This is an archive of the discontinued LLVM Phabricator instance.

[libomptarget] [amdgpu] Change default number of teams per computation unit
ClosedPublic

Authored by dhruvachak on Mar 19 2021, 7:00 PM.

Details

Summary

This patch is related to https://reviews.llvm.org/D98832. Based on discussions there, I decided to separate out the teams default as this patch. This change is to increase the number of teams per computation unit so as to provide more wavefronts for hiding latency. This change improves performance for some programs, including 20-50% for some Stream benchmarks.

Diff Detail

Event Timeline

dhruvachak created this revision.Mar 19 2021, 7:00 PM
dhruvachak requested review of this revision.Mar 19 2021, 7:00 PM
Herald added a project: Restricted Project. · View Herald TranscriptMar 19 2021, 7:00 PM
JonChesterfield accepted this revision.Jun 29 2021, 2:04 PM

LG, thanks

This revision is now accepted and ready to land.Jun 29 2021, 2:04 PM
This revision was landed with ongoing or failed builds.Jun 29 2021, 3:35 PM
This revision was automatically updated to reflect the committed changes.