[PATCH] D117798: [X86] Remove __builtin_ia32_pmax/min intrinsics and use generic __builtin_elementwise_max/min

Simon Pilgrim via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Fri Jan 21 02:44:19 PST 2022


RKSimon added a comment.

Thanks!



================
Comment at: clang/lib/Headers/avx512bwintrin.h:894
   return (__m512i)__builtin_ia32_selectb_512((__mmask64)__M,
                                              (__v64qi)_mm512_min_epu8(__A, __B),
                                              (__v64qi)_mm512_setzero_si512());
----------------
pengfei wrote:
> Should we change the type here too? The same below.
I don't think so - `__builtin_ia32_selectb_512` uses `__v64qi` everywhere else, and ignores signedness. It only matters inside the min/max builtins


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D117798/new/

https://reviews.llvm.org/D117798



More information about the cfe-commits mailing list