*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2015 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2017 by Delphix. All rights reserved.
* Copyright 2015 Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
+ * Copyright 2013 Saso Kiselkov. All rights reserved.
+ * Copyright (c) 2017 Datto Inc.
*/
#include <sys/zfs_context.h>
#include <sys/zil.h>
#include <sys/vdev_impl.h>
#include <sys/vdev_file.h>
+#include <sys/vdev_raidz.h>
#include <sys/metaslab.h>
#include <sys/uberblock_impl.h>
#include <sys/txg.h>
#include <sys/ddt.h>
#include <sys/kstat.h>
#include "zfs_prop.h"
-#include "zfeature_common.h"
+#include <sys/zfeature.h>
+#include "qat_compress.h"
/*
* SPA locking
*/
unsigned long zfs_deadman_synctime_ms = 1000000ULL;
+/*
+ * Check time in milliseconds. This defines the frequency at which we check
+ * for hung I/O.
+ */
+unsigned long zfs_deadman_checktime_ms = 5000ULL;
+
/*
* By default the deadman is enabled.
*/
* it is possible to run the pool completely out of space, causing it to
* be permanently read-only.
*
+ * Note that on very small pools, the slop space will be larger than
+ * 3.2%, in an effort to have it be at least spa_min_slop (128MB),
+ * but we never allow it to be more than half the pool size.
+ *
* See also the comments in zfs_space_check_t.
*/
int spa_slop_shift = 5;
+uint64_t spa_min_slop = 128 * 1024 * 1024;
/*
* ==========================================================================
static void
spa_config_lock_init(spa_t *spa)
{
- int i;
-
- for (i = 0; i < SCL_LOCKS; i++) {
+ for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
mutex_init(&scl->scl_lock, NULL, MUTEX_DEFAULT, NULL);
cv_init(&scl->scl_cv, NULL, CV_DEFAULT, NULL);
static void
spa_config_lock_destroy(spa_t *spa)
{
- int i;
-
- for (i = 0; i < SCL_LOCKS; i++) {
+ for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
mutex_destroy(&scl->scl_lock);
cv_destroy(&scl->scl_cv);
int
spa_config_tryenter(spa_t *spa, int locks, void *tag, krw_t rw)
{
- int i;
-
- for (i = 0; i < SCL_LOCKS; i++) {
+ for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
spa_config_enter(spa_t *spa, int locks, void *tag, krw_t rw)
{
int wlocks_held = 0;
- int i;
ASSERT3U(SCL_LOCKS, <, sizeof (wlocks_held) * NBBY);
- for (i = 0; i < SCL_LOCKS; i++) {
+ for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (scl->scl_writer == curthread)
wlocks_held |= (1 << i);
void
spa_config_exit(spa_t *spa, int locks, void *tag)
{
- int i;
-
- for (i = SCL_LOCKS - 1; i >= 0; i--) {
+ for (int i = SCL_LOCKS - 1; i >= 0; i--) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
int
spa_config_held(spa_t *spa, int locks, krw_t rw)
{
- int i, locks_held = 0;
+ int locks_held = 0;
- for (i = 0; i < SCL_LOCKS; i++) {
+ for (int i = 0; i < SCL_LOCKS; i++) {
spa_config_lock_t *scl = &spa->spa_config_lock[i];
if (!(locks & (1 << i)))
continue;
{
spa_t *spa = arg;
+ /* Disable the deadman if the pool is suspended. */
+ if (spa_suspended(spa))
+ return;
+
zfs_dbgmsg("slow spa_sync: started %llu seconds ago, calls %llu",
(gethrtime() - spa->spa_sync_starttime) / NANOSEC,
++spa->spa_deadman_calls);
if (zfs_deadman_enabled)
vdev_deadman(spa->spa_root_vdev);
- spa->spa_deadman_tqid = taskq_dispatch_delay(system_taskq,
+ spa->spa_deadman_tqid = taskq_dispatch_delay(system_delay_taskq,
spa_deadman, spa, TQ_SLEEP, ddi_get_lbolt() +
- NSEC_TO_TICK(spa->spa_deadman_synctime));
+ MSEC_TO_TICK(zfs_deadman_checktime_ms));
}
/*
{
spa_t *spa;
spa_config_dirent_t *dp;
- int t;
- int i;
ASSERT(MUTEX_HELD(&spa_namespace_lock));
mutex_init(&spa->spa_history_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_proc_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_props_lock, NULL, MUTEX_DEFAULT, NULL);
+ mutex_init(&spa->spa_cksum_tmpls_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_scrub_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_suspend_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_vdev_top_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&spa->spa_feat_stats_lock, NULL, MUTEX_DEFAULT, NULL);
+ mutex_init(&spa->spa_alloc_lock, NULL, MUTEX_DEFAULT, NULL);
cv_init(&spa->spa_async_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_evicting_os_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_scrub_io_cv, NULL, CV_DEFAULT, NULL);
cv_init(&spa->spa_suspend_cv, NULL, CV_DEFAULT, NULL);
- for (t = 0; t < TXG_SIZE; t++)
+ for (int t = 0; t < TXG_SIZE; t++)
bplist_create(&spa->spa_free_bplist[t]);
(void) strlcpy(spa->spa_name, name, sizeof (spa->spa_name));
if (altroot)
spa->spa_root = spa_strdup(altroot);
+ avl_create(&spa->spa_alloc_tree, zio_bookmark_compare,
+ sizeof (zio_t), offsetof(zio_t, io_alloc_node));
+
/*
* Every pool starts with the default cachefile
*/
spa->spa_min_ashift = INT_MAX;
spa->spa_max_ashift = 0;
+ /* Reset cached value */
+ spa->spa_dedup_dspace = ~0ULL;
+
/*
* As a pool is being created, treat all features as disabled by
* setting SPA_FEATURE_DISABLED for all entries in the feature
* refcount cache.
