This is an archive of the discontinued LLVM Phabricator instance.

AMDGPU: Allow clustering flat memory operations
ClosedPublic

Authored by arsenm on May 23 2016, 8:46 PM.

Details

Reviewers
tstellarAMD

Diff Detail

Event Timeline

arsenm updated this revision to Diff 58184.May 23 2016, 8:46 PM
arsenm retitled this revision from to AMDGPU: Allow clustering flat memory operations.
arsenm updated this object.
arsenm added a reviewer: tstellarAMD.
arsenm added a subscriber: llvm-commits.

Are there tests for this?

Are there tests for this?

I don't think this does much because there is no offset. It seems like a flaw in the scheduler/this hook that it will only schedule based on the immediate offset in the instruction and not based on the offset in the mem operand

tstellarAMD accepted this revision.Feb 1 2017, 10:59 AM
This revision is now accepted and ready to land.Feb 1 2017, 10:59 AM
arsenm closed this revision.Feb 1 2017, 12:34 PM

r293809