5 XFS is a high performance journaling filesystem which originated
6 on the SGI IRIX platform. It is completely multi-threaded, can
7 support large files and large filesystems, extended attributes,
8 variable block sizes, is extent based, and makes extensive use of
9 Btrees (directories, extents, free space) to aid both performance
12 Refer to the documentation at http://oss.sgi.com/projects/xfs/
13 for further details. This implementation is on-disk compatible
14 with the IRIX version of XFS.
20 When mounting an XFS filesystem, the following options are accepted.
23 Sets the buffered I/O end-of-file preallocation size when
24 doing delayed allocation writeout (default size is 64KiB).
25 Valid values for this option are page size (typically 4KiB)
26 through to 1GiB, inclusive, in power-of-2 increments.
29 The options enable/disable (default is disabled for backward
30 compatibility on-disk) an "opportunistic" improvement to be
31 made in the way inline extended attributes are stored on-disk.
32 When the new form is used for the first time (by setting or
33 removing extended attributes) the on-disk superblock feature
34 bit field will be updated to reflect this format being in use.
37 Enables the use of block layer write barriers for writes into
38 the journal and unwritten extent conversion. This allows for
39 drive level write caching to be enabled, for devices that
40 support write barriers.
43 Enable the DMAPI (Data Management API) event callouts.
44 Use with the "mtpt" option.
46 grpid/bsdgroups and nogrpid/sysvgroups
47 These options define what group ID a newly created file gets.
48 When grpid is set, it takes the group ID of the directory in
49 which it is created; otherwise (the default) it takes the fsgid
50 of the current process, unless the directory has the setgid bit
51 set, in which case it takes the gid from the parent directory,
52 and also gets the setgid bit set if it is a directory itself.
55 In memory inode hashes have been removed, so this option has
56 no function as of August 2007. Option is deprecated.
59 When ikeep is specified, XFS does not delete empty inode clusters
60 and keeps them around on disk. ikeep is the traditional XFS
61 behaviour. When noikeep is specified, empty inode clusters
62 are returned to the free space pool. The default is noikeep.
65 Indicates that XFS is allowed to create inodes at any location
66 in the filesystem, including those which will result in inode
67 numbers occupying more than 32 bits of significance. This is
68 provided for backwards compatibility, but causes problems for
69 backup applications that cannot handle large inode numbers.
72 If "nolargeio" is specified, the optimal I/O reported in
73 st_blksize by stat(2) will be as small as possible to allow user
74 applications to avoid inefficient read/modify/write I/O.
75 If "largeio" specified, a filesystem that has a "swidth" specified
76 will return the "swidth" value (in bytes) in st_blksize. If the
77 filesystem does not have a "swidth" specified but does specify
78 an "allocsize" then "allocsize" (in bytes) will be returned
80 If neither of these two options are specified, then filesystem
81 will behave as if "nolargeio" was specified.
84 Set the number of in-memory log buffers. Valid numbers range
86 The default value is 8 buffers for filesystems with a
87 blocksize of 64KiB, 4 buffers for filesystems with a blocksize
88 of 32KiB, 3 buffers for filesystems with a blocksize of 16KiB
89 and 2 buffers for all other configurations. Increasing the
90 number of buffers may increase performance on some workloads
91 at the cost of the memory used for the additional log buffers
92 and their associated control structures.
95 Set the size of each in-memory log buffer.
96 Size may be specified in bytes, or in kilobytes with a "k" suffix.
97 Valid sizes for version 1 and version 2 logs are 16384 (16k) and
98 32768 (32k). Valid sizes for version 2 logs also include
99 65536 (64k), 131072 (128k) and 262144 (256k).
100 The default value for machines with more than 32MiB of memory
101 is 32768, machines with less memory use 16384 by default.
103 logdev=device and rtdev=device
104 Use an external log (metadata journal) and/or real-time device.
105 An XFS filesystem has up to three parts: a data section, a log
106 section, and a real-time section. The real-time section is
107 optional, and the log section can be separate from the data
108 section or contained within it.
111 Use with the "dmapi" option. The value specified here will be
112 included in the DMAPI mount event, and should be the path of
113 the actual mountpoint that is used.
116 Data allocations will not be aligned at stripe unit boundaries.
119 Access timestamps are not updated when a file is read.
122 The filesystem will be mounted without running log recovery.
123 If the filesystem was not cleanly unmounted, it is likely to
124 be inconsistent when mounted in "norecovery" mode.
125 Some files or directories may not be accessible because of this.
126 Filesystems mounted "norecovery" must be mounted read-only or
130 Don't check for double mounted file systems using the file system uuid.
131 This is useful to mount LVM snapshot volumes.
134 Make O_SYNC writes implement true O_SYNC. WITHOUT this option,
135 Linux XFS behaves as if an "osyncisdsync" option is used,
136 which will make writes to files opened with the O_SYNC flag set
137 behave as if the O_DSYNC flag had been used instead.
