Index: stable/10/cddl/contrib/opensolaris/cmd/zfs/zfs.8 =================================================================== --- stable/10/cddl/contrib/opensolaris/cmd/zfs/zfs.8 (revision 268646) +++ stable/10/cddl/contrib/opensolaris/cmd/zfs/zfs.8 (revision 268647) @@ -1,3509 +1,3556 @@ '\" te .\" Copyright (c) 2013, Martin Matuska . .\" All Rights Reserved. .\" .\" The contents of this file are subject to the terms of the .\" Common Development and Distribution License (the "License"). .\" You may not use this file except in compliance with the License. .\" .\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE .\" or http://www.opensolaris.org/os/licensing. .\" See the License for the specific language governing permissions .\" and limitations under the License. .\" .\" When distributing Covered Code, include this CDDL HEADER in each .\" file and include the License file at usr/src/OPENSOLARIS.LICENSE. .\" If applicable, add the following below this CDDL HEADER, with the .\" fields enclosed by brackets "[]" replaced with your own identifying .\" information: Portions Copyright [yyyy] [name of copyright owner] .\" .\" Copyright (c) 2010, Sun Microsystems, Inc. All Rights Reserved. -.\" Copyright (c) 2013 by Delphix. All rights reserved. +.\" Copyright (c) 2014 by Delphix. All rights reserved. .\" Copyright (c) 2011, Pawel Jakub Dawidek .\" Copyright (c) 2012, Glen Barber .\" Copyright (c) 2012, Bryan Drewery .\" Copyright (c) 2013 by Saso Kiselkov. All rights reserved. .\" Copyright (c) 2013 Nexenta Systems, Inc. All Rights Reserved. .\" Copyright (c) 2014, Joyent, Inc. All rights reserved. .\" Copyright (c) 2013, Steven Hartland .\" Copyright (c) 2014, Xin LI .\" .\" $FreeBSD$ .\" -.Dd April 23, 2014 +.Dd May 27, 2014 .Dt ZFS 8 .Os .Sh NAME .Nm zfs .Nd configures ZFS file systems .Sh SYNOPSIS .Nm .Op Fl \&? .Nm .Cm create .Op Fl pu .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... Ar filesystem .Nm .Cm create .Op Fl ps .Op Fl b Ar blocksize .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Fl V .Ar size volume .Nm .Cm destroy .Op Fl fnpRrv .Ar filesystem Ns | Ns Ar volume .Nm .Cm destroy .Op Fl dnpRrv .Sm off .Ar filesystem Ns | Ns volume .Ns @snap .Op % Ns Ar snap .Op , Ns Ar snap Op % Ns Ar snap .Op , Ns ... .Sm on .Nm .Cm destroy .Ar filesystem Ns | Ns Ar volume Ns # Ns Ar bookmark .Nm .Cm snapshot Ns | Ns Cm snap .Op Fl r .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Ar filesystem@snapname Ns | Ns Ar volume@snapname .Ar filesystem@snapname Ns | Ns Ar volume@snapname Ns ... .Nm .Cm rollback .Op Fl rRf .Ar snapshot .Nm .Cm clone .Op Fl p .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Ar snapshot filesystem Ns | Ns Ar volume .Nm .Cm promote .Ar clone-filesystem .Nm .Cm rename .Op Fl f .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Nm .Cm rename .Op Fl f .Fl p .Ar filesystem Ns | Ns Ar volume .Ar filesystem Ns | Ns Ar volume .Nm .Cm rename .Fl r .Ar snapshot snapshot .Nm .Cm rename .Fl u .Op Fl p .Ar filesystem filesystem .Nm .Cm list .Op Fl r Ns | Ns Fl d Ar depth .Op Fl Hp .Op Fl o Ar property Ns Oo , Ns property Ns Oc Ns ... .Op Fl t Ar type Ns Oo , Ns type Ns Oc Ns ... .Oo Fl s Ar property Oc Ns ... .Oo Fl S Ar property Oc Ns ... .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Nm .Cm set .Ar property Ns = Ns Ar value .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Nm .Cm get .Op Fl r Ns | Ns Fl d Ar depth .Op Fl Hp .Op Fl o Ar all | field Ns Oo , Ns Ar field Oc Ns ... .Op Fl t Ar type Ns Oo Ns , Ar type Oc Ns ... .Op Fl s Ar source Ns Oo Ns , Ns Ar source Oc Ns ... .Ar all | property Ns Oo Ns , Ns Ar property Oc Ns ... .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Nm .Cm inherit .Op Fl rS .Ar property .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Nm .Cm upgrade .Op Fl v .Nm .Cm upgrade .Op Fl r .Op Fl V Ar version .Fl a | Ar filesystem .Nm .Cm userspace .Op Fl Hinp .Op Fl o Ar field Ns Oo , Ns Ar field Oc Ns ... .Oo Fl s Ar field Oc Ns ... .Oo Fl S Ar field Oc Ns ... .Op Fl t Ar type Ns Oo Ns , Ns Ar type Oc Ns ... .Ar filesystem Ns | Ns Ar snapshot .Nm .Cm groupspace .Op Fl Hinp .Op Fl o Ar field Ns Oo , Ns field Oc Ns ... .Oo Fl s Ar field Oc Ns ... .Oo Fl S Ar field Oc Ns ... .Op Fl t Ar type Ns Oo Ns , Ns Ar type Oc Ns ... .Ar filesystem Ns | Ns Ar snapshot .Nm .Cm mount .Nm .Cm mount .Op Fl vO .Op Fl o Ar property Ns Oo , Ns Ar property Oc Ns ... .Fl a | Ar filesystem .Nm .Cm unmount Ns | Ns Cm umount .Op Fl f .Fl a | Ar filesystem Ns | Ns Ar mountpoint .Nm .Cm share .Fl a | Ar filesystem .Nm .Cm unshare .Fl a | Ar filesystem Ns | Ns Ar mountpoint .Nm .Cm bookmark .Ar snapshot .Ar bookmark .Nm .Cm send .Op Fl DnPpRv .Op Fl i Ar snapshot | Fl I Ar snapshot .Ar snapshot .Nm .Cm send .Op Fl i Ar snapshot Ns | Ns bookmark .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Nm .Cm receive Ns | Ns Cm recv .Op Fl vnFu .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Nm .Cm receive Ns | Ns Cm recv .Op Fl vnFu .Op Fl d | e .Ar filesystem .Nm .Cm allow .Ar filesystem Ns | Ns Ar volume .Nm .Cm allow .Op Fl ldug .Ar user Ns | Ns Ar group Ns Oo Ns , Ns Ar user Ns | Ns Ar group Oc Ns ... .Ar perm Ns | Ns Ar @setname Ns .Oo Ns , Ns Ar perm Ns | Ns Ar @setname Oc Ns ... .Ar filesystem Ns | Ns Ar volume .Nm .Cm allow .Op Fl ld .Fl e Ns | Ns Cm everyone .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Nm .Cm allow .Fl c .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Nm .Cm allow .Fl s .Ar @setname .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Nm .Cm unallow .Op Fl rldug .Ar user Ns | Ns Ar group Ns Oo Ns , Ns Ar user Ns | Ns Ar group Oc Ns ... .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Nm .Cm unallow .Op Fl rld .Fl e Ns | Ns Cm everyone .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Nm .Cm unallow .Op Fl r .Fl c .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Nm .Cm unallow .Op Fl r .Fl s .Ar @setname .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Nm .Cm hold .Op Fl r .Ar tag snapshot Ns ... .Nm .Cm holds .Op Fl r .Ar snapshot Ns ... .Nm .Cm release .Op Fl r .Ar tag snapshot Ns ... .Nm .Cm diff .Op Fl FHt .Ar snapshot .Op Ar snapshot Ns | Ns Ar filesystem .Nm .Cm jail .Ar jailid Ns | Ns Ar jailname filesystem .Nm .Cm unjail .Ar jailid Ns | Ns Ar jailname filesystem .Sh DESCRIPTION The .Nm command configures .Tn ZFS datasets within a .Tn ZFS storage pool, as described in .Xr zpool 8 . A dataset is identified by a unique path within the .Tn ZFS namespace. For example: .Bd -ragged -offset 4n .No pool/ Ns Brq filesystem,volume,snapshot .Ed .Pp where the maximum length of a dataset name is .Dv MAXNAMELEN (256 bytes). .Pp A dataset can be one of the following: .Bl -hang -width 12n .It Sy file system A .Tn ZFS dataset of type .Em filesystem can be mounted within the standard system namespace and behaves like other file systems. While .Tn ZFS file systems are designed to be .Tn POSIX compliant, known issues exist that prevent compliance in some cases. Applications that depend on standards conformance might fail due to nonstandard behavior when checking file system free space. .It Sy volume A logical volume exported as a raw or block device. This type of dataset should only be used under special circumstances. File systems are typically used in most environments. .It Sy snapshot A read-only version of a file system or volume at a given point in time. It is specified as .Em filesystem@name or .Em volume@name . .El .Ss ZFS File System Hierarchy A .Tn ZFS storage pool is a logical collection of devices that provide space for datasets. A storage pool is also the root of the .Tn ZFS file system hierarchy. .Pp The root of the pool can be accessed as a file system, such as mounting and unmounting, taking snapshots, and setting properties. The physical storage characteristics, however, are managed by the .Xr zpool 8 command. .Pp See .Xr zpool 8 for more information on creating and administering pools. .Ss Snapshots A snapshot is a read-only copy of a file system or volume. Snapshots can be created extremely quickly, and initially consume no additional space within the pool. As data within the active dataset changes, the snapshot consumes more data than would otherwise be shared with the active dataset. .Pp Snapshots can have arbitrary names. Snapshots of volumes can be cloned or rolled back, but cannot be accessed independently. .Pp File system snapshots can be accessed under the .Pa \&.zfs/snapshot directory in the root of the file system. Snapshots are automatically mounted on demand and may be unmounted at regular intervals. The visibility of the .Pa \&.zfs directory can be controlled by the .Sy snapdir property. .Ss Clones A clone is a writable volume or file system whose initial contents are the same as another dataset. As with snapshots, creating a clone is nearly instantaneous, and initially consumes no additional space. .Pp Clones can only be created from a snapshot. When a snapshot is cloned, it creates an implicit dependency between the parent and child. Even though the clone is created somewhere else in the dataset hierarchy, the original snapshot cannot be destroyed as long as a clone exists. The .Sy origin property exposes this dependency, and the .Cm destroy command lists any such dependencies, if they exist. .Pp The clone parent-child dependency relationship can be reversed by using the .Cm promote subcommand. This causes the "origin" file system to become a clone of the specified file system, which makes it possible to destroy the file system that the clone was created from. .Ss Mount Points Creating a .Tn ZFS file system is a simple operation, so the number of file systems per system is likely to be numerous. To cope with this, .Tn ZFS automatically manages mounting and unmounting file systems without the need to edit the .Pa /etc/fstab file. All automatically managed file systems are mounted by .Tn ZFS at boot time. .Pp By default, file systems are mounted under .Pa /path , where .Ar path is the name of the file system in the .Tn ZFS namespace. Directories are created and destroyed as needed. .Pp A file system can also have a mount point set in the .Sy mountpoint property. This directory is created as needed, and .Tn ZFS automatically mounts the file system when the .Qq Nm Cm mount Fl a command is invoked (without editing .Pa /etc/fstab ) . The .Sy mountpoint property can be inherited, so if .Em pool/home has a mount point of .Pa /home , then .Em pool/home/user automatically inherits a mount point of .Pa /home/user . .Pp A file system .Sy mountpoint property of .Cm none prevents the file system from being mounted. .Pp If needed, .Tn ZFS file systems can also be managed with traditional tools .Pq Xr mount 8 , Xr umount 8 , Xr fstab 5 . If a file system's mount point is set to .Cm legacy , .Tn ZFS makes no attempt to manage the file system, and the administrator is responsible for mounting and unmounting the file system. .Ss Jails .No A Tn ZFS dataset can be attached to a jail by using the .Qq Nm Cm jail subcommand. You cannot attach a dataset to one jail and the children of the same dataset to another jails. To allow management of the dataset from within a jail, the .Sy jailed property has to be set and the jail needs access to the .Pa /dev/zfs device. The .Sy quota property cannot be changed from within a jail. See .Xr jail 8 for information on how to allow mounting .Tn ZFS datasets from within a jail. .Pp .No A Tn ZFS dataset can be detached from a jail using the .Qq Nm Cm unjail subcommand. .Pp After a dataset is attached to a jail and the jailed property is set, a jailed file system cannot be mounted outside the jail, since the jail administrator might have set the mount point to an unacceptable value. .Ss Deduplication Deduplication is the process for removing redundant data at the block-level, reducing the total amount of data stored. If a file system has the .Cm dedup property enabled, duplicate data blocks are removed synchronously. The result is that only unique data is stored and common components are shared among files. .Ss Native Properties Properties are divided into two types, native properties and user-defined (or "user") properties. Native properties either export internal statistics or control .Tn ZFS behavior. In addition, native properties are either editable or read-only. User properties have no effect on .Tn ZFS behavior, but you can use them to annotate datasets in a way that is meaningful in your environment. For more information about user properties, see the .Qq Sx User Properties section, below. .Pp Every dataset has a set of properties that export statistics about the dataset as well as control various behaviors. Properties are inherited from the parent unless overridden by the child. Some properties apply only to certain types of datasets (file systems, volumes, or snapshots). .Pp The values of numeric properties can be specified using human-readable suffixes (for example, .Sy k , KB , M , Gb , and so forth, up to .Sy Z for zettabyte). The following are all valid (and equal) specifications: .Bd -ragged -offset 4n 1536M, 1.5g, 1.50GB .Ed .Pp The values of non-numeric properties are case sensitive and must be lowercase, except for .Sy mountpoint , sharenfs , No and Sy sharesmb . .Pp The following native properties consist of read-only statistics about the dataset. These properties can be neither set, nor inherited. Native properties apply to all dataset types unless otherwise noted. .Bl -tag -width 2n .It Sy available The amount of space available to the dataset and all its children, assuming that there is no other activity in the pool. Because space is shared within a pool, availability can be limited by any number of factors, including physical pool size, quotas, reservations, or other datasets within the pool. .Pp This property can also be referred to by its shortened column name, .Sy avail . .It Sy compressratio For non-snapshots, the compression ratio achieved for the .Sy used space of this dataset, expressed as a multiplier. The .Sy used property includes descendant datasets, and, for clones, does not include the space shared with the origin snapshot. For snapshots, the .Sy compressratio is the same as the .Sy refcompressratio property. Compression can be turned on by running: .Qq Nm Cm set compression=on Ar dataset The default value is .Cm off . .It Sy creation The time this dataset was created. .It Sy clones For snapshots, this property is a comma-separated list of filesystems or volumes which are clones of this snapshot. The clones' .Sy origin property is this snapshot. If the .Sy clones property is not empty, then this snapshot can not be destroyed (even with the .Fl r or .Fl f options). .It Sy defer_destroy This property is .Cm on if the snapshot has been marked for deferred destroy by using the .Qq Nm Cm destroy -d command. Otherwise, the property is .Cm off . .It Sy filesystem_count The total number of filesystems and volumes that exist under this location in the dataset tree. This value is only available when a .Sy filesystem_limit has been set somewhere in the tree under which the dataset resides. .It Sy logicalreferenced The amount of space that is .Qq logically accessible by this dataset. See the .Sy referenced property. The logical space ignores the effect of the .Sy compression and .Sy copies properties, giving a quantity closer to the amount of data that applications see. However, it does include space consumed by metadata. .Pp This property can also be referred to by its shortened column name, .Sy lrefer . .It Sy logicalused The amount of space that is .Qq logically consumed by this dataset and all its descendents. See the .Sy used property. The logical space ignores the effect of the .Sy compression and .Sy copies properties, giving a quantity closer to the amount of data that applications see. .Pp This property can also be referred to by its shortened column name, .Sy lused . .It Sy mounted For file systems, indicates whether the file system is currently mounted. This property can be either .Cm yes or .Cm no . .It Sy origin For cloned file systems or volumes, the snapshot from which the clone was created. See also the .Sy clones property. .It Sy referenced The amount of data that is accessible by this dataset, which may or may not be shared with other datasets in the pool. When a snapshot or clone is created, it initially references the same amount of space as the file system or snapshot it was created from, since its contents are identical. .Pp This property can also be referred to by its shortened column name, .Sy refer . .It Sy refcompressratio The compression ratio achieved for the .Sy referenced space of this dataset, expressed as a multiplier. See also the .Sy compressratio property. .It Sy snapshot_count The total number of snapshots that exist under this location in the dataset tree. This value is only available when a .Sy snapshot_limit has been set somewhere in the tree under which the dataset resides. .It Sy type The type of dataset: .Sy filesystem , volume , No or Sy snapshot . .It Sy used The amount of space consumed by this dataset and all its descendents. This is the value that is checked against this dataset's quota and reservation. The space used does not include this dataset's reservation, but does take into account the reservations of any descendent datasets. The amount of space that a dataset consumes from its parent, as well as the amount of space that are freed if this dataset is recursively destroyed, is the greater of its space used and its reservation. .Pp When snapshots (see the .Qq Sx Snapshots section) are created, their space is initially shared between the snapshot and the file system, and possibly with previous snapshots. As the file system changes, space that was previously shared becomes unique to the snapshot, and counted in the snapshot's space used. Additionally, deleting snapshots can increase the amount of space unique to (and used by) other snapshots. .Pp The amount of space used, available, or referenced does not take into account pending changes. Pending changes are generally accounted for within a few seconds. Committing a change to a disk using .Xr fsync 2 or .Sy O_SYNC does not necessarily guarantee that the space usage information is updated immediately. .It Sy usedby* The .Sy usedby* properties decompose the .Sy used properties into the various reasons that space is used. Specifically, .Sy used No = .Sy usedbysnapshots + usedbydataset + usedbychildren + usedbyrefreservation . These properties are only available for datasets created with .Tn ZFS pool version 13 pools and higher. .It Sy usedbysnapshots The amount of space consumed by snapshots of this dataset. In particular, it is the amount of space that would be freed if all of this dataset's snapshots were destroyed. Note that this is not simply the sum of the snapshots' .Sy used properties because space can be shared by multiple snapshots. .It Sy usedbydataset The amount of space used by this dataset itself, which would be freed if the dataset were destroyed (after first removing any .Sy refreservation and destroying any necessary snapshots or descendents). .It Sy usedbychildren The amount of space used by children of this dataset, which would be freed if all the dataset's children were destroyed. .It Sy usedbyrefreservation The amount of space used by a .Sy refreservation set on this dataset, which would be freed if the .Sy refreservation was removed. .It Sy userused@ Ns Ar user The amount of space consumed by the specified user in this dataset. Space is charged to the owner of each file, as displayed by .Qq Nm ls Fl l . The amount of space charged is displayed by .Qq Nm du and .Qq Nm ls Fl s . See the .Qq Nm Cm userspace subcommand for more information. .Pp Unprivileged users can access only their own space usage. The root user, or a user who has been granted the .Sy userused privilege with .Qq Nm Cm allow , can access everyone's usage. .Pp The .Sy userused@ Ns ... properties are not displayed by .Qq Nm Cm get all . The user's name must be appended after the .Sy @ symbol, using one of the following forms: .Bl -bullet -offset 2n .It POSIX name (for example, .Em joe ) .It POSIX numeric ID (for example, .Em 1001 ) .El .It Sy userrefs This property is set to the number of user holds on this snapshot. User holds are set by using the .Qq Nm Cm hold command. .It Sy groupused@ Ns Ar group The amount of space consumed by the specified group in this dataset. Space is charged to the group of each file, as displayed by .Nm ls Fl l . See the .Sy userused@ Ns Ar user property for more information. .Pp Unprivileged users can only access their own groups' space usage. The root user, or a user who has been granted the .Sy groupused privilege with .Qq Nm Cm allow , can access all groups' usage. .It Sy volblocksize Ns = Ns Ar blocksize For volumes, specifies the block size of the volume. The .Ar blocksize cannot be changed once the volume has been written, so it should be set at volume creation time. The default .Ar blocksize for volumes is 8 Kbytes. Any power of 2 from 512 bytes to 128 Kbytes is valid. .Pp This property can also be referred to by its shortened column name, .Sy volblock . .It Sy written The amount of .Sy referenced space written to this dataset since the previous snapshot. .It Sy written@ Ns Ar snapshot The amount of .Sy referenced space written to this dataset since the specified snapshot. This is the space that is referenced by this dataset but was not referenced by the specified snapshot. .Pp The .Ar snapshot may be specified as a short snapshot name (just the part after the .Sy @ ) , in which case it will be interpreted as a snapshot in the same filesystem as this dataset. The .Ar snapshot may be a full snapshot name .Pq Em filesystem@snapshot , which for clones may be a snapshot in the origin's filesystem (or the origin of the origin's filesystem, etc). .El .Pp The following native properties can be used to change the behavior of a .Tn ZFS dataset. .Bl -tag -width 2n .It Xo .Sy aclinherit Ns = Ns Cm discard | .Cm noallow | .Cm restricted | .Cm passthrough | .Cm passthrough-x .Xc Controls how .Tn ACL entries are inherited when files and directories are created. A file system with an .Sy aclinherit property of .Cm discard does not inherit any .Tn ACL entries. A file system with an .Sy aclinherit property value of .Cm noallow only inherits inheritable .Tn ACL entries that specify "deny" permissions. The property value .Cm restricted (the default) removes the .Em write_acl and .Em write_owner permissions when the .Tn ACL entry is inherited. A file system with an .Sy aclinherit property value of .Cm passthrough inherits all inheritable .Tn ACL entries without any modifications made to the .Tn ACL entries when they are inherited. A file system with an .Sy aclinherit property value of .Cm passthrough-x has the same meaning as .Cm passthrough , except that the .Em owner@ , group@ , No and Em everyone@ Tn ACE Ns s inherit the execute permission only if the file creation mode also requests the execute bit. .Pp When the property value is set to .Cm passthrough , files are created with a mode determined by the inheritable .Tn ACE Ns s. If no inheritable .Tn ACE Ns s exist that affect the mode, then the mode is set in accordance to the requested mode from the application. .It Sy aclmode Ns = Ns Cm discard | groupmask | passthrough | restricted Controls how an .Tn ACL is modified during .Xr chmod 2 . A file system with an .Sy aclmode property of .Cm discard (the default) deletes all .Tn ACL entries that do not represent the mode of the file. An .Sy aclmode property of .Cm groupmask reduces permissions granted in all .Em ALLOW entries found in the .Tn ACL such that they are no greater than the group permissions specified by .Xr chmod 2 . A file system with an .Sy aclmode property of .Cm passthrough indicates that no changes are made to the .Tn ACL other than creating or updating the necessary .Tn ACL entries to represent the new mode of the file or directory. An .Sy aclmode property of .Cm restricted will cause the .Xr chmod 2 operation to return an error when used on any file or directory which has a non-trivial .Tn ACL whose entries can not be represented by a mode. .Xr chmod 2 is required to change the set user ID, set group ID, or sticky bits on a file or directory, as they do not have equivalent .Tn ACL entries. In order to use .Xr chmod 2 on a file or directory with a non-trivial .Tn ACL when .Sy aclmode is set to .Cm restricted , you must first remove all .Tn ACL entries which do not represent the current mode. .It Sy atime Ns = Ns Cm on | off Controls whether the access time for files is updated when they are read. Turning this property off avoids producing write traffic when reading files and can result in significant performance gains, though it might confuse mailers and other similar utilities. The default value is .Cm on . .It Sy canmount Ns = Ns Cm on | off | noauto If this property is set to .Cm off , the file system cannot be mounted, and is ignored by .Qq Nm Cm mount Fl a . Setting this property to .Cm off is similar to setting the .Sy mountpoint property to .Cm none , except that the dataset still has a normal .Sy mountpoint property, which can be inherited. Setting this property to .Cm off allows datasets to be used solely as a mechanism to inherit properties. One example of setting .Sy canmount Ns = Ns Cm off is to have two datasets with the same .Sy mountpoint , so that the children of both datasets appear in the same directory, but might have different inherited characteristics. .Pp When the .Cm noauto value is set, a dataset can only be mounted and unmounted explicitly. The dataset is not mounted automatically when the dataset is created or imported, nor is it mounted by the .Qq Nm Cm mount Fl a command or unmounted by the .Qq Nm Cm umount Fl a command. .Pp This property is not inherited. .It Sy checksum Ns = Ns Cm on | off | fletcher2 | fletcher4 | sha256 | noparity Controls the checksum used to verify data integrity. The default value is .Cm on , which automatically selects an appropriate algorithm (currently, .Cm fletcher4 , but this may change in future releases). The value .Cm off disables integrity checking on user data. The value .Cm noparity not only disables integrity but also disables maintaining parity for user data. This setting is used internally by a dump device residing on a RAID-Z pool and should not be used by any other dataset. Disabling checksums is .Em NOT a recommended practice. .It Sy compression Ns = Ns Cm on | off | lzjb | gzip | gzip- Ns Ar N | zle | Cm lz4 Controls the compression algorithm used for this dataset. The .Cm lzjb compression algorithm is optimized for performance while providing decent data compression. Setting compression to .Cm on uses the .Cm lzjb compression algorithm. The .Cm gzip compression algorithm uses the same compression as the .Xr gzip 1 command. You can specify the .Cm gzip level by using the value .Cm gzip- Ns Ar N where .Ar N is an integer from 1 (fastest) to 9 (best compression ratio). Currently, .Cm gzip is equivalent to .Cm gzip-6 (which is also the default for .Xr gzip 1 ) . The .Cm zle compression algorithm compresses runs of zeros. .Pp The .Sy lz4 compression algorithm is a high-performance replacement for the .Sy lzjb algorithm. It features significantly faster compression and decompression, as well as a moderately higher compression ratio than .Sy lzjb , but can only be used on pools with the .Sy lz4_compress feature set to .Sy enabled . See .Xr zpool-features 7 for details on ZFS feature flags and the .Sy lz4_compress feature. .Pp This property can also be referred to by its shortened column name .Cm compress . Changing this property affects only newly-written data. .It Sy copies Ns = Ns Cm 1 | 2 | 3 Controls the number of copies of data stored for this dataset. These copies are in addition to any redundancy provided by the pool, for example, mirroring or RAID-Z. The copies are stored on different disks, if possible. The space used by multiple copies is charged to the associated file and dataset, changing the .Sy used property and counting against quotas and reservations. .Pp Changing this property only affects newly-written data. Therefore, set this property at file system creation time by using the .Fl o Cm copies= Ns Ar N option. .It Sy dedup Ns = Ns Cm on | off | verify | sha256 Ns Op Cm ,verify Configures deduplication for a dataset. The default value is .Cm off . The default deduplication checksum is .Cm sha256 (this may change in the future). When .Sy dedup is enabled, the checksum defined here overrides the .Sy checksum property. Setting the value to .Cm verify has the same effect as the setting .Cm sha256,verify . .Pp If set to .Cm verify , .Tn ZFS will do a byte-to-byte comparsion in case of two blocks having the same signature to make sure the block contents are identical. .It Sy devices Ns = Ns Cm on | off The .Sy devices property is currently not supported on .Fx . .It Sy exec Ns = Ns Cm on | off Controls whether processes can be executed from within this file system. The default value is .Cm on . .It Sy mlslabel Ns = Ns Ar label | Cm none The .Sy mlslabel property is currently not supported on .Fx . .It Sy filesystem_limit Ns = Ns Ar count | Cm none Limits the number of filesystems and volumes that can exist under this point in the dataset tree. The limit is not enforced if the user is allowed to change the limit. Setting a .Sy filesystem_limit on a descendent of a filesystem that already has a .Sy filesystem_limit does not override the ancestor's .Sy filesystem_limit , but rather imposes an additional limit. This feature must be enabled to be used .Po see .Xr zpool-features 7 .Pc . .It Sy mountpoint Ns = Ns Ar path | Cm none | legacy Controls the mount point used for this file system. See the .Qq Sx Mount Points section for more information on how this property is used. .Pp When the .Sy mountpoint property is changed for a file system, the file system and any children that inherit the mount point are unmounted. If the new value is .Cm legacy , then they remain unmounted. Otherwise, they are automatically remounted in the new location if the property was previously .Cm legacy or .Cm none , or if they were mounted before the property was changed. In addition, any shared file systems are unshared and shared in the new location. .It Sy nbmand Ns = Ns Cm on | off The .Sy nbmand property is currently not supported on .Fx . .It Sy primarycache Ns = Ns Cm all | none | metadata Controls what is cached in the primary cache (ARC). If this property is set to .Cm all , then both user data and metadata is cached. If this property is set to .Cm none , then neither user data nor metadata is cached. If this property is set to .Cm metadata , then only metadata is cached. The default value is .Cm all . .It Sy quota Ns = Ns Ar size | Cm none Limits the amount of space a dataset and its descendents can consume. This property enforces a hard limit on the amount of space used. This includes all space consumed by descendents, including file systems and snapshots. Setting a quota on a descendent of a dataset that already has a quota does not override the ancestor's quota, but rather imposes an additional limit. .Pp Quotas cannot be set on volumes, as the .Sy volsize property acts as an implicit quota. .It Sy snapshot_limit Ns = Ns Ar count | Cm none Limits the number of snapshots that can be created on a dataset and its descendents. Setting a .Sy snapshot_limit on a descendent of a dataset that already has a .Sy snapshot_limit does not override the ancestor's .Sy snapshot_limit , but rather imposes an additional limit. The limit is not enforced if the user is allowed to change the limit. For example, this means that recursive snapshots taken from the global zone are counted against each delegated dataset within a jail. This feature must be enabled to be used .Po see .Xr zpool-features 7 .Pc . .It Sy userquota@ Ns Ar user Ns = Ns Ar size | Cm none Limits the amount of space consumed by the specified user. Similar to the .Sy refquota property, the .Sy userquota space calculation does not include space that is used by descendent datasets, such as snapshots and clones. User space consumption is identified by the .Sy userspace@ Ns Ar user property. .Pp Enforcement of user quotas may be delayed by several seconds. This delay means that a user might exceed their quota before the system notices that they are over quota and begins to refuse additional writes with the .Em EDQUOT error message. See the .Cm userspace subcommand for more information. .Pp Unprivileged users can only access their own groups' space usage. The root user, or a user who has been granted the .Sy userquota privilege with .Qq Nm Cm allow , can get and set everyone's quota. .Pp This property is not available on volumes, on file systems before version 4, or on pools before version 15. The .Sy userquota@ Ns ... properties are not displayed by .Qq Nm Cm get all . The user's name must be appended after the .Sy @ symbol, using one of the following forms: .Bl -bullet -offset 2n .It POSIX name (for example, .Em joe ) .It POSIX numeric ID (for example, .Em 1001 ) .El .It Sy groupquota@ Ns Ar group Ns = Ns Ar size | Cm none Limits the amount of space consumed by the specified group. Group space consumption is identified by the .Sy userquota@ Ns Ar user property. .Pp Unprivileged users can access only their own groups' space