/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2014 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2017 by Delphix. All rights reserved.
+ * Copyright (c) 2014 Integros [integros.com]
+ * Copyright 2016 Nexenta Systems, Inc.
+ * Copyright (c) 2017 Lawrence Livermore National Security, LLC.
+ * Copyright (c) 2015, 2017, Intel Corporation.
*/
#include <stdio.h>
#include <sys/arc.h>
#include <sys/ddt.h>
#include <sys/zfeature.h>
+#include <sys/abd.h>
+#include <sys/blkptr.h>
+#include <sys/dsl_crypt.h>
#include <zfs_comutil.h>
-#undef ZFS_MAXNAMELEN
#include <libzfs.h>
+#include "zdb.h"
+
#define ZDB_COMPRESS_NAME(idx) ((idx) < ZIO_COMPRESS_FUNCTIONS ? \
zio_compress_table[(idx)].ci_name : "UNKNOWN")
#define ZDB_CHECKSUM_NAME(idx) ((idx) < ZIO_CHECKSUM_FUNCTIONS ? \
zio_checksum_table[(idx)].ci_name : "UNKNOWN")
-#define ZDB_OT_NAME(idx) ((idx) < DMU_OT_NUMTYPES ? \
- dmu_ot[(idx)].ot_name : DMU_OT_IS_VALID(idx) ? \
- dmu_ot_byteswap[DMU_OT_BYTESWAP(idx)].ob_name : "UNKNOWN")
#define ZDB_OT_TYPE(idx) ((idx) < DMU_OT_NUMTYPES ? (idx) : \
- (((idx) == DMU_OTN_ZAP_DATA || (idx) == DMU_OTN_ZAP_METADATA) ? \
- DMU_OT_ZAP_OTHER : DMU_OT_NUMTYPES))
+ (idx) == DMU_OTN_ZAP_DATA || (idx) == DMU_OTN_ZAP_METADATA ? \
+ DMU_OT_ZAP_OTHER : \
+ (idx) == DMU_OTN_UINT64_DATA || (idx) == DMU_OTN_UINT64_METADATA ? \
+ DMU_OT_UINT64_OTHER : DMU_OT_NUMTYPES)
+
+static char *
+zdb_ot_name(dmu_object_type_t type)
+{
+ if (type < DMU_OT_NUMTYPES)
+ return (dmu_ot[type].ot_name);
+ else if ((type & DMU_OT_NEWTYPE) &&
+ ((type & DMU_OT_BYTESWAP_MASK) < DMU_BSWAP_NUMFUNCS))
+ return (dmu_ot_byteswap[type & DMU_OT_BYTESWAP_MASK].ob_name);
+ else
+ return ("UNKNOWN");
+}
-#ifndef lint
+extern int reference_tracking_enable;
extern int zfs_recover;
extern uint64_t zfs_arc_max, zfs_arc_meta_limit;
-#else
-int zfs_recover;
-uint64_t zfs_arc_max, zfs_arc_meta_limit;
-#endif
+extern int zfs_vdev_async_read_max_active;
-const char cmdname[] = "zdb";
+static const char cmdname[] = "zdb";
uint8_t dump_opt[256];
typedef void object_viewer_t(objset_t *, uint64_t, void *data, size_t size);
-extern void dump_intent_log(zilog_t *);
uint64_t *zopt_object = NULL;
-int zopt_objects = 0;
+static unsigned zopt_objects = 0;
libzfs_handle_t *g_zfs;
uint64_t max_inflight = 1000;
usage(void)
{
(void) fprintf(stderr,
- "Usage: %s [-CumMdibcsDvhLXFPA] [-t txg] [-e [-p path...]] "
- "[-U config] [-I inflight I/Os] poolname [object...]\n"
- " %s [-divPA] [-e -p path...] [-U config] dataset "
- "[object...]\n"
- " %s -mM [-LXFPA] [-t txg] [-e [-p path...]] [-U config] "
- "poolname [vdev [metaslab...]]\n"
- " %s -R [-A] [-e [-p path...]] poolname "
- "vdev:offset:size[:flags]\n"
- " %s -S [-PA] [-e [-p path...]] [-U config] poolname\n"
- " %s -l [-uA] device\n"
- " %s -C [-A] [-U config]\n\n",
- cmdname, cmdname, cmdname, cmdname, cmdname, cmdname, cmdname);
+ "Usage:\t%s [-AbcdDFGhiLMPsvX] [-e [-V] [-p <path> ...]] "
+ "[-I <inflight I/Os>]\n"
+ "\t\t[-o <var>=<value>]... [-t <txg>] [-U <cache>] [-x <dumpdir>]\n"
+ "\t\t[<poolname> [<object> ...]]\n"
+ "\t%s [-AdiPv] [-e [-V] [-p <path> ...]] [-U <cache>] <dataset>\n"
+ "\t\t[<object> ...]\n"
+ "\t%s -C [-A] [-U <cache>]\n"
+ "\t%s -l [-Aqu] <device>\n"
+ "\t%s -m [-AFLPX] [-e [-V] [-p <path> ...]] [-t <txg>] "
+ "[-U <cache>]\n\t\t<poolname> [<vdev> [<metaslab> ...]]\n"
+ "\t%s -O <dataset> <path>\n"
+ "\t%s -R [-A] [-e [-V] [-p <path> ...]] [-U <cache>]\n"
+ "\t\t<poolname> <vdev>:<offset>:<size>[:<flags>]\n"
+ "\t%s -E [-A] word0:word1:...:word15\n"
+ "\t%s -S [-AP] [-e [-V] [-p <path> ...]] [-U <cache>] "
+ "<poolname>\n\n",
+ cmdname, cmdname, cmdname, cmdname, cmdname, cmdname, cmdname,
+ cmdname, cmdname);
(void) fprintf(stderr, " Dataset name must include at least one "
"separator character '/' or '@'\n");
(void) fprintf(stderr, " If object numbers are specified, only "
"those objects are dumped\n\n");
(void) fprintf(stderr, " Options to control amount of output:\n");
- (void) fprintf(stderr, " -u uberblock\n");
- (void) fprintf(stderr, " -d dataset(s)\n");
- (void) fprintf(stderr, " -i intent logs\n");
- (void) fprintf(stderr, " -C config (or cachefile if alone)\n");
- (void) fprintf(stderr, " -h pool history\n");
(void) fprintf(stderr, " -b block statistics\n");
- (void) fprintf(stderr, " -m metaslabs\n");
- (void) fprintf(stderr, " -M metaslab groups\n");
(void) fprintf(stderr, " -c checksum all metadata (twice for "
"all data) blocks\n");
- (void) fprintf(stderr, " -s report stats on zdb's I/O\n");
+ (void) fprintf(stderr, " -C config (or cachefile if alone)\n");
+ (void) fprintf(stderr, " -d dataset(s)\n");
(void) fprintf(stderr, " -D dedup statistics\n");
- (void) fprintf(stderr, " -S simulate dedup to measure effect\n");
- (void) fprintf(stderr, " -v verbose (applies to all others)\n");
- (void) fprintf(stderr, " -l dump label contents\n");
+ (void) fprintf(stderr, " -E decode and display block from an "
+ "embedded block pointer\n");
+ (void) fprintf(stderr, " -h pool history\n");
+ (void) fprintf(stderr, " -i intent logs\n");
+ (void) fprintf(stderr, " -l read label contents\n");
(void) fprintf(stderr, " -L disable leak tracking (do not "
"load spacemaps)\n");
+ (void) fprintf(stderr, " -m metaslabs\n");
+ (void) fprintf(stderr, " -M metaslab groups\n");
+ (void) fprintf(stderr, " -O perform object lookups by path\n");
(void) fprintf(stderr, " -R read and display block from a "
- "device\n\n");
+ "device\n");
+ (void) fprintf(stderr, " -s report stats on zdb's I/O\n");
+ (void) fprintf(stderr, " -S simulate dedup to measure effect\n");
+ (void) fprintf(stderr, " -v verbose (applies to all "
+ "others)\n\n");
(void) fprintf(stderr, " Below options are intended for use "
- "with other options (except -l):\n");
+ "with other options:\n");
(void) fprintf(stderr, " -A ignore assertions (-A), enable "
"panic recovery (-AA) or both (-AAA)\n");
- (void) fprintf(stderr, " -F attempt automatic rewind within "
- "safe range of transaction groups\n");
- (void) fprintf(stderr, " -U <cachefile_path> -- use alternate "
- "cachefile\n");
- (void) fprintf(stderr, " -X attempt extreme rewind (does not "
- "work with dataset)\n");
(void) fprintf(stderr, " -e pool is exported/destroyed/"
"has altroot/not in a cachefile\n");
+ (void) fprintf(stderr, " -F attempt automatic rewind within "
+ "safe range of transaction groups\n");
+ (void) fprintf(stderr, " -G dump zfs_dbgmsg buffer before "
+ "exiting\n");
+ (void) fprintf(stderr, " -I <number of inflight I/Os> -- "
+ "specify the maximum number of\n "
+ "checksumming I/Os [default is 200]\n");
+ (void) fprintf(stderr, " -o <variable>=<value> set global "
+ "variable to an unsigned 32-bit integer\n");
(void) fprintf(stderr, " -p <path> -- use one or more with "
"-e to specify path to vdev dir\n");
(void) fprintf(stderr, " -P print numbers in parseable form\n");
+ (void) fprintf(stderr, " -q don't print label contents\n");
(void) fprintf(stderr, " -t <txg> -- highest txg to use when "
"searching for uberblocks\n");
- (void) fprintf(stderr, " -I <number of inflight I/Os> -- "
- "specify the maximum number of checksumming I/Os "
- "[default is 200]\n");
+ (void) fprintf(stderr, " -u uberblock\n");
+ (void) fprintf(stderr, " -U <cachefile_path> -- use alternate "
+ "cachefile\n");
+ (void) fprintf(stderr, " -V do verbatim import\n");
+ (void) fprintf(stderr, " -x <dumpdir> -- "
+ "dump all read blocks into specified directory\n");
+ (void) fprintf(stderr, " -X attempt extreme rewind (does not "
+ "work with dataset)\n");
(void) fprintf(stderr, "Specify an option more than once (e.g. -bb) "
"to make only that option verbose\n");
(void) fprintf(stderr, "Default is to dump everything non-verbosely\n");
exit(1);
}
+static void
+dump_debug_buffer(void)
+{
+ if (dump_opt['G']) {
+ (void) printf("\n");
+ zfs_dbgmsg_print("zdb");
+ }
+}
+
/*
* Called for usage errors that are discovered after a call to spa_open(),
* dmu_bonus_hold(), or pool_match(). abort() is called for other errors.
