[PATCH] D59149: [LV] move useEmulatedMaskMemRefHack() functionality to TTI.

[Hideki Saito via Phabricator via llvm-commits]

hsaito created this revision.
hsaito added reviewers: rengolin, markus, craig.topper, hfinkel.
Herald added subscribers: llvm-commits, jdoerfert, rkruppe, hiraditya.
Herald added a project: LLVM.

This is a long overdue response to the review comments in https://reviews.llvm.org/D43208#inline-382175.

D43208 <https://reviews.llvm.org/D43208> created a generic "hack" to essentially disable vectorization if the masked memref requires emulation, to "match" the preexisting behavior.
This patch moves it to TTI so that this can be moved towards the better cost modeling, one target at a time.

Thanks, @markus for reminding me.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D59149

Files:
  llvm/include/llvm/Analysis/TargetTransformInfo.h
  llvm/include/llvm/Analysis/TargetTransformInfoImpl.h
  llvm/lib/Analysis/TargetTransformInfo.cpp
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D59149.189902.patch
Type: text/x-patch
Size: 6044 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190308/a4df047a/attachment.bin>