*/
- for (i = 0; i < SPA_FEATURES; i++) {
+ for (int i = 0; i < SPA_FEATURES; i++) {
spa->spa_feat_refcount_cache[i] = SPA_FEATURE_DISABLED;
}
spa_remove(spa_t *spa)
{
spa_config_dirent_t *dp;
- int t;
ASSERT(MUTEX_HELD(&spa_namespace_lock));
ASSERT(spa->spa_state == POOL_STATE_UNINITIALIZED);
kmem_free(dp, sizeof (spa_config_dirent_t));
}
+ avl_destroy(&spa->spa_alloc_tree);
list_destroy(&spa->spa_config_list);
nvlist_free(spa->spa_label_features);
spa_stats_destroy(spa);
spa_config_lock_destroy(spa);
- for (t = 0; t < TXG_SIZE; t++)
+ for (int t = 0; t < TXG_SIZE; t++)
bplist_destroy(&spa->spa_free_bplist[t]);
+ zio_checksum_templates_free(spa);
+
cv_destroy(&spa->spa_async_cv);
cv_destroy(&spa->spa_evicting_os_cv);
cv_destroy(&spa->spa_proc_cv);
cv_destroy(&spa->spa_scrub_io_cv);
cv_destroy(&spa->spa_suspend_cv);
+ mutex_destroy(&spa->spa_alloc_lock);
mutex_destroy(&spa->spa_async_lock);
mutex_destroy(&spa->spa_errlist_lock);
mutex_destroy(&spa->spa_errlog_lock);
mutex_destroy(&spa->spa_history_lock);
mutex_destroy(&spa->spa_proc_lock);
mutex_destroy(&spa->spa_props_lock);
+ mutex_destroy(&spa->spa_cksum_tmpls_lock);
mutex_destroy(&spa->spa_scrub_lock);
mutex_destroy(&spa->spa_suspend_lock);
mutex_destroy(&spa->spa_vdev_top_lock);
int aux_count;
} spa_aux_t;
-static int
+static inline int
spa_aux_compare(const void *a, const void *b)
{
- const spa_aux_t *sa = a;
- const spa_aux_t *sb = b;
+ const spa_aux_t *sa = (const spa_aux_t *)a;
+ const spa_aux_t *sb = (const spa_aux_t *)b;
- if (sa->aux_guid < sb->aux_guid)
- return (-1);
- else if (sa->aux_guid > sb->aux_guid)
- return (1);
- else
- return (0);
+ return (AVL_CMP(sa->aux_guid, sb->aux_guid));
}
void
void
spa_vdev_config_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error, char *tag)
{
+ ASSERT(MUTEX_HELD(&spa_namespace_lock));
+
int config_changed = B_FALSE;
- ASSERT(MUTEX_HELD(&spa_namespace_lock));
ASSERT(txg > spa_last_synced_txg(spa));
spa->spa_pending_vdev = NULL;
spa_vdev_state_exit(spa_t *spa, vdev_t *vd, int error)
{
boolean_t config_changed = B_FALSE;
+ vdev_t *vdev_top;
+
+ if (vd == NULL || vd == spa->spa_root_vdev) {
+ vdev_top = spa->spa_root_vdev;
+ } else {
+ vdev_top = vd->vdev_top;
+ }
if (vd != NULL || error == 0)
- vdev_dtl_reassess(vd ? vd->vdev_top : spa->spa_root_vdev,
- 0, 0, B_FALSE);
+ vdev_dtl_reassess(vdev_top, 0, 0, B_FALSE);
if (vd != NULL) {
- vdev_state_dirty(vd->vdev_top);
+ if (vd != spa->spa_root_vdev)
+ vdev_state_dirty(vdev_top);
+
config_changed = B_TRUE;
spa->spa_config_generation++;
}
ASSERT(range != 0);
+ if (range == 1)
+ return (0);
+
(void) random_get_pseudo_bytes((void *)&r, sizeof (uint64_t));
return (r % range);
char type[256];
char *checksum = NULL;
char *compress = NULL;
+ char *crypt_type = NULL;
if (bp != NULL) {
if (BP_GET_TYPE(bp) & DMU_OT_NEWTYPE) {
(void) strlcpy(type, dmu_ot[BP_GET_TYPE(bp)].ot_name,
sizeof (type));
}
+ if (BP_IS_ENCRYPTED(bp)) {
+ crypt_type = "encrypted";
+ } else if (BP_IS_AUTHENTICATED(bp)) {
