1.. SPDX-License-Identifier: GPL-2.0 2 3====================== 4The SGI XFS Filesystem 5====================== 6 7XFS is a high performance journaling filesystem which originated 8on the SGI IRIX platform. It is completely multi-threaded, can 9support large files and large filesystems, extended attributes, 10variable block sizes, is extent based, and makes extensive use of 11Btrees (directories, extents, free space) to aid both performance 12and scalability. 13 14Refer to the documentation at https://xfs.wiki.kernel.org/ 15for further details. This implementation is on-disk compatible 16with the IRIX version of XFS. 17 18 19Mount Options 20============= 21 22When mounting an XFS filesystem, the following options are accepted. 23 24 allocsize=size 25 Sets the buffered I/O end-of-file preallocation size when 26 doing delayed allocation writeout (default size is 64KiB). 27 Valid values for this option are page size (typically 4KiB) 28 through to 1GiB, inclusive, in power-of-2 increments. 29 30 The default behaviour is for dynamic end-of-file 31 preallocation size, which uses a set of heuristics to 32 optimise the preallocation size based on the current 33 allocation patterns within the file and the access patterns 34 to the file. Specifying a fixed ``allocsize`` value turns off 35 the dynamic behaviour. 36 37 discard or nodiscard (default) 38 Enable/disable the issuing of commands to let the block 39 device reclaim space freed by the filesystem. This is 40 useful for SSD devices, thinly provisioned LUNs and virtual 41 machine images, but may have a performance impact. 42 43 Note: It is currently recommended that you use the ``fstrim`` 44 application to ``discard`` unused blocks rather than the ``discard`` 45 mount option because the performance impact of this option 46 is quite severe. 47 48 grpid/bsdgroups or nogrpid/sysvgroups (default) 49 These options define what group ID a newly created file 50 gets. When ``grpid`` is set, it takes the group ID of the 51 directory in which it is created; otherwise it takes the 52 ``fsgid`` of the current process, unless the directory has the 53 ``setgid`` bit set, in which case it takes the ``gid`` from the 54 parent directory, and also gets the ``setgid`` bit set if it is 55 a directory itself. 56 57 filestreams 58 Make the data allocator use the filestreams allocation mode 59 across the entire filesystem rather than just on directories 60 configured to use it. 61 62 inode32 or inode64 (default) 63 When ``inode32`` is specified, it indicates that XFS limits 64 inode creation to locations which will not result in inode 65 numbers with more than 32 bits of significance. 66 67 When ``inode64`` is specified, it indicates that XFS is allowed 68 to create inodes at any location in the filesystem, 69 including those which will result in inode numbers occupying 70 more than 32 bits of significance. 71 72 ``inode32`` is provided for backwards compatibility with older 73 systems and applications, since 64 bits inode numbers might 74 cause problems for some applications that cannot handle 75 large inode numbers. If applications are in use which do 76 not handle inode numbers bigger than 32 bits, the ``inode32`` 77 option should be specified. 78 79 largeio or nolargeio (default) 80 If ``nolargeio`` is specified, the optimal I/O reported in 81 ``st_blksize`` by **stat(2)** will be as small as possible to allow 82 user applications to avoid inefficient read/modify/write 83 I/O. This is typically the page size of the machine, as 84 this is the granularity of the page cache. 85 86 If ``largeio`` is specified, a filesystem that was created with a 87 ``swidth`` specified will return the ``swidth`` value (in bytes) 88 in ``st_blksize``. If the filesystem does not have a ``swidth`` 89 specified but does specify an ``allocsize`` then ``allocsize`` 90 (in bytes) will be returned instead. Otherwise the behaviour 91 is the same as if ``nolargeio`` was specified. 92 93 logbufs=value 94 Set the number of in-memory log buffers. Valid numbers 95 range from 2-8 inclusive. 96 97 The default value is 8 buffers. 98 99 If the memory cost of 8 log buffers is too high on small 100 systems, then it may be reduced at some cost to performance 101 on metadata intensive workloads. The ``logbsize`` option below 102 controls the size of each buffer and so is also relevant to 103 this case. 104 105 lifetime (default) or nolifetime 106 Enable data placement based on write life time hints provided 107 by the user. This turns on co-allocation of data of similar 108 life times when statistically favorable to reduce garbage 109 collection cost. 110 111 These options are only available for zoned rt file systems. 