Merge illumos revisions 13572, 13573, 13574:

Rev. 13572:
disk sync write perf regression when slog is used post oi_148 [1]

Rev. 13573:
crash during reguid causes stale config [2]
allow and unallow missing from zpool history since removal of pyzfs [5]

Rev. 13574:
leaking a vdev when removing an l2cache device [3]
memory leak when adding a file-based l2arc device [4]
leak in ZFS from metaslab_group_create and zfs_ereport_checksum [6]

References:
https://www.illumos.org/issues/1909 [1]
https://www.illumos.org/issues/1949 [2]
https://www.illumos.org/issues/1951 [3]
https://www.illumos.org/issues/1952 [4]
https://www.illumos.org/issues/1953 [5]
https://www.illumos.org/issues/1954 [6]

Obtained from:	illumos (issues #1909, #1949, #1951, #1952, #1953, #1954)
MFC after:	2 weeks
This commit is contained in:
Martin Matuska 2012-01-24 23:09:54 +00:00
parent bf318c6aa2
commit 538251bbf6
8 changed files with 52 additions and 24 deletions

View File

@ -4213,7 +4213,7 @@ zfs_get_fsacl(zfs_handle_t *zhp, nvlist_t **nvl)
(void) strlcpy(zc.zc_name, zhp->zfs_name, ZFS_MAXNAMELEN);
if (zfs_ioctl(hdl, ZFS_IOC_GET_FSACL, &zc) != 0) {
if (ioctl(hdl->libzfs_fd, ZFS_IOC_GET_FSACL, &zc) != 0) {
(void) snprintf(errbuf, sizeof (errbuf),
dgettext(TEXT_DOMAIN, "cannot get permissions on '%s'"),
zc.zc_name);

View File

@ -20,7 +20,7 @@
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2011 by Delphix. All rights reserved.
* Copyright (c) 2012 by Delphix. All rights reserved.
*/
#include <sys/zfs_context.h>
@ -36,7 +36,7 @@
* avoid having to load lots of space_maps in a given txg. There are,
* however, some cases where we want to avoid "fast" ganging and instead
* we want to do an exhaustive search of all metaslabs on this device.
* Currently we don't allow any gang or dump device related allocations
* Currently we don't allow any gang, zil, or dump device related allocations
* to "fast" gang.
*/
#define CAN_FASTGANG(flags) \

View File

@ -1073,8 +1073,10 @@ spa_unload(spa_t *spa)
}
spa->spa_spares.sav_count = 0;
for (i = 0; i < spa->spa_l2cache.sav_count; i++)
for (i = 0; i < spa->spa_l2cache.sav_count; i++) {
vdev_clear_stats(spa->spa_l2cache.sav_vdevs[i]);
vdev_free(spa->spa_l2cache.sav_vdevs[i]);
}
if (spa->spa_l2cache.sav_vdevs) {
kmem_free(spa->spa_l2cache.sav_vdevs,
spa->spa_l2cache.sav_count * sizeof (void *));
@ -1302,11 +1304,13 @@ spa_load_l2cache(spa_t *spa)
vd = oldvdevs[i];
if (vd != NULL) {
ASSERT(vd->vdev_isl2cache);
if (spa_l2cache_exists(vd->vdev_guid, &pool) &&
pool != 0ULL && l2arc_vdev_present(vd))
l2arc_remove_vdev(vd);
(void) vdev_close(vd);
spa_l2cache_remove(vd);
vdev_clear_stats(vd);
vdev_free(vd);
}
}
@ -1949,7 +1953,7 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config,
*/
if (type != SPA_IMPORT_ASSEMBLE) {
spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
error = vdev_validate(rvd);
error = vdev_validate(rvd, mosconfig);
spa_config_exit(spa, SCL_ALL, FTAG);
if (error != 0)
@ -2818,6 +2822,7 @@ spa_validate_aux_devs(spa_t *spa, nvlist_t *nvroot, uint64_t crtxg, int mode,
if ((strcmp(config, ZPOOL_CONFIG_L2CACHE) == 0) &&
strcmp(vd->vdev_ops->vdev_op_type, VDEV_TYPE_DISK) != 0) {
error = ENOTBLK;
vdev_free(vd);
goto out;
}
#endif
@ -2927,10 +2932,6 @@ spa_l2cache_drop(spa_t *spa)
if (spa_l2cache_exists(vd->vdev_guid, &pool) &&
pool != 0ULL && l2arc_vdev_present(vd))
l2arc_remove_vdev(vd);
if (vd->vdev_isl2cache)
spa_l2cache_remove(vd);
vdev_clear_stats(vd);
(void) vdev_close(vd);
}
}
@ -3929,7 +3930,7 @@ spa_vdev_attach(spa_t *spa, uint64_t guid, nvlist_t *nvroot, int replacing)
pvd = oldvd->vdev_parent;
if ((error = spa_config_parse(spa, &newrootvd, nvroot, NULL, 0,
VDEV_ALLOC_ADD)) != 0)
VDEV_ALLOC_ATTACH)) != 0)
return (spa_vdev_exit(spa, NULL, txg, EINVAL));
if (newrootvd->vdev_children != 1)

View File

@ -20,6 +20,7 @@
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2012 by Delphix. All rights reserved.
*/
#ifndef _SYS_VDEV_H
@ -48,7 +49,7 @@ extern boolean_t zfs_nocacheflush;
extern int vdev_open(vdev_t *);
extern void vdev_open_children(vdev_t *);
extern boolean_t vdev_uses_zvols(vdev_t *);
extern int vdev_validate(vdev_t *);
extern int vdev_validate(vdev_t *, boolean_t);
extern void vdev_close(vdev_t *);
extern int vdev_create(vdev_t *, uint64_t txg, boolean_t isreplace);
extern void vdev_reopen(vdev_t *);

