[all-commits] [llvm/llvm-project] 676c7c: [OpenMP] Added the support for cache line size 256...

[Shilei Tian via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 676c7cb0c0d4b66affde3ff7fc566c7a5aaa7246
      https://github.com/llvm/llvm-project/commit/676c7cb0c0d4b66affde3ff7fc566c7a5aaa7246
  Author: Shilei Tian <tianshilei1992 at gmail.com>
  Date:   2021-01-09 (Sat, 09 Jan 2021)

  Changed paths:
    M openmp/runtime/CMakeLists.txt
    M openmp/runtime/cmake/LibompGetArchitecture.cmake
    M openmp/runtime/cmake/LibompUtils.cmake
    M openmp/runtime/cmake/config-ix.cmake
    M openmp/runtime/src/kmp_config.h.cmake

  Log Message:
  -----------
  [OpenMP] Added the support for cache line size 256 for A64FX

Fugaku supercomputer is built with the Fujitsu A64FX microprocessor, whose cache line is 256. In current libomp, we only have cache line size 128 for PPC64 and otherwise 64. This patch added the support of cache line 256 for A64FX. It's worth noting that although A64FX is a variant of AArch64, this property is not shared. As a result, in light of UCX source code (https://github.com/openucx/ucx/blob/392443ab92626412605dee1572056f79c897c6c3/src/ucs/arch/aarch64/cpu.c#L17), we can only determine by checking whether the CPU is FUJITSU A64FX.

Reviewed By: jdoerfert, Hahnfeld

Differential Revision: https://reviews.llvm.org/D93169