I'm sorry but this is uncalled for. Some people are excessively crass
when they don't like something. Having a completely magic /dev appears to
be perfectly acceptable to plenty of people. It's a kernel option not a
requirement, let it be. In point A you bespeak evil of hard coded things
and yet you want a hard coded template in point D.
As for point E, no. Flat out no. The current design of devfs is heavily
changed due to a lot of influence from Linus and related about how it should
be done. And for inflicting it upon you, again it is an option you
certainly are not required to implement either in part or whole. It is not
the perfect solution everyone wants but Linus has accepted it for the time
being.
-d
> OK, since I did a lot of wading through the mount-related code lately, let
> me add a couple of things:
> a) fixed *WHAM* pathnames *WHAM* are fucking evil *WHAM*
> b) there is no real need to mount shmfs to get the things working.
> Provable: I've done that.
> c) the whole concept of monolytic devfs is a bullshit, excusable
> only because we lack union-mounts. As soon as they are in the devfs will
> be heading out. There _is_ a point in a driver providing a tiny filesystem
> with right set of devices and user union-mounting it on /dev,
> /dev/tape, /dev/hamster/duct/tape, whatever. But 'register a set of stuff
> into the devfs' thing is a bad design.
> d) funny as it sounds, in-kernel template for root is not a bad
> thing. Provided the presense of union-mount, again. It would make for more
> regular code in initialization sequence.
> e) yes, Richard, we all know that you've got your own vision of
> the way things should be done. That's OK, just don't inflict it on those
> who don't want it, will you? Oh, and search on the net for "Mr. Bill's
> DEC-20".
-
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/
This archive was generated by hypermail 2b29 : Fri Mar 31 2000 - 21:00:24 EST