mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Yury Norov <yury.norov@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	linux-kernel@vger.kernel.org,
	Valentin Schneider <vschneid@redhat.com>,
	Mel Gorman <mgorman@suse.de>,
	Steven Rostedt <rostedt@goodmis.org>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Ben Segall <bsegall@google.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	Shuah Khan <skhan@linuxfoundation.org>
Subject: Re: [RFC PATCH v1 4/6] sched: NUMA-aware per-memory-map concurrency IDs
Date: Fri, 23 Aug 2024 16:45:38 -0400	[thread overview]
Message-ID: <67dde65a-75fa-4c34-a8a7-02260c394bf2@efficios.com> (raw)
In-Reply-To: <ZsjtuTsKuS1k1RK_@yury-ThinkPad>

On 2024-08-23 22:14, Yury Norov wrote:
> On Fri, Aug 23, 2024 at 02:59:44PM -0400, Mathieu Desnoyers wrote:
>> The issue addressed by this change is the non-locality of NUMA accesses
>> to data structures indexed by concurrency IDs: for example, in a
>> scenario where a process has two threads, and they periodically run one
>> after the other on different NUMA nodes, each will be assigned mm_cid=0.
>> As a consequence, they will end up accessing the same pages, and thus at
>> least one of the threads will need to perform remote NUMA accesses,
>> which is inefficient.
>>
>> That being said, the same issue theoretically exists due to false
>> sharing of cache lines by threads running on after another on different
> 
> running one after another you mean?

Yes, you are correct. I will fix this typo for the next round,

Thanks,

Mathieu

> 
>> cores/CPUs within a single NUMA node, but the extent of the performance
>> impact is lesser than remote NUMA accesses.
>>
>> Solve this by making the rseq concurrency ID (mm_cid) NUMA-aware. On
>> NUMA systems, when a NUMA-aware concurrency ID is observed by user-space
>> to be associated with a NUMA node, guarantee that it never changes NUMA
>> node unless either a kernel-level NUMA configuration change happens, or
>> scheduler migrations end up migrating tasks across NUMA nodes.
>>
>> There is a tradeoff between NUMA locality and compactness of the
>> concurrency ID allocation. Favor compactness over NUMA locality when
>> the scheduler migrates tasks across NUMA nodes, as this does not cause
>> the frequent remote NUMA accesses behavior. This is done by limiting the
>> concurrency ID range to minimum between the number of threads belonging
>> to the process and the number of allowed CPUs.
>>
>> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
>> Cc: Peter Zijlstra <peterz@infradead.org>
>> Cc: Ingo Molnar <mingo@redhat.com>
>> Cc: Valentin Schneider <vschneid@redhat.com>
>> Cc: Mel Gorman <mgorman@suse.de>
>> Cc: Steven Rostedt <rostedt@goodmis.org>
>> Cc: Vincent Guittot <vincent.guittot@linaro.org>
>> Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
>> Cc: Ben Segall <bsegall@google.com>
>> ---
>> Changes since v0:
>> - Rename "notandnot" to "nor".

-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com


  reply	other threads:[~2024-08-23 20:46 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-23 18:59 [RFC PATCH v1 0/6] sched: NUMA-aware " Mathieu Desnoyers
2024-08-23 18:59 ` [RFC PATCH v1 1/6] lib: Clarify comment on top of find_next_andnot_bit Mathieu Desnoyers
2024-08-23 19:20   ` Yury Norov
2024-08-23 18:59 ` [RFC PATCH v1 2/6] lib: Implement find_{first,next,nth}_nor_bit, find_first_andnot_bit Mathieu Desnoyers
2024-08-23 19:19   ` Yury Norov
2024-08-23 20:51     ` Mathieu Desnoyers
2024-08-28 18:24     ` Mathieu Desnoyers
2024-08-28 18:41       ` Yury Norov
2024-08-23 18:59 ` [RFC PATCH v1 3/6] cpumask: Implement cpumask_{first,next}_{nor,andnot} Mathieu Desnoyers
2024-08-23 19:18   ` Yury Norov
2024-08-23 20:49     ` Mathieu Desnoyers
2024-08-23 18:59 ` [RFC PATCH v1 4/6] sched: NUMA-aware per-memory-map concurrency IDs Mathieu Desnoyers
2024-08-23 20:14   ` Yury Norov
2024-08-23 20:45     ` Mathieu Desnoyers [this message]
2024-08-23 18:59 ` [RFC PATCH v1 5/6] selftests/rseq: x86: Implement rseq_load_u32_u32 Mathieu Desnoyers
2024-08-23 18:59 ` [RFC PATCH v1 6/6] selftests/rseq: Implement NUMA node id vs mm_cid invariant test Mathieu Desnoyers

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=67dde65a-75fa-4c34-a8a7-02260c394bf2@efficios.com \
    --to=mathieu.desnoyers@efficios.com \
    --cc=bsegall@google.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=skhan@linuxfoundation.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.com \
    --cc=yury.norov@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®