Archive-Team
/
zfs
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 103 Wiki Activity
zfs/module/zfs/zfs_znode.c

2179 lines
55 KiB
C
Raw Normal View History

Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* 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
*/
/*
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
* Copyright (c) 2012, 2014 by Delphix. All rights reserved.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
/* Portions Copyright 2007 Jeremy Teo */
#ifdef _KERNEL
#include <sys/types.h>
#include <sys/param.h>
#include <sys/time.h>
#include <sys/systm.h>
#include <sys/sysmacros.h>
#include <sys/resource.h>
#include <sys/mntent.h>
#include <sys/mkdev.h>
#include <sys/u8_textprep.h>
#include <sys/dsl_dataset.h>
#include <sys/vfs.h>
#include <sys/vfs_opreg.h>
#include <sys/vnode.h>
#include <sys/file.h>
#include <sys/kmem.h>
#include <sys/errno.h>
#include <sys/unistd.h>
#include <sys/mode.h>
#include <sys/atomic.h>
#include <vm/pvn.h>
#include "fs/fs_subr.h"
#include <sys/zfs_dir.h>
#include <sys/zfs_acl.h>
#include <sys/zfs_ioctl.h>
#include <sys/zfs_rlock.h>
#include <sys/zfs_fuid.h>
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
#include <sys/zfs_vnops.h>
Add .zfs control directory Add support for the .zfs control directory. This was accomplished by leveraging as much of the existing ZFS infrastructure as posible and updating it for Linux as required. The bulk of the core functionality is now all there with the following limitations. *) The .zfs/snapshot directory automount support requires a 2.6.37 or newer kernel. The exception is RHEL6.2 which has backported the d_automount patches. *) Creating/destroying/renaming snapshots with mkdir/rmdir/mv in the .zfs/snapshot directory works as expected. However, this functionality is only available to root until zfs delegations are finished. * mkdir - create a snapshot * rmdir - destroy a snapshot * mv - rename a snapshot The following issues are known defeciences, but we expect them to be addressed by future commits. *) Add automount support for kernels older the 2.6.37. This should be possible using follow_link() which is what Linux did before. *) Accessing the .zfs/snapshot directory via NFS is not yet possible. The majority of the ground work for this is complete. However, finishing this work will require resolving some lingering integration issues with the Linux NFS kernel server. *) The .zfs/shares directory exists but no futher smb functionality has yet been implemented. Contributions-by: Rohan Puri <rohan.puri15@gmail.com> Contributiobs-by: Andrew Barnes <barnes333@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #173
2011-11-11 07:15:53 +00:00
#include <sys/zfs_ctldir.h>
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
#include <sys/dnode.h>
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
#include <sys/fs/zfs.h>
#include <sys/kidmap.h>
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
#include <sys/zpl.h>
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
#endif /* _KERNEL */
#include <sys/dmu.h>
Illumos 5027 - zfs large block support 5027 zfs large block support Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com> Reviewed by: Richard Elling <richard.elling@richardelling.com> Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/5027 https://github.com/illumos/illumos-gate/commit/b515258 Porting Notes: * Included in this patch is a tiny ISP2() cleanup in zio_init() from Illumos 5255. * Unlike the upstream Illumos commit this patch does not impose an arbitrary 128K block size limit on volumes. Volumes, like filesystems, are limited by the zfs_max_recordsize=1M module option. * By default the maximum record size is limited to 1M by the module option zfs_max_recordsize. This value may be safely increased up to 16M which is the largest block size supported by the on-disk format. At the moment, 1M blocks clearly offer a significant performance improvement but the benefits of going beyond this for the majority of workloads are less clear. * The illumos version of this patch increased DMU_MAX_ACCESS to 32M. This was determined not to be large enough when using 16M blocks because the zfs_make_xattrdir() function will fail (EFBIG) when assigning a TX. This was immediately observed under Linux because all newly created files must have a security xattr created and that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M. * On 32-bit platforms a hard limit of 1M is set for blocks due to the limited virtual address space. We should be able to relax this one the ABD patches are merged. Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #354
2014-11-03 20:15:08 +00:00
#include <sys/dmu_objset.h>
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
#include <sys/dmu_tx.h>
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
#include <sys/refcount.h>
#include <sys/stat.h>
#include <sys/zap.h>
#include <sys/zfs_znode.h>
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
#include <sys/sa.h>
#include <sys/zfs_sa.h>
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
#include <sys/zfs_stat.h>
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
#include "zfs_prop.h"
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
#include "zfs_comutil.h"
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* Define ZNODE_STATS to turn on statistic gathering. By default, it is only
* turned on when DEBUG is also defined.
*/
#ifdef DEBUG
#define ZNODE_STATS
#endif /* DEBUG */
#ifdef ZNODE_STATS
#define ZNODE_STAT_ADD(stat) ((stat)++)
#else
#define ZNODE_STAT_ADD(stat) /* nothing */
#endif /* ZNODE_STATS */
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* Functions needed for userland (ie: libzpool) are not put under
* #ifdef_KERNEL; the rest of the functions have dependencies
* (such as VFS logic) that will not compile easily in userland.
*/
#ifdef _KERNEL
Rebase master to b117
2009-07-02 22:44:48 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
static kmem_cache_t *znode_cache = NULL;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
static kmem_cache_t *znode_hold_cache = NULL;
Add zfs_object_mutex_size module option Add a zfs_object_mutex_size module option to facilitate resizing the the per-dataset znode mutex array. Increasing this value may help make the deadlock described in #4106 less common, but this is not a proper fix. This patch is primarily to aid debugging and analysis. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Issue #4106
2015-12-18 20:19:14 +00:00
unsigned int zfs_object_mutex_size = ZFS_OBJ_MTX_SZ;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*ARGSUSED*/
static int
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_znode_cache_constructor(void *buf, void *arg, int kmflags)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
znode_t *zp = buf;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
inode_init_once(ZTOI(zp));
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
list_link_init(&zp->z_link_node);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_init(&zp->z_lock, NULL, MUTEX_DEFAULT, NULL);
rw_init(&zp->z_parent_lock, NULL, RW_DEFAULT, NULL);
Identify locks flagged by lockdep When running a kernel with CONFIG_LOCKDEP=y, lockdep reports possible recursive locking in some cases and possible circular locking dependency in others, within the SPL and ZFS modules. This patch uses a mutex type defined in SPL, MUTEX_NOLOCKDEP, to mark such mutexes when they are initialized. This mutex type causes attempts to take or release those locks to be wrapped in lockdep_off() and lockdep_on() calls to silence the dependency checker and allow the use of lock_stats to examine contention. For RW locks, it uses an analogous lock type, RW_NOLOCKDEP. The goal is that these locks are ultimately changed back to type MUTEX_DEFAULT or RW_DEFAULT, after the locks are annotated to reflect their relationship (e.g. z_name_lock below) or any real problem with the lock dependencies are fixed. Some of the affected locks are: tc_open_lock: ============= This is an array of locks, all with same name, which txg_quiesce must take all of in order to move txg to next state. All default to the same lockdep class, and so to lockdep appears recursive. zp->z_name_lock: ================ In zfs_rmdir, dzp = znode for the directory (input to zfs_dirent_lock) zp = znode for the entry being removed (output of zfs_dirent_lock) zfs_rmdir()->zfs_dirent_lock() takes z_name_lock in dzp zfs_rmdir() takes z_name_lock in zp Since both dzp and zp are type znode_t, the locks have the same default class, and lockdep considers it a possible recursive lock attempt. l->l_rwlock: ============ zap_expand_leaf() sometimes creates two new zap leaf structures, via these call paths: zap_deref_leaf()->zap_get_leaf_byblk()->zap_leaf_open() zap_expand_leaf()->zap_create_leaf()->zap_expand_leaf()->zap_create_leaf() Because both zap_leaf_open() and zap_create_leaf() initialize l->l_rwlock in their (separate) leaf structures, the lockdep class is the same, and the linux kernel believes these might both be the same lock, and emits a possible recursive lock warning. Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3895
2015-10-15 20:08:27 +00:00
rw_init(&zp->z_name_lock, NULL, RW_NOLOCKDEP, NULL);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_init(&zp->z_acl_lock, NULL, MUTEX_DEFAULT, NULL);
Implement SA based xattrs The current ZFS implementation stores xattrs on disk using a hidden directory. In this directory a file name represents the xattr name and the file contexts are the xattr binary data. This approach is very flexible and allows for arbitrarily large xattrs. However, it also suffers from a significant performance penalty. Accessing a single xattr can requires up to three disk seeks. 1) Lookup the dnode object. 2) Lookup the dnodes's xattr directory object. 3) Lookup the xattr object in the directory. To avoid this performance penalty Linux filesystems such as ext3 and xfs try to store the xattr as part of the inode on disk. When the xattr is to large to store in the inode then a single external block is allocated for them. In practice most xattrs are small and this approach works well. The addition of System Attributes (SA) to zfs provides us a clean way to make this optimization. When the dataset property 'xattr=sa' is set then xattrs will be preferentially stored as System Attributes. This allows tiny xattrs (~100 bytes) to be stored with the dnode and up to 64k of xattrs to be stored in the spill block. If additional xattr space is required, which is unlikely under Linux, they will be stored using the traditional directory approach. This optimization results in roughly a 3x performance improvement when accessing xattrs which brings zfs roughly to parity with ext4 and xfs (see table below). When multiple xattrs are stored per-file the performance improvements are even greater because all of the xattrs stored in the spill block will be cached. However, by default SA based xattrs are disabled in the Linux port to maximize compatibility with other implementations. If you do enable SA based xattrs then they will not be visible on platforms which do not support this feature. ---------------------------------------------------------------------- Time in seconds to get/set one xattr of N bytes on 100,000 files ------+--------------------------------+------------------------------ | setxattr | getxattr bytes | ext4 xfs zfs-dir zfs-sa | ext4 xfs zfs-dir zfs-sa ------+--------------------------------+------------------------------ 1 | 2.33 31.88 21.50 4.57 | 2.35 2.64 6.29 2.43 32 | 2.79 30.68 21.98 4.60 | 2.44 2.59 6.78 2.48 256 | 3.25 31.99 21.36 5.92 | 2.32 2.71 6.22 3.14 1024 | 3.30 32.61 22.83 8.45 | 2.40 2.79 6.24 3.27 4096 | 3.57 317.46 22.52 10.73 | 2.78 28.62 6.90 3.94 16384 | n/a 2342.39 34.30 19.20 | n/a 45.44 145.90 7.55 65536 | n/a 2941.39 128.15 131.32* | n/a 141.92 256.85 262.12* Legend: * ext4 - Stock RHEL6.1 ext4 mounted with '-o user_xattr'. * xfs - Stock RHEL6.1 xfs mounted with default options. * zfs-dir - Directory based xattrs only. * zfs-sa - Prefer SAs but spill in to directories as needed, a trailing * indicates overflow in to directories occured. NOTE: Ext4 supports 4096 bytes of xattr name/value pairs per file. NOTE: XFS and ZFS have no limit on xattr name/value pairs per file. NOTE: Linux limits individual name/value pairs to 65536 bytes. NOTE: All setattr/getattr's were done after dropping the cache. NOTE: All tests were run against a single hard drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #443
2011-10-24 23:55:20 +00:00
rw_init(&zp->z_xattr_lock, NULL, RW_DEFAULT, NULL);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
zfs_rlock_init(&zp->z_range_lock);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zp->z_dirlocks = NULL;
Rebase master to b121
2009-08-18 18:43:27 +00:00
zp->z_acl_cached = NULL;
Implement SA based xattrs The current ZFS implementation stores xattrs on disk using a hidden directory. In this directory a file name represents the xattr name and the file contexts are the xattr binary data. This approach is very flexible and allows for arbitrarily large xattrs. However, it also suffers from a significant performance penalty. Accessing a single xattr can requires up to three disk seeks. 1) Lookup the dnode object. 2) Lookup the dnodes's xattr directory object. 3) Lookup the xattr object in the directory. To avoid this performance penalty Linux filesystems such as ext3 and xfs try to store the xattr as part of the inode on disk. When the xattr is to large to store in the inode then a single external block is allocated for them. In practice most xattrs are small and this approach works well. The addition of System Attributes (SA) to zfs provides us a clean way to make this optimization. When the dataset property 'xattr=sa' is set then xattrs will be preferentially stored as System Attributes. This allows tiny xattrs (~100 bytes) to be stored with the dnode and up to 64k of xattrs to be stored in the spill block. If additional xattr space is required, which is unlikely under Linux, they will be stored using the traditional directory approach. This optimization results in roughly a 3x performance improvement when accessing xattrs which brings zfs roughly to parity with ext4 and xfs (see table below). When multiple xattrs are stored per-file the performance improvements are even greater because all of the xattrs stored in the spill block will be cached. However, by default SA based xattrs are disabled in the Linux port to maximize compatibility with other implementations. If you do enable SA based xattrs then they will not be visible on platforms which do not support this feature. ---------------------------------------------------------------------- Time in seconds to get/set one xattr of N bytes on 100,000 files ------+--------------------------------+------------------------------ | setxattr | getxattr bytes | ext4 xfs zfs-dir zfs-sa | ext4 xfs zfs-dir zfs-sa ------+--------------------------------+------------------------------ 1 | 2.33 31.88 21.50 4.57 | 2.35 2.64 6.29 2.43 32 | 2.79 30.68 21.98 4.60 | 2.44 2.59 6.78 2.48 256 | 3.25 31.99 21.36 5.92 | 2.32 2.71 6.22 3.14 1024 | 3.30 32.61 22.83 8.45 | 2.40 2.79 6.24 3.27 4096 | 3.57 317.46 22.52 10.73 | 2.78 28.62 6.90 3.94 16384 | n/a 2342.39 34.30 19.20 | n/a 45.44 145.90 7.55 65536 | n/a 2941.39 128.15 131.32* | n/a 141.92 256.85 262.12* Legend: * ext4 - Stock RHEL6.1 ext4 mounted with '-o user_xattr'. * xfs - Stock RHEL6.1 xfs mounted with default options. * zfs-dir - Directory based xattrs only. * zfs-sa - Prefer SAs but spill in to directories as needed, a trailing * indicates overflow in to directories occured. NOTE: Ext4 supports 4096 bytes of xattr name/value pairs per file. NOTE: XFS and ZFS have no limit on xattr name/value pairs per file. NOTE: Linux limits individual name/value pairs to 65536 bytes. NOTE: All setattr/getattr's were done after dropping the cache. NOTE: All tests were run against a single hard drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #443
2011-10-24 23:55:20 +00:00
zp->z_xattr_cached = NULL;
Fix unlinked file cannot do xattr operations Currently, doing things like fsetxattr(2) on an unlinked file will result in ENODATA. There's two places that cause this: zfs_dirent_lock and zfs_zget. The fix in zfs_dirent_lock is pretty straightforward. In zfs_zget though, we need it to not return error when the zp is unlinked. This is a pretty big change in behavior, but skimming through all the callers, I don't think this change would cause any problem. Also there's nothing preventing z_unlinked from being set after the z_lock mutex is dropped before but before zfs_zget returns anyway. The rest of the stuff is to make sure we don't log xattr stuff when owner is unlinked. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-10-13 00:30:46 +00:00
zp->z_xattr_parent = 0;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
zp->z_moved = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (0);
}
/*ARGSUSED*/
static void
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_znode_cache_destructor(void *buf, void *arg)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
znode_t *zp = buf;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
ASSERT(!list_link_active(&zp->z_link_node));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_destroy(&zp->z_lock);
rw_destroy(&zp->z_parent_lock);
rw_destroy(&zp->z_name_lock);
mutex_destroy(&zp->z_acl_lock);
Implement SA based xattrs The current ZFS implementation stores xattrs on disk using a hidden directory. In this directory a file name represents the xattr name and the file contexts are the xattr binary data. This approach is very flexible and allows for arbitrarily large xattrs. However, it also suffers from a significant performance penalty. Accessing a single xattr can requires up to three disk seeks. 1) Lookup the dnode object. 2) Lookup the dnodes's xattr directory object. 3) Lookup the xattr object in the directory. To avoid this performance penalty Linux filesystems such as ext3 and xfs try to store the xattr as part of the inode on disk. When the xattr is to large to store in the inode then a single external block is allocated for them. In practice most xattrs are small and this approach works well. The addition of System Attributes (SA) to zfs provides us a clean way to make this optimization. When the dataset property 'xattr=sa' is set then xattrs will be preferentially stored as System Attributes. This allows tiny xattrs (~100 bytes) to be stored with the dnode and up to 64k of xattrs to be stored in the spill block. If additional xattr space is required, which is unlikely under Linux, they will be stored using the traditional directory approach. This optimization results in roughly a 3x performance improvement when accessing xattrs which brings zfs roughly to parity with ext4 and xfs (see table below). When multiple xattrs are stored per-file the performance improvements are even greater because all of the xattrs stored in the spill block will be cached. However, by default SA based xattrs are disabled in the Linux port to maximize compatibility with other implementations. If you do enable SA based xattrs then they will not be visible on platforms which do not support this feature. ---------------------------------------------------------------------- Time in seconds to get/set one xattr of N bytes on 100,000 files ------+--------------------------------+------------------------------ | setxattr | getxattr bytes | ext4 xfs zfs-dir zfs-sa | ext4 xfs zfs-dir zfs-sa ------+--------------------------------+------------------------------ 1 | 2.33 31.88 21.50 4.57 | 2.35 2.64 6.29 2.43 32 | 2.79 30.68 21.98 4.60 | 2.44 2.59 6.78 2.48 256 | 3.25 31.99 21.36 5.92 | 2.32 2.71 6.22 3.14 1024 | 3.30 32.61 22.83 8.45 | 2.40 2.79 6.24 3.27 4096 | 3.57 317.46 22.52 10.73 | 2.78 28.62 6.90 3.94 16384 | n/a 2342.39 34.30 19.20 | n/a 45.44 145.90 7.55 65536 | n/a 2941.39 128.15 131.32* | n/a 141.92 256.85 262.12* Legend: * ext4 - Stock RHEL6.1 ext4 mounted with '-o user_xattr'. * xfs - Stock RHEL6.1 xfs mounted with default options. * zfs-dir - Directory based xattrs only. * zfs-sa - Prefer SAs but spill in to directories as needed, a trailing * indicates overflow in to directories occured. NOTE: Ext4 supports 4096 bytes of xattr name/value pairs per file. NOTE: XFS and ZFS have no limit on xattr name/value pairs per file. NOTE: Linux limits individual name/value pairs to 65536 bytes. NOTE: All setattr/getattr's were done after dropping the cache. NOTE: All tests were run against a single hard drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #443
2011-10-24 23:55:20 +00:00
rw_destroy(&zp->z_xattr_lock);
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
zfs_rlock_destroy(&zp->z_range_lock);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
ASSERT(zp->z_dirlocks == NULL);
Rebase master to b121
2009-08-18 18:43:27 +00:00
ASSERT(zp->z_acl_cached == NULL);
Implement SA based xattrs The current ZFS implementation stores xattrs on disk using a hidden directory. In this directory a file name represents the xattr name and the file contexts are the xattr binary data. This approach is very flexible and allows for arbitrarily large xattrs. However, it also suffers from a significant performance penalty. Accessing a single xattr can requires up to three disk seeks. 1) Lookup the dnode object. 2) Lookup the dnodes's xattr directory object. 3) Lookup the xattr object in the directory. To avoid this performance penalty Linux filesystems such as ext3 and xfs try to store the xattr as part of the inode on disk. When the xattr is to large to store in the inode then a single external block is allocated for them. In practice most xattrs are small and this approach works well. The addition of System Attributes (SA) to zfs provides us a clean way to make this optimization. When the dataset property 'xattr=sa' is set then xattrs will be preferentially stored as System Attributes. This allows tiny xattrs (~100 bytes) to be stored with the dnode and up to 64k of xattrs to be stored in the spill block. If additional xattr space is required, which is unlikely under Linux, they will be stored using the traditional directory approach. This optimization results in roughly a 3x performance improvement when accessing xattrs which brings zfs roughly to parity with ext4 and xfs (see table below). When multiple xattrs are stored per-file the performance improvements are even greater because all of the xattrs stored in the spill block will be cached. However, by default SA based xattrs are disabled in the Linux port to maximize compatibility with other implementations. If you do enable SA based xattrs then they will not be visible on platforms which do not support this feature. ---------------------------------------------------------------------- Time in seconds to get/set one xattr of N bytes on 100,000 files ------+--------------------------------+------------------------------ | setxattr | getxattr bytes | ext4 xfs zfs-dir zfs-sa | ext4 xfs zfs-dir zfs-sa ------+--------------------------------+------------------------------ 1 | 2.33 31.88 21.50 4.57 | 2.35 2.64 6.29 2.43 32 | 2.79 30.68 21.98 4.60 | 2.44 2.59 6.78 2.48 256 | 3.25 31.99 21.36 5.92 | 2.32 2.71 6.22 3.14 1024 | 3.30 32.61 22.83 8.45 | 2.40 2.79 6.24 3.27 4096 | 3.57 317.46 22.52 10.73 | 2.78 28.62 6.90 3.94 16384 | n/a 2342.39 34.30 19.20 | n/a 45.44 145.90 7.55 65536 | n/a 2941.39 128.15 131.32* | n/a 141.92 256.85 262.12* Legend: * ext4 - Stock RHEL6.1 ext4 mounted with '-o user_xattr'. * xfs - Stock RHEL6.1 xfs mounted with default options. * zfs-dir - Directory based xattrs only. * zfs-sa - Prefer SAs but spill in to directories as needed, a trailing * indicates overflow in to directories occured. NOTE: Ext4 supports 4096 bytes of xattr name/value pairs per file. NOTE: XFS and ZFS have no limit on xattr name/value pairs per file. NOTE: Linux limits individual name/value pairs to 65536 bytes. NOTE: All setattr/getattr's were done after dropping the cache. NOTE: All tests were run against a single hard drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #443
2011-10-24 23:55:20 +00:00
ASSERT(zp->z_xattr_cached == NULL);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
}
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
static int
zfs_znode_hold_cache_constructor(void *buf, void *arg, int kmflags)
{
znode_hold_t *zh = buf;
mutex_init(&zh->zh_lock, NULL, MUTEX_DEFAULT, NULL);
refcount_create(&zh->zh_refcount);
zh->zh_obj = ZFS_NO_OBJECT;
return (0);
}
static void
zfs_znode_hold_cache_destructor(void *buf, void *arg)
{
znode_hold_t *zh = buf;
mutex_destroy(&zh->zh_lock);
refcount_destroy(&zh->zh_refcount);
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
void
zfs_znode_init(void)
{
/*
Allocate zfs_znode_cache on the Linux slab The Linux slab, in general, performs better than the SPl slab in cases where a lot of objects are allocated and fragmentation is likely present. This patch fixes pathologically bad behavior in cases where the ARC is filled with mostly metadata and a user program needs to allocate and dirty enough memory which would require an insignificant amount of the ARC to be reclaimed. If zfs_znode_cache is on the SPL slab, the system may spin for a very long time trying to reclaim sufficient memory. If it is on the Linux slab, the behavior has been observed to be much more predictible; the memory is reclaimed more efficiently. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #3283
2015-04-14 05:06:40 +00:00
* Initialize zcache. The KMC_SLAB hint is used in order that it be
* backed by kmalloc() when on the Linux slab in order that any
* wait_on_bit() operations on the related inode operate properly.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
ASSERT(znode_cache == NULL);
znode_cache = kmem_cache_create("zfs_znode_cache",
sizeof (znode_t), 0, zfs_znode_cache_constructor,
Allocate zfs_znode_cache on the Linux slab The Linux slab, in general, performs better than the SPl slab in cases where a lot of objects are allocated and fragmentation is likely present. This patch fixes pathologically bad behavior in cases where the ARC is filled with mostly metadata and a user program needs to allocate and dirty enough memory which would require an insignificant amount of the ARC to be reclaimed. If zfs_znode_cache is on the SPL slab, the system may spin for a very long time trying to reclaim sufficient memory. If it is on the Linux slab, the behavior has been observed to be much more predictible; the memory is reclaimed more efficiently. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #3283
2015-04-14 05:06:40 +00:00
zfs_znode_cache_destructor, NULL, NULL, NULL, KMC_SLAB);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
ASSERT(znode_hold_cache == NULL);
znode_hold_cache = kmem_cache_create("zfs_znode_hold_cache",
sizeof (znode_hold_t), 0, zfs_znode_hold_cache_constructor,
zfs_znode_hold_cache_destructor, NULL, NULL, NULL, 0);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
void
zfs_znode_fini(void)
{
/*
* Cleanup zcache
*/
if (znode_cache)
kmem_cache_destroy(znode_cache);
znode_cache = NULL;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
if (znode_hold_cache)
kmem_cache_destroy(znode_hold_cache);
znode_hold_cache = NULL;
}
/*
* The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to
* serialize access to a znode and its SA buffer while the object is being
* created or destroyed. This kind of locking would normally reside in the
* znode itself but in this case that's impossible because the znode and SA
* buffer may not yet exist. Therefore the locking is handled externally
* with an array of mutexs and AVLs trees which contain per-object locks.
