Mailing List Archive

Yes I'm running lirc with mceusb controller.


Sent from my Samsung GALAXY S4 on the Telstra mobile network

<div>-------- Original message --------</div><div>From: Kirk Bocek <t004@kbocek.com> </div><div>Date:04/06/2015 1:37 AM (GMT+10:00) </div><div>To: "User discussion about ATrpms.net" <atrpms-users@atrpms.net> </div><div>Subject: [ATrpms-users] MythTV 0.27.4-31 Upgrade Success </div><div>
</div>Upgrade finally succeeded! Thank you Stephen.

My difficulty was 1. that your qt48-x11 package still needs the base
distro qt-x11 package, 2. yes there are incompatibilities with the qt47
packages although it is possible to have them co-installed and 3. a few
things were needed from EPEL. A few other packages needed to be
installed/upgraded that were not needed for Myth 0.24 but I can't recall
what they were.

Now my problem is with LIRC on my frontends. Stephen, are you running
LIRC on CentOS 6?

I updated my frontends to the latest 6.6 and your
nvidia-graphics-legacy-340 packages. But there are no LIRC packages for
the current kernel. I tried the LIRC 0.9.0 packages from EPEL with no
success:

http://lists.centos.org/pipermail/centos/2015-June/152589.html

Do you have any thoughts?

Thanks.


_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

I'll have a look a bit later on. What controller are you using. 


Sent from my Samsung GALAXY S4 on the Telstra mobile network

<div>-------- Original message --------</div><div>From: Kirk Bocek <t004@kbocek.com> </div><div>Date:04/06/2015 1:37 AM (GMT+10:00) </div><div>To: "User discussion about ATrpms.net" <atrpms-users@atrpms.net> </div><div>Subject: [ATrpms-users] MythTV 0.27.4-31 Upgrade Success </div><div>
</div>Upgrade finally succeeded! Thank you Stephen.

My difficulty was 1. that your qt48-x11 package still needs the base
distro qt-x11 package, 2. yes there are incompatibilities with the qt47
packages although it is possible to have them co-installed and 3. a few
things were needed from EPEL. A few other packages needed to be
installed/upgraded that were not needed for Myth 0.24 but I can't recall
what they were.

Now my problem is with LIRC on my frontends. Stephen, are you running
LIRC on CentOS 6?

I updated my frontends to the latest 6.6 and your
nvidia-graphics-legacy-340 packages. But there are no LIRC packages for
the current kernel. I tried the LIRC 0.9.0 packages from EPEL with no
success:

http://lists.centos.org/pipermail/centos/2015-June/152589.html

Do you have any thoughts?

Thanks.


_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On 6/3/2015 12:26 PM, Stephen Collier wrote:

I'll have a look a bit later on. What controller are you using. 

The driver identifies it as an Media Center Ed. eHome Infrared Remote Transceiver


IR NEC protocol handler initialized
IR RC5(x) protocol handler initialized
IR RC6 protocol handler initialized
Registered IR keymap rc-rc6-mce
input: Media Center Ed. eHome Infrared Remote Transceiver (0471:0815) as /devices/pci0000:00/0000:00:1d.3/usb5/5-1/5-1:1.0/rc/rc0/input14
rc0: Media Center Ed. eHome Infrared Remote Transceiver (0471:0815) as /devices/pci0000:00/0000:00:1d.3/usb5/5-1/5-1:1.0/rc/rc0
mceusb 5-1:1.0: Registered Philips eHome Infrared Transceiver on usb5:2
usbcore: registered new interface driver mceusb
IR JVC protocol handler initialized
IR Sony protocol handler initialized
lirc_dev: IR Remote Control driver registered, major 247
rc rc0: lirc_dev: driver ir-lirc-codec (mceusb) registered at minor = 0
IR LIRC bridge handler initialized
  alloc irq_desc for 32 on node -1
  alloc kstat_irqs on node -1

When I run irw I see in /var/log/messages a bunch of:
Jun  3 07:02:48 marble lircd-0.9.0[1501]: failed on bit 13
Jun  3 07:02:48 marble lircd-0.9.0[1501]: failed on bit 9

So *something* is getting through.

