Menu
▾
▴
lirc-list
|
From: Hal <sou...@ha...> - 2007-03-30 12:38:03
|
What alternatives have people used? I'm about ready to stop beating my head against the wall with lirc. Wireless keyboard, of course, but that's pretty bulky and may not have enough range. Buy a capture card that comes with it's own remote. Then I have yet another remote kicking around, plus I still probalby have problems making it work with linux. Any better solutions? Hal |
|
From: Paul B. <peb...@sa...> - 2007-03-30 13:37:59
|
Hal wrote: > What alternatives have people used? I'm about ready to stop beating my > head against the wall with lirc. > > Wireless keyboard, of course, but that's pretty bulky and may not have > enough range. > > Buy a capture card that comes with it's own remote. Then I have yet > another remote kicking around, plus I still probalby have problems > making it work with linux. > > Any better solutions? From previous posts, I assume that you are having trouble with getting your homemade IR receiver to work. If you are considering buying new hardware to solve the problem, then you could consider buying an IR receiver. Over time, I have used three different solutions to control my MythTV frontend. First, I used an IR keyboard. In addition to having the keyboard, I trained my universal remote to send the IR keyboard commands so that I did not always need the IR keyboard. However, this method did not give me the control that I wanted. Since different applications (MythTV, MPlayer and Xine) used different keys for the same function, I could not program my IR remote to seamlessly control the three different applications. Second, I used the Ira receiver <http://www.home-electro.com/ira3.php>. It uses LIRC's irman driver. Once I generated lircrc file for a remote that I had, it worked fine. However, I did not like the remote that I had. I preferred this solution to the IR keyboard solution, because using LIRC allowed me to seamlessly control MythTV, MPlayer and Xine through their native LIRC support. Now, I use a Microsoft Windows Media Center Remote <http://www.newegg.com/Product/Product.asp?Item=N82E16880100851>. It uses the LIRC's mceusb2 driver and LIRC's lirc_mceusb2 and lirc_dev kernel modules. I prefer this solution to the Ira receiver solution, because the receiver uses USB rather than RS232 and because the remote is better than any of my existing remotes for controlling MythTV, MPlayer and Xine. |
|
From: Hal <sou...@ha...> - 2007-03-30 23:11:40
|
R. G. Newbury wrote: > Hal wrote: > >> What alternatives have people used? I'm about ready to stop beating >> my head against the wall with lirc. >> >> Wireless keyboard, of course, but that's pretty bulky and may not >> have enough range. >> >> Buy a capture card that comes with it's own remote. Then I have yet >> another remote kicking around, plus I still probalby have problems >> making it work with linux. >> >> Any better solutions? >> >> Hal > > > Well the lirc documentation sucks, but it you have a reasonable remote > that is supported then the error is at your side of the keyboard. Not > that you are necessarily doing something BAD, you are just not doing > something RIGHT. > > What is your reciever and what is your remote? > > Have you downloaded the lirc tarball and run the little configurator > program to select your hardware? > > Have you tried just doing a ./configure --all instead? > > Does it then make, and make install? > > At this point you have the back end stuff prepared ready for > installation. > > If you got to here, then it is a matter of modprobing the bits into > place. modprobe lirc_i2c, modprobe lirc_dev and modprobe > lirc_your_unit_here. > > Or put 'alias char_major_61 lirc_your_driver_here' in your > modprobe.conf and reboot. > > > The back end hardware should now work if lircd is started > > YOu should have a /dev/lirc....if not fix your udev rules. > You may need to do a symlink from /dev/lirc0 > > At this point, you need to run lircd from a console..(It should be put > into a script in init.d to start automatically, later). > > Run mode2 and your remote's kepresses should be visible. > > THERE the hardware layer works. > > Now, build/ borrow an lircd.conf file, or use irw to capture the > keypresses (or is it irrrecord...going from memor). The key codes are > matched up with the key names in this file. The key nameds are used to > select actions to perform in the lircrc file. The key names must match > in both....ie PageUp in one cannot be Page+ in the other. > > Build your lircrc file..Lots around for things like mythtv. > > That's it. > > Geoff > > Oh, I don't doubt that the problem is correctable. I wish I could at least get an error message to chase down. The receiver is home-brew, TSOP-1738 based. The lirc software was installed using Synaptec apt-get on PCLinuxOS. Software startup looks like modprobe lirc_dev ln -s /dev/lirc/0 /dev/lirc0 setserial /dev/ttyS0 uart none modprobe lirc_serial irq=4 io=0x3f8 lircd -d /dev/lirc0 /etc/lircd.conf --nodaemon No errors or warnings in the console or /var/log/messages. Then mode2 -d /dev/lirc0 in a different window. Nothing happens. Everything is being done as root at this point. I can see that the hardware is getting power, and when I point a remote at it and press buttons, I can see the voltage dip at the DCD input. Using statserial instead of the lirc software, I can see that the DCD bit changes state. It's almost certainly some configuration issue in the software. I've been over and over the docs, but I can't see anything wrong. In the absence of error messages, I've hit a wall. Hal |