va_end(ap);
(void) fprintf(stderr, "\n");
+ dump_debug_buffer();
+
exit(1);
}
}
static void
-zdb_nicenum(uint64_t num, char *buf)
+zdb_nicenum(uint64_t num, char *buf, size_t buflen)
{
if (dump_opt['P'])
- (void) sprintf(buf, "%llu", (longlong_t)num);
+ (void) snprintf(buf, buflen, "%llu", (longlong_t)num);
else
- nicenum(num, buf);
+ nicenum(num, buf, sizeof (buf));
}
-const char histo_stars[] = "****************************************";
-const int histo_width = sizeof (histo_stars) - 1;
+static const char histo_stars[] = "****************************************";
+static const uint64_t histo_width = sizeof (histo_stars) - 1;
static void
dump_histogram(const uint64_t *histo, int size, int offset)
}
/*ARGSUSED*/
-void
+static void
dump_uint8(objset_t *os, uint64_t object, void *data, size_t size)
{
}
zap_cursor_t zc;
zap_attribute_t attr;
void *prop;
- int i;
+ unsigned i;
dump_zap_stats(os, object);
(void) printf("\n");
uint64_t i;
char bytes[32], comp[32], uncomp[32];
+ /* make sure the output won't get truncated */
+ CTASSERT(sizeof (bytes) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (comp) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (uncomp) >= NN_NUMBUF_SZ);
+
if (bpop == NULL)
return;
- zdb_nicenum(bpop->bpo_bytes, bytes);
- zdb_nicenum(bpop->bpo_comp, comp);
- zdb_nicenum(bpop->bpo_uncomp, uncomp);
+ zdb_nicenum(bpop->bpo_bytes, bytes, sizeof (bytes));
+ zdb_nicenum(bpop->bpo_comp, comp, sizeof (comp));
+ zdb_nicenum(bpop->bpo_uncomp, uncomp, sizeof (uncomp));
(void) printf("\t\tnum_blkptrs = %llu\n",
(u_longlong_t)bpop->bpo_num_blkptrs);
dump_bpobj_subobjs(objset_t *os, uint64_t object, void *data, size_t size)
{
dmu_object_info_t doi;
- uint64_t i;
+ int64_t i;
VERIFY0(dmu_object_info(os, object, &doi));
uint64_t *subobjs = kmem_alloc(doi.doi_max_offset, KM_SLEEP);
}
for (i = 0; i <= last_nonzero; i++) {
- (void) printf("\t%llu\n", (longlong_t)subobjs[i]);
+ (void) printf("\t%llu\n", (u_longlong_t)subobjs[i]);
}
kmem_free(subobjs, doi.doi_max_offset);
}
zap_cursor_t zc;
zap_attribute_t attr;
uint16_t *layout_attrs;
- int i;
+ unsigned i;
dump_zap_stats(os, object);
(void) printf("\n");
zap_cursor_fini(&zc);
}
-int
+static int
get_dtl_refcount(vdev_t *vd)
{
int refcount = 0;
- int c;
if (vd->vdev_ops->vdev_op_leaf) {
space_map_t *sm = vd->vdev_dtl_sm;
return (0);
}
- for (c = 0; c < vd->vdev_children; c++)
+ for (unsigned c = 0; c < vd->vdev_children; c++)
refcount += get_dtl_refcount(vd->vdev_child[c]);
return (refcount);
}
-int
+static int
get_metaslab_refcount(vdev_t *vd)
{
int refcount = 0;
- int c, m;
- if (vd->vdev_top == vd && !vd->vdev_removing) {
- for (m = 0; m < vd->vdev_ms_count; m++) {
+ if (vd->vdev_top == vd) {
+ for (uint64_t m = 0; m < vd->vdev_ms_count; m++) {
space_map_t *sm = vd->vdev_ms[m]->ms_sm;
if (sm != NULL &&
refcount++;
}
}
- for (c = 0; c < vd->vdev_children; c++)
+ for (unsigned c = 0; c < vd->vdev_children; c++)
refcount += get_metaslab_refcount(vd->vdev_child[c]);
return (refcount);
}
+static int
+get_obsolete_refcount(vdev_t *vd)
+{
+ int refcount = 0;
+
+ uint64_t obsolete_sm_obj = vdev_obsolete_sm_object(vd);
+ if (vd->vdev_top == vd && obsolete_sm_obj != 0) {
+ dmu_object_info_t doi;
+ VERIFY0(dmu_object_info(vd->vdev_spa->spa_meta_objset,
+ obsolete_sm_obj, &doi));
+ if (doi.doi_bonus_size == sizeof (space_map_phys_t)) {
+ refcount++;
+ }
+ } else {
+ ASSERT3P(vd->vdev_obsolete_sm, ==, NULL);
+ ASSERT3U(obsolete_sm_obj, ==, 0);
+ }
+ for (unsigned c = 0; c < vd->vdev_children; c++) {
+ refcount += get_obsolete_refcount(vd->vdev_child[c]);
+ }
+
+ return (refcount);
+}
+
+static int
+get_prev_obsolete_spacemap_refcount(spa_t *spa)
+{
+ uint64_t prev_obj =
+ spa->spa_condensing_indirect_phys.scip_prev_obsolete_sm_object;
+ if (prev_obj != 0) {
+ dmu_object_info_t doi;
+ VERIFY0(dmu_object_info(spa->spa_meta_objset, prev_obj, &doi));
+ if (doi.doi_bonus_size == sizeof (space_map_phys_t)) {
+ return (1);
+ }
+ }
+ return (0);
+}
+
static int
verify_spacemap_refcounts(spa_t *spa)
{
&expected_refcount);
actual_refcount = get_dtl_refcount(spa->spa_root_vdev);
actual_refcount += get_metaslab_refcount(spa->spa_root_vdev);
+ actual_refcount += get_obsolete_refcount(spa->spa_root_vdev);
+ actual_refcount += get_prev_obsolete_spacemap_refcount(spa);
if (expected_refcount != actual_refcount) {
(void) printf("space map refcount mismatch: expected %lld != "
dump_spacemap(objset_t *os, space_map_t *sm)
{
uint64_t alloc, offset, entry;
- char *ddata[] = { "ALLOC", "FREE", "CONDENSE", "INVALID",
- "INVALID", "INVALID", "INVALID", "INVALID" };
+ const char *ddata[] = { "ALLOC", "FREE", "CONDENSE", "INVALID",
+ "INVALID", "INVALID", "INVALID", "INVALID" };
if (sm == NULL)
return;
+ (void) printf("space map object %llu:\n",
+ (longlong_t)sm->sm_phys->smp_object);
+ (void) printf(" smp_objsize = 0x%llx\n",
+ (longlong_t)sm->sm_phys->smp_objsize);
+ (void) printf(" smp_alloc = 0x%llx\n",
+ (longlong_t)sm->sm_phys->smp_alloc);
+
/*
* Print out the freelist entries in both encoded and decoded form.
*/
avl_tree_t *t = &msp->ms_size_tree;
int free_pct = range_tree_space(rt) * 100 / msp->ms_size;
- zdb_nicenum(metaslab_block_maxsize(msp), maxbuf);
+ /* max sure nicenum has enough space */
+ CTASSERT(sizeof (maxbuf) >= NN_NUMBUF_SZ);
+
+ zdb_nicenum(metaslab_block_maxsize(msp), maxbuf, sizeof (maxbuf));
(void) printf("\t %25s %10lu %7s %6s %4s %4d%%\n",
"segments", avl_numnodes(t), "maxsize", maxbuf,
space_map_t *sm = msp->ms_sm;
char freebuf[32];
- zdb_nicenum(msp->ms_size - space_map_allocated(sm), freebuf);
+ zdb_nicenum(msp->ms_size - space_map_allocated(sm), freebuf,
+ sizeof (freebuf));
(void) printf(
"\tmetaslab %6llu offset %12llx spacemap %6llu free %5s\n",
if (dump_opt['d'] > 5 || dump_opt['m'] > 3) {
ASSERT(msp->ms_size == (1ULL << vd->vdev_ms_shift));
- mutex_enter(&msp->ms_lock);
dump_spacemap(spa->spa_meta_objset, msp->ms_sm);
- mutex_exit(&msp->ms_lock);
}
}
vdev_t *rvd = spa->spa_root_vdev;
metaslab_class_t *mc = spa_normal_class(spa);
uint64_t fragmentation;
- int c;
metaslab_class_histogram_verify(mc);
- for (c = 0; c < rvd->vdev_children; c++) {
+ for (unsigned c = 0; c < rvd->vdev_children; c++) {
vdev_t *tvd = rvd->vdev_child[c];
metaslab_group_t *mg = tvd->vdev_mg;
dump_histogram(mc->mc_histogram, RANGE_TREE_HISTOGRAM_SIZE, 0);
}
+static void
+print_vdev_indirect(vdev_t *vd)
+{
+ vdev_indirect_config_t *vic = &vd->vdev_indirect_config;
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+ vdev_indirect_births_t *vib = vd->vdev_indirect_births;
+
+ if (vim == NULL) {
+ ASSERT3P(vib, ==, NULL);
+ return;
+ }
+
+ ASSERT3U(vdev_indirect_mapping_object(vim), ==,
+ vic->vic_mapping_object);
+ ASSERT3U(vdev_indirect_births_object(vib), ==,
+ vic->vic_births_object);
+
+ (void) printf("indirect births obj %llu:\n",
+ (longlong_t)vic->vic_births_object);
+ (void) printf(" vib_count = %llu\n",
+ (longlong_t)vdev_indirect_births_count(vib));
+ for (uint64_t i = 0; i < vdev_indirect_births_count(vib); i++) {
+ vdev_indirect_birth_entry_phys_t *cur_vibe =
+ &vib->vib_entries[i];
+ (void) printf("\toffset %llx -> txg %llu\n",
+ (longlong_t)cur_vibe->vibe_offset,
+ (longlong_t)cur_vibe->vibe_phys_birth_txg);
+ }
+ (void) printf("\n");
+
+ (void) printf("indirect mapping obj %llu:\n",
+ (longlong_t)vic->vic_mapping_object);
+ (void) printf(" vim_max_offset = 0x%llx\n",
+ (longlong_t)vdev_indirect_mapping_max_offset(vim));
+ (void) printf(" vim_bytes_mapped = 0x%llx\n",
+ (longlong_t)vdev_indirect_mapping_bytes_mapped(vim));
+ (void) printf(" vim_count = %llu\n",
+ (longlong_t)vdev_indirect_mapping_num_entries(vim));
+
+ if (dump_opt['d'] <= 5 && dump_opt['m'] <= 3)
+ return;
+
+ uint32_t *counts = vdev_indirect_mapping_load_obsolete_counts(vim);
+
+ for (uint64_t i = 0; i < vdev_indirect_mapping_num_entries(vim); i++) {
+ vdev_indirect_mapping_entry_phys_t *vimep =
+ &vim->vim_entries[i];
+ (void) printf("\t<%llx:%llx:%llx> -> "
+ "<%llx:%llx:%llx> (%x obsolete)\n",
+ (longlong_t)vd->vdev_id,
+ (longlong_t)DVA_MAPPING_GET_SRC_OFFSET(vimep),
+ (longlong_t)DVA_GET_ASIZE(&vimep->vimep_dst),
+ (longlong_t)DVA_GET_VDEV(&vimep->vimep_dst),
+ (longlong_t)DVA_GET_OFFSET(&vimep->vimep_dst),
+ (longlong_t)DVA_GET_ASIZE(&vimep->vimep_dst),
+ counts[i]);
+ }
+ (void) printf("\n");
+
+ uint64_t obsolete_sm_object = vdev_obsolete_sm_object(vd);
+ if (obsolete_sm_object != 0) {