usage. The root user, or a user who has been granted the .Sy groupquota privilege with .Qq Nm Cm allow , can get and set all groups' quotas. .It Sy readonly Ns = Ns Cm on | off Controls whether this dataset can be modified. The default value is .Cm off . .It Sy recordsize Ns = Ns Ar size Specifies a suggested block size for files in the file system. This property is designed solely for use with database workloads that access files in fixed-size records. .Tn ZFS automatically tunes block sizes according to internal algorithms optimized for typical access patterns. .Pp For databases that create very large files but access them in small random chunks, these algorithms may be suboptimal. Specifying a .Sy recordsize greater than or equal to the record size of the database can result in significant performance gains. Use of this property for general purpose file systems is strongly discouraged, and may adversely affect performance. .Pp The size specified must be a power of two greater than or equal to 512 and less than or equal to 128 Kbytes. .Pp Changing the file system's .Sy recordsize affects only files created afterward; existing files are unaffected. .Pp This property can also be referred to by its shortened column name, .Sy recsize . +.It Sy redundant_metadata Ns = Ns Cm all | most +Controls what types of metadata are stored redundantly. +ZFS stores an extra copy of metadata, so that if a single block is corrupted, +the amount of user data lost is limited. +This extra copy is in addition to any redundancy provided at the pool level +.Pq e.g. by mirroring or RAID-Z , +and is in addition to an extra copy specified by the +.Sy copies +property +.Pq up to a total of 3 copies . +For example if the pool is mirrored, +.Cm copies Ns = Ns Ar 2 , +and +.Cm redundant_metadata Ns = Ns Ar most , +then ZFS +stores 6 copies of most metadata, and 4 copies of data and some +metadata. +.Pp +When set to +.Cm all , +ZFS stores an extra copy of all metadata. +If a +single on-disk block is corrupt, at worst a single block of user data +.Po which is +.Cm recordsize +bytes long +can be lost. +.Pc +.Pp +When set to +.Cm most , +ZFS stores an extra copy of most types of +metadata. +This can improve performance of random writes, because less +metadata must be written. +In practice, at worst about 100 blocks +.Po of +.Cm recordsize +bytes each +.Pc +of user data can be lost if a single +on-disk block is corrupt. +The exact behavior of which metadata blocks +are stored redundantly may change in future releases. +.Pp +The default value is +.Cm all . .It Sy refquota Ns = Ns Ar size | Cm none Limits the amount of space a dataset can consume. This property enforces a hard limit on the amount of space used. This hard limit does not include space used by descendents, including file systems and snapshots. .It Sy refreservation Ns = Ns Ar size | Cm none The minimum amount of space guaranteed to a dataset, not including its descendents. When the amount of space used is below this value, the dataset is treated as if it were taking up the amount of space specified by .Sy refreservation . The .Sy refreservation reservation is accounted for in the parent datasets' space used, and counts against the parent datasets' quotas and reservations. .Pp If .Sy refreservation is set, a snapshot is only allowed if there is enough free pool space outside of this reservation to accommodate the current number of "referenced" bytes in the dataset. .Pp This property can also be referred to by its shortened column name, .Sy refreserv . .It Sy reservation Ns = Ns Ar size | Cm none The minimum amount of space guaranteed to a dataset and its descendents. When the amount of space used is below this value, the dataset is treated as if it were taking up the amount of space specified by its reservation. Reservations are accounted for in the parent datasets' space used, and count against the parent datasets' quotas and reservations. .Pp This property can also be referred to by its shortened column name, .Sy reserv . .It Sy secondarycache Ns = Ns Cm all | none | metadata Controls what is cached in the secondary cache (L2ARC). If this property is set to .Cm all , then both user data and metadata is cached. If this property is set to .Cm none , then neither user data nor metadata is cached. If this property is set to .Cm metadata , then only metadata is cached. The default value is .Cm all . .It Sy setuid Ns = Ns Cm on | off Controls whether the .No set- Ns Tn UID bit is respected for the file system. The default value is .Cm on . .It Sy sharesmb Ns = Ns Cm on | off | Ar opts The .Sy sharesmb property currently has no effect on .Fx . .It Sy sharenfs Ns = Ns Cm on | off | Ar opts Controls whether the file system is shared via .Tn NFS , and what options are used. A file system with a .Sy sharenfs property of .Cm off is managed the traditional way via .Xr exports 5 . Otherwise, the file system is automatically shared and unshared with the .Qq Nm Cm share and .Qq Nm Cm unshare commands. If the property is set to .Cm on no .Tn NFS export options are used. Otherwise, .Tn NFS export options are equivalent to the contents of this property. The export options may be comma-separated. See .Xr exports 5 for a list of valid options. .Pp When the .Sy sharenfs property is changed for a dataset, the .Xr mountd 8 daemon is reloaded. .It Sy logbias Ns = Ns Cm latency | throughput Provide a hint to .Tn ZFS about handling of synchronous requests in this dataset. If .Sy logbias is set to .Cm latency (the default), .Tn ZFS will use pool log devices (if configured) to handle the requests at low latency. If .Sy logbias is set to .Cm throughput , .Tn ZFS will not use configured pool log devices. .Tn ZFS will instead optimize synchronous operations for global pool throughput and efficient use of resources. .It Sy snapdir Ns = Ns Cm hidden | visible Controls whether the .Pa \&.zfs directory is hidden or visible in the root of the file system as discussed in the .Qq Sx Snapshots section. The default value is .Cm hidden . .It Sy sync Ns = Ns Cm standard | always | disabled Controls the behavior of synchronous requests (e.g. .Xr fsync 2 , O_DSYNC). This property accepts the following values: .Bl -tag -offset 4n -width 8n .It Sy standard This is the POSIX specified behavior of ensuring all synchronous requests are written to stable storage and all devices are flushed to ensure data is not cached by device controllers (this is the default). .It Sy always All file system transactions are written and flushed before their system calls return. This has a large performance penalty. .It Sy disabled Disables synchronous requests. File system transactions are only committed to stable storage periodically. This option will give the highest performance. However, it is very dangerous as .Tn ZFS would be ignoring the synchronous transaction demands of applications such as databases or .Tn NFS . Administrators should only use this option when the risks are understood. .El .It Sy volsize Ns = Ns Ar size For volumes, specifies the logical size of the volume. By default, creating a volume establishes a reservation of equal size. For storage pools with a version number of 9 or higher, a .Sy refreservation is set instead. Any changes to .Sy volsize are reflected in an equivalent change to the reservation (or .Sy refreservation ) . The .Sy volsize can only be set to a multiple of .Cm volblocksize , and cannot be zero. .Pp The reservation is kept equal to the volume's logical size to prevent unexpected behavior for consumers. Without the reservation, the volume could run out of space, resulting in undefined behavior or data corruption, depending on how the volume is used. These effects can also occur when the volume size is changed while it is in use (particularly when shrinking the size). Extreme care should be used when adjusting the volume size. .Pp Though not recommended, a "sparse volume" (also known as "thin provisioning") can be created by specifying the .Fl s option to the .Qq Nm Cm create Fl V command, or by changing the reservation after the volume has been created. A "sparse volume" is a volume where the reservation is less then the volume size. Consequently, writes to a sparse volume can fail with .Sy ENOSPC when the pool is low on space. For a sparse volume, changes to .Sy volsize are not reflected in the reservation. .It Sy volmode Ns = Ns Cm default | geom | dev | none This property specifies how volumes should be exposed to the OS. Setting it to .Sy geom exposes volumes as .Xr geom 4 providers, providing maximal functionality. Setting it to .Sy dev exposes volumes only as cdev device in devfs. Such volumes can be accessed only as raw disk device files, i.e. they can not be partitioned, mounted, participate in RAIDs, etc, but they are faster, and in some use scenarios with untrusted consumer, such as NAS or VM storage, can be more safe. Volumes with property set to .Sy none are not exposed outside ZFS, but can be snapshoted, cloned, replicated, etc, that can be suitable for backup purposes. Value .Sy default means that volumes exposition is controlled by system-wide sysctl/tunable .Va vfs.zfs.vol.mode , where .Sy geom , .Sy dev and .Sy none are encoded as 1, 2 and 3 respectively. The default values is .Sy geom . This property can be changed any time, but so far it is processed only during volume creation and pool import. .It Sy vscan Ns = Ns Cm off | on The .Sy vscan property is currently not supported on .Fx . .It Sy xattr Ns = Ns Cm off | on The .Sy xattr property is currently not supported on .Fx . .It Sy jailed Ns = Ns Cm off | on Controls whether the dataset is managed from a jail. See the .Qq Sx Jails section for more information. The default value is .Cm off . .El .Pp The following three properties cannot be changed after the file system is created, and therefore, should be set when the file system is created. If the properties are not set with the .Qq Nm Cm create or .Nm zpool Cm create commands, these properties are inherited from the parent dataset. If the parent dataset lacks these properties due to having been created prior to these features being supported, the new file system will have the default values for these properties. .Bl -tag -width 4n .It Sy casesensitivity Ns = Ns Cm sensitive | insensitive | mixed Indicates whether the file name matching algorithm used by the file system should be case-sensitive, case-insensitive, or allow a combination of both styles of matching. The default value for the .Sy casesensitivity property is .Cm sensitive . Traditionally, UNIX and POSIX file systems have case-sensitive file names. .Pp The .Cm mixed value for the .Sy casesensitivity property indicates that the file system can support requests for both case-sensitive and case-insensitive matching behavior. .It Sy normalization Ns = Ns Cm none | formC | formD | formKC | formKD Indicates whether the file system should perform a .Sy unicode normalization of file names whenever two file names are compared, and which normalization algorithm should be used. File names are always stored unmodified, names are normalized as part of any comparison process. If this property is set to a legal value other than .Cm none , and the .Sy utf8only property was left unspecified, the .Sy utf8only property is automatically set to .Cm on . The default value of the .Sy normalization property is .Cm none . This property cannot be changed after the file system is created. .It Sy utf8only Ns = Ns Cm on | off Indicates whether the file system should reject file names that include characters that are not present in the .Sy UTF-8 character code set. If this property is explicitly set to .Cm off , the normalization property must either not be explicitly set or be set to .Cm none . The default value for the .Sy utf8only property is .Cm off . This property cannot be changed after the file system is created. .El .Pp The .Sy casesensitivity , normalization , No and Sy utf8only properties are also new permissions that can be assigned to non-privileged users by using the .Tn ZFS delegated administration feature. .Ss Temporary Mount Point Properties When a file system is mounted, either through .Xr mount 8 for legacy mounts or the .Qq Nm Cm mount command for normal file systems, its mount options are set according to its properties. The correlation between properties and mount options is as follows: .Bl -column -offset 4n "PROPERTY" "MOUNT OPTION" .It "PROPERTY MOUNT OPTION" .It "atime atime/noatime" .It "exec exec/noexec" .It "readonly ro/rw" .It "setuid suid/nosuid" .El .Pp In addition, these options can be set on a per-mount basis using the .Fl o option, without affecting the property that is stored on disk. The values specified on the command line override the values stored in the dataset. These properties are reported as "temporary" by the .Qq Nm Cm get command. If the properties are changed while the dataset is mounted, the new setting overrides any temporary settings. .Ss User Properties In addition to the standard native properties, .Tn ZFS supports arbitrary user properties. User properties have no effect on .Tn ZFS behavior, but applications or administrators can use them to annotate datasets (file systems, volumes, and snapshots). .Pp User property names must contain a colon .Pq Sy \&: character to distinguish them from native properties. They may contain lowercase letters, numbers, and the following punctuation characters: colon .Pq Sy \&: , dash .Pq Sy \&- , period .Pq Sy \&. and underscore .Pq Sy \&_ . The expected convention is that the property name is divided into two portions such as .Em module Ns Sy \&: Ns Em property , but this namespace is not enforced by .Tn ZFS . User property names can be at most 256 characters, and cannot begin with a dash .Pq Sy \&- . .Pp When making programmatic use of user properties, it is strongly suggested to use a reversed .Tn DNS domain name for the .Ar module component of property names to reduce the chance that two independently-developed packages use the same property name for different purposes. Property names beginning with .Em com.sun are reserved for use by Sun Microsystems. .Pp The values of user properties are arbitrary strings, are always inherited, and are never validated. All of the commands that operate on properties .Po .Qq Nm Cm list , .Qq Nm Cm get , .Qq Nm Cm set and so forth .Pc can be used to manipulate both native properties and user properties. Use the .Qq Nm Cm inherit command to clear a user property. If the property is not defined in any parent dataset, it is removed entirely. Property values are limited to 1024 characters. .Sh SUBCOMMANDS All subcommands that modify state are logged persistently to the pool in their original form. .Bl -tag -width 2n .It Xo .Nm .Op Fl \&? .Xc .Pp Displays a help message. .It Xo .Nm .Cm create .Op Fl pu .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Ar filesystem .Xc .Pp Creates a new .Tn ZFS file system. The file system is automatically mounted according to the .Sy mountpoint property inherited from the parent. .Bl -tag -width indent .It Fl p Creates all the non-existing parent datasets. Datasets created in this manner are automatically mounted according to the .Sy mountpoint property inherited from their parent. Any property specified on the command line using the .Fl o option is ignored. If the target filesystem already exists, the operation completes successfully. .It Fl u Newly created file system is not mounted. .It Fl o Ar property Ns = Ns Ar value Sets the specified property as if the command .Qq Nm Cm set Ar property Ns = Ns Ar value was invoked at the same time the dataset was created. Any editable .Tn ZFS property can also be set at creation time. Multiple .Fl o options can be specified. An error results if the same property is specified in multiple .Fl o options. .El .It Xo .Nm .Cm create .Op Fl ps .Op Fl b Ar blocksize .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Fl V .Ar size volume .Xc .Pp Creates a volume of the given size. The volume is exported as a block device in .Pa /dev/zvol/path , where .Ar path is the name of the volume in the .Tn ZFS namespace. The size represents the logical size as exported by the device. By default, a reservation of equal size is created. .Pp .Ar size is automatically rounded up to the nearest 128 Kbytes to ensure that the volume has an integral number of blocks regardless of .Ar blocksize . .Bl -tag -width indent .It Fl p Creates all the non-existing parent datasets. Datasets created in this manner are automatically mounted according to the .Sy mountpoint property inherited from their parent. Any property specified on the command line using the .Fl o option is ignored. If the target filesystem already exists, the operation completes successfully. .It Fl s Creates a sparse volume with no reservation. See .Sy volsize in the .Qq Sx Native Properties section for more information about sparse volumes. .It Fl b Ar blocksize Equivalent to .Fl o Cm volblocksize Ns = Ns Ar blocksize . If this option is specified in conjunction with .Fl o Cm volblocksize , the resulting behavior is undefined. .It Fl o Ar property Ns = Ns Ar value Sets the specified property as if the .Qq Nm Cm set Ar property Ns = Ns Ar value command was invoked at the same time the dataset was created. Any editable .Tn ZFS property can also be set at creation time. Multiple .Fl o options can be specified. An error results if the same property is specified in multiple .Fl o options. .El .It Xo .Nm .Cm destroy .Op Fl fnpRrv .Ar filesystem Ns | Ns Ar volume .Xc .Pp Destroys the given dataset. By default, the command unshares any file systems that are currently shared, unmounts any file systems that are currently mounted, and refuses to destroy a dataset that has active dependents (children or clones). .Bl -tag -width indent .It Fl r Recursively destroy all children. .It Fl R Recursively destroy all dependents, including cloned file systems outside the target hierarchy. .It Fl f Force an unmount of any file systems using the .Qq Nm Cm unmount Fl f command. This option has no effect on non-file systems or unmounted file systems. .It Fl n Do a dry-run ("No-op") deletion. No data will be deleted. This is useful in conjunction with the .Fl v or .Fl p flags to determine what data would be deleted. .It Fl p Print machine-parsable verbose information about the deleted data. .It Fl v Print verbose information about the deleted data. .El .Pp Extreme care should be taken when applying either the .Fl r or the .Fl R options, as they can destroy large portions of a pool and cause unexpected behavior for mounted file systems in use. .It Xo .Nm .Cm destroy .Op Fl dnpRrv .Sm off .Ar snapshot .Op % Ns Ar snapname .Op , Ns ... .Sm on .Xc .Pp The given snapshots are destroyed immediately if and only if the .Qq Nm Cm destroy command without the .Fl d option would have destroyed it. Such immediate destruction would occur, for example, if the snapshot had no clones and the user-initiated reference count were zero. .Pp If a snapshot does not qualify for immediate destruction, it is marked for deferred deletion. In this state, it exists as a usable, visible snapshot until both of the preconditions listed above are met, at which point it is destroyed. .Pp An inclusive range of snapshots may be specified by separating the first and last snapshots with a percent sign .Pq Sy % . The first and/or last snapshots may be left blank, in which case the filesystem's oldest or newest snapshot will be implied. .Pp Multiple snapshots (or ranges of snapshots) of the same filesystem or volume may be specified in a comma-separated list of snapshots. Only the snapshot's short name (the part after the .Sy @ ) should be specified when using a range or comma-separated list to identify multiple snapshots. .Bl -tag -width indent .It Fl r Destroy (or mark for deferred deletion) all snapshots with this name in descendent file systems. .It Fl R Recursively destroy all clones of these snapshots, including the clones, snapshots, and children. If this flag is specified, the .Op fl d flag will have no effect. .It Fl n Do a dry-run ("No-op") deletion. No data will be deleted. This is useful in conjunction with the .Fl v or .Fl p flags to determine what data would be deleted. .It Fl p Print machine-parsable verbose information about the deleted data. .It Fl v Print verbose information about the deleted data. .It Fl d Defer snapshot deletion. .El .Pp Extreme care should be taken when applying either the .Fl r or the .Fl R options, as they can destroy large portions of a pool and cause unexpected behavior for mounted file systems in use. .It Xo .Nm .Cm destroy .Ar filesystem Ns | Ns Ar volume Ns # Ns Ar bookmark .Xc .Pp The given bookmark is destroyed. .It Xo .Nm .Cm snapshot Ns | Ns Cm snap .Op Fl r .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Ar filesystem@snapname Ns | Ns volume@snapname .Ar filesystem@snapname Ns | Ns volume@snapname Ns ... .Xc .Pp Creates snapshots with the given names. All previous modifications by successful system calls to the file system are part of the snapshots. Snapshots are taken atomically, so that all snapshots correspond to the same moment in time. See the .Qq Sx Snapshots section for details. .Bl -tag -width indent .It Fl r Recursively create snapshots of all descendent datasets .It Fl o Ar property Ns = Ns Ar value Sets the specified property; see .Qq Nm Cm create for details. .El .It Xo .Nm .Cm rollback .Op Fl rRf .Ar snapshot .Xc .Pp Roll back the given dataset to a previous snapshot. When a dataset is rolled back, all data that has changed since the snapshot is discarded, and the dataset reverts to the state at the time of the snapshot. By default, the command refuses to roll back to a snapshot other than the most recent one. In order to do so, all intermediate snapshots and bookmarks must be destroyed by specifying the .Fl r option. .Pp The .Fl rR options do not recursively destroy the child snapshots of a recursive snapshot. Only direct snapshots of the specified filesystem are destroyed by either of these options. To completely roll back a recursive snapshot, you must rollback the individual child snapshots. .Bl -tag -width indent .It Fl r Destroy any snapshots and bookmarks more recent than the one specified. .It Fl R Destroy any more recent snapshots and bookmarks, as well as any clones of those snapshots. .It Fl f Used with the .Fl R option to force an unmount of any clone file systems that are to be destroyed. .El .It Xo .Nm .Cm clone .Op Fl p .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ... .Ar snapshot filesystem Ns | Ns Ar volume .Xc .Pp Creates a clone of the given snapshot. See the .Qq Sx Clones section for details. The target dataset can be located anywhere in the .Tn ZFS hierarchy, and is created as the same type as the original. .Bl -tag -width indent .It Fl p Creates all the non-existing parent datasets. Datasets created in this manner are automatically mounted according to the .Sy mountpoint property inherited from their parent. If the target filesystem or volume already exists, the operation completes successfully. .It Fl o Ar property Ns = Ns Ar value Sets the specified property; see .Qq Nm Cm create for details. .El .It Xo .Nm .Cm promote .Ar clone-filesystem .Xc .Pp Promotes a clone file system to no longer be dependent on its "origin" snapshot. This makes it possible to destroy the file system that the clone was created from. The clone parent-child dependency relationship is reversed, so that the origin file system becomes a clone of the specified file system. .Pp The snapshot that was cloned, and any snapshots previous to this snapshot, are now owned by the promoted clone. The space they use moves from the origin file system to the promoted clone, so enough space must be available to accommodate these snapshots. No new space is consumed by this operation, but the space accounting is adjusted. The promoted clone must not have any conflicting snapshot names of its own. The .Cm rename subcommand can be used to rename any conflicting snapshots. .It Xo .Nm .Cm rename .Op Fl f .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Xc .It Xo .Nm .Cm rename .Op Fl f .Fl p .Ar filesystem Ns | Ns Ar volume .Ar filesystem Ns | Ns Ar volume .Xc .It Xo .Nm .Cm rename .Fl u .Op Fl p .Ar filesystem filesystem .Xc .Pp Renames the given dataset. The new target can be located anywhere in the .Tn ZFS hierarchy, with the exception of snapshots. Snapshots can only be renamed within the parent file system or volume. When renaming a snapshot, the parent file system of the snapshot does not need to be specified as part of the second argument. Renamed file systems can inherit new mount points, in which case they are unmounted and remounted at the new mount point. .Bl -tag -width indent .It Fl p Creates all the nonexistent parent datasets. Datasets created in this manner are automatically mounted according to the .Sy mountpoint property inherited from their parent. .It Fl u Do not remount file systems during rename. If a file system's .Sy mountpoint property is set to .Cm legacy or .Cm none , file system is not unmounted even if this option is not given. .It Fl f Force unmount any filesystems that need to be unmounted in the process. This flag has no effect if used together with the .Fl u flag. .El .It Xo .Nm .Cm rename .Fl r .Ar snapshot snapshot .Xc .Pp Recursively rename the snapshots of all descendent datasets. Snapshots are the only dataset that can be renamed recursively. .It Xo .Nm .Cm list .Op Fl r Ns | Ns Fl d Ar depth .Op Fl Hp .Op Fl o Ar property Ns Oo , Ns Ar property Oc Ns ... .Op Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... .Oo Fl s Ar property Oc Ns ... .Oo Fl S Ar property Oc Ns ... .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Xc .Pp Lists the property information for the given datasets in tabular form. If specified, you can list property information by the absolute pathname or the relative pathname. By default, all file systems and volumes are displayed. Snapshots are displayed if the .Sy listsnaps property is .Cm on (the default is .Cm off ) . The following fields are displayed, .Sy name , used , available , referenced , mountpoint . .Bl -tag -width indent .It Fl r Recursively display any children of the dataset on the command line. .It Fl d Ar depth Recursively display any children of the dataset, limiting the recursion to .Ar depth . A depth of .Sy 1 will display only the dataset and its direct children. .It Fl H Used for scripting mode. Do not print headers and separate fields by a single tab instead of arbitrary white space. .It Fl p Display numbers in parsable (exact) values. .It Fl o Ar property Ns Oo , Ns Ar property Oc Ns ... A comma-separated list of properties to display. The property must be: .Bl -bullet -offset 2n .It One of the properties described in the .Qq Sx Native Properties section .It A user property .It The value .Cm name to display the dataset name .It The value .Cm space to display space usage properties on file systems and volumes. This is a shortcut for specifying .Fl o .Sy name,avail,used,usedsnap,usedds,usedrefreserv,usedchild .Fl t .Sy filesystem,volume syntax. .El .It Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... A comma-separated list of types to display, where .Ar type is one of .Sy filesystem , snapshot , snap , volume , bookmark , No or Sy all . For example, specifying .Fl t Cm snapshot displays only snapshots. .It Fl s Ar property A property for sorting the output by column in ascending order based on the value of the property. The property must be one of the properties described in the .Qq Sx Properties section, or the special value .Cm name to sort by the dataset name. Multiple properties can be specified at one time using multiple .Fl s property options. Multiple .Fl s options are evaluated from left to right in decreasing order of importance. .Pp The following is a list of sorting criteria: .Bl -bullet -offset 2n .It Numeric types sort in numeric order. .It String types sort in alphabetical order. .It Types inappropriate for a row sort that row to the literal bottom, regardless of the specified ordering. .It If no sorting options are specified the existing behavior of .Qq Nm Cm list is preserved. .El .It Fl S Ar property Same as the .Fl s option, but sorts by property in descending order. .El .It Xo .Nm .Cm set .Ar property Ns = Ns Ar value .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Xc .Pp Sets the property to the given value for each dataset. Only some properties can be edited. See the "Properties" section for more information on what properties can be set and acceptable values. Numeric values can be specified as exact values, or in a human-readable form with a suffix of .Sy B , K , M , G , T , P , E , Z (for bytes, kilobytes, megabytes, gigabytes, terabytes, petabytes, exabytes, or zettabytes, respectively). User properties can be set on snapshots. For more information, see the .Qq Sx User Properties section. .It Xo .Nm .Cm get .Op Fl r Ns | Ns Fl d Ar depth .Op Fl Hp .Op Fl o Ar all | field Ns Oo , Ns Ar field Oc Ns ... .Op Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... .Op Fl s Ar source Ns Oo , Ns Ar source Oc Ns ... .Ar all | property Ns Oo , Ns Ar property Oc Ns ... .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Xc .Pp Displays properties for the given datasets. If no datasets are specified, then the command displays properties for all datasets on the system. For each property, the following columns are displayed: .Pp .Bl -hang -width "property" -offset indent -compact .It name Dataset name .It property Property name .It value Property value .It source Property source. Can either be local, default, temporary, inherited, or none (\&-). .El .Pp All columns except the .Sy RECEIVED column are displayed by default. The columns to display can be specified by using the .Fl o option. This command takes a comma-separated list of properties as described in the .Qq Sx Native Properties and .Qq Sx User Properties sections. .Pp The special value .Cm all can be used to display all properties that apply to the given dataset's type (filesystem, volume, snapshot, or bookmark). .Bl -tag -width indent .It Fl r Recursively display properties for any children. .It Fl d Ar depth Recursively display any children of the dataset, limiting the recursion to .Ar depth . A depth of .Sy 1 will display only the dataset and its direct children. .It Fl H Display output in a form more easily parsed by scripts. Any headers are omitted, and fields are explicitly separated by a single tab instead of an arbitrary amount of space. .It Fl p Display numbers in parsable (exact) values. .It Fl o Cm all | Ar field Ns Oo , Ns Ar field Oc Ns ... A comma-separated list of columns to display. Supported values are .Sy name,property,value,received,source . Default values are .Sy name,property,value,source . The keyword .Cm all specifies all columns. .It Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... A comma-separated list of types to display, where .Ar type is one of .Sy filesystem , snapshot , volume , No or Sy all . For example, specifying .Fl t Cm snapshot displays only snapshots. .It Fl s Ar source Ns Oo , Ns Ar source Oc Ns ... A comma-separated list of sources to display. Those properties coming from a source other than those in this list are ignored. Each source must be one of the following: .Sy local,default,inherited,temporary,received,none . The default value is all sources. .El .It Xo .Nm .Cm inherit .Op Fl rS .Ar property .