> On Wed, 26 Nov 1997, Richard Masoner wrote:
>
> > > 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,
> >
> > Nice idea; unforunately the number of bits in the minor number is
> > already a limiting factor in Linux.
>
> We could implement /dev in the same way we implement /proc...
Yes, and I'd love it as it is great on AIX already. Probably make it
/proc/dev though. Those who want it can use it by a symlin to /dev. The
other once not. You must also allow to add further symlinks to /proc/dev
(for /dev/mouse and such) + probably adding further entries by hand with
mkdev (such that someone can add another device node for the same device,
but with different permissions.). A tool to read out the current usermods
(owner ship and permissions) into a plain text file to be executed at boot
time to reestablish the same setup would be nice.
On contrary to zillions IMHO bloated and pointless stuff in /proc also
always being kind of a security issue this would a really good thing to
have. Still, the kernel shall not impose any policies, hence enforcing
standard names and such this way, may not be liked everywhere.
All this has been discussed for ages. So, either someone code it or stop
this FAQ thread now.
Note though, that this is not related to the bit width issue at all. A
single modification to mknod and e2fs and anyone can do the same with an
e2fs mounted /dev. Alas ANY program likeing to stat() one of the /dev
entries will get confused and need recompilation, if not fixing. THIS
is the major issue about more dev_t bits.
Michael.
(eowmob@exp-math.uni-essen.de or eowmob@pollux.exp-math.uni-essen.de
Please do not use my vm or de0hrz1a accounts anymore. In case of real
problems reaching me try mat42b@spi.power.uni-essen.de instead.)