From: Peter Zijlstra <peterz@infradead.org>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: linux-kernel@vger.kernel.org,
"André Almeida" <andrealmeid@igalia.com>,
"Darren Hart" <dvhart@infradead.org>,
"Davidlohr Bueso" <dave@stgolabs.net>,
"Ingo Molnar" <mingo@redhat.com>,
"Juri Lelli" <juri.lelli@redhat.com>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Valentin Schneider" <vschneid@redhat.com>,
"Waiman Long" <longman@redhat.com>
Subject: Re: [PATCH v8 03/15] futex: Add basic infrastructure for local task local hash.
Date: Mon, 3 Feb 2025 15:27:43 +0100 [thread overview]
Message-ID: <20250203142743.GI7145@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20250203135935.440018-4-bigeasy@linutronix.de>
On Mon, Feb 03, 2025 at 02:59:23PM +0100, Sebastian Andrzej Siewior wrote:
> The futex hashmap is system wide and shared by random tasks. Each slot
> is hashed based on its address and VMA. Due to randomized VMAs (and
> memory allocations) the same logical lock (pointer) can end up in a
> different hash bucket on each invocation of the application. This in
> turn means that different applications may share a hash bucket on the
> first invocation but not on the second an it is not always clear which
> applications will be involved. This can result in high latency's to
> acquire the futex_hash_bucket::lock especially if the lock owner is
> limited to a CPU and not be effectively PI boosted.
>
> Introduce a task local hash map. The hashmap can be allocated via
> prctl(PR_FUTEX_HASH, PR_FUTEX_HASH_SET_SLOTS, 0)
>
> The `0' argument allocates a default number of 16 slots, a higher number
> can be specified if desired. The current upper limit is 131072.
Hmm, I would expect 0 to disable the local thing.
Now, I realize this is somewhat tricky, since there might be futexes
inside it. But mapping 0 to some default value seems.. well, odd.
next prev parent reply other threads:[~2025-02-03 14:27 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-03 13:59 [PATCH v8 00/15] futex: Add support task local hash maps Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 01/15] rcuref: Avoid false positive "imbalanced put" report Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 02/15] futex: Create helper function to initialize a hash slot Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 03/15] futex: Add basic infrastructure for local task local hash Sebastian Andrzej Siewior
2025-02-03 14:27 ` Peter Zijlstra [this message]
2025-02-03 15:51 ` Sebastian Andrzej Siewior
2025-02-04 10:34 ` Peter Zijlstra
2025-02-05 8:39 ` Sebastian Andrzej Siewior
2025-02-07 9:41 ` Juri Lelli
2025-02-07 11:00 ` Sebastian Andrzej Siewior
2025-02-07 11:06 ` Peter Zijlstra
2025-02-07 14:47 ` Juri Lelli
2025-02-03 14:29 ` Peter Zijlstra
2025-02-03 14:41 ` Peter Zijlstra
2025-02-03 15:39 ` Peter Zijlstra
2025-02-03 15:52 ` Sebastian Andrzej Siewior
2025-02-04 8:41 ` Peter Zijlstra
2025-02-04 9:28 ` Thomas Gleixner
2025-02-03 13:59 ` [PATCH v8 04/15] futex: Allow automatic allocation of process wide futex hash Sebastian Andrzej Siewior
2025-02-03 14:36 ` Peter Zijlstra
2025-02-03 15:54 ` Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 05/15] futex: Hash only the address for private futexes Sebastian Andrzej Siewior
2025-02-03 14:41 ` Peter Zijlstra
2025-02-03 13:59 ` [PATCH v8 06/15] futex: Move private hashing into its own function Sebastian Andrzej Siewior
2025-02-04 9:34 ` Peter Zijlstra
2025-02-05 7:51 ` Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 07/15] futex: Decrease the waiter count before the unlock operation Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 08/15] futex: Prepare for reference counting of the process private hash end of operation Sebastian Andrzej Siewior
2025-02-04 9:49 ` Peter Zijlstra
2025-02-05 7:54 ` Sebastian Andrzej Siewior
2025-02-05 9:37 ` Peter Zijlstra
2025-02-03 13:59 ` [PATCH v8 09/15] futex: Re-evaluate the hash bucket after dropping the lock Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 10/15] futex: Introduce futex_get_locked_hb() Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 11/15] futex: Acquire a hash reference in futex_wait_multiple_setup() Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 12/15] futex: Allow to re-allocate the private local hash Sebastian Andrzej Siewior
2025-02-04 11:05 ` Peter Zijlstra
2025-02-05 8:00 ` Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 13/15] futex: Resize local futex hash table based on number of threads Sebastian Andrzej Siewior
2025-02-04 10:21 ` Peter Zijlstra
2025-02-05 8:05 ` Sebastian Andrzej Siewior
2025-02-07 9:07 ` Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 14/15] futex: Use a hashmask instead of hashsize Sebastian Andrzej Siewior
2025-02-03 13:59 ` [PATCH v8 15/15] futex: Avoid allocating new local hash if there is something pending Sebastian Andrzej Siewior
2025-02-04 15:14 ` [PATCH v8 00/15] futex: Add support task local hash maps Peter Zijlstra
2025-02-05 8:46 ` Sebastian Andrzej Siewior
2025-02-05 12:20 ` Sebastian Andrzej Siewior
2025-02-05 12:52 ` Peter Zijlstra
2025-02-05 16:52 ` Sebastian Andrzej Siewior
2025-02-20 15:12 ` Peter Zijlstra
2025-02-20 15:57 ` Sebastian Andrzej Siewior
2025-02-21 16:00 ` Sebastian Andrzej Siewior
2025-02-21 19:21 ` Peter Zijlstra
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=20250203142743.GI7145@noisy.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=andrealmeid@igalia.com \
--cc=bigeasy@linutronix.de \
--cc=dave@stgolabs.net \
--cc=dvhart@infradead.org \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=longman@redhat.com \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=vschneid@redhat.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