112 113 logbsize=value 114 Set the size of each in-memory log buffer. The size may be 115 specified in bytes, or in kilobytes with a "k" suffix. 116 Valid sizes for version 1 and version 2 logs are 16384 (16k) 117 and 32768 (32k). Valid sizes for version 2 logs also 118 include 65536 (64k), 131072 (128k) and 262144 (256k). The 119 logbsize must be an integer multiple of the log 120 stripe unit configured at **mkfs(8)** time. 121 122 The default value for version 1 logs is 32768, while the 123 default value for version 2 logs is MAX(32768, log_sunit). 124 125 logdev=device and rtdev=device 126 Use an external log (metadata journal) and/or real-time device. 127 An XFS filesystem has up to three parts: a data section, a log 128 section, and a real-time section. The real-time section is 129 optional, and the log section can be separate from the data 130 section or contained within it. 131 132 max_atomic_write=value 133 Set the maximum size of an atomic write. The size may be 134 specified in bytes, in kilobytes with a "k" suffix, in megabytes 135 with a "m" suffix, or in gigabytes with a "g" suffix. The size 136 cannot be larger than the maximum write size, larger than the 137 size of any allocation group, or larger than the size of a 138 remapping operation that the log can complete atomically. 139 140 The default value is to set the maximum I/O completion size 141 to allow each CPU to handle one at a time. 142 143 max_open_zones=value 144 Specify the max number of zones to keep open for writing on a 145 zoned rt device. Many open zones aids file data separation 146 but may impact performance on HDDs. 147 148 If ``max_open_zones`` is not specified, the value is determined 149 by the capabilities and the size of the zoned rt device. 150 151 noalign 152 Data allocations will not be aligned at stripe unit 153 boundaries. This is only relevant to filesystems created 154 with non-zero data alignment parameters (``sunit``, ``swidth``) by 155 **mkfs(8)**. 156 157 norecovery 158 The filesystem will be mounted without running log recovery. 159 If the filesystem was not cleanly unmounted, it is likely to 160 be inconsistent when mounted in ``norecovery`` mode. 161 Some files or directories may not be accessible because of this. 162 Filesystems mounted ``norecovery`` must be mounted read-only or 163 the mount will fail. 164 165 nouuid 166 Don't check for double mounted file systems using the file 167 system ``uuid``. This is useful to mount LVM snapshot volumes, 168 and often used in combination with ``norecovery`` for mounting 169 read-only snapshots. 170 171 noquota 172 Forcibly turns off all quota accounting and enforcement 173 within the filesystem. 174 175 uquota/usrquota/uqnoenforce/quota 176 User disk quota accounting enabled, and limits (optionally) 177 enforced. Refer to **xfs_quota(8)** for further details. 178 179 gquota/grpquota/gqnoenforce 180 Group disk quota accounting enabled and limits (optionally) 181 enforced. Refer to **xfs_quota(8)** for further details. 182 183 pquota/prjquota/pqnoenforce 184 Project disk quota accounting enabled and limits (optionally) 185 enforced. Refer to **xfs_quota(8)** for further details. 186 187 sunit=value and swidth=value 188 Used to specify the stripe unit and width for a RAID device 189 or a stripe volume. "value" must be specified in 512-byte 190 block units. These options are only relevant to filesystems 191 that were created with non-zero data alignment parameters. 192 193 The ``sunit`` and ``swidth`` parameters specified must be compatible 194 with the existing filesystem alignment characteristics. In 195 general, that means the only valid changes to ``sunit`` are 196 increasing it by a power-of-2 multiple. Valid ``swidth`` values 197 are any integer multiple of a valid ``sunit`` value. 198 199 Typically the only time these mount options are necessary if 200 after an underlying RAID device has had its geometry 201 modified, such as adding a new disk to a RAID5 lun and 202 reshaping it. 203 204 swalloc 205 Data allocations will be rounded up to stripe width boundaries 206 when the current end of file is being extended and the file 207 size is larger than the stripe width size. 208 209 wsync 210 When specified, all filesystem namespace operations are 211 executed synchronously. This ensures that when the namespace 212 operation (create, unlink, etc) completes, the change to the 213 namespace is on stable storage. This is useful in HA setups 214 where failover must not result in clients seeing 215 inconsistent namespace presentation during or after a 216 failover event. 