1xfs_io(8) System Manager's Manual xfs_io(8)
2
6 xfs_io - debug the I/O path of an XFS filesystem
7
9 xfs_io [ -adfimrRstxT ] [ -c cmd ] ... [ -C cmd ] ... [ -p prog ] [
10 file ]
11 xfs_io -V
12
14 xfs_io is a debugging tool like xfs_db(8), but is aimed at examining
15 the regular file I/O paths rather than the raw XFS volume itself.
16 These code paths include not only the obvious read/write/mmap inter‐
17 faces for manipulating files, but also cover all of the XFS extensions
18 (such as space preallocation, additional inode flags, etc).
19
21 xfs_io commands may be run interactively (the default) or as arguments
22 on the command line. Interactive mode always runs commands on the cur‐
23 rent open file, whilst commands run from the command line may be
24 repeated on all open files rather than just the current open file. In
25 general, open file iteration will occur for commands that operate on
26 file content or state. In contrast, commands that operate on filesystem
27 or system-wide state will only be run on the current file regardless of
28 how many files are currently open. Multiple arguments may be given on
29 the command line and they are run in the sequence given. The program
30 exits one all commands have been run.
31 -c cmd Run the specified command on all currently open files. To
33 maintain compatibility with historical usage, commands that
34 can not be run on all open files will still be run but only
35 execute once on the current open file. Multiple -c arguments
36 may be given and may be interleaved on the command line in
37 any order with -C commands.
38 -C cmd Run the specified command only on the current open file.
40 Multiple -C arguments may be given and may be interleaved on
41 the command line in any order with -c commands.
42 -p prog Set the program name for prompts and some error messages, the
44 default value is xfs_io.
45 -f Create file if it does not already exist.
47 -r Open file read-only, initially. This is required if file is
49 immutable or append-only.
50 -i Start an idle thread. The purpose of this idle thread is to
52 test io from a multi threaded process. With single threaded
53 process, the file table is not shared and file structs are
54 not reference counted. Spawning an idle thread can help
55 detecting file struct reference leaks.
56 -x Expert mode. Dangerous commands are only available in this
58 mode. These commands also tend to require additional privi‐
59 leges.
60 -V Prints the version number and exits.
62 The other open(2) options described below are also available from the
64 command line.
65
67 xfs_io maintains a number of open files and memory mappings. Files can
68 be initially opened on the command line (optionally), and additional
69 files can also be opened later.
70 xfs_io commands can be broken up into three groups. Some commands are
72 aimed at doing regular file I/O - read, write, sync, space prealloca‐
73 tion, etc.
74 The second set of commands exist for manipulating memory mapped regions
76 of a file - mapping, accessing, storing, unmapping, flushing, etc.
77 The remaining commands are for the navigation and display of data
79 structures relating to the open files, mappings, and the filesystems
80 where they reside.
81 Many commands have extensive online help. Use the help command for more
83 details on any command.
84
86 file [ N ]
87 Display a list of all open files and (optionally) switch to an
88 alternate current open file.
89 open [[ -acdfrstRTPL ] path ]
91 Closes the current file, and opens the file specified by path
92 instead. Without any arguments, displays statistics about the
93 current file - see the stat command.
94 -a opens append-only (O_APPEND).
95 -d opens for direct I/O (O_DIRECT).
96 -f creates the file if it doesn't already exist (O_CREAT).
97 -r opens read-only (O_RDONLY).
98 -s opens for synchronous I/O (O_SYNC).
99 -t truncates on open (O_TRUNC).
100 -n opens in non-blocking mode if possible (O_NONBLOCK).
101 -T create a temporary file not linked into the filesystem
102 namespace (O_TMPFILE). The pathname passed must refer to
103 a directory which is treated as virtual parent for the
104 newly created invisible file. Can not be used together
105 with the -r option.
106 -R marks the file as a realtime XFS file after opening it,
107 if it is not already marked as such.
108 -P opens the path as a referent only (O_PATH). This is
109 incompatible with other flags specifying other O_xxx
110 flags apart from -L.
111 -L doesn't follow symlinks (O_NOFOLLOW). This is incompati‐
112 ble with other flags specifying other O_xxx flags apart
113 from -P.
114 o See the open command.
116 close Closes the current open file, marking the next open file as cur‐
118 rent (if one exists).
