linux-kernel - Re: [PATCH] nokia n900: update dts with camera support

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: <20170828150725.nvrr2c3a2lir5xic@valkosipuli.retiisi.org.uk>
Date:   Mon, 28 Aug 2017 18:07:25 +0300
From:   Sakari Ailus <sakari.ailus@....fi>
To:     Pavel Machek <pavel@....cz>
Cc:     Sebastian Reichel <sre@...nel.org>, pali.rohar@...il.com,
        kernel list <linux-kernel@...r.kernel.org>,
        linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>,
        linux-omap@...r.kernel.org, tony@...mide.com, khilman@...nel.org,
        aaro.koskinen@....fi, ivo.g.dimitrov.75@...il.com,
        patrikbachan@...il.com, serge@...lyn.com, abcloriens@...il.com
Subject: Re: [PATCH] nokia n900: update dts with camera support

Hi Pavel,

On Mon, Aug 28, 2017 at 03:59:59PM +0200, Pavel Machek wrote:
> Hi!
> 
> > > > Well, strobe property above already uses = <0>/<1> format, as do
> > > > others.
> > >
> > > > Problem with "false = property does not exist" is that you don't know
> > > > if it is "someone forgot to define it" or "someone made a typo" or
> > > > "dts is too old to know about this property" or "the property indeed
> > > > should be false"...
> > >
> > > As this is an established practice, I think we should follow it for bool
> > > properties.
> > 
> > Yes it's common practice, there is also device_property_read_bool()
> > to get the value.
> > 
> > > We could change the existing ones, too, and leave some extra checks in
> > > place to handle old dtbs.
> > 
> > The following should be downward compatible:
> > 
> > var = read_bool();
> > if (var && !read_int())
> >     var = false;
> > 
> > Btw. DT people should be CC'd for DT binding additions/changes.
> 
> Ok, so for now we do crc=1 by default. Now we'd like to have crc
> support configurable in the dts. But if we just introduce "crc;"
> option, it will break old dts users. We could introduce "no-crc;" and
> that would work in this particular case, but will break when we want
> different defaults at different devices.
> 
> Anyway, introducing "no-crc;" seems pretty ugly to me. I'd rather do
> "crc=<0>;", in a similar way we handle other options now.

The same issue actually exists for CSI-2: the CRC is typically enabled and
I think you can at least disable the check (in OMAP3 ISP). Do we have a
need to disable it at the moment for any purpose? I think I've seen this
was disabled somewhere but it may just as well be a thoughtlessly written
configuration.

-- 
Regards,

Sakari Ailus
e-mail: sakari.ailus@....fi