+ objset_t *mos = vd->vdev_spa->spa_meta_objset;
+ (void) printf("obsolete space map object %llu:\n",
+ (u_longlong_t)obsolete_sm_object);
+ ASSERT(vd->vdev_obsolete_sm != NULL);
+ ASSERT3U(space_map_object(vd->vdev_obsolete_sm), ==,
+ obsolete_sm_object);
+ dump_spacemap(mos, vd->vdev_obsolete_sm);
+ (void) printf("\n");
+ }
+}
+
static void
dump_metaslabs(spa_t *spa)
{
vd = rvd->vdev_child[c];
print_vdev_metaslab_header(vd);
+ print_vdev_indirect(vd);
+
for (m = 0; m < vd->vdev_ms_count; m++)
dump_metaslab(vd->vdev_ms[m]);
(void) printf("\n");
{
const ddt_phys_t *ddp = dde->dde_phys;
const ddt_key_t *ddk = &dde->dde_key;
- char *types[4] = { "ditto", "single", "double", "triple" };
+ const char *types[4] = { "ditto", "single", "double", "triple" };
char blkbuf[BP_SPRINTF_LEN];
blkptr_t blk;
int p;
{
ddt_histogram_t ddh_total;
ddt_stat_t dds_total;
- enum zio_checksum c;
- enum ddt_type type;
- enum ddt_class class;
- bzero(&ddh_total, sizeof (ddt_histogram_t));
- bzero(&dds_total, sizeof (ddt_stat_t));
+ bzero(&ddh_total, sizeof (ddh_total));
+ bzero(&dds_total, sizeof (dds_total));
- for (c = 0; c < ZIO_CHECKSUM_FUNCTIONS; c++) {
+ for (enum zio_checksum c = 0; c < ZIO_CHECKSUM_FUNCTIONS; c++) {
ddt_t *ddt = spa->spa_ddt[c];
- for (type = 0; type < DDT_TYPES; type++) {
- for (class = 0; class < DDT_CLASSES;
+ for (enum ddt_type type = 0; type < DDT_TYPES; type++) {
+ for (enum ddt_class class = 0; class < DDT_CLASSES;
class++) {
dump_ddt(ddt, type, class);
}
{
spa_t *spa = vd->vdev_spa;
boolean_t required;
- char *name[DTL_TYPES] = { "missing", "partial", "scrub", "outage" };
+ const char *name[DTL_TYPES] = { "missing", "partial", "scrub",
+ "outage" };
char prefix[256];
- int c, t;
spa_vdev_state_enter(spa, SCL_NONE);
required = vdev_dtl_required(vd);
vd->vdev_parent ? vd->vdev_ops->vdev_op_type : spa_name(spa),
required ? "DTL-required" : "DTL-expendable");
- for (t = 0; t < DTL_TYPES; t++) {
+ for (int t = 0; t < DTL_TYPES; t++) {
range_tree_t *rt = vd->vdev_dtl[t];
if (range_tree_space(rt) == 0)
continue;
(void) snprintf(prefix, sizeof (prefix), "\t%*s%s",
indent + 2, "", name[t]);
- mutex_enter(rt->rt_lock);
range_tree_walk(rt, dump_dtl_seg, prefix);
- mutex_exit(rt->rt_lock);
if (dump_opt['d'] > 5 && vd->vdev_children == 0)
dump_spacemap(spa->spa_meta_objset,
vd->vdev_dtl_sm);
}
- for (c = 0; c < vd->vdev_children; c++)
+ for (unsigned c = 0; c < vd->vdev_children; c++)
dump_dtl(vd->vdev_child[c], indent + 4);
}
dump_history(spa_t *spa)
{
nvlist_t **events = NULL;
- char buf[SPA_MAXBLOCKSIZE];
+ char *buf;
uint64_t resid, len, off = 0;
uint_t num = 0;
int error;
struct tm t;
char tbuf[30];
char internalstr[MAXPATHLEN];
- int i;
+
+ if ((buf = malloc(SPA_OLD_MAXBLOCKSIZE)) == NULL) {
+ (void) fprintf(stderr, "%s: unable to allocate I/O buffer\n",
+ __func__);
+ return;
+ }
do {
- len = sizeof (buf);
+ len = SPA_OLD_MAXBLOCKSIZE;
if ((error = spa_history_get(spa, &off, &len, buf)) != 0) {
(void) fprintf(stderr, "Unable to read history: "
"error %d\n", error);
+ free(buf);
return;
}
} while (len != 0);
(void) printf("\nHistory:\n");
- for (i = 0; i < num; i++) {
+ for (unsigned i = 0; i < num; i++) {
uint64_t time, txg, ievent;
char *cmd, *intstr;
boolean_t printed = B_FALSE;
dump_nvlist(events[i], 2);
}
}
+ free(buf);
}
/*ARGSUSED*/
print_indirect(bp, zb, dnp);
if (BP_GET_LEVEL(bp) > 0 && !BP_IS_HOLE(bp)) {
- uint32_t flags = ARC_WAIT;
+ arc_flags_t flags = ARC_FLAG_WAIT;
int i;
blkptr_t *cbp;
int epb = BP_GET_LSIZE(bp) >> SPA_BLKPTRSHIFT;
}
if (!err)
ASSERT3U(fill, ==, BP_GET_FILL(bp));
- (void) arc_buf_remove_ref(buf, &buf);
+ arc_buf_destroy(buf, &buf);
}
return (err);
time_t crtime;
char nice[32];
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (nice) >= NN_NUMBUF_SZ);
+
if (dd == NULL)
return;
(u_longlong_t)dd->dd_origin_obj);
(void) printf("\t\tchild_dir_zapobj = %llu\n",
(u_longlong_t)dd->dd_child_dir_zapobj);
- zdb_nicenum(dd->dd_used_bytes, nice);
+ zdb_nicenum(dd->dd_used_bytes, nice, sizeof (nice));
(void) printf("\t\tused_bytes = %s\n", nice);
- zdb_nicenum(dd->dd_compressed_bytes, nice);
+ zdb_nicenum(dd->dd_compressed_bytes, nice, sizeof (nice));
(void) printf("\t\tcompressed_bytes = %s\n", nice);
- zdb_nicenum(dd->dd_uncompressed_bytes, nice);
+ zdb_nicenum(dd->dd_uncompressed_bytes, nice, sizeof (nice));
(void) printf("\t\tuncompressed_bytes = %s\n", nice);
- zdb_nicenum(dd->dd_quota, nice);
+ zdb_nicenum(dd->dd_quota, nice, sizeof (nice));
(void) printf("\t\tquota = %s\n", nice);
- zdb_nicenum(dd->dd_reserved, nice);
+ zdb_nicenum(dd->dd_reserved, nice, sizeof (nice));
(void) printf("\t\treserved = %s\n", nice);
(void) printf("\t\tprops_zapobj = %llu\n",
(u_longlong_t)dd->dd_props_zapobj);
(u_longlong_t)dd->dd_flags);
#define DO(which) \
- zdb_nicenum(dd->dd_used_breakdown[DD_USED_ ## which], nice); \
+ zdb_nicenum(dd->dd_used_breakdown[DD_USED_ ## which], nice, \
+ sizeof (nice)); \
(void) printf("\t\tused_breakdown[" #which "] = %s\n", nice)
DO(HEAD);
DO(SNAP);
char used[32], compressed[32], uncompressed[32], unique[32];
char blkbuf[BP_SPRINTF_LEN];
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (used) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (compressed) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (uncompressed) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (unique) >= NN_NUMBUF_SZ);
+
if (ds == NULL)
return;
ASSERT(size == sizeof (*ds));
crtime = ds->ds_creation_time;
- zdb_nicenum(ds->ds_referenced_bytes, used);
- zdb_nicenum(ds->ds_compressed_bytes, compressed);
- zdb_nicenum(ds->ds_uncompressed_bytes, uncompressed);
- zdb_nicenum(ds->ds_unique_bytes, unique);
+ zdb_nicenum(ds->ds_referenced_bytes, used, sizeof (used));
+ zdb_nicenum(ds->ds_compressed_bytes, compressed, sizeof (compressed));
+ zdb_nicenum(ds->ds_uncompressed_bytes, uncompressed,
+ sizeof (uncompressed));
+ zdb_nicenum(ds->ds_unique_bytes, unique, sizeof (unique));
snprintf_blkptr(blkbuf, sizeof (blkbuf), &ds->ds_bp);
(void) printf("\t\tdir_obj = %llu\n",
}
static void
-dump_bptree(objset_t *os, uint64_t obj, char *name)
+dump_bptree(objset_t *os, uint64_t obj, const char *name)
{
char bytes[32];
bptree_phys_t *bt;
dmu_buf_t *db;
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (bytes) >= NN_NUMBUF_SZ);
+
if (dump_opt['d'] < 3)
return;
VERIFY3U(0, ==, dmu_bonus_hold(os, obj, FTAG, &db));
bt = db->db_data;
- zdb_nicenum(bt->bt_bytes, bytes);
+ zdb_nicenum(bt->bt_bytes, bytes, sizeof (bytes));
(void) printf("\n %s: %llu datasets, %s\n",
name, (unsigned long long)(bt->bt_end - bt->bt_begin), bytes);
dmu_buf_rele(db, FTAG);
}
static void
-dump_full_bpobj(bpobj_t *bpo, char *name, int indent)
+dump_full_bpobj(bpobj_t *bpo, const char *name, int indent)
{
char bytes[32];
char comp[32];
char uncomp[32];
uint64_t i;
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (bytes) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (comp) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (uncomp) >= NN_NUMBUF_SZ);
+
if (dump_opt['d'] < 3)
return;
- zdb_nicenum(bpo->bpo_phys->bpo_bytes, bytes);
+ zdb_nicenum(bpo->bpo_phys->bpo_bytes, bytes, sizeof (bytes));
if (bpo->bpo_havesubobj && bpo->bpo_phys->bpo_subobjs != 0) {
- zdb_nicenum(bpo->bpo_phys->bpo_comp, comp);
- zdb_nicenum(bpo->bpo_phys->bpo_uncomp, uncomp);
+ zdb_nicenum(bpo->bpo_phys->bpo_comp, comp, sizeof (comp));
+ zdb_nicenum(bpo->bpo_phys->bpo_uncomp, uncomp, sizeof (uncomp));
(void) printf(" %*s: object %llu, %llu local blkptrs, "
"%llu subobjs in object, %llu, %s (%s/%s comp)\n",
indent * 8, name,
continue;
}
dump_full_bpobj(&subbpo, "subobj", indent + 1);
+ bpobj_close(&subbpo);
}
} else {
(void) printf(" %*s: object %llu, %llu blkptrs, %s\n",
char comp[32];
char uncomp[32];
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (bytes) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (comp) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (uncomp) >= NN_NUMBUF_SZ);
+
if (dump_opt['d'] < 3)
return;
return;
}
- zdb_nicenum(dl->dl_phys->dl_used, bytes);
- zdb_nicenum(dl->dl_phys->dl_comp, comp);
- zdb_nicenum(dl->dl_phys->dl_uncomp, uncomp);
+ zdb_nicenum(dl->dl_phys->dl_used, bytes, sizeof (bytes));
+ zdb_nicenum(dl->dl_phys->dl_comp, comp, sizeof (comp));
+ zdb_nicenum(dl->dl_phys->dl_uncomp, uncomp, sizeof (uncomp));
(void) printf("\n Deadlist: %s (%s/%s comp)\n",
bytes, comp, uncomp);
static avl_tree_t idx_tree;
static avl_tree_t domain_tree;