+ crypt_type = "authenticated";
+ } else if (BP_HAS_INDIRECT_MAC_CKSUM(bp)) {
+ crypt_type = "indirect-MAC";
+ } else {
+ crypt_type = "unencrypted";
+ }
if (!BP_IS_EMBEDDED(bp)) {
checksum =
zio_checksum_table[BP_GET_CHECKSUM(bp)].ci_name;
}
SNPRINTF_BLKPTR(snprintf, ' ', buf, buflen, bp, type, checksum,
- compress);
+ crypt_type, compress);
}
void
* lowercase hexadecimal numbers that don't overflow.
*/
uint64_t
-strtonum(const char *str, char **nptr)
+zfs_strtonum(const char *str, char **nptr)
{
uint64_t val = 0;
char c;
return (spa->spa_syncing_txg);
}
+/*
+ * Return the last txg where data can be dirtied. The final txgs
+ * will be used to just clear out any deferred frees that remain.
+ */
+uint64_t
+spa_final_dirty_txg(spa_t *spa)
+{
+ return (spa->spa_final_txg - TXG_DEFER_SIZE);
+}
+
pool_state_t
spa_state(spa_t *spa)
{
return (spa->spa_freeze_txg);
}
-/* ARGSUSED */
+/*
+ * Return the inflated asize for a logical write in bytes. This is used by the
+ * DMU to calculate the space a logical write will require on disk.
+ * If lsize is smaller than the largest physical block size allocatable on this
+ * pool we use its value instead, since the write will end up using the whole
+ * block anyway.
+ */
uint64_t
-spa_get_asize(spa_t *spa, uint64_t lsize)
+spa_get_worst_case_asize(spa_t *spa, uint64_t lsize)
{
- return (lsize * spa_asize_inflation);
+ if (lsize == 0)
+ return (0); /* No inflation needed */
+ return (MAX(lsize, 1 << spa->spa_max_ashift) * spa_asize_inflation);
}
/*
* Return the amount of slop space in bytes. It is 1/32 of the pool (3.2%),
- * or at least 32MB.
+ * or at least 128MB, unless that would cause it to be more than half the
+ * pool size.
*
* See the comment above spa_slop_shift for details.
*/
uint64_t
-spa_get_slop_space(spa_t *spa) {
+spa_get_slop_space(spa_t *spa)
+{
uint64_t space = spa_get_dspace(spa);
- return (MAX(space >> spa_slop_shift, SPA_MINDEVSIZE >> 1));
+ return (MAX(space >> spa_slop_shift, MIN(space >> 1, spa_min_slop)));
}
uint64_t
bp_get_dsize_sync(spa_t *spa, const blkptr_t *bp)
{
uint64_t dsize = 0;
- int d;
- for (d = 0; d < BP_GET_NDVAS(bp); d++)
+ for (int d = 0; d < BP_GET_NDVAS(bp); d++)
dsize += dva_get_dsize_sync(spa, &bp->blk_dva[d]);
return (dsize);
bp_get_dsize(spa_t *spa, const blkptr_t *bp)
{
uint64_t dsize = 0;
- int d;
spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER);
- for (d = 0; d < BP_GET_NDVAS(bp); d++)
+ for (int d = 0; d < BP_GET_NDVAS(bp); d++)
dsize += dva_get_dsize_sync(spa, &bp->blk_dva[d]);
spa_config_exit(spa, SCL_VDEV, FTAG);
int s;
s = strcmp(s1->spa_name, s2->spa_name);
- if (s > 0)
- return (1);
- if (s < 0)
- return (-1);
- return (0);
+
+ return (AVL_ISIGN(s));
}
void
refcount_init();
unique_init();
range_tree_init();
+ metaslab_alloc_trace_init();
ddt_init();
zio_init();
dmu_init();
zil_init();
vdev_cache_stat_init();
+ vdev_mirror_stat_init();
+ vdev_raidz_math_init();
+ vdev_file_init();
zfs_prop_init();
zpool_prop_init();
