Add a new LLVM switch -profile-sample-block-accurate to trust zero block counts for branches. Currently we leave out such zero counts when annotating branch weight metadata, which would lead to weights being considered as unknown.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
Comment Actions
This turns out to improve HHVM (https://github.com/facebook/hhvm) RPS by 0.1%, while reducing .text size by 4%, using AutoFDO. I hope to get it in first so we can do more evaluation. Eventually either add a new driver switch just like fprofile-sample-accurate, or merge it into fprofile-sample-accurate if it turns out to be helpful in general.