138 This can result in better performance without compromising
140 However if this option is not in effect, timestamp updates from
141 O_SYNC writes can be lost if the system crashes.
142 If timestamp updates are critical, use the osyncisosync option.
144 uquota/usrquota/uqnoenforce/quota
145 User disk quota accounting enabled, and limits (optionally)
146 enforced. Refer to xfs_quota(8) for further details.
148 gquota/grpquota/gqnoenforce
149 Group disk quota accounting enabled and limits (optionally)
150 enforced. Refer to xfs_quota(8) for further details.
152 pquota/prjquota/pqnoenforce
153 Project disk quota accounting enabled and limits (optionally)
154 enforced. Refer to xfs_quota(8) for further details.
156 sunit=value and swidth=value
157 Used to specify the stripe unit and width for a RAID device or
158 a stripe volume. "value" must be specified in 512-byte block
160 If this option is not specified and the filesystem was made on
161 a stripe volume or the stripe width or unit were specified for
162 the RAID device at mkfs time, then the mount system call will
163 restore the value from the superblock. For filesystems that
164 are made directly on RAID devices, these options can be used
165 to override the information in the superblock if the underlying
166 disk layout changes after the filesystem has been created.
167 The "swidth" option is required if the "sunit" option has been
168 specified, and must be a multiple of the "sunit" value.
171 Data allocations will be rounded up to stripe width boundaries
172 when the current end of file is being extended and the file
173 size is larger than the stripe width size.
179 The following sysctls are available for the XFS filesystem:
181 fs.xfs.stats_clear (Min: 0 Default: 0 Max: 1)
182 Setting this to "1" clears accumulated XFS statistics
183 in /proc/fs/xfs/stat. It then immediately resets to "0".
185 fs.xfs.xfssyncd_centisecs (Min: 100 Default: 3000 Max: 720000)
186 The interval at which the xfssyncd thread flushes metadata
187 out to disk. This thread will flush log activity out, and
188 do some processing on unlinked inodes.
190 fs.xfs.xfsbufd_centisecs (Min: 50 Default: 100 Max: 3000)
191 The interval at which xfsbufd scans the dirty metadata buffers list.
193 fs.xfs.age_buffer_centisecs (Min: 100 Default: 1500 Max: 720000)
194 The age at which xfsbufd flushes dirty metadata buffers to disk.
196 fs.xfs.error_level (Min: 0 Default: 3 Max: 11)
197 A volume knob for error reporting when internal errors occur.
198 This will generate detailed messages & backtraces for filesystem
199 shutdowns, for example. Current threshold values are:
205 fs.xfs.panic_mask (Min: 0 Default: 0 Max: 127)
206 Causes certain error conditions to call BUG(). Value is a bitmask;
207 AND together the tags which represent errors which should cause panics:
210 XFS_PTAG_IFLUSH 0x00000001
211 XFS_PTAG_LOGRES 0x00000002
212 XFS_PTAG_AILDELETE 0x00000004
213 XFS_PTAG_ERROR_REPORT 0x00000008
214 XFS_PTAG_SHUTDOWN_CORRUPT 0x00000010
215 XFS_PTAG_SHUTDOWN_IOERROR 0x00000020
216 XFS_PTAG_SHUTDOWN_LOGERROR 0x00000040
218 This option is intended for debugging only.
220 fs.xfs.irix_symlink_mode (Min: 0 Default: 0 Max: 1)
221 Controls whether symlinks are created with mode 0777 (default)
222 or whether their mode is affected by the umask (irix mode).
224 fs.xfs.irix_sgid_inherit (Min: 0 Default: 0 Max: 1)
225 Controls files created in SGID directories.
226 If the group ID of the new file does not match the effective group
227 ID or one of the supplementary group IDs of the parent dir, the
228 ISGID bit is cleared if the irix_sgid_inherit compatibility sysctl
231 fs.xfs.restrict_chown (Min: 0 Default: 1 Max: 1)
232 Controls whether unprivileged users can use chown to "give away"
233 a file to another user.
235 fs.xfs.inherit_sync (Min: 0 Default: 1 Max: 1)
236 Setting this to "1" will cause the "sync" flag set
237 by the xfs_io(8) chattr command on a directory to be
238 inherited by files in that directory.
240 fs.xfs.inherit_nodump (Min: 0 Default: 1 Max: 1)
241 Setting this to "1" will cause the "nodump" flag set
242 by the xfs_io(8) chattr command on a directory to be
243 inherited by files in that directory.
245 fs.xfs.inherit_noatime (Min: 0 Default: 1 Max: 1)
246 Setting this to "1" will cause the "noatime" flag set
247 by the xfs_io(8) chattr command on a directory to be
248 inherited by files in that directory.
250 fs.xfs.inherit_nosymlinks (Min: 0 Default: 1 Max: 1)
251 Setting this to "1" will cause the "nosymlinks" flag set
252 by the xfs_io(8) chattr command on a directory to be
253 inherited by files in that directory.
255 fs.xfs.rotorstep (Min: 1 Default: 1 Max: 256)
256 In "inode32" allocation mode, this option determines how many
257 files the allocator attempts to allocate in the same allocation
258 group before moving to the next allocation group. The intent
259 is to control the rate at which the allocator moves between
260 allocation groups when allocating extents for new files.