[PATCH] D26686: [X86][AVX512] Replace lossless i32/u64 to f64 conversion intrinsics with generic IR

[Simon Pilgrim via cfe-commits]

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

Both the (V)CVTDQ2PD (i32 to f64) and (V)CVTDQ2PD (u32 to f64) conversion instructions are lossless and can be safely represented as generic __builtin_convertvector calls instead of x86 intrinsics without affecting final codegen.

This patch removes the clang builtins and their use in the headers - a future patch will deal with removing the llvm intrinsics.

This is an extension patch to https://reviews.llvm.org/D20528 which dealt with the equivalent sse/avx cases.


Repository:
  rL LLVM

https://reviews.llvm.org/D26686

Files:
  include/clang/Basic/BuiltinsX86.def
  lib/Headers/avx512fintrin.h
  lib/Headers/avx512vlintrin.h
  test/CodeGen/avx512f-builtins.c
  test/CodeGen/avx512vl-builtins.c

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D26686.78038.patch
Type: text/x-patch
Size: 18467 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20161115/1faed05c/attachment-0001.bin>