rpms
/
glibc
7
0
Fork 1
Code Issues Pull Requests Releases Activity

Fix temporal threshold calculations (#2213907) 

Resolves: #221390
This commit is contained in:
DJ Delorie 2023-07-26 22:45:32 -04:00
parent 4a20a2b4f5
commit 8c4d8a0e5b
7 changed files with 1035 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
glibc-rh2213907-1.patch Normal file
View File

@ -0,0 +1,45 @@
From ed2f9dc9420c4c61436328778a70459d0a35556a Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Mon, 8 May 2023 22:10:20 -0500
Subject: [PATCH] x86: Use 64MB as nt-store threshold if no cacheinfo [BZ
#30429]
Content-type: text/plain; charset=UTF-8
If `non_temporal_threshold` is below `minimum_non_temporal_threshold`,
it almost certainly means we failed to read the systems cache info.
In this case, rather than defaulting the minimum correct value, we
should default to a value that gets at least reasonable
performance. 64MB is chosen conservatively to be at the very high
end. This should never cause non-temporal stores when, if we had read
cache info, we wouldn't have otherwise.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
---
sysdeps/x86/dl-cacheinfo.h | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
index ec88945b39..877e73d700 100644
--- a/sysdeps/x86/dl-cacheinfo.h
+++ b/sysdeps/x86/dl-cacheinfo.h
@@ -749,8 +749,16 @@ dl_init_cacheinfo (struct cpu_features *cpu_features)
reflected in the manual. */
unsigned long int maximum_non_temporal_threshold = SIZE_MAX >> 4;
unsigned long int minimum_non_temporal_threshold = 0x4040;
+
+ /* If `non_temporal_threshold` less than `minimum_non_temporal_threshold`
+ it most likely means we failed to detect the cache info. We don't want
+ to default to `minimum_non_temporal_threshold` as such a small value,
+ while correct, has bad performance. We default to 64MB as reasonable
+ default bound. 64MB is likely conservative in that most/all systems would
+ choose a lower value so it should never forcing non-temporal stores when
+ they otherwise wouldn't be used. */
if (non_temporal_threshold < minimum_non_temporal_threshold)
- non_temporal_threshold = minimum_non_temporal_threshold;
+ non_temporal_threshold = 64 * 1024 * 1024;
else if (non_temporal_threshold > maximum_non_temporal_threshold)
non_temporal_threshold = maximum_non_temporal_threshold;
--
2.39.3

