Messages in this thread | | | Date | Wed, 19 Jul 2000 11:17:40 -0500 (CDT) | From | "Forever shall I be." <> | Subject | Re: devfs question |
| |
Khimenko Victor wrote:
> > > On Tue, 18 Jul 2000, Richard Gooch wrote: > > > Andrew Pimlott writes: > > > On Tue, Jul 18, 2000 at 06:20:07PM -0400, Richard Gooch wrote: > > > > Andrew Pimlott writes: > > > > > Perhaps best to look for a purely numerical last component, and in > > > > > that case take the last two components. > > > > > > > > That would break the old names. "/dev/tty1" would not match to "tty1". > > > > > > I'm sorry, I meant fall back to the old logic if the last component > > > of the path is not purely numerical. > > > > You mean if all characters after the last '/' are numeric, match on > > the last two path components? > > Yeah, I guess that would work. It seems a little hacky, but then again > > so are the other solutions. > > > In fact I think it's pretty good solution :-) You need ability to > distingush /dev/pts/0 , /dev/vc/0 and [perhaps] /devfs/x/y/z/0 or > /usb/A/B/C/0 while retaining compatibility with old scheme (where > /dev/tty1 is just tty1 ). It works: > > /dev/tty1 => tty1 > /dev/pts/0 => pts/0 > /dev/vc/0 => vc/0 > /devfs/x/y/z/0 => /devfs/x/y/z/0 > /usb/A/B/C/0 => /usb/A/B/C/0 > > All is pretty clear: if your device is in /dev then strip /dev and use > path relative to /dev, if not - use full name. To me it looks better then > any other scheme (it's possible to make a clash if you'll create > /dev/devfs/x/y/z/0 AND you'll use /dev//devfs/x/y/z/0 in inittab; in this > situation answer is simple: DON'T DO IT! And if you are using full > canonical pathnames clash is not possible anymore with ANY structure of > device names). >
Maybe I'm not seeing something here, but wouldn't it be better to just not chomp the path at all? i.e.
/dev/tty1 -> /dev/tty1 /dev/vc/0 -> /dev/vc/0
and perhaps just compare the last characters of the names listed in securetty, for compatibility..
-- Zinx Verituse <zinx@linuxfreak.com> gpg (id 921B1558) (fp 5746 73A1 2184 A27A 9EC0 EDCC E132 BCEF 921B 1558)
- 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/
| |