MIME-Version: 1.0
In-Reply-To: <1491527002.4166.148.camel@kernel.crashing.org>
References: <1491462281.2815.47.camel@neuling.org> <CAL_JsqKe7SdO+Bg219v02Jve86zGs=f-2R0AHnhCtOKjb_bB3A@mail.gmail.com>
 <1491527002.4166.148.camel@kernel.crashing.org>
From: Rob Herring <robh@kernel.org>
Date: Fri, 7 Apr 2017 09:03:28 -0500
Message-ID: <CAL_JsqJ4U8B=E7AiJ_+SSKKKFcWLEx0uD2+Dkp67CLKRX-PCDg@mail.gmail.com>
Subject: Re: tty crash in tty_ldisc_receive_buf()
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michael Neuling <mikey@neuling.org>, Al Viro <viro@zeniv.linux.org.uk>,
        johan Hovold <johan@kernel.org>,
        Peter Hurley <peter@hurleysoftware.com>,
        Wang YanQing <udknight@gmail.com>,
        Alexander Popov <alex.popov@linux.com>,
        Mikulas Patocka <mpatocka@redhat.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 625
Lines: 21

On Thu, Apr 6, 2017 at 8:03 PM, Benjamin Herrenschmidt
<benh@kernel.crashing.org> wrote:
> On Thu, 2017-04-06 at 08:28 -0500, Rob Herring wrote:
>>
>> Can you try this one [1].
>>
>> Rob
>>
>> [1] https://lkml.org/lkml/2017/3/23/569
>
> Unless I missed something, that patch doesn't change barriers or
> locking, so I don't see how it could fix anything ...

Only because that's a code path that I changed recently and the
locking and ordering appears to be tricky.

> Mind you we haven't quite figured out exactly what's happening here.

Is this with a serial port? There were some changes to serial_core close in 4.9.

Rob