Cyclom Multi-Port Cards
Gert Doering (gert@greenie.muc.de)
Wed, 25 Nov 1998 09:34:10 +0100
Hi,
On Wed, Nov 25, 1998 at 01:28:30AM -0600, Robert Canary wrote:
> > setserial is NOT recommended AT ALL. Mgetty and most other useful
> Yes, that's pretty much what they are saying.
No, it's not. They explicitely recommend using setserial to get 57600 and
115200 bps.
> > programs can use 57600 and 115200 perfectly, and using setserial just
> > confuses the issue.
>
> Hmmmm, If a port defaults to .... se` 38.4k baud, then it seems to me
> that no matter what the application is putting out the port will only do
> 38.4k baud.
Well, no. The application can always change the port speed. Setserial
will only influence what the port will do if the application requests
38400 (will that be "38400", "57600" or "115200"). As modern applications
can request 57600 and 115200 directly, there is no need for setserial
anymore.
> function mgetty(reset_CPU_to_port_speed){
> case in "reset_CPU_to_port_speed"
> cannot)
> setserial /dev/ttyC0 spd_vhi
> msg="mgetty cannot control port to CPU speeds, using setserial"
unnecessary, as mgetty can *always* set any speed it wants. This is what
I keep saying: "setserial spd_[v]hi" is not needed anymore. Forget about
its existance.
[..]
> Does mgetty reset the CPU-to-port speed? If it does then what's the
> problem
It doesn't reset "setserial" follies, 'cause this is very much nonportable
and I refuse to add code for it. It *does* log a warning message if you
have used setserial, though.
> > Linux 2.3 will drop spd_hi and spd_vhi.
> Main line Linux ditributors are still passing out 2.0.xx
I was just saying that using setserial now is a bad idea, because it will
stop working RSN...
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany gert@greenie.muc.de
fax: +49-89-35655025 gert.doering@physik.tu-muenchen.de