223
glibc-rh2213907-2.patch Normal file
View File

@ -0,0 +1,223 @@
From af992e7abdc9049714da76cae1e5e18bc4838fb8 Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Wed, 7 Jun 2023 13:18:01 -0500
Subject: [PATCH] x86: Increase `non_temporal_threshold` to roughly `sizeof_L3
/ 4`
Content-type: text/plain; charset=UTF-8
Current `non_temporal_threshold` set to roughly '3/4 * sizeof_L3 /
ncores_per_socket'. This patch updates that value to roughly
'sizeof_L3 / 4`
The original value (specifically dividing the `ncores_per_socket`) was
done to limit the amount of other threads' data a `memcpy`/`memset`
could evict.
Dividing by 'ncores_per_socket', however leads to exceedingly low
non-temporal thresholds and leads to using non-temporal stores in
cases where REP MOVSB is multiple times faster.
Furthermore, non-temporal stores are written directly to main memory
so using it at a size much smaller than L3 can place soon to be
accessed data much further away than it otherwise could be. As well,
modern machines are able to detect streaming patterns (especially if
REP MOVSB is used) and provide LRU hints to the memory subsystem. This
in affect caps the total amount of eviction at 1/cache_associativity,
far below meaningfully thrashing the entire cache.
As best I can tell, the benchmarks that lead this small threshold
where done comparing non-temporal stores versus standard cacheable
stores. A better comparison (linked below) is to be REP MOVSB which,
on the measure systems, is nearly 2x faster than non-temporal stores
at the low-end of the previous threshold, and within 10% for over
100MB copies (well past even the current threshold). In cases with a
low number of threads competing for bandwidth, REP MOVSB is ~2x faster
up to `sizeof_L3`.
The divisor of `4` is a somewhat arbitrary value. From benchmarks it
seems Skylake and Icelake both prefer a divisor of `2`, but older CPUs
such as Broadwell prefer something closer to `8`. This patch is meant
to be followed up by another one to make the divisor cpu-specific, but
in the meantime (and for easier backporting), this patch settles on
`4` as a middle-ground.
Benchmarks comparing non-temporal stores, REP MOVSB, and cacheable
stores where done using:
https://github.com/goldsteinn/memcpy-nt-benchmarks
Sheets results (also available in pdf on the github):
https://docs.google.com/spreadsheets/d/e/2PACX-1vS183r0rW_jRX6tG_E90m9qVuFiMbRIJvi5VAE8yYOvEOIEEc3aSNuEsrFbuXw5c3nGboxMmrupZD7K/pubhtml
Reviewed-by: DJ Delorie <dj@redhat.com>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
---
sysdeps/x86/dl-cacheinfo.h | 70 +++++++++++++++++++++++---------------
1 file changed, 43 insertions(+), 27 deletions(-)
[diff rebased by DJ]
diff -rup a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
--- a/sysdeps/x86/dl-cacheinfo.h 2023-07-25 00:38:39.386831871 -0400
+++ b/sysdeps/x86/dl-cacheinfo.h 2023-07-25 00:38:40.372870369 -0400
@@ -408,7 +408,7 @@ handle_zhaoxin (int name)
}
static void
-get_common_cache_info (long int *shared_ptr, unsigned int *threads_ptr,
+get_common_cache_info (long int *shared_ptr, long int * shared_per_thread_ptr, unsigned int *threads_ptr,
long int core)
{
unsigned int eax;
@@ -427,6 +427,7 @@ get_common_cache_info (long int *shared_
unsigned int family = cpu_features->basic.family;
unsigned int model = cpu_features->basic.model;
long int shared = *shared_ptr;
+ long int shared_per_thread = *shared_per_thread_ptr;
unsigned int threads = *threads_ptr;
bool inclusive_cache = true;
bool support_count_mask = true;
@@ -442,6 +443,7 @@ get_common_cache_info (long int *shared_
/* Try L2 otherwise. */
level = 2;
shared = core;
+ shared_per_thread = core;
threads_l2 = 0;
threads_l3 = -1;
}
@@ -598,29 +600,28 @@ get_common_cache_info (long int *shared_
}
else
{
-intel_bug_no_cache_info:
- /* Assume that all logical threads share the highest cache
- level. */
- threads
- = ((cpu_features->features[CPUID_INDEX_1].cpuid.ebx >> 16)
- & 0xff);
- }
-
- /* Cap usage of highest cache level to the number of supported
- threads. */
- if (shared > 0 && threads > 0)
- shared /= threads;
+ intel_bug_no_cache_info:
+ /* Assume that all logical threads share the highest cache
+ level. */
+ threads = ((cpu_features->features[CPUID_INDEX_1].cpuid.ebx >> 16)
+ & 0xff);
+
+ /* Get per-thread size of highest level cache. */
+ if (shared_per_thread > 0 && threads > 0)
+ shared_per_thread /= threads;
+ }
