{
    "mode": "man",
    "parameter": "xfs",
    "section": "5",
    "url": "https://www.chedong.com/phpMan.php/man/xfs/5/json",
    "generated": "2026-11-08T02:31:59Z",
    "sections": {
        "NAME": {
            "content": "xfs - layout, mount options, and supported file attributes for the XFS filesystem\n",
            "subsections": []
        },
        "DESCRIPTION": {
            "content": "An  XFS  filesystem  can  reside  on a regular disk partition or on a logical volume.  An XFS\nfilesystem has up to three parts: a data section, a log section, and a realtime section.  Us‐\ning the default mkfs.xfs(8) options, the realtime section is absent, and the log area is con‐\ntained within the data section.  The log section can be either separate from the data section\nor contained within it.  The filesystem sections are divided into a certain number of blocks,\nwhose size is specified at mkfs.xfs(8) time with the -b option.\n\nThe data section contains all the filesystem metadata (inodes, directories, indirect  blocks)\nas  well  as the user file data for ordinary (non-realtime) files and the log area if the log\nis internal to the data section.  The data section is divided into  a  number  of  allocation\ngroups.  The number and size of the allocation groups are chosen by mkfs.xfs(8) so that there\nis  normally  a small number of equal-sized groups.  The number of allocation groups controls\nthe amount of parallelism available in file and block allocation.   It  should  be  increased\nfrom  the default if there is sufficient memory and a lot of allocation activity.  The number\nof allocation groups should not be set very high, since this can cause large amounts  of  CPU\ntime to be used by the filesystem, especially when the filesystem is nearly full.  More allo‐\ncation groups are added (of the original size) when xfsgrowfs(8) is run.\n\nThe  log section (or area, if it is internal to the data section) is used to store changes to\nfilesystem metadata while the filesystem is running until those changes are made to the  data\nsection.   It  is  written  sequentially  during normal operation and read only during mount.\nWhen mounting a filesystem after a crash, the log is read to complete operations that were in\nprogress at the time of the crash.\n\nThe realtime section is used to store the data of realtime files.  These  files  had  an  at‐\ntribute  bit  set  through  xfsctl(3) after file creation, before any data was written to the\nfile.  The realtime section is divided into a number of extents of fixed size  (specified  at\nmkfs.xfs(8)  time).   Each file in the realtime section has an extent size that is a multiple\nof the realtime section extent size.\n\nEach allocation group contains several data structures.  The first sector  contains  the  su‐\nperblock.   For  allocation  groups after the first, the superblock is just a copy and is not\nupdated after mkfs.xfs(8).  The next three sectors contain information for  block  and  inode\nallocation within the allocation group.  Also contained within each allocation group are data\nstructures to locate free blocks and inodes; these are located through the header structures.\n\nEach XFS filesystem is labeled with a Universal Unique Identifier (UUID).  The UUID is stored\nin  every allocation group header and is used to help distinguish one XFS filesystem from an‐\nother, therefore you should avoid using dd(1) or other  block-by-block  copying  programs  to\ncopy  XFS  filesystems.   If two XFS filesystems on the same machine have the same UUID, xfs‐\ndump(8) may become confused when doing incremental and resumed dumps.  xfsdump(8) and  xfsre‐\nstore(8) are recommended for making copies of XFS filesystems.\n",
            "subsections": []
        },
        "OPERATIONS": {
            "content": "Some  functionality  specific to the XFS filesystem is accessible to applications through the\nxfsctl(3) and by-handle (see openbyhandle(3)) interfaces.\n",
            "subsections": []
        },
        "MOUNT OPTIONS": {
            "content": "The following XFS-specific mount options may be used when mounting an XFS  filesystem.  Other\ngeneric options may be used as well; refer to the mount(8) manual page for more details.\n",
            "subsections": [
                {
                    "name": "allocsize=size",
                    "content": "Sets  the  buffered  I/O  end-of-file preallocation size when doing delayed allocation\nwriteout. Valid values for this option are page size (typically 4KiB) through to 1GiB,\ninclusive, in power-of-2 increments.\n\nThe default behavior is for dynamic end-of-file preallocation size, which uses  a  set\nof  heuristics to optimise the preallocation size based on the current allocation pat‐\nterns within the file and the access patterns to the file. Specifying a  fixed  alloc‐\nsize value turns off the dynamic behavior.\n\nattr2|noattr2\nNote:  These  options have been deprecated as of kernel v5.10; The noattr2 option will\nbe removed no earlier than in September 2025 and attr2 option will  be  immutable  de‐\nfault.\n\nThe options enable/disable an \"opportunistic\" improvement to be made in the way inline\nextended  attributes are stored on-disk.  When the new form is used for the first time\nwhen attr2 is selected (either when setting or removing extended attributes)  the  on-\ndisk superblock feature bit field will be updated to reflect this format being in use.\n\nThe  default  behavior  is determined by the on-disk feature bit indicating that attr2\nbehavior is active. If either mount option it set, then that becomes the  new  default\nused by the filesystem.\n\nCRC  enabled  filesystems  always use the attr2 format, and so will reject the noattr2\nmount option if it is set.\n"
