[PATCH] D38671: [X86][AVX512] lowering shuffle i/f intrinsic - llvm part

[jina via Phabricator via llvm-commits]

jina.nahias added inline comments.


================
Comment at: lib/Target/X86/X86ISelLowering.cpp:29084
 
+  // Attempt to use SHUF128 for masked X86ISD::VPERM2X128
+  // from this paterrn:
----------------
craig.topper wrote:
> You should do this in combineBitcastForMaskedOp. It was already created to fix shuffles to work with masking.
this function is called only if the select operand is bitcast, but in 64x2 intrinsics there are no bitcast and this function wont be called. 


https://reviews.llvm.org/D38671