[mythtv-users] USB IR Blaster: close but no cigar

[Jarod Wilson]

On Thu, Apr 15, 2010 at 2:11 PM, Ron Garrison <ron.garrison at gmail.com> wrote:
> On Thu, Apr 15, 2010 at 7:48 AM, Greg Woods <greg at gregandeva.net> wrote:
>> Damn! I had such high hopes....
>>
>> This didn't work. I just recompiled lirc from CVS, and my blaster still
>> does not work. I am sure of:
>>
>> 1) I am running the correct version of lircd
>> 2) I am running the correct version of the lirc_mceusb module

How absolutely certain are you of this? What does 'modinfo lirc_mceusb' say?

>> 3) I am running the correct version of irw/irsend
>> 4) irw works (so the receiver works)
>> 5) irsend reports no errors
>> 6) The blaster does not work (no visible flash, no flash on the receiver
>> when irsend is run with the blaster held in front of it)
>>
>> Anything I can do to debug this and figure out what else I may have done
>> wrong?
>>
>> If it matters, this system is still at Fedora 10. Could the kernel be
>> too old for this new module? (Probably not since it loads fine, but I am
>> grasping at straws now).

Given that it loads fine and irw reports back keys from the remote,
yeah, kernel version doesn't matter here.

> Load your modules with debug = 1 and compare the messages you get with
> the source code.  This is how I discovered I was running the wrong
> version.  Anyhow, it will hopefully point you in the right direction.
> If you post your dmesg output I could at least see how it compares
> with my debug output.

Indeed, loading lirc_mceusb with debug=1 should be fairly verbose,
showing raw data traffic inbound and outbound, complete with some
decoding of some of the mce protocol packets now. Should be pretty
obvious right off the bat from spew in dmesg if you really are running
the latest code or something older. :)

-- 
Jarod Wilson
jarod at wilsonet.com