[PATCH] D88396: [X86] Replace movaps with movups when avx is enabled.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: llvm/lib/Target/X86/CMakeLists.txt:54
   X86EvexToVex.cpp
+  X86MovapsToMovups.cpp
   X86LegalizerInfo.cpp
----------------
sorting


================
Comment at: llvm/lib/Target/X86/X86MovapsToMovups.cpp:1
+//===- X86MovapsToMovups.cpp ----------------------------------------------===//
+//
----------------
Very minor issue - but this isn't just movups - how about X86UnalignedVectorMoves.cpp ?


================
Comment at: llvm/lib/Target/X86/X86MovapsToMovups.cpp:13
+/// If the address is not aligned, movups can run without raising exception.
+/// So movups is prefered.
+//
----------------
"So unaligned load/stores may be preferred if hardware exceptions can't be trusted."?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88396/new/

https://reviews.llvm.org/D88396