Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
In-Reply-To: <87y3ekafcw.fsf@ashishki-desk.ger.corp.intel.com>
References: <000000000000ccdf0605708a9ccc@google.com> <87y3ekafcw.fsf@ashishki-desk.ger.corp.intel.com>
From:   Dmitry Vyukov <dvyukov@google.com>
Date:   Mon, 9 Jul 2018 13:02:21 +0200
Message-ID: <CACT4Y+Yd+8qSFZVU1LydP7gaRSRiZ7qTceri+TkVVCcpOeP+YA@mail.gmail.com>
Subject: Re: KASAN: use-after-free Write in _free_event
To:     Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc:     syzbot <syzbot+a24c397a29ad22d86c98@syzkaller.appspotmail.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        jolsa@redhat.com, LKML <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@redhat.com>,
        Namhyung Kim <namhyung@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        syzkaller-bugs <syzkaller-bugs@googlegroups.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jul 9, 2018 at 12:11 PM, Alexander Shishkin
<alexander.shishkin@linux.intel.com> wrote:
> syzbot <syzbot+a24c397a29ad22d86c98@syzkaller.appspotmail.com> writes:
>
>> Hello,
>>
>> syzbot found the following crash on:
>>
>> HEAD commit:    b2d44d145d2a Merge tag '4.18-rc3-smb3fixes' of git://git.s..
>> git tree:       upstream
>> console output: https://syzkaller.appspot.com/x/log.txt?x=178cf50c400000
>> kernel config:  https://syzkaller.appspot.com/x/.config?x=2ca6c7a31d407f86
>> dashboard link: https://syzkaller.appspot.com/bug?extid=a24c397a29ad22d86c98
>> compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
>>
>> Unfortunately, I don't have any reproducer for this crash yet.
>
> Is there a chance of getting a reproducer for this one?

There is a chance. Frequently syzbot finds a reproducer after some
time. What's the chance in this case I don't know. Since it happened
only once so far, probably not too high.

But there seems to be a good hint in the KASAN report: task was freed
right inside of fork/copy_process, probably some error happened, but
it seems to have been registered in some global list already and
perf_release discovered it there. Does it make sense? Was it
registered? What should have been prevented the task alive at the time
of access?