freebsd-dev/module/zfs/dmu.c

2116 lines
52 KiB
C
Raw Normal View History

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
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2011, 2016 by Delphix. All rights reserved.
* Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
* Copyright (c) 2013, Joyent, Inc. All rights reserved.
* Copyright (c) 2014, Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2015 by Chunwei Chen. All rights reserved.
2008-11-20 20:01:55 +00:00
*/
#include <sys/dmu.h>
#include <sys/dmu_impl.h>
#include <sys/dmu_tx.h>
#include <sys/dbuf.h>
#include <sys/dnode.h>
#include <sys/zfs_context.h>
#include <sys/dmu_objset.h>
#include <sys/dmu_traverse.h>
#include <sys/dsl_dataset.h>
#include <sys/dsl_dir.h>
#include <sys/dsl_pool.h>
#include <sys/dsl_synctask.h>
#include <sys/dsl_prop.h>
#include <sys/dmu_zfetch.h>
#include <sys/zfs_ioctl.h>
#include <sys/zap.h>
#include <sys/zio_checksum.h>
#include <sys/zio_compress.h>
#include <sys/sa.h>
#include <sys/zfeature.h>
2008-11-20 20:01:55 +00:00
#ifdef _KERNEL
#include <sys/vmsystm.h>
#include <sys/zfs_znode.h>
2008-11-20 20:01:55 +00:00
#endif
/*
* Enable/disable nopwrite feature.
*/
int zfs_nopwrite_enabled = 1;
2008-11-20 20:01:55 +00:00
const dmu_object_type_info_t dmu_ot[DMU_OT_NUMTYPES] = {
{ DMU_BSWAP_UINT8, TRUE, "unallocated" },
{ DMU_BSWAP_ZAP, TRUE, "object directory" },
{ DMU_BSWAP_UINT64, TRUE, "object array" },
{ DMU_BSWAP_UINT8, TRUE, "packed nvlist" },
{ DMU_BSWAP_UINT64, TRUE, "packed nvlist size" },
{ DMU_BSWAP_UINT64, TRUE, "bpobj" },
{ DMU_BSWAP_UINT64, TRUE, "bpobj header" },
{ DMU_BSWAP_UINT64, TRUE, "SPA space map header" },
{ DMU_BSWAP_UINT64, TRUE, "SPA space map" },
{ DMU_BSWAP_UINT64, TRUE, "ZIL intent log" },
{ DMU_BSWAP_DNODE, TRUE, "DMU dnode" },
{ DMU_BSWAP_OBJSET, TRUE, "DMU objset" },
{ DMU_BSWAP_UINT64, TRUE, "DSL directory" },
{ DMU_BSWAP_ZAP, TRUE, "DSL directory child map"},
{ DMU_BSWAP_ZAP, TRUE, "DSL dataset snap map" },
{ DMU_BSWAP_ZAP, TRUE, "DSL props" },
{ DMU_BSWAP_UINT64, TRUE, "DSL dataset" },
{ DMU_BSWAP_ZNODE, TRUE, "ZFS znode" },
{ DMU_BSWAP_OLDACL, TRUE, "ZFS V0 ACL" },
{ DMU_BSWAP_UINT8, FALSE, "ZFS plain file" },
{ DMU_BSWAP_ZAP, TRUE, "ZFS directory" },
{ DMU_BSWAP_ZAP, TRUE, "ZFS master node" },
{ DMU_BSWAP_ZAP, TRUE, "ZFS delete queue" },
{ DMU_BSWAP_UINT8, FALSE, "zvol object" },
{ DMU_BSWAP_ZAP, TRUE, "zvol prop" },
{ DMU_BSWAP_UINT8, FALSE, "other uint8[]" },
{ DMU_BSWAP_UINT64, FALSE, "other uint64[]" },
{ DMU_BSWAP_ZAP, TRUE, "other ZAP" },
{ DMU_BSWAP_ZAP, TRUE, "persistent error log" },
{ DMU_BSWAP_UINT8, TRUE, "SPA history" },
{ DMU_BSWAP_UINT64, TRUE, "SPA history offsets" },
{ DMU_BSWAP_ZAP, TRUE, "Pool properties" },
{ DMU_BSWAP_ZAP, TRUE, "DSL permissions" },
{ DMU_BSWAP_ACL, TRUE, "ZFS ACL" },
{ DMU_BSWAP_UINT8, TRUE, "ZFS SYSACL" },
{ DMU_BSWAP_UINT8, TRUE, "FUID table" },
{ DMU_BSWAP_UINT64, TRUE, "FUID table size" },
{ DMU_BSWAP_ZAP, TRUE, "DSL dataset next clones"},
{ DMU_BSWAP_ZAP, TRUE, "scan work queue" },
{ DMU_BSWAP_ZAP, TRUE, "ZFS user/group used" },
{ DMU_BSWAP_ZAP, TRUE, "ZFS user/group quota" },
{ DMU_BSWAP_ZAP, TRUE, "snapshot refcount tags"},
{ DMU_BSWAP_ZAP, TRUE, "DDT ZAP algorithm" },
{ DMU_BSWAP_ZAP, TRUE, "DDT statistics" },
{ DMU_BSWAP_UINT8, TRUE, "System attributes" },
{ DMU_BSWAP_ZAP, TRUE, "SA master node" },
{ DMU_BSWAP_ZAP, TRUE, "SA attr registration" },
{ DMU_BSWAP_ZAP, TRUE, "SA attr layouts" },
{ DMU_BSWAP_ZAP, TRUE, "scan translations" },
{ DMU_BSWAP_UINT8, FALSE, "deduplicated block" },
{ DMU_BSWAP_ZAP, TRUE, "DSL deadlist map" },
{ DMU_BSWAP_UINT64, TRUE, "DSL deadlist map hdr" },
{ DMU_BSWAP_ZAP, TRUE, "DSL dir clones" },
{ DMU_BSWAP_UINT64, TRUE, "bpobj subobj" }
};
const dmu_object_byteswap_info_t dmu_ot_byteswap[DMU_BSWAP_NUMFUNCS] = {
{ byteswap_uint8_array, "uint8" },
{ byteswap_uint16_array, "uint16" },
{ byteswap_uint32_array, "uint32" },
{ byteswap_uint64_array, "uint64" },
{ zap_byteswap, "zap" },
{ dnode_buf_byteswap, "dnode" },
{ dmu_objset_byteswap, "objset" },
{ zfs_znode_byteswap, "znode" },
{ zfs_oldacl_byteswap, "oldacl" },
{ zfs_acl_byteswap, "acl" }
2008-11-20 20:01:55 +00:00
};
OpenZFS 7004 - dmu_tx_hold_zap() does dnode_hold() 7x on same object Using a benchmark which has 32 threads creating 2 million files in the same directory, on a machine with 16 CPU cores, I observed poor performance. I noticed that dmu_tx_hold_zap() was using about 30% of all CPU, and doing dnode_hold() 7 times on the same object (the ZAP object that is being held). dmu_tx_hold_zap() keeps a hold on the dnode_t the entire time it is running, in dmu_tx_hold_t:txh_dnode, so it would be nice to use the dnode_t that we already have in hand, rather than repeatedly calling dnode_hold(). To do this, we need to pass the dnode_t down through all the intermediate calls that dmu_tx_hold_zap() makes, making these routines take the dnode_t* rather than an objset_t* and a uint64_t object number. In particular, the following routines will need to have analogous *_by_dnode() variants created: dmu_buf_hold_noread() dmu_buf_hold() zap_lookup() zap_lookup_norm() zap_count_write() zap_lockdir() zap_count_write() This can improve performance on the benchmark described above by 100%, from 30,000 file creations per second to 60,000. (This improvement is on top of that provided by working around the object allocation issue. Peak performance of ~90,000 creations per second was observed with 8 CPUs; adding CPUs past that decreased performance due to lock contention.) The CPU used by dmu_tx_hold_zap() was reduced by 88%, from 340 CPU-seconds to 40 CPU-seconds. Sponsored by: Intel Corp. Signed-off-by: Matthew Ahrens <mahrens@delphix.com> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/7004 OpenZFS-commit: https://github.com/openzfs/openzfs/pull/109 Closes #4641 Closes #4972
2016-07-20 22:42:13 +00:00
int
dmu_buf_hold_noread_by_dnode(dnode_t *dn, uint64_t offset,
void *tag, dmu_buf_t **dbp)
{
uint64_t blkid;
dmu_buf_impl_t *db;
blkid = dbuf_whichblock(dn, 0, offset);
rw_enter(&dn->dn_struct_rwlock, RW_READER);
db = dbuf_hold(dn, blkid, tag);
rw_exit(&dn->dn_struct_rwlock);
if (db == NULL) {
*dbp = NULL;
return (SET_ERROR(EIO));
}
*dbp = &db->db;
return (0);
}
2008-11-20 20:01:55 +00:00
int
dmu_buf_hold_noread(objset_t *os, uint64_t object, uint64_t offset,
void *tag, dmu_buf_t **dbp)
2008-11-20 20:01:55 +00:00
{
dnode_t *dn;
uint64_t blkid;
dmu_buf_impl_t *db;
int err;
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
blkid = dbuf_whichblock(dn, 0, offset);
2008-11-20 20:01:55 +00:00
rw_enter(&dn->dn_struct_rwlock, RW_READER);
db = dbuf_hold(dn, blkid, tag);
rw_exit(&dn->dn_struct_rwlock);
dnode_rele(dn, FTAG);
2008-11-20 20:01:55 +00:00
if (db == NULL) {
*dbp = NULL;
return (SET_ERROR(EIO));
}
*dbp = &db->db;
return (err);
}
OpenZFS 7004 - dmu_tx_hold_zap() does dnode_hold() 7x on same object Using a benchmark which has 32 threads creating 2 million files in the same directory, on a machine with 16 CPU cores, I observed poor performance. I noticed that dmu_tx_hold_zap() was using about 30% of all CPU, and doing dnode_hold() 7 times on the same object (the ZAP object that is being held). dmu_tx_hold_zap() keeps a hold on the dnode_t the entire time it is running, in dmu_tx_hold_t:txh_dnode, so it would be nice to use the dnode_t that we already have in hand, rather than repeatedly calling dnode_hold(). To do this, we need to pass the dnode_t down through all the intermediate calls that dmu_tx_hold_zap() makes, making these routines take the dnode_t* rather than an objset_t* and a uint64_t object number. In particular, the following routines will need to have analogous *_by_dnode() variants created: dmu_buf_hold_noread() dmu_buf_hold() zap_lookup() zap_lookup_norm() zap_count_write() zap_lockdir() zap_count_write() This can improve performance on the benchmark described above by 100%, from 30,000 file creations per second to 60,000. (This improvement is on top of that provided by working around the object allocation issue. Peak performance of ~90,000 creations per second was observed with 8 CPUs; adding CPUs past that decreased performance due to lock contention.) The CPU used by dmu_tx_hold_zap() was reduced by 88%, from 340 CPU-seconds to 40 CPU-seconds. Sponsored by: Intel Corp. Signed-off-by: Matthew Ahrens <mahrens@delphix.com> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/7004 OpenZFS-commit: https://github.com/openzfs/openzfs/pull/109 Closes #4641 Closes #4972
2016-07-20 22:42:13 +00:00
int
dmu_buf_hold_by_dnode(dnode_t *dn, uint64_t offset,
void *tag, dmu_buf_t **dbp, int flags)
{
int err;
int db_flags = DB_RF_CANFAIL;
if (flags & DMU_READ_NO_PREFETCH)
db_flags |= DB_RF_NOPREFETCH;
err = dmu_buf_hold_noread_by_dnode(dn, offset, tag, dbp);
if (err == 0) {
dmu_buf_impl_t *db = (dmu_buf_impl_t *)(*dbp);
err = dbuf_read(db, NULL, db_flags);
if (err != 0) {
dbuf_rele(db, tag);
*dbp = NULL;
}
}
return (err);
}
int
dmu_buf_hold(objset_t *os, uint64_t object, uint64_t offset,
void *tag, dmu_buf_t **dbp, int flags)
{
int err;
int db_flags = DB_RF_CANFAIL;
if (flags & DMU_READ_NO_PREFETCH)
db_flags |= DB_RF_NOPREFETCH;
err = dmu_buf_hold_noread(os, object, offset, tag, dbp);
if (err == 0) {
dmu_buf_impl_t *db = (dmu_buf_impl_t *)(*dbp);
err = dbuf_read(db, NULL, db_flags);
if (err != 0) {
2008-11-20 20:01:55 +00:00
dbuf_rele(db, tag);
*dbp = NULL;
2008-11-20 20:01:55 +00:00
}
}
return (err);
}
int
dmu_bonus_max(void)
{
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
return (DN_OLD_MAX_BONUSLEN);
2008-11-20 20:01:55 +00:00
}
int
dmu_set_bonus(dmu_buf_t *db_fake, int newsize, dmu_tx_t *tx)
2008-11-20 20:01:55 +00:00
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
int error;
2008-11-20 20:01:55 +00:00
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
if (dn->dn_bonus != db) {
error = SET_ERROR(EINVAL);
} else if (newsize < 0 || newsize > db_fake->db_size) {
error = SET_ERROR(EINVAL);
} else {
dnode_setbonuslen(dn, newsize, tx);
error = 0;
}
DB_DNODE_EXIT(db);
return (error);
2008-11-20 20:01:55 +00:00
}
int
dmu_set_bonustype(dmu_buf_t *db_fake, dmu_object_type_t type, dmu_tx_t *tx)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
int error;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
if (!DMU_OT_IS_VALID(type)) {
error = SET_ERROR(EINVAL);
} else if (dn->dn_bonus != db) {
error = SET_ERROR(EINVAL);
} else {
dnode_setbonus_type(dn, type, tx);
error = 0;
}
DB_DNODE_EXIT(db);
return (error);
}
dmu_object_type_t
dmu_get_bonustype(dmu_buf_t *db_fake)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
dmu_object_type_t type;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
type = dn->dn_bonustype;
DB_DNODE_EXIT(db);
return (type);
}
int
dmu_rm_spill(objset_t *os, uint64_t object, dmu_tx_t *tx)
{
dnode_t *dn;
int error;
error = dnode_hold(os, object, FTAG, &dn);
dbuf_rm_spill(dn, tx);
rw_enter(&dn->dn_struct_rwlock, RW_WRITER);
dnode_rm_spill(dn, tx);
rw_exit(&dn->dn_struct_rwlock);
dnode_rele(dn, FTAG);
return (error);
}
2008-11-20 20:01:55 +00:00
/*
* returns ENOENT, EIO, or 0.
