Re: XON/XOFF communication problem
From: Floyd Davidson (floyd_at_barrow.com)
Date: 11/06/03
- Next message: David Douthitt: "Re: Are sysadmins going to jump the RH ship?"
- Previous message: Michael Heiming: "Re: Are sysadmins going to jump the RH ship?"
- In reply to: Andrej T.: "Re: XON/XOFF communication problem"
- Next in thread: Andrej T.: "Re: XON/XOFF communication problem"
- Reply: Andrej T.: "Re: XON/XOFF communication problem"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 05 Nov 2003 14:37:46 -0900
"Andrej T." <andrej@rikom.si> wrote:
>Floyd Davidson wrote:
>> The cable schematic looks fine. What kind of cable is it physically?
>> And how long is it? At 115.2kbps you can't expect it to work over
>> any significant length. More than a few feet probably won't work.
>
>The cable is 1m long so I guess this shouldn't be the
Heh, at 1m it's hard to imagine it wouldn't work with wet
string... :-)
>problem. And it is an UTP cable, the same that I'm using when
Is this a cable you made?
Twisted pair is a _bad_ idea for RS-232 cabling, unless it is
done just exactly right. For example, the TxD and RxD is *not*
a balanced pair and using one pair of a UTP cable for those two
leads will result in crosstalk between them. The TxD and RxD
leads should each be paired with some lead that will *not* be
changing, such as DCD, RING, DTR, or whatever. The same goes
for CTS/RTS, except since that operates (generally) at a much
lower rate there won't be nearly the same problem.
I'm not sure if a 1m cable would be long enough to exhibit this
or not, but it might be. On the other hand, I'm certain that
at 38.4Kbps it wouldn't be a problem.
>connecting the VT 510 terminal to an Etherlite (which btw. works
>fine, except that the schematic is different).
How is the schematic different???
>> What happens at speed higher than 19.2Kbps? Do you get the same
>> kind of errors?
>
>Yes, over 19.2Kbps I get the same errors then with XON/XOFF
>communication. I will test the cable again with the script that
>opens the port just once and will write the expiriances in the
>next post.
>
>> You did change the gettydefs to enable hardware flow control,
>> right? That is the -CRTSCTS entry.
>
>Before running the script I have set the port with the command :
>"stty -F /dev/ttyS0 -ixon -ixoff crtscts". But as I said before
>I will test the cable again to be sure!
Hmmm... you may have to get that into the same process that
opens the port.
(
stty -ixon -ixoff crtscts
while read input < filename
do
echo $input
done
) < /dev/ttyS0 > /dev/ttyS0
>But it's still a mystery to me why the XON/XOFF communication
>isn't working... How can I locate the author of the XON/XOFF
>communication code?
I have no idea!
>Is there a standard that says that XON/XOFF
>communications is only working at certain speeds?
Nope.
>I know this is
>an old protocol, but I would like to know if this is possible in
>Linux or not.
I'm amazed that you can't get anything working. I don't
remember ever trying using XON/XOFF, but I have setup Linux
systems to do computer to computer serial port comm at 115.2Kbps
and didn't have a problem. It's been awhile though.
I may just give it a try, just to verify for you that it can (or
can't be done). It kinda fits into another project... I just
wrote 5 different "terminal programs" in 5 different ways, to
demonstrate how various types of serial IO is done (async,
canonical, two processes, two threads, etc. etc). I've only been
testing them by talking to a modem at 57.6Kbps. One purported
"valuable" use of this type of program is to use the terminal
program on a separate computer, connected to the development
system by a null modem, to aid in testing serial comm programs
under development. Hence I really should verify that the
programs work on my hardware at the highest rate that I can get
the serial ports to run at.
I'll see if I can find a null modem and get back...
-- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com
- Next message: David Douthitt: "Re: Are sysadmins going to jump the RH ship?"
- Previous message: Michael Heiming: "Re: Are sysadmins going to jump the RH ship?"
- In reply to: Andrej T.: "Re: XON/XOFF communication problem"
- Next in thread: Andrej T.: "Re: XON/XOFF communication problem"
- Reply: Andrej T.: "Re: XON/XOFF communication problem"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|