View File

@ -261,6 +261,7 @@ typedef struct vdev_label {
#define VDEV_ALLOC_L2CACHE 3
#define VDEV_ALLOC_ROOTPOOL 4
#define VDEV_ALLOC_SPLIT 5
#define VDEV_ALLOC_ATTACH 6
/*
* Allocate or free a vdev

View File

@ -499,7 +499,7 @@ vdev_alloc(spa_t *spa, vdev_t **vdp, nvlist_t *nv, vdev_t *parent, uint_t id,
&vd->vdev_removing);
}
if (parent && !parent->vdev_parent) {
if (parent && !parent->vdev_parent && alloctype != VDEV_ALLOC_ATTACH) {
ASSERT(alloctype == VDEV_ALLOC_LOAD ||
alloctype == VDEV_ALLOC_ADD ||
alloctype == VDEV_ALLOC_SPLIT ||
@ -675,6 +675,8 @@ vdev_top_transfer(vdev_t *svd, vdev_t *tvd)
svd->vdev_ms_shift = 0;
svd->vdev_ms_count = 0;
if (tvd->vdev_mg)
ASSERT3P(tvd->vdev_mg, ==, svd->vdev_mg);
tvd->vdev_mg = svd->vdev_mg;
tvd->vdev_ms = svd->vdev_ms;
@ -1294,13 +1296,18 @@ vdev_open(vdev_t *vd)
* contents. This needs to be done before vdev_load() so that we don't
* inadvertently do repair I/Os to the wrong device.
*
* If 'strict' is false ignore the spa guid check. This is necessary because
* if the machine crashed during a re-guid the new guid might have been written
* to all of the vdev labels, but not the cached config. The strict check
* will be performed when the pool is opened again using the mos config.
*
* This function will only return failure if one of the vdevs indicates that it
* has since been destroyed or exported. This is only possible if
* /etc/zfs/zpool.cache was readonly at the time. Otherwise, the vdev state
* will be updated but the function will return 0.
*/
int
vdev_validate(vdev_t *vd)
vdev_validate(vdev_t *vd, boolean_t strict)
{
spa_t *spa = vd->vdev_spa;
nvlist_t *label;
@ -1308,7 +1315,7 @@ vdev_validate(vdev_t *vd)
uint64_t state;
for (int c = 0; c < vd->vdev_children; c++)
if (vdev_validate(vd->vdev_child[c]) != 0)
if (vdev_validate(vd->vdev_child[c], strict) != 0)
return (EBADF);
/*
@ -1338,8 +1345,9 @@ vdev_validate(vdev_t *vd)
return (0);
}
if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_GUID,
&guid) != 0 || guid != spa_guid(spa)) {
if (strict && (nvlist_lookup_uint64(label,
ZPOOL_CONFIG_POOL_GUID, &guid) != 0 ||
guid != spa_guid(spa))) {
vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN,
VDEV_AUX_CORRUPT_DATA);
nvlist_free(label);
@ -1501,7 +1509,7 @@ vdev_reopen(vdev_t *vd)
!l2arc_vdev_present(vd))
l2arc_add_vdev(spa, vd);
} else {
(void) vdev_validate(vd);
(void) vdev_validate(vd, B_TRUE);
}
/*

View File

@ -23,6 +23,10 @@
* Use is subject to license terms.
*/
/*
* Copyright (c) 2012 by Delphix. All rights reserved.
*/
#include <sys/spa.h>
#include <sys/spa_impl.h>
#include <sys/vdev.h>
@ -709,6 +713,10 @@ zfs_ereport_start_checksum(spa_t *spa, vdev_t *vd,
if (report->zcr_ereport == NULL) {
report->zcr_free(report->zcr_cbdata, report->zcr_cbinfo);
if (report->zcr_ckinfo != NULL) {
kmem_free(report->zcr_ckinfo,
sizeof (*report->zcr_ckinfo));
}
kmem_free(report, sizeof (*report));
return;
}

View File

@ -20,7 +20,7 @@
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2011 by Delphix. All rights reserved.
* Copyright (c) 2012 by Delphix. All rights reserved.
*/
#include <sys/zfs_context.h>
@ -2235,13 +2235,22 @@ zio_alloc_zil(spa_t *spa, uint64_t txg, blkptr_t *new_bp, blkptr_t *old_bp,
ASSERT(txg > spa_syncing_txg(spa));
if (use_slog)
/*
* ZIL blocks are always contiguous (i.e. not gang blocks) so we
* set the METASLAB_GANG_AVOID flag so that they don't "fast gang"
* when allocating them.
*/
if (use_slog) {
error = metaslab_alloc(spa, spa_log_class(spa), size,
new_bp, 1, txg, old_bp, METASLAB_HINTBP_AVOID);
new_bp, 1, txg, old_bp,
METASLAB_HINTBP_AVOID | METASLAB_GANG_AVOID);
}
if (error)
if (error) {
error = metaslab_alloc(spa, spa_normal_class(spa), size,
new_bp, 1, txg, old_bp, METASLAB_HINTBP_AVOID);
new_bp, 1, txg, old_bp,
METASLAB_HINTBP_AVOID | METASLAB_GANG_AVOID);
}
if (error == 0) {
BP_SET_LSIZE(new_bp, size);