Return to BSD News archive
Newsgroups: comp.os.386bsd.development,poster
Path: sserve!newshost.anu.edu.au!munnari.oz.au!network.ucsd.edu!usc!howland.reston.ans.net!newsserver.jvnc.net!gmd.de!mururoa!veit
From: veit@mururoa.gmd.de (Holger Veit)
Subject: Re: Some ideas on the driver interface (New idea!)
Message-ID: <1993Mar22.153537.6212@gmd.de>
Sender: veit@mururoa (Holger Veit)
Nntp-Posting-Host: mururoa.gmd.de
Organization: GMD - German National Research Center for Computer Science
References: <1993Mar17.122612.5929@neptune.inf.ethz.ch> <1o9l9u$nn@tricky.wft.stack.urc.tue.nl> <1993Mar22.135730.8173@neptune.inf.ethz.ch>
Date: Mon, 22 Mar 1993 15:35:37 GMT
Lines: 38
In article <1993Mar22.135730.8173@neptune.inf.ethz.ch>, weingart@inf.ethz.ch (Tobias Weingartner) writes:
|> In article <1o9l9u$nn@tricky.wft.stack.urc.tue.nl>, michael@wft.stack.urc.tue.nl (Michael Brouwer) writes:
|> > In article <1993Mar17.122612.5929@neptune.inf.ethz.ch> weingart@inf.ethz.ch
|> > (Tobias Weingartner) writes:
|> > > There. Simple no? Instead of making mount(2) take a special file,
|> > > make it take a string representing the device. The "dev" device would
|> > > be mounted on /dev, and emulate like it was a real /dev. This way
|> > > only the devices that were configured into the kernel would even
|> > > need to show up in "dev".
|> >
|> > This is a really nice idea, but you'll have to think of a way to implement
|> > things like `chgrp uucp /dev/com01'; `chmod 660 /dev/com01', or do you suggest
|> > rebuilding the kernel to modify `/dev' protection?
|>
|> No, the kernel (or driver) emulates the 'dev' device just like /dev. If you
|> do a chmod or chown or whatever, it will get reflected on the device.
|>
|> Hmm, makes me wonder what an rm or mknod would do... ;-)
I believe, 'rm' would remove a device from the table that is exported into the
world, it wouldn't remove the device itself or clear the [bc]devsw entry (or the
alternative representation of this). 'mknod' would create such a link between the
real driver structure and the visible filesystem. You need a two layering system
actually: The table containing the physical existing devices, and a logical
layer which maps the name space to them. The vfs calls operate on the logical
layer only.
Holger
|> +41'41'33'25'40 * 'finger weingart@tau.inf.ethz.ch' *
--
Dr. Holger Veit | INTERNET: Holger.Veit@gmd.de
| | / GMD-SET German National Research | Phone: (+49) 2241 14 2448
|__| / Center for Computer Science | Fax: (+49) 2241 14 2342
| | / P.O. Box 13 16 | Three lines Signature space
| |/ Schloss Birlinghoven | available for rent. Nearly
DW-5205 St. Augustin, Germany | unused, good conditions