csj <[EMAIL PROTECTED]> writes: > But, let's say I just want to play with devfs, could I still go back to > my old disk-based setup? Could I still boot my old devfsd-disabled > kernel after using a devfsd-enabled kernel?
Sure. Having devfs enabled in the kernel does not require you to use devfs. Whatever files you currently have in /dev will not get touched by devfs. However, if devfs is enabled in the kernel, then you can mount -t devfs - /dev Or even mount -t devfs - /devfs Then, you can look in that directory and see a bunch of magic entries that devfs has created (in much the same way that /proc has a bunch of magic entries). You won't see your original /dev files, because you've mounted a new filesystem over top of that directory, so whatever was there before is hidden. If you then decide you don't like devfs, just unmount it. Whatever device nodes happened to be in /dev before you mounted devfs over top of /dev will still be there after you unmount. Similarly, if you boot a kernel without devfs support, obviously you won't be able to mount devfs on top of /dev, so whatever was in /dev before will be there. Mounting on /devfs doesn't buy you much, because most software looks at /dev for its devices. But it's useful if you want to see what devfs will create for you. There's nothing magic about the fact that device nodes are in /dev. A device is just a special file in the filesystem. There's also nothing magic about the filenames in /dev, which is why devfs can refer to /dev/hda1 as /dev/discs/disc0/part1. -- Dave Carrigan ([EMAIL PROTECTED]) | Yow! I'm EMOTIONAL now because I UNIX-Apache-Perl-Linux-Firewalls-LDAP-C-DNS | have MERCHANDISING CLOUT!! Seattle, WA, USA | http://www.rudedog.org/ |