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot Ns ... .Xc .Pp Clears the specified property, causing it to be inherited from an ancestor. If no ancestor has the property set, then the default value is used. See the .Qq Sx Properties section for a listing of default values, and details on which properties can be inherited. .Bl -tag -width indent .It Fl r Recursively inherit the given property for all children. .It Fl S For properties with a received value, revert to this value. This flag has no effect on properties that do not have a received value. .El .It Xo .Nm .Cm upgrade .Op Fl v .Xc .Pp Displays a list of file systems that are not the most recent version. .Bl -tag -width indent .It Fl v Displays .Tn ZFS filesystem versions supported by the current software. The current .Tn ZFS filesystem version and all previous supported versions are displayed, along with an explanation of the features provided with each version. .El .It Xo .Nm .Cm upgrade .Op Fl r .Op Fl V Ar version .Fl a | Ar filesystem .Xc .Pp Upgrades file systems to a new on-disk version. Once this is done, the file systems will no longer be accessible on systems running older versions of the software. .Qq Nm Cm send streams generated from new snapshots of these file systems cannot be accessed on systems running older versions of the software. .Pp In general, the file system version is independent of the pool version. See .Xr zpool 8 for information on the .Nm zpool Cm upgrade command. .Pp In some cases, the file system version and the pool version are interrelated and the pool version must be upgraded before the file system version can be upgraded. .Bl -tag -width indent .It Fl r Upgrade the specified file system and all descendent file systems. .It Fl V Ar version Upgrade to the specified .Ar version . If the .Fl V flag is not specified, this command upgrades to the most recent version. This option can only be used to increase the version number, and only up to the most recent version supported by this software. .It Fl a Upgrade all file systems on all imported pools. .It Ar filesystem Upgrade the specified file system. .El .It Xo .Nm .Cm userspace .Op Fl Hinp .Op Fl o Ar field Ns Oo , Ns Ar field Oc Ns ... .Oo Fl s Ar field Oc Ns ... .Oo Fl S Ar field Oc Ns ... .Op Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... .Ar filesystem Ns | Ns Ar snapshot .Xc .Pp Displays space consumed by, and quotas on, each user in the specified filesystem or snapshot. This corresponds to the .Sy userused@ Ns Ar user and .Sy userquota@ Ns Ar user properties. .Bl -tag -width indent .It Fl n Print numeric ID instead of user/group name. .It Fl H Do not print headers, use tab-delimited output. .It Fl p Use exact (parsable) numeric output. .It Fl o Ar field Ns Oo , Ns Ar field Oc Ns ... Display only the specified fields from the following set: .Sy type,name,used,quota . The default is to display all fields. .It Fl s Ar field Sort output by this field. The .Fl s and .Fl S flags may be specified multiple times to sort first by one field, then by another. The default is .Fl s Cm type Fl s Cm name . .It Fl S Ar field Sort by this field in reverse order. See .Fl s . .It Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... Print only the specified types from the following set: .Sy all,posixuser,smbuser,posixgroup,smbgroup . .Pp The default is .Fl t Cm posixuser,smbuser . .Pp The default can be changed to include group types. .It Fl i Translate SID to POSIX ID. This flag currently has no effect on .Fx . .El .It Xo .Nm .Cm groupspace .Op Fl Hinp .Op Fl o Ar field Ns Oo , Ns Ar field Oc Ns ... .Oo Fl s Ar field Oc Ns ... .Oo Fl S Ar field Oc Ns ... .Op Fl t Ar type Ns Oo , Ns Ar type Oc Ns ... .Ar filesystem Ns | Ns Ar snapshot .Xc .Pp Displays space consumed by, and quotas on, each group in the specified filesystem or snapshot. This subcommand is identical to .Qq Nm Cm userspace , except that the default types to display are .Fl t Sy posixgroup,smbgroup . .It Xo .Nm .Cm mount .Xc .Pp Displays all .Tn ZFS file systems currently mounted. .Bl -tag -width indent .It Fl f .El .It Xo .Nm .Cm mount .Op Fl vO .Op Fl o Ar property Ns Oo , Ns Ar property Oc Ns ... .Fl a | Ar filesystem .Xc .Pp Mounts .Tn ZFS file systems. .Bl -tag -width indent .It Fl v Report mount progress. .It Fl O Perform an overlay mount. Overlay mounts are not supported on .Fx . .It Fl o Ar property Ns Oo , Ns Ar property Oc Ns ... An optional, comma-separated list of mount options to use temporarily for the duration of the mount. See the .Qq Sx Temporary Mount Point Properties section for details. .It Fl a Mount all available .Tn ZFS file systems. This command may be executed on .Fx system startup by .Pa /etc/rc.d/zfs . For more information, see variable .Va zfs_enable in .Xr rc.conf 5 . .It Ar filesystem Mount the specified filesystem. .El .It Xo .Nm .Cm unmount Ns | Ns Cm umount .Op Fl f .Fl a | Ar filesystem Ns | Ns Ar mountpoint .Xc .Pp Unmounts currently mounted .Tn ZFS file systems. .Bl -tag -width indent .It Fl f Forcefully unmount the file system, even if it is currently in use. .It Fl a Unmount all available .Tn ZFS file systems. .It Ar filesystem | mountpoint Unmount the specified filesystem. The command can also be given a path to a .Tn ZFS file system mount point on the system. .El .It Xo .Nm .Cm share .Fl a | Ar filesystem .Xc .Pp Shares .Tn ZFS file systems that have the .Sy sharenfs property set. .Bl -tag -width indent .It Fl a Share all .Tn ZFS file systems that have the .Sy sharenfs property set. This command may be executed on .Fx system startup by .Pa /etc/rc.d/zfs . For more information, see variable .Va zfs_enable in .Xr rc.conf 5 . .It Ar filesystem Share the specified filesystem according to the .Tn sharenfs property. File systems are shared when the .Tn sharenfs property is set. .El .It Xo .Nm .Cm unshare .Fl a | Ar filesystem Ns | Ns Ar mountpoint .Xc .Pp Unshares .Tn ZFS file systems that have the .Tn sharenfs property set. .Bl -tag -width indent .It Fl a Unshares .Tn ZFS file systems that have the .Sy sharenfs property set. This command may be executed on .Fx system shutdown by .Pa /etc/rc.d/zfs . For more information, see variable .Va zfs_enable in .Xr rc.conf 5 . .It Ar filesystem | mountpoint Unshare the specified filesystem. The command can also be given a path to a .Tn ZFS file system shared on the system. .El .It Xo .Nm .Cm bookmark .Ar snapshot .Ar bookmark .Xc .Pp Creates a bookmark of the given snapshot. Bookmarks mark the point in time when the snapshot was created, and can be used as the incremental source for a .Qq Nm Cm send command. .Pp This feature must be enabled to be used. See .Xr zpool-features 7 for details on ZFS feature flags and the .Sy bookmark feature. .It Xo .Nm .Cm send .Op Fl DnPpRv .Op Fl i Ar snapshot | Fl I Ar snapshot .Ar snapshot .Xc .Pp Creates a stream representation of the last .Ar snapshot argument (not part of .Fl i or .Fl I ) which is written to standard output. The output can be redirected to a file or to a different system (for example, using .Xr ssh 1 ) . By default, a full stream is generated. .Bl -tag -width indent .It Fl i Ar snapshot Generate an incremental stream from the first .Ar snapshot Pq the incremental source to the second .Ar snapshot Pq the incremental target . The incremental source can be specified as the last component of the snapshot name .Pq the Em @ No character and following and it is assumed to be from the same file system as the incremental target. .Pp If the destination is a clone, the source may be the origin snapshot, which must be fully specified (for example, .Cm pool/fs@origin , not just .Cm @origin ) . .It Fl I Ar snapshot Generate a stream package that sends all intermediary snapshots from the first .Ar snapshot to the second .Ar snapshot . For example, .Ic -I @a fs@d is similar to .Ic -i @a fs@b; -i @b fs@c; -i @c fs@d . The incremental source may be specified as with the .Fl i option. .It Fl R Generate a replication stream package, which will replicate the specified filesystem, and all descendent file systems, up to the named snapshot. When received, all properties, snapshots, descendent file systems, and clones are preserved. .Pp If the .Fl i or .Fl I flags are used in conjunction with the .Fl R flag, an incremental replication stream is generated. The current values of properties, and current snapshot and file system names are set when the stream is received. If the .Fl F flag is specified when this stream is received, snapshots and file systems that do not exist on the sending side are destroyed. .It Fl D Generate a deduplicated stream. Blocks which would have been sent multiple times in the send stream will only be sent once. The receiving system must also support this feature to receive a deduplicated stream. This flag can be used regardless of the dataset's .Sy dedup property, but performance will be much better if the filesystem uses a dedup-capable checksum (eg. .Sy sha256 ) . .It Fl p Include the dataset's properties in the stream. This flag is implicit when .Fl R is specified. The receiving system must also support this feature. .It Fl n Do a dry-run ("No-op") send. Do not generate any actual send data. This is useful in conjunction with the .Fl v or .Fl P flags to determine what data will be sent. .It Fl P Print machine-parsable verbose information about the stream package generated. .It Fl v Print verbose information about the stream package generated. This information includes a per-second report of how much data has been sent. .El .Pp The format of the stream is committed. You will be able to receive your streams on future versions of .Tn ZFS . .It Xo .Nm .Cm send .Op Fl i Ar snapshot Ns | Ns Ar bookmark .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Xc .Pp Generate a send stream, which may be of a filesystem, and may be incremental from a bookmark. If the destination is a filesystem or volume, the pool must be read-only, or the filesystem must not be mounted. When the stream generated from a filesystem or volume is received, the default snapshot name will be .Pq --head-- . .Bl -tag -width indent .It Fl i Ar snapshot Ns | Ns bookmark Generate an incremental send stream. The incremental source must be an earlier snapshot in the destination's history. It will commonly be an earlier snapshot in the destination's filesystem, in which case it can be specified as the last component of the name .Pq the Em # No or Em @ No character and following . .Pp If the incremental target is a clone, the incremental source can be the origin snapshot, or an earlier snapshot in the origin's filesystem, or the origin's origin, etc. .El .It Xo .Nm .Cm receive Ns | Ns Cm recv .Op Fl vnFu .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot .Xc .It Xo .Nm .Cm receive Ns | Ns Cm recv .Op Fl vnFu .Op Fl d | e .Ar filesystem .Xc .Pp Creates a snapshot whose contents are as specified in the stream provided on standard input. If a full stream is received, then a new file system is created as well. Streams are created using the .Qq Nm Cm send subcommand, which by default creates a full stream. .Qq Nm Cm recv can be used as an alias for .Qq Nm Cm receive . .Pp If an incremental stream is received, then the destination file system must already exist, and its most recent snapshot must match the incremental stream's source. For .Sy zvol Ns s, the destination device link is destroyed and recreated, which means the .Sy zvol cannot be accessed during the .Sy receive operation. .Pp When a snapshot replication package stream that is generated by using the .Qq Nm Cm send Fl R command is received, any snapshots that do not exist on the sending location are destroyed by using the .Qq Nm Cm destroy Fl d command. .Pp The name of the snapshot (and file system, if a full stream is received) that this subcommand creates depends on the argument type and the .Fl d or .Fl e option. .Pp If the argument is a snapshot name, the specified .Ar snapshot is created. If the argument is a file system or volume name, a snapshot with the same name as the sent snapshot is created within the specified .Ar filesystem or .Ar volume . If the .Fl d or .Fl e option is specified, the snapshot name is determined by appending the sent snapshot's name to the specified .Ar filesystem . If the .Fl d option is specified, all but the pool name of the sent snapshot path is appended (for example, .Sy b/c@1 appended from sent snapshot .Sy a/b/c@1 ) , and if the .Fl e option is specified, only the tail of the sent snapshot path is appended (for example, .Sy c@1 appended from sent snapshot .Sy a/b/c@1 ) . In the case of .Fl d , any file systems needed to replicate the path of the sent snapshot are created within the specified file system. .Bl -tag -width indent .It Fl d Use the full sent snapshot path without the first element (without pool name) to determine the name of the new snapshot as described in the paragraph above. .It Fl e Use only the last element of the sent snapshot path to determine the name of the new snapshot as described in the paragraph above. .It Fl u File system that is associated with the received stream is not mounted. .It Fl v Print verbose information about the stream and the time required to perform the receive operation. .It Fl n Do not actually receive the stream. This can be useful in conjunction with the .Fl v option to verify the name the receive operation would use. .It Fl F Force a rollback of the file system to the most recent snapshot before performing the receive operation. If receiving an incremental replication stream (for example, one generated by .Qq Nm Cm send Fl R Fi iI ) , destroy snapshots and file systems that do not exist on the sending side. .El .It Xo .Nm .Cm allow .Ar filesystem Ns | Ns Ar volume .Xc .Pp Displays permissions that have been delegated on the specified filesystem or volume. See the other forms of .Qq Nm Cm allow for more information. .It Xo .Nm .Cm allow .Op Fl ldug .Ar user Ns | Ns Ar group Ns Oo Ns , Ns Ar user Ns | Ns Ar group Oc Ns ... .Ar perm Ns | Ns Ar @setname Ns .Oo Ns , Ns Ar perm Ns | Ns Ar @setname Oc Ns ... .Ar filesystem Ns | Ns Ar volume .Xc .It Xo .Nm .Cm allow .Op Fl ld .Fl e Ns | Ns Cm everyone .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Xc .Pp Delegates .Tn ZFS administration permission for the file systems to non-privileged users. .Bl -tag -width indent .It Xo .Op Fl ug .Ar user Ns | Ns Ar group Ns Oo , Ar user Ns | Ns Ar group Oc Ns ... .Xc Specifies to whom the permissions are delegated. Multiple entities can be specified as a comma-separated list. If neither of the .Fl ug options are specified, then the argument is interpreted preferentially as the keyword .Cm everyone , then as a user name, and lastly as a group name. To specify a user or group named .Qq everyone , use the .Fl u or .Fl g options. To specify a group with the same name as a user, use the .Fl g option. .It Op Fl e Ns | Ns Cm everyone Specifies that the permissions be delegated to .Qq everyone . .It Xo .Ar perm Ns | Ns Ar @setname Ns Oo , Ns Ar perm Ns | Ns Ar @setname Oc Ns ... .Xc The permissions to delegate. Multiple permissions may be specified as a comma-separated list. Permission names are the same as .Tn ZFS subcommand and property names. See the property list below. Property set names, which begin with an at sign .Pq Sy @ , may be specified. See the .Fl s form below for details. .It Xo .Op Fl ld .Ar filesystem Ns | Ns Ar volume .Xc Specifies where the permissions are delegated. If neither of the .Fl ld options are specified, or both are, then the permissions are allowed for the file system or volume, and all of its descendents. If only the .Fl l option is used, then is allowed "locally" only for the specified file system. If only the .Fl d option is used, then is allowed only for the descendent file systems. .El .Pp Permissions are generally the ability to use a .Tn ZFS subcommand or change a .Tn ZFS property. The following permissions are available: .Bl -column -offset 4n "secondarycache" "subcommand" .It NAME Ta TYPE Ta NOTES .It allow Ta subcommand Ta Must Xo also have the permission that is being allowed .Xc .It clone Ta subcommand Ta Must Xo also have the 'create' ability and 'mount' ability in the origin file system .Xc .It create Ta subcommand Ta Must also have the 'mount' ability .It destroy Ta subcommand Ta Must also have the 'mount' ability .It diff Ta subcommand Ta Allows lookup of paths within a dataset given an object number, and the ability to create snapshots necessary to 'zfs diff' .It hold Ta subcommand Ta Allows adding a user hold to a snapshot .It mount Ta subcommand Ta Allows mount/umount of Tn ZFS No datasets .It promote Ta subcommand Ta Must Xo also have the 'mount' and 'promote' ability in the origin file system .Xc .It receive Ta subcommand Ta Must also have the 'mount' and 'create' ability .It release Ta subcommand Ta Allows Xo releasing a user hold which might destroy the snapshot .Xc .It rename Ta subcommand Ta Must Xo also have the 'mount' and 'create' ability in the new parent .Xc .It rollback Ta subcommand Ta Must also have the 'mount' ability .It send Ta subcommand .It share Ta subcommand Ta Allows Xo sharing file systems over the .Tn NFS protocol .Xc .It snapshot Ta subcommand Ta Must also have the 'mount' ability .It groupquota Ta other Ta Allows accessing any groupquota@... property .It groupused Ta other Ta Allows reading any groupused@... property .It userprop Ta other Ta Allows changing any user property .It userquota Ta other Ta Allows accessing any userquota@... property .It userused Ta other Ta Allows reading any userused@... property .It aclinherit Ta property .It aclmode Ta property .It atime Ta property .It canmount Ta property .It casesensitivity Ta property .It checksum Ta property .It compression Ta property .It copies Ta property .It dedup Ta property .It devices Ta property .It exec Ta property .It filesystem_limit Ta property .It logbias Ta property .It jailed Ta property .It mlslabel Ta property .It mountpoint Ta property .It nbmand Ta property .It normalization Ta property .It primarycache Ta property .It quota Ta property .It readonly Ta property .It recordsize Ta property .It refquota Ta property .It refreservation Ta property .It reservation Ta property .It secondarycache Ta property .It setuid Ta property .It sharenfs Ta property .It sharesmb Ta property .It snapdir Ta property .It snapshot_limit Ta property .It sync Ta property .It utf8only Ta property .It version Ta property .It volblocksize Ta property .It volsize Ta property .It vscan Ta property .It xattr Ta property .El .It Xo .Nm .Cm allow .Fl c .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Xc .Pp Sets "create time" permissions. These permissions are granted (locally) to the creator of any newly-created descendent file system. .It Xo .Nm .Cm allow .Fl s .Ar @setname .Ar perm Ns | Ns Ar @setname Ns Op Ns , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... .Ar filesystem Ns | Ns Ar volume .Xc .Pp Defines or adds permissions to a permission set. The set can be used by other .Qq Nm Cm allow commands for the specified file system and its descendents. Sets are evaluated dynamically, so changes to a set are immediately reflected. Permission sets follow the same naming restrictions as ZFS file systems, but the name must begin with an "at sign" .Pq Sy @ , and can be no more than 64 characters long. .It Xo .Nm .Cm unallow .Op Fl rldug .Ar user Ns | Ns Ar group Ns Oo Ns , Ns Ar user Ns | Ns Ar group Oc Ns ... .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Xc .It Xo .Nm .Cm unallow .Op Fl rld .Fl e Ns | Ns Cm everyone .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Xc .It Xo .Nm .Cm unallow .Op Fl r .Fl c .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Xc .Pp Removes permissions that were granted with the .Qq Nm Cm allow command. No permissions are explicitly denied, so other permissions granted are still in effect. For example, if the permission is granted by an ancestor. If no permissions are specified, then all permissions for the specified .Ar user , group , No or everyone are removed. Specifying .Cm everyone .Po or using the Fl e option .Pc only removes the permissions that were granted to everyone , not all permissions for every user and group. See the .Qq Nm Cm allow command for a description of the .Fl ldugec options. .Bl -tag -width indent .It Fl r Recursively remove the permissions from this file system and all descendents. .El .It Xo .Nm .Cm unallow .Op Fl r .Fl s .Ar @setname .Oo Ar perm Ns | Ns Ar @setname Ns Op , Ns Ar perm Ns | Ns Ar @setname Ns .Ns ... Oc .Ar filesystem Ns | Ns Ar volume .Xc .Pp Removes permissions from a permission set. If no permissions are specified, then all permissions are removed, thus removing the set entirely. .It Xo .Nm .Cm hold .Op Fl r .Ar tag snapshot Ns ... .Xc .Pp Adds a single reference, named with the .Ar tag argument, to the specified snapshot or snapshots. Each snapshot has its own tag namespace, and tags must be unique within that space. .Pp If a hold exists on a snapshot, attempts to destroy that snapshot by using the .Qq Nm Cm destroy command returns .Em EBUSY . .Bl -tag -width indent .It Fl r Specifies that a hold with the given tag is applied recursively to the snapshots of all descendent file systems. .El .It Xo .Nm .Cm holds .Op Fl r .Ar snapshot Ns ... .Xc .Pp Lists all existing user references for the given snapshot or snapshots. .Bl -tag -width indent .It Fl r Lists the holds that are set on the named descendent snapshots, in addition to listing the holds on the named snapshot. .El .It Xo .Nm .Cm release .Op Fl r .Ar tag snapshot Ns ... .Xc .Pp Removes a single reference, named with the .Ar tag argument, from the specified snapshot or snapshots. The tag must already exist for each snapshot. .Bl -tag -width indent .It Fl r Recursively releases a hold with the given tag on the snapshots of all descendent file systems. .El .It Xo .Nm .Cm diff .Op Fl FHt .Ar snapshot .Op Ar snapshot Ns | Ns Ar filesystem .Xc .Pp Display the difference between a snapshot of a given filesystem and another snapshot of that filesystem from a later time or the current contents of the filesystem. The first column is a character indicating the type of change, the other columns indicate pathname, new pathname .Pq in case of rename , change in link count, and optionally file type and/or change time. .Pp The types of change are: .Bl -column -offset 2n indent .It \&- Ta path was removed .It \&+ Ta path was added .It \&M Ta path was modified .It \&R Ta path was renamed .El .Bl -tag -width indent .It Fl F Display an indication of the type of file, in a manner similar to the .Fl F option of .Xr ls 1 . .Bl -column -offset 2n indent .It \&B Ta block device .It \&C Ta character device .It \&F Ta regular file .It \&/ Ta directory .It \&@ Ta symbolic link .It \&= Ta socket .It \&> Ta door (not supported on Fx ) .It \&| Ta named pipe (not supported on Fx ) .It \&P Ta event port (not supported on Fx ) .El .It Fl H Give more parsable tab-separated output, without header lines and without arrows. .It Fl t Display the path's inode change time as the first column of output. .El .It Xo .Nm .Cm jail .Ar jailid filesystem .Xc .Pp Attaches the specified .Ar filesystem to the jail identified by JID .Ar jailid . From now on this file system tree can be managed from within a jail if the .Sy jailed property has been set. To use this functionality, the jail needs the .Va allow.mount and .Va allow.mount.zfs parameters set to 1 and the .Va enforce_statfs parameter set to a value lower than 2. .Pp See .Xr jail 8 for more information on managing jails and configuring the parameters above. .It Xo .Nm .Cm unjail .Ar jailid filesystem .Xc .Pp Detaches the specified .Ar filesystem from the jail identified by JID .Ar jailid . .El .Sh EXIT STATUS The following exit values are returned: .Bl -tag -offset 2n -width 2n .It 0 Successful completion. .It 1 An error occurred. .It 2 Invalid command line options were specified. .El .Sh EXAMPLES .Bl -tag -width 0n .It Sy Example 1 No Creating a Tn ZFS No File System Hierarchy .Pp The following commands create a file system named .Em pool/home and a file system named .Em pool/home/bob . The mount point .Pa /home is set for the parent file system, and is automatically inherited by the child file system. .Bd -literal -offset 2n .Li # Ic zfs create pool/home .Li # Ic zfs set mountpoint=/home pool/home .Li # Ic zfs create pool/home/bob .Ed .It Sy Example 2 No Creating a Tn ZFS No Snapshot .Pp The following command creates a snapshot named .Sy yesterday . This snapshot is mounted on demand in the .Pa \&.zfs/snapshot directory at the root of the .Em pool/home/bob file system. .Bd -literal -offset 2n .Li # Ic zfs snapshot pool/home/bob@yesterday .Ed .It Sy Example 3 No Creating and Destroying Multiple Snapshots .Pp The following command creates snapshots named .Em yesterday of .Em pool/home and all of its descendent file systems. Each snapshot is mounted on demand in the .Pa \&.zfs/snapshot directory at the root of its file system. The second command destroys the newly created snapshots. .Bd -literal -offset 2n .Li # Ic zfs snapshot -r pool/home@yesterday .Li # Ic zfs destroy -r pool/home@yesterday .Ed .It Sy Example 4 No Disabling and Enabling File System Compression .Pp The following command disables the .Sy compression property for all file systems under .Em pool/home . The next command explicitly enables .Sy compression for .Em pool/home/anne . .Bd -literal -offset 2n .Li # Ic zfs set compression=off pool/home .Li # Ic zfs set compression=on pool/home/anne .Ed .It Sy Example 5 No Listing Tn ZFS No Datasets .Pp The following command lists all active file systems and volumes in the system. Snapshots are displayed if the .Sy listsnaps property is .Cm on . The default is .Cm off . See .Xr zpool 8 for more information on pool properties. .Bd -literal -offset 2n .Li # Ic zfs list NAME USED AVAIL REFER MOUNTPOINT pool 450K 457G 18K /pool pool/home 315K 457G 21K /home pool/home/anne 18K 457G 18K /home/anne pool/home/bob 276K 457G 276K /home/bob .Ed .It Sy Example 6 No Setting a Quota on a Tn ZFS No File System .Pp The following command sets a quota of 50 Gbytes for .Em pool/home/bob . .Bd -literal -offset 2n .Li # Ic zfs set quota=50G pool/home/bob .Ed .It Sy Example 7 No Listing Tn ZFS No Properties .Pp The following command lists all properties for .Em pool/home/bob . .Bd -literal -offset 2n .Li # Ic zfs get all pool/home/bob NAME PROPERTY VALUE SOURCE pool/home/bob type filesystem - pool/home/bob creation Tue Jul 21 15:53 2009 - pool/home/bob used 21K - pool/home/bob available 20.0G - pool/home/bob referenced 21K - pool/home/bob compressratio 1.00x - pool/home/bob mounted yes - pool/home/bob quota 20G local pool/home/bob reservation none default pool/home/bob recordsize 128K default pool/home/bob mountpoint /home/bob default pool/home/bob sharenfs off default pool/home/bob checksum on default pool/home/bob compression on local pool/home/bob atime on default pool/home/bob devices on default pool/home/bob exec on default pool/home/bob filesystem_limit none default pool/home/bob setuid on default pool/home/bob readonly off default pool/home/bob jailed off default pool/home/bob snapdir hidden default pool/home/bob snapshot_limit none default pool/home/bob aclmode discard default pool/home/bob aclinherit restricted default pool/home/bob canmount on default pool/home/bob xattr on default pool/home/bob copies 1 default pool/home/bob version 5 - pool/home/bob utf8only off - pool/home/bob normalization none - pool/home/bob casesensitivity sensitive - pool/home/bob vscan off default pool/home/bob nbmand off default pool/home/bob sharesmb off default pool/home/bob refquota none default pool/home/bob refreservation none default pool/home/bob primarycache all default pool/home/bob secondarycache all default pool/home/bob usedbysnapshots 0 - pool/home/bob usedbydataset 21K - pool/home/bob usedbychildren 0 - pool/home/bob usedbyrefreservation 0 - pool/home/bob logbias latency default pool/home/bob dedup off default pool/home/bob mlslabel - pool/home/bob sync standard default pool/home/bob refcompressratio 1.00x - .Ed .Pp The following command gets a single property value. .Bd -literal -offset 2n .Li # Ic zfs get -H -o value compression pool/home/bob on .Ed .Pp The following command lists all properties with local settings for .Em pool/home/bob . .Bd -literal -offset 2n .Li # Ic zfs get -s local -o name,property,value all pool/home/bob NAME PROPERTY VALUE pool/home/bob quota 20G pool/home/bob compression on .Ed .It Sy Example 8 No Rolling Back a Tn ZFS No File System .Pp The following command reverts the contents of .Em pool/home/anne to the snapshot named .Em yesterday , deleting all intermediate snapshots. .Bd -literal -offset 2n .Li # Ic zfs rollback -r pool/home/anne@yesterday .Ed .It Sy Example 9 No Creating a Tn ZFS No Clone .Pp The following command creates a writable file system whose initial contents are the same as .Em pool/home/bob@yesterday . .Bd -literal -offset 2n .Li # Ic zfs clone pool/home/bob@yesterday pool/clone .Ed .It Sy Example 10 No Promoting a Tn ZFS No Clone .Pp The following commands illustrate how to test out changes to a file system, and then replace the original file system with the changed one, using clones, clone promotion, and renaming: .Bd -literal -offset 2n .Li # Ic zfs create pool/project/production .Ed .Pp Populate .Pa /pool/project/production with data and continue with the following commands: .Bd -literal -offset 2n .Li # Ic zfs snapshot pool/project/production@today .Li # Ic zfs clone pool/project/production@today pool/project/beta .Ed .Pp Now make changes to .Pa /pool/project/beta and continue with the following commands: .Bd -literal -offset 2n .Li # Ic zfs promote pool/project/beta .Li # Ic zfs rename pool/project/production pool/project/legacy .Li # Ic zfs rename pool/project/beta pool/project/production .Ed .Pp Once the legacy version is no longer needed, it can be destroyed. .Bd -literal -offset 2n .Li # Ic zfs destroy pool/project/legacy .Ed .It Sy Example 11 No Inheriting Tn ZFS No Properties .Pp The following command causes .Em pool/home/bob and .Em pool/home/anne to inherit the .Sy checksum property from their parent. .Bd -literal -offset 2n .Li # Ic zfs inherit checksum pool/home/bob pool/home/anne .Ed .It Sy Example 12 No Remotely Replicating Tn ZFS No Data .Pp The following commands send a full stream and then an incremental stream to a remote machine, restoring them into .Sy poolB/received/fs@a and .Sy poolB/received/fs@b , respectively. .Sy poolB must contain the file system .Sy poolB/received , and must not initially contain .Sy poolB/received/fs . .Bd -literal -offset 2n .Li # Ic zfs send pool/fs@a | ssh host zfs receive poolB/received/fs@a .Li # Ic zfs send -i a pool/fs@b | ssh host zfs receive poolB/received/fs .Ed .It Xo .Sy Example 13 Using the .Qq zfs receive -d Option .Xc .Pp The following command sends a full stream of .Sy poolA/fsA/fsB@snap to a remote machine, receiving it into .Sy poolB/received/fsA/fsB@snap . The .Sy fsA/fsB@snap portion of the received snapshot's name is determined from the name of the sent snapshot. .Sy poolB must contain the file system .Sy poolB/received . If .Sy poolB/received/fsA does not exist, it is created as an empty file system. .Bd -literal -offset 2n .Li # Ic zfs send poolA/fsA/fsB@snap | ssh host zfs receive -d poolB/received .Ed .It Sy Example 14 No Setting User Properties .Pp The following example sets the user-defined .Sy com.example:department property for a dataset. .Bd -literal -offset 2n .Li # Ic zfs set com.example:department=12345 tank/accounting .Ed .It Sy Example 15 No Performing a Rolling Snapshot .Pp The following example shows how to maintain a history of snapshots with a consistent naming scheme. To keep a week's worth of snapshots, the user destroys the oldest snapshot, renames the remaining snapshots, and then creates a new snapshot, as follows: .Bd -literal -offset 2n .Li # Ic zfs destroy -r pool/users@7daysago .Li # Ic zfs rename -r pool/users@6daysago @7daysago .Li # Ic zfs rename -r pool/users@5daysago @6daysago .Li # Ic zfs rename -r pool/users@4daysago @5daysago .Li # Ic zfs rename -r pool/users@3daysago @4daysago .Li # Ic zfs rename -r pool/users@2daysago @3daysago .Li # Ic zfs rename -r pool/users@yesterday @2daysago .Li # Ic zfs rename -r pool/users@today @yesterday .Li # Ic zfs snapshot -r pool/users@today .Ed .It Xo .Sy Example 16 Setting .Qq sharenfs Property Options on a ZFS File System .Xc .Pp The following command shows how to set .Sy sharenfs property options to enable root access for a specific network on the .Em tank/home file system. The contents of the .Sy sharenfs property are valid .Xr exports 5 options. .Bd -literal -offset 2n .Li # Ic zfs set sharenfs="maproot=root,network 192.168.0.0/24" tank/home .Ed .Pp Another way to write this command with the same result is: .Bd -literal -offset 2n .Li # Ic set zfs sharenfs="-maproot=root -network 192.168.0.0/24" tank/home .Ed .It Xo .Sy Example 17 Delegating .Tn ZFS Administration Permissions on a .Tn ZFS Dataset .Xc .Pp The following example shows how to set permissions so that user .Em cindys can create, destroy, mount, and take snapshots on .Em tank/cindys . The permissions on .Em tank/cindys are also displayed. .Bd -literal -offset 2n .Li # Ic zfs allow cindys create,destroy,mount,snapshot tank/cindys .Li # Ic zfs allow tank/cindys ------------------------------------------------------------- Local+Descendent permissions on (tank/cindys) user cindys create,destroy,mount,snapshot ------------------------------------------------------------- .Ed .It Sy Example 18 No Delegating Create Time Permissions on a Tn ZFS No Dataset .Pp The following example shows how to grant anyone in the group .Em staff to create file systems in .Em tank/users . This syntax also allows staff members to destroy their own file systems, but not destroy anyone else's file system. The permissions on .Em tank/users are also displayed. .Bd -literal -offset 2n .Li # Ic zfs allow staff create,mount tank/users .Li # Ic zfs allow -c destroy tank/users .Li # Ic zfs allow tank/users ------------------------------------------------------------- Create time permissions on (tank/users) create,destroy Local+Descendent permissions on (tank/users) group staff create,mount ------------------------------------------------------------- .Ed .It Xo .Sy Example 19 Defining and Granting a Permission Set on a .Tn ZFS Dataset .Xc .Pp The following example shows how to define and grant a permission set on the .Em tank/users file system. The permissions on .Em tank/users are also displayed. .Bd -literal -offset 2n .Li # Ic zfs allow -s @pset create,destroy,snapshot,mount tank/users .Li # Ic zfs allow staff @pset tank/users .Li # Ic zfs allow tank/users ------------------------------------------------------------- Permission sets on (tank/users) @pset create,destroy,mount,snapshot Create time permissions on (tank/users) create,destroy Local+Descendent permissions on (tank/users) group staff @pset,create,mount ------------------------------------------------------------- .Ed .It Sy Example 20 No Delegating Property Permissions on a Tn ZFS No Dataset .Pp The following example shows to grant the ability to set quotas and reservations on the .Sy users/home file system. The permissions on .Sy users/home are also displayed. .Bd -literal -offset 