diff options
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2017-03-11 00:29:27 +0000 |
|---|---|---|
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2017-03-11 00:29:27 +0000 |
| commit | 79da2a7698ab2e86d6eef12859d879ccfb98f46e (patch) | |
| tree | 59f6e3745f81cc3221b443b98c5ae1079ba1ad53 /llvm/test/CodeGen/AMDGPU/clamp.ll | |
| parent | ee8a4f51c4f01de2eba8a018539905aa674f8dff (diff) | |
| download | bcm5719-llvm-79da2a7698ab2e86d6eef12859d879ccfb98f46e.tar.gz bcm5719-llvm-79da2a7698ab2e86d6eef12859d879ccfb98f46e.zip | |
[AMDGPU] Remove getBidirectionalReasonRank
This method inverts the Reason field of a scheduling candidate.
It does right comparison between RegCritical and RegExcess, but
everything else is broken. In fact it can prefer less strong reason
such as Weak over RegCritical because Weak > -RegCritical.
The CandReason enum is properly sorted, so just remove artificial
ranking.
Differential Revision: https://reviews.llvm.org/D30557
llvm-svn: 297536
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/clamp.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/clamp.ll | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/clamp.ll b/llvm/test/CodeGen/AMDGPU/clamp.ll index 6a78290f9a8..9735c7074be 100644 --- a/llvm/test/CodeGen/AMDGPU/clamp.ll +++ b/llvm/test/CodeGen/AMDGPU/clamp.ll @@ -51,8 +51,8 @@ define amdgpu_kernel void @v_clamp_negabs_f32(float addrspace(1)* %out, float ad } ; GCN-LABEL: {{^}}v_clamp_negzero_f32: -; GCN: {{buffer|flat}}_load_dword [[A:v[0-9]+]] -; GCN: v_bfrev_b32_e32 [[SIGNBIT:v[0-9]+]], 1 +; GCN-DAG: {{buffer|flat}}_load_dword [[A:v[0-9]+]] +; GCN-DAG: v_bfrev_b32_e32 [[SIGNBIT:v[0-9]+]], 1 ; GCN: v_med3_f32 v{{[0-9]+}}, [[A]], [[SIGNBIT]], 1.0 define amdgpu_kernel void @v_clamp_negzero_f32(float addrspace(1)* %out, float addrspace(1)* %aptr) #0 { %tid = call i32 @llvm.amdgcn.workitem.id.x() |
