[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <883a5a4f-b34e-689c-2fbd-7bf03db532eb@redhat.com>
Date: Sun, 20 Aug 2023 12:24:33 -0400
From: Waiman Long <longman@...hat.com>
To: Qais Yousef <qyousef@...alina.io>, stable@...r.kernel.org
Cc: Juri Lelli <juri.lelli@...hat.com>, Tejun Heo <tj@...nel.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Peter Zijlstra <peterz@...radead.org>,
Vincent Guittot <vincent.guittot@...aro.org>,
Ingo Molnar <mingo@...nel.org>, Hao Luo <haoluo@...gle.com>,
John Stultz <jstultz@...gle.com>, cgroups@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/6] Backport rework of deadline bandwidth restoration for
5.10.y
On 8/20/23 11:21, Qais Yousef wrote:
> This is a backport of the series that fixes the way deadline bandwidth
> restoration is done which is causing noticeable delay on resume path. It also
> converts the cpuset lock back into a mutex which some users on Android too.
> I lack the details but AFAIU the read/write semaphore was slower on high
> contention.
Note that it was a percpu rwsem before this patch series. It was not a
regular rwsem. Percpu rwsem isn't designed to handle high write lock
contention. A regular rwsem should be similar to mutex in performance
when handling high write lock contention.
Cheers,
Longman
Powered by blists - more mailing lists