static boolean_t fuid_table_loaded;
-static boolean_t sa_loaded;
-sa_attr_type_t *sa_attr_table;
+static objset_t *sa_os = NULL;
+static sa_attr_type_t *sa_attr_table = NULL;
+
+static int
+open_objset(const char *path, dmu_objset_type_t type, void *tag, objset_t **osp)
+{
+ int err;
+ uint64_t sa_attrs = 0;
+ uint64_t version = 0;
+
+ VERIFY3P(sa_os, ==, NULL);
+ err = dmu_objset_own(path, type, B_TRUE, B_FALSE, tag, osp);
+ if (err != 0) {
+ (void) fprintf(stderr, "failed to own dataset '%s': %s\n", path,
+ strerror(err));
+ return (err);
+ }
+
+ if (dmu_objset_type(*osp) == DMU_OST_ZFS && !(*osp)->os_encrypted) {
+ (void) zap_lookup(*osp, MASTER_NODE_OBJ, ZPL_VERSION_STR,
+ 8, 1, &version);
+ if (version >= ZPL_VERSION_SA) {
+ (void) zap_lookup(*osp, MASTER_NODE_OBJ, ZFS_SA_ATTRS,
+ 8, 1, &sa_attrs);
+ }
+ err = sa_setup(*osp, sa_attrs, zfs_attr_table, ZPL_END,
+ &sa_attr_table);
+ if (err != 0) {
+ (void) fprintf(stderr, "sa_setup failed: %s\n",
+ strerror(err));
+ dmu_objset_disown(*osp, B_FALSE, tag);
+ *osp = NULL;
+ }
+ }
+ sa_os = *osp;
+
+ return (0);
+}
+
+static void
+close_objset(objset_t *os, void *tag)
+{
+ VERIFY3P(os, ==, sa_os);
+ if (os->os_sa != NULL)
+ sa_tear_down(os);
+ dmu_objset_disown(os, B_FALSE, tag);
+ sa_attr_table = NULL;
+ sa_os = NULL;
+}
static void
fuid_table_destroy(void)
int idx = 0;
int error;
- if (!sa_loaded) {
- uint64_t sa_attrs = 0;
- uint64_t version;
-
- VERIFY(zap_lookup(os, MASTER_NODE_OBJ, ZPL_VERSION_STR,
- 8, 1, &version) == 0);
- if (version >= ZPL_VERSION_SA) {
- VERIFY(zap_lookup(os, MASTER_NODE_OBJ, ZFS_SA_ATTRS,
- 8, 1, &sa_attrs) == 0);
- }
- if ((error = sa_setup(os, sa_attrs, zfs_attr_table,
- ZPL_END, &sa_attr_table)) != 0) {
- (void) printf("sa_setup failed errno %d, can't "
- "display znode contents\n", error);
- return;
- }
- sa_loaded = B_TRUE;
- }
-
+ VERIFY3P(os, ==, sa_os);
if (sa_handle_get(os, object, NULL, SA_HDL_PRIVATE, &hdl)) {
(void) printf("Failed to get handle for SA znode\n");
return;
return;
}
- error = zfs_obj_to_path(os, object, path, sizeof (path));
- if (error != 0) {
- (void) snprintf(path, sizeof (path), "\?\?\?<object#%llu>",
- (u_longlong_t)object);
- }
- if (dump_opt['d'] < 3) {
- (void) printf("\t%s\n", path);
- (void) sa_handle_destroy(hdl);
- return;
- }
-
z_crtime = (time_t)crtm[0];
z_atime = (time_t)acctm[0];
z_mtime = (time_t)modtm[0];
z_ctime = (time_t)chgtm[0];
- (void) printf("\tpath %s\n", path);
+ if (dump_opt['d'] > 4) {
+ error = zfs_obj_to_path(os, object, path, sizeof (path));
+ if (error != 0) {
+ (void) snprintf(path, sizeof (path),
+ "\?\?\?<object#%llu>", (u_longlong_t)object);
+ }
+ (void) printf("\tpath %s\n", path);
+ }
dump_uidgid(os, uid, gid);
(void) printf("\tatime %s", ctime(&z_atime));
(void) printf("\tmtime %s", ctime(&z_mtime));
(void) printf("\tparent %llu\n", (u_longlong_t)parent);
(void) printf("\tlinks %llu\n", (u_longlong_t)links);
(void) printf("\tpflags %llx\n", (u_longlong_t)pflags);
+ if (dmu_objset_projectquota_enabled(os) && (pflags & ZFS_PROJID)) {
+ uint64_t projid;
+
+ if (sa_lookup(hdl, sa_attr_table[ZPL_PROJID], &projid,
+ sizeof (uint64_t)) == 0)
+ (void) printf("\tprojid %llu\n", (u_longlong_t)projid);
+ }
if (sa_lookup(hdl, sa_attr_table[ZPL_XATTR], &xattr,
sizeof (uint64_t)) == 0)
(void) printf("\txattr %llu\n", (u_longlong_t)xattr);
dump_packed_nvlist, /* FUID nvlist size */
dump_zap, /* DSL dataset next clones */
dump_zap, /* DSL scrub queue */
- dump_zap, /* ZFS user/group used */
- dump_zap, /* ZFS user/group quota */
+ dump_zap, /* ZFS user/group/project used */
+ dump_zap, /* ZFS user/group/project quota */
dump_zap, /* snapshot refcount tags */
dump_ddt_zap, /* DDT ZAP object */
dump_zap, /* DDT statistics */
};
static void
-dump_object(objset_t *os, uint64_t object, int verbosity, int *print_header)
+dump_object(objset_t *os, uint64_t object, int verbosity, int *print_header,
+ uint64_t *dnode_slots_used)
{
dmu_buf_t *db = NULL;
dmu_object_info_t doi;
dnode_t *dn;
+ boolean_t dnode_held = B_FALSE;
void *bonus = NULL;
size_t bsize = 0;
- char iblk[32], dblk[32], lsize[32], asize[32], fill[32];
+ char iblk[32], dblk[32], lsize[32], asize[32], fill[32], dnsize[32];
char bonus_size[32];
char aux[50];
int error;
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (iblk) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (dblk) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (lsize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (asize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (bonus_size) >= NN_NUMBUF_SZ);
+
if (*print_header) {
- (void) printf("\n%10s %3s %5s %5s %5s %5s %6s %s\n",
- "Object", "lvl", "iblk", "dblk", "dsize", "lsize",
- "%full", "type");
+ (void) printf("\n%10s %3s %5s %5s %5s %6s %5s %6s %s\n",
+ "Object", "lvl", "iblk", "dblk", "dsize", "dnsize",
+ "lsize", "%full", "type");
*print_header = 0;
}
if (object == 0) {
dn = DMU_META_DNODE(os);
+ dmu_object_info_from_dnode(dn, &doi);
} else {
- error = dmu_bonus_hold(os, object, FTAG, &db);
+ /*
+ * Encrypted datasets will have sensitive bonus buffers
+ * encrypted. Therefore we cannot hold the bonus buffer and
+ * must hold the dnode itself instead.
+ */
+ error = dmu_object_info(os, object, &doi);
if (error)
- fatal("dmu_bonus_hold(%llu) failed, errno %u",
- object, error);
- bonus = db->db_data;
- bsize = db->db_size;
- dn = DB_DNODE((dmu_buf_impl_t *)db);
- }
- dmu_object_info_from_dnode(dn, &doi);
-
- zdb_nicenum(doi.doi_metadata_block_size, iblk);
- zdb_nicenum(doi.doi_data_block_size, dblk);
- zdb_nicenum(doi.doi_max_offset, lsize);
- zdb_nicenum(doi.doi_physical_blocks_512 << 9, asize);
- zdb_nicenum(doi.doi_bonus_size, bonus_size);
+ fatal("dmu_object_info() failed, errno %u", error);
+
+ if (os->os_encrypted &&
+ DMU_OT_IS_ENCRYPTED(doi.doi_bonus_type)) {
+ error = dnode_hold(os, object, FTAG, &dn);
+ if (error)
+ fatal("dnode_hold() failed, errno %u", error);
+ dnode_held = B_TRUE;
+ } else {
+ error = dmu_bonus_hold(os, object, FTAG, &db);
+ if (error)
+ fatal("dmu_bonus_hold(%llu) failed, errno %u",
+ object, error);
+ bonus = db->db_data;
+ bsize = db->db_size;
+ dn = DB_DNODE((dmu_buf_impl_t *)db);
+ }
+ }
+
+ if (dnode_slots_used)
+ *dnode_slots_used = doi.doi_dnodesize / DNODE_MIN_SIZE;
+
+ zdb_nicenum(doi.doi_metadata_block_size, iblk, sizeof (iblk));
+ zdb_nicenum(doi.doi_data_block_size, dblk, sizeof (dblk));
+ zdb_nicenum(doi.doi_max_offset, lsize, sizeof (lsize));
+ zdb_nicenum(doi.doi_physical_blocks_512 << 9, asize, sizeof (asize));
+ zdb_nicenum(doi.doi_bonus_size, bonus_size, sizeof (bonus_size));
+ zdb_nicenum(doi.doi_dnodesize, dnsize, sizeof (dnsize));
(void) sprintf(fill, "%6.2f", 100.0 * doi.doi_fill_count *
doi.doi_data_block_size / (object == 0 ? DNODES_PER_BLOCK : 1) /
doi.doi_max_offset);
aux[0] = '\0';
if (doi.doi_checksum != ZIO_CHECKSUM_INHERIT || verbosity >= 6) {
- (void) snprintf(aux + strlen(aux), sizeof (aux), " (K=%s)",
- ZDB_CHECKSUM_NAME(doi.doi_checksum));
+ (void) snprintf(aux + strlen(aux), sizeof (aux) - strlen(aux),
+ " (K=%s)", ZDB_CHECKSUM_NAME(doi.doi_checksum));
}
if (doi.doi_compress != ZIO_COMPRESS_INHERIT || verbosity >= 6) {
- (void) snprintf(aux + strlen(aux), sizeof (aux), " (Z=%s)",
- ZDB_COMPRESS_NAME(doi.doi_compress));
+ (void) snprintf(aux + strlen(aux), sizeof (aux) - strlen(aux),
+ " (Z=%s)", ZDB_COMPRESS_NAME(doi.doi_compress));
}
- (void) printf("%10lld %3u %5s %5s %5s %5s %6s %s%s\n",
+ (void) printf("%10lld %3u %5s %5s %5s %6s %5s %6s %s%s\n",
(u_longlong_t)object, doi.doi_indirection, iblk, dblk,
- asize, lsize, fill, ZDB_OT_NAME(doi.doi_type), aux);
+ asize, dnsize, lsize, fill, zdb_ot_name(doi.doi_type), aux);
if (doi.doi_bonus_type != DMU_OT_NONE && verbosity > 3) {
- (void) printf("%10s %3s %5s %5s %5s %5s %6s %s\n",
- "", "", "", "", "", bonus_size, "bonus",
- ZDB_OT_NAME(doi.doi_bonus_type));
+ (void) printf("%10s %3s %5s %5s %5s %5s %5s %6s %s\n",
+ "", "", "", "", "", "", bonus_size, "bonus",
+ zdb_ot_name(doi.doi_bonus_type));
}
if (verbosity >= 4) {
- (void) printf("\tdnode flags: %s%s%s\n",
+ (void) printf("\tdnode flags: %s%s%s%s\n",
(dn->dn_phys->dn_flags & DNODE_FLAG_USED_BYTES) ?
"USED_BYTES " : "",
(dn->dn_phys->dn_flags & DNODE_FLAG_USERUSED_ACCOUNTED) ?
"USERUSED_ACCOUNTED " : "",
+ (dn->dn_phys->dn_flags & DNODE_FLAG_USEROBJUSED_ACCOUNTED) ?
+ "USEROBJUSED_ACCOUNTED " : "",
(dn->dn_phys->dn_flags & DNODE_FLAG_SPILL_BLKPTR) ?
