From: Thomas Gleixner <tglx@linutronix.de>
To: Yury Norov <yury.norov@gmail.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Gabriele Monaco <gmonaco@redhat.com>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Michael Jeanson <mjeanson@efficios.com>,
Jens Axboe <axboe@kernel.dk>,
"Paul E. McKenney" <paulmck@kernel.org>,
"Gautham R. Shenoy" <gautham.shenoy@amd.com>,
Florian Weimer <fweimer@redhat.com>,
Tim Chen <tim.c.chen@intel.com>,
TCMalloc Team <tcmalloc-eng@google.com>
Subject: Re: [patch 07/19] cpumask: Introduce cpumask_or_weight()
Date: Tue, 21 Oct 2025 22:21:30 +0200 [thread overview]
Message-ID: <87ms5kxayd.ffs@tglx> (raw)
In-Reply-To: <aO_iiKKVyKSlXeF2@yury>
Yury!
On Wed, Oct 15 2025 at 14:06, Yury Norov wrote:
> On Wed, Oct 15, 2025 at 01:41:50PM -0400, Yury Norov wrote:
> Ok, I see now. You want to do a regular cpumask_or(), but return the
> hweight() of the result, instead of a boolean.
>
> The cpumask_or_weight() may be really confused with cpumask_weight_or().
> Can you try considering a different naming? (I am seemingly can't.)
the only thing I came up with was cpumask_or_and_weight(), but that
sounded odd too. cpumask_or_and_calc_weight() perhaps.
> Can you describe the performance impact you've mentioned in the commit
> message in more details?
It's sparing the second loop with the related memory reads. It's about
10-20% faster for a 4k CPU mask (64 iterations) depending on the machine
I test on.
As this is invoked with runqueue lock held, there is definitely a desire
to spare as much cycles as possible.
Thanks,
tglx
next prev parent reply other threads:[~2025-10-21 20:21 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-15 17:29 [patch 00/19] sched: Rewrite MM CID management Thomas Gleixner
2025-10-15 17:29 ` [patch 01/19] sched/mmcid: Revert the complex " Thomas Gleixner
2025-10-15 17:29 ` [patch 02/19] sched/mmcid: Use proper data structures Thomas Gleixner
2025-10-15 17:29 ` [patch 03/19] sched/mmcid: Cacheline align MM CID storage Thomas Gleixner
2025-10-15 17:29 ` [patch 04/19] sched: Fixup whitespace damage Thomas Gleixner
2025-10-15 17:29 ` [patch 05/19] sched/mmcid: Move scheduler code out of global header Thomas Gleixner
2025-10-15 17:29 ` [patch 06/19] sched/mmcid: Prevent pointless work in mm_update_cpus_allowed() Thomas Gleixner
2025-10-17 11:12 ` Peter Zijlstra
2025-10-17 12:49 ` Thomas Gleixner
2025-10-17 17:58 ` Peter Zijlstra
2025-10-17 18:19 ` Peter Zijlstra
2025-10-19 20:32 ` Thomas Gleixner
2025-10-20 8:22 ` Peter Zijlstra
2025-10-21 18:25 ` Thomas Gleixner
2025-10-15 17:29 ` [patch 07/19] cpumask: Introduce cpumask_or_weight() Thomas Gleixner
2025-10-15 17:41 ` Yury Norov
2025-10-15 18:06 ` Yury Norov
2025-10-21 20:21 ` Thomas Gleixner [this message]
2025-10-21 19:34 ` Thomas Gleixner
2025-10-15 17:29 ` [patch 08/19] sched/mmcid: Use cpumask_or_weight() Thomas Gleixner
2025-10-15 17:29 ` [patch 09/19] sched/mmcid: Convert mm CID mask to a bitmap Thomas Gleixner
2025-10-15 17:29 ` [patch 10/19] signal: Move MMCID exit out of sighand lock Thomas Gleixner
2025-10-15 17:29 ` [patch 11/19] sched/mmcid: Move initialization out of line Thomas Gleixner
2025-10-15 17:29 ` [patch 12/19] sched/mmcid: Provide precomputed maximal value Thomas Gleixner
2025-10-15 17:29 ` [patch 13/19] sched/mmcid: Serialize sched_mm_cid_fork()/exit() with a mutex Thomas Gleixner
2025-10-15 17:29 ` [patch 14/19] sched/mmcid: Introduce per task/CPU ownership infrastrcuture Thomas Gleixner
2025-10-15 17:29 ` [patch 15/19] sched/mmcid: Provide new scheduler CID mechanism Thomas Gleixner
2025-10-15 17:29 ` [patch 16/19] sched/mmcid: Provide CID ownership mode fixup functions Thomas Gleixner
2025-10-20 6:34 ` Thomas Gleixner
2025-10-20 9:13 ` Peter Zijlstra
2025-10-20 9:16 ` Peter Zijlstra
2025-10-20 9:27 ` Peter Zijlstra
2025-10-21 18:27 ` Thomas Gleixner
2025-10-15 17:29 ` [patch 17/19] irqwork: Move data struct to a types header Thomas Gleixner
2025-10-15 17:29 ` [patch 18/19] sched/mmcid: Implement deferred mode change Thomas Gleixner
2025-10-15 17:30 ` [patch 19/19] sched/mmcid: Switch over to the new mechanism Thomas Gleixner
2025-10-17 7:09 ` [patch 00/19] sched: Rewrite MM CID management Thomas Gleixner
2025-10-17 11:31 ` Florian Weimer
2025-10-17 12:56 ` Thomas Gleixner
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=87ms5kxayd.ffs@tglx \
--to=tglx@linutronix.de \
--cc=axboe@kernel.dk \
--cc=fweimer@redhat.com \
--cc=gautham.shenoy@amd.com \
--cc=gmonaco@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mjeanson@efficios.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=tcmalloc-eng@google.com \
--cc=tim.c.chen@intel.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