Print this page
NEX-9752 backport illumos 6950 ARC should cache compressed data
Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Reviewed by: Yuri Pankov <yuri.pankov@nexenta.com>
6950 ARC should cache compressed data
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: Don Brady <don.brady@intel.com>
Reviewed by: Richard Elling <Richard.Elling@RichardElling.com>
Approved by: Richard Lowe <richlowe@richlowe.net>
NEX-7603 Back port OpenZFS #188 Create tunable to ignore hole_birth
feature
Reviewed by: Roman Strashkin <roman.strashkin@nexenta.com>
Reviewed by: Dan Fields <dan.fields@nexenta.com>
2605 want to resume interrupted zfs send
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: Richard Elling <Richard.Elling@RichardElling.com>
Reviewed by: Xin Li <delphij@freebsd.org>
Reviewed by: Arne Jansen <sensille@gmx.net>
Approved by: Dan McDonald <danmcd@omniti.com>
NEX-4582 update wrc test cases for allow to use write back cache per tree of datasets
Reviewed by: Steve Peng <steve.peng@nexenta.com>
Reviewed by: Alex Aizman <alex.aizman@nexenta.com>
5960 zfs recv should prefetch indirect blocks
5925 zfs receive -o origin=
Reviewed by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
NEX-4207 WRC and dedup on the same pool cause system-panic
Reviewed by: Alek Pinchuk <alek.pinchuk@nexenta.com>
Reviewed by: Alex Aizman <alex.aizman@nexenta.com>
NEX-4193 WRC does not migrate data that belong to intermediate snapshots
Reviewed by: Alek Pinchuk <alek.pinchuk@nexenta.com>
Reviewed by: Alex Aizman <alex.aizman@nexenta.com>
NEX-3710 WRC improvements and bug-fixes
* refactored WRC move-logic to use zio kmem_cashes
* replace size and compression fields by blk_prop field
(the same in blkptr_t) to little reduce size of wrc_block_t
and use similar macros as for blkptr_t to get PSIZE, LSIZE
and COMPRESSION
* make CPU more happy by reduce atomic calls
* removed unused code
* fixed naming of variables
* fixed possible system panic after restart system
with enabled WRC
* fixed a race that causes system panic
Reviewed by: Alek Pinchuk <alek@nexenta.com>
Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com>
NEX-3558 KRRP Integration
4459 Typo in itadm(1m) usage message: delete-inititator
Reviewed by: Milan Jurik <milan.jurik@xylab.cz>
Reviewed by: Marcel Telka <marcel@telka.sk>
Approved by: Robert Mustacchi <rm@joyent.com>
4504 traverse_visitbp: visit DMU_GROUPUSED_OBJECT before DMU_USERUSED_OBJECT
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Andriy Gapon <andriy.gapon@hybridcluster.com>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Approved by: Robert Mustacchi <rm@joyent.com>
4391 panic system rather than corrupting pool if we hit bug 4390
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Gordon Ross <gwr@nexenta.com>
4370 avoid transmitting holes during zfs send
4371 DMU code clean up
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Approved by: Garrett D'Amore <garrett@damore.org>
re #12619 rb4429 More dp->dp_config_rwlock holds
re #12585 rb4049 ZFS++ work port - refactoring to improve separation of open/closed code, bug fixes, performance improvements - open code
Bug 11205: add missing libzfs_closed_stubs.c to fix opensource-only build.
ZFS plus work: special vdevs, cos, cos/vdev properties
@@ -18,10 +18,11 @@
*
* CDDL HEADER END
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Copyright 2015 Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2012, 2016 by Delphix. All rights reserved.
*/
#include <sys/zfs_context.h>
#include <sys/dmu_objset.h>
@@ -54,10 +55,11 @@
typedef struct traverse_data {
spa_t *td_spa;
uint64_t td_objset;
blkptr_t *td_rootbp;
uint64_t td_min_txg;
+ uint64_t td_max_txg;
zbookmark_phys_t *td_resume;
int td_flags;
prefetch_data_t *td_pfd;
boolean_t td_paused;
uint64_t td_hole_birth_enabled_txg;
@@ -189,11 +191,12 @@
* some children will not be needed (and in fact may have already
* been freed).
