Re: [PATCH] Input: xpad - sync supported devices with 360Controller

[Benjamin Valentin <benpicco@googlemail.com>]

On Sat, 6 May 2017 14:19:58 -0700
Cameron Gutman <aicommander@gmail.com> wrote:
 
> If the device type is uncertain, we should definitely not add them to
> the the list. Doing so may regress them by overriding the
> autodetection done in xpad_probe().
>
> […]
>
> There's really no reason to add an entry like "Generic Controller".
> Gamepads work perfectly fine without being included in this list, as
> long as their VIDs are in xpad_table[], since we already derive the
> correct XTYPE for unknown pads using the USB interface descriptor in
> xpad_probe().
> 
> The only value that this table has is for gamepads that need quirks
> like MAP_TRIGGERS_TO_BUTTONS and similar, and simply having a nicer
> name to display than "Generic X-Box pad". If we don't have either, I
> don't think it's worth adding them here.

Ah I didn't know this! That's reassuring as otherwise it would have been
an impossible battle to track down all the Xbox compatible controllers
and add them to an ever growing list :)

It looks like 360Controller also detecs wheels and flight sticks at run-time,
at least for Xbox One peripherals [0]. I don't have hardware to test this
unfortunately.

> > This patch depends on my previous series which sorts the USB IDs
> > and syncs them with the xboxdrv userspace driver.  
> 
> At this point, I'd bundle all 4 patches into a v2 of your initial
> series to make it easier for Dmitry to keep them all straight.

As he already merged the three first ones, I guess I'll just resend this one
with the generic/uncertain ones stripped.

[0] https://github.com/360Controller/360Controller/blob/master/360Controller/Controller.cpp#L587