*/
int
dmu_bonus_hold(objset_t *os, uint64_t object, void *tag, dmu_buf_t **dbp)
{
dnode_t *dn;
dmu_buf_impl_t *db;
int error;
error = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (error)
return (error);
rw_enter(&dn->dn_struct_rwlock, RW_READER);
if (dn->dn_bonus == NULL) {
rw_exit(&dn->dn_struct_rwlock);
rw_enter(&dn->dn_struct_rwlock, RW_WRITER);
if (dn->dn_bonus == NULL)
dbuf_create_bonus(dn);
}
db = dn->dn_bonus;
/* as long as the bonus buf is held, the dnode will be held */
if (refcount_add(&db->db_holds, tag) == 1) {
2008-11-20 20:01:55 +00:00
VERIFY(dnode_add_ref(dn, db));
atomic_inc_32(&dn->dn_dbufs_count);
}
/*
* Wait to drop dn_struct_rwlock until after adding the bonus dbuf's
* hold and incrementing the dbuf count to ensure that dnode_move() sees
* a dnode hold for every dbuf.
*/
rw_exit(&dn->dn_struct_rwlock);
2008-11-20 20:01:55 +00:00
dnode_rele(dn, FTAG);
VERIFY(0 == dbuf_read(db, NULL, DB_RF_MUST_SUCCEED | DB_RF_NOPREFETCH));
2008-11-20 20:01:55 +00:00
*dbp = &db->db;
return (0);
}
/*
* returns ENOENT, EIO, or 0.
*
* This interface will allocate a blank spill dbuf when a spill blk
* doesn't already exist on the dnode.
*
* if you only want to find an already existing spill db, then
* dmu_spill_hold_existing() should be used.
*/
int
dmu_spill_hold_by_dnode(dnode_t *dn, uint32_t flags, void *tag, dmu_buf_t **dbp)
{
dmu_buf_impl_t *db = NULL;
int err;
if ((flags & DB_RF_HAVESTRUCT) == 0)
rw_enter(&dn->dn_struct_rwlock, RW_READER);
db = dbuf_hold(dn, DMU_SPILL_BLKID, tag);
if ((flags & DB_RF_HAVESTRUCT) == 0)
rw_exit(&dn->dn_struct_rwlock);
ASSERT(db != NULL);
err = dbuf_read(db, NULL, flags);
if (err == 0)
*dbp = &db->db;
else
dbuf_rele(db, tag);
return (err);
}
int
dmu_spill_hold_existing(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)bonus;
dnode_t *dn;
int err;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
if (spa_version(dn->dn_objset->os_spa) < SPA_VERSION_SA) {
err = SET_ERROR(EINVAL);
} else {
rw_enter(&dn->dn_struct_rwlock, RW_READER);
if (!dn->dn_have_spill) {
err = SET_ERROR(ENOENT);
} else {
err = dmu_spill_hold_by_dnode(dn,
DB_RF_HAVESTRUCT | DB_RF_CANFAIL, tag, dbp);
}
rw_exit(&dn->dn_struct_rwlock);
}
DB_DNODE_EXIT(db);
return (err);
}
int
dmu_spill_hold_by_bonus(dmu_buf_t *bonus, void *tag, dmu_buf_t **dbp)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)bonus;
dnode_t *dn;
int err;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
err = dmu_spill_hold_by_dnode(dn, DB_RF_CANFAIL, tag, dbp);
DB_DNODE_EXIT(db);
return (err);
}
2008-11-20 20:01:55 +00:00
/*
* Note: longer-term, we should modify all of the dmu_buf_*() interfaces
* to take a held dnode rather than <os, object> -- the lookup is wasteful,
* and can induce severe lock contention when writing to several files
* whose dnodes are in the same block.
*/
static int
2009-07-02 22:44:48 +00:00
dmu_buf_hold_array_by_dnode(dnode_t *dn, uint64_t offset, uint64_t length,
boolean_t read, void *tag, int *numbufsp, dmu_buf_t ***dbpp, uint32_t flags)
2008-11-20 20:01:55 +00:00
{
dmu_buf_t **dbp;
uint64_t blkid, nblks, i;
2009-07-02 22:44:48 +00:00
uint32_t dbuf_flags;
2008-11-20 20:01:55 +00:00
int err;
zio_t *zio;
ASSERT(length <= DMU_MAX_ACCESS);
/*
* Note: We directly notify the prefetch code of this read, so that
* we can tell it about the multi-block read. dbuf_read() only knows
* about the one block it is accessing.
*/
dbuf_flags = DB_RF_CANFAIL | DB_RF_NEVERWAIT | DB_RF_HAVESTRUCT |
DB_RF_NOPREFETCH;
2008-11-20 20:01:55 +00:00
rw_enter(&dn->dn_struct_rwlock, RW_READER);
if (dn->dn_datablkshift) {
int blkshift = dn->dn_datablkshift;
nblks = (P2ROUNDUP(offset + length, 1ULL << blkshift) -
P2ALIGN(offset, 1ULL << blkshift)) >> blkshift;
2008-11-20 20:01:55 +00:00
} else {
if (offset + length > dn->dn_datablksz) {
zfs_panic_recover("zfs: accessing past end of object "
"%llx/%llx (size=%u access=%llu+%llu)",
(longlong_t)dn->dn_objset->
os_dsl_dataset->ds_object,
(longlong_t)dn->dn_object, dn->dn_datablksz,
(longlong_t)offset, (longlong_t)length);
2009-08-18 18:43:27 +00:00
rw_exit(&dn->dn_struct_rwlock);
return (SET_ERROR(EIO));
2008-11-20 20:01:55 +00:00
}
nblks = 1;
}
dbp = kmem_zalloc(sizeof (dmu_buf_t *) * nblks, KM_SLEEP);
2008-11-20 20:01:55 +00:00
zio = zio_root(dn->dn_objset->os_spa, NULL, NULL, ZIO_FLAG_CANFAIL);
blkid = dbuf_whichblock(dn, 0, offset);
2008-11-20 20:01:55 +00:00
for (i = 0; i < nblks; i++) {
dmu_buf_impl_t *db = dbuf_hold(dn, blkid + i, tag);
2008-11-20 20:01:55 +00:00
if (db == NULL) {
rw_exit(&dn->dn_struct_rwlock);
dmu_buf_rele_array(dbp, nblks, tag);
zio_nowait(zio);
return (SET_ERROR(EIO));
2008-11-20 20:01:55 +00:00
}
2008-11-20 20:01:55 +00:00
/* initiate async i/o */
if (read)
2009-07-02 22:44:48 +00:00
(void) dbuf_read(db, zio, dbuf_flags);
2008-11-20 20:01:55 +00:00
dbp[i] = &db->db;
}
if ((flags & DMU_READ_NO_PREFETCH) == 0 &&
DNODE_META_IS_CACHEABLE(dn) && length <= zfetch_array_rd_sz) {
dmu_zfetch(&dn->dn_zfetch, blkid, nblks,
read && DNODE_IS_CACHEABLE(dn));
}
2008-11-20 20:01:55 +00:00
rw_exit(&dn->dn_struct_rwlock);
/* wait for async i/o */
err = zio_wait(zio);
if (err) {
dmu_buf_rele_array(dbp, nblks, tag);
return (err);
}
/* wait for other io to complete */
if (read) {
for (i = 0; i < nblks; i++) {
dmu_buf_impl_t *db = (dmu_buf_impl_t *)dbp[i];
mutex_enter(&db->db_mtx);
while (db->db_state == DB_READ ||
db->db_state == DB_FILL)
cv_wait(&db->db_changed, &db->db_mtx);
if (db->db_state == DB_UNCACHED)
err = SET_ERROR(EIO);
2008-11-20 20:01:55 +00:00
mutex_exit(&db->db_mtx);
if (err) {
dmu_buf_rele_array(dbp, nblks, tag);
return (err);
}
}
}
*numbufsp = nblks;
*dbpp = dbp;
return (0);
}
static int
dmu_buf_hold_array(objset_t *os, uint64_t object, uint64_t offset,
uint64_t length, int read, void *tag, int *numbufsp, dmu_buf_t ***dbpp)
{
dnode_t *dn;
int err;
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
err = dmu_buf_hold_array_by_dnode(dn, offset, length, read, tag,
2009-07-02 22:44:48 +00:00
numbufsp, dbpp, DMU_READ_PREFETCH);
2008-11-20 20:01:55 +00:00
dnode_rele(dn, FTAG);
return (err);
}
int
dmu_buf_hold_array_by_bonus(dmu_buf_t *db_fake, uint64_t offset,
uint64_t length, boolean_t read, void *tag, int *numbufsp,
dmu_buf_t ***dbpp)
2008-11-20 20:01:55 +00:00
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
2008-11-20 20:01:55 +00:00
int err;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
2008-11-20 20:01:55 +00:00
err = dmu_buf_hold_array_by_dnode(dn, offset, length, read, tag,
2009-07-02 22:44:48 +00:00
numbufsp, dbpp, DMU_READ_PREFETCH);
DB_DNODE_EXIT(db);
2008-11-20 20:01:55 +00:00
return (err);
}
void
dmu_buf_rele_array(dmu_buf_t **dbp_fake, int numbufs, void *tag)
{
int i;
dmu_buf_impl_t **dbp = (dmu_buf_impl_t **)dbp_fake;
if (numbufs == 0)
return;
for (i = 0; i < numbufs; i++) {
if (dbp[i])
dbuf_rele(dbp[i], tag);
}
kmem_free(dbp, sizeof (dmu_buf_t *) * numbufs);
}
Illumos #4045 write throttle & i/o scheduler performance work 4045 zfs write throttle & i/o scheduler performance work 1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync read, sync write, async read, async write, and scrub/resilver. The scheduler issues a number of concurrent i/os from each class to the device. Once a class has been selected, an i/o is selected from this class using either an elevator algorithem (async, scrub classes) or FIFO (sync classes). The number of concurrent async write i/os is tuned dynamically based on i/o load, to achieve good sync i/o latency when there is not a high load of writes, and good write throughput when there is. See the block comment in vdev_queue.c (reproduced below) for more details. 2. The write throttle (dsl_pool_tempreserve_space() and txg_constrain_throughput()) is rewritten to produce much more consistent delays when under constant load. The new write throttle is based on the amount of dirty data, rather than guesses about future performance of the system. When there is a lot of dirty data, each transaction (e.g. write() syscall) will be delayed by the same small amount. This eliminates the "brick wall of wait" that the old write throttle could hit, causing all transactions to wait several seconds until the next txg opens. One of the keys to the new write throttle is decrementing the amount of dirty data as i/o completes, rather than at the end of spa_sync(). Note that the write throttle is only applied once the i/o scheduler is issuing the maximum number of outstanding async writes. See the block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for more details. This diff has several other effects, including: * the commonly-tuned global variable zfs_vdev_max_pending has been removed; use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead. * the size of each txg (meaning the amount of dirty data written, and thus the time it takes to write out) is now controlled differently. There is no longer an explicit time goal; the primary determinant is amount of dirty data. Systems that are under light or medium load will now often see that a txg is always syncing, but the impact to performance (e.g. read latency) is minimal. Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this. * zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression, checksum, etc. This improves latency by not allowing these CPU-intensive tasks to consume all CPU (on machines with at least 4 CPU's; the percentage is rounded up). --matt APPENDIX: problems with the current i/o scheduler The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem with this is that if there are always i/os pending, then certain classes of i/os can see very long delays. For example, if there are always synchronous reads outstanding, then no async writes will be serviced until they become "past due". One symptom of this situation is that each pass of the txg sync takes at least several seconds (typically 3 seconds). If many i/os become "past due" (their deadline is in the past), then we must service all of these overdue i/os before any new i/os. This happens when we enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in the future. If we can't complete all the i/os in 2.5 seconds (e.g. because there were always reads pending), then these i/os will become past due. Now we must service all the "async" writes (which could be hundreds of megabytes) before we service any reads, introducing considerable latency to synchronous i/os (reads or ZIL writes). Notes on porting to ZFS on Linux: - zio_t gained new members io_physdone and io_phys_children. Because object caches in the Linux port call the constructor only once at allocation time, objects may contain residual data when retrieved from the cache. Therefore zio_create() was updated to zero out the two new fields. - vdev_mirror_pending() relied on the depth of the per-vdev pending queue (vq->vq_pending_tree) to select the least-busy leaf vdev to read from. This tree has been replaced by vq->vq_active_tree which is now used for the same purpose. - vdev_queue_init() used the value of zfs_vdev_max_pending to determine the number of vdev I/O buffers to pre-allocate. That global no longer exists, so we instead use the sum of the *_max_active values for each of the five I/O classes described above. - The Illumos implementation of dmu_tx_delay() delays a transaction by sleeping in condition variable embedded in the thread (curthread->t_delay_cv). We do not have an equivalent CV to use in Linux, so this change replaced the delay logic with a wrapper called zfs_sleep_until(). This wrapper could be adopted upstream and in other downstream ports to abstract away operating system-specific delay logic. - These tunables are added as module parameters, and descriptions added to the zfs-module-parameters.5 man page. spa_asize_inflation zfs_deadman_synctime_ms zfs_vdev_max_active zfs_vdev_async_write_active_min_dirty_percent zfs_vdev_async_write_active_max_dirty_percent zfs_vdev_async_read_max_active zfs_vdev_async_read_min_active zfs_vdev_async_write_max_active zfs_vdev_async_write_min_active zfs_vdev_scrub_max_active zfs_vdev_scrub_min_active zfs_vdev_sync_read_max_active zfs_vdev_sync_read_min_active zfs_vdev_sync_write_max_active zfs_vdev_sync_write_min_active zfs_dirty_data_max_percent zfs_delay_min_dirty_percent zfs_dirty_data_max_max_percent zfs_dirty_data_max zfs_dirty_data_max_max zfs_dirty_data_sync zfs_delay_scale The latter four have type unsigned long, whereas they are uint64_t in Illumos. This accommodates Linux's module_param() supported types, but means they may overflow on 32-bit architectures. The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most likely to overflow on 32-bit systems, since they express physical RAM sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to 2^32 which does overflow. To resolve that, this port instead initializes it in arc_init() to 25% of physical RAM, and adds the tunable zfs_dirty_data_max_max_percent to override that percentage. While this solution doesn't completely avoid the overflow issue, it should be a reasonable default for most systems, and the minority of affected systems can work around the issue by overriding the defaults. - Fixed reversed logic in comment above zfs_delay_scale declaration. - Clarified comments in vdev_queue.c regarding when per-queue minimums take effect. - Replaced dmu_tx_write_limit in the dmu_tx kstat file with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts how many times a transaction has been delayed because the pool dirty data has exceeded zfs_delay_min_dirty_percent. The latter counts how many times the pool dirty data has exceeded zfs_dirty_data_max (which we expect to never happen). - The original patch would have regressed the bug fixed in zfsonlinux/zfs@c418410, which prevented users from setting the zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE. A similar fix is added to vdev_queue_aggregate(). - In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the heap instead of the stack. In Linux we can't afford such large structures on the stack. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Brendan Gregg <brendan.gregg@joyent.com> Approved by: Robert Mustacchi <rm@joyent.com> References: http://www.illumos.org/issues/4045 illumos/illumos-gate@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
/*
* Issue prefetch i/os for the given blocks. If level is greater than 0, the
* indirect blocks prefeteched will be those that point to the blocks containing
* the data starting at offset, and continuing to offset + len.
