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
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
|
.TH PDISK 8 "20 December 1996" "MkLinux DR2" "Linux Programmer's Manual"
.SH NAME
pdisk \- Apple partition table editor for Linux
.SH SYNOPSIS
.B pdisk
.B "[\-h|\--help] [\-v|\--version] [\-l|\--list [name ...]]"
.br
.B pdisk
.B "[\-r|\--readonly]"
device ...
.SH DESCRIPTION
.B pdisk
is a menu driven program which partitions disks using the standard Apple
disk partitioning scheme described in "Inside Macintosh: Devices".
It does not support the intel/dos partitioning scheme supported by
.BR fdisk .
The
.I device
is usually one of the following:
.nf
.RS
/dev/sda
/dev/sdb
/dev/sdc
/dev/sdd
/dev/sde
/dev/sdf
/dev/sdg
/dev/hda
/dev/hdb
.RE
.fi
MkLinux interprets device names differently than standard Linux.
In MkLinux /dev/sda is the device at SCSI id 0, /dev/sdb is the device at SCSI
id 1, and so on.
In standard Linux /dev/sda is the first hard disk on the SCSI bus (i.e. the
one with the lowest id), /dev/sdb is the second hard disk, and so on.
The
.I partition
is a
.I device
name followed by a partition number.
The partition number is the index (starting from one) of the partition
map entry in the partition map.
For example,
.B /dev/sda2
is the partition described by the second entry in the partiton map on /dev/sda.
.SH OPTIONS
.TP
.B \-v | \--version
Prints version number of the
.B pdisk
program.
.TP
.B \-h | \--help
Prints a rather lame set of help messages for the
.B pdisk
program.
.TP
.B \-l | \--list
If no
.IR name s
are present then lists the partition tables for
.BR /dev/sda ,
.BR /dev/sdb ,
.BR /dev/sdc ,
.BR /dev/sdd ,
.BR /dev/sde ,
.BR /dev/sdf ,
and
.BR /dev/sdg .
Otherwise, lists the partition tables for the specified
.IR name s.
.TP
.B \-r | \--readonly
Prevents
.B pdisk
from writing to the device.
.SH "Editing Partition Tables"
An argument which is simply the name of a
.I device
indicates that
.B pdisk
should edit the partition table of that device.
The current top level editing commands are:
.nf
.RS
h command help
p print the partition table
P (print ordered by base address)
i initialize partition map
s change size of partition map
c create new partition
C (create with type also specified)
d delete a partition
r reorder partition entry in map
w write the partition table
q quit without saving changes
.RE
.fi
Commands which take arguments prompt for each argument in turn.
You can also type any number of the arguments separated by spaces
and those prompts will be skipped.
The only exception to typeahead are the confirmation prompts on the
.B i
and
.B w
commands.
The idea being that if we expect you to confirm the decision we
shouldn't undermine that by allowing you to be precipitate about it.
Partitions are always specified by their number,
which the index of the partition entry in the partition map.
Most of the commands will change the index numbers of all partitions
after the affected partition.
You are advised to print the table as frequently as necessary.
Creating more than fifteen partitions is not advised.
There is currently a bug in the some (all?) of the kernels which causes
access to the whole disk fail if more than fifteen partitions are in the map.
The
.B c
(create new partition) command is the only one with complicated arguments.
The first argument is the base address (in blocks) of the partition.
Besides a raw number, you can also specify a partition number followed
by the letter 'p' to indicate that the first block of the new partition should
be the same as the first block of that existing free space partition.
The second argument is the length of the partition in blocks.
This can be a raw number or can be a partition number followed by the
letter 'p' to use the size of that partition or can be a number followed
by 'k', 'm', or 'g' to indicate the size in kilobytes, megabytes, or gigabytes
respectively.
(These are powers of 1024, of course, not powers of 1000.)
The last argument is the name of the partition.
This can be a single word without quotes, or a string surrounded by
single or double quotes.
The
.B C
command is identical to the
.B c
command, with the addition of a partition type argument after the
other arguments.
The
.B r
(reorder) command allows the index number of partitions to be changed.
The index numbers are constrained to be a contiguous sequence.
The
.B i
(initalize) command prompts for the size of the device.
This was done to get around a bug in the kernel where it reports the wrong
size for the device.
The
.B w
(write) command does write the partition map out,
but there is currently a bug in the interaction between MkLinux and Mach
which causes the partition map not to be reinterpreted.
In order to use the new partition map you must reboot.
.SH BUGS
Some people believe there should really be just one disk partitioning utility.
.br
.B pdisk
should be able to create HFS partitions that work.
.br
Even more help should be available during user input.
.SH "SEE ALSO"
.BR fdisk (8),
.BR mkswap (8),
.BR mkfs (8)
.SH AUTHOR
Eryk Vershen (eryk@apple.com)
|