What I'd like to know is how you did your install. This is using the EPEL packages.

$rpm -qa |grep lirc
lirc-0.9.0-8.el6.x86_64
lirc-libs-0.9.0-8.el6.x86_64

I'm also running MythTV on ContOS6. I hope this helps.

On Wed, 03 Jun 2015 14:15:55 -0700
Kirk Bocek <t004@kbocek.com> wrote:

> The driver identifies it as an Media Center Ed. eHome Infrared Remote
> Transceiver

I have a similar device---a USB IR receiver. There's also a IR
receiver on my tuner which I'm not using.

> IR NEC protocol handler initialized
> IR RC5(x) protocol handler initialized
> IR RC6 protocol handler initialized
> Registered IR keymap rc-rc6-mce
> input: Media Center Ed. eHome Infrared Remote Transceiver (0471:0815)
> as /devices/pci0000:00/0000:00:1d.3/usb5/5-1/5-1:1.0/rc/rc0/input14

This means the kernel recognized the receiver as an input device,
similar to a keyboard. I think you want to use the devinput driver
for lirc. http://lirc.org/html/devinput.html
You can check for the receiver in /proc/bus/input/devices.

I set LIRC_DRIVER="devinput" and LIRC_DEVICE in /etc/sysconfig/lirc an
copies the generic devinput config to lird.conf.

> rc0: Media Center Ed. eHome Infrared Remote Transceiver (0471:0815)
> as /devices/pci0000:00/0000:00:1d.3/usb5/5-1/5-1:1.0/rc/rc0 mceusb
> 5-1:1.0: Registered Philips eHome Infrared Transceiver on usb5:2
> usbcore: registered new interface driver mceusb IR JVC protocol
> handler initialized IR Sony protocol handler initialized lirc_dev: IR
> Remote Control driver registered, major 247 rc rc0: lirc_dev: driver
> ir-lirc-codec (mceusb) registered at minor = 0 IR LIRC bridge handler
> initialized alloc irq_desc for 32 on node -1
>   alloc kstat_irqs on node -1
>
> When I run irw I see in /var/log/messages a bunch of:
> Jun  3 07:02:48 marble lircd-0.9.0[1501]: failed on bit 13
> Jun  3 07:02:48 marble lircd-0.9.0[1501]: failed on bit 9
>
> So *something* is getting through.
>
> What I'd like to know is how you did your install. This is using the
> EPEL packages.
>
> $rpm -qa |grep lirc
> lirc-0.9.0-8.el6.x86_64
> lirc-libs-0.9.0-8.el6.x86_64

I have the same packages. It's been running this way for a while
without problems.

Previously, I had a different USB remote which pretended to be a
keyboard and I couldn't quite get lirc to work with it. It can be
hit-or-miss as to weather the cheap Windows remotes work with Linux.

Jim

_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On 6/3/2015 7:15 PM, James Szinger wrote:
> I'm also running MythTV on ContOS6. I hope this helps.
>

Absolutely, we seem to be few and far between.

> I have a similar device---a USB IR receiver. There's also a IR
> receiver on my tuner which I'm not using.
>
>> IR NEC protocol handler initialized
>> IR RC5(x) protocol handler initialized
>> IR RC6 protocol handler initialized
>> Registered IR keymap rc-rc6-mce
>> input: Media Center Ed. eHome Infrared Remote Transceiver (0471:0815)
>> as /devices/pci0000:00/0000:00:1d.3/usb5/5-1/5-1:1.0/rc/rc0/input14
> This means the kernel recognized the receiver as an input device,
> similar to a keyboard. I think you want to use the devinput driver
> for lirc. http://lirc.org/html/devinput.html
> You can check for the receiver in /proc/bus/input/devices.
>
> I set LIRC_DRIVER="devinput" and LIRC_DEVICE in /etc/sysconfig/lirc an
> copies the generic devinput config to lird.conf.

