[PATCH] D52121: [X86] Fold (movmsk (setne (and X, (1 << C)), 0)) -> (movmsk (X << C))

[Simon Pilgrim via Phabricator via llvm-commits]

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

LGTM with one minor



================
Comment at: lib/Target/X86/X86ISelLowering.cpp:38823
+      Src.getOperand(0).getValueType() == Src.getValueType() &&
+      Src.getValueType().getScalarSizeInBits() >= 16 &&
+      cast<CondCodeSDNode>(Src.getOperand(2))->get() == ISD::SETNE &&
----------------
Safer to use >= 32 since movmsk vXi16 doesn't exist.


https://reviews.llvm.org/D52121