[PATCH] D37893: [X86] Prefer VPERMQ over VPERM2F128 for any unary shuffle, not just the ones that can be done with a insertf128

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: lib/Target/X86/X86ISelLowering.cpp:12163
+    if (Subtarget.hasAVX2() && V2.isUndef())
+      return SDValue();
+
----------------
Does this even need to be inside the zero test? If V2 is undef, then V1 should not be zero. Can we place it before lowerVectorShuffleAsBlend?


https://reviews.llvm.org/D37893