lkml.org 
[lkml]   [2000]   [Jan]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 2.3.40-pre1: Partition Enumeration
Hi guys,

Just to clarify - when I mailed this yesterday I understood clearly that
to fix this issue is absolutely impossible - but I expect no less than
ability to do the impossible from linux-kernel people (otherwise I would
just fix it myself and send a patch).

So, is there a way to do the impossible without something ugly like:

a) declare Intel Architecture as preferred (well, Intel manuals kind of
say it is so why not), i.e. admit arch-dependent code in
fs/partitions/

b) declare msdos partition table as Standard (Microsoft thinks it is
so why not :)

The problem with Linux is that it treats all kinds of partitions as equal,
i.e. it does the right thing in terms of OS design. But that doesn't help
the user who has to modify /etc/rc.d/rc.sysinit to generate /etc/fstab on
the fly (using fdisk -l, /proc/partitions, rdev etc) to be able to boot
kernels with different partition support without fsck complaining and
mounts failing. And the fact that / is mounted readonly at that stage
doesn't help either (but can be worked around using ramdisks)

Regards,
Tigran.

On Thu, 13 Jan 2000, Tigran Aivazian wrote:

> Hi,
>
> I am not sure the current algorithm for enumerating 3rd party
> partitions is perfect. It makes it very hard to switch between
> booting kernels with and without support for those partitions. E.g. I have
> 2.2.12 from Red Hat 6.1 that has no support for UnixWare and FreeBSD
> partitions and now use 2.3.40-pre1 with support for those two and all my
> normal (i.e. Linux ext2) partitions get shifted around. That;s not good.
>
> To make it clear, have a look at these:
>
> # fdisk -l /dev/hda
>
> Disk /dev/hda: 255 heads, 63 sectors, 3467 cylinders
> Units = cylinders of 16065 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hda1 1 9 72261 83 Linux
> /dev/hda2 10 264 2048287+ a5 BSD/386
> /dev/hda3 * 265 902 5124735 63 GNU HURD or SysV
> /dev/hda4 903 3467 20603362+ 5 Extended
> /dev/hda5 903 968 530113+ 82 Linux swap
> /dev/hda6 969 1733 6144831 83 Linux
> /dev/hda7 1734 2116 3076416 6 FAT16
> /dev/hda8 2117 3467 10851876 83 Linux
> # grep hda /proc/partitions
> 3 0 27852048 hda
> 3 1 72261 hda1
> 3 2 2048287 hda2
> 3 3 5124735 hda3
> 3 4 1 hda4
> 3 5 4795402 hda5
> 3 6 208845 hda6
> 3 7 17 hda7
> 3 8 8015 hda8
> 3 9 24097 hda9
> 3 10 8032 hda10
> 3 11 530113 hda11
> 3 12 6144831 hda12
> 3 13 3076416 hda13
> 3 14 10851876 hda14
> 3 15 1843200 hda15
> 3 16 205087 hda16
>
> see, my rootfs switched from /dev/hda6 to /dev/hda12.
>
>
> Regards,
> Tigran.
>
>


-
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/

\
 
 \ /
  Last update: 2005-03-22 13:55    [from the cache]
©2003-2011 Jasper Spaans