[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZerlUU4kAv580rLh@gmail.com>
Date: Fri, 8 Mar 2024 11:15:45 +0100
From: Ingo Molnar <mingo@...nel.org>
To: Vincent Guittot <vincent.guittot@...aro.org>
Cc: Valentin Schneider <vschneid@...hat.com>, linux-kernel@...r.kernel.org,
Shrikanth Hegde <sshegde@...ux.ibm.com>,
Peter Zijlstra <peterz@...radead.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [PATCH 6/9] sched/balancing: Update run_rebalance_domains()
comments
* Vincent Guittot <vincent.guittot@...aro.org> wrote:
> > """
> > The run_rebalance_domains() softirq handler is triggered via SCHED_SOFTIRQ
> > from two places:
> >
> > - directly from trigger_load_balance() in scheduler_tick(), for periodic
> > load balance
> >
> > - indirectly from kick_ilb() (invoked down the scheduler_tick() too), which
> > issues an SMP cross-call to nohz_csd_func() which will itself raise the
> > softirq, for NOHZ idle balancing.
>
> I'm not sure that we should provide too many details of the path as
> this might change in the future. What about the below ?
>
> - directly from the local scheduler_tick() for periodic load balance
>
> - indirectly from a remote scheduler_tick() for NOHZ idle balancing
> through the SMP cross-call nohz_csd_func()
Okay - I updated it to:
/*
* This softirq handler is triggered via SCHED_SOFTIRQ from two places:
*
* - directly from the local scheduler_tick() for periodic load balancing
*
* - indirectly from a remote scheduler_tick() for NOHZ idle balancing
* through the SMP cross-call nohz_csd_func()
*/
static __latent_entropy void run_rebalance_domains(struct softirq_action *h)
Does this work with everyone?
Thanks,
Ingo
Powered by blists - more mailing lists