PERFORMANCE.README
1GENERAL INFORMATION ON PERFORMANCE TESTS
2----------------------------------------
3
4The following performance tests were based on two file sets: the
5liveCD filesystem from the Ubuntu liveCD (Warty release), and the
6liveCD filesystem from the Damn Small Linux liveCD (release 0.8.4).
7The Ubuntu liveCD filesystem was used to test filesystem performance
8from CDROM and hard disk for Zisofs, Cloop, Squashfs 2.0 and Squashfs2.1.
9CRAMFS filesystem performance could not be tested for this filesystem
10bacause it exceeds the maximum supported size of CRAMFS. To test
11CRAMFS performance against Squashfs, the liveCD filesystem from
12Damn Small Linux was used.
13
14NOTE: the usual warnings apply to these results, they are provided for
15illustrative purposes only, and due to different hardware and/or file data, you
16may obtain different results. As such the results are provided "as is" without
17any warranty (either express or implied) and you assume all risks as to their
18quality and accuracy.
19
201. Ubuntu liveCD performance tests
21
22 ext3 uncompressed size 1.4 GB
23 Zisofs compressed size 589.81 MB
24 Cloop compressed size 471.89 MB
25 Squashfs2.0 compressed size 448.58 MB
26 Squashfs2.1 compressed size 448.58 MB
27
281.1 Performance tests from CDROM
29
301.1.1 Directory Lookup performance
31
32 Time taken to perform "ls -lR --color=alawys | cat > /dev/null" on filesystem
33 mounted from CDROM
34
35 Zisofs 49.88 seconds (User 2.60 secs, Sys 11.19 secs)
36 Cloop 20.80 seconds (User 2.71 secs, Sys 13.50 secs)
37 Squashfs2.0 16.56 seconds (User 2.42 secs, Sys 10.37 secs)
38 Squashfs2.1 10.14 seconds (User 2.48 secs, Sys 4.44 secs)
39
401.1.2 Sequential I/O performance
41
42 Time taken to perform "tar cf - | cat > /dev/null" on filesystem mounted
43 from CDROM
44
45 Zisofs 27 minutes 28.54 seconds (User 3.00 secs, Sys 1 min 4.80 secs)
46 Cloop 5 minutes 55.72 seconds (User 2.90 secs, Sys 3 min 37.90 secs)
47 Squashfs2.0 5 minutes 20.87 seconds (User 2.33 secs, Sys 56.98 secs)
48 Squashfs2.1 5 minutes 15.46 seconds (user 2.28 secs, Sys 51.12 secs)
49
501.1.3 Random I/O performance
51
52 Random access pattern generated by "find /mnt -type f -printf "%s %p\n" | sort
53 -g | awk '{ printf $2 }' > /tmp/sort
54
55 Time taken to perform "cpio -o --quiet -H newc < /tmp/sort > /dev/null"
56 on filesystem mounted from CDROM
57
58 Zisofs 101 minutes 29.65 seconds (User 5.33 secs, Sys 1 min 17.20 secs)
59 Cloop 35 minutes 27.51 seconds (user 5.93 secs, Sys 4 mins 30.23 secs)
60 Squashfs2.0 21 minutes 53.05 seconds (user 5.71 secs, Sys 2 mins 36.59 secs)
61 Squashfs2.1 21 minutes 46.99 seconds (User 5.80 secs, Sys 2 mins 31.88 secs)
62
63
641.2 Performance tests from Hard disk
65
661.2.1 Directory Lookup performance
67
68 Time taken to perform "ls -lR --color=alawys | cat > /dev/null" on filesystem
69 mounted from Hard disk
70
71 Zisofs 17.29 seconds (User 2.62 secs, Sys 11.08 secs)
72 Cloop 16.46 seconds (User 2.63 secs, Sys 13.41 secs)
73 Squashfs2.0 13.75 seconds (User 2.44 secs, Sys 11.00 secs)
74 Squashfs2.1 6.94 seconds (User 2.44 secs, Sys 4.48 secs)
75
761.2.2 Sequential I/O performance
77
78 Time taken to perform "tar cf - | cat > /dev/null" on filesystem mounted
79 from Hard disk
80
81 Zisofs 1 minute 21.47 seconds (User 2.73 secs, Sys 54.44 secs)
82 Cloop 1 minute 34.06 seconds (user 2.85 secs, Sys 1 min 12.13 secs)
83 Squashfs2.0 1 minute 21.22 seconds (User 2.42 secs, Sys 56.21 secs)
84 Squashfs2.1 1 minute 15.46 seconds (User 2.36 secs, Sys 49.78 secs)
85
861.2.3 Random I/O performance
87
88 Random access pattern generated by "find /mnt -type f -printf "%s %p\n" | sort
89 -g | awk '{ printf $2 }' > /tmp/sort
90
91 Time taken to perform "cpio -o --quiet -H newc < /tmp/sort > /dev/null"
92 on filesystem mounted from Hard disk
93
94 Zisofs 11 minutes 13.64 seconds (User 5.08 secs, Sys 52.62 secs)
95 Cloop 5 minutes 37.93 seconds (user 6 secs, Sys 2 mins 22.38 secs)
96 Squashfs2.0 5 minutes 7.11 seconds (user 5.63 secs, Sys 2 mins 35.23 secs)
97 Squashfs2.1 5 minutes 1.87 seconds (User 5.71 secs, Sys 2 mins 29.98 secs)
98
99
1002. Damn Small Linux liveCD performance tests
101
102 ext3 uncompressed size 126 MB
103 CRAMFS compressed size 52.19 MB
104 Squashfs2.0 compressed size 46.52 MB
105 Squashfs2.1 compressed size 46.52 MB
106
1072.1 Performance tests from CDROM
108
1092.1.1 Directory Lookup performance
110
111 Time taken to perform "ls -lR --color=alawys | cat > /dev/null" on filesystem
112 mounted from CDROM
113
114 CRAMFS 10.85 seconds (User 0.39 secs, Sys 0.98 secs)
115 Squashfs2.0 2.97 seconds (User 0.36 secs, Sys 2.15 secs)
116 Squashfs2.1 2.43 seconds (User 0.40 secs, Sys 1.42 secs)
117
1182.1.2 Sequential I/O performance
119
120 Time taken to perform "tar cf - | cat > /dev/null" on filesystem mounted
121 from CDROM
122
123 CRAMFS 55.38 seconds (User 0.34 secs, Sys 6.98 secs)
124 Squashfs2.0 35.99 seconds (User 0.30 secs, Sys 6.35 secs)
125 Squashfs2.1 33.83 seconds (User 0.26 secs, Sys 5.56 secs)
126
1272.1.3 Random I/O performance
128
129 Random access pattern generated by "find /mnt -type f -printf "%s %p\n" | sort
130 -g | awk '{ printf $2 }' > /tmp/sort
131
132 Time taken to perform "cpio -o --quiet -H newc < /tmp/sort > /dev/null"
133 on filesystem mounted from CDROM
134
135
136 CRAMFS 3 minutes 1.68 seconds (User 0.54 secs, Sys 9.51 secs)
137 Squashfs2.0 1 minute 39.45 seconds (User 0.57 secs, Sys 13.14 secs)
138 Squashfs2.1 1 minute 38.41 seconds (User 0.58 secs, Sys 13.08 secs)
139