*
* In zfs_znode_hold_enter() a per-object lock is created as needed, inserted
* in to the correct AVL tree and finally the per-object lock is held. In
* zfs_znode_hold_exit() the process is reversed. The per-object lock is
* released, removed from the AVL tree and destroyed if there are no waiters.
*
* This scheme has two important properties:
*
* 1) No memory allocations are performed while holding one of the z_hold_locks.
* This ensures evict(), which can be called from direct memory reclaim, will
* never block waiting on a z_hold_locks which just happens to have hashed
* to the same index.
*
* 2) All locks used to serialize access to an object are per-object and never
* shared. This minimizes lock contention without creating a large number
* of dedicated locks.
*
* On the downside it does require znode_lock_t structures to be frequently
* allocated and freed. However, because these are backed by a kmem cache
* and very short lived this cost is minimal.
*/
int
zfs_znode_hold_compare(const void *a, const void *b)
{
Performance optimization of AVL tree comparator functions perf: 2.75x faster ddt_entry_compare() First 256bits of ddt_key_t is a block checksum, which are expected to be close to random data. Hence, on average, comparison only needs to look at first few bytes of the keys. To reduce number of conditional jump instructions, the result is computed as: sign(memcmp(k1, k2)). Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} , which is computed efficiently. Synthetic performance evaluation of original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R) CPU E5-2660 v3: old 6.85789 s new 2.49089 s perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare() Compute the result directly instead of using conditionals perf: zfs_range_compare() Speedup between 1.1x - 2.5x, depending on compiler version and optimization level. perf: spa_error_entry_compare() `bcmp()` is not suitable for comparator use. Use `memcmp()` instead. perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare() perf: 2.8x faster zil_bp_compare() perf: 2.8x faster mze_compare() perf: faster dbuf_compare() perf: faster compares in spa_misc perf: 2.8x faster layout_hash_compare() perf: 2.8x faster space_reftree_compare() perf: libzfs: faster avl tree comparators perf: guid_compare() perf: dsl_deadlist_compare() perf: perm_set_compare() perf: 2x faster range_tree_seg_compare() perf: faster unique_compare() perf: faster vdev_cache _compare() perf: faster vdev_uberblock_compare() perf: faster fuid _compare() perf: faster zfs_znode_hold_compare() Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Signed-off-by: Richard Elling <richard.elling@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5033
2016-08-27 18:12:53 +00:00
const znode_hold_t *zh_a = (const znode_hold_t *)a;
const znode_hold_t *zh_b = (const znode_hold_t *)b;
return (AVL_CMP(zh_a->zh_obj, zh_b->zh_obj));
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
}
boolean_t
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_held(zfsvfs_t *zfsvfs, uint64_t obj)
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
{
znode_hold_t *zh, search;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
int i = ZFS_OBJ_HASH(zfsvfs, obj);
Close possible zfs_znode_held() race Check if the lock is held while holding the z_hold_locks() lock. This prevents a possible use-after-free bug for callers which are not holding the lock. There currently are no such callers so this can't cause a problem today but it has been fixed regardless. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4244 Issue #4124
2016-01-19 18:41:21 +00:00
boolean_t held;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
search.zh_obj = obj;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_enter(&zfsvfs->z_hold_locks[i]);
zh = avl_find(&zfsvfs->z_hold_trees[i], &search, NULL);
Close possible zfs_znode_held() race Check if the lock is held while holding the z_hold_locks() lock. This prevents a possible use-after-free bug for callers which are not holding the lock. There currently are no such callers so this can't cause a problem today but it has been fixed regardless. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4244 Issue #4124
2016-01-19 18:41:21 +00:00
held = (zh && MUTEX_HELD(&zh->zh_lock)) ? B_TRUE : B_FALSE;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_exit(&zfsvfs->z_hold_locks[i]);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
Close possible zfs_znode_held() race Check if the lock is held while holding the z_hold_locks() lock. This prevents a possible use-after-free bug for callers which are not holding the lock. There currently are no such callers so this can't cause a problem today but it has been fixed regardless. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4244 Issue #4124
2016-01-19 18:41:21 +00:00
return (held);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
}
static znode_hold_t *
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_enter(zfsvfs_t *zfsvfs, uint64_t obj)
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
{
znode_hold_t *zh, *zh_new, search;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
int i = ZFS_OBJ_HASH(zfsvfs, obj);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
boolean_t found = B_FALSE;
zh_new = kmem_cache_alloc(znode_hold_cache, KM_SLEEP);
zh_new->zh_obj = obj;
search.zh_obj = obj;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_enter(&zfsvfs->z_hold_locks[i]);
zh = avl_find(&zfsvfs->z_hold_trees[i], &search, NULL);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
if (likely(zh == NULL)) {
zh = zh_new;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
avl_add(&zfsvfs->z_hold_trees[i], zh);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
} else {
ASSERT3U(zh->zh_obj, ==, obj);
found = B_TRUE;
}
refcount_add(&zh->zh_refcount, NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_exit(&zfsvfs->z_hold_locks[i]);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
if (found == B_TRUE)
kmem_cache_free(znode_hold_cache, zh_new);
ASSERT(MUTEX_NOT_HELD(&zh->zh_lock));
ASSERT3S(refcount_count(&zh->zh_refcount), >, 0);
mutex_enter(&zh->zh_lock);
return (zh);
}
static void
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs_t *zfsvfs, znode_hold_t *zh)
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
int i = ZFS_OBJ_HASH(zfsvfs, zh->zh_obj);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
boolean_t remove = B_FALSE;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
ASSERT(zfs_znode_held(zfsvfs, zh->zh_obj));
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
ASSERT3S(refcount_count(&zh->zh_refcount), >, 0);
mutex_exit(&zh->zh_lock);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_enter(&zfsvfs->z_hold_locks[i]);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
if (refcount_remove(&zh->zh_refcount, NULL) == 0) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
avl_remove(&zfsvfs->z_hold_trees[i], zh);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
remove = B_TRUE;
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_exit(&zfsvfs->z_hold_locks[i]);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
if (remove == B_TRUE)
kmem_cache_free(znode_hold_cache, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
int
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_create_share_dir(zfsvfs_t *zfsvfs, dmu_tx_t *tx)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Renamed HAVE_SHARE ifdefs to HAVE_SMB_SHARE. The remaining code that is guarded by HAVE_SHARE ifdefs is related to the .zfs/shares functionality which is currently not available on Linux. On Solaris the .zfs/shares directory can be used to set permissions for SMB shares. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2011-07-02 19:34:12 +00:00
#ifdef HAVE_SMB_SHARE
Rebase master to b117
2009-07-02 22:44:48 +00:00
zfs_acl_ids_t acl_ids;
vattr_t vattr;
znode_t *sharezp;
vnode_t *vp;
znode_t *zp;
int error;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase master to b117
2009-07-02 22:44:48 +00:00
vattr.va_mask = AT_MODE|AT_UID|AT_GID|AT_TYPE;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
vattr.va_mode = S_IFDIR | 0555;
Rebase master to b117
2009-07-02 22:44:48 +00:00
vattr.va_uid = crgetuid(kcred);
vattr.va_gid = crgetgid(kcred);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Change KM_PUSHPAGE -> KM_SLEEP By marking DMU transaction processing contexts with PF_FSTRANS we can revert the KM_PUSHPAGE -> KM_SLEEP changes. This brings us back in line with upstream. In some cases this means simply swapping the flags back. For others fnvlist_alloc() was replaced by nvlist_alloc(..., KM_PUSHPAGE) and must be reverted back to fnvlist_alloc() which assumes KM_SLEEP. The one place KM_PUSHPAGE is kept is when allocating ARC buffers which allows us to dip in to reserved memory. This is again the same as upstream. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-11-21 00:09:39 +00:00
sharezp = kmem_cache_alloc(znode_cache, KM_SLEEP);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
sharezp->z_moved = 0;
Rebase master to b117
2009-07-02 22:44:48 +00:00
sharezp->z_unlinked = 0;
sharezp->z_atime_dirty = 0;
sharezp->z_zfsvfs = zfsvfs;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sharezp->z_is_sa = zfsvfs->z_use_sa;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase master to b117
2009-07-02 22:44:48 +00:00
vp = ZTOV(sharezp);
vn_reinit(vp);
vp->v_type = VDIR;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase master to b117
2009-07-02 22:44:48 +00:00
VERIFY(0 == zfs_acl_ids_create(sharezp, IS_ROOT_NODE, &vattr,
kcred, NULL, &acl_ids));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zfs_mknode(sharezp, &vattr, tx, kcred, IS_ROOT_NODE, &zp, &acl_ids);
Rebase master to b117
2009-07-02 22:44:48 +00:00
ASSERT3P(zp, ==, sharezp);
ASSERT(!vn_in_dnlc(ZTOV(sharezp))); /* not valid to move */
POINTER_INVALIDATE(&sharezp->z_zfsvfs);
error = zap_add(zfsvfs->z_os, MASTER_NODE_OBJ,
ZFS_SHARES_DIR, 8, 1, &sharezp->z_id, tx);
zfsvfs->z_shares_dir = sharezp->z_id;
zfs_acl_ids_free(&acl_ids);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
// ZTOV(sharezp)->v_count = 0;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sa_handle_destroy(sharezp->z_sa_hdl);
Rebase master to b117
2009-07-02 22:44:48 +00:00
kmem_cache_free(znode_cache, sharezp);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase master to b117
2009-07-02 22:44:48 +00:00
return (error);
VFS: Wrap with HAVE_SHARE Certain NFS/SMB share functionality is not yet in place. These functions used to be wrapped with the generic HAVE_ZPL to prevent them from being compiled. I still don't want them compiled but I'm working toward eliminating the use of HAVE_ZPL. So I'm just renaming the wrapper here to HAVE_SHARE. They still won't be compiled until all the share issues are worked through. Share support is the last missing piece from zfs_ioctl.c.
2011-01-07 19:51:27 +00:00
#else
return (0);
Renamed HAVE_SHARE ifdefs to HAVE_SMB_SHARE. The remaining code that is guarded by HAVE_SHARE ifdefs is related to the .zfs/shares functionality which is currently not available on Linux. On Solaris the .zfs/shares directory can be used to set permissions for SMB shares. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2011-07-02 19:34:12 +00:00
#endif /* HAVE_SMB_SHARE */
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
static void
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_sa_init(zfsvfs_t *zfsvfs, znode_t *zp,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_buf_t *db, dmu_object_type_t obj_type, sa_handle_t *sa_hdl)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
ASSERT(zfs_znode_held(zfsvfs, zp->z_id));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_enter(&zp->z_lock);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ASSERT(zp->z_sa_hdl == NULL);
ASSERT(zp->z_acl_cached == NULL);
if (sa_hdl == NULL) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
VERIFY(0 == sa_handle_get_from_db(zfsvfs->z_os, db, zp,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
SA_HDL_SHARED, &zp->z_sa_hdl));
} else {
zp->z_sa_hdl = sa_hdl;
sa_set_userp(sa_hdl, zp);
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_is_sa = (obj_type == DMU_OT_SA) ? B_TRUE : B_FALSE;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_exit(&zp->z_lock);
}
void
zfs_znode_dmu_fini(znode_t *zp)
{
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
ASSERT(zfs_znode_held(ZTOZSB(zp), zp->z_id) || zp->z_unlinked ||
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
RW_WRITE_HELD(&ZTOZSB(zp)->z_teardown_inactive_lock));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sa_handle_destroy(zp->z_sa_hdl);
zp->z_sa_hdl = NULL;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
/*
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
* Called by new_inode() to allocate a new inode.
*/
int
zfs_inode_alloc(struct super_block *sb, struct inode **ip)
{
znode_t *zp;
Change KM_PUSHPAGE -> KM_SLEEP By marking DMU transaction processing contexts with PF_FSTRANS we can revert the KM_PUSHPAGE -> KM_SLEEP changes. This brings us back in line with upstream. In some cases this means simply swapping the flags back. For others fnvlist_alloc() was replaced by nvlist_alloc(..., KM_PUSHPAGE) and must be reverted back to fnvlist_alloc() which assumes KM_SLEEP. The one place KM_PUSHPAGE is kept is when allocating ARC buffers which allows us to dip in to reserved memory. This is again the same as upstream. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-11-21 00:09:39 +00:00
zp = kmem_cache_alloc(znode_cache, KM_SLEEP);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
*ip = ZTOI(zp);
return (0);
}
/*
* Called in multiple places when an inode should be destroyed.
*/
void
zfs_inode_destroy(struct inode *ip)
{
znode_t *zp = ITOZ(ip);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_enter(&zfsvfs->z_znodes_lock);
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
if (list_link_active(&zp->z_link_node)) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
list_remove(&zfsvfs->z_all_znodes, zp);
zfsvfs->z_nr_znodes--;
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_exit(&zfsvfs->z_znodes_lock);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (zp->z_acl_cached) {
zfs_acl_free(zp->z_acl_cached);
zp->z_acl_cached = NULL;
}
Implement SA based xattrs The current ZFS implementation stores xattrs on disk using a hidden directory. In this directory a file name represents the xattr name and the file contexts are the xattr binary data. This approach is very flexible and allows for arbitrarily large xattrs. However, it also suffers from a significant performance penalty. Accessing a single xattr can requires up to three disk seeks. 1) Lookup the dnode object. 2) Lookup the dnodes's xattr directory object. 3) Lookup the xattr object in the directory. To avoid this performance penalty Linux filesystems such as ext3 and xfs try to store the xattr as part of the inode on disk. When the xattr is to large to store in the inode then a single external block is allocated for them. In practice most xattrs are small and this approach works well. The addition of System Attributes (SA) to zfs provides us a clean way to make this optimization. When the dataset property 'xattr=sa' is set then xattrs will be preferentially stored as System Attributes. This allows tiny xattrs (~100 bytes) to be stored with the dnode and up to 64k of xattrs to be stored in the spill block. If additional xattr space is required, which is unlikely under Linux, they will be stored using the traditional directory approach. This optimization results in roughly a 3x performance improvement when accessing xattrs which brings zfs roughly to parity with ext4 and xfs (see table below). When multiple xattrs are stored per-file the performance improvements are even greater because all of the xattrs stored in the spill block will be cached. However, by default SA based xattrs are disabled in the Linux port to maximize compatibility with other implementations. If you do enable SA based xattrs then they will not be visible on platforms which do not support this feature. ---------------------------------------------------------------------- Time in seconds to get/set one xattr of N bytes on 100,000 files ------+--------------------------------+------------------------------ | setxattr | getxattr bytes | ext4 xfs zfs-dir zfs-sa | ext4 xfs zfs-dir zfs-sa ------+--------------------------------+------------------------------ 1 | 2.33 31.88 21.50 4.57 | 2.35 2.64 6.29 2.43 32 | 2.79 30.68 21.98 4.60 | 2.44 2.59 6.78 2.48 256 | 3.25 31.99 21.36 5.92 | 2.32 2.71 6.22 3.14 1024 | 3.30 32.61 22.83 8.45 | 2.40 2.79 6.24 3.27 4096 | 3.57 317.46 22.52 10.73 | 2.78 28.62 6.90 3.94 16384 | n/a 2342.39 34.30 19.20 | n/a 45.44 145.90 7.55 65536 | n/a 2941.39 128.15 131.32* | n/a 141.92 256.85 262.12* Legend: * ext4 - Stock RHEL6.1 ext4 mounted with '-o user_xattr'. * xfs - Stock RHEL6.1 xfs mounted with default options. * zfs-dir - Directory based xattrs only. * zfs-sa - Prefer SAs but spill in to directories as needed, a trailing * indicates overflow in to directories occured. NOTE: Ext4 supports 4096 bytes of xattr name/value pairs per file. NOTE: XFS and ZFS have no limit on xattr name/value pairs per file. NOTE: Linux limits individual name/value pairs to 65536 bytes. NOTE: All setattr/getattr's were done after dropping the cache. NOTE: All tests were run against a single hard drive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #443
2011-10-24 23:55:20 +00:00
if (zp->z_xattr_cached) {
nvlist_free(zp->z_xattr_cached);
zp->z_xattr_cached = NULL;
}
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
kmem_cache_free(znode_cache, zp);
}
static void
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_inode_set_ops(zfsvfs_t *zfsvfs, struct inode *ip)
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
{
Don't store rdev in SA for FIFOs and sockets Update the handling of named pipes and sockets to be consistent with other platforms with regard to the rdev attribute. While all ZFS ipmlementations store the rdev for device files in a system attribute (SA), this is not the case for FIFOs and sockets. Indeed, Linux always passes rdev=0 to mknod() for FIFOs and sockets, so the value is not needed. Add an ASSERT that rdev==0 for FIFOs and sockets to detect if the expected behavior ever changes. Closes #216
2011-05-09 19:31:56 +00:00
uint64_t rdev = 0;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
switch (ip->i_mode & S_IFMT) {
case S_IFREG:
ip->i_op = &zpl_inode_operations;
ip->i_fop = &zpl_file_operations;
ip->i_mapping->a_ops = &zpl_address_space_operations;
break;
case S_IFDIR:
ip->i_op = &zpl_dir_inode_operations;
ip->i_fop = &zpl_dir_file_operations;
ITOZ(ip)->z_zn_prefetch = B_TRUE;
break;
case S_IFLNK:
ip->i_op = &zpl_symlink_inode_operations;
break;
Don't store rdev in SA for FIFOs and sockets Update the handling of named pipes and sockets to be consistent with other platforms with regard to the rdev attribute. While all ZFS ipmlementations store the rdev for device files in a system attribute (SA), this is not the case for FIFOs and sockets. Indeed, Linux always passes rdev=0 to mknod() for FIFOs and sockets, so the value is not needed. Add an ASSERT that rdev==0 for FIFOs and sockets to detect if the expected behavior ever changes. Closes #216
2011-05-09 19:31:56 +00:00
/*
* rdev is only stored in a SA only for device files.
