Hi,
On 03-06-15 22:03, Andrea Venturi wrote:
hello,
i started testing the 4.1rcX kernel
from https://github.com/jwrdegoede/linux-sunxi/blob/sunxi-wip/ to check how
the sunxi-codec was working on my Cubieboard A10 board.
i found some pitfall that's why i'm writing here for a feedback. BTW i know
it's been mainly developed and tested on A20 Cubietruck so more testing on
different HW should help iron out all the quirks, right?
Right, more testing is always appreciated.
after compiling the kernel, i had to add the proper codec entry in the
sun4-a10.dtsi and i borrowed the one from the A20 dtsi here:
https://github.com/jwrdegoede/linux-sunxi/blob/sunxi-wip/arch/arm/boot/dts/sun7i-a20.dtsi
i've got wav samples from:
http://www-mmsp.ece.mcgill.ca/documents/AudioFormats/WAVE/Samples.html
the sample i'm using is 16 bit stereo with a man voice on the left channel
and a woman voice on the right (as far as i can understand!):
http://www-mmsp.ece.mcgill.ca/documents/AudioFormats/WAVE/Samples/AFsp/M1F1-int16-AFsp.wav
when i tested it the first time, with default "card mixer setting", no
audio, no joy! :-), with aplay hanging and mplayer spitting out
...
AO: [alsa] 8000Hz 2ch s16le (2 bytes per sample)
Video: no video
Starting playback...
A: 0.0 (00.0) of 2.0 (02.0) ??,?%
Audio device got stuck!
A: 0.0 (00.0) of 2.0 (02.0) ??,?%
Audio device got stuck!
then, mripard on #linux-sunxi told me to modify at least the interrupt line
that was incorrect as they are different between A20 and A10 dtsi:
A20: interrupts = <0 30 4>; A10: interrupts = <30>;
BTW i've been told that this codec has been tested also on Mele A1000
that's A10 based STB too, but i failed to see any trace of it in the
sunxi-wip so don't know if this report has been done.
Right the patches for the Mele A1000 where still floating around on the
list, I've picked them up now and added them to my sunxi-wip branch,
so you may want to give things another try copying the relevant bits from
the sun4i-a10-a1000.dts file to the sun4i-a10-cubieboard.dts one.
anywa it turns out mripard was also right about trying to move the alsa
controls to start playback; indeed toggling the alsamixer setting, i
started listening music, at least from the left earphone.
the alsamixer setting are a but puzzling. you can have a look of a working
setup in this screenshot here: http://picpaste.com/sunxi-codec-mono.png
i started moving the knobs up and down, but i could just move from mute to
left ear and back;
* when it was written "mute" on alsamixer "DAC output" control, the audio
was playing on the left ear
* on the left ear i can hear both man and woman (mixed) when the "Right
Mixer" is on Right,
* on the left i can hear only the woman (that should be right channel)
when Right Mixer is Disabled.
so the mixer is not very right on my setup.. i know from the A10 user
manual that the "analog mixing" is quite a complex audio routing. i'll
check better if the alsa audio controls names are mapping the Allwinner
device scheme, but in any case it's a bit difficult to understand what the
controls do without "RTFM" (that from a user perspective is always a
nightmare! :-)
I've not looked into the mapping of the controls at all yet, we could
certainly use some help with this!
moreover when moving control to the position where it was muted, again on
mplayer console i could spot messages of "audio device is stuck" , and
again "toggling back" the "DAC output" control, the music starting again..
i suppose anyway that toggling alsa mixer controls should never "stop" the
DAC from sinking samples coming from the application level, right?
Right.
Regards,
Hans
--
You received this message because you are subscribed to the Google Groups
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to linux-sunxi+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
