niansa-misc/linux
1
0
Fork 0
mirror of synced 2025-03-06 20:59:54 +01:00
Code Issues Projects Releases Packages Wiki Activity
Linux kernel source tree
1309170 commits 2 branches 875 tags 3.5 GiB
C 97.5%
Assembly 1%
Shell 0.6%
Python 0.3%
Makefile 0.3%
Find a file
Mathieu Desnoyers 7e019dcc47 sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads 
commit 223baf9d17 ("sched: Fix performance regression introduced by mm_cid")
introduced a per-mm/cpu current concurrency id (mm_cid), which keeps
a reference to the concurrency id allocated for each CPU. This reference
expires shortly after a 100ms delay.

These per-CPU references keep the per-mm-cid data cache-local in
situations where threads are running at least once on each CPU within
each 100ms window, thus keeping the per-cpu reference alive.

However, intermittent workloads behaving in bursts spaced by more than
100ms on each CPU exhibit bad cache locality and degraded performance
compared to purely per-cpu data indexing, because concurrency IDs are
allocated over various CPUs and cores, therefore losing cache locality
of the associated data.

Introduce the following changes to improve per-mm-cid cache locality:

- Add a "recent_cid" field to the per-mm/cpu mm_cid structure to keep
  track of which mm_cid value was last used, and use it as a hint to
  attempt re-allocating the same concurrency ID the next time this
  mm/cpu needs to allocate a concurrency ID,

- Add a per-mm CPUs allowed mask, which keeps track of the union of
  CPUs allowed for all threads belonging to this mm. This cpumask is
  only set during the lifetime of the mm, never cleared, so it
  represents the union of all the CPUs allowed since the beginning of
  the mm lifetime (note that the mm_cpumask() is really arch-specific
  and tailored to the TLB flush needs, and is thus _not_ a viable
  approach for this),

- Add a per-mm nr_cpus_allowed to keep track of the weight of the
  per-mm CPUs allowed mask (for fast access),

- Add a per-mm max_nr_cid to keep track of the highest number of
  concurrency IDs allocated for the mm. This is used for expanding the
  concurrency ID allocation within the upper bound defined by:

    min(mm->nr_cpus_allowed, mm->mm_users)

  When the next unused CID value reaches this threshold, stop trying
  to expand the cid allocation and use the first available cid value
  instead.

  Spreading allocation to use all the cid values within the range

    [ 0, min(mm->nr_cpus_allowed, mm->mm_users) - 1 ]

  improves cache locality while preserving mm_cid compactness within the
  expected user limits,

- In __mm_cid_try_get, only return cid values within the range
  [ 0, mm->nr_cpus_allowed ] rather than [ 0, nr_cpu_ids ]. This
  prevents allocating cids above the number of allowed cpus in
  rare scenarios where cid allocation races with a concurrent
  remote-clear of the per-mm/cpu cid. This improvement is made
  possible by the addition of the per-mm CPUs allowed mask,

- In sched_mm_cid_migrate_to, use mm->nr_cpus_allowed rather than
  t->nr_cpus_allowed. This criterion was really meant to compare
  the number of mm->mm_users to the number of CPUs allowed for the
  entire mm. Therefore, the prior comparison worked fine when all
  threads shared the same CPUs allowed mask, but not so much in
  scenarios where those threads have different masks (e.g. each
  thread pinned to a single CPU). This improvement is made
  possible by the addition of the per-mm CPUs allowed mask.

* Benchmarks

Each thread increments 16kB worth of 8-bit integers in bursts, with
a configurable delay between each thread's execution. Each thread run
one after the other (no threads run concurrently). The order of
thread execution in the sequence is random. The thread execution
sequence begins again after all threads have executed. The 16kB areas
are allocated with rseq_mempool and indexed by either cpu_id, mm_cid
(not cache-local), or cache-local mm_cid. Each thread is pinned to its
own core.

Testing configurations:

8-core/1-L3:        Use 8 cores within a single L3
24-core/24-L3:      Use 24 cores, 1 core per L3
192-core/24-L3:     Use 192 cores (all cores in the system)
384-thread/24-L3:   Use 384 HW threads (all HW threads in the system)

Intermittent workload delays between threads: 200ms, 10ms.

Hardware:

CPU(s):                   384
  On-line CPU(s) list:    0-383
Vendor ID:                AuthenticAMD
  Model name:             AMD EPYC 9654 96-Core Processor
    Thread(s) per core:   2
    Core(s) per socket:   96
    Socket(s):            2
Caches (sum of all):
  L1d:                    6 MiB (192 instances)
  L1i:                    6 MiB (192 instances)
  L2:                     192 MiB (192 instances)
  L3:                     768 MiB (24 instances)

Each result is an average of 5 test runs. The cache-local speedup
is calculated as: (cache-local mm_cid) / (mm_cid).

