linux-kernel - Re: [PATCH 1/5] Input: ambakmi

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: <20170326152354.GA11555@leoy-linaro>
Date:   Sun, 26 Mar 2017 23:23:54 +0800
From:   Leo Yan <leo.yan@...aro.org>
To:     Russell King - ARM Linux <linux@...linux.org.uk>
Cc:     Dmitry Torokhov <dmitry.torokhov@...il.com>,
        Alessandro Zummo <a.zummo@...ertech.it>,
        Alexandre Belloni <alexandre.belloni@...e-electrons.com>,
        Linus Walleij <linus.walleij@...aro.org>,
        Baptiste Reynal <b.reynal@...tualopensystems.com>,
        Alex Williamson <alex.williamson@...hat.com>,
        Jaroslav Kysela <perex@...ex.cz>,
        Takashi Iwai <tiwai@...e.com>, linux-input@...r.kernel.org,
        linux-kernel@...r.kernel.org, rtc-linux@...glegroups.com,
        linux-arm-kernel@...ts.infradead.org, kvm@...r.kernel.org,
        alsa-devel@...a-project.org
Subject: Re: [PATCH 1/5] Input: ambakmi - Convert to use devm_*()

On Sun, Mar 26, 2017 at 03:50:24PM +0100, Russell King - ARM Linux wrote:
> On Sun, Mar 26, 2017 at 10:41:50PM +0800, Leo Yan wrote:
> > Convert driver to use devm_*() APIs so rely on driver model core layer
> > to manage resources. This eliminates error path boilerplate and makes
> > code neat.
> > 
> > This patch also fixes the memory leakage for 'kmi->io' when remove
> > module.
> 
> No, it is not leaked, in fact, your patch introduces a use-after-free
> bug.
> 
> kmi->io is of type "struct serio", and this structure has an embedded
> "struct device".  The lifetime of any structure containing a "struct
> device" is controlled by the lifetime of the struct device.
> 
> Once serio_register_port() has been called on it, it is up to the
> serio layer to free this structure.
> 
> Therefore, your patch creates a bug, and so it gets a NAK.  There is no
> resource leakage here.

Thanks for reviewing. Now I understood and please ignore this patch
and patch 0005.

Thanks,
Leo Yan