Linux Infrared Remote Control

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sunday 28 March 2004 17:50, Christoph Bartelmus wrote:
> Volunteers, step forward...
Sorry, at the moment I have no time at all. This is my last school year and=
 I=20
already have problems with certain subjects :/

I wanted to have more time.
=2D --=20
=46lameeyes <dgp85@...>
You can find LIRC for 2.6 kernels at
http://flameeyes.web.ctonet.it/
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFAZvqie2h1+2mHVWMRAgXsAKCqKk8KF7PAv8ltzxNjQda0ckRVeQCgtLkD
9p117I6gQvlwJsjzMAkQq6Y=3D
=3DLwpl
=2D----END PGP SIGNATURE-----

Hi!

Christoph Bartelmus "lirc@..." wrote:
[...]
> I have changed lirc_serial, lirc_sir, lirc_parallel, lirc_dev and
> lirc_gpio to *compile* with 2.6.3.

Feedback after 1 week: one success report.

The state of the following modules is currently unknown and they  
probably will have to be ported to 2.6:

lirc_atiusb
lirc_bt829
lirc_i2c
lirc_it87
lirc_mceusb

There are currently 600 people subscribed to this list. I hope there are  
some people among them who can invest some time and make these modules  
work in 2.6.

Christoph

Christoph,
 I'll be most willing to help out if only I knew what
"lirc_serial: Unknown symbol module_refcount" means.

 If you can send me a link or explain it would be
great,

Thanks,
Shaul

--- Christoph Bartelmus <lirc@...> wrote:
> Hi!
> 
> Christoph Bartelmus "lirc@..." wrote:
> [...]
> > I have changed lirc_serial, lirc_sir,
> lirc_parallel, lirc_dev and
> > lirc_gpio to *compile* with 2.6.3.
> 
> Feedback after 1 week: one success report.
> 
> The state of the following modules is currently
> unknown and they  
> probably will have to be ported to 2.6:
> 
> lirc_atiusb
> lirc_bt829
> lirc_i2c
> lirc_it87
> lirc_mceusb
> 
> There are currently 600 people subscribed to this
> list. I hope there are  
> some people among them who can invest some time and
> make these modules  
> work in 2.6.
> 
> Christoph
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux
> Tutorials
> Free Linux tutorial presented by Daniel Robbins,
> President and CEO of
> GenToo technologies. Learn everything from
> fundamentals to system
>
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

>>>>> "Christoph" == Christoph Bartelmus <lirc@...> writes:

 Christoph> Hi!
 Christoph> Christoph Bartelmus "lirc@..." wrote:
 Christoph> [...]
 >> I have changed lirc_serial, lirc_sir, lirc_parallel, lirc_dev and
 >> lirc_gpio to *compile* with 2.6.3.

 Christoph> Feedback after 1 week: one success report.

 Christoph> The state of the following modules is currently unknown
 Christoph> and they probably will have to be ported to 2.6:

 Christoph> lirc_atiusb lirc_bt829 lirc_i2c lirc_it87 lirc_mceusb

lirc_i2c works fine here. I guess folks don't complain unless there is
a problem. hope this helps. 

Thanks

Hi A.,

on 04 Apr 04 at 12:10, you wrote:
[...]
> lirc_i2c works fine here. I guess folks don't complain unless there is
> a problem. hope this helps.

LIRC from CVS does not compile when using lirc_i2c.
If you mean the Flameeyes' patches: I appreciate his efforts, but for me  
the patches have introduced more new problems than they solved.

Christoph

Christoph,
 I've recompiled the kernel and now I can modprobe
lirc_serial (thnx!).

 The problem still remains - I don't get the +10v from
the RTS in the RS232 connection,

Shaul

p.s. I'll be more than happy to send you anything you
ask for (I am using 2.6.4 with gentoo patches)

--- Christoph Bartelmus <lirc@...> wrote:
> Hi A.,
> 
> on 04 Apr 04 at 12:10, you wrote:
> [...]
> > lirc_i2c works fine here. I guess folks don't
> complain unless there is
> > a problem. hope this helps.
> 
> LIRC from CVS does not compile when using lirc_i2c.
> If you mean the Flameeyes' patches: I appreciate his
> efforts, but for me  
> the patches have introduced more new problems than
> they solved.
> 
> Christoph
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux
> Tutorials
> Free Linux tutorial presented by Daniel Robbins,
> President and CEO of
> GenToo technologies. Learn everything from
> fundamentals to system
>
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

----- Original Message -----
From: "A. P. Kennedy" <spamtree@...>
Cc: <lirc-list@...>
Sent: Sunday, April 04, 2004 12:10 PM


