This is an archive of the discontinued LLVM Phabricator instance.

AMDGPU/SI: Implement a work-around for smrd corrupting vccz bit
ClosedPublic

Authored by tstellarAMD on Jan 29 2016, 7:34 AM.

Details

Summary

We will hit this once we have enabled uniform branches. The
smrd-vccz-bug.ll test will be added with the uniform branch commit.

Diff Detail

Event Timeline

tstellarAMD retitled this revision from to AMDGPU/SI: Implement a work-around for smrd corrupting vccz bit.
tstellarAMD updated this object.
tstellarAMD added reviewers: arsenm, mareko.
tstellarAMD added a subscriber: llvm-commits.
arsenm added inline comments.Jan 29 2016, 8:35 PM
lib/Target/AMDGPU/SIInsertWaits.cpp
523

Do we need to worry about modifications to vcc_lo/vcc_hi?

536

Typo heere

tstellarAMD marked an inline comment as done.Feb 5 2016, 11:31 AM
tstellarAMD added inline comments.
lib/Target/AMDGPU/SIInsertWaits.cpp
523

modifiesRegister() considers sub-regs too.

arsenm accepted this revision.Feb 5 2016, 11:38 AM
arsenm edited edge metadata.

LGTM

This revision is now accepted and ready to land.Feb 5 2016, 11:38 AM
This revision was automatically updated to reflect the committed changes.