From: Peter Zijlstra <peterz@infradead.org>
To: Raistlin <raistlin@linux.it>
Cc: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
Steven Rostedt <rostedt@goodmis.org>,
Chris Friesen <cfriesen@nortel.com>,
Frederic Weisbecker <fweisbec@gmail.com>,
Darren Hart <darren@dvhart.com>, Henrik Austad <henrik@austad.us>,
Johan Eker <johan.eker@ericsson.com>,
"p.faure" <p.faure@akatech.ch>,
linux-kernel <linux-kernel@vger.kernel.org>,
Claudio Scordino <claudio@evidence.eu.com>,
michael trimarchi <trimarchi@retis.sssup.it>,
Fabio Checconi <fabio@gandalf.sssup.it>,
Tommaso Cucinotta <t.cucinotta@sssup.it>,
Juri Lelli <juri.lelli@gmail.com>,
Nicola Manica <nicola.manica@gmail.com>,
Luca Abeni <luca.abeni@unitn.it>
Subject: Re: [RFC][PATCH 10/11] sched: add bandwidth management for sched_dl.
Date: Wed, 14 Apr 2010 12:09:48 +0200 [thread overview]
Message-ID: <1271239788.32749.15.camel@laptop> (raw)
In-Reply-To: <1267385230.13676.101.camel@Palantir>
On Sun, 2010-02-28 at 20:27 +0100, Raistlin wrote:
> @@ -2063,6 +2210,30 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd)
> return delta < (s64)sysctl_sched_migration_cost;
> }
>
> +/*
> + * When dealing with a -deadline task, we have to check if moving it to
> + * a new CPU is possible or not. In fact, this is only true iff there
> + * is enough bandwidth available on such CPU, otherwise we want the
> + * whole migration progedure to fail over.
> + */
> +static inline
> +bool __set_task_cpu_dl(struct task_struct *p, unsigned int cpu)
> +{
> + struct dl_bandwidth *dl_b = task_dl_bandwidth(p);
> +
> + raw_spin_lock(&dl_b->dl_runtime_lock);
> + if (dl_b->dl_bw < dl_b->dl_total_bw[cpu] + p->dl.dl_bw) {
> + raw_spin_unlock(&dl_b->dl_runtime_lock);
> +
> + return 0;
> + }
> + dl_b->dl_total_bw[task_cpu(p)] -= p->dl.dl_bw;
> + dl_b->dl_total_bw[cpu] += p->dl.dl_bw;
> + raw_spin_unlock(&dl_b->dl_runtime_lock);
> +
> + return 1;
> +}
> +
> void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
> {
> #ifdef CONFIG_SCHED_DEBUG
> @@ -2077,6 +2248,9 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
> trace_sched_migrate_task(p, new_cpu);
>
> if (task_cpu(p) != new_cpu) {
> + if (task_has_dl_policy(p) && !__set_task_cpu_dl(p, new_cpu))
> + return;
> +
> p->se.nr_migrations++;
> perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0);
> }
Yikes!!, I'm not sure we can sanely deal with set_task_cpu() doing that.
I'd much rather see us never attempting set_task_cpu() when we know its
not going to be possible.
That also means that things like set_cpus_allowed_ptr() /
sys_sched_setaffinity() will need to propagate the error back to their
users, which in turn will need to be able to cope.
next prev parent reply other threads:[~2010-04-15 7:29 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-28 19:06 [RFC][PATCH 0/11] sched: SCHED_DEADLINE v2 Raistlin
2010-02-28 19:15 ` [RFC][PATCH 01/11] sched: add sched_class->task_dead Raistlin
2010-02-28 19:17 ` [RFC][PATCH 02/11] sched: SCHED_DEADLINE policy implementation Raistlin
2010-04-13 18:22 ` Peter Zijlstra
2010-04-13 18:22 ` Peter Zijlstra
2010-04-13 18:22 ` Peter Zijlstra
2010-04-13 18:55 ` Steven Rostedt
2010-04-15 7:34 ` Peter Zijlstra
2010-04-13 18:22 ` Peter Zijlstra
2010-04-13 18:22 ` Peter Zijlstra
2010-02-28 19:18 ` [RFC][PATCH 03/11] sched: add extended scheduling interface Raistlin
2010-02-28 19:19 ` [RFC][PATCH 04/11] sched: add resource limits for -deadline tasks Raistlin
2010-02-28 19:20 ` [RFC][PATCH 05/11] sched: add a syscall to wait for the next instance Raistlin
2010-02-28 19:22 ` [RFC][PATCH 06/11] sched: add the sched-debug bits for sched_dl Raistlin
2010-02-28 19:23 ` [RFC][PATCH 07/11] sched: add latency tracing for -deadline tasks Raistlin
2010-02-28 19:24 ` [RFC][PATCH 08/11] sched: send SIGXCPU at -deadline task overruns Raistlin
2010-04-13 18:22 ` Peter Zijlstra
2010-04-13 19:32 ` Oleg Nesterov
2010-02-28 19:26 ` [RFC][PATCH 09/11] sched: first draft of deadline inheritance Raistlin
2010-04-14 8:25 ` Peter Zijlstra
2010-04-14 9:45 ` Peter Zijlstra
2010-02-28 19:27 ` [RFC][PATCH 10/11] sched: add bandwidth management for sched_dl Raistlin
2010-04-14 10:09 ` Peter Zijlstra [this message]
2010-02-28 19:28 ` [RFC][PATCH 11/11] sched: add sched_dl documentation Raistlin
2010-04-14 10:17 ` [RFC][PATCH 0/11] sched: SCHED_DEADLINE v2 Peter Zijlstra
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=1271239788.32749.15.camel@laptop \
--to=peterz@infradead.org \
--cc=cfriesen@nortel.com \
--cc=claudio@evidence.eu.com \
--cc=darren@dvhart.com \
--cc=fabio@gandalf.sssup.it \
--cc=fweisbec@gmail.com \
--cc=henrik@austad.us \
--cc=johan.eker@ericsson.com \
--cc=juri.lelli@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luca.abeni@unitn.it \
--cc=mingo@elte.hu \
--cc=nicola.manica@gmail.com \
--cc=p.faure@akatech.ch \
--cc=raistlin@linux.it \
--cc=rostedt@goodmis.org \
--cc=t.cucinotta@sssup.it \
--cc=tglx@linutronix.de \
--cc=trimarchi@retis.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).