Re: lots of suspicious RCU traces

[Paul E. McKenney]

On Wed, Oct 24, 2012 at 12:17:16PM -0700, Sergey Senozhatsky wrote:
> On (10/24/12 20:52), Oleg Nesterov wrote:
> > On 10/24, Sergey Senozhatsky wrote:
> > >
> > > On (10/24/12 20:06), Oleg Nesterov wrote:
> > > > On 10/24, Sergey Senozhatsky wrote:
> > > > >
> > > > > small question,
> > > > >
> > > > > ptrace_notify() and forward calls are able to both indirectly and directly call schedule(),
> > > > > /* direct call from ptrace_stop()*/,
> > > > > should, in this case, rcu_user_enter() be called before tracehook_report_syscall_exit(regs, step)
> > > > > and ptrace chain?
> > > >
> > > > Well, I don't really understand this magic... but why?
> > > >
> > >
> > > My understanding is (I may be wrong)
> > 
> > Oh, I bet I have much more chances to be wrong ;)
> > 
> > > that we can schedule() from ptrace chain to
> > 
> > I don't understand how ptrace chain differs from, say, audit_syscall_exit().
> > There is nothing special in ptrace_stop() in this respect.
> >
> 
> hm.
> 
> > > some arbitrary task, which will continue its execution from the point where RCU assumes
> > > CPU as not idle, while CPU in fact still in idle state -- no one said rcu_idle_exit()
> > 
> > confused... of course it would be wrong if syscall_trace_leave() is
> > called when CPU is considered idle,
> > 
> 
> sorry, I meant idle from RCU point of view:
> 
> int rcu_is_cpu_idle(void)
> {
>          return !rcu_dynticks_nesting;
> }

Hmmm...  This reproduces on UP builds, then?

> > > if so, does the same apply to in_user?
> > 
> > Not sure we understand each other. But I believe that ->in_user should be
> > already false when syscall_trace_leave() is called.
> 
> oh, my apology. I was very wrong about this.

Frederic, thoughts?

							Thanx, Paul

--
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/