I am currently going around with a gentleman on the MythTV list who is
confused because the mceusb driver is creating both /dev/lirc0 and the
/dev/input/eventXX entries. He thinks one or the other should be created.

So can you still do multiple remotes with the devinput interface? So you
don't have a specific remote in /etc/lirc/lircd.conf?

>
> What I'd like to know is how you did your install. This is using the
> EPEL packages.
>
> $rpm -qa |grep lirc
> lirc-0.9.0-8.el6.x86_64
> lirc-libs-0.9.0-8.el6.x86_64
> I have the same packages. It's been running this way for a while
> without problems.
>
> Previously, I had a different USB remote which pretended to be a
> keyboard and I couldn't quite get lirc to work with it. It can be
> hit-or-miss as to weather the cheap Windows remotes work with Linux.
>

Okay, good, just gotta figure this out.

_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On Wed, 03 Jun 2015 19:25:20 -0700
Kirk Bocek <t004@kbocek.com> wrote:
>
> I am currently going around with a gentleman on the MythTV list who
> is confused because the mceusb driver is creating both /dev/lirc0 and
> the /dev/input/eventXX entries. He thinks one or the other should be
> created.

This doesn't match my experience. I just plugged a USB DVB adapter
into my Fedora laptop and it created both a /dev/lirc0 and
a /dev/input/eventXX entry. I'me not sure it matters much, since I
don't refer to the /dev/ entry at all in my lirc config.

>
> So can you still do multiple remotes with the devinput interface? So
> you don't have a specific remote in /etc/lirc/lircd.conf?

I don't know, since I never tried it. Maybe one would have to run
multiple lirc daemons.

For my current setup I specify the device using the
LIRC_DEVICE='name="Media Center Ed...."' syntax.


Jim

_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On 6/4/2015 6:31 AM, James Szinger wrote:
> On Wed, 03 Jun 2015 19:25:20 -0700
> Kirk Bocek <t004@kbocek.com> wrote:
>> I am currently going around with a gentleman on the MythTV list who
>> is confused because the mceusb driver is creating both /dev/lirc0 and
>> the /dev/input/eventXX entries. He thinks one or the other should be
>> created.
> This doesn't match my experience. I just plugged a USB DVB adapter
> into my Fedora laptop and it created both a /dev/lirc0 and
> a /dev/input/eventXX entry. I'me not sure it matters much, since I
> don't refer to the /dev/ entry at all in my lirc config.

I think this guy was wrong.

>> So can you still do multiple remotes with the devinput interface? So
>> you don't have a specific remote in /etc/lirc/lircd.conf?
> I don't know, since I never tried it. Maybe one would have to run
> multiple lirc daemons.
>
> For my current setup I specify the device using the
> LIRC_DEVICE='name="Media Center Ed...."' syntax.
>
>

So with 0.8.7 I was using the LIRC interface. It sounds like with 0.9.0
I need to move to the devinput interface. /etc/sysconfig/lirc:

LIRC_DRIVER="devinput"
LIRC_DEVICE="/dev/input/event14"

But you have the udev rule to provide a stable /dev/input name. Would
you share that please.

I'm also told that for testing I use ir-keytable instead of irw.