> >>>>> "Christoph" == Christoph Bartelmus <lirc@...> writes:
>
>  Christoph> Hi!
>  Christoph> Christoph Bartelmus "lirc@..." wrote:
>  Christoph> [...]
>  >> I have changed lirc_serial, lirc_sir, lirc_parallel, lirc_dev and
>  >> lirc_gpio to *compile* with 2.6.3.
>
>  Christoph> Feedback after 1 week: one success report.
>
>  Christoph> The state of the following modules is currently unknown
>  Christoph> and they probably will have to be ported to 2.6:
>
>  Christoph> lirc_atiusb lirc_bt829 lirc_i2c lirc_it87 lirc_mceusb
>
> lirc_i2c works fine here. I guess folks don't complain unless there is
> a problem. hope this helps.
>
> Thanks

What's your secret?  I'm having problems with lirc_i2c.  I'm on kernel
2.6.4.  I had to use flameeyes patch just to get stuff to compile and now
only one client can connect to lirc and even after that one has connected
and disconnected it can't reconnect until I rmmod lirc_i2c and modprobe
lirc_i2c.

Malcolm

Hi!

Shaul Kedem "shaul_kedem@..." wrote:
[...]
>  I've recompiled the kernel and now I can modprobe
> lirc_serial (thnx!).
>
>  The problem still remains - I don't get the +10v from
> the RTS in the RS232 connection,

You should start a client that actually opens /dev/lirc.

Christoph

Hi!

malcolm "lirc@..." wrote:
[...]
> What's your secret?  I'm having problems with lirc_i2c.  I'm on kernel
> 2.6.4.  I had to use flameeyes patch just to get stuff to compile and now
> only one client can connect to lirc and even after that one has connected
> and disconnected it can't reconnect until I rmmod lirc_i2c and modprobe
> lirc_i2c.

This has been fixed in LIRC CVS two weeks ago. Of course lirc_i2c from  
CVS still does not work with 2.6.

Christoph

Christoph,
 Thanks - but that doesn't help, I tried running
mode2, irrecord and lircd but none works,

 Does the polarity and current change only if there is
a valid circuit attached or is it unrelated ?

Please advise,
Shaul

--- Christoph Bartelmus <lirc@...> wrote:
> Hi!
> 
> Shaul Kedem "shaul_kedem@..." wrote:
> [...]
> >  I've recompiled the kernel and now I can modprobe
> > lirc_serial (thnx!).
> >
> >  The problem still remains - I don't get the +10v
> from
> > the RTS in the RS232 connection,
> 
> You should start a client that actually opens
> /dev/lirc.
> 
> Christoph
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux
> Tutorials
> Free Linux tutorial presented by Daniel Robbins,
> President and CEO of
> GenToo technologies. Learn everything from
> fundamentals to system
>
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

----- Original Message ----- 
From: "malcolm" <lirc@...>
To: <lirc-list@...>
Sent: Sunday, April 04, 2004 4:01 PM
Subject: Re: 2.6 support



> What's your secret?  I'm having problems with lirc_i2c.  I'm on kernel
> 2.6.4.  I had to use flameeyes patch just to get stuff to compile and now
> only one client can connect to lirc and even after that one has connected
> and disconnected it can't reconnect until I rmmod lirc_i2c and modprobe
> lirc_i2c.
>
> Malcolm

I wanted to make a correction here (I know false reportings can make it very
hard for programmers).  I did more testing.
I found after modprobe lirc_i2c lots of clients can connect (not just one)
and use lirc provided the very first client that connected does not
disconnect.  As soon as it disconnects things break and no one else can
connect or use lirc until I rmmod lirc_i2c and insmod lirc_i2c.

Thanks
Malcolm

Hi!

Shaul Kedem "shaul_kedem@..." wrote:
[...]
>  Thanks - but that doesn't help, I tried running
> mode2, irrecord and lircd but none works,
>
>  Does the polarity and current change only if there is
> a valid circuit attached or is it unrelated ?

No, the pin will be set regardless of if there is a circuit attached.

The probability that this problem is related to 2.6 is near zero.

Christoph

Hi,

On Sunday 04 April 2004 12.47, Christoph Bartelmus wrote:
> > I have changed lirc_serial, lirc_sir, lirc_parallel, lirc_dev and
> > lirc_gpio to *compile* with 2.6.3.

lirc_serial and lirc_dev works with 2.6.4, but they are both being reported as 
unsafe in the logs:
Module lirc_serial cannot be unloaded due to unsafe usage in 
include/linux/module.h:516

Also, when using make install the .o-files are installed 
to /lib/modules/<kver>/misc instead of the .ko-files. But maybe that's a 
feature ;)

I had some problems with setup.sh. See attached patch for solution to my 
problem. Without it, setup.sh doesn't save my choices in DriverOptions and 
ConfigSoftware. I'm using lirc-cvs from today and cdialog 0.9b-20040316.

