linux-kernel - Re: [PATCH v3] mfd: syscon: Support early initialization

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: <20140423151932.GD12304@sirena.org.uk>
Date:	Wed, 23 Apr 2014 16:19:32 +0100
From:	Mark Brown <broonie@...nel.org>
To:	Michal Simek <monstr@...str.eu>
Cc:	Lee Jones <lee.jones@...aro.org>,
	Michal Simek <michal.simek@...inx.com>,
	linux-kernel@...r.kernel.org, pankaj.dubey@...sung.com,
	Samuel Ortiz <sameo@...ux.intel.com>, arnd@...db.de
Subject: Re: [PATCH v3] mfd: syscon: Support early initialization

On Wed, Apr 23, 2014 at 03:04:28PM +0200, Michal Simek wrote:

> In early_syscon_probe np->data contains pointer to struct syscon which
> stores base, regmap and res - driver private data for system controller.
> (init in early_syscon_probe)
> This is the way I am aware of how to share driver private data without pdev.
> Maybe there is better way how to do it that's why please let me know
> if you are aware about it.

The main issue I'm seeing is that the fact that the pointer in the node
is being used without a reference being held; I'd expect something to
explicitly own the reference.  Either hold the reference somewhere and
hand it off between the various stages or free the syscon object and
allocate a new one each time it's used prior to the device being
instantiated.

Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)