Re: PROPOSAL: /proc/dev

[Richard Gooch]

James Mastros writes:
> On 1 Jan 1998, H. Peter Anvin wrote:
> > >
> > >   Hi, all. My recent work on the /proc/mtrr interface has gotten me
> > > thinking about an old discussion: configuring /dev entries. Instead=
>  of
> > > complicated schemes for determining what devices are available and
> > > automagically creating /dev entries, how about creating entries in
> > > /proc/dev instead?
> > >=20
> >=20
> > Not again...
> >=20
> > 	-hpa
> Why not?  If you could give us some good reasons why not then we probab=
> ly
> will drop it.  But several times, I have heard this come up, and never =
> a
> good reason why it shouldn't be done.  The two reasons that I have hear=
> d are:
> 1)  Memory usage -- so if you can't spare the memory, don't use it!

I haven't yet seen any reason why /proc/dev should take more meory
than the existing scheme.

> 2)  Lack of user configurablity (IE different permissions, different na=
> mes)
>     -- so we have callbacks through a userspace daemon.  Or we let peop=
> le
>     write really simple kernelspace modules.  (Basicly a big select and=
>  a
>     function call or three with constant arguments for each case.)

/etc/modules.conf:
options	pcwd def_prot=0666 name="mywatchdog"

should solve these problems. "def_prot" and "name" can be
standardised. A little bit of magic in an include file can even make
all this automatic.

				Regards,

					Richard....