Intermittent workload delay: 200ms

                     per-cpu     mm_cid    cache-local mm_cid    cache-local speedup
                         (ns)      (ns)                  (ns)
8-core/1-L3             1374      19289                  1336            14.4x
24-core/24-L3           2423      26721                  1594            16.7x
192-core/24-L3          2291      15826                  2153             7.3x
384-thread/24-L3        1874      13234                  1907             6.9x

Intermittent workload delay: 10ms

                     per-cpu     mm_cid    cache-local mm_cid    cache-local speedup
                         (ns)      (ns)                  (ns)
8-core/1-L3               662       756                   686             1.1x
24-core/24-L3            1378      3648                  1035             3.5x
192-core/24-L3           1439     10833                  1482             7.3x
384-thread/24-L3         1503     10570                  1556             6.8x

[ This deprecates the prior "sched: NUMA-aware per-memory-map concurrency IDs"
  patch series with a simpler and more general approach. ]

[ This patch applies on top of v6.12-rc1. ]

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Marco Elver <elver@google.com>
Link: https://lore.kernel.org/lkml/20240823185946.418340-1-mathieu.desnoyers@efficios.com/
2024-10-14 12:52:40 +02:00
arch ARM64: 2024-10-06 10:53:28 -07:00
block block-6.12-20241004 2024-10-04 10:43:44 -07:00
certs sign-file,extract-cert: use pkcs11 provider for OPENSSL MAJOR >= 3 2024-09-20 19:52:48 +03:00
crypto move asm/unaligned.h to linux/unaligned.h 2024-10-02 17:23:23 -04:00
Documentation platform-drivers-x86 for v6.12-2 2024-10-06 11:11:01 -07:00
drivers platform-drivers-x86 for v6.12-2 2024-10-06 11:11:01 -07:00
fs sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads 2024-10-14 12:52:40 +02:00
include sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads 2024-10-14 12:52:40 +02:00
init cfi: encode cfi normalized integers + kasan/gcov bug in Kconfig 2024-09-26 21:27:27 +02:00
io_uring io_uring/net: harden multishot termination case for recv 2024-09-30 08:26:59 -06:00
ipc struct fd layout change (and conversion to accessor helpers) 2024-09-23 09:35:36 -07:00
kernel sched: Improve cache locality of RSEQ concurrency IDs for intermittent workloads 2024-10-14 12:52:40 +02:00
lib slab fixes for 6.12-rc1 2024-10-04 12:05:39 -07:00
LICENSES LICENSES: add 0BSD license text 2024-09-01 20:43:24 -07:00
mm mm, slab: suppress warnings in test_leak_destroy kunit test 2024-10-02 16:28:46 +02:00
net Including fixes from ieee802154, bluetooth and netfilter. 2024-10-03 09:44:00 -07:00
rust rust: kunit: use C-string literals to clean warning 2024-10-01 23:46:42 +02:00
samples [tree-wide] finally take no_llseek out 2024-09-27 08:18:43 -07:00
scripts kbuild: deb-pkg: Remove blank first line from maint scripts 2024-10-07 02:36:38 +09:00
security hardening fixes for v6.12-rc2 2024-10-05 10:19:14 -07:00
sound sound fixes for 6.12-rc2 2024-10-04 11:29:46 -07:00
tools ARM64: 2024-10-06 10:53:28 -07:00
usr initramfs: shorten cmd_initfs in usr/Makefile 2024-07-16 01:07:52 +09:00
virt sched/fair: Fix external p->on_rq users 2024-10-14 09:14:35 +02:00
.clang-format clang-format: Update with v6.11-rc1's for_each macro list 2024-08-02 13:20:31 +02:00
.cocciconfig scripts: add Linux .cocciconfig for coccinelle 2016-07-22 12:13:39 +02:00
.editorconfig .editorconfig: remove trim_trailing_whitespace option 2024-06-13 16:47:52 +02:00
.get_maintainer.ignore Add Jeff Kirsher to .get_maintainer.ignore 2024-03-08 11:36:54 +00:00
.gitattributes .gitattributes: set diff driver for Rust source code files 2023-05-31 17:48:25 +02:00
.gitignore Kbuild updates for v6.12 2024-09-24 13:02:06 -07:00
.mailmap Summary 2024-09-24 11:08:40 -07:00
.rustfmt.toml rust: add .rustfmt.toml 2022-09-28 09:02:20 +02:00
COPYING COPYING: state that all contributions really are covered by this file 2020-02-10 13:32:20 -08:00
CREDITS MAINTAINERS: Mark powerpc spufs as orphaned 2024-08-19 21:27:56 +10:00
Kbuild Kbuild updates for v6.1 2022-10-10 12:00:45 -07:00
Kconfig kbuild: ensure full rebuild when the compiler is updated 2020-05-12 13:28:33 +09:00
MAINTAINERS platform-drivers-x86 for v6.12-2 2024-10-06 11:11:01 -07:00
Makefile Linux 6.12-rc2 2024-10-06 15:32:27 -07:00
README README: Fix spelling 2024-03-18 03:36:32 -06:00

README 

Linux kernel
============

There are several guides for kernel developers and users. These guides can
be rendered in a number of formats, like HTML and PDF. Please read
Documentation/admin-guide/README.rst first.

In order to build the documentation, use ``make htmldocs`` or
``make pdfdocs``.  The formatted documentation can also be read online at:

    https://www.kernel.org/doc/html/latest/

There are various text files in the Documentation/ subdirectory,
several of them using the reStructuredText markup notation.

Please read the Documentation/process/changes.rst file, as it contains the
requirements for building and running the kernel, and information about
the problems which may result by upgrading your kernel.