Re: PUBLIC CHALLENGE: (was RE: devfs again, (was RE: USB device alloc ation) )

david parsons (o.r.c@p.e.l.l.p.o.r.t.l.a.n.d.o.r.u.s)
8 Oct 1999 14:11:15 -0700


In article <linux.kernel.Pine.BSF.4.05.9910080513480.50036-100000@idiom.com>,
M Carling <m@idiom.com> wrote:
>
>David Parsons wrote:
>> Well, the best way to fix the implementation would then be to
>> include it in 2.3.x and let the legion of kernel hackers fix the
>> problems with it.
>
>While I'm not sure there is really anything about devfs that needs fixing,
>if there is then surely it can and should be fixed before including it in
>the kernel.

There's one massive missing feature in devfs: I can't union-mount it
over an ext2 fs and have it inherit and propagate permissions and
ownerships down to the underlying layer. Devfs works just fine
without it (If there were problems with it, I'd not have been able to
do my patches to the memory detection code, because major number
creep breaks too many things otherwise.)

There are people on the list who, at least from what they write,
might be able to put this functionality in without even breaking a
sweat, but until devfs is in the kernel (at a whopping 33k space
penalty -- 12k in bzImage -- as of 2.3.19) it's a bit optimistic
to assume that they'll carefully hand-patch their kernels, add
the unionfs, and tweak vfs to fit for every single kernel release
from now until the time the core team votes to allow this feature
in.

If 2.3 is now feature frozen, this is distressing. It means that
I'll have to not port Mastodon to 2.4, but will have to stick with
2.0.28+logo, because I'm not going to upgrade to a kernel that
doesn't use such a vital piece of functionality.

____
david parsons \bi/ Of course, I'm fully expecting 2.5 won't support
\/ binfmt_aout...

-
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/