DSP TCH task : time to get/put FAACH

Sylvain Munaut 246tnt at gmail.com
Fri Oct 22 21:47:16 UTC 2010

Some news from the front on this :

>  * When to put / read data for FACCH.
>   [snip]

Ok, I've accepted it's a requirement of the DSP ... (I looked into the
DSP code to make sure it was) ...

>  * TCH/H support: AFAIK, I did everything that should be required for
> it to work:

After fixing some bugs, and some debugging sessions with steve|m (thanks btw :):
 - Half subchannel 0: Call establishement / FACCH works. But no audio.
Steve confirmed that audio _is_ sent from the mobile, by recording it
with the racal and then playing it back on a phone with stock
firmware. So it's a downlink issue ...

 - Half subchannel 1: FACCH works, but can't establish a call anyway
... The problem seems to be in the upper layer. My current theory is
that we hit some race condition somewhere and the slight timing
difference between subchannel 0 and subchannel 1 triggers it in one
case but not the other ...

   Basically on FACCH:
   - we send the SABM
   - we get the UA back
   - we send the ASSIGNEMENT COMPLETE N(S)=0, N(R)=0
   - we get RR N(R)=1
   - we send 3 times RR N(R)=0  ( <-- this is weird )
   - then lapdm doesn't xmit anything from the upper layer anymore.



More information about the baseband-devel mailing list