Re: cpu_hotplug vs oom_notify_list: possible circular locking dependency detected

From	David Rientjes <rientjes@google.com>
Newsgroups	linux.kernel
Subject	Re: cpu_hotplug vs oom_notify_list: possible circular locking dependency detected
Date	2015-07-16 23:10 +0200
Message-ID	<pMYzN-1uX-41@gated-at.bofh.it> (permalink)
References	<pLn9g-7fr-9@gated-at.bofh.it> <pMh1M-5xP-1@gated-at.bofh.it> <pMhOa-6FT-7@gated-at.bofh.it> <pMi7x-72M-23@gated-at.bofh.it> <pMDlE-4rS-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

[Multipart message — attachments visible in raw view] - view raw

On Wed, 15 Jul 2015, Paul E. McKenney wrote:

> On Tue, Jul 14, 2015 at 04:48:24PM -0700, David Rientjes wrote:
> > On Tue, 14 Jul 2015, Paul E. McKenney wrote:
> > 
> > > commit a1992f2f3b8e174d740a8f764d0d51344bed2eed
> > > Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > > Date:   Tue Jul 14 16:24:14 2015 -0700
> > > 
> > >     rcu: Don't disable CPU hotplug during OOM notifiers
> > >     
> > >     RCU's rcu_oom_notify() disables CPU hotplug in order to stabilize the
> > >     list of online CPUs, which it traverses.  However, this is completely
> > >     pointless because smp_call_function_single() will quietly fail if invoked
> > >     on an offline CPU.  Because the count of requests is incremented in the
> > >     rcu_oom_notify_cpu() function that is remotely invoked, everything works
> > >     nicely even in the face of concurrent CPU-hotplug operations.
> > >     
> > >     Furthermore, in recent kernels, invoking get_online_cpus() from an OOM
> > >     notifier can result in deadlock.  This commit therefore removes the
> > >     call to get_online_cpus() and put_online_cpus() from rcu_oom_notify().
> > >     
> > >     Reported-by: Marcin Ślusarz <marcin.slusarz@gmail.com>
> > >     Reported-by: David Rientjes <rientjes@google.com>
> > >     Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > 
> > Acked-by: David Rientjes <rientjes@google.com>
> 
> Thank you!
> 
> Any news on whether or not it solves the problem?
> 

Marcin, is your lockdep violation reproducible?  If so, does this patch 
fix it?

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: cpu_hotplug vs oom_notify_list: possible circular locking  dependency detected "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-07-15 01:30 +0200
  Re: cpu_hotplug vs oom_notify_list: possible circular locking  dependency detected David Rientjes <rientjes@google.com> - 2015-07-15 01:50 +0200
    Re: cpu_hotplug vs oom_notify_list: possible circular locking  dependency detected David Rientjes <rientjes@google.com> - 2015-07-16 23:10 +0200

csiph-web