*/
if (td->td_resume != NULL && !ZB_IS_ZERO(td->td_resume))
return;
- if (BP_IS_HOLE(bp) || bp->blk_birth <= td->td_min_txg)
+ if (BP_IS_HOLE(bp) || bp->blk_birth <= td->td_min_txg ||
+ bp->blk_birth >= td->td_max_txg)
return;
if (BP_GET_LEVEL(bp) == 0 && BP_GET_TYPE(bp) != DMU_OT_DNODE)
return;
(void) arc_read(NULL, td->td_spa, bp, NULL, NULL,
@@ -255,13 +258,15 @@
* Note that the meta-dnode cannot be reallocated.
*/
if (!send_holes_without_birth_time &&
(!td->td_realloc_possible ||
zb->zb_object == DMU_META_DNODE_OBJECT) &&
- td->td_hole_birth_enabled_txg <= td->td_min_txg)
+ (td->td_hole_birth_enabled_txg <= td->td_min_txg ||
+ td->td_hole_birth_enabled_txg > td->td_max_txg))
return (0);
- } else if (bp->blk_birth <= td->td_min_txg) {
+ } else if (bp->blk_birth <= td->td_min_txg ||
+ bp->blk_birth >= td->td_max_txg) {
return (0);
}
if (pd != NULL && !pd->pd_exited && prefetch_needed(pd, bp)) {
uint64_t size = BP_GET_LSIZE(bp);
@@ -284,10 +289,13 @@
if (td->td_flags & TRAVERSE_PRE) {
err = td->td_func(td->td_spa, NULL, bp, zb, dnp,
td->td_arg);
if (err == TRAVERSE_VISIT_NO_CHILDREN)
return (0);
+ /* handle pausing at a common point */
+ if (err == ERESTART)
+ td->td_paused = B_TRUE;
if (err != 0)
goto post;
}
if (BP_GET_LEVEL(bp) > 0) {
@@ -418,10 +426,16 @@
(dnp->dn_indblkshift - SPA_BLKPTRSHIFT);
}
td->td_paused = B_TRUE;
}
+ /* if we walked over all bp bookmark must be cleared */
+ if (!err && !td->td_paused && td->td_resume != NULL &&
+ bp == td->td_rootbp && td->td_pfd != NULL) {
+ bzero(td->td_resume, sizeof (*td->td_resume));
+ }
+
return (err);
}
static void
prefetch_dnode_metadata(traverse_data_t *td, const dnode_phys_t *dnp,
@@ -544,12 +558,12 @@
* NB: dataset must not be changing on-disk (eg, is a snapshot or we are
* in syncing context).
*/
static int
traverse_impl(spa_t *spa, dsl_dataset_t *ds, uint64_t objset, blkptr_t *rootbp,
- uint64_t txg_start, zbookmark_phys_t *resume, int flags,
- blkptr_cb_t func, void *arg)
+ uint64_t txg_start, uint64_t txg_finish, zbookmark_phys_t *resume,
+ int flags, blkptr_cb_t func, void *arg)
{
traverse_data_t td;
prefetch_data_t pd = { 0 };
zbookmark_phys_t czb;
int err;
@@ -559,10 +573,11 @@
td.td_spa = spa;
td.td_objset = objset;
td.td_rootbp = rootbp;
td.td_min_txg = txg_start;
+ td.td_max_txg = txg_finish;
td.td_resume = resume;
td.td_func = func;
td.td_arg = arg;
td.td_pfd = &pd;
td.td_flags = flags;
@@ -629,11 +644,12 @@
traverse_dataset_resume(dsl_dataset_t *ds, uint64_t txg_start,
zbookmark_phys_t *resume,
int flags, blkptr_cb_t func, void *arg)
{
return (traverse_impl(ds->ds_dir->dd_pool->dp_spa, ds, ds->ds_object,
- &dsl_dataset_phys(ds)->ds_bp, txg_start, resume, flags, func, arg));
+ &dsl_dataset_phys(ds)->ds_bp, txg_start, UINT64_MAX, resume, flags,
+ func, arg));
}
int
traverse_dataset(dsl_dataset_t *ds, uint64_t txg_start,
int flags, blkptr_cb_t func, void *arg)
@@ -645,33 +661,36 @@
traverse_dataset_destroyed(spa_t *spa, blkptr_t *blkptr,
uint64_t txg_start, zbookmark_phys_t *resume, int flags,
blkptr_cb_t func, void *arg)
{
return (traverse_impl(spa, NULL, ZB_DESTROYED_OBJSET,
- blkptr, txg_start, resume, flags, func, arg));
+ blkptr, txg_start, UINT64_MAX, resume, flags, func, arg));
}
/*
* NB: pool must not be changing on-disk (eg, from zdb or sync context).
