OSDN Git Service

[OpenMP] Added the support for cache line size 256 for A64FX
authorShilei Tian <tianshilei1992@gmail.com>
Sat, 9 Jan 2021 16:58:37 +0000 (11:58 -0500)
committerShilei Tian <tianshilei1992@gmail.com>
Sat, 9 Jan 2021 16:58:47 +0000 (11:58 -0500)
commit676c7cb0c0d4b66affde3ff7fc566c7a5aaa7246
treeaa6ae7759238fce5cfd053e2952dd3f3a7f475b9
parent524d8fa9a5a5428628a21a91016a52a54a9fe838
[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
openmp/runtime/CMakeLists.txt
openmp/runtime/cmake/LibompGetArchitecture.cmake
openmp/runtime/cmake/LibompUtils.cmake
openmp/runtime/cmake/config-ix.cmake
openmp/runtime/src/kmp_config.h.cmake