[PATCH] D107904: [LegalizeTypes][VP] Add widening support for binary VP ops

[Fraser Cormack via Phabricator via llvm-commits]

frasercrmck marked an inline comment as done.
frasercrmck added inline comments.


================
Comment at: llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp:3250
+  // may not actually need widening, leading it be split along with the VP
+  // operation. This could lead to an infinite split/widen loop. We only handle
+  // the case where the mask needs widening to an identically-sized type as the
----------------
craig.topper wrote:
> This deserves a FIXME. AVX512 may want to use VP for predication and this would need to be fixed.
Good point. Added now.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107904/new/

https://reviews.llvm.org/D107904