This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU] Refactor MUBUF/MTBUF TD instruction definitions
ClosedPublic

Authored by vpykhtin on Sep 7 2016, 7:21 AM.

Details

Summary

Done the same way as previous.

It turns out target "real" definitions can be shortened using NAME value inside multiclass, see atomic reals.

Lit tests passed.

Diff Detail

Repository
rL LLVM

Event Timeline

vpykhtin updated this revision to Diff 70533.Sep 7 2016, 7:21 AM
vpykhtin retitled this revision from to [AMDGPU] Refactor MUFUB/MTBUD TD instruction definitions.
vpykhtin updated this object.
vpykhtin set the repository for this revision to rL LLVM.
vpykhtin added a project: Restricted Project.
vpykhtin added a subscriber: nhaustov.
vpykhtin retitled this revision from [AMDGPU] Refactor MUFUB/MTBUD TD instruction definitions to [AMDGPU] Refactor MUBUF/MTBUF TD instruction definitions.Sep 7 2016, 7:26 AM
artem.tamazov accepted this revision.Sep 8 2016, 3:37 AM
artem.tamazov edited edge metadata.

LGTM.

This revision is now accepted and ready to land.Sep 8 2016, 3:37 AM
vpykhtin updated this revision to Diff 70859.Sep 9 2016, 10:17 AM
vpykhtin edited edge metadata.
vpykhtin removed rL LLVM as the repository for this revision.

rebase to the latest tip.

tstellarAMD accepted this revision.Sep 9 2016, 1:33 PM
tstellarAMD edited edge metadata.

OCL/OGL tests pass. LGTM.

This revision was automatically updated to reflect the committed changes.