[mythtv-users] still having remote troubles with Fedora 14/Harmony

[Jarod Wilson]

On Jan 25, 2011, at 11:23 PM, Jarod Wilson wrote:

> On Tue, Jan 25, 2011 at 5:31 PM, Jarod Wilson <jarod at wilsonet.com> wrote:
>> On Jan 22, 2011, at 12:42 PM, Greg Woods wrote:
> ...
>>> OK, I finally got time to try this. Downloaded and installed the x86_64
>>> version and booted it. I would say that things are improved, but not
>>> completely fixed. While it is now a bit easier to avoid double keys, it
>>> still happens if I am the least bit careless about how long I hold the
>>> key down. The Streamzap is still better than the Harmony. I think the
>>> Streamzap is almost wife-usable, but the Harmony is still way too
>>> sensitive.
>>> 
>>> Short answer: improved, but still more of a problem than it used to be
>>> before I upgraded from F13 to F14.
>> 
>> So I think the "way too sensitive" may be "we actually report more
>> accurately what signals we've seen", which needs to then be suppressed
>> and/or filtered in userspace, either with the suppress_repeat param
>> in lircd.conf or using repeat=X in .lircrc. At least, as far as I can
>> tell, the same issues that were in the mceusb driver are either already
>> fixed in streamzap, or were never issues, due to differences in the
>> device driver buffer parsing routines. But I still need to go and do
>> some actual hardware prodding myself.
> 
> Crap. Just tried my own streamzap hardware. Things are definitely off a
> touch. Rather than reporting a keypress and repeat(s), its reporting as a
> new keypress every time, which explains your description of "way too
> sensitive". I'll try to get it sorted out as soon as I can.

And now I have a fix:

http://git.linuxtv.org/jarod/linux-2.6-ir.git?a=commitdiff;h=7249db782c552ea89dc7cd6e5ae0e2c30ca984a9

Might be a few days before its in a Fedora kernel build though, since I have
a few other IR things I'd like to poke at a bit more before doing another
kernel build.

-- 
Jarod Wilson
jarod at wilsonet.com