Re: hung_task checking and sys_sync

[Daniel Walker <dwalker@fifo99.com>]

On Tue, Jun 12, 2012 at 03:45:20PM -0700, Mandeep Baines wrote:
> On Tue, Jun 12, 2012 at 3:34 PM, Daniel Walker <dwalker@fifo99.com> wrote:
> > On Tue, Jun 12, 2012 at 03:29:12PM -0700, Mandeep Singh Baines wrote:
> >>
> >> But the time is not unbounded. You could mask the hung_task_detector for
> >> this case but then you lose the ability to catch bugs in this code path.
> >>
> >> The timeout is configurable via /proc/sys/kernel/hung_task_timeout_secs.
> >> Can you bump up the value at boot via sysctl.conf?
> >
> > Maybe, but I'm wondering if these types should just be stopped because Andrew
> > had complained about them already.
> >
> 
> Fair enough. Actually, internally I had a patch where we'd use a task
> flag to disable and enable the hang check but the approach in the
> patch you pointed me to seems better.

I'm not really in love with it actually.. It's not ifdef'd for one, but
it's also changing potentially good kernel behavior to avoid warnings.
 
> >> > Has there been any commit that disable these messages bdi_sched_wait?
> >> >
> >>
> >> No. There is no mechanism to disable hung_task for a specific code path.
> >> We do skip processes if PF_PROZEN or PF_FROZEN_SKIP is set but that is
> >> really a different situation where the wait is unbounded.
> >
> > There is presidence for this type of change,
> >
> > Author: Mark Lord <kernel@teksavvy.com>
> > Date:   Fri Sep 24 09:51:13 2010 -0400
> >
> >    block: Prevent hang_check firing during long I/O
> >
> >    During long I/O operations, the hang_check timer may fire,
> >    trigger stack dumps that unnecessarily alarm the user.
> >
> >    Eg.  hdparm --security-erase NULL /dev/sdb  ## can take *hours* to complete
> >
> >    So, if hang_check is armed, we should wake up periodically
> >    to prevent it from triggering.  This patch uses a wake-up interval
> >    equal to half the hang_check timer period, which keeps overhead low enough.
> >
> >    Signed-off-by: Mark Lord <mlord@pobox.com>
> >    Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
> >
> 
> Interesting. I wasn't aware of this patch. Maybe we could abstract
> this approach via wait_for_completion_no_hang_check().

Could be .. You could put a stack structure into a list of tasks that
should be ignored prior to the task sleeping. Then when the thread wakes
the stack structure could be removed. Then that list get checked
during the hung task checking.

Daniel