..ad9f97e
--- /dev/null
+++ b/llvm/test/Transforms/MemCpyOpt/pr29105.ll
@@ -0,0 +1,38 @@
+; RUN: opt -memcpyopt -instcombine -S %s | FileCheck %s
+%Foo = type { [2048 x i64] }
+
+; CHECK-LABEL: @baz(
+; CHECK-NOT: call void @llvm.memcpy
+define void @baz() unnamed_addr #0 {
+entry-block:
+  %x.sroa.0 = alloca [2048 x i64], align 8
+  %tmp0 = alloca [2048 x i64], align 8
+  %0 = bitcast [2048 x i64]* %tmp0 to i8*
+  %tmp2 = alloca %Foo, align 8
+  %x.sroa.0.0..sroa_cast6 = bitcast [2048 x i64]* %x.sroa.0 to i8*
+  call void @llvm.lifetime.start(i64 16384, i8* %x.sroa.0.0..sroa_cast6)
+  call void @llvm.lifetime.start(i64 16384, i8* %0)
+  call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 16384, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %x.sroa.0.0..sroa_cast6, i8* %0, i64 16384, i32 8, i1 false)
+  call void @llvm.lifetime.end(i64 16384, i8* %0)
+  %1 = bitcast %Foo* %tmp2 to i8*
+  call void @llvm.lifetime.start(i64 16384, i8* %1)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %x.sroa.0.0..sroa_cast6, i64 16384, i32 8, i1 false)
+  call void @bar(%Foo* noalias nocapture nonnull dereferenceable(16384) %tmp2)
+  call void @llvm.lifetime.end(i64 16384, i8* %1)
+  call void @llvm.lifetime.end(i64 16384, i8* %x.sroa.0.0..sroa_cast6)
+  ret void
+}
+
+declare void @llvm.lifetime.start(i64, i8* nocapture) #1
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1) #1
+
+declare void @llvm.lifetime.end(i64, i8* nocapture) #1
+
+declare void @bar(%Foo* noalias nocapture readonly dereferenceable(16384)) unnamed_addr #0
+
+declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i32, i1) #1
+
+attributes #0 = { uwtable }
+attributes #1 = { argmemonly nounwind }