[PATCH] [X86][SSE] Vectorized i64 uniform constant SRA shifts

[Simon Pilgrim]

In http://reviews.llvm.org/D9645#169914, @delena wrote:

> You should generate another code for SSE 4.1 and AVX, According to my previous comment.


Thanks - I have better support for pmovsx* coming in an upcoming later patch - it will be much easier to implement once this patch is already done.


REPOSITORY
  rL LLVM

http://reviews.llvm.org/D9645

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/