}
/* Account for non-inclusive L2 and L3 caches. */
if (!inclusive_cache)
{
if (threads_l2 > 0)
- core /= threads_l2;
+ shared_per_thread += core / threads_l2;
shared += core;
}
*shared_ptr = shared;
+ *shared_per_thread_ptr = shared_per_thread;
*threads_ptr = threads;
}
@@ -630,6 +631,7 @@ dl_init_cacheinfo (struct cpu_features *
/* Find out what brand of processor. */
long int data = -1;
long int shared = -1;
+ long int shared_per_thread = -1;
long int core = -1;
unsigned int threads = 0;
unsigned long int level1_icache_size = -1;
@@ -650,6 +652,7 @@ dl_init_cacheinfo (struct cpu_features *
data = handle_intel (_SC_LEVEL1_DCACHE_SIZE, cpu_features);
core = handle_intel (_SC_LEVEL2_CACHE_SIZE, cpu_features);
shared = handle_intel (_SC_LEVEL3_CACHE_SIZE, cpu_features);
+ shared_per_thread = shared;
level1_icache_size
= handle_intel (_SC_LEVEL1_ICACHE_SIZE, cpu_features);
@@ -673,13 +676,14 @@ dl_init_cacheinfo (struct cpu_features *
level4_cache_size
= handle_intel (_SC_LEVEL4_CACHE_SIZE, cpu_features);
- get_common_cache_info (&shared, &threads, core);
+ get_common_cache_info (&shared, &shared_per_thread, &threads, core);
}
else if (cpu_features->basic.kind == arch_kind_zhaoxin)
{
data = handle_zhaoxin (_SC_LEVEL1_DCACHE_SIZE);
core = handle_zhaoxin (_SC_LEVEL2_CACHE_SIZE);
shared = handle_zhaoxin (_SC_LEVEL3_CACHE_SIZE);
+ shared_per_thread = shared;
level1_icache_size = handle_zhaoxin (_SC_LEVEL1_ICACHE_SIZE);
level1_icache_linesize = handle_zhaoxin (_SC_LEVEL1_ICACHE_LINESIZE);
@@ -693,13 +697,14 @@ dl_init_cacheinfo (struct cpu_features *
level3_cache_assoc = handle_zhaoxin (_SC_LEVEL3_CACHE_ASSOC);
level3_cache_linesize = handle_zhaoxin (_SC_LEVEL3_CACHE_LINESIZE);
- get_common_cache_info (&shared, &threads, core);
+ get_common_cache_info (&shared, &shared_per_thread, &threads, core);
}
else if (cpu_features->basic.kind == arch_kind_amd)
{
data = handle_amd (_SC_LEVEL1_DCACHE_SIZE, cpu_features);
core = handle_amd (_SC_LEVEL2_CACHE_SIZE, cpu_features);
shared = handle_amd (_SC_LEVEL3_CACHE_SIZE, cpu_features);
+ shared_per_thread = shared;
level1_icache_size = handle_amd (_SC_LEVEL1_ICACHE_SIZE, cpu_features);
level1_icache_linesize
@@ -721,6 +726,9 @@ dl_init_cacheinfo (struct cpu_features *
if (shared <= 0)
/* No shared L3 cache. All we have is the L2 cache. */
shared = core;
+
+ if (shared_per_thread <= 0)
+ shared_per_thread = shared;
}
cpu_features->level1_icache_size = level1_icache_size;
@@ -736,17 +744,25 @@ dl_init_cacheinfo (struct cpu_features *
cpu_features->level3_cache_linesize = level3_cache_linesize;
cpu_features->level4_cache_size = level4_cache_size;
- /* The default setting for the non_temporal threshold is 3/4 of one
- thread's share of the chip's cache. For most Intel and AMD processors
- with an initial release date between 2017 and 2020, a thread's typical
- share of the cache is from 500 KBytes to 2 MBytes. Using the 3/4
- threshold leaves 125 KBytes to 500 KBytes of the thread's data
- in cache after a maximum temporal copy, which will maintain
- in cache a reasonable portion of the thread's stack and other
- active data. If the threshold is set higher than one thread's
- share of the cache, it has a substantial risk of negatively
- impacting the performance of other threads running on the chip. */
- unsigned long int non_temporal_threshold = shared * 3 / 4;
+ /* The default setting for the non_temporal threshold is 1/4 of size
+ of the chip's cache. For most Intel and AMD processors with an
+ initial release date between 2017 and 2023, a thread's typical
+ share of the cache is from 18-64MB. Using the 1/4 L3 is meant to
+ estimate the point where non-temporal stores begin out-competing
+ REP MOVSB. As well the point where the fact that non-temporal
+ stores are forced back to main memory would already occurred to the
+ majority of the lines in the copy. Note, concerns about the
+ entire L3 cache being evicted by the copy are mostly alleviated
+ by the fact that modern HW detects streaming patterns and
+ provides proper LRU hints so that the maximum thrashing
+ capped at 1/associativity. */
+ unsigned long int non_temporal_threshold = shared / 4;
+ /* 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))
+ non_temporal_threshold = shared_per_thread * 3 / 4;
/* 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
if that operation cannot overflow. Minimum of 0x4040 (16448) because the