*/
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
case S_IFCHR:
case S_IFBLK:
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
(void) sa_lookup(ITOZ(ip)->z_sa_hdl, SA_ZPL_RDEV(zfsvfs), &rdev,
Add logic to try and recover an inode with an invalid mode When an inode is detected with invalid mode bits the safe thing to do is panic the system. This indicates a problem with the contents of a dnode and it should never be possible. This is the default behavior. Unfortunately, due to flaws in the system attribute (SA) implementation (on all platforms) it was possible that ZFS could create a damaged dnode. This was a rare issue which only impacted dnodes which used a spill block. Normally only symlinks and files with ACLs would require a spill block. However, if the dataset had the xattr=sa property set and extended attributes were used this problem could occur. As of the 0.6.4 tag the root cause of this issue has been fixed. For pools which are exhibiting this damage the 'zfs_recover=1' module option may be set. This will cause ZFS to interpret the dnode with invalid mode bits as a normal file. This may allow the files to be accessed for recovery purposes. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3548
2015-07-13 21:51:59 +00:00
sizeof (rdev));
Don't store rdev in SA for FIFOs and sockets Update the handling of named pipes and sockets to be consistent with other platforms with regard to the rdev attribute. While all ZFS ipmlementations store the rdev for device files in a system attribute (SA), this is not the case for FIFOs and sockets. Indeed, Linux always passes rdev=0 to mknod() for FIFOs and sockets, so the value is not needed. Add an ASSERT that rdev==0 for FIFOs and sockets to detect if the expected behavior ever changes. Closes #216
2011-05-09 19:31:56 +00:00
/*FALLTHROUGH*/
case S_IFIFO:
case S_IFSOCK:
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
init_special_inode(ip, ip->i_mode, rdev);
ip->i_op = &zpl_special_inode_operations;
break;
default:
Add logic to try and recover an inode with an invalid mode When an inode is detected with invalid mode bits the safe thing to do is panic the system. This indicates a problem with the contents of a dnode and it should never be possible. This is the default behavior. Unfortunately, due to flaws in the system attribute (SA) implementation (on all platforms) it was possible that ZFS could create a damaged dnode. This was a rare issue which only impacted dnodes which used a spill block. Normally only symlinks and files with ACLs would require a spill block. However, if the dataset had the xattr=sa property set and extended attributes were used this problem could occur. As of the 0.6.4 tag the root cause of this issue has been fixed. For pools which are exhibiting this damage the 'zfs_recover=1' module option may be set. This will cause ZFS to interpret the dnode with invalid mode bits as a normal file. This may allow the files to be accessed for recovery purposes. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3548
2015-07-13 21:51:59 +00:00
zfs_panic_recover("inode %llu has invalid mode: 0x%x\n",
(u_longlong_t)ip->i_ino, ip->i_mode);
/* Assume the inode is a file and attempt to continue */
ip->i_mode = S_IFREG | 0644;
ip->i_op = &zpl_inode_operations;
ip->i_fop = &zpl_file_operations;
ip->i_mapping->a_ops = &zpl_address_space_operations;
break;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
}
}
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
void
zfs_set_inode_flags(znode_t *zp, struct inode *ip)
{
/*
* Linux and Solaris have different sets of file attributes, so we
* restrict this conversion to the intersection of the two.
*/
Use inode_set_flags when available Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-16 21:54:51 +00:00
#ifdef HAVE_INODE_SET_FLAGS
unsigned int flags = 0;
if (zp->z_pflags & ZFS_IMMUTABLE)
flags |= S_IMMUTABLE;
if (zp->z_pflags & ZFS_APPENDONLY)
flags |= S_APPEND;
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
Use inode_set_flags when available Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-16 21:54:51 +00:00
inode_set_flags(ip, flags, S_IMMUTABLE|S_APPEND);
#else
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
if (zp->z_pflags & ZFS_IMMUTABLE)
ip->i_flags |= S_IMMUTABLE;
else
ip->i_flags &= ~S_IMMUTABLE;
if (zp->z_pflags & ZFS_APPENDONLY)
ip->i_flags |= S_APPEND;
else
ip->i_flags &= ~S_APPEND;
Use inode_set_flags when available Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-16 21:54:51 +00:00
#endif
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
}
Enable lazytime semantic for atime Linux 4.0 introduces lazytime. The idea is that when we update the atime, we delay writing it to disk for as long as it is reasonably possible. When lazytime is enabled, dirty_inode will be called with only I_DIRTY_TIME flag whenever i_atime is updated. So under such condition, we will set z_atime_dirty. We will only write it to disk if file is closed, inode is evicted or setattr is called. Ideally, we should also write it whenever SA is going to be updated, but it is left for future improvement. There's one thing that we should take care of now that we allow i_atime to be dirty. In original implementation, whenever SA is modified, zfs_inode_update will be called to overwrite every thing in inode. This will cause dirty i_atime to be discarded. We fix this by don't overwrite i_atime in zfs_inode_update. We only overwrite i_atime when allocating new inode or doing zfs_rezget with zfs_inode_update_new. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-31 23:52:03 +00:00
/*
* Update the embedded inode given the znode. We should work toward
* eliminating this function as soon as possible by removing values
* which are duplicated between the znode and inode. If the generic
* inode has the correct field it should be used, and the ZFS code
* updated to access the inode. This can be done incrementally.
*/
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
void
zfs_inode_update(znode_t *zp)
Enable lazytime semantic for atime Linux 4.0 introduces lazytime. The idea is that when we update the atime, we delay writing it to disk for as long as it is reasonably possible. When lazytime is enabled, dirty_inode will be called with only I_DIRTY_TIME flag whenever i_atime is updated. So under such condition, we will set z_atime_dirty. We will only write it to disk if file is closed, inode is evicted or setattr is called. Ideally, we should also write it whenever SA is going to be updated, but it is left for future improvement. There's one thing that we should take care of now that we allow i_atime to be dirty. In original implementation, whenever SA is modified, zfs_inode_update will be called to overwrite every thing in inode. This will cause dirty i_atime to be discarded. We fix this by don't overwrite i_atime in zfs_inode_update. We only overwrite i_atime when allocating new inode or doing zfs_rezget with zfs_inode_update_new. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-31 23:52:03 +00:00
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs;
Enable lazytime semantic for atime Linux 4.0 introduces lazytime. The idea is that when we update the atime, we delay writing it to disk for as long as it is reasonably possible. When lazytime is enabled, dirty_inode will be called with only I_DIRTY_TIME flag whenever i_atime is updated. So under such condition, we will set z_atime_dirty. We will only write it to disk if file is closed, inode is evicted or setattr is called. Ideally, we should also write it whenever SA is going to be updated, but it is left for future improvement. There's one thing that we should take care of now that we allow i_atime to be dirty. In original implementation, whenever SA is modified, zfs_inode_update will be called to overwrite every thing in inode. This will cause dirty i_atime to be discarded. We fix this by don't overwrite i_atime in zfs_inode_update. We only overwrite i_atime when allocating new inode or doing zfs_rezget with zfs_inode_update_new. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-31 23:52:03 +00:00
struct inode *ip;
uint32_t blksize;
u_longlong_t i_blocks;
ASSERT(zp != NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs = ZTOZSB(zp);
Enable lazytime semantic for atime Linux 4.0 introduces lazytime. The idea is that when we update the atime, we delay writing it to disk for as long as it is reasonably possible. When lazytime is enabled, dirty_inode will be called with only I_DIRTY_TIME flag whenever i_atime is updated. So under such condition, we will set z_atime_dirty. We will only write it to disk if file is closed, inode is evicted or setattr is called. Ideally, we should also write it whenever SA is going to be updated, but it is left for future improvement. There's one thing that we should take care of now that we allow i_atime to be dirty. In original implementation, whenever SA is modified, zfs_inode_update will be called to overwrite every thing in inode. This will cause dirty i_atime to be discarded. We fix this by don't overwrite i_atime in zfs_inode_update. We only overwrite i_atime when allocating new inode or doing zfs_rezget with zfs_inode_update_new. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-31 23:52:03 +00:00
ip = ZTOI(zp);
/* Skip .zfs control nodes which do not exist on disk. */
if (zfsctl_is_node(ip))
return;
dmu_object_size_from_db(sa_get_db(zp->z_sa_hdl), &blksize, &i_blocks);
spin_lock(&ip->i_lock);
ip->i_blocks = i_blocks;
i_size_write(ip, zp->z_size);
spin_unlock(&ip->i_lock);
}
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
/*
* Construct a znode+inode and initialize.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*
* This does not do a call to dmu_set_user() that is
* up to the caller to do, in case you don't want to
* return the znode
*/
static znode_t *
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_alloc(zfsvfs_t *zfsvfs, dmu_buf_t *db, int blksz,
Kill zp->z_xattr_parent to prevent pinning zp->z_xattr_parent will pin the parent. This will cause huge issue when unlink a file with xattr. Because the unlinked file is pinned, it will never get purged immediately. And because of that, the xattr stuff will never be marked as unlinked. So the whole unlinked stuff will stay there until shrink cache or umount. This change partially reverts e89260a. This is safe because only the zp->z_xattr_parent optimization is removed, zpl_xattr_security_init() is still called from the zpl outside the inode lock. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlop <chris@onthe.net.au> Issue #4359 Issue #3508 Issue #4413 Issue #4827
2016-07-06 00:24:36 +00:00
dmu_object_type_t obj_type, uint64_t obj, sa_handle_t *hdl)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
znode_t *zp;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
struct inode *ip;
Use local variable to read zp->z_mode When accessing the zp->z_mode through the SA bulk interface we expect that 64-bits are available to hold the result. However, on 32-bit platforms mode_t will only be 32-bits so we cannot pass it to SA_ADD_BULK_ATTR(). Instead a local uint64_t variable must be used and the result assigned to zp->z_mode. This went unnoticed on 32-bit little endian platforms because the bytes happen to end up in the correct 32-bits. But on big endian platforms like Sparc the zp->z_mode will always end up set to zero. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: marku89 <mar42@kola.li> Issue #1700
2014-01-07 22:16:46 +00:00
uint64_t mode;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t parent;
Kill znode->z_gen field This field is a duplicate of the inode->i_generation, so just kill it. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4538 Closes #4654
2016-04-18 19:08:53 +00:00
uint64_t tmp_gen;
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
uint64_t links;
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
uint64_t z_uid, z_gid;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
uint64_t atime[2], mtime[2], ctime[2];
sa_bulk_attr_t bulk[11];
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
int count = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
ASSERT(zfsvfs != NULL);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
ip = new_inode(zfsvfs->z_sb);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (ip == NULL)
return (NULL);
VFS: Integrate zfs_znode_alloc() Under Linux the convention for filesystem specific data structure is to embed it along with the generic vfs data structure. This differs significantly from Solaris. Since we want to integrates as cleanly with the Linux VFS as possible. This changes modifies zfs_znode_alloc() to allocate a znode with an embedded inode for use with the generic VFS. This is done by calling iget_locked() which will allocate a new inode if needed by calling sb->alloc_inode(). This function allocates enough memory for a znode_t by returns a pointer to the inode structure for Linux's VFS. This function is also responsible for setting the callback znode->z_set_ops_inodes() which is used to register the correct handlers for the inode.
2011-01-05 21:16:12 +00:00
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
zp = ITOZ(ip);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
ASSERT(zp->z_dirlocks == NULL);
Add .zfs control directory Add support for the .zfs control directory. This was accomplished by leveraging as much of the existing ZFS infrastructure as posible and updating it for Linux as required. The bulk of the core functionality is now all there with the following limitations. *) The .zfs/snapshot directory automount support requires a 2.6.37 or newer kernel. The exception is RHEL6.2 which has backported the d_automount patches. *) Creating/destroying/renaming snapshots with mkdir/rmdir/mv in the .zfs/snapshot directory works as expected. However, this functionality is only available to root until zfs delegations are finished. * mkdir - create a snapshot * rmdir - destroy a snapshot * mv - rename a snapshot The following issues are known defeciences, but we expect them to be addressed by future commits. *) Add automount support for kernels older the 2.6.37. This should be possible using follow_link() which is what Linux did before. *) Accessing the .zfs/snapshot directory via NFS is not yet possible. The majority of the ground work for this is complete. However, finishing this work will require resolving some lingering integration issues with the Linux NFS kernel server. *) The .zfs/shares directory exists but no futher smb functionality has yet been implemented. Contributions-by: Rohan Puri <rohan.puri15@gmail.com> Contributiobs-by: Andrew Barnes <barnes333@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #173
2011-11-11 07:15:53 +00:00
ASSERT3P(zp->z_acl_cached, ==, NULL);
ASSERT3P(zp->z_xattr_cached, ==, NULL);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
zp->z_moved = 0;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_sa_hdl = NULL;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zp->z_unlinked = 0;
zp->z_atime_dirty = 0;
zp->z_mapcnt = 0;
zp->z_id = db->db_object;
zp->z_blksz = blksz;
zp->z_seq = 0x7A4653;
zp->z_sync_cnt = 0;
Add .zfs control directory Add support for the .zfs control directory. This was accomplished by leveraging as much of the existing ZFS infrastructure as posible and updating it for Linux as required. The bulk of the core functionality is now all there with the following limitations. *) The .zfs/snapshot directory automount support requires a 2.6.37 or newer kernel. The exception is RHEL6.2 which has backported the d_automount patches. *) Creating/destroying/renaming snapshots with mkdir/rmdir/mv in the .zfs/snapshot directory works as expected. However, this functionality is only available to root until zfs delegations are finished. * mkdir - create a snapshot * rmdir - destroy a snapshot * mv - rename a snapshot The following issues are known defeciences, but we expect them to be addressed by future commits. *) Add automount support for kernels older the 2.6.37. This should be possible using follow_link() which is what Linux did before. *) Accessing the .zfs/snapshot directory via NFS is not yet possible. The majority of the ground work for this is complete. However, finishing this work will require resolving some lingering integration issues with the Linux NFS kernel server. *) The .zfs/shares directory exists but no futher smb functionality has yet been implemented. Contributions-by: Rohan Puri <rohan.puri15@gmail.com> Contributiobs-by: Andrew Barnes <barnes333@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #173
2011-11-11 07:15:53 +00:00
zp->z_is_mapped = B_FALSE;
zp->z_is_ctldir = B_FALSE;
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
zp->z_is_stale = B_FALSE;
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
zp->z_range_lock.zr_size = &zp->z_size;
zp->z_range_lock.zr_blksz = &zp->z_blksz;
zp->z_range_lock.zr_max_blksz = &ZTOZSB(zp)->z_max_blksz;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_sa_init(zfsvfs, zp, db, obj_type, hdl);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MODE(zfsvfs), NULL, &mode, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GEN(zfsvfs), NULL, &tmp_gen, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zfsvfs), NULL,
&zp->z_size, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs), NULL, &links, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&zp->z_pflags, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_PARENT(zfsvfs), NULL,
VFS: Integrate zfs_znode_alloc() Under Linux the convention for filesystem specific data structure is to embed it along with the generic vfs data structure. This differs significantly from Solaris. Since we want to integrates as cleanly with the Linux VFS as possible. This changes modifies zfs_znode_alloc() to allocate a znode with an embedded inode for use with the generic VFS. This is done by calling iget_locked() which will allocate a new inode if needed by calling sb->alloc_inode(). This function allocates enough memory for a znode_t by returns a pointer to the inode structure for Linux's VFS. This function is also responsible for setting the callback znode->z_set_ops_inodes() which is used to register the correct handlers for the inode.
2011-01-05 21:16:12 +00:00
&parent, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_UID(zfsvfs), NULL, &z_uid, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GID(zfsvfs), NULL, &z_gid, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_ATIME(zfsvfs), NULL, &atime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zfsvfs), NULL, &mtime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs), NULL, &ctime, 16);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Use cstyle -cpP in `make cstyle` check Enable picky cstyle checks and resolve the new warnings. The vast majority of the changes needed were to handle minor issues with whitespace formatting. This patch contains no functional changes. Non-whitespace changes are as follows: * 8 times ; to { } in for/while loop * fix missing ; in cmd/zed/agents/zfs_diagnosis.c * comment (confim -> confirm) * change endline , to ; in cmd/zpool/zpool_main.c * a number of /* BEGIN CSTYLED */ /* END CSTYLED */ blocks * /* CSTYLED */ markers * change == 0 to ! * ulong to unsigned long in module/zfs/dsl_scan.c * rearrangement of module_param lines in module/zfs/metaslab.c * add { } block around statement after for_each_online_node Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Håkan Johansson <f96hajo@chalmers.se> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5465
2016-12-12 18:46:26 +00:00
if (sa_bulk_lookup(zp->z_sa_hdl, bulk, count) != 0 || tmp_gen == 0) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (hdl == NULL)
sa_handle_destroy(zp->z_sa_hdl);
Fix fail path in zfs_znode_alloc When sa_bulk_lookup() fails, unlock_new_inode() will spit out a WARNING. It will also recursive deadlock on ZFS_OBJ_HOLD_ENTER in zfs_zinactive(). Since we never call insert_inode_locked in fail path, I_NEW is never set, the inode is never hashed. So unlock_new_inode() can be safely remove it. We set z_sa_hdl to NULL in fail path so that iput path will stop at zfs_inactive() without entering zfs_zinactive(). This way we can avoid the deadlock and prevent double sa_handle_destroy(). Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #3899
2015-10-09 19:27:01 +00:00
zp->z_sa_hdl = NULL;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
goto error;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
VFS: Integrate zfs_znode_alloc() Under Linux the convention for filesystem specific data structure is to embed it along with the generic vfs data structure. This differs significantly from Solaris. Since we want to integrates as cleanly with the Linux VFS as possible. This changes modifies zfs_znode_alloc() to allocate a znode with an embedded inode for use with the generic VFS. This is done by calling iget_locked() which will allocate a new inode if needed by calling sb->alloc_inode(). This function allocates enough memory for a znode_t by returns a pointer to the inode structure for Linux's VFS. This function is also responsible for setting the callback znode->z_set_ops_inodes() which is used to register the correct handlers for the inode.
2011-01-05 21:16:12 +00:00
Refactor inode->i_mode management Refactor the code in such a way so that inode->i_mode is being set at the same time zp->z_mode is being changed. This has the effect of keeping both in sync without relying on zfs_inode_update. Reviewed-by: Richard Laager <rlaager@wiktel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Closes #5158
2016-09-27 21:08:52 +00:00
zp->z_mode = ip->i_mode = mode;
Kill znode->z_gen field This field is a duplicate of the inode->i_generation, so just kill it. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4538 Closes #4654
2016-04-18 19:08:53 +00:00
ip->i_generation = (uint32_t)tmp_gen;
Move assignment of i_blkbits field Currently i_blkbits is always set to SPA_MINBLOCKSHIFT every time zfs_inode_update_impl is called. Since this value never changes move its assignment to at inode creation time. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4906
2016-07-29 17:02:59 +00:00
ip->i_blkbits = SPA_MINBLOCKSHIFT;
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
set_nlink(ip, (uint32_t)links);
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
zfs_uid_write(ip, z_uid);
zfs_gid_write(ip, z_gid);
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
zfs_set_inode_flags(zp, ip);
Use local variable to read zp->z_mode When accessing the zp->z_mode through the SA bulk interface we expect that 64-bits are available to hold the result. However, on 32-bit platforms mode_t will only be 32-bits so we cannot pass it to SA_ADD_BULK_ATTR(). Instead a local uint64_t variable must be used and the result assigned to zp->z_mode. This went unnoticed on 32-bit little endian platforms because the bytes happen to end up in the correct 32-bits. But on big endian platforms like Sparc the zp->z_mode will always end up set to zero. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: marku89 <mar42@kola.li> Issue #1700
2014-01-07 22:16:46 +00:00
Fix unlinked file cannot do xattr operations Currently, doing things like fsetxattr(2) on an unlinked file will result in ENODATA. There's two places that cause this: zfs_dirent_lock and zfs_zget. The fix in zfs_dirent_lock is pretty straightforward. In zfs_zget though, we need it to not return error when the zp is unlinked. This is a pretty big change in behavior, but skimming through all the callers, I don't think this change would cause any problem. Also there's nothing preventing z_unlinked from being set after the z_lock mutex is dropped before but before zfs_zget returns anyway. The rest of the stuff is to make sure we don't log xattr stuff when owner is unlinked. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-10-13 00:30:46 +00:00
/* Cache the xattr parent id */
if (zp->z_pflags & ZFS_XATTR)
zp->z_xattr_parent = parent;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
ZFS_TIME_DECODE(&ip->i_atime, atime);
ZFS_TIME_DECODE(&ip->i_mtime, mtime);
ZFS_TIME_DECODE(&ip->i_ctime, ctime);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
ip->i_ino = obj;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
zfs_inode_update(zp);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_inode_set_ops(zfsvfs, ip);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
/*
* The only way insert_inode_locked() can fail is if the ip->i_ino
* number is already hashed for this super block. This can never
* happen because the inode numbers map 1:1 with the object numbers.
