Re: [PATCH v1 0/7] HID: playstation: DS4: LED bugfix, third-party gamepad support

[Max Staudt]

On 1/25/24 07:24, Roderick Colenbrander wrote:
> Sorry for the late reply. I had glanced over them, but didn't have an
> opportunity for a detailed review yet.

No worries, thank you for giving the patches a chance.


> My general fear is a balance between supporting clone devices vs
> reliability. This driver is heavily used in devices (phones, tablets,
> TVs, cars). There have been bug reports in the past and just getting
> the fixes downstream takes a lot of time (e.g. Android devices).

I understand this very well, which is why I hope I've kept the patches small enough to be alright to follow. If you have a bad feeling about something in particular, please let me know, and maybe we can find a better solution or alleviate concerns.

On my end, I am working (currently as a hobby) on an appliance that aims to support as many controllers as possible, hence I know the choices you face in order to keep user expectations in check, and user experience consistent. See my comment on patch 7/7 - what if someone uses a third-party controller that happens to work by chance, and a kernel update breaks stuff?

So I am with you on focusing on the original devices, which have predictable behaviour and quality. That said, hopefully some or all of these patches are trivial enough to be included upstream. I feel that it's beautiful to plug random stuff into a Linux box and to find that it just works :)


> One of the key things I really would like to see enhanced are the unit
> tests (hid-tools / kernel side now). To really make sure we emulate
> behavior of these other devices well. The tricky part is that they
> don't always support all the HID requests of the real device (which is
> weird as the game console does use those HID reports and others and I
> don't know how it would have worked there).

I've been wondering the same, but without a PS4 of my own, I didn't try such controllers on a real console. That would be interesting indeed!


> That's in general the key feedback about the tests. A question for
> Max: do you have access to all the devices being added?

I don't have all devices I've ever tested, but I do have at least one reproducer for each patch in this series.



Thanks,

Max