linux-kernel - Re: [PATCH v2 0/4] gpio: fix SRCU bugs

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: <CAMRc=McXr_3OD6Uf+PKD-wLQfJKJSWes3fh_ZD9fnrPyF9GoSQ@mail.gmail.com>
Date: Wed, 14 Feb 2024 20:08:33 +0100
From: Bartosz Golaszewski <brgl@...ev.pl>
To: paulmck@...nel.org
Cc: Linus Walleij <linus.walleij@...aro.org>, Kent Gibson <warthog618@...il.com>, 
	Alex Elder <elder@...aro.org>, Geert Uytterhoeven <geert+renesas@...der.be>, 
	Andy Shevchenko <andriy.shevchenko@...ux.intel.com>, Wolfram Sang <wsa@...-dreams.de>, 
	Mark Brown <broonie@...nel.org>, Dan Carpenter <dan.carpenter@...aro.org>, 
	linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org, 
	Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH v2 0/4] gpio: fix SRCU bugs

On Wed, Feb 14, 2024 at 7:44 PM Paul E. McKenney <paulmck@...nel.org> wrote:
>
> On Wed, Feb 14, 2024 at 09:44:15AM +0100, Bartosz Golaszewski wrote:
> > From: Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
> >
> > Here are four fixes to some bugs in recent SRCU changes. The first one fixes
> > an actual race condition. The other three just make lockdep happy.
>
> For 1/4-3/4:
>
> Acked-by: Paul E. McKenney <paulmck@...nel.org>
>
> For 4/4, you are playing with fire, but I will assume that you know what
> you are doing.  ;-)
>

Up until this rework, this gdev->chip pointer could go from under any
user at any point. Now we have this gpio_device wrapper that provides
an entry point to using the chip safely while protected by the SRCU
read lock. Anyone who is still accessing gpio_chip directly (and not
being the GPIO provider themselves) is asking for trouble. There's
however no point in spamming lockdep splats in this case. I may end up
adding a warning to these routines.

Unfortunately, it's hard to fix 15 years of technical debt. :(

Thanks for the Acks.
Bartosz

[snip]