This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU] Don't shrink VOP3 instructions pre-RA on GFX10+
ClosedPublic

Authored by foad on Sep 13 2022, 5:22 AM.

Details

Reviewers
Joe_Nash
Group Reviewers
Restricted Project
Commits
rG2e8863b6a11f: [AMDGPU] Don't shrink VOP3 instructions pre-RA on GFX10+
Summary

In GFX10, there is no advantage to shrinking these instructions pre-RA,
so this just saves a bit of work.

In GFX11 there is an advantage to *not* shrinking them pre-RA, because
the register classes for 16-bit operands are less restrictive in the
VOP3 form than in the shrunk form. This patch is a prerequisite for
actually setting up those register classes correctly for 16-bit vs
non-16-bit operands.

Diff Detail

Event Timeline

foad created this revision.Sep 13 2022, 5:22 AM
Herald added a project: Restricted Project. · View Herald TranscriptSep 13 2022, 5:22 AM
foad requested review of this revision.Sep 13 2022, 5:22 AM
Herald added a project: Restricted Project. · View Herald TranscriptSep 13 2022, 5:22 AM

This is an alternative to the SIShrinkInstructions changes in D133723.

Joe_Nash accepted this revision.Sep 13 2022, 10:31 AM
Joe_Nash added a subscriber: Joe_Nash.

LGTM.

This revision is now accepted and ready to land.Sep 13 2022, 10:31 AM
This revision was landed with ongoing or failed builds.Sep 13 2022, 12:30 PM
This revision was automatically updated to reflect the committed changes.