Re: Linux-2.0.1 (patch in serial.c is a problem)

Carsten Paeth (calle@calle.in-berlin.de)
Fri, 5 Jul 1996 14:54:18 +0200 (MET DST)

>
> On Wed, 3 Jul 1996, Carsten Paeth wrote:
>
> > The patch in serial.c breaks all getty's (mgetty/agetty/uugetty)
> > on serial lines, because CLOCAL is not long set by default.
> > I undo this patch for me, now it is working.
>
> I just upgraded a system to 2.0.1 last night. It uses Cyclades ports to
> answer 50 modems. I noticed no problems. What exactly do you mean by
> "breaks all getty's"? I'm using agetty on the VC's and mgetty on the
> ttyC's.
>
> ------------------------------------------------------------------
> Jon Lewis | Mime attachments are OK
> jlewis@inorganic5.fdt.net | But please ask before sending
> http://inorganic5.fdt.net | unsolicited huge files.
> ________Finger jlewis@inorganic5.fdt.net for PGP public key_______
>

The mgetty/agetty hang in open(2) because CLOCAL is not set.
getty doesn't hang, but if login try's to reopen the serial line
in hangs.

The relevat part, was only changed in serial.c NOT in cyclades.c
or an other serial driver.

new serial.c:2758:
serial_driver.init_termios.c_lflag &=~ (ISIG | ICANON | ECHO);
serial_driver.flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_RESET_TERMIOS;

old serial.c:2758:
serial_driver.init_termios.c_cflag =
B9600 | CS8 | CREAD | HUPCL | CLOCAL;
serial_driver.flags = TTY_DRIVER_REAL_RAW;

cyclades.c:2792:
cy_serial_driver.init_termios.c_cflag =
B9600 | CS8 | CREAD | HUPCL | CLOCAL;
cy_serial_driver.flags = TTY_DRIVER_REAL_RAW;

calle

-- 
calle@calle.in-berlin.de