projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f848798) | patch
[LoopCacheAnalysis][NFC] Add a test case for improved loop cache analysis cost calcul...
authorCongzhe Cao <congzhe.cao@huawei.com>
Wed, 4 May 2022 21:09:13 +0000 (17:09 -0400)
committerCongzheUalberta <congzhecao@gmail.com>
Wed, 4 May 2022 21:13:10 +0000 (17:13 -0400)
commit5e004fb787698440a387750db7f8028e7cb14cfc
tree0921c46071c5693e1abda6187265102225a8e911tree | snapshot
parentf848798b7d3f59cf1f4de4187618eaad10b0ae86commit | diff
[LoopCacheAnalysis][NFC] Add a test case for improved loop cache analysis cost calculation

Added a motivating test case for D123400 where the loopnest has a
suboptimal loop order j-i-k. After D123400 we ensure that the order
of loop cache analysis output is loop i-j-k, despite the suboptimal
order in the original loopnest.

Reviewed By: bmahjour, #loopoptwg

Differential Revision: https://reviews.llvm.org/D122776
llvm/test/Analysis/LoopCacheAnalysis/PowerPC/single-store.ll diff | blob | history
Domain: System / Toolchain;