Index: lib/Transforms/Utils/MemorySSA.cpp =================================================================== --- lib/Transforms/Utils/MemorySSA.cpp +++ lib/Transforms/Utils/MemorySSA.cpp @@ -1333,7 +1333,10 @@ VersionStack.pop_back(); ++PopEpoch; } - + // We have to track the size after we first popped, because we may + // see later pushes before we see uses, and we need to be able to + // tell if any LowerBounds pointed into old parts of the stack. + unsigned SizeAfterPop = VersionStack.size(); for (MemoryAccess &MA : *Accesses) { auto *MU = dyn_cast(&MA); if (!MU) { @@ -1356,7 +1359,7 @@ LocInfo.PopEpoch = PopEpoch; LocInfo.StackEpoch = StackEpoch; // If the lower bound was in the info we popped, we have to reset it. - if (LocInfo.LowerBound >= VersionStack.size()) { + if (LocInfo.LowerBound >= SizeAfterPop) { // Reset the lower bound of things to check. // TODO: Some day we should be able to reset to last kill, rather than // 0. Index: test/Transforms/Util/MemorySSA/pr28880.ll =================================================================== --- /dev/null +++ test/Transforms/Util/MemorySSA/pr28880.ll @@ -0,0 +1,54 @@ +; RUN: opt -basicaa -print-memoryssa -verify-memoryssa -analyze < %s 2>&1 | FileCheck %s +; RUN: opt -aa-pipeline=basic-aa -passes='print,verify' -disable-output < %s 2>&1 | FileCheck %s + +; This testcase is reduced from SingleSource/Benchmarks/Misc/fbench.c +; It is testing to make sure that the MemorySSA use optimizer +; comes up with right answers when dealing with multiple MemoryLocations +; over different blocks. See PR28880 for more details. +@global = external hidden unnamed_addr global double, align 8 +@global.1 = external hidden unnamed_addr global double, align 8 + +; Function Attrs: nounwind ssp uwtable +define hidden fastcc void @hoge() unnamed_addr #0 { +bb: + br i1 undef, label %bb1, label %bb2 + +bb1: ; preds = %bb +; These accesses should not conflict. +; CHECK: 1 = MemoryDef(liveOnEntry) +; 1 = MemoryDef(liveOnEntry) +; CHECK-NEXT: store double undef, double* @global, align 8 + store double undef, double* @global, align 8 +; CHECK: MemoryUse(liveOnEntry) +; MemoryUse(liveOnEntry) +; CHECK-NEXT: %tmp = load double, double* @global.1, align 8 + %tmp = load double, double* @global.1, align 8 + unreachable + +bb2: ; preds = %bb + br label %bb3 + +bb3: ; preds = %bb2 + br i1 undef, label %bb4, label %bb6 + +bb4: ; preds = %bb3 +; These accesses should conflict. +; CHECK: 2 = MemoryDef(liveOnEntry) +; 2 = MemoryDef(liveOnEntry) +; CHECK-NEXT: store double 0.000000e+00, double* @global.1, align 8 + store double 0.000000e+00, double* @global.1, align 8 +; CHECK: MemoryUse(2) +; MemoryUse(2) +; CHECK-NEXT: %tmp5 = load double, double* @global.1, align 8 + %tmp5 = load double, double* @global.1, align 8 + unreachable + +bb6: ; preds = %bb3 + unreachable +} + +attributes #0 = { nounwind ssp uwtable "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="core2" "target-features"="+cx16,+fxsr,+mmx,+sse,+sse2,+sse3,+ssse3" "unsafe-fp-math"="false" "use-soft-float"="false" } + +!llvm.ident = !{!0} + +!0 = !{!"Apple LLVM version 8.0.0 (clang-800.0.35)"}