[lkml]   [1998]   [Aug]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: DevFS vs. normal /dev (was DEVFSv50 and /dev/fb? (or /dev/fb/? ???))

On Mon, 10 Aug 1998, Horst von Brand wrote:

> How do you reference, say, /dev/fd0 if the floppy module isn't even loaded?
> You need to do that so kmod loads the module, as things stand now.
> --
> Dr. Horst H. von Brand
> Departamento de Informatica Fono: +56 32 654431
> Universidad Tecnica Federico Santa Maria +56 32 654239
> Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513

This is an issue I raised the other day. Currently, to enable dynamic
device loading, you must use the magic tar stuff to create the device
in advance with the correct permissions owner, and group. There is
a fundamental "chicken and the egg" problem with the auto-creation
of device special nodes and use of kmod to auto-load modules.
Now in _theory_ Devfs could intercept any query for the existance
of a device special file to then use a daemon to automagically create
device special files with the right permissions owner, and group,
and then if the permissions are correct, tell kmod to load the
module. Then if it doesn't work, you get the same behavior as now
when a device isn't supported by the kernel. And in theory, Devfs
could then, when the last reference to a particular device special
file is closed, tell the user space daemon to blow away the file.

Of course, it doesn't do this now.


Erik B. Andersen Web:
--This message was written using 73% post-consumer electrons--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:43    [W:0.066 / U:6.752 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site