475
glibc-rh2213907-3.patch Normal file
View File

@ -0,0 +1,475 @@
From f193ea20eddc6cef84cba54cf1a647204ee6a86b Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Wed, 7 Jun 2023 13:18:02 -0500
Subject: [PATCH] x86: Refactor Intel `init_cpu_features`
Content-type: text/plain; charset=UTF-8
This patch should have no affect on existing functionality.
The current code, which has a single switch for model detection and
setting prefered features, is difficult to follow/extend. The cases
use magic numbers and many microarchitectures are missing. This makes
it difficult to reason about what is implemented so far and/or
how/where to add support for new features.
This patch splits the model detection and preference setting stages so
that CPU preferences can be set based on a complete list of available
microarchitectures, rather than based on model magic numbers.
Reviewed-by: DJ Delorie <dj@redhat.com>
---
sysdeps/x86/cpu-features.c | 390 +++++++++++++++++++++++++++++--------
1 file changed, 309 insertions(+), 81 deletions(-)
diff --git a/sysdeps/x86/cpu-features.c b/sysdeps/x86/cpu-features.c
index 0a99efdb28..d52a718e92 100644
--- a/sysdeps/x86/cpu-features.c
+++ b/sysdeps/x86/cpu-features.c
@@ -417,6 +417,216 @@ _Static_assert (((index_arch_Fast_Unaligned_Load
== index_arch_Fast_Copy_Backward)),
"Incorrect index_arch_Fast_Unaligned_Load");
+
+/* Intel Family-6 microarch list. */
+enum
+{
+ /* Atom processors. */
+ INTEL_ATOM_BONNELL,
+ INTEL_ATOM_SILVERMONT,
+ INTEL_ATOM_AIRMONT,
+ INTEL_ATOM_GOLDMONT,
+ INTEL_ATOM_GOLDMONT_PLUS,
+ INTEL_ATOM_SIERRAFOREST,
+ INTEL_ATOM_GRANDRIDGE,
+ INTEL_ATOM_TREMONT,
+
+ /* Bigcore processors. */
+ INTEL_BIGCORE_MEROM,
+ INTEL_BIGCORE_PENRYN,
+ INTEL_BIGCORE_DUNNINGTON,
+ INTEL_BIGCORE_NEHALEM,
+ INTEL_BIGCORE_WESTMERE,
+ INTEL_BIGCORE_SANDYBRIDGE,
+ INTEL_BIGCORE_IVYBRIDGE,
+ INTEL_BIGCORE_HASWELL,
+ INTEL_BIGCORE_BROADWELL,
+ INTEL_BIGCORE_SKYLAKE,
+ INTEL_BIGCORE_KABYLAKE,
+ INTEL_BIGCORE_COMETLAKE,
+ INTEL_BIGCORE_SKYLAKE_AVX512,
+ INTEL_BIGCORE_CANNONLAKE,
+ INTEL_BIGCORE_ICELAKE,
+ INTEL_BIGCORE_TIGERLAKE,
+ INTEL_BIGCORE_ROCKETLAKE,
+ INTEL_BIGCORE_SAPPHIRERAPIDS,
+ INTEL_BIGCORE_RAPTORLAKE,
+ INTEL_BIGCORE_EMERALDRAPIDS,
+ INTEL_BIGCORE_METEORLAKE,
+ INTEL_BIGCORE_LUNARLAKE,
+ INTEL_BIGCORE_ARROWLAKE,
+ INTEL_BIGCORE_GRANITERAPIDS,
+
+ /* Mixed (bigcore + atom SOC). */
+ INTEL_MIXED_LAKEFIELD,
+ INTEL_MIXED_ALDERLAKE,
+
+ /* KNL. */
+ INTEL_KNIGHTS_MILL,
+ INTEL_KNIGHTS_LANDING,
+
+ /* Unknown. */
+ INTEL_UNKNOWN,
+};
+
+static unsigned int
+intel_get_fam6_microarch (unsigned int model,
+ __attribute__ ((unused)) unsigned int stepping)
+{
+ switch (model)
+ {
+ case 0x1C:
+ case 0x26:
+ return INTEL_ATOM_BONNELL;
+ case 0x27:
+ case 0x35:
+ case 0x36:
+ /* Really Saltwell, but Saltwell is just a die shrink of Bonnell
+ (microarchitecturally identical). */
+ return INTEL_ATOM_BONNELL;
+ case 0x37:
+ case 0x4A:
+ case 0x4D:
+ case 0x5D:
+ return INTEL_ATOM_SILVERMONT;
+ case 0x4C:
+ case 0x5A:
+ case 0x75:
+ return INTEL_ATOM_AIRMONT;
+ case 0x5C:
+ case 0x5F:
+ return INTEL_ATOM_GOLDMONT;
+ case 0x7A:
+ return INTEL_ATOM_GOLDMONT_PLUS;
+ case 0xAF:
+ return INTEL_ATOM_SIERRAFOREST;
+ case 0xB6:
+ return INTEL_ATOM_GRANDRIDGE;
+ case 0x86:
+ case 0x96:
+ case 0x9C:
+ return INTEL_ATOM_TREMONT;
+ case 0x0F:
+ case 0x16:
+ return INTEL_BIGCORE_MEROM;
+ case 0x17:
+ return INTEL_BIGCORE_PENRYN;
+ case 0x1D:
+ return INTEL_BIGCORE_DUNNINGTON;
+ case 0x1A:
+ case 0x1E:
+ case 0x1F:
+ case 0x2E:
+ return INTEL_BIGCORE_NEHALEM;
+ case 0x25:
+ case 0x2C:
+ case 0x2F:
+ return INTEL_BIGCORE_WESTMERE;
+ case 0x2A:
+ case 0x2D:
+ return INTEL_BIGCORE_SANDYBRIDGE;
+ case 0x3A:
+ case 0x3E:
+ return INTEL_BIGCORE_IVYBRIDGE;
+ case 0x3C:
+ case 0x3F:
+ case 0x45:
+ case 0x46:
+ return INTEL_BIGCORE_HASWELL;
+ case 0x3D:
+ case 0x47:
+ case 0x4F:
+ case 0x56:
+ return INTEL_BIGCORE_BROADWELL;
+ case 0x4E:
+ case 0x5E:
+ return INTEL_BIGCORE_SKYLAKE;
+ case 0x8E:
+ /*
+ Stepping = {9}