*
* The one exception is rolling back a mounted file system, but in
* this case all the active inode are unhashed during the rollback.
*/
VERIFY3S(insert_inode_locked(ip), ==, 0);
Call d_instantiate before unlocking inode Under Linux a dentry referencing an inode must be instantiated before the inode is unlocked. To accomplish this without overly modifing the core ZFS code the dentry it passed via the vattr_t. There are cases such as replay when a dentry is not available. In which case it is obviously not initialized at inode creation time, if a dentry is needed it will be spliced as when required via d_lookup().
2011-03-30 06:04:39 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_enter(&zfsvfs->z_znodes_lock);
list_insert_tail(&zfsvfs->z_all_znodes, zp);
zfsvfs->z_nr_znodes++;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
membar_producer();
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_exit(&zfsvfs->z_znodes_lock);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
unlock_new_inode(ip);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (zp);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
error:
iput(ip);
cstyle: Resolve C style issues The vast majority of these changes are in Linux specific code. They are the result of not having an automated style checker to validate the code when it was originally written. Others were caused when the common code was slightly adjusted for Linux. This patch contains no functional changes. It only refreshes the code to conform to style guide. Everyone submitting patches for inclusion upstream should now run 'make checkstyle' and resolve any warning prior to opening a pull request. The automated builders have been updated to fail a build if when 'make checkstyle' detects an issue. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1821
2013-11-01 19:26:11 +00:00
return (NULL);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Fix zil_commit() NULL dereference Update the current code to ensure inodes are never dirtied if they are part of a read-only file system or snapshot. If they do somehow get dirtied an attempt will make made to write them to disk. In the case of snapshots, which don't have a ZIL, this will result in a NULL dereference in zil_commit(). Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2405
2014-07-15 20:29:57 +00:00
/*
* Safely mark an inode dirty. Inodes which are part of a read-only
* file system or snapshot may not be dirtied.
*/
void
zfs_mark_inode_dirty(struct inode *ip)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ITOZSB(ip);
Fix zil_commit() NULL dereference Update the current code to ensure inodes are never dirtied if they are part of a read-only file system or snapshot. If they do somehow get dirtied an attempt will make made to write them to disk. In the case of snapshots, which don't have a ZIL, this will result in a NULL dereference in zil_commit(). Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2405
2014-07-15 20:29:57 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if (zfs_is_readonly(zfsvfs) || dmu_objset_is_snapshot(zfsvfs->z_os))
Fix zil_commit() NULL dereference Update the current code to ensure inodes are never dirtied if they are part of a read-only file system or snapshot. If they do somehow get dirtied an attempt will make made to write them to disk. In the case of snapshots, which don't have a ZIL, this will result in a NULL dereference in zil_commit(). Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2405
2014-07-15 20:29:57 +00:00
return;
mark_inode_dirty(ip);
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
static uint64_t empty_xattr;
static uint64_t pad[4];
static zfs_acl_phys_t acl_phys;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* Create a new DMU object to hold a zfs znode.
*
* IN: dzp - parent directory for new znode
* vap - file attributes for new znode
* tx - dmu transaction id for zap operations
* cr - credentials of caller
* flag - flags:
* IS_ROOT_NODE - new object will be root
* IS_XATTR - new object is an attribute
* bonuslen - length of bonus buffer
* setaclp - File/Dir initial ACL
* fuidp - Tracks fuid allocation.
*
* OUT: zpp - allocated znode
*
*/
void
zfs_mknode(znode_t *dzp, vattr_t *vap, dmu_tx_t *tx, cred_t *cr,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint_t flag, znode_t **zpp, zfs_acl_ids_t *acl_ids)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t crtime[2], atime[2], mtime[2], ctime[2];
uint64_t mode, size, links, parent, pflags;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
uint64_t dzp_pflags = 0;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t rdev = 0;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(dzp);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_buf_t *db;
Linux 4.18 compat: inode timespec -> timespec64 Commit torvalds/linux@95582b0 changes the inode i_atime, i_mtime, and i_ctime members form timespec's to timespec64's to make them 2038 safe. As part of this change the current_time() function was also updated to return the timespec64 type. Resolve this issue by introducing a new inode_timespec_t type which is defined to match the timespec type used by the inode. It should be used when working with inode timestamps to ensure matching types. The timestruc_t type under Illumos was used in a similar fashion but was specified to always be a timespec_t. Rather than incorrectly define this type all timespec_t types have been replaced by the new inode_timespec_t type. Finally, the kernel and user space 'sys/time.h' headers were aligned with each other. They define as appropriate for the context several constants as macros and include static inline implementation of gethrestime(), gethrestime_sec(), and gethrtime(). Reviewed-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #7643 Backported-by: Richard Yao <ryao@gentoo.org>
2018-08-12 22:22:03 +00:00
inode_timespec_t now;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
uint64_t gen, obj;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
int bonuslen;
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
int dnodesize;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sa_handle_t *sa_hdl;
dmu_object_type_t obj_type;
Conserve stack in zfs_mkdir() Move the sa_attrs array from the stack to the heap to minimize stack space usage.
2011-01-07 21:47:47 +00:00
sa_bulk_attr_t *sa_attrs;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
int cnt = 0;
zfs_acl_locator_cb_t locate = { 0 };
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
znode_hold_t *zh;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if (zfsvfs->z_replay) {
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
obj = vap->va_nodeid;
now = vap->va_ctime; /* see zfs_replay_create() */
gen = vap->va_nblocks; /* ditto */
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
dnodesize = vap->va_fsid; /* ditto */
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
obj = 0;
gethrestime(&now);
gen = dmu_tx_get_txg(tx);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
dnodesize = dmu_objset_dnodesize(zfsvfs->z_os);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
if (dnodesize == 0)
dnodesize = DNODE_MIN_SIZE;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
obj_type = zfsvfs->z_use_sa ? DMU_OT_SA : DMU_OT_ZNODE;
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
bonuslen = (obj_type == DMU_OT_SA) ?
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
DN_BONUS_SIZE(dnodesize) : ZFS_OLD_ZNODE_PHYS_SIZE;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* Create a new DMU object.
*/
/*
* There's currently no mechanism for pre-reading the blocks that will
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
* be needed to allocate a new object, so we accept the small chance
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
* that there will be an i/o error and we will fail one of the
* assertions below.
*/
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (S_ISDIR(vap->va_mode)) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if (zfsvfs->z_replay) {
VERIFY0(zap_create_claim_norm_dnsize(zfsvfs->z_os, obj,
zfsvfs->z_norm, DMU_OT_DIRECTORY_CONTENTS,
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
obj_type, bonuslen, dnodesize, tx));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
obj = zap_create_norm_dnsize(zfsvfs->z_os,
zfsvfs->z_norm, DMU_OT_DIRECTORY_CONTENTS,
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
obj_type, bonuslen, dnodesize, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
} else {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if (zfsvfs->z_replay) {
VERIFY0(dmu_object_claim_dnsize(zfsvfs->z_os, obj,
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
DMU_OT_PLAIN_FILE_CONTENTS, 0,
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
obj_type, bonuslen, dnodesize, tx));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
obj = dmu_object_alloc_dnsize(zfsvfs->z_os,
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
DMU_OT_PLAIN_FILE_CONTENTS, 0,
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
obj_type, bonuslen, dnodesize, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zh = zfs_znode_hold_enter(zfsvfs, obj);
Fix dnode allocation race When performing concurrent object allocations using the new multi-threaded allocator and large dnodes it's possible to allocate overlapping large dnodes. This case should have been handled by detecting an error returned by dnode_hold_impl(). But that logic only checked the returned dnp was not-NULL, and the dnp variable was not reset to NULL when retrying. Resolve this issue by properly checking the return value of dnode_hold_impl(). Additionally, it was possible that dnode_hold_impl() would misreport a dnode as free when it was in fact in use. This could occurs for two reasons: * The per-slot zrl_lock must be held over the entire critical section which includes the alloc/free until the new dnode is assigned to children_dnodes. Additionally, all of the zrl_lock's in the range must be held to protect moving dnodes. * The dn->dn_ot_type cannot be solely relied upon to check the type. When allocating a new dnode its type will be DMU_OT_NONE after dnode_create(). Only latter when dnode_allocate() is called will it transition to the new type. This means there's a window when allocating where it can mistaken for a free dnode. Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Tony Hutter <hutter2@llnl.gov> Reviewed-by: Olaf Faaland <faaland1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #6414 Closes #6439
2017-08-08 15:38:53 +00:00
VERIFY0(sa_buf_hold(zfsvfs->z_os, obj, NULL, &db));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* If this is the root, fix up the half-initialized parent pointer
* to reference the just-allocated physical data area.
*/
if (flag & IS_ROOT_NODE) {
dzp->z_id = obj;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
} else {
dzp_pflags = dzp->z_pflags;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
/*
* If parent is an xattr, so am I.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (dzp_pflags & ZFS_XATTR) {
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
flag |= IS_XATTR;
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if (zfsvfs->z_use_fuids)
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
pflags = ZFS_ARCHIVE | ZFS_AV_MODIFIED;
else
pflags = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (S_ISDIR(vap->va_mode)) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
size = 2; /* contents ("." and "..") */
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
links = 2;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
} else {
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
size = 0;
Add support for O_TMPFILE Linux 3.11 add O_TMPFILE to open(2), which allow creating an unlinked file on supported filesystem. It's basically doing open(2) and unlink(2) atomically. The filesystem support is added through i_op->tmpfile. We basically copy the create operation except we get rid of the link and name related stuff and add the new node to unlinked set. We also add support for linkat(2) to link tmpfile. However, since all previous file operation will skip ZIL, we force a txg_wait_synced to make sure we are sync safe. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-01-26 20:29:46 +00:00
links = (flag & IS_TMPFILE) ? 0 : 1;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Don't store rdev in SA for FIFOs and sockets Update the handling of named pipes and sockets to be consistent with other platforms with regard to the rdev attribute. While all ZFS ipmlementations store the rdev for device files in a system attribute (SA), this is not the case for FIFOs and sockets. Indeed, Linux always passes rdev=0 to mknod() for FIFOs and sockets, so the value is not needed. Add an ASSERT that rdev==0 for FIFOs and sockets to detect if the expected behavior ever changes. Closes #216
2011-05-09 19:31:56 +00:00
if (S_ISBLK(vap->va_mode) || S_ISCHR(vap->va_mode))
Remove rdev packing Remove custom code to pack/unpack dev_t's. Under Linux all dev_t's are an unsigned 32-bit value even on 64-bit platforms. The lower 20 bits are used for the minor number and the upper 12 for the major number. This means if your importing a pool from Solaris you may get strange major/minor numbers. But it doesn't really matter because even if we add compatibility code to translate the encoded Solaris major/minor they won't do you any good under Linux. You will still need to recreate the dev_t with a major/minor which maps to reserved major numbers used under Linux. Dropping this code also resolves 32-bit builds by removing the offending 32-bit compatibility code.
2011-02-23 23:13:03 +00:00
rdev = vap->va_rdev;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
parent = dzp->z_id;
mode = acl_ids->z_mode;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (flag & IS_XATTR)
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
pflags |= ZFS_XATTR;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
/*
* No execs denied will be deterimed when zfs_mode_compute() is called.
*/
pflags |= acl_ids->z_aclp->z_hints &
(ZFS_ACL_TRIVIAL|ZFS_INHERIT_ACE|ZFS_ACL_AUTO_INHERIT|
ZFS_ACL_DEFAULTED|ZFS_ACL_PROTECTED);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&now, crtime);
ZFS_TIME_ENCODE(&now, ctime);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (vap->va_mask & ATTR_ATIME) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&vap->va_atime, atime);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&now, atime);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (vap->va_mask & ATTR_MTIME) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&vap->va_mtime, mtime);
} else {
ZFS_TIME_ENCODE(&now, mtime);
}
/* Now add in all of the "SA" attributes */
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
VERIFY(0 == sa_handle_get_from_db(zfsvfs->z_os, db, NULL, SA_HDL_SHARED,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&sa_hdl));
/*
* Setup the array of attributes to be replaced/set on the new file
*
* order for DMU_OT_ZNODE is critical since it needs to be constructed
* in the old znode_phys_t format. Don't change this ordering
*/
Change KM_PUSHPAGE -> KM_SLEEP By marking DMU transaction processing contexts with PF_FSTRANS we can revert the KM_PUSHPAGE -> KM_SLEEP changes. This brings us back in line with upstream. In some cases this means simply swapping the flags back. For others fnvlist_alloc() was replaced by nvlist_alloc(..., KM_PUSHPAGE) and must be reverted back to fnvlist_alloc() which assumes KM_SLEEP. The one place KM_PUSHPAGE is kept is when allocating ARC buffers which allows us to dip in to reserved memory. This is again the same as upstream. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-11-21 00:09:39 +00:00
sa_attrs = kmem_alloc(sizeof (sa_bulk_attr_t) * ZPL_END, KM_SLEEP);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (obj_type == DMU_OT_ZNODE) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ATIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &atime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &mtime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &ctime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CRTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &crtime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GEN(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &gen, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MODE(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &mode, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_SIZE(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &size, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PARENT(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &parent, 8);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MODE(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &mode, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_SIZE(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &size, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GEN(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &gen, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_UID(zfsvfs),
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
NULL, &acl_ids->z_fuid, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GID(zfsvfs),
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
NULL, &acl_ids->z_fgid, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PARENT(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &parent, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_FLAGS(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &pflags, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ATIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &atime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &mtime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &ctime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CRTIME(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &crtime, 16);
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_LINKS(zfsvfs), NULL, &links, 8);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (obj_type == DMU_OT_ZNODE) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_XATTR(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&empty_xattr, 8);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (obj_type == DMU_OT_ZNODE ||
Don't store rdev in SA for FIFOs and sockets Update the handling of named pipes and sockets to be consistent with other platforms with regard to the rdev attribute. While all ZFS ipmlementations store the rdev for device files in a system attribute (SA), this is not the case for FIFOs and sockets. Indeed, Linux always passes rdev=0 to mknod() for FIFOs and sockets, so the value is not needed. Add an ASSERT that rdev==0 for FIFOs and sockets to detect if the expected behavior ever changes. Closes #216
2011-05-09 19:31:56 +00:00
(S_ISBLK(vap->va_mode) || S_ISCHR(vap->va_mode))) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_RDEV(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &rdev, 8);
}
if (obj_type == DMU_OT_ZNODE) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_FLAGS(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &pflags, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_UID(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&acl_ids->z_fuid, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GID(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&acl_ids->z_fgid, 8);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PAD(zfsvfs), NULL, pad,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sizeof (uint64_t) * 4);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ZNODE_ACL(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&acl_phys, sizeof (zfs_acl_phys_t));
} else if (acl_ids->z_aclp->z_version >= ZFS_ACL_VERSION_FUID) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_DACL_COUNT(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&acl_ids->z_aclp->z_acl_count, 8);
locate.cb_aclp = acl_ids->z_aclp;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_DACL_ACES(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zfs_acl_data_locator, &locate,
acl_ids->z_aclp->z_acl_bytes);
mode = zfs_mode_compute(mode, acl_ids->z_aclp, &pflags,
acl_ids->z_fuid, acl_ids->z_fgid);
}
VERIFY(sa_replace_all_by_template(sa_hdl, sa_attrs, cnt, tx) == 0);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (!(flag & IS_ROOT_NODE)) {
Retry zfs_znode_alloc() in zfs_mknode() For historical reasons zfs_mknode() was written such that it could never fail. This poses a problem for Linux since zfs_znode_alloc() could potentually failure due to low memory. Handle this gracefully by retrying zfs_znode_alloc() until it succeeds, direct reclaim will eventually be able to allocate memory. Reviewed-by: George Melikov <mail@gmelikov.ru> Reviewed-by: loli10K <ezomori.nozomu@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5535 Closes #5908
2017-03-24 01:26:50 +00:00
/*
* The call to zfs_znode_alloc() may fail if memory is low
* via the call path: alloc_inode() -> inode_init_always() ->
* security_inode_alloc() -> inode_alloc_security(). Since
* the existing code is written such that zfs_mknode() can
* not fail retry until sufficient memory has been reclaimed.
*/
do {
*zpp = zfs_znode_alloc(zfsvfs, db, 0, obj_type, obj,
sa_hdl);
} while (*zpp == NULL);
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
VERIFY(*zpp != NULL);
VERIFY(dzp != NULL);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
/*
* If we are creating the root node, the "parent" we
* passed in is the znode for the root.
