[Gta04-owner] The sound level changes with or without remote USB cable

[EdorFaus]

Hi,

So, to summarize our findings so far:

You can always hear what the other end says.

When using the handsfree profile (which means software routing):
- they cannot hear you, regardless of USB cable connection.

When using the earpiece profile (which I assume means hardware routing):
- when the USB cable is not connected, they cannot hear you
- when the USB cable is connected, they can hear you
- both of these cases have identical ALSA states.


This is... odd, to say the least.

Hm. I think the next thing I'd try would be to record some audio from 
the mic, without involving phone calls, just to check if the mic-to-SoC 
audio path works. If it does, the problem is apparently closer to the 
modem than to the audio chip or mic (unless it's a system problem).

Simply using arecord didn't work for me when I tried it myself, so I 
found this[1] page that explains how to do it. Unfortunately, trying 
that just gave me a high-pitched sound after the third-last line (at 
which point I had to abort the attempt), so something's apparently weird 
about my setup, and I can't really experiment with it right now, but you 
may have better luck.

[1] http://projects.goldelico.com/p/gta04-kernel/page/Sound/#wikititle_5



As far as I know (correct me if I'm wrong), the mic is connected to the 
audio codec which is connected to both the modem and the SoC, and the 
SoC is also connected to the modem.

Using hardware routing, only the mic, codec and modem are part of that 
audio path, while with software routing, the SoC (and its software) is 
in between the codec and modem.

Assuming the codec's state is fully represented by the ALSA state, which 
did not change between the working and non-working cases, the codec 
chip's state is apparently fine.

That leaves the mic or modem, or possibly the interconnects between them.

The mic does not *have* a state as such, AFAIK, being a fairly simple 
part that's hardwired to the codec.

Which leaves either the modem state, or a hardware issue.

And, frankly, I'm starting to suspect a hardware issue regardless of the 
result of the above test, as I don't see why the USB (power) connected 
state should affect the modem state (or the state of the other parts for 
that matter), while hardware issues can be affected by even very small 
changes in such things...

-Frode