Re: devfs patch v3

Dave Cinege (dcinege@psychosis.com)
Mon, 12 Jan 98 05:29:15 -0500


On Mon, 12 Jan 1998 01:44:25 -0800 (PST), H. Peter Anvin wrote:

>> >Not if you put it on initrd.
>>
>> As the guy that just added some new initrd features, which makes it much
>> easier to deal with (as does your syslinux ; > ), I'll be the first
>> to say I don't feel this is reasonable. What would you do? Play with images
>> or archives everytime there needs to be a boot change? I personally don't
>> even like having to run LILO with every kernel change.
>>
>> Something like this seems more like a Q&D hack instead of doing it
>> right. MHO
>
>No boot change... it is a bloody setup program... maybe I'm completely
>misunderstand what you want.

Maybe, so let me go over it again. ext2, msdos, and I'm sure other
FS's have a volume label. When the filesystem (ext2) driver
is loaded, it should scan the superblocks of the partitions for this volume
label, and then make them available via dev symlinks/devfs/proc or
something. Then in fstab instead of /dev/sda8 I can spec /dev/usr1 if the
volume label of /dev/sda8 is 'usr1'. Follow me thus far?

Up to this point it can all be done in user space. The problem is mounting
the root. We always have to spec the actual device for the root, if we do
notpull this info out with the kernel. Now you don't need the kernel to
create all the symlinks(whatever) just search for the label that matches the
one speced for the root. Then I can spec /dev/root-drive with the boot
loader, and linux it will always try to mount the drive with the label
'root-drive' as the root.

The reason I thought of this was because after repartitioning one day I
broke linux. The partition itself *never moved*, but it was now /dev/sda11
instead of /dev/sda10, and I was left with an unbootable system that was not
very easy to fix.

>I think having to run an install program or script -- including LILO
>-- on a kernel change is a quite reasonable thing.

Reasonable, yes, but I still don't like it : >
Lately I've been spoiled by being able to just drop new kernels on a
syslinux disk. It would be nice to have something like that for ext2. That
would be great because with that and the above scheme you could throw the
damn root partition anywhere and it would come right up so long as you named
it correctly.

-------------------------------------------------------------------------
http://www.psychosis.com/emc/ Elite MicroComputers 908-541-4214
http://www.psychosis.com/linux-router/ Linux Router Project

"The only way to remain cutting edge, is to design for things that do not yet
exist." -- Me : )