diff options
| -rw-r--r-- | llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp | 2 | ||||
| -rw-r--r-- | llvm/test/Transforms/MemCpyOpt/profitable-memset.ll | 20 |
2 files changed, 21 insertions, 1 deletions
diff --git a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp index a51204ab2e3..d8f6c2a36cf 100644 --- a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp +++ b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp @@ -185,7 +185,7 @@ bool MemsetRange::isProfitableToUseMemset(const DataLayout &DL) const { // size. If so, check to see whether we will end up actually reducing the // number of stores used. unsigned Bytes = unsigned(End-Start); - unsigned MaxIntSize = DL.getLargestLegalIntTypeSize(); + unsigned MaxIntSize = DL.getLargestLegalIntTypeSize() / 8; if (MaxIntSize == 0) MaxIntSize = 1; unsigned NumPointerStores = Bytes / MaxIntSize; diff --git a/llvm/test/Transforms/MemCpyOpt/profitable-memset.ll b/llvm/test/Transforms/MemCpyOpt/profitable-memset.ll new file mode 100644 index 00000000000..bbef1244ee3 --- /dev/null +++ b/llvm/test/Transforms/MemCpyOpt/profitable-memset.ll @@ -0,0 +1,20 @@ +; RUN: opt < %s -memcpyopt -S | FileCheck %s + +target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128" + +; CHECK-LABEL: @foo( +; CHECK-NOT: store +; CHECK: call void @llvm.memset.p0i8.i64(i8* %2, i8 0, i64 8, i32 2, i1 false) + +define void @foo(i64* nocapture %P) { +entry: + %0 = bitcast i64* %P to i16* + %arrayidx = getelementptr inbounds i16, i16* %0, i64 1 + %1 = bitcast i16* %arrayidx to i32* + %arrayidx1 = getelementptr inbounds i16, i16* %0, i64 3 + store i16 0, i16* %0, align 2 + store i32 0, i32* %1, align 4 + store i16 0, i16* %arrayidx1, align 2 + ret void +} + |