1402.2 Performance tests from Hard disk
141
1422.2.1 Directory Lookup performance
143
144 Time taken to perform "ls -lR --color=alawys | cat > /dev/null" on filesystem
145 mounted from Hard disk
146
147 CRAMFS 1.77 seconds (User 0.53 secs, Sys 1.21 secs)
148 Squashfs2.0 2.67 seconds (User 0.41 secs, Sys 2.25 secs)
149 Squashfs2.1 1.87 seconds (User 0.41 secs, Sys 1.46 secs)
150
1512.2.2 Sequential I/O performance
152
153 Time taken to perform "tar cf - | cat > /dev/null" on filesystem mounted
154 from Hard disk
155
156 CRAMFS 6.80 seconds (User 0.36 secs, Sys 6.02 secs)
157 Squashfs2.0 7.23 seconds (User 0.29 secs, Sys 6.62 secs)
158 Squashfs2.1 6.53 seconds (User 0.31 secs, Sys 5.82 secs)
159
1602.2.3 Random I/O performance
161
162 Random access pattern generated by "find /mnt -type f -printf "%s %p\n" | sort
163 -g | awk '{ printf $2 }' > /tmp/sort
164
165 Time taken to perform "cpio -o --quiet -H newc < /tmp/sort > /dev/null"
166 on filesystem mounted from Hard disk
167
168
169 CRAMFS 28.55 seconds (User 0.49 secs, Sys 6.49 secs)
170 Squashfs2.0 25.44 seconds (User 0.58 secs, Sys 13.17 secs)
171 Squashfs2.1 24.72 seconds (User 0.56 secs, Sys 13.15 secs)
172
README-2.0
1NOTE: This the original README for version 2.0. It is retained as it
2contains information about the fragment design. A description of the new 2.0
3mksquashfs options has been added to the main README file, and that
4file should now be consulted for these.
5
6 SQUASHFS 2.0 - A squashed read-only filesystem for Linux
7
8 Copyright 2004 Phillip Lougher (plougher@users.sourceforge.net)
9
10 Released under the GPL licence (version 2 or later).
11
12Welcome to the final release of Squashfs version 2.0! A lot of changes to the
13filesystem have been made under the bonnet (hood). Squashfs 2.0 uses fragment
14blocks and larger blocks (64K) to improve compression ratio by about 5 - 20%
15over Squashfs 1.0 depending on the files being compressed. Using fragment
16blocks allows Squashfs 2.0 to achieve better compression than cloop and similar
17compression to tgz files while retaining the I/O efficiency of a compressed
18filesystem.
19
20Detailed changes:
21
221. Squashfs 2.0 has added the concept of fragment blocks (see later discussion).
23 Files smaller than the file block size (64K in Squashfs 2.0) and optionally
24 the remainder of files that do not fit fully into a block (i.e. the last 32K
25 in a 96K file) are packed into shared fragments and compressed together.
26 This achieves on average 5 - 20% better compression than Squashfs 1.x.
27
282. The maximum block size has been increased to 64K.
29
303. The maximum number of UIDs has been increased to 256 (from 48 in 1.x).
31
324. The maximum number of GIDs has been increased to 256 (from 15 in 1.x).
33
345. New mksquashfs -all-root, -root-owned, -force-uid, and -force-gid
35 options. These allow the uids/gids of files in the generated
36 filesystem to be specified, overriding the uids/gids in the
37 source filesystem.
38
396. Initrds are now supported for kernels 2.6.x.
40
417. Removal of sleep_on() function call in 2.6.x patch, to allow Squashfs
42 to work on the Fedora rc2 kernel.
43
448. AMD64, check-data and gid bug fixes.
45
469. Numerous small bug fixes have been made.
47
4810. New patch for Linux 2.6.7.
49
50
51New Squashfs 2.0 options
52------------------------
53
54-noF or -noFragmentCompression
55
56 Do not compress the fragments. Added for compatibility with noI and
57 noD, probably not that useful.
58
59-no-fragments
60
61 Do not use fragment blocks, and rather generate a filesystem
62 similar to a Squashfs 1.x filesystem. It will of course still
63 be a Squashfs 2.0 filesystem but without fragments, and so
64 it won't be mountable on a Squashfs 1.x system.
65
66-always-use-fragments
67
68 By default only small files less than the block size are packed into
69 fragment blocks. The ends of files which do not fit fully into a block,
70 are NOT by default packed into fragments. To illustrate this, a
71 100K file has an initial 64K block and a 36K remainder. This
72 36K remainder is not packed into a fragment by default. This is
73 because to do so leads to a 10 - 20% drop in sequential I/O
74 performance, as a disk head seek is needed to seek to the initial
75 file data and another disk seek is need to seek to the fragment
76 block.
77
78 Specify this option if you want file remainders to be packed into
79 fragment blocks. Doing so may increase the compression obtained
80 BUT at the expense of I/O speed.
81
82-no-duplicates
83
84 Do not detect duplicate files.
85
86-all-root
87-root-owned
88
89 These options (both do exactly the same thing), force all file
90 uids/gids in the generated Squashfs filesystem to be root.
91 This allows root owned filesystems to be built without root access
92 on the host machine.
93
94-force-uid uid
95
96 This option forces all files in the generated Squashfs filesystem to
97 be owned by the specified uid. The uid can be specified either by
98 name (i.e. "root") or by number.
99
100-force-gid gid
101
102 This option forces all files in the generated Squashfs filesystem to
103 be group owned by the specified gid. The gid can be specified either by
104 name (i.e. "root") or by number.
105
106
107Compression improvements example
108--------------------------------
109
110The following is the compression results obtained compressing the 2.6.6
111linux kernel source using CRAMFS, Cloop (with iso filesystem), Squashfs 1.3 and
112Squashfs 2.0 (results generated using big-endian filesystems).
113
114In decreasing order of size:
115
116 CRAMFS 62791680 bytes (59.9M)
117 Squashfs 1.x 51351552 bytes (48.9M)
118 Cloop 46118681 bytes (44.0M)
119 Squashfs 2.0 45604854 bytes (43.5M)
120
121
122The Squashfs 1.x filesystem is 12.6% larger than the new 2.0 filesystem.
123The cloop filesystem is 1.1% larger than the Squashfs 2.0 filesystem.
124
125
126Fragment blocks in Squashfs 2.0
127-------------------------------
128
129Squashfs like all other compressed filesystems compresses files individually
130on a block by block basis. This is performed to allow mounting and
131de-compression of files on a block by block basis without requiring the entire
132filesystem to be decompressed. This is in contrast to data-based compression
133schemes which compress without understanding the underlying filesystem (i.e.
134cloop and tgz files) and which, therefore, do not compress files individually.
135Each approach has advantages and disadvantages, data-based systems have better
136compression because compression is always performed at the maximum block size
137(64K in cloop) irrespective of the size of each file (which could be less than
138the block size). Compressed filesystems tend to be faster at I/O because
139they understand the filesystem and therefore employ better caching stategies
140and read less un-needed data from the filesystem.
141
142Fragment blocks in Squashfs 2.0 solves this problem by packing files (and
143optionally the ends of files) which are smaller than the block size into
144shared blocks, which are compressed together. For example five files each of
14510K will be packed into one shared fragment of 50K and compressed together,
146rather than being compressed in five 10K blocks.
147
148This scheme produces a hybrid filesystem, retaining the I/O efficiency
149of a compressed filesystem, while obtaining the compression efficiency
150of data-based schemes by compressing small files together.
151
152
153Squashfs 1.x and Squashfs 2.0 compatibility
154-------------------------------------------
155
156Appending to Squashfs 1.x filesystems is not supported. If you wish to append
157to 1.x filesystems, then either use the original mksquashfs, or convert them
158to Squashfs 2.0 by mounting the filesystem and running the 2.0 mksquashfs
159on the mounted filesystem.
160
161Mounting Squashfs 1.x filesystems IS supported by the 2.0 kernel patch.
162
README-2.0-AMD64
1Information for amd64 users
2---------------------------
3
4All previous releases of Squashfs (2.0-alpha and older) generate incorrect
5filesystems on amd64 machines. These filesystems work correctly on amd64
6machines, but cannot be mounted on non-amd64 machines. Likewise, filesystems
7generated on non amd64 machines could not be mounted on amd64 machines.
8This bug was caused by the different size of the "time_t" definition used in
9SquashFS filesystem structures.
10
11This bug is now fixed in this release. However, all amd64 filesystems
12generated by previous releases will not be mountable on amd64 machines
13with this release. If you have pre-existing amd64 generated filesystems,
14it is important that you recreate the filesystem. This can be performed
15by mounting the filesystem using a kernel with the original patch
16(i.e. a 2.0-alpha or older patch) and running the SquashFS 2.0
17(i.e. this release) mksquashfs tool to create a new SquashFS filesystem.
18
19
README-2.1
1 SQUASHFS 2.1 - A squashed read-only filesystem for Linux
2
3 Copyright 2004 Phillip Lougher (plougher@users.sourceforge.net)
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs version 2.1-r2. Squashfs 2.1 introduces indexed
8directories which considerably speed up directory lookup (ls, find etc.) for
9directories which are greater than 8K in size. All directories are now also
10sorted alphabetically which further speeds up directory lookup. Many smaller
11improvements have also been made to this release, please see the CHANGES file
12entry for detailed changes.
13
141. DIRECTORY SPEED IMPROVEMENT EXAMPLES
15---------------------------------------
16
17To give an indication of the directory speed improvements a number of test
18results are shown here. There is in addition a new PERFORMANCE.README file
19which gives details of I/O and lookup performance for Squashfs 2.1 against
20the Zisofs, Cloop and CRAMFS filesystems.
21
22example 1:
23
24Filesystems generated from a single directory of 72,784 files (2.6 MB
25directory size). Each file is 10 bytes in size (the test is directory
26lookup and so the file size isn't an issue). The ext3 uncompressed
27directory size is 288 MB (presumably because of one file per block).
28
29Zisofs compressed size 153.50 MB
30Cloop (isofs) compressed size 1.74 MB
31Squashfs2.1 compressed size 612 KB (0.60 MB)
32
33Time taken to perform "ls -lR --color=always | cat > /dev/null" on
34filesystems mounted on hard disk.
35
36Zisofs 35 minutes 7.895 seconds (User 7.868 secs, Sys 34 mins 5.621 secs)
37Cloop 35 minutes 12.765 seconds (User 7.771 secs, Sys 34 mins 3.869 secs)
38Squashfs2.1 19 seconds (User 5.119 secs, Sys 14.547 secs)
39
40example 2:
41
42Filesystems were generated from the Ubuntu Warty livecd (original uncompressed
43size on ext3 is 1.4 GB).
44
45Zisofs compressed size 589.81 MB
46Cloop (isofs) compressed size 471.19 MB
47Squashfs2.0 compressed size 448.58 MB
48Squashfs2.1 compressed size 448.58 MB
49
50Time taken to perform "ls -lR --color=always | cat > /dev/null" on
51filesystems mounted on hard disk.
52
53Zisofs 49.875 seconds (User time 2.589 secs, Sys 11.194 secs)
54Cloop 20.797 seconds (User time 2.706 secs, Sys 13.496 secs)
55Squashfs2.0 16.556 seconds (User time 2.424 secs, Sys 10.371 secs)
56Squashfs2.1 10.143 seconds (User time 2.475 secs, Sys 4.440 secs)
57
58
59NOTE: the usual warnings apply to these results, they are provided for
60illustrative purposes only, and due to different hardware and/or file data, you
61may obtain different results. As such the results are provided "as is" without
62any warranty (either express or implied) and you assume all risks as to their
63quality and accuracy.
64
652. NEW MKSQUASHFS OPTIONS
66-------------------------
67
68There is only one extra option "-2.0". This tells mksquashfs to generate
69a filesystem which is mountable with Squashfs version 2.0.
70
713. APPENDING AND MOUNTING SQUASHFS 2.0 FILESYSTEMS
72--------------------------------------------------
73
74Mounting 2.0 filesystems is supported by Squashfs 2.1. In addition
75mksquashfs v2.1 can append to 2.0 filesystems, although the generated
76filesystem will still be a 2.0 filesystem.
77
784. DONATIONS
79------------
80
81If you find Squashfs useful then please consider making a donation,
82particularly if you use Squashfs in a commercial product. Please consider
83giving something back especially if you're making money from it.
84
85Off the Squashfs subject somewhat I'm currently looking for another
86job doing Linux kernel or filesystems work. If you know of any such
87work that can be performed from the UK then please get in touch. Thanks.
88
README-3.0
1 SQUASHFS 3.0 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2006 Phillip Lougher <phillip@lougher.org.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to the first release of Squashfs version 3.0. Squashfs 3.0 has the
8the following improvements to 2.x.
9
10 1. Filesystems are no longer limited to 4 GB. In
11 theory 2^64 or 4 exabytes is now supported.
12
13 2. Files are no longer limited to 4 GB. In theory the maximum
14 file size is 4 exabytes.
15
16 3. Metadata (inode table and directory tables) are no longer
17 restricted to 16 Mbytes.
18
19 4. Hardlinks are now suppported.
20
21 5. Nlink counts are now supported.
22
23 6. Readdir now returns '.' and '..' entries.
24
25 7. Special support for files larger than 256 MB has been added to
26 the Squashfs kernel code for faster read access.
27
28 8. Inode numbers are now stored within the inode rather than being
29 computed from inode location on disk (this is not so much an
30 improvement, but a change forced by the previously listed
31 improvements).
32
33There is a new Unsquashfs utility (in squashfs-tools) than can be used to
34decompress a filesystem without mounting it.
35
36Squashfs 3.0 supports 2.x filesystems. Support for 1.x filesystems
37will be added in the future.
38
391. UNSQUASHFS
40-------------
41
42Unsquashfs has the following options:
43
44SYNTAX: unsquashfs [-ls | -dest] filesystem
45 -version print version, licence and copyright information
46 -info print files as they are unsquashed
47 -ls list filesystem only
48 -dest <pathname> unsquash to <pathname>, default "squashfs-root"
49
50The "-ls" option can be used to list the contents of a filesystem without
51decompressing the filesystem data itself.
52
53The "-info" option forces Unsquashfs to print each file as it is decompressed.
54
55The "-dest" option specifies the directory that is used to decompress
56the filesystem data. If this option is not given then the filesystem is
57decompressed to the directory "squashfs-root" in the current working directory.
58
59Unsquashfs can decompress 3.0 filesystems. Support for 2.x and 1.x
60filesystems will be added in the future.
61
README-3.1
1 SQUASHFS 3.1 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2006 Phillip Lougher <phillip@lougher.org.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs version 3.1-r2. Squashfs 3.1 has major improvements to
8the Squashfs tools (Mksquashfs and Unsquashfs), some major bug fixes, new
9kernel patches, and various other smaller improvements and bug fixes.
10Please see the CHANGES file for a detailed list.
11
121. MKSQUASHFS
13-------------
14
15Mksquashfs has been rewritten and it is now multi-threaded. It offers
16the following improvements:
17
181. Parallel compression. By default as many compression and fragment
19compression threads are created as there are available processors.
20This significantly speeds up performance on SMP systems.
21
222. File input and filesystem output is peformed in parallel on separate
23threads to maximise I/O performance. Even on single processor systems
24this speeds up performance by at least 10%.
25
263. Appending has been significantly improved, and files within the
27filesystem being appended to are no longer scanned and checksummed. This
28significantly improves append time for large filesystems.
29
304. File duplicate checking has been optimised, and split into two separate
31phases. Only files which are considered possible duplicates after the
32first phase are checksummed and cached in memory.
33
345. The use of swap memory was found to significantly impact performance. The
35amount of memory used to cache the file is now a command line option, by default
36this is 512 Mbytes.
37
381.1 NEW COMMAND LINE OPTIONS
39----------------------------
40
41The new Mksquashfs program has a couple of extra command line options
42which can be used to control the new features:
43
44-processors <processors>
45
46This specifies the number of processors used by Mksquashfs.
47By default this is the number of available processors.
48
49-read_queue <size in Mbytes>
50
51This specifies the size of the file input queue used by the reader thread.
52This defaults to 64 Mbytes.
53
54-write_queue <size in Mbytes>
55
56This specifies the size of the filesystem output queue used by the
57writer thread. It also specifies the maximum cache used in file
58duplicate detection (the output queue is shared between these tasks).
59This defaults to 512 Mbytes.
60
611.2 PERFORMANCE RESULTS
62-----------------------
63
64The following results give an indication of the speed improvements. Two
65example filesystems were tested, a liveCD filesystem (about 1.8 Gbytes
66uncompressed), and my home directory consisting largely of text files
67(about 1.3 Gbytes uncompressed). Tests were run on a single core
68and a dual core system.
69
70Dual Core (AMDx2 3800+) system:
71Source directories on ext3.
72
73LiveCD, old mksquashfs:
74
75real 11m48.401s
76user 9m27.056s
77sys 0m15.281s
78
79LiveCD, new par_mksquashfs:
80
81real 4m8.736s
82user 7m11.771s
83sys 0m27.749s
84
85"Home", old mksquashfs:
86
87real 4m34.360s
88user 3m54.007s
89sys 0m32.155s
90
91"Home", new par_mksquashfs:
92
93real 1m27.381s
94user 2m7.304s
95sys 0m17.234s
96
97Single Core PowerBook (PowerPC G4 1.5 GHz Ubuntu Linux)
98Source directories on ext3.
99
100LiveCD, old mksquashs:
101
102real 11m38.472s
103user 9m6.137s
104sys 0m23.799s
105
106LiveCD, par_mksquashfs:
107
108real 10m5.572s
109user 8m59.921s
110sys 0m16.145s
111
112"Home", old mksquashfs:
113
114real 3m42.298s
115user 2m49.478s
116sys 0m13.675s
117
118"Home", new par_mksquashfs:
119
120real 3m9.178s
121user 2m50.699s
122sys 0m9.069s
123
124I'll be interested in any performance results obtained, especially from SMP
125machines larger than my dual-core AMD box, as this will give an indication of
126the scalability of the code. Obviously, I'm also interested in any problems,
127deadlocks, low performance etc.
128
1292. UNSQUASHFS
130-------------
131
132Unsquashfs now allows you to specify the filename or directory that is to be
133extracted from the Squashfs filesystem, rather than always extracting the
134entire filesystem. It also has a new "-force" option, and all options can be
135specified in a short form (-i rather than -info).
136
137The Unsquashfs usage info is now:
138
139SYNTAX: ./unsquashfs [options] filesystem [directory or file to extract]
140 -v[ersion] print version, licence and copyright information
141 -i[nfo] print files as they are unsquashed
142 -l[s] list filesystem only
143 -d[est] <pathname> unsquash to <pathname>, default "squashfs-root"
144 -f[orce] if file already exists then overwrite
145
146To extract a subset of the filesystem, the filename or directory
147tree that is to be extracted can now be specified on the command line. The
148file/directory should be specified using the full path to the file/directory
149as it appears within the Squashfs filesystem. The file/directory will also be
150extracted to that position within the specified destination directory.
151
152The new "-force" option forces Unsquashfs to output to the destination
153directory even if files or directories already exist. This allows you
154to update an existing directory tree, or to Unsquashfs to a partially
155filled directory. Without the "-force" option, Unsquashfs will
156refuse to overwrite any existing files, or to create any directories if they
157already exist. This is done to protect data in case of mistakes, and
158so the "-force" option should be used with caution.
159
README-3.2
1 SQUASHFS 3.2 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2007 Phillip Lougher <phillip@lougher.org.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs version 3.2. Squashfs 3.2 has support for NFS exporting,
8some improvements to the Squashfs tools (Mksquashfs and Unsquashfs), some
9major bug fixes, new kernel patches, and various other smaller improvements
10and bug fixes. Please see the CHANGES file for a detailed list.
11
121. MKSQUASHFS
13-------------
14
15New command line options:
16
17-no-exports
18
19 Squashfs now supports NFS exports. By default the additional
20 information necessary is added to the filesystem by Mksquashfs. If you
21 do not wish this extra information, then this option can be specified.
22 This will save a couple of bytes per file, and the filesystem
23 will be identical to Squashfs 3.1.
24
25-no-progress
26
27 Mksquashfs by default now displays a progress bar. This option disables
28 it.
29
302. UNSQUASHFS
31-------------
32
33 Unsquashfs now supports Squashfs 2.x filesystems.
34
README-3.3
1 SQUASHFS 3.3 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2007 Phillip Lougher <phillip@lougher.demon.co.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to another release of Squashfs. This is the 22nd release in just
8over five years of work. Squashfs 3.3 has lots of nice improvements,
9both to the filesystem itself (bigger blocks, and sparse files), but
10also to the Squashfs-tools Mksquashfs and Unsquashfs. As usual the
11CHANGES file has a detailed list of all the improvements.
12
13Following is a description of the changes to the Squashfs tools, usage
14guides to the new options, and a summary of the new options.
15
161. MKSQUASHFS - EXTENDED EXCLUDE FILE HANDLING
17----------------------------------------------
18
191. Extended wildcard pattern matching now supported in exclude files
20
21 Enabled by specifying -wildcards option
22
23 Supports both anchored and non-anchored exclude files.
24
251.1 Anchored excludes
26
27 Similar to existing exclude files except with wildcards. Exclude
28 file matches from root of source directories.
29
30 Examples:
31
32 1. mksquashfs example image.sqsh -wildcards -e 'test/*.gz'
33
34 Exclude all files matching "*.gz" in the top level directory "test".
35
36 2. mksquashfs example image.sqsh -wildcards -e '*/[Tt]est/example*'
37
38 Exclude all files beginning with "example" inside directories called
39 "Test" or "test", that occur inside any top level directory.
40
41 Using extended wildcards, negative matching is also possible.
42
43 3. mksquashfs example image.sqsh -wildcards -e 'test/!(*data*).gz'
44
45 Exclude all files matching "*.gz" in top level directory "test",
46 except those with "data" in the name.
47
481.2 Non-anchored excludes
49
50 By default excludes match from the top level directory, but it is
51 often useful to exclude a file matching anywhere in the source directories.
52 For this non-anchored excludes can be used, specified by pre-fixing the
53 exclude with "...".
54
55 Examples:
56
57 1. mksquashfs example image.sqsh -wildcards -e '... *.gz'
58
59 Exclude files matching "*.gz" anywhere in the source directories.
60 For example this will match "example.gz", "test/example.gz", and
61 "test/test/example.gz".
62
63 2. mksquashfs example image.sqsh -wildcards -e '... [Tt]est/*.gz'
64
65 Exclude files matching "*.gz" inside directories called "Test" or
66 "test" that occur anywhere in the source directories.
67
68 Again, using extended wildcards, negative matching is also possible.
69
70 3. mksquashfs example image.sqsh -wildcards -e '... !(*data*).gz'
71
72 Exclude all files matching "*.gz" anywhere in the source directories,
73 except those with "data" in the name.
74
752. Regular expression pattern matching now supported in exclude files
76
77 Enabled by specifying -regex option. Identical behaviour to wild
78card pattern matching, except patterns are considered to be regular
79expressions.
80
81 Supports both anchored and non-anchored exclude files.
82
83
842. MKSQUASHFS - NEW RECOVERY FILE FEATURE
85-----------------------------------------
86
87Recovery files are now created when appending to existing Squashfs
88filesystems. This allows the original filesystem to be recovered
89if Mksquashfs aborts unexpectedly (i.e. power failure).
90
91The recovery files are called squashfs_recovery_xxx_yyy, where
92"xxx" is the name of the filesystem being appended to, and "yyy" is a
93number to guarantee filename uniqueness (the PID of the parent Mksquashfs
94process).
95
96Normally if Mksquashfs exits correctly the recovery file is deleted to
97avoid cluttering the filesystem. If Mksquashfs aborts, the "-recover"
98option can be used to recover the filesystem, giving the previously
99created recovery file as a parameter, i.e.
100
101mksquashfs dummy image.sqsh -recover squashfs_recovery_image.sqsh_1234
102
103The writing of the recovery file can be disabled by specifying the
104"-no-recovery" option.
105
106
1073. UNSQUASHFS - EXTENDED EXTRACT FILE HANDLING
108----------------------------------------------
109
1101. Multiple extract files can now be specified on the command line, and the
111files/directories to be extracted can now also be given in a file.
112
113To specify a file containing the extract files use the "-e[f]" option.
114
1152. Extended wildcard pattern matching now supported in extract files
116
117 Enabled by default. Similar to existing extract files except with
118wildcards.
119
120 Examples:
121
122 1. unsquashfs image.sqsh 'test/*.gz'
123
124 Extract all files matching "*.gz" in the top level directory "test".
125
126 2. unsquashfs image.sqsh '[Tt]est/example*'
127
128 Extract all files beginning with "example" inside top level directories
129 called "Test" or "test".
130
131 Using extended wildcards, negative matching is also possible.
132
133 3. unsquashfs image.sqsh 'test/!(*data*).gz'
134
135 Extract all files matching "*.gz" in top level directory "test",
136 except those with "data" in the name.
