AMDGPU: Use inbounds when calculating workitem offset

When promoting allocas to LDS, we know we are indexing
into a specific area just created, and the calculation
will also never overflow.

Also emit some of the muls as nsw nuw, because instcombine
infers this already from the range metadata. I think
putting this on the other adds and muls might be OK too,
but I'm not 100% sure.

llvm-svn: 259545

1 files changed, 7 insertions, 6 deletions

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp b/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
index 16676eceeee..cc09ced8f6e 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
@@ -485,17 +485,18 @@ void AMDGPUPromoteAlloca::visitAlloca(AllocaInst &I) {
   Value *TIdY = getWorkitemID(Builder, 1);
   Value *TIdZ = getWorkitemID(Builder, 2);
 
-  Value *Tmp0 = Builder.CreateMul(TCntY, TCntZ);
+  Value *Tmp0 = Builder.CreateMul(TCntY, TCntZ, "", true, true);
   Tmp0 = Builder.CreateMul(Tmp0, TIdX);
-  Value *Tmp1 = Builder.CreateMul(TIdY, TCntZ);
+  Value *Tmp1 = Builder.CreateMul(TIdY, TCntZ, "", true, true);
   Value *TID = Builder.CreateAdd(Tmp0, Tmp1);
   TID = Builder.CreateAdd(TID, TIdZ);
 
-  std::vector<Value*> Indices;
-  Indices.push_back(Constant::getNullValue(Type::getInt32Ty(Mod->getContext())));
-  Indices.push_back(TID);
+  Value *Indices[] = {
+    Constant::getNullValue(Type::getInt32Ty(Mod->getContext())),
+    TID
+  };
 
-  Value *Offset = Builder.CreateGEP(GVTy, GV, Indices);
+  Value *Offset = Builder.CreateInBoundsGEP(GVTy, GV, Indices);
   I.mutateType(Offset->getType());
   I.replaceAllUsesWith(Offset);
   I.eraseFromParent();