From: Frederic Weisbecker <fweisbec@gmail.com>
To: Rik van Riel <riel@redhat.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>,
Wanpeng Li <kernellwp@gmail.com>,
linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [BUG nohz]: wrong user and system time accounting
Date: Thu, 30 Mar 2017 00:54:30 +0200 [thread overview]
Message-ID: <20170329225428.GC23895@lerouge> (raw)
In-Reply-To: <1490818125.28917.11.camel@redhat.com>
(Adding Thomas in Cc)
On Wed, Mar 29, 2017 at 04:08:45PM -0400, Rik van Riel wrote:
> On Wed, 2017-03-29 at 13:16 -0400, Luiz Capitulino wrote:
> > On Tue, 28 Mar 2017 13:24:06 -0400
> > Luiz Capitulino <lcapitulino@redhat.com> wrote:
> >
> > > 1. In my tracing I'm seeing that sometimes (always?) the
> > > time interval between two timer interrupts is less than 1ms
> >
> > I think that's the root cause.
> >
> > In this trace, we see the following:
> >
> > 1. On CPU15, we transition from user-space to kernel-space because
> > of a timer interrupt (it's the tick)
> >
> > 2. vtimer_delta() returns 0, because jiffies didn't change since the
> > last accounting
> >
> > 3. While CPU15 is executing in kernel-space, jiffies is updated
> > by CPU0
> >
> > 4. When going back to user-space, vtime_delta() returns non-zero
> > and the whole time is accounted for system time (observe how
> > the cputime parameter in account_system_time() is less than 1ms)
>
> In other words, the tick on cpu0 is aligned
> with the tick on the nohz_full cpus, and
> jiffies is advanced while the nohz_full cpus
> with an active tick happen to be in kernel
> mode?
Ah you found out faster than me :-)
> Frederic, can you think of any reason why
> the tick on nohz_full CPUs would end up aligned
> with the tick on cpu0, instead of running at some
> random offset?
tick_init_jiffy_update() takes that decision to align all ticks.
I'm not sure why. I don't see anything that could depend on that
wide tick synchronization. The jiffies update itself relies on ktime
to check when to update it. So even if the tick fires a bit later
on CPU 1 than on CPU 0, the jiffies updates should stay coherent and
should never exceed 999us delay in the worst case (for HZ=1000)
Now I might overlook something.
>
> A random offset, or better yet a somewhat randomized
> tick length to make sure that simultaneous ticks are
> fairly rare and the vtime sampling does not end up
> "in phase" with the jiffies incrementing, could make
> the accounting work right again.
>
> Of course, that assumes the above hypothesis is correct :)
I'm not sure that randomizing the tick start per CPU would be a
right solution. Somewhere in the world you can be sure the tick
randomization of some nohz_full CPU will coincide with the tick
of CPU 0 :o)
Or we could force that tick on nohz_full CPUs to be far from
CPU 0's tick... I'm not sure such a solution would be accepted though.
next prev parent reply other threads:[~2017-03-29 22:54 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-23 20:55 [BUG nohz]: wrong user and system time accounting Luiz Capitulino
2017-03-24 0:56 ` Rik van Riel
2017-03-24 1:05 ` Luiz Capitulino
2017-03-24 1:08 ` Rik van Riel
2017-03-24 1:39 ` Luiz Capitulino
2017-03-27 5:33 ` lkml
2017-03-24 1:52 ` Wanpeng Li
2017-03-24 3:56 ` Luiz Capitulino
2017-03-27 1:56 ` Wanpeng Li
2017-03-27 17:35 ` Rik van Riel
2017-03-28 7:19 ` Wanpeng Li
[not found] ` <20170328132406.7d23579c@redhat.com>
[not found] ` <20170328161454.4a5d9e8b@redhat.com>
2017-03-28 21:01 ` Rik van Riel
2017-03-28 21:26 ` Luiz Capitulino
2017-03-29 9:56 ` Wanpeng Li
2017-03-29 12:56 ` Frederic Weisbecker
2017-03-28 21:24 ` Rik van Riel
2017-03-28 21:30 ` Luiz Capitulino
[not found] ` <20170329131656.1d6cb743@redhat.com>
2017-03-29 20:08 ` Rik van Riel
2017-03-29 22:54 ` Frederic Weisbecker [this message]
2017-03-30 12:57 ` Rik van Riel
2017-03-30 1:58 ` Wanpeng Li
2017-03-30 12:40 ` Frederic Weisbecker
2017-03-30 13:19 ` Mike Galbraith
2017-03-30 4:27 ` Mike Galbraith
2017-03-30 6:47 ` Wanpeng Li
2017-03-30 11:52 ` Wanpeng Li
2017-03-30 12:33 ` Mike Galbraith
2017-03-30 13:38 ` Frederic Weisbecker
2017-03-30 13:59 ` Wanpeng Li
2017-03-30 14:18 ` Frederic Weisbecker
2017-03-30 21:25 ` Luiz Capitulino
2017-03-31 20:09 ` Luiz Capitulino
2017-03-31 23:24 ` Frederic Weisbecker
2017-04-01 3:11 ` Luiz Capitulino
2017-04-03 15:23 ` Frederic Weisbecker
2017-04-03 19:06 ` Luiz Capitulino
2017-04-04 17:36 ` Luiz Capitulino
2017-04-05 14:26 ` Rik van Riel
2017-04-11 11:03 ` Wanpeng Li
2017-04-11 11:36 ` Peter Zijlstra
2017-04-11 11:43 ` Wanpeng Li
2017-04-11 14:22 ` Thomas Gleixner
2017-04-12 13:18 ` Frederic Weisbecker
2017-04-12 14:57 ` Thomas Gleixner
2017-04-12 15:14 ` Frederic Weisbecker
2017-04-13 4:31 ` Wanpeng Li
2017-04-13 13:32 ` Frederic Weisbecker
2017-05-02 10:01 ` Wanpeng Li
2017-05-15 8:17 ` Wanpeng Li
2017-06-29 17:22 ` Frederic Weisbecker
2017-03-30 12:51 ` Frederic Weisbecker
2017-03-30 13:02 ` Rik van Riel
2017-03-30 13:35 ` Mike Galbraith
2017-04-03 14:40 ` Frederic Weisbecker
2017-04-04 7:32 ` Mike Galbraith
2017-03-30 13:44 ` Frederic Weisbecker
[not found] ` <20170329221700.GB23895@lerouge>
2017-03-29 22:46 ` Wanpeng Li
2017-03-30 2:14 ` Luiz Capitulino
2017-03-30 12:27 ` Wanpeng Li
2017-03-27 18:38 ` Luiz Capitulino
2017-03-28 5:28 ` Wanpeng Li
2017-03-28 13:44 ` Luiz Capitulino
2017-03-29 13:04 ` Frederic Weisbecker
2017-03-29 13:14 ` Rik van Riel
2017-03-29 13:23 ` Luiz Capitulino
2017-03-29 21:12 ` Frederic Weisbecker
2017-03-30 1:48 ` Luiz Capitulino
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=20170329225428.GC23895@lerouge \
--to=fweisbec@gmail.com \
--cc=kernellwp@gmail.com \
--cc=lcapitulino@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=riel@redhat.com \
--cc=tglx@linutronix.de \
/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).