diff options
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2018-11-19 19:58:13 +0000 |
|---|---|---|
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2018-11-19 19:58:13 +0000 |
| commit | 8bafbae889638bc4bc2c82fa56787d254de2d088 (patch) | |
| tree | 174bfbc1c0adc8814e193e756b6f28022a2de1db /llvm/lib/Target/AMDGPU | |
| parent | d677c3fc6199aa375808a6250517b118ba66b40a (diff) | |
| download | bcm5719-llvm-8bafbae889638bc4bc2c82fa56787d254de2d088.tar.gz bcm5719-llvm-8bafbae889638bc4bc2c82fa56787d254de2d088.zip | |
[AMDGPU] Restored selection of scalar_to_vector (v2x16)
This works if DAG combiner is enabled, but without combining
we cannot select scalar_to_vector of <2 x half> and <2 x i16>.
Differential Revision: https://reviews.llvm.org/D54718
llvm-svn: 347259
Diffstat (limited to 'llvm/lib/Target/AMDGPU')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/SIInstructions.td | 18 |
1 files changed, 9 insertions, 9 deletions
diff --git a/llvm/lib/Target/AMDGPU/SIInstructions.td b/llvm/lib/Target/AMDGPU/SIInstructions.td index 9b0edf4c4e0..5f78a107275 100644 --- a/llvm/lib/Target/AMDGPU/SIInstructions.td +++ b/llvm/lib/Target/AMDGPU/SIInstructions.td @@ -1537,15 +1537,15 @@ def : GCNPat < } // End SubtargetPredicate = HasVOP3PInsts -// def : GCNPat < -// (v2f16 (scalar_to_vector f16:$src0)), -// (COPY $src0) -// >; - -// def : GCNPat < -// (v2i16 (scalar_to_vector i16:$src0)), -// (COPY $src0) -// >; +def : GCNPat < + (v2f16 (scalar_to_vector f16:$src0)), + (COPY $src0) +>; + +def : GCNPat < + (v2i16 (scalar_to_vector i16:$src0)), + (COPY $src0) +>; def : GCNPat < (v4i16 (scalar_to_vector i16:$src0)), |
