From eacb1020aabac23b4a833f4c9fcde289937a7ba7 Mon Sep 17 00:00:00 2001 From: Farhana Aleen Date: Mon, 28 May 2018 18:15:11 +0000 Subject: [AMDGPU] Re-enabled 128bit wide-vector generation for local addr space by default. Summary: Bug reported here https://bugs.freedesktop.org/show_bug.cgi?id=105464 found to be resolved by some other fixes. Author: FarhanaAleen llvm-svn: 333380 --- llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll | 3 +-- llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'llvm/test/Transforms/LoadStoreVectorizer/AMDGPU') diff --git a/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll b/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll index 19fc44bb6c8..5eb3b25c1dc 100644 --- a/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll +++ b/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll @@ -504,8 +504,7 @@ define amdgpu_kernel void @merge_local_store_2_constants_i32_align_2(i32 addrspa } ; CHECK-LABEL: @merge_local_store_4_constants_i32 -; CHECK: store <2 x i32> , <2 x i32> addrspace(3)* -; CHECK: store <2 x i32> , <2 x i32> addrspace(3)* +; CHECK: store <4 x i32> , <4 x i32> addrspace(3)* define amdgpu_kernel void @merge_local_store_4_constants_i32(i32 addrspace(3)* %out) #0 { %out.gep.1 = getelementptr i32, i32 addrspace(3)* %out, i32 1 %out.gep.2 = getelementptr i32, i32 addrspace(3)* %out, i32 2 diff --git a/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll b/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll index 8a78f3d7e9b..b684ca8c12c 100644 --- a/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll +++ b/llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll @@ -29,11 +29,10 @@ define amdgpu_kernel void @no_crash(i32 %arg) { ; longest chain vectorized ; CHECK-LABEL: @interleave_get_longest -; CHECK: load <2 x i32> +; CHECK: load <4 x i32> ; CHECK: load i32 ; CHECK: store <2 x i32> zeroinitializer ; CHECK: load i32 -; CHECK: load <2 x i32> ; CHECK: load i32 ; CHECK: load i32 -- cgit v1.2.3