137
1383. Regular expression pattern matching now supported in extract files
139
140 Enabled by specifying -r[egex] option. Identical behaviour to wild
141card pattern matching, except patterns are considered to be regular
142expressions.
143
1444. UNSQUASHFS - EXTENDED FILENAME PRINTING
145------------------------------------------
146
147Filename printing has been enhanced and Unquashfs can now display filenames
148with file attributes ('ls -l' style output).
149
150New options:
151
152 -ll[s]
153
154 list filesystem with file attributes, but don't unsquash
155
156 -li[nfo]
157
158 print files as they are unsquashed with file attributes
159
160
1615. UNSQUASHFS - MISCELLANEOUS OPTIONS
162-------------------------------------
163
164 -s[tat]
165
166 Display the filesystem superblock information. This is useful to
167 discover the filesystem version, byte ordering, whether it has an
168 NFS export table, and what options were used to compress
169 the filesystem.
170
README-4.0
1 SQUASHFS 4.0 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2009 Phillip Lougher <phillip@lougher.demon.co.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs 4.0. This is an initial tools only release to
8support users of the 2.6.29 kernel, following the mainlining of Squashfs
9earlier this year.
10
11Later releases will probably contain kernel patches supporting 4.0
12layouts for earlier kernels.
13
14New Mksquashfs options
15----------------------
16
17Mksquashfs now supports pseudo files, these allow fake directories, character
18and block devices to be specified and added to the Squashfs filesystem being
19built, rather than requiring them to be present in the source directories.
20This, for example, allows device nodes to be added to the filesystem without
21requiring root access.
22
23Two options are supported, -p allows one pseudo file to be specified on the
24command line, and -pf allows a pseudo file to be specified containing a
25list of pseduo definitions, one per line.
26
27Pseudo device nodes are specified using 7 arguments
28
29Filename type mode uid gid major minor
30
31Where type is either
32 b - for block devices, and
33 c - for character devices
34
35mode is the octal mode specifier, similar to that expected by chmod.
36
37Uid and gid can be either specified as a decimal number, or by name.
38
39For example:
40
41/dev/chr_dev c 666 root root 100 1
42/dev/blk_dev b 444 0 0 200 200
43
44Directories are specified using 5 arguments
45
46Filename type mode uid gid
47
48Where type is d.
49
README-4.1
1 SQUASHFS 4.1 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2010 Phillip Lougher <phillip@lougher.demon.co.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs 4.1. This is a tools only release, support for Squashfs
8file systems is in mainline (2.6.29 and later).
9
10New features in Squashfs-tools 4.1
11----------------------------------
12
13 1. Support for extended attributes
14 2. Support for LZMA and LZO compression
15 3. New pseudo file features
16
17Compatiblity
18------------
19
20Mksquashfs 4.1 generates 4.0 filesystems. These filesystems are fully
21compatible/interchangable with filesystems generated by Mksquashfs 4.0 and are
22mountable on 2.6.29 and later kernels.
23
24Extended attributes (xattrs)
25----------------------------
26
27Squashfs file systems now have extended attribute support. The
28extended attribute implementation has the following features:
29
301. Layout can store up to 2^48 bytes of compressed xattr data.
312. Number of xattrs per inode unlimited.
323. Total size of xattr data per inode 2^48 bytes of compressed data.
334. Up to 4 Gbytes of data per xattr value.
345. Inline and out-of-line xattr values supported for higher performance
35 in xattr scanning (listxattr & getxattr), and to allow xattr value
36 de-duplication.
376. Both whole inode xattr duplicate detection and individual xattr value
38 duplicate detection supported. These can obviously nest, file C's
39 xattrs can be a complete duplicate of file B, and file B's xattrs
40 can be a partial duplicate of file A.
417. Xattr name prefix types stored, allowing the redundant "user.", "trusted."
42 etc. characters to be eliminated and more concisely stored.
438. Support for files, directories, symbolic links, device nodes, fifos
44 and sockets.
45
46Extended attribute support is in 2.6.35 and later kernels. File systems
47with extended attributes can be mounted on 2.6.29 and later kernels, the
48extended attributes will be ignored with a warning.
49
50LZMA and LZO compression
51------------------------
52
53Squashfs now supports LZMA and LZO compression.
54
55LZO support is in 2.6.36 and newer kernels. LZMA is not yet in mainline.
56
57New Mksquashfs options
58----------------------
59
60-comp <comp>
61
62 Select <comp> compression.
63
64 The compression algorithms supported by the build of Mksquashfs can be
65 found by typing mksquashfs without any arguments. The compressors available
66 are displayed at the end of the help message, e.g.
67
68 Compressors available:
69 gzip (default)
70 lzma
71 lzo
72
73 The default compression used when -comp isn't specified on the command line
74 is indicated by "(default)".
75
76-no-xattrs
77 Don't store extended attributes
78
79-xattrs
80 Store extended attributes
81
82 The default behaviour of Mksquashfs with respect to extended attribute
83 storage is build time selectable. The Mksquashfs help message indicates
84 whether extended attributes are stored or not, e.g.
85
86 -no-xattrs don't store extended attributes
87 -xattrs store extended attributes (default)
88
89 shows that extended attributes are stored by default, and can be disabled
90 by the -no-xattrs option.
91
92 -no-xattrs don't store extended attributes (default)
93 -xattrs store extended attributes
94
95 shows that extended attributes are not stored by default, storage can be
96 enabled by the -xattrs option.
97
98
99-noX
100-noXattrCompression
101 Don't compress extended attributes
102
103
104New Unsquashfs options
105----------------------
106
107-n[o-xattrs]
108 Don't extract xattrs in filesystem
109
110-x[attrs]
111 Extract xattrs in filesystem
112
113 The default behaviour of Unsquashfs with respect to extended attributes
114 is build time selectable. The Unsquashfs help message indicates whether
115 extended attributes are stored or not, e.g.
116
117 -no[-xattrs] don't extract xattrs in file system
118 -x[attrs] extract xattrs in file system (default)
119
120 shows that xattrs are extracted by default.
121
122 -no[-xattrs] don't extract xattrs in file system (default)
123 -x[attrs] extract xattrs in file system
124
125 shows that xattrs are not extracted by default.
126
127
128New pseudo file support
129-----------------------
130
131Mksquashfs supports pseudo files, these allow fake files, directories, character
132and block devices to be specified and added to the Squashfs filesystem being
133built, rather than requiring them to be present in the source directories.
134This, for example, allows device nodes to be added to the filesystem without
135requiring root access.
