[PATCH] D27879: [AVX-512][InstCombine] Teach InstCombine to turn scalar add/sub/mul/div with rounding intrinsics into normal IR operations if the rounding mode is CUR_DIRECTION.

Craig Topper via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Sat Dec 17 10:19:01 PST 2016


craig.topper created this revision.
craig.topper added reviewers: delena, RKSimon, zvi.
craig.topper added a subscriber: llvm-commits.

I only do this for unmasked cases for now because isel is failing to fold the mask. I'll try to fix that soon.

I'll do the same thing for packed add/sub/mul/div in a future patch.


https://reviews.llvm.org/D27879

Files:
  lib/Transforms/InstCombine/InstCombineCalls.cpp
  test/Transforms/InstCombine/x86-avx512.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D27879.81845.patch
Type: text/x-patch
Size: 18059 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161217/12e10613/attachment.bin>


More information about the llvm-commits mailing list