docs/Partition-ends-after-end-of-disk.txt


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

        Cylinder/Head/Size Hell Is Not Over 

            pixel@mandrakesoft.com 
         $Date$


Nowadays, you can boot Linux without ever relying on CHS
(Cylinder/Head/Size):

- CHS information on partition table is obsolete and not used anymore.
It was causing the 1024th cylinder limitation:
  - disallowing partitions above 8GiG. Problem overcome using 28bits Linear
  Block Addressing (LBA) in partition table.
  - disallowing to boot on partition above 8GiG (due to int13 limitation used
  in bootloaders). Problem overcome using int13 extensions (EDD)

Logical CHS (Cylinder/Head/Size) and physical CHS are not related anymore.
Especially the size of a hard drive is not a multiple of the number of
Cylinders.


The problem is that many tools still rely somewhat heavily on CHS information.
- most tools adjust partitions on cylinder boundary
- some tools compute the size of the disk based on CHS (eg: fdisk)


Resulting problems in various tools:

- last partition not cylinder boundary adjusted, and overrun CHS computed size
  - fdisk
    > Total allocated sectors %d greater than the maximum %d
  - cfdisk (need confirmation)
    > FATAL ERROR: Bad primary partition %d: Partition ends after end-of-disk
    or
    > FATAL ERROR: Bad logical partition %d: Partition ends after end-of-disk
  - Partition Magic (version?)
    > Error #109: Partition ends after end of disk
  - Windows <= 98se, Windows Me
    (doesn't like it, what error message?)

  Tools handling it correctly:
  - parted and so diskdruid in RedHat >= 7.2 (or is it 7.x?)
  - diskdrake >= Mandrake 8.2
  - Windows XP (need confirmation)

  Tools using the full size of the disk, and so causing pbs with fdisk/cfdisk/...:
  - diskdrake == Mandrake 8.2

- disagreement between CHS and linear address
  - parted
    > "Warning: Unable to align partition properly.  This probably means that
    > another partitioning tool generated an
    > incorrect partition table, because it didn't have the correct BIOS
    > geometry.  It is safe to ignore,but
    > ignoring may cause (fixable) problems with some boot loaders.
    > Ignore Cancel ? C
  - lilo
    > Warning: Kernel & BIOS return differing head/sector geometries for device 0x80
    >     Kernel: 38792 cylinders, 16 heads, 63 sectors
    >       BIOS: 1023 cylinders, 255 heads, 63 sectors
    > Device 0x2100: Inconsistent partition table, 2nd entry
    >   CHS address in PT:  48:0:1  -->  LBA (771120)
    >   LBA address in PT:  48384  -->  CHS (3:3:1)
    > The partition table is *NOT* being adjusted.