[PATCH] D11886: [InstCombine] Move SSE2/AVX2 arithmetic vector shift folding to instcombiner

[Andrea Di Biagio via llvm-commits]

andreadb added a comment.

Hi Simon,

I saw that Michael and David already reviewed your patch.
If you address their comments then the patch looks good to me too. Thanks!

In future, we should also move the target specific combine rules on sse/avx blend intrinsic calls from 'PerformINTRINSIC_WO_CHAINCombine' to InstCombine.

Thanks,
-Andrea


Repository:
  rL LLVM

http://reviews.llvm.org/D11886