"SPILL_BLKPTR" : "");
(void) printf("\tdnode maxblkid: %llu\n",
(longlong_t)dn->dn_phys->dn_maxblkid);
- object_viewer[ZDB_OT_TYPE(doi.doi_bonus_type)](os, object,
- bonus, bsize);
- object_viewer[ZDB_OT_TYPE(doi.doi_type)](os, object, NULL, 0);
+ if (!dnode_held) {
+ object_viewer[ZDB_OT_TYPE(doi.doi_bonus_type)](os,
+ object, bonus, bsize);
+ } else {
+ (void) printf("\t\t(bonus encrypted)\n");
+ }
+
+ if (!os->os_encrypted || !DMU_OT_IS_ENCRYPTED(doi.doi_type)) {
+ object_viewer[ZDB_OT_TYPE(doi.doi_type)](os, object,
+ NULL, 0);
+ } else {
+ (void) printf("\t\t(object encrypted)\n");
+ }
+
*print_header = 1;
}
for (;;) {
char segsize[32];
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (segsize) >= NN_NUMBUF_SZ);
error = dnode_next_offset(dn,
0, &start, minlvl, blkfill, 0);
if (error)
end = start;
error = dnode_next_offset(dn,
DNODE_FIND_HOLE, &end, minlvl, blkfill, 0);
- zdb_nicenum(end - start, segsize);
+ zdb_nicenum(end - start, segsize, sizeof (segsize));
(void) printf("\t\tsegment [%016llx, %016llx)"
" size %5s\n", (u_longlong_t)start,
(u_longlong_t)end, segsize);
if (db != NULL)
dmu_buf_rele(db, FTAG);
+ if (dnode_held)
+ dnode_rele(dn, FTAG);
}
-static char *objset_types[DMU_OST_NUMTYPES] = {
+static const char *objset_types[DMU_OST_NUMTYPES] = {
"NONE", "META", "ZPL", "ZVOL", "OTHER", "ANY" };
static void
uint64_t refdbytes, usedobjs, scratch;
char numbuf[32];
char blkbuf[BP_SPRINTF_LEN + 20];
- char osname[MAXNAMELEN];
- char *type = "UNKNOWN";
+ char osname[ZFS_MAX_DATASET_NAME_LEN];
+ const char *type = "UNKNOWN";
int verbosity = dump_opt['d'];
int print_header = 1;
- int i, error;
+ unsigned i;
+ int error;
+ uint64_t total_slots_used = 0;
+ uint64_t max_slot_used = 0;
+ uint64_t dnode_slots;
+
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (numbuf) >= NN_NUMBUF_SZ);
dsl_pool_config_enter(dmu_objset_pool(os), FTAG);
dmu_objset_fast_stat(os, &dds);
ASSERT3U(usedobjs, ==, BP_GET_FILL(os->os_rootbp));
- zdb_nicenum(refdbytes, numbuf);
+ zdb_nicenum(refdbytes, numbuf, sizeof (numbuf));
if (verbosity >= 4) {
(void) snprintf(blkbuf, sizeof (blkbuf), ", rootbp ");
if (zopt_objects != 0) {
for (i = 0; i < zopt_objects; i++)
dump_object(os, zopt_object[i], verbosity,
- &print_header);
+ &print_header, NULL);
(void) printf("\n");
return;
}
if (dump_opt['i'] != 0 || verbosity >= 2)
dump_intent_log(dmu_objset_zil(os));
- if (dmu_objset_ds(os) != NULL)
- dump_deadlist(&dmu_objset_ds(os)->ds_deadlist);
+ if (dmu_objset_ds(os) != NULL) {
+ dsl_dataset_t *ds = dmu_objset_ds(os);
+ dump_deadlist(&ds->ds_deadlist);
+
+ if (dsl_dataset_remap_deadlist_exists(ds)) {
+ (void) printf("ds_remap_deadlist:\n");
+ dump_deadlist(&ds->ds_remap_deadlist);
+ }
+ }
if (verbosity < 2)
return;
if (BP_IS_HOLE(os->os_rootbp))
return;
- dump_object(os, 0, verbosity, &print_header);
+ dump_object(os, 0, verbosity, &print_header, NULL);
object_count = 0;
if (DMU_USERUSED_DNODE(os) != NULL &&
DMU_USERUSED_DNODE(os)->dn_type != 0) {
- dump_object(os, DMU_USERUSED_OBJECT, verbosity, &print_header);
- dump_object(os, DMU_GROUPUSED_OBJECT, verbosity, &print_header);
+ dump_object(os, DMU_USERUSED_OBJECT, verbosity, &print_header,
+ NULL);
+ dump_object(os, DMU_GROUPUSED_OBJECT, verbosity, &print_header,
+ NULL);
}
+ if (DMU_PROJECTUSED_DNODE(os) != NULL &&
+ DMU_PROJECTUSED_DNODE(os)->dn_type != 0)
+ dump_object(os, DMU_PROJECTUSED_OBJECT, verbosity,
+ &print_header, NULL);
+
object = 0;
while ((error = dmu_object_next(os, &object, B_FALSE, 0)) == 0) {
- dump_object(os, object, verbosity, &print_header);
+ dump_object(os, object, verbosity, &print_header, &dnode_slots);
object_count++;
+ total_slots_used += dnode_slots;
+ max_slot_used = object + dnode_slots - 1;
}
- ASSERT3U(object_count, ==, usedobjs);
+ (void) printf("\n");
+
+ (void) printf(" Dnode slots:\n");
+ (void) printf("\tTotal used: %10llu\n",
+ (u_longlong_t)total_slots_used);
+ (void) printf("\tMax used: %10llu\n",
+ (u_longlong_t)max_slot_used);
+ (void) printf("\tPercent empty: %10lf\n",
+ (double)(max_slot_used - total_slots_used)*100 /
+ (double)max_slot_used);
(void) printf("\n");
(void) fprintf(stderr, "dmu_object_next() = %d\n", error);
abort();
}
+
+ ASSERT3U(object_count, ==, usedobjs);
}
static void
(void) printf("\tguid_sum = %llu\n", (u_longlong_t)ub->ub_guid_sum);
(void) printf("\ttimestamp = %llu UTC = %s",
(u_longlong_t)ub->ub_timestamp, asctime(localtime(×tamp)));
- if (dump_opt['u'] >= 3) {
+
+ (void) printf("\tmmp_magic = %016llx\n",
+ (u_longlong_t)ub->ub_mmp_magic);
+ if (ub->ub_mmp_magic == MMP_MAGIC)
+ (void) printf("\tmmp_delay = %0llu\n",
+ (u_longlong_t)ub->ub_mmp_delay);
+
+ if (dump_opt['u'] >= 4) {
char blkbuf[BP_SPRINTF_LEN];
snprintf_blkptr(blkbuf, sizeof (blkbuf), &ub->ub_rootbp);
(void) printf("\trootbp = %s\n", blkbuf);
nvlist_free(config);
}
-#define ZDB_MAX_UB_HEADER_SIZE 32
+/*
+ * ZFS label nvlist stats
+ */
+typedef struct zdb_nvl_stats {
+ int zns_list_count;
+ int zns_leaf_count;
+ size_t zns_leaf_largest;
+ size_t zns_leaf_total;
+ nvlist_t *zns_string;
+ nvlist_t *zns_uint64;
+ nvlist_t *zns_boolean;
+} zdb_nvl_stats_t;
static void
-dump_label_uberblocks(vdev_label_t *lbl, uint64_t ashift)
+collect_nvlist_stats(nvlist_t *nvl, zdb_nvl_stats_t *stats)
{
- vdev_t vd;
- vdev_t *vdp = &vd;
- char header[ZDB_MAX_UB_HEADER_SIZE];
- int i;
+ nvlist_t *list, **array;
+ nvpair_t *nvp = NULL;
+ char *name;
+ uint_t i, items;
- vd.vdev_ashift = ashift;
- vdp->vdev_top = vdp;
+ stats->zns_list_count++;
- for (i = 0; i < VDEV_UBERBLOCK_COUNT(vdp); i++) {
- uint64_t uoff = VDEV_UBERBLOCK_OFFSET(vdp, i);
- uberblock_t *ub = (void *)((char *)lbl + uoff);
+ while ((nvp = nvlist_next_nvpair(nvl, nvp)) != NULL) {
+ name = nvpair_name(nvp);
- if (uberblock_verify(ub))
- continue;
- (void) snprintf(header, ZDB_MAX_UB_HEADER_SIZE,
- "Uberblock[%d]\n", i);
- dump_uberblock(ub, header, "");
+ switch (nvpair_type(nvp)) {
+ case DATA_TYPE_STRING:
+ fnvlist_add_string(stats->zns_string, name,
+ fnvpair_value_string(nvp));
+ break;
+ case DATA_TYPE_UINT64:
+ fnvlist_add_uint64(stats->zns_uint64, name,
+ fnvpair_value_uint64(nvp));
+ break;
+ case DATA_TYPE_BOOLEAN:
+ fnvlist_add_boolean(stats->zns_boolean, name);
+ break;
+ case DATA_TYPE_NVLIST:
+ if (nvpair_value_nvlist(nvp, &list) == 0)
+ collect_nvlist_stats(list, stats);
+ break;
+ case DATA_TYPE_NVLIST_ARRAY:
+ if (nvpair_value_nvlist_array(nvp, &array, &items) != 0)
+ break;
+
+ for (i = 0; i < items; i++) {
+ collect_nvlist_stats(array[i], stats);
+
+ /* collect stats on leaf vdev */
+ if (strcmp(name, "children") == 0) {
+ size_t size;
+
+ (void) nvlist_size(array[i], &size,
+ NV_ENCODE_XDR);
+ stats->zns_leaf_total += size;
+ if (size > stats->zns_leaf_largest)
+ stats->zns_leaf_largest = size;
+ stats->zns_leaf_count++;
+ }
+ }
+ break;
+ default:
+ (void) printf("skip type %d!\n", (int)nvpair_type(nvp));
+ }
}
}
static void
-dump_label(const char *dev)
+dump_nvlist_stats(nvlist_t *nvl, size_t cap)
{
- int fd;
- vdev_label_t label;
- char *path, *buf = label.vl_vdev_phys.vp_nvlist;
- size_t buflen = sizeof (label.vl_vdev_phys.vp_nvlist);
- struct stat64 statbuf;
- uint64_t psize, ashift;
- int len = strlen(dev) + 1;
- int l;
-
- if (strncmp(dev, "/dev/dsk/", 9) == 0) {
- len++;
- path = malloc(len);
- (void) snprintf(path, len, "%s%s", "/dev/rdsk/", dev + 9);
- } else {
- path = strdup(dev);
+ zdb_nvl_stats_t stats = { 0 };
+ size_t size, sum = 0, total;
+ size_t noise;
+
+ /* requires nvlist with non-unique names for stat collection */
+ VERIFY0(nvlist_alloc(&stats.zns_string, 0, 0));
+ VERIFY0(nvlist_alloc(&stats.zns_uint64, 0, 0));
+ VERIFY0(nvlist_alloc(&stats.zns_boolean, 0, 0));
+ VERIFY0(nvlist_size(stats.zns_boolean, &noise, NV_ENCODE_XDR));
+
+ (void) printf("\n\nZFS Label NVList Config Stats:\n");
+
+ VERIFY0(nvlist_size(nvl, &total, NV_ENCODE_XDR));
+ (void) printf(" %d bytes used, %d bytes free (using %4.1f%%)\n\n",
+ (int)total, (int)(cap - total), 100.0 * total / cap);
+
+ collect_nvlist_stats(nvl, &stats);
+
+ VERIFY0(nvlist_size(stats.zns_uint64, &size, NV_ENCODE_XDR));
+ size -= noise;
+ sum += size;
+ (void) printf("%12s %4d %6d bytes (%5.2f%%)\n", "integers:",
+ (int)fnvlist_num_pairs(stats.zns_uint64),
+ (int)size, 100.0 * size / total);
+
+ VERIFY0(nvlist_size(stats.zns_string, &size, NV_ENCODE_XDR));
+ size -= noise;
+ sum += size;
+ (void) printf("%12s %4d %6d bytes (%5.2f%%)\n", "strings:",
+ (int)fnvlist_num_pairs(stats.zns_string),
+ (int)size, 100.0 * size / total);
+
+ VERIFY0(nvlist_size(stats.zns_boolean, &size, NV_ENCODE_XDR));
+ size -= noise;
+ sum += size;
+ (void) printf("%12s %4d %6d bytes (%5.2f%%)\n", "booleans:",
+ (int)fnvlist_num_pairs(stats.zns_boolean),
