linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Juri Lelli <juri.lelli@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: tglx@linutronix.de, mingo@redhat.com, rostedt@goodmis.org,
	cfriesen@nortel.com, oleg@redhat.com, fweisbec@gmail.com,
	darren@dvhart.com, johan.eker@ericsson.com, p.faure@akatech.ch,
	linux-kernel@vger.kernel.org, claudio@evidence.eu.com,
	michael@amarulasolutions.com, fchecconi@gmail.com,
	tommaso.cucinotta@sssup.it, nicola.manica@disi.unitn.it,
	luca.abeni@unitn.it, dhaval.giani@gmail.com, hgu1972@gmail.com,
	paulmck@linux.vnet.ibm.com, raistlin@linux.it,
	insop.song@ericsson.com, liming.wang@windriver.com
Subject: Re: [PATCH 05/16] sched: SCHED_DEADLINE policy implementation.
Date: Mon, 23 Apr 2012 17:34:54 +0200	[thread overview]
Message-ID: <4F95769E.70303@gmail.com> (raw)
In-Reply-To: <1335191104.28150.153.camel@twins>

On 04/23/2012 04:25 PM, Peter Zijlstra wrote:
> On Fri, 2012-04-06 at 09:14 +0200, Juri Lelli wrote:
>> +/*
>> + * This is the bandwidth enforcement timer callback. If here, we know
>> + * a task is not on its dl_rq, since the fact that the timer was running
>> + * means the task is throttled and needs a runtime replenishment.
>> + *
>> + * However, what we actually do depends on the fact the task is active,
>> + * (it is on its rq) or has been removed from there by a call to
>> + * dequeue_task_dl(). In the former case we must issue the runtime
>> + * replenishment and add the task back to the dl_rq; in the latter, we just
>> + * do nothing but clearing dl_throttled, so that runtime and deadline
>> + * updating (and the queueing back to dl_rq) will be done by the
>> + * next call to enqueue_task_dl().
>
> OK, so that comment isn't entirely clear to me, how can that timer still
> be active when the task isn't? You start the timer when you throttle it,
> at that point it cannot in fact dequeue itself anymore.
>
> The only possibility I see is the one mentioned with the dl_task() check
> below, that someone else called sched_setscheduler() on it.
>

Ok, I was also stuck at this point when I first reviewed v3.
Then I convinced myself that, even if probably always true,
the p->on_rq check would prevent weird situations like for
example: by the time I block on a mutex, go to sleep or whatever,
I am throttled, then the dl_timer fires and I'm still !on_rq.
But I didn't see this happening ever actually...

>> + */
>> +static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
>> +{
>> +       unsigned long flags;
>> +       struct sched_dl_entity *dl_se = container_of(timer,
>> +                                                    struct sched_dl_entity,
>> +                                                    dl_timer);
>> +       struct task_struct *p = dl_task_of(dl_se);
>> +       struct rq *rq = task_rq_lock(p,&flags);
>> +
>> +       /*
>> +        * We need to take care of a possible races here. In fact, the
>> +        * task might have changed its scheduling policy to something
>> +        * different from SCHED_DEADLINE (through sched_setscheduler()).
>> +        */
>> +       if (!dl_task(p))
>> +               goto unlock;
>> +
>> +       dl_se->dl_throttled = 0;
>> +       if (p->on_rq) {
>> +               enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
>> +               if (task_has_dl_policy(rq->curr))
>> +                       check_preempt_curr_dl(rq, p, 0);
>> +               else
>> +                       resched_task(rq->curr);
>> +       }
>
> So I can't see how that cannot be true.
>
>> +unlock:
>> +       task_rq_unlock(rq, p,&flags);
>> +
>> +       return HRTIMER_NORESTART;
>> +}

  reply	other threads:[~2012-04-23 15:35 UTC|newest]

