[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

fhahn added a comment. Thanks! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122126/new/ https://reviews.llvm.org/D122126 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG3ed9f603fd59: [LoopVectorize] Dont interleave when the number of runtime checks exceeds the… (authored by TiehuZhang, committed by mdchen).

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

TiehuZhang added a comment. In D122126#3515070 , @fhahn wrote: > Still LGTM, thanks! The remaining suggestion can be addressed directly before > committing the patch. Thanks, @fhahn! I'll add the precommit test when committing the patch Repository:

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

fhahn accepted this revision. fhahn added a comment. Still LGTM, thanks! The remaining suggestion can be addressed directly before committing the patch. Comment at: llvm/lib/Transforms/Vectorize/LoopVectorizationPlanner.h:324 + /// Check if the number of runtime checks

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

TiehuZhang updated this revision to Diff 429110. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122126/new/ https://reviews.llvm.org/D122126 Files: llvm/lib/Transforms/Vectorize/LoopVectorizationPlanner.h

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

fhahn accepted this revision. fhahn added a comment. LGTM with additional suggestions inline, thanks! Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:10462 +} else { + ORE->emit([&]() { +return OptimizationRemarkAnalysisAliasing( I

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

TiehuZhang added a comment. The code has been updated since accept. Please review it again. Thank you very much! @fhahn @dmgreen Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:10460 IC = CM.selectInterleaveCount(VF.Width, *VF.Cost.getValue()); +if

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

TiehuZhang updated this revision to Diff 428930. TiehuZhang added a comment. (Updated) Difference with accepted version: Move memory runtime checks to processLoop to control both VF and IC Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122126/new/

[PATCH] D122126: [LoopVectorize] Don't interleave when the number of runtime checks exceeds the threshold

TiehuZhang updated this revision to Diff 427300. TiehuZhang added a comment. Herald added a project: clang. Herald added a subscriber: cfe-commits. Fix the failed case (optimization-remark-options.c), because the remark info should be updated Repository: rG LLVM Github Monorepo CHANGES