You wrote
> After being up for about 3 days, Emacs started putting `^@' at the
...
>
> Further investigation showed it wasn't Emacs' fault. The child process
...
> Another surprise is that when simply opening /dev/ptyp1 and reading, the
> `read' call returned EIO until /dev/ttyp1 had been opened for the first
> time. (Both `cat < /dev/ptyp1' and tracing Emacs confirmed this). When
...
> I can't reproduce this, but I have seen the same problem with older
> kernels. I don't understand enough of the tty code to find whatever is
> causing it.
...
> fine, so I can't reproduce the bug described above.
>
> At least I know how to work around the problem now: search for a rogue
> process that still has /dev/ttyp1 (or whatever) open, and kill it.
I think I've seen similar problems for many, many revisions. I use
GNU screen all the time; sometimes it will fail to open a pty for a new
screen (no error message) after some other screen is killed. It seems
to happpen especially when I abruptly kill a screen during I/O.
If I then open an xterm that uses the pty that screen couldn't get
to work, typing in it results in spaced characters like this:
helix:[jf]~% l s - l / h o m e
This looks consistent with what you're describing above.
I've had limited success fixing this problem with the reset program
and other random things; sometimes it goes away for a while. However,
I recently "corrupted" a pty and it was useless (with -EIO when I
tried to open it) after that.
This seems like a pretty serious bug. Comments?
_
__ .(_
||
/
-----BEGIN PGP SIGNATURE-----
Version: 2.6.3
Charset: noconv
iQBFAwUBMbJT6caxFdsA0XPZAQE3fAF/YkNEYsdOxrC9tbs5Ay+iXmg9Cs75anRb
xZ59oTXuZyGLqsqLpsuTsJgBHSfYhvq/
=ozl0
-----END PGP SIGNATURE-----