On 12/19/2011 03:20 PM, Somebody in the thread at some point said:
On 12/19/2011 01:16 PM, Somebody in the thread at some point said:
About Panda UCM, last week I noticed though that there is different wiring on input side between 4430 Panda (line in goes to analogue headset mic inputs on twl6040) and 4460 Panda (line in goes to FM analogue inputs). So this week I plan to make changes to the onboard audio driver to change its alsa card name based on what board it's on, with a view to elaborating out the UCM directories so the recording case is handled correctly depending on the board.
That's my concern too. The first thing is we need directories under /usr/share/alsa/ucm with the same audio card names on both panda boards. The second thing is to have correct configurations for both boards. I don't have a 4460 board, Andy, can you help on giving the ucm configurations for it?
Yes I'm hoping to sort it out. I'll give you the patches on UCM when they're done.
Okay thanks to Seb for sending the udev and 4430 changes, adapted a bit by me, along with my changes to ucm settings and a kernel patch to create a different audio device name for PandaES ('PandaES'), pulse is kind of working mostly here.
The attached files need going into the rootfs at:
/lib/udev/rules.d/90-pulseaudio-omap.rules /usr/share/pulseaudio/alsa-mixer/profile-sets/4430.conf
and unpack the tarball in /usr/share/alsa/ucm.
It's able to dynamically switch between audio outputs here using pulse interface, but it crashes (recoverably) sometimes doing that.
I wasn't able to convince the pulse UI to do anything with the 'record' UCM setting, although I adapted it for Panda / PandaES routing.
The matching kernel patch you need is now in tilt-3.1 and tilt-linux-linaro-3.1 which Linaro packages.
-Andy