diff options
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/load-local-f32.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/load-local-f32.ll | 11 |
1 files changed, 8 insertions, 3 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/load-local-f32.ll b/llvm/test/CodeGen/AMDGPU/load-local-f32.ll index 3d8720888c7..c2722717fb3 100644 --- a/llvm/test/CodeGen/AMDGPU/load-local-f32.ll +++ b/llvm/test/CodeGen/AMDGPU/load-local-f32.ll @@ -2,7 +2,7 @@ ; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SICIVI,FUNC %s ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefixes=EG,FUNC %s -; Testing for ds_read_128 +; Testing for ds_read/write_128 ; RUN: llc -march=amdgcn -mcpu=tahiti -amdgpu-ds128 < %s | FileCheck -check-prefixes=SI,FUNC %s ; RUN: llc -march=amdgcn -mcpu=tonga -amdgpu-ds128 < %s | FileCheck -check-prefixes=CIVI,FUNC %s ; RUN: llc -march=amdgcn -mcpu=gfx900 -amdgpu-ds128 < %s | FileCheck -check-prefixes=CIVI,FUNC %s @@ -127,17 +127,22 @@ entry: ret void } -; Tests if ds_read_b128 gets generated for the 16 byte aligned load. +; Tests if ds_read/write_b128 gets generated for the 16 byte aligned load. ; FUNC-LABEL: {{^}}local_v4f32_to_128: + ; SI-NOT: ds_read_b128 +; SI-NOT: ds_write_b128 + ; CIVI: ds_read_b128 +; CIVI: ds_write_b128 + ; EG: LDS_READ_RET ; EG: LDS_READ_RET ; EG: LDS_READ_RET ; EG: LDS_READ_RET define amdgpu_kernel void @local_v4f32_to_128(<4 x float> addrspace(3)* %out, <4 x float> addrspace(3)* %in) { %ld = load <4 x float>, <4 x float> addrspace(3)* %in, align 16 - store <4 x float> %ld, <4 x float> addrspace(3)* %out + store <4 x float> %ld, <4 x float> addrspace(3)* %out, align 16 ret void } |
