Page MenuHomePhabricator

[AArch64][GlobalISel] Enable memcpy family combines on minsize functions

Authored by paquette on May 10 2021, 2:12 PM.



The combines in tryCombineMemCpyFamily have heuristics (e.g. TLI.getMaxStoresPerMemset) which consider size. So, theoretically, enabling these combines on minsize functions shouldn't be harmful.

With this enabled we save 0.9% geomean on CTMark at -Oz, and 5.1% on Bullet. There are no code size regressions.

Diff Detail

Event Timeline

paquette created this revision.May 10 2021, 2:12 PM
paquette requested review of this revision.May 10 2021, 2:12 PM
Herald added a project: Restricted Project. · View Herald TranscriptMay 10 2021, 2:12 PM
aemerson accepted this revision.May 10 2021, 3:04 PM

Wow, not running this at all for -Oz? LGTM.

This revision is now accepted and ready to land.May 10 2021, 3:04 PM
This revision was landed with ongoing or failed builds.May 10 2021, 3:25 PM
This revision was automatically updated to reflect the committed changes.