+ -> Amberlake
+ Stepping = {10}
+ -> Coffeelake
+ Stepping = {11, 12}
+ -> Whiskeylake
+ else
+ -> Kabylake
+
+ All of these are derivatives of Kabylake (Skylake client).
+ */
+ return INTEL_BIGCORE_KABYLAKE;
+ case 0x9E:
+ /*
+ Stepping = {10, 11, 12, 13}
+ -> Coffeelake
+ else
+ -> Kabylake
+
+ Coffeelake is a derivatives of Kabylake (Skylake client).
+ */
+ return INTEL_BIGCORE_KABYLAKE;
+ case 0xA5:
+ case 0xA6:
+ return INTEL_BIGCORE_COMETLAKE;
+ case 0x66:
+ return INTEL_BIGCORE_CANNONLAKE;
+ case 0x55:
+ /*
+ Stepping = {6, 7}
+ -> Cascadelake
+ Stepping = {11}
+ -> Cooperlake
+ else
+ -> Skylake-avx512
+
+ These are all microarchitecturally indentical, so use
+ Skylake-avx512 for all of them.
+ */
+ return INTEL_BIGCORE_SKYLAKE_AVX512;
+ case 0x6A:
+ case 0x6C:
+ case 0x7D:
+ case 0x7E:
+ case 0x9D:
+ return INTEL_BIGCORE_ICELAKE;
+ case 0x8C:
+ case 0x8D:
+ return INTEL_BIGCORE_TIGERLAKE;
+ case 0xA7:
+ return INTEL_BIGCORE_ROCKETLAKE;
+ case 0x8F:
+ return INTEL_BIGCORE_SAPPHIRERAPIDS;
+ case 0xB7:
+ case 0xBA:
+ case 0xBF:
+ return INTEL_BIGCORE_RAPTORLAKE;
+ case 0xCF:
+ return INTEL_BIGCORE_EMERALDRAPIDS;
+ case 0xAA:
+ case 0xAC:
+ return INTEL_BIGCORE_METEORLAKE;
+ case 0xbd:
+ return INTEL_BIGCORE_LUNARLAKE;
+ case 0xc6:
+ return INTEL_BIGCORE_ARROWLAKE;
+ case 0xAD:
+ case 0xAE:
+ return INTEL_BIGCORE_GRANITERAPIDS;
+ case 0x8A:
+ return INTEL_MIXED_LAKEFIELD;
+ case 0x97:
+ case 0x9A:
+ case 0xBE:
+ return INTEL_MIXED_ALDERLAKE;
+ case 0x85:
+ return INTEL_KNIGHTS_MILL;
+ case 0x57:
+ return INTEL_KNIGHTS_LANDING;
+ default:
+ return INTEL_UNKNOWN;
+ }
+}
+
static inline void
init_cpu_features (struct cpu_features *cpu_features)
{
@@ -453,129 +663,147 @@ init_cpu_features (struct cpu_features *cpu_features)
if (family == 0x06)
{
model += extended_model;
- switch (model)
+ unsigned int microarch
+ = intel_get_fam6_microarch (model, stepping);
+
+ switch (microarch)
{
- case 0x1c:
- case 0x26:
- /* BSF is slow on Atom. */
+ /* Atom / KNL tuning. */
+ case INTEL_ATOM_BONNELL:
+ /* BSF is slow on Bonnell. */
cpu_features->preferred[index_arch_Slow_BSF]
- |= bit_arch_Slow_BSF;
+ |= bit_arch_Slow_BSF;
break;
- case 0x57:
- /* Knights Landing. Enable Silvermont optimizations. */
-
- case 0x7a:
- /* Unaligned load versions are faster than SSSE3
- on Goldmont Plus. */
-
- case 0x5c:
- case 0x5f:
/* Unaligned load versions are faster than SSSE3
- on Goldmont. */
+ on Airmont, Silvermont, Goldmont, and Goldmont Plus. */
+ case INTEL_ATOM_AIRMONT:
+ case INTEL_ATOM_SILVERMONT:
+ case INTEL_ATOM_GOLDMONT:
+ case INTEL_ATOM_GOLDMONT_PLUS:
- case 0x4c:
- case 0x5a:
- case 0x75:
- /* Airmont is a die shrink of Silvermont. */
+ /* Knights Landing. Enable Silvermont optimizations. */
+ case INTEL_KNIGHTS_LANDING:
- case 0x37:
- case 0x4a:
- case 0x4d:
- case 0x5d:
- /* Unaligned load versions are faster than SSSE3
- on Silvermont. */
cpu_features->preferred[index_arch_Fast_Unaligned_Load]
- |= (bit_arch_Fast_Unaligned_Load
- | bit_arch_Fast_Unaligned_Copy
- | bit_arch_Prefer_PMINUB_for_stringop
- | bit_arch_Slow_SSE4_2);
+ |= (bit_arch_Fast_Unaligned_Load
+ | bit_arch_Fast_Unaligned_Copy
+ | bit_arch_Prefer_PMINUB_for_stringop
+ | bit_arch_Slow_SSE4_2);
break;
- case 0x86:
- case 0x96:
- case 0x9c:
+ case INTEL_ATOM_TREMONT:
/* Enable rep string instructions, unaligned load, unaligned
- copy, pminub and avoid SSE 4.2 on Tremont. */
+ copy, pminub and avoid SSE 4.2 on Tremont. */
cpu_features->preferred[index_arch_Fast_Rep_String]
- |= (bit_arch_Fast_Rep_String
- | bit_arch_Fast_Unaligned_Load
- | bit_arch_Fast_Unaligned_Copy
- | bit_arch_Prefer_PMINUB_for_stringop
- | bit_arch_Slow_SSE4_2);
+ |= (bit_arch_Fast_Rep_String
+ | bit_arch_Fast_Unaligned_Load
+ | bit_arch_Fast_Unaligned_Copy
+ | bit_arch_Prefer_PMINUB_for_stringop
+ | bit_arch_Slow_SSE4_2);
break;
+ /*
+ Default tuned Knights microarch.
+ case INTEL_KNIGHTS_MILL:
+ */
+
+ /*
+ Default tuned atom microarch.
+ case INTEL_ATOM_SIERRAFOREST:
