[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <5b8d92f6-0d55-4b43-844c-696e71266978@redhat.com>
Date: Fri, 15 Nov 2024 19:40:13 -0500
From: Waiman Long <llong@...hat.com>
To: Michal Koutný <mkoutny@...e.com>
Cc: Tejun Heo <tj@...nel.org>, Johannes Weiner <hannes@...xchg.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>, Valentin Schneider <vschneid@...hat.com>,
Frederic Weisbecker <frederic@...nel.org>, linux-kernel@...r.kernel.org,
cgroups@...r.kernel.org, Costa Shulyupin <cshulyup@...hat.com>,
Daniel Wagner <dwagner@...e.de>
Subject: Re: [PATCH v2 1/2] sched/isolation: Exclude dynamically isolated CPUs
from housekeeping masks
On 11/15/24 10:45 AM, Michal Koutný wrote:
> (Also someone mentioned that this could share lots of code with CPU
> offlining/onlining.)
Yes, that is true. The simplest way to do that is to offline the CPUs to
be isolated, change the housekeeping masks and then online those CPUs
again. That is good for managing a single isolated partition. However,
Daniel had told me that CPU hotplug code could cause latency spike in
existing isolated CPUs. That could be a problem if we have more than one
isolated partitions to manage. So more investigation will be needed in this.
This is still the direction we are going initially, but first we need to
enable dynamic changes to the housekeeping masks first.
Cheers,
Longman
Powered by blists - more mailing lists