Re: [PATCH v4 2/4] dt-bindings: media: add Maxim MAX96714 GMSL2 Deserializer

[Conor Dooley <conor@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 2064 bytes --]

On Fri, Mar 08, 2024 at 04:48:16PM +0100, Julien Massot wrote:
> 
> 
> On 3/8/24 16:07, Conor Dooley wrote:
> > On Fri, Mar 08, 2024 at 03:08:12PM +0100, Julien Massot wrote:
> > > On 3/7/24 20:21, Conor Dooley wrote:
> > > > On Tue, Mar 05, 2024 at 04:26:06PM +0100, Julien Massot wrote:
> > > > > Add DT bindings for Maxim MAX96714 GMSL2 Deserializer.
> > > > > 
> > > > > Signed-off-by: Julien Massot <julien.massot@collabora.com>
> > > > > ---
> > > > > Change since v3:
> > > > >    - Renamed file to maxim,max96714.yaml dropped the 'f' suffix
> > > > 
> > > > Why? The filename should match the compatible, which /does/ have an f.
> > > All the work has been done on MAX96714F variant of this Maxim GMSL2
> > > deserializer.
> > > The driver and the binding remain suitable for all variants of this chipset,
> > > since they share the same
> > > register mapping, similar features etc..
> > > 
> > > MAX96714 exists in different variant: MAX96714 / MAX96714F / MAX96714K that
> > > will be easy
> > > to add support for this binding and driver later.
> > 
> > Either document the non-f version if it really is that similar, using
> > all of the same properties, or name the file after the version you've
> > actually documented. I don't see why this particular case should be
> > given an exception to how bindings are named.
> > 
> > What is the actual difference between the f and non f versions? Is it
> > visible to software?
> 
> Yes there are a few differences visible to the software, for example the
> GMSL
> link rate since MAX96714 support 6 and 3 Gbps, while MAX96714F only supports
> 3Gbps.
> the registers map is the same, but a few values are not possible with the
> 'f' version.
> 
> I will add a compatible for the non 'f' version, and will do the same for
> the max96717 binding.

It's not immediately clear if that means that the f version should be a
fallback for the non-f version, but sounds like it could be if the
difference is purely that there's a reduced set of link rates or
similar.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]