*/
*zpp = dzp;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
(*zpp)->z_sa_hdl = sa_hdl;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
(*zpp)->z_pflags = pflags;
Refactor inode->i_mode management Refactor the code in such a way so that inode->i_mode is being set at the same time zp->z_mode is being changed. This has the effect of keeping both in sync without relying on zfs_inode_update. Reviewed-by: Richard Laager <rlaager@wiktel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Closes #5158
2016-09-27 21:08:52 +00:00
(*zpp)->z_mode = ZTOI(*zpp)->i_mode = mode;
Implement large_dnode pool feature Justification ------------- This feature adds support for variable length dnodes. Our motivation is to eliminate the overhead associated with using spill blocks. Spill blocks are used to store system attribute data (i.e. file metadata) that does not fit in the dnode's bonus buffer. By allowing a larger bonus buffer area the use of a spill block can be avoided. Spill blocks potentially incur an additional read I/O for every dnode in a dnode block. As a worst case example, reading 32 dnodes from a 16k dnode block and all of the spill blocks could issue 33 separate reads. Now suppose those dnodes have size 1024 and therefore don't need spill blocks. Then the worst case number of blocks read is reduced to from 33 to two--one per dnode block. In practice spill blocks may tend to be co-located on disk with the dnode blocks so the reduction in I/O would not be this drastic. In a badly fragmented pool, however, the improvement could be significant. ZFS-on-Linux systems that make heavy use of extended attributes would benefit from this feature. In particular, ZFS-on-Linux supports the xattr=sa dataset property which allows file extended attribute data to be stored in the dnode bonus buffer as an alternative to the traditional directory-based format. Workloads such as SELinux and the Lustre distributed filesystem often store enough xattr data to force spill bocks when xattr=sa is in effect. Large dnodes may therefore provide a performance benefit to such systems. Other use cases that may benefit from this feature include files with large ACLs and symbolic links with long target names. Furthermore, this feature may be desirable on other platforms in case future applications or features are developed that could make use of a larger bonus buffer area. Implementation -------------- The size of a dnode may be a multiple of 512 bytes up to the size of a dnode block (currently 16384 bytes). A dn_extra_slots field was added to the current on-disk dnode_phys_t structure to describe the size of the physical dnode on disk. The 8 bits for this field were taken from the zero filled dn_pad2 field. The field represents how many "extra" dnode_phys_t slots a dnode consumes in its dnode block. This convention results in a value of 0 for 512 byte dnodes which preserves on-disk format compatibility with older software. Similarly, the in-memory dnode_t structure has a new dn_num_slots field to represent the total number of dnode_phys_t slots consumed on disk. Thus dn->dn_num_slots is 1 greater than the corresponding dnp->dn_extra_slots. This difference in convention was adopted because, unlike on-disk structures, backward compatibility is not a concern for in-memory objects, so we used a more natural way to represent size for a dnode_t. The default size for newly created dnodes is determined by the value of a new "dnodesize" dataset property. By default the property is set to "legacy" which is compatible with older software. Setting the property to "auto" will allow the filesystem to choose the most suitable dnode size. Currently this just sets the default dnode size to 1k, but future code improvements could dynamically choose a size based on observed workload patterns. Dnodes of varying sizes can coexist within the same dataset and even within the same dnode block. For example, to enable automatically-sized dnodes, run # zfs set dnodesize=auto tank/fish The user can also specify literal values for the dnodesize property. These are currently limited to powers of two from 1k to 16k. The power-of-2 limitation is only for simplicity of the user interface. Internally the implementation can handle any multiple of 512 up to 16k, and consumers of the DMU API can specify any legal dnode value. The size of a new dnode is determined at object allocation time and stored as a new field in the znode in-memory structure. New DMU interfaces are added to allow the consumer to specify the dnode size that a newly allocated object should use. Existing interfaces are unchanged to avoid having to update every call site and to preserve compatibility with external consumers such as Lustre. The new interfaces names are given below. The versions of these functions that don't take a dnodesize parameter now just call the _dnsize() versions with a dnodesize of 0, which means use the legacy dnode size. New DMU interfaces: dmu_object_alloc_dnsize() dmu_object_claim_dnsize() dmu_object_reclaim_dnsize() New ZAP interfaces: zap_create_dnsize() zap_create_norm_dnsize() zap_create_flags_dnsize() zap_create_claim_norm_dnsize() zap_create_link_dnsize() The constant DN_MAX_BONUSLEN is renamed to DN_OLD_MAX_BONUSLEN. The spa_maxdnodesize() function should be used to determine the maximum bonus length for a pool. These are a few noteworthy changes to key functions: * The prototype for dnode_hold_impl() now takes a "slots" parameter. When the DNODE_MUST_BE_FREE flag is set, this parameter is used to ensure the hole at the specified object offset is large enough to hold the dnode being created. The slots parameter is also used to ensure a dnode does not span multiple dnode blocks. In both of these cases, if a failure occurs, ENOSPC is returned. Keep in mind, these failure cases are only possible when using DNODE_MUST_BE_FREE. If the DNODE_MUST_BE_ALLOCATED flag is set, "slots" must be 0. dnode_hold_impl() will check if the requested dnode is already consumed as an extra dnode slot by an large dnode, in which case it returns ENOENT. * The function dmu_object_alloc() advances to the next dnode block if dnode_hold_impl() returns an error for a requested object. This is because the beginning of the next dnode block is the only location it can safely assume to either be a hole or a valid starting point for a dnode. * dnode_next_offset_level() and other functions that iterate through dnode blocks may no longer use a simple array indexing scheme. These now use the current dnode's dn_num_slots field to advance to the next dnode in the block. This is to ensure we properly skip the current dnode's bonus area and don't interpret it as a valid dnode. zdb --- The zdb command was updated to display a dnode's size under the "dnsize" column when the object is dumped. For ZIL create log records, zdb will now display the slot count for the object. ztest ----- Ztest chooses a random dnodesize for every newly created object. The random distribution is more heavily weighted toward small dnodes to better simulate real-world datasets. Unused bonus buffer space is filled with non-zero values computed from the object number, dataset id, offset, and generation number. This helps ensure that the dnode traversal code properly skips the interior regions of large dnodes, and that these interior regions are not overwritten by data belonging to other dnodes. A new test visits each object in a dataset. It verifies that the actual dnode size matches what was stored in the ztest block tag when it was created. It also verifies that the unused bonus buffer space is filled with the expected data patterns. ZFS Test Suite -------------- Added six new large dnode-specific tests, and integrated the dnodesize property into existing tests for zfs allow and send/recv. Send/Receive ------------ ZFS send streams for datasets containing large dnodes cannot be received on pools that don't support the large_dnode feature. A send stream with large dnodes sets a DMU_BACKUP_FEATURE_LARGE_DNODE flag which will be unrecognized by an incompatible receiving pool so that the zfs receive will fail gracefully. While not implemented here, it may be possible to generate a backward-compatible send stream from a dataset containing large dnodes. The implementation may be tricky, however, because the send object record for a large dnode would need to be resized to a 512 byte dnode, possibly kicking in a spill block in the process. This means we would need to construct a new SA layout and possibly register it in the SA layout object. The SA layout is normally just sent as an ordinary object record. But if we are constructing new layouts while generating the send stream we'd have to build the SA layout object dynamically and send it at the end of the stream. For sending and receiving between pools that do support large dnodes, the drr_object send record type is extended with a new field to store the dnode slot count. This field was repurposed from unused padding in the structure. ZIL Replay ---------- The dnode slot count is stored in the uppermost 8 bits of the lr_foid field. The bits were unused as the object id is currently capped at 48 bits. Resizing Dnodes --------------- It should be possible to resize a dnode when it is dirtied if the current dnodesize dataset property differs from the dnode's size, but this functionality is not currently implemented. Clearly a dnode can only grow if there are sufficient contiguous unused slots in the dnode block, but it should always be possible to shrink a dnode. Growing dnodes may be useful to reduce fragmentation in a pool with many spill blocks in use. Shrinking dnodes may be useful to allow sending a dataset to a pool that doesn't support the large_dnode feature. Feature Reference Counting -------------------------- The reference count for the large_dnode pool feature tracks the number of datasets that have ever contained a dnode of size larger than 512 bytes. The first time a large dnode is created in a dataset the dataset is converted to an extensible dataset. This is a one-way operation and the only way to decrement the feature count is to destroy the dataset, even if the dataset no longer contains any large dnodes. The complexity of reference counting on a per-dnode basis was too high, so we chose to track it on a per-dataset basis similarly to the large_block feature. Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3542
2016-03-17 01:25:34 +00:00
(*zpp)->z_dnodesize = dnodesize;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (obj_type == DMU_OT_ZNODE ||
acl_ids->z_aclp->z_version < ZFS_ACL_VERSION_FUID) {
Illumos 4370, 4371 4370 avoid transmitting holes during zfs send 4371 DMU code clean up Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net> Approved by: Garrett D'Amore <garrett@damore.org>a References: https://www.illumos.org/issues/4370 https://www.illumos.org/issues/4371 https://github.com/illumos/illumos-gate/commit/43466aa Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2529
2013-12-09 18:37:51 +00:00
VERIFY0(zfs_aclset_common(*zpp, acl_ids->z_aclp, cr, tx));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
}
cstyle: Resolve C style issues The vast majority of these changes are in Linux specific code. They are the result of not having an automated style checker to validate the code when it was originally written. Others were caused when the common code was slightly adjusted for Linux. This patch contains no functional changes. It only refreshes the code to conform to style guide. Everyone submitting patches for inclusion upstream should now run 'make checkstyle' and resolve any warning prior to opening a pull request. The automated builders have been updated to fail a build if when 'make checkstyle' detects an issue. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1821
2013-11-01 19:26:11 +00:00
kmem_free(sa_attrs, sizeof (sa_bulk_attr_t) * ZPL_END);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
/*
Illumos #3742 3742 zfs comments need cleaner, more consistent style Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Christopher Siden <christopher.siden@delphix.com> References: https://www.illumos.org/issues/3742 illumos/illumos-gate@f7170741490edba9d1d9c697c177c887172bc741 Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. The change to zfs_vfsops.c was dropped because it involves zfs_mount_label_policy, which does not exist in the Linux port.
2013-06-11 17:12:34 +00:00
* Update in-core attributes. It is assumed the caller will be doing an
* sa_bulk_update to push the changes out.
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
*/
void
zfs_xvattr_set(znode_t *zp, xvattr_t *xvap, dmu_tx_t *tx)
{
xoptattr_t *xoap;
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
boolean_t update_inode = B_FALSE;
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
xoap = xva_getxoptattr(xvap);
ASSERT(xoap);
if (XVA_ISSET_REQ(xvap, XAT_CREATETIME)) {
uint64_t times[2];
ZFS_TIME_ENCODE(&xoap->xoa_createtime, times);
(void) sa_update(zp->z_sa_hdl, SA_ZPL_CRTIME(ZTOZSB(zp)),
&times, sizeof (times), tx);
XVA_SET_RTN(xvap, XAT_CREATETIME);
}
if (XVA_ISSET_REQ(xvap, XAT_READONLY)) {
ZFS_ATTR_SET(zp, ZFS_READONLY, xoap->xoa_readonly,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_READONLY);
}
if (XVA_ISSET_REQ(xvap, XAT_HIDDEN)) {
ZFS_ATTR_SET(zp, ZFS_HIDDEN, xoap->xoa_hidden,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_HIDDEN);
}
if (XVA_ISSET_REQ(xvap, XAT_SYSTEM)) {
ZFS_ATTR_SET(zp, ZFS_SYSTEM, xoap->xoa_system,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_SYSTEM);
}
if (XVA_ISSET_REQ(xvap, XAT_ARCHIVE)) {
ZFS_ATTR_SET(zp, ZFS_ARCHIVE, xoap->xoa_archive,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_ARCHIVE);
}
if (XVA_ISSET_REQ(xvap, XAT_IMMUTABLE)) {
ZFS_ATTR_SET(zp, ZFS_IMMUTABLE, xoap->xoa_immutable,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_IMMUTABLE);
Refactor updating of immutable/appendonly flags Move the synchronization of inode/znode i_flgas/pflags into the respective internal zfs function. This is mostly mechanical work and shouldn't introduce any functional changes. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #5223
2016-10-05 21:47:29 +00:00
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
update_inode = B_TRUE;
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
}
if (XVA_ISSET_REQ(xvap, XAT_NOUNLINK)) {
ZFS_ATTR_SET(zp, ZFS_NOUNLINK, xoap->xoa_nounlink,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_NOUNLINK);
}
if (XVA_ISSET_REQ(xvap, XAT_APPENDONLY)) {
ZFS_ATTR_SET(zp, ZFS_APPENDONLY, xoap->xoa_appendonly,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_APPENDONLY);
Refactor updating of immutable/appendonly flags Move the synchronization of inode/znode i_flgas/pflags into the respective internal zfs function. This is mostly mechanical work and shouldn't introduce any functional changes. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #5223
2016-10-05 21:47:29 +00:00
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
update_inode = B_TRUE;
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
}
if (XVA_ISSET_REQ(xvap, XAT_NODUMP)) {
ZFS_ATTR_SET(zp, ZFS_NODUMP, xoap->xoa_nodump,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_NODUMP);
}
if (XVA_ISSET_REQ(xvap, XAT_OPAQUE)) {
ZFS_ATTR_SET(zp, ZFS_OPAQUE, xoap->xoa_opaque,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_OPAQUE);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_QUARANTINED)) {
ZFS_ATTR_SET(zp, ZFS_AV_QUARANTINED,
xoap->xoa_av_quarantined, zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_AV_QUARANTINED);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_MODIFIED)) {
ZFS_ATTR_SET(zp, ZFS_AV_MODIFIED, xoap->xoa_av_modified,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_AV_MODIFIED);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_SCANSTAMP)) {
zfs_sa_set_scanstamp(zp, xvap, tx);
XVA_SET_RTN(xvap, XAT_AV_SCANSTAMP);
}
if (XVA_ISSET_REQ(xvap, XAT_REPARSE)) {
ZFS_ATTR_SET(zp, ZFS_REPARSE, xoap->xoa_reparse,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_REPARSE);
}
if (XVA_ISSET_REQ(xvap, XAT_OFFLINE)) {
ZFS_ATTR_SET(zp, ZFS_OFFLINE, xoap->xoa_offline,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_OFFLINE);
}
if (XVA_ISSET_REQ(xvap, XAT_SPARSE)) {
ZFS_ATTR_SET(zp, ZFS_SPARSE, xoap->xoa_sparse,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_SPARSE);
}
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
if (update_inode)
zfs_set_inode_flags(zp, ZTOI(zp));
Drop HAVE_XVATTR macros When I began work on the Posix layer it immediately became clear to me that to integrate cleanly with the Linux VFS certain Solaris specific things would have to go. One of these things was to elimate as many Solaris specific types from the ZPL layer as possible. They would be replaced with their Linux equivalents. This would not only be good for performance, but for the general readability and health of the code. The Solaris and Linux VFS are different beasts and should be treated as such. Most of the code remains common for constructing transactions and such, but there are subtle and important differenced which need to be repsected. This policy went quite for for certain types such as the vnode_t, and it initially seemed to be working out well for the vattr_t. There was a relatively small amount of related xvattr_t code I was forced to comment out with HAVE_XVATTR. But it didn't look that hard to come back soon and replace it all with a native Linux type. However, after going doing this path with xvattr some distance it clear that this code was woven in the ZPL more deeply than I thought. In particular its hooks went very deep in to the ZPL replay code and replacing it would not be as easy as I originally thought. Rather than continue persuing replacing and removing this code I've taken a step back and reevaluted things. This commit reverts many of my previous commits which removed xvattr related code. It restores much of the code to its original upstream state and now relies on improved xvattr_t support in the zfs package itself. The result of this is that much of the code which I had commented out, which accidentally broke things like replay, is now back in place and working. However, there may be a small performance impact for getattr/setattr operations because they now require a translation from native Linux to Solaris types. For now that's a price I'm willing to pay. Once everything is completely functional we can revisting the issue of removing the vattr_t/xvattr_t types. Closes #111
2011-03-01 20:24:09 +00:00
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
int
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_zget(zfsvfs_t *zfsvfs, uint64_t obj_num, znode_t **zpp)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
dmu_object_info_t doi;
dmu_buf_t *db;
znode_t *zp;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
znode_hold_t *zh;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
int err;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sa_handle_t *hdl;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*zpp = NULL;
Fix deadlock in zfs_zget() zfsonlinux/zfs#180 occurred because of a race between inode eviction and zfs_zget(). zfsonlinux/zfs@36df284 tried to address it by making a call to the VFS to learn whether an inode is being evicted. If it was being evicted the operation was retried after dropping and reacquiring the relevant resources. Unfortunately, this introduced another deadlock. INFO: task kworker/u24:6:891 blocked for more than 120 seconds. Tainted: P O 3.13.6 #1 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. kworker/u24:6 D ffff88107fcd2e80 0 891 2 0x00000000 Workqueue: writeback bdi_writeback_workfn (flush-zfs-5) ffff8810370ff950 0000000000000002 ffff88103853d940 0000000000012e80 ffff8810370fffd8 0000000000012e80 ffff88103853d940 ffff880f5c8be098 ffff88107ffb6950 ffff8810370ff980 ffff88103a9a5b78 0000000000000000 Call Trace: [<ffffffff813dd1d4>] schedule+0x24/0x70 [<ffffffff8115fc09>] __wait_on_freeing_inode+0x99/0xc0 [<ffffffff8115fdd8>] find_inode_fast+0x78/0xb0 [<ffffffff811608c5>] ilookup+0x65/0xd0 [<ffffffffa035c5ab>] zfs_zget+0xdb/0x260 [zfs] [<ffffffffa03589d6>] zfs_get_data+0x46/0x340 [zfs] [<ffffffffa035fee1>] zil_add_block+0xa31/0xc00 [zfs] [<ffffffffa0360642>] zil_commit+0x12/0x20 [zfs] [<ffffffffa036a6e4>] zpl_putpage+0x174/0x840 [zfs] [<ffffffff811071ec>] do_writepages+0x1c/0x40 [<ffffffff8116df2b>] __writeback_single_inode+0x3b/0x2b0 [<ffffffff8116ecf7>] writeback_sb_inodes+0x247/0x420 [<ffffffff8116f5f3>] wb_writeback+0xe3/0x320 [<ffffffff81170b8e>] bdi_writeback_workfn+0xfe/0x490 [<ffffffff8106072c>] process_one_work+0x16c/0x490 [<ffffffff810613f3>] worker_thread+0x113/0x390 [<ffffffff81066edf>] kthread+0xdf/0x100 This patch implements the original fix in a slightly different manner in order to avoid both deadlocks. Instead of relying on a call to ilookup() which can block in __wait_on_freeing_inode() the return value from igrab() is used. This gives us the information that ilookup() provided without the risk of a deadlock. Alternately, this race could be closed by registering an sops->drop_inode() callback. The callback would need to detect the active SA hold thereby informing the VFS that this inode should not be evicted. Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #180
2014-03-25 19:41:18 +00:00
again:
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zh = zfs_znode_hold_enter(zfsvfs, obj_num);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
err = sa_buf_hold(zfsvfs->z_os, obj_num, NULL, &db);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (err) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (err);
}
dmu_object_info_from_db(db, &doi);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (doi.doi_bonus_type != DMU_OT_SA &&
(doi.doi_bonus_type != DMU_OT_ZNODE ||
(doi.doi_bonus_type == DMU_OT_ZNODE &&
doi.doi_bonus_size < sizeof (znode_phys_t)))) {
sa_buf_rele(db, NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
return (SET_ERROR(EINVAL));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
hdl = dmu_buf_get_user(db);
if (hdl != NULL) {
Fixed a use-after-free bug in zfs_zget(). Fixed a bug where zfs_zget could access a stale znode pointer when the inode had already been removed from the inode cache via iput -> iput_final -> ... -> zfs_zinactive but the corresponding SA handle was still alive. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #180
2011-04-14 20:07:24 +00:00
zp = sa_get_userdata(hdl);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Revert "Fixed a use-after-free bug in zfs_zget()." This reverts commit 36df284366caa77cb40083d2e6bcce02274e2f05. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-04-03 21:26:16 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
* Since "SA" does immediate eviction we
* should never find a sa handle that doesn't
* know about the znode.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ASSERT3P(zp, !=, NULL);
mutex_enter(&zp->z_lock);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
ASSERT3U(zp->z_id, ==, obj_num);
Fix unlinked file cannot do xattr operations Currently, doing things like fsetxattr(2) on an unlinked file will result in ENODATA. There's two places that cause this: zfs_dirent_lock and zfs_zget. The fix in zfs_dirent_lock is pretty straightforward. In zfs_zget though, we need it to not return error when the zp is unlinked. This is a pretty big change in behavior, but skimming through all the callers, I don't think this change would cause any problem. Also there's nothing preventing z_unlinked from being set after the z_lock mutex is dropped before but before zfs_zget returns anyway. The rest of the stuff is to make sure we don't log xattr stuff when owner is unlinked. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-10-13 00:30:46 +00:00
/*
* If igrab() returns NULL the VFS has independently
* determined the inode should be evicted and has
* called iput_final() to start the eviction process.
* The SA handle is still valid but because the VFS
* requires that the eviction succeed we must drop
* our locks and references to allow the eviction to
* complete. The zfs_zget() may then be retried.
*
* This unlikely case could be optimized by registering
* a sops->drop_inode() callback. The callback would
* need to detect the active SA hold thereby informing
* the VFS that this inode should not be evicted.
*/
if (igrab(ZTOI(zp)) == NULL) {
mutex_exit(&zp->z_lock);
sa_buf_rele(db, NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Fix unlinked file cannot do xattr operations Currently, doing things like fsetxattr(2) on an unlinked file will result in ENODATA. There's two places that cause this: zfs_dirent_lock and zfs_zget. The fix in zfs_dirent_lock is pretty straightforward. In zfs_zget though, we need it to not return error when the zp is unlinked. This is a pretty big change in behavior, but skimming through all the callers, I don't think this change would cause any problem. Also there's nothing preventing z_unlinked from being set after the z_lock mutex is dropped before but before zfs_zget returns anyway. The rest of the stuff is to make sure we don't log xattr stuff when owner is unlinked. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-10-13 00:30:46 +00:00
/* inode might need this to finish evict */
cond_resched();
goto again;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Fix unlinked file cannot do xattr operations Currently, doing things like fsetxattr(2) on an unlinked file will result in ENODATA. There's two places that cause this: zfs_dirent_lock and zfs_zget. The fix in zfs_dirent_lock is pretty straightforward. In zfs_zget though, we need it to not return error when the zp is unlinked. This is a pretty big change in behavior, but skimming through all the callers, I don't think this change would cause any problem. Also there's nothing preventing z_unlinked from being set after the z_lock mutex is dropped before but before zfs_zget returns anyway. The rest of the stuff is to make sure we don't log xattr stuff when owner is unlinked. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-10-13 00:30:46 +00:00
*zpp = zp;
err = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_exit(&zp->z_lock);
Fix locking order in zfs_zget() Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-03-25 21:49:47 +00:00
sa_buf_rele(db, NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (err);
}
/*
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
* Not found create new znode/vnode but only if file exists.
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
*
* There is a small window where zfs_vget() could
* find this object while a file create is still in
* progress. This is checked for in zfs_znode_alloc()
*
* if zfs_znode_alloc() fails it will drop the hold on the
* bonus buffer.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zp = zfs_znode_alloc(zfsvfs, db, doi.doi_data_block_size,
Kill zp->z_xattr_parent to prevent pinning zp->z_xattr_parent will pin the parent. This will cause huge issue when unlink a file with xattr. Because the unlinked file is pinned, it will never get purged immediately. And because of that, the xattr stuff will never be marked as unlinked. So the whole unlinked stuff will stay there until shrink cache or umount. This change partially reverts e89260a. This is safe because only the zp->z_xattr_parent optimization is removed, zpl_xattr_security_init() is still called from the zpl outside the inode lock. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlop <chris@onthe.net.au> Issue #4359 Issue #3508 Issue #4413 Issue #4827
2016-07-06 00:24:36 +00:00
doi.doi_bonus_type, obj_num, NULL);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (zp == NULL) {
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
err = SET_ERROR(ENOENT);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
} else {
*zpp = zp;
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
return (err);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
int
zfs_rezget(znode_t *zp)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
dmu_object_info_t doi;
dmu_buf_t *db;
uint64_t obj_num = zp->z_id;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t mode;
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
uint64_t links;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
sa_bulk_attr_t bulk[10];
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
int err;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
int count = 0;
uint64_t gen;
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
uint64_t z_uid, z_gid;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
uint64_t atime[2], mtime[2], ctime[2];
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
znode_hold_t *zh;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Skip ctldir znode in zfs_rezget to fix snapdir issues Skip ctldir in zfs_rezget, otherwise they will always get invalidated. This will cause funny behaviour for the mounted snapdirs. Especially for Linux >= 3.18, d_invalidate will detach the mountpoint and prevent anyone automount it again as long as someone is still using the detached mount. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4514 Closes #4661 Closes #4672
2016-05-20 00:04:27 +00:00
/*
* skip ctldir, otherwise they will always get invalidated. This will
* cause funny behaviour for the mounted snapdirs. Especially for
* Linux >= 3.18, d_invalidate will detach the mountpoint and prevent
* anyone automount it again as long as someone is still using the
* detached mount.
