Re: [RFC, PATCH] HID: hid-sony, override usbhid_output_raw_report for Sixaxis

[Alan Ott <alan@signal11.us>]

On 10/14/2010 03:48 AM, Antonio Ospite wrote:
>
> The point about overriding hid_output_raw_report() is not about
> sixaxis_set_operational_usb() sending a control message, but about any
> _output_ (not feature) report the user would like to send via
> hidraw_write(), the sixaxis will accept those only if sent on the
> control endpoint (it will discard them if sent on the interrupt ep,
> like the current usbhid code does) maybe I didn't state that clearly,
> sorry.
>    

Ok, that makes a lot more sense then. You're saying it's not about the 
code _in_ your driver, it's about userspace code accessing the sixaxis 
via hidraw.

My personal preference is that I like the override idea better than the 
quirk. An override will be slightly faster, because of not having to 
check for the quirk in non-sixaxis cases.

> About sixaxis_set_operational_usb(), I don't remember _any_ change to
> it except its name, but I guess we can very well make it call
> hid_output_raw_report(), just only after your hidraw feature report
> patches land to mainline.
>
>    

sixaxis_set_operational_usb() doesn't bother me. I was confused about 
the original intent of your patch.

Alan.