[PATCH] D18094: [x86, AVX] replace masked load with full vector load when possible

[Ashutosh Nema via llvm-commits]

ashutosh.nema added a comment.

Thanks Sanjay for working on this.

Changes looks OK to me, just a minor comment.


================
Comment at: lib/Target/X86/X86ISelLowering.cpp:27342
@@ -27341,6 +27341,3 @@
 
-/// Convert a masked load with a constant mask into a masked load and a select.
-/// This allows the select operation to use a faster kind of shuffle instruction
-/// (for example, vblendvps -> vblendps).
 static SDValue
 combineMaskedLoadConstantMask(MaskedLoadSDNode *ML, SelectionDAG &DAG,
----------------
Please retain and add comments here.


http://reviews.llvm.org/D18094