Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1507336

Re: hit a KASan bug related to Perf during stress test

From Oleg Nesterov <oleg@redhat.com>
Newsgroups linux.kernel
Subject Re: hit a KASan bug related to Perf during stress test
Date 2016-10-24 18:00 +0200
Message-ID <svPPj-1bs-9@gated-at.bofh.it> (permalink)
References (5 earlier) <svMHM-7B9-25@gated-at.bofh.it> <svMHM-7B9-23@gated-at.bofh.it> <svNu9-87q-27@gated-at.bofh.it> <svOzU-lR-15@gated-at.bofh.it> <svPFE-17R-19@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

On 10/24, Oleg Nesterov wrote:
>
> On 10/24, Peter Zijlstra wrote:
> >
> > --- a/kernel/events/core.c
> > +++ b/kernel/events/core.c
> > @@ -1257,7 +1257,14 @@ static u32 perf_event_pid(struct perf_event *event, struct task_struct *p)
> >  	if (event->parent)
> >  		event = event->parent;
> >
> > -	return task_tgid_nr_ns(p, event->ns);
> > +	/*
> > +	 * It is possible the task already got unhashed, in which case we
> > +	 * cannot determine the current->group_leader/real_parent.
> > +	 *
> > +	 * Also, report -1 to indicate unhashed, so as not to confused with
> > +	 * 0 for the idle task.
> > +	 */
> > +	return pid_alive(p) ? task_tgid_nr_ns(p, event->ns) : ~0;
> >  }
>
> Yes, but this _looks_ racy unless p == current. I mean, pid_alive() makes
> task_tgid_nr_ns() safe, but task_tgid_nr_ns() still can return zero _if_
> it can race with the exiting task.
>
> >  static u32 perf_event_tid(struct perf_event *event, struct task_struct *p)
> > @@ -1268,7 +1275,7 @@ static u32 perf_event_tid(struct perf_event *event, struct task_struct *p)
> >  	if (event->parent)
> >  		event = event->parent;
> >
> > -	return task_pid_nr_ns(p, event->ns);
> > +	return pid_alive(p) ? task_pid_nr_ns(p, event->ns) : ~0;
>
> The same.
>
> However. At first glance the only case when p != current is copy_process(),
> right? And in this case the new child can't go away. So I think this patch
> is fine.

Actually there is another case, comm_write() -> perf_event_comm_output(). It
checks same_thread_group(current, p), so we can only race with the exiting
sub-thread. perf_event_pid() can't return zero, perf_event_tid() can.

And I personally think we do not care and your patch is fine anyway ;)

Oleg.

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

Thread

Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 12:00 +0200
  Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 13:20 +0200
    Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 13:30 +0200
      Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 13:40 +0200
        Re: hit a KASan bug related to Perf during stress test Jiri Olsa <jolsa@redhat.com> - 2016-10-24 14:10 +0200
          Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:20 +0200
    Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 13:30 +0200
      Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 14:10 +0200
        Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 14:20 +0200
          Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:30 +0200
            Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:40 +0200
              Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 15:30 +0200
                Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 15:50 +0200
                Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 16:20 +0200
                Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 16:40 +0200
                Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 17:50 +0200
                Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 18:00 +0200
                RE: hit a KASan bug related to Perf during stress test "Ni, BaoleX" <baolex.ni@intel.com> - 2016-10-25 09:00 +0200
                Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-25 11:30 +0200
                Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-25 16:50 +0200
                Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-26 11:10 +0200
                Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-26 18:20 +0200
            Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 14:40 +0200
        Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:20 +0200
    Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:20 +0200
      Re: hit a KASan bug related to Perf during stress test Peter Zijlstra <peterz@infradead.org> - 2016-10-24 14:30 +0200
      Re: hit a KASan bug related to Perf during stress test Oleg Nesterov <oleg@redhat.com> - 2016-10-24 14:30 +0200

csiph-web