119 c See the close command.
121 chmod -r | -w
123 Change the mode of the currently open file. The -r option will
124 set the file permissions to read-only (0444), whilst the -w
125 option will set the file permissions to read-write (0644). This
126 allows xfs_io to set up mismatches between the file permissions
127 and the open file descriptor read/write mode to exercise permis‐
128 sion checks inside various syscalls.
129 pread [ -b bsize ] [ -v ] [ -FBR [ -Z seed ] ] [ -V vectors ] offset
131 length
132 Reads a range of bytes in a specified blocksize from the given
133 offset.
134 -b can be used to set the blocksize into which the read(2)
135 requests will be split. The default blocksize is 4096
136 bytes.
137 -v dump the contents of the buffer after reading, by default
138 only the count of bytes actually read is dumped.
139 -F read the buffers in a forwards sequential direction.
140 -B read the buffers in a reserve sequential direction.
141 -R read the buffers in the give range in a random order.
142 -Z seed
143 specify the random number seed used for random reads.
144 -V vectors
145 Use the vectored IO read syscall preadv(2) with a number
146 of blocksize length iovecs. The number of iovecs is set
147 by the vectors parameter.
148 r See the pread command.
150 pwrite [ -i file ] [ -dDwNOW ] [ -s skip ] [ -b size ] [ -S seed ] [
152 -FBR [ -Z zeed ] ] [ -V vectors ] offset length
153 Writes a range of bytes in a specified blocksize from the given
154 offset. The bytes written can be either a set pattern or read
155 in from another file before writing.
156 -i allows an input file to be specified as the source of the
157 data to be written.
158 -d causes direct I/O, rather than the usual buffered I/O, to
159 be used when reading the input file.
160 -w call fdatasync(2) once all writes are complete (included
161 in timing results)
162 -N Perform the pwritev2(2) call with RWF_NOWAIT.
163 -D Perform the pwritev2(2) call with RWF_DSYNC.
164 -O perform pwrite once and return the (maybe partial) bytes
165 written.
166 -W call fsync(2) once all writes are complete (included in
167 timing results)
168 -s specifies the number of bytes to skip from the start of
169 the input file before starting to read.
170 -b used to set the blocksize into which the write(2)
171 requests will be split. The default blocksize is 4096
172 bytes.
173 -S used to set the (repeated) fill pattern which is used
174 when the data to write is not coming from a file. The
175 default buffer fill pattern value is 0xcdcdcdcd.
176 -F write the buffers in a forwards sequential direction.
177 -B write the buffers in a reserve sequential direction.
178 -R write the buffers in the give range in a random order.
179 -Z seed
180 specify the random number seed used for random write
181 -V vectors
182 Use the vectored IO write syscall pwritev(2) with a num‐
183 ber of blocksize length iovecs. The number of iovecs is
184 set by the vectors parameter.
185 w See the pwrite command.
187 bmap [ -adelpv ] [ -n nx ]
189 Prints the block mapping for the current open file. Refer to the
190 xfs_bmap(8) manual page for complete documentation.
191 fiemap [ -alv ] [ -n nx ] [ offset [ len ]]
193 Prints the block mapping for the current open file using the
194 fiemap ioctl. Options behave as described in the xfs_bmap(8)
195 manual page.
196 Optionally, this command also supports passing the start offset
198 from where to begin the mapping and the length of that region.
199 The kernel will return any full extents which intersect with the
200 requested range, and the fiemap command will print them in their
201 entirety. If the requested range starts or ends in a hole,
202 fiemap will print the hole, truncated to the requested range.
203 extsize [ -R | -D ] [ value ]
205 Display and/or modify the preferred extent size used when allo‐
206 cating space for the currently open file. If the -R option is
207 specified, a recursive descent is performed for all directory
208 entries below the currently open file (-D can be used to
209 restrict the output to directories only). If the target file is
210 a directory, then the inherited extent size is set for that
211 directory (new files created in that directory inherit that
212 extent size). The value should be specified in bytes, or using
213 one of the usual units suffixes (k, m, g, b, etc). The extent
214 size is always reported in units of bytes.
