linux-kernel - Re: [PATCH 2/7] iio: light: tsl2583: change functions to only have a single exit point

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]

Date:   Wed, 19 Oct 2016 15:51:25 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Brian Masney <masneyb@...tation.org>
Cc:     devel@...verdev.osuosl.org, lars@...afoo.de,
        linux-iio@...r.kernel.org, gregkh@...uxfoundation.org,
        linux-kernel@...r.kernel.org, pmeerw@...erw.net, knaack.h@....de,
        jic23@...nel.org
Subject: Re: [PATCH 2/7] iio: light: tsl2583: change functions to only have a
 single exit point

On Wed, Oct 19, 2016 at 08:38:16AM -0400, Brian Masney wrote:
> On Wed, Oct 19, 2016 at 02:08:59PM +0300, Dan Carpenter wrote:
> > On Wed, Oct 19, 2016 at 06:32:05AM -0400, Brian Masney wrote:
> > > Change the following functions to only have a single exit point:
> > > taos_i2c_read(), taos_als_calibrate(), taos_chip_on(),
> > > taos_gain_store(), taos_gain_available_show(), taos_luxtable_store()
> > > and taos_probe().
> > > 
> > 
> > What's the point of this?  This style of code just makes things more
> > complicated and leads to "forgot the error code" bugs.  People think
> > that it future proofs the code in case we add locking but I have looked
> > into this and it has minimal if any impact at preventing locking bugs.
> 
> The reason that I did this was due to the locking that I added later in
> the patch series. Each function would only have a single call to
> mutex_unlock(). I should have mentioned that in my message.

This kind of trick does not work in real life.  I have looked at it.
In theory, it should work but in real life, empirically, it does not
and it introduces additional new bugs.

People are always looking for a magic bullet but there is no such thing.
The best approach is just to write the simplest, clearest code that you
can.

regards,
dan carpenter