Subject: Re: troubles with serial console support....
To: None <port-i386@netbsd.org>
From: Greg A. Woods <woods@most.weird.com>
List: port-i386
Date: 01/03/1999 11:04:24
[ On Sun, January 3, 1999 at 12:29:19 (+0100), Martin Husemann wrote: ]
> Subject: Re: troubles with serial console support....
>
> > Unfortunately I don't have a data scope, or even a working oscilloscope
> > at the moment, though my trusty RS-232 breakout box is hinting to me
> > that the PC is dropping RTS after accepting the first character of
> > input
> 
> I use a three wire cabling, as my DEC VT420 doesn't do RTS/CTS.
> This works fine. 

In my case the problem isn't with the cable or the null modem
configuration.  As I mentioned I tried every combination of such things.
The problem is that the PC really doesn't want to receive characters.
That it says so by lowering RTS is essentially irrelevant.
Unfortunately ignoring RTS doesn't help get characters through to the PC.

> What happens if someone switches off the console when you have RTS/CTS
> handshake? You'll probably block your system rock solid at the next
> kernel printf...

Hmmm, yes, that's probably something that should be mentioned in the
boot_console(8) manual page....  I'll add it to the copy I send with my
proposal.  Thanks for the reminder!

Normally in a production environment serial consoles are used on servers
that are either safely tucked into a locked computer room.  In my case
their console cables are normally connected to a terminal server or
similar as well.  Anyone who turns that device off will suffer the same
penalty they would receive for turning off the server itself.

I do want to use the DIRECT_SERIAL option in my production kernels,
partly to avoid this problem, and since that's the only way to get
XON/XOFF flow control support currently.  (I will not do without flow
control on the consoles, which tying RTS/CTS up would result in with the
BIOS I/O option.)

Also, in modern day machines with syslog it's rarely helpful to print
messages to the console, especially when that console is not a printing
terminal, and it's either multiplexed on a terminal server or 

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>