Okay,

I think that both tuner signal and tuner level go straight to the DSP, so no way to claim an IRQ on raising edge or falling edge on those pins.

Pin 4 of the serial connector on the back of the Empeg itself is momentarily high right at the beginning of the power up sequence, then falls to -5V and stays there.

The various pins (CTS - headlight sense, DCD - telephone mute, and ignition sense) can be read via GPLR. empeg_power.c shows how to do this.

Incedentally, it would appear that the RDS data is processed by the DSP and then made available to one of the GPIO pins - see empeg_rds.c

The source says that SerialDTR is GPIO23. I tried to make pin 4 of the serial connector on the back ot the Empeg go +5V by doing this: "GPSR=EMPEG_SERIALDTR" ...in hijack.c, just below the initialisation code for I2C fan. Didn't appear to have any effect. Also tried clearing the register (in case the output is inverted).

I can conclude that either GPIO23 isn't conencted to pin 4 or something else is changing it back again. I've grepped the source, but can't find anything that might be chaning it back. There was reference to both the debug LED and SIR (serial IR?) - so perhaps GPIO23 is used elsewhere and is not connected to pin4. If this is the case, what is raising pin4 at boot-up?

So, is there any way to:
a) set any of the pins on the on-board serial connector to +5V (or greater) in software
b) generate an interrupt on a raising or falling edge on either the tuner_level or tuner_signal line