lib/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2025-08-08 17:42:12 +03:00
Code Activity

x86: Add seperate non-temporal tunable for memset

Browse Source 
The tuning for non-temporal stores for memset vs memcpy is not always
the same. This includes both the exact value and whether non-temporal
stores are profitable at all for a given arch.

This patch add `x86_memset_non_temporal_threshold`. Currently we
disable non-temporal stores for non Intel vendors as the only
benchmarks showing its benefit have been on Intel hardware.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
This commit is contained in:
Noah Goldstein
2024-05-24 12:38:51 -05:00
parent 5bf0ab8057
commit 46b5e98ef6
7 changed files with 49 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
sysdeps/x86/include/cpu-features.h
View File

@@ -944,8 +944,10 @@ struct cpu_features
/* Shared cache size for use in memory and string routines, typically
L2 or L3 size. */
unsigned long int shared_cache_size;
/* Threshold to use non temporal store. */
/* Threshold to use non temporal store in memmove. */
unsigned long int non_temporal_threshold;
/* Threshold to use non temporal store in memset. */
unsigned long int memset_non_temporal_threshold;
/* Threshold to use "rep movsb". */
unsigned long int rep_movsb_threshold;
/* Threshold to stop using "rep movsb". */