*/
int
-traverse_pool(spa_t *spa, uint64_t txg_start, int flags,
- blkptr_cb_t func, void *arg)
+traverse_pool(spa_t *spa, uint64_t txg_start, uint64_t txg_finish, int flags,
+ blkptr_cb_t func, void *arg, zbookmark_phys_t *zb)
{
- int err;
+ int err = 0, lasterr = 0;
dsl_pool_t *dp = spa_get_dsl(spa);
objset_t *mos = dp->dp_meta_objset;
boolean_t hard = (flags & TRAVERSE_HARD);
/* visit the MOS */
+ if (!zb || (zb->zb_objset == 0 && zb->zb_object == 0)) {
err = traverse_impl(spa, NULL, 0, spa_get_rootblkptr(spa),
- txg_start, NULL, flags, func, arg);
+ txg_start, txg_finish, NULL, flags, func, arg);
if (err != 0)
return (err);
+ }
/* visit each dataset */
- for (uint64_t obj = 1; err == 0;
+ for (uint64_t obj = (zb && !ZB_IS_ZERO(zb))? zb->zb_objset : 1;
+ err == 0 || (err != ESRCH && hard);
err = dmu_object_next(mos, &obj, B_FALSE, txg_start)) {
dmu_object_info_t doi;
err = dmu_object_info(mos, obj, &doi);
if (err != 0) {
@@ -680,27 +699,73 @@
break;
}
if (doi.doi_bonus_type == DMU_OT_DSL_DATASET) {
dsl_dataset_t *ds;
+ objset_t *os;
+ boolean_t os_is_snapshot = B_FALSE;
uint64_t txg = txg_start;
+ uint64_t ctxg;
+ uint64_t max_txg = txg_finish;
dsl_pool_config_enter(dp, FTAG);
err = dsl_dataset_hold_obj(dp, obj, FTAG, &ds);
dsl_pool_config_exit(dp, FTAG);
if (err != 0) {
if (hard)
continue;
break;
}
- if (dsl_dataset_phys(ds)->ds_prev_snap_txg > txg)
+
+ dsl_pool_config_enter(dp, FTAG);
+ err = dmu_objset_from_ds(ds, &os);
+ if (err == 0)
+ os_is_snapshot = dmu_objset_is_snapshot(os);
+
+ dsl_pool_config_exit(dp, FTAG);
+ if (err != 0) {
+ dsl_dataset_rele(ds, FTAG);
+ if (hard)
+ continue;
+ break;
+ }
+ ctxg = dsl_dataset_phys(ds)->ds_creation_txg;
+
+ /* uplimited traverse walks over shapshots only */
+ if (max_txg != UINT64_MAX && !os_is_snapshot) {
+ dsl_dataset_rele(ds, FTAG);
+ continue;
+ }
+ if (max_txg != UINT64_MAX && ctxg >= max_txg) {
+ dsl_dataset_rele(ds, FTAG);
+ continue;
+ }
+ if (os_is_snapshot && ctxg <= txg_start) {
+ dsl_dataset_rele(ds, FTAG);
+ continue;
+ }
+ if (max_txg == UINT64_MAX &&
+ dsl_dataset_phys(ds)->ds_prev_snap_txg > txg)
txg = dsl_dataset_phys(ds)->ds_prev_snap_txg;
- err = traverse_dataset(ds, txg, flags, func, arg);
+ if (txg > max_txg)
+ max_txg = txg;
+ err = traverse_impl(spa, ds, ds->ds_object,
+ &dsl_dataset_phys(ds)->ds_bp,
+ txg, max_txg, zb, flags, func, arg);
dsl_dataset_rele(ds, FTAG);
- if (err != 0)
+ if (err != 0) {
+ if (!hard)
+ return (err);
+ lasterr = err;
+ }
+ if (zb && !ZB_IS_ZERO(zb))
break;
}
}
- if (err == ESRCH)
+ if (err == ESRCH) {
+ /* zero bookmark means we are done */
+ if (zb)
+ bzero(zb, sizeof (*zb));
err = 0;
- return (err);
+ }
+ return (err != 0 ? err : lasterr);
}