diff --git a/llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp b/llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp --- a/llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp +++ b/llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp @@ -203,9 +203,14 @@ NewTy, New, Idx, New->getName() + ".sub"); IC.InsertNewInstBefore(GEP, *It); + // Gracefully handle allocas in other address spaces. + Value *NewV = GEP; + if (AI.getType()->getPointerAddressSpace() != GEP->getType()->getPointerAddressSpace()) + NewV = IC.Builder.CreateAddrSpaceCast(GEP, AI.getType()); + // Now make everything use the getelementptr instead of the original // allocation. - return IC.replaceInstUsesWith(AI, GEP); + return IC.replaceInstUsesWith(AI, NewV); } } diff --git a/llvm/test/Transforms/InstCombine/alloca-in-non-alloca-as.ll b/llvm/test/Transforms/InstCombine/alloca-in-non-alloca-as.ll new file mode 100644 --- /dev/null +++ b/llvm/test/Transforms/InstCombine/alloca-in-non-alloca-as.ll @@ -0,0 +1,19 @@ +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; RUN: opt < %s -instcombine -S | FileCheck %s + +; Gracefully handle the alloca that is not in the alloca AS (=5) + +target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-G1-ni:7" +target triple = "amdgcn-amd-amdhsa" + +define weak amdgpu_kernel void @__omp_offloading_802_ea0109_main_l8(i32* %a) { +; CHECK-LABEL: @__omp_offloading_802_ea0109_main_l8( +; CHECK-NEXT: .master: +; CHECK-NEXT: ret void +; +.master: + %0 = alloca i8, i64 8, align 1 + %a_on_stack = bitcast i8* %0 to i32** + store i32* undef, i32** %a_on_stack, align 8 + ret void +}