+ (int)size, 100.0 * size / total);
+
+ size = total - sum; /* treat remainder as nvlist overhead */
+ (void) printf("%12s %4d %6d bytes (%5.2f%%)\n\n", "nvlists:",
+ stats.zns_list_count, (int)size, 100.0 * size / total);
+
+ if (stats.zns_leaf_count > 0) {
+ size_t average = stats.zns_leaf_total / stats.zns_leaf_count;
+
+ (void) printf("%12s %4d %6d bytes average\n", "leaf vdevs:",
+ stats.zns_leaf_count, (int)average);
+ (void) printf("%24d bytes largest\n",
+ (int)stats.zns_leaf_largest);
+
+ if (dump_opt['l'] >= 3 && average > 0)
+ (void) printf(" space for %d additional leaf vdevs\n",
+ (int)((cap - total) / average));
}
+ (void) printf("\n");
- if ((fd = open64(path, O_RDONLY)) < 0) {
- (void) printf("cannot open '%s': %s\n", path, strerror(errno));
- free(path);
- exit(1);
- }
+ nvlist_free(stats.zns_string);
+ nvlist_free(stats.zns_uint64);
+ nvlist_free(stats.zns_boolean);
+}
- if (fstat64_blk(fd, &statbuf) != 0) {
- (void) printf("failed to stat '%s': %s\n", path,
- strerror(errno));
- free(path);
- (void) close(fd);
- exit(1);
+typedef struct cksum_record {
+ zio_cksum_t cksum;
+ boolean_t labels[VDEV_LABELS];
+ avl_node_t link;
+} cksum_record_t;
+
+static int
+cksum_record_compare(const void *x1, const void *x2)
+{
+ const cksum_record_t *l = (cksum_record_t *)x1;
+ const cksum_record_t *r = (cksum_record_t *)x2;
+ int arraysize = ARRAY_SIZE(l->cksum.zc_word);
+ int difference;
+
+ for (int i = 0; i < arraysize; i++) {
+ difference = AVL_CMP(l->cksum.zc_word[i], r->cksum.zc_word[i]);
+ if (difference)
+ break;
}
- psize = statbuf.st_size;
- psize = P2ALIGN(psize, (uint64_t)sizeof (vdev_label_t));
+ return (difference);
+}
- for (l = 0; l < VDEV_LABELS; l++) {
- nvlist_t *config = NULL;
+static cksum_record_t *
+cksum_record_alloc(zio_cksum_t *cksum, int l)
+{
+ cksum_record_t *rec;
- (void) printf("--------------------------------------------\n");
- (void) printf("LABEL %d\n", l);
- (void) printf("--------------------------------------------\n");
+ rec = umem_zalloc(sizeof (*rec), UMEM_NOFAIL);
+ rec->cksum = *cksum;
+ rec->labels[l] = B_TRUE;
- if (pread64(fd, &label, sizeof (label),
- vdev_label_offset(psize, l, 0)) != sizeof (label)) {
- (void) printf("failed to read label %d\n", l);
- continue;
- }
+ return (rec);
+}
- if (nvlist_unpack(buf, buflen, &config, 0) != 0) {
- (void) printf("failed to unpack label %d\n", l);
- ashift = SPA_MINBLOCKSHIFT;
- } else {
- nvlist_t *vdev_tree = NULL;
+static cksum_record_t *
+cksum_record_lookup(avl_tree_t *tree, zio_cksum_t *cksum)
+{
+ cksum_record_t lookup = { .cksum = *cksum };
+ avl_index_t where;
- dump_nvlist(config, 4);
- if ((nvlist_lookup_nvlist(config,
- ZPOOL_CONFIG_VDEV_TREE, &vdev_tree) != 0) ||
- (nvlist_lookup_uint64(vdev_tree,
- ZPOOL_CONFIG_ASHIFT, &ashift) != 0))
- ashift = SPA_MINBLOCKSHIFT;
- nvlist_free(config);
- }
- if (dump_opt['u'])
- dump_label_uberblocks(&label, ashift);
+ return (avl_find(tree, &lookup, &where));
+}
+
+static cksum_record_t *
+cksum_record_insert(avl_tree_t *tree, zio_cksum_t *cksum, int l)
+{
+ cksum_record_t *rec;
+
+ rec = cksum_record_lookup(tree, cksum);
+ if (rec) {
+ rec->labels[l] = B_TRUE;
+ } else {
+ rec = cksum_record_alloc(cksum, l);
+ avl_add(tree, rec);
}
- free(path);
- (void) close(fd);
+ return (rec);
}
-static uint64_t num_large_blocks;
-
-/*ARGSUSED*/
static int
-dump_one_dir(const char *dsname, void *arg)
+first_label(cksum_record_t *rec)
{
- int error;
- objset_t *os;
+ for (int i = 0; i < VDEV_LABELS; i++)
+ if (rec->labels[i])
+ return (i);
+
+ return (-1);
+}
+
+static void
+print_label_numbers(char *prefix, cksum_record_t *rec)
+{
+ printf("%s", prefix);
+ for (int i = 0; i < VDEV_LABELS; i++)
+ if (rec->labels[i] == B_TRUE)
+ printf("%d ", i);
+ printf("\n");
+}
+
+#define MAX_UBERBLOCK_COUNT (VDEV_UBERBLOCK_RING >> UBERBLOCK_SHIFT)
+
+typedef struct label {
+ vdev_label_t label;
+ nvlist_t *config_nv;
+ cksum_record_t *config;
+ cksum_record_t *uberblocks[MAX_UBERBLOCK_COUNT];
+ boolean_t header_printed;
+ boolean_t read_failed;
+} label_t;
+
+static void
+print_label_header(label_t *label, int l)
+{
+
+ if (dump_opt['q'])
+ return;
+
+ if (label->header_printed == B_TRUE)
+ return;
+
+ (void) printf("------------------------------------\n");
+ (void) printf("LABEL %d\n", l);
+ (void) printf("------------------------------------\n");
+
+ label->header_printed = B_TRUE;
+}
+
+static void
+dump_config_from_label(label_t *label, size_t buflen, int l)
+{
+ if (dump_opt['q'])
+ return;
+
+ if ((dump_opt['l'] < 3) && (first_label(label->config) != l))
+ return;
+
+ print_label_header(label, l);
+ dump_nvlist(label->config_nv, 4);
+ print_label_numbers(" labels = ", label->config);
+
+ if (dump_opt['l'] >= 2)
+ dump_nvlist_stats(label->config_nv, buflen);
+}
+
+#define ZDB_MAX_UB_HEADER_SIZE 32
+
+static void
+dump_label_uberblocks(label_t *label, uint64_t ashift, int label_num)
+{
+
+ vdev_t vd;
+ char header[ZDB_MAX_UB_HEADER_SIZE];
+
+ vd.vdev_ashift = ashift;
+ vd.vdev_top = &vd;
+
+ for (int i = 0; i < VDEV_UBERBLOCK_COUNT(&vd); i++) {
+ uint64_t uoff = VDEV_UBERBLOCK_OFFSET(&vd, i);
+ uberblock_t *ub = (void *)((char *)&label->label + uoff);
+ cksum_record_t *rec = label->uberblocks[i];
+
+ if (rec == NULL) {
+ if (dump_opt['u'] >= 2) {
+ print_label_header(label, label_num);
+ (void) printf(" Uberblock[%d] invalid\n", i);
+ }
+ continue;
+ }
+
+ if ((dump_opt['u'] < 3) && (first_label(rec) != label_num))
+ continue;
+
+ if ((dump_opt['u'] < 4) &&
+ (ub->ub_mmp_magic == MMP_MAGIC) && ub->ub_mmp_delay &&
+ (i >= VDEV_UBERBLOCK_COUNT(&vd) - MMP_BLOCKS_PER_LABEL))
+ continue;
+
+ print_label_header(label, label_num);
+ (void) snprintf(header, ZDB_MAX_UB_HEADER_SIZE,
+ " Uberblock[%d]\n", i);
+ dump_uberblock(ub, header, "");
+ print_label_numbers(" labels = ", rec);
+ }
+}
+
+static char curpath[PATH_MAX];
+
+/*
+ * Iterate through the path components, recursively passing
+ * current one's obj and remaining path until we find the obj
+ * for the last one.
+ */
+static int
+dump_path_impl(objset_t *os, uint64_t obj, char *name)
+{
+ int err;
+ int header = 1;
+ uint64_t child_obj;
+ char *s;
+ dmu_buf_t *db;
+ dmu_object_info_t doi;
+
+ if ((s = strchr(name, '/')) != NULL)
+ *s = '\0';
+ err = zap_lookup(os, obj, name, 8, 1, &child_obj);
+
+ (void) strlcat(curpath, name, sizeof (curpath));
+
+ if (err != 0) {
+ (void) fprintf(stderr, "failed to lookup %s: %s\n",
+ curpath, strerror(err));
+ return (err);
+ }
+
+ child_obj = ZFS_DIRENT_OBJ(child_obj);
+ err = sa_buf_hold(os, child_obj, FTAG, &db);
+ if (err != 0) {
+ (void) fprintf(stderr,
+ "failed to get SA dbuf for obj %llu: %s\n",
+ (u_longlong_t)child_obj, strerror(err));
+ return (EINVAL);
+ }
+ dmu_object_info_from_db(db, &doi);
+ sa_buf_rele(db, FTAG);
+
+ if (doi.doi_bonus_type != DMU_OT_SA &&
+ doi.doi_bonus_type != DMU_OT_ZNODE) {
+ (void) fprintf(stderr, "invalid bonus type %d for obj %llu\n",
+ doi.doi_bonus_type, (u_longlong_t)child_obj);
+ return (EINVAL);
+ }
+
+ if (dump_opt['v'] > 6) {
+ (void) printf("obj=%llu %s type=%d bonustype=%d\n",
+ (u_longlong_t)child_obj, curpath, doi.doi_type,
+ doi.doi_bonus_type);
+ }
+
+ (void) strlcat(curpath, "/", sizeof (curpath));
+
+ switch (doi.doi_type) {
+ case DMU_OT_DIRECTORY_CONTENTS:
+ if (s != NULL && *(s + 1) != '\0')
+ return (dump_path_impl(os, child_obj, s + 1));
+ /*FALLTHROUGH*/
+ case DMU_OT_PLAIN_FILE_CONTENTS:
+ dump_object(os, child_obj, dump_opt['v'], &header, NULL);
+ return (0);
+ default:
+ (void) fprintf(stderr, "object %llu has non-file/directory "
+ "type %d\n", (u_longlong_t)obj, doi.doi_type);
+ break;
+ }
+
+ return (EINVAL);
+}
+
+/*
+ * Dump the blocks for the object specified by path inside the dataset.
+ */
+static int
+dump_path(char *ds, char *path)
+{
+ int err;
+ objset_t *os;
+ uint64_t root_obj;
+
+ err = open_objset(ds, DMU_OST_ZFS, FTAG, &os);
+ if (err != 0)
+ return (err);
+
+ err = zap_lookup(os, MASTER_NODE_OBJ, ZFS_ROOT_OBJ, 8, 1, &root_obj);
+ if (err != 0) {
+ (void) fprintf(stderr, "can't lookup root znode: %s\n",
+ strerror(err));
+ dmu_objset_disown(os, B_FALSE, FTAG);
+ return (EINVAL);
+ }
+
+ (void) snprintf(curpath, sizeof (curpath), "dataset=%s path=/", ds);
+
+ err = dump_path_impl(os, root_obj, path);
+
+ close_objset(os, FTAG);
+ return (err);
+}
+
+static int
+dump_label(const char *dev)
+{
+ char path[MAXPATHLEN];
+ label_t labels[VDEV_LABELS];
+ uint64_t psize, ashift;
+ struct stat64 statbuf;
+ boolean_t config_found = B_FALSE;
+ boolean_t error = B_FALSE;
+ avl_tree_t config_tree;
+ avl_tree_t uberblock_tree;
+ void *node, *cookie;
+ int fd;
+
+ bzero(labels, sizeof (labels));
+
+ /*
+ * Check if we were given absolute path and use it as is.
