linux-kernel - Re: process 'stuck' at exit.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20131210163925.6f5ecad1@gandalf.local.home>
Date:	Tue, 10 Dec 2013 16:39:25 -0500
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Thomas Gleixner <tglx@...utronix.de>
Cc:	Darren Hart <dvhart@...ux.intel.com>,
	Dave Jones <davej@...hat.com>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrea Arcangeli <aarcange@...hat.com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Mel Gorman <mgorman@...e.de>, Oleg Nesterov <oleg@...hat.com>
Subject: Re: process 'stuck' at exit.

On Tue, 10 Dec 2013 22:28:01 +0100 (CET)
Thomas Gleixner <tglx@...utronix.de> wrote:

> On Tue, 10 Dec 2013, Darren Hart wrote:
> > On Tue, 2013-12-10 at 15:55 -0500, Dave Jones wrote:
> > > On Tue, Dec 10, 2013 at 09:34:24PM +0100, Thomas Gleixner wrote:
> > >  
> > >  > >  > PS: Oleg - the whole thread is on lkml. Ping me if you need more context.
> > >  > > 
> > >  > > btw, I've left the machine in that state, and will for as long as necesary
> > >  > > in case someone has any ideas for further tracing experiments.
> > >  > 
> > >  > Can you gather a trace with the function tracer? That will tell us
> > >  > what the thing is actually doing.
> > > 
> > > Feed me command lines, and I'll see what I can do.
> > > 
> > > 	Dave
> > > 
> > 
> > So let's see if we can determine the call chain inside futex_requeue, if
> > there is one. We want to see if we are calling the following functions:
> > 
> > futex.c: free_pi_state
> > futex.c: get_futex_value_locked
> > futex.c: futex_proxy_trylock_atomic
> > futex.c: lookup_pi_state
> > futex.c: fault_in_user_writeable
> > 
> > I'm hoping we can make use of ftrace here?
> > 
> > Consider:
> > 
> > $ trace-cmd record -P <PID> -p function -l '*futex*'

Note the above just picks functions that have "futex" in its name, and
will not include all the functions you listed above.

> > 
> > And to try outside of futex:
> > 
> > $ trace-cmd record -P <PID> -p function
> > 
> > Then ^C after a few seconds. The trace will be in trace.dat
> 
> We definitely want the latter. The futex code gets compiled into some
> badly traceable mess.

Yeah, lets use all functions. If its in a loop, we should see exactly
what is happening, with full function tracing.

You may be able to simplify it to just:

	trace-cmd record -p function sleep 10

and then read the trace. Nasty loopers usually dominate these types of
traces so no need to filter on them in the recording.

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