215 cowextsize [ -R | -D ] [ value ]
217 Display and/or modify the preferred copy-on-write extent size
218 used when allocating space for the currently open file. If the
219 -R option is specified, a recursive descent is performed for all
220 directory entries below the currently open file (-D can be used
221 to restrict the output to directories only). If the target file
222 is a directory, then the inherited CoW extent size is set for
223 that directory (new files created in that directory inherit that
224 CoW extent size). The value should be specified in bytes, or
225 using one of the usual units suffixes (k, m, g, b, etc). The
226 extent size is always reported in units of bytes.
227 allocsp size 0
229 Sets the size of the file to size and zeroes any additional
230 space allocated using the XFS_IOC_ALLOCSP/XFS_IOC_FREESP system
231 call described in the xfsctl(3) manual page. allocsp and freesp
232 do exactly the same thing.
233 freesp size 0
235 See the allocsp command.
236 fadvise [ -r | -s | [[ -d | -n | -w ] offset length ]]
238 On platforms which support it, allows hints be given to the sys‐
239 tem regarding the expected I/O patterns on the file. The range
240 arguments are required by some advise commands ([*] below), and
241 the others must have no range arguments. With no arguments, the
242 POSIX_FADV_NORMAL advice is implied (default readahead).
243 -d the data will not be accessed again in the near future
244 (POSIX_FADV_DONTNEED[*]).
245 -n data will be accessed once and not be reused
246 (POSIX_FADV_NOREUSE[*]).
247 -r expect access to data in random order (POSIX_FADV_RAN‐
248 DOM), which sets readahead to zero.
249 -s expect access to data in sequential order
250 (POSIX_FADV_SEQUENTIAL), which doubles the default reada‐
251 head on the file.
252 -w advises the specified data will be needed again
253 (POSIX_FADV_WILLNEED[*]) which forces the maximum reada‐
254 head.
255 fdatasync
257 Calls fdatasync(2) to flush the file's in-core data to disk.
258 fsync Calls fsync(2) to flush all in-core file state to disk.
260 s See the fsync command.
262 sync_range [ -a | -b | -w ] offset length
264 On platforms which support it, allows control of syncing a range
265 of the file to disk. With no options, SYNC_FILE_RANGE_WRITE is
266 implied on the range supplied.
267 -a wait for IO in the given range to finish after writing
268 (SYNC_FILE_RANGE_WAIT_AFTER).
269 -b wait for IO in the given range to finish before writing
270 (SYNC_FILE_RANGE_WAIT_BEFORE).
271 -w start writeback of dirty data in the given range
272 (SYNC_FILE_RANGE_WRITE).
273 sync Calls sync(2) to flush all filesystems' in-core data to disk.
275 syncfs Calls syncfs(2) to flush this filesystem's in-core data to disk.
277 resvsp offset length
279 Allocates reserved, unwritten space for part of a file using the
280 XFS_IOC_RESVSP system call described in the xfsctl(3) manual
281 page.
282 unresvsp offset length
284 Frees reserved space for part of a file using the
285 XFS_IOC_UNRESVSP system call described in the xfsctl(3) manual
286 page.
287 falloc [ -k ] offset length
289 Allocates reserved, unwritten space for part of a file using the
290 fallocate routine as described in the fallocate(2) manual page.
291 -k will set the FALLOC_FL_KEEP_SIZE flag as described in
292 fallocate(2).
293 fcollapse offset length
295 Call fallocate with FALLOC_FL_COLLAPSE_RANGE flag as described
296 in the fallocate(2) manual page to de-allocates blocks and elim‐
297 inates the hole created in this process by shifting data blocks
298 into the hole.
299 finsert offset length
301 Call fallocate with FALLOC_FL_INSERT_RANGE flag as described in
302 the fallocate(2) manual page to create the hole by shifting data
303 blocks.
304 fpunch offset length
306 Punches (de-allocates) blocks in the file by calling fallocate
307 with the FALLOC_FL_PUNCH_HOLE flag as described in the fallo‐
308 cate(2) manual page.
309 funshare offset length
311 Call fallocate with FALLOC_FL_UNSHARE_RANGE flag as described in
312 the fallocate(2) manual page to unshare all shared blocks within
313 the range.
314 fzero [ -k ] offset length
316 Call fallocate with FALLOC_FL_ZERO_RANGE flag as described in
317 the fallocate(2) manual page to allocate and zero blocks within
318 the range. With the -k option, use the FALLOC_FL_KEEP_SIZE flag
319 as well.
