This is an archive of the discontinued LLVM Phabricator instance.

[PowerPC][AIX][RFC] Generate inlined quadword lock free atomic operations
AbandonedPublic

Authored by lkail on Jun 1 2021, 3:41 AM.

Details

Reviewers
nemanjai
jsji
shchenz
jfb
xingxue
Group Reviewers
Restricted Project
Summary

Generate inlined quadword lock free atomic operations for PPC64 on AIX. It keeps compatible with current i128 implementation for PPC64. Need more input for if this patch is at a proper direction.

Diff Detail

Event Timeline

lkail created this revision.Jun 1 2021, 3:41 AM
lkail requested review of this revision.Jun 1 2021, 3:41 AM
Herald added a project: Restricted Project. · View Herald TranscriptJun 1 2021, 3:41 AM
jsji added a reviewer: xingxue.Jun 1 2021, 6:55 AM
jsji added a comment.Jun 17 2021, 2:42 PM

I would prefer we implementing it using AtomicExpand , not using AtomicExpansionKind::LLSC , but with AtomicExpansionKind::MaskedIntrinsic.

lkail abandoned this revision.Jun 22 2021, 10:46 PM

All related work moved to https://reviews.llvm.org/D103614