linux-kernel - Re: [BUG] 3.13-rc1 kernel crash when enable all tracepoints

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: <CAGdX0WGVDX9FzBL2tJUTx3oHBPopK74OOrFzThR_7m3GxQUs3Q@mail.gmail.com>
Date:	Tue, 26 Nov 2013 15:20:29 +0800
From:	Jovi Zhangwei <jovi.zhangwei@...il.com>
To:	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
Cc:	Steven Rostedt <rostedt@...dmis.org>,
	Oleg Nesterov <oleg@...hat.com>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [BUG] 3.13-rc1 kernel crash when enable all tracepoints

On Tue, Nov 26, 2013 at 2:33 PM, Masami Hiramatsu
<masami.hiramatsu.pt@...achi.com> wrote:
> (2013/11/26 13:38), Jovi Zhangwei wrote:
>> Hi,
>>
>> I'm not sure this issue already be fixed or not, it can be reproduced
>> permanently.
>>
>> (I didn't use git-bisect yet, you guys might can understand it quickly)
>>
>> #echo 1 > /sys/kernel/debug/tracing/events/enable
>
> Thanks for reporting. I could reproduce it.
>
> To narrow this down, I tried to run the below command
>
> [tracing]# for i in events/*/*/enable ; do echo $i; echo 1 > $i; done
>
> And it ran through the end without any problem.
> Hm, next I checked the difference of available_events and set_event.
>
> [tracing]# diff available_events set_event
> 283d282
> < ftrace:function
>
> So, I guess it was caused by enabling ftrace:function, and
> it is unable to do that via set_event, nor events/ftrace/enable
> I'm not sure how, but it seems that ftrace:function can be
> enabled by the events/enable.
>
I'm not sure this relate with ftrace:function event, from the
crash log, it seems more like caused by lock events.

Now I only enable lock events (lockdep compiled in my box), below
two commands both can crash system.

#echo 1 > /sys/kernel/debug/tracing/events/lock/lock_acquire/enable
or.
#echo 1 > /sys/kernel/debug/tracing/events/lock/lock_release/enable

According to the log, it seems like a recursion tracing bug.

register lock event -> jump_label_update -> text_poke_bp ->
on_each_cpu -> local_apic_timer_interrupt -> ktime_get_update_offsets
-> lock_release

(Perhaps the crash you reproduced is another crash? similar crash log?)

Jovi
--
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/