This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU] move hasAGPRs/hasVGPRs into header
ClosedPublic

Authored by rampitec on Sep 29 2021, 12:52 PM.

Details

Summary

It is now very simple and can go right into the header
allowing optimizer to combine callers, such as isVGPRClass
and similar.

It does not need anything from the TRI itself anymore, so
make it static class member along with the callers.

Diff Detail

Event Timeline

rampitec created this revision.Sep 29 2021, 12:52 PM
rampitec requested review of this revision.Sep 29 2021, 12:52 PM
Herald added a project: Restricted Project. · View Herald TranscriptSep 29 2021, 12:52 PM
Herald added a subscriber: wdng. · View Herald Transcript
This revision is now accepted and ready to land.Sep 29 2021, 7:30 PM

For the record, I have measured check-llvm-codegen-amdgpu with a release build, it seems to speedup BE compilation around 5% in average.

This revision was landed with ongoing or failed builds.Sep 30 2021, 10:34 AM
This revision was automatically updated to reflect the committed changes.
foad added a subscriber: piotr.Oct 1 2021, 12:28 AM

For the record, I have measured check-llvm-codegen-amdgpu with a release build, it seems to speedup BE compilation around 5% in average.

That is amazing!

+@piotr for awareness.

For the record, I have measured check-llvm-codegen-amdgpu with a release build, it seems to speedup BE compilation around 5% in average.

That is amazing!

+@piotr for awareness.

Again for the record, compiler was built with the clang.