// Erik 

-- 
Thinking of Emacs as just an editor is like thinking of the space
shuttle as just another part of the public transportation system.
  -- Dave Kenny, comp.emacs

Erik Johansson <lirc@...>
http://www.lysator.liu.se/~erre/

Christoph,
 Would you be kind enough as to point me to where this
change in the pin polarity/etc. happens in the code? I
would like to learn how you did it and whether it
works on my machine...

 Thanks,
Shaul

--- Christoph Bartelmus <lirc@...> wrote:
> Hi!
> 
> Shaul Kedem "shaul_kedem@..." wrote:
> [...]
> >  Thanks - but that doesn't help, I tried running
> > mode2, irrecord and lircd but none works,
> >
> >  Does the polarity and current change only if
> there is
> > a valid circuit attached or is it unrelated ?
> 
> No, the pin will be set regardless of if there is a
> circuit attached.
> 
> The probability that this problem is related to 2.6
> is near zero.
> 
> Christoph
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux
> Tutorials
> Free Linux tutorial presented by Daniel Robbins,
> President and CEO of
> GenToo technologies. Learn everything from
> fundamentals to system
>
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

>>>>> "malcolm" == malcolm  <lirc@...> writes:

 >>  lirc_i2c works fine here. I guess folks don't complain unless
 >> there is a problem. hope this helps.
 >> 
 >> Thanks

 malcolm> What's your secret? I'm having problems with lirc_i2c. I'm
 malcolm> on kernel 2.6.4. I had to use flameeyes patch just to get
 malcolm> stuff to compile and now only one client can connect to lirc
 malcolm> and even after that one has connected and disconnected it
 malcolm> can't reconnect until I rmmod lirc_i2c and modprobe
 malcolm> lirc_i2c.

I used flameeyes patch with 2.6.5. The Kconfig patch did not apply so
applied that one by hand. 

I used the following patch that some nice person posted on this list
last week to correct the problem with reconnecting you described
above. Hope this helps. My system is working perfectly with a
hauppauge 350 and mythtv. 

--- lirc_dev.c.org  2004-03-31 08:29:20.000000000 -0600
+++ lirc_dev.c  2004-03-31 08:29:34.000000000 -0600
@@ -156,7 +156,7 @@
            } else {
                interruptible_sleep_on(ir->p.get_queue(ir->p.data));
            }
-           if (ir->shutdown || !ir->open) {
+           if (ir->shutdown) {
                break;
            }
            if (!add_to_buf(ir)) {

***********************************************

Good luck,

Alan

Hi!

Shaul Kedem "shaul_kedem@..." wrote:
[...]
>  Would you be kind enough as to point me to where this
> change in the pin polarity/etc. happens in the code?

It's in init_port().

Christoph

Hi!

Erik Johansson "lirc@..." wrote:
[...]
> I had some problems with setup.sh. See attached patch for solution to my
> problem. Without it, setup.sh doesn't save my choices in DriverOptions and
> ConfigSoftware. I'm using lirc-cvs from today and cdialog 0.9b-20040316.

Your patch breaks things for most other people. I think this might be a  
bug in your dialog version. It does not happen with dialog
0.9a-20020309a. But I have put a work-around in CVS anyway.

Christoph

Hi,
 Testing lirc_serial on 2.6.5 with the gentoo patches
version ("gentoo-dev-sources") I am happy to report
that the current version of lirc_serial in the cvs
made my transmitter work and I can see the +10v that
should be supplied for the receiver, I got some
technical difficulties right now but I hope I will be
able to test the serial receive some time soon,

 so, just to clarify: without having anything to check
with (no "real" codes because the receiver doesn't
work yet) I can say that I saw the "improved
transmitter" blink :-)

Shaul

--- Christoph Bartelmus <lirc@...> wrote:
> Hi!
> 
> Shaul Kedem "shaul_kedem@..." wrote:
> [...]
> >  Would you be kind enough as to point me to where
> this
> > change in the pin polarity/etc. happens in the
> code?
> 
> It's in init_port().
> 
> Christoph
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux
> Tutorials
> Free Linux tutorial presented by Daniel Robbins,
> President and CEO of
> GenToo technologies. Learn everything from
> fundamentals to system
>
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click


__________________________________
Do you Yahoo!?
Yahoo! Tax Center - File online by April 15th
http://taxes.yahoo.com/filing.html