FCDEV3B bring-up status

[Mychaela Falconia]

Hi Serg,

> It worked once with different SIM card, but it is definitely not a hardware
> issue because Magnetite build works every time with the same SIM card.
> Maybe L1/DSP is configured a bit differently and if my calibration values
> are a bit off, Citrine may not be able to detect carriers.

I don't see how the quality of calibration could be the cause of the
problem you are having with Citrine, as your log shows that the fw
does not even attempt to connect to a cell, i.e., the upper layers of
the G23M protocol stack are not commanding L1 to make a network sync
attempt.  And because L1 is not being commanded to do the operation in
question, I don't see how it could be an L1 or DSP problem either -
instead it must be some glitch higher up in the protocol stack code.

The G23M PS used in Citrine is not too polished, i.e., it may still
have bugs that need to be shaken out, and the issue you are having
looks like one of them.  However, my preferred approach to debugging
these G23M PS problems is to debug them first in the TCS2/TCS3 hybrid
config of Magnetite (see doc/Modem-configs in the fc-magnetite source
tree), and then port to Citrine if there is commercial demand for the
latter.

Thus you might want to try building Magnetite in the hybrid config
(different from the l1reconst config I currently use and recommend for
production use), and see if the SIM that's giving you trouble in
Citrine works in Magnetite TCS2/TCS3 hybrid or not - this experiment
would narrow the search for the bug.

M~