[PATCH] D15787: [X86][AVX512] Lower broadcast inrtrinsics

[Asaf Badouh via llvm-commits]

AsafBadouh created this revision.
AsafBadouh added reviewers: delena, igorb, m_zuckerman.
AsafBadouh added a subscriber: llvm-commits.
AsafBadouh set the repository for this revision to rL LLVM.

lower broadcast<type>x<vector> to shuffles.
there are two cases:
1. src is 128 bits and dest is 512 bits:
  in this case we will lower it to shuffle with imm = 0.
2. src is 256 bit and dest is 512 bits:
  in this case we will lower it to shuffle with imm = 01000100b (0x44)
  that way we will broadcast the 256bit source:
  ymm[0,1,2,3] => zmm[0,1,2,3,0,1,2,3]
  then it will mask it with the pass thru value (in case it's mask op).





Repository:
  rL LLVM

http://reviews.llvm.org/D15787

Files:
  ../llvm/include/llvm/IR/IntrinsicsX86.td
  ../llvm/lib/Target/X86/X86ISelLowering.cpp
  ../llvm/lib/Target/X86/X86IntrinsicsInfo.h
  ../llvm/test/CodeGen/X86/avx512-intrinsics.ll
  ../llvm/test/CodeGen/X86/avx512dq-intrinsics.ll
  ../llvm/test/CodeGen/X86/avx512dqvl-intrinsics.ll
  ../llvm/test/CodeGen/X86/avx512vl-intrinsics.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D15787.43657.patch
Type: text/x-patch
Size: 20261 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20151227/35f63dce/attachment.bin>