[PATCH] D44267: Remove SRAs from v16i8 multiply lowering on sse2 targets

[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



================
Comment at: lib/Target/X86/X86ISelLowering.cpp:22322
     } else {
-      const int ShufMask[] = {-1, 0, -1, 1, -1, 2, -1, 3,
-                              -1, 4, -1, 5, -1, 6, -1, 7};
+      // We're going mask off the low byte of each result element of the mullw,
+      // so it doesn't matter what's in the high byte of each 16-bit element.
----------------
going to mask


================
Comment at: lib/Target/X86/X86ISelLowering.cpp:22342
     } else {
-      const int ShufMask[] = {-1, 8,  -1, 9,  -1, 10, -1, 11,
-                              -1, 12, -1, 13, -1, 14, -1, 15};
+      // We're going mask off the low byte of each result element of the mullw,
+      // so it doesn't matter what's in the high byte of each 16-bit element.
----------------
going to mask


https://reviews.llvm.org/D44267