2n .Li # Ic zfs allow cindys quota,reservation users/home .Li # Ic zfs allow cindys ------------------------------------------------------------- Local+Descendent permissions on (users/home) user cindys quota,reservation ------------------------------------------------------------- .Li # Ic su - cindys .Li cindys% Ic zfs set quota=10G users/home/marks .Li cindys% Ic zfs get quota users/home/marks NAME PROPERTY VALUE SOURCE users/home/marks quota 10G local .Ed .It Sy Example 21 No Removing ZFS Delegated Permissions on a Tn ZFS No Dataset .Pp The following example shows how to remove the snapshot permission from the .Em staff group on the .Em tank/users file system. The permissions on .Em tank/users are also displayed. .Bd -literal -offset 2n .Li # Ic zfs unallow staff snapshot tank/users .Li # Ic zfs allow tank/users ------------------------------------------------------------- Permission sets on (tank/users) @pset create,destroy,mount,snapshot Create time permissions on (tank/users) create,destroy Local+Descendent permissions on (tank/users) group staff @pset,create,mount ------------------------------------------------------------- .Ed .It Sy Example 22 Showing the differences between a snapshot and a ZFS Dataset .Pp The following example shows how to see what has changed between a prior snapshot of a ZFS Dataset and its current state. The .Fl F option is used to indicate type information for the files affected. .Bd -literal -offset 2n .Li # Ic zfs diff tank/test@before tank/test M / /tank/test/ M F /tank/test/linked (+1) R F /tank/test/oldname -> /tank/test/newname - F /tank/test/deleted + F /tank/test/created M F /tank/test/modified .Ed .El .Sh SEE ALSO .Xr chmod 2 , .Xr fsync 2 , .Xr exports 5 , .Xr fstab 5 , .Xr rc.conf 5 , .Xr jail 8 , .Xr mount 8 , .Xr umount 8 , .Xr zpool 8 .Sh AUTHORS This manual page is a .Xr mdoc 7 reimplementation of the .Tn OpenSolaris manual page .Em zfs(1M) , modified and customized for .Fx and licensed under the Common Development and Distribution License .Pq Tn CDDL . .Pp The .Xr mdoc 7 implementation of this manual page was initially written by .An Martin Matuska Aq mm@FreeBSD.org . Index: stable/10/sys/cddl/contrib/opensolaris/common/zfs/zfs_prop.c =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/common/zfs/zfs_prop.c (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/common/zfs/zfs_prop.c (revision 268647) @@ -1,655 +1,666 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2011, 2014 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. * Copyright (c) 2013, Joyent, Inc. All rights reserved. */ /* Portions Copyright 2010 Robert Milkowski */ #include #include #include #include #include #include #include "zfs_prop.h" #include "zfs_deleg.h" #if defined(_KERNEL) #include #else #include #include #include #endif static zprop_desc_t zfs_prop_table[ZFS_NUM_PROPS]; /* Note this is indexed by zfs_userquota_prop_t, keep the order the same */ const char *zfs_userquota_prop_prefixes[] = { "userused@", "userquota@", "groupused@", "groupquota@" }; zprop_desc_t * zfs_prop_get_table(void) { return (zfs_prop_table); } void zfs_prop_init(void) { static zprop_index_t checksum_table[] = { { "on", ZIO_CHECKSUM_ON }, { "off", ZIO_CHECKSUM_OFF }, { "fletcher2", ZIO_CHECKSUM_FLETCHER_2 }, { "fletcher4", ZIO_CHECKSUM_FLETCHER_4 }, { "sha256", ZIO_CHECKSUM_SHA256 }, { "noparity", ZIO_CHECKSUM_NOPARITY }, { NULL } }; static zprop_index_t dedup_table[] = { { "on", ZIO_CHECKSUM_ON }, { "off", ZIO_CHECKSUM_OFF }, { "verify", ZIO_CHECKSUM_ON | ZIO_CHECKSUM_VERIFY }, { "sha256", ZIO_CHECKSUM_SHA256 }, { "sha256,verify", ZIO_CHECKSUM_SHA256 | ZIO_CHECKSUM_VERIFY }, { NULL } }; static zprop_index_t compress_table[] = { { "on", ZIO_COMPRESS_ON }, { "off", ZIO_COMPRESS_OFF }, { "lzjb", ZIO_COMPRESS_LZJB }, { "gzip", ZIO_COMPRESS_GZIP_6 }, /* gzip default */ { "gzip-1", ZIO_COMPRESS_GZIP_1 }, { "gzip-2", ZIO_COMPRESS_GZIP_2 }, { "gzip-3", ZIO_COMPRESS_GZIP_3 }, { "gzip-4", ZIO_COMPRESS_GZIP_4 }, { "gzip-5", ZIO_COMPRESS_GZIP_5 }, { "gzip-6", ZIO_COMPRESS_GZIP_6 }, { "gzip-7", ZIO_COMPRESS_GZIP_7 }, { "gzip-8", ZIO_COMPRESS_GZIP_8 }, { "gzip-9", ZIO_COMPRESS_GZIP_9 }, { "zle", ZIO_COMPRESS_ZLE }, { "lz4", ZIO_COMPRESS_LZ4 }, { NULL } }; static zprop_index_t snapdir_table[] = { { "hidden", ZFS_SNAPDIR_HIDDEN }, { "visible", ZFS_SNAPDIR_VISIBLE }, { NULL } }; static zprop_index_t acl_mode_table[] = { { "discard", ZFS_ACL_DISCARD }, { "groupmask", ZFS_ACL_GROUPMASK }, { "passthrough", ZFS_ACL_PASSTHROUGH }, { "restricted", ZFS_ACL_RESTRICTED }, { NULL } }; static zprop_index_t acl_inherit_table[] = { { "discard", ZFS_ACL_DISCARD }, { "noallow", ZFS_ACL_NOALLOW }, { "restricted", ZFS_ACL_RESTRICTED }, { "passthrough", ZFS_ACL_PASSTHROUGH }, { "secure", ZFS_ACL_RESTRICTED }, /* bkwrd compatability */ { "passthrough-x", ZFS_ACL_PASSTHROUGH_X }, { NULL } }; static zprop_index_t case_table[] = { { "sensitive", ZFS_CASE_SENSITIVE }, { "insensitive", ZFS_CASE_INSENSITIVE }, { "mixed", ZFS_CASE_MIXED }, { NULL } }; static zprop_index_t copies_table[] = { { "1", 1 }, { "2", 2 }, { "3", 3 }, { NULL } }; /* * Use the unique flags we have to send to u8_strcmp() and/or * u8_textprep() to represent the various normalization property * values. */ static zprop_index_t normalize_table[] = { { "none", 0 }, { "formD", U8_TEXTPREP_NFD }, { "formKC", U8_TEXTPREP_NFKC }, { "formC", U8_TEXTPREP_NFC }, { "formKD", U8_TEXTPREP_NFKD }, { NULL } }; static zprop_index_t version_table[] = { { "1", 1 }, { "2", 2 }, { "3", 3 }, { "4", 4 }, { "5", 5 }, { "current", ZPL_VERSION }, { NULL } }; static zprop_index_t boolean_table[] = { { "off", 0 }, { "on", 1 }, { NULL } }; static zprop_index_t logbias_table[] = { { "latency", ZFS_LOGBIAS_LATENCY }, { "throughput", ZFS_LOGBIAS_THROUGHPUT }, { NULL } }; static zprop_index_t canmount_table[] = { { "off", ZFS_CANMOUNT_OFF }, { "on", ZFS_CANMOUNT_ON }, { "noauto", ZFS_CANMOUNT_NOAUTO }, { NULL } }; static zprop_index_t cache_table[] = { { "none", ZFS_CACHE_NONE }, { "metadata", ZFS_CACHE_METADATA }, { "all", ZFS_CACHE_ALL }, { NULL } }; static zprop_index_t sync_table[] = { { "standard", ZFS_SYNC_STANDARD }, { "always", ZFS_SYNC_ALWAYS }, { "disabled", ZFS_SYNC_DISABLED }, { NULL } }; static zprop_index_t volmode_table[] = { { "default", ZFS_VOLMODE_DEFAULT }, { "geom", ZFS_VOLMODE_GEOM }, { "dev", ZFS_VOLMODE_DEV }, { "none", ZFS_VOLMODE_NONE }, { NULL } }; + static zprop_index_t redundant_metadata_table[] = { + { "all", ZFS_REDUNDANT_METADATA_ALL }, + { "most", ZFS_REDUNDANT_METADATA_MOST }, + { NULL } + }; + /* inherit index properties */ + zprop_register_index(ZFS_PROP_REDUNDANT_METADATA, "redundant_metadata", + ZFS_REDUNDANT_METADATA_ALL, + PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, + "all | most", "REDUND_MD", + redundant_metadata_table); zprop_register_index(ZFS_PROP_SYNC, "sync", ZFS_SYNC_STANDARD, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "standard | always | disabled", "SYNC", sync_table); zprop_register_index(ZFS_PROP_CHECKSUM, "checksum", ZIO_CHECKSUM_DEFAULT, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "on | off | fletcher2 | fletcher4 | sha256", "CHECKSUM", checksum_table); zprop_register_index(ZFS_PROP_DEDUP, "dedup", ZIO_CHECKSUM_OFF, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "on | off | verify | sha256[,verify]", "DEDUP", dedup_table); zprop_register_index(ZFS_PROP_COMPRESSION, "compression", ZIO_COMPRESS_DEFAULT, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "on | off | lzjb | gzip | gzip-[1-9] | zle | lz4", "COMPRESS", compress_table); zprop_register_index(ZFS_PROP_SNAPDIR, "snapdir", ZFS_SNAPDIR_HIDDEN, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "hidden | visible", "SNAPDIR", snapdir_table); zprop_register_index(ZFS_PROP_ACLMODE, "aclmode", ZFS_ACL_DISCARD, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "discard | groupmask | passthrough | restricted", "ACLMODE", acl_mode_table); zprop_register_index(ZFS_PROP_ACLINHERIT, "aclinherit", ZFS_ACL_RESTRICTED, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "discard | noallow | restricted | passthrough | passthrough-x", "ACLINHERIT", acl_inherit_table); zprop_register_index(ZFS_PROP_COPIES, "copies", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "1 | 2 | 3", "COPIES", copies_table); zprop_register_index(ZFS_PROP_PRIMARYCACHE, "primarycache", ZFS_CACHE_ALL, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT | ZFS_TYPE_VOLUME, "all | none | metadata", "PRIMARYCACHE", cache_table); zprop_register_index(ZFS_PROP_SECONDARYCACHE, "secondarycache", ZFS_CACHE_ALL, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT | ZFS_TYPE_VOLUME, "all | none | metadata", "SECONDARYCACHE", cache_table); zprop_register_index(ZFS_PROP_LOGBIAS, "logbias", ZFS_LOGBIAS_LATENCY, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "latency | throughput", "LOGBIAS", logbias_table); zprop_register_index(ZFS_PROP_VOLMODE, "volmode", ZFS_VOLMODE_DEFAULT, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT | ZFS_TYPE_VOLUME, "default | geom | dev | none", "VOLMODE", volmode_table); /* inherit index (boolean) properties */ zprop_register_index(ZFS_PROP_ATIME, "atime", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "on | off", "ATIME", boolean_table); zprop_register_index(ZFS_PROP_DEVICES, "devices", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "DEVICES", boolean_table); zprop_register_index(ZFS_PROP_EXEC, "exec", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "EXEC", boolean_table); zprop_register_index(ZFS_PROP_SETUID, "setuid", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "SETUID", boolean_table); zprop_register_index(ZFS_PROP_READONLY, "readonly", 0, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "on | off", "RDONLY", boolean_table); zprop_register_index(ZFS_PROP_ZONED, "jailed", 0, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "on | off", "JAILED", boolean_table); zprop_register_index(ZFS_PROP_XATTR, "xattr", 1, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "XATTR", boolean_table); zprop_register_index(ZFS_PROP_VSCAN, "vscan", 0, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "on | off", "VSCAN", boolean_table); zprop_register_index(ZFS_PROP_NBMAND, "nbmand", 0, PROP_INHERIT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "NBMAND", boolean_table); /* default index properties */ zprop_register_index(ZFS_PROP_VERSION, "version", 0, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "1 | 2 | 3 | 4 | 5 | current", "VERSION", version_table); zprop_register_index(ZFS_PROP_CANMOUNT, "canmount", ZFS_CANMOUNT_ON, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM, "on | off | noauto", "CANMOUNT", canmount_table); /* readonly index (boolean) properties */ zprop_register_index(ZFS_PROP_MOUNTED, "mounted", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM, "yes | no", "MOUNTED", boolean_table); zprop_register_index(ZFS_PROP_DEFER_DESTROY, "defer_destroy", 0, PROP_READONLY, ZFS_TYPE_SNAPSHOT, "yes | no", "DEFER_DESTROY", boolean_table); /* set once index properties */ zprop_register_index(ZFS_PROP_NORMALIZE, "normalization", 0, PROP_ONETIME, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "none | formC | formD | formKC | formKD", "NORMALIZATION", normalize_table); zprop_register_index(ZFS_PROP_CASE, "casesensitivity", ZFS_CASE_SENSITIVE, PROP_ONETIME, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "sensitive | insensitive | mixed", "CASE", case_table); /* set once index (boolean) properties */ zprop_register_index(ZFS_PROP_UTF8ONLY, "utf8only", 0, PROP_ONETIME, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_SNAPSHOT, "on | off", "UTF8ONLY", boolean_table); /* string properties */ zprop_register_string(ZFS_PROP_ORIGIN, "origin", NULL, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "ORIGIN"); zprop_register_string(ZFS_PROP_CLONES, "clones", NULL, PROP_READONLY, ZFS_TYPE_SNAPSHOT, "[,...]", "CLONES"); zprop_register_string(ZFS_PROP_MOUNTPOINT, "mountpoint", "/", PROP_INHERIT, ZFS_TYPE_FILESYSTEM, " | legacy | none", "MOUNTPOINT"); zprop_register_string(ZFS_PROP_SHARENFS, "sharenfs", "off", PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "on | off | share(1M) options", "SHARENFS"); zprop_register_string(ZFS_PROP_TYPE, "type", NULL, PROP_READONLY, ZFS_TYPE_DATASET | ZFS_TYPE_BOOKMARK, "filesystem | volume | snapshot | bookmark", "TYPE"); zprop_register_string(ZFS_PROP_SHARESMB, "sharesmb", "off", PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "on | off | sharemgr(1M) options", "SHARESMB"); zprop_register_string(ZFS_PROP_MLSLABEL, "mlslabel", ZFS_MLSLABEL_DEFAULT, PROP_INHERIT, ZFS_TYPE_DATASET, "", "MLSLABEL"); /* readonly number properties */ zprop_register_number(ZFS_PROP_USED, "used", 0, PROP_READONLY, ZFS_TYPE_DATASET, "", "USED"); zprop_register_number(ZFS_PROP_AVAILABLE, "available", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "AVAIL"); zprop_register_number(ZFS_PROP_REFERENCED, "referenced", 0, PROP_READONLY, ZFS_TYPE_DATASET, "", "REFER"); zprop_register_number(ZFS_PROP_COMPRESSRATIO, "compressratio", 0, PROP_READONLY, ZFS_TYPE_DATASET, "<1.00x or higher if compressed>", "RATIO"); zprop_register_number(ZFS_PROP_REFRATIO, "refcompressratio", 0, PROP_READONLY, ZFS_TYPE_DATASET, "<1.00x or higher if compressed>", "REFRATIO"); zprop_register_number(ZFS_PROP_VOLBLOCKSIZE, "volblocksize", ZVOL_DEFAULT_BLOCKSIZE, PROP_ONETIME, ZFS_TYPE_VOLUME, "512 to 128k, power of 2", "VOLBLOCK"); zprop_register_number(ZFS_PROP_USEDSNAP, "usedbysnapshots", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "USEDSNAP"); zprop_register_number(ZFS_PROP_USEDDS, "usedbydataset", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "USEDDS"); zprop_register_number(ZFS_PROP_USEDCHILD, "usedbychildren", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "USEDCHILD"); zprop_register_number(ZFS_PROP_USEDREFRESERV, "usedbyrefreservation", 0, PROP_READONLY, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "USEDREFRESERV"); zprop_register_number(ZFS_PROP_USERREFS, "userrefs", 0, PROP_READONLY, ZFS_TYPE_SNAPSHOT, "", "USERREFS"); zprop_register_number(ZFS_PROP_WRITTEN, "written", 0, PROP_READONLY, ZFS_TYPE_DATASET, "", "WRITTEN"); zprop_register_number(ZFS_PROP_LOGICALUSED, "logicalused", 0, PROP_READONLY, ZFS_TYPE_DATASET, "", "LUSED"); zprop_register_number(ZFS_PROP_LOGICALREFERENCED, "logicalreferenced", 0, PROP_READONLY, ZFS_TYPE_DATASET, "", "LREFER"); /* default number properties */ zprop_register_number(ZFS_PROP_QUOTA, "quota", 0, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM, " | none", "QUOTA"); zprop_register_number(ZFS_PROP_RESERVATION, "reservation", 0, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, " | none", "RESERV"); zprop_register_number(ZFS_PROP_VOLSIZE, "volsize", 0, PROP_DEFAULT, ZFS_TYPE_VOLUME, "", "VOLSIZE"); zprop_register_number(ZFS_PROP_REFQUOTA, "refquota", 0, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM, " | none", "REFQUOTA"); zprop_register_number(ZFS_PROP_REFRESERVATION, "refreservation", 0, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, " | none", "REFRESERV"); zprop_register_number(ZFS_PROP_FILESYSTEM_LIMIT, "filesystem_limit", UINT64_MAX, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM, " | none", "FSLIMIT"); zprop_register_number(ZFS_PROP_SNAPSHOT_LIMIT, "snapshot_limit", UINT64_MAX, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, " | none", "SSLIMIT"); zprop_register_number(ZFS_PROP_FILESYSTEM_COUNT, "filesystem_count", UINT64_MAX, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM, "", "FSCOUNT"); zprop_register_number(ZFS_PROP_SNAPSHOT_COUNT, "snapshot_count", UINT64_MAX, PROP_DEFAULT, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME, "", "SSCOUNT"); /* inherit number properties */ zprop_register_number(ZFS_PROP_RECORDSIZE, "recordsize", SPA_MAXBLOCKSIZE, PROP_INHERIT, ZFS_TYPE_FILESYSTEM, "512 to 128k, power of 2", "RECSIZE"); /* hidden properties */ zprop_register_hidden(ZFS_PROP_CREATETXG, "createtxg", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET | ZFS_TYPE_BOOKMARK, "CREATETXG"); zprop_register_hidden(ZFS_PROP_NUMCLONES, "numclones", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_SNAPSHOT, "NUMCLONES"); zprop_register_hidden(ZFS_PROP_NAME, "name", PROP_TYPE_STRING, PROP_READONLY, ZFS_TYPE_DATASET | ZFS_TYPE_BOOKMARK, "NAME"); zprop_register_hidden(ZFS_PROP_ISCSIOPTIONS, "iscsioptions", PROP_TYPE_STRING, PROP_INHERIT, ZFS_TYPE_VOLUME, "ISCSIOPTIONS"); zprop_register_hidden(ZFS_PROP_STMF_SHAREINFO, "stmf_sbd_lu", PROP_TYPE_STRING, PROP_INHERIT, ZFS_TYPE_VOLUME, "STMF_SBD_LU"); zprop_register_hidden(ZFS_PROP_GUID, "guid", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET | ZFS_TYPE_BOOKMARK, "GUID"); zprop_register_hidden(ZFS_PROP_USERACCOUNTING, "useraccounting", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET, "USERACCOUNTING"); zprop_register_hidden(ZFS_PROP_UNIQUE, "unique", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET, "UNIQUE"); zprop_register_hidden(ZFS_PROP_OBJSETID, "objsetid", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET, "OBJSETID"); zprop_register_hidden(ZFS_PROP_INCONSISTENT, "inconsistent", PROP_TYPE_NUMBER, PROP_READONLY, ZFS_TYPE_DATASET, "INCONSISTENT"); /* oddball properties */ zprop_register_impl(ZFS_PROP_CREATION, "creation", PROP_TYPE_NUMBER, 0, NULL, PROP_READONLY, ZFS_TYPE_DATASET | ZFS_TYPE_BOOKMARK, "", "CREATION", B_FALSE, B_TRUE, NULL); } boolean_t zfs_prop_delegatable(zfs_prop_t prop) { zprop_desc_t *pd = &zfs_prop_table[prop]; /* The mlslabel property is never delegatable. */ if (prop == ZFS_PROP_MLSLABEL) return (B_FALSE); return (pd->pd_attr != PROP_READONLY); } /* * Given a zfs dataset property name, returns the corresponding property ID. */ zfs_prop_t zfs_name_to_prop(const char *propname) { return (zprop_name_to_prop(propname, ZFS_TYPE_DATASET)); } /* * For user property names, we allow all lowercase alphanumeric characters, plus * a few useful punctuation characters. */ static int valid_char(char c) { return ((c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || c == '-' || c == '_' || c == '.' || c == ':'); } /* * Returns true if this is a valid user-defined property (one with a ':'). */ boolean_t zfs_prop_user(const char *name) { int i; char c; boolean_t foundsep = B_FALSE; for (i = 0; i < strlen(name); i++) { c = name[i]; if (!valid_char(c)) return (B_FALSE); if (c == ':') foundsep = B_TRUE; } if (!foundsep) return (B_FALSE); return (B_TRUE); } /* * Returns true if this is a valid userspace-type property (one with a '@'). * Note that after the @, any character is valid (eg, another @, for SID * user@domain). */ boolean_t zfs_prop_userquota(const char *name) { zfs_userquota_prop_t prop; for (prop = 0; prop < ZFS_NUM_USERQUOTA_PROPS; prop++) { if (strncmp(name, zfs_userquota_prop_prefixes[prop], strlen(zfs_userquota_prop_prefixes[prop])) == 0) { return (B_TRUE); } } return (B_FALSE); } /* * Returns true if this is a valid written@ property. * Note that after the @, any character is valid (eg, another @, for * written@pool/fs@origin). */ boolean_t zfs_prop_written(const char *name) { static const char *prefix = "written@"; return (strncmp(name, prefix, strlen(prefix)) == 0); } /* * Tables of index types, plus functions to convert between the user view * (strings) and internal representation (uint64_t). */ int zfs_prop_string_to_index(zfs_prop_t prop, const char *string, uint64_t *index) { return (zprop_string_to_index(prop, string, index, ZFS_TYPE_DATASET)); } int zfs_prop_index_to_string(zfs_prop_t prop, uint64_t index, const char **string) { return (zprop_index_to_string(prop, index, string, ZFS_TYPE_DATASET)); } uint64_t zfs_prop_random_value(zfs_prop_t prop, uint64_t seed) { return (zprop_random_value(prop, seed, ZFS_TYPE_DATASET)); } /* * Returns TRUE if the property applies to any of the given dataset types. */ boolean_t zfs_prop_valid_for_type(int prop, zfs_type_t types) { return (zprop_valid_for_type(prop, types)); } zprop_type_t zfs_prop_get_type(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_proptype); } /* * Returns TRUE if the property is readonly. */ boolean_t zfs_prop_readonly(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_attr == PROP_READONLY || zfs_prop_table[prop].pd_attr == PROP_ONETIME); } /* * Returns TRUE if the property is only allowed to be set once. */ boolean_t zfs_prop_setonce(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_attr == PROP_ONETIME); } const char * zfs_prop_default_string(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_strdefault); } uint64_t zfs_prop_default_numeric(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_numdefault); } /* * Given a dataset property ID, returns the corresponding name. * Assuming the zfs dataset property ID is valid. */ const char * zfs_prop_to_name(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_name); } /* * Returns TRUE if the property is inheritable. */ boolean_t zfs_prop_inheritable(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_attr == PROP_INHERIT || zfs_prop_table[prop].pd_attr == PROP_ONETIME); } #ifndef _KERNEL /* * Returns a string describing the set of acceptable values for the given * zfs property, or NULL if it cannot be set. */ const char * zfs_prop_values(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_values); } /* * Returns TRUE if this property is a string type. Note that index types * (compression, checksum) are treated as strings in userland, even though they * are stored numerically on disk. */ int zfs_prop_is_string(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_proptype == PROP_TYPE_STRING || zfs_prop_table[prop].pd_proptype == PROP_TYPE_INDEX); } /* * Returns the column header for the given property. Used only in * 'zfs list -o', but centralized here with the other property information. */ const char * zfs_prop_column_name(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_colname); } /* * Returns whether the given property should be displayed right-justified for * 'zfs list'. */ boolean_t zfs_prop_align_right(zfs_prop_t prop) { return (zfs_prop_table[prop].pd_rightalign); } #endif Index: stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c (revision 268647) @@ -1,1912 +1,1925 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2012, 2014 by Delphix. All rights reserved. */ /* Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ /* Copyright (c) 2013, Joyent, Inc. All rights reserved. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef _KERNEL #include #include #endif /* * Enable/disable nopwrite feature. */ int zfs_nopwrite_enabled = 1; SYSCTL_DECL(_vfs_zfs); TUNABLE_INT("vfs.zfs.nopwrite_enabled", &zfs_nopwrite_enabled); SYSCTL_INT(_vfs_zfs, OID_AUTO, nopwrite_enabled, CTLFLAG_RDTUN, &zfs_nopwrite_enabled, 0, "Enable nopwrite feature"); const dmu_object_type_info_t dmu_ot[DMU_OT_NUMTYPES] = { { DMU_BSWAP_UINT8, TRUE, "unallocated" }, { DMU_BSWAP_ZAP, TRUE, "object directory" }, { DMU_BSWAP_UINT64, TRUE, "object array" }, { DMU_BSWAP_UINT8, TRUE, "packed nvlist" }, { DMU_BSWAP_UINT64, TRUE, "packed nvlist size" }, { DMU_BSWAP_UINT64, TRUE, "bpobj" }, { DMU_BSWAP_UINT64, TRUE, "bpobj header" }, { DMU_BSWAP_UINT64, TRUE, "SPA space map header" }, { DMU_BSWAP_UINT64, TRUE, "SPA space map" }, { DMU_BSWAP_UINT64, TRUE, "ZIL intent log" }, { DMU_BSWAP_DNODE, TRUE, "DMU dnode" }, { DMU_BSWAP_OBJSET, TRUE, "DMU objset" }, { DMU_BSWAP_UINT64, TRUE, "DSL directory" }, { DMU_BSWAP_ZAP, TRUE, "DSL directory child map"}, { DMU_BSWAP_ZAP, TRUE, "DSL dataset snap map" }, { DMU_BSWAP_ZAP, TRUE, "DSL props" }, { DMU_BSWAP_UINT64, TRUE, "DSL dataset" }, { DMU_BSWAP_ZNODE, TRUE, "ZFS znode" }, { DMU_BSWAP_OLDACL, TRUE, "ZFS V0 ACL" }, { DMU_BSWAP_UINT8, FALSE, "ZFS plain file" }, { DMU_BSWAP_ZAP, TRUE, "ZFS directory" }, { DMU_BSWAP_ZAP, TRUE, "ZFS master node" }, { DMU_BSWAP_ZAP, TRUE, "ZFS delete queue" }, { DMU_BSWAP_UINT8, FALSE, "zvol object" }, { DMU_BSWAP_ZAP, TRUE, "zvol prop" }, { DMU_BSWAP_UINT8, FALSE, "other uint8[]" }, { DMU_BSWAP_UINT64, FALSE, "other uint64[]" }, { DMU_BSWAP_ZAP, TRUE, "other ZAP" }, { DMU_BSWAP_ZAP, TRUE, "persistent error log" }, { DMU_BSWAP_UINT8, TRUE, "SPA history" }, { DMU_BSWAP_UINT64, TRUE, "SPA history offsets" }, { DMU_BSWAP_ZAP, TRUE, "Pool properties" }, { DMU_BSWAP_ZAP, TRUE, "DSL permissions" }, { DMU_BSWAP_ACL, TRUE, "ZFS ACL" }, { DMU_BSWAP_UINT8, TRUE, "ZFS SYSACL" }, { DMU_BSWAP_UINT8, TRUE, "FUID table" }, { DMU_BSWAP_UINT64, TRUE, "FUID table size" }, { DMU_BSWAP_ZAP, TRUE, "DSL dataset next clones"}, { DMU_BSWAP_ZAP, TRUE, "scan work queue" }, { DMU_BSWAP_ZAP, TRUE, "ZFS user/group used" }, { DMU_BSWAP_ZAP, TRUE, "ZFS user/group quota" }, { DMU_BSWAP_ZAP, TRUE, "snapshot refcount tags"}, { DMU_BSWAP_ZAP, TRUE, "DDT ZAP algorithm" }, { DMU_BSWAP_ZAP, TRUE, "DDT statistics" }, { DMU_BSWAP_UINT8, TRUE, "System attributes" }, { DMU_BSWAP_ZAP, TRUE, "SA master node" }, { DMU_BSWAP_ZAP, TRUE, "SA attr registration" }, { DMU_BSWAP_ZAP, TRUE, "SA attr layouts" }, { DMU_BSWAP_ZAP, TRUE, "scan translations" }, { DMU_BSWAP_UINT8, FALSE, "deduplicated block" }, { DMU_BSWAP_ZAP, TRUE, "DSL deadlist map" }, { DMU_BSWAP_UINT64, TRUE, "DSL deadlist map hdr" }, { DMU_BSWAP_ZAP, TRUE, "DSL dir clones" }, { DMU_BSWAP_UINT64, TRUE, "bpobj subobj" } }; const dmu_object_byteswap_info_t dmu_ot_byteswap[DMU_BSWAP_NUMFUNCS] = { { byteswap_uint8_array, "uint8" }, { byteswap_uint16_array, "uint16" }, { byteswap_uint32_array, "uint32" }, { byteswap_uint64_array, "uint64" }, { zap_byteswap, "zap" }, { dnode_buf_byteswap, "dnode" }, { dmu_objset_byteswap, "objset" }, { zfs_znode_byteswap, "znode" }, { zfs_oldacl_byteswap, "oldacl" }, { zfs_acl_byteswap, "acl" } }; int dmu_buf_hold(objset_t *os, uint64_t object, uint64_t offset, void *tag, dmu_buf_t **dbp, int flags) { dnode_t *dn; uint64_t blkid; dmu_buf_impl_t *db; int err; int db_flags = DB_RF_CANFAIL; if (flags & DMU_READ_NO_PREFETCH) db_flags |= DB_RF_NOPREFETCH; err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); blkid = dbuf_whichblock(dn, offset); rw_enter(&dn->dn_struct_rwlock, RW_READER); db = dbuf_hold(dn, blkid, tag); rw_exit(&dn->dn_struct_rwlock); if (db == NULL) { err = SET_ERROR(EIO); } else { err = dbuf_read(db, NULL, db_flags); if (err) { dbuf_rele(db, tag); db = NULL; } } dnode_rele(dn, FTAG); *dbp = &db->db; /* NULL db plus first field offset is NULL */ return (err); } int dmu_bonus_max(void) { return (DN_MAX_BONUSLEN); } int dmu_set_bonus(dmu_buf_t *db_fake, int newsize, dmu_tx_t *tx) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; dnode_t *dn; int error; DB_DNODE_ENTER(db); dn = DB_DNODE(db); if (dn->dn_bonus != db) { error = SET_ERROR(EINVAL); } else if (newsize < 0 || newsize > db_fake->db_size) { error = SET_ERROR(EINVAL); } else { dnode_setbonuslen(dn, newsize, tx); error = 0; } DB_DNODE_EXIT(db); return (error); } int dmu_set_bonustype(dmu_buf_t *db_fake, dmu_object_type_t type, dmu_tx_t *tx) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; dnode_t *dn; int error; DB_DNODE_ENTER(db); dn = DB_DNODE(db); if (!DMU_OT_IS_VALID(type)) { error = SET_ERROR(EINVAL); } else if (dn->dn_bonus != db) { error = SET_ERROR(EINVAL); } else { dnode_setbonus_type(dn, type, tx); error = 0; } DB_DNODE_EXIT(db); return (error); } dmu_object_type_t dmu_get_bonustype(dmu_buf_t *db_fake) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; dnode_t *dn; dmu_object_type_t type; DB_DNODE_ENTER(db); dn = DB_DNODE(db); type = dn->dn_bonustype; DB_DNODE_EXIT(db); return (type); } int dmu_rm_spill(objset_t *os, uint64_t object, dmu_tx_t *tx) { dnode_t *dn; int error; error = dnode_hold(os, object, FTAG, &dn); dbuf_rm_spill(dn, tx); rw_enter(&dn->dn_struct_rwlock, RW_WRITER); dnode_rm_spill(dn, tx); rw_exit(&dn->dn_struct_rwlock); dnode_rele(dn, FTAG); return (error); } /* * returns ENOENT, EIO, or 0. */ int dmu_bonus_hold(objset_t *os, uint64_t object, void *tag, dmu_buf_t **dbp) { dnode_t *dn; dmu_buf_impl_t *db; int error; error = dnode_hold(os, object, FTAG, &dn); if (error) return (error); rw_enter(&dn->dn_struct_rwlock, RW_READER); if (dn->dn_bonus == NULL) { rw_exit(&dn->dn_struct_rwlock); rw_enter(&dn->dn_struct_rwlock, RW_WRITER); if (dn->dn_bonus == NULL) dbuf_create_bonus(dn); } db = dn->dn_bonus; /* as long as the bonus buf is held, the dnode will be held */ if (refcount_add(&db->db_holds, tag) == 1) { VERIFY(dnode_add_ref(dn, db)); (void) atomic_inc_32_nv(&dn->dn_dbufs_count); } /* * Wait to drop dn_struct_rwlock until after adding the bonus dbuf's * hold and incrementing the dbuf count to ensure that dnode_move() sees * a dnode hold for every dbuf. */ rw_exit(&dn->dn_struct_rwlock); dnode_rele(dn, FTAG); VERIFY(0 == dbuf_read(db, NULL, DB_RF_MUST_SUCCEED | DB_RF_NOPREFETCH)); *dbp = &db->db; return (0); } /* * returns ENOENT, EIO, or 0. * * This interface will allocate a blank spill dbuf when a spill blk * doesn't already exist on the dnode. * * if you only want to find an already existing spill db, then * dmu_spill_hold_existing() should be used. */ int dmu_spill_hold_by_dnode(dnode_t *dn, uint32_t flags, void *tag, dmu_buf_t **dbp) { dmu_buf_impl_t *db = NULL; int err; if ((flags & DB_RF_HAVESTRUCT) == 0) rw_enter(&dn->dn_struct_rwlock, RW_READER); db = dbuf_hold(dn, DMU_SPILL_BLKID, tag); if ((flags & DB_RF_HAVESTRUCT) == 0) rw_exit(&dn->dn_struct_rwlock); ASSERT(db != NULL); err = dbuf_read(db, NULL, flags); if (err == 0) *dbp = &db->db; else dbuf_rele(db, tag); return (err); } int dmu_spill_hold_existing(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)bonus; dnode_t *dn; int err; DB_DNODE_ENTER(db); dn = DB_DNODE(db); if (spa_version(dn->dn_objset->os_spa) < SPA_VERSION_SA) { err = SET_ERROR(EINVAL); } else { rw_enter(&dn->dn_struct_rwlock, RW_READER); if (!dn->dn_have_spill) { err = SET_ERROR(ENOENT); } else { err = dmu_spill_hold_by_dnode(dn, DB_RF_HAVESTRUCT | DB_RF_CANFAIL, tag, dbp); } rw_exit(&dn->dn_struct_rwlock); } DB_DNODE_EXIT(db); return (err); } int dmu_spill_hold_by_bonus(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)bonus; dnode_t *dn; int err; DB_DNODE_ENTER(db); dn = DB_DNODE(db); err = dmu_spill_hold_by_dnode(dn, DB_RF_CANFAIL, tag, dbp); DB_DNODE_EXIT(db); return (err); } /* * Note: longer-term, we should modify all of the dmu_buf_*() interfaces * to take a held dnode rather than -- the lookup is wasteful, * and can induce severe lock contention when writing to several files * whose dnodes are in the same block. */ static int dmu_buf_hold_array_by_dnode(dnode_t *dn, uint64_t offset, uint64_t length, int read, void *tag, int *numbufsp, dmu_buf_t ***dbpp, uint32_t flags) { dmu_buf_t **dbp; uint64_t blkid, nblks, i; uint32_t dbuf_flags; int err; zio_t *zio; ASSERT(length <= DMU_MAX_ACCESS); dbuf_flags = DB_RF_CANFAIL | DB_RF_NEVERWAIT | DB_RF_HAVESTRUCT; if (flags & DMU_READ_NO_PREFETCH || length > zfetch_array_rd_sz) dbuf_flags |= DB_RF_NOPREFETCH; rw_enter(&dn->dn_struct_rwlock, RW_READER); if (dn->dn_datablkshift) { int blkshift = dn->dn_datablkshift; nblks = (P2ROUNDUP(offset+length, 1ULL<> blkshift; } else { if (offset + length > dn->dn_datablksz) { zfs_panic_recover("zfs: accessing past end of object " "%llx/%llx (size=%u access=%llu+%llu)", (longlong_t)dn->dn_objset-> os_dsl_dataset->ds_object, (longlong_t)dn->dn_object, dn->dn_datablksz, (longlong_t)offset, (longlong_t)length); rw_exit(&dn->dn_struct_rwlock); return (SET_ERROR(EIO)); } nblks = 1; } dbp = kmem_zalloc(sizeof (dmu_buf_t *) * nblks, KM_SLEEP); zio = zio_root(dn->dn_objset->os_spa, NULL, NULL, ZIO_FLAG_CANFAIL); blkid = dbuf_whichblock(dn, offset); for (i = 0; i < nblks; i++) { dmu_buf_impl_t *db = dbuf_hold(dn, blkid+i, tag); if (db == NULL) { rw_exit(&dn->dn_struct_rwlock); dmu_buf_rele_array(dbp, nblks, tag); zio_nowait(zio); return (SET_ERROR(EIO)); } /* initiate async i/o */ if (read) (void) dbuf_read(db, zio, dbuf_flags); #ifdef _KERNEL else curthread->td_ru.ru_oublock++; #endif dbp[i] = &db->db; } rw_exit(&dn->dn_struct_rwlock); /* wait for async i/o */ err = zio_wait(zio); if (err) { dmu_buf_rele_array(dbp, nblks, tag); return (err); } /* wait for other io to complete */ if (read) { for (i = 0; i < nblks; i++) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)dbp[i]; mutex_enter(&db->db_mtx); while (db->db_state == DB_READ || db->db_state == DB_FILL) cv_wait(&db->db_changed, &db->db_mtx); if (db->db_state == DB_UNCACHED) err = SET_ERROR(EIO); mutex_exit(&db->db_mtx); if (err) { dmu_buf_rele_array(dbp, nblks, tag); return (err); } } } *numbufsp = nblks; *dbpp = dbp; return (0); } static int dmu_buf_hold_array(objset_t *os, uint64_t object, uint64_t offset, uint64_t length, int read, void *tag, int *numbufsp, dmu_buf_t ***dbpp) { dnode_t *dn; int err; err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); err = dmu_buf_hold_array_by_dnode(dn, offset, length, read, tag, numbufsp, dbpp, DMU_READ_PREFETCH); dnode_rele(dn, FTAG); return (err); } int dmu_buf_hold_array_by_bonus(dmu_buf_t *db_fake, uint64_t offset, uint64_t length, int read, void *tag, int *numbufsp, dmu_buf_t ***dbpp) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; dnode_t *dn; int err; DB_DNODE_ENTER(db); dn = DB_DNODE(db); err = dmu_buf_hold_array_by_dnode(dn, offset, length, read, tag, numbufsp, dbpp, DMU_READ_PREFETCH); DB_DNODE_EXIT(db); return (err); } void dmu_buf_rele_array(dmu_buf_t **dbp_fake, int numbufs, void *tag) { int i; dmu_buf_impl_t **dbp = (dmu_buf_impl_t **)dbp_fake; if (numbufs == 0) return; for (i = 0; i < numbufs; i++) { if (dbp[i]) dbuf_rele(dbp[i], tag); } kmem_free(dbp, sizeof (dmu_buf_t *) * numbufs); } /* * Issue prefetch i/os for the given blocks. * * Note: The assumption is that we *know* these blocks will be needed * almost immediately. Therefore, the prefetch i/os will be issued at * ZIO_PRIORITY_SYNC_READ * * Note: indirect blocks and other metadata will be read synchronously, * causing this function to block if they are not already cached. */ void dmu_prefetch(objset_t *os, uint64_t object, uint64_t offset, uint64_t len) { dnode_t *dn; uint64_t blkid; int nblks, err; if (zfs_prefetch_disable) return; if (len == 0) { /* they're interested in the bonus buffer */ dn = DMU_META_DNODE(os); if (object == 0 || object >= DN_MAX_OBJECT) return; rw_enter(&dn->dn_struct_rwlock, RW_READER); blkid = dbuf_whichblock(dn, object * sizeof (dnode_phys_t)); dbuf_prefetch(dn, blkid, ZIO_PRIORITY_SYNC_READ); rw_exit(&dn->dn_struct_rwlock); return; } /* * XXX - Note, if the dnode for the requested object is not * already cached, we will do a *synchronous* read in the * dnode_hold() call. The same is true for any indirects. */ err = dnode_hold(os, object, FTAG, &dn); if (err != 0) return; rw_enter(&dn->dn_struct_rwlock, RW_READER); if (dn->dn_datablkshift) { int blkshift = dn->dn_datablkshift; nblks = (P2ROUNDUP(offset + len, 1 << blkshift) - P2ALIGN(offset, 1 << blkshift)) >> blkshift; } else { nblks = (offset < dn->dn_datablksz); } if (nblks != 0) { blkid = dbuf_whichblock(dn, offset); for (int i = 0; i < nblks; i++) dbuf_prefetch(dn, blkid + i, ZIO_PRIORITY_SYNC_READ); } rw_exit(&dn->dn_struct_rwlock); dnode_rele(dn, FTAG); } /* * Get the next "chunk" of file data to free. We traverse the file from * the end so that the file gets shorter over time (if we crashes in the * middle, this will leave us in a better state). We find allocated file * data by simply searching the allocated level 1 indirects. * * On input, *start should be the first offset that does not need to be * freed (e.g. "offset + length"). On return, *start will be the first * offset that should be freed. */ static int get_next_chunk(dnode_t *dn, uint64_t *start, uint64_t minimum) { uint64_t maxblks = DMU_MAX_ACCESS >> (dn->dn_indblkshift + 1); /* bytes of data covered by a level-1 indirect block */ uint64_t iblkrange = dn->dn_datablksz * EPB(dn->dn_indblkshift, SPA_BLKPTRSHIFT); ASSERT3U(minimum, <=, *start); if (*start - minimum <= iblkrange * maxblks) { *start = minimum; return (0); } ASSERT(ISP2(iblkrange)); for (uint64_t blks = 0; *start > minimum && blks < maxblks; blks++) { int err; /* * dnode_next_offset(BACKWARDS) will find an allocated L1 * indirect block at or before the input offset. We must * decrement *start so that it is at the end of the region * to search. */ (*start)--; err = dnode_next_offset(dn, DNODE_FIND_BACKWARDS, start, 2, 1, 0); /* if there are no indirect blocks before start, we are done */ if (err == ESRCH) { *start = minimum; break; } else if (err != 0) { return (err); } /* set start to the beginning of this L1 indirect */ *start = P2ALIGN(*start, iblkrange); } if (*start < minimum) *start = minimum; return (0); } static int dmu_free_long_range_impl(objset_t *os, dnode_t *dn, uint64_t offset, uint64_t length) { uint64_t object_size = (dn->dn_maxblkid + 1) * dn->dn_datablksz; int err; if (offset >= object_size) return (0); if (length == DMU_OBJECT_END || offset + length > object_size) length = object_size - offset; while (length != 0) { uint64_t chunk_end, chunk_begin; chunk_end = chunk_begin = offset + length; /* move chunk_begin backwards to the beginning of this chunk */ err = get_next_chunk(dn, &chunk_begin, offset); if (err) return (err); ASSERT3U(chunk_begin, >=, offset); ASSERT3U(chunk_begin, <=, chunk_end); dmu_tx_t *tx = dmu_tx_create(os); dmu_tx_hold_free(tx, dn->dn_object, chunk_begin, chunk_end - chunk_begin); err = dmu_tx_assign(tx, TXG_WAIT); if (err) { dmu_tx_abort(tx); return (err); } dnode_free_range(dn, chunk_begin, chunk_end - chunk_begin, tx); dmu_tx_commit(tx); length -= chunk_end - chunk_begin; } return (0); } int dmu_free_long_range(objset_t *os, uint64_t object, uint64_t offset, uint64_t length) { dnode_t *dn; int err; err = dnode_hold(os, object, FTAG, &dn); if (err != 0) return (err); err = dmu_free_long_range_impl(os, dn, offset, length); /* * It is important to zero out the maxblkid when freeing the entire * file, so that (a) subsequent calls to dmu_free_long_range_impl() * will take the fast path, and (b) dnode_reallocate() can verify * that the entire file has been freed. */ if (err == 0 && offset == 0 && length == DMU_OBJECT_END) dn->dn_maxblkid = 0; dnode_rele(dn, FTAG); return (err); } int dmu_free_long_object(objset_t *os, uint64_t object) { dmu_tx_t *tx; int err; err = dmu_free_long_range(os, object, 0, DMU_OBJECT_END); if (err != 0) return (err); tx = dmu_tx_create(os); dmu_tx_hold_bonus(tx, object); dmu_tx_hold_free(tx, object, 0, DMU_OBJECT_END); err = dmu_tx_assign(tx, TXG_WAIT); if (err == 0) { err = dmu_object_free(os, object, tx); dmu_tx_commit(tx); } else { dmu_tx_abort(tx); } return (err); } int dmu_free_range(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, dmu_tx_t *tx) { dnode_t *dn; int err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); ASSERT(offset < UINT64_MAX); ASSERT(size == -1ULL || size <= UINT64_MAX - offset); dnode_free_range(dn, offset, size, tx); dnode_rele(dn, FTAG); return (0); } int dmu_read(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, void *buf, uint32_t flags) { dnode_t *dn; dmu_buf_t **dbp; int numbufs, err; err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); /* * Deal with odd block sizes, where there can't be data past the first * block. If we ever do the tail block optimization, we will need to * handle that here as well. */ if (dn->dn_maxblkid == 0) { int newsz = offset > dn->dn_datablksz ? 0 : MIN(size, dn->dn_datablksz - offset); bzero((char *)buf + newsz, size - newsz); size = newsz; } while (size > 0) { uint64_t mylen = MIN(size, DMU_MAX_ACCESS / 2); int i; /* * NB: we could do this block-at-a-time, but it's nice * to be reading in parallel. */ err = dmu_buf_hold_array_by_dnode(dn, offset, mylen, TRUE, FTAG, &numbufs, &dbp, flags); if (err) break; for (i = 0; i < numbufs; i++) { int tocpy; int bufoff; dmu_buf_t *db = dbp[i]; ASSERT(size > 0); bufoff = offset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); bcopy((char *)db->db_data + bufoff, buf, tocpy); offset += tocpy; size -= tocpy; buf = (char *)buf + tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); } dnode_rele(dn, FTAG); return (err); } void dmu_write(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, const void *buf, dmu_tx_t *tx) { dmu_buf_t **dbp; int numbufs, i; if (size == 0) return; VERIFY(0 == dmu_buf_hold_array(os, object, offset, size, FALSE, FTAG, &numbufs, &dbp)); for (i = 0; i < numbufs; i++) { int tocpy; int bufoff; dmu_buf_t *db = dbp[i]; ASSERT(size > 0); bufoff = offset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size); if (tocpy == db->db_size) dmu_buf_will_fill(db, tx); else dmu_buf_will_dirty(db, tx); bcopy(buf, (char *)db->db_data + bufoff, tocpy); if (tocpy == db->db_size) dmu_buf_fill_done(db, tx); offset += tocpy; size -= tocpy; buf = (char *)buf + tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); } void dmu_prealloc(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, dmu_tx_t *tx) { dmu_buf_t **dbp; int numbufs, i; if (size == 0) return; VERIFY(0 == dmu_buf_hold_array(os, object, offset, size, FALSE, FTAG, &numbufs, &dbp)); for (i = 0; i < numbufs; i++) { dmu_buf_t *db = dbp[i]; dmu_buf_will_not_fill(db, tx); } dmu_buf_rele_array(dbp, numbufs, FTAG); } /* * DMU support for xuio */ kstat_t *xuio_ksp = NULL; int dmu_xuio_init(xuio_t *xuio, int nblk) { dmu_xuio_t *priv; uio_t *uio = &xuio->xu_uio; uio->uio_iovcnt = nblk; uio->uio_iov = kmem_zalloc(nblk * sizeof (iovec_t), KM_SLEEP); priv = kmem_zalloc(sizeof (dmu_xuio_t), KM_SLEEP); priv->cnt = nblk; priv->bufs = kmem_zalloc(nblk * sizeof (arc_buf_t *), KM_SLEEP); priv->iovp = uio->uio_iov; XUIO_XUZC_PRIV(xuio) = priv; if (XUIO_XUZC_RW(xuio) == UIO_READ) XUIOSTAT_INCR(xuiostat_onloan_rbuf, nblk); else XUIOSTAT_INCR(xuiostat_onloan_wbuf, nblk); return (0); } void dmu_xuio_fini(xuio_t *xuio) { dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio); int nblk = priv->cnt; kmem_free(priv->iovp, nblk * sizeof (iovec_t)); kmem_free(priv->bufs, nblk * sizeof (arc_buf_t *)); kmem_free(priv, sizeof (dmu_xuio_t)); if (XUIO_XUZC_RW(xuio) == UIO_READ) XUIOSTAT_INCR(xuiostat_onloan_rbuf, -nblk); else XUIOSTAT_INCR(xuiostat_onloan_wbuf, -nblk); } /* * Initialize iov[priv->next] and priv->bufs[priv->next] with { off, n, abuf } * and increase priv->next by 1. */ int dmu_xuio_add(xuio_t *xuio, arc_buf_t *abuf, offset_t off, size_t n) { struct iovec *iov; uio_t *uio = &xuio->xu_uio; dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio); int i = priv->next++; ASSERT(i < priv->cnt); ASSERT(off + n <= arc_buf_size(abuf)); iov = uio->uio_iov + i; iov->iov_base = (char *)abuf->b_data + off; iov->iov_len = n; priv->bufs[i] = abuf; return (0); } int dmu_xuio_cnt(xuio_t *xuio) { dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio); return (priv->cnt); } arc_buf_t * dmu_xuio_arcbuf(xuio_t *xuio, int i) { dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio); ASSERT(i < priv->cnt); return (priv->bufs[i]); } void dmu_xuio_clear(xuio_t *xuio, int i) { dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio); ASSERT(i < priv->cnt); priv->bufs[i] = NULL; } static void xuio_stat_init(void) { xuio_ksp = kstat_create("zfs", 0, "xuio_stats", "misc", KSTAT_TYPE_NAMED, sizeof (xuio_stats) / sizeof (kstat_named_t), KSTAT_FLAG_VIRTUAL); if (xuio_ksp != NULL) { xuio_ksp->ks_data = &xuio_stats; kstat_install(xuio_ksp); } } static void xuio_stat_fini(void) { if (xuio_ksp != NULL) { kstat_delete(xuio_ksp); xuio_ksp = NULL; } } void xuio_stat_wbuf_copied() { XUIOSTAT_BUMP(xuiostat_wbuf_copied); } void xuio_stat_wbuf_nocopy() { XUIOSTAT_BUMP(xuiostat_wbuf_nocopy); } #ifdef _KERNEL int dmu_read_uio(objset_t *os, uint64_t object, uio_t *uio, uint64_t size) { dmu_buf_t **dbp; int numbufs, i, err; xuio_t *xuio = NULL; /* * NB: we could do this block-at-a-time, but it's nice * to be reading in parallel. */ err = dmu_buf_hold_array(os, object, uio->uio_loffset, size, TRUE, FTAG, &numbufs, &dbp); if (err) return (err); #ifdef UIO_XUIO if (uio->uio_extflg == UIO_XUIO) xuio = (xuio_t *)uio; #endif for (i = 0; i < numbufs; i++) { int tocpy; int bufoff; dmu_buf_t *db = dbp[i]; ASSERT(size > 0); bufoff = uio->uio_loffset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); if (xuio) { dmu_buf_impl_t *dbi = (dmu_buf_impl_t *)db; arc_buf_t *dbuf_abuf = dbi->db_buf; arc_buf_t *abuf = dbuf_loan_arcbuf(dbi); err = dmu_xuio_add(xuio, abuf, bufoff, tocpy); if (!err) { uio->uio_resid -= tocpy; uio->uio_loffset += tocpy; } if (abuf == dbuf_abuf) XUIOSTAT_BUMP(xuiostat_rbuf_nocopy); else XUIOSTAT_BUMP(xuiostat_rbuf_copied); } else { err = uiomove((char *)db->db_data + bufoff, tocpy, UIO_READ, uio); } if (err) break; size -= tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); return (err); } static int dmu_write_uio_dnode(dnode_t *dn, uio_t *uio, uint64_t size, dmu_tx_t *tx) { dmu_buf_t **dbp; int numbufs; int err = 0; int i; err = dmu_buf_hold_array_by_dnode(dn, uio->uio_loffset, size, FALSE, FTAG, &numbufs, &dbp, DMU_READ_PREFETCH); if (err) return (err); for (i = 0; i < numbufs; i++) { int tocpy; int bufoff; dmu_buf_t *db = dbp[i]; ASSERT(size > 0); bufoff = uio->uio_loffset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size); if (tocpy == db->db_size) dmu_buf_will_fill(db, tx); else dmu_buf_will_dirty(db, tx); /* * XXX uiomove could block forever (eg. nfs-backed * pages). There needs to be a uiolockdown() function * to lock the pages in memory, so that uiomove won't * block. */ err = uiomove((char *)db->db_data + bufoff, tocpy, UIO_WRITE, uio); if (tocpy == db->db_size) dmu_buf_fill_done(db, tx); if (err) break; size -= tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); return (err); } int dmu_write_uio_dbuf(dmu_buf_t *zdb, uio_t *uio, uint64_t size, dmu_tx_t *tx) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)zdb; dnode_t *dn; int err; if (size == 0) return (0); DB_DNODE_ENTER(db); dn = DB_DNODE(db); err = dmu_write_uio_dnode(dn, uio, size, tx); DB_DNODE_EXIT(db); return (err); } int dmu_write_uio(objset_t *os, uint64_t object, uio_t *uio, uint64_t size, dmu_tx_t *tx) { dnode_t *dn; int err; if (size == 0) return (0); err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); err = dmu_write_uio_dnode(dn, uio, size, tx); dnode_rele(dn, FTAG); return (err); } #ifdef sun int dmu_write_pages(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, page_t *pp, dmu_tx_t *tx) { dmu_buf_t **dbp; int numbufs, i; int err; if (size == 0) return (0); err = dmu_buf_hold_array(os, object, offset, size, FALSE, FTAG, &numbufs, &dbp); if (err) return (err); for (i = 0; i < numbufs; i++) { int tocpy, copied, thiscpy; int bufoff; dmu_buf_t *db = dbp[i]; caddr_t va; ASSERT(size > 0); ASSERT3U(db->db_size, >=, PAGESIZE); bufoff = offset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size); if (tocpy == db->db_size) dmu_buf_will_fill(db, tx); else dmu_buf_will_dirty(db, tx); for (copied = 0; copied < tocpy; copied += PAGESIZE) { ASSERT3U(pp->p_offset, ==, db->db_offset + bufoff); thiscpy = MIN(PAGESIZE, tocpy - copied); va = zfs_map_page(pp, S_READ); bcopy(va, (char *)db->db_data + bufoff, thiscpy); zfs_unmap_page(pp, va); pp = pp->p_next; bufoff += PAGESIZE; } if (tocpy == db->db_size) dmu_buf_fill_done(db, tx); offset += tocpy; size -= tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); return (err); } #else int dmu_write_pages(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, vm_page_t *ma, dmu_tx_t *tx) { dmu_buf_t **dbp; struct sf_buf *sf; int numbufs, i; int err; if (size == 0) return (0); err = dmu_buf_hold_array(os, object, offset, size, FALSE, FTAG, &numbufs, &dbp); if (err) return (err); for (i = 0; i < numbufs; i++) { int tocpy, copied, thiscpy; int bufoff; dmu_buf_t *db = dbp[i]; caddr_t va; ASSERT(size > 0); ASSERT3U(db->db_size, >=, PAGESIZE); bufoff = offset - db->db_offset; tocpy = (int)MIN(db->db_size - bufoff, size); ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size); if (tocpy == db->db_size) dmu_buf_will_fill(db, tx); else dmu_buf_will_dirty(db, tx); for (copied = 0; copied < tocpy; copied += PAGESIZE) { ASSERT3U(ptoa((*ma)->pindex), ==, db->db_offset + bufoff); thiscpy = MIN(PAGESIZE, tocpy - copied); va = zfs_map_page(*ma, &sf); bcopy(va, (char *)db->db_data + bufoff, thiscpy); zfs_unmap_page(sf); ma += 1; bufoff += PAGESIZE; } if (tocpy == db->db_size) dmu_buf_fill_done(db, tx); offset += tocpy; size -= tocpy; } dmu_buf_rele_array(dbp, numbufs, FTAG); return (err); } #endif /* sun */ #endif /* * Allocate a loaned anonymous arc buffer. */ arc_buf_t * dmu_request_arcbuf(dmu_buf_t *handle, int size) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)handle; return (arc_loan_buf(db->db_objset->os_spa, size)); } /* * Free a loaned arc buffer. */ void dmu_return_arcbuf(arc_buf_t *buf) { arc_return_buf(buf, FTAG); VERIFY(arc_buf_remove_ref(buf, FTAG)); } /* * When possible directly assign passed loaned arc buffer to a dbuf. * If this is not possible copy the contents of passed arc buf via * dmu_write(). */ void dmu_assign_arcbuf(dmu_buf_t *handle, uint64_t offset, arc_buf_t *buf, dmu_tx_t *tx) { dmu_buf_impl_t *dbuf = (dmu_buf_impl_t *)handle; dnode_t *dn; dmu_buf_impl_t *db; uint32_t blksz = (uint32_t)arc_buf_size(buf); uint64_t blkid; DB_DNODE_ENTER(dbuf); dn = DB_DNODE(dbuf); rw_enter(&dn->dn_struct_rwlock, RW_READER); blkid = dbuf_whichblock(dn, offset); VERIFY((db = dbuf_hold(dn, blkid, FTAG)) != NULL); rw_exit(&dn->dn_struct_rwlock); DB_DNODE_EXIT(dbuf); if (offset == db->db.db_offset && blksz == db->db.db_size) { dbuf_assign_arcbuf(db, buf, tx); dbuf_rele(db, FTAG); } else { objset_t *os; uint64_t object; DB_DNODE_ENTER(dbuf); dn = DB_DNODE(dbuf); os = dn->dn_objset; object = dn->dn_object; DB_DNODE_EXIT(dbuf); dbuf_rele(db, FTAG); dmu_write(os, object, offset, blksz, buf->b_data, tx); dmu_return_arcbuf(buf); XUIOSTAT_BUMP(xuiostat_wbuf_copied); } } typedef struct { dbuf_dirty_record_t *dsa_dr; dmu_sync_cb_t *dsa_done; zgd_t *dsa_zgd; dmu_tx_t *dsa_tx; } dmu_sync_arg_t; /* ARGSUSED */ static void dmu_sync_ready(zio_t *zio, arc_buf_t *buf, void *varg) { dmu_sync_arg_t *dsa = varg; dmu_buf_t *db = dsa->dsa_zgd->zgd_db; blkptr_t *bp = zio->io_bp; if (zio->io_error == 0) { if (BP_IS_HOLE(bp)) { /* * A block of zeros may compress to a hole, but the * block size still needs to be known for replay. */ BP_SET_LSIZE(bp, db->db_size); } else { ASSERT(BP_GET_LEVEL(bp) == 0); bp->blk_fill = 1; } } } static void dmu_sync_late_arrival_ready(zio_t *zio) { dmu_sync_ready(zio, NULL, zio->io_private); } /* ARGSUSED */ static void dmu_sync_done(zio_t *zio, arc_buf_t *buf, void *varg) { dmu_sync_arg_t *dsa = varg; dbuf_dirty_record_t *dr = dsa->dsa_dr; dmu_buf_impl_t *db = dr->dr_dbuf; mutex_enter(&db->db_mtx); ASSERT(dr->dt.dl.dr_override_state == DR_IN_DMU_SYNC); if (zio->io_error == 0) { dr->dt.dl.dr_nopwrite = !!(zio->io_flags & ZIO_FLAG_NOPWRITE); if (dr->dt.dl.dr_nopwrite) { blkptr_t *bp = zio->io_bp; blkptr_t *bp_orig = &zio->io_bp_orig; uint8_t chksum = BP_GET_CHECKSUM(bp_orig); ASSERT(BP_EQUAL(bp, bp_orig)); ASSERT(zio->io_prop.zp_compress != ZIO_COMPRESS_OFF); ASSERT(zio_checksum_table[chksum].ci_dedup); } dr->dt.dl.dr_overridden_by = *zio->io_bp; dr->dt.dl.dr_override_state = DR_OVERRIDDEN; dr->dt.dl.dr_copies = zio->io_prop.zp_copies; if (BP_IS_HOLE(&dr->dt.dl.dr_overridden_by)) BP_ZERO(&dr->dt.dl.dr_overridden_by); } else { dr->dt.dl.dr_override_state = DR_NOT_OVERRIDDEN; } cv_broadcast(&db->db_changed); mutex_exit(&db->db_mtx); dsa->dsa_done(dsa->dsa_zgd, zio->io_error); kmem_free(dsa, sizeof (*dsa)); } static void dmu_sync_late_arrival_done(zio_t *zio) { blkptr_t *bp = zio->io_bp; dmu_sync_arg_t *dsa = zio->io_private; blkptr_t *bp_orig = &zio->io_bp_orig; if (zio->io_error == 0 && !BP_IS_HOLE(bp)) { /* * If we didn't allocate a new block (i.e. ZIO_FLAG_NOPWRITE) * then there is nothing to do here. Otherwise, free the * newly allocated block in this txg. */ if (zio->io_flags & ZIO_FLAG_NOPWRITE) { ASSERT(BP_EQUAL(bp, bp_orig)); } else { ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig)); ASSERT(zio->io_bp->blk_birth == zio->io_txg); ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa)); zio_free(zio->io_spa, zio->io_txg, zio->io_bp); } } dmu_tx_commit(dsa->dsa_tx); dsa->dsa_done(dsa->dsa_zgd, zio->io_error); kmem_free(dsa, sizeof (*dsa)); } static int dmu_sync_late_arrival(zio_t *pio, objset_t *os, dmu_sync_cb_t *done, zgd_t *zgd, zio_prop_t *zp, zbookmark_t *zb) { dmu_sync_arg_t *dsa; dmu_tx_t *tx; tx = dmu_tx_create(os); dmu_tx_hold_space(tx, zgd->zgd_db->db_size); if (dmu_tx_assign(tx, TXG_WAIT) != 0) { dmu_tx_abort(tx); /* Make zl_get_data do txg_waited_synced() */ return (SET_ERROR(EIO)); } dsa = kmem_alloc(sizeof (dmu_sync_arg_t), KM_SLEEP); dsa->dsa_dr = NULL; dsa->dsa_done = done; dsa->dsa_zgd = zgd; dsa->dsa_tx = tx; zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx), zgd->zgd_bp, zgd->zgd_db->db_data, zgd->zgd_db->db_size, zp, dmu_sync_late_arrival_ready, NULL, dmu_sync_late_arrival_done, dsa, ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, zb)); return (0); } /* * Intent log support: sync the block associated with db to disk. * N.B. and XXX: the caller is responsible for making sure that the * data isn't changing while dmu_sync() is writing it. * * Return values: * * EEXIST: this txg has already been synced, so there's nothing to do. * The caller should not log the write. * * ENOENT: the block was dbuf_free_range()'d, so there's nothing to do. * The caller should not log the write. * * EALREADY: this block is already in the process of being synced. * The caller should track its progress (somehow). * * EIO: could not do the I/O. * The caller should do a txg_wait_synced(). * * 0: the I/O has been initiated. * The caller should log this blkptr in the done callback. * It is possible that the I/O will fail, in which case * the error will be reported to the done callback and * propagated to pio from zio_done(). */ int dmu_sync(zio_t *pio, uint64_t txg, dmu_sync_cb_t *done, zgd_t *zgd) { blkptr_t *bp = zgd->zgd_bp; dmu_buf_impl_t *db = (dmu_buf_impl_t *)zgd->zgd_db; objset_t *os = db->db_objset; dsl_dataset_t *ds = os->os_dsl_dataset; dbuf_dirty_record_t *dr; dmu_sync_arg_t *dsa; zbookmark_t zb; zio_prop_t zp; dnode_t *dn; ASSERT(pio != NULL); ASSERT(txg != 0); SET_BOOKMARK(&zb, ds->ds_object, db->db.db_object, db->db_level, db->db_blkid); DB_DNODE_ENTER(db); dn = DB_DNODE(db); dmu_write_policy(os, dn, db->db_level, WP_DMU_SYNC, &zp); DB_DNODE_EXIT(db); /* * If we're frozen (running ziltest), we always need to generate a bp. */ if (txg > spa_freeze_txg(os->os_spa)) return (dmu_sync_late_arrival(pio, os, done, zgd, &zp, &zb)); /* * Grabbing db_mtx now provides a barrier between dbuf_sync_leaf() * and us. If we determine that this txg is not yet syncing, * but it begins to sync a moment later, that's OK because the * sync thread will block in dbuf_sync_leaf() until we drop db_mtx. */ mutex_enter(&db->db_mtx); if (txg <= spa_last_synced_txg(os->os_spa)) { /* * This txg has already synced. There's nothing to do. */ mutex_exit(&db->db_mtx); return (SET_ERROR(EEXIST)); } if (txg <= spa_syncing_txg(os->os_spa)) { /* * This txg is currently syncing, so we can't mess with * the dirty record anymore; just write a new log block. */ mutex_exit(&db->db_mtx); return (dmu_sync_late_arrival(pio, os, done, zgd, &zp, &zb)); } dr = db->db_last_dirty; while (dr && dr->dr_txg != txg) dr = dr->dr_next; if (dr == NULL) { /* * There's no dr for this dbuf, so it must have been freed. * There's no need to log writes to freed blocks, so we're done. */ mutex_exit(&db->db_mtx); return (SET_ERROR(ENOENT)); } ASSERT(dr->dr_next == NULL || dr->dr_next->dr_txg < txg); /* * Assume the on-disk data is X, the current syncing data is Y, * and the current in-memory data is Z (currently in dmu_sync). * X and Z are identical but Y is has been modified. Normally, * when X and Z are the same we will perform a nopwrite but if Y * is different we must disable nopwrite since the resulting write * of Y to disk can free the block containing X. If we allowed a * nopwrite to occur the block pointing to Z would reference a freed * block. Since this is a rare case we simplify this by disabling * nopwrite if the current dmu_sync-ing dbuf has been modified in * a previous transaction. */ if (dr->dr_next) zp.zp_nopwrite = B_FALSE; ASSERT(dr->dr_txg == txg); if (dr->dt.dl.dr_override_state == DR_IN_DMU_SYNC || dr->dt.dl.dr_override_state == DR_OVERRIDDEN) { /* * We have already issued a sync write for this buffer, * or this buffer has already been synced. It could not * have been dirtied since, or we would have cleared the state. */ mutex_exit(&db->db_mtx); return (SET_ERROR(EALREADY)); } ASSERT(dr->dt.dl.dr_override_state == DR_NOT_OVERRIDDEN); dr->dt.dl.dr_override_state = DR_IN_DMU_SYNC; mutex_exit(&db->db_mtx); dsa = kmem_alloc(sizeof (dmu_sync_arg_t), KM_SLEEP); dsa->dsa_dr = dr; dsa->dsa_done = done; dsa->dsa_zgd = zgd; dsa->dsa_tx = NULL; zio_nowait(arc_write(pio, os->os_spa, txg, bp, dr->dt.dl.dr_data, DBUF_IS_L2CACHEABLE(db), DBUF_IS_L2COMPRESSIBLE(db), &zp, dmu_sync_ready, NULL, dmu_sync_done, dsa, ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, &zb)); return (0); } int dmu_object_set_blocksize(objset_t *os, uint64_t object, uint64_t size, int ibs, dmu_tx_t *tx) { dnode_t *dn; int err; err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); err = dnode_set_blksz(dn, size, ibs, tx); dnode_rele(dn, FTAG); return (err); } void dmu_object_set_checksum(objset_t *os, uint64_t object, uint8_t checksum, dmu_tx_t *tx) { dnode_t *dn; /* XXX assumes dnode_hold will not get an i/o error */ (void) dnode_hold(os, object, FTAG, &dn); ASSERT(checksum < ZIO_CHECKSUM_FUNCTIONS); dn->dn_checksum = checksum; dnode_setdirty(dn, tx); dnode_rele(dn, FTAG); } void dmu_object_set_compress(objset_t *os, uint64_t object, uint8_t compress, dmu_tx_t *tx) { dnode_t *dn; /* XXX assumes dnode_hold will not get an i/o error */ (void) dnode_hold(os, object, FTAG, &dn); ASSERT(compress < ZIO_COMPRESS_FUNCTIONS); dn->dn_compress = compress; dnode_setdirty(dn, tx); dnode_rele(dn, FTAG); } int zfs_mdcomp_disable = 0; TUNABLE_INT("vfs.zfs.mdcomp_disable", &zfs_mdcomp_disable); SYSCTL_INT(_vfs_zfs, OID_AUTO, mdcomp_disable, CTLFLAG_RW, &zfs_mdcomp_disable, 0, "Disable metadata compression"); +/* + * When the "redundant_metadata" property is set to "most", only indirect + * blocks of this level and higher will have an additional ditto block. + */ +int zfs_redundant_metadata_most_ditto_level = 2; + void dmu_write_policy(objset_t *os, dnode_t *dn, int level, int wp, zio_prop_t *zp) { dmu_object_type_t type = dn ? dn->dn_type : DMU_OT_OBJSET; boolean_t ismd = (level > 0 || DMU_OT_IS_METADATA(type) || (wp & WP_SPILL)); enum zio_checksum checksum = os->os_checksum; enum zio_compress compress = os->os_compress; enum zio_checksum dedup_checksum = os->os_dedup_checksum; boolean_t dedup = B_FALSE; boolean_t nopwrite = B_FALSE; boolean_t dedup_verify = os->os_dedup_verify; int copies = os->os_copies; /* * We maintain different write policies for each of the following * types of data: * 1. metadata * 2. preallocated blocks (i.e. level-0 blocks of a dump device) * 3. all other level 0 blocks */ if (ismd) { /* * XXX -- we should design a compression algorithm * that specializes in arrays of bps. */ compress = zfs_mdcomp_disable ? ZIO_COMPRESS_EMPTY : ZIO_COMPRESS_LZJB; /* * Metadata always gets checksummed. If the data * checksum is multi-bit correctable, and it's not a * ZBT-style checksum, then it's suitable for metadata * as well. Otherwise, the metadata checksum defaults * to fletcher4. */ if (zio_checksum_table[checksum].ci_correctable < 1 || zio_checksum_table[checksum].ci_eck) checksum = ZIO_CHECKSUM_FLETCHER_4; + + if (os->os_redundant_metadata == ZFS_REDUNDANT_METADATA_ALL || + (os->os_redundant_metadata == + ZFS_REDUNDANT_METADATA_MOST && + (level >= zfs_redundant_metadata_most_ditto_level || + DMU_OT_IS_METADATA(type) || (wp & WP_SPILL)))) + copies++; } else if (wp & WP_NOFILL) { ASSERT(level == 0); /* * If we're writing preallocated blocks, we aren't actually * writing them so don't set any policy properties. These * blocks are currently only used by an external subsystem * outside of zfs (i.e. dump) and not written by the zio * pipeline. */ compress = ZIO_COMPRESS_OFF; checksum = ZIO_CHECKSUM_NOPARITY; } else { compress = zio_compress_select(dn->dn_compress, compress); checksum = (dedup_checksum == ZIO_CHECKSUM_OFF) ? zio_checksum_select(dn->dn_checksum, checksum) : dedup_checksum; /* * Determine dedup setting. If we are in dmu_sync(), * we won't actually dedup now because that's all * done in syncing context; but we do want to use the * dedup checkum. If the checksum is not strong * enough to ensure unique signatures, force * dedup_verify. */ if (dedup_checksum != ZIO_CHECKSUM_OFF) { dedup = (wp & WP_DMU_SYNC) ? B_FALSE : B_TRUE; if (!zio_checksum_table[checksum].ci_dedup) dedup_verify = B_TRUE; } /* * Enable nopwrite if we have a cryptographically secure * checksum that has no known collisions (i.e. SHA-256) * and compression is enabled. We don't enable nopwrite if * dedup is enabled as the two features are mutually exclusive. */ nopwrite = (!dedup && zio_checksum_table[checksum].ci_dedup && compress != ZIO_COMPRESS_OFF && zfs_nopwrite_enabled); } zp->zp_checksum = checksum; zp->zp_compress = compress; zp->zp_type = (wp & WP_SPILL) ? dn->dn_bonustype : type; zp->zp_level = level; - zp->zp_copies = MIN(copies + ismd, spa_max_replication(os->os_spa)); + zp->zp_copies = MIN(copies, spa_max_replication(os->os_spa)); zp->zp_dedup = dedup; zp->zp_dedup_verify = dedup && dedup_verify; zp->zp_nopwrite = nopwrite; } int dmu_offset_next(objset_t *os, uint64_t object, boolean_t hole, uint64_t *off) { dnode_t *dn; int i, err; err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); /* * Sync any current changes before * we go trundling through the block pointers. */ for (i = 0; i < TXG_SIZE; i++) { if (list_link_active(&dn->dn_dirty_link[i])) break; } if (i != TXG_SIZE) { dnode_rele(dn, FTAG); txg_wait_synced(dmu_objset_pool(os), 0); err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); } err = dnode_next_offset(dn, (hole ? DNODE_FIND_HOLE : 0), off, 1, 1, 0); dnode_rele(dn, FTAG); return (err); } void dmu_object_info_from_dnode(dnode_t *dn, dmu_object_info_t *doi) { dnode_phys_t *dnp; rw_enter(&dn->dn_struct_rwlock, RW_READER); mutex_enter(&dn->dn_mtx); dnp = dn->dn_phys; doi->doi_data_block_size = dn->dn_datablksz; doi->doi_metadata_block_size = dn->dn_indblkshift ? 1ULL << dn->dn_indblkshift : 0; doi->doi_type = dn->dn_type; doi->doi_bonus_type = dn->dn_bonustype; doi->doi_bonus_size = dn->dn_bonuslen; doi->doi_indirection = dn->dn_nlevels; doi->doi_checksum = dn->dn_checksum; doi->doi_compress = dn->dn_compress; doi->doi_physical_blocks_512 = (DN_USED_BYTES(dnp) + 256) >> 9; doi->doi_max_offset = (dn->dn_maxblkid + 1) * dn->dn_datablksz; doi->doi_fill_count = 0; for (int i = 0; i < dnp->dn_nblkptr; i++) doi->doi_fill_count += dnp->dn_blkptr[i].blk_fill; mutex_exit(&dn->dn_mtx); rw_exit(&dn->dn_struct_rwlock); } /* * Get information on a DMU object. * If doi is NULL, just indicates whether the object exists. */ int dmu_object_info(objset_t *os, uint64_t object, dmu_object_info_t *doi) { dnode_t *dn; int err = dnode_hold(os, object, FTAG, &dn); if (err) return (err); if (doi != NULL) dmu_object_info_from_dnode(dn, doi); dnode_rele(dn, FTAG); return (0); } /* * As above, but faster; can be used when you have a held dbuf in hand. */ void dmu_object_info_from_db(dmu_buf_t *db_fake, dmu_object_info_t *doi) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; DB_DNODE_ENTER(db); dmu_object_info_from_dnode(DB_DNODE(db), doi); DB_DNODE_EXIT(db); } /* * Faster still when you only care about the size. * This is specifically optimized for zfs_getattr(). */ void dmu_object_size_from_db(dmu_buf_t *db_fake, uint32_t *blksize, u_longlong_t *nblk512) { dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake; dnode_t *dn; DB_DNODE_ENTER(db); dn = DB_DNODE(db); *blksize = dn->dn_datablksz; /* add 1 for dnode space */ *nblk512 = ((DN_USED_BYTES(dn->dn_phys) + SPA_MINBLOCKSIZE/2) >> SPA_MINBLOCKSHIFT) + 1; DB_DNODE_EXIT(db); } void byteswap_uint64_array(void *vbuf, size_t size) { uint64_t *buf = vbuf; size_t count = size >> 3; int i; ASSERT((size & 7) == 0); for (i = 0; i < count; i++) buf[i] = BSWAP_64(buf[i]); } void byteswap_uint32_array(void *vbuf, size_t size) { uint32_t *buf = vbuf; size_t count = size >> 2; int i; ASSERT((size & 3) == 0); for (i = 0; i < count; i++) buf[i] = BSWAP_32(buf[i]); } void byteswap_uint16_array(void *vbuf, size_t size) { uint16_t *buf = vbuf; size_t count = size >> 1; int i; ASSERT((size & 1) == 0); for (i = 0; i < count; i++) buf[i] = BSWAP_16(buf[i]); } /* ARGSUSED */ void byteswap_uint8_array(void *vbuf, size_t size) { } void dmu_init(void) { zfs_dbgmsg_init(); sa_cache_init(); xuio_stat_init(); dmu_objset_init(); dnode_init(); dbuf_init(); zfetch_init(); zio_compress_init(); l2arc_init(); arc_init(); } void dmu_fini(void) { arc_fini(); /* arc depends on l2arc, so arc must go first */ l2arc_fini(); zfetch_fini(); zio_compress_fini(); dbuf_fini(); dnode_fini(); dmu_objset_fini(); xuio_stat_fini(); sa_cache_fini(); zfs_dbgmsg_fini(); } Index: stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c (revision 268647) @@ -1,1786 +1,1809 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2012, 2014 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. * Copyright (c) 2013, Joyent, Inc. All rights reserved. */ /* Portions Copyright 2010 Robert Milkowski */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include /* * Needed to close a window in dnode_move() that allows the objset to be freed * before it can be safely accessed. */ krwlock_t os_lock; void dmu_objset_init(void) { rw_init(&os_lock, NULL, RW_DEFAULT, NULL); } void dmu_objset_fini(void) { rw_destroy(&os_lock); } spa_t * dmu_objset_spa(objset_t *os) { return (os->os_spa); } zilog_t * dmu_objset_zil(objset_t *os) { return (os->os_zil); } dsl_pool_t * dmu_objset_pool(objset_t *os) { dsl_dataset_t *ds; if ((ds = os->os_dsl_dataset) != NULL && ds->ds_dir) return (ds->ds_dir->dd_pool); else return (spa_get_dsl(os->os_spa)); } dsl_dataset_t * dmu_objset_ds(objset_t *os) { return (os->os_dsl_dataset); } dmu_objset_type_t dmu_objset_type(objset_t *os) { return (os->os_phys->os_type); } void dmu_objset_name(objset_t *os, char *buf) { dsl_dataset_name(os->os_dsl_dataset, buf); } uint64_t dmu_objset_id(objset_t *os) { dsl_dataset_t *ds = os->os_dsl_dataset; return (ds ? ds->ds_object : 0); } -uint64_t +zfs_sync_type_t dmu_objset_syncprop(objset_t *os) { return (os->os_sync); } -uint64_t +zfs_logbias_op_t dmu_objset_logbias(objset_t *os) { return (os->os_logbias); } static void checksum_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance should have been done by now. */ ASSERT(newval != ZIO_CHECKSUM_INHERIT); os->os_checksum = zio_checksum_select(newval, ZIO_CHECKSUM_ON_VALUE); } static void compression_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance and range checking should have been done by now. */ ASSERT(newval != ZIO_COMPRESS_INHERIT); os->os_compress = zio_compress_select(newval, ZIO_COMPRESS_ON_VALUE); } static void copies_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance and range checking should have been done by now. */ ASSERT(newval > 0); ASSERT(newval <= spa_max_replication(os->os_spa)); os->os_copies = newval; } static void dedup_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; spa_t *spa = os->os_spa; enum zio_checksum checksum; /* * Inheritance should have been done by now. */ ASSERT(newval != ZIO_CHECKSUM_INHERIT); checksum = zio_checksum_dedup_select(spa, newval, ZIO_CHECKSUM_OFF); os->os_dedup_checksum = checksum & ZIO_CHECKSUM_MASK; os->os_dedup_verify = !!