217 218 errortag=tagname 219 When specified, enables the error inject tag named "tagname" with the 220 default frequency. Can be specified multiple times to enable multiple 221 errortags. Specifying this option on remount will reset the error tag 222 to the default value if it was set to any other value before. 223 This option is only supported when CONFIG_XFS_DEBUG is enabled, and 224 will not be reflected in /proc/self/mounts. 225 226Deprecation of V4 Format 227======================== 228 229The V4 filesystem format lacks certain features that are supported by 230the V5 format, such as metadata checksumming, strengthened metadata 231verification, and the ability to store timestamps past the year 2038. 232Because of this, the V4 format is deprecated. All users should upgrade 233by backing up their files, reformatting, and restoring from the backup. 234 235Administrators and users can detect a V4 filesystem by running xfs_info 236against a filesystem mountpoint and checking for a string containing 237"crc=". If no such string is found, please upgrade xfsprogs to the 238latest version and try again. 239 240The deprecation will take place in two parts. Support for mounting V4 241filesystems can now be disabled at kernel build time via Kconfig option. 242These options were changed to default to no in September 2025. In 243September 2030, support will be removed from the codebase entirely. 244 245Note: Distributors may choose to withdraw V4 format support earlier than 246the dates listed above. 247 248Deprecated Mount Options 249======================== 250 251============================ ================ 252 Name Removal Schedule 253============================ ================ 254Mounting with V4 filesystem September 2030 255Mounting ascii-ci filesystem September 2030 256============================ ================ 257 258 259Removed Mount Options 260===================== 261 262=========================== ======= 263 Name Removed 264=========================== ======= 265 delaylog/nodelaylog v4.0 266 ihashsize v4.0 267 irixsgid v4.0 268 osyncisdsync/osyncisosync v4.0 269 barrier v4.19 270 nobarrier v4.19 271 ikeep/noikeep v6.18 272 attr2/noattr2 v6.18 273=========================== ======= 274 275sysctls 276======= 277 278The following sysctls are available for the XFS filesystem: 279 280 fs.xfs.stats_clear (Min: 0 Default: 0 Max: 1) 281 Setting this to "1" clears accumulated XFS statistics 282 in /proc/fs/xfs/stat. It then immediately resets to "0". 283 284 fs.xfs.xfssyncd_centisecs (Min: 100 Default: 3000 Max: 720000) 285 The interval at which the filesystem flushes metadata 286 out to disk and runs internal cache cleanup routines. 287 288 fs.xfs.filestream_centisecs (Min: 1 Default: 3000 Max: 360000) 289 The interval at which the filesystem ages filestreams cache 290 references and returns timed-out AGs back to the free stream 291 pool. 292 293 fs.xfs.speculative_prealloc_lifetime 294 (Units: seconds Min: 1 Default: 300 Max: 86400) 295 The interval at which the background scanning for inodes 296 with unused speculative preallocation runs. The scan 297 removes unused preallocation from clean inodes and releases 298 the unused space back to the free pool. 299 300 fs.xfs.error_level (Min: 0 Default: 3 Max: 11) 301 A volume knob for error reporting when internal errors occur. 302 This will generate detailed messages & backtraces for filesystem 303 shutdowns, for example. Current threshold values are: 304 305 XFS_ERRLEVEL_OFF: 0 306 XFS_ERRLEVEL_LOW: 1 307 XFS_ERRLEVEL_HIGH: 5 308 309 fs.xfs.panic_mask (Min: 0 Default: 0 Max: 511) 310 Causes certain error conditions to call BUG(). Value is a bitmask; 311 OR together the tags which represent errors which should cause panics: 312 313 XFS_NO_PTAG 0 314 XFS_PTAG_IFLUSH 0x00000001 315 XFS_PTAG_LOGRES 0x00000002 316 XFS_PTAG_AILDELETE 0x00000004 317 XFS_PTAG_ERROR_REPORT 0x00000008 318 XFS_PTAG_SHUTDOWN_CORRUPT 0x00000010 319 XFS_PTAG_SHUTDOWN_IOERROR 0x00000020 320 XFS_PTAG_SHUTDOWN_LOGERROR 0x00000040 321 XFS_PTAG_FSBLOCK_ZERO 0x00000080 322 XFS_PTAG_VERIFIER_ERROR 0x00000100 323 324 This option is intended for debugging only. 325 326 fs.xfs.inherit_sync (Min: 0 Default: 1 Max: 1) 327 Setting this to "1" will cause the "sync" flag set 328 by the **xfs_io(8)** chattr command on a directory to be 329 inherited by files in that directory. 330 331 fs.xfs.inherit_nodump (Min: 0 Default: 1 Max: 1) 332 Setting this to "1" will cause the "nodump" flag set 333 by the **xfs_io(8)** chattr command on a directory to be 334 inherited by files in that directory. 335 336 fs.xfs.inherit_noatime (Min: 0 Default: 1 Max: 1) 337 Setting this to "1" will cause the "noatime" flag set 338 by the **xfs_io(8)** chattr command on a directory to be 339 inherited by files in that directory. 