Package: util-linux Version: 2.12p-4sarge1 Looks like it is time for another Mad System Administrator bug(s):
The display of varying disk slice types is inconsistant. From what I can
tell, it appears that the unit display of MSDOS partitions reports end
sector-1, rather than end sector. In cylinder display mode start cylinder
reporting is off by one.
Attached are two mostly equivalent sectors. One is using a Sun disk label
the other is an MS-DOS partition table. Both have two actual disk slices
that line up, `fdisk -l` of these report differing numbers:
Disk /dev/hd# (Sun disk label): 16 heads, 63 sectors, 19374 cylinders
Units = cylinders of 1008 * 512 bytes
Device Flag Start End Blocks Id System
/dev/hd#2 u 17436 19374 976752 82 Linux swap
/dev/hd#3 u 0 19374 9764496 5 Whole disk
/dev/hd#8 0 17436 8787744 83 Linux native
Disk /dev/hd# (Sun disk label): 16 heads, 63 sectors, 19374 cylinders
Units = sectors of 1 * 512 bytes
Device Flag Start End Blocks Id System
/dev/hd#2 u 17575488 19528992 976752 82 Linux swap
/dev/hd#3 u 0 19528992 9764496 5 Whole disk
/dev/hd#8 0 17575488 8787744 83 Linux native
Disk /dev/hd#: 10.2 GB, 10245537792 bytes
16 heads, 63 sectors/track, 19852 cylinders
Units = cylinders of 1008 * 512 = 516096 bytes
Device Boot Start End Blocks Id System
/dev/hd#2 17437 19374 976752 82 Linux swap / Solaris
/dev/hd#4 * 1 17436 8787744 83 Linux
Partition table entries are not in disk order
Disk /dev/hd#: 10.2 GB, 10245537792 bytes
16 heads, 63 sectors/track, 19852 cylinders, total 20010816 sectors
Units = sectors of 1 * 512 = 512 bytes
Device Boot Start End Blocks Id System
/dev/hd#2 17575488 19528991 976752 82 Linux swap / Solaris
/dev/hd#4 * 0 17575487 8787744 83 Linux
Partition table entries are not in disk order
I'm sure they line up, because I confirmed that the second was mountable
using either table and the first was usable as swap. I must argue that
the end sector listing from the MSDOS-style is wrong because the end
number is what the Sun-style actually contains, and the block count is
correct in both cases. I'd also argue cynlinders start from zero so the
start cylinder reporting is one too high for MSDOS-style.
Also it was impossible to create the MSDOS-style table inside `fdisk`,
some creativity was required. Though changing between the styles is not a
common event, both were quite serious problems for doing the conversion.
--
(\___(\___(\______ --=> 8-) EHM <=-- ______/)___/)___/)
\BS ( | [EMAIL PROTECTED] PGP 8881EF59 | ) /
\_CS\ | _____ -O #include <stddisclaimer.h> O- _____ | / _/
\___\_|_/82 04 A1 3C C7 B1 37 2A*E3 6E 84 DA 97 4C 40 E6\_|_/___/
sun.raw
Description: Binary data
dos.raw
Description: Binary data
Disk /dev/hd# (Sun disk label): 16 heads, 63 sectors, 19374 cylinders
Units = cylinders of 1008 * 512 bytes
Device Flag Start End Blocks Id System
/boot/boot.hdb2 u 17436 19374 976752 82 Linux swap
/boot/boot.hdb3 u 0 19374 9764496 5 Whole disk
/boot/boot.hdb8 0 17436 8787744 83 Linux native
Disk /dev/hd#: 10.2 GB, 10245537792 bytes 16 heads, 63 sectors/track, 19852 cylinders Units = cylinders of 1008 * 512 = 516096 bytes Device Boot Start End Blocks Id System /dev/hdb2 17437 19374 976752 82 Linux swap / Solaris /dev/hdb4 * 1 17436 8787744 83 Linux Partition table entries are not in disk order
Disk /dev/hd# (Sun disk label): 16 heads, 63 sectors, 19374 cylinders
Units = sectors of 1 * 512 bytes
Device Flag Start End Blocks Id System
/boot/boot.hdb2 u 17575488 19528992 976752 82 Linux swap
/boot/boot.hdb3 u 0 19528992 9764496 5 Whole disk
/boot/boot.hdb8 0 17575488 8787744 83 Linux native
Disk /dev/hd#: 10.2 GB, 10245537792 bytes 16 heads, 63 sectors/track, 19852 cylinders, total 20010816 sectors Units = sectors of 1 * 512 = 512 bytes Device Boot Start End Blocks Id System /dev/hdb2 17575488 19528991 976752 82 Linux swap / Solaris /dev/hdb4 * 0 17575487 8787744 83 Linux Partition table entries are not in disk order
