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

[Simon Pilgrim]

Hi andreadb, delena, qcolombet,

This patch adds vectorization support for uniform constant i64 arithmetic shift right operators - this is particularly useful for improving sign extension to i64 types.

REPOSITORY
  rL LLVM

http://reviews.llvm.org/D9645

Files:
  lib/Target/X86/X86ISelLowering.cpp
  lib/Target/X86/X86TargetTransformInfo.cpp
  test/Analysis/CostModel/X86/testshiftashr.ll
  test/CodeGen/X86/vec_int_to_fp.ll
  test/CodeGen/X86/vector-sext.ll
  test/CodeGen/X86/vshift-3.ll
  test/CodeGen/X86/widen_conv-2.ll

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D9645.25426.patch
Type: text/x-patch
Size: 30455 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150510/f499e060/attachment.bin>