Re: [PATCH v4] softlockup: remove hung_task_check_count

[Frederic Weisbecker]

On Mon, Jan 26, 2009 at 04:30:55PM -0800, Mandeep Singh Baines wrote:
> Peter Zijlstra (peterz@xxxxxxxxxxxxx) wrote:
> > On Mon, 2009-01-26 at 09:36 -0800, Mandeep Baines wrote:
> > 
> > > Unfortunately, this can't be done for hung_task. It writes to the
> > > task_struct here:
> > 
> > Don't top post!
> > 
> > > static void check_hung_task(struct task_struct *t, unsigned long now,
> > >                             unsigned long timeout)
> > > {
> > >         unsigned long switch_count = t->nvcsw + t->nivcsw;
> > > 
> > >         if (t->flags & PF_FROZEN)
> > >                 return;
> > > 
> > >         if (switch_count != t->last_switch_count || !t->last_switch_timestamp) {
> > >                 t->last_switch_count = switch_count;
> > >                 t->last_switch_timestamp = now;
> > >                 return;
> > >         }
> > > 
> > > It is able to get away with using only a read_lock because no one else
> > > reads or writes to these fields.
> > 
> > How would RCU be different here?
> > 
> 
> My bad, RCU wouldn't be any different. I misunderstood how RCU works. Just
> spent the morning reading the LWN 3-part series on RCU and I think I'm able to
> grok it now;)
> 
> Below is a patch to hung_task which removes the hung_task_check_count and
> converts the read_locks to RCU.
> 
> Thanks Frédéric and Peter!
> 
> ---
> To avoid holding the tasklist lock too long, hung_task_check_count was used
> as an upper bound on the number of tasks that are checked by hung_task.
> This patch removes the hung_task_check_count sysctl.
> 
> Instead of checking a limited number of tasks, all tasks are checked. To
> avoid holding the CPU for too long, need_resched() is checked often. To
> avoid blocking out writers, the read_lock has been converted to an
> rcu_read_lock().
> 
> It is safe convert to an rcu_read_lock() because the tasks and thread_group
> lists are both protected by list_*_rcu() operations. The worst that can
> happen is that hung_task will update last_switch_timestamp field of a DEAD
> task.
> 
> The design was proposed by Frédéric Weisbecker. Peter Zijlstra suggested
> the use of RCU.
> 
> Signed-off-by: Mandeep Singh Baines <msb@xxxxxxxxxx>
> ---
>  include/linux/sched.h |    1 -
>  kernel/hung_task.c    |   12 +++---------
>  kernel/sysctl.c       |    9 ---------
>  3 files changed, 3 insertions(+), 19 deletions(-)
> 
> diff --git a/include/linux/sched.h b/include/linux/sched.h
> index f2f94d5..278121c 100644
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@ -315,7 +315,6 @@ static inline void touch_all_softlockup_watchdogs(void)
>  
>  #ifdef CONFIG_DETECT_HUNG_TASK
>  extern unsigned int  sysctl_hung_task_panic;
> -extern unsigned long sysctl_hung_task_check_count;
>  extern unsigned long sysctl_hung_task_timeout_secs;
>  extern unsigned long sysctl_hung_task_warnings;
>  extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
> diff --git a/kernel/hung_task.c b/kernel/hung_task.c
> index ba8ccd4..7d67350 100644
> --- a/kernel/hung_task.c
> +++ b/kernel/hung_task.c
> @@ -17,11 +17,6 @@
>  #include <linux/sysctl.h>
>  
>  /*
> - * Have a reasonable limit on the number of tasks checked:
> - */
> -unsigned long __read_mostly sysctl_hung_task_check_count = 1024;
> -
> -/*
>   * Zero means infinite timeout - no checking done:
>   */
>  unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120;
> @@ -116,7 +111,6 @@ static void check_hung_task(struct task_struct *t, unsigned long now,
>   */
>  static void check_hung_uninterruptible_tasks(unsigned long timeout)
>  {
> -	int max_count = sysctl_hung_task_check_count;
>  	unsigned long now = get_timestamp();
>  	struct task_struct *g, *t;
>  
> @@ -127,16 +121,16 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout)
>  	if (test_taint(TAINT_DIE) || did_panic)
>  		return;
>  
> -	read_lock(&tasklist_lock);
> +	rcu_read_lock();
>  	do_each_thread(g, t) {
> -		if (!--max_count)
> +		if (need_resched())
>  			goto unlock;
>  		/* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */
>  		if (t->state == TASK_UNINTERRUPTIBLE)
>  			check_hung_task(t, now, timeout);
>  	} while_each_thread(g, t);
>   unlock:
> -	read_unlock(&tasklist_lock);
> +	rcu_read_unlock();
>  }
>  
>  static void update_poll_jiffies(void)
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 2481ed3..16526a2 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -820,15 +820,6 @@ static struct ctl_table kern_table[] = {
>  	},
>  	{
>  		.ctl_name	= CTL_UNNUMBERED,
> -		.procname	= "hung_task_check_count",
> -		.data		= &sysctl_hung_task_check_count,
> -		.maxlen		= sizeof(unsigned long),
> -		.mode		= 0644,
> -		.proc_handler	= &proc_doulongvec_minmax,
> -		.strategy	= &sysctl_intvec,
> -	},
> -	{
> -		.ctl_name	= CTL_UNNUMBERED,
>  		.procname	= "hung_task_timeout_secs",
>  		.data		= &sysctl_hung_task_timeout_secs,
>  		.maxlen		= sizeof(unsigned long),
> -- 
> 1.5.4.5
> 


That looks good :-)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/