Messages in this thread | | | From | Andrzej Krzysztofowicz <> | Subject | Re: 2.3.40-pre1: Partition Enumeration | Date | Sat, 15 Jan 2000 10:03:54 +0100 (CET) |
| |
> > > > see, my rootfs switched from /dev/hda6 to /dev/hda12. > > > > This is old-known problem and it seems there is no good solution of it. > > The only I know is (potentially) devfs naming convention. > > > > Changing the sequence of second-level partition initialization doesn't solve > > this problem. (it only moves problem from extended to BSD/Unixware/Solaris > > partitions) > > No, but it is the best possible approximation to a stable numbering. > For most people (who do not have both solaris and unixware, compiling > even kernels with solaris support and odd kernels with unixware support) > the right order suffices to eliminate all problems.
People not using BSD/Unixware/Solaris partitions do not have anu problem at all. But if you use any of {BSD,Unixware,Solaris} together with DOS-extended it does not solve your problem: add/remove any logical drive in the extended partition and all your {BSD,Unixware,Solaris} slices become renumbered...
> The problem is independent of naming - there is a tree-structured space > here, and the kernel has to flatten it, assigning minor numbers. > In user space it is easy to populate /dev/unixware with device nodes > p1, p2, ... that are symlinks to /dev/hda12, /dev/hda13, ...
Same with DOS-extended/logical drives... But the problem is the devices/symlinks need to be created during kernel start by devfs or another animal - it does no matter.
> (or use mount by label or uuid).
This is better solution, but unfortunately not all filesystems have/support label/uuid ...
-- ======================================================================= Andrzej M. Krzysztofowicz ankry@mif.pg.gda.pl tel. (0-58) 347 14 61 Wydz.Fizyki Technicznej i Matematyki Stosowanej Politechniki Gdanskiej
- 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/
| |