[PATCH] D39704: [X86] [CodeGen] Compiler not using SHLD/SHRD instructions when doing double shift pattern combine for 16bit or 8bit arguments (PR35155)

[Konstantin Belochapka via Phabricator via llvm-commits]

kbelochapka created this revision.

Use SHLD/SHRD instruction when performing double shift pattern combine for any unsigned integer type.

template<typename T, typename SH> 
T shld(T x, T, y, SH bits) {

  return (x << bits)  | ( y >> (sizeof(T) - bits))

}
T  - any of unsigned integer type
SH  - any of signed or unsigned integer type


https://reviews.llvm.org/D39704

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/CodeGen/X86/shift-double-x86_64.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D39704.121794.patch
Type: text/x-patch
Size: 42264 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171106/2ec8feda/attachment-0001.bin>