                },
                {
                    "name": "dax=value",
                    "content": "Set CPU direct access (DAX) behavior for the current filesystem. This mount option ac‐\ncepts the following values:\n\n\"dax=inode\" DAX will be enabled only on regular files with FSXFLAGDAX applied.\n\n\"dax=never\" DAX will not be enabled for any files. FSXFLAGDAX will be ignored.\n\n\"dax=always\" DAX will be enabled for all regular files, regardless of the FSXFLAGDAX\nstate.\n\nIf no option is used when mounting a  filesystem  stored  on  a  DAX  capable  device,\ndax=inode will be used as default.\n\nFor  details  regarding DAX behavior in kernel, please refer to kernel's documentation\nat filesystems/dax.txt\n\ndiscard|nodiscard\nEnable/disable the issuing of commands to let the block device reclaim space freed  by\nthe  filesystem.   This is useful for SSD devices, thinly provisioned LUNs and virtual\nmachine images, but may have a performance impact.\n\nNote: It is currently recommended that you use the fstrim application to  discard  un‐\nused  blocks  rather  than  the discard mount option because the performance impact of\nthis option is quite severe.  For this reason, nodiscard is the default.\n\ngrpid|bsdgroups|nogrpid|sysvgroups\nThese options define what group ID a newly created file gets.  When grpid is  set,  it\ntakes the group ID of the directory in which it is created; otherwise it takes the fs‐\ngid of the current process, unless the directory has the setgid bit set, in which case\nit  takes the gid from the parent directory, and also gets the setgid bit set if it is\na directory itself.\n"
                },
                {
                    "name": "filestreams",
                    "content": "Make the data allocator use the filestreams allocation mode across the entire filesys‐\ntem rather than just on directories configured to use it.\n\nikeep|noikeep\nNote: These options have been deprecated as of kernel v5.10; The noikeep  option  will\nbe  removed  no  earlier than in September 2025 and ikeep option will be immutable de‐\nfault.\n\n\nWhen ikeep is specified, XFS does not delete  empty  inode  clusters  and  keeps  them\naround  on  disk.  When noikeep is specified, empty inode clusters are returned to the\nfree space pool.  noikeep is the default.\n\ninode32|inode64\nWhen inode32 is specified, it indicates that XFS limits inode  creation  to  locations\nwhich will not result in inode numbers with more than 32 bits of significance.\n\nWhen  inode64  is  specified, it indicates that XFS is allowed to create inodes at any\nlocation in the filesystem, including those which will result in inode numbers occupy‐\ning more than 32 bits of significance.\n\ninode32 is provided for backwards compatibility with older systems  and  applications,\nsince  64  bits  inode  numbers might cause problems for some applications that cannot\nhandle large inode numbers.  If applications are in use which do not handle inode num‐\nbers bigger than 32 bits, the inode32 option should be specified.\n\nFor kernel v3.7 and later, inode64 is the default.\n\nlargeio|nolargeio\nIf \"nolargeio\" is specified, the optimal I/O reported in stblksize by stat(2) will be\nas small as possible to allow user applications to avoid inefficient read/modify/write\nI/O.  This is typically the page size of the machine, as this is  the  granularity  of\nthe page cache.\n\nIf  \"largeio\"  specified, a filesystem that was created with a \"swidth\" specified will\nreturn the \"swidth\" value (in bytes) in stblksize. If the filesystem does not have  a\n\"swidth\" specified but does specify an \"allocsize\" then \"allocsize\" (in bytes) will be\nreturned  instead. Otherwise the behavior is the same as if \"nolargeio\" was specified.\nnolargeio is the default.\n"
                },
                {
                    "name": "logbufs=value",
                    "content": "Set the number of in-memory log buffers.  Valid numbers range from 2–8 inclusive.\n\nThe default value is 8 buffers.\n\nIf the memory cost of 8 log buffers is too high on small systems, then it may  be  re‐\nduced at some cost to performance on metadata intensive workloads. The logbsize option\nbelow controls the size of each buffer and so is also relevant to this case.\n"
                },
                {
                    "name": "logbsize=value",
                    "content": "Set  the size of each in-memory log buffer.  The size may be specified in bytes, or in\nkibibytes (KiB) with a \"k\" suffix.  Valid sizes for version 1 and version 2  logs  are\n16384  (value=16k) and 32768 (value=32k).  Valid sizes for version 2 logs also include\n65536 (value=64k), 131072 (value=128k) and 262144 (value=256k). The logbsize  must  be\nan integer multiple of the log stripe unit configured at mkfs time.\n\nThe  default  value for version 1 logs is 32768, while the default value for version 2\nlogs is max(32768, logsunit).\n\nlogdev=device and rtdev=device\nUse an external log (metadata journal) and/or real-time device.  An XFS filesystem has\nup to three parts: a data section, a log section, and a real-time section.  The  real-\ntime section is optional, and the log section can be separate from the data section or\ncontained within it.\n"
                },
                {
                    "name": "noalign",
