[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAB0TPYGLW-bCRBTj--Y3xrgridzzR0V4Q+OCa6wv+yCi7RDaBA@mail.gmail.com>
Date: Tue, 24 Apr 2018 10:30:49 +0200
From: Martijn Coenen <maco@...roid.com>
To: Jan Kara <jack@...e.cz>
Cc: syzbot <syzbot+e38306788a2e7102a3b6@...kaller.appspotmail.com>,
amir73il@...il.com, Dan Carpenter <dan.carpenter@...cle.com>,
dwindsor@...il.com, elena.reshetova@...el.com,
LKML <linux-kernel@...r.kernel.org>, mszeredi@...hat.com,
syzkaller-bugs <syzkaller-bugs@...glegroups.com>
Subject: Re: INFO: task hung in fsnotify_mark_destroy_workfn
On Wed, Apr 18, 2018 at 11:36 AM, Jan Kara <jack@...e.cz> wrote:
> OK, so we are waiting for the grace period on fsnotify_mark_srcu. Seems
> like someone is holding fsnotify_mark_srcu too long or srcu period cannot
> finish for some other reason. However the reproducer basically contains
> only one binder ioctl and I have no idea how that's connected with fsnotify
> in any way.
Yeah, binder isn't involved with fsnotify at all.
> So either the reproducer is wrong, or binder is corrupting
> memory and fsnotify is just a victim, or something like that...
>From the dumped stacks I don't see how binder is involved in that task
getting hung up, so either it's not related to binder after all, or it
is indeed some form of corruption, but then I wouldn't expect it to
reproduce in the same way every time (or otherwise expect KASAN to
catch it). I'll try to see if I can repro it locally.
Martijn
>
> Honza
> --
> Jan Kara <jack@...e.com>
> SUSE Labs, CR
Powered by blists - more mailing lists