|
From: Hal <sou...@ha...> - 2007-03-31 21:00:12
|
A number of *very* good points here... R. G. Newbury wrote: > >Hal wrote: > > The receiver is home-brew, TSOP-1738 based. The lirc software was > installed using Synaptec apt-get on PCLinuxOS. Software startup looks > like > > > > modprobe lirc_dev > > ln -s /dev/lirc/0 /dev/lirc0 > > setserial /dev/ttyS0 uart none > > modprobe lirc_serial irq=4 io=0x3f8 > > STOP HERE: DO NOT RUN LIRCD YET! > > Are you sure that these are in the correct order? I have never touched > a serial receiver but ISTR that there are some weird 'gotchas' > involved in ensuring that the right program has control of the serial > port (ie, a modem has not grabbed it or something like that). > > What happens if you: > > setserial first. > modprobe lirc_dev > modprobe lirc_serial > Stop and check what is in /dev > > ln -s /dev/lirc/0 /dev/lirc0 (point here being that it is possible > that udev does not create the /dev/lirc/0 UNTIL the lirc_serial module > is loaded. Check the udev rule to see *exactly* what class of thing is > being checked to fire the udev rule which creates the /dev/lirc(/0). I > do not think that lirc_dev will do this: I think it is an underlying > common module that does not directly interact with hardware (could be > wrong!). I suspect that lirc_serial does the hardware bit, and talks > to /dev/lirc(/x). You are correct that /dev/lirc/0 is being created after 'modprobe lirc_serial'. Even though the link apeared to be functional anyway, I prefer your suggested statement order. > And the lirc modules generally look for a /dev/lirc NOT a /dev/lirc0. I've seen references both ways. The problem is that since /dev/lirc already exists as a directory, I can't create a link by the same name. > Stop after the modprobe lirc_serial and look at what has been created. > I am assuming that you are getting a /dev/lirc/0 but I think you might > be better thereafter if the link is /dev/lirc (since you may be > missing a '-d /dev/lirc0' somewhere. > > Down below you have attempted to run mode2. Mode2 and lircd are > equivalents in terms of layering. Whoa. Major eye-opener there. I didn't realize that at all. > Try mode2 HERE. > before: > > lircd -d /dev/lirc0 /etc/lircd.conf --nodaemon Unfortunately mode2 is still silent. > > No errors or warnings in the console or /var/log/messages. > Have you looked at dmesg? Although if something is NOT being created, > there would be no message.. By habit I tend to look in /var/log/messages. dmesg reports lirc_dev: IR Remote Control driver registered, at major 61 lirc_serial: auto-detected active high receiver lirc_dev: lirc_register_plugin: sample_rate: 0 Oh, and /etc/modprobe.conf includes alias char-major-61 lirc_serial options lirc_serial io=0x3f8 irq=4 I also tried after the modprobes cd /dev mv lirc lircd ln -s lircd/0 lirc so that /dev/lirc would be the device. No joy. You mentioned udev rules. The contents of /etc/udev/rules.d/lirc.rules: KERNEL="lirc[0-9]*", NAME="lirc/%n" I have zero experience with udev rules, so this doesn't really tell me much. Seems reasonable enough to my untrained eye, I guess. You provided so much useful information it got my hopes up. But today still isn't the big day, I guess. I really appreciate the help, though. Hal |
|
From: <li...@ba...> - 2007-04-01 07:47:22
|
Hi! Hal "sou...@ha..." wrote: [...] > lirc_dev: IR Remote Control driver registered, at major 61 > lirc_serial: auto-detected active high receiver This usually means: - your receiver is broken or - your port is broken or - your receiver is not connected or - your receiver is connected to the wrong port Christoph |
|
From: Hal <sou...@ha...> - 2007-04-02 01:14:16
|
Christoph Bartelmus wrote: >>lirc_dev: IR Remote Control driver registered, at major 61 >>lirc_serial: auto-detected active high receiver >> >> > >This usually means: >- your receiver is broken or >- your port is broken or >- your receiver is not connected or >- your receiver is connected to the wrong port > > > Can you expound on that? Those dmesg entries seem pretty common from Googling around. Plus, the positive results from statserial indicate working hardware. Hal |
|
From: Hal <sou...@ha...> - 2007-04-01 23:47:46
|