(checksum & ZIO_CHECKSUM_VERIFY); } static void primary_cache_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance and range checking should have been done by now. */ ASSERT(newval == ZFS_CACHE_ALL || newval == ZFS_CACHE_NONE || newval == ZFS_CACHE_METADATA); os->os_primary_cache = newval; } static void secondary_cache_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance and range checking should have been done by now. */ ASSERT(newval == ZFS_CACHE_ALL || newval == ZFS_CACHE_NONE || newval == ZFS_CACHE_METADATA); os->os_secondary_cache = newval; } static void sync_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; /* * Inheritance and range checking should have been done by now. */ ASSERT(newval == ZFS_SYNC_STANDARD || newval == ZFS_SYNC_ALWAYS || newval == ZFS_SYNC_DISABLED); os->os_sync = newval; if (os->os_zil) zil_set_sync(os->os_zil, newval); } static void +redundant_metadata_changed_cb(void *arg, uint64_t newval) +{ + objset_t *os = arg; + + /* + * Inheritance and range checking should have been done by now. + */ + ASSERT(newval == ZFS_REDUNDANT_METADATA_ALL || + newval == ZFS_REDUNDANT_METADATA_MOST); + + os->os_redundant_metadata = newval; +} + +static void logbias_changed_cb(void *arg, uint64_t newval) { objset_t *os = arg; ASSERT(newval == ZFS_LOGBIAS_LATENCY || newval == ZFS_LOGBIAS_THROUGHPUT); os->os_logbias = newval; if (os->os_zil) zil_set_logbias(os->os_zil, newval); } void dmu_objset_byteswap(void *buf, size_t size) { objset_phys_t *osp = buf; ASSERT(size == OBJSET_OLD_PHYS_SIZE || size == sizeof (objset_phys_t)); dnode_byteswap(&osp->os_meta_dnode); byteswap_uint64_array(&osp->os_zil_header, sizeof (zil_header_t)); osp->os_type = BSWAP_64(osp->os_type); osp->os_flags = BSWAP_64(osp->os_flags); if (size == sizeof (objset_phys_t)) { dnode_byteswap(&osp->os_userused_dnode); dnode_byteswap(&osp->os_groupused_dnode); } } int dmu_objset_open_impl(spa_t *spa, dsl_dataset_t *ds, blkptr_t *bp, objset_t **osp) { objset_t *os; int i, err; ASSERT(ds == NULL || MUTEX_HELD(&ds->ds_opening_lock)); os = kmem_zalloc(sizeof (objset_t), KM_SLEEP); os->os_dsl_dataset = ds; os->os_spa = spa; os->os_rootbp = bp; if (!BP_IS_HOLE(os->os_rootbp)) { uint32_t aflags = ARC_WAIT; zbookmark_t zb; SET_BOOKMARK(&zb, ds ? ds->ds_object : DMU_META_OBJSET, ZB_ROOT_OBJECT, ZB_ROOT_LEVEL, ZB_ROOT_BLKID); if (DMU_OS_IS_L2CACHEABLE(os)) aflags |= ARC_L2CACHE; if (DMU_OS_IS_L2COMPRESSIBLE(os)) aflags |= ARC_L2COMPRESS; dprintf_bp(os->os_rootbp, "reading %s", ""); err = arc_read(NULL, spa, os->os_rootbp, arc_getbuf_func, &os->os_phys_buf, ZIO_PRIORITY_SYNC_READ, ZIO_FLAG_CANFAIL, &aflags, &zb); if (err != 0) { kmem_free(os, sizeof (objset_t)); /* convert checksum errors into IO errors */ if (err == ECKSUM) err = SET_ERROR(EIO); return (err); } /* Increase the blocksize if we are permitted. */ if (spa_version(spa) >= SPA_VERSION_USERSPACE && arc_buf_size(os->os_phys_buf) < sizeof (objset_phys_t)) { arc_buf_t *buf = arc_buf_alloc(spa, sizeof (objset_phys_t), &os->os_phys_buf, ARC_BUFC_METADATA); bzero(buf->b_data, sizeof (objset_phys_t)); bcopy(os->os_phys_buf->b_data, buf->b_data, arc_buf_size(os->os_phys_buf)); (void) arc_buf_remove_ref(os->os_phys_buf, &os->os_phys_buf); os->os_phys_buf = buf; } os->os_phys = os->os_phys_buf->b_data; os->os_flags = os->os_phys->os_flags; } else { int size = spa_version(spa) >= SPA_VERSION_USERSPACE ? sizeof (objset_phys_t) : OBJSET_OLD_PHYS_SIZE; os->os_phys_buf = arc_buf_alloc(spa, size, &os->os_phys_buf, ARC_BUFC_METADATA); os->os_phys = os->os_phys_buf->b_data; bzero(os->os_phys, size); } /* * Note: the changed_cb will be called once before the register * func returns, thus changing the checksum/compression from the * default (fletcher2/off). Snapshots don't need to know about * checksum/compression/copies. */ if (ds) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_PRIMARYCACHE), primary_cache_changed_cb, os); if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_SECONDARYCACHE), secondary_cache_changed_cb, os); } if (!dsl_dataset_is_snapshot(ds)) { if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_CHECKSUM), checksum_changed_cb, os); } if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_COMPRESSION), compression_changed_cb, os); } if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_COPIES), copies_changed_cb, os); } if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_DEDUP), dedup_changed_cb, os); } if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_LOGBIAS), logbias_changed_cb, os); } if (err == 0) { err = dsl_prop_register(ds, zfs_prop_to_name(ZFS_PROP_SYNC), sync_changed_cb, os); } + if (err == 0) { + err = dsl_prop_register(ds, + zfs_prop_to_name( + ZFS_PROP_REDUNDANT_METADATA), + redundant_metadata_changed_cb, os); + } } if (err != 0) { VERIFY(arc_buf_remove_ref(os->os_phys_buf, &os->os_phys_buf)); kmem_free(os, sizeof (objset_t)); return (err); } } else if (ds == NULL) { /* It's the meta-objset. */ os->os_checksum = ZIO_CHECKSUM_FLETCHER_4; os->os_compress = ZIO_COMPRESS_LZJB; os->os_copies = spa_max_replication(spa); os->os_dedup_checksum = ZIO_CHECKSUM_OFF; - os->os_dedup_verify = 0; - os->os_logbias = 0; - os->os_sync = 0; + os->os_dedup_verify = B_FALSE; + os->os_logbias = ZFS_LOGBIAS_LATENCY; + os->os_sync = ZFS_SYNC_STANDARD; os->os_primary_cache = ZFS_CACHE_ALL; os->os_secondary_cache = ZFS_CACHE_ALL; } if (ds == NULL || !dsl_dataset_is_snapshot(ds)) os->os_zil_header = os->os_phys->os_zil_header; os->os_zil = zil_alloc(os, &os->os_zil_header); for (i = 0; i < TXG_SIZE; i++) { list_create(&os->os_dirty_dnodes[i], sizeof (dnode_t), offsetof(dnode_t, dn_dirty_link[i])); list_create(&os->os_free_dnodes[i], sizeof (dnode_t), offsetof(dnode_t, dn_dirty_link[i])); } list_create(&os->os_dnodes, sizeof (dnode_t), offsetof(dnode_t, dn_link)); list_create(&os->os_downgraded_dbufs, sizeof (dmu_buf_impl_t), offsetof(dmu_buf_impl_t, db_link)); mutex_init(&os->os_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&os->os_obj_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&os->os_user_ptr_lock, NULL, MUTEX_DEFAULT, NULL); DMU_META_DNODE(os) = dnode_special_open(os, &os->os_phys->os_meta_dnode, DMU_META_DNODE_OBJECT, &os->os_meta_dnode); if (arc_buf_size(os->os_phys_buf) >= sizeof (objset_phys_t)) { DMU_USERUSED_DNODE(os) = dnode_special_open(os, &os->os_phys->os_userused_dnode, DMU_USERUSED_OBJECT, &os->os_userused_dnode); DMU_GROUPUSED_DNODE(os) = dnode_special_open(os, &os->os_phys->os_groupused_dnode, DMU_GROUPUSED_OBJECT, &os->os_groupused_dnode); } /* * We should be the only thread trying to do this because we * have ds_opening_lock */ if (ds) { mutex_enter(&ds->ds_lock); ASSERT(ds->ds_objset == NULL); ds->ds_objset = os; mutex_exit(&ds->ds_lock); } *osp = os; return (0); } int dmu_objset_from_ds(dsl_dataset_t *ds, objset_t **osp) { int err = 0; mutex_enter(&ds->ds_opening_lock); *osp = ds->ds_objset; if (*osp == NULL) { err = dmu_objset_open_impl(dsl_dataset_get_spa(ds), ds, dsl_dataset_get_blkptr(ds), osp); } mutex_exit(&ds->ds_opening_lock); return (err); } /* * Holds the pool while the objset is held. Therefore only one objset * can be held at a time. */ int dmu_objset_hold(const char *name, void *tag, objset_t **osp) { dsl_pool_t *dp; dsl_dataset_t *ds; int err; err = dsl_pool_hold(name, tag, &dp); if (err != 0) return (err); err = dsl_dataset_hold(dp, name, tag, &ds); if (err != 0) { dsl_pool_rele(dp, tag); return (err); } err = dmu_objset_from_ds(ds, osp); if (err != 0) { dsl_dataset_rele(ds, tag); dsl_pool_rele(dp, tag); } return (err); } /* * dsl_pool must not be held when this is called. * Upon successful return, there will be a longhold on the dataset, * and the dsl_pool will not be held. */ int dmu_objset_own(const char *name, dmu_objset_type_t type, boolean_t readonly, void *tag, objset_t **osp) { dsl_pool_t *dp; dsl_dataset_t *ds; int err; err = dsl_pool_hold(name, FTAG, &dp); if (err != 0) return (err); err = dsl_dataset_own(dp, name, tag, &ds); if (err != 0) { dsl_pool_rele(dp, FTAG); return (err); } err = dmu_objset_from_ds(ds, osp); dsl_pool_rele(dp, FTAG); if (err != 0) { dsl_dataset_disown(ds, tag); } else if (type != DMU_OST_ANY && type != (*osp)->os_phys->os_type) { dsl_dataset_disown(ds, tag); return (SET_ERROR(EINVAL)); } else if (!readonly && dsl_dataset_is_snapshot(ds)) { dsl_dataset_disown(ds, tag); return (SET_ERROR(EROFS)); } return (err); } void dmu_objset_rele(objset_t *os, void *tag) { dsl_pool_t *dp = dmu_objset_pool(os); dsl_dataset_rele(os->os_dsl_dataset, tag); dsl_pool_rele(dp, tag); } /* * When we are called, os MUST refer to an objset associated with a dataset * that is owned by 'tag'; that is, is held and long held by 'tag' and ds_owner * == tag. We will then release and reacquire ownership of the dataset while * holding the pool config_rwlock to avoid intervening namespace or ownership * changes may occur. * * This exists solely to accommodate zfs_ioc_userspace_upgrade()'s desire to * release the hold on its dataset and acquire a new one on the dataset of the * same name so that it can be partially torn down and reconstructed. */ void dmu_objset_refresh_ownership(objset_t *os, void *tag) { dsl_pool_t *dp; dsl_dataset_t *ds, *newds; char name[MAXNAMELEN]; ds = os->os_dsl_dataset; VERIFY3P(ds, !=, NULL); VERIFY3P(ds->ds_owner, ==, tag); VERIFY(dsl_dataset_long_held(ds)); dsl_dataset_name(ds, name); dp = dmu_objset_pool(os); dsl_pool_config_enter(dp, FTAG); dmu_objset_disown(os, tag); VERIFY0(dsl_dataset_own(dp, name, tag, &newds)); VERIFY3P(newds, ==, os->os_dsl_dataset); dsl_pool_config_exit(dp, FTAG); } void dmu_objset_disown(objset_t *os, void *tag) { dsl_dataset_disown(os->os_dsl_dataset, tag); } void dmu_objset_evict_dbufs(objset_t *os) { dnode_t *dn; mutex_enter(&os->os_lock); /* process the mdn last, since the other dnodes have holds on it */ list_remove(&os->os_dnodes, DMU_META_DNODE(os)); list_insert_tail(&os->os_dnodes, DMU_META_DNODE(os)); /* * Find the first dnode with holds. We have to do this dance * because dnode_add_ref() only works if you already have a * hold. If there are no holds then it has no dbufs so OK to * skip. */ for (dn = list_head(&os->os_dnodes); dn && !dnode_add_ref(dn, FTAG); dn = list_next(&os->os_dnodes, dn)) continue; while (dn) { dnode_t *next_dn = dn; do { next_dn = list_next(&os->os_dnodes, next_dn); } while (next_dn && !dnode_add_ref(next_dn, FTAG)); mutex_exit(&os->os_lock); dnode_evict_dbufs(dn); dnode_rele(dn, FTAG); mutex_enter(&os->os_lock); dn = next_dn; } mutex_exit(&os->os_lock); } void dmu_objset_evict(objset_t *os) { dsl_dataset_t *ds = os->os_dsl_dataset; for (int t = 0; t < TXG_SIZE; t++) ASSERT(!dmu_objset_is_dirty(os, t)); if (ds) { if (!dsl_dataset_is_snapshot(ds)) { VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_CHECKSUM), checksum_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_COMPRESSION), compression_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_COPIES), copies_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_DEDUP), dedup_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_LOGBIAS), logbias_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_SYNC), sync_changed_cb, os)); + VERIFY0(dsl_prop_unregister(ds, + zfs_prop_to_name(ZFS_PROP_REDUNDANT_METADATA), + redundant_metadata_changed_cb, os)); } VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_PRIMARYCACHE), primary_cache_changed_cb, os)); VERIFY0(dsl_prop_unregister(ds, zfs_prop_to_name(ZFS_PROP_SECONDARYCACHE), secondary_cache_changed_cb, os)); } if (os->os_sa) sa_tear_down(os); dmu_objset_evict_dbufs(os); dnode_special_close(&os->os_meta_dnode); if (DMU_USERUSED_DNODE(os)) { dnode_special_close(&os->os_userused_dnode); dnode_special_close(&os->os_groupused_dnode); } zil_free(os->os_zil); ASSERT3P(list_head(&os->os_dnodes), ==, NULL); VERIFY(arc_buf_remove_ref(os->os_phys_buf, &os->os_phys_buf)); /* * This is a barrier to prevent the objset from going away in * dnode_move() until we can safely ensure that the objset is still in * use. We consider the objset valid before the barrier and invalid * after the barrier. */ rw_enter(&os_lock, RW_READER); rw_exit(&os_lock); mutex_destroy(&os->os_lock); mutex_destroy(&os->os_obj_lock); mutex_destroy(&os->os_user_ptr_lock); kmem_free(os, sizeof (objset_t)); } timestruc_t dmu_objset_snap_cmtime(objset_t *os) { return (dsl_dir_snap_cmtime(os->os_dsl_dataset->ds_dir)); } /* called from dsl for meta-objset */ objset_t * dmu_objset_create_impl(spa_t *spa, dsl_dataset_t *ds, blkptr_t *bp, dmu_objset_type_t type, dmu_tx_t *tx) { objset_t *os; dnode_t *mdn; ASSERT(dmu_tx_is_syncing(tx)); if (ds != NULL) VERIFY0(dmu_objset_from_ds(ds, &os)); else VERIFY0(dmu_objset_open_impl(spa, NULL, bp, &os)); mdn = DMU_META_DNODE(os); dnode_allocate(mdn, DMU_OT_DNODE, 1 << DNODE_BLOCK_SHIFT, DN_MAX_INDBLKSHIFT, DMU_OT_NONE, 0, tx); /* * We don't want to have to increase the meta-dnode's nlevels * later, because then we could do it in quescing context while * we are also accessing it in open context. * * This precaution is not necessary for the MOS (ds == NULL), * because the MOS is only updated in syncing context. * This is most fortunate: the MOS is the only objset that * needs to be synced multiple times as spa_sync() iterates * to convergence, so minimizing its dn_nlevels matters. */ if (ds != NULL) { int levels = 1; /* * Determine the number of levels necessary for the meta-dnode * to contain DN_MAX_OBJECT dnodes. */ while ((uint64_t)mdn->dn_nblkptr << (mdn->dn_datablkshift + (levels - 1) * (mdn->dn_indblkshift - SPA_BLKPTRSHIFT)) < DN_MAX_OBJECT * sizeof (dnode_phys_t)) levels++; mdn->dn_next_nlevels[tx->tx_txg & TXG_MASK] = mdn->dn_nlevels = levels; } ASSERT(type != DMU_OST_NONE); ASSERT(type != DMU_OST_ANY); ASSERT(type < DMU_OST_NUMTYPES); os->os_phys->os_type = type; if (dmu_objset_userused_enabled(os)) { os->os_phys->os_flags |= OBJSET_FLAG_USERACCOUNTING_COMPLETE; os->os_flags = os->os_phys->os_flags; } dsl_dataset_dirty(ds, tx); return (os); } typedef struct dmu_objset_create_arg { const char *doca_name; cred_t *doca_cred; void (*doca_userfunc)(objset_t *os, void *arg, cred_t *cr, dmu_tx_t *tx); void *doca_userarg; dmu_objset_type_t doca_type; uint64_t doca_flags; } dmu_objset_create_arg_t; /*ARGSUSED*/ static int dmu_objset_create_check(void *arg, dmu_tx_t *tx) { dmu_objset_create_arg_t *doca = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dir_t *pdd; const char *tail; int error; if (strchr(doca->doca_name, '@') != NULL) return (SET_ERROR(EINVAL)); error = dsl_dir_hold(dp, doca->doca_name, FTAG, &pdd, &tail); if (error != 0) return (error); if (tail == NULL) { dsl_dir_rele(pdd, FTAG); return (SET_ERROR(EEXIST)); } error = dsl_fs_ss_limit_check(pdd, 1, ZFS_PROP_FILESYSTEM_LIMIT, NULL, doca->doca_cred); dsl_dir_rele(pdd, FTAG); return (error); } static void dmu_objset_create_sync(void *arg, dmu_tx_t *tx) { dmu_objset_create_arg_t *doca = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dir_t *pdd; const char *tail; dsl_dataset_t *ds; uint64_t obj; blkptr_t *bp; objset_t *os; VERIFY0(dsl_dir_hold(dp, doca->doca_name, FTAG, &pdd, &tail)); obj = dsl_dataset_create_sync(pdd, tail, NULL, doca->doca_flags, doca->doca_cred, tx); VERIFY0(dsl_dataset_hold_obj(pdd->dd_pool, obj, FTAG, &ds)); bp = dsl_dataset_get_blkptr(ds); os = dmu_objset_create_impl(pdd->dd_pool->dp_spa, ds, bp, doca->doca_type, tx); if (doca->doca_userfunc != NULL) { doca->doca_userfunc(os, doca->doca_userarg, doca->doca_cred, tx); } spa_history_log_internal_ds(ds, "create", tx, ""); dsl_dataset_rele(ds, FTAG); dsl_dir_rele(pdd, FTAG); } int dmu_objset_create(const char *name, dmu_objset_type_t type, uint64_t flags, void (*func)(objset_t *os, void *arg, cred_t *cr, dmu_tx_t *tx), void *arg) { dmu_objset_create_arg_t doca; doca.doca_name = name; doca.doca_cred = CRED(); doca.doca_flags = flags; doca.doca_userfunc = func; doca.doca_userarg = arg; doca.doca_type = type; return (dsl_sync_task(name, dmu_objset_create_check, dmu_objset_create_sync, &doca, 5)); } typedef struct dmu_objset_clone_arg { const char *doca_clone; const char *doca_origin; cred_t *doca_cred; } dmu_objset_clone_arg_t; /*ARGSUSED*/ static int dmu_objset_clone_check(void *arg, dmu_tx_t *tx) { dmu_objset_clone_arg_t *doca = arg; dsl_dir_t *pdd; const char *tail; int error; dsl_dataset_t *origin; dsl_pool_t *dp = dmu_tx_pool(tx); if (strchr(doca->doca_clone, '@') != NULL) return (SET_ERROR(EINVAL)); error = dsl_dir_hold(dp, doca->doca_clone, FTAG, &pdd, &tail); if (error != 0) return (error); if (tail == NULL) { dsl_dir_rele(pdd, FTAG); return (SET_ERROR(EEXIST)); } /* You can't clone across pools. */ if (pdd->dd_pool != dp) { dsl_dir_rele(pdd, FTAG); return (SET_ERROR(EXDEV)); } error = dsl_fs_ss_limit_check(pdd, 1, ZFS_PROP_FILESYSTEM_LIMIT, NULL, doca->doca_cred); if (error != 0) { dsl_dir_rele(pdd, FTAG); return (SET_ERROR(EDQUOT)); } dsl_dir_rele(pdd, FTAG); error = dsl_dataset_hold(dp, doca->doca_origin, FTAG, &origin); if (error != 0) return (error); /* You can't clone across pools. */ if (origin->ds_dir->dd_pool != dp) { dsl_dataset_rele(origin, FTAG); return (SET_ERROR(EXDEV)); } /* You can only clone snapshots, not the head datasets. */ if (!dsl_dataset_is_snapshot(origin)) { dsl_dataset_rele(origin, FTAG); return (SET_ERROR(EINVAL)); } dsl_dataset_rele(origin, FTAG); return (0); } static void dmu_objset_clone_sync(void *arg, dmu_tx_t *tx) { dmu_objset_clone_arg_t *doca = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dir_t *pdd; const char *tail; dsl_dataset_t *origin, *ds; uint64_t obj; char namebuf[MAXNAMELEN]; VERIFY0(dsl_dir_hold(dp, doca->doca_clone, FTAG, &pdd, &tail)); VERIFY0(dsl_dataset_hold(dp, doca->doca_origin, FTAG, &origin)); obj = dsl_dataset_create_sync(pdd, tail, origin, 0, doca->doca_cred, tx); VERIFY0(dsl_dataset_hold_obj(pdd->dd_pool, obj, FTAG, &ds)); dsl_dataset_name(origin, namebuf); spa_history_log_internal_ds(ds, "clone", tx, "origin=%s (%llu)", namebuf, origin->ds_object); dsl_dataset_rele(ds, FTAG); dsl_dataset_rele(origin, FTAG); dsl_dir_rele(pdd, FTAG); } int dmu_objset_clone(const char *clone, const char *origin) { dmu_objset_clone_arg_t doca; doca.doca_clone = clone; doca.doca_origin = origin; doca.doca_cred = CRED(); return (dsl_sync_task(clone, dmu_objset_clone_check, dmu_objset_clone_sync, &doca, 5)); } int dmu_objset_snapshot_one(const char *fsname, const char *snapname) { int err; char *longsnap = kmem_asprintf("%s@%s", fsname, snapname); nvlist_t *snaps = fnvlist_alloc(); fnvlist_add_boolean(snaps, longsnap); strfree(longsnap); err = dsl_dataset_snapshot(snaps, NULL, NULL); fnvlist_free(snaps); return (err); } static void dmu_objset_sync_dnodes(list_t *list, list_t *newlist, dmu_tx_t *tx) { dnode_t *dn; while (dn = list_head(list)) { ASSERT(dn->dn_object != DMU_META_DNODE_OBJECT); ASSERT(dn->dn_dbuf->db_data_pending); /* * Initialize dn_zio outside dnode_sync() because the * meta-dnode needs to set it ouside dnode_sync(). */ dn->dn_zio = dn->dn_dbuf->db_data_pending->dr_zio; ASSERT(dn->dn_zio); ASSERT3U(dn->dn_nlevels, <=, DN_MAX_LEVELS); list_remove(list, dn); if (newlist) { (void) dnode_add_ref(dn, newlist); list_insert_tail(newlist, dn); } dnode_sync(dn, tx); } } /* ARGSUSED */ static void dmu_objset_write_ready(zio_t *zio, arc_buf_t *abuf, void *arg) { blkptr_t *bp = zio->io_bp; objset_t *os = arg; dnode_phys_t *dnp = &os->os_phys->os_meta_dnode; ASSERT3P(bp, ==, os->os_rootbp); ASSERT3U(BP_GET_TYPE(bp), ==, DMU_OT_OBJSET); ASSERT0(BP_GET_LEVEL(bp)); /* * Update rootbp fill count: it should be the number of objects * allocated in the object set (not counting the "special" * objects that are stored in the objset_phys_t -- the meta * dnode and user/group accounting objects). */ bp->blk_fill = 0; for (int i = 0; i < dnp->dn_nblkptr; i++) bp->blk_fill += dnp->dn_blkptr[i].blk_fill; } /* ARGSUSED */ static void dmu_objset_write_done(zio_t *zio, arc_buf_t *abuf, void *arg) { blkptr_t *bp = zio->io_bp; blkptr_t *bp_orig = &zio->io_bp_orig; objset_t *os = arg; if (zio->io_flags & ZIO_FLAG_IO_REWRITE) { ASSERT(BP_EQUAL(bp, bp_orig)); } else { dsl_dataset_t *ds = os->os_dsl_dataset; dmu_tx_t *tx = os->os_synctx; (void) dsl_dataset_block_kill(ds, bp_orig, tx, B_TRUE); dsl_dataset_block_born(ds, bp, tx); } } /* called from dsl */ void dmu_objset_sync(objset_t *os, zio_t *pio, dmu_tx_t *tx) { int txgoff; zbookmark_t zb; zio_prop_t zp; zio_t *zio; list_t *list; list_t *newlist = NULL; dbuf_dirty_record_t *dr; dprintf_ds(os->os_dsl_dataset, "txg=%llu\n", tx->tx_txg); ASSERT(dmu_tx_is_syncing(tx)); /* XXX the write_done callback should really give us the tx... */ os->os_synctx = tx; if (os->os_dsl_dataset == NULL) { /* * This is the MOS. If we have upgraded, * spa_max_replication() could change, so reset * os_copies here. */ os->os_copies = spa_max_replication(os->os_spa); } /* * Create the root block IO */ SET_BOOKMARK(&zb, os->os_dsl_dataset ? os->os_dsl_dataset->ds_object : DMU_META_OBJSET, ZB_ROOT_OBJECT, ZB_ROOT_LEVEL, ZB_ROOT_BLKID); arc_release(os->os_phys_buf, &os->os_phys_buf); dmu_write_policy(os, NULL, 0, 0, &zp); zio = arc_write(pio, os->os_spa, tx->tx_txg, os->os_rootbp, os->os_phys_buf, DMU_OS_IS_L2CACHEABLE(os), DMU_OS_IS_L2COMPRESSIBLE(os), &zp, dmu_objset_write_ready, NULL, dmu_objset_write_done, os, ZIO_PRIORITY_ASYNC_WRITE, ZIO_FLAG_MUSTSUCCEED, &zb); /* * Sync special dnodes - the parent IO for the sync is the root block */ DMU_META_DNODE(os)->dn_zio = zio; dnode_sync(DMU_META_DNODE(os), tx); os->os_phys->os_flags = os->os_flags; if (DMU_USERUSED_DNODE(os) && DMU_USERUSED_DNODE(os)->dn_type != DMU_OT_NONE) { DMU_USERUSED_DNODE(os)->dn_zio = zio; dnode_sync(DMU_USERUSED_DNODE(os), tx); DMU_GROUPUSED_DNODE(os)->dn_zio = zio; dnode_sync(DMU_GROUPUSED_DNODE(os), tx); } txgoff = tx->tx_txg & TXG_MASK; if (dmu_objset_userused_enabled(os)) { newlist = &os->os_synced_dnodes; /* * We must create the list here because it uses the * dn_dirty_link[] of this txg. */ list_create(newlist, sizeof (dnode_t), offsetof(dnode_t, dn_dirty_link[txgoff])); } dmu_objset_sync_dnodes(&os->os_free_dnodes[txgoff], newlist, tx); dmu_objset_sync_dnodes(&os->os_dirty_dnodes[txgoff], newlist, tx); list = &DMU_META_DNODE(os)->dn_dirty_records[txgoff]; while (dr = list_head(list)) { ASSERT0(dr->dr_dbuf->db_level); list_remove(list, dr); if (dr->dr_zio) zio_nowait(dr->dr_zio); } /* * Free intent log blocks up to this tx. */ zil_sync(os->os_zil, tx); os->os_phys->os_zil_header = os->os_zil_header; zio_nowait(zio); } boolean_t dmu_objset_is_dirty(objset_t *os, uint64_t txg) { return (!list_is_empty(&os->os_dirty_dnodes[txg & TXG_MASK]) || !list_is_empty(&os->os_free_dnodes[txg & TXG_MASK])); } static objset_used_cb_t *used_cbs[DMU_OST_NUMTYPES]; void dmu_objset_register_type(dmu_objset_type_t ost, objset_used_cb_t *cb) { used_cbs[ost] = cb; } boolean_t dmu_objset_userused_enabled(objset_t *os) { return (spa_version(os->os_spa) >= SPA_VERSION_USERSPACE && used_cbs[os->os_phys->os_type] != NULL && DMU_USERUSED_DNODE(os) != NULL); } static void do_userquota_update(objset_t *os, uint64_t used, uint64_t flags, uint64_t user, uint64_t group, boolean_t subtract, dmu_tx_t *tx) { if ((flags & DNODE_FLAG_USERUSED_ACCOUNTED)) { int64_t delta = DNODE_SIZE + used; if (subtract) delta = -delta; VERIFY3U(0, ==, zap_increment_int(os, DMU_USERUSED_OBJECT, user, delta, tx)); VERIFY3U(0, ==, zap_increment_int(os, DMU_GROUPUSED_OBJECT, group, delta, tx)); } } void dmu_objset_do_userquota_updates(objset_t *os, dmu_tx_t *tx) { dnode_t *dn; list_t *list = &os->os_synced_dnodes; ASSERT(list_head(list) == NULL || dmu_objset_userused_enabled(os)); while (dn = list_head(list)) { int flags; ASSERT(!DMU_OBJECT_IS_SPECIAL(dn->dn_object)); ASSERT(dn->dn_phys->dn_type == DMU_OT_NONE || dn->dn_phys->dn_flags & DNODE_FLAG_USERUSED_ACCOUNTED); /* Allocate the user/groupused objects if necessary. */ if (DMU_USERUSED_DNODE(os)->dn_type == DMU_OT_NONE) { VERIFY(0 == zap_create_claim(os, DMU_USERUSED_OBJECT, DMU_OT_USERGROUP_USED, DMU_OT_NONE, 0, tx)); VERIFY(0 == zap_create_claim(os, DMU_GROUPUSED_OBJECT, DMU_OT_USERGROUP_USED, DMU_OT_NONE, 0, tx)); } /* * We intentionally modify the zap object even if the * net delta is zero. Otherwise * the block of the zap obj could be shared between * datasets but need to be different between them after * a bprewrite. */ flags = dn->dn_id_flags; ASSERT(flags); if (flags & DN_ID_OLD_EXIST) { do_userquota_update(os, dn->dn_oldused, dn->dn_oldflags, dn->dn_olduid, dn->dn_oldgid, B_TRUE, tx); } if (flags & DN_ID_NEW_EXIST) { do_userquota_update(os, DN_USED_BYTES(dn->dn_phys), dn->dn_phys->dn_flags, dn->dn_newuid, dn->dn_newgid, B_FALSE, tx); } mutex_enter(&dn->dn_mtx); dn->dn_oldused = 0; dn->dn_oldflags = 0; if (dn->dn_id_flags & DN_ID_NEW_EXIST) { dn->dn_olduid = dn->dn_newuid; dn->dn_oldgid = dn->dn_newgid; dn->dn_id_flags |= DN_ID_OLD_EXIST; if (dn->dn_bonuslen == 0) dn->dn_id_flags |= DN_ID_CHKED_SPILL; else dn->dn_id_flags |= DN_ID_CHKED_BONUS; } dn->dn_id_flags &= ~(DN_ID_NEW_EXIST); mutex_exit(&dn->dn_mtx); list_remove(list, dn); dnode_rele(dn, list); } } /* * Returns a pointer to data to find uid/gid from * * If a dirty record for transaction group that is syncing can't * be found then NULL is returned. In the NULL case it is assumed * the uid/gid aren't changing. */ static void * dmu_objset_userquota_find_data(dmu_buf_impl_t *db, dmu_tx_t *tx) { dbuf_dirty_record_t *dr, **drp; void *data; if (db->db_dirtycnt == 0) return (db->db.db_data); /* Nothing is changing */ for (drp = &db->db_last_dirty; (dr = *drp) != NULL; drp = &dr->dr_next) if (dr->dr_txg == tx->tx_txg) break; if (dr == NULL) { data = NULL; } else { dnode_t *dn; DB_DNODE_ENTER(dr->dr_dbuf); dn = DB_DNODE(dr->dr_dbuf); if (dn->dn_bonuslen == 0 && dr->dr_dbuf->db_blkid == DMU_SPILL_BLKID) data = dr->dt.dl.dr_data->b_data; else data = dr->dt.dl.dr_data; DB_DNODE_EXIT(dr->dr_dbuf); } return (data); } void dmu_objset_userquota_get_ids(dnode_t *dn, boolean_t before, dmu_tx_t *tx) { objset_t *os = dn->dn_objset; void *data = NULL; dmu_buf_impl_t *db = NULL; uint64_t *user = NULL; uint64_t *group = NULL; int flags = dn->dn_id_flags; int error; boolean_t have_spill = B_FALSE; if (!dmu_objset_userused_enabled(dn->dn_objset)) return; if (before && (flags & (DN_ID_CHKED_BONUS|DN_ID_OLD_EXIST| DN_ID_CHKED_SPILL))) return; if (before && dn->dn_bonuslen != 0) data = DN_BONUS(dn->dn_phys); else if (!before && dn->dn_bonuslen != 0) { if (dn->dn_bonus) { db = dn->dn_bonus; mutex_enter(&db->db_mtx); data = dmu_objset_userquota_find_data(db, tx); } else { data = DN_BONUS(dn->dn_phys); } } else if (dn->dn_bonuslen == 0 && dn->dn_bonustype == DMU_OT_SA) { int rf = 0; if (RW_WRITE_HELD(&dn->dn_struct_rwlock)) rf |= DB_RF_HAVESTRUCT; error = dmu_spill_hold_by_dnode(dn, rf | DB_RF_MUST_SUCCEED, FTAG, (dmu_buf_t **)&db); ASSERT(error == 0); mutex_enter(&db->db_mtx); data = (before) ? db->db.db_data : dmu_objset_userquota_find_data(db, tx); have_spill = B_TRUE; } else { mutex_enter(&dn->dn_mtx); dn->dn_id_flags |= DN_ID_CHKED_BONUS; mutex_exit(&dn->dn_mtx); return; } if (before) { ASSERT(data); user = &dn->dn_olduid; group = &dn->dn_oldgid; } else if (data) { user = &dn->dn_newuid; group = &dn->dn_newgid; } /* * Must always call the callback in case the object * type has changed and that type isn't an object type to track */ error = used_cbs[os->os_phys->os_type](dn->dn_bonustype, data, user, group); /* * Preserve existing uid/gid when the callback can't determine * what the new uid/gid are and the callback returned EEXIST. * The EEXIST error tells us to just use the existing uid/gid. * If we don't know what the old values are then just assign * them to 0, since that is a new file being created. */ if (!before && data == NULL && error == EEXIST) { if (flags & DN_ID_OLD_EXIST) { dn->dn_newuid = dn->dn_olduid; dn->dn_newgid = dn->dn_oldgid; } else { dn->dn_newuid = 0; dn->dn_newgid = 0; } error = 0; } if (db) mutex_exit(&db->db_mtx); mutex_enter(&dn->dn_mtx); if (error == 0 && before) dn->dn_id_flags |= DN_ID_OLD_EXIST; if (error == 0 && !before) dn->dn_id_flags |= DN_ID_NEW_EXIST; if (have_spill) { dn->dn_id_flags |= DN_ID_CHKED_SPILL; } else { dn->dn_id_flags |= DN_ID_CHKED_BONUS; } mutex_exit(&dn->dn_mtx); if (have_spill) dmu_buf_rele((dmu_buf_t *)db, FTAG); } boolean_t dmu_objset_userspace_present(objset_t *os) { return (os->os_phys->os_flags & OBJSET_FLAG_USERACCOUNTING_COMPLETE); } int dmu_objset_userspace_upgrade(objset_t *os) { uint64_t obj; int err = 0; if (dmu_objset_userspace_present(os)) return (0); if (!dmu_objset_userused_enabled(os)) return (SET_ERROR(ENOTSUP)); if (dmu_objset_is_snapshot(os)) return (SET_ERROR(EINVAL)); /* * We simply need to mark every object dirty, so that it will be * synced out and now accounted. If this is called * concurrently, or if we already did some work before crashing, * that's fine, since we track each object's accounted state * independently. */ for (obj = 0; err == 0; err = dmu_object_next(os, &obj, FALSE, 0)) { dmu_tx_t *tx; dmu_buf_t *db; int objerr; if (issig(JUSTLOOKING) && issig(FORREAL)) return (SET_ERROR(EINTR)); objerr = dmu_bonus_hold(os, obj, FTAG, &db); if (objerr != 0) continue; tx = dmu_tx_create(os); dmu_tx_hold_bonus(tx, obj); objerr = dmu_tx_assign(tx, TXG_WAIT); if (objerr != 0) { dmu_tx_abort(tx); continue; } dmu_buf_will_dirty(db, tx); dmu_buf_rele(db, FTAG); dmu_tx_commit(tx); } os->os_flags |= OBJSET_FLAG_USERACCOUNTING_COMPLETE; txg_wait_synced(dmu_objset_pool(os), 0); return (0); } void dmu_objset_space(objset_t *os, uint64_t *refdbytesp, uint64_t *availbytesp, uint64_t *usedobjsp, uint64_t *availobjsp) { dsl_dataset_space(os->os_dsl_dataset, refdbytesp, availbytesp, usedobjsp, availobjsp); } uint64_t dmu_objset_fsid_guid(objset_t *os) { return (dsl_dataset_fsid_guid(os->os_dsl_dataset)); } void dmu_objset_fast_stat(objset_t *os, dmu_objset_stats_t *stat) { stat->dds_type = os->os_phys->os_type; if (os->os_dsl_dataset) dsl_dataset_fast_stat(os->os_dsl_dataset, stat); } void dmu_objset_stats(objset_t *os, nvlist_t *nv) { ASSERT(os->os_dsl_dataset || os->os_phys->os_type == DMU_OST_META); if (os->os_dsl_dataset != NULL) dsl_dataset_stats(os->os_dsl_dataset, nv); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_TYPE, os->os_phys->os_type); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USERACCOUNTING, dmu_objset_userspace_present(os)); } int dmu_objset_is_snapshot(objset_t *os) { if (os->os_dsl_dataset != NULL) return (dsl_dataset_is_snapshot(os->os_dsl_dataset)); else return (B_FALSE); } int dmu_snapshot_realname(objset_t *os, char *name, char *real, int maxlen, boolean_t *conflict) { dsl_dataset_t *ds = os->os_dsl_dataset; uint64_t ignored; if (ds->ds_phys->ds_snapnames_zapobj == 0) return (SET_ERROR(ENOENT)); return (zap_lookup_norm(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_phys->ds_snapnames_zapobj, name, 8, 1, &ignored, MT_FIRST, real, maxlen, conflict)); } int dmu_snapshot_list_next(objset_t *os, int namelen, char *name, uint64_t *idp, uint64_t *offp, boolean_t *case_conflict) { dsl_dataset_t *ds = os->os_dsl_dataset; zap_cursor_t cursor; zap_attribute_t attr; ASSERT(dsl_pool_config_held(dmu_objset_pool(os))); if (ds->ds_phys->ds_snapnames_zapobj == 0) return (SET_ERROR(ENOENT)); zap_cursor_init_serialized(&cursor, ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_phys->ds_snapnames_zapobj, *offp); if (zap_cursor_retrieve(&cursor, &attr) != 0) { zap_cursor_fini(&cursor); return (SET_ERROR(ENOENT)); } if (strlen(attr.za_name) + 1 > namelen) { zap_cursor_fini(&cursor); return (SET_ERROR(ENAMETOOLONG)); } (void) strcpy(name, attr.za_name); if (idp) *idp = attr.za_first_integer; if (case_conflict) *case_conflict = attr.za_normalization_conflict; zap_cursor_advance(&cursor); *offp = zap_cursor_serialize(&cursor); zap_cursor_fini(&cursor); return (0); } int dmu_dir_list_next(objset_t *os, int namelen, char *name, uint64_t *idp, uint64_t *offp) { dsl_dir_t *dd = os->os_dsl_dataset->ds_dir; zap_cursor_t cursor; zap_attribute_t attr; /* there is no next dir on a snapshot! */ if (os->os_dsl_dataset->ds_object != dd->dd_phys->dd_head_dataset_obj) return (SET_ERROR(ENOENT)); zap_cursor_init_serialized(&cursor, dd->dd_pool->dp_meta_objset, dd->dd_phys->dd_child_dir_zapobj, *offp); if (zap_cursor_retrieve(&cursor, &attr) != 0) { zap_cursor_fini(&cursor); return (SET_ERROR(ENOENT)); } if (strlen(attr.za_name) + 1 > namelen) { zap_cursor_fini(&cursor); return (SET_ERROR(ENAMETOOLONG)); } (void) strcpy(name, attr.za_name); if (idp) *idp = attr.za_first_integer; zap_cursor_advance(&cursor); *offp = zap_cursor_serialize(&cursor); zap_cursor_fini(&cursor); return (0); } /* * Find objsets under and including ddobj, call func(ds) on each. */ int dmu_objset_find_dp(dsl_pool_t *dp, uint64_t ddobj, int func(dsl_pool_t *, dsl_dataset_t *, void *), void *arg, int flags) { dsl_dir_t *dd; dsl_dataset_t *ds; zap_cursor_t zc; zap_attribute_t *attr; uint64_t thisobj; int err; ASSERT(dsl_pool_config_held(dp)); err = dsl_dir_hold_obj(dp, ddobj, NULL, FTAG, &dd); if (err != 0) return (err); /* Don't visit hidden ($MOS & $ORIGIN) objsets. */ if (dd->dd_myname[0] == '$') { dsl_dir_rele(dd, FTAG); return (0); } thisobj = dd->dd_phys->dd_head_dataset_obj; attr = kmem_alloc(sizeof (zap_attribute_t), KM_SLEEP); /* * Iterate over all children. */ if (flags & DS_FIND_CHILDREN) { for (zap_cursor_init(&zc, dp->dp_meta_objset, dd->dd_phys->dd_child_dir_zapobj); zap_cursor_retrieve(&zc, attr) == 0; (void) zap_cursor_advance(&zc)) { ASSERT3U(attr->za_integer_length, ==, sizeof (uint64_t)); ASSERT3U(attr->za_num_integers, ==, 1); err = dmu_objset_find_dp(dp, attr->za_first_integer, func, arg, flags); if (err != 0) break; } zap_cursor_fini(&zc); if (err != 0) { dsl_dir_rele(dd, FTAG); kmem_free(attr, sizeof (zap_attribute_t)); return (err); } } /* * Iterate over all snapshots. */ if (flags & DS_FIND_SNAPSHOTS) { dsl_dataset_t *ds; err = dsl_dataset_hold_obj(dp, thisobj, FTAG, &ds); if (err == 0) { uint64_t snapobj = ds->ds_phys->ds_snapnames_zapobj; dsl_dataset_rele(ds, FTAG); for (zap_cursor_init(&zc, dp->dp_meta_objset, snapobj); zap_cursor_retrieve(&zc, attr) == 0; (void) zap_cursor_advance(&zc)) { ASSERT3U(attr->za_integer_length, ==, sizeof (uint64_t)); ASSERT3U(attr->za_num_integers, ==, 1); err = dsl_dataset_hold_obj(dp, attr->za_first_integer, FTAG, &ds); if (err != 0) break; err = func(dp, ds, arg); dsl_dataset_rele(ds, FTAG); if (err != 0) break; } zap_cursor_fini(&zc); } } dsl_dir_rele(dd, FTAG); kmem_free(attr, sizeof (zap_attribute_t)); if (err != 0) return (err); /* * Apply to self. */ err = dsl_dataset_hold_obj(dp, thisobj, FTAG, &ds); if (err != 0) return (err); err = func(dp, ds, arg); dsl_dataset_rele(ds, FTAG); return (err); } /* * Find all objsets under name, and for each, call 'func(child_name, arg)'. * The dp_config_rwlock must not be held when this is called, and it * will not be held when the callback is called. * Therefore this function should only be used when the pool is not changing * (e.g. in syncing context), or the callback can deal with the possible races. */ static int dmu_objset_find_impl(spa_t *spa, const char *name, int func(const char *, void *), void *arg, int flags) { dsl_dir_t *dd; dsl_pool_t *dp = spa_get_dsl(spa); dsl_dataset_t *ds; zap_cursor_t zc; zap_attribute_t *attr; char *child; uint64_t thisobj; int err; dsl_pool_config_enter(dp, FTAG); err = dsl_dir_hold(dp, name, FTAG, &dd, NULL); if (err != 0) { dsl_pool_config_exit(dp, FTAG); return (err); } /* Don't visit hidden ($MOS & $ORIGIN) objsets. */ if (dd->dd_myname[0] == '$') { dsl_dir_rele(dd, FTAG); dsl_pool_config_exit(dp, FTAG); return (0); } thisobj = dd->dd_phys->dd_head_dataset_obj; attr = kmem_alloc(sizeof (zap_attribute_t), KM_SLEEP); /* * Iterate over all children. */ if (flags & DS_FIND_CHILDREN) { for (zap_cursor_init(&zc, dp->dp_meta_objset, dd->dd_phys->dd_child_dir_zapobj); zap_cursor_retrieve(&zc, attr) == 0; (void) zap_cursor_advance(&zc)) { ASSERT3U(attr->za_integer_length, ==, sizeof (uint64_t)); ASSERT3U(attr->za_num_integers, ==, 1); child = kmem_asprintf("%s/%s", name, attr->za_name); dsl_pool_config_exit(dp, FTAG); err = dmu_objset_find_impl(spa, child, func, arg, flags); dsl_pool_config_enter(dp, FTAG); strfree(child); if (err != 0) break; } zap_cursor_fini(&zc); if (err != 0) { dsl_dir_rele(dd, FTAG); dsl_pool_config_exit(dp, FTAG); kmem_free(attr, sizeof (zap_attribute_t)); return (err); } } /* * Iterate over all snapshots. */ if (flags & DS_FIND_SNAPSHOTS) { err = dsl_dataset_hold_obj(dp, thisobj, FTAG, &ds); if (err == 0) { uint64_t snapobj = ds->ds_phys->ds_snapnames_zapobj; dsl_dataset_rele(ds, FTAG); for (zap_cursor_init(&zc, dp->dp_meta_objset, snapobj); zap_cursor_retrieve(&zc, attr) == 0; (void) zap_cursor_advance(&zc)) { ASSERT3U(attr->za_integer_length, ==, sizeof (uint64_t)); ASSERT3U(attr->za_num_integers, ==, 1); child = kmem_asprintf("%s@%s", name, attr->za_name); dsl_pool_config_exit(dp, FTAG); err = func(child, arg); dsl_pool_config_enter(dp, FTAG); strfree(child); if (err != 0) break; } zap_cursor_fini(&zc); } } dsl_dir_rele(dd, FTAG); kmem_free(attr, sizeof (zap_attribute_t)); dsl_pool_config_exit(dp, FTAG); if (err != 0) return (err); /* Apply to self. */ return (func(name, arg)); } /* * See comment above dmu_objset_find_impl(). */ int dmu_objset_find(char *name, int func(const char *, void *), void *arg, int flags) { spa_t *spa; int error; error = spa_open(name, &spa, FTAG); if (error != 0) return (error); error = dmu_objset_find_impl(spa, name, func, arg, flags); spa_close(spa, FTAG); return (error); } void dmu_objset_set_user(objset_t *os, void *user_ptr) { ASSERT(MUTEX_HELD(&os->os_user_ptr_lock)); os->os_user_ptr = user_ptr; } void * dmu_objset_get_user(objset_t *os) { ASSERT(MUTEX_HELD(&os->os_user_ptr_lock)); return (os->os_user_ptr); } /* * Determine name of filesystem, given name of snapshot. * buf must be at least MAXNAMELEN bytes */ int dmu_fsname(const char *snapname, char *buf) { char *atp = strchr(snapname, '@'); if (atp == NULL) return (SET_ERROR(EINVAL)); if (atp - snapname >= MAXNAMELEN) return (SET_ERROR(ENAMETOOLONG)); (void) strlcpy(buf, snapname, atp - snapname + 1); return (0); } Index: stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu.h =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu.h (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu.h (revision 268647) @@ -1,825 +1,825 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2011, 2014 by Delphix. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. * Copyright (c) 2012, Joyent, Inc. All rights reserved. * Copyright 2013 DEY Storage Systems, Inc. */ /* Portions Copyright 2010 Robert Milkowski */ #ifndef _SYS_DMU_H #define _SYS_DMU_H /* * This file describes the interface that the DMU provides for its * consumers. * * The DMU also interacts with the SPA. That interface is described in * dmu_spa.h. */ #include #include #include #include #include #ifdef __cplusplus extern "C" { #endif struct uio; struct xuio; struct page; struct vnode; struct spa; struct zilog; struct zio; struct blkptr; struct zap_cursor; struct dsl_dataset; struct dsl_pool; struct dnode; struct drr_begin; struct drr_end; struct zbookmark; struct spa; struct nvlist; struct arc_buf; struct zio_prop; struct sa_handle; struct file; typedef struct objset objset_t; typedef struct dmu_tx dmu_tx_t; typedef struct dsl_dir dsl_dir_t; typedef enum dmu_object_byteswap { DMU_BSWAP_UINT8, DMU_BSWAP_UINT16, DMU_BSWAP_UINT32, DMU_BSWAP_UINT64, DMU_BSWAP_ZAP, DMU_BSWAP_DNODE, DMU_BSWAP_OBJSET, DMU_BSWAP_ZNODE, DMU_BSWAP_OLDACL, DMU_BSWAP_ACL, /* * Allocating a new byteswap type number makes the on-disk format * incompatible with any other format that uses the same number. * * Data can usually be structured to work with one of the * DMU_BSWAP_UINT* or DMU_BSWAP_ZAP types. */ DMU_BSWAP_NUMFUNCS } dmu_object_byteswap_t; #define DMU_OT_NEWTYPE 0x80 #define DMU_OT_METADATA 0x40 #define DMU_OT_BYTESWAP_MASK 0x3f /* * Defines a uint8_t object type. Object types specify if the data * in the object is metadata (boolean) and how to byteswap the data * (dmu_object_byteswap_t). */ #define DMU_OT(byteswap, metadata) \ (DMU_OT_NEWTYPE | \ ((metadata) ? DMU_OT_METADATA : 0) | \ ((byteswap) & DMU_OT_BYTESWAP_MASK)) #define DMU_OT_IS_VALID(ot) (((ot) & DMU_OT_NEWTYPE) ? \ ((ot) & DMU_OT_BYTESWAP_MASK) < DMU_BSWAP_NUMFUNCS : \ (ot) < DMU_OT_NUMTYPES) #define DMU_OT_IS_METADATA(ot) (((ot) & DMU_OT_NEWTYPE) ? \ ((ot) & DMU_OT_METADATA) : \ dmu_ot[(ot)].ot_metadata) #define DMU_OT_BYTESWAP(ot) (((ot) & DMU_OT_NEWTYPE) ? \ ((ot) & DMU_OT_BYTESWAP_MASK) : \ dmu_ot[(ot)].ot_byteswap) typedef enum dmu_object_type { DMU_OT_NONE, /* general: */ DMU_OT_OBJECT_DIRECTORY, /* ZAP */ DMU_OT_OBJECT_ARRAY, /* UINT64 */ DMU_OT_PACKED_NVLIST, /* UINT8 (XDR by nvlist_pack/unpack) */ DMU_OT_PACKED_NVLIST_SIZE, /* UINT64 */ DMU_OT_BPOBJ, /* UINT64 */ DMU_OT_BPOBJ_HDR, /* UINT64 */ /* spa: */ DMU_OT_SPACE_MAP_HEADER, /* UINT64 */ DMU_OT_SPACE_MAP, /* UINT64 */ /* zil: */ DMU_OT_INTENT_LOG, /* UINT64 */ /* dmu: */ DMU_OT_DNODE, /* DNODE */ DMU_OT_OBJSET, /* OBJSET */ /* dsl: */ DMU_OT_DSL_DIR, /* UINT64 */ DMU_OT_DSL_DIR_CHILD_MAP, /* ZAP */ DMU_OT_DSL_DS_SNAP_MAP, /* ZAP */ DMU_OT_DSL_PROPS, /* ZAP */ DMU_OT_DSL_DATASET, /* UINT64 */ /* zpl: */ DMU_OT_ZNODE, /* ZNODE */ DMU_OT_OLDACL, /* Old ACL */ DMU_OT_PLAIN_FILE_CONTENTS, /* UINT8 */ DMU_OT_DIRECTORY_CONTENTS, /* ZAP */ DMU_OT_MASTER_NODE, /* ZAP */ DMU_OT_UNLINKED_SET, /* ZAP */ /* zvol: */ DMU_OT_ZVOL, /* UINT8 */ DMU_OT_ZVOL_PROP, /* ZAP */ /* other; for testing only! */ DMU_OT_PLAIN_OTHER, /* UINT8 */ DMU_OT_UINT64_OTHER, /* UINT64 */ DMU_OT_ZAP_OTHER, /* ZAP */ /* new object types: */ DMU_OT_ERROR_LOG, /* ZAP */ DMU_OT_SPA_HISTORY, /* UINT8 */ DMU_OT_SPA_HISTORY_OFFSETS, /* spa_his_phys_t */ DMU_OT_POOL_PROPS, /* ZAP */ DMU_OT_DSL_PERMS, /* ZAP */ DMU_OT_ACL, /* ACL */ DMU_OT_SYSACL, /* SYSACL */ DMU_OT_FUID, /* FUID table (Packed NVLIST UINT8) */ DMU_OT_FUID_SIZE, /* FUID table size UINT64 */ DMU_OT_NEXT_CLONES, /* ZAP */ DMU_OT_SCAN_QUEUE, /* ZAP */ DMU_OT_USERGROUP_USED, /* ZAP */ DMU_OT_USERGROUP_QUOTA, /* ZAP */ DMU_OT_USERREFS, /* ZAP */ DMU_OT_DDT_ZAP, /* ZAP */ DMU_OT_DDT_STATS, /* ZAP */ DMU_OT_SA, /* System attr */ DMU_OT_SA_MASTER_NODE, /* ZAP */ DMU_OT_SA_ATTR_REGISTRATION, /* ZAP */ DMU_OT_SA_ATTR_LAYOUTS, /* ZAP */ DMU_OT_SCAN_XLATE, /* ZAP */ DMU_OT_DEDUP, /* fake dedup BP from ddt_bp_create() */ DMU_OT_DEADLIST, /* ZAP */ DMU_OT_DEADLIST_HDR, /* UINT64 */ DMU_OT_DSL_CLONES, /* ZAP */ DMU_OT_BPOBJ_SUBOBJ, /* UINT64 */ /* * Do not allocate new object types here. Doing so makes the on-disk * format incompatible with any other format that uses the same object * type number. * * When creating an object which does not have one of the above types * use the DMU_OTN_* type with the correct byteswap and metadata * values. * * The DMU_OTN_* types do not have entries in the dmu_ot table, * use the DMU_OT_IS_METDATA() and DMU_OT_BYTESWAP() macros instead * of indexing into dmu_ot directly (this works for both DMU_OT_* types * and DMU_OTN_* types). */ DMU_OT_NUMTYPES, /* * Names for valid types declared with DMU_OT(). */ DMU_OTN_UINT8_DATA = DMU_OT(DMU_BSWAP_UINT8, B_FALSE), DMU_OTN_UINT8_METADATA = DMU_OT(DMU_BSWAP_UINT8, B_TRUE), DMU_OTN_UINT16_DATA = DMU_OT(DMU_BSWAP_UINT16, B_FALSE), DMU_OTN_UINT16_METADATA = DMU_OT(DMU_BSWAP_UINT16, B_TRUE), DMU_OTN_UINT32_DATA = DMU_OT(DMU_BSWAP_UINT32, B_FALSE), DMU_OTN_UINT32_METADATA = DMU_OT(DMU_BSWAP_UINT32, B_TRUE), DMU_OTN_UINT64_DATA = DMU_OT(DMU_BSWAP_UINT64, B_FALSE), DMU_OTN_UINT64_METADATA = DMU_OT(DMU_BSWAP_UINT64, B_TRUE), DMU_OTN_ZAP_DATA = DMU_OT(DMU_BSWAP_ZAP, B_FALSE), DMU_OTN_ZAP_METADATA = DMU_OT(DMU_BSWAP_ZAP, B_TRUE), } dmu_object_type_t; typedef enum txg_how { TXG_WAIT = 1, TXG_NOWAIT, TXG_WAITED, } txg_how_t; void byteswap_uint64_array(void *buf, size_t size); void byteswap_uint32_array(void *buf, size_t size); void byteswap_uint16_array(void *buf, size_t size); void byteswap_uint8_array(void *buf, size_t size); void zap_byteswap(void *buf, size_t size); void zfs_oldacl_byteswap(void *buf, size_t size); void zfs_acl_byteswap(void *buf, size_t size); void zfs_znode_byteswap(void *buf, size_t size); #define DS_FIND_SNAPSHOTS (1<<0) #define DS_FIND_CHILDREN (1<<1) /* * The maximum number of bytes that can be accessed as part of one * operation, including metadata. */ #define DMU_MAX_ACCESS (10<<20) /* 10MB */ #define DMU_MAX_DELETEBLKCNT (20480) /* ~5MB of indirect blocks */ #define DMU_USERUSED_OBJECT (-1ULL) #define DMU_GROUPUSED_OBJECT (-2ULL) #define DMU_DEADLIST_OBJECT (-3ULL) /* * artificial blkids for bonus buffer and spill blocks */ #define DMU_BONUS_BLKID (-1ULL) #define DMU_SPILL_BLKID (-2ULL) /* * Public routines to create, destroy, open, and close objsets. */ int dmu_objset_hold(const char *name, void *tag, objset_t **osp); int dmu_objset_own(const char *name, dmu_objset_type_t type, boolean_t readonly, void *tag, objset_t **osp); void dmu_objset_rele(objset_t *os, void *tag); void dmu_objset_disown(objset_t *os, void *tag); int dmu_objset_open_ds(struct dsl_dataset *ds, objset_t **osp); void dmu_objset_evict_dbufs(objset_t *os); int dmu_objset_create(const char *name, dmu_objset_type_t type, uint64_t flags, void (*func)(objset_t *os, void *arg, cred_t *cr, dmu_tx_t *tx), void *arg); int dmu_get_recursive_snaps_nvl(char *fsname, const char *snapname, struct nvlist *snaps); int dmu_objset_clone(const char *name, const char *origin); int dsl_destroy_snapshots_nvl(struct nvlist *snaps, boolean_t defer, struct nvlist *errlist); int dmu_objset_snapshot_one(const char *fsname, const char *snapname); int dmu_objset_snapshot_tmp(const char *, const char *, int); int dmu_objset_find(char *name, int func(const char *, void *), void *arg, int flags); void dmu_objset_byteswap(void *buf, size_t size); int dsl_dataset_rename_snapshot(const char *fsname, const char *oldsnapname, const char *newsnapname, boolean_t recursive); typedef struct dmu_buf { uint64_t db_object; /* object that this buffer is part of */ uint64_t db_offset; /* byte offset in this object */ uint64_t db_size; /* size of buffer in bytes */ void *db_data; /* data in buffer */ } dmu_buf_t; typedef void dmu_buf_evict_func_t(struct dmu_buf *db, void *user_ptr); /* * The names of zap entries in the DIRECTORY_OBJECT of the MOS. */ #define DMU_POOL_DIRECTORY_OBJECT 1 #define DMU_POOL_CONFIG "config" #define DMU_POOL_FEATURES_FOR_WRITE "features_for_write" #define DMU_POOL_FEATURES_FOR_READ "features_for_read" #define DMU_POOL_FEATURE_DESCRIPTIONS "feature_descriptions" #define DMU_POOL_FEATURE_ENABLED_TXG "feature_enabled_txg" #define DMU_POOL_ROOT_DATASET "root_dataset" #define DMU_POOL_SYNC_BPOBJ "sync_bplist" #define DMU_POOL_ERRLOG_SCRUB "errlog_scrub" #define DMU_POOL_ERRLOG_LAST "errlog_last" #define DMU_POOL_SPARES "spares" #define DMU_POOL_DEFLATE "deflate" #define DMU_POOL_HISTORY "history" #define DMU_POOL_PROPS "pool_props" #define DMU_POOL_L2CACHE "l2cache" #define DMU_POOL_TMP_USERREFS "tmp_userrefs" #define DMU_POOL_DDT "DDT-%s-%s-%s" #define DMU_POOL_DDT_STATS "DDT-statistics" #define DMU_POOL_CREATION_VERSION "creation_version" #define DMU_POOL_SCAN "scan" #define DMU_POOL_FREE_BPOBJ "free_bpobj" #define DMU_POOL_BPTREE_OBJ "bptree_obj" #define DMU_POOL_EMPTY_BPOBJ "empty_bpobj" /* * Allocate an object from this objset. The range of object numbers * available is (0, DN_MAX_OBJECT). Object 0 is the meta-dnode. * * The transaction must be assigned to a txg. The newly allocated * object will be "held" in the transaction (ie. you can modify the * newly allocated object in this transaction). * * dmu_object_alloc() chooses an object and returns it in *objectp. * * dmu_object_claim() allocates a specific object number. If that * number is already allocated, it fails and returns EEXIST. * * Return 0 on success, or ENOSPC or EEXIST as specified above. */ uint64_t dmu_object_alloc(objset_t *os, dmu_object_type_t ot, int blocksize, dmu_object_type_t bonus_type, int bonus_len, dmu_tx_t *tx); int dmu_object_claim(objset_t *os, uint64_t object, dmu_object_type_t ot, int blocksize, dmu_object_type_t bonus_type, int bonus_len, dmu_tx_t *tx); int dmu_object_reclaim(objset_t *os, uint64_t object, dmu_object_type_t ot, int blocksize, dmu_object_type_t bonustype, int bonuslen); /* * Free an object from this objset. * * The object's data will be freed as well (ie. you don't need to call * dmu_free(object, 0, -1, tx)). * * The object need not be held in the transaction. * * If there are any holds on this object's buffers (via dmu_buf_hold()), * or tx holds on the object (via dmu_tx_hold_object()), you can not * free it; it fails and returns EBUSY. * * If the object is not allocated, it fails and returns ENOENT. * * Return 0 on success, or EBUSY or ENOENT as specified above. */ int dmu_object_free(objset_t *os, uint64_t object, dmu_tx_t *tx); /* * Find the next allocated or free object. * * The objectp parameter is in-out. It will be updated to be the next * object which is allocated. Ignore objects which have not been * modified since txg. * * XXX Can only be called on a objset with no dirty data. * * Returns 0 on success, or ENOENT if there are no more objects. */ int dmu_object_next(objset_t *os, uint64_t *objectp, boolean_t hole, uint64_t txg); /* * Set the data blocksize for an object. * * The object cannot have any blocks allcated beyond the first. If * the first block is allocated already, the new size must be greater * than the current block size. If these conditions are not met, * ENOTSUP will be returned. * * Returns 0 on success, or EBUSY if there are any holds on the object * contents, or ENOTSUP as described above. */ int dmu_object_set_blocksize(objset_t *os, uint64_t object, uint64_t size, int ibs, dmu_tx_t *tx); /* * Set the checksum property on a dnode. The new checksum algorithm will * apply to all newly written blocks; existing blocks will not be affected. */ void dmu_object_set_checksum(objset_t *os, uint64_t object, uint8_t checksum, dmu_tx_t *tx); /* * Set the compress property on a dnode. The new compression algorithm will * apply to all newly written blocks; existing blocks will not be affected. */ void dmu_object_set_compress(objset_t *os, uint64_t object, uint8_t compress, dmu_tx_t *tx); /* * Decide how to write a block: checksum, compression, number of copies, etc. */ #define WP_NOFILL 0x1 #define WP_DMU_SYNC 0x2 #define WP_SPILL 0x4 void dmu_write_policy(objset_t *os, struct dnode *dn, int level, int wp, struct zio_prop *zp); /* * The bonus data is accessed more or less like a regular buffer. * You must dmu_bonus_hold() to get the buffer, which will give you a * dmu_buf_t with db_offset==-1ULL, and db_size = the size of the bonus * data. As with any normal buffer, you must call dmu_buf_read() to * read db_data, dmu_buf_will_dirty() before modifying it, and the * object must be held in an assigned transaction before calling * dmu_buf_will_dirty. You may use dmu_buf_set_user() on the bonus * buffer as well. You must release your hold with dmu_buf_rele(). * * Returns ENOENT, EIO, or 0. */ int dmu_bonus_hold(objset_t *os, uint64_t object, void *tag, dmu_buf_t **); int dmu_bonus_max(void); int dmu_set_bonus(dmu_buf_t *, int, dmu_tx_t *); int dmu_set_bonustype(dmu_buf_t *, dmu_object_type_t, dmu_tx_t *); dmu_object_type_t dmu_get_bonustype(dmu_buf_t *); int dmu_rm_spill(objset_t *, uint64_t, dmu_tx_t *); /* * Special spill buffer support used by "SA" framework */ int dmu_spill_hold_by_bonus(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp); int dmu_spill_hold_by_dnode(struct dnode *dn, uint32_t flags, void *tag, dmu_buf_t **dbp); int dmu_spill_hold_existing(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp); /* * Obtain the DMU buffer from the specified object which contains the * specified offset. dmu_buf_hold() puts a "hold" on the buffer, so * that it will remain in memory. You must release the hold with * dmu_buf_rele(). You musn't access the dmu_buf_t after releasing your * hold. You must have a hold on any dmu_buf_t* you pass to the DMU. * * You must call dmu_buf_read, dmu_buf_will_dirty, or dmu_buf_will_fill * on the returned buffer before reading or writing the buffer's * db_data. The comments for those routines describe what particular * operations are valid after calling them. * * The object number must be a valid, allocated object number. */ int dmu_buf_hold(objset_t *os, uint64_t object, uint64_t offset, void *tag, dmu_buf_t **, int flags); void dmu_buf_add_ref(dmu_buf_t *db, void* tag); void dmu_buf_rele(dmu_buf_t *db, void *tag); uint64_t dmu_buf_refcount(dmu_buf_t *db); /* * dmu_buf_hold_array holds the DMU buffers which contain all bytes in a * range of an object. A pointer to an array of dmu_buf_t*'s is * returned (in *dbpp). * * dmu_buf_rele_array releases the hold on an array of dmu_buf_t*'s, and * frees the array. The hold on the array of buffers MUST be released * with dmu_buf_rele_array. You can NOT release the hold on each buffer * individually with dmu_buf_rele. */ int dmu_buf_hold_array_by_bonus(dmu_buf_t *db, uint64_t offset, uint64_t length, int read, void *tag, int *numbufsp, dmu_buf_t ***dbpp); void dmu_buf_rele_array(dmu_buf_t **, int numbufs, void *tag); /* * Returns NULL on success, or the existing user ptr if it's already * been set. * * user_ptr is for use by the user and can be obtained via dmu_buf_get_user(). * * user_data_ptr_ptr should be NULL, or a pointer to a pointer which * will be set to db->db_data when you are allowed to access it. Note * that db->db_data (the pointer) can change when you do dmu_buf_read(), * dmu_buf_tryupgrade(), dmu_buf_will_dirty(), or dmu_buf_will_fill(). * *user_data_ptr_ptr will be set to the new value when it changes. * * If non-NULL, pageout func will be called when this buffer is being * excised from the cache, so that you can clean up the data structure * pointed to by user_ptr. * * dmu_evict_user() will call the pageout func for all buffers in a * objset with a given pageout func. */ void *dmu_buf_set_user(dmu_buf_t *db, void *user_ptr, void *user_data_ptr_ptr, dmu_buf_evict_func_t *pageout_func); /* * set_user_ie is the same as set_user, but request immediate eviction * when hold count goes to zero. */ void *dmu_buf_set_user_ie(dmu_buf_t *db, void *user_ptr, void *user_data_ptr_ptr, dmu_buf_evict_func_t *pageout_func); void *dmu_buf_update_user(dmu_buf_t *db_fake, void *old_user_ptr, void *user_ptr, void *user_data_ptr_ptr, dmu_buf_evict_func_t *pageout_func); void dmu_evict_user(objset_t *os, dmu_buf_evict_func_t *func); /* * Returns the user_ptr set with dmu_buf_set_user(), or NULL if not set. */ void *dmu_buf_get_user(dmu_buf_t *db); /* * Returns the blkptr associated with this dbuf, or NULL if not set. */ struct blkptr *dmu_buf_get_blkptr(dmu_buf_t *db); /* * Indicate that you are going to modify the buffer's data (db_data). * * The transaction (tx) must be assigned to a txg (ie. you've called * dmu_tx_assign()). The buffer's object must be held in the tx * (ie. you've called dmu_tx_hold_object(tx, db->db_object)). */ void dmu_buf_will_dirty(dmu_buf_t *db, dmu_tx_t *tx); /* * Tells if the given dbuf is freeable. */ boolean_t dmu_buf_freeable(dmu_buf_t *); /* * You must create a transaction, then hold the objects which you will * (or might) modify as part of this transaction. Then you must assign * the transaction to a transaction group. Once the transaction has * been assigned, you can modify buffers which belong to held objects as * part of this transaction. You can't modify buffers before the * transaction has been assigned; you can't modify buffers which don't * belong to objects which this transaction holds; you can't hold * objects once the transaction has been assigned. You may hold an * object which you are going to free (with dmu_object_free()), but you * don't have to. * * You can abort the transaction before it has been assigned. * * Note that you may hold buffers (with dmu_buf_hold) at any time, * regardless of transaction state. */ #define DMU_NEW_OBJECT (-1ULL) #define DMU_OBJECT_END (-1ULL) dmu_tx_t *dmu_tx_create(objset_t *os); void dmu_tx_hold_write(dmu_tx_t *tx, uint64_t object, uint64_t off, int len); void dmu_tx_hold_free(dmu_tx_t *tx, uint64_t object, uint64_t off, uint64_t len); void dmu_tx_hold_zap(dmu_tx_t *tx, uint64_t object, int add, const char *name); void dmu_tx_hold_bonus(dmu_tx_t *tx, uint64_t object); void dmu_tx_hold_spill(dmu_tx_t *tx, uint64_t object); void dmu_tx_hold_sa(dmu_tx_t *tx, struct sa_handle *hdl, boolean_t may_grow); void dmu_tx_hold_sa_create(dmu_tx_t *tx, int total_size); void dmu_tx_abort(dmu_tx_t *tx); int dmu_tx_assign(dmu_tx_t *tx, enum txg_how txg_how); void dmu_tx_wait(dmu_tx_t *tx); void dmu_tx_commit(dmu_tx_t *tx); /* * To register a commit callback, dmu_tx_callback_register() must be called. * * dcb_data is a pointer to caller private data that is passed on as a * callback parameter. The caller is responsible for properly allocating and * freeing it. * * When registering a callback, the transaction must be already created, but * it cannot be committed or aborted. It can be assigned to a txg or not. * * The callback will be called after the transaction has been safely written * to stable storage and will also be called if the dmu_tx is aborted. * If there is any error which prevents the transaction from being committed to * disk, the callback will be called with a value of error != 0. */ typedef void dmu_tx_callback_func_t(void *dcb_data, int error); void dmu_tx_callback_register(dmu_tx_t *tx, dmu_tx_callback_func_t *dcb_func, void *dcb_data); /* * Free up the data blocks for a defined range of a file. If size is * -1, the range from offset to end-of-file is freed. */ int dmu_free_range(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, dmu_tx_t *tx); int dmu_free_long_range(objset_t *os, uint64_t object, uint64_t offset, uint64_t size); int dmu_free_long_object(objset_t *os, uint64_t object); /* * Convenience functions. * * Canfail routines will return 0 on success, or an errno if there is a * nonrecoverable I/O error. */ #define DMU_READ_PREFETCH 0 /* prefetch */ #define DMU_READ_NO_PREFETCH 1 /* don't prefetch */ int dmu_read(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, void *buf, uint32_t flags); void dmu_write(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, const void *buf, dmu_tx_t *tx); void dmu_prealloc(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, dmu_tx_t *tx); int dmu_read_uio(objset_t *os, uint64_t object, struct uio *uio, uint64_t size); int dmu_write_uio(objset_t *os, uint64_t object, struct uio *uio, uint64_t size, dmu_tx_t *tx); int dmu_write_uio_dbuf(dmu_buf_t *zdb, struct uio *uio, uint64_t size, dmu_tx_t *tx); #ifdef _KERNEL #ifdef sun int dmu_write_pages(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, struct page *pp, dmu_tx_t *tx); #else int dmu_write_pages(objset_t *os, uint64_t object, uint64_t offset, uint64_t size, struct vm_page **ppa, dmu_tx_t *tx); #endif #endif struct arc_buf *dmu_request_arcbuf(dmu_buf_t *handle, int size); void dmu_return_arcbuf(struct arc_buf *buf); void dmu_assign_arcbuf(dmu_buf_t *handle, uint64_t offset, struct arc_buf *buf, dmu_tx_t *tx); int dmu_xuio_init(struct xuio *uio, int niov); void dmu_xuio_fini(struct xuio *uio); int dmu_xuio_add(struct xuio *uio, struct arc_buf *abuf, offset_t off, size_t n); int dmu_xuio_cnt(struct xuio *uio); struct arc_buf *dmu_xuio_arcbuf(struct xuio *uio, int i); void dmu_xuio_clear(struct xuio *uio, int i); void xuio_stat_wbuf_copied(); void xuio_stat_wbuf_nocopy(); extern int zfs_prefetch_disable; /* * Asynchronously try to read in the data. */ void dmu_prefetch(objset_t *os, uint64_t object, uint64_t offset, uint64_t len); typedef struct dmu_object_info { /* All sizes are in bytes unless otherwise indicated. */ uint32_t doi_data_block_size; uint32_t doi_metadata_block_size; dmu_object_type_t doi_type; dmu_object_type_t doi_bonus_type; uint64_t doi_bonus_size; uint8_t doi_indirection; /* 2 = dnode->indirect->data */ uint8_t doi_checksum; uint8_t doi_compress; uint8_t doi_pad[5]; uint64_t doi_physical_blocks_512; /* data + metadata, 512b blks */ uint64_t doi_max_offset; uint64_t doi_fill_count; /* number of non-empty blocks */ } dmu_object_info_t; typedef void arc_byteswap_func_t(void *buf, size_t size); typedef struct dmu_object_type_info { dmu_object_byteswap_t ot_byteswap; boolean_t ot_metadata; char *ot_name; } dmu_object_type_info_t; typedef struct dmu_object_byteswap_info { arc_byteswap_func_t *ob_func; char *ob_name; } dmu_object_byteswap_info_t; extern const dmu_object_type_info_t dmu_ot[DMU_OT_NUMTYPES]; extern const dmu_object_byteswap_info_t dmu_ot_byteswap[DMU_BSWAP_NUMFUNCS]; /* * Get information on a DMU object. * * Return 0 on success or ENOENT if object is not allocated. * * If doi is NULL, just indicates whether the object exists. */ int dmu_object_info(objset_t *os, uint64_t object, dmu_object_info_t *doi); /* Like dmu_object_info, but faster if you have a held dnode in hand. */ void dmu_object_info_from_dnode(struct dnode *dn, dmu_object_info_t *doi); /* Like dmu_object_info, but faster if you have a held dbuf in hand. */ void dmu_object_info_from_db(dmu_buf_t *db, dmu_object_info_t *doi); /* * Like dmu_object_info_from_db, but faster still when you only care about * the size. This is specifically optimized for zfs_getattr(). */ void dmu_object_size_from_db(dmu_buf_t *db, uint32_t *blksize, u_longlong_t *nblk512); typedef struct dmu_objset_stats { uint64_t dds_num_clones; /* number of clones of this */ uint64_t dds_creation_txg; uint64_t dds_guid; dmu_objset_type_t dds_type; uint8_t dds_is_snapshot; uint8_t dds_inconsistent; char dds_origin[MAXNAMELEN]; } dmu_objset_stats_t; /* * Get stats on a dataset. */ void dmu_objset_fast_stat(objset_t *os, dmu_objset_stats_t *stat); /* * Add entries to the nvlist for all the objset's properties. See * zfs_prop_table[] and zfs(1m) for details on the properties. */ void dmu_objset_stats(objset_t *os, struct nvlist *nv); /* * Get the space usage statistics for statvfs(). * * refdbytes is the amount of space "referenced" by this objset. * availbytes is the amount of space available to this objset, taking * into account quotas & reservations, assuming that no other objsets * use the space first. These values correspond to the 'referenced' and * 'available' properties, described in the zfs(1m) manpage. * * usedobjs and availobjs are the number of objects currently allocated, * and available. */ void dmu_objset_space(objset_t *os, uint64_t *refdbytesp, uint64_t *availbytesp, uint64_t *usedobjsp, uint64_t *availobjsp); /* * The fsid_guid is a 56-bit ID that can change to avoid collisions. * (Contrast with the ds_guid which is a 64-bit ID that will never * change, so there is a small probability that it will collide.) */ uint64_t dmu_objset_fsid_guid(objset_t *os); /* * Get the [cm]time for an objset's snapshot dir */ timestruc_t dmu_objset_snap_cmtime(objset_t *os); int dmu_objset_is_snapshot(objset_t *os); extern struct spa *dmu_objset_spa(objset_t *os); extern struct zilog *dmu_objset_zil(objset_t *os); extern struct dsl_pool *dmu_objset_pool(objset_t *os); extern struct dsl_dataset *dmu_objset_ds(objset_t *os); extern void dmu_objset_name(objset_t *os, char *buf); extern dmu_objset_type_t dmu_objset_type(objset_t *os); extern uint64_t dmu_objset_id(objset_t *os); -extern uint64_t dmu_objset_syncprop(objset_t *os); -extern uint64_t dmu_objset_logbias(objset_t *os); +extern zfs_sync_type_t dmu_objset_syncprop(objset_t *os); +extern zfs_logbias_op_t dmu_objset_logbias(objset_t *os); extern int dmu_snapshot_list_next(objset_t *os, int namelen, char *name, uint64_t *id, uint64_t *offp, boolean_t *case_conflict); extern int dmu_snapshot_realname(objset_t *os, char *name, char *real, int maxlen, boolean_t *conflict); extern int dmu_dir_list_next(objset_t *os, int namelen, char *name, uint64_t *idp, uint64_t *offp); typedef int objset_used_cb_t(dmu_object_type_t bonustype, void *bonus, uint64_t *userp, uint64_t *groupp); extern void dmu_objset_register_type(dmu_objset_type_t ost, objset_used_cb_t *cb); extern void dmu_objset_set_user(objset_t *os, void *user_ptr); extern void *dmu_objset_get_user(objset_t *os); /* * Return the txg number for the given assigned transaction. */ uint64_t dmu_tx_get_txg(dmu_tx_t *tx); /* * Synchronous write. * If a parent zio is provided this function initiates a write on the * provided buffer as a child of the parent zio. * In the absence of a parent zio, the write is completed synchronously. * At write completion, blk is filled with the bp of the written block. * Note that while the data covered by this function will be on stable * storage when the write completes this new data does not become a * permanent part of the file until the associated transaction commits. */ /* * {zfs,zvol,ztest}_get_done() args */ typedef struct zgd { struct zilog *zgd_zilog; struct blkptr *zgd_bp; dmu_buf_t *zgd_db; struct rl *zgd_rl; void *zgd_private; } zgd_t; typedef void dmu_sync_cb_t(zgd_t *arg, int error); int dmu_sync(struct zio *zio, uint64_t txg, dmu_sync_cb_t *done, zgd_t *zgd); /* * Find the next hole or data block in file starting at *off * Return found offset in *off. Return ESRCH for end of file. */ int dmu_offset_next(objset_t *os, uint64_t object, boolean_t hole, uint64_t *off); /* * Initial setup and final teardown. */ extern void dmu_init(void); extern void dmu_fini(void); typedef void (*dmu_traverse_cb_t)(objset_t *os, void *arg, struct blkptr *bp, uint64_t object, uint64_t offset, int len); void dmu_traverse_objset(objset_t *os, uint64_t txg_start, dmu_traverse_cb_t cb, void *arg); int dmu_diff(const char *tosnap_name, const char *fromsnap_name, struct file *fp, offset_t *offp); /* CRC64 table */ #define ZFS_CRC64_POLY 0xC96C5795D7870F42ULL /* ECMA-182, reflected form */ extern uint64_t zfs_crc64_table[256]; extern int zfs_mdcomp_disable; #ifdef __cplusplus } #endif #endif /* _SYS_DMU_H */ Index: stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu_objset.h =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu_objset.h (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu_objset.h (revision 268647) @@ -1,178 +1,179 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2012 by Delphix. All rights reserved. + * Copyright (c) 2012, 