[PATCH] D20861: [AVX512] add missing integer to float conversion

Asaf Badouh via cfe-commits cfe-commits at lists.llvm.org
Wed Jun 1 05:46:07 PDT 2016


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

I used SSE intrinsics to implement some of the conversion.
for example:

```

#define _mm_cvtss_i32 _mm_cvtss_si32

```
Backend will generate avx512 asm in cases we compile with avx512f flag.


Repository:
  rL LLVM

http://reviews.llvm.org/D20861

Files:
  tools/clang/lib/Headers/avx512fintrin.h
  tools/clang/test/CodeGen/avx512f-builtins.c

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D20861.59210.patch
Type: text/x-patch
Size: 5728 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20160601/0d7eeaf7/attachment.bin>


More information about the cfe-commits mailing list