SHMEM_SET_CACHE_LINE_INV(3)
β¨ SHMEM_SET_CACHE_LINE_INV(3) β¨
okay bestie, this is the Open MPI manual page for SHMEM_SET_CACHE_LINE_INV(3) β yes, the cache-line-invalidation crew is here, and the whole vibe is βlegacy architecture support but make it cute.β π¦π
πΈ NAME πΈ
The official roster is: shmem_clear_cache_inv(3), shmem_set_cache_inv(3), shmem_set_cache_line_inv(3), shmem_udcflush(3), shmem_udcflush_line(3) β Controls data cache utilities. Periodt. π
π SYNOPSIS π
C or C++:
#include <mpp/shmem.h>
void shmem_clear_cache_inv(void)
void shmem_clear_cache_line_inv(void *target)
void shmem_set_cache_inv(void)
void shmem_set_cache_line_inv(void *target)
void shmem_udcflush(void)
void shmem_udcflush_line(void *target)
Fortran:
INCLUDE "mpp/shmem.fh"
CALL SHMEM_CLEAR_CACHE_INV
CALL SHMEM_CLEAR_CACHE_LINE_INV(target)
CALL SHMEM_SET_CACHE_INV
CALL SHMEM_SET_CACHE_LINE_INV(target)
CALL SHMEM_UDCFLUSH
CALL SHMEM_UDCFLUSH_LINE(target)
π¦ DESCRIPTION π¦
The argument that gets handed to the cache line control routines is:
targetβ a data object that is local to the processing element (PE).targetcan be of any noncharacter type. If youβre using Fortran, it can be of any kind. π
Now for the function-by-function glow-up:
shmem_clear_cache_invβ disables automatic cache coherency mode previously enabled byshmem_set_cache_invorshmem_set_cache_line_inv.shmem_clear_cache_line_invβ disables automatic cache coherency mode for the cache line associated with the address oftargetonly.shmem_set_cache_invβ enables the OpenSHMEM API to automatically decide the best strategy for cache coherency.shmem_set_cache_line_invβ enables automatic cache coherency mode for the cache line associated with the address oftargetonly.shmem_clear_cache_invβ yes, the page says it again, and Iβm keeping it because every detail matters: disables automatic cache coherency mode previously enabled byshmem_set_cache_invorshmem_set_cache_line_inv. πshmem_udcflushβ makes the entire user data cache coherent.shmem_udcflush_lineβ makes coherent the cache line that corresponds with the address specified bytarget.
π NOTES π
These routines have been retained for improved backward compatibility with legacy architectures. No cap β theyβre here for the legacy architecture girlies. π«Άβ¨
π SEE ALSO
intro_shmem(3)shmem_put(3)shmem_swap(3)
π COPYRIGHT π
2003-2026, The Open MPI Community
okay final sign-off, bestie: this Open MPI manual page is dated May 14, 2026, and the page name is SHMEM_SET_CACHE_LINE_INV(3) β mwah, bye πβ¨
π¦ where this page comes from
- arch rolling: nvhpc-comm-libs 26.9-1 Β· LicenseRef-NVIDIA-HPC-SDK
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms