1 .. SPDX-License-Identifier: GPL-2.0
2
3 ================================
4 Optimized MPEG Filesystem (OMFS)
5 ================================
6
7 Overview
8 ========
9
10 OMFS is a filesystem created by SonicBlue for
11 and Rio Karma MP3 player. The filesystem is e
12 block sizes from 2k to 8k, with hash-based dir
13 filesystem driver may be used to read and writ
14 devices.
15
16 Note, it is not recommended that this FS be us
17 filesystem for your own streaming media device
18 will likely perform better.
19
20 More information is available at:
21
22 http://linux-karma.sf.net/
23
24 Various utilities, including mkomfs and omfsck
25 omfsprogs, available at:
26
27 https://bobcopeland.com/karma/
28
29 Instructions are included in its README.
30
31 Options
32 =======
33
34 OMFS supports the following mount-time options
35
36 ============ ===========================
37 uid=n make all files owned by spe
38 gid=n make all files owned by spe
39 umask=xxx set permission umask to xxx
40 fmask=xxx set umask to xxx for files
41 dmask=xxx set umask to xxx for direct
42 ============ ===========================
43
44 Disk format
45 ===========
46
47 OMFS discriminates between "sysblocks" and nor
48 group consists of super block information, fil
49 and extents. Each sysblock has a header conta
50 sysblock, and may be mirrored in successive bl
51 have a smaller size than a data block, but sin
52 same 64-bit block number, any remaining space
53 unused.
54
55 Sysblock header information::
56
57 struct omfs_header {
58 __be64 h_self; /*
59 __be32 h_body_size; /*
60 __be16 h_crc; /*
61 char h_fill1[2];
62 u8 h_version; /*
63 char h_type; /*
64 u8 h_magic; /*
65 u8 h_check_xor; /*
66 __be32 h_fill2;
67 };
68
69 Files and directories are both represented by
70
71 struct omfs_inode {
72 struct omfs_header i_head; /*
73 __be64 i_parent; /*
74 __be64 i_sibling; /*
75 __be64 i_ctime; /*
76 char i_fill1[35];
77 char i_type; /*
78 __be32 i_fill2;
79 char i_fill3[64];
80 char i_name[OMFS_NAMELEN]; /*
81 __be64 i_size; /*
82 };
83
84 Directories in OMFS are implemented as a large
85 hashed then prepended into the bucket list beg
86 Lookup requires hashing the filename, then see
87 until a match is found on i_name. Empty bucke
88 pointers with all-1s (~0).
89
90 A file is an omfs_inode structure followed by
91 OMFS_EXTENT_START::
92
93 struct omfs_extent_entry {
94 __be64 e_cluster; /*
95 __be64 e_blocks; /*
96 };
97
98 struct omfs_extent {
99 __be64 e_next; /*
100 __be32 e_extent_count; /*
101 __be32 e_fill;
102 struct omfs_extent_entry e_entry;
103 };
104
105 Each extent holds the block offset followed by
106 the extent. The final extent in each table is
107 being ~0 and e_blocks being ones'-complement o
108 in the table.
109
110 If this table overflows, a continuation inode
111 e_next. These have a header but lack the rest
112
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.