[PATCH] D56011: [x86] lower extracted fadd/fsub to horizontal vector math

Tue Jan 1 10:15:27 PST 2019

spatel updated this revision to Diff 179798.
spatel added a comment.

Patch updated:
No functional diff from the previous rev, but rebased after pre-committing the move/rename of the helper with rL350193 <https://reviews.llvm.org/rL350193>.

One thing I'm not sure about - do we want to enable this for AVX512-sized vectors too? Ie, can we assert that the extract is from 128/256/512-bit instead of checking that explicitly?

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56011/new/

https://reviews.llvm.org/D56011

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/CodeGen/X86/haddsub-undef.ll
  test/CodeGen/X86/haddsub.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D56011.179798.patch
Type: text/x-patch
Size: 30154 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190101/416d2bdb/attachment.bin>