Illumos #4045 write throttle & i/o scheduler performance work 4045 zfs write throttle & i/o scheduler performance work 1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync read, sync write, async read, async write, and scrub/resilver. The scheduler issues a number of concurrent i/os from each class to the device. Once a class has been selected, an i/o is selected from this class using either an elevator algorithem (async, scrub classes) or FIFO (sync classes). The number of concurrent async write i/os is tuned dynamically based on i/o load, to achieve good sync i/o latency when there is not a high load of writes, and good write throughput when there is. See the block comment in vdev_queue.c (reproduced below) for more details. 2. The write throttle (dsl_pool_tempreserve_space() and txg_constrain_throughput()) is rewritten to produce much more consistent delays when under constant load. The new write throttle is based on the amount of dirty data, rather than guesses about future performance of the system. When there is a lot of dirty data, each transaction (e.g. write() syscall) will be delayed by the same small amount. This eliminates the "brick wall of wait" that the old write throttle could hit, causing all transactions to wait several seconds until the next txg opens. One of the keys to the new write throttle is decrementing the amount of dirty data as i/o completes, rather than at the end of spa_sync(). Note that the write throttle is only applied once the i/o scheduler is issuing the maximum number of outstanding async writes. See the block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for more details. This diff has several other effects, including: * the commonly-tuned global variable zfs_vdev_max_pending has been removed; use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead. * the size of each txg (meaning the amount of dirty data written, and thus the time it takes to write out) is now controlled differently. There is no longer an explicit time goal; the primary determinant is amount of dirty data. Systems that are under light or medium load will now often see that a txg is always syncing, but the impact to performance (e.g. read latency) is minimal. Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this. * zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression, checksum, etc. This improves latency by not allowing these CPU-intensive tasks to consume all CPU (on machines with at least 4 CPU's; the percentage is rounded up). --matt APPENDIX: problems with the current i/o scheduler The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem with this is that if there are always i/os pending, then certain classes of i/os can see very long delays. For example, if there are always synchronous reads outstanding, then no async writes will be serviced until they become "past due". One symptom of this situation is that each pass of the txg sync takes at least several seconds (typically 3 seconds). If many i/os become "past due" (their deadline is in the past), then we must service all of these overdue i/os before any new i/os. This happens when we enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in the future. If we can't complete all the i/os in 2.5 seconds (e.g. because there were always reads pending), then these i/os will become past due. Now we must service all the "async" writes (which could be hundreds of megabytes) before we service any reads, introducing considerable latency to synchronous i/os (reads or ZIL writes). Notes on porting to ZFS on Linux: - zio_t gained new members io_physdone and io_phys_children. Because object caches in the Linux port call the constructor only once at allocation time, objects may contain residual data when retrieved from the cache. Therefore zio_create() was updated to zero out the two new fields. - vdev_mirror_pending() relied on the depth of the per-vdev pending queue (vq->vq_pending_tree) to select the least-busy leaf vdev to read from. This tree has been replaced by vq->vq_active_tree which is now used for the same purpose. - vdev_queue_init() used the value of zfs_vdev_max_pending to determine the number of vdev I/O buffers to pre-allocate. That global no longer exists, so we instead use the sum of the *_max_active values for each of the five I/O classes described above. - The Illumos implementation of dmu_tx_delay() delays a transaction by sleeping in condition variable embedded in the thread (curthread->t_delay_cv). We do not have an equivalent CV to use in Linux, so this change replaced the delay logic with a wrapper called zfs_sleep_until(). This wrapper could be adopted upstream and in other downstream ports to abstract away operating system-specific delay logic. - These tunables are added as module parameters, and descriptions added to the zfs-module-parameters.5 man page. spa_asize_inflation zfs_deadman_synctime_ms zfs_vdev_max_active zfs_vdev_async_write_active_min_dirty_percent zfs_vdev_async_write_active_max_dirty_percent zfs_vdev_async_read_max_active zfs_vdev_async_read_min_active zfs_vdev_async_write_max_active zfs_vdev_async_write_min_active zfs_vdev_scrub_max_active zfs_vdev_scrub_min_active zfs_vdev_sync_read_max_active zfs_vdev_sync_read_min_active zfs_vdev_sync_write_max_active zfs_vdev_sync_write_min_active zfs_dirty_data_max_percent zfs_delay_min_dirty_percent zfs_dirty_data_max_max_percent zfs_dirty_data_max zfs_dirty_data_max_max zfs_dirty_data_sync zfs_delay_scale The latter four have type unsigned long, whereas they are uint64_t in Illumos. This accommodates Linux's module_param() supported types, but means they may overflow on 32-bit architectures. The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most likely to overflow on 32-bit systems, since they express physical RAM sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to 2^32 which does overflow. To resolve that, this port instead initializes it in arc_init() to 25% of physical RAM, and adds the tunable zfs_dirty_data_max_max_percent to override that percentage. While this solution doesn't completely avoid the overflow issue, it should be a reasonable default for most systems, and the minority of affected systems can work around the issue by overriding the defaults. - Fixed reversed logic in comment above zfs_delay_scale declaration. - Clarified comments in vdev_queue.c regarding when per-queue minimums take effect. - Replaced dmu_tx_write_limit in the dmu_tx kstat file with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts how many times a transaction has been delayed because the pool dirty data has exceeded zfs_delay_min_dirty_percent. The latter counts how many times the pool dirty data has exceeded zfs_dirty_data_max (which we expect to never happen). - The original patch would have regressed the bug fixed in zfsonlinux/zfs@c418410, which prevented users from setting the zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE. A similar fix is added to vdev_queue_aggregate(). - In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the heap instead of the stack. In Linux we can't afford such large structures on the stack. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Brendan Gregg <brendan.gregg@joyent.com> Approved by: Robert Mustacchi <rm@joyent.com> References: http://www.illumos.org/issues/4045 illumos/illumos-gate@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
*
* Note that if the indirect blocks above the blocks being prefetched are not in
* cache, they will be asychronously read in.
Illumos #4045 write throttle & i/o scheduler performance work 4045 zfs write throttle & i/o scheduler performance work 1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync read, sync write, async read, async write, and scrub/resilver. The scheduler issues a number of concurrent i/os from each class to the device. Once a class has been selected, an i/o is selected from this class using either an elevator algorithem (async, scrub classes) or FIFO (sync classes). The number of concurrent async write i/os is tuned dynamically based on i/o load, to achieve good sync i/o latency when there is not a high load of writes, and good write throughput when there is. See the block comment in vdev_queue.c (reproduced below) for more details. 2. The write throttle (dsl_pool_tempreserve_space() and txg_constrain_throughput()) is rewritten to produce much more consistent delays when under constant load. The new write throttle is based on the amount of dirty data, rather than guesses about future performance of the system. When there is a lot of dirty data, each transaction (e.g. write() syscall) will be delayed by the same small amount. This eliminates the "brick wall of wait" that the old write throttle could hit, causing all transactions to wait several seconds until the next txg opens. One of the keys to the new write throttle is decrementing the amount of dirty data as i/o completes, rather than at the end of spa_sync(). Note that the write throttle is only applied once the i/o scheduler is issuing the maximum number of outstanding async writes. See the block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for more details. This diff has several other effects, including: * the commonly-tuned global variable zfs_vdev_max_pending has been removed; use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead. * the size of each txg (meaning the amount of dirty data written, and thus the time it takes to write out) is now controlled differently. There is no longer an explicit time goal; the primary determinant is amount of dirty data. Systems that are under light or medium load will now often see that a txg is always syncing, but the impact to performance (e.g. read latency) is minimal. Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this. * zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression, checksum, etc. This improves latency by not allowing these CPU-intensive tasks to consume all CPU (on machines with at least 4 CPU's; the percentage is rounded up). --matt APPENDIX: problems with the current i/o scheduler The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem with this is that if there are always i/os pending, then certain classes of i/os can see very long delays. For example, if there are always synchronous reads outstanding, then no async writes will be serviced until they become "past due". One symptom of this situation is that each pass of the txg sync takes at least several seconds (typically 3 seconds). If many i/os become "past due" (their deadline is in the past), then we must service all of these overdue i/os before any new i/os. This happens when we enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in the future. If we can't complete all the i/os in 2.5 seconds (e.g. because there were always reads pending), then these i/os will become past due. Now we must service all the "async" writes (which could be hundreds of megabytes) before we service any reads, introducing considerable latency to synchronous i/os (reads or ZIL writes). Notes on porting to ZFS on Linux: - zio_t gained new members io_physdone and io_phys_children. Because object caches in the Linux port call the constructor only once at allocation time, objects may contain residual data when retrieved from the cache. Therefore zio_create() was updated to zero out the two new fields. - vdev_mirror_pending() relied on the depth of the per-vdev pending queue (vq->vq_pending_tree) to select the least-busy leaf vdev to read from. This tree has been replaced by vq->vq_active_tree which is now used for the same purpose. - vdev_queue_init() used the value of zfs_vdev_max_pending to determine the number of vdev I/O buffers to pre-allocate. That global no longer exists, so we instead use the sum of the *_max_active values for each of the five I/O classes described above. - The Illumos implementation of dmu_tx_delay() delays a transaction by sleeping in condition variable embedded in the thread (curthread->t_delay_cv). We do not have an equivalent CV to use in Linux, so this change replaced the delay logic with a wrapper called zfs_sleep_until(). This wrapper could be adopted upstream and in other downstream ports to abstract away operating system-specific delay logic. - These tunables are added as module parameters, and descriptions added to the zfs-module-parameters.5 man page. spa_asize_inflation zfs_deadman_synctime_ms zfs_vdev_max_active zfs_vdev_async_write_active_min_dirty_percent zfs_vdev_async_write_active_max_dirty_percent zfs_vdev_async_read_max_active zfs_vdev_async_read_min_active zfs_vdev_async_write_max_active zfs_vdev_async_write_min_active zfs_vdev_scrub_max_active zfs_vdev_scrub_min_active zfs_vdev_sync_read_max_active zfs_vdev_sync_read_min_active zfs_vdev_sync_write_max_active zfs_vdev_sync_write_min_active zfs_dirty_data_max_percent zfs_delay_min_dirty_percent zfs_dirty_data_max_max_percent zfs_dirty_data_max zfs_dirty_data_max_max zfs_dirty_data_sync zfs_delay_scale The latter four have type unsigned long, whereas they are uint64_t in Illumos. This accommodates Linux's module_param() supported types, but means they may overflow on 32-bit architectures. The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most likely to overflow on 32-bit systems, since they express physical RAM sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to 2^32 which does overflow. To resolve that, this port instead initializes it in arc_init() to 25% of physical RAM, and adds the tunable zfs_dirty_data_max_max_percent to override that percentage. While this solution doesn't completely avoid the overflow issue, it should be a reasonable default for most systems, and the minority of affected systems can work around the issue by overriding the defaults. - Fixed reversed logic in comment above zfs_delay_scale declaration. - Clarified comments in vdev_queue.c regarding when per-queue minimums take effect. - Replaced dmu_tx_write_limit in the dmu_tx kstat file with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts how many times a transaction has been delayed because the pool dirty data has exceeded zfs_delay_min_dirty_percent. The latter counts how many times the pool dirty data has exceeded zfs_dirty_data_max (which we expect to never happen). - The original patch would have regressed the bug fixed in zfsonlinux/zfs@c418410, which prevented users from setting the zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE. A similar fix is added to vdev_queue_aggregate(). - In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the heap instead of the stack. In Linux we can't afford such large structures on the stack. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Brendan Gregg <brendan.gregg@joyent.com> Approved by: Robert Mustacchi <rm@joyent.com> References: http://www.illumos.org/issues/4045 illumos/illumos-gate@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
*/
2008-11-20 20:01:55 +00:00
void
dmu_prefetch(objset_t *os, uint64_t object, int64_t level, uint64_t offset,
uint64_t len, zio_priority_t pri)
2008-11-20 20:01:55 +00:00
{
dnode_t *dn;
uint64_t blkid;
