[PATCH] D14113: [X86][AVX512] add convert float to half

[Asaf Badouh via llvm-commits]

AsafBadouh created this revision.
AsafBadouh added reviewers: igorb, delena.
AsafBadouh added a subscriber: llvm-commits.

convert float to half with mask/maskz for the reg to reg version and mask for the reg to mem version (there is no maskz version for reg to mem).

made naming fix in GCC builtin for cvtph2ps in its 128bit width version (in order to be align with GCC) 


http://reviews.llvm.org/D14113

Files:
  ../commit/llvm_clean/include/llvm/IR/IntrinsicsX86.td
  ../commit/llvm_clean/lib/Target/X86/X86InstrAVX512.td
  ../commit/llvm_clean/lib/Target/X86/X86InstrFragmentsSIMD.td
  ../commit/llvm_clean/lib/Target/X86/X86IntrinsicsInfo.h
  ../commit/llvm_clean/test/CodeGen/X86/avx512vl-intrinsics.ll
  ../commit/llvm_clean/test/MC/X86/avx512-encodings.s
  ../commit/llvm_clean/test/MC/X86/x86-64-avx512f_vl.s

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D14113.38528.patch
Type: text/x-patch
Size: 17318 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20151027/2b7f0c4c/attachment.bin>