136
137Mksquashfs 4.1 adds support for "dynamic pseudo files" and a modify operation.
138Dynamic pseudo files allow files to be dynamically created when Mksquashfs
139is run, their contents being the result of running a command or piece of
140shell script. The modifiy operation allows the mode/uid/gid of an existing
141file in the source filesystem to be modified.
142
143Two Mksquashfs options are supported, -p allows one pseudo file to be specified
144on the command line, and -pf allows a pseudo file to be specified containing a
145list of pseduo definitions, one per line.
146
147Pseudo operations
148-----------------
149
1501. Creating a dynamic file
151--------------------------
152
153Pseudo definition
154
155Filename f mode uid gid command
156
157mode is the octal mode specifier, similar to that expected by chmod.
158
159uid and gid can be either specified as a decimal number, or by name.
160
161command can be an executable or a piece of shell script, and it is executed
162by running "/bin/sh -c command". The stdout becomes the contents of
163"Filename".
164
165Examples:
166
167Running a basic command
168-----------------------
169
170/somedir/dmesg f 444 root root dmesg
171
172creates a file "/somedir/dmesg" containing the output from dmesg.
173
174Executing shell script
175----------------------
176
177RELEASE f 444 root root \
178 if [ ! -e /tmp/ver ]; then \
179 echo 0 > /tmp/ver; \
180 fi; \
181 ver=`cat /tmp/ver`; \
182 ver=$((ver +1)); \
183 echo $ver > /tmp/ver; \
184 echo -n `cat /tmp/release`; \
185 echo "-dev #"$ver `date` "Build host" `hostname`
186
187Creates a file RELEASE containing the release name, date, build host, and
188an incrementing version number. The incrementing version is a side-effect
189of executing the shell script, and ensures every time Mksquashfs is run a
190new version number is used without requiring any other shell scripting.
191
192The above example also shows that commands can be split across multiple lines
193using "\". Obviously as the script will be presented to the shell as a single
194line, a semicolon is need to separate individual shell commands within the
195shell script.
196
197Reading from a device (or fifo/named socket)
198--------------------------------------------
199
200input f 444 root root dd if=/dev/sda1 bs=1024 count=10
201
202Copies 10K from the device /dev/sda1 into the file input. Ordinarily Mksquashfs
203given a device, fifo, or named socket will place that special file within the
204Squashfs filesystem, the above allows input from these special files to be
205captured and placed in the Squashfs filesystem.
206
2072. Creating a block or character device
208---------------------------------------
209
210Pseudo definition
211
212Filename type mode uid gid major minor
213
214Where type is either
215 b - for block devices, and
216 c - for character devices
217
218mode is the octal mode specifier, similar to that expected by chmod.
219
220uid and gid can be either specified as a decimal number, or by name.
221
222For example:
223
224/dev/chr_dev c 666 root root 100 1
225/dev/blk_dev b 666 0 0 200 200
226
227creates a character device "/dev/chr_dev" with major:minor 100:1 and
228a block device "/dev/blk_dev" with major:minor 200:200, both with root
229uid/gid and a mode of rw-rw-rw.
230
2313. Creating a directory
232-----------------------
233
234Pseudo definition
235
236Filename d mode uid gid
237
238mode is the octal mode specifier, similar to that expected by chmod.
239
240uid and gid can be either specified as a decimal number, or by name.
241
242For example:
243
244/pseudo_dir d 666 root root
245
246creates a directory "/pseudo_dir" with root uid/gid and mode of rw-rw-rw.
247
2484. Modifying attributes of an existing file
249-------------------------------------------
250
251Pseudo definition
252
253Filename m mode uid gid
254
255mode is the octal mode specifier, similar to that expected by chmod.
256
257uid and gid can be either specified as a decimal number, or by name.
258
259For example:
260
261dmesg m 666 root root
262
263Changes the attributes of the file "dmesg" in the filesystem to have
264root uid/gid and a mode of rw-rw-rw, overriding the attributes obtained
265from the source filesystem.
266
README-4.2
1 SQUASHFS 4.2 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2011 Phillip Lougher <phillip@lougher.demon.co.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs 4.2. This is a tools only release, support for Squashfs
8filesystems is in mainline (2.6.29 and later).
9
10New features in Squashfs-tools 4.2
11----------------------------------
12
13 1. Support for XZ compression
14 2. Support for compressor specific options
15
16Compatiblity
17------------
18
19Mksquashfs 4.2 generates 4.0 filesystems. These filesystems are fully
20compatible/interchangable with filesystems generated by Mksquashfs 4.0 and are
21mountable on 2.6.29 and later kernels.
22
23XZ compression
24--------------
25
26Squashfs now supports XZ compression.
27
28XZ support is in 2.6.38 and newer kernels.
29
30New Mksquashfs options
31----------------------
32
33-X<compressor-option>
34
35 Compression algorithms can now support compression specific options. These
36options are prefixed by -X, and are passed to the compressor for handling.
37
38 The compression specific options supported by each compressor can be
39found by typing mksquashfs without any arguments. They are displayed at the
40end of the help message, e.g.
41
42Compressors available and compressor specific options:
43 gzip (no options) (default)
44 lzo (no options)
45 xz
46 -Xbcj filter1,filter2,...,filterN
47 Compress using filter1,filter2,...,filterN in turn
48 (in addition to no filter), and choose the best compression.
49 Available filters: x86, arm, armthumb, powerpc, sparc, ia64
50 -Xdict-size <dict-size>
51 Use <dict-size> as the XZ dictionary size. The dictionary size
52 can be specified as a percentage of the block size, or as an
53 absolute value. The dictionary size must be less than or equal
54 to the block size and 8192 bytes or larger. It must also be
55 storable in the xz header as either 2^n or as 2^n+2^(n+1).
56 Example dict-sizes are 75%, 50%, 37.5%, 25%, or 32K, 16K, 8K
57 etc.
58
README-4.3
1 SQUASHFS 4.3 - A squashed read-only filesystem for Linux
2
3 Copyright 2002-2014 Phillip Lougher <phillip@lougher.demon.co.uk>
4
5 Released under the GPL licence (version 2 or later).
6
7Welcome to Squashfs 4.3. This is the first release in over 3 years, and
8there are substantial improvements to stability, new compression options
9and compressors, speed optimisations, and new options for Mksquashfs/Unsquashfs.
10
11This is a tools only release, support for Squashfs filesystems is
12in mainline (2.6.29 and later).
