On 02/03/2025 05:20, Jonathan Cameron wrote:

On Thu, 27 Feb 2025 09:46:06 +0200
Matti Vaittinen [off-list ref] wrote:

quoted

On 26/02/2025 18:10, David Lechner wrote:

quoted

On 2/26/25 12:28 AM, Matti Vaittinen wrote:

...

So today the situation is we have all the options in tree and we aren't
really in a position to drop any of them:

Sure. I am only really interested whether we want to prefer some 
approach for (majority of) new drivers. Furthermore, I believe there 
will always be corner cases and oddities which won't fit to the 'de 
facto' model. That doesn't mean we shouldn't help those which don't have 
such 'oddities' to work with some generic code.

Hindsight is a wonderful thing.  I'm not sure on what policy we should have
gone for, but now we are kind of stuck with this slightly messy situation.

Sorry if my comments came out as criticism. It was not intention, I just 
try to justify the helpers by trying to think what new drivers should 
prefer.

Helper wise if it expands usefulness we may want a bool parameter to say
if we skip the missing or not + make sure a max expected channel is provided
(might already be - I didn't check!)

This far it only had (optional) maximum channel ID for sanity checking 
(useful for callers which use the ID to index an array). The bool 
parameter would also require a parameter specifying the number of 
expected channels. That'd make 3 parameters which may be used or unused.

I don't think I saw existing code which would have used these 
parameters. It might be cleaner to add new APIs when we get such 
use-cases. That should simplify the use for current cases.

Thank You for the long explanation of current system + the history :) I 
appreciate your guidance!

Yours,
	-- Matti