320 zero offset length
322 Call xfsctl with XFS_IOC_ZERO_RANGE as described in the
323 xfsctl(3) manual page to allocate and zero blocks within the
324 range.
325 truncate offset
327 Truncates the current file at the given offset using ftrun‐
328 cate(2).
329 sendfile -i srcfile | -f N [ offset length ]
331 On platforms which support it, allows a direct in-kernel copy
332 between two file descriptors. The current open file is the tar‐
333 get, the source must be specified as another open file (-f) or
334 by path (-i).
335 readdir [ -v ] [ -o offset ] [ -l length ]
337 Read a range of directory entries from a given offset of a
338 directory.
339 -v verbose mode - dump dirent content as defined in read‐
340 dir(3)
341 -o specify starting offset
342 -l specify total length to read (in bytes)
343 seek -a | -d | -h [ -r ] [ -s ] offset
345 On platforms that support the lseek(2) SEEK_DATA and SEEK_HOLE
346 options, display the offsets of the specified segments.
347 -a Display both data and hole segments starting at the spec‐
348 ified offset.
349 -d Display the data segment starting at the specified off‐
350 set.
351 -h Display the hole segment starting at the specified off‐
352 set.
353 -r Recursively display all the specified segments starting
354 at the specified offset.
355 -s Display the starting lseek(2) offset. This offset will be
356 a calculated value when both data and holes are displayed
357 together or performing a recusively display.
358 reflink [ -C ] [ -q ] src_file [src_offset dst_offset length]
360 On filesystems that support the FICLONERANGE or
361 BTRFS_IOC_CLONE_RANGE ioctls, map length bytes at offset
362 dst_offset in the open file to the same physical blocks that are
363 mapped at offset src_offset in the file src_file , replacing any
364 contents that may already have been there. If a program writes
365 into a reflinked block range of either file, the dirty blocks
366 will be cloned, written to, and remapped ("copy on write") in
367 the affected file, leaving the other file(s) unchanged. If
368 src_offset, dst_offset, and length are omitted, all contents of
369 src_file will be reflinked into the open file.
370 -C Print timing statistics in a condensed format.
371 -q Do not print timing statistics at all.
372 dedupe [ -C ] [ -q ] src_file src_offset dst_offset length
374 On filesystems that support the FIDEDUPERANGE or
375 BTRFS_IOC_FILE_EXTENT_SAME ioctls, map length bytes at offset
376 dst_offset in the open file to the same physical blocks that are
377 mapped at offset src_offset in the file src_file , but only if
378 the contents of both ranges are identical. This is known as
379 block-based deduplication. If a program writes into a reflinked
380 block range of either file, the dirty blocks will be cloned,
381 written to, and remapped ("copy on write") in the affected file,
382 leaving the other file(s) unchanged.
383 -C Print timing statistics in a condensed format.
384 -q Do not print timing statistics at all.
385 copy_range [ -s src_offset ] [ -d dst_offset ] [ -l length ] src_file |
387 -f N
388 On filesystems that support the copy_file_range(2) system call,
389 copies data from the source file into the current open file.
390 The source must be specified either by path (src_file) or as
391 another open file (-f). If length is not specified, this com‐
392 mand copies data from src_offset to the end of src_file into the
393 dst_file at dst_offset.
394 -s Copy data from src_file beginning from src_offset.
395 -d Copy data into the open file beginning at dst_offset.
396 -l Copy up to length bytes of data.
397 swapext donor_file
399 Swaps extent forks between files. The current open file is the
400 target. The donor file is specified by path. Note that file data
401 is not copied (file content moves with the fork(s)).
402 set_encpolicy [ -c mode ] [ -n mode ] [ -f flags ] [ -v version ] [
404 keyspec ]
405 On filesystems that support encryption, assign an encryption
406 policy to the current file. keyspec is a hex string which spec‐
407 ifies the encryption key to use. For v1 encryption policies,
408 keyspec must be a 16-character hex string (8 bytes). For v2
409 policies, keyspec must be a 32-character hex string (16 bytes).
410 If unspecified, an all-zeroes value is used.
411 -c mode
412 contents encryption mode (e.g. AES-256-XTS)
413 -n mode
414 filenames encryption mode (e.g. AES-256-CTS)
415 -f flags
416 policy flags (numeric)
417 -v version
418 policy version. Defaults to 1 or 2 depending on the
419 length of keyspec; or to 1 if keyspec is unspecified.
420 get_encpolicy [ -1 ] [ -t ]
422 On filesystems that support encryption, display the encryption
423 policy of the current file.
424 -1 Use only the old ioctl to get the encryption policy.
425 This only works if the file has a v1 encryption policy.
426 -t Test whether v2 encryption policies are supported.
427 Prints "supported", "unsupported", or an error message.
428 add_enckey [ -d descriptor ]
430 On filesystems that support encryption, add an encryption key to
431 the filesystem containing the currently open file. The key in
432 binary (typically 64 bytes long) is read from standard input.
433 -d descriptor
434 key descriptor, as a 16-character hex string (8 bytes).
435 If given, the key will be available for use by v1 encryp‐
436 tion policies that use this descriptor. Otherwise, the
437 key is added as a v2 policy key, and on success the
438 resulting "key identifier" will be printed.
439 rm_enckey [ -a ] keyspec
441 On filesystems that support encryption, remove an encryption key
442 from the filesystem containing the currently open file. keyspec
443 is a hex string specifying the key to remove, as a 16-character
444 "key descriptor" or a 32-character "key identifier".
445 -a Remove the key for all users who have added it, not just
446 the current user. This is a privileged operation.
447 enckey_status keyspec
449 On filesystems that support encryption, display the status of an
450 encryption key. keyspec is a hex string specifying the key for
451 which to display the status, as a 16-character "key descriptor"
452 or a 32-character "key identifier".
453 lsattr [ -R | -D | -a | -v ]
455 List extended inode flags on the currently open file. If the -R
456 option is specified, a recursive descent is performed for all
457 directory entries below the currently open file (-D can be used
458 to restrict the output to directories only). This is a depth
459 first descent, it does not follow symlinks and it also does not
460 cross mount points.
461 chattr [ -R | -D ] [ +/-riasAdtPneEfSxC ]
463 Change extended inode flags on the currently open file. The -R
464 and -D options have the same meaning as above. The mapping
465 between each letter and the inode flags (refer to xfsctl(3) for
466 the full list) is available via the help command.
467 flink path
469 Link the currently open file descriptor into the filesystem
470 namespace.
471 stat [ -v|-r ]
473 Selected statistics from stat(2) and the XFS_IOC_GETXATTR system
474 call on the current file. If the -v option is specified, the
475 atime (last access), mtime (last modify), and ctime (last
476 change) timestamps are also displayed. The -r option dumps raw
477 fields from the stat structure.
478 statx [ -v|-r ][ -m basic | -m all | -m <mask> ][ -FD ]
480 Selected statistics from stat(2) and the XFS_IOC_GETXATTR system
481 call on the current file.
482 -v Show timestamps.
483 -r Dump raw statx structure values.
484 -m basic
485 Set the field mask for the statx call to
486 STATX_BASIC_STATS.
487 -m all
488 Set the the field mask for the statx call to STATX_ALL
489 (default).
490 -m <mask>
491 Specify a numeric field mask for the statx call.
492 -F Force the attributes to be synced with the server.
493 -D Don't sync attributes with the server.
494 chproj [ -R|-D ]
496 Modifies the project identifier associated with the current
497 path. The -R option will recursively descend if the current path
498 is a directory. The -D option will also recursively descend,
499 only setting modifying projects on subdirectories. See the
500 xfs_quota(8) manual page for more information about project
501 identifiers.
502 lsproj [ -R|-D ]
504 Displays the project identifier associated with the current
505 path. The -R and -D options behave as described above, in
506 chproj.
507 parent [ -cpv ]
509 By default this command prints out the parent inode numbers,
510 inode generation numbers and basenames of all the hardlinks
511 which point to the inode of the current file.
512 -p the output is similar to the default output except path‐
513 names up to the mount-point are printed out instead of
514 the component name.
515 -c the file's filesystem will check all the parent
516 attributes for consistency.
517 -v verbose output will be printed.
518 [NOTE: Not currently operational on Linux.]
