[mythtv-users] Joy and pain (Daryl McDonald)

NickS nicks272 at yahoo.co.uk
Fri Feb 14 12:48:32 UTC 2014




> looking at device '/devices/platform/lirc_serial.0/lirc/lirc0':
> KERNEL=="lirc0"
> SUBSYSTEM=="lirc"
> DRIVER==""
>
> looking at parent device '/devices/platform/lirc_serial.0':
> KERNELS=="lirc_serial.0"
> SUBSYSTEMS=="platform"
> DRIVERS=="lirc_serial"

From this output I would create the following rule:

KERNEL=="lirc*", KERNELS="lirc_serial.0", SYMLINK+="lircSerial"

The first part makes sure the rule matches the device at the lirc level of the device tree. Previously when I (and possibly you) queried (with udevadm info) against "event?" and tried to match with  KERNEL=="event*"  my udev rule would create a symlink
 but it seemed to be at the wrong level to work with lircd. If your setup works temporarily with /dev/lirc1 then that's the device you need to query against using:

udevadm test -a -p  $(udevadm info -q path -n /dev/lirc1)

and KERNEL="lirc*" is the level your udev rule should match against.

The second part matches with "lirc_serial.0" which appears to be the only uniquely identfying piece of info further up the device tree, so we match the correct device regardless of the lirc number.


The other issue I came across when testing the udev rule was that the filename in /etc/udev/rules.d made a big difference. Initially I'd created an 86-remote.rules, the file was picked up but the rule didn't take effect. Naming it z21-persistent-local.rules seemed to do the trick.

Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mythtv.org/pipermail/mythtv-users/attachments/20140214/b2a0c7c0/attachment.html>


More information about the mythtv-users mailing list