2014 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ /* Portions Copyright 2010 Robert Milkowski */ #ifndef _SYS_DMU_OBJSET_H #define _SYS_DMU_OBJSET_H #include #include #include #include #include #include #include #include #ifdef __cplusplus extern "C" { #endif extern krwlock_t os_lock; struct dsl_pool; struct dsl_dataset; struct dmu_tx; #define OBJSET_PHYS_SIZE 2048 #define OBJSET_OLD_PHYS_SIZE 1024 #define OBJSET_BUF_HAS_USERUSED(buf) \ (arc_buf_size(buf) > OBJSET_OLD_PHYS_SIZE) #define OBJSET_FLAG_USERACCOUNTING_COMPLETE (1ULL<<0) typedef struct objset_phys { dnode_phys_t os_meta_dnode; zil_header_t os_zil_header; uint64_t os_type; uint64_t os_flags; char os_pad[OBJSET_PHYS_SIZE - sizeof (dnode_phys_t)*3 - sizeof (zil_header_t) - sizeof (uint64_t)*2]; dnode_phys_t os_userused_dnode; dnode_phys_t os_groupused_dnode; } objset_phys_t; struct objset { /* Immutable: */ struct dsl_dataset *os_dsl_dataset; spa_t *os_spa; arc_buf_t *os_phys_buf; objset_phys_t *os_phys; /* * The following "special" dnodes have no parent and are exempt from * dnode_move(), but they root their descendents in this objset using * handles anyway, so that all access to dnodes from dbufs consistently * uses handles. */ dnode_handle_t os_meta_dnode; dnode_handle_t os_userused_dnode; dnode_handle_t os_groupused_dnode; zilog_t *os_zil; /* can change, under dsl_dir's locks: */ - uint8_t os_checksum; - uint8_t os_compress; + enum zio_checksum os_checksum; + enum zio_compress os_compress; uint8_t os_copies; - uint8_t os_dedup_checksum; - uint8_t os_dedup_verify; - uint8_t os_logbias; - uint8_t os_primary_cache; - uint8_t os_secondary_cache; - uint8_t os_sync; + enum zio_checksum os_dedup_checksum; + boolean_t os_dedup_verify; + zfs_logbias_op_t os_logbias; + zfs_cache_type_t os_primary_cache; + zfs_cache_type_t os_secondary_cache; + zfs_sync_type_t os_sync; + zfs_redundant_metadata_type_t os_redundant_metadata; /* no lock needed: */ struct dmu_tx *os_synctx; /* XXX sketchy */ blkptr_t *os_rootbp; zil_header_t os_zil_header; list_t os_synced_dnodes; uint64_t os_flags; /* Protected by os_obj_lock */ kmutex_t os_obj_lock; uint64_t os_obj_next; /* Protected by os_lock */ kmutex_t os_lock; list_t os_dirty_dnodes[TXG_SIZE]; list_t os_free_dnodes[TXG_SIZE]; list_t os_dnodes; list_t os_downgraded_dbufs; /* stuff we store for the user */ kmutex_t os_user_ptr_lock; void *os_user_ptr; sa_os_t *os_sa; }; #define DMU_META_OBJSET 0 #define DMU_META_DNODE_OBJECT 0 #define DMU_OBJECT_IS_SPECIAL(obj) ((int64_t)(obj) <= 0) #define DMU_META_DNODE(os) ((os)->os_meta_dnode.dnh_dnode) #define DMU_USERUSED_DNODE(os) ((os)->os_userused_dnode.dnh_dnode) #define DMU_GROUPUSED_DNODE(os) ((os)->os_groupused_dnode.dnh_dnode) #define DMU_OS_IS_L2CACHEABLE(os) \ ((os)->os_secondary_cache == ZFS_CACHE_ALL || \ (os)->os_secondary_cache == ZFS_CACHE_METADATA) #define DMU_OS_IS_L2COMPRESSIBLE(os) (zfs_mdcomp_disable == B_FALSE) /* called from zpl */ int dmu_objset_hold(const char *name, void *tag, objset_t **osp); int dmu_objset_own(const char *name, dmu_objset_type_t type, boolean_t readonly, void *tag, objset_t **osp); void dmu_objset_refresh_ownership(objset_t *os, void *tag); void dmu_objset_rele(objset_t *os, void *tag); void dmu_objset_disown(objset_t *os, void *tag); int dmu_objset_from_ds(struct dsl_dataset *ds, objset_t **osp); void dmu_objset_stats(objset_t *os, nvlist_t *nv); void dmu_objset_fast_stat(objset_t *os, dmu_objset_stats_t *stat); void dmu_objset_space(objset_t *os, uint64_t *refdbytesp, uint64_t *availbytesp, uint64_t *usedobjsp, uint64_t *availobjsp); uint64_t dmu_objset_fsid_guid(objset_t *os); int dmu_objset_find_dp(struct dsl_pool *dp, uint64_t ddobj, int func(struct dsl_pool *, struct dsl_dataset *, void *), void *arg, int flags); int dmu_objset_prefetch(const char *name, void *arg); void dmu_objset_evict_dbufs(objset_t *os); timestruc_t dmu_objset_snap_cmtime(objset_t *os); /* called from dsl */ void dmu_objset_sync(objset_t *os, zio_t *zio, dmu_tx_t *tx); boolean_t dmu_objset_is_dirty(objset_t *os, uint64_t txg); objset_t *dmu_objset_create_impl(spa_t *spa, struct dsl_dataset *ds, blkptr_t *bp, dmu_objset_type_t type, dmu_tx_t *tx); int dmu_objset_open_impl(spa_t *spa, struct dsl_dataset *ds, blkptr_t *bp, objset_t **osp); void dmu_objset_evict(objset_t *os); void dmu_objset_do_userquota_updates(objset_t *os, dmu_tx_t *tx); void dmu_objset_userquota_get_ids(dnode_t *dn, boolean_t before, dmu_tx_t *tx); boolean_t dmu_objset_userused_enabled(objset_t *os); int dmu_objset_userspace_upgrade(objset_t *os); boolean_t dmu_objset_userspace_present(objset_t *os); int dmu_fsname(const char *snapname, char *buf); void dmu_objset_init(void); void dmu_objset_fini(void); #ifdef __cplusplus } #endif #endif /* _SYS_DMU_OBJSET_H */ Index: stable/10/sys/cddl/contrib/opensolaris/uts/common/sys/fs/zfs.h =================================================================== --- stable/10/sys/cddl/contrib/opensolaris/uts/common/sys/fs/zfs.h (revision 268646) +++ stable/10/sys/cddl/contrib/opensolaris/uts/common/sys/fs/zfs.h (revision 268647) @@ -1,949 +1,955 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2011, 2014 by Delphix. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. * Copyright (c) 2013, Joyent, Inc. All rights reserved. * Copyright (c) 2012, Martin Matuska . All rights reserved. */ /* Portions Copyright 2010 Robert Milkowski */ #ifndef _SYS_FS_ZFS_H #define _SYS_FS_ZFS_H #include #include #include #ifdef __cplusplus extern "C" { #endif /* * Types and constants shared between userland and the kernel. */ /* * Each dataset can be one of the following types. These constants can be * combined into masks that can be passed to various functions. */ typedef enum { ZFS_TYPE_FILESYSTEM = (1 << 0), ZFS_TYPE_SNAPSHOT = (1 << 1), ZFS_TYPE_VOLUME = (1 << 2), ZFS_TYPE_POOL = (1 << 3), ZFS_TYPE_BOOKMARK = (1 << 4) } zfs_type_t; typedef enum dmu_objset_type { DMU_OST_NONE, DMU_OST_META, DMU_OST_ZFS, DMU_OST_ZVOL, DMU_OST_OTHER, /* For testing only! */ DMU_OST_ANY, /* Be careful! */ DMU_OST_NUMTYPES } dmu_objset_type_t; #define ZFS_TYPE_DATASET \ (ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME | ZFS_TYPE_SNAPSHOT) #define ZAP_MAXNAMELEN 256 #define ZAP_MAXVALUELEN (1024 * 8) #define ZAP_OLDMAXVALUELEN 1024 /* * Dataset properties are identified by these constants and must be added to * the end of this list to ensure that external consumers are not affected * by the change. If you make any changes to this list, be sure to update * the property table in usr/src/common/zfs/zfs_prop.c. */ typedef enum { ZFS_PROP_TYPE, ZFS_PROP_CREATION, ZFS_PROP_USED, ZFS_PROP_AVAILABLE, ZFS_PROP_REFERENCED, ZFS_PROP_COMPRESSRATIO, ZFS_PROP_MOUNTED, ZFS_PROP_ORIGIN, ZFS_PROP_QUOTA, ZFS_PROP_RESERVATION, ZFS_PROP_VOLSIZE, ZFS_PROP_VOLBLOCKSIZE, ZFS_PROP_RECORDSIZE, ZFS_PROP_MOUNTPOINT, ZFS_PROP_SHARENFS, ZFS_PROP_CHECKSUM, ZFS_PROP_COMPRESSION, ZFS_PROP_ATIME, ZFS_PROP_DEVICES, ZFS_PROP_EXEC, ZFS_PROP_SETUID, ZFS_PROP_READONLY, ZFS_PROP_ZONED, ZFS_PROP_SNAPDIR, ZFS_PROP_ACLMODE, ZFS_PROP_ACLINHERIT, ZFS_PROP_CREATETXG, /* not exposed to the user */ ZFS_PROP_NAME, /* not exposed to the user */ ZFS_PROP_CANMOUNT, ZFS_PROP_ISCSIOPTIONS, /* not exposed to the user */ ZFS_PROP_XATTR, ZFS_PROP_NUMCLONES, /* not exposed to the user */ ZFS_PROP_COPIES, ZFS_PROP_VERSION, ZFS_PROP_UTF8ONLY, ZFS_PROP_NORMALIZE, ZFS_PROP_CASE, ZFS_PROP_VSCAN, ZFS_PROP_NBMAND, ZFS_PROP_SHARESMB, ZFS_PROP_REFQUOTA, ZFS_PROP_REFRESERVATION, ZFS_PROP_GUID, ZFS_PROP_PRIMARYCACHE, ZFS_PROP_SECONDARYCACHE, ZFS_PROP_USEDSNAP, ZFS_PROP_USEDDS, ZFS_PROP_USEDCHILD, ZFS_PROP_USEDREFRESERV, ZFS_PROP_USERACCOUNTING, /* not exposed to the user */ ZFS_PROP_STMF_SHAREINFO, /* not exposed to the user */ ZFS_PROP_DEFER_DESTROY, ZFS_PROP_USERREFS, ZFS_PROP_LOGBIAS, ZFS_PROP_UNIQUE, /* not exposed to the user */ ZFS_PROP_OBJSETID, /* not exposed to the user */ ZFS_PROP_DEDUP, ZFS_PROP_MLSLABEL, ZFS_PROP_SYNC, ZFS_PROP_REFRATIO, ZFS_PROP_WRITTEN, ZFS_PROP_CLONES, ZFS_PROP_LOGICALUSED, ZFS_PROP_LOGICALREFERENCED, ZFS_PROP_INCONSISTENT, /* not exposed to the user */ ZFS_PROP_VOLMODE, ZFS_PROP_FILESYSTEM_LIMIT, ZFS_PROP_SNAPSHOT_LIMIT, ZFS_PROP_FILESYSTEM_COUNT, ZFS_PROP_SNAPSHOT_COUNT, + ZFS_PROP_REDUNDANT_METADATA, ZFS_NUM_PROPS } zfs_prop_t; typedef enum { ZFS_PROP_USERUSED, ZFS_PROP_USERQUOTA, ZFS_PROP_GROUPUSED, ZFS_PROP_GROUPQUOTA, ZFS_NUM_USERQUOTA_PROPS } zfs_userquota_prop_t; extern const char *zfs_userquota_prop_prefixes[ZFS_NUM_USERQUOTA_PROPS]; /* * Pool properties are identified by these constants and must be added to the * end of this list to ensure that external consumers are not affected * by the change. If you make any changes to this list, be sure to update * the property table in usr/src/common/zfs/zpool_prop.c. */ typedef enum { ZPOOL_PROP_NAME, ZPOOL_PROP_SIZE, ZPOOL_PROP_CAPACITY, ZPOOL_PROP_ALTROOT, ZPOOL_PROP_HEALTH, ZPOOL_PROP_GUID, ZPOOL_PROP_VERSION, ZPOOL_PROP_BOOTFS, ZPOOL_PROP_DELEGATION, ZPOOL_PROP_AUTOREPLACE, ZPOOL_PROP_CACHEFILE, ZPOOL_PROP_FAILUREMODE, ZPOOL_PROP_LISTSNAPS, ZPOOL_PROP_AUTOEXPAND, ZPOOL_PROP_DEDUPDITTO, ZPOOL_PROP_DEDUPRATIO, ZPOOL_PROP_FREE, ZPOOL_PROP_ALLOCATED, ZPOOL_PROP_READONLY, ZPOOL_PROP_COMMENT, ZPOOL_PROP_EXPANDSZ, ZPOOL_PROP_FREEING, ZPOOL_NUM_PROPS } zpool_prop_t; /* Small enough to not hog a whole line of printout in zpool(1M). */ #define ZPROP_MAX_COMMENT 32 #define ZPROP_CONT -2 #define ZPROP_INVAL -1 #define ZPROP_VALUE "value" #define ZPROP_SOURCE "source" typedef enum { ZPROP_SRC_NONE = 0x1, ZPROP_SRC_DEFAULT = 0x2, ZPROP_SRC_TEMPORARY = 0x4, ZPROP_SRC_LOCAL = 0x8, ZPROP_SRC_INHERITED = 0x10, ZPROP_SRC_RECEIVED = 0x20 } zprop_source_t; #define ZPROP_SRC_ALL 0x3f #define ZPROP_SOURCE_VAL_RECVD "$recvd" #define ZPROP_N_MORE_ERRORS "N_MORE_ERRORS" /* * Dataset flag implemented as a special entry in the props zap object * indicating that the dataset has received properties on or after * SPA_VERSION_RECVD_PROPS. The first such receive blows away local properties * just as it did in earlier versions, and thereafter, local properties are * preserved. */ #define ZPROP_HAS_RECVD "$hasrecvd" typedef enum { ZPROP_ERR_NOCLEAR = 0x1, /* failure to clear existing props */ ZPROP_ERR_NORESTORE = 0x2 /* failure to restore props on error */ } zprop_errflags_t; typedef int (*zprop_func)(int, void *); /* * Properties to be set on the root file system of a new pool * are stuffed into their own nvlist, which is then included in * the properties nvlist with the pool properties. */ #define ZPOOL_ROOTFS_PROPS "root-props-nvl" /* * Dataset property functions shared between libzfs and kernel. */ const char *zfs_prop_default_string(zfs_prop_t); uint64_t zfs_prop_default_numeric(zfs_prop_t); boolean_t zfs_prop_readonly(zfs_prop_t); boolean_t zfs_prop_inheritable(zfs_prop_t); boolean_t zfs_prop_setonce(zfs_prop_t); const char *zfs_prop_to_name(zfs_prop_t); zfs_prop_t zfs_name_to_prop(const char *); boolean_t zfs_prop_user(const char *); boolean_t zfs_prop_userquota(const char *); int zfs_prop_index_to_string(zfs_prop_t, uint64_t, const char **); int zfs_prop_string_to_index(zfs_prop_t, const char *, uint64_t *); uint64_t zfs_prop_random_value(zfs_prop_t, uint64_t seed); boolean_t zfs_prop_valid_for_type(int, zfs_type_t); /* * Pool property functions shared between libzfs and kernel. */ zpool_prop_t zpool_name_to_prop(const char *); const char *zpool_prop_to_name(zpool_prop_t); const char *zpool_prop_default_string(zpool_prop_t); uint64_t zpool_prop_default_numeric(zpool_prop_t); boolean_t zpool_prop_readonly(zpool_prop_t); boolean_t zpool_prop_feature(const char *); boolean_t zpool_prop_unsupported(const char *name); int zpool_prop_index_to_string(zpool_prop_t, uint64_t, const char **); int zpool_prop_string_to_index(zpool_prop_t, const char *, uint64_t *); uint64_t zpool_prop_random_value(zpool_prop_t, uint64_t seed); /* * Definitions for the Delegation. */ typedef enum { ZFS_DELEG_WHO_UNKNOWN = 0, ZFS_DELEG_USER = 'u', ZFS_DELEG_USER_SETS = 'U', ZFS_DELEG_GROUP = 'g', ZFS_DELEG_GROUP_SETS = 'G', ZFS_DELEG_EVERYONE = 'e', ZFS_DELEG_EVERYONE_SETS = 'E', ZFS_DELEG_CREATE = 'c', ZFS_DELEG_CREATE_SETS = 'C', ZFS_DELEG_NAMED_SET = 's', ZFS_DELEG_NAMED_SET_SETS = 'S' } zfs_deleg_who_type_t; typedef enum { ZFS_DELEG_NONE = 0, ZFS_DELEG_PERM_LOCAL = 1, ZFS_DELEG_PERM_DESCENDENT = 2, ZFS_DELEG_PERM_LOCALDESCENDENT = 3, ZFS_DELEG_PERM_CREATE = 4 } zfs_deleg_inherit_t; #define ZFS_DELEG_PERM_UID "uid" #define ZFS_DELEG_PERM_GID "gid" #define ZFS_DELEG_PERM_GROUPS "groups" #define ZFS_MLSLABEL_DEFAULT "none" #define ZFS_SMB_ACL_SRC "src" #define ZFS_SMB_ACL_TARGET "target" typedef enum { ZFS_CANMOUNT_OFF = 0, ZFS_CANMOUNT_ON = 1, ZFS_CANMOUNT_NOAUTO = 2 } zfs_canmount_type_t; typedef enum { ZFS_LOGBIAS_LATENCY = 0, ZFS_LOGBIAS_THROUGHPUT = 1 } zfs_logbias_op_t; typedef enum zfs_share_op { ZFS_SHARE_NFS = 0, ZFS_UNSHARE_NFS = 1, ZFS_SHARE_SMB = 2, ZFS_UNSHARE_SMB = 3 } zfs_share_op_t; typedef enum zfs_smb_acl_op { ZFS_SMB_ACL_ADD, ZFS_SMB_ACL_REMOVE, ZFS_SMB_ACL_RENAME, ZFS_SMB_ACL_PURGE } zfs_smb_acl_op_t; typedef enum zfs_cache_type { ZFS_CACHE_NONE = 0, ZFS_CACHE_METADATA = 1, ZFS_CACHE_ALL = 2 } zfs_cache_type_t; typedef enum { ZFS_SYNC_STANDARD = 0, ZFS_SYNC_ALWAYS = 1, ZFS_SYNC_DISABLED = 2 } zfs_sync_type_t; typedef enum { ZFS_VOLMODE_DEFAULT = 0, ZFS_VOLMODE_GEOM = 1, ZFS_VOLMODE_DEV = 2, ZFS_VOLMODE_NONE = 3 } zfs_volmode_t; + +typedef enum { + ZFS_REDUNDANT_METADATA_ALL, + ZFS_REDUNDANT_METADATA_MOST +} zfs_redundant_metadata_type_t; /* * On-disk version number. */ #define SPA_VERSION_1 1ULL #define SPA_VERSION_2 2ULL #define SPA_VERSION_3 3ULL #define SPA_VERSION_4 4ULL #define SPA_VERSION_5 5ULL #define SPA_VERSION_6 6ULL #define SPA_VERSION_7 7ULL #define SPA_VERSION_8 8ULL #define SPA_VERSION_9 9ULL #define SPA_VERSION_10 10ULL #define SPA_VERSION_11 11ULL #define SPA_VERSION_12 12ULL #define SPA_VERSION_13 13ULL #define SPA_VERSION_14 14ULL #define SPA_VERSION_15 15ULL #define SPA_VERSION_16 16ULL #define SPA_VERSION_17 17ULL #define SPA_VERSION_18 18ULL #define SPA_VERSION_19 19ULL #define SPA_VERSION_20 20ULL #define SPA_VERSION_21 21ULL #define SPA_VERSION_22 22ULL #define SPA_VERSION_23 23ULL #define SPA_VERSION_24 24ULL #define SPA_VERSION_25 25ULL #define SPA_VERSION_26 26ULL #define SPA_VERSION_27 27ULL #define SPA_VERSION_28 28ULL #define SPA_VERSION_5000 5000ULL /* * When bumping up SPA_VERSION, make sure GRUB ZFS understands the on-disk * format change. Go to usr/src/grub/grub-0.97/stage2/{zfs-include/, fsys_zfs*}, * and do the appropriate changes. Also bump the version number in * usr/src/grub/capability. */ #define SPA_VERSION SPA_VERSION_5000 #define SPA_VERSION_STRING "5000" /* * Symbolic names for the changes that caused a SPA_VERSION switch. * Used in the code when checking for presence or absence of a feature. * Feel free to define multiple symbolic names for each version if there * were multiple changes to on-disk structures during that version. * * NOTE: When checking the current SPA_VERSION in your code, be sure * to use spa_version() since it reports the version of the * last synced uberblock. Checking the in-flight version can * be dangerous in some cases. */ #define SPA_VERSION_INITIAL SPA_VERSION_1 #define SPA_VERSION_DITTO_BLOCKS SPA_VERSION_2 #define SPA_VERSION_SPARES SPA_VERSION_3 #define SPA_VERSION_RAIDZ2 SPA_VERSION_3 #define SPA_VERSION_BPOBJ_ACCOUNT SPA_VERSION_3 #define SPA_VERSION_RAIDZ_DEFLATE SPA_VERSION_3 #define SPA_VERSION_DNODE_BYTES SPA_VERSION_3 #define SPA_VERSION_ZPOOL_HISTORY SPA_VERSION_4 #define SPA_VERSION_GZIP_COMPRESSION SPA_VERSION_5 #define SPA_VERSION_BOOTFS SPA_VERSION_6 #define SPA_VERSION_SLOGS SPA_VERSION_7 #define SPA_VERSION_DELEGATED_PERMS SPA_VERSION_8 #define SPA_VERSION_FUID SPA_VERSION_9 #define SPA_VERSION_REFRESERVATION SPA_VERSION_9 #define SPA_VERSION_REFQUOTA SPA_VERSION_9 #define SPA_VERSION_UNIQUE_ACCURATE SPA_VERSION_9 #define SPA_VERSION_L2CACHE SPA_VERSION_10 #define SPA_VERSION_NEXT_CLONES SPA_VERSION_11 #define SPA_VERSION_ORIGIN SPA_VERSION_11 #define SPA_VERSION_DSL_SCRUB SPA_VERSION_11 #define SPA_VERSION_SNAP_PROPS SPA_VERSION_12 #define SPA_VERSION_USED_BREAKDOWN SPA_VERSION_13 #define SPA_VERSION_PASSTHROUGH_X SPA_VERSION_14 #define SPA_VERSION_USERSPACE SPA_VERSION_15 #define SPA_VERSION_STMF_PROP SPA_VERSION_16 #define SPA_VERSION_RAIDZ3 SPA_VERSION_17 #define SPA_VERSION_USERREFS SPA_VERSION_18 #define SPA_VERSION_HOLES SPA_VERSION_19 #define SPA_VERSION_ZLE_COMPRESSION SPA_VERSION_20 #define SPA_VERSION_DEDUP SPA_VERSION_21 #define SPA_VERSION_RECVD_PROPS SPA_VERSION_22 #define SPA_VERSION_SLIM_ZIL SPA_VERSION_23 #define SPA_VERSION_SA SPA_VERSION_24 #define SPA_VERSION_SCAN SPA_VERSION_25 #define SPA_VERSION_DIR_CLONES SPA_VERSION_26 #define SPA_VERSION_DEADLISTS SPA_VERSION_26 #define SPA_VERSION_FAST_SNAP SPA_VERSION_27 #define SPA_VERSION_MULTI_REPLACE SPA_VERSION_28 #define SPA_VERSION_BEFORE_FEATURES SPA_VERSION_28 #define SPA_VERSION_FEATURES SPA_VERSION_5000 #define SPA_VERSION_IS_SUPPORTED(v) \ (((v) >= SPA_VERSION_INITIAL && (v) <= SPA_VERSION_BEFORE_FEATURES) || \ ((v) >= SPA_VERSION_FEATURES && (v) <= SPA_VERSION)) /* * ZPL version - rev'd whenever an incompatible on-disk format change * occurs. This is independent of SPA/DMU/ZAP versioning. You must * also update the version_table[] and help message in zfs_prop.c. * * When changing, be sure to teach GRUB how to read the new format! * See usr/src/grub/grub-0.97/stage2/{zfs-include/,fsys_zfs*} */ #define ZPL_VERSION_1 1ULL #define ZPL_VERSION_2 2ULL #define ZPL_VERSION_3 3ULL #define ZPL_VERSION_4 4ULL #define ZPL_VERSION_5 5ULL #define ZPL_VERSION ZPL_VERSION_5 #define ZPL_VERSION_STRING "5" #define ZPL_VERSION_INITIAL ZPL_VERSION_1 #define ZPL_VERSION_DIRENT_TYPE ZPL_VERSION_2 #define ZPL_VERSION_FUID ZPL_VERSION_3 #define ZPL_VERSION_NORMALIZATION ZPL_VERSION_3 #define ZPL_VERSION_SYSATTR ZPL_VERSION_3 #define ZPL_VERSION_USERSPACE ZPL_VERSION_4 #define ZPL_VERSION_SA ZPL_VERSION_5 /* Rewind request information */ #define ZPOOL_NO_REWIND 1 /* No policy - default behavior */ #define ZPOOL_NEVER_REWIND 2 /* Do not search for best txg or rewind */ #define ZPOOL_TRY_REWIND 4 /* Search for best txg, but do not rewind */ #define ZPOOL_DO_REWIND 8 /* Rewind to best txg w/in deferred frees */ #define ZPOOL_EXTREME_REWIND 16 /* Allow extreme measures to find best txg */ #define ZPOOL_REWIND_MASK 28 /* All the possible rewind bits */ #define ZPOOL_REWIND_POLICIES 31 /* All the possible policy bits */ typedef struct zpool_rewind_policy { uint32_t zrp_request; /* rewind behavior requested */ uint64_t zrp_maxmeta; /* max acceptable meta-data errors */ uint64_t zrp_maxdata; /* max acceptable data errors */ uint64_t zrp_txg; /* specific txg to load */ } zpool_rewind_policy_t; /* * The following are configuration names used in the nvlist describing a pool's * configuration. */ #define ZPOOL_CONFIG_VERSION "version" #define ZPOOL_CONFIG_POOL_NAME "name" #define ZPOOL_CONFIG_POOL_STATE "state" #define ZPOOL_CONFIG_POOL_TXG "txg" #define ZPOOL_CONFIG_POOL_GUID "pool_guid" #define ZPOOL_CONFIG_CREATE_TXG "create_txg" #define ZPOOL_CONFIG_TOP_GUID "top_guid" #define ZPOOL_CONFIG_VDEV_TREE "vdev_tree" #define ZPOOL_CONFIG_TYPE "type" #define ZPOOL_CONFIG_CHILDREN "children" #define ZPOOL_CONFIG_ID "id" #define ZPOOL_CONFIG_GUID "guid" #define ZPOOL_CONFIG_PATH "path" #define ZPOOL_CONFIG_DEVID "devid" #define ZPOOL_CONFIG_METASLAB_ARRAY "metaslab_array" #define ZPOOL_CONFIG_METASLAB_SHIFT "metaslab_shift" #define ZPOOL_CONFIG_ASHIFT "ashift" #define ZPOOL_CONFIG_ASIZE "asize" #define ZPOOL_CONFIG_DTL "DTL" #define ZPOOL_CONFIG_SCAN_STATS "scan_stats" /* not stored on disk */ #define ZPOOL_CONFIG_VDEV_STATS "vdev_stats" /* not stored on disk */ #define ZPOOL_CONFIG_WHOLE_DISK "whole_disk" #define ZPOOL_CONFIG_ERRCOUNT "error_count" #define ZPOOL_CONFIG_NOT_PRESENT "not_present" #define ZPOOL_CONFIG_SPARES "spares" #define ZPOOL_CONFIG_IS_SPARE "is_spare" #define ZPOOL_CONFIG_NPARITY "nparity" #define ZPOOL_CONFIG_HOSTID "hostid" #define ZPOOL_CONFIG_HOSTNAME "hostname" #define ZPOOL_CONFIG_LOADED_TIME "initial_load_time" #define ZPOOL_CONFIG_UNSPARE "unspare" #define ZPOOL_CONFIG_PHYS_PATH "phys_path" #define ZPOOL_CONFIG_IS_LOG "is_log" #define ZPOOL_CONFIG_L2CACHE "l2cache" #define ZPOOL_CONFIG_HOLE_ARRAY "hole_array" #define ZPOOL_CONFIG_VDEV_CHILDREN "vdev_children" #define ZPOOL_CONFIG_IS_HOLE "is_hole" #define ZPOOL_CONFIG_DDT_HISTOGRAM "ddt_histogram" #define ZPOOL_CONFIG_DDT_OBJ_STATS "ddt_object_stats" #define ZPOOL_CONFIG_DDT_STATS "ddt_stats" #define ZPOOL_CONFIG_SPLIT "splitcfg" #define ZPOOL_CONFIG_ORIG_GUID "orig_guid" #define ZPOOL_CONFIG_SPLIT_GUID "split_guid" #define ZPOOL_CONFIG_SPLIT_LIST "guid_list" #define ZPOOL_CONFIG_REMOVING "removing" #define ZPOOL_CONFIG_RESILVER_TXG "resilver_txg" #define ZPOOL_CONFIG_COMMENT "comment" #define ZPOOL_CONFIG_SUSPENDED "suspended" /* not stored on disk */ #define ZPOOL_CONFIG_TIMESTAMP "timestamp" /* not stored on disk */ #define ZPOOL_CONFIG_BOOTFS "bootfs" /* not stored on disk */ #define ZPOOL_CONFIG_MISSING_DEVICES "missing_vdevs" /* not stored on disk */ #define ZPOOL_CONFIG_LOAD_INFO "load_info" /* not stored on disk */ #define ZPOOL_CONFIG_REWIND_INFO "rewind_info" /* not stored on disk */ #define ZPOOL_CONFIG_UNSUP_FEAT "unsup_feat" /* not stored on disk */ #define ZPOOL_CONFIG_ENABLED_FEAT "enabled_feat" /* not stored on disk */ #define ZPOOL_CONFIG_CAN_RDONLY "can_rdonly" /* not stored on disk */ #define ZPOOL_CONFIG_FEATURES_FOR_READ "features_for_read" #define ZPOOL_CONFIG_FEATURE_STATS "feature_stats" /* not stored on disk */ /* * The persistent vdev state is stored as separate values rather than a single * 'vdev_state' entry. This is because a device can be in multiple states, such * as offline and degraded. */ #define ZPOOL_CONFIG_OFFLINE "offline" #define ZPOOL_CONFIG_FAULTED "faulted" #define ZPOOL_CONFIG_DEGRADED "degraded" #define ZPOOL_CONFIG_REMOVED "removed" #define ZPOOL_CONFIG_FRU "fru" #define ZPOOL_CONFIG_AUX_STATE "aux_state" /* Rewind policy parameters */ #define ZPOOL_REWIND_POLICY "rewind-policy" #define ZPOOL_REWIND_REQUEST "rewind-request" #define ZPOOL_REWIND_REQUEST_TXG "rewind-request-txg" #define ZPOOL_REWIND_META_THRESH "rewind-meta-thresh" #define ZPOOL_REWIND_DATA_THRESH "rewind-data-thresh" /* Rewind data discovered */ #define ZPOOL_CONFIG_LOAD_TIME "rewind_txg_ts" #define ZPOOL_CONFIG_LOAD_DATA_ERRORS "verify_data_errors" #define ZPOOL_CONFIG_REWIND_TIME "seconds_of_rewind" #define VDEV_TYPE_ROOT "root" #define VDEV_TYPE_MIRROR "mirror" #define VDEV_TYPE_REPLACING "replacing" #define VDEV_TYPE_RAIDZ "raidz" #define VDEV_TYPE_DISK "disk" #define VDEV_TYPE_FILE "file" #define VDEV_TYPE_MISSING "missing" #define VDEV_TYPE_HOLE "hole" #define VDEV_TYPE_SPARE "spare" #define VDEV_TYPE_LOG "log" #define VDEV_TYPE_L2CACHE "l2cache" /* * This is needed in userland to report the minimum necessary device size. */ #define SPA_MINDEVSIZE (64ULL << 20) /* * The location of the pool configuration repository, shared between kernel and * userland. */ #define ZPOOL_CACHE "/boot/zfs/zpool.cache" /* * vdev states are ordered from least to most healthy. * A vdev that's CANT_OPEN or below is considered unusable. */ typedef enum vdev_state { VDEV_STATE_UNKNOWN = 0, /* Uninitialized vdev */ VDEV_STATE_CLOSED, /* Not currently open */ VDEV_STATE_OFFLINE, /* Not allowed to open */ VDEV_STATE_REMOVED, /* Explicitly removed from system */ VDEV_STATE_CANT_OPEN, /* Tried to open, but failed */ VDEV_STATE_FAULTED, /* External request to fault device */ VDEV_STATE_DEGRADED, /* Replicated vdev with unhealthy kids */ VDEV_STATE_HEALTHY /* Presumed good */ } vdev_state_t; #define VDEV_STATE_ONLINE VDEV_STATE_HEALTHY /* * vdev aux states. When a vdev is in the CANT_OPEN state, the aux field * of the vdev stats structure uses these constants to distinguish why. */ typedef enum vdev_aux { VDEV_AUX_NONE, /* no error */ VDEV_AUX_OPEN_FAILED, /* ldi_open_*() or vn_open() failed */ VDEV_AUX_CORRUPT_DATA, /* bad label or disk contents */ VDEV_AUX_NO_REPLICAS, /* insufficient number of replicas */ VDEV_AUX_BAD_GUID_SUM, /* vdev guid sum doesn't match */ VDEV_AUX_TOO_SMALL, /* vdev size is too small */ VDEV_AUX_BAD_LABEL, /* the label is OK but invalid */ VDEV_AUX_VERSION_NEWER, /* on-disk version is too new */ VDEV_AUX_VERSION_OLDER, /* on-disk version is too old */ VDEV_AUX_UNSUP_FEAT, /* unsupported features */ VDEV_AUX_SPARED, /* hot spare used in another pool */ VDEV_AUX_ERR_EXCEEDED, /* too many errors */ VDEV_AUX_IO_FAILURE, /* experienced I/O failure */ VDEV_AUX_BAD_LOG, /* cannot read log chain(s) */ VDEV_AUX_EXTERNAL, /* external diagnosis */ VDEV_AUX_SPLIT_POOL, /* vdev was split off into another pool */ VDEV_AUX_ASHIFT_TOO_BIG /* vdev's min block size is too large */ } vdev_aux_t; /* * pool state. The following states are written to disk as part of the normal * SPA lifecycle: ACTIVE, EXPORTED, DESTROYED, SPARE, L2CACHE. The remaining * states are software abstractions used at various levels to communicate * pool state. */ typedef enum pool_state { POOL_STATE_ACTIVE = 0, /* In active use */ POOL_STATE_EXPORTED, /* Explicitly exported */ POOL_STATE_DESTROYED, /* Explicitly destroyed */ POOL_STATE_SPARE, /* Reserved for hot spare use */ POOL_STATE_L2CACHE, /* Level 2 ARC device */ POOL_STATE_UNINITIALIZED, /* Internal spa_t state */ POOL_STATE_UNAVAIL, /* Internal libzfs state */ POOL_STATE_POTENTIALLY_ACTIVE /* Internal libzfs state */ } pool_state_t; /* * Scan Functions. */ typedef enum pool_scan_func { POOL_SCAN_NONE, POOL_SCAN_SCRUB, POOL_SCAN_RESILVER, POOL_SCAN_FUNCS } pool_scan_func_t; /* * ZIO types. Needed to interpret vdev statistics below. */ typedef enum zio_type { ZIO_TYPE_NULL = 0, ZIO_TYPE_READ, ZIO_TYPE_WRITE, ZIO_TYPE_FREE, ZIO_TYPE_CLAIM, ZIO_TYPE_IOCTL, ZIO_TYPES } zio_type_t; /* * Pool statistics. Note: all fields should be 64-bit because this * is passed between kernel and userland as an nvlist uint64 array. */ typedef struct pool_scan_stat { /* values stored on disk */ uint64_t pss_func; /* pool_scan_func_t */ uint64_t pss_state; /* dsl_scan_state_t */ uint64_t pss_start_time; /* scan start time */ uint64_t pss_end_time; /* scan end time */ uint64_t pss_to_examine; /* total bytes to scan */ uint64_t pss_examined; /* total examined bytes */ uint64_t pss_to_process; /* total bytes to process */ uint64_t pss_processed; /* total processed bytes */ uint64_t pss_errors; /* scan errors */ /* values not stored on disk */ uint64_t pss_pass_exam; /* examined bytes per scan pass */ uint64_t pss_pass_start; /* start time of a scan pass */ } pool_scan_stat_t; typedef enum dsl_scan_state { DSS_NONE, DSS_SCANNING, DSS_FINISHED, DSS_CANCELED, DSS_NUM_STATES } dsl_scan_state_t; /* * Vdev statistics. Note: all fields should be 64-bit because this * is passed between kernel and userland as an nvlist uint64 array. */ typedef struct vdev_stat { hrtime_t vs_timestamp; /* time since vdev load */ uint64_t vs_state; /* vdev state */ uint64_t vs_aux; /* see vdev_aux_t */ uint64_t vs_alloc; /* space allocated */ uint64_t vs_space; /* total capacity */ uint64_t vs_dspace; /* deflated capacity */ uint64_t vs_rsize; /* replaceable dev size */ uint64_t vs_esize; /* expandable dev size */ uint64_t vs_ops[ZIO_TYPES]; /* operation count */ uint64_t vs_bytes[ZIO_TYPES]; /* bytes read/written */ uint64_t vs_read_errors; /* read errors */ uint64_t vs_write_errors; /* write errors */ uint64_t vs_checksum_errors; /* checksum errors */ uint64_t vs_self_healed; /* self-healed bytes */ uint64_t vs_scan_removing; /* removing? */ uint64_t vs_scan_processed; /* scan processed bytes */ uint64_t vs_configured_ashift; /* TLV vdev_ashift */ uint64_t vs_logical_ashift; /* vdev_logical_ashift */ uint64_t vs_physical_ashift; /* vdev_physical_ashift */ } vdev_stat_t; #define VDEV_STAT_VALID(field, uint64_t_field_count) \ ((uint64_t_field_count * sizeof(uint64_t)) >= \ (offsetof(vdev_stat_t, field) + sizeof(((vdev_stat_t *)NULL)->field))) /* * DDT statistics. Note: all fields should be 64-bit because this * is passed between kernel and userland as an nvlist uint64 array. */ typedef struct ddt_object { uint64_t ddo_count; /* number of elments in ddt */ uint64_t ddo_dspace; /* size of ddt on disk */ uint64_t ddo_mspace; /* size of ddt in-core */ } ddt_object_t; typedef struct ddt_stat { uint64_t dds_blocks; /* blocks */ uint64_t dds_lsize; /* logical size */ uint64_t dds_psize; /* physical size */ uint64_t dds_dsize; /* deflated allocated size */ uint64_t dds_ref_blocks; /* referenced blocks */ uint64_t dds_ref_lsize; /* referenced lsize * refcnt */ uint64_t dds_ref_psize; /* referenced psize * refcnt */ uint64_t dds_ref_dsize; /* referenced dsize * refcnt */ } ddt_stat_t; typedef struct ddt_histogram { ddt_stat_t ddh_stat[64]; /* power-of-two histogram buckets */ } ddt_histogram_t; #define ZVOL_DRIVER "zvol" #define ZFS_DRIVER "zfs" #define ZFS_DEV_NAME "zfs" #define ZFS_DEV "/dev/" ZFS_DEV_NAME /* general zvol path */ #define ZVOL_DIR "/dev/zvol" /* expansion */ #define ZVOL_PSEUDO_DEV "/devices/pseudo/zfs@0:" /* for dump and swap */ #define ZVOL_FULL_DEV_DIR ZVOL_DIR "/dsk/" #define ZVOL_FULL_RDEV_DIR ZVOL_DIR "/rdsk/" #define ZVOL_PROP_NAME "name" #define ZVOL_DEFAULT_BLOCKSIZE 8192 /* * /dev/zfs ioctl numbers. */ typedef enum zfs_ioc { ZFS_IOC_FIRST = 0, ZFS_IOC_POOL_CREATE = ZFS_IOC_FIRST, ZFS_IOC_POOL_DESTROY, ZFS_IOC_POOL_IMPORT, ZFS_IOC_POOL_EXPORT, ZFS_IOC_POOL_CONFIGS, ZFS_IOC_POOL_STATS, ZFS_IOC_POOL_TRYIMPORT, ZFS_IOC_POOL_SCAN, ZFS_IOC_POOL_FREEZE, ZFS_IOC_POOL_UPGRADE, ZFS_IOC_POOL_GET_HISTORY, ZFS_IOC_VDEV_ADD, ZFS_IOC_VDEV_REMOVE, ZFS_IOC_VDEV_SET_STATE, ZFS_IOC_VDEV_ATTACH, ZFS_IOC_VDEV_DETACH, ZFS_IOC_VDEV_SETPATH, ZFS_IOC_VDEV_SETFRU, ZFS_IOC_OBJSET_STATS, ZFS_IOC_OBJSET_ZPLPROPS, ZFS_IOC_DATASET_LIST_NEXT, ZFS_IOC_SNAPSHOT_LIST_NEXT, ZFS_IOC_SET_PROP, ZFS_IOC_CREATE, ZFS_IOC_DESTROY, ZFS_IOC_ROLLBACK, ZFS_IOC_RENAME, ZFS_IOC_RECV, ZFS_IOC_SEND, ZFS_IOC_INJECT_FAULT, ZFS_IOC_CLEAR_FAULT, ZFS_IOC_INJECT_LIST_NEXT, ZFS_IOC_ERROR_LOG, ZFS_IOC_CLEAR, ZFS_IOC_PROMOTE, ZFS_IOC_DESTROY_SNAPS, ZFS_IOC_SNAPSHOT, ZFS_IOC_DSOBJ_TO_DSNAME, ZFS_IOC_OBJ_TO_PATH, ZFS_IOC_POOL_SET_PROPS, ZFS_IOC_POOL_GET_PROPS, ZFS_IOC_SET_FSACL, ZFS_IOC_GET_FSACL, ZFS_IOC_SHARE, ZFS_IOC_INHERIT_PROP, ZFS_IOC_SMB_ACL, ZFS_IOC_USERSPACE_ONE, ZFS_IOC_USERSPACE_MANY, ZFS_IOC_USERSPACE_UPGRADE, ZFS_IOC_HOLD, ZFS_IOC_RELEASE, ZFS_IOC_GET_HOLDS, ZFS_IOC_OBJSET_RECVD_PROPS, ZFS_IOC_VDEV_SPLIT, ZFS_IOC_NEXT_OBJ, ZFS_IOC_DIFF, ZFS_IOC_TMP_SNAPSHOT, ZFS_IOC_OBJ_TO_STATS, ZFS_IOC_JAIL, ZFS_IOC_UNJAIL, ZFS_IOC_POOL_REGUID, ZFS_IOC_SPACE_WRITTEN, ZFS_IOC_SPACE_SNAPS, ZFS_IOC_SEND_PROGRESS, ZFS_IOC_POOL_REOPEN, ZFS_IOC_LOG_HISTORY, ZFS_IOC_SEND_NEW, ZFS_IOC_SEND_SPACE, ZFS_IOC_CLONE, ZFS_IOC_BOOKMARK, ZFS_IOC_GET_BOOKMARKS, ZFS_IOC_DESTROY_BOOKMARKS, ZFS_IOC_LAST } zfs_ioc_t; /* * Internal SPA load state. Used by FMA diagnosis engine. */ typedef enum { SPA_LOAD_NONE, /* no load in progress */ SPA_LOAD_OPEN, /* normal open */ SPA_LOAD_IMPORT, /* import in progress */ SPA_LOAD_TRYIMPORT, /* tryimport in progress */ SPA_LOAD_RECOVER, /* recovery requested */ SPA_LOAD_ERROR /* load failed */ } spa_load_state_t; /* * Bookmark name values. */ #define ZPOOL_ERR_LIST "error list" #define ZPOOL_ERR_DATASET "dataset" #define ZPOOL_ERR_OBJECT "object" #define HIS_MAX_RECORD_LEN (MAXPATHLEN + MAXPATHLEN + 1) /* * The following are names used in the nvlist describing * the pool's history log. */ #define ZPOOL_HIST_RECORD "history record" #define ZPOOL_HIST_TIME "history time" #define ZPOOL_HIST_CMD "history command" #define ZPOOL_HIST_WHO "history who" #define ZPOOL_HIST_ZONE "history zone" #define ZPOOL_HIST_HOST "history hostname" #define ZPOOL_HIST_TXG "history txg" #define ZPOOL_HIST_INT_EVENT "history internal event" #define ZPOOL_HIST_INT_STR "history internal str" #define ZPOOL_HIST_INT_NAME "internal_name" #define ZPOOL_HIST_IOCTL "ioctl" #define ZPOOL_HIST_INPUT_NVL "in_nvl" #define ZPOOL_HIST_OUTPUT_NVL "out_nvl" #define ZPOOL_HIST_DSNAME "dsname" #define ZPOOL_HIST_DSID "dsid" /* * Flags for ZFS_IOC_VDEV_SET_STATE */ #define ZFS_ONLINE_CHECKREMOVE 0x1 #define ZFS_ONLINE_UNSPARE 0x2 #define ZFS_ONLINE_FORCEFAULT 0x4 #define ZFS_ONLINE_EXPAND 0x8 #define ZFS_OFFLINE_TEMPORARY 0x1 /* * Flags for ZFS_IOC_POOL_IMPORT */ #define ZFS_IMPORT_NORMAL 0x0 #define ZFS_IMPORT_VERBATIM 0x1 #define ZFS_IMPORT_ANY_HOST 0x2 #define ZFS_IMPORT_MISSING_LOG 0x4 #define ZFS_IMPORT_ONLY 0x8 /* * Sysevent payload members. ZFS will generate the following sysevents with the * given payloads: * * ESC_ZFS_RESILVER_START * ESC_ZFS_RESILVER_END * ESC_ZFS_POOL_DESTROY * ESC_ZFS_POOL_REGUID * * ZFS_EV_POOL_NAME DATA_TYPE_STRING * ZFS_EV_POOL_GUID DATA_TYPE_UINT64 * * ESC_ZFS_VDEV_REMOVE * ESC_ZFS_VDEV_CLEAR * ESC_ZFS_VDEV_CHECK * * ZFS_EV_POOL_NAME DATA_TYPE_STRING * ZFS_EV_POOL_GUID DATA_TYPE_UINT64 * ZFS_EV_VDEV_PATH DATA_TYPE_STRING (optional) * ZFS_EV_VDEV_GUID DATA_TYPE_UINT64 */ #define ZFS_EV_POOL_NAME "pool_name" #define ZFS_EV_POOL_GUID "pool_guid" #define ZFS_EV_VDEV_PATH "vdev_path" #define ZFS_EV_VDEV_GUID "vdev_guid" #ifdef __cplusplus } #endif #endif /* _SYS_FS_ZFS_H */ Index: stable/10 =================================================================== --- stable/10 (revision 268646) +++ stable/10 (revision 268647) Property changes on: stable/10 ___________________________________________________________________ Modified: svn:mergeinfo ## -0,0 +0,1 ## Merged /head:r266771