mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Preeti U Murthy <preeti@linux.vnet.ibm.com>,
	Christoph Lameter <cl@linux.com>, Ingo Molnar <mingo@kernel.org>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	Rik van Riel <riel@redhat.com>
Subject: Re: [PATCH 08/10] posix-cpu-timers: Migrate to use new tick dependency mask model
Date: Mon, 3 Aug 2015 21:07:40 +0200	[thread overview]
Message-ID: <20150803190740.GB25159@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <20150803173936.GC26022@lerouge>

On Mon, Aug 03, 2015 at 07:39:37PM +0200, Frederic Weisbecker wrote:
> On Mon, Aug 03, 2015 at 07:12:43PM +0200, Peter Zijlstra wrote:
> > On Fri, Jul 31, 2015 at 04:49:55PM +0200, Frederic Weisbecker wrote:
> > > Instead of doing a per signal dependency, I'm going to use a per task
> > > one.
> > 
> > Urgh, does this mean you'll keep the horrid tick_nohz_task_switch()
> > thing?
> 
> I thought I would drop it, but now that I think about it more, I think I
> need to keep it because if we enqueue a posix timer to a sleeping task
> and that task later wakes up, we must restart the tick, and that is only
> possible with a check on context switch :-(
> 
> This current patchset removed the need for that with a global dependency
> for posix timers: as long as there is one enqueued we keep the tick. But
> Chris and Luiz fear that Tilera users have posix timers on housekeepers.
> They also suggested we offline the posix timers. I fear it's going to be
> a high overhead as it means polling on the target task context of execution.
> Unless we move the task itself to housekeepers...

At least do something like the below, that irq save/restore is expensive
and can be trivially avoided.

Also, tick_nohz_full_kick() checks that tick_nohz_full_cpu() thing
again.

---
 kernel/sched/core.c      | 2 +-
 kernel/time/tick-sched.c | 9 +--------
 2 files changed, 2 insertions(+), 9 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 4d34035bb3ee..57d1af7c0660 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2471,6 +2471,7 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	vtime_task_switch(prev);
 	finish_arch_switch(prev);
 	perf_event_task_sched_in(prev, current);
+	tick_nohz_task_switch();
 	finish_lock_switch(rq, prev);
 	finish_arch_post_lock_switch();
 
@@ -2489,7 +2490,6 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 		put_task_struct(prev);
 	}
 
-	tick_nohz_task_switch();
 	return rq;
 }
 
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 3319e16f31e5..9f2225ef230c 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -260,18 +260,11 @@ void tick_nohz_full_kick_all(void)
  */
 void __tick_nohz_task_switch(void)
 {
-	unsigned long flags;
-
-	local_irq_save(flags);
-
 	if (!tick_nohz_full_cpu(smp_processor_id()))
-		goto out;
+		return;
 
 	if (tick_nohz_tick_stopped() && !can_stop_full_tick())
 		tick_nohz_full_kick();
-
-out:
-	local_irq_restore(flags);
 }
 
 /* Parse the boot-time nohz CPU list from the kernel parameters. */

  reply	other threads:[~2015-08-03 19:07 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-23 16:42 [PATCH 00/10] nohz: Tick dependency mask v2 Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 01/10] nohz: Remove idle task special case Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 02/10] nohz: Restart nohz full tick from irq exit Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 03/10] nohz: Move tick_nohz_restart_sched_tick() above its users Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 04/10] nohz: Remove useless argument on tick_nohz_task_switch() Frederic Weisbecker
2015-08-03 12:39   ` Peter Zijlstra
2015-08-03 12:49     ` Frederic Weisbecker
2015-08-03 13:04       ` Peter Zijlstra
2015-07-23 16:42 ` [PATCH 05/10] nohz: New tick dependency mask Frederic Weisbecker
2015-07-24 16:55   ` Chris Metcalf
2015-07-24 17:16     ` Frederic Weisbecker
2015-07-24 17:43       ` Chris Metcalf
2015-08-03 12:48         ` Peter Zijlstra
2015-08-03 12:43   ` Peter Zijlstra
2015-08-03 13:05     ` Frederic Weisbecker
2015-08-03 13:24       ` Peter Zijlstra
2015-08-03 13:49         ` Frederic Weisbecker
2015-08-03 12:57   ` Peter Zijlstra
2015-08-03 13:09     ` Frederic Weisbecker
2015-08-03 13:29       ` Peter Zijlstra
2015-08-03 13:55         ` Frederic Weisbecker
2015-08-03 14:11           ` Peter Zijlstra
2015-07-23 16:42 ` [PATCH 06/10] perf: Migrate perf to use new tick dependency mask model Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 07/10] sched: Migrate sched " Frederic Weisbecker
2015-07-23 16:55   ` Frederic Weisbecker
2015-07-24 16:56   ` Chris Metcalf
2015-07-29 13:01     ` Frederic Weisbecker
2015-08-03 14:00   ` Peter Zijlstra
2015-08-03 14:50     ` Frederic Weisbecker
2015-08-03 17:09       ` Peter Zijlstra
2015-08-03 17:30         ` Frederic Weisbecker
2015-08-04  7:41           ` Peter Zijlstra
2015-08-10 14:02             ` Juri Lelli
2015-08-10 14:16               ` Frederic Weisbecker
2015-08-10 14:28                 ` Peter Zijlstra
2015-08-10 15:11                   ` Peter Zijlstra
2015-08-10 15:29                     ` Frederic Weisbecker
2015-08-10 15:43                       ` Juri Lelli
2015-08-10 16:41                       ` Peter Zijlstra
2015-08-10 15:33                 ` Christoph Lameter
2015-07-23 16:42 ` [PATCH 08/10] posix-cpu-timers: Migrate " Frederic Weisbecker
2015-07-24 16:57   ` Chris Metcalf
2015-07-29 13:23     ` Frederic Weisbecker
2015-07-29 17:24       ` Chris Metcalf
2015-07-30  0:44         ` Frederic Weisbecker
2015-07-30 14:31           ` Luiz Capitulino
2015-07-30 14:46             ` Frederic Weisbecker
2015-07-30 19:35           ` Chris Metcalf
2015-07-30 19:45             ` Frederic Weisbecker
2015-07-30 19:52               ` Chris Metcalf
2015-07-31 14:49                 ` Frederic Weisbecker
2015-08-03 15:59                   ` Chris Metcalf
2015-08-03 18:01                     ` Frederic Weisbecker
2015-08-03 17:12                   ` Peter Zijlstra
2015-08-03 17:39                     ` Frederic Weisbecker
2015-08-03 19:07                       ` Peter Zijlstra [this message]
2015-08-06 17:13                       ` Chris Metcalf
2015-07-23 16:42 ` [PATCH 09/10] sched-clock: " Frederic Weisbecker
2015-07-23 16:42 ` [PATCH 10/10] nohz: Remove task switch obsolete tick dependency check Frederic Weisbecker

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=20150803190740.GB25159@twins.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=cl@linux.com \
    --cc=cmetcalf@ezchip.com \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=preeti@linux.vnet.ibm.com \
    --cc=riel@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=viresh.kumar@linaro.org \
    /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

all inboxes | Powered by JetHome®