This is an archive of the discontinued LLVM Phabricator instance.

AVX-512: Fix for PR28175 - Scalar code optimization.
ClosedPublic

Authored by delena on Sep 12 2016, 6:53 AM.

Details

Summary

Optimized (truncate (assertzext x) to i1) and anyext i1 to i8/16/32.
Optimization of this patterns is a one more step towards i1 optimization on AVX-512.

Diff Detail

Repository
rL LLVM

Event Timeline

delena updated this revision to Diff 71002.Sep 12 2016, 6:53 AM
delena retitled this revision from to AVX-512: Fix for PR28175 - Scalar code optimization..
delena updated this object.
delena added a reviewer: igorb.
delena set the repository for this revision to rL LLVM.
delena added a subscriber: llvm-commits.
igorb added inline comments.Sep 12 2016, 7:48 AM
../test/CodeGen/X86/avx512-fsel.ll
1

could you please commit the test before the change

delena updated this revision to Diff 71042.Sep 12 2016, 12:27 PM
igorb accepted this revision.Sep 13 2016, 12:33 AM
igorb edited edge metadata.

LGTM

This revision is now accepted and ready to land.Sep 13 2016, 12:33 AM
This revision was automatically updated to reflect the committed changes.