source: src/linux/universal/linux-4.9/fs/squashfs/Kconfig @ 31630

Last change on this file since 31630 was 31630, checked in by brainslayer, 10 days ago

new fs is compatible with standard squashfs, just smaller

File size: 7.3 KB
Line 
1config SQUASHFS
2        tristate "SquashFS 4.0 - Squashed file system support"
3        depends on BLOCK
4        help
5          Saying Y here includes support for SquashFS 4.0 (a Compressed
6          Read-Only File System).  Squashfs is a highly compressed read-only
7          filesystem for Linux.  It uses zlib, lzo or xz compression to
8          compress both files, inodes and directories.  Inodes in the system
9          are very small and all blocks are packed to minimise data overhead.
10          Block sizes greater than 4K are supported up to a maximum of 1 Mbytes
11          (default block size 128K).  SquashFS 4.0 supports 64 bit filesystems
12          and files (larger than 4GB), full uid/gid information, hard links and
13          timestamps.
14
15          Squashfs is intended for general read-only filesystem use, for
16          archival use (i.e. in cases where a .tar.gz file may be used), and in
17          embedded systems where low overhead is needed.  Further information
18          and tools are available from http://squashfs.sourceforge.net.
19
20          If you want to compile this as a module ( = code which can be
21          inserted in and removed from the running kernel whenever you want),
22          say M here.  The module will be called squashfs.  Note that the root
23          file system (the one containing the directory /) cannot be compiled
24          as a module.
25
26          If unsure, say N.
27
28choice
29        prompt "File decompression options"
30        depends on SQUASHFS
31        help
32          Squashfs now supports two options for decompressing file
33          data.  Traditionally Squashfs has decompressed into an
34          intermediate buffer and then memcopied it into the page cache.
35          Squashfs now supports the ability to decompress directly into
36          the page cache.
37
38          If unsure, select "Decompress file data into an intermediate buffer"
39
40config SQUASHFS_FILE_CACHE
41        bool "Decompress file data into an intermediate buffer"
42        help
43          Decompress file data into an intermediate buffer and then
44          memcopy it into the page cache.
45
46config SQUASHFS_FILE_DIRECT
47        bool "Decompress files directly into the page cache"
48        help
49          Directly decompress file data into the page cache.
50          Doing so can significantly improve performance because
51          it eliminates a memcpy and it also removes the lock contention
52          on the single buffer.
53
54endchoice
55
56choice
57        prompt "Decompressor parallelisation options"
58        depends on SQUASHFS
59        help
60          Squashfs now supports three parallelisation options for
61          decompression.  Each one exhibits various trade-offs between
62          decompression performance and CPU and memory usage.
63
64          If in doubt, select "Single threaded compression"
65
66config SQUASHFS_DECOMP_SINGLE
67        bool "Single threaded compression"
68        help
69          Traditionally Squashfs has used single-threaded decompression.
70          Only one block (data or metadata) can be decompressed at any
71          one time.  This limits CPU and memory usage to a minimum.
72
73config SQUASHFS_DECOMP_MULTI
74        bool "Use multiple decompressors for parallel I/O"
75        help
76          By default Squashfs uses a single decompressor but it gives
77          poor performance on parallel I/O workloads when using multiple CPU
78          machines due to waiting on decompressor availability.
79
80          If you have a parallel I/O workload and your system has enough memory,
81          using this option may improve overall I/O performance.
82
83          This decompressor implementation uses up to two parallel
84          decompressors per core.  It dynamically allocates decompressors
85          on a demand basis.
86
87config SQUASHFS_DECOMP_MULTI_PERCPU
88        bool "Use percpu multiple decompressors for parallel I/O"
89        help
90          By default Squashfs uses a single decompressor but it gives
91          poor performance on parallel I/O workloads when using multiple CPU
92          machines due to waiting on decompressor availability.
93
94          This decompressor implementation uses a maximum of one
95          decompressor per core.  It uses percpu variables to ensure
96          decompression is load-balanced across the cores.
97