_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On 6/4/2015 7:18 AM, Kirk Bocek wrote:
>
>
> On 6/4/2015 6:31 AM, James Szinger wrote:
>> On Wed, 03 Jun 2015 19:25:20 -0700
>> Kirk Bocek <t004@kbocek.com> wrote:
>>> I am currently going around with a gentleman on the MythTV list who
>>> is confused because the mceusb driver is creating both /dev/lirc0 and
>>> the /dev/input/eventXX entries. He thinks one or the other should be
>>> created.
>> This doesn't match my experience. I just plugged a USB DVB adapter
>> into my Fedora laptop and it created both a /dev/lirc0 and
>> a /dev/input/eventXX entry. I'me not sure it matters much, since I
>> don't refer to the /dev/ entry at all in my lirc config.
>
> I think this guy was wrong.
>
>>> So can you still do multiple remotes with the devinput interface? So
>>> you don't have a specific remote in /etc/lirc/lircd.conf?
>> I don't know, since I never tried it. Maybe one would have to run
>> multiple lirc daemons.
>>
>> For my current setup I specify the device using the
>> LIRC_DEVICE='name="Media Center Ed...."' syntax.
>>
>>
>
> So with 0.8.7 I was using the LIRC interface. It sounds like with
> 0.9.0 I need to move to the devinput interface. /etc/sysconfig/lirc:
>
> LIRC_DRIVER="devinput"
> LIRC_DEVICE="/dev/input/event14"
>
> But you have the udev rule to provide a stable /dev/input name. Would
> you share that please.
>
> I'm also told that for testing I use ir-keytable instead of irw.
>

Argh... still no output from ir-keytable.

$ir-keytable
Found /sys/class/rc/rc4/ (/dev/input/event14) with:
Driver mceusb, table rc-rc6-mce
Supported protocols: NEC RC-5 RC-6 JVC SONY LIRC
Enabled protocols: RC-6
Name: Media Center Ed. eHome Infrared
bus: 3, vendor/product: 0471:0815, version: 0x0000
Repeat delay = 500 ms, repeat period = 33 ms

$ir-keytable -t -d /dev/input/event14
Testing events. Please, press CTRL-C to abort.
^C

ir-keytable -r -d /dev/input/event14
Enabled protocols: NEC RC-5 RC-6 JVC SONY SANYO LIRC RC-5-SZ SHARP
MCE_KBD XMP

I have the devinput config file in /etc/lirc/lircd.conf.


_______________________________________________
atrpms-users mailing list
atrpms-users@atrpms.net
http://lists.atrpms.net/mailman/listinfo/atrpms-users

On 6/4/2015 9:10 AM, Kirk Bocek wrote:

Argh... still no output from ir-keytable.

$ir-keytable
Found /sys/class/rc/rc4/ (/dev/input/event14) with:
        Driver mceusb, table rc-rc6-mce
        Supported protocols: NEC RC-5 RC-6 JVC SONY LIRC
        Enabled protocols: RC-6
        Name: Media Center Ed. eHome Infrared
        bus: 3, vendor/product: 0471:0815, version: 0x0000
        Repeat delay = 500 ms, repeat period = 33 ms

$ir-keytable -t -d /dev/input/event14
Testing events. Please, press CTRL-C to abort.
^C

ir-keytable -r -d /dev/input/event14
Enabled protocols: NEC RC-5 RC-6 JVC SONY SANYO LIRC RC-5-SZ SHARP MCE_KBD XMP

I have the devinput config file in /etc/lirc/lircd.conf.

After all that hassle I think this just boils down to a problem with the remote configuration file I was using. I used irrecord to start a partial config file and viola! irw produces output:

0000000000000001 00 BTN_SELECT recorded.conf
0000000000000001 00 BTN_SELECT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000003 00 BTN_LEFT recorded.conf

Note that I did this after *undoing* the devinput configuration:

LIRC_DRIVER=""
LIRC_DEVICE=""

Now the configuration I have been using for *years* is a Scientific Atlanta 3250. Now there *is* a configuration in the database for this remote but it doesn't appear to work with LIRC 0.9.0. Seems like everything in the database has been generated well before 0.8.6.

http://lirc.sourceforge.net/remotes/scientific_atlanta/3250"]http://lirc.sourceforge.net/remotes/scientific_atlanta/3250

So the /dev/lirc0 interface is working. What was not working was my remote configuration from 0.8.7.