Illumos #4045 write throttle & i/o scheduler performance work 4045 zfs write throttle & i/o scheduler performance work 1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync read, sync write, async read, async write, and scrub/resilver. The scheduler issues a number of concurrent i/os from each class to the device. Once a class has been selected, an i/o is selected from this class using either an elevator algorithem (async, scrub classes) or FIFO (sync classes). The number of concurrent async write i/os is tuned dynamically based on i/o load, to achieve good sync i/o latency when there is not a high load of writes, and good write throughput when there is. See the block comment in vdev_queue.c (reproduced below) for more details. 2. The write throttle (dsl_pool_tempreserve_space() and txg_constrain_throughput()) is rewritten to produce much more consistent delays when under constant load. The new write throttle is based on the amount of dirty data, rather than guesses about future performance of the system. When there is a lot of dirty data, each transaction (e.g. write() syscall) will be delayed by the same small amount. This eliminates the "brick wall of wait" that the old write throttle could hit, causing all transactions to wait several seconds until the next txg opens. One of the keys to the new write throttle is decrementing the amount of dirty data as i/o completes, rather than at the end of spa_sync(). Note that the write throttle is only applied once the i/o scheduler is issuing the maximum number of outstanding async writes. See the block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for more details. This diff has several other effects, including: * the commonly-tuned global variable zfs_vdev_max_pending has been removed; use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead. * the size of each txg (meaning the amount of dirty data written, and thus the time it takes to write out) is now controlled differently. There is no longer an explicit time goal; the primary determinant is amount of dirty data. Systems that are under light or medium load will now often see that a txg is always syncing, but the impact to performance (e.g. read latency) is minimal. Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this. * zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression, checksum, etc. This improves latency by not allowing these CPU-intensive tasks to consume all CPU (on machines with at least 4 CPU's; the percentage is rounded up). --matt APPENDIX: problems with the current i/o scheduler The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem with this is that if there are always i/os pending, then certain classes of i/os can see very long delays. For example, if there are always synchronous reads outstanding, then no async writes will be serviced until they become "past due". One symptom of this situation is that each pass of the txg sync takes at least several seconds (typically 3 seconds). If many i/os become "past due" (their deadline is in the past), then we must service all of these overdue i/os before any new i/os. This happens when we enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in the future. If we can't complete all the i/os in 2.5 seconds (e.g. because there were always reads pending), then these i/os will become past due. Now we must service all the "async" writes (which could be hundreds of megabytes) before we service any reads, introducing considerable latency to synchronous i/os (reads or ZIL writes). Notes on porting to ZFS on Linux: - zio_t gained new members io_physdone and io_phys_children. Because object caches in the Linux port call the constructor only once at allocation time, objects may contain residual data when retrieved from the cache. Therefore zio_create() was updated to zero out the two new fields. - vdev_mirror_pending() relied on the depth of the per-vdev pending queue (vq->vq_pending_tree) to select the least-busy leaf vdev to read from. This tree has been replaced by vq->vq_active_tree which is now used for the same purpose. - vdev_queue_init() used the value of zfs_vdev_max_pending to determine the number of vdev I/O buffers to pre-allocate. That global no longer exists, so we instead use the sum of the *_max_active values for each of the five I/O classes described above. - The Illumos implementation of dmu_tx_delay() delays a transaction by sleeping in condition variable embedded in the thread (curthread->t_delay_cv). We do not have an equivalent CV to use in Linux, so this change replaced the delay logic with a wrapper called zfs_sleep_until(). This wrapper could be adopted upstream and in other downstream ports to abstract away operating system-specific delay logic. - These tunables are added as module parameters, and descriptions added to the zfs-module-parameters.5 man page. spa_asize_inflation zfs_deadman_synctime_ms zfs_vdev_max_active zfs_vdev_async_write_active_min_dirty_percent zfs_vdev_async_write_active_max_dirty_percent zfs_vdev_async_read_max_active zfs_vdev_async_read_min_active zfs_vdev_async_write_max_active zfs_vdev_async_write_min_active zfs_vdev_scrub_max_active zfs_vdev_scrub_min_active zfs_vdev_sync_read_max_active zfs_vdev_sync_read_min_active zfs_vdev_sync_write_max_active zfs_vdev_sync_write_min_active zfs_dirty_data_max_percent zfs_delay_min_dirty_percent zfs_dirty_data_max_max_percent zfs_dirty_data_max zfs_dirty_data_max_max zfs_dirty_data_sync zfs_delay_scale The latter four have type unsigned long, whereas they are uint64_t in Illumos. This accommodates Linux's module_param() supported types, but means they may overflow on 32-bit architectures. The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most likely to overflow on 32-bit systems, since they express physical RAM sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to 2^32 which does overflow. To resolve that, this port instead initializes it in arc_init() to 25% of physical RAM, and adds the tunable zfs_dirty_data_max_max_percent to override that percentage. While this solution doesn't completely avoid the overflow issue, it should be a reasonable default for most systems, and the minority of affected systems can work around the issue by overriding the defaults. - Fixed reversed logic in comment above zfs_delay_scale declaration. - Clarified comments in vdev_queue.c regarding when per-queue minimums take effect. - Replaced dmu_tx_write_limit in the dmu_tx kstat file with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts how many times a transaction has been delayed because the pool dirty data has exceeded zfs_delay_min_dirty_percent. The latter counts how many times the pool dirty data has exceeded zfs_dirty_data_max (which we expect to never happen). - The original patch would have regressed the bug fixed in zfsonlinux/zfs@c418410, which prevented users from setting the zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE. A similar fix is added to vdev_queue_aggregate(). - In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the heap instead of the stack. In Linux we can't afford such large structures on the stack. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Brendan Gregg <brendan.gregg@joyent.com> Approved by: Robert Mustacchi <rm@joyent.com> References: http://www.illumos.org/issues/4045 illumos/illumos-gate@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
int nblks, err;
2008-11-20 20:01:55 +00:00
if (len == 0) { /* they're interested in the bonus buffer */
dn = DMU_META_DNODE(os);
2008-11-20 20:01:55 +00:00
if (object == 0 || object >= DN_MAX_OBJECT)
return;
rw_enter(&dn->dn_struct_rwlock, RW_READER);
blkid = dbuf_whichblock(dn, level,
object * sizeof (dnode_phys_t));
dbuf_prefetch(dn, level, blkid, pri, 0);
2008-11-20 20:01:55 +00:00
rw_exit(&dn->dn_struct_rwlock);
return;
}
/*
* XXX - Note, if the dnode for the requested object is not
* already cached, we will do a *synchronous* read in the
* dnode_hold() call. The same is true for any indirects.
*/
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err != 0)
return;
rw_enter(&dn->dn_struct_rwlock, RW_READER);
/*
* offset + len - 1 is the last byte we want to prefetch for, and offset
* is the first. Then dbuf_whichblk(dn, level, off + len - 1) is the
* last block we want to prefetch, and dbuf_whichblock(dn, level,
* offset) is the first. Then the number we need to prefetch is the
* last - first + 1.
*/
if (level > 0 || dn->dn_datablkshift != 0) {
nblks = dbuf_whichblock(dn, level, offset + len - 1) -
dbuf_whichblock(dn, level, offset) + 1;
2008-11-20 20:01:55 +00:00
} else {
nblks = (offset < dn->dn_datablksz);
}
if (nblks != 0) {
Illumos #4045 write throttle & i/o scheduler performance work 4045 zfs write throttle & i/o scheduler performance work 1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync read, sync write, async read, async write, and scrub/resilver. The scheduler issues a number of concurrent i/os from each class to the device. Once a class has been selected, an i/o is selected from this class using either an elevator algorithem (async, scrub classes) or FIFO (sync classes). The number of concurrent async write i/os is tuned dynamically based on i/o load, to achieve good sync i/o latency when there is not a high load of writes, and good write throughput when there is. See the block comment in vdev_queue.c (reproduced below) for more details. 2. The write throttle (dsl_pool_tempreserve_space() and txg_constrain_throughput()) is rewritten to produce much more consistent delays when under constant load. The new write throttle is based on the amount of dirty data, rather than guesses about future performance of the system. When there is a lot of dirty data, each transaction (e.g. write() syscall) will be delayed by the same small amount. This eliminates the "brick wall of wait" that the old write throttle could hit, causing all transactions to wait several seconds until the next txg opens. One of the keys to the new write throttle is decrementing the amount of dirty data as i/o completes, rather than at the end of spa_sync(). Note that the write throttle is only applied once the i/o scheduler is issuing the maximum number of outstanding async writes. See the block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for more details. This diff has several other effects, including: * the commonly-tuned global variable zfs_vdev_max_pending has been removed; use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead. * the size of each txg (meaning the amount of dirty data written, and thus the time it takes to write out) is now controlled differently. There is no longer an explicit time goal; the primary determinant is amount of dirty data. Systems that are under light or medium load will now often see that a txg is always syncing, but the impact to performance (e.g. read latency) is minimal. Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this. * zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression, checksum, etc. This improves latency by not allowing these CPU-intensive tasks to consume all CPU (on machines with at least 4 CPU's; the percentage is rounded up). --matt APPENDIX: problems with the current i/o scheduler The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem with this is that if there are always i/os pending, then certain classes of i/os can see very long delays. For example, if there are always synchronous reads outstanding, then no async writes will be serviced until they become "past due". One symptom of this situation is that each pass of the txg sync takes at least several seconds (typically 3 seconds). If many i/os become "past due" (their deadline is in the past), then we must service all of these overdue i/os before any new i/os. This happens when we enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in the future. If we can't complete all the i/os in 2.5 seconds (e.g. because there were always reads pending), then these i/os will become past due. Now we must service all the "async" writes (which could be hundreds of megabytes) before we service any reads, introducing considerable latency to synchronous i/os (reads or ZIL writes). Notes on porting to ZFS on Linux: - zio_t gained new members io_physdone and io_phys_children. Because object caches in the Linux port call the constructor only once at allocation time, objects may contain residual data when retrieved from the cache. Therefore zio_create() was updated to zero out the two new fields. - vdev_mirror_pending() relied on the depth of the per-vdev pending queue (vq->vq_pending_tree) to select the least-busy leaf vdev to read from. This tree has been replaced by vq->vq_active_tree which is now used for the same purpose. - vdev_queue_init() used the value of zfs_vdev_max_pending to determine the number of vdev I/O buffers to pre-allocate. That global no longer exists, so we instead use the sum of the *_max_active values for each of the five I/O classes described above. - The Illumos implementation of dmu_tx_delay() delays a transaction by sleeping in condition variable embedded in the thread (curthread->t_delay_cv). We do not have an equivalent CV to use in Linux, so this change replaced the delay logic with a wrapper called zfs_sleep_until(). This wrapper could be adopted upstream and in other downstream ports to abstract away operating system-specific delay logic. - These tunables are added as module parameters, and descriptions added to the zfs-module-parameters.5 man page. spa_asize_inflation zfs_deadman_synctime_ms zfs_vdev_max_active zfs_vdev_async_write_active_min_dirty_percent zfs_vdev_async_write_active_max_dirty_percent zfs_vdev_async_read_max_active zfs_vdev_async_read_min_active zfs_vdev_async_write_max_active zfs_vdev_async_write_min_active zfs_vdev_scrub_max_active zfs_vdev_scrub_min_active zfs_vdev_sync_read_max_active zfs_vdev_sync_read_min_active zfs_vdev_sync_write_max_active zfs_vdev_sync_write_min_active zfs_dirty_data_max_percent zfs_delay_min_dirty_percent zfs_dirty_data_max_max_percent zfs_dirty_data_max zfs_dirty_data_max_max zfs_dirty_data_sync zfs_delay_scale The latter four have type unsigned long, whereas they are uint64_t in Illumos. This accommodates Linux's module_param() supported types, but means they may overflow on 32-bit architectures. The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most likely to overflow on 32-bit systems, since they express physical RAM sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to 2^32 which does overflow. To resolve that, this port instead initializes it in arc_init() to 25% of physical RAM, and adds the tunable zfs_dirty_data_max_max_percent to override that percentage. While this solution doesn't completely avoid the overflow issue, it should be a reasonable default for most systems, and the minority of affected systems can work around the issue by overriding the defaults. - Fixed reversed logic in comment above zfs_delay_scale declaration. - Clarified comments in vdev_queue.c regarding when per-queue minimums take effect. - Replaced dmu_tx_write_limit in the dmu_tx kstat file with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts how many times a transaction has been delayed because the pool dirty data has exceeded zfs_delay_min_dirty_percent. The latter counts how many times the pool dirty data has exceeded zfs_dirty_data_max (which we expect to never happen). - The original patch would have regressed the bug fixed in zfsonlinux/zfs@c418410, which prevented users from setting the zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE. A similar fix is added to vdev_queue_aggregate(). - In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the heap instead of the stack. In Linux we can't afford such large structures on the stack. Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Adam Leventhal <ahl@delphix.com> Reviewed by: Christopher Siden <christopher.siden@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Brendan Gregg <brendan.gregg@joyent.com> Approved by: Robert Mustacchi <rm@joyent.com> References: http://www.illumos.org/issues/4045 illumos/illumos-gate@69962b5647e4a8b9b14998733b765925381b727e Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1913
2013-08-29 03:01:20 +00:00
int i;
blkid = dbuf_whichblock(dn, level, offset);
2008-11-20 20:01:55 +00:00
for (i = 0; i < nblks; i++)
dbuf_prefetch(dn, level, blkid + i, pri, 0);
2008-11-20 20:01:55 +00:00
}
rw_exit(&dn->dn_struct_rwlock);
dnode_rele(dn, FTAG);
}
2009-08-18 18:43:27 +00:00
/*
* Get the next "chunk" of file data to free. We traverse the file from
* the end so that the file gets shorter over time (if we crashes in the
* middle, this will leave us in a better state). We find allocated file
* data by simply searching the allocated level 1 indirects.
*
* On input, *start should be the first offset that does not need to be
* freed (e.g. "offset + length"). On return, *start will be the first
* offset that should be freed.
2009-08-18 18:43:27 +00:00
*/
static int
get_next_chunk(dnode_t *dn, uint64_t *start, uint64_t minimum)
{
uint64_t maxblks = DMU_MAX_ACCESS >> (dn->dn_indblkshift + 1);
/* bytes of data covered by a level-1 indirect block */
2009-08-18 18:43:27 +00:00
uint64_t iblkrange =
dn->dn_datablksz * EPB(dn->dn_indblkshift, SPA_BLKPTRSHIFT);
uint64_t blks;
ASSERT3U(minimum, <=, *start);
if (*start - minimum <= iblkrange * maxblks) {
*start = minimum;
return (0);
}
2009-08-18 18:43:27 +00:00
ASSERT(ISP2(iblkrange));
for (blks = 0; *start > minimum && blks < maxblks; blks++) {
int err;
/*
* dnode_next_offset(BACKWARDS) will find an allocated L1
* indirect block at or before the input offset. We must
* decrement *start so that it is at the end of the region
* to search.
*/
(*start)--;
err = dnode_next_offset(dn,
2009-08-18 18:43:27 +00:00
DNODE_FIND_BACKWARDS, start, 2, 1, 0);
/* if there are no indirect blocks before start, we are done */
2009-08-18 18:43:27 +00:00
if (err == ESRCH) {
*start = minimum;
break;
} else if (err != 0) {
return (err);
2009-08-18 18:43:27 +00:00
}
/* set start to the beginning of this L1 indirect */
2009-08-18 18:43:27 +00:00
*start = P2ALIGN(*start, iblkrange);
}
if (*start < minimum)
*start = minimum;
return (0);
}
static int
dmu_free_long_range_impl(objset_t *os, dnode_t *dn, uint64_t offset,
uint64_t length)
{
uint64_t object_size;
int err;
if (dn == NULL)
return (SET_ERROR(EINVAL));
object_size = (dn->dn_maxblkid + 1) * dn->dn_datablksz;
if (offset >= object_size)
return (0);
if (length == DMU_OBJECT_END || offset + length > object_size)
length = object_size - offset;
while (length != 0) {
uint64_t chunk_end, chunk_begin;
dmu_tx_t *tx;
chunk_end = chunk_begin = offset + length;
/* move chunk_begin backwards to the beginning of this chunk */
err = get_next_chunk(dn, &chunk_begin, offset);
if (err)
return (err);
ASSERT3U(chunk_begin, >=, offset);
ASSERT3U(chunk_begin, <=, chunk_end);
tx = dmu_tx_create(os);
dmu_tx_hold_free(tx, dn->dn_object,
chunk_begin, chunk_end - chunk_begin);
/*
* Mark this transaction as typically resulting in a net
* reduction in space used.