340 341 fs.xfs.inherit_nosymlinks (Min: 0 Default: 1 Max: 1) 342 Setting this to "1" will cause the "nosymlinks" flag set 343 by the **xfs_io(8)** chattr command on a directory to be 344 inherited by files in that directory. 345 346 fs.xfs.inherit_nodefrag (Min: 0 Default: 1 Max: 1) 347 Setting this to "1" will cause the "nodefrag" flag set 348 by the **xfs_io(8)** chattr command on a directory to be 349 inherited by files in that directory. 350 351 fs.xfs.rotorstep (Min: 1 Default: 1 Max: 256) 352 In "inode32" allocation mode, this option determines how many 353 files the allocator attempts to allocate in the same allocation 354 group before moving to the next allocation group. The intent 355 is to control the rate at which the allocator moves between 356 allocation groups when allocating extents for new files. 357 358Deprecated Sysctls 359================== 360 361None currently. 362 363Removed Sysctls 364=============== 365 366========================================== ======= 367 Name Removed 368========================================== ======= 369 fs.xfs.xfsbufd_centisec v4.0 370 fs.xfs.age_buffer_centisecs v4.0 371 fs.xfs.irix_symlink_mode v6.18 372 fs.xfs.irix_sgid_inherit v6.18 373 fs.xfs.speculative_cow_prealloc_lifetime v6.18 374========================================== ======= 375 376Error handling 377============== 378 379XFS can act differently according to the type of error found during its 380operation. The implementation introduces the following concepts to the error 381handler: 382 383 -failure speed: 384 Defines how fast XFS should propagate an error upwards when a specific 385 error is found during the filesystem operation. It can propagate 386 immediately, after a defined number of retries, after a set time period, 387 or simply retry forever. 388 389 -error classes: 390 Specifies the subsystem the error configuration will apply to, such as 391 metadata IO or memory allocation. Different subsystems will have 392 different error handlers for which behaviour can be configured. 393 394 -error handlers: 395 Defines the behavior for a specific error. 396 397The filesystem behavior during an error can be set via ``sysfs`` files. Each 398error handler works independently - the first condition met by an error handler 399for a specific class will cause the error to be propagated rather than reset and 400retried. 401 402The action taken by the filesystem when the error is propagated is context 403dependent - it may cause a shut down in the case of an unrecoverable error, 404it may be reported back to userspace, or it may even be ignored because 405there's nothing useful we can with the error or anyone we can report it to (e.g. 406during unmount). 407 408The configuration files are organized into the following hierarchy for each 409mounted filesystem: 410 411 /sys/fs/xfs/<dev>/error/<class>/<error>/ 412 413Where: 414 <dev> 415 The short device name of the mounted filesystem. This is the same device 416 name that shows up in XFS kernel error messages as "XFS(<dev>): ..." 417 418 <class> 419 The subsystem the error configuration belongs to. As of 4.9, the defined 420 classes are: 421 422 - "metadata": applies metadata buffer write IO 423 424 <error> 425 The individual error handler configurations. 426 427 428Each filesystem has "global" error configuration options defined in their top 429level directory: 430 431 /sys/fs/xfs/<dev>/error/ 432 433 fail_at_unmount (Min: 0 Default: 1 Max: 1) 434 Defines the filesystem error behavior at unmount time. 435 436 If set to a value of 1, XFS will override all other error configurations 437 during unmount and replace them with "immediate fail" characteristics. 438 i.e. no retries, no retry timeout. This will always allow unmount to 439 succeed when there are persistent errors present. 440 441 If set to 0, the configured retry behaviour will continue until all 442 retries and/or timeouts have been exhausted. This will delay unmount 443 completion when there are persistent errors, and it may prevent the 444 filesystem from ever unmounting fully in the case of "retry forever" 445 handler configurations. 446 447 Note: there is no guarantee that fail_at_unmount can be set while an 448 unmount is in progress. It is possible that the ``sysfs`` entries are 449 removed by the unmounting filesystem before a "retry forever" error 450 handler configuration causes unmount to hang, and hence the filesystem 451 must be configured appropriately before unmount begins to prevent 452 unmount hangs. 