On Wed, Jun 12, 2002 at 10:07:40PM -0400, Joe Landman wrote: | On Wed, 2002-06-12 at 21:13, Tom Allison wrote: | | > devfs: v1.10 (20020120) Richard Gooch | > devfs: boot_options: 0x0 | | No matter what else is a problem, the above is also very much a problem. | | Just say no to devfs.
I've been using devfs for months. I really like it. It makes it much easier to identify problems with "missing" devices. (I can't read my cd, but /dev/cdrom is there. It doesn't matter if the inode is on disk.) Besides, you can have devfs compiled (the pre-packaged kernels do) and not enabled. When I boot I get these relevant lines : devfs: boot_options: 0x1 Mounted devfs on /dev If that second line is missing, then devfs isn't even being used on Tom's system. | It mucks with device things in ways that cause | exactly the sort of problem you have. I did find the root= semantics to be unintuitive. I use the old /dev/hda1 notation for the root=. After boot (after devfsd is started) it doesn't matter which I use. It's just a name, nothing more nothing less. | When you get this "VFS: Unable to mount root fs on ..." error, this | generally means that the kernel was told root (the /) is on /dev/hdaX | when in fact, it is on /dev/hdaY (Y != X). Either that or you're missing a critical driver, or you don't have a root= parameter (happened to me once due to a typo). (or maybe some new condition I haven't seen before) | Note how devfs can contribute to this, as root may really be on | /dev/ide/bus/0/part/1 or whatever sillyness it forces. The name is (largely) irrelevant. The major:minor modes are the same. (however, I found that I was unable to use root=/dev/ide/host0/bus0/target0/lun0/part1, but can (and do) use root=/dev/hda1) HTH, -D -- (http://www.thinkgeek.com/images/products/zoom/no-clue.jpg) > SELECT * FROM users WHERE clue > 0 0 rows returned http://dman.ddts.net/~dman/
pgpVvaUocYTHT.pgp
Description: PGP signature