*/
if (zp->z_is_ctldir)
return (0);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zh = zfs_znode_hold_enter(zfsvfs, obj_num);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
mutex_enter(&zp->z_acl_lock);
if (zp->z_acl_cached) {
zfs_acl_free(zp->z_acl_cached);
zp->z_acl_cached = NULL;
}
mutex_exit(&zp->z_acl_lock);
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
Revert "Fix z_xattr_lock/z_teardown_lock lock inversion" This reverts commit 6b32ef572f754efc3f9edb20d022450f8e6b02d9.
2015-12-21 17:12:37 +00:00
rw_enter(&zp->z_xattr_lock, RW_WRITER);
Fix 'zfs rollback' on mounted file systems Rolling back a mounted filesystem with open file handles and cached dentries+inodes never worked properly in ZoL. The major issue was that Linux provides no easy mechanism for modules to invalidate the inode cache for a file system. Because of this it was possible that an inode from the previous filesystem would not get properly dropped from the cache during rolling back. Then a new inode with the same inode number would be create and collide with the existing cached inode. Ideally this would trigger an VERIFY() but in practice the error wasn't handled and it would just NULL reference. Luckily, this issue can be resolved by sprucing up the existing Solaris zfs_rezget() functionality for the Linux VFS. The way it works now is that when a file system is rolled back all the cached inodes will be traversed and refetched from disk. If a version of the cached inode exists on disk the in-core copy will be updated accordingly. If there is no match for that object on disk it will be unhashed from the inode cache and marked as stale. This will effectively make the inode unfindable for lookups allowing the inode number to be immediately recycled. The inode will then only be accessible from the cached dentries. Subsequent dentry lookups which reference a stale inode will result in the dentry being invalidated. Once invalidated the dentry will drop its reference on the inode allowing it to be safely pruned from the cache. Special care is taken for negative dentries since they do not reference any inode. These dentires will be invalidate based on when they were added to the dentry cache. Entries added before the last rollback will be invalidate to prevent them from masking real files in the dataset. Two nice side effects of this fix are: * Removes the dependency on spl_invalidate_inodes(), it can now be safely removed from the SPL when we choose to do so. * zfs_znode_alloc() no longer requires a dentry to be passed. This effectively reverts this portition of the code to its upstream counterpart. The dentry is not instantiated more correctly in the Linux ZPL layer. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #795
2013-01-16 00:41:09 +00:00
if (zp->z_xattr_cached) {
nvlist_free(zp->z_xattr_cached);
zp->z_xattr_cached = NULL;
}
rw_exit(&zp->z_xattr_lock);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ASSERT(zp->z_sa_hdl == NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
err = sa_buf_hold(zfsvfs->z_os, obj_num, NULL, &db);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (err) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (err);
}
dmu_object_info_from_db(db, &doi);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (doi.doi_bonus_type != DMU_OT_SA &&
(doi.doi_bonus_type != DMU_OT_ZNODE ||
(doi.doi_bonus_type == DMU_OT_ZNODE &&
doi.doi_bonus_size < sizeof (znode_phys_t)))) {
sa_buf_rele(db, NULL);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
return (SET_ERROR(EINVAL));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_sa_init(zfsvfs, zp, db, doi.doi_bonus_type, NULL);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
/* reload cached values */
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GEN(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&gen, sizeof (gen));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&zp->z_size, sizeof (zp->z_size));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs), NULL,
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
&links, sizeof (links));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&zp->z_pflags, sizeof (zp->z_pflags));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_UID(zfsvfs), NULL,
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
&z_uid, sizeof (z_uid));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GID(zfsvfs), NULL,
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
&z_gid, sizeof (z_gid));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MODE(zfsvfs), NULL,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&mode, sizeof (mode));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_ATIME(zfsvfs), NULL,
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
&atime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zfsvfs), NULL,
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
&mtime, 16);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs), NULL,
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
&ctime, 16);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (sa_bulk_lookup(zp->z_sa_hdl, bulk, count)) {
zfs_znode_dmu_fini(zp);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
return (SET_ERROR(EIO));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
}
Refactor inode->i_mode management Refactor the code in such a way so that inode->i_mode is being set at the same time zp->z_mode is being changed. This has the effect of keeping both in sync without relying on zfs_inode_update. Reviewed-by: Richard Laager <rlaager@wiktel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Closes #5158
2016-09-27 21:08:52 +00:00
zp->z_mode = ZTOI(zp)->i_mode = mode;
Remove znode's z_uid/z_gid member Remove duplicate z_uid/z_gid member which are also held in the generic vfs inode struct. This is done by first removing the members from struct znode and then using the KUID_TO_SUID/KGID_TO_SGID macros to access the respective member from struct inode. In cases where the uid/gids are being marshalled from/to disk, use the newly introduced zfs_(uid|gid)_(read|write) functions to properly save the uids rather than the internal kernel representation. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4685 Issue #227
2016-05-22 11:15:57 +00:00
zfs_uid_write(ZTOI(zp), z_uid);
zfs_gid_write(ZTOI(zp), z_gid);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
ZFS_TIME_DECODE(&ZTOI(zp)->i_atime, atime);
ZFS_TIME_DECODE(&ZTOI(zp)->i_mtime, mtime);
ZFS_TIME_DECODE(&ZTOI(zp)->i_ctime, ctime);
Kill znode->z_gen field This field is a duplicate of the inode->i_generation, so just kill it. Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4538 Closes #4654
2016-04-18 19:08:53 +00:00
if (gen != ZTOI(zp)->i_generation) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zfs_znode_dmu_fini(zp);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
return (SET_ERROR(EIO));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
zp->z_unlinked = (ZTOI(zp)->i_nlink == 0);
set_nlink(ZTOI(zp), (uint32_t)links);
Fix i_flags issue caused by 64c688d Fix zfs_xvattr_set to set S_IMMUTABLE and S_APPEND flags correctly. Reinstate zfs_set_inode_flags and use it when zfs_xvatter_set and also when setting up inode in zfs_znode_alloc and zfs_rezget. Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
2016-12-14 22:18:53 +00:00
zfs_set_inode_flags(zp, ZTOI(zp));
Use native inode->i_nlink instead of znode->z_links A mostly mechanical change, taking into account i_nlink is 32 bits vs ZFS's 64 bit on-disk link count. We revert "xattr dir doesn't get purged during iput" (ddae16a) as this is a more Linux-integrated fix for the same issue. In addition, setting the initial link count on a new node has been changed from setting one less than required in zfs_mknode() then incrementing to the correct count in zfs_link_create() (which was somewhat bizarre in the first place), to setting the correct count in zfs_mknode() and not incrementing it in zfs_link_create(). This both means we no longer set the link count in sa_bulk_update() twice (once for the initial incorrect count then again for the correct count), as well as adhering to the Linux requirement of not incrementing a zero link count without I_LINKABLE (see linux commit f4e0c30c). Signed-off-by: Chris Dunlop <chris@onthe.net.au> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Closes #4838 Issue #227
2016-07-14 14:44:38 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zp->z_blksz = doi.doi_data_block_size;
Enable lazytime semantic for atime Linux 4.0 introduces lazytime. The idea is that when we update the atime, we delay writing it to disk for as long as it is reasonably possible. When lazytime is enabled, dirty_inode will be called with only I_DIRTY_TIME flag whenever i_atime is updated. So under such condition, we will set z_atime_dirty. We will only write it to disk if file is closed, inode is evicted or setattr is called. Ideally, we should also write it whenever SA is going to be updated, but it is left for future improvement. There's one thing that we should take care of now that we allow i_atime to be dirty. In original implementation, whenever SA is modified, zfs_inode_update will be called to overwrite every thing in inode. This will cause dirty i_atime to be discarded. We fix this by don't overwrite i_atime in zfs_inode_update. We only overwrite i_atime when allocating new inode or doing zfs_rezget with zfs_inode_update_new. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-31 23:52:03 +00:00
zp->z_atime_dirty = 0;
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
zfs_inode_update(zp);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (0);
}
void
zfs_znode_delete(znode_t *zp, dmu_tx_t *tx)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
objset_t *os = zfsvfs->z_os;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
uint64_t obj = zp->z_id;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
uint64_t acl_obj = zfs_external_acl(zp);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
znode_hold_t *zh;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zh = zfs_znode_hold_enter(zfsvfs, obj);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (acl_obj) {
VERIFY(!zp->z_is_sa);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
VERIFY(0 == dmu_object_free(os, acl_obj, tx));
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
}
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
VERIFY(0 == dmu_object_free(os, obj, tx));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zfs_znode_dmu_fini(zp);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
void
zfs_zinactive(znode_t *zp)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
uint64_t z_id = zp->z_id;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
znode_hold_t *zh;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ASSERT(zp->z_sa_hdl);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
Fix evict() deadlock Now that KM_SLEEP is not defined as GFP_NOFS there is the possibility of synchronous reclaim deadlocks. These deadlocks never existed in the original OpenSolaris code because all memory reclaim on Solaris is done asyncronously. Linux does both synchronous (direct) and asynchronous (indirect) reclaim. This commit addresses a deadlock caused by inode eviction. A KM_SLEEP allocation may trigger direct memory reclaim and shrink the inode cache. This can occur while a mutex in the array of ZFS_OBJ_HOLD mutexes is held. Through the ->shrink_icache_memory()->evict()->zfs_inactive()-> zfs_zinactive() call path the same mutex may be reacquired resulting in a deadlock. To avoid this deadlock the process must not reacquire the mutex when it is already holding it. This is a reasonable fix for now but longer term the ZFS_OBJ_HOLD mutex locking should be reevaluated. This infrastructure already prevents us from ever using the Linux lock dependency analysis tools, and it may limit scalability.
2011-03-21 17:19:30 +00:00
* Don't allow a zfs_zget() while were trying to release this znode.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zh = zfs_znode_hold_enter(zfsvfs, z_id);
Fix evict() deadlock Now that KM_SLEEP is not defined as GFP_NOFS there is the possibility of synchronous reclaim deadlocks. These deadlocks never existed in the original OpenSolaris code because all memory reclaim on Solaris is done asyncronously. Linux does both synchronous (direct) and asynchronous (indirect) reclaim. This commit addresses a deadlock caused by inode eviction. A KM_SLEEP allocation may trigger direct memory reclaim and shrink the inode cache. This can occur while a mutex in the array of ZFS_OBJ_HOLD mutexes is held. Through the ->shrink_icache_memory()->evict()->zfs_inactive()-> zfs_zinactive() call path the same mutex may be reacquired resulting in a deadlock. To avoid this deadlock the process must not reacquire the mutex when it is already holding it. This is a reasonable fix for now but longer term the ZFS_OBJ_HOLD mutex locking should be reevaluated. This infrastructure already prevents us from ever using the Linux lock dependency analysis tools, and it may limit scalability.
2011-03-21 17:19:30 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_enter(&zp->z_lock);
/*
* If this was the last reference to a file with no links,
* remove the file from the file system.
*/
if (zp->z_unlinked) {
mutex_exit(&zp->z_lock);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zfs_rmnode(zp);
return;
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
mutex_exit(&zp->z_lock);
zfs_znode_dmu_fini(zp);
Fix evict() deadlock Now that KM_SLEEP is not defined as GFP_NOFS there is the possibility of synchronous reclaim deadlocks. These deadlocks never existed in the original OpenSolaris code because all memory reclaim on Solaris is done asyncronously. Linux does both synchronous (direct) and asynchronous (indirect) reclaim. This commit addresses a deadlock caused by inode eviction. A KM_SLEEP allocation may trigger direct memory reclaim and shrink the inode cache. This can occur while a mutex in the array of ZFS_OBJ_HOLD mutexes is held. Through the ->shrink_icache_memory()->evict()->zfs_inactive()-> zfs_zinactive() call path the same mutex may be reacquired resulting in a deadlock. To avoid this deadlock the process must not reacquire the mutex when it is already holding it. This is a reasonable fix for now but longer term the ZFS_OBJ_HOLD mutex locking should be reevaluated. This infrastructure already prevents us from ever using the Linux lock dependency analysis tools, and it may limit scalability.
2011-03-21 17:19:30 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfs_znode_hold_exit(zfsvfs, zh);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Implement relatime. Add the "relatime" property. When set to "on", a file's atime will only be updated if the existing atime at least a day old or if the existing ctime or mtime has been updated since the last access. This behavior is compatible with the Linux "relatime" mount option. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2064 Closes #1917
2014-01-18 19:00:53 +00:00
static inline int
zfs_compare_timespec(struct timespec *t1, struct timespec *t2)
{
if (t1->tv_sec < t2->tv_sec)
return (-1);
if (t1->tv_sec > t2->tv_sec)
return (1);
return (t1->tv_nsec - t2->tv_nsec);
}
/*
* Prepare to update znode time stamps.
*
* IN: zp - znode requiring timestamp update
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
* flag - ATTR_MTIME, ATTR_CTIME flags
Implement relatime. Add the "relatime" property. When set to "on", a file's atime will only be updated if the existing atime at least a day old or if the existing ctime or mtime has been updated since the last access. This behavior is compatible with the Linux "relatime" mount option. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2064 Closes #1917
2014-01-18 19:00:53 +00:00
*
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
* OUT: zp - z_seq
Implement relatime. Add the "relatime" property. When set to "on", a file's atime will only be updated if the existing atime at least a day old or if the existing ctime or mtime has been updated since the last access. This behavior is compatible with the Linux "relatime" mount option. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2064 Closes #1917
2014-01-18 19:00:53 +00:00
* mtime - new mtime
* ctime - new ctime
*
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
* Note: We don't update atime here, because we rely on Linux VFS to do
* atime updating.
Implement relatime. Add the "relatime" property. When set to "on", a file's atime will only be updated if the existing atime at least a day old or if the existing ctime or mtime has been updated since the last access. This behavior is compatible with the Linux "relatime" mount option. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2064 Closes #1917
2014-01-18 19:00:53 +00:00
*/
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
void
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zfs_tstamp_update_setup(znode_t *zp, uint_t flag, uint64_t mtime[2],
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
uint64_t ctime[2])
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Linux 4.18 compat: inode timespec -> timespec64 Commit torvalds/linux@95582b0 changes the inode i_atime, i_mtime, and i_ctime members form timespec's to timespec64's to make them 2038 safe. As part of this change the current_time() function was also updated to return the timespec64 type. Resolve this issue by introducing a new inode_timespec_t type which is defined to match the timespec type used by the inode. It should be used when working with inode timestamps to ensure matching types. The timestruc_t type under Illumos was used in a similar fashion but was specified to always be a timespec_t. Rather than incorrectly define this type all timespec_t types have been replaced by the new inode_timespec_t type. Finally, the kernel and user space 'sys/time.h' headers were aligned with each other. They define as appropriate for the context several constants as macros and include static inline implementation of gethrestime(), gethrestime_sec(), and gethrtime(). Reviewed-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #7643 Backported-by: Richard Yao <ryao@gentoo.org>
2018-08-12 22:22:03 +00:00
inode_timespec_t now;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
gethrestime(&now);
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
zp->z_seq++;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (flag & ATTR_MTIME) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&now, mtime);
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
ZFS_TIME_DECODE(&(ZTOI(zp)->i_mtime), mtime);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (ZTOZSB(zp)->z_use_fuids) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_pflags |= (ZFS_ARCHIVE |
ZFS_AV_MODIFIED);
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (flag & ATTR_CTIME) {
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
ZFS_TIME_ENCODE(&now, ctime);
Refactor generic inode time updating ZFS doesn't provide a custom update_time method meaning it delegates this job to the generic VFS layer. The only time when it needs to set the various *time values is when the inode is being marshalled to/from the disk. Do this by moving the relevant code from zfs_inode_update_impl to zfs_node_alloc and zfs_rezget. As a result from this change it is no longer necessary to have multiple versions of the zfs_inode_update function - so just nuke them and leave only one. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com> Issue #227 Closes #4916
2016-08-01 20:02:25 +00:00
ZFS_TIME_DECODE(&(ZTOI(zp)->i_ctime), ctime);
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (ZTOZSB(zp)->z_use_fuids)
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_pflags |= ZFS_ARCHIVE;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
}
/*
* Grow the block size for a file.
*
* IN: zp - znode of file to free data in.
* size - requested block size
* tx - open transaction.
*
* NOTE: this function assumes that the znode is write locked.
*/
void
zfs_grow_blocksize(znode_t *zp, uint64_t size, dmu_tx_t *tx)
{
int error;
u_longlong_t dummy;
if (size <= zp->z_blksz)
return;
/*
* If the file size is already greater than the current blocksize,
* we will not grow. If there is more than one block in a file,
* the blocksize cannot change.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (zp->z_blksz && zp->z_size > zp->z_blksz)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
error = dmu_object_set_blocksize(ZTOZSB(zp)->z_os, zp->z_id,
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
size, 0, tx);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (error == ENOTSUP)
return;
Illumos #3006 3006 VERIFY[S,U,P] and ASSERT[S,U,P] frequently check if first argument is zero Reviewed by Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by George Wilson <george.wilson@delphix.com> Approved by Eric Schrock <eric.schrock@delphix.com> References: illumos/illumos-gate@fb09f5aad449c97fe309678f3f604982b563a96f https://illumos.org/issues/3006 Requires: zfsonlinux/spl@1c6d149feb4033e4a56fb987004edc5d45288bcb Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1509
2013-05-10 21:17:03 +00:00
ASSERT0(error);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/* What blocksize did we actually get? */
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_object_size_from_db(sa_get_db(zp->z_sa_hdl), &zp->z_blksz, &dummy);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
/*
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
* Increase the file length
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*
* IN: zp - znode of file to free data in.
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
* end - new end-of-file
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
* RETURN: 0 on success, error code on failure
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
static int
zfs_extend(znode_t *zp, uint64_t end)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
dmu_tx_t *tx;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
rl_t *rl;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
uint64_t newblksz;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
int error;
/*
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
* We will change zp_size, lock the whole file.
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
*/
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
rl = zfs_range_lock(&zp->z_range_lock, 0, UINT64_MAX, RL_WRITER);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* Nothing to do if file already at desired length.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (end <= zp->z_size) {
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zfs_range_unlock(rl);
return (0);
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
tx = dmu_tx_create(zfsvfs->z_os);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
if (end > zp->z_blksz &&
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
(!ISP2(zp->z_blksz) || zp->z_blksz < zfsvfs->z_max_blksz)) {
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* We are growing the file past the current block size.
*/
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
if (zp->z_blksz > ZTOZSB(zp)->z_max_blksz) {
Illumos 5027 - zfs large block support 5027 zfs large block support Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com> Reviewed by: Richard Elling <richard.elling@richardelling.com> Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/5027 https://github.com/illumos/illumos-gate/commit/b515258 Porting Notes: * Included in this patch is a tiny ISP2() cleanup in zio_init() from Illumos 5255. * Unlike the upstream Illumos commit this patch does not impose an arbitrary 128K block size limit on volumes. Volumes, like filesystems, are limited by the zfs_max_recordsize=1M module option. * By default the maximum record size is limited to 1M by the module option zfs_max_recordsize. This value may be safely increased up to 16M which is the largest block size supported by the on-disk format. At the moment, 1M blocks clearly offer a significant performance improvement but the benefits of going beyond this for the majority of workloads are less clear. * The illumos version of this patch increased DMU_MAX_ACCESS to 32M. This was determined not to be large enough when using 16M blocks because the zfs_make_xattrdir() function will fail (EFBIG) when assigning a TX. This was immediately observed under Linux because all newly created files must have a security xattr created and that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M. * On 32-bit platforms a hard limit of 1M is set for blocks due to the limited virtual address space. We should be able to relax this one the ABD patches are merged. Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #354
2014-11-03 20:15:08 +00:00
/*
* File's blocksize is already larger than the
* "recordsize" property. Only let it grow to
* the next power of 2.
