From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>,
Donald Buczek <buczek@molgen.mpg.de>,
Paul Menzel <pmenzel@molgen.mpg.de>,
dvteam@molgen.mpg.de, linux-mm@kvack.org,
linux-kernel@vger.kernel.org,
Josh Triplett <josh@joshtriplett.org>
Subject: Re: INFO: rcu_sched detected stalls on CPUs/tasks with `kswapd` and `mem_cgroup_shrink_node`
Date: Thu, 1 Dec 2016 10:42:52 -0800 [thread overview]
Message-ID: <20161201184252.GP3924@linux.vnet.ibm.com> (raw)
In-Reply-To: <20161201180953.GO3045@worktop.programming.kicks-ass.net>
On Thu, Dec 01, 2016 at 07:09:53PM +0100, Peter Zijlstra wrote:
> On Thu, Dec 01, 2016 at 08:59:18AM -0800, Paul E. McKenney wrote:
> > On Thu, Dec 01, 2016 at 05:36:14PM +0100, Peter Zijlstra wrote:
> > > Well, with the above change cond_resched() is already sufficient, no?
> >
> > Maybe. Right now, cond_resched_rcu_qs() gets a quiescent state to
> > the RCU core in less than one jiffy, with my other change, this becomes
> > a handful of jiffies depending on HZ and NR_CPUS. I expect this
> > increase to a handful of jiffies to be a non-event.
> >
> > After my upcoming patch, cond_resched() will get a quiescent state to
> > the RCU core in about ten seconds. While I am am not all that nervous
> > about the increase from less than a jiffy to a handful of jiffies,
> > increasing to ten seconds via cond_resched() does make me quite nervous.
> > Past experience indicates that someone's kernel will likely be fatally
> > inconvenienced by this magnitude of change.
> >
> > Or am I misunderstanding what you are proposing?
>
> No, that is indeed what I was proposing. Hurm.. OK let me ponder that a
> bit. There might be a few games we can play with !PREEMPT to avoid IPIs.
>
> Thing is, I'm slightly uncomfortable with de-coupling rcu-sched from
> actual schedule() calls.
OK, what is the source of your discomfort?
There are several intermediate levels of evasive action:
0. If there is another runnable task and certain other conditions
are met, cond_resched() will invoke schedule(), which will
provide an RCU quiescent state.
1. All cond_resched_rcu_qs() invocations increment the CPU's
rcu_qs_ctr per-CPU variable, which is treated by later
invocations of RCU core as a quiescent state. (I have
a patch queued that causes RCU to ignore changes to this
counter until the grace period is a few jiffies old.)
In this case, the rcu_node locks plus smp_mb__after_unlock_lock()
provide the needed ordering.
2. If any cond_resched_rcu_qs() sees that an expedited grace
period is waiting on the current CPU, it invokes rcu_sched_qs()
to force RCU to see the quiescent state. (To your point,
rcu_sched_qs() is normally called from schedule(), but also
from the scheduling-clock interrupt when it interrupts
usermode or idle.)
Again, the rcu_node locks plus smp_mb__after_unlock_lock()
provide the needed ordering.
3. If the grace period extends for more than 50 milliseconds
(by default, tunable), all subsequent cond_resched_rcu_qs()
invocations on that CPU turn into momentary periods of
idleness from RCU's viewpoint. (Atomically add 2 to the
dyntick-idle counter.)
Here, the atomic increment is surrounded by smp_mb__*_atomic()
to provide the needed ordering, which should be a good substitute
for actually passing through schedule().
4. If the grace period extends for more than 21 seconds (by default),
we emit an RCU CPU stall warning and then do a resched_cpu().
I am proposing also doing a resched_cpu() halfway to RCU CPU
stall-warning time.
5. An RCU-sched expedited grace period does a local resched_cpu()
from its IPI handler to force the CPU through a quiescent
state. (Yes, I could just invoke resched_cpu() from the
task orchestrating the expedited grace period, but this approach
allows more common code between RCU-preempt and RCU-sched
expedited grace periods.)
> > > In fact, by doing the IPI thing we get the entire cond_resched*()
> > > family, and we could add the should_resched() guard to
> > > cond_resched_rcu().
> >
> > So that cond_resched_rcu_qs() looks something like this, in order
> > to avoid the function call in the case where the scheduler has nothing
> > to do?
>
> I was actually thinking of this:
Oh! I had forgotten about cond_resched_rcu(), and thought you did a typo.