519 utimes atime_sec atime_nsec mtime_sec mtime_nsec
521 The utimes command changes the atime and mtime of the current
522 file. sec uses UNIX timestamp notation and is the seconds
523 elapsed since 1970-01-01 00:00:00 UTC. nsec is the nanoseconds
524 since the sec. This value needs to be in the range 0-999999999
525 with UTIME_NOW and UTIME_OMIT being exceptions. Each (sec,
526 nsec) pair constitutes a single timestamp value.
527
531 mmap [ N | [[ -rwxS ] [-s size ] offset length ]]
532 With no arguments, mmap shows the current mappings. Specifying a
533 single numeric argument N sets the current mapping. If two argu‐
534 ments are specified (a range specified by offset and length), a
535 new mapping is created spanning the range, and the protection
536 mode can be given as a combination of PROT_READ (-r), PROT_WRITE
537 (-w), and PROT_EXEC (-x). The mapping will be created with the
538 MAP_SHARED flag by default, or with the Linux specific (MAP_SYNC
539 | MAP_SHARED_VALIDATE) flags if -S is given. -s size is used to
540 do a mmap(size) && munmap(size) operation at first, try to
541 reserve some extendible free memory space, if size is bigger
542 than length parameter. But there's not guarantee that the memory
543 after length ( up to size ) will stay free. e.g. "mmap -rw -s
544 8192 1024" will mmap 0 ~ 1024 bytes memory, but try to reserve
545 1024 ~ 8192 free space(no guarantee). This free space will help‐
546 ful for "mremap 8192" without MREMAP_MAYMOVE flag.
547 mm See the mmap command.
549 mremap [ -f <new_address> ] [ -m ] new_length
551 Changes the current mapping size to new_length. Whether the
552 mapping may be moved is controlled by the flags passed;
553 MREMAP_FIXED (-f), or MREMAP_MAYMOVE (-m). new_length specifies
554 a page-aligned address to which the mapping must be moved. It
555 can be set to 139946004389888, 4096k or 1g etc.
556 mrm See the mremap command.
558 munmap Unmaps the current memory mapping.
560 mu See the munmap command.
562 mread [ -f | -v ] [ -r ] [ offset length ]
564 Accesses a segment of the current memory mapping, optionally
565 dumping it to the standard output stream (with -v or -f option)
566 for inspection. The accesses are performed sequentially from the
567 start offset by default, but can also be done from the end back‐
568 wards through the mapping if the -r option in specified. The
569 two verbose modes differ only in the relative offsets they dis‐
570 play, the -f option is relative to file start, whereas -v shows
571 offsets relative to the start of the mapping.
572 mr See the mread command.
574 mwrite [ -r ] [ -S seed ] [ offset length ]
576 Stores a byte into memory for a range within a mapping. The
577 default stored value is 'X', repeated to fill the range speci‐
578 fied, but this can be changed using the -S option. The memory
579 stores are performed sequentially from the start offset by
580 default, but can also be done from the end backwards through the
581 mapping if the -r option in specified.
582 mw See the mwrite command.
584 msync [ -i ] [ -a | -s ] [ offset length ]
586 Writes all modified copies of pages over the specified range (or
587 entire mapping if no range specified) to their backing storage
588 locations. Also, optionally invalidates (-i) so that subsequent
589 references to the pages will be obtained from their backing
590 storage locations (instead of cached copies). The flush can be
591 done synchronously (-s) or asynchronously (-a).
592 ms See the msync command.
594 madvise [ -d | -r | -s | -w ] [ offset length ]
596 Modifies page cache behavior when operating on the current map‐
597 ping. The range arguments are required by some advise commands
598 ([*] below). With no arguments, the POSIX_MADV_NORMAL advice is
599 implied (default readahead).
600 -d the pages will not be needed (POSIX_MADV_DONTNEED[*]).
601 -r expect random page references (POSIX_MADV_RANDOM), which
602 sets readahead to zero.
603 -s expect sequential page references (POSIX_MADV_SEQUEN‐
604 TIAL), which doubles the default readahead on the file.
605 -w advises the specified pages will be needed again
606 (POSIX_MADV_WILLNEED[*]) which forces the maximum reada‐
607 head.
608 mincore
610 Dumps a list of pages or ranges of pages that are currently in
611 core, for the current memory mapping.
