diff options
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2018-02-24 01:32:32 +0000 |
|---|---|---|
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2018-02-24 01:32:32 +0000 |
| commit | fa48c496e2ccde292b74b5dc45aa29c65e3748be (patch) | |
| tree | 821f35f9135999480b15bcfb994d1aaf021c9369 /llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll | |
| parent | c22867f5fc960970d413a15cfe47a2e7379c6e29 (diff) | |
| download | bcm5719-llvm-fa48c496e2ccde292b74b5dc45aa29c65e3748be.tar.gz bcm5719-llvm-fa48c496e2ccde292b74b5dc45aa29c65e3748be.zip | |
[AMDGPU] Shrinking V_SUBBREV_U32
V_SUBBREV_U32 is a commute opcode for V_SUBB_U32. However, when
we try to commute V_SUBB_U32 in order to shrink it we do not then
process V_SUBBREV_U32 and it stay VOP3. This is fixed.
Differential Revision: https://reviews.llvm.org/D43699
llvm-svn: 326011
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll b/llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll index 9e47c7d3449..f94b2fa3aa3 100644 --- a/llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll +++ b/llvm/test/CodeGen/AMDGPU/combine-cond-add-sub.ll @@ -19,8 +19,8 @@ bb: } ; GCN-LABEL: {{^}}sub1: -; GCN: v_cmp_gt_u32_e{{32|64}} [[CC:[^,]+]], v{{[0-9]+}}, v{{[0-9]+}} -; GCN: v_subb_u32_e{{32|64}} v{{[0-9]+}}, {{[^,]+}}, v{{[0-9]+}}, 0, [[CC]] +; GCN: v_cmp_gt_u32_e32 vcc, v{{[0-9]+}}, v{{[0-9]+}} +; GCN: v_subbrev_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc ; GCN-NOT: v_cndmask define amdgpu_kernel void @sub1(i32 addrspace(1)* nocapture %arg) { @@ -134,8 +134,8 @@ bb: } ; GCN-LABEL: {{^}}sext_flclass: -; GCN: v_cmp_class_f32_e{{32|64}} [[CC:[^,]+]], -; GCN: v_subb_u32_e{{32|64}} v{{[0-9]+}}, {{[^,]+}}, v{{[0-9]+}}, 0, [[CC]] +; GCN: v_cmp_class_f32_e32 vcc, +; GCN: v_subbrev_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc ; GCN-NOT: v_cndmask define amdgpu_kernel void @sext_flclass(i32 addrspace(1)* nocapture %arg, float %x) { |