Thread overview: 129+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-04-06  7:14 [RFC][PATCH 00/16] sched: SCHED_DEADLINE v4 Juri Lelli
2012-04-06  7:14 ` [PATCH 01/16] sched: add sched_class->task_dead Juri Lelli
2012-04-08 17:49   ` Oleg Nesterov
2012-04-08 18:09     ` Juri Lelli
2012-04-06  7:14 ` [PATCH 02/16] sched: add extended scheduling interface Juri Lelli
2012-04-06  7:14 ` [PATCH 03/16] sched: SCHED_DEADLINE data structures Juri Lelli
2012-04-23  9:08   ` Peter Zijlstra
2012-04-23  9:47     ` Juri Lelli
2012-04-23  9:49       ` Peter Zijlstra
2012-04-23  9:55         ` Juri Lelli
2012-04-23 10:12           ` Peter Zijlstra
2012-04-23  9:13   ` Peter Zijlstra
2012-04-23  9:28     ` Juri Lelli
2012-04-23  9:30   ` Peter Zijlstra
2012-04-23  9:36     ` Juri Lelli
2012-04-23  9:39       ` Peter Zijlstra
2012-04-23  9:34   ` Peter Zijlstra
2012-04-23 10:16     ` Juri Lelli
2012-04-23 10:28       ` Peter Zijlstra
2012-04-23 10:33         ` Juri Lelli
2012-04-06  7:14 ` [PATCH 04/16] sched: SCHED_DEADLINE SMP-related " Juri Lelli
2012-04-06  7:14 ` [PATCH 05/16] sched: SCHED_DEADLINE policy implementation Juri Lelli
2012-04-11  3:06   ` Steven Rostedt
2012-04-11  6:54     ` Juri Lelli
2012-04-11 13:41   ` Steven Rostedt
2012-04-11 13:55     ` Juri Lelli
2012-04-23 10:15   ` Peter Zijlstra
2012-04-23 10:18     ` Juri Lelli
2012-04-23 10:31   ` Peter Zijlstra
2012-04-23 10:37     ` Juri Lelli
2012-04-23 21:25       ` Tommaso Cucinotta
2012-04-23 21:45         ` Peter Zijlstra
2012-04-23 23:25           ` Tommaso Cucinotta
2012-04-24  6:29             ` Dario Faggioli
2012-04-24  6:52               ` Juri Lelli
2012-04-23 11:32   ` Peter Zijlstra
2012-04-23 12:13     ` Juri Lelli
2012-04-23 12:22       ` Peter Zijlstra
2012-04-23 13:37         ` Juri Lelli
2012-04-23 14:01           ` Peter Zijlstra
2012-04-23 11:34   ` Peter Zijlstra
2012-04-23 11:57     ` Juri Lelli
2012-04-23 11:55   ` Peter Zijlstra
2012-04-23 14:43     ` Juri Lelli
2012-04-23 15:11       ` Peter Zijlstra
2012-04-23 21:55     ` Tommaso Cucinotta
2012-04-23 21:58       ` Peter Zijlstra
2012-04-23 23:21         ` Tommaso Cucinotta
2012-04-24  9:50           ` Peter Zijlstra
2012-04-24  1:03         ` Steven Rostedt
2012-04-23 14:11   ` Peter Zijlstra
2012-04-23 14:25   ` Peter Zijlstra
2012-04-23 15:34     ` Juri Lelli [this message]
2012-04-23 14:35   ` Peter Zijlstra
2012-04-23 15:39     ` Juri Lelli
2012-04-23 15:43       ` Peter Zijlstra
2012-04-23 16:41         ` Juri Lelli
     [not found]           ` <4F95D41F.5060700@sssup.it>
2012-04-24  7:21             ` Juri Lelli
2012-04-24  9:00               ` Peter Zijlstra
2012-05-15 10:10         ` Juri Lelli
2012-04-23 15:15   ` Peter Zijlstra
2012-04-23 15:37     ` Juri Lelli
2012-04-06  7:14 ` [PATCH 06/16] sched: SCHED_DEADLINE push and pull logic Juri Lelli
2012-04-06 13:39   ` Hillf Danton
2012-04-06 17:31     ` Juri Lelli
2012-04-07  2:32       ` Hillf Danton
2012-04-07  7:46         ` Dario Faggioli
2012-04-08 20:20         ` Juri Lelli
2012-04-09 12:28           ` Hillf Danton
2012-04-10  8:11             ` Juri Lelli
2012-04-11 15:57               ` Steven Rostedt
2012-04-11 16:00           ` Steven Rostedt
2012-04-11 16:09             ` Juri Lelli
2012-04-11 14:10     ` Steven Rostedt
2012-04-12 12:28       ` Hillf Danton
2012-04-12 12:51         ` Steven Rostedt
2012-04-12 12:56           ` Hillf Danton