+ case INTEL_ATOM_GRANDRIDGE:
+ */
+
+ /* Bigcore/Default Tuning. */
default:
/* Unknown family 0x06 processors. Assuming this is one
of Core i3/i5/i7 processors if AVX is available. */
if (!CPU_FEATURES_CPU_P (cpu_features, AVX))
break;
/* Fall through. */
-
- case 0x1a:
- case 0x1e:
- case 0x1f:
- case 0x25:
- case 0x2c:
- case 0x2e:
- case 0x2f:
+ case INTEL_BIGCORE_NEHALEM:
+ case INTEL_BIGCORE_WESTMERE:
/* Rep string instructions, unaligned load, unaligned copy,
and pminub are fast on Intel Core i3, i5 and i7. */
cpu_features->preferred[index_arch_Fast_Rep_String]
- |= (bit_arch_Fast_Rep_String
- | bit_arch_Fast_Unaligned_Load
- | bit_arch_Fast_Unaligned_Copy
- | bit_arch_Prefer_PMINUB_for_stringop);
+ |= (bit_arch_Fast_Rep_String
+ | bit_arch_Fast_Unaligned_Load
+ | bit_arch_Fast_Unaligned_Copy
+ | bit_arch_Prefer_PMINUB_for_stringop);
break;
+
+ /*
+ Default tuned Bigcore microarch.
+ case INTEL_BIGCORE_SANDYBRIDGE:
+ case INTEL_BIGCORE_IVYBRIDGE:
+ case INTEL_BIGCORE_HASWELL:
+ case INTEL_BIGCORE_BROADWELL:
+ case INTEL_BIGCORE_SKYLAKE:
+ case INTEL_BIGCORE_KABYLAKE:
+ case INTEL_BIGCORE_COMETLAKE:
+ case INTEL_BIGCORE_SKYLAKE_AVX512:
+ case INTEL_BIGCORE_CANNONLAKE:
+ case INTEL_BIGCORE_ICELAKE:
+ case INTEL_BIGCORE_TIGERLAKE:
+ case INTEL_BIGCORE_ROCKETLAKE:
+ case INTEL_BIGCORE_RAPTORLAKE:
+ case INTEL_BIGCORE_METEORLAKE:
+ case INTEL_BIGCORE_LUNARLAKE:
+ case INTEL_BIGCORE_ARROWLAKE:
+ case INTEL_BIGCORE_SAPPHIRERAPIDS:
+ case INTEL_BIGCORE_EMERALDRAPIDS:
+ case INTEL_BIGCORE_GRANITERAPIDS:
+ */
+
+ /*
+ Default tuned Mixed (bigcore + atom SOC).
+ case INTEL_MIXED_LAKEFIELD:
+ case INTEL_MIXED_ALDERLAKE:
+ */
}
- /* Disable TSX on some processors to avoid TSX on kernels that
- weren't updated with the latest microcode package (which
- disables broken feature by default). */
- switch (model)
+ /* Disable TSX on some processors to avoid TSX on kernels that
+ weren't updated with the latest microcode package (which
+ disables broken feature by default). */
+ switch (microarch)
{
- case 0x55:
+ case INTEL_BIGCORE_SKYLAKE_AVX512:
+ /* 0x55 (Skylake-avx512) && stepping <= 5 disable TSX. */
if (stepping <= 5)
goto disable_tsx;
break;
- case 0x8e:
- /* NB: Although the errata documents that for model == 0x8e,
- only 0xb stepping or lower are impacted, the intention of
- the errata was to disable TSX on all client processors on
- all steppings. Include 0xc stepping which is an Intel
- Core i7-8665U, a client mobile processor. */
- case 0x9e:
+
+ case INTEL_BIGCORE_KABYLAKE:
+ /* NB: Although the errata documents that for model == 0x8e
+ (kabylake skylake client), only 0xb stepping or lower are
+ impacted, the intention of the errata was to disable TSX on
+ all client processors on all steppings. Include 0xc
+ stepping which is an Intel Core i7-8665U, a client mobile
+ processor. */
if (stepping > 0xc)
break;
/* Fall through. */
- case 0x4e:
- case 0x5e:
- {
+ case INTEL_BIGCORE_SKYLAKE:
/* Disable Intel TSX and enable RTM_ALWAYS_ABORT for
processors listed in:
https://www.intel.com/content/www/us/en/support/articles/000059422/processors.html
*/
-disable_tsx:
+ disable_tsx:
CPU_FEATURE_UNSET (cpu_features, HLE);
CPU_FEATURE_UNSET (cpu_features, RTM);
CPU_FEATURE_SET (cpu_features, RTM_ALWAYS_ABORT);
- }
- break;
- case 0x3f:
- /* Xeon E7 v3 with stepping >= 4 has working TSX. */
- if (stepping >= 4)
break;
- /* Fall through. */
- case 0x3c:
- case 0x45:
- case 0x46:
- /* Disable Intel TSX on Haswell processors (except Xeon E7 v3
- with stepping >= 4) to avoid TSX on kernels that weren't
- updated with the latest microcode package (which disables
- broken feature by default). */
- CPU_FEATURE_UNSET (cpu_features, RTM);
- break;
+
+ case INTEL_BIGCORE_HASWELL:
+ /* Xeon E7 v3 (model == 0x3f) with stepping >= 4 has working
+ TSX. Haswell also include other model numbers that have
+ working TSX. */
+ if (model == 0x3f && stepping >= 4)
+ break;
+
+ CPU_FEATURE_UNSET (cpu_features, RTM);
+ break;
}
}
--
2.39.3

