1 .. SPDX-License-Identifier: GPL-2.0
2
3 ===================
4 The QNX6 Filesystem
5 ===================
6
7 The qnx6fs is used by newer QNX operating syst
8 It got introduced in QNX 6.4.0 and is used def
9
10 Option
11 ======
12
13 mmi_fs Mount filesystem as used for e
14
15 Specification
16 =============
17
18 qnx6fs shares many properties with traditional
19 concepts of blocks, inodes and directories.
20
21 On QNX it is possible to create little endian
22 This feature makes it possible to create and u
23 for the target (QNX is used on quite a range o
24 running on a different endianness.
25
26 The Linux driver handles endianness transparen
27
28 Blocks
29 ------
30
31 The space in the device or file is split up in
32 size of 512, 1024, 2048 or 4096, which is deci
33 created.
34
35 Blockpointers are 32bit, so the maximum space
36 2^32 * 4096 bytes or 16TB
37
38 The superblocks
39 ---------------
40
41 The superblock contains all global information
42 Each qnx6fs got two superblocks, each one havi
43 That serial number is used to identify the "ac
44 In write mode with reach new snapshot (after e
45 serial of the new master superblock is increas
46
47 So basically the snapshot functionality is rea
48 update of the serial number. Before updating t
49 are done by copying all modified blocks during
50 (or period) and building up a new (stable) fil
51 inactive superblock.
52
53 Each superblock holds a set of root inodes for
54 parts. (Inode, Bitmap and Longfilenames)
55 Each of these root nodes holds information lik
56 data and the addressing levels in that specifi
57 If the level value is 0, up to 16 direct block
58 node.
59
60 Level 1 adds an additional indirect addressing
61 addressing block holds up to blocksize / 4 byt
62 Level 2 adds an additional indirect addressing
63 to 16 * 256 * 256 = 1048576 blocks that can be
64
65 Unused block pointers are always set to ~0 - r
66 indirect addressing blocks or inodes.
67
68 Data leaves are always on the lowest level. So
69 tree levels.
70
71 The first Superblock is located at 0x2000. (0x
72 The Audi MMI 3G first superblock directly star
73
74 Second superblock position can either be calcu
75 information (total number of filesystem blocks
76 device address, zeroing the last 3 bytes and t
77 that address.
78
79 0x1000 is the size reserved for each superbloc
80 blocksize of the filesystem.
81
82 Inodes
83 ------
84
85 Each object in the filesystem is represented b
86 The inode structure contains pointers to the f
87 the data held in the object and all of the met
88 its longname. (filenames longer than 27 charac
89 The metadata about an object includes the perm
90 size, number of blocks used, access time, chan
91
92 Object mode field is POSIX format. (which make
93
94 There are also pointers to the first 16 blocks
95 addressed with 16 direct blocks.
96
97 For more than 16 blocks an indirect addressing
98 used. (scheme is the same as the one used for
99
100 The filesize is stored 64bit. Inode counting s
101 filename inodes start with 0)
102
103 Directories
104 -----------
105
106 A directory is a filesystem object and has an
107 It is a specially formatted file containing re
108 name with an inode number.
109
110 '.' inode number points to the directory inode
111
112 '..' inode number points to the parent directo
113
114 Eeach filename record additionally got a filen
115
116 One special case are long filenames or subdire
117
118 These got set a filename length field of 0xff
119 record plus the longfile inode number also sto
120
121 With that longfilename inode number, the longf
122 starting with the superblock longfilename root
123
124 Special files
125 -------------
126
127 Symbolic links are also filesystem objects wit
128 bit in the inode mode field identifying them a
129
130 The directory entry file inode pointer points
131
132 Hard links got an inode, a directory entry, bu
133 no block pointers and the directory file recor
134 inode.
135
136 Character and block special devices do not exi
137 are handled by the QNX kernel/drivers and crea
138 underlying filesystem.
139
140 Long filenames
141 --------------
142
143 Long filenames are stored in a separate addres
144 is the longfilename root node in the active su
145
146 Each data block (tree leaves) holds one long f
147 limited to 510 bytes. The first two starting b
148 for the actual filename.
149
150 If that structure shall fit for all allowed bl
151 is a limit of 510 bytes for the actual filenam
152
153 Bitmap
154 ------
155
156 The qnx6fs filesystem allocation bitmap is sto
157 root node in the superblock and each bit in th
158 filesystem block.
159
160 The first block is block 0, which starts 0x100
161 So for a normal qnx6fs 0x3000 (bootblock + sup
162 address at which block 0 is located.
163
164 Bits at the end of the last bitmap block are s
165 smaller than addressing space in the bitmap.
166
167 Bitmap system area
168 ------------------
169
170 The bitmap itself is divided into three parts.
171
172 First the system area, that is split into two
173
174 Then userspace.
175
176 The requirement for a static, fixed preallocat
177 qnx6fs deals with writes.
178
179 Each superblock got its own half of the system
180 always uses blocks from the lower half while s
181 blocks represented by the upper half bitmap sy
182
183 Bitmap blocks, Inode blocks and indirect addre
184 tree structures are treated as system blocks.
185
186 The rational behind that is that a write reque
187 (system area of the inactive - resp. lower ser
188 at the same time there is still a complete sta
189 other half of the system area.
190
191 When finished with writing (a sync write is co
192 time or a filesystem sync is requested), seria
193 superblock atomically is increased and the fs
194 stable declared - superblock.
195
196 For all data outside the system area, blocks a
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.