This is an archive of the discontinued LLVM Phabricator instance.

[SCEV][NFC] Win some compile time from mass forgetMemoizedResults
ClosedPublic

Authored by mkazantsev on Oct 22 2021, 12:20 AM.

Details

Summary

Mass forgetMemoizedResults can be done more efficiently than bunch
of individual invocations of helper because we can traverse maps being
updated just once, rather than doing this for each invidivual SCEV.

Should be NFC and supposedly improves compile time.

Diff Detail

Event Timeline

mkazantsev created this revision.Oct 22 2021, 12:20 AM
mkazantsev requested review of this revision.Oct 22 2021, 12:20 AM
Herald added a project: Restricted Project. · View Herald TranscriptOct 22 2021, 12:20 AM
reames accepted this revision.Oct 22 2021, 10:10 AM

LGTM

llvm/lib/Analysis/ScalarEvolution.cpp
12786

Personally, I'd inline this into the single caller as having it out of line doesn't seem to add any value, but I don't hugely care either way.

This revision is now accepted and ready to land.Oct 22 2021, 10:10 AM
This revision was landed with ongoing or failed builds.Oct 25 2021, 12:20 AM
This revision was automatically updated to reflect the committed changes.