[PATCH] D19661: [X86] Also try to zero elts when lowering v32i8 shuffle with PSHUFB.
Ahmed Bougacha via llvm-commits
llvm-commits at lists.llvm.org
Fri May 27 12:26:30 PDT 2016
ab updated this revision to Diff 58824.
ab added a comment.
I did a final review before committing, but realized I missed the v16i16 case.
Here's an alternative patch, with a generalized PSHUFB lowering, and "zeroable mask" computation. WDYT?
I went ahead and renamed lowerVectorShuffleAsPSHUFB in r271024.
http://reviews.llvm.org/D19661
Files:
lib/Target/X86/X86ISelLowering.cpp
test/CodeGen/X86/vector-shuffle-256-v16.ll
test/CodeGen/X86/vector-shuffle-256-v32.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D19661.58824.patch
Type: text/x-patch
Size: 8445 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160527/86d778a0/attachment.bin>
More information about the llvm-commits
mailing list