*/
dmu_tx_mark_netfree(tx);
err = dmu_tx_assign(tx, TXG_WAIT);
if (err) {
dmu_tx_abort(tx);
return (err);
}
dnode_free_range(dn, chunk_begin, chunk_end - chunk_begin, tx);
dmu_tx_commit(tx);
length -= chunk_end - chunk_begin;
}
return (0);
}
int
dmu_free_long_range(objset_t *os, uint64_t object,
uint64_t offset, uint64_t length)
{
dnode_t *dn;
int err;
err = dnode_hold(os, object, FTAG, &dn);
if (err != 0)
return (err);
err = dmu_free_long_range_impl(os, dn, offset, length);
/*
* It is important to zero out the maxblkid when freeing the entire
* file, so that (a) subsequent calls to dmu_free_long_range_impl()
* will take the fast path, and (b) dnode_reallocate() can verify
* that the entire file has been freed.
*/
if (err == 0 && offset == 0 && length == DMU_OBJECT_END)
dn->dn_maxblkid = 0;
dnode_rele(dn, FTAG);
return (err);
}
int
dmu_free_long_object(objset_t *os, uint64_t object)
{
dmu_tx_t *tx;
int err;
err = dmu_free_long_range(os, object, 0, DMU_OBJECT_END);
if (err != 0)
return (err);
tx = dmu_tx_create(os);
dmu_tx_hold_bonus(tx, object);
dmu_tx_hold_free(tx, object, 0, DMU_OBJECT_END);
dmu_tx_mark_netfree(tx);
err = dmu_tx_assign(tx, TXG_WAIT);
if (err == 0) {
err = dmu_object_free(os, object, tx);
dmu_tx_commit(tx);
} else {
dmu_tx_abort(tx);
}
return (err);
}
2008-11-20 20:01:55 +00:00
int
dmu_free_range(objset_t *os, uint64_t object, uint64_t offset,
uint64_t size, dmu_tx_t *tx)
{
dnode_t *dn;
int err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
ASSERT(offset < UINT64_MAX);
ASSERT(size == -1ULL || size <= UINT64_MAX - offset);
dnode_free_range(dn, offset, size, tx);
dnode_rele(dn, FTAG);
return (0);
}
int
dmu_read(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
2009-07-02 22:44:48 +00:00
void *buf, uint32_t flags)
2008-11-20 20:01:55 +00:00
{
dnode_t *dn;
dmu_buf_t **dbp;
2009-08-18 18:43:27 +00:00
int numbufs, err;
2008-11-20 20:01:55 +00:00
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
/*
* Deal with odd block sizes, where there can't be data past the first
* block. If we ever do the tail block optimization, we will need to
* handle that here as well.
*/
2009-08-18 18:43:27 +00:00
if (dn->dn_maxblkid == 0) {
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
uint64_t newsz = offset > dn->dn_datablksz ? 0 :
2008-11-20 20:01:55 +00:00
MIN(size, dn->dn_datablksz - offset);
bzero((char *)buf + newsz, size - newsz);
size = newsz;
}
while (size > 0) {
uint64_t mylen = MIN(size, DMU_MAX_ACCESS / 2);
2009-08-18 18:43:27 +00:00
int i;
2008-11-20 20:01:55 +00:00
/*
* NB: we could do this block-at-a-time, but it's nice
* to be reading in parallel.
*/
err = dmu_buf_hold_array_by_dnode(dn, offset, mylen,
2009-07-02 22:44:48 +00:00
TRUE, FTAG, &numbufs, &dbp, flags);
2008-11-20 20:01:55 +00:00
if (err)
break;
for (i = 0; i < numbufs; i++) {
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
uint64_t tocpy;
int64_t bufoff;
2008-11-20 20:01:55 +00:00
dmu_buf_t *db = dbp[i];
ASSERT(size > 0);
bufoff = offset - db->db_offset;
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
tocpy = MIN(db->db_size - bufoff, size);
2008-11-20 20:01:55 +00:00
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
(void) memcpy(buf, (char *)db->db_data + bufoff, tocpy);
2008-11-20 20:01:55 +00:00
offset += tocpy;
size -= tocpy;
buf = (char *)buf + tocpy;
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
}
dnode_rele(dn, FTAG);
return (err);
}
void
dmu_write(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
const void *buf, dmu_tx_t *tx)
{
dmu_buf_t **dbp;
int numbufs, i;
if (size == 0)
return;
VERIFY0(dmu_buf_hold_array(os, object, offset, size,
2008-11-20 20:01:55 +00:00
FALSE, FTAG, &numbufs, &dbp));
for (i = 0; i < numbufs; i++) {
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
uint64_t tocpy;
int64_t bufoff;
2008-11-20 20:01:55 +00:00
dmu_buf_t *db = dbp[i];
ASSERT(size > 0);
bufoff = offset - db->db_offset;
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
tocpy = MIN(db->db_size - bufoff, size);
2008-11-20 20:01:55 +00:00
ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size);
if (tocpy == db->db_size)
dmu_buf_will_fill(db, tx);
else
dmu_buf_will_dirty(db, tx);
(void) memcpy((char *)db->db_data + bufoff, buf, tocpy);
2008-11-20 20:01:55 +00:00
if (tocpy == db->db_size)
dmu_buf_fill_done(db, tx);
offset += tocpy;
size -= tocpy;
buf = (char *)buf + tocpy;
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
}
void
dmu_prealloc(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
dmu_tx_t *tx)
{
dmu_buf_t **dbp;
int numbufs, i;
if (size == 0)
return;
VERIFY(0 == dmu_buf_hold_array(os, object, offset, size,
FALSE, FTAG, &numbufs, &dbp));
for (i = 0; i < numbufs; i++) {
dmu_buf_t *db = dbp[i];
dmu_buf_will_not_fill(db, tx);
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
}
void
dmu_write_embedded(objset_t *os, uint64_t object, uint64_t offset,
void *data, uint8_t etype, uint8_t comp, int uncompressed_size,
int compressed_size, int byteorder, dmu_tx_t *tx)
{
dmu_buf_t *db;
ASSERT3U(etype, <, NUM_BP_EMBEDDED_TYPES);
ASSERT3U(comp, <, ZIO_COMPRESS_FUNCTIONS);
VERIFY0(dmu_buf_hold_noread(os, object, offset,
FTAG, &db));
dmu_buf_write_embedded(db,
data, (bp_embedded_type_t)etype, (enum zio_compress)comp,
uncompressed_size, compressed_size, byteorder, tx);
dmu_buf_rele(db, FTAG);
}
/*
* DMU support for xuio
*/
kstat_t *xuio_ksp = NULL;
typedef struct xuio_stats {
/* loaned yet not returned arc_buf */
kstat_named_t xuiostat_onloan_rbuf;
kstat_named_t xuiostat_onloan_wbuf;
/* whether a copy is made when loaning out a read buffer */
kstat_named_t xuiostat_rbuf_copied;
kstat_named_t xuiostat_rbuf_nocopy;
/* whether a copy is made when assigning a write buffer */
kstat_named_t xuiostat_wbuf_copied;
kstat_named_t xuiostat_wbuf_nocopy;
} xuio_stats_t;
static xuio_stats_t xuio_stats = {
{ "onloan_read_buf", KSTAT_DATA_UINT64 },
{ "onloan_write_buf", KSTAT_DATA_UINT64 },
{ "read_buf_copied", KSTAT_DATA_UINT64 },
{ "read_buf_nocopy", KSTAT_DATA_UINT64 },
{ "write_buf_copied", KSTAT_DATA_UINT64 },
{ "write_buf_nocopy", KSTAT_DATA_UINT64 }
};
#define XUIOSTAT_INCR(stat, val) \
atomic_add_64(&xuio_stats.stat.value.ui64, (val))
#define XUIOSTAT_BUMP(stat) XUIOSTAT_INCR(stat, 1)
int
dmu_xuio_init(xuio_t *xuio, int nblk)
{
dmu_xuio_t *priv;
uio_t *uio = &xuio->xu_uio;
uio->uio_iovcnt = nblk;
uio->uio_iov = kmem_zalloc(nblk * sizeof (iovec_t), KM_SLEEP);
priv = kmem_zalloc(sizeof (dmu_xuio_t), KM_SLEEP);
priv->cnt = nblk;
priv->bufs = kmem_zalloc(nblk * sizeof (arc_buf_t *), KM_SLEEP);
priv->iovp = (iovec_t *)uio->uio_iov;
XUIO_XUZC_PRIV(xuio) = priv;
if (XUIO_XUZC_RW(xuio) == UIO_READ)
XUIOSTAT_INCR(xuiostat_onloan_rbuf, nblk);
else
XUIOSTAT_INCR(xuiostat_onloan_wbuf, nblk);
return (0);
}
void
dmu_xuio_fini(xuio_t *xuio)
{
dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio);
int nblk = priv->cnt;
kmem_free(priv->iovp, nblk * sizeof (iovec_t));
kmem_free(priv->bufs, nblk * sizeof (arc_buf_t *));
kmem_free(priv, sizeof (dmu_xuio_t));
if (XUIO_XUZC_RW(xuio) == UIO_READ)
XUIOSTAT_INCR(xuiostat_onloan_rbuf, -nblk);
else
XUIOSTAT_INCR(xuiostat_onloan_wbuf, -nblk);
}
/*
* Initialize iov[priv->next] and priv->bufs[priv->next] with { off, n, abuf }
* and increase priv->next by 1.
*/
int
dmu_xuio_add(xuio_t *xuio, arc_buf_t *abuf, offset_t off, size_t n)
{
struct iovec *iov;
uio_t *uio = &xuio->xu_uio;
dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio);
int i = priv->next++;
ASSERT(i < priv->cnt);
ASSERT(off + n <= arc_buf_lsize(abuf));
iov = (iovec_t *)uio->uio_iov + i;
iov->iov_base = (char *)abuf->b_data + off;
iov->iov_len = n;
priv->bufs[i] = abuf;
return (0);
}
int
dmu_xuio_cnt(xuio_t *xuio)
{
dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio);
return (priv->cnt);
}
arc_buf_t *
dmu_xuio_arcbuf(xuio_t *xuio, int i)
{
dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio);
ASSERT(i < priv->cnt);
return (priv->bufs[i]);
}
void
dmu_xuio_clear(xuio_t *xuio, int i)
{
dmu_xuio_t *priv = XUIO_XUZC_PRIV(xuio);
ASSERT(i < priv->cnt);
priv->bufs[i] = NULL;
}
static void
xuio_stat_init(void)
{
xuio_ksp = kstat_create("zfs", 0, "xuio_stats", "misc",
KSTAT_TYPE_NAMED, sizeof (xuio_stats) / sizeof (kstat_named_t),
KSTAT_FLAG_VIRTUAL);
if (xuio_ksp != NULL) {
xuio_ksp->ks_data = &xuio_stats;
kstat_install(xuio_ksp);
}
}
static void
xuio_stat_fini(void)
{
if (xuio_ksp != NULL) {
kstat_delete(xuio_ksp);
xuio_ksp = NULL;
}
}
void
xuio_stat_wbuf_copied()
{
XUIOSTAT_BUMP(xuiostat_wbuf_copied);
}
void
xuio_stat_wbuf_nocopy()
{
XUIOSTAT_BUMP(xuiostat_wbuf_nocopy);
}
2008-11-20 20:01:55 +00:00
#ifdef _KERNEL
static int
dmu_read_uio_dnode(dnode_t *dn, uio_t *uio, uint64_t size)
{
dmu_buf_t **dbp;
int numbufs, i, err;
xuio_t *xuio = NULL;
/*
* NB: we could do this block-at-a-time, but it's nice
* to be reading in parallel.
*/
err = dmu_buf_hold_array_by_dnode(dn, uio->uio_loffset, size,
TRUE, FTAG, &numbufs, &dbp, 0);
if (err)
return (err);
for (i = 0; i < numbufs; i++) {
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
uint64_t tocpy;
int64_t bufoff;
dmu_buf_t *db = dbp[i];
ASSERT(size > 0);
bufoff = uio->uio_loffset - db->db_offset;
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
tocpy = MIN(db->db_size - bufoff, size);
if (xuio) {
dmu_buf_impl_t *dbi = (dmu_buf_impl_t *)db;
arc_buf_t *dbuf_abuf = dbi->db_buf;
arc_buf_t *abuf = dbuf_loan_arcbuf(dbi);
err = dmu_xuio_add(xuio, abuf, bufoff, tocpy);
if (!err) {
uio->uio_resid -= tocpy;
uio->uio_loffset += tocpy;
}
if (abuf == dbuf_abuf)
XUIOSTAT_BUMP(xuiostat_rbuf_nocopy);
else
XUIOSTAT_BUMP(xuiostat_rbuf_copied);
} else {
err = uiomove((char *)db->db_data + bufoff, tocpy,
UIO_READ, uio);
}
if (err)
break;
size -= tocpy;
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
return (err);
}
/*
* Read 'size' bytes into the uio buffer.
* From object zdb->db_object.
* Starting at offset uio->uio_loffset.
*
* If the caller already has a dbuf in the target object
* (e.g. its bonus buffer), this routine is faster than dmu_read_uio(),
* because we don't have to find the dnode_t for the object.
*/
int
dmu_read_uio_dbuf(dmu_buf_t *zdb, uio_t *uio, uint64_t size)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)zdb;
dnode_t *dn;
int err;
if (size == 0)
return (0);
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
err = dmu_read_uio_dnode(dn, uio, size);
DB_DNODE_EXIT(db);
return (err);
}
/*
* Read 'size' bytes into the uio buffer.
* From the specified object
* Starting at offset uio->uio_loffset.
