I think this manages to not break the DAG handling with the divergent
predicates because the stadalone divergent patterns end up with a
higher priority than the pattern on the instruction definition.
The 16-bit versions don't work yet.
Paths
| Differential D64725
AMDGPU/GlobalISel: Select G_SHL ClosedPublic Authored by arsenm on Jul 15 2019, 5:54 AM.
Details
Diff Detail Event TimelineHerald added subscribers: Petar.Avramovic, t-tye, tpr and 6 others. · View Herald TranscriptJul 15 2019, 5:54 AM Comment Actions This looks OK to me, but I'm not as familiar with how this will impact SDAG. Might want to have someone else take a look too. This revision is now accepted and ready to land.Jul 16 2019, 2:42 AM
Revision Contents
Diff 209821 lib/Target/AMDGPU/SOPInstructions.td
lib/Target/AMDGPU/VOP2Instructions.td
lib/Target/AMDGPU/VOP3Instructions.td
test/CodeGen/AMDGPU/GlobalISel/inst-select-shl.mir
|