linux-kernel - Re: [PATCH v2 01/38] backlight/bd6107: Compare against struct fb

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]

Date:   Tue, 13 Jun 2023 09:31:22 +0200
From:   Thomas Zimmermann <tzimmermann@...e.de>
To:     "Ruhl, Michael J" <michael.j.ruhl@...el.com>,
        "daniel@...ll.ch" <daniel@...ll.ch>,
        "javierm@...hat.com" <javierm@...hat.com>,
        "sam@...nborg.org" <sam@...nborg.org>,
        "deller@....de" <deller@....de>,
        "geert+renesas@...der.be" <geert+renesas@...der.be>,
        "lee@...nel.org" <lee@...nel.org>,
        "daniel.thompson@...aro.org" <daniel.thompson@...aro.org>,
        "jingoohan1@...il.com" <jingoohan1@...il.com>,
        "dan.carpenter@...aro.org" <dan.carpenter@...aro.org>
Cc:     "linux-fbdev@...r.kernel.org" <linux-fbdev@...r.kernel.org>,
        Laurent Pinchart <laurent.pinchart+renesas@...asonboard.com>,
        "linux-sh@...r.kernel.org" <linux-sh@...r.kernel.org>,
        "linux-staging@...ts.linux.dev" <linux-staging@...ts.linux.dev>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "dri-devel@...ts.freedesktop.org" <dri-devel@...ts.freedesktop.org>,
        "stable@...r.kernel.org" <stable@...r.kernel.org>,
        "linux-omap@...r.kernel.org" <linux-omap@...r.kernel.org>
Subject: Re: [PATCH v2 01/38] backlight/bd6107: Compare against struct
 fb_info.device

Hi

Am 12.06.23 um 17:17 schrieb Ruhl, Michael J:
[...]
> 
> Thomas,
> 
> Looking at the fb.h file I see:
> 
> 	struct device *device;		/* This is the parent */
> 	struct device *dev;		/* This is this fb device */
> 
> Is this documentation "correct"?  If so, how does that match what you are doing here?

The comments are correct. Let's go through what's happening here.

The field 'device' is the Linux device (platform_device, pci_dev, etc.) 
and 'dev' is the fbdev character device that is /dev/fb*.

We set 'device' where we allocate the fb_info in framebuffer_alloc()

https://elixir.bootlin.com/linux/v6.3/source/drivers/video/fbdev/core/fbsysfs.c#L57

and we set 'dev' when we register the chrdev within register_framebuffer().

https://elixir.bootlin.com/linux/v6.3/source/drivers/video/fbdev/core/fbmem.c#L1555

(And the point of this patch series is to make the chrdev optional.)

The problem with bd6107 is that is misses the part where it registers 
the platform device. The driver appears to be unused.

But gpio_backlight from patches 3 and 4 works. The architecture code 
sets the 'fbdev' field from a platform-device structure at

https://elixir.bootlin.com/linux/v6.3/source/arch/sh/boards/mach-ecovec24/setup.c#L389

and later creates the platform device as part of

https://elixir.bootlin.com/linux/v6.3/source/arch/sh/boards/mach-ecovec24/setup.c#L1483

It will be used with the sh-mobile fbdev driver and become the 'device' 
field there.

In the backlight code, the gpio_backlight driver copies the fbdev field at

https://elixir.bootlin.com/linux/v6.3/source/drivers/video/backlight/gpio_backlight.c#L62

to later use it incorrectly in .check_fb. Hence, the helper has to 
compare the platform device to the 'device' field, not the 'dev' field; 
which is being fixed by this patchset.

The two other drivers, bd6107 and lv5207lp, have the same bug.

Best regards
Thomas


> 
> Thanks,
> 
> M
> 
>> }
>>
>> static const struct backlight_ops bd6107_backlight_ops = {
>> --
>> 2.41.0
> 

-- 
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nuernberg, Germany
GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
HRB 36809 (AG Nuernberg)

Download attachment "OpenPGP_signature" of type "application/pgp-signature" (841 bytes)