*/
int
dmu_read_uio(objset_t *os, uint64_t object, uio_t *uio, uint64_t size)
{
dnode_t *dn;
int err;
if (size == 0)
return (0);
err = dnode_hold(os, object, FTAG, &dn);
if (err)
return (err);
err = dmu_read_uio_dnode(dn, uio, size);
dnode_rele(dn, FTAG);
return (err);
}
static int
dmu_write_uio_dnode(dnode_t *dn, uio_t *uio, uint64_t size, dmu_tx_t *tx)
{
dmu_buf_t **dbp;
int numbufs;
int err = 0;
int i;
err = dmu_buf_hold_array_by_dnode(dn, uio->uio_loffset, size,
FALSE, FTAG, &numbufs, &dbp, DMU_READ_PREFETCH);
if (err)
return (err);
for (i = 0; i < numbufs; i++) {
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
uint64_t tocpy;
int64_t bufoff;
dmu_buf_t *db = dbp[i];
ASSERT(size > 0);
bufoff = uio->uio_loffset - db->db_offset;
dmu: fix integer overflows The params to the functions are uint64_t, but the offsets to memcpy / bcopy are calculated using 32bit ints. This patch changes them to also be uint64_t so there isnt an overflow. PaX's Size Overflow caught this when formatting a zvol. Gentoo bug: #546490 PAX: offset: 1ffffb000 db->db_offset: 1ffffa000 db->db_size: 2000 size: 5000 PAX: size overflow detected in function dmu_read /var/tmp/portage/sys-fs/zfs-kmod-0.6.3-r1/work/zfs-zfs-0.6.3/module/zfs/../../module/zfs/dmu.c:781 cicus.366_146 max, count: 15 CPU: 1 PID: 2236 Comm: zvol/10 Tainted: P O 3.17.7-hardened-r1 #1 Call Trace: [<ffffffffa0382ee8>] ? dsl_dataset_get_holds+0x9d58/0x343ce [zfs] [<ffffffff81a59c88>] dump_stack+0x4e/0x7a [<ffffffffa0393c2a>] ? dsl_dataset_get_holds+0x1aa9a/0x343ce [zfs] [<ffffffff81206696>] report_size_overflow+0x36/0x40 [<ffffffffa02dba2b>] dmu_read+0x52b/0x920 [zfs] [<ffffffffa0373ad1>] zrl_is_locked+0x7d1/0x1ce0 [zfs] [<ffffffffa0364cd2>] zil_clean+0x9d2/0xc00 [zfs] [<ffffffffa0364f21>] zil_commit+0x21/0x30 [zfs] [<ffffffffa0373fe1>] zrl_is_locked+0xce1/0x1ce0 [zfs] [<ffffffff81a5e2c7>] ? __schedule+0x547/0xbc0 [<ffffffffa01582e6>] taskq_cancel_id+0x2a6/0x5b0 [spl] [<ffffffff81103eb0>] ? wake_up_state+0x20/0x20 [<ffffffffa0158150>] ? taskq_cancel_id+0x110/0x5b0 [spl] [<ffffffff810f7ff4>] kthread+0xc4/0xe0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 [<ffffffff81a62fa4>] ret_from_fork+0x74/0xa0 [<ffffffff810f7f30>] ? kthread_create_on_node+0x170/0x170 Signed-off-by: Jason Zaman <jason@perfinion.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #3333
2015-04-30 12:20:38 +00:00
tocpy = MIN(db->db_size - bufoff, size);
ASSERT(i == 0 || i == numbufs-1 || tocpy == db->db_size);
if (tocpy == db->db_size)
dmu_buf_will_fill(db, tx);
else
dmu_buf_will_dirty(db, tx);
/*
* XXX uiomove could block forever (eg.nfs-backed
* pages). There needs to be a uiolockdown() function
* to lock the pages in memory, so that uiomove won't
* block.
*/
err = uiomove((char *)db->db_data + bufoff, tocpy,
UIO_WRITE, uio);
if (tocpy == db->db_size)
dmu_buf_fill_done(db, tx);
if (err)
break;
size -= tocpy;
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
return (err);
}
/*
* Write 'size' bytes from the uio buffer.
* To object zdb->db_object.
* Starting at offset uio->uio_loffset.
*
* If the caller already has a dbuf in the target object
* (e.g. its bonus buffer), this routine is faster than dmu_write_uio(),
* because we don't have to find the dnode_t for the object.
*/
int
dmu_write_uio_dbuf(dmu_buf_t *zdb, uio_t *uio, uint64_t size,
dmu_tx_t *tx)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)zdb;
dnode_t *dn;
int err;
if (size == 0)
return (0);
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
err = dmu_write_uio_dnode(dn, uio, size, tx);
DB_DNODE_EXIT(db);
return (err);
}
/*
* Write 'size' bytes from the uio buffer.
* To the specified object.
* Starting at offset uio->uio_loffset.
*/
int
dmu_write_uio(objset_t *os, uint64_t object, uio_t *uio, uint64_t size,
dmu_tx_t *tx)
{
dnode_t *dn;
int err;
if (size == 0)
return (0);
err = dnode_hold(os, object, FTAG, &dn);
if (err)
return (err);
err = dmu_write_uio_dnode(dn, uio, size, tx);
dnode_rele(dn, FTAG);
return (err);
}
#endif /* _KERNEL */
2008-11-20 20:01:55 +00:00
2009-07-02 22:44:48 +00:00
/*
* Allocate a loaned anonymous arc buffer.
*/
arc_buf_t *
dmu_request_arcbuf(dmu_buf_t *handle, int size)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)handle;
2009-07-02 22:44:48 +00:00
return (arc_loan_buf(db->db_objset->os_spa, B_FALSE, size));
2009-07-02 22:44:48 +00:00
}
/*
* Free a loaned arc buffer.
*/
void
dmu_return_arcbuf(arc_buf_t *buf)
{
arc_return_buf(buf, FTAG);
OpenZFS 6950 - ARC should cache compressed data Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Matt Ahrens <mahrens@delphix.com> Reviewed by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Tom Caputi <tcaputi@datto.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Ported by: David Quigley <david.quigley@intel.com> This review covers the reading and writing of compressed arc headers, sharing data between the arc_hdr_t and the arc_buf_t, and the implementation of a new dbuf cache to keep frequently access data uncompressed. I've added a new member to l1 arc hdr called b_pdata. The b_pdata always hangs off the arc_buf_hdr_t (if an L1 hdr is in use) and points to the physical block for that DVA. The physical block may or may not be compressed. If compressed arc is enabled and the block on-disk is compressed, then the b_pdata will match the block on-disk and remain compressed in memory. If the block on disk is not compressed, then neither will the b_pdata. Lastly, if compressed arc is disabled, then b_pdata will always be an uncompressed version of the on-disk block. Typically the arc will cache only the arc_buf_hdr_t and will aggressively evict any arc_buf_t's that are no longer referenced. This means that the arc will primarily have compressed blocks as the arc_buf_t's are considered overhead and are always uncompressed. When a consumer reads a block we first look to see if the arc_buf_hdr_t is cached. If the hdr is cached then we allocate a new arc_buf_t and decompress the b_pdata contents into the arc_buf_t's b_data. If the hdr already has a arc_buf_t, then we will allocate an additional arc_buf_t and bcopy the uncompressed contents from the first arc_buf_t to the new one. Writing to the compressed arc requires that we first discard the b_pdata since the physical block is about to be rewritten. The new data contents will be passed in via an arc_buf_t (uncompressed) and during the I/O pipeline stages we will copy the physical block contents to a newly allocated b_pdata. When an l2arc is inuse it will also take advantage of the b_pdata. Now the l2arc will always write the contents of b_pdata to the l2arc. This means that when compressed arc is enabled that the l2arc blocks are identical to those stored in the main data pool. This provides a significant advantage since we can leverage the bp's checksum when reading from the l2arc to determine if the contents are valid. If the compressed arc is disabled, then we must first transform the read block to look like the physical block in the main data pool before comparing the checksum and determining it's valid. OpenZFS-issue: https://www.illumos.org/issues/6950 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7fc10f0 Issue #5078
2016-06-02 04:04:53 +00:00
arc_buf_destroy(buf, FTAG);
2009-07-02 22:44:48 +00:00
}
/*
* When possible directly assign passed loaned arc buffer to a dbuf.
* If this is not possible copy the contents of passed arc buf via
* dmu_write().
*/
void
dmu_assign_arcbuf(dmu_buf_t *handle, uint64_t offset, arc_buf_t *buf,
dmu_tx_t *tx)
{
dmu_buf_impl_t *dbuf = (dmu_buf_impl_t *)handle;
dnode_t *dn;
2009-07-02 22:44:48 +00:00
dmu_buf_impl_t *db;
uint32_t blksz = (uint32_t)arc_buf_lsize(buf);
2009-07-02 22:44:48 +00:00
uint64_t blkid;
DB_DNODE_ENTER(dbuf);
dn = DB_DNODE(dbuf);
2009-07-02 22:44:48 +00:00
rw_enter(&dn->dn_struct_rwlock, RW_READER);
blkid = dbuf_whichblock(dn, 0, offset);
2009-07-02 22:44:48 +00:00
VERIFY((db = dbuf_hold(dn, blkid, FTAG)) != NULL);
rw_exit(&dn->dn_struct_rwlock);
DB_DNODE_EXIT(dbuf);
2009-07-02 22:44:48 +00:00
/*
* We can only assign if the offset is aligned, the arc buf is the
* same size as the dbuf, and the dbuf is not metadata.
*/
if (offset == db->db.db_offset && blksz == db->db.db_size) {
2009-07-02 22:44:48 +00:00
dbuf_assign_arcbuf(db, buf, tx);
dbuf_rele(db, FTAG);
} else {
objset_t *os;
uint64_t object;
/* compressed bufs must always be assignable to their dbuf */
ASSERT3U(arc_get_compression(buf), ==, ZIO_COMPRESS_OFF);
ASSERT(!(buf->b_flags & ARC_BUF_FLAG_COMPRESSED));
DB_DNODE_ENTER(dbuf);
dn = DB_DNODE(dbuf);
os = dn->dn_objset;
object = dn->dn_object;
DB_DNODE_EXIT(dbuf);
2009-07-02 22:44:48 +00:00
dbuf_rele(db, FTAG);
dmu_write(os, object, offset, blksz, buf->b_data, tx);
2009-07-02 22:44:48 +00:00
dmu_return_arcbuf(buf);
XUIOSTAT_BUMP(xuiostat_wbuf_copied);
2009-07-02 22:44:48 +00:00
}
}
2008-11-20 20:01:55 +00:00
typedef struct {
dbuf_dirty_record_t *dsa_dr;
dmu_sync_cb_t *dsa_done;
zgd_t *dsa_zgd;
dmu_tx_t *dsa_tx;
2008-11-20 20:01:55 +00:00
} dmu_sync_arg_t;
/* ARGSUSED */
static void
dmu_sync_ready(zio_t *zio, arc_buf_t *buf, void *varg)
{
dmu_sync_arg_t *dsa = varg;
dmu_buf_t *db = dsa->dsa_zgd->zgd_db;
blkptr_t *bp = zio->io_bp;
if (zio->io_error == 0) {
if (BP_IS_HOLE(bp)) {
/*
* A block of zeros may compress to a hole, but the
* block size still needs to be known for replay.
*/
BP_SET_LSIZE(bp, db->db_size);
} else if (!BP_IS_EMBEDDED(bp)) {
ASSERT(BP_GET_LEVEL(bp) == 0);
bp->blk_fill = 1;
}
}
}
static void
dmu_sync_late_arrival_ready(zio_t *zio)
{
dmu_sync_ready(zio, NULL, zio->io_private);
}
2008-11-20 20:01:55 +00:00
/* ARGSUSED */
static void
dmu_sync_done(zio_t *zio, arc_buf_t *buf, void *varg)
{
dmu_sync_arg_t *dsa = varg;
dbuf_dirty_record_t *dr = dsa->dsa_dr;
2008-11-20 20:01:55 +00:00
dmu_buf_impl_t *db = dr->dr_dbuf;
mutex_enter(&db->db_mtx);
ASSERT(dr->dt.dl.dr_override_state == DR_IN_DMU_SYNC);
if (zio->io_error == 0) {
dr->dt.dl.dr_nopwrite = !!(zio->io_flags & ZIO_FLAG_NOPWRITE);
if (dr->dt.dl.dr_nopwrite) {
ASSERTV(blkptr_t *bp = zio->io_bp);
ASSERTV(blkptr_t *bp_orig = &zio->io_bp_orig);
ASSERTV(uint8_t chksum = BP_GET_CHECKSUM(bp_orig));
ASSERT(BP_EQUAL(bp, bp_orig));
ASSERT(zio->io_prop.zp_compress != ZIO_COMPRESS_OFF);
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
ASSERT(zio_checksum_table[chksum].ci_flags &
ZCHECKSUM_FLAG_NOPWRITE);
}
dr->dt.dl.dr_overridden_by = *zio->io_bp;
dr->dt.dl.dr_override_state = DR_OVERRIDDEN;
dr->dt.dl.dr_copies = zio->io_prop.zp_copies;
/*
* Old style holes are filled with all zeros, whereas
* new-style holes maintain their lsize, type, level,
* and birth time (see zio_write_compress). While we
* need to reset the BP_SET_LSIZE() call that happened
* in dmu_sync_ready for old style holes, we do *not*
* want to wipe out the information contained in new
* style holes. Thus, only zero out the block pointer if
* it's an old style hole.
*/
if (BP_IS_HOLE(&dr->dt.dl.dr_overridden_by) &&
dr->dt.dl.dr_overridden_by.blk_birth == 0)
BP_ZERO(&dr->dt.dl.dr_overridden_by);
} else {
dr->dt.dl.dr_override_state = DR_NOT_OVERRIDDEN;
}
2008-11-20 20:01:55 +00:00
cv_broadcast(&db->db_changed);
mutex_exit(&db->db_mtx);
dsa->dsa_done(dsa->dsa_zgd, zio->io_error);
2008-11-20 20:01:55 +00:00
kmem_free(dsa, sizeof (*dsa));
}
static void
dmu_sync_late_arrival_done(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
dmu_sync_arg_t *dsa = zio->io_private;
ASSERTV(blkptr_t *bp_orig = &zio->io_bp_orig);
if (zio->io_error == 0 && !BP_IS_HOLE(bp)) {
/*
* If we didn't allocate a new block (i.e. ZIO_FLAG_NOPWRITE)
* then there is nothing to do here. Otherwise, free the
* newly allocated block in this txg.
*/
if (zio->io_flags & ZIO_FLAG_NOPWRITE) {
ASSERT(BP_EQUAL(bp, bp_orig));
} else {
ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig));
ASSERT(zio->io_bp->blk_birth == zio->io_txg);
ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa));
zio_free(zio->io_spa, zio->io_txg, zio->io_bp);
}
}
dmu_tx_commit(dsa->dsa_tx);
dsa->dsa_done(dsa->dsa_zgd, zio->io_error);
kmem_free(dsa, sizeof (*dsa));
}
static int
dmu_sync_late_arrival(zio_t *pio, objset_t *os, dmu_sync_cb_t *done, zgd_t *zgd,
zio_prop_t *zp, zbookmark_phys_t *zb)
{
dmu_sync_arg_t *dsa;
dmu_tx_t *tx;
tx = dmu_tx_create(os);
dmu_tx_hold_space(tx, zgd->zgd_db->db_size);
if (dmu_tx_assign(tx, TXG_WAIT) != 0) {
dmu_tx_abort(tx);
/* Make zl_get_data do txg_waited_synced() */
return (SET_ERROR(EIO));
}
dsa = kmem_alloc(sizeof (dmu_sync_arg_t), KM_SLEEP);
dsa->dsa_dr = NULL;
dsa->dsa_done = done;
dsa->dsa_zgd = zgd;
dsa->dsa_tx = tx;
zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx),
zgd->zgd_bp, zgd->zgd_db->db_data, zgd->zgd_db->db_size,
zgd->zgd_db->db_size, zp, dmu_sync_late_arrival_ready, NULL,
NULL, dmu_sync_late_arrival_done, dsa, ZIO_PRIORITY_SYNC_WRITE,
ZIO_FLAG_CANFAIL, zb));
return (0);
2008-11-20 20:01:55 +00:00
}
/*
* Intent log support: sync the block associated with db to disk.
