SQUASHFS/squashfs-tools-4.4/RELEASE-READMEs/README-2.0

   1 NOTE:  This the original README for version 2.0.  It is retained as it
   2 contains information about the fragment design.  A description of the new 2.0
   3 mksquashfs options has been added to the main README file, and that
   4 file 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
  12 Welcome to the final release of Squashfs version 2.0!  A lot of changes to the
  13 filesystem have been made under the bonnet (hood).  Squashfs 2.0 uses fragment
  14 blocks and larger blocks (64K) to improve compression ratio by about 5 - 20%
  15 over Squashfs 1.0 depending on the files being compressed.  Using fragment
  16 blocks allows Squashfs 2.0 to achieve better compression than cloop and similar
  17 compression to tgz files while retaining the I/O efficiency of a compressed
  18 filesystem.
  19
  20 Detailed changes:
  21
  22 1. 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
  28 2. The maximum block size has been increased to 64K.
  29
  30 3. The maximum number of UIDs has been increased to 256 (from 48 in 1.x).
  31
  32 4. The maximum number of GIDs has been increased to 256 (from 15 in 1.x).
  33
  34 5. 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
  39 6. Initrds are now supported for kernels 2.6.x.
  40
  41 7. Removal of sleep_on() function call in 2.6.x patch, to allow Squashfs
  42    to work on the Fedora rc2 kernel.
  43
  44 8. AMD64, check-data and gid bug fixes.
  45
  46 9. Numerous small bug fixes have been made.
  47
  48 10. New patch for Linux 2.6.7.
  49
  50
  51 New 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
 107 Compression improvements example
 108 --------------------------------
 109
 110 The following is the compression results obtained compressing the 2.6.6
 111 linux kernel source using CRAMFS, Cloop (with iso filesystem), Squashfs 1.3 and
 112 Squashfs 2.0 (results generated using big-endian filesystems).
 113
 114 In 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
 122 The Squashfs 1.x filesystem is 12.6% larger than the new 2.0 filesystem.
 123 The cloop filesystem is 1.1% larger than the Squashfs 2.0 filesystem.
 124
 125
 126 Fragment blocks in Squashfs 2.0
 127 -------------------------------
 128
 129 Squashfs like all other compressed filesystems compresses files individually
 130 on a block by block basis.  This is performed to allow mounting and
 131 de-compression of files on a block by block basis without requiring the entire
 132 filesystem to be decompressed.  This is in contrast to data-based compression
 133 schemes which compress without understanding the underlying filesystem (i.e.
 134 cloop and tgz files) and which, therefore, do not compress files individually.
 135 Each approach has advantages and disadvantages, data-based systems have better
 136 compression 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
 138 the block size).  Compressed filesystems tend to be faster at I/O because
 139 they understand the filesystem and therefore employ better caching stategies
 140 and read less un-needed data from the filesystem.
 141
 142 Fragment blocks in Squashfs 2.0 solves this problem by packing files (and
 143 optionally the ends of files) which are smaller than the block size into
 144 shared blocks, which are compressed together.  For example five files each of
 145 10K will be packed into one shared fragment of 50K and compressed together,
 146 rather than being compressed in five 10K blocks.
 147
 148 This scheme produces a hybrid filesystem, retaining the I/O efficiency
 149 of a compressed filesystem, while obtaining the compression efficiency
 150 of data-based schemes by compressing small files together.
 151
 152
 153 Squashfs 1.x and Squashfs 2.0 compatibility
 154 -------------------------------------------
 155
 156 Appending to Squashfs 1.x filesystems is not supported.  If you wish to append
 157 to 1.x filesystems, then either use the original mksquashfs, or convert them
 158 to Squashfs 2.0 by mounting the filesystem and running the 2.0 mksquashfs
 159 on the mounted filesystem.
 160
 161 Mounting Squashfs 1.x filesystems IS supported by the 2.0 kernel patch.