*/
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
ASSERT(!ISP2(zp->z_blksz));
Illumos 5027 - zfs large block support 5027 zfs large block support Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com> Reviewed by: Richard Elling <richard.elling@richardelling.com> Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/5027 https://github.com/illumos/illumos-gate/commit/b515258 Porting Notes: * Included in this patch is a tiny ISP2() cleanup in zio_init() from Illumos 5255. * Unlike the upstream Illumos commit this patch does not impose an arbitrary 128K block size limit on volumes. Volumes, like filesystems, are limited by the zfs_max_recordsize=1M module option. * By default the maximum record size is limited to 1M by the module option zfs_max_recordsize. This value may be safely increased up to 16M which is the largest block size supported by the on-disk format. At the moment, 1M blocks clearly offer a significant performance improvement but the benefits of going beyond this for the majority of workloads are less clear. * The illumos version of this patch increased DMU_MAX_ACCESS to 32M. This was determined not to be large enough when using 16M blocks because the zfs_make_xattrdir() function will fail (EFBIG) when assigning a TX. This was immediately observed under Linux because all newly created files must have a security xattr created and that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M. * On 32-bit platforms a hard limit of 1M is set for blocks due to the limited virtual address space. We should be able to relax this one the ABD patches are merged. Ported-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #354
2014-11-03 20:15:08 +00:00
newblksz = MIN(end, 1 << highbit64(zp->z_blksz));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
newblksz = MIN(end, ZTOZSB(zp)->z_max_blksz);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
dmu_tx_hold_write(tx, zp->z_id, 0, newblksz);
} else {
newblksz = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Illumos #4347 ZPL can use dmu_tx_assign(TXG_WAIT) Fix a lock contention issue by allowing threads not holding ZPL locks to block when waiting to assign a transaction. Porting Notes: zfs_putpage() still uses TXG_NOWAIT, unlike the upstream version. This case may be a contention point just like zfs_write(), however it is not safe to block here since it may be called during memory reclaim. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Dan McDonald <danmcd@nexenta.com> Reviewed by: Boris Protopopov <boris.protopopov@nexenta.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://www.illumos.org/issues/4347 illumos/illumos-gate@e722410c49fe67cbf0f639cbcc288bd6cbcf7dd1 Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-11-22 23:13:18 +00:00
error = dmu_tx_assign(tx, TXG_WAIT);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
if (error) {
dmu_tx_abort(tx);
zfs_range_unlock(rl);
return (error);
}
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
if (newblksz)
zfs_grow_blocksize(zp, newblksz, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_size = end;
Prototype/structure update for Linux I appologize in advance why to many things ended up in this commit. When it could be seperated in to a whole series of commits teasing that all apart now would take considerable time and I'm not sure there's much merrit in it. As such I'll just summerize the intent of the changes which are all (or partly) in this commit. Broadly the intent is to remove as much Solaris specific code as possible and replace it with native Linux equivilants. More specifically: 1) Replace all instances of zfsvfs_t with zfs_sb_t. While the type is largely the same calling it private super block data rather than a zfsvfs is more consistent with how Linux names this. While non critical it makes the code easier to read when your thinking in Linux friendly VFS terms. 2) Replace vnode_t with struct inode. The Linux VFS doesn't have the notion of a vnode and there's absolutely no good reason to create one. There are in fact several good reasons to remove it. It just adds overhead on Linux if we were to manage one, it conplicates the code, and it likely will lead to bugs so there's a good change it will be out of date. The code has been updated to remove all need for this type. 3) Replace all vtype_t's with umode types. Along with this shift all uses of types to mode bits. The Solaris code would pass a vtype which is redundant with the Linux mode. Just update all the code to use the Linux mode macros and remove this redundancy. 4) Remove using of vn_* helpers and replace where needed with inode helpers. The big example here is creating iput_aync to replace vn_rele_async. Other vn helpers will be addressed as needed but they should be be emulated. They are a Solaris VFS'ism and should simply be replaced with Linux equivilants. 5) Update znode alloc/free code. Under Linux it's common to embed the inode specific data with the inode itself. This removes the need for an extra memory allocation. In zfs this information is called a znode and it now embeds the inode with it. Allocators have been updated accordingly. 6) Minimal integration with the vfs flags for setting up the super block and handling mount options has been added this code will need to be refined but functionally it's all there. This will be the first and last of these to large to review commits.
2011-02-08 19:16:06 +00:00
VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_SIZE(ZTOZSB(zp)),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
&zp->z_size, sizeof (zp->z_size), tx));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_range_unlock(rl);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
dmu_tx_commit(tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
return (0);
}
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
/*
* zfs_zero_partial_page - Modeled after update_pages() but
* with different arguments and semantics for use by zfs_freesp().
*
* Zeroes a piece of a single page cache entry for zp at offset
* start and length len.
*
* Caller must acquire a range lock on the file for the region
* being zeroed in order that the ARC and page cache stay in sync.
*/
static void
zfs_zero_partial_page(znode_t *zp, uint64_t start, uint64_t len)
{
struct address_space *mp = ZTOI(zp)->i_mapping;
struct page *pp;
int64_t off;
void *pb;
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
ASSERT((start & PAGE_MASK) == ((start + len - 1) & PAGE_MASK));
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
off = start & (PAGE_SIZE - 1);
start &= PAGE_MASK;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
pp = find_lock_page(mp, start >> PAGE_SHIFT);
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
if (pp) {
if (mapping_writably_mapped(mp))
flush_dcache_page(pp);
pb = kmap(pp);
bzero(pb + off, len);
kunmap(pp);
if (mapping_writably_mapped(mp))
flush_dcache_page(pp);
mark_page_accessed(pp);
SetPageUptodate(pp);
ClearPageError(pp);
unlock_page(pp);
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
put_page(pp);
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
}
}
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* Free space in a file.
*
* IN: zp - znode of file to free data in.
* off - start of section to free.
* len - length of section to free.
*
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
* RETURN: 0 on success, error code on failure
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
*/
static int
zfs_free_range(znode_t *zp, uint64_t off, uint64_t len)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
rl_t *rl;
int error;
/*
* Lock the range being freed.
*/
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
rl = zfs_range_lock(&zp->z_range_lock, off, len, RL_WRITER);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* Nothing to do if file already at desired length.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (off >= zp->z_size) {
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_range_unlock(rl);
return (0);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (off + len > zp->z_size)
len = zp->z_size - off;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
error = dmu_free_long_range(zfsvfs->z_os, zp->z_id, off, len);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
/*
* Zero partial page cache entries. This must be done under a
* range lock in order to keep the ARC and page cache in sync.
*/
if (zp->z_is_mapped) {
loff_t first_page, last_page, page_len;
loff_t first_page_offset, last_page_offset;
/* first possible full page in hole */
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
first_page = (off + PAGE_SIZE - 1) >> PAGE_SHIFT;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
/* last page of hole */
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
last_page = (off + len) >> PAGE_SHIFT;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
/* offset of first_page */
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
first_page_offset = first_page << PAGE_SHIFT;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
/* offset of last_page */
Linux 4.6 compat: PAGE_CACHE_SIZE removal As described in torvalds/linux@4a2d057e the macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were originally introduced to make it possible to add bigger chunks to the page cache. This never panned out and it has therefore been removed from the kernel. ZFS has been updated to use the PAGE_{SIZE,SHIFT,MASK,ALIGN} macros and calls to page_cache_release() have been replaced with put_page(). There was no need to introduce a configure check for this because these interfaces have existed for a very long time. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Closes #4489
2016-04-05 19:39:37 +00:00
last_page_offset = last_page << PAGE_SHIFT;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
Perform whole-page page truncation for hole-punching under a range lock As an attempt to perform the page truncation more optimally, the hole-punching support added in 223df0161fad50f53a8fa5ffeea8cc4f8137d522 truncated performed the operation in two steps: first, sub-page "stubs" were zeroed under the range lock in zfs_free_range() using the new zfs_zero_partial_page() function and then the whole pages were truncated within zfs_freesp(). This left a window of opportunity during which the full pages could be touched. This patch closes the window by moving the whole-page truncation into zfs_free_range() under the range lock. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2733
2014-09-26 04:40:41 +00:00
/* truncate whole pages */
if (last_page_offset > first_page_offset) {
truncate_inode_pages_range(ZTOI(zp)->i_mapping,
first_page_offset, last_page_offset - 1);
}
/* truncate sub-page ranges */
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
if (first_page > last_page) {
/* entire punched area within a single page */
zfs_zero_partial_page(zp, off, len);
} else {
/* beginning of punched area at the end of a page */
page_len = first_page_offset - off;
if (page_len > 0)
zfs_zero_partial_page(zp, off, page_len);
/* end of punched area at the beginning of a page */
page_len = off + len - last_page_offset;
if (page_len > 0)
zfs_zero_partial_page(zp, last_page_offset,
page_len);
}
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
zfs_range_unlock(rl);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
return (error);
}
/*
* Truncate a file
*
* IN: zp - znode of file to free data in.
* end - new end-of-file.
*
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
* RETURN: 0 on success, error code on failure
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
*/
static int
zfs_trunc(znode_t *zp, uint64_t end)
{
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
dmu_tx_t *tx;
rl_t *rl;
int error;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
sa_bulk_attr_t bulk[2];
int count = 0;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* We will change zp_size, lock the whole file.
*/
Remove dummy znode from zvol_state struct zvol_state contains a dummy znode, which is around 1KB on x64, only for zfs_range_lock. But in reality, other than z_range_lock and z_range_avl, zfs_range_lock only need znode on regular file, which means we add 1KB on a structure and gain nothing. In this patch, we remove the dummy znode for zvol_state. In order to do that, we also need to refactor zfs_range_lock a bit. We move z_range_lock and z_range_avl pair out of znode_t to form zfs_rlock_t. This new struct replaces znode_t as the main handle inside the range lock functions. We also add pointers to z_size, z_blksz, and z_max_blksz so range lock code doesn't depend on znode_t. This allows non-ZPL consumers like Lustre to use the range locks with their equivalent znode_t structure. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Boris Protopopov <boris.protopopov@actifio.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4510
2016-04-11 21:53:48 +00:00
rl = zfs_range_lock(&zp->z_range_lock, 0, UINT64_MAX, RL_WRITER);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* Nothing to do if file already at desired length.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if (end >= zp->z_size) {
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_range_unlock(rl);
return (0);
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
error = dmu_free_long_range(zfsvfs->z_os, zp->z_id, end, -1);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
if (error) {
zfs_range_unlock(rl);
return (error);
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
tx = dmu_tx_create(zfsvfs->z_os);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
dmu_tx_mark_netfree(tx);
zfs_trunc() should use dmu_tx_assign(tx, TXG_WAIT) As part of the write throttle & i/o schedule performance work the zfs_trunc() function should have been updated to use TXG_WAIT. Using TXG_WAIT ensures that the tx will be part of the next txg. If TXG_NOWAIT is used and retried for ERESTART errors then the tx can suffer from starvation. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Closes #2488
2014-07-21 17:19:25 +00:00
error = dmu_tx_assign(tx, TXG_WAIT);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
if (error) {
dmu_tx_abort(tx);
zfs_range_unlock(rl);
return (error);
}
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
zp->z_size = end;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zfsvfs),
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
NULL, &zp->z_size, sizeof (zp->z_size));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (end == 0) {
zp->z_pflags &= ~ZFS_SPARSE;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs),
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
NULL, &zp->z_pflags, 8);
}
VERIFY(sa_bulk_update(zp->z_sa_hdl, bulk, count, tx) == 0);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
dmu_tx_commit(tx);
Rebase master to b108
2009-02-18 20:51:31 +00:00
zfs_range_unlock(rl);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (0);
}
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
/*
* Free space in a file
*
* IN: zp - znode of file to free data in.
* off - start of range
* len - end of range (0 => EOF)
* flag - current file open mode flags.
* log - TRUE if this action should be logged
*
Illumos 4950 - files sometimes can't be removed from a full filesystem 4950 files sometimes can't be removed from a full filesystem Reviewed by: Adam Leventhal <adam.leventhal@delphix.com> Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Sebastien Roy <sebastien.roy@delphix.com> Reviewed by: Boris Protopopov <bprotopopov@hotmail.com> Approved by: Dan McDonald <danmcd@omniti.com> References: https://www.illumos.org/issues/4950 https://github.com/illumos/illumos-gate/commit/4bb7380 Porting notes: - ZoL currently does not log discards to zvols, so the portion of this patch that modifies the discard logging to mark it as freeing space has been discarded. 2. may_delete_now had been removed from zfs_remove() in ZoL. It has been reintroduced. 3. We do not try to emulate vnodes, so the following lines are not valid on Linux: mutex_enter(&vp->v_lock); may_delete_now = vp->v_count == 1 && !vn_has_cached_data(vp); mutex_exit(&vp->v_lock); This has been replaced with: mutex_enter(&zp->z_lock); may_delete_now = atomic_read(&ip->i_count) == 1 && !(zp->z_is_mapped); mutex_exit(&zp->z_lock); Ported-by: Richard Yao <richard.yao@clusterhq.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-07-07 19:49:36 +00:00
* RETURN: 0 on success, error code on failure
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
*/
int
zfs_freesp(znode_t *zp, uint64_t off, uint64_t len, int flag, boolean_t log)
{
dmu_tx_t *tx;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs = ZTOZSB(zp);
zilog_t *zilog = zfsvfs->z_log;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t mode;
uint64_t mtime[2], ctime[2];
sa_bulk_attr_t bulk[3];
int count = 0;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
int error;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
if ((error = sa_lookup(zp->z_sa_hdl, SA_ZPL_MODE(zfsvfs), &mode,
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
sizeof (mode))) != 0)
return (error);
if (off > zp->z_size) {
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
error = zfs_extend(zp, off+len);
if (error == 0 && log)
goto log;
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
goto out;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
}
if (len == 0) {
error = zfs_trunc(zp, off);
} else {
if ((error = zfs_free_range(zp, off, len)) == 0 &&
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
off + len > zp->z_size)
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
error = zfs_extend(zp, off+len);
}
if (error || !log)
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
goto out;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
log:
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
tx = dmu_tx_create(zfsvfs->z_os);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
Illumos #4347 ZPL can use dmu_tx_assign(TXG_WAIT) Fix a lock contention issue by allowing threads not holding ZPL locks to block when waiting to assign a transaction. Porting Notes: zfs_putpage() still uses TXG_NOWAIT, unlike the upstream version. This case may be a contention point just like zfs_write(), however it is not safe to block here since it may be called during memory reclaim. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Dan McDonald <danmcd@nexenta.com> Reviewed by: Boris Protopopov <boris.protopopov@nexenta.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://www.illumos.org/issues/4347 illumos/illumos-gate@e722410c49fe67cbf0f639cbcc288bd6cbcf7dd1 Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2013-11-22 23:13:18 +00:00
error = dmu_tx_assign(tx, TXG_WAIT);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
if (error) {
dmu_tx_abort(tx);
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
goto out;
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
}
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zfsvfs), NULL, mtime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs), NULL, ctime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs),
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
NULL, &zp->z_pflags, 8);
Fix atime handling and relatime The problem for atime: We have 3 places for atime: inode->i_atime, znode->z_atime and SA. And its handling is a mess. A huge part of mess regarding atime comes from zfs_tstamp_update_setup, zfs_inode_update, and zfs_getattr, which behave inconsistently with those three values. zfs_tstamp_update_setup clears z_atime_dirty unconditionally as long as you don't pass ATTR_ATIME. Which means every write(2) operation which only updates ctime and mtime will cause atime changes to not be written to disk. Also zfs_inode_update from write(2) will replace inode->i_atime with what's inside SA(stale). But doesn't touch z_atime. So after read(2) and write(2). You'll have i_atime(stale), z_atime(new), SA(stale) and z_atime_dirty=0. Now, if you do stat(2), zfs_getattr will actually replace i_atime with what's inside, z_atime. So you will have now you'll have i_atime(new), z_atime(new), SA(stale) and z_atime_dirty=0. These will all gone after umount. And you'll leave with a stale atime. The problem for relatime: We do have a relatime config inside ZFS dataset, but how it should interact with the mount flag MS_RELATIME is not well defined. It seems it wanted relatime mount option to override the dataset config by showing it as temporary in `zfs get`. But at the same time, `zfs set relatime=on|off` would also seems to want to override the mount option. Not to mention that MS_RELATIME flag is actually never passed into ZFS, so it never really worked. How Linux handles atime: The Linux kernel actually handles atime completely in VFS, except for writing it to disk. So if we remove the atime handling in ZFS, things would just work, no matter it's strictatime, relatime, noatime, or even O_NOATIME. And whenever VFS updates the i_atime, it will notify the underlying filesystem via sb->dirty_inode(). And also there's one thing to note about atime flags like MS_RELATIME and other flags like MS_NODEV, etc. They are mount point flags rather than filesystem(sb) flags. Since native linux filesystem can be mounted at multiple places at the same time, they can all have different atime settings. So these flags are never passed down to filesystem drivers. What this patch tries to do: We remove znode->z_atime, since we won't gain anything from it. We remove most of the atime handling and leave it to VFS. The only thing we do with atime is to write it when dirty_inode() or setattr() is called. We also add file_accessed() in zpl_read() since it's not provided in vfs_read(). After this patch, only the MS_RELATIME flag will have effect. The setting in dataset won't do anything. We will make zfstuil to mount ZFS with MS_RELATIME set according to the setting in dataset in future patch. Signed-off-by: Chunwei Chen <david.chen@osnexus.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4482
2016-03-30 00:53:34 +00:00
zfs_tstamp_update_setup(zp, CONTENT_MODIFIED, mtime, ctime);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
zfs_log_truncate(zilog, tx, TX_TRUNCATE, zp, off, len);
dmu_tx_commit(tx);
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
VFS: Add zfs_inode_update() helper For the moment we have left ZFS unchanged and it updates many values as part of the znode. However, some of these values should be set in the inode. For the moment this is handled by adding a function called zfs_inode_update() which updates the inode based on the znode. This is considered a workaround until we can systematically go through the ZFS code and have it directly update the inode. At which point zfs_update_inode() can be dropped entirely. Keeping two copies of the same data isn't only inefficient it's a breeding ground for bugs.
2011-01-05 22:27:30 +00:00
zfs_inode_update(zp);
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
error = 0;
out:
/*
* Truncate the page cache - for file truncate operations, use
* the purpose-built API for truncations. For punching operations,
Perform whole-page page truncation for hole-punching under a range lock As an attempt to perform the page truncation more optimally, the hole-punching support added in 223df0161fad50f53a8fa5ffeea8cc4f8137d522 truncated performed the operation in two steps: first, sub-page "stubs" were zeroed under the range lock in zfs_free_range() using the new zfs_zero_partial_page() function and then the whole pages were truncated within zfs_freesp(). This left a window of opportunity during which the full pages could be touched. This patch closes the window by moving the whole-page truncation into zfs_free_range() under the range lock. Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #2733
2014-09-26 04:40:41 +00:00
* the truncation is handled under a range lock in zfs_free_range.
Implement fallocate FALLOC_FL_PUNCH_HOLE Add support for the FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE mode of fallocate(2). Mimic the behavior of other native file systems such as ext4 in cases where the file might be extended. If the offset is beyond the end of the file, return success without changing the file. If the extent of the punched hole would extend the file, only the existing tail of the file is punched. Add the zfs_zero_partial_page() function, modeled after update_page(), to handle zeroing partial pages in a hole-punching operation. It must be used under a range lock for the requested region in order that the ARC and page cache stay in sync. Move the existing page cache truncation via truncate_setsize() into zfs_freesp() for better source structure compatibility with upstream code. Add page cache truncation to zfs_freesp() and zfs_free_range() to handle hole punching. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Closes #2619
2014-08-20 22:35:13 +00:00
*/
if (len == 0)
truncate_setsize(ZTOI(zp), off);
return (error);
Rebase to OpenSolaris b103, in the process we are removing any code which did not originate from the OpenSolaris source. These changes will be reintroduced in topic branches for easier tracking
2008-12-03 20:09:06 +00:00
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
void
zfs_create_fs(objset_t *os, cred_t *cr, nvlist_t *zplprops, dmu_tx_t *tx)
{
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
struct super_block *sb;
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs_t *zfsvfs;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
uint64_t moid, obj, sa_obj, version;
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
uint64_t sense = ZFS_CASE_SENSITIVE;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
uint64_t norm = 0;
nvpair_t *elem;
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
int size;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
int error;
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
int i;
znode_t *rootzp = NULL;
vattr_t vattr;
znode_t *zp;
zfs_acl_ids_t acl_ids;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* First attempt to create master node.
*/
/*
* In an empty objset, there are no blocks to read and thus
* there can be no i/o errors (which we assert below).
*/
moid = MASTER_NODE_OBJ;
error = zap_create_claim(os, moid, DMU_OT_MASTER_NODE,
DMU_OT_NONE, 0, tx);
ASSERT(error == 0);
/*
* Set starting attributes.
*/
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
version = zfs_zpl_version_map(spa_version(dmu_objset_spa(os)));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
elem = NULL;
while ((elem = nvlist_next_nvpair(zplprops, elem)) != NULL) {
/* For the moment we expect all zpl props to be uint64_ts */
uint64_t val;
char *name;
ASSERT(nvpair_type(elem) == DATA_TYPE_UINT64);
VERIFY(nvpair_value_uint64(elem, &val) == 0);
name = nvpair_name(elem);
if (strcmp(name, zfs_prop_to_name(ZFS_PROP_VERSION)) == 0) {
Rebase master to b117
2009-07-02 22:44:48 +00:00
if (val < version)
version = val;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
} else {
error = zap_update(os, moid, name, 8, 1, &val, tx);
}
ASSERT(error == 0);
if (strcmp(name, zfs_prop_to_name(ZFS_PROP_NORMALIZE)) == 0)
norm = val;
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
else if (strcmp(name, zfs_prop_to_name(ZFS_PROP_CASE)) == 0)
sense = val;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
ASSERT(version != 0);
Rebase master to b117
2009-07-02 22:44:48 +00:00
error = zap_update(os, moid, ZPL_VERSION_STR, 8, 1, &version, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
/*
* Create zap object used for SA attribute registration
*/
if (version >= ZPL_VERSION_SA) {
sa_obj = zap_create(os, DMU_OT_SA_MASTER_NODE,
DMU_OT_NONE, 0, tx);
error = zap_add(os, moid, ZFS_SA_ATTRS, 8, 1, &sa_obj, tx);
ASSERT(error == 0);
} else {
sa_obj = 0;
}
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
/*
* Create a delete queue.
