[PATCH] [x86] Begin a significant overhaul of how vector lowering is done in the x86 backend.

[David Majnemer]

================
Comment at: lib/Target/X86/X86ISelLowering.cpp:6781
@@ +6780,3 @@
+
+// NB: We rely heavily on "undef" masks preserving the input lane.
+static SDValue getV4ShuffleImmForMask(ArrayRef<int> Mask, SelectionDAG &DAG) {
----------------
Jim Grosbach wrote:
> "NB?"
It is probably an abbreviation for [[ http://en.wikipedia.org/wiki/Nota_bene | nota bene ]]

http://reviews.llvm.org/D4225