linux-kernel - Re: [PATCH 1/2] media: ir-kbd-i2c: prevent potential NULL pointer access

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] [day] [month] [year] [list]

Message-ID: <20190725104453.5kqusisyg44zbbcy@gofer.mess.org>
Date:   Thu, 25 Jul 2019 11:44:53 +0100
From:   Sean Young <sean@...s.org>
To:     Wolfram Sang <wsa@...-dreams.de>
Cc:     Wolfram Sang <wsa+renesas@...g-engineering.com>,
        linux-i2c@...r.kernel.org,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] media: ir-kbd-i2c: prevent potential NULL pointer
 access

Hi Wolfram,

On Thu, Jul 25, 2019 at 09:55:38AM +0200, Wolfram Sang wrote:
> Hi Sean,
> 
> thanks for the review!
> 
> On Thu, Jul 25, 2019 at 06:12:02AM +0100, Sean Young wrote:
> > On Mon, Jul 22, 2019 at 07:26:31PM +0200, Wolfram Sang wrote:
> > > i2c_new_dummy() can fail returning a NULL pointer. The code does not
> > > bail out in this case and the returned pointer is blindly used.
> > 
> > I don't see how. The existing code tries to set up the tx part; if
> > i2c_new_dummy() return NULL then the rcdev is registered without tx,
> > and tx_c is never used.
> 
> Yes, you are totally right. I missed that the send_block function is
> also only called iff zilog_init succeeded. Thanks for the heads up and
> sorry for the noise.

Not at all, thank you for the patch.

> > > Convert
> > > to devm_i2c_new_dummy_device() which returns an ERR_PTR and also bail
> > > out when failing the validity check.
> > 
> > Possibly I was being overly cautious with not bailing out if tx can't
> > be registered; moving to devm is probably a good idea. However the
> > commit message is misleading, because the existing code has no
> > NULL pointer access.
> 
> Yep, I will resend with a proper commit message. Technically, there is
> no need to bail out anymore because there is no NULL pointer access. My
> tendency is now to not bail out and keep the old behaviour (registering
> without tx). What do you think?

Since I write this code I've got pretty much every model with this zilog
transmitter/receiver, and they all work fine, including different firmware
versions. If there is a problem it would be nice to hear about it, and
not silently swallow the error. I think.

Thanks,

Sean