+ * Otherwise if the provided vdev name doesn't point to a file,
+ * try prepending expected disk paths and partition numbers.
+ */
+ (void) strlcpy(path, dev, sizeof (path));
+ if (dev[0] != '/' && stat64(path, &statbuf) != 0) {
+ int error;
+
+ error = zfs_resolve_shortname(dev, path, MAXPATHLEN);
+ if (error == 0 && zfs_dev_is_whole_disk(path)) {
+ if (zfs_append_partition(path, MAXPATHLEN) == -1)
+ error = ENOENT;
+ }
+
+ if (error || (stat64(path, &statbuf) != 0)) {
+ (void) printf("failed to find device %s, try "
+ "specifying absolute path instead\n", dev);
+ return (1);
+ }
+ }
+
+ if ((fd = open64(path, O_RDONLY)) < 0) {
+ (void) printf("cannot open '%s': %s\n", path, strerror(errno));
+ exit(1);
+ }
+
+ if (fstat64_blk(fd, &statbuf) != 0) {
+ (void) printf("failed to stat '%s': %s\n", path,
+ strerror(errno));
+ (void) close(fd);
+ exit(1);
+ }
+
+ if (S_ISBLK(statbuf.st_mode) && ioctl(fd, BLKFLSBUF) != 0)
+ (void) printf("failed to invalidate cache '%s' : %s\n", path,
+ strerror(errno));
+
+ avl_create(&config_tree, cksum_record_compare,
+ sizeof (cksum_record_t), offsetof(cksum_record_t, link));
+ avl_create(&uberblock_tree, cksum_record_compare,
+ sizeof (cksum_record_t), offsetof(cksum_record_t, link));
+
+ psize = statbuf.st_size;
+ psize = P2ALIGN(psize, (uint64_t)sizeof (vdev_label_t));
+ ashift = SPA_MINBLOCKSHIFT;
+
+ /*
+ * 1. Read the label from disk
+ * 2. Unpack the configuration and insert in config tree.
+ * 3. Traverse all uberblocks and insert in uberblock tree.
+ */
+ for (int l = 0; l < VDEV_LABELS; l++) {
+ label_t *label = &labels[l];
+ char *buf = label->label.vl_vdev_phys.vp_nvlist;
+ size_t buflen = sizeof (label->label.vl_vdev_phys.vp_nvlist);
+ nvlist_t *config;
+ cksum_record_t *rec;
+ zio_cksum_t cksum;
+ vdev_t vd;
+
+ if (pread64(fd, &label->label, sizeof (label->label),
+ vdev_label_offset(psize, l, 0)) != sizeof (label->label)) {
+ if (!dump_opt['q'])
+ (void) printf("failed to read label %d\n", l);
+ label->read_failed = B_TRUE;
+ error = B_TRUE;
+ continue;
+ }
+
+ label->read_failed = B_FALSE;
+
+ if (nvlist_unpack(buf, buflen, &config, 0) == 0) {
+ nvlist_t *vdev_tree = NULL;
+ size_t size;
+
+ if ((nvlist_lookup_nvlist(config,
+ ZPOOL_CONFIG_VDEV_TREE, &vdev_tree) != 0) ||
+ (nvlist_lookup_uint64(vdev_tree,
+ ZPOOL_CONFIG_ASHIFT, &ashift) != 0))
+ ashift = SPA_MINBLOCKSHIFT;
+
+ if (nvlist_size(config, &size, NV_ENCODE_XDR) != 0)
+ size = buflen;
+
+ fletcher_4_native_varsize(buf, size, &cksum);
+ rec = cksum_record_insert(&config_tree, &cksum, l);
+
+ label->config = rec;
+ label->config_nv = config;
+ config_found = B_TRUE;
+ } else {
+ error = B_TRUE;
+ }
+
+ vd.vdev_ashift = ashift;
+ vd.vdev_top = &vd;
+
+ for (int i = 0; i < VDEV_UBERBLOCK_COUNT(&vd); i++) {
+ uint64_t uoff = VDEV_UBERBLOCK_OFFSET(&vd, i);
+ uberblock_t *ub = (void *)((char *)label + uoff);
+
+ if (uberblock_verify(ub))
+ continue;
+
+ fletcher_4_native_varsize(ub, sizeof (*ub), &cksum);
+ rec = cksum_record_insert(&uberblock_tree, &cksum, l);
+
+ label->uberblocks[i] = rec;
+ }
+ }
+
+ /*
+ * Dump the label and uberblocks.
+ */
+ for (int l = 0; l < VDEV_LABELS; l++) {
+ label_t *label = &labels[l];
+ size_t buflen = sizeof (label->label.vl_vdev_phys.vp_nvlist);
+
+ if (label->read_failed == B_TRUE)
+ continue;
+
+ if (label->config_nv) {
+ dump_config_from_label(label, buflen, l);
+ } else {
+ if (!dump_opt['q'])
+ (void) printf("failed to unpack label %d\n", l);
+ }
+
+ if (dump_opt['u'])
+ dump_label_uberblocks(label, ashift, l);
+
+ nvlist_free(label->config_nv);
+ }
+
+ cookie = NULL;
+ while ((node = avl_destroy_nodes(&config_tree, &cookie)) != NULL)
+ umem_free(node, sizeof (cksum_record_t));
+
+ cookie = NULL;
+ while ((node = avl_destroy_nodes(&uberblock_tree, &cookie)) != NULL)
+ umem_free(node, sizeof (cksum_record_t));
+
+ avl_destroy(&config_tree);
+ avl_destroy(&uberblock_tree);
+
+ (void) close(fd);
+
+ return (config_found == B_FALSE ? 2 :
+ (error == B_TRUE ? 1 : 0));
+}
+
+static uint64_t dataset_feature_count[SPA_FEATURES];
+static uint64_t remap_deadlist_count = 0;
+
+/*ARGSUSED*/
+static int
+dump_one_dir(const char *dsname, void *arg)
+{
+ int error;
+ objset_t *os;
+ spa_feature_t f;
+
+ error = open_objset(dsname, DMU_OST_ANY, FTAG, &os);
+ if (error != 0)
+ return (0);
+
+ for (f = 0; f < SPA_FEATURES; f++) {
+ if (!dmu_objset_ds(os)->ds_feature_inuse[f])
+ continue;
+ ASSERT(spa_feature_table[f].fi_flags &
+ ZFEATURE_FLAG_PER_DATASET);
+ dataset_feature_count[f]++;
+ }
+
+ if (dsl_dataset_remap_deadlist_exists(dmu_objset_ds(os))) {
+ remap_deadlist_count++;
+ }
- error = dmu_objset_own(dsname, DMU_OST_ANY, B_TRUE, FTAG, &os);
- if (error) {
- (void) printf("Could not open %s, error %d\n", dsname, error);
- return (0);
- }
- if (dmu_objset_ds(os)->ds_large_blocks)
- num_large_blocks++;
dump_dir(os);
- dmu_objset_disown(os, FTAG);
+ close_objset(os, FTAG);
fuid_table_destroy();
- sa_loaded = B_FALSE;
return (0);
}
#define ZDB_OT_OTHER (DMU_OT_NUMTYPES + 2)
#define ZDB_OT_TOTAL (DMU_OT_NUMTYPES + 3)
-static char *zdb_ot_extname[] = {
+static const char *zdb_ot_extname[] = {
"deferred free",
"dedup ditto",
"other",
typedef struct zdb_cb {
zdb_blkstats_t zcb_type[ZB_TOTAL + 1][ZDB_OT_TOTAL + 1];
+ uint64_t zcb_removing_size;
uint64_t zcb_dedup_asize;
uint64_t zcb_dedup_blocks;
uint64_t zcb_embedded_blocks[NUM_BP_EMBEDDED_TYPES];
uint64_t zcb_embedded_histogram[NUM_BP_EMBEDDED_TYPES]
- [BPE_PAYLOAD_SIZE];
+ [BPE_PAYLOAD_SIZE + 1];
uint64_t zcb_start;
- uint64_t zcb_lastprint;
+ hrtime_t zcb_lastprint;
uint64_t zcb_totalasize;
uint64_t zcb_errors[256];
int zcb_readfails;
int zcb_haderrors;
spa_t *zcb_spa;
+ uint32_t **zcb_vd_obsolete_counts;
} zdb_cb_t;
static void
* SPA_OLD_MAXBLOCKSIZE; larger blocks go into the last,
* "other", bucket.
*/
- int idx = BP_GET_PSIZE(bp) >> SPA_MINBLOCKSHIFT;
+ unsigned idx = BP_GET_PSIZE(bp) >> SPA_MINBLOCKSHIFT;
idx = MIN(idx, SPA_OLD_MAXBLOCKSIZE / SPA_MINBLOCKSIZE + 1);
zb->zb_psize_histogram[idx]++;
zdb_cb_t *zcb = zio->io_private;
zbookmark_phys_t *zb = &zio->io_bookmark;
- zio_data_buf_free(zio->io_data, zio->io_size);
+ abd_free(zio->io_abd);
mutex_enter(&spa->spa_scrub_lock);
- spa->spa_scrub_inflight--;
+ spa->spa_load_verify_ios--;
cv_broadcast(&spa->spa_scrub_io_cv);
if (ioerr && !(zio->io_flags & ZIO_FLAG_SPECULATIVE)) {
dmu_object_type_t type;
boolean_t is_metadata;
+ if (bp == NULL)
+ return (0);
+
if (dump_opt['b'] >= 5 && bp->blk_birth > 0) {
char blkbuf[BP_SPRINTF_LEN];
snprintf_blkptr(blkbuf, sizeof (blkbuf), bp);
if (!BP_IS_EMBEDDED(bp) &&
(dump_opt['c'] > 1 || (dump_opt['c'] && is_metadata))) {
size_t size = BP_GET_PSIZE(bp);
- void *data = zio_data_buf_alloc(size);
+ abd_t *abd = abd_alloc(size, B_FALSE);
int flags = ZIO_FLAG_CANFAIL | ZIO_FLAG_SCRUB | ZIO_FLAG_RAW;
/* If it's an intent log block, failure is expected. */
flags |= ZIO_FLAG_SPECULATIVE;
mutex_enter(&spa->spa_scrub_lock);
- while (spa->spa_scrub_inflight > max_inflight)
+ while (spa->spa_load_verify_ios > max_inflight)
cv_wait(&spa->spa_scrub_io_cv, &spa->spa_scrub_lock);
- spa->spa_scrub_inflight++;
+ spa->spa_load_verify_ios++;
mutex_exit(&spa->spa_scrub_lock);
- zio_nowait(zio_read(NULL, spa, bp, data, size,
+ zio_nowait(zio_read(NULL, spa, bp, abd, size,
zdb_blkptr_done, zcb, ZIO_PRIORITY_ASYNC_READ, flags, zb));
}
zcb->zcb_readfails = 0;
- if (dump_opt['b'] < 5 &&
- gethrtime() > zcb->zcb_lastprint + NANOSEC) {
+ /* only call gethrtime() every 100 blocks */
+ static int iters;
+ if (++iters > 100)
+ iters = 0;
+ else
+ return (0);
+
+ if (dump_opt['b'] < 5 && gethrtime() > zcb->zcb_lastprint + NANOSEC) {
uint64_t now = gethrtime();
char buf[10];
uint64_t bytes = zcb->zcb_type[ZB_TOTAL][ZDB_OT_TOTAL].zb_asize;
int sec_remaining =
(zcb->zcb_totalasize - bytes) / 1024 / kb_per_sec;
- zfs_nicenum(bytes, buf, sizeof (buf));
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (buf) >= NN_NUMBUF_SZ);
+
+ zfs_nicebytes(bytes, buf, sizeof (buf));
(void) fprintf(stderr,
"\r%5s completed (%4dMB/s) "
"estimated time remaining: %uhr %02umin %02usec ",
NULL /* alloc */
};
+/* ARGSUSED */
+static void
+claim_segment_impl_cb(uint64_t inner_offset, vdev_t *vd, uint64_t offset,
+ uint64_t size, void *arg)
+{
+ /*
+ * This callback was called through a remap from
+ * a device being removed. Therefore, the vdev that
+ * this callback is applied to is a concrete
+ * vdev.
