[PATCH] D65010: [AMDGPU] Fixed occupancy calculation for gfx10
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Jul 19 13:29:40 PDT 2019
arsenm accepted this revision.
arsenm added inline comments.
This revision is now accepted and ready to land.
================
Comment at: lib/Target/AMDGPU/AMDGPUSubtarget.cpp:595
+ unsigned MaxWaves = getMaxWavesPerEU();
+ unsigned Granula = getVGPRAllocGranule();
+ if (VGPRs < Granula)
----------------
s/Granula/Granule
================
Comment at: lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp:258-260
+ if (!isGFX10(*STI))
+ return 10;
+ return 20;
----------------
return isGFX10 ? 20 : 10
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D65010/new/
https://reviews.llvm.org/D65010
More information about the llvm-commits
mailing list