[PATCH] D130830: Don't widen shuffle element with AVX512

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:16802
+        SVOp->getMask() == Mask)
+      return SDValue();
+
----------------
LuoYuanke wrote:
> Fix crashing.
We really need to avoid creating nodes if we're going to bail - can this be merged into the CanUseSublanes check above? Or just a relaxed comparison of Mask and InLaneMask ?


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:17572
+        SVOp->getMask() == Mask)
+      return SDValue();
+
----------------
LuoYuanke wrote:
> Fix crashing.
Again - don't create a shuffle if we're not going to use it.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130830/new/

https://reviews.llvm.org/D130830