All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stanislav Meduna <stano@meduna.org>
To: "linux-rt-users@vger.kernel.org" <linux-rt-users@vger.kernel.org>
Subject: Re: hrtimer: interrupt took 6742 ns, then RT throttling and hung machine for nearly 2 seconds
Date: Mon, 15 Apr 2013 13:54:01 +0200	[thread overview]
Message-ID: <516BEA59.4050303@meduna.org> (raw)
In-Reply-To: <516BDE52.90200@meduna.org>

On 15.04.2013 13:02, Stanislav Meduna wrote:

> Apr 15 10:14:57 lnx kernel: [56281.700293] hrtimer: interrupt took 6742 ns

The /proc/timer_list follows. What worries me a bit are the retries -
they are slowly rising (~1 / sec for nr_retries) even if the machine
is basically idle. Is it normal?

The rise only happens if our application is running, but it is
basically the only thing in system that is actively using
the hrtimers, so this does not really say much. Tweaking
the priority of ksoftirqd has no effect.


Timer List Version: v0.6
HRTIMER_MAX_CLOCK_BASES: 3
now at 342575389945882 nsecs

cpu: 0
 clock 0:
  .base:       c04d7f68
  .index:      0
  .resolution: 1 nsecs
  .get_time:   ktime_get
  .offset:     0 nsecs
active timers:
 #0: tick_cpu_sched, tick_sched_timer, S:01
 # expires at 342575390000000-342575390000000 nsecs [in 54118 to 54118
nsecs]
 #1: <cd9bb680>, timerfd_tmrproc, S:01
 # expires at 342575390284003-342575390284003 nsecs [in 338121 to 338121
nsecs]
 #2: <cc6cff48>, hrtimer_wakeup, S:01
 # expires at 342575390421978-342575390421978 nsecs [in 476096 to 476096
nsecs]
... 16 more hrtimer_wakeup follow ...

 clock 1:
  .base:       c04d7fa0
  .index:      1
  .resolution: 1 nsecs
  .get_time:   ktime_get_real
  .offset:     1365682957275692841 nsecs
active timers:
 clock 2:
  .base:       c04d7fd8
  .index:      2
  .resolution: 1 nsecs
  .get_time:   ktime_get_boottime
  .offset:     0 nsecs
active timers:
  .expires_next   : 342575390284003 nsecs
  .hres_active    : 1
  .nr_events      : 556415337
  .nr_retries     : 1124752
  .nr_hangs       : 1
  .max_hang_time  : 6742 nsecs
  .nohz_mode      : 2
  .idle_tick      : 342575389000000 nsecs
  .tick_stopped   : 0
  .idle_jiffies   : 342275388
  .idle_calls     : 514502119
  .idle_sleeps    : 510841382
  .idle_entrytime : 342575388177611 nsecs
  .idle_waketime  : 342575388177611 nsecs
  .idle_exittime  : 342575388184375 nsecs
  .idle_sleeptime : 301580528758450 nsecs
  .iowait_sleeptime: 21756268924 nsecs
  .last_jiffies   : 342275388
  .next_jiffies   : 342275421
  .idle_expires   : 342575421000000 nsecs
jiffies: 342275390


Tick Device: mode:     1
Per CPU device: 0
Clock Event Device: hpet
 max_delta_ns:   149983013276
 min_delta_ns:   13409
 mult:           61496111
 shift:          32
 mode:           3
 next_event:     342575390284003 nsecs
 set_next_event: hpet_legacy_next_event
 set_mode:       hpet_legacy_set_mode
 event_handler:  hrtimer_interrupt
 retries:        9359


Regards
-- 
                                      Stano


  reply	other threads:[~2013-04-15 11:54 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-15 11:02 hrtimer: interrupt took 6742 ns, then RT throttling and hung machine for nearly 2 seconds Stanislav Meduna
2013-04-15 11:54 ` Stanislav Meduna [this message]
2013-04-15 12:22 ` Thomas Gleixner
2013-04-15 12:56   ` Stanislav Meduna
2013-04-17 15:46     ` timerfd and softirqd [Was: Re: hrtimer: interrupt took 6742 ns, then RT throttling and hung machine for nearly 2 seconds] Stanislav Meduna
2013-04-18  9:11       ` timerfd read does not return [Was: Re: timerfd and softirqd] Stanislav Meduna
2013-04-19 19:53         ` Stanislav Meduna
2013-04-22  7:35           ` [PATCH] Re: timerfd read does not return Stanislav Meduna
2013-04-22  8:55             ` Stanislav Meduna
2013-04-27  8:34         ` timerfd read does not return - was probably fixed in 3.4.38 Stanislav Meduna
2013-04-28 11:53           ` Carsten Emde
2013-04-29  8:43             ` Stanislav Meduna
2013-05-02 20:02           ` Steven Rostedt
2013-05-10 12:42           ` timerfd read does not return - some traces Stanislav Meduna
2013-05-12 17:31             ` Stanislav Meduna
2013-05-12 23:20             ` timerfd read does not return - hangs inside put_user Stanislav Meduna
2013-05-13  8:05               ` timerfd read does not return - caused by MM fault Stanislav Meduna
2013-05-14  8:31                 ` Livelock in handle_pte_fault [Was: Re: timerfd read does not return] Stanislav Meduna
2013-11-25 10:36                   ` Vijay Katoch

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=516BEA59.4050303@meduna.org \
    --to=stano@meduna.org \
    --cc=linux-rt-users@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.