Linux-Hams archive - May 1998: Re: Baycom problem solved.

Re: Baycom problem solved.

Gerd Rthig Sekt. Medizin SS97 (bpdzm@mailit.tunk.net)
Mon, 4 May 1998 02:57:50 +0200 (DFT)


Hello Alex, hello all,

reading the Linux-HAMs mailinglist or its archive you could avoid such
troubles.
About a year ago, we had exactly the same problem, and Thomas Sailer, the
author of the BayCom and SoundModem driver, wrote a very good explanation
for that behaviour.
I agree with you that there should be a message telling you about weird
UARTS, if possible.
This problem is also mentioned in my translation of Terry Dawson's
AX25-HOWTO into German on

http://www.ardos.de/gerd/axhowto.html

along with a solution that helps you out with some of the dodgy UARTS.
Sometimes, the command sequence

/sbin/setserial /dev/{your serial port} uart 16550A skip_test
/sbin/setserial /dev/{your serial port} uart none

before initializing the BayCom driver (which we compiled as a module)
will do the trick.

> Admittedly, I haven't managed to connect to anything yet, but I can see
> all the packets going past, and it keys the transmitter up and generates a
> signal on the audio output whenever I try to transmit a packet. I think
> it's probably just something simple in the setup of the audio output
> stage, possibly the volume pot set wrong. I will have to borrow a mate's
> handheld on Tuesday so I can listen to what kind of a signal it is
> generating.

Possibly, the TXDELAY is set too low. Try to set TXD to higher values,
maybe this already helps. Also, setting a higher value for TXTAIL (@T2
parameter on TNC2) may be handy.

> I wonder if there is any way the baycom driver could be made to detect
> these evil deformed UART chips, and inform the user of the problem, so he
> won't have to go through what I did? The symptoms are that the driver
> detects and initialises the UART okay, and thinks everything is fine and
> dandy, but no interrupts are coming through. Perhaps the driver could time
> out if it doesn't recieve any interrupts from the UART after a set amount
> of time, and complain that either the interrupts are set up wrong or you
> have a dodgy UART. There should at least be a brief mention of the problem
> in the documentation for the driver, and probably the AX25 howto.
>
Good idea, I fully agree with you.

Cheers

Gerd