zpool_feature_init();
spa_config_load();
l2arc_start();
+ qat_init();
}
void
spa_evict_all();
+ vdev_file_fini();
vdev_cache_stat_fini();
+ vdev_mirror_stat_fini();
+ vdev_raidz_math_fini();
zil_fini();
dmu_fini();
zio_fini();
ddt_fini();
+ metaslab_alloc_trace_fini();
range_tree_fini();
unique_fini();
refcount_fini();
fm_fini();
+ qat_fini();
avl_destroy(&spa_namespace_avl);
avl_destroy(&spa_spare_avl);
{
/* data not stored on disk */
spa->spa_scan_pass_start = gethrestime_sec();
+ if (dsl_scan_is_paused_scrub(spa->spa_dsl_pool->dp_scan))
+ spa->spa_scan_pass_scrub_pause = spa->spa_scan_pass_start;
+ else
+ spa->spa_scan_pass_scrub_pause = 0;
+ spa->spa_scan_pass_scrub_spent_paused = 0;
spa->spa_scan_pass_exam = 0;
vdev_scan_stat_init(spa->spa_root_vdev);
}
/* data not stored on disk */
ps->pss_pass_start = spa->spa_scan_pass_start;
ps->pss_pass_exam = spa->spa_scan_pass_exam;
+ ps->pss_pass_scrub_pause = spa->spa_scan_pass_scrub_pause;
+ ps->pss_pass_scrub_spent_paused = spa->spa_scan_pass_scrub_spent_paused;
return (0);
}
return (SPA_OLD_MAXBLOCKSIZE);
}
+int
+spa_maxdnodesize(spa_t *spa)
+{
+ if (spa_feature_is_enabled(spa, SPA_FEATURE_LARGE_DNODE))
+ return (DNODE_MAX_SIZE);
+ else
+ return (DNODE_MIN_SIZE);
+}
+
+boolean_t
+spa_multihost(spa_t *spa)
+{
+ return (spa->spa_multihost ? B_TRUE : B_FALSE);
+}
+
+unsigned long
+spa_get_hostid(void)
+{
+ unsigned long myhostid;
+
+#ifdef _KERNEL
+ myhostid = zone_get_hostid(NULL);
+#else /* _KERNEL */
+ /*
+ * We're emulating the system's hostid in userland, so
+ * we can't use zone_get_hostid().
+ */
+ (void) ddi_strtoul(hw_serial, NULL, 10, &myhostid);
+#endif /* _KERNEL */
+
+ return (myhostid);
+}
+
#if defined(_KERNEL) && defined(HAVE_SPL)
/* Namespace manipulation */
EXPORT_SYMBOL(spa_lookup);
EXPORT_SYMBOL(spa_state);
EXPORT_SYMBOL(spa_load_state);
EXPORT_SYMBOL(spa_freeze_txg);
-EXPORT_SYMBOL(spa_get_asize);
EXPORT_SYMBOL(spa_get_dspace);
EXPORT_SYMBOL(spa_update_dspace);
EXPORT_SYMBOL(spa_deflate);
EXPORT_SYMBOL(spa_delegation);
EXPORT_SYMBOL(spa_meta_objset);
EXPORT_SYMBOL(spa_maxblocksize);
+EXPORT_SYMBOL(spa_maxdnodesize);
/* Miscellaneous support routines */
EXPORT_SYMBOL(spa_rename);
EXPORT_SYMBOL(spa_is_root);
EXPORT_SYMBOL(spa_writeable);
EXPORT_SYMBOL(spa_mode);
-
EXPORT_SYMBOL(spa_namespace_lock);
+/* BEGIN CSTYLED */
module_param(zfs_flags, uint, 0644);
MODULE_PARM_DESC(zfs_flags, "Set additional debugging flags");
module_param(zfs_deadman_synctime_ms, ulong, 0644);
MODULE_PARM_DESC(zfs_deadman_synctime_ms, "Expiration time in milliseconds");
+module_param(zfs_deadman_checktime_ms, ulong, 0644);
+MODULE_PARM_DESC(zfs_deadman_checktime_ms,
+ "Dead I/O check interval in milliseconds");
+
module_param(zfs_deadman_enabled, int, 0644);
MODULE_PARM_DESC(zfs_deadman_enabled, "Enable deadman timer");
module_param(spa_slop_shift, int, 0644);
MODULE_PARM_DESC(spa_slop_shift, "Reserved free space in pool");
+/* END CSTYLED */
#endif
projects / mirror_zfs.git / blobdiff