diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2015-08-29 07:16:50 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2015-08-29 07:16:50 +0000 |
| commit | e4d0c142e8e9cf8b456c06f5ce9641d89b67ac86 (patch) | |
| tree | 5476ba4a50871c8be1f624710b2c042bfe6fc32f /llvm/test/CodeGen/AMDGPU/sub.ll | |
| parent | 9a32cd3d3b79f94aa7e9d016a5e6eeb2650dd8a8 (diff) | |
| download | bcm5719-llvm-e4d0c142e8e9cf8b456c06f5ce9641d89b67ac86.tar.gz bcm5719-llvm-e4d0c142e8e9cf8b456c06f5ce9641d89b67ac86.zip | |
AMDGPU: Add sdst operand to VOP2b instructions
The VOP3 encoding of these allows any SGPR pair for the i1
output, but this was forced before to always use vcc.
This doesn't yet try to use this, but does add the operand
to the definitions so the main change is adding vcc to the
output of the VOP2 encoding.
llvm-svn: 246358
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/sub.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/sub.ll | 14 |
1 files changed, 7 insertions, 7 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/sub.ll b/llvm/test/CodeGen/AMDGPU/sub.ll index b7fba0efa5b..9f9446a4e60 100644 --- a/llvm/test/CodeGen/AMDGPU/sub.ll +++ b/llvm/test/CodeGen/AMDGPU/sub.ll @@ -7,7 +7,7 @@ declare i32 @llvm.r600.read.tidig.x() readnone ; FUNC-LABEL: {{^}}test_sub_i32: ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; SI: v_subrev_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} +; SI: v_subrev_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} define void @test_sub_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) { %b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1 %a = load i32, i32 addrspace(1)* %in @@ -22,8 +22,8 @@ define void @test_sub_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) { ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} define void @test_sub_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) { %b_ptr = getelementptr <2 x i32>, <2 x i32> addrspace(1)* %in, i32 1 @@ -40,10 +40,10 @@ define void @test_sub_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1) ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} -; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} +; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}} define void @test_sub_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) { %b_ptr = getelementptr <4 x i32>, <4 x i32> addrspace(1)* %in, i32 1 |