This is a remote that I could use with a cheap programmable remote *and* that had the A, B, C, D buttons I've come to depend on. I copied the base config and it caused errors on loading requiring the removal of the TV/VCR section.

So I either need to pick a new remote that is both programmable on my physical remote and has A, B, C, D buttons or I need to irrecord a new setup off of the generic RC-5 configuration.

On 6/5/2015 8:33 AM, Kirk Bocek wrote:

On 6/4/2015 9:10 AM, Kirk Bocek wrote:

Argh... still no output from ir-keytable.

$ir-keytable
Found /sys/class/rc/rc4/ (/dev/input/event14) with:
        Driver mceusb, table rc-rc6-mce
        Supported protocols: NEC RC-5 RC-6 JVC SONY LIRC
        Enabled protocols: RC-6
        Name: Media Center Ed. eHome Infrared
        bus: 3, vendor/product: 0471:0815, version: 0x0000
        Repeat delay = 500 ms, repeat period = 33 ms

$ir-keytable -t -d /dev/input/event14
Testing events. Please, press CTRL-C to abort.
^C

ir-keytable -r -d /dev/input/event14
Enabled protocols: NEC RC-5 RC-6 JVC SONY SANYO LIRC RC-5-SZ SHARP MCE_KBD XMP

I have the devinput config file in /etc/lirc/lircd.conf.

After all that hassle I think this just boils down to a problem with the remote configuration file I was using. I used irrecord to start a partial config file and viola! irw produces output:

0000000000000001 00 BTN_SELECT recorded.conf
0000000000000001 00 BTN_SELECT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000002 00 BTN_RIGHT recorded.conf
0000000000000003 00 BTN_LEFT recorded.conf

Note that I did this after *undoing* the devinput configuration:

LIRC_DRIVER=""
LIRC_DEVICE=""

Now the configuration I have been using for *years* is a Scientific Atlanta 3250. Now there *is* a configuration in the database for this remote but it doesn't appear to work with LIRC 0.9.0. Seems like everything in the database has been generated well before 0.8.6.

http://lirc.sourceforge.net/remotes/scientific_atlanta/3250"]http://lirc.sourceforge.net/remotes/scientific_atlanta/3250

So the /dev/lirc0 interface is working. What was not working was my remote configuration from 0.8.7.

This is a remote that I could use with a cheap programmable remote *and* that had the A, B, C, D buttons I've come to depend on. I copied the base config and it caused errors on loading requiring the removal of the TV/VCR section.

So I either need to pick a new remote that is both programmable on my physical remote and has A, B, C, D buttons or I need to irrecord a new setup off of the generic RC-5 configuration.

Yup, it did boil down to my remote configuration file. My physical remote was programmable as an MCE remote. I swapped in the mceusb configuration file and everything is now just working.

irw:
000000037ff07bdf 00 KEY_LEFT mceusb
000000037ff07bdf 01 KEY_LEFT mceusb
000000037ff07bdf 02 KEY_LEFT mceusb
000000037ff07bdd 00 KEY_OK mceusb
000000037ff07bdd 01 KEY_OK mceusb
000000037ff07bdd 02 KEY_OK mceusb
000000037ff07bde 00 KEY_RIGHT mceusb
000000037ff07bde 01 KEY_RIGHT mceusb
000000037ff07bde 02 KEY_RIGHT mceusb

I spent some time editing ~/.mythtv/lircrc for Myth and ~/.lircrc for mplayer putting in all the new namespace strings and everything is just working. Yay!

A user on the mythtv-users list stated the older configuration files are compatible. Clearly this is not the case. I even tried directly using the Scientific Atlanta file without changes. It generated an error on the first section of the file. And obviously the rest of it just wasn't working.

I think sticking to the files in the lirc-remotes-0.9.0 package will be safer. And anything in the online database will require lots of testing.

Thank you everyone for your help and suggestions.