From 6374dceef2bf72456144184cb70bdd216a744d61 Mon Sep 17 00:00:00 2001 From: May Date: Wed, 15 Jan 2025 10:42:33 +0800 Subject: [PATCH 3/3] x86: Set default non_temporal_threshold for Zhaoxin processors Current 'non_temporal_threshold' set to 'non_temporal_threshold_lowbound' on Zhaoxin processors without ERMS. The default 'non_temporal_threshold_lowbound' is too small for the KH-40000 and KX-7000 Zhaoxin processors, this patch updates the value to 'shared / cachesize_non_temporal_divisor'. Signed-off-by: May --- sysdeps/x86/cpu-features.c | 1 + sysdeps/x86/dl-cacheinfo.h | 6 ++++-- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/sysdeps/x86/cpu-features.c b/sysdeps/x86/cpu-features.c index 43b5f562..f752ebd2 100644 --- a/sysdeps/x86/cpu-features.c +++ b/sysdeps/x86/cpu-features.c @@ -949,6 +949,7 @@ https://www.intel.com/content/www/us/en/support/articles/000059422/processors.ht /* Yongfeng and Shijidadao mircoarch tuning. */ case 0x5b: + cpu_features->cachesize_non_temporal_divisor = 2; case 0x6b: cpu_features->preferred[index_arch_AVX_Fast_Unaligned_Load] &= ~bit_arch_AVX_Fast_Unaligned_Load; diff --git a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h index 6c774042..bd2a9122 100644 --- a/sysdeps/x86/dl-cacheinfo.h +++ b/sysdeps/x86/dl-cacheinfo.h @@ -940,8 +940,10 @@ dl_init_cacheinfo (struct cpu_features *cpu_features) /* If no ERMS, we use the per-thread L3 chunking. Normal cacheable stores run a higher risk of actually thrashing the cache as they don't have a HW LRU hint. As well, their performance in highly parallel situations is - noticeably worse. */ - if (!CPU_FEATURE_USABLE_P (cpu_features, ERMS)) + noticeably worse. Zhaoxin processors are an exception, the lowbound is not + suitable for them based on actual test data. */ + if (!CPU_FEATURE_USABLE_P (cpu_features, ERMS) + && cpu_features->basic.kind != arch_kind_zhaoxin) non_temporal_threshold = non_temporal_threshold_lowbound; /* SIZE_MAX >> 4 because memmove-vec-unaligned-erms right-shifts the value of 'x86_non_temporal_threshold' by `LOG_4X_MEMCPY_THRESH` (4) and it is best -- 2.27.0