linux-kernel - Re: lockdep warning: module_mutex vs. ftrace

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 for Android: free password hash cracker in your pocket

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

Date:	Tue, 16 Jun 2009 13:56:54 -0700 (PDT)
From:	Daniel J Blueman <daniel.blueman@...il.com>
To:	Steven Rostedt <srostedt@...hat.com>
Cc:	Linux Kernel <linux-kernel@...r.kernel.org>
Subject: Re: lockdep warning: module_mutex vs. ftrace_lock

On Apr 13, 3:10 pm, Steven Rostedt <srost...@...hat.com> wrote:
> On Tue, 2009-04-07 at 20:30 +0200, Frederic Weisbecker wrote:
> > On Tue, Apr 07, 2009 at 06:34:56PM +0200, Miklos Szeredi wrote:
> > > =======================================================
> > > [ INFO: possible circular locking dependency detected ]
> > > 2.6.29-09854-gd508afb-dirty #6
> > > -------------------------------------------------------
> > > modprobe/3184 is trying to acquire lock:
> > >  (ftrace_lock){+.+.+.}, at: [<ffffffff80277e71>] ftrace_convert_nops+0x32/0x267
>
> > > but task is already holding lock:
> > >  (module_mutex){+.+.+.}, at: [<ffffffff8026a539>] sys_init_module+0x3f/0x1d3
>
> > > which lock already depends on the new lock.
>
> > > the existing dependency chain (in reverse order) is:
>
> > > -> #1 (module_mutex){+.+.+.}:
> > >        [<ffffffff8026134a>] __lock_acquire+0x12fe/0x1668
> > >        [<ffffffff802617a2>] lock_acquire+0xee/0x112
> > >        [<ffffffff804fe0a6>] mutex_lock_nested+0x4f/0x305
> > >        [<ffffffff8026709e>] module_update_tracepoints+0x1c/0x73
> > >        [<ffffffff80274f74>] tracepoint_update_probes+0x21/0x23
> > >        [<ffffffff8027510a>] tracepoint_probe_register+0x4a/0x68
> > >        [<ffffffff80277cae>] register_ftrace_graph+0x2a9/0x30f
> > >        [<ffffffff8027f928>] trace_selftest_startup_function_graph+0x2e/0xbb
> > >        [<ffffffff8027fdc5>] register_tracer+0x151/0x26e
> > >        [<ffffffff8070b942>] init_graph_trace+0x10/0x12
> > >        [<ffffffff80209066>] do_one_initcall+0x5b/0x135
> > >        [<ffffffff806fa5e4>] kernel_init+0x12f/0x185
> > >        [<ffffffff8020c2ba>] child_rip+0xa/0x20
> > >        [<ffffffffffffffff>] 0xffffffffffffffff
>
> > > -> #0 (ftrace_lock){+.+.+.}:
> > >        [<ffffffff80261067>] __lock_acquire+0x101b/0x1668
> > >        [<ffffffff802617a2>] lock_acquire+0xee/0x112
> > >        [<ffffffff804fe0a6>] mutex_lock_nested+0x4f/0x305
> > >        [<ffffffff80277e71>] ftrace_convert_nops+0x32/0x267
> > >        [<ffffffff802780bd>] ftrace_init_module+0x17/0x19
> > >        [<ffffffff8026a011>] load_module+0x1122/0x160b
> > >        [<ffffffff8026a556>] sys_init_module+0x5c/0x1d3
> > >        [<ffffffff8020b21b>] system_call_fastpath+0x16/0x1b
> > >        [<ffffffffffffffff>] 0xffffffffffffffff
>
> > Ah that's this ftrace/modules lock dependency that Steven told me about
> > recently. Steven, do you know more about it now?
>
> Yeah, I think I submitted a patch to fix that. I'll check my repo.

This is still live in 2.6.30 proper.

Let me know if you'd like further testing on your patch, or if you
can't find locate it.

Thanks,
  Daniel
--
Daniel J Blueman
--
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/