From: [EMAIL PROTECTED] (Jeremy Hall)
Date: Mon, 20 Sep 1999 23:33:53 -0400 (EDT)
hmmm it's a libraries thing. I did
/usr/sbin/chroot /dead /usr/bin/cu -l /dev/ttyS1 -s 9600
/dead is the root filesystem of another machine. Obviously this messes
with ttyS1 on that machine while I have ttyS1 on pongo open, but this
indicates to me that somehow the libraries are not happy about this tty
stuff. What do I do?
Err.. no. Unix devices are local to the machine, and /dev/ttyS1 is
merely an inode which states "open the device corresponding to major
number 4, minor number 65" on the local machine. Doing a chroot to
another machine isn't going to mess with anything other than the local
system.
If the above works, then you may be right that it's likely an issue with
either your shared libraries. It could also be an issue with the actual
/usr/bin/cu.
Over the years, various dialout programs have done enough "squinky"
things that I no longer trust them. I tend to use kermit, since it lets
me control all of the serial parameters directly via "set" commands, and
it doesn't do anything extraneous that I didn't ask it to do.
As far as why some device works fine under a 2.0 kernel and not under
2.2 kernel, another general reason why it might not work is that in the
2.2 kernel, we no longer autoprobe irq's by default. This was getting
things wrong too often, and was a real support nightmare for me. Since
most people using COM 1-4 devices usually used the default IRQ,
ironically I got less support questions after I disabled IRQ
autoprobing. So you might want to check the hardware serial
configuration using the setserial command "setserial /dev/ttyS0" and
make sure that it's what you expect.
- Ted
-
To unsubscribe from this list: send the line "unsubscribe linux-serial" in
the body of a message to [EMAIL PROTECTED]