* N.B. and XXX: the caller is responsible for making sure that the
* data isn't changing while dmu_sync() is writing it.
*
* Return values:
*
* EEXIST: this txg has already been synced, so there's nothing to do.
2008-11-20 20:01:55 +00:00
* The caller should not log the write.
*
* ENOENT: the block was dbuf_free_range()'d, so there's nothing to do.
* The caller should not log the write.
*
* EALREADY: this block is already in the process of being synced.
* The caller should track its progress (somehow).
*
* EIO: could not do the I/O.
* The caller should do a txg_wait_synced().
2008-11-20 20:01:55 +00:00
*
* 0: the I/O has been initiated.
* The caller should log this blkptr in the done callback.
* It is possible that the I/O will fail, in which case
* the error will be reported to the done callback and
* propagated to pio from zio_done().
2008-11-20 20:01:55 +00:00
*/
int
dmu_sync(zio_t *pio, uint64_t txg, dmu_sync_cb_t *done, zgd_t *zgd)
2008-11-20 20:01:55 +00:00
{
blkptr_t *bp = zgd->zgd_bp;
dmu_buf_impl_t *db = (dmu_buf_impl_t *)zgd->zgd_db;
objset_t *os = db->db_objset;
dsl_dataset_t *ds = os->os_dsl_dataset;
2008-11-20 20:01:55 +00:00
dbuf_dirty_record_t *dr;
dmu_sync_arg_t *dsa;
zbookmark_phys_t zb;
zio_prop_t zp;
dnode_t *dn;
2008-11-20 20:01:55 +00:00
ASSERT(pio != NULL);
2008-11-20 20:01:55 +00:00
ASSERT(txg != 0);
SET_BOOKMARK(&zb, ds->ds_object,
db->db.db_object, db->db_level, db->db_blkid);
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
dmu_write_policy(os, dn, db->db_level, WP_DMU_SYNC,
ZIO_COMPRESS_INHERIT, &zp);
DB_DNODE_EXIT(db);
2008-11-20 20:01:55 +00:00
/*
* If we're frozen (running ziltest), we always need to generate a bp.
2008-11-20 20:01:55 +00:00
*/
if (txg > spa_freeze_txg(os->os_spa))
return (dmu_sync_late_arrival(pio, os, done, zgd, &zp, &zb));
2008-11-20 20:01:55 +00:00
/*
* Grabbing db_mtx now provides a barrier between dbuf_sync_leaf()
* and us. If we determine that this txg is not yet syncing,
* but it begins to sync a moment later, that's OK because the
* sync thread will block in dbuf_sync_leaf() until we drop db_mtx.
2008-11-20 20:01:55 +00:00
*/
mutex_enter(&db->db_mtx);
if (txg <= spa_last_synced_txg(os->os_spa)) {
2008-11-20 20:01:55 +00:00
/*
* This txg has already synced. There's nothing to do.
2008-11-20 20:01:55 +00:00
*/
mutex_exit(&db->db_mtx);
return (SET_ERROR(EEXIST));
2008-11-20 20:01:55 +00:00
}
if (txg <= spa_syncing_txg(os->os_spa)) {
/*
* This txg is currently syncing, so we can't mess with
* the dirty record anymore; just write a new log block.
*/
mutex_exit(&db->db_mtx);
return (dmu_sync_late_arrival(pio, os, done, zgd, &zp, &zb));
2008-11-20 20:01:55 +00:00
}
dr = db->db_last_dirty;
while (dr && dr->dr_txg != txg)
2008-11-20 20:01:55 +00:00
dr = dr->dr_next;
if (dr == NULL) {
2008-11-20 20:01:55 +00:00
/*
* There's no dr for this dbuf, so it must have been freed.
2008-11-20 20:01:55 +00:00
* There's no need to log writes to freed blocks, so we're done.
*/
mutex_exit(&db->db_mtx);
return (SET_ERROR(ENOENT));
2008-11-20 20:01:55 +00:00
}
ASSERT(dr->dr_next == NULL || dr->dr_next->dr_txg < txg);
/*
* Assume the on-disk data is X, the current syncing data (in
* txg - 1) is Y, and the current in-memory data is Z (currently
* in dmu_sync).
*
* We usually want to perform a nopwrite if X and Z are the
* same. However, if Y is different (i.e. the BP is going to
* change before this write takes effect), then a nopwrite will
* be incorrect - we would override with X, which could have
* been freed when Y was written.
*
* (Note that this is not a concern when we are nop-writing from
* syncing context, because X and Y must be identical, because
* all previous txgs have been synced.)
*
* Therefore, we disable nopwrite if the current BP could change
* before this TXG. There are two ways it could change: by
* being dirty (dr_next is non-NULL), or by being freed
* (dnode_block_freed()). This behavior is verified by
* zio_done(), which VERIFYs that the override BP is identical
* to the on-disk BP.
*/
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
if (dr->dr_next != NULL || dnode_block_freed(dn, db->db_blkid))
zp.zp_nopwrite = B_FALSE;
DB_DNODE_EXIT(db);
2008-11-20 20:01:55 +00:00
ASSERT(dr->dr_txg == txg);
if (dr->dt.dl.dr_override_state == DR_IN_DMU_SYNC ||
dr->dt.dl.dr_override_state == DR_OVERRIDDEN) {
2008-11-20 20:01:55 +00:00
/*
* We have already issued a sync write for this buffer,
* or this buffer has already been synced. It could not
2008-11-20 20:01:55 +00:00
* have been dirtied since, or we would have cleared the state.
*/
mutex_exit(&db->db_mtx);
return (SET_ERROR(EALREADY));
2008-11-20 20:01:55 +00:00
}
ASSERT(dr->dt.dl.dr_override_state == DR_NOT_OVERRIDDEN);
2008-11-20 20:01:55 +00:00
dr->dt.dl.dr_override_state = DR_IN_DMU_SYNC;
mutex_exit(&db->db_mtx);
dsa = kmem_alloc(sizeof (dmu_sync_arg_t), KM_SLEEP);
dsa->dsa_dr = dr;
dsa->dsa_done = done;
dsa->dsa_zgd = zgd;
dsa->dsa_tx = NULL;
zio_nowait(arc_write(pio, os->os_spa, txg,
bp, dr->dt.dl.dr_data, DBUF_IS_L2CACHEABLE(db),
OpenZFS 6950 - ARC should cache compressed data Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Matt Ahrens <mahrens@delphix.com> Reviewed by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Tom Caputi <tcaputi@datto.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Ported by: David Quigley <david.quigley@intel.com> This review covers the reading and writing of compressed arc headers, sharing data between the arc_hdr_t and the arc_buf_t, and the implementation of a new dbuf cache to keep frequently access data uncompressed. I've added a new member to l1 arc hdr called b_pdata. The b_pdata always hangs off the arc_buf_hdr_t (if an L1 hdr is in use) and points to the physical block for that DVA. The physical block may or may not be compressed. If compressed arc is enabled and the block on-disk is compressed, then the b_pdata will match the block on-disk and remain compressed in memory. If the block on disk is not compressed, then neither will the b_pdata. Lastly, if compressed arc is disabled, then b_pdata will always be an uncompressed version of the on-disk block. Typically the arc will cache only the arc_buf_hdr_t and will aggressively evict any arc_buf_t's that are no longer referenced. This means that the arc will primarily have compressed blocks as the arc_buf_t's are considered overhead and are always uncompressed. When a consumer reads a block we first look to see if the arc_buf_hdr_t is cached. If the hdr is cached then we allocate a new arc_buf_t and decompress the b_pdata contents into the arc_buf_t's b_data. If the hdr already has a arc_buf_t, then we will allocate an additional arc_buf_t and bcopy the uncompressed contents from the first arc_buf_t to the new one. Writing to the compressed arc requires that we first discard the b_pdata since the physical block is about to be rewritten. The new data contents will be passed in via an arc_buf_t (uncompressed) and during the I/O pipeline stages we will copy the physical block contents to a newly allocated b_pdata. When an l2arc is inuse it will also take advantage of the b_pdata. Now the l2arc will always write the contents of b_pdata to the l2arc. This means that when compressed arc is enabled that the l2arc blocks are identical to those stored in the main data pool. This provides a significant advantage since we can leverage the bp's checksum when reading from the l2arc to determine if the contents are valid. If the compressed arc is disabled, then we must first transform the read block to look like the physical block in the main data pool before comparing the checksum and determining it's valid. OpenZFS-issue: https://www.illumos.org/issues/6950 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7fc10f0 Issue #5078
2016-06-02 04:04:53 +00:00
&zp, dmu_sync_ready, NULL, NULL, dmu_sync_done, dsa,
ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, &zb));
return (0);
2008-11-20 20:01:55 +00:00
}
int
dmu_object_set_blocksize(objset_t *os, uint64_t object, uint64_t size, int ibs,
dmu_tx_t *tx)
{
dnode_t *dn;
int err;
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
err = dnode_set_blksz(dn, size, ibs, tx);
dnode_rele(dn, FTAG);
return (err);
}
void
dmu_object_set_checksum(objset_t *os, uint64_t object, uint8_t checksum,
dmu_tx_t *tx)
{
dnode_t *dn;
/*
* Send streams include each object's checksum function. This
* check ensures that the receiving system can understand the
* checksum function transmitted.
*/
ASSERT3U(checksum, <, ZIO_CHECKSUM_LEGACY_FUNCTIONS);
VERIFY0(dnode_hold(os, object, FTAG, &dn));
ASSERT3U(checksum, <, ZIO_CHECKSUM_FUNCTIONS);
2008-11-20 20:01:55 +00:00
dn->dn_checksum = checksum;
dnode_setdirty(dn, tx);
dnode_rele(dn, FTAG);
}
void
dmu_object_set_compress(objset_t *os, uint64_t object, uint8_t compress,
dmu_tx_t *tx)
{
dnode_t *dn;
/*
* Send streams include each object's compression function. This
* check ensures that the receiving system can understand the
* compression function transmitted.
*/
ASSERT3U(compress, <, ZIO_COMPRESS_LEGACY_FUNCTIONS);
VERIFY0(dnode_hold(os, object, FTAG, &dn));
2008-11-20 20:01:55 +00:00
dn->dn_compress = compress;
dnode_setdirty(dn, tx);
dnode_rele(dn, FTAG);
}
int zfs_mdcomp_disable = 0;
/*
* When the "redundant_metadata" property is set to "most", only indirect
* blocks of this level and higher will have an additional ditto block.
*/
int zfs_redundant_metadata_most_ditto_level = 2;
void
dmu_write_policy(objset_t *os, dnode_t *dn, int level, int wp,
enum zio_compress override_compress, zio_prop_t *zp)
{
dmu_object_type_t type = dn ? dn->dn_type : DMU_OT_OBJSET;
boolean_t ismd = (level > 0 || DMU_OT_IS_METADATA(type) ||
(wp & WP_SPILL));
enum zio_checksum checksum = os->os_checksum;
enum zio_compress compress = os->os_compress;
enum zio_checksum dedup_checksum = os->os_dedup_checksum;
boolean_t dedup = B_FALSE;
boolean_t nopwrite = B_FALSE;
boolean_t dedup_verify = os->os_dedup_verify;
int copies = os->os_copies;
/*
* We maintain different write policies for each of the following
* types of data:
* 1. metadata
* 2. preallocated blocks (i.e. level-0 blocks of a dump device)
* 3. all other level 0 blocks
*/
if (ismd) {
if (zfs_mdcomp_disable) {
compress = ZIO_COMPRESS_EMPTY;
} else {
/*
* XXX -- we should design a compression algorithm
* that specializes in arrays of bps.
*/
compress = zio_compress_select(os->os_spa,
ZIO_COMPRESS_ON, ZIO_COMPRESS_ON);
}
/*
* Metadata always gets checksummed. If the data
* checksum is multi-bit correctable, and it's not a
* ZBT-style checksum, then it's suitable for metadata
* as well. Otherwise, the metadata checksum defaults
* to fletcher4.
*/
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
if (!(zio_checksum_table[checksum].ci_flags &
ZCHECKSUM_FLAG_METADATA) ||
(zio_checksum_table[checksum].ci_flags &
ZCHECKSUM_FLAG_EMBEDDED))
checksum = ZIO_CHECKSUM_FLETCHER_4;
if (os->os_redundant_metadata == ZFS_REDUNDANT_METADATA_ALL ||
(os->os_redundant_metadata ==
ZFS_REDUNDANT_METADATA_MOST &&
(level >= zfs_redundant_metadata_most_ditto_level ||
DMU_OT_IS_METADATA(type) || (wp & WP_SPILL))))
copies++;
} else if (wp & WP_NOFILL) {
ASSERT(level == 0);
/*
* If we're writing preallocated blocks, we aren't actually
* writing them so don't set any policy properties. These
* blocks are currently only used by an external subsystem
* outside of zfs (i.e. dump) and not written by the zio
* pipeline.
*/
compress = ZIO_COMPRESS_OFF;
checksum = ZIO_CHECKSUM_OFF;
} else {
compress = zio_compress_select(os->os_spa, dn->dn_compress,
compress);
checksum = (dedup_checksum == ZIO_CHECKSUM_OFF) ?
zio_checksum_select(dn->dn_checksum, checksum) :
dedup_checksum;
/*
* Determine dedup setting. If we are in dmu_sync(),
* we won't actually dedup now because that's all
* done in syncing context; but we do want to use the
* dedup checkum. If the checksum is not strong
* enough to ensure unique signatures, force
* dedup_verify.
*/
if (dedup_checksum != ZIO_CHECKSUM_OFF) {
dedup = (wp & WP_DMU_SYNC) ? B_FALSE : B_TRUE;
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
if (!(zio_checksum_table[checksum].ci_flags &
ZCHECKSUM_FLAG_DEDUP))
dedup_verify = B_TRUE;
}
/*
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
* Enable nopwrite if we have secure enough checksum
* algorithm (see comment in zio_nop_write) and
* compression is enabled. We don't enable nopwrite if
* dedup is enabled as the two features are mutually
* exclusive.
