linux-kernel - Re: [PATCH v2] i2c: Use u8 type in i2c transfer calls

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: <20221020113541.0000490e@huawei.com>
Date:   Thu, 20 Oct 2022 11:35:41 +0100
From:   Jonathan Cameron <Jonathan.Cameron@...wei.com>
To:     Wolfram Sang <wsa-dev@...g-engineering.com>
CC:     Jason Gerecke <killertofu@...il.com>,
        Andy Shevchenko <andy.shevchenko@...il.com>,
        linux-i2c <linux-i2c@...r.kernel.org>,
        linux-iio <linux-iio@...r.kernel.org>,
        Jonathan Cameron <jic23@...nel.org>,
        "Lars-Peter Clausen" <lars@...afoo.de>,
        LKML <linux-kernel@...r.kernel.org>,
        Ping Cheng <pinglinux@...il.com>,
        "Tobita, Tatsunosuke" <tatsunosuke.tobita@...om.com>,
        Jason Gerecke <jason.gerecke@...om.com>,
        Ping Cheng <ping.cheng@...om.com>
Subject: Re: [PATCH v2] i2c: Use u8 type in i2c transfer calls

On Wed, 19 Oct 2022 23:37:51 +0200
Wolfram Sang <wsa-dev@...g-engineering.com> wrote:

> > I spent a little time trying to put together a Coccinelle script to
> > take care of everything but I eventually realized the size of the task
> > was larger than I was comfortable with. In particular, even though I
> > might be able to put together a script, I worry I don't have a good
> > way to test the resulting treewide changes to avoid regression.  
> 
> The coccinelle scripts are one thing. I am quite familiar with it, so I
> regard this as "work but doable". My main headache is that I am not sure
> about the best way to upstream the result. I'd like to avoid a flag-day
> where all drivers across all subsystems need to be converted, but I
> don't really see a way around it. Preparing such a branch and make sure
> it does not regress is quite some work on a moving target.

Horrendous though it is, you 'could' take it via a void * intermediate
step.   That way all the warnings will disappear (I think).
You then move all the callers to providing u8 * then switch the function
to that.  Could happen over several cycles with coccicheck moaning about
any new entries in the meantime.

Jonathan


> 
>