                    "content": "Data  allocations will not be aligned at stripe unit boundaries. This is only relevant\nto filesystems created with non-zero data  alignment  parameters  (sunit,  swidth)  by\nmkfs.\n"
                },
                {
                    "name": "norecovery",
                    "content": "The  filesystem  will  be mounted without running log recovery.  If the filesystem was\nnot cleanly unmounted, it is likely to be inconsistent when  mounted  in  \"norecovery\"\nmode.   Some  files or directories may not be accessible because of this.  Filesystems\nmounted \"norecovery\" must be mounted read-only or the mount will fail.\n\nnouuid Don't check for double mounted file systems using the file system uuid.  This is  use‐\nful to mount LVM snapshot volumes, and often used in combination with \"norecovery\" for\nmounting read-only snapshots.\n"
                },
                {
                    "name": "noquota",
                    "content": "Forcibly turns off all quota accounting and enforcement within the filesystem.\n"
                },
                {
                    "name": "uquota/usrquota/quota/uqnoenforce/qnoenforce",
                    "content": "User  disk  quota  accounting  enabled,  and  limits  (optionally) enforced.  Refer to\nxfsquota(8) for further details.\n"
                },
                {
                    "name": "gquota/grpquota/gqnoenforce",
                    "content": "Group disk quota accounting  enabled  and  limits  (optionally)  enforced.   Refer  to\nxfsquota(8) for further details.\n"
                },
                {
                    "name": "pquota/prjquota/pqnoenforce",
                    "content": "Project  disk  quota  accounting  enabled  and limits (optionally) enforced.  Refer to\nxfsquota(8) for further details.\n\nsunit=value and swidth=value\nUsed to specify the stripe unit and width for  a  RAID  device  or  a  stripe  volume.\n\"value\"  must be specified in 512-byte block units. These options are only relevant to\nfilesystems that were created with non-zero data alignment parameters.\n\nThe sunit and swidth  parameters  specified  must  be  compatible  with  the  existing\nfilesystem  alignment  characteristics.  In general, that means the only valid changes\nto sunit are increasing it by a power-of-2 multiple. Valid swidth values are any inte‐\nger multiple of a valid sunit value.\n\nTypically the only time these mount options are necessary if after an underlying  RAID\ndevice  has  had  it's geometry modified, such as adding a new disk to a RAID5 lun and\nreshaping it.\n"
                },
                {
                    "name": "swalloc",
                    "content": "Data allocations will be rounded up to stripe width boundaries when the current end of\nfile is being extended and the file size is larger than the stripe width size.\n\nwsync  When specified, all filesystem namespace operations are executed  synchronously.  This\nensures  that when the namespace operation (create, unlink, etc) completes, the change\nto the namespace is on stable storage. This is useful in HA setups where failover must\nnot result in clients seeing inconsistent namespace presentation  during  or  after  a\nfailover event.\n"
                }
            ]
        },
        "REMOVED MOUNT OPTIONS": {
            "content": "The  following mount options have been removed from the kernel, and will yield mount failures\nif specified.  Mount options are deprecated for a significant period time prior to removal.\nName                        Removed\n----                        -------\ndelaylog/nodelaylog         v4.0\nihashsize                   v4.0\nirixsgid                    v4.0\nosyncisdsync/osyncisosync   v4.0\nbarrier/nobarrier           v4.19\n",
            "subsections": []
        },
        "FILE ATTRIBUTES": {
            "content": "The XFS filesystem supports setting the following file attributes on Linux systems using  the\nchattr(1) utility:\n\na - append only\n\nA - no atime updates\n\nd - no dump\n\ni - immutable\n\nS - synchronous updates\n\nFor descriptions of these attribute flags, please refer to the chattr(1) man page.\n",
            "subsections": []
        },
        "SEE ALSO": {
            "content": "chattr(1),  xfsctl(3),  mount(8),  mkfs.xfs(8), xfsinfo(8), xfsadmin(8), xfsdump(8), xfsre‐\nstore(8).\n\nxfs(5)",
            "subsections": []
        }
    },
    "summary": "xfs - layout, mount options, and supported file attributes for the XFS filesystem",
    "flags": [],
    "examples": [],
    "see_also": [
        {
            "name": "chattr",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/chattr/1/json"
        },
        {
            "name": "xfsctl",
            "section": "3",
            "url": "https://www.chedong.com/phpMan.php/man/xfsctl/3/json"
        },
        {
            "name": "mount",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/mount/8/json"
        },
        {
            "name": "mkfs.xfs",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/mkfs.xfs/8/json"
        },
        {
            "name": "xfsinfo",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/xfsinfo/8/json"
        },
        {
            "name": "xfsadmin",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/xfsadmin/8/json"
        },
        {
            "name": "xfsdump",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/xfsdump/8/json"
        },
        {
            "name": "store",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/store/8/json"
        }
    ]
}