linux-kernel - Re: [PATCH] platform/x86: int3472: Fix double free of GPIO device during unregister

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: <aQDXfk1BZIxD4H8S@smile.fi.intel.com>
Date: Tue, 28 Oct 2025 16:47:26 +0200
From: Andy Shevchenko <andriy.shevchenko@...el.com>
To: Hans de Goede <hansg@...nel.org>
Cc: Dan Scally <dan.scally@...asonboard.com>,
	Qiu Wenbo <qiuwenbo@...me.org>, Daniel Scally <djrscally@...il.com>,
	Qiu Wenbo <qiuwenbo@...insec.com.cn>,
	platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
	Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>,
	Sakari Ailus <sakari.ailus@...ux.intel.com>,
	Andy Shevchenko <andy@...nel.org>
Subject: Re: [PATCH] platform/x86: int3472: Fix double free of GPIO device
 during unregister

On Tue, Oct 28, 2025 at 03:36:56PM +0100, Hans de Goede wrote:
> On 28-Oct-25 12:38 PM, Andy Shevchenko wrote:
> > On Tue, Oct 28, 2025 at 11:09:12AM +0000, Dan Scally wrote:
> >> On 28/10/2025 10:54, Andy Shevchenko wrote:
> >>> On Tue, Oct 28, 2025 at 11:38:00AM +0100, Hans de Goede wrote:
> >>>> On 28-Oct-25 11:02 AM, Andy Shevchenko wrote:
> >>>>> On Tue, Oct 28, 2025 at 08:55:07AM +0000, Dan Scally wrote:
> >>>>>> On 24/10/2025 06:05, Qiu Wenbo wrote:

...

> >>>>>> However the Fixes tag I wonder about; devm_gpiod_get() will also result in a
> >>>>>> call to gpiod_put() when the module is unloaded; doesn't that mean that the
> >>>>>> same issue  will occur before that commit?
> >>>>>
> >>>>> Actually a good question! To me sounds like it's a bug(?) in regulator code.
> >>>>> It must not release resources it didn't acquire. This sounds like a clear
> >>>>> layering violation.
> >>>>
> >>>> I think the problem is that when it comes from devicetree it is acquired
> >>>> by the regulator core.
> >>>
> >>> Hmm... I probably missed that, but I failed to see this. Any pointers?
> >>
> >> They can come through the struct regulator_desc.of_parse_cb(), which is called in
> >> regulator_of_init_data(), from regulator_register(). For example: https://elixir.bootlin.com/linux/v6.17.5/source/drivers/power/supply/mt6370-charger.c#L234>
> > 
> > Ah, thank you, Dan, for the pointers. Indeed, that's how it's done. Hmm, still
> > why can't we let the regulator consumer to decide when to clean the resource?
> > I think this is an attempt to have a refcounting against shared GPIO resource
> > and it should be done in the GPIOLIB (if not yet). In regulator that put
> > call should probably be conditional (based on the source of GPIO request).
> 
> Fixing this sounds like a somewhat big undertaking. In the mean time
> I think we should move forward with this patch to fix the immediate
> issue with the double free.

And I am not objecting to apply this as you may see in the tags given so far.
In regard to the undertaking it seems in Bart's (GPIOLIB maintainer) TODO list.

-- 
With Best Regards,
Andy Shevchenko