Re: possible deadlock in blkdev_reread_part

[Dmitry Vyukov]

On Thu, Sep 13, 2018 at 3:43 PM, Tetsuo Handa
<penguin-kernel@xxxxxxxxxxxxxxxxxxx> wrote:
> On 2018/09/13 21:58, Dmitry Vyukov wrote:
>> On Wed, May 2, 2018 at 1:23 PM, Dmitry Vyukov <dvyukov@xxxxxxxxxx> wrote:
>>> On Wed, May 2, 2018 at 12:30 PM, Tetsuo Handa
>>> <penguin-kernel@xxxxxxxxxxxxxxxxxxx> wrote:
>>>> Dmitry Vyukov wrote:
>>>>>> syzbot is reporting various bugs which involve /dev/loopX.
>>>>>> Two of them
>>>>>>
>>>>>>   INFO: rcu detected stall in lo_ioctl
>>>>>>   https://syzkaller.appspot.com/bug?id=7b49fb610af9cca78c24e9f796f2e8b0d5573997
>>>>>>
>>>>>>   general protection fault in lo_ioctl (2)
>>>>>>   https://syzkaller.appspot.com/bug?id=f3cfe26e785d85f9ee259f385515291d21bd80a3
>>>>>
>>>>> /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
>>>>>
>>>>> Now there is a repro for this one. I've pushed it to kernel mailing lists:
>>>>>
>>>>> https://groups.google.com/d/msg/syzkaller-bugs/c8KUcTAzTvA/3o_7g6-tAwAJ
>>>>
>>>> OK, thanks. But among loop related reports, this will be a dup of
>>>> "INFO: rcu detected stall in blkdev_ioctl" which already has C reproducer.
>>>> Should we merge them?
>>>
>>> Yes, sure, I will take care of it.
>>
>> 1. I forgot to take care of it.
>>
>> 2. "INFO: rcu detected stall in blkdev_ioctl" was fixed 3 months ago:
>> https://syzkaller.appspot.com/bug?id=1f7b710f4110f225aed1f4263ec2b98b8dbd472e
>>
>> but this bug still happens up until now:
>> https://syzkaller.appspot.com/bug?id=bf154052f0eea4bc7712499e4569505907d15889
>>
>> So this is a different bug.
>
> I'm not sure what you are talking about.
> RCU stall and lockdep warning are obviously different bugs.

Ah, OK, I misinterpreted your "But among loop related reports, this
will be a dup of "INFO: rcu detected stall in blkdev_ioctl"".
So this is just an independent bug that still happens.

> Regarding lockdep warning on loop module, we are still waiting for Jens to
> come up a better alternative than
> http://lkml.kernel.org/r/1527297408-4428-1-git-send-email-penguin-kernel@xxxxxxxxxxxxxxxxxxx .
> Since no alternative was proposed, I think we should start testing my patch.
>