*/
Rebase master to b117
2009-07-02 22:44:48 +00:00
obj = zap_create(os, DMU_OT_UNLINKED_SET, DMU_OT_NONE, 0, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Rebase master to b117
2009-07-02 22:44:48 +00:00
error = zap_add(os, moid, ZFS_UNLINKED_SET, 8, 1, &obj, tx);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
ASSERT(error == 0);
Rebase master to b117
2009-07-02 22:44:48 +00:00
/*
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
* Create root znode. Create minimal znode/inode/zfsvfs/sb
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
* to allow zfs_mknode to work.
Rebase master to b117
2009-07-02 22:44:48 +00:00
*/
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
vattr.va_mask = ATTR_MODE|ATTR_UID|ATTR_GID;
vattr.va_mode = S_IFDIR|0755;
vattr.va_uid = crgetuid(cr);
vattr.va_gid = crgetgid(cr);
Change KM_PUSHPAGE -> KM_SLEEP By marking DMU transaction processing contexts with PF_FSTRANS we can revert the KM_PUSHPAGE -> KM_SLEEP changes. This brings us back in line with upstream. In some cases this means simply swapping the flags back. For others fnvlist_alloc() was replaced by nvlist_alloc(..., KM_PUSHPAGE) and must be reverted back to fnvlist_alloc() which assumes KM_SLEEP. The one place KM_PUSHPAGE is kept is when allocating ARC buffers which allows us to dip in to reserved memory. This is again the same as upstream. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-11-21 00:09:39 +00:00
rootzp = kmem_cache_alloc(znode_cache, KM_SLEEP);
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
rootzp->z_moved = 0;
rootzp->z_unlinked = 0;
rootzp->z_atime_dirty = 0;
rootzp->z_is_sa = USE_SA(version, os);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs = kmem_zalloc(sizeof (zfsvfs_t), KM_SLEEP);
zfsvfs->z_os = os;
zfsvfs->z_parent = zfsvfs;
zfsvfs->z_version = version;
zfsvfs->z_use_fuids = USE_FUIDS(version, os);
zfsvfs->z_use_sa = USE_SA(version, os);
zfsvfs->z_norm = norm;
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
Change KM_PUSHPAGE -> KM_SLEEP By marking DMU transaction processing contexts with PF_FSTRANS we can revert the KM_PUSHPAGE -> KM_SLEEP changes. This brings us back in line with upstream. In some cases this means simply swapping the flags back. For others fnvlist_alloc() was replaced by nvlist_alloc(..., KM_PUSHPAGE) and must be reverted back to fnvlist_alloc() which assumes KM_SLEEP. The one place KM_PUSHPAGE is kept is when allocating ARC buffers which allows us to dip in to reserved memory. This is again the same as upstream. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2014-11-21 00:09:39 +00:00
sb = kmem_zalloc(sizeof (struct super_block), KM_SLEEP);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
sb->s_fs_info = zfsvfs;
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
ZTOI(rootzp)->i_sb = sb;
error = sa_setup(os, sa_obj, zfs_attr_table, ZPL_END,
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
&zfsvfs->z_attr_table);
Rebase master to b117
2009-07-02 22:44:48 +00:00
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
ASSERT(error == 0);
Rebase master to b117
2009-07-02 22:44:48 +00:00
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
/*
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
* Fold case on file systems that are always or sometimes case
* insensitive.
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
*/
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
if (sense == ZFS_CASE_INSENSITIVE || sense == ZFS_CASE_MIXED)
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs->z_norm |= U8_TEXTPREP_TOUPPER;
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_init(&zfsvfs->z_znodes_lock, NULL, MUTEX_DEFAULT, NULL);
list_create(&zfsvfs->z_all_znodes, sizeof (znode_t),
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
offsetof(znode_t, z_link_node));
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
size = MIN(1 << (highbit64(zfs_object_mutex_size)-1), ZFS_OBJ_MTX_MAX);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
zfsvfs->z_hold_size = size;
zfsvfs->z_hold_trees = vmem_zalloc(sizeof (avl_tree_t) * size,
KM_SLEEP);
zfsvfs->z_hold_locks = vmem_zalloc(sizeof (kmutex_t) * size, KM_SLEEP);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
for (i = 0; i != size; i++) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
avl_create(&zfsvfs->z_hold_trees[i], zfs_znode_hold_compare,
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
sizeof (znode_hold_t), offsetof(znode_hold_t, zh_node));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
mutex_init(&zfsvfs->z_hold_locks[i], NULL, MUTEX_DEFAULT, NULL);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
}
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
VERIFY(0 == zfs_acl_ids_create(rootzp, IS_ROOT_NODE, &vattr,
cr, NULL, &acl_ids));
zfs_mknode(rootzp, &vattr, tx, cr, IS_ROOT_NODE, &zp, &acl_ids);
ASSERT3P(zp, ==, rootzp);
error = zap_add(os, moid, ZFS_ROOT_OBJ, 8, 1, &rootzp->z_id, tx);
ASSERT(error == 0);
zfs_acl_ids_free(&acl_ids);
Add linux kernel disk support Native Linux vdev disk interfaces Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:45:02 +00:00
Use zfs_mknode() to create dataset root Long, long, long ago when the effort to port ZFS was begun the zfs_create_fs() function was heavily modified to remove all of its VFS dependencies. This allowed Lustre to use the dataset without us having to spend the time porting all the required VFS code. Fast-forward several years and we now have all the VFS code in place but are still relying on the modified zfs_create_fs(). This isn't required anymore and we can now use zfs_mknode() to create the root znode for the filesystem. This commit reverts the contents of zfs_create_fs() to largely match the upstream OpenSolaris code. There have been minor modifications to accomidate the Linux VFS but that is all. This code fixes issue #116 by bootstraping enough of the VFS data structures so we can rely on zfs_mknode() to create the root directory. This ensures it is created properly with support for system attributes. Previously it wasn't which is why it behaved differently that all other directories when modified. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #116
2011-07-20 23:50:22 +00:00
atomic_set(&ZTOI(rootzp)->i_count, 0);
sa_handle_destroy(rootzp->z_sa_hdl);
kmem_cache_free(znode_cache, rootzp);
/*
* Create shares directory
*/
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
error = zfs_create_share_dir(zfsvfs, tx);
Rebase master to b117
2009-07-02 22:44:48 +00:00
ASSERT(error == 0);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
for (i = 0; i != size; i++) {
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
avl_destroy(&zfsvfs->z_hold_trees[i]);
mutex_destroy(&zfsvfs->z_hold_locks[i]);
Fix zsb->z_hold_mtx deadlock The zfs_znode_hold_enter() / zfs_znode_hold_exit() functions are used to serialize access to a znode and its SA buffer while the object is being created or destroyed. This kind of locking would normally reside in the znode itself but in this case that's impossible because the znode and SA buffer may not yet exist. Therefore the locking is handled externally with an array of mutexs and AVLs trees which contain per-object locks. In zfs_znode_hold_enter() a per-object lock is created as needed, inserted in to the correct AVL tree and finally the per-object lock is held. In zfs_znode_hold_exit() the process is reversed. The per-object lock is released, removed from the AVL tree and destroyed if there are no waiters. This scheme has two important properties: 1) No memory allocations are performed while holding one of the z_hold_locks. This ensures evict(), which can be called from direct memory reclaim, will never block waiting on a z_hold_locks which just happens to have hashed to the same index. 2) All locks used to serialize access to an object are per-object and never shared. This minimizes lock contention without creating a large number of dedicated locks. On the downside it does require znode_lock_t structures to be frequently allocated and freed. However, because these are backed by a kmem cache and very short lived this cost is minimal. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #4106
2015-12-22 21:47:38 +00:00
}
Fix usage of zsb after free Caught by code inspection, the variable zsb was referenced after being freed. Move the kmem_free() to the end of the function. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2011-09-09 17:24:55 +00:00
Add missing *_destroy/*_fini calls The proposed debugging enhancements in zfsonlinux/spl#587 identified the following missing *_destroy/*_fini calls. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Gvozden Neskovic <neskovic@gmail.com> Closes #5428
2016-11-26 20:30:44 +00:00
mutex_destroy(&zfsvfs->z_znodes_lock);
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
vmem_free(zfsvfs->z_hold_trees, sizeof (avl_tree_t) * size);
vmem_free(zfsvfs->z_hold_locks, sizeof (kmutex_t) * size);
Fix usage of zsb after free Caught by code inspection, the variable zsb was referenced after being freed. Move the kmem_free() to the end of the function. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2011-09-09 17:24:55 +00:00
kmem_free(sb, sizeof (struct super_block));
Rename zfs_sb_t -> zfsvfs_t The use of zfs_sb_t instead of zfsvfs_t results in unnecessary conflicts with the upstream source. Change all instances of zfs_sb_t to zfsvfs_t including updating the variables names. Whenever possible the code was updated to be consistent with hope it appears in the upstream OpenZFS source. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2017-03-08 00:21:37 +00:00
kmem_free(zfsvfs, sizeof (zfsvfs_t));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
#endif /* _KERNEL */
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
static int
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
zfs_sa_setup(objset_t *osp, sa_attr_type_t **sa_table)
{
uint64_t sa_obj = 0;
int error;
error = zap_lookup(osp, MASTER_NODE_OBJ, ZFS_SA_ATTRS, 8, 1, &sa_obj);
if (error != 0 && error != ENOENT)
return (error);
error = sa_setup(osp, sa_obj, zfs_attr_table, ZPL_END, sa_table);
return (error);
}
static int
zfs_grab_sa_handle(objset_t *osp, uint64_t obj, sa_handle_t **hdlp,
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
dmu_buf_t **db, void *tag)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
dmu_object_info_t doi;
int error;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
if ((error = sa_buf_hold(osp, obj, tag, db)) != 0)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (error);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
dmu_object_info_from_db(*db, &doi);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
if ((doi.doi_bonus_type != DMU_OT_SA &&
doi.doi_bonus_type != DMU_OT_ZNODE) ||
Fix gcc c90 compliance warnings Fix non-c90 compliant code, for the most part these changes simply deal with where a particular variable is declared. Under c90 it must alway be done at the very start of a block. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 16:52:39 +00:00
(doi.doi_bonus_type == DMU_OT_ZNODE &&
doi.doi_bonus_size < sizeof (znode_phys_t))) {
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
sa_buf_rele(*db, tag);
Illumos #3598 3598 want to dtrace when errors are generated in zfs Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Approved by: Garrett D'Amore <garrett@damore.org> References: https://www.illumos.org/issues/3598 illumos/illumos-gate@be6fd75a69ae679453d9cda5bff3326111e6d1ca Ported-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #1775 Porting notes: 1. include/sys/zfs_context.h has been modified to render some new macros inert until dtrace is available on Linux. 2. Linux-specific changes have been adapted to use SET_ERROR(). 3. I'm NOT happy about this change. It does nothing but ugly up the code under Linux. Unfortunately we need to take it to avoid more merge conflicts in the future. -Brian
2013-03-08 18:41:28 +00:00
return (SET_ERROR(ENOTSUP));
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
error = sa_handle_get(osp, obj, NULL, SA_HDL_PRIVATE, hdlp);
if (error != 0) {
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
sa_buf_rele(*db, tag);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
return (error);
}
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
return (0);
}
void
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(sa_handle_t *hdl, dmu_buf_t *db, void *tag)
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
{
sa_handle_destroy(hdl);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
sa_buf_rele(db, tag);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
}
/*
* Given an object number, return its parent object number and whether
* or not the object is an extended attribute directory.
*/
static int
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
zfs_obj_to_pobj(objset_t *osp, sa_handle_t *hdl, sa_attr_type_t *sa_table,
uint64_t *pobjp, int *is_xattrdir)
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
{
uint64_t parent;
uint64_t pflags;
uint64_t mode;
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
uint64_t parent_mode;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
sa_bulk_attr_t bulk[3];
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
sa_handle_t *sa_hdl;
dmu_buf_t *sa_db;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
int count = 0;
int error;
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_PARENT], NULL,
&parent, sizeof (parent));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_FLAGS], NULL,
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
&pflags, sizeof (pflags));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_MODE], NULL,
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
&mode, sizeof (mode));
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if ((error = sa_bulk_lookup(hdl, bulk, count)) != 0)
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
return (error);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
/*
* When a link is removed its parent pointer is not changed and will
* be invalid. There are two cases where a link is removed but the
* file stays around, when it goes to the delete queue and when there
* are additional links.
*/
error = zfs_grab_sa_handle(osp, parent, &sa_hdl, &sa_db, FTAG);
if (error != 0)
return (error);
error = sa_lookup(sa_hdl, ZPL_MODE, &parent_mode, sizeof (parent_mode));
zfs_release_sa_handle(sa_hdl, sa_db, FTAG);
if (error != 0)
return (error);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
*is_xattrdir = ((pflags & ZFS_XATTR) != 0) && S_ISDIR(mode);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
/*
* Extended attributes can be applied to files, directories, etc.
* Otherwise the parent must be a directory.
*/
if (!*is_xattrdir && !S_ISDIR(parent_mode))
return (EINVAL);
*pobjp = parent;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (0);
}
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
/*
* Given an object number, return some zpl level statistics
*/
static int
zfs_obj_to_stats_impl(sa_handle_t *hdl, sa_attr_type_t *sa_table,
zfs_stat_t *sb)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
{
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
sa_bulk_attr_t bulk[4];
int count = 0;
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_MODE], NULL,
&sb->zs_mode, sizeof (sb->zs_mode));
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_GEN], NULL,
&sb->zs_gen, sizeof (sb->zs_gen));
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_LINKS], NULL,
&sb->zs_links, sizeof (sb->zs_links));
SA_ADD_BULK_ATTR(bulk, count, sa_table[ZPL_CTIME], NULL,
&sb->zs_ctime, sizeof (sb->zs_ctime));
return (sa_bulk_lookup(hdl, bulk, count));
}
static int
zfs_obj_to_path_impl(objset_t *osp, uint64_t obj, sa_handle_t *hdl,
sa_attr_type_t *sa_table, char *buf, int len)
{
sa_handle_t *sa_hdl;
sa_handle_t *prevhdl = NULL;
dmu_buf_t *prevdb = NULL;
dmu_buf_t *sa_db = NULL;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
char *path = buf + len - 1;
int error;
*path = '\0';
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
sa_hdl = hdl;
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
for (;;) {
Fix uninitialized variables When compiling on an ARM device using gcc 4.7.3 several variables in the zfs_obj_to_path_impl() function were flagged as uninitialized. To resolve the warnings explicitly initialize them to zero. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1716
2013-09-13 20:10:36 +00:00
uint64_t pobj = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
char component[MAXNAMELEN + 2];
size_t complen;
Fix uninitialized variables When compiling on an ARM device using gcc 4.7.3 several variables in the zfs_obj_to_path_impl() function were flagged as uninitialized. To resolve the warnings explicitly initialize them to zero. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1716
2013-09-13 20:10:36 +00:00
int is_xattrdir = 0;
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (prevdb)
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(prevhdl, prevdb, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
Illumos 3139 - zdb dies when it tries to determine path of unlinked file 3139 zdb dies when it tries to determine path of unlinked file Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com> Reviewed by: Christopher Siden <chris.siden@delphix.com> Reviewed by: Eric Schrock <eric.schrock@delphix.com> Approved by: Dan McDonald <danmcd@nexenta.com> References: https://github.com/illumos/illumos-gate/commit/1ce39b5 https://www.illumos.org/issues/3139 Ported-by: kernelOfTruth kerneloftruth@gmail.com Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2015-12-31 15:41:52 +00:00
if ((error = zfs_obj_to_pobj(osp, sa_hdl, sa_table, &pobj,
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
&is_xattrdir)) != 0)
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
break;
if (pobj == obj) {
if (path[0] != '/')
*--path = '/';
break;
}
component[0] = '/';
if (is_xattrdir) {
(void) sprintf(component + 1, "<xattrdir>");
} else {
error = zap_value_search(osp, pobj, obj,
ZFS_DIRENT_OBJ(-1ULL), component + 1);
if (error != 0)
break;
}
complen = strlen(component);
path -= complen;
ASSERT(path >= buf);
bcopy(component, path, complen);
obj = pobj;
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (sa_hdl != hdl) {
prevhdl = sa_hdl;
prevdb = sa_db;
}
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
error = zfs_grab_sa_handle(osp, obj, &sa_hdl, &sa_db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (error != 0) {
sa_hdl = prevhdl;
sa_db = prevdb;
break;
}
}
if (sa_hdl != NULL && sa_hdl != hdl) {
ASSERT(sa_db != NULL);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(sa_hdl, sa_db, FTAG);
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
}
if (error == 0)
(void) memmove(buf, path, buf + len - path);
Update core ZFS code from build 121 to build 141.
2010-05-28 20:45:14 +00:00
Initial Linux ZFS GIT Repo
2008-11-20 20:01:55 +00:00
return (error);
}
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
int
zfs_obj_to_path(objset_t *osp, uint64_t obj, char *buf, int len)
{
sa_attr_type_t *sa_table;
sa_handle_t *hdl;
dmu_buf_t *db;
int error;
error = zfs_sa_setup(osp, &sa_table);
if (error != 0)
return (error);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
error = zfs_grab_sa_handle(osp, obj, &hdl, &db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (error != 0)
return (error);
error = zfs_obj_to_path_impl(osp, obj, hdl, sa_table, buf, len);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(hdl, db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
return (error);
}
int
zfs_obj_to_stats(objset_t *osp, uint64_t obj, zfs_stat_t *sb,
char *buf, int len)
{
char *path = buf + len - 1;
sa_attr_type_t *sa_table;
sa_handle_t *hdl;
dmu_buf_t *db;
int error;
*path = '\0';
error = zfs_sa_setup(osp, &sa_table);
if (error != 0)
return (error);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
error = zfs_grab_sa_handle(osp, obj, &hdl, &db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
if (error != 0)
return (error);
error = zfs_obj_to_stats_impl(hdl, sa_table, sb);
if (error != 0) {
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(hdl, db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
return (error);
}
error = zfs_obj_to_path_impl(osp, obj, hdl, sa_table, buf, len);
Illumos #xxx: zdb -vvv broken after zfs diff integration References to Illumos issue and patch: - https://github.com/illumos/illumos-gate/commit/163eb7ff Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #340
2011-07-26 23:38:27 +00:00
zfs_release_sa_handle(hdl, db, FTAG);
Update to onnv_147 This is the last official OpenSolaris tag before the public development tree was closed.
2010-08-26 21:24:34 +00:00
return (error);
}
Add linux kernel module support Setup linux kernel module support, this includes: - zfs context for kernel/user - kernel module build system integration - kernel module macros - kernel module symbol export - kernel module options Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:49:16 +00:00
#if defined(_KERNEL) && defined(HAVE_SPL)
EXPORT_SYMBOL(zfs_create_fs);
EXPORT_SYMBOL(zfs_obj_to_path);
Add zfs_object_mutex_size module option Add a zfs_object_mutex_size module option to facilitate resizing the the per-dataset znode mutex array. Increasing this value may help make the deadlock described in #4106 less common, but this is not a proper fix. This patch is primarily to aid debugging and analysis. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Issue #4106
2015-12-18 20:19:14 +00:00
Use cstyle -cpP in `make cstyle` check Enable picky cstyle checks and resolve the new warnings. The vast majority of the changes needed were to handle minor issues with whitespace formatting. This patch contains no functional changes. Non-whitespace changes are as follows: * 8 times ; to { } in for/while loop * fix missing ; in cmd/zed/agents/zfs_diagnosis.c * comment (confim -> confirm) * change endline , to ; in cmd/zpool/zpool_main.c * a number of /* BEGIN CSTYLED */ /* END CSTYLED */ blocks * /* CSTYLED */ markers * change == 0 to ! * ulong to unsigned long in module/zfs/dsl_scan.c * rearrangement of module_param lines in module/zfs/metaslab.c * add { } block around statement after for_each_online_node Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Håkan Johansson <f96hajo@chalmers.se> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5465
2016-12-12 18:46:26 +00:00
/* CSTYLED */
Add zfs_object_mutex_size module option Add a zfs_object_mutex_size module option to facilitate resizing the the per-dataset znode mutex array. Increasing this value may help make the deadlock described in #4106 less common, but this is not a proper fix. This patch is primarily to aid debugging and analysis. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Tim Chase <tim@chase2k.com> Issue #4106
2015-12-18 20:19:14 +00:00
module_param(zfs_object_mutex_size, uint, 0644);
MODULE_PARM_DESC(zfs_object_mutex_size, "Size of znode hold array");
Add linux kernel module support Setup linux kernel module support, this includes: - zfs context for kernel/user - kernel module build system integration - kernel module macros - kernel module symbol export - kernel module options Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-08-26 18:49:16 +00:00
#endif