lkml.org 
[lkml]   [1997]   [Nov]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: TTY changes to 2.1.65
On Tue, 25 Nov 1997, Theodore Y. Ts'o wrote:
> From: Jes Degn Soerensen <jds@kom.auc.dk>
>
> I see no reason for wasting all these majors on this and IMHO it is a
> lot better to use dynamic allocation of serial devices, thus if one
> replaces a serial board with a different brand it is not necessary to
> go and change the entire software setup.
>
> It's something to think about, but doing this in a way that doesn't
> break compatibility with either (a) existing systems' /dev directories
> and (b) all of the existing tty device drivers will be a wee bit
> challenging. It'll probably be a while before I get to this one,
> actually.

A possible intermediate solution would be a "generic" driver, with its
own major number, which is a run-time configurable thin wrapper around
"real" serial driver syscalls.

The implementation should be fairly simple... all syscalls to devices
with the "generic" major would get multiplexed (based upon their
minor) to the "real" major handler, possible with the minor number
offset or mapped in some other way. This could be applied in general
to any character device, or even block devices for that matter...
(another way of looking at the SCSI numbering problem?)

Of course, there's added overhead on every syscall using the "generic"
device. But that's a trade-off some might be willing to make... (so,
_how_ often do you change serial boards?)

Just an idea.
Adam
--
Things look so bad everywhere Adam D. Bradley artdodge@cs.bu.edu
In this whole world what is fair Boston University Computer Science
We walk blind and we try to see Ph.D. student and Linux hacker
Falling behind in what could be ----> Bring me a Higher Love ----> <><

\
 
 \ /
  Last update: 2005-03-22 13:40    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site