[PATCH] D50600: AMDGPU: Use splat vectors for undefs when folding canonicalize

[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 except comment typo.



================
Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:7016
+      // than the normal qNaN. If it's a register, prefer 0.0 since that's
+      // cheaper to use amd may be free with a packed operation.
+      if (NewElts[0].isUndef()) {
----------------
typo: and


https://reviews.llvm.org/D50600