13
14Changes in Squashfs-tools 4.3
15-----------------------------
16
171. Stability improvements. Better checking of user input for out of
18 range/invalid values. Better handling of corrupted Squashfs filesystems
19 (Mksquashfs append mode, and Unsquashfs). Better handling of buffer
20 overflow/underflow.
21
222. GZIP compressor now supports compression options, allowing different
23 compression levels to be used.
24
253. Rewritten LZO compressor with compression options, allowing different
26 LZO algorithms and different compression levels to be used.
27
284. New LZ4 compressor (note not yet in mainline kernel)
29
305. Better default memory usage for Mksquashfs. Mksquashfs by default now
31 uses 25% of physical memory.
32
336. Duplicate checking in Mksquashfs further optimised. With certain
34 "problem filesystems" greater than 2x performance improvement.
35 Filesystems with a lot of duplicates should see at least 10-20% speed
36 improvement.
37
387. The -stat option in Unsquashfs now displays the compression options
39 used to generate the original filesystem. Previously -stat only displayed
40 the compression algorithm used.
41
428. The file being compressed/uncompressed in Mksquashfs/Unsquashfs is now
43 displayed if CTRL-\ (SIGQUIT from keyboard) typed.
44
459. The status of the internal queues/caches in Mksquashfs/Unsquashfs is
46 now displayed if CTRL-\ (SIGQUIT from keyboard) is typed twice within
47 one second. Normally only useful for "power users", but it can be
48 used to discover if there's any bottlenecks affecting performance
49 (the bottleneck will normally be the compressors/fragment compressors).
50
5110. Miscellaneous new options for Mksquashfs/Unsquashfs to fine tune behaviour.
52
5311. Fixes for CVE-2012-4024 and CVE-2012-4025.
54
55Compatiblity
56------------
57
58Mksquashfs 4.3 generates 4.0 filesystems. These filesystems are fully
59compatible/interchangable with filesystems generated by Mksquashfs 4.0 and are
60mountable on 2.6.29 and later kernels.
61
62Compressors
63-----------
64
65New compression options and compressors are now supported.
66
67The new options and compressors are:
68
691. gzip
70 -Xcompression-level <compression-level>
71 <compression-level> should be 1 .. 9 (default 9)
72 -Xwindow-size <window-size>
73 <window-size> should be 8 .. 15 (default 15)
74 -Xstrategy strategy1,strategy2,...,strategyN
75 Compress using strategy1,strategy2,...,strategyN in turn
76 and choose the best compression.
77 Available strategies: default, filtered, huffman_only,
78 run_length_encoded and fixed
79
802. lzo
81 -Xalgorithm <algorithm>
82 Where <algorithm> is one of:
83 lzo1x_1
84 lzo1x_1_11
85 lzo1x_1_12
86 lzo1x_1_15
87 lzo1x_999 (default)
88 -Xcompression-level <compression-level>
89 <compression-level> should be 1 .. 9 (default 8)
90 Only applies to lzo1x_999 algorithm
91
923. lz4
93 -Xhc
94 Compress using LZ4 High Compression
95
96The compression specific options are, obviously, specific to the compressor
97in question, and you should read the compressor documentation and check
98their web sites to understand their behaviour.
99
100In general the defaults used by Mksquashfs for each compressor are optimised
101to give the best performance for each compressor, where what constitutes
102best depends on the compressor. For gzip/xz best means highest compression
103(trying multiple filters/strategies can improve compression, but this is
104extremely expensive computationally, and hence, not suitable for the defaults),
105for LZO/LZ4 best means a tradeoff between compression and (de)-compression
106overhead (LZO/LZ4 by definition are intended for weaker processors).
107
108New Mksquashfs options
109----------------------
110
1111. -mem <size>
112
113 Set the amount of memory used by Mksquashfs to <size> bytes. G/M and K
114 post-fixes are supported.
115
116 By default Mksquashfs uses 25% of the physical memory. Increasing
117 this with the -mem option can increase performance (note it does not have
118 any effect on compression). Reducing it can prevent thrashing if the
119 system is busy and there is not 25% of physical memory free (again, note
120 it does not have any effect on compression).
121
1222. -exit-on-error
123
124 By default Mksquashfs treats certain errors as benign, if these
125 errors occur Mksquashfs prints the error on the console but continues.
126 These errors are typically failure to read a file from the source filesystem.
127 This is deliberate, in many cases users prefer Mksquashfs to flag
128 the error but continue rather than abort what may be hours of compression.
129
130 But there are times where failure to read any file is considered critical,
131 and users (especially in the case of automated scripts where the
132 errors output to the console may be missed) prefer Mksquashfs to exit.
133
134 The new -exit-on-error option can be used in this scenario. This option
135 makes Mksquashfs treat all benign errors as fatal.
136
1373. -progress
138
139 By default if -info is specified, the progress bar is disabled as it gets
140 in the way. Occasionally you might want the progress bar enabled whilst
141 -info is enabled. This option forces Mksquashfs to output the progress
142 bar when -info is specified.
143
1444. -Xhelp
145
146 Display the usage text for the currently selected compressor.
147
148New Unsquashfs options
149----------------------
150
1511. -u[ser-xattrs]
152
153 Only write user xattrs. This forces Unsquashfs to ignore system xattrs.
154 This is useful when Unsquashing a filesystem as a non-root user, and the
155 filesystem contains system xattrs which are only writable by root.
156
157Major bugs fixed
158----------------
159
1601. If Mksquashfs ran out of space in the destination filesystem, this
161 would not cause Mksquashfs to immediately abort, and Mksquashfs would
162 continue to process the source filesystem. Mksquashfs now immediately
163 aborts on out of space in the destination filesystem.
164
1652. Unsquashfs ignored the maximum number of open files limit, and if that
166 was lower than the default limit for Linux, it would run out of file
167 descriptors. Unsquashfs now limits the number of open files to the
168 limit currently in force (e.g. specified by setrlimit).
169
1703. If huge numbers of dynamic pseudo files were specified, Mksquashfs
171 could exceed the maximum number of open files limit. This was because
172 Mksquashfs created all the dynamic file processes up front before
173 commencing source filesystem reading and compression. Mksquashfs
174 now creates the dynamic file processes on demand whilst reading
175 and compressing the source filesystem, thus limiting the number of
176 dynamic pseudo file processes in existence at any one time.
177
1784. When outputting Unsquashfs used to set the permissions of directories
179 as it recursively descended. This in hindsight had an obvious oversight,
180 if a directory had only read permission (or was otherwise restricted), then
181 Unsquashfs would fail to write its contents when descending into it. Fixed
182 by setting directory permissions as Unsquashfs recursively unwinds.
183