Re: Migrating to larger numbers

Richard Gooch (rgooch@atnf.csiro.au)
Tue, 8 Jun 1999 16:28:48 +1000


Nate Riffe writes:
> On Mon, 7 Jun 1999, Dan Hollis wrote:
>
> > On Tue, 8 Jun 1999, Nathan Hand wrote:
> > > If devfs is so bad, could we have reasons why. Policy isn't a good
> > > reason (the policy's already in Documentation/devices.txt).
> >
> > Fear. Ignorance. Resistance to change. Not having read the documentation
> > or actually tried devfs. Etc.
>
> What exactly does hpa refer to when he says 'policy'? It's pretty vague
> to me.

One man's policy is another man's consistency...

> I've been using devfs for a while now. The only real problems I
> have with it are that changes to ownership and permissions are not
> inherently persistent and the chicken and the egg problem between
> device files and modules.

Persistence: that's currently pushed to user space, which is
consistent with one particular ideology.

Chooks and eggs: this is not a problem. Module autoloading works quite
well. All that is required is that /etc/modules.conf is correctly
configured, and attempts to access non-existent device nodes will
automagically cause the driver to be loaded and the nodes suddenly
appear. And before anyone complains about having to configure
/etc/modules.conf, I'll point out that you have to do that anyway to
get module autoloading to work.

Regards,

Richard....

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/