2012-04-12 13:35             ` Steven Rostedt
2012-04-12 13:41               ` Hillf Danton
2012-04-11 16:07   ` Steven Rostedt
2012-04-11 16:11     ` Juri Lelli
2012-04-11 16:14   ` Steven Rostedt
2012-04-19 13:44     ` Juri Lelli
2012-04-11 16:21   ` Steven Rostedt
2012-04-11 16:24     ` Juri Lelli
2012-04-11 16:33   ` Steven Rostedt
2012-04-24 13:15     ` Peter Zijlstra
2012-04-24 18:50       ` Steven Rostedt
2012-04-24 18:53         ` Peter Zijlstra
2012-04-24 19:01           ` Steven Rostedt
2012-04-11 17:25   ` Steven Rostedt
2012-04-11 17:48     ` Juri Lelli
2012-04-06  7:14 ` [PATCH 07/16] sched: SCHED_DEADLINE avg_update accounting Juri Lelli
2012-04-06  7:14 ` [PATCH 08/16] sched: add period support for -deadline tasks Juri Lelli
2012-04-11 20:32   ` Steven Rostedt
2012-04-11 21:56     ` Juri Lelli
2012-04-11 22:13     ` Tommaso Cucinotta
2012-04-12  0:19       ` Steven Rostedt
2012-04-12  6:39     ` Luca Abeni
2012-04-06  7:14 ` [PATCH 09/16] sched: add schedstats " Juri Lelli
2012-04-06  7:14 ` [PATCH 10/16] sched: add resource limits " Juri Lelli
2012-04-24 15:07   ` Peter Zijlstra
2012-04-24 15:22     ` Juri Lelli
2012-04-24 16:27       ` Peter Zijlstra
2012-04-24 17:14         ` Juri Lelli
2012-04-06  7:14 ` [PATCH 11/16] sched: add latency tracing " Juri Lelli
2012-04-11 21:03   ` Steven Rostedt
2012-04-12  7:16     ` Juri Lelli
2012-04-16 15:51     ` Daniel Vacek
2012-04-16 19:56       ` Steven Rostedt
2012-04-16 21:31         ` Daniel Vacek
2012-04-06  7:14 ` [PATCH 12/16] rtmutex: turn the plist into an rb-tree Juri Lelli
2012-04-11 21:11   ` Steven Rostedt
2012-04-22 14:28     ` Juri Lelli
2012-04-23  8:33     ` Peter Zijlstra
2012-04-23 11:37       ` Steven Rostedt
2012-04-06  7:14 ` [PATCH 13/16] sched: drafted deadline inheritance logic Juri Lelli
2012-04-12  2:42   ` Steven Rostedt
2012-04-22 14:04     ` Juri Lelli
2012-04-23  8:39     ` Peter Zijlstra
2012-04-06  7:14 ` [PATCH 14/16] sched: add bandwidth management for sched_dl Juri Lelli
2012-04-06  7:14 ` [PATCH 15/16] sched: speed up -dl pushes with a push-heap Juri Lelli
2012-04-06  7:14 ` [PATCH 16/16] sched: add sched_dl documentation Juri Lelli
2012-04-06  8:25 ` [RFC][PATCH 00/16] sched: SCHED_DEADLINE v4 Luca Abeni
2012-04-07  9:25   ` Tadeus Prastowo
2012-04-06 11:07 ` Dario Faggioli
2012-04-11 14:17 ` Steven Rostedt
2012-04-11 14:28   ` Juri Lelli
  -- strict thread matches above, loose matches on Subject: below --
2012-04-23 12:31 [PATCH 05/16] sched: SCHED_DEADLINE policy implementation cucinotta

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4F95769E.70303@gmail.com \
    --to=juri.lelli@gmail.com \
    --cc=cfriesen@nortel.com \
    --cc=claudio@evidence.eu.com \
    --cc=darren@dvhart.com \
    --cc=dhaval.giani@gmail.com \
    --cc=fchecconi@gmail.com \
    --cc=fweisbec@gmail.com \
    --cc=hgu1972@gmail.com \
    --cc=insop.song@ericsson.com \
    --cc=johan.eker@ericsson.com \
    --cc=liming.wang@windriver.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luca.abeni@unitn.it \
    --cc=michael@amarulasolutions.com \
    --cc=mingo@redhat.com \
    --cc=nicola.manica@disi.unitn.it \
    --cc=oleg@redhat.com \
    --cc=p.faure@akatech.ch \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=raistlin@linux.it \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    --cc=tommaso.cucinotta@sssup.it \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).