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
References: <20190719135352.GF4240@sasha-vm> <20190807114649.fjfaj4oytcxaua7o@linutronix.de>
 <CADLLry6a9a0TKOEEPxOW_DS+XXwk5qUuaH+W9cmbLwvudXAV8A@mail.gmail.com> <alpine.DEB.2.21.1908071452350.24014@nanos.tec.linutronix.de>
In-Reply-To: <alpine.DEB.2.21.1908071452350.24014@nanos.tec.linutronix.de>
From:   Austin Kim <austindh.kim@gmail.com>
Date:   Wed, 7 Aug 2019 22:33:31 +0900
Message-ID: <CADLLry5bKjVB5BM7HVF8wOPXkQWpQ-6xM=F78MDRHhpCxE=uBw@mail.gmail.com>
Subject: Re: NULL ptr deref in wq_worker_sleeping on 4.19
To:     Thomas Gleixner <tglx@linutronix.de>
Cc:     Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Sasha Levin <sashal@kernel.org>, peterz@infradead.org,
        mingo@kernel.org, tj@kernel.org, jiangshanlai@gmail.com,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        =?UTF-8?B?I+yEnO2YleynhA==?= <mike.seo@lge.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Is it possible for you to generate kernel patch only with **kernel
log** for 'use-after-free' case?
With kernel log, we only have limited information, so it is hard to
debug 'use-after-free' case with this signature.

For this matter, kernel should be compiled with below config.
CONFIG_SLUB_DEBUG=3Dy
CONFIG_SLUB_DEBUG_ON=3Dy

2019=EB=85=84 8=EC=9B=94 7=EC=9D=BC (=EC=88=98) =EC=98=A4=ED=9B=84 9:56, Th=
omas Gleixner <tglx@linutronix.de>=EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=B1:
>
> On Wed, 7 Aug 2019, Austin Kim wrote:
>
>   A: Because it messes up the order in which people normally read text.
>   Q: Why is top-posting such a bad thing?
>   A: Top-posting.
>   Q: What is the most annoying thing in e-mail?
>
>   A: No.
>   Q: Should I include quotations after my reply?
>
>   http://daringfireball.net/2007/07/on_top
>
> > I wonder what kinds of workqueue is used in case of this panic.
> >
> > If system workqueue(system_wq) is used for this case, it would be a
> > help to replace it with high priority workqueue(system_highpri_wq). If
> > panic disappers with high priority workqueue(system_highpri_wq), we
> > would think about another scenario.
>
> How would that help? As Sebastian explained, something overwrote memory o=
r
> it is a Use After Free. How would a high priority workqueue 'fix' that?
>
> You need to find the root cause, which is either memory corruption or a u=
se
> after free.
>
> Thanks,
>
>         tglx