178
glibc-rh2213907-4.patch Normal file
View File

@ -0,0 +1,178 @@
From 180897c161a171d8ef0faee1c6c9fd6b57d8b13b Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Wed, 7 Jun 2023 13:18:03 -0500
Subject: [PATCH] x86: Make the divisor in setting `non_temporal_threshold` cpu
specific
Content-type: text/plain; charset=UTF-8
Different systems prefer a different divisors.
From benchmarks[1] so far the following divisors have been found:
ICX : 2
SKX : 2
BWD : 8
For Intel, we are generalizing that BWD and older prefers 8 as a
divisor, and SKL and newer prefers 2. This number can be further tuned
as benchmarks are run.
[1]: https://github.com/goldsteinn/memcpy-nt-benchmarks
Reviewed-by: DJ Delorie <dj@redhat.com>
---
sysdeps/x86/cpu-features.c | 31 ++++++++++++++++++++---------
sysdeps/x86/dl-cacheinfo.h | 32 ++++++++++++++++++------------
sysdeps/x86/dl-diagnostics-cpu.c | 11 ++++++----
sysdeps/x86/include/cpu-features.h | 3 +++
4 files changed, 51 insertions(+), 26 deletions(-)
[DJ - edited for ABI compatibility]
diff -rup a/sysdeps/x86/cpu-features.c b/sysdeps/x86/cpu-features.c
--- a/sysdeps/x86/cpu-features.c 2023-07-26 17:56:19.679300711 -0400
+++ b/sysdeps/x86/cpu-features.c 2023-07-28 15:27:00.336324265 -0400
@@ -35,6 +35,9 @@ extern void TUNABLE_CALLBACK (set_x86_sh
# endif
#endif
+unsigned long int __rtld_global_ro_cachesize_non_temporal_divisor
+ attribute_hidden;
+
#if CET_ENABLED
# include <dl-cet.h>
#endif
@@ -614,6 +617,7 @@ init_cpu_features (struct cpu_features *
unsigned int stepping = 0;
enum cpu_features_kind kind;
+ __rtld_global_ro_cachesize_non_temporal_divisor = 4;
#if !HAS_CPUID
if (__get_cpuid_max (0, 0) == 0)
{
@@ -694,13 +698,13 @@ init_cpu_features (struct cpu_features *
/* Bigcore/Default Tuning. */
default:
+ default_tuning:
/* Unknown family 0x06 processors. Assuming this is one
of Core i3/i5/i7 processors if AVX is available. */
if (!CPU_FEATURES_CPU_P (cpu_features, AVX))
break;
- /* Fall through. */
- case INTEL_BIGCORE_NEHALEM:
- case INTEL_BIGCORE_WESTMERE:
+
+ enable_modern_features:
/* Rep string instructions, unaligned load, unaligned copy,
and pminub are fast on Intel Core i3, i5 and i7. */
cpu_features->preferred[index_arch_Fast_Rep_String]
@@ -710,12 +714,23 @@ init_cpu_features (struct cpu_features *
| bit_arch_Prefer_PMINUB_for_stringop);
break;
- /*
- Default tuned Bigcore microarch.
+ case INTEL_BIGCORE_NEHALEM:
+ case INTEL_BIGCORE_WESTMERE:
+ /* Older CPUs prefer non-temporal stores at lower threshold. */
+ __rtld_global_ro_cachesize_non_temporal_divisor = 8;
+ goto enable_modern_features;
+
+ /* Older Bigcore microarch (smaller non-temporal store
+ threshold). */
case INTEL_BIGCORE_SANDYBRIDGE:
case INTEL_BIGCORE_IVYBRIDGE:
case INTEL_BIGCORE_HASWELL:
case INTEL_BIGCORE_BROADWELL:
+ __rtld_global_ro_cachesize_non_temporal_divisor = 8;
+ goto default_tuning;
+
+ /* Newer Bigcore microarch (larger non-temporal store
+ threshold). */
case INTEL_BIGCORE_SKYLAKE:
case INTEL_BIGCORE_KABYLAKE:
case INTEL_BIGCORE_COMETLAKE:
@@ -731,13 +746,14 @@ init_cpu_features (struct cpu_features *
case INTEL_BIGCORE_SAPPHIRERAPIDS:
case INTEL_BIGCORE_EMERALDRAPIDS:
case INTEL_BIGCORE_GRANITERAPIDS:
- */
+ __rtld_global_ro_cachesize_non_temporal_divisor = 2;
+ goto default_tuning;
- /*
- Default tuned Mixed (bigcore + atom SOC).
+ /* Default tuned Mixed (bigcore + atom SOC). */
case INTEL_MIXED_LAKEFIELD:
case INTEL_MIXED_ALDERLAKE:
- */
+ __rtld_global_ro_cachesize_non_temporal_divisor = 2;
+ goto default_tuning;
}
/* Disable TSX on some processors to avoid TSX on kernels that
diff -rup a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
--- a/sysdeps/x86/dl-cacheinfo.h 2023-07-26 17:56:18.662261475 -0400
+++ b/sysdeps/x86/dl-cacheinfo.h 2023-07-26 17:56:20.756342261 -0400
@@ -744,19 +744,25 @@ dl_init_cacheinfo (struct cpu_features *
cpu_features->level3_cache_linesize = level3_cache_linesize;
cpu_features->level4_cache_size = level4_cache_size;
- /* The default setting for the non_temporal threshold is 1/4 of size
- of the chip's cache. For most Intel and AMD processors with an
- initial release date between 2017 and 2023, a thread's typical
- share of the cache is from 18-64MB. Using the 1/4 L3 is meant to
- estimate the point where non-temporal stores begin out-competing
- REP MOVSB. As well the point where the fact that non-temporal
- stores are forced back to main memory would already occurred to the
- majority of the lines in the copy. Note, concerns about the
- entire L3 cache being evicted by the copy are mostly alleviated
- by the fact that modern HW detects streaming patterns and
- provides proper LRU hints so that the maximum thrashing
- capped at 1/associativity. */
- unsigned long int non_temporal_threshold = shared / 4;
+ unsigned long int cachesize_non_temporal_divisor
+ = __rtld_global_ro_cachesize_non_temporal_divisor;
+ if (cachesize_non_temporal_divisor <= 0)
+ cachesize_non_temporal_divisor = 4;
+
+ /* The default setting for the non_temporal threshold is [1/8, 1/2] of size
+ of the chip's cache (depending on `cachesize_non_temporal_divisor` which
+ is microarch specific. The defeault is 1/4). For most Intel and AMD
+ processors with an initial release date between 2017 and 2023, a thread's
+ typical share of the cache is from 18-64MB. Using a reasonable size
+ fraction of L3 is meant to estimate the point where non-temporal stores
+ begin out-competing REP MOVSB. As well the point where the fact that
+ non-temporal stores are forced back to main memory would already occurred
+ to the majority of the lines in the copy. Note, concerns about the entire
+ L3 cache being evicted by the copy are mostly alleviated by the fact that
+ modern HW detects streaming patterns and provides proper LRU hints so that
+ the maximum thrashing capped at 1/associativity. */
+ unsigned long int non_temporal_threshold
+ = shared / cachesize_non_temporal_divisor;
/* 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
diff -rup a/sysdeps/x86/dl-diagnostics-cpu.c b/sysdeps/x86/dl-diagnostics-cpu.c
--- a/sysdeps/x86/dl-diagnostics-cpu.c 2021-08-01 21:33:43.000000000 -0400
+++ b/sysdeps/x86/dl-diagnostics-cpu.c 2023-07-26 17:56:20.761342454 -0400
@@ -117,4 +117,6 @@ _dl_diagnostics_cpu (void)
+ sizeof (cpu_features->level4_cache_size)
== sizeof (*cpu_features),
"last cpu_features field has been printed");
+ print_cpu_features_value ("cachesize_non_temporal_divisor",
+ __rtld_global_ro_cachesize_non_temporal_divisor);
}
diff -rup a/sysdeps/x86/include/cpu-features.h b/sysdeps/x86/include/cpu-features.h
--- a/sysdeps/x86/include/cpu-features.h 2021-08-01 21:33:43.000000000 -0400
+++ b/sysdeps/x86/include/cpu-features.h 2023-07-27 13:51:52.081494751 -0400
@@ -919,6 +919,10 @@ struct cpu_features
unsigned long int level4_cache_size;
};
+/* When no user non_temporal_threshold is specified. We default to
+ cachesize / cachesize_non_temporal_divisor. */
+extern unsigned long int __rtld_global_ro_cachesize_non_temporal_divisor;
+
/* Get a pointer to the CPU features structure. */
extern const struct cpu_features *_dl_x86_get_cpu_features (void)
__attribute__ ((pure));