612
615 bulkstat [ -a agno ] [ -d ] [ -e endino ] [ -n batchsize ] [ -s
616 startino ] [ -v version"]
617 Display raw stat information about a bunch of inodes in an XFS
618 filesystem. Options are as follows:
619 -a agno
620 Display only results from the given allocation group.
621 If not specified, all results returned will be dis‐
622 played.
623 -d Print debugging information about call results.
624 -e endino
625 Stop displaying records when this inode number is
626 reached. Defaults to stopping when the system call
627 stops returning results.
628 -n batchsize
629 Retrieve at most this many records per call. Defaults
630 to 4,096.
631 -s startino
632 Display inode allocation records starting with this
633 inode. Defaults to the first inode in the filesystem.
634 If the given inode is not allocated, results will
635 begin with the next allocated inode in the filesystem.
636 -v version
637 Use a particular version of the kernel interface.
638 Currently supported versions are 1 and 5.
639 bulkstat_single [ -d ] [ -v version ] [ inum... | special... ]
641 Display raw stat information about individual inodes in an XFS
642 filesystem. The -d and -v options are the same as the bulkstat
643 command. Arguments must be inode numbers or any of the special
644 values:
645 root Display information about the root directory inode.
646 freeze Suspend all write I/O requests to the filesystem of the current
648 file. Only available in expert mode and requires privileges.
649 thaw Undo the effects of a filesystem freeze operation. Only avail‐
651 able in expert mode and requires privileges.
652 inject [ tag ]
654 Inject errors into a filesystem to observe filesystem behavior
655 at specific points under adverse conditions. Without the tag
656 argument, displays the list of error tags available. Only
657 available in expert mode and requires privileges.
658 resblks [ blocks ]
660 Get and/or set count of reserved filesystem blocks using the
661 XFS_IOC_GET_RESBLKS or XFS_IOC_SET_RESBLKS system calls. Note
662 -- this can be useful for exercising out of space behavior.
663 Only available in expert mode and requires privileges.
664 shutdown [ -f ]
666 Force the filesystem to shut down, preventing any further IO.
667 XFS and other filesystems implement this functionality, although
668 implementation details may differ slightly. Only available in
669 expert mode and requires privileges.
670 By default, the filesystem will not attempt to flush completed
672 transactions to disk before shutting down the filesystem. This
673 simulates a disk failure or crash.
674 -f Force the filesystem to flush all completed transactions
675 to disk before shutting down, matching XFS behavior when
676 critical corruption is encountered.
677 statfs Selected statistics from statfs(2) and the XFS_IOC_FSGEOMETRY
679 system call on the filesystem where the current file resides.
680 inode [ [ -n ] number ] [ -v ]
682 The inode command queries physical information about an inode.
683 With no arguments, it will return 1 or 0, indicating whether or
684 not any inode numbers greater than 32 bits are currently in use
685 in the filesystem. If given an inode number as an argument, the
686 command will return the same inode number if it is in use, or 0
687 if not. With -n number , the next used inode number after this
688 number will be returned, or zero if the supplied inode number is
689 the highest one in use. With -v the command will also report the
690 number of bits (32 or 64) used by the inode number printed in
691 the result; if no inode number was specified on the command
692 line, the maximum possible inode number in the system will be
693 printed along with its size.
694 inumbers [ -a agno ] [ -d ] [ -e endino ] [ -n batchsize ] [ -s
696 startino ] [ -v version ]
697 Prints allocation information about groups of inodes in an XFS
698 filesystem. Callers can use this information to figure out
699 which inodes are allocated. Options are as follows:
700 -a agno
701 Display only results from the given allocation group.
702 If not specified, all results returned will be dis‐
703 played.
704 -d Print debugging information about call results.
705 -e endino
706 Stop displaying records when this inode number is
707 reached. Defaults to stopping when the system call
708 stops returning results.
709 -n batchsize
710 Retrieve at most this many records per call. Defaults
711 to 4,096.
712 -s startino
713 Display inode allocation records starting with this
714 inode. Defaults to the first inode in the filesystem.
715 If the given inode is not allocated, results will
716 begin with the next allocated inode in the filesystem.
