[PATCH] D63882: AMDGPU: Custom lower vector_shuffle for v4i16/v4f16

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec accepted this revision.
rampitec added a comment.
This revision is now accepted and ready to land.

LGTM with small suggestion.



================
Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:4675
+  assert(Elt % 2 == 0);
+  return Mask[Elt + 1] == Mask[Elt] + 1 && (Mask[Elt] % 2 == 0);
+}
----------------
It's better to swap conditions. That way you will not read beyond the array even if accidentally pass an odd index.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63882/new/

https://reviews.llvm.org/D63882