[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220330143748.37f4804c@gandalf.local.home>
Date: Wed, 30 Mar 2022 14:37:48 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Chengming Zhou <zhouchengming@...edance.com>, mingo@...hat.com,
juri.lelli@...hat.com, vincent.guittot@...aro.org,
dietmar.eggemann@....com, bsegall@...gle.com, mgorman@...e.de,
bristot@...hat.com, linux-kernel@...r.kernel.org,
duanxiongchun@...edance.com, songmuchun@...edance.com,
Frederic Weisbecker <fweisbec@...il.com>
Subject: Re: [External] Re: [PATCH] sched/fair: fix broken bandwidth control
with nohz_full
On Wed, 30 Mar 2022 20:23:27 +0200
Peter Zijlstra <peterz@...radead.org> wrote:
> > Chengming brought up VMs. That's a case to want to control the bandwidth,
> > but also not interrupt them with timer interrupts when they are running as
> > the top priority task on a CPU.
>
> It's CFS, there is nothing top priority about that.
If there's only one task running on a CPU, even with CFS, why do we need a
tick?
Look, we have a host that is doing nothing but running VMs. Could be a
single VM. The host doesn't even have disk, it just runs on from initrd,
and the user is actually just logging into a guest.
Why should the host be triggering a tick when running this VM, which is the
only thing that is running on the host? Yeah, it's CFS, or does CFS have to
have a tick? even when there's nothing else on the CPU?
-- Steve
Powered by blists - more mailing lists