Skip to content

[MemCpyOpt] Fix infinite loop in memset+memcpy fold #98638

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 15, 2024
Merged

[MemCpyOpt] Fix infinite loop in memset+memcpy fold #98638

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1247,6 +1247,15 @@ bool MemCpyOptPass::processMemSetMemCpyDependence(MemCpyInst *MemCpy,
if (!BAA.isMustAlias(MemSet->getDest(), MemCpy->getDest()))
return false;

// Don't perform the transform if src_size may be zero. In that case, the
// transform is essentially a complex no-op and may lead to an infinite
// loop if BasicAA is smart enough to understand that dst and dst + src_size
// are still MustAlias after the transform.
Value *SrcSize = MemCpy->getLength();
if (!isKnownNonZero(SrcSize,
SimplifyQuery(MemCpy->getDataLayout(), DT, AC, MemCpy)))
return false;

// Check that src and dst of the memcpy aren't the same. While memcpy
// operands cannot partially overlap, exact equality is allowed.
if (isModSet(BAA.getModRefInfo(MemCpy, MemoryLocation::getForSource(MemCpy))))
Expand All @@ -1263,7 +1272,6 @@ bool MemCpyOptPass::processMemSetMemCpyDependence(MemCpyInst *MemCpy,
// Use the same i8* dest as the memcpy, killing the memset dest if different.
Value *Dest = MemCpy->getRawDest();
Value *DestSize = MemSet->getLength();
Value *SrcSize = MemCpy->getLength();

if (mayBeVisibleThroughUnwinding(Dest, MemSet, MemCpy))
return false;
Expand Down Expand Up @@ -1677,8 +1685,7 @@ bool MemCpyOptPass::processMemCpy(MemCpyInst *M, BasicBlock::iterator &BBI) {
return true;
}

// If the size is zero, remove the memcpy. This also prevents infinite loops
// in processMemSetMemCpyDependence, which is a no-op for zero-length memcpys.
// If the size is zero, remove the memcpy.
if (isZeroSize(M->getLength())) {
++BBI;
eraseInstruction(M);
Expand Down
15 changes: 15 additions & 0 deletions llvm/test/Transforms/MemCpyOpt/memcpy-zero-size.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,3 +34,18 @@ define void @pr64886(i64 %len, ptr noalias %p) {
call void @llvm.memcpy.p0.p0.i64(ptr inttoptr (i64 -1 to ptr), ptr %p, i64 poison, i1 false)
ret void
}

define void @pr98610(ptr %p, ptr noalias %p2) {
; CHECK-LABEL: @pr98610(
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr [[P:%.*]], i8 0, i64 1, i1 false)
; CHECK-NEXT: [[ZERO_EXT:%.*]] = zext i32 0 to i64
; CHECK-NEXT: [[MUL:%.*]] = mul i64 [[ZERO_EXT]], 1
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[P]], ptr [[P2:%.*]], i64 [[MUL]], i1 false)
; CHECK-NEXT: ret void
;
call void @llvm.memset.p0.i64(ptr %p, i8 0, i64 1, i1 false)
%zero.ext = zext i32 0 to i64
%mul = mul i64 %zero.ext, 1
call void @llvm.memcpy.p0.p0.i64(ptr %p, ptr %p2, i64 %mul, i1 false)
ret void
}
13 changes: 9 additions & 4 deletions llvm/test/Transforms/MemCpyOpt/memset-memcpy-dbgloc.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,15 +8,20 @@ declare void @llvm.memset.p0.i64(ptr nocapture, i8, i64, i1)
declare void @llvm.memcpy.p0.p0.i64(ptr nocapture, ptr nocapture readonly, i64, i1)

define void @test_constant(i64 %src_size, ptr %dst, i64 %dst_size, i8 %c) !dbg !5 {
; CHECK-LABEL: @test_constant(
; CHECK-NEXT: [[TMP1:%.*]] = icmp ule i64 [[DST_SIZE:%.*]], [[SRC_SIZE:%.*]], !dbg [[DBG11:![0-9]+]]
; CHECK-LABEL: define void @test_constant(
; CHECK-SAME: i64 [[SRC_SIZE:%.*]], ptr [[DST:%.*]], i64 [[DST_SIZE:%.*]], i8 [[C:%.*]]) !dbg [[DBG5:![0-9]+]] {
; CHECK-NEXT: [[NON_ZERO:%.*]] = icmp ne i64 [[SRC_SIZE]], 0
; CHECK-NEXT: call void @llvm.assume(i1 [[NON_ZERO]])
; CHECK-NEXT: [[TMP1:%.*]] = icmp ule i64 [[DST_SIZE]], [[SRC_SIZE]], !dbg [[DBG11:![0-9]+]]
; CHECK-NEXT: [[TMP2:%.*]] = sub i64 [[DST_SIZE]], [[SRC_SIZE]], !dbg [[DBG11]]
; CHECK-NEXT: [[TMP3:%.*]] = select i1 [[TMP1]], i64 0, i64 [[TMP2]], !dbg [[DBG11]]
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i8, ptr [[DST:%.*]], i64 [[SRC_SIZE]], !dbg [[DBG11]]
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 1 [[TMP4]], i8 [[C:%.*]], i64 [[TMP3]], i1 false), !dbg [[DBG11]]
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i8, ptr [[DST]], i64 [[SRC_SIZE]], !dbg [[DBG11]]
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 1 [[TMP4]], i8 [[C]], i64 [[TMP3]], i1 false), !dbg [[DBG11]]
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[DST]], ptr @C, i64 [[SRC_SIZE]], i1 false), !dbg [[DBG12:![0-9]+]]
; CHECK-NEXT: ret void, !dbg [[DBG13:![0-9]+]]
;
%non.zero = icmp ne i64 %src_size, 0
call void @llvm.assume(i1 %non.zero)
call void @llvm.memset.p0.i64(ptr %dst, i8 %c, i64 %dst_size, i1 false), !dbg !11
call void @llvm.memcpy.p0.p0.i64(ptr %dst, ptr @C, i64 %src_size, i1 false), !dbg !12
ret void, !dbg !13
Expand Down
Loading
Loading