[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20240620232430.GB439156@cmpxchg.org>
Date: Thu, 20 Jun 2024 19:24:30 -0400
From: Johannes Weiner <hannes@...xchg.org>
To: John Stultz <jstultz@...gle.com>
Cc: LKML <linux-kernel@...r.kernel.org>, Ingo Molnar <mingo@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Juri Lelli <juri.lelli@...hat.com>,
Vincent Guittot <vincent.guittot@...aro.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>,
Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
Daniel Bristot de Oliveira <bristot@...hat.com>,
Valentin Schneider <vschneid@...hat.com>,
Suren Baghdasaryan <surenb@...gle.com>,
Chengming Zhou <zhouchengming@...edance.com>,
Thomas Gleixner <tglx@...utronix.de>,
Frederic Weisbecker <frederic@...nel.org>,
Qais Yousef <qyousef@...alina.io>,
Joel Fernandes <joel@...lfernandes.org>, kernel-team@...roid.com,
Chengming Zhou <chengming.zhou@...ux.dev>,
Jimmy Shiu <jimmyshiu@...gle.com>
Subject: Re: [PATCH v2] sched: Move psi_account_irqtime() out of
update_rq_clock_task() hotpath
On Thu, Jun 20, 2024 at 10:32:57AM -0700, John Stultz wrote:
> It was reported that in moving to 6.1, a larger then 10%
> regression was seen in the performance of
> clock_gettime(CLOCK_THREAD_CPUTIME_ID,...).
>
> Using a simple reproducer, I found:
> 5.10:
> 100000000 calls in 24345994193 ns => 243.460 ns per call
> 100000000 calls in 24288172050 ns => 242.882 ns per call
> 100000000 calls in 24289135225 ns => 242.891 ns per call
>
> 6.1:
> 100000000 calls in 28248646742 ns => 282.486 ns per call
> 100000000 calls in 28227055067 ns => 282.271 ns per call
> 100000000 calls in 28177471287 ns => 281.775 ns per call
>
> The cause of this was finally narrowed down to the addition of
> psi_account_irqtime() in update_rq_clock_task(), in commit
> 52b1364ba0b1 ("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ
> pressure").
>
> In my initial attempt to resolve this, I leaned towards moving
> all accounting work out of the clock_gettime() call path, but it
> wasn't very pretty, so it will have to wait for a later deeper
> rework. Instead, Peter shared this approach:
>
> Rework psi_account_irqtime() to use its own psi_irq_time base
> for accounting, and move it out of the hotpath, calling it
> instead from sched_tick() and __schedule().
>
> In testing this, we found the importance of ensuring
> psi_account_irqtime() is run under the rq_lock, which Johannes
> Weiner helpfully explained, so also add some lockdep annotations
> to make that requirement clear.
>
> With this change the performance is back in-line with 5.10:
> 6.1+fix:
> 100000000 calls in 24297324597 ns => 242.973 ns per call
> 100000000 calls in 24318869234 ns => 243.189 ns per call
> 100000000 calls in 24291564588 ns => 242.916 ns per call
>
> Cc: Ingo Molnar <mingo@...hat.com>
> Cc: Peter Zijlstra <peterz@...radead.org>
> Cc: Juri Lelli <juri.lelli@...hat.com>
> Cc: Vincent Guittot <vincent.guittot@...aro.org>
> Cc: Dietmar Eggemann <dietmar.eggemann@....com>
> Cc: Steven Rostedt <rostedt@...dmis.org>
> Cc: Ben Segall <bsegall@...gle.com>
> Cc: Mel Gorman <mgorman@...e.de>
> Cc: Daniel Bristot de Oliveira <bristot@...hat.com>
> Cc: Valentin Schneider <vschneid@...hat.com>
> Cc: Johannes Weiner <hannes@...xchg.org>
> Cc: Suren Baghdasaryan <surenb@...gle.com>
> Cc: Chengming Zhou <zhouchengming@...edance.com>
> Cc: Thomas Gleixner <tglx@...utronix.de>
> Cc: Frederic Weisbecker <frederic@...nel.org>
> Cc: Qais Yousef <qyousef@...alina.io>
> Cc: Joel Fernandes <joel@...lfernandes.org>
> Cc: kernel-team@...roid.com
> Reviewed-by: Chengming Zhou <chengming.zhou@...ux.dev>
> Reviewed-by: Qais Yousef <qyousef@...alina.io>
> Fixes: 52b1364ba0b1 ("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure")
> Originally-by: Peter Zijlstra <peterz@...radead.org>
> Reported-by: Jimmy Shiu <jimmyshiu@...gle.com>
> Signed-off-by: John Stultz <jstultz@...gle.com>
Excellent.
Acked-by: Johannes Weiner <hannes@...xchg.org>
Powered by blists - more mailing lists