[GTALUG] Actual ttyS0 MIA
Stewart C. Russell
scruss at gmail.com
Sat May 9 19:25:36 EDT 2020
On 2020-05-07 4:09 p.m., Alvin Starr via talk wrote:
>
> The headers and cables tended to be pretty standard.
Thanks, Alvin. TIL I learned that they're actually not. Well, there's
one standard, and there's the thing that ASUS uses. Guess who bought a
"standard" cable but has an ASUS motherboard?
The "standard" way assumes an IDC 10-way header on the motherboard and
an IDC 9-pin RS-232 connector on the other end. Because of the
difference in pin ordering, a straight-through ribbon cable ends up
mapping header pins 1-9 to 1, 6, 2, 7, 3, 8, 4, 9.
1 2 3 4 5 6 7 8 9
1 2 3 4 5
6 7 8 9
You can see the pinout for this on a weird little embedded 386SX clone I
have (ZF MicroSystems OEMmodule):
https://archive.org/details/zf_systemcard_technical/page/10/mode/2up
ASUS, bless 'em, decided that they should be different and mapping
header pins 1-9 to RS-232 connector pins 1-9. More logical, maybe, but
it means you can never use IDC 9-pin connectors with an ASUS board.
I've rewired it and all is well.
> There are a number of programs for managing serial ports but I have had reasonably good luck with minicom.
minicom's usually my go-to program too. One of my devices insists on
talking 7M1, and minicom can't set that as a default. I had to compile
the venerable C-Kermit for out-the-box mark parity support. C-Kermit
doesn't come as an Ubuntu package, and building it is a bit special.
CUNY stopped supporting it (and Frank) in 2011, so development has
slowed way down. A shame, 'cos you can bootstrap Kermit using almost
anything and end up with a robust 8-bit file transfer path.
cheers,
Stewart
More information about the talk
mailing list