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

[Sanjay Patel via Phabricator via llvm-commits]

spatel marked an inline comment as done.
spatel added inline comments.


================
Comment at: test/CodeGen/X86/haddsub.ll:1034
 
 ; 512-bit vectors, float/double, fadd/fsub
 
----------------
This wasn't clear from the patch update comment: we have 512-bit tests and AVX512 RUN lines. However, the output is identical to regular AVX1 because we're just extracting the low 128-bits.


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

https://reviews.llvm.org/D56011