*/
OpenZFS 4185 - add new cryptographic checksums to ZFS: SHA-512, Skein, Edon-R Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com> Reviewed by: Richard Lowe <richlowe@richlowe.net> Approved by: Garrett D'Amore <garrett@damore.org> Ported by: Tony Hutter <hutter2@llnl.gov> OpenZFS-issue: https://www.illumos.org/issues/4185 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/45818ee Porting Notes: This code is ported on top of the Illumos Crypto Framework code: https://github.com/zfsonlinux/zfs/pull/4329/commits/b5e030c8dbb9cd393d313571dee4756fbba8c22d The list of porting changes includes: - Copied module/icp/include/sha2/sha2.h directly from illumos - Removed from module/icp/algs/sha2/sha2.c: #pragma inline(SHA256Init, SHA384Init, SHA512Init) - Added 'ctx' to lib/libzfs/libzfs_sendrecv.c:zio_checksum_SHA256() since it now takes in an extra parameter. - Added CTASSERT() to assert.h from for module/zfs/edonr_zfs.c - Added skein & edonr to libicp/Makefile.am - Added sha512.S. It was generated from sha512-x86_64.pl in Illumos. - Updated ztest.c with new fletcher_4_*() args; used NULL for new CTX argument. - In icp/algs/edonr/edonr_byteorder.h, Removed the #if defined(__linux) section to not #include the non-existant endian.h. - In skein_test.c, renane NULL to 0 in "no test vector" array entries to get around a compiler warning. - Fixup test files: - Rename <sys/varargs.h> -> <varargs.h>, <strings.h> -> <string.h>, - Remove <note.h> and define NOTE() as NOP. - Define u_longlong_t - Rename "#!/usr/bin/ksh" -> "#!/bin/ksh -p" - Rename NULL to 0 in "no test vector" array entries to get around a compiler warning. - Remove "for isa in $($ISAINFO); do" stuff - Add/update Makefiles - Add some userspace headers like stdio.h/stdlib.h in places of sys/types.h. - EXPORT_SYMBOL *_Init/*_Update/*_Final... routines in ICP modules. - Update scripts/zfs2zol-patch.sed - include <sys/sha2.h> in sha2_impl.h - Add sha2.h to include/sys/Makefile.am - Add skein and edonr dirs to icp Makefile - Add new checksums to zpool_get.cfg - Move checksum switch block from zfs_secpolicy_setprop() to zfs_check_settable() - Fix -Wuninitialized error in edonr_byteorder.h on PPC - Fix stack frame size errors on ARM32 - Don't unroll loops in Skein on 32-bit to save stack space - Add memory barriers in sha2.c on 32-bit to save stack space - Add filetest_001_pos.ksh checksum sanity test - Add option to write psudorandom data in file_write utility
2016-06-15 22:47:05 +00:00
nopwrite = (!dedup && (zio_checksum_table[checksum].ci_flags &
ZCHECKSUM_FLAG_NOPWRITE) &&
compress != ZIO_COMPRESS_OFF && zfs_nopwrite_enabled);
}
zp->zp_checksum = checksum;
/*
* If we're writing a pre-compressed buffer, the compression type we use
* must match the data. If it hasn't been compressed yet, then we should
* use the value dictated by the policies above.
*/
zp->zp_compress = override_compress != ZIO_COMPRESS_INHERIT
? override_compress : compress;
ASSERT3U(zp->zp_compress, !=, ZIO_COMPRESS_INHERIT);
zp->zp_type = (wp & WP_SPILL) ? dn->dn_bonustype : type;
zp->zp_level = level;
zp->zp_copies = MIN(copies, spa_max_replication(os->os_spa));
zp->zp_dedup = dedup;
zp->zp_dedup_verify = dedup && dedup_verify;
zp->zp_nopwrite = nopwrite;
}
2008-11-20 20:01:55 +00:00
int
dmu_offset_next(objset_t *os, uint64_t object, boolean_t hole, uint64_t *off)
{
dnode_t *dn;
int i, err;
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
/*
* Sync any current changes before
* we go trundling through the block pointers.
*/
for (i = 0; i < TXG_SIZE; i++) {
if (list_link_active(&dn->dn_dirty_link[i]))
break;
}
if (i != TXG_SIZE) {
dnode_rele(dn, FTAG);
txg_wait_synced(dmu_objset_pool(os), 0);
err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
}
err = dnode_next_offset(dn, (hole ? DNODE_FIND_HOLE : 0), off, 1, 1, 0);
2008-11-20 20:01:55 +00:00
dnode_rele(dn, FTAG);
return (err);
}
void
__dmu_object_info_from_dnode(dnode_t *dn, dmu_object_info_t *doi)
2008-11-20 20:01:55 +00:00
{
dnode_phys_t *dnp = dn->dn_phys;
int i;
2008-11-20 20:01:55 +00:00
doi->doi_data_block_size = dn->dn_datablksz;
doi->doi_metadata_block_size = dn->dn_indblkshift ?
1ULL << dn->dn_indblkshift : 0;
doi->doi_type = dn->dn_type;
doi->doi_bonus_type = dn->dn_bonustype;
doi->doi_bonus_size = dn->dn_bonuslen;
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
doi->doi_dnodesize = dn->dn_num_slots << DNODE_SHIFT;
2008-11-20 20:01:55 +00:00
doi->doi_indirection = dn->dn_nlevels;
doi->doi_checksum = dn->dn_checksum;
doi->doi_compress = dn->dn_compress;
2014-09-12 03:28:35 +00:00
doi->doi_nblkptr = dn->dn_nblkptr;
doi->doi_physical_blocks_512 = (DN_USED_BYTES(dnp) + 256) >> 9;
doi->doi_max_offset = (dn->dn_maxblkid + 1) * dn->dn_datablksz;
doi->doi_fill_count = 0;
for (i = 0; i < dnp->dn_nblkptr; i++)
doi->doi_fill_count += BP_GET_FILL(&dnp->dn_blkptr[i]);
}
void
dmu_object_info_from_dnode(dnode_t *dn, dmu_object_info_t *doi)
{
rw_enter(&dn->dn_struct_rwlock, RW_READER);
mutex_enter(&dn->dn_mtx);
__dmu_object_info_from_dnode(dn, doi);
2008-11-20 20:01:55 +00:00
mutex_exit(&dn->dn_mtx);
rw_exit(&dn->dn_struct_rwlock);
}
/*
* Get information on a DMU object.
* If doi is NULL, just indicates whether the object exists.
*/
int
dmu_object_info(objset_t *os, uint64_t object, dmu_object_info_t *doi)
{
dnode_t *dn;
int err = dnode_hold(os, object, FTAG, &dn);
2008-11-20 20:01:55 +00:00
if (err)
return (err);
if (doi != NULL)
dmu_object_info_from_dnode(dn, doi);
dnode_rele(dn, FTAG);
return (0);
}
/*
* As above, but faster; can be used when you have a held dbuf in hand.
*/
void
dmu_object_info_from_db(dmu_buf_t *db_fake, dmu_object_info_t *doi)
2008-11-20 20:01:55 +00:00
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
DB_DNODE_ENTER(db);
dmu_object_info_from_dnode(DB_DNODE(db), doi);
DB_DNODE_EXIT(db);
2008-11-20 20:01:55 +00:00
}
/*
* Faster still when you only care about the size.
* This is specifically optimized for zfs_getattr().
*/
void
dmu_object_size_from_db(dmu_buf_t *db_fake, uint32_t *blksize,
u_longlong_t *nblk512)
2008-11-20 20:01:55 +00:00
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
2008-11-20 20:01:55 +00:00
*blksize = dn->dn_datablksz;
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
/* add in number of slots used for the dnode itself */
2008-11-20 20:01:55 +00:00
*nblk512 = ((DN_USED_BYTES(dn->dn_phys) + SPA_MINBLOCKSIZE/2) >>
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
SPA_MINBLOCKSHIFT) + dn->dn_num_slots;
DB_DNODE_EXIT(db);
}
void
dmu_object_dnsize_from_db(dmu_buf_t *db_fake, int *dnsize)
{
dmu_buf_impl_t *db = (dmu_buf_impl_t *)db_fake;
dnode_t *dn;
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
*dnsize = dn->dn_num_slots << DNODE_SHIFT;
DB_DNODE_EXIT(db);
2008-11-20 20:01:55 +00:00
}
void
byteswap_uint64_array(void *vbuf, size_t size)
{
uint64_t *buf = vbuf;
size_t count = size >> 3;
int i;
ASSERT((size & 7) == 0);
for (i = 0; i < count; i++)
buf[i] = BSWAP_64(buf[i]);
}
void
byteswap_uint32_array(void *vbuf, size_t size)
{
uint32_t *buf = vbuf;
size_t count = size >> 2;
int i;
ASSERT((size & 3) == 0);
for (i = 0; i < count; i++)
buf[i] = BSWAP_32(buf[i]);
}
void
byteswap_uint16_array(void *vbuf, size_t size)
{
uint16_t *buf = vbuf;
size_t count = size >> 1;
int i;
ASSERT((size & 1) == 0);
for (i = 0; i < count; i++)
buf[i] = BSWAP_16(buf[i]);
}
/* ARGSUSED */
void
byteswap_uint8_array(void *vbuf, size_t size)
{
}
void
dmu_init(void)
{
zfs_dbgmsg_init();
sa_cache_init();
xuio_stat_init();
dmu_objset_init();
2008-11-20 20:01:55 +00:00
dnode_init();
zfetch_init();
dmu_tx_init();
2008-11-20 20:01:55 +00:00
l2arc_init();
arc_init();
OpenZFS 6950 - ARC should cache compressed data Authored by: George Wilson <george.wilson@delphix.com> Reviewed by: Prakash Surya <prakash.surya@delphix.com> Reviewed by: Dan Kimmel <dan.kimmel@delphix.com> Reviewed by: Matt Ahrens <mahrens@delphix.com> Reviewed by: Paul Dagnelie <pcd@delphix.com> Reviewed by: Tom Caputi <tcaputi@datto.com> Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov> Ported by: David Quigley <david.quigley@intel.com> This review covers the reading and writing of compressed arc headers, sharing data between the arc_hdr_t and the arc_buf_t, and the implementation of a new dbuf cache to keep frequently access data uncompressed. I've added a new member to l1 arc hdr called b_pdata. The b_pdata always hangs off the arc_buf_hdr_t (if an L1 hdr is in use) and points to the physical block for that DVA. The physical block may or may not be compressed. If compressed arc is enabled and the block on-disk is compressed, then the b_pdata will match the block on-disk and remain compressed in memory. If the block on disk is not compressed, then neither will the b_pdata. Lastly, if compressed arc is disabled, then b_pdata will always be an uncompressed version of the on-disk block. Typically the arc will cache only the arc_buf_hdr_t and will aggressively evict any arc_buf_t's that are no longer referenced. This means that the arc will primarily have compressed blocks as the arc_buf_t's are considered overhead and are always uncompressed. When a consumer reads a block we first look to see if the arc_buf_hdr_t is cached. If the hdr is cached then we allocate a new arc_buf_t and decompress the b_pdata contents into the arc_buf_t's b_data. If the hdr already has a arc_buf_t, then we will allocate an additional arc_buf_t and bcopy the uncompressed contents from the first arc_buf_t to the new one. Writing to the compressed arc requires that we first discard the b_pdata since the physical block is about to be rewritten. The new data contents will be passed in via an arc_buf_t (uncompressed) and during the I/O pipeline stages we will copy the physical block contents to a newly allocated b_pdata. When an l2arc is inuse it will also take advantage of the b_pdata. Now the l2arc will always write the contents of b_pdata to the l2arc. This means that when compressed arc is enabled that the l2arc blocks are identical to those stored in the main data pool. This provides a significant advantage since we can leverage the bp's checksum when reading from the l2arc to determine if the contents are valid. If the compressed arc is disabled, then we must first transform the read block to look like the physical block in the main data pool before comparing the checksum and determining it's valid. OpenZFS-issue: https://www.illumos.org/issues/6950 OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7fc10f0 Issue #5078
2016-06-02 04:04:53 +00:00
dbuf_init();
2008-11-20 20:01:55 +00:00
}
void
dmu_fini(void)
{
arc_fini(); /* arc depends on l2arc, so arc must go first */
l2arc_fini();
dmu_tx_fini();
zfetch_fini();
2008-11-20 20:01:55 +00:00
dbuf_fini();
dnode_fini();
dmu_objset_fini();
xuio_stat_fini();
sa_cache_fini();
zfs_dbgmsg_fini();
2008-11-20 20:01:55 +00:00
}
#if defined(_KERNEL) && defined(HAVE_SPL)
EXPORT_SYMBOL(dmu_bonus_hold);
EXPORT_SYMBOL(dmu_buf_hold_array_by_bonus);
EXPORT_SYMBOL(dmu_buf_rele_array);
EXPORT_SYMBOL(dmu_prefetch);
EXPORT_SYMBOL(dmu_free_range);
EXPORT_SYMBOL(dmu_free_long_range);
EXPORT_SYMBOL(dmu_free_long_object);
EXPORT_SYMBOL(dmu_read);
EXPORT_SYMBOL(dmu_write);
EXPORT_SYMBOL(dmu_prealloc);
EXPORT_SYMBOL(dmu_object_info);
EXPORT_SYMBOL(dmu_object_info_from_dnode);
EXPORT_SYMBOL(dmu_object_info_from_db);
EXPORT_SYMBOL(dmu_object_size_from_db);
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
EXPORT_SYMBOL(dmu_object_dnsize_from_db);
EXPORT_SYMBOL(dmu_object_set_blocksize);
EXPORT_SYMBOL(dmu_object_set_checksum);
EXPORT_SYMBOL(dmu_object_set_compress);
EXPORT_SYMBOL(dmu_write_policy);
EXPORT_SYMBOL(dmu_sync);
EXPORT_SYMBOL(dmu_request_arcbuf);
EXPORT_SYMBOL(dmu_return_arcbuf);
EXPORT_SYMBOL(dmu_assign_arcbuf);
EXPORT_SYMBOL(dmu_buf_hold);
EXPORT_SYMBOL(dmu_ot);
module_param(zfs_mdcomp_disable, int, 0644);
MODULE_PARM_DESC(zfs_mdcomp_disable, "Disable meta data compression");
module_param(zfs_nopwrite_enabled, int, 0644);
MODULE_PARM_DESC(zfs_nopwrite_enabled, "Enable NOP writes");
#endif