98endchoice
99
100config SQUASHFS_XATTR
101        bool "Squashfs XATTR support"
102        depends on SQUASHFS
103        help
104          Saying Y here includes support for extended attributes (xattrs).
105          Xattrs are name:value pairs associated with inodes by
106          the kernel or by users (see the attr(5) manual page).
107
108          If unsure, say N.
109
110config SQUASHFS_ZLIB
111        bool "Include support for ZLIB compressed file systems"
112        depends on SQUASHFS
113        select ZLIB_INFLATE
114        default y
115        help
116          ZLIB compression is the standard compression used by Squashfs
117          file systems.  It offers a good trade-off between compression
118          achieved and the amount of CPU time and memory necessary to
119          compress and decompress.
120
121          If unsure, say Y.
122
123config SQUASHFS_LZ4
124        bool "Include support for LZ4 compressed file systems"
125        depends on SQUASHFS
126        select LZ4_DECOMPRESS
127        help
128          Saying Y here includes support for reading Squashfs file systems
129          compressed with LZ4 compression.  LZ4 compression is mainly
130          aimed at embedded systems with slower CPUs where the overheads
131          of zlib are too high.
132
133          LZ4 is not the standard compression used in Squashfs and so most
134          file systems will be readable without selecting this option.
135
136          If unsure, say N.
137
138config SQUASHFS_LZO
139        bool "Include support for LZO compressed file systems"
140        depends on SQUASHFS
141        select LZO_DECOMPRESS
142        help
143          Saying Y here includes support for reading Squashfs file systems
144          compressed with LZO compression.  LZO compression is mainly
145          aimed at embedded systems with slower CPUs where the overheads
146          of zlib are too high.
147
148          LZO is not the standard compression used in Squashfs and so most
149          file systems will be readable without selecting this option.
150
151          If unsure, say N.
152
153config SQUASHFS_XZ
154        bool "Include support for XZ compressed file systems"
155        depends on SQUASHFS
156        select XZ_DEC
157        help
158          Saying Y here includes support for reading Squashfs file systems
159          compressed with XZ compression.  XZ gives better compression than
160          the default zlib compression, at the expense of greater CPU and
161          memory overhead.
162
163          XZ is not the standard compression used in Squashfs and so most
164          file systems will be readable without selecting this option.
165
166          If unsure, say N.
167
168config SQUASHFS_4K_DEVBLK_SIZE
169        bool "Use 4K device block size?"
170        depends on SQUASHFS
171        help
172          By default Squashfs sets the dev block size (sb_min_blocksize)
173          to 1K or the smallest block size supported by the block device
174          (if larger).  This, because blocks are packed together and
175          unaligned in Squashfs, should reduce latency.
176
177          This, however, gives poor performance on MTD NAND devices where
178          the optimal I/O size is 4K (even though the devices can support
179          smaller block sizes).
180
181          Using a 4K device block size may also improve overall I/O
182          performance for some file access patterns (e.g. sequential
183          accesses of files in filesystem order) on all media.
184
185          Setting this option will force Squashfs to use a 4K device block
186          size by default.
187
188          If unsure, say N.
189
190config SQUASHFS_EMBEDDED
191        bool "Additional option for memory-constrained systems"
192        depends on SQUASHFS
193        help
194          Saying Y here allows you to specify cache size.
195
196          If unsure, say N.
197
198config SQUASHFS_FRAGMENT_CACHE_SIZE
199        int "Number of fragments cached" if SQUASHFS_EMBEDDED
200        depends on SQUASHFS
201        default "3"
202        help
203          By default SquashFS caches the last 3 fragments read from
204          the filesystem.  Increasing this amount may mean SquashFS
205          has to re-read fragments less often from disk, at the expense
206          of extra system memory.  Decreasing this amount will mean
207          SquashFS uses less memory at the expense of extra reads from disk.
208
209          Note there must be at least one cached fragment.  Anything
210          much more than three will probably not make much difference.
Note: See TracBrowser for help on using the repository browser.