453 454Each filesystem has specific error class handlers that define the error 455propagation behaviour for specific errors. There is also a "default" error 456handler defined, which defines the behaviour for all errors that don't have 457specific handlers defined. Where multiple retry constraints are configured for 458a single error, the first retry configuration that expires will cause the error 459to be propagated. The handler configurations are found in the directory: 460 461 /sys/fs/xfs/<dev>/error/<class>/<error>/ 462 463 max_retries (Min: -1 Default: Varies Max: INTMAX) 464 Defines the allowed number of retries of a specific error before 465 the filesystem will propagate the error. The retry count for a given 466 error context (e.g. a specific metadata buffer) is reset every time 467 there is a successful completion of the operation. 468 469 Setting the value to "-1" will cause XFS to retry forever for this 470 specific error. 471 472 Setting the value to "0" will cause XFS to fail immediately when the 473 specific error is reported. 474 475 Setting the value to "N" (where 0 < N < Max) will make XFS retry the 476 operation "N" times before propagating the error. 477 478 retry_timeout_seconds (Min: -1 Default: Varies Max: 1 day) 479 Define the amount of time (in seconds) that the filesystem is 480 allowed to retry its operations when the specific error is 481 found. 482 483 Setting the value to "-1" will allow XFS to retry forever for this 484 specific error. 485 486 Setting the value to "0" will cause XFS to fail immediately when the 487 specific error is reported. 488 489 Setting the value to "N" (where 0 < N < Max) will allow XFS to retry the 490 operation for up to "N" seconds before propagating the error. 491 492**Note:** The default behaviour for a specific error handler is dependent on both 493the class and error context. For example, the default values for 494"metadata/ENODEV" are "0" rather than "-1" so that this error handler defaults 495to "fail immediately" behaviour. This is done because ENODEV is a fatal, 496unrecoverable error no matter how many times the metadata IO is retried. 497 498Workqueue Concurrency 499===================== 500 501XFS uses kernel workqueues to parallelize metadata update processes. This 502enables it to take advantage of storage hardware that can service many IO 503operations simultaneously. This interface exposes internal implementation 504details of XFS, and as such is explicitly not part of any userspace API/ABI 505guarantee the kernel may give userspace. These are undocumented features of 506the generic workqueue implementation XFS uses for concurrency, and they are 507provided here purely for diagnostic and tuning purposes and may change at any 508time in the future. 509 510The control knobs for a filesystem's workqueues are organized by task at hand 511and the short name of the data device. They all can be found in: 512 513 /sys/bus/workqueue/devices/${task}!${device} 514 515================ =========== 516 Task Description 517================ =========== 518 xfs_iwalk-$pid Inode scans of the entire filesystem. Currently limited to 519 mount time quotacheck. 520 xfs-gc Background garbage collection of disk space that have been 521 speculatively allocated beyond EOF or for staging copy on 522 write operations. 523================ =========== 524 525For example, the knobs for the quotacheck workqueue for /dev/nvme0n1 would be 526found in /sys/bus/workqueue/devices/xfs_iwalk-1111!nvme0n1/. 527 528The interesting knobs for XFS workqueues are as follows: 529 530============ =========== 531 Knob Description 532============ =========== 533 max_active Maximum number of background threads that can be started to 534 run the work. 535 cpumask CPUs upon which the threads are allowed to run. 536 nice Relative priority of scheduling the threads. These are the 537 same nice levels that can be applied to userspace processes. 538============ =========== 539 540Zoned Filesystems 541================= 542 543For zoned file systems, the following attributes are exposed in: 544 545 /sys/fs/xfs/<dev>/zoned/ 546 547 max_open_zones (Min: 1 Default: Varies Max: UINTMAX) 548 This read-only attribute exposes the maximum number of open zones 549 available for data placement. The value is determined at mount time and 550 is limited by the capabilities of the backing zoned device, file system 551 size and the max_open_zones mount option. 552 553 zonegc_low_space (Min: 0 Default: 0 Max: 100) 554 Define a percentage for how much of the unused space that GC should keep 555 available for writing. A high value will reclaim more of the space 556 occupied by unused blocks, creating a larger buffer against write 557 bursts at the cost of increased write amplification. Regardless 558 of this value, garbage collection will always aim to free a minimum 559 amount of blocks to keep max_open_zones open for data placement purposes. 560