Gentoo Archives: gentoo-user

From: Mick <michaelkintzios@×××××.com>
To: gentoo-user@l.g.o
Subject: Re: [gentoo-user] Boot partitions (WAS: migrating disks (from mounts to disklabels
Date: Tue, 23 Nov 2010 20:09:23
Message-Id: 201011232008.18778.michaelkintzios@gmail.com
In Reply to: Re: [gentoo-user] Boot partitions (WAS: migrating disks (from mounts to disklabels by David W Noon
1 On Tuesday 23 November 2010 17:18:56 David W Noon wrote:
2 > On Tue, 23 Nov 2010 17:20:02 +0100, Peter Humphrey wrote about Re:
3 > [gentoo-user] Boot partitions (WAS: migrating disks (from mounts to
4 >
5 > disklabels:
6 > >On Sunday 21 November 2010 16:22:15 David W Noon wrote:
7 > >> What I suspect is in the remainder of that space is a hidden primary
8 > >> partition containing a "transparent" bootstrap that augments the BIOS
9 > >> and permits booting from a logical/extended partition. This would be
10 > >> similar to the old OS/2 Boot Manager, although that was hardly
11 > >> transparent. This hidden partition was probably placed there by
12 > >> cfdisk when you first partitioned the drive and started it with an
13 > >> extended partition. The OS/2 FDISK.COM did something similar when
14 > >> the first partition on a drive was not a primary (including Boot
15 > >> Manager).
16 > >>
17 > >> A forensic examination of that area would be of interest.
18 > >
19 > >Including attempting creation of three more primary partitions. If
20 > >your hunch is right, the last will not be created as the allowable
21 > >four exist already.
22 >
23 > No, only 1 primary and Neil's extended partition would currently
24 > exist. So it should be possible to create 2 more primaries, if there
25 > were space on the disk.
26 >
27 > >I don't believe your hunch is right though - it's just too complex to
28 > >be worthwhile for any but a very few customers.
29 >
30 > Not at all complex. OS/2 has had the option to work that way since
31 > about 1987 on 80286 hardware (and written by Microsoft).
32 >
33 > >Nobody's answered my parenthetical question though: why do SCSI and
34 > >IDE interfaces allow different total numbers of partitions?
35 >
36 > They don't.
37 >
38 > The number of primary/extended partitions is limited to 4, and the
39 > number of logical drives is limited by the size of the extended
40 > partition. If you can fit more partitions onto one drive than another,
41 > it is because either the drive is bigger or the partitions are smaller.
42
43 Errm, not exactly. SCSI/SATAs are limited to 15 (inc. one extended partition)
44 and old (legacy driven) IDEs are limited to some 63 partitions if I recall
45 correctly. If you use the new libata I think you only get 15 partitions for
46 SATA/PATA.
47
48 I know, because I remember some years ago getting a bit over-enthusiastic with
49 a new SATA drive and fdisk only to end up with partitions that I couldn't
50 mount ... O_o
51 --
52 Regards,
53 Mick

Attachments

File name MIME type
signature.asc application/pgp-signature