linux-kernel - Re: [PATCH] ARM: davinci_all_defconfig: set CONFIG_DAVINCI

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <36d2c0a8-e723-8f5d-3d1f-71c09ead54c5@ti.com>
Date:   Tue, 16 Jan 2018 19:32:34 +0530
From:   Sekhar Nori <nsekhar@...com>
To:     Adam Ford <aford173@...il.com>
CC:     David Lechner <david@...hnology.com>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ARM: davinci_all_defconfig: set CONFIG_DAVINCI_WATCHDOG=y

On Tuesday 16 January 2018 05:47 PM, Adam Ford wrote:
> On Mon, Jan 15, 2018 at 11:59 PM, Sekhar Nori <nsekhar@...com> wrote:
>> On Monday 15 January 2018 10:59 PM, David Lechner wrote:
>>> This changes CONFIG_DAVINCI_WATCHDOG from a module to a compiled-in
>>> option. Since the reset function has been moved out of the mach code in
>>> commit 0808d3260456 ("ARM: davinci: remove watchdog reset") and into the
>>> watchdog driver, devices cannot reboot unless the watchdog driver is
>>> loaded, so make it a compiled-in option so that we can always reboot, even
>>> when modules are not loaded.
>>>
>>> Cc: Sekhar Nori <nsekhar@...com>
>>> Suggested-by: Adam Ford <aford173@...il.com>
>>> Signed-off-by: David Lechner <david@...hnology.com>
>>
>> Hmm, we already depend on modules to load correctly for a lot of
>> functionality. Why should reboot be an exception? In general, unless the
>> driver is needed for loading rootfile system, I would keep it as a module.
> 
> I would argue that his patch corrects a regression introduced by the
> patch series.  Before the series, 'reboot' worked properly without the
> watchdog being loaded as a module.  With this patch series 'reboot'
> stops functioning which I would argue is a regression.  Making the
> watchdog part of the kernel fixes the broken reboot regression.  I
> would request it be accepted.

Okay, thats a convincing argument. We are going to shift to watchdog
driver for reboot only with v4.17, so will add it to my queue.

Thanks,
Sekhar