> Have you tried mode2 with different device parameters? > > mode2 --device=/dev/lirc > mode2 --device=/dev/lirc/0 > mode2 --device=/dev/lirc0 > I have tried 'mode2 -d /dev/lirc/0' to bypass the softlink, with no affect. 'mode2 -d /dev/lirc' gives a "Not a character device" error. >> I also tried after the modprobes >> >> cd /dev >> mv lirc lircd >> ln -s lircd/0 lirc > > This makes no sense. You should not need to symlink except possibly > for ease (ie /dev/lirc0 symm'ed by /dev/lirc ) But you cannot symlink > UP.. > > The second line renames dev/licc to /dev/lircd ....Why? > > The third line makes a symlink from lircd/0 (where did THAT come from: > was it the /dev/lirc you just renamed, with a subfolder /o??) to > lirc..WHY? > > Trash this stuff. > No, no, this is just something I went ahead and tried since your last message. It was the only way to have the device addressed at /dev/lirc instead of /dev/lirc0. And predictably it did *not* help. >> so that /dev/lirc would be the device. No joy. > > > What DID you have in the first place? > The softlink /dev/lirc0 points to /dev/lirc/0. > So try 'mode2 --=device=/dev/lirc/0' > As I said, I have tried mode2 -d /dev/lirc/0. See why I'm ready give it up? Hal |
|
From: Hal <sou...@ha...> - 2007-04-02 14:08:42
|
> Ok then, lets back up a bit. > Clean boot, do nothing (but let the system load modules and udev do > its bit, what do you find in /dev I'm still loading modules from the command line for troubleshooting purposes, but having done that... > You should have: > > /dev/lircd and, if I follow you, /dev/lircd gets created by lircd, which I'm not getting to yet. Still trying to get output from mode2. So no, /dev/lircd does not exist. > /dev/lirc/0 > /dev/lirc (as symlinlk from /dev/lirc/0 /dev/lirc is a directory entry, so the character device is at /dev/lirc0 -> /dev/lirc/0. > If you do not have a /dev/lircd which sounds like the case, to me, > then no wonder mode2 will not work. If you have no /dev/lircd then I > would *guess* that your options are wrong, or the module does not or > cannot grab the serial port. This presents a chicken-and-egg problem. /dev/lircd is created by running lircd, but lircd shouldn't be run simultaneously with mode2. (Actually, they *will* both run without complaint. But I still don't get output from mode2.) /dev/lircd, when it exists, is a socket. mode2 reads from a character device. Are you sure about that dependency? Hal |
|
From: Hal <sou...@ha...> - 2007-04-02 14:29:47
|
EUREKA!!!! Serendipity can be a wonderful thing. Following your thought about options being wrong, I decided to experiment with the options in the modprobe command line, rather than rely on the contents of modprobe.conf. From a post describing someopne else's travails, I copied/pasted 'modprobe lirc_serial type=0 irq=4 io=0x3f8' and suddenly got output from mode2! It took a few minutes to figure out why. It seems that there is this double-super-secret option to lirc_serial, 'type=0', the absence of which was causing my problems. As far as Google and I can tell, this information doesn't exist anywhere on www.lirc.org. I found it completely by accident. Thanks so much for your help. Hopefully theings will be smoother from this point on. Hal |
|
From: Hal <sou...@ha...> - 2007-04-02 21:47:11
|
> Now do everyone a favour, write up a small howto page and post it on > the mythtv wiki, with a good title including the words 'lirc serial'. > > Geoff > Good idea. If I help one person avoid this headache, it'll be worthwhile. But I think I'll wait until the whole system is working top-to-bottom, ftrst. :-) Hal |
|
From: <li...@ba...> - 2007-04-03 07:23:19
|
Hi! Hal "sou...@ha..." wrote: >> Now do everyone a favour, write up a small howto page and post it on >> the mythtv wiki, with a good title including the words 'lirc serial'. > Good idea. If I help one person avoid this headache, it'll be worthwhile. It would be even more interesting why you compiled the driver for the wrong device in the first place. Christoph |
|
From: Hal <sou...@ha...> - 2007-04-03 12:41:57
|
>It would be even more interesting why you compiled the driver for the >wrong device in the first place. > >Christoph > > > I didn't compile the driver at all. I installed it using the Synaptic tool from witihin PClinuxOS. Hal |
|
From: Jason L. E. <in...@da...> - 2007-04-03 13:21:43
|
you guys should change the subject on this thread. it's insulting. -Jason Ellison On Tue, 3 Apr 2007, Hal wrote: > > >It would be even more interesting why you compiled the driver for the > >wrong device in the first place. > > > >Christoph > > > > > > > I didn't compile the driver at all. I installed it using the Synaptic > tool from witihin PClinuxOS. > > Hal > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > |
|
From: <li...@ba...> - 2007-04-04 14:39:15
|
Hi! Hal "sou...@ha..." wrote: >> It would be even more interesting why you compiled the driver for the >> wrong device in the first place. >> > I didn't compile the driver at all. I installed it using the Synaptic > tool from witihin PClinuxOS. Then you should let the maintainer of the package know that his package is broken. Christoph |
