This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU] Make more use of madmk/fmamk instructions
ClosedPublic

Authored by foad on Mar 1 2022, 9:16 AM.

Details

Summary

In convertToThreeAddress handle VOP2 mac/fmac instructions with a
literal src0 operand, since these are prime candidates for
converting to madmk/fmamk.

Previously this would only happen if src0 (or src1) was a register
defined by a move-immediate instruction, but in many cases these
operands have already been folded because SIFoldOperands runs
before TwoAddressInstructionPass.

Diff Detail

Event Timeline

foad created this revision.Mar 1 2022, 9:16 AM
foad requested review of this revision.Mar 1 2022, 9:16 AM
Herald added a project: Restricted Project. · View Herald TranscriptMar 1 2022, 9:16 AM
arsenm accepted this revision.Mar 1 2022, 9:30 AM
This revision is now accepted and ready to land.Mar 1 2022, 9:30 AM
This revision was landed with ongoing or failed builds.Mar 2 2022, 2:33 AM
This revision was automatically updated to reflect the committed changes.
Herald added a project: Restricted Project. · View Herald TranscriptMar 2 2022, 2:33 AM