[mythtv-users] (Slightly ot) mce remote stops responding

[Stephen P. Villano]

On 11/14/14 3:46 AM, Chris Simmons wrote:
> On 13/11/14 16:38, Chris Simmons wrote:
>> I've upgraded my frontend from fedora 19 to fedora 20 and something
>> odd has happened with the MCE remote.  It works fine initially then
>> at some point (often after waking from suspend) it will initially
>> work but after a few key presses just stop responding.  The light
>> still flashes on the receiver, lsusb still shows it, everything looks
>> normal but nothing comes out.  I've not found any errors being logged
>> anywhere yet either which doesn't give me much to go on.
>>
>> The closest thing I've found is this below, similar symptoms in that
>> plugging it into a different port will fix things (until it occurs
>> again).  I've turned off usb autosuspend but that didn't fix things. 
>> Note that the machine doesn't have USB 3 ports.
>>
>> https://github.com/OpenELEC/OpenELEC.tv/issues/3515
>>
>> Any ideas?
>>
>> The only think I can think of is to try making it rmmod the mceusb
>> driver on suspend and see if that helps.
>>
>> Cheers,
>>
>> Chris Simmons.
>> _______________________________________________
>> mythtv-users mailing list
>> mythtv-users at mythtv.org
>> http://lists.mythtv.org/mailman/listinfo/mythtv-users
>> http://wiki.mythtv.org/Mailing_List_etiquette
>> MythTV Forums: https://forum.mythtv.org
> Thanks for the replies but its not the batteries nor focus.  It went
> wrong again last night so I did this.
>
> - exit frontend
> - stop lircd (if only because it eats the events so xev doesn't see them)
> - run xev
> - nothing
> - rmmod mceusb
> - modprobe mceusb
> - xev shows remove springs back to life
>
> I've got another machine which is very similar, same os, same receiver
> but the problem doesn't occur.  Swapping the receivers/remotes didn't
> change anything.   I don't suspend this other machine so much as its
> used less frequently so its usually turned off rather than suspended
> but still.  Makes me wonder whether its a bug in the USB hub driver as
> they have different motherboards?
>
> Still it looks like rmmod'ing mceusb before suspend might fix things,
> I tried it and lircd died but I'm sure I can do something about that.
>
> I'd be grateful of further input if anyone has any bright ideas :)
>
> Cheers,
>
> Chris.
> _______________________________________________
Is mceusb in memory on resume? If so, it would appear to be tolerating
suspend poorly.
In that case, rmmod before suspend and modprobe on resume should work.
So, stopping lircd, rmmod mceusb on suspend, modprobe and start lircd
upon resume should work.
Hardware implementation varies, so drivers vary. Some hardware has known
bugs and some have suspected bugs, some of which are unpatchable. So,
the different motherboards could be the cause, either at a driver level
or hardware level. Determining which would be more along the lines of
working with a kernel driver developer to ascertain which, if either is
the cause of the issue.