49
glibc-rh2213907-5.patch Normal file
View File

@ -0,0 +1,49 @@
From 47f747217811db35854ea06741be3685e8bbd44d Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Mon, 17 Jul 2023 23:14:33 -0500
Subject: [PATCH] x86: Fix slight bug in `shared_per_thread` cache size
calculation.
Content-type: text/plain; charset=UTF-8
After:
```
commit af992e7abdc9049714da76cae1e5e18bc4838fb8
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Wed Jun 7 13:18:01 2023 -0500
x86: Increase `non_temporal_threshold` to roughly `sizeof_L3 / 4`
```
Split `shared` (cumulative cache size) from `shared_per_thread` (cache
size per socket), the `shared_per_thread` *can* be slightly off from
the previous calculation.
Previously we added `core` even if `threads_l2` was invalid, and only
used `threads_l2` to divide `core` if it was present. The changed
version only included `core` if `threads_l2` was valid.
This change restores the old behavior if `threads_l2` is invalid by
adding the entire value of `core`.
Reviewed-by: DJ Delorie <dj@redhat.com>
---
sysdeps/x86/dl-cacheinfo.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
index c98fa57a7b..43be2c1229 100644
--- a/sysdeps/x86/dl-cacheinfo.h
+++ b/sysdeps/x86/dl-cacheinfo.h
@@ -614,8 +614,8 @@ get_common_cache_info (long int *shared_ptr, long int * shared_per_thread_ptr, u
/* Account for non-inclusive L2 and L3 caches. */
if (!inclusive_cache)
{
- if (threads_l2 > 0)
- shared_per_thread += core / threads_l2;
+ long int core_per_thread = threads_l2 > 0 ? (core / threads_l2) : core;
+ shared_per_thread += core_per_thread;
shared += core;
}
--
2.39.3

55
glibc-rh2213907-6.patch Normal file
View File

@ -0,0 +1,55 @@
From 8b9a0af8ca012217bf90d1dc0694f85b49ae09da Mon Sep 17 00:00:00 2001
From: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Tue, 18 Jul 2023 10:27:59 -0500
Subject: [PATCH] [PATCH v1] x86: Use `3/4*sizeof(per-thread-L3)` as low bound
for NT threshold.
Content-type: text/plain; charset=UTF-8
On some machines we end up with incomplete cache information. This can
make the new calculation of `sizeof(total-L3)/custom-divisor` end up
lower than intended (and lower than the prior value). So reintroduce
the old bound as a lower bound to avoid potentially regressing code
where we don't have complete information to make the decision.
Reviewed-by: DJ Delorie <dj@redhat.com>
---
sysdeps/x86/dl-cacheinfo.h | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
[diff rebased by DJ]
diff -rup a/sysdeps/x86/dl-cacheinfo.h b/sysdeps/x86/dl-cacheinfo.h
--- a/sysdeps/x86/dl-cacheinfo.h 2023-07-25 00:38:43.343986368 -0400
+++ b/sysdeps/x86/dl-cacheinfo.h 2023-07-25 00:38:44.336025100 -0400
@@ -751,8 +751,8 @@ dl_init_cacheinfo (struct cpu_features *
/* The default setting for the non_temporal threshold is [1/8, 1/2] of size
of the chip's cache (depending on `cachesize_non_temporal_divisor` which
- is microarch specific. The defeault is 1/4). For most Intel and AMD
- processors with an initial release date between 2017 and 2023, a thread's
+ is microarch specific. The default is 1/4). For most Intel processors
+ with an initial release date between 2017 and 2023, a thread's
typical share of the cache is from 18-64MB. Using a reasonable size
fraction of L3 is meant to estimate the point where non-temporal stores
begin out-competing REP MOVSB. As well the point where the fact that
@@ -763,12 +763,21 @@ dl_init_cacheinfo (struct cpu_features *
the maximum thrashing capped at 1/associativity. */
unsigned long int non_temporal_threshold
= shared / cachesize_non_temporal_divisor;
+
+ /* If the computed non_temporal_threshold <= 3/4 * per-thread L3, we most
+ likely have incorrect/incomplete cache info in which case, default to
+ 3/4 * per-thread L3 to avoid regressions. */
+ unsigned long int non_temporal_threshold_lowbound
+ = shared_per_thread * 3 / 4;
+ if (non_temporal_threshold < non_temporal_threshold_lowbound)
+ non_temporal_threshold = non_temporal_threshold_lowbound;
+
/* 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))
- non_temporal_threshold = shared_per_thread * 3 / 4;
+ 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
if that operation cannot overflow. Minimum of 0x4040 (16448) because the

11
glibc.spec
View File

@ -155,7 +155,7 @@ end \
Summary: The GNU libc libraries
Name: glibc
Version: %{glibcversion}
Release: 78%{?dist}
Release: 79%{?dist}
# In general, GPLv2+ is used by programs, LGPLv2+ is used for
# libraries.
@ -737,6 +737,12 @@ Patch496: glibc-rh2224349.patch
Patch497: glibc-rh2224289-3.patch
Patch498: glibc-rh2224504-1.patch
Patch499: glibc-rh2224504-2.patch
Patch500: glibc-rh2213907-1.patch
Patch501: glibc-rh2213907-2.patch
Patch502: glibc-rh2213907-3.patch
Patch503: glibc-rh2213907-4.patch
Patch504: glibc-rh2213907-5.patch
Patch505: glibc-rh2213907-6.patch
##############################################################################
# Continued list of core "glibc" package information:
@ -2894,6 +2900,9 @@ update_gconv_modules_cache ()
%endif
%changelog
* Tue Aug 8 2023 DJ Delorie <dj@redhat.com> - 2.34-79
- Fix temporal threshold calculations (#2213907)
* Fri Aug 4 2023 Florian Weimer <fweimer@redhat.com> - 2.34-78
- Ignore symbolic link change on /etc/nsswitch.conf (#2229156)