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> diff --git a/include/linux/sched.h b/include/linux/sched.h
> index 2d0c82e1d348..2dc7d8056b2a 100644
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@ -3374,9 +3374,11 @@ static inline int signal_pending_state(long state, struct task_struct *p)
> static inline void cond_resched_rcu(void)
> {
> #if defined(CONFIG_DEBUG_ATOMIC_SLEEP) || !defined(CONFIG_PREEMPT_RCU)
> - rcu_read_unlock();
> - cond_resched();
> - rcu_read_lock();
> + if (should_resched(1)) {
> + rcu_read_unlock();
> + cond_resched();
> + rcu_read_lock();
> + }
> #endif
> }
>
>
next prev parent reply other threads:[~2016-12-01 18:42 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <24c226a5-1a4a-173e-8b4e-5107a2baac04@molgen.mpg.de>
2016-11-08 12:22 ` INFO: rcu_sched detected stalls on CPUs/tasks with `kswapd` and `mem_cgroup_shrink_node` Paul Menzel
2016-11-08 17:03 ` Paul E. McKenney
2016-11-08 17:38 ` Paul Menzel
2016-11-08 18:39 ` Paul E. McKenney
2016-11-16 17:01 ` Paul Menzel
2016-11-16 17:30 ` Paul E. McKenney
2016-11-21 13:41 ` Michal Hocko
2016-11-21 14:01 ` Paul E. McKenney
2016-11-21 14:18 ` Michal Hocko
2016-11-21 14:29 ` Paul E. McKenney
2016-11-21 15:35 ` Donald Buczek
2016-11-24 10:15 ` Michal Hocko
2016-11-24 18:50 ` Donald Buczek
2016-11-27 9:37 ` Paul Menzel
2016-11-27 5:32 ` Christopher S. Aker
2016-11-27 9:19 ` Donald Buczek
2016-11-28 11:04 ` Michal Hocko
2016-11-28 12:26 ` Paul Menzel
2016-11-30 10:28 ` Donald Buczek
2016-11-30 11:09 ` Michal Hocko
2016-11-30 11:43 ` Donald Buczek
2016-12-02 9:14 ` Donald Buczek
2016-12-06 8:32 ` Donald Buczek
2016-11-30 11:53 ` Paul E. McKenney
2016-11-30 11:54 ` Paul E. McKenney
2016-11-30 12:31 ` Paul Menzel
2016-11-30 14:31 ` Paul E. McKenney
2016-11-30 13:19 ` Michal Hocko
2016-11-30 14:29 ` Paul E. McKenney
2016-11-30 16:38 ` Peter Zijlstra
2016-11-30 17:02 ` Paul E. McKenney
2016-11-30 17:05 ` Michal Hocko
2016-11-30 17:23 ` Paul E. McKenney
2016-11-30 17:34 ` Michal Hocko
2016-11-30 17:50 ` Peter Zijlstra
2016-11-30 19:40 ` Paul E. McKenney
2016-12-01 5:30 ` Peter Zijlstra
2016-12-01 12:40 ` Paul E. McKenney
2016-12-01 16:36 ` Peter Zijlstra
2016-12-01 16:59 ` Paul E. McKenney
2016-12-01 18:09 ` Peter Zijlstra
2016-12-01 18:42 ` Paul E. McKenney [this message]
2016-12-01 18:49 ` Peter Zijlstra
[not found] <20161125212000.GI31360@linux.vnet.ibm.com>
[not found] ` <20161128095825.GI14788@dhcp22.suse.cz>
[not found] ` <20161128105425.GY31360@linux.vnet.ibm.com>
[not found] ` <3a4242cb-0198-0a3b-97ae-536fb5ff83ec@kernelpanic.ru>
[not found] ` <20161128143435.GC3924@linux.vnet.ibm.com>
[not found] ` <eba1571e-f7a8-09b3-5516-c2bc35b38a83@kernelpanic.ru>
[not found] ` <20161128150509.GG3924@linux.vnet.ibm.com>
[not found] ` <66fd50e1-a922-846a-f427-7654795bd4b5@kernelpanic.ru>
[not found] ` <20161130174802.GM18432@dhcp22.suse.cz>
[not found] ` <fd34243c-2ebf-c14b-55e6-684a9dc614e7@kernelpanic.ru>
[not found] ` <20161130182552.GN18432@dhcp22.suse.cz>
2016-12-01 18:10 ` Boris Zhmurov
2016-12-01 19:39 ` Paul E. McKenney
2016-12-02 9:37 ` Michal Hocko
2016-12-02 13:52 ` Paul E. McKenney
2016-12-02 16:39 ` Boris Zhmurov
2016-12-02 16:44 ` Paul E. McKenney
2016-12-02 17:02 ` Michal Hocko
2016-12-02 17:15 ` Paul E. McKenney
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=20161201184252.GP3924@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=buczek@molgen.mpg.de \
--cc=dvteam@molgen.mpg.de \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=peterz@infradead.org \
--cc=pmenzel@molgen.mpg.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).