Dr. H. Nikolaus Schaller
hns at goldelico.com
Fri Sep 2 08:20:13 CEST 2011
Am 31.08.2011 um 21:51 schrieb Dr. H. Nikolaus Schaller:
> Hi all,
> again it is time for some status.
> 2. OPTION GTM601
> we have not yet had success to access the PCM interface
> from the Linux side. The problem is that we don't understand
> ALSA SoC good enough to add a separate McBSP channel
> that either becomes visible as a separate sound card or
> as another channel or subdevice of the existing sound card
> for Earpiece, Microphone etc.
> If there is someone on this list who is knowledgable about
> the ALSA SoC architecture and drivers, please let's discuss.
Here we still need some support.
> What I have tested is to map the PCM interface pins to
> GPIOs and look through them if there are signals. Good news:
> yes, there is the PCM clock and the frame sync. But on
> my two boards I can't see PCM data. Tomorrow the team
> will meet and we will analyse two more boards.
> This can again be a solderin issue because the missing
> signals are in one corner of the LGA and LGS modules
> may have soldering problems by PCB warping.
We had to inspect only one board and could see clock and
data signals change when doing a phone call. This means
the hardware is basically ok and just my board has a production
So if we get the ALSA SoC driver fixed, we can do telephone
And, we have other good news. We have found the reason
why the USB-OTG port isn't working reliably. The reason
is that the ESD filter chip is specified for USB 1.0 only and
deteriorates the USB 2.0 signals too much. Knowing this,
we still wonder that we had the interface partially working...
The solution is to skip this chip in the early adopter production
run. What we loose is ESD protection on the USB connector,
so please be careful...
Since there appears to be no drop-in replacement chip,
we will finally solve this in the next board layout (e.g. with a
More information about the Gta04-owner