[PATCH] D41096: [X86] Initial support for prefer-vector-width function attribute

[Craig Topper via Phabricator via llvm-commits]

craig.topper added a comment.

Also doesn't this mean that we have to explicitly force +avx512-wide-vectors somewhere if the user passes "-mavx512f" with no -march?

  def FeatureAVX512WideVectors : SubtargetFeature<"avx512-wide-vectors", "HasAVX512WideVectors", "true", "Enable use of 512-bit Vector Registers", [FeatureAVX512]>;


https://reviews.llvm.org/D41096