717 -v version
718 Use a particular version of the kernel interface.
719 Currently supported versions are 1 and 5.
720 scrub type [ agnumber | ino gen ]
722 Scrub internal XFS filesystem metadata. The type parameter
723 specifies which type of metadata to scrub. For AG metadata, one
724 AG number must be specified. For file metadata, the scrub is
725 applied to the open file unless the inode number and generation
726 number are specified.
727 repair type [ agnumber | ino gen ]
729 Repair internal XFS filesystem metadata. The type parameter
730 specifies which type of metadata to repair. For AG metadata,
731 one AG number must be specified. For file metadata, the repair
732 is applied to the open file unless the inode number and genera‐
733 tion number are specified.
734 label [ -c | -s label ]
736 On filesystems that support online label manipulation, get, set,
737 or clear the filesystem label. With no options, print the cur‐
738 rent filesystem label. The -c option clears the filesystem
739 label by setting it to the null string. The -s label option
740 sets the filesystem label to label. If the label is longer than
741 the filesystem will accept, xfs_io will print an error message.
742 XFS filesystem labels can be at most 12 characters long.
743 fsmap [ -d | -l | -r ] [ -m | -v ] [ -n nx ] [ start ] [ end ]
745 Prints the mapping of disk blocks used by the filesystem hosting
746 the current file. The map lists each extent used by files,
747 allocation group metadata, journalling logs, and static filesys‐
748 tem metadata, as well as any regions that are unused. Each line
749 of the listings takes the following form:
750 extent: major:minor [startblock..endblock]: owner startoff‐
752 set..endoffset length
753 Static filesystem metadata, allocation group metadata, btrees,
755 journalling logs, and free space are marked by replacing the
756 startoffset..endoffset with the appropriate marker. All blocks,
757 offsets, and lengths are specified in units of 512-byte blocks,
758 no matter what the filesystem's block size is. The optional
759 start and end arguments can be used to constrain the output to a
760 particular range of disk blocks. If these two options are spec‐
761 ified, exactly one of -d, -l, or -r must also be set.
762 -d Display only extents from the data device. This
763 option only applies for XFS filesystems.
764 -l Display only extents from the external log device.
765 This option only applies to XFS filesystems.
766 -r Display only extents from the realtime device. This
767 option only applies to XFS filesystems.
768 -m Display results in a machine readable format (CSV).
769 This option is not compatible with the -v flag. The
770 columns of the output are: extent number, device
771 major, device minor, physical start, physical end,
772 owner, offset start, offset end, length. The start,
773 end, and length numbers are provided in units of 512b.
774 The owner field is a special string that takes the
775 form:
776 inode_%lld_data
778 for inode data.
779 inode_%lld_data_bmbt
780 for inode data extent maps.
781 inode_%lld_attr
782 for inode extended attribute data.
783 inode_%lld_attr_bmbt
784 for inode extended attribute extent maps.
785 special_%u:%u
786 for other filesystem metadata.
787 -n num_extents
790 If this option is given, fsmap obtains the extent list
791 of the file in groups of num_extents extents. In the
792 absence of -n, fsmap queries the system for extents in
793 groups of 131,072 records.
794 -v Shows verbose information. When this flag is speci‐
796 fied, additional AG specific information is appended
797 to each line in the following form:
798 agno (startagblock..endagblock) nblocks flags
800 A second -v option will print out the flags legend.
802 This option is not compatible with the -m flag.
803
807 help [ command ]
808 Display a brief description of one or all commands.
809 print Display a list of all open files and memory mapped regions. The
811 current file and current mapping are distinguishable from any
812 others.
813 p See the print command.
815 quit Exit xfs_io.
817 q See the quit command.
819 log_writes -d device -m mark
821 Create a mark named mark in the dm-log-writes log specified by
822 device. This is intended to be equivalent to the shell command:
823 dmsetup message device 0 mark mark
825 lw See the log_writes command.
827 crc32cselftest
829 Test the internal crc32c implementation to make sure that it
830 computes results correctly.
831
833 mkfs.xfs(8), xfsctl(3), xfs_bmap(8), xfs_db(8), xfs(5), fdatasync(2),
834 fstat(2), fstatfs(2), fsync(2), ftruncate(2), futimens(3), mmap(2),
835 msync(2), open(2), pread(2), pwrite(2), readdir(3), dmsetup(8).
836 xfs_io(8)