Re: dvb-usb-remote woes [was: HID: ignore afatech 9016]

[Pekka Sarnila <sarnila@adit.fi>]

Ok, that was it, I had the old firmware. Now it works with the patch. I 
can now see that the afateck driver does not use the HID 
interface/endpoint (1/83) at all, but vendor specific bulk endpoint (0/81).

The fact that manufacturers have started to more more use the usb vendor 
class instead of the HID class is probably partly a consequence of HID 
specification being poorly designed. Oh, well. More work for driver writers.

Anyway, I'm still of the opinion that the ir-to-code translate should be 
per remote controller not per tv-stick (ideally loaded by kernel from 
the userspace table, and easily modifiable by userspace program).

All in all the remote controller stack is IMHO a real mess: different 
level translates (in kernel and in user space), different user program 
interfaces (keyboard with or without the special keys (many programs 
recognize only the standard keyboard keys), lircd direct, lircd/dcop). 
Really easiest is to modify kernel to get what you want. For average 
user the task of getting remote to do what they want is close to mission 
impossible. Hope someone would do something about this.


Pekka


Antti Palosaari wrote:
> On 02/01/2010 09:42 PM, Jiri Slaby wrote:
> 
>> On 02/01/2010 07:28 PM, Jiri Kosina wrote:
>>
>>> On Mon, 1 Feb 2010, Pekka Sarnila wrote:
>>>
>>>> I pulled few days ago latest
>>>>
>>>>     
>>>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>>>>
>>>> and compiled it. Everything works fine including the tv-stick and the
>>>> remote. However I get:
>>>>
>>>>    <3>af9015: command failed:255
>>>>    <3>dvb-usb: error while querying for an remote control event.
>>
>>
>> Yes, I saw this quite recently too. For me it appears when it is booted
>> up with the stick in. It's still to be fixed.
> 
> 
> I suspect you are using old firmware, 4.65.0.0 probably, that does not 
> support remote polling and thus this 255 errors seen.
> 
> regards
> Antti