+ */
+ ASSERT(vdev_is_concrete(vd));
+
+ VERIFY0(metaslab_claim_impl(vd, offset, size,
+ spa_first_txg(vd->vdev_spa)));
+}
+
+static void
+claim_segment_cb(void *arg, uint64_t offset, uint64_t size)
+{
+ vdev_t *vd = arg;
+
+ vdev_indirect_ops.vdev_op_remap(vd, offset, size,
+ claim_segment_impl_cb, NULL);
+}
+
+/*
+ * After accounting for all allocated blocks that are directly referenced,
+ * we might have missed a reference to a block from a partially complete
+ * (and thus unused) indirect mapping object. We perform a secondary pass
+ * through the metaslabs we have already mapped and claim the destination
+ * blocks.
+ */
+static void
+zdb_claim_removing(spa_t *spa, zdb_cb_t *zcb)
+{
+ if (spa->spa_vdev_removal == NULL)
+ return;
+
+ spa_config_enter(spa, SCL_CONFIG, FTAG, RW_READER);
+
+ spa_vdev_removal_t *svr = spa->spa_vdev_removal;
+ vdev_t *vd = svr->svr_vdev;
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+
+ for (uint64_t msi = 0; msi < vd->vdev_ms_count; msi++) {
+ metaslab_t *msp = vd->vdev_ms[msi];
+
+ if (msp->ms_start >= vdev_indirect_mapping_max_offset(vim))
+ break;
+
+ ASSERT0(range_tree_space(svr->svr_allocd_segs));
+
+ if (msp->ms_sm != NULL) {
+ VERIFY0(space_map_load(msp->ms_sm,
+ svr->svr_allocd_segs, SM_ALLOC));
+
+ /*
+ * Clear everything past what has been synced,
+ * because we have not allocated mappings for it yet.
+ */
+ range_tree_clear(svr->svr_allocd_segs,
+ vdev_indirect_mapping_max_offset(vim),
+ msp->ms_sm->sm_start + msp->ms_sm->sm_size -
+ vdev_indirect_mapping_max_offset(vim));
+ }
+
+ zcb->zcb_removing_size +=
+ range_tree_space(svr->svr_allocd_segs);
+ range_tree_vacate(svr->svr_allocd_segs, claim_segment_cb, vd);
+ }
+
+ spa_config_exit(spa, SCL_CONFIG, FTAG);
+}
+
+/*
+ * vm_idxp is an in-out parameter which (for indirect vdevs) is the
+ * index in vim_entries that has the first entry in this metaslab. On
+ * return, it will be set to the first entry after this metaslab.
+ */
+static void
+zdb_leak_init_ms(metaslab_t *msp, uint64_t *vim_idxp)
+{
+ metaslab_group_t *mg = msp->ms_group;
+ vdev_t *vd = mg->mg_vd;
+ vdev_t *rvd = vd->vdev_spa->spa_root_vdev;
+
+ mutex_enter(&msp->ms_lock);
+ metaslab_unload(msp);
+
+ /*
+ * We don't want to spend the CPU manipulating the size-ordered
+ * tree, so clear the range_tree ops.
+ */
+ msp->ms_tree->rt_ops = NULL;
+
+ (void) fprintf(stderr,
+ "\rloading vdev %llu of %llu, metaslab %llu of %llu ...",
+ (longlong_t)vd->vdev_id,
+ (longlong_t)rvd->vdev_children,
+ (longlong_t)msp->ms_id,
+ (longlong_t)vd->vdev_ms_count);
+
+ /*
+ * For leak detection, we overload the metaslab ms_tree to
+ * contain allocated segments instead of free segments. As a
+ * result, we can't use the normal metaslab_load/unload
+ * interfaces.
+ */
+ if (vd->vdev_ops == &vdev_indirect_ops) {
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+ for (; *vim_idxp < vdev_indirect_mapping_num_entries(vim);
+ (*vim_idxp)++) {
+ vdev_indirect_mapping_entry_phys_t *vimep =
+ &vim->vim_entries[*vim_idxp];
+ uint64_t ent_offset = DVA_MAPPING_GET_SRC_OFFSET(vimep);
+ uint64_t ent_len = DVA_GET_ASIZE(&vimep->vimep_dst);
+ ASSERT3U(ent_offset, >=, msp->ms_start);
+ if (ent_offset >= msp->ms_start + msp->ms_size)
+ break;
+
+ /*
+ * Mappings do not cross metaslab boundaries,
+ * because we create them by walking the metaslabs.
+ */
+ ASSERT3U(ent_offset + ent_len, <=,
+ msp->ms_start + msp->ms_size);
+ range_tree_add(msp->ms_tree, ent_offset, ent_len);
+ }
+ } else if (msp->ms_sm != NULL) {
+ VERIFY0(space_map_load(msp->ms_sm, msp->ms_tree, SM_ALLOC));
+ }
+
+ if (!msp->ms_loaded) {
+ msp->ms_loaded = B_TRUE;
+ }
+ mutex_exit(&msp->ms_lock);
+}
+
+/* ARGSUSED */
+static int
+increment_indirect_mapping_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx)
+{
+ zdb_cb_t *zcb = arg;
+ spa_t *spa = zcb->zcb_spa;
+ vdev_t *vd;
+ const dva_t *dva = &bp->blk_dva[0];
+
+ ASSERT(!dump_opt['L']);
+ ASSERT3U(BP_GET_NDVAS(bp), ==, 1);
+
+ spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER);
+ vd = vdev_lookup_top(zcb->zcb_spa, DVA_GET_VDEV(dva));
+ ASSERT3P(vd, !=, NULL);
+ spa_config_exit(spa, SCL_VDEV, FTAG);
+
+ ASSERT(vd->vdev_indirect_config.vic_mapping_object != 0);
+ ASSERT3P(zcb->zcb_vd_obsolete_counts[vd->vdev_id], !=, NULL);
+
+ vdev_indirect_mapping_increment_obsolete_count(
+ vd->vdev_indirect_mapping,
+ DVA_GET_OFFSET(dva), DVA_GET_ASIZE(dva),
+ zcb->zcb_vd_obsolete_counts[vd->vdev_id]);
+
+ return (0);
+}
+
+static uint32_t *
+zdb_load_obsolete_counts(vdev_t *vd)
+{
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+ spa_t *spa = vd->vdev_spa;
+ spa_condensing_indirect_phys_t *scip =
+ &spa->spa_condensing_indirect_phys;
+ uint32_t *counts;
+
+ EQUIV(vdev_obsolete_sm_object(vd) != 0, vd->vdev_obsolete_sm != NULL);
+ counts = vdev_indirect_mapping_load_obsolete_counts(vim);
+ if (vd->vdev_obsolete_sm != NULL) {
+ vdev_indirect_mapping_load_obsolete_spacemap(vim, counts,
+ vd->vdev_obsolete_sm);
+ }
+ if (scip->scip_vdev == vd->vdev_id &&
+ scip->scip_prev_obsolete_sm_object != 0) {
+ space_map_t *prev_obsolete_sm = NULL;
+ VERIFY0(space_map_open(&prev_obsolete_sm, spa->spa_meta_objset,
+ scip->scip_prev_obsolete_sm_object, 0, vd->vdev_asize, 0));
+ space_map_update(prev_obsolete_sm);
+ vdev_indirect_mapping_load_obsolete_spacemap(vim, counts,
+ prev_obsolete_sm);
+ space_map_close(prev_obsolete_sm);
+ }
+ return (counts);
+}
+
static void
zdb_ddt_leak_init(spa_t *spa, zdb_cb_t *zcb)
{
- ddt_bookmark_t ddb = { 0 };
+ ddt_bookmark_t ddb;
ddt_entry_t dde;
int error;
int p;
+ bzero(&ddb, sizeof (ddb));
while ((error = ddt_walk(spa, &ddb, &dde)) == 0) {
blkptr_t blk;
ddt_phys_t *ddp = dde.dde_phys;
zdb_leak_init(spa_t *spa, zdb_cb_t *zcb)
{
zcb->zcb_spa = spa;
- uint64_t c, m;
+ uint64_t c;
if (!dump_opt['L']) {
+ dsl_pool_t *dp = spa->spa_dsl_pool;
vdev_t *rvd = spa->spa_root_vdev;
+
+ /*
+ * We are going to be changing the meaning of the metaslab's
+ * ms_tree. Ensure that the allocator doesn't try to
+ * use the tree.
+ */
+ spa->spa_normal_class->mc_ops = &zdb_metaslab_ops;
+ spa->spa_log_class->mc_ops = &zdb_metaslab_ops;
+
+ zcb->zcb_vd_obsolete_counts =
+ umem_zalloc(rvd->vdev_children * sizeof (uint32_t *),
+ UMEM_NOFAIL);
+
for (c = 0; c < rvd->vdev_children; c++) {
vdev_t *vd = rvd->vdev_child[c];
- for (m = 0; m < vd->vdev_ms_count; m++) {
- metaslab_t *msp = vd->vdev_ms[m];
- mutex_enter(&msp->ms_lock);
- metaslab_unload(msp);
+ uint64_t vim_idx = 0;
+
+ ASSERT3U(c, ==, vd->vdev_id);
+
+ /*
+ * Note: we don't check for mapping leaks on
+ * removing vdevs because their ms_tree's are
+ * used to look for leaks in allocated space.
+ */
+ if (vd->vdev_ops == &vdev_indirect_ops) {
+ zcb->zcb_vd_obsolete_counts[c] =
+ zdb_load_obsolete_counts(vd);
/*
- * For leak detection, we overload the metaslab
- * ms_tree to contain allocated segments
- * instead of free segments. As a result,
- * we can't use the normal metaslab_load/unload
- * interfaces.
+ * Normally, indirect vdevs don't have any
+ * metaslabs. We want to set them up for
+ * zio_claim().
*/
- if (msp->ms_sm != NULL) {
- (void) fprintf(stderr,
- "\rloading space map for "
- "vdev %llu of %llu, "
- "metaslab %llu of %llu ...",
- (longlong_t)c,
- (longlong_t)rvd->vdev_children,
- (longlong_t)m,
- (longlong_t)vd->vdev_ms_count);
-
- msp->ms_ops = &zdb_metaslab_ops;
- VERIFY0(space_map_load(msp->ms_sm,
- msp->ms_tree, SM_ALLOC));
- msp->ms_loaded = B_TRUE;
- }
- mutex_exit(&msp->ms_lock);
+ VERIFY0(vdev_metaslab_init(vd, 0));
+ }
+
+ for (uint64_t m = 0; m < vd->vdev_ms_count; m++) {
+ zdb_leak_init_ms(vd->vdev_ms[m], &vim_idx);
+ }
+ if (vd->vdev_ops == &vdev_indirect_ops) {
+ ASSERT3U(vim_idx, ==,
+ vdev_indirect_mapping_num_entries(
+ vd->vdev_indirect_mapping));
}
}
(void) fprintf(stderr, "\n");
+
+ if (bpobj_is_open(&dp->dp_obsolete_bpobj)) {
+ ASSERT(spa_feature_is_enabled(spa,
+ SPA_FEATURE_DEVICE_REMOVAL));
+ (void) bpobj_iterate_nofree(&dp->dp_obsolete_bpobj,
+ increment_indirect_mapping_cb, zcb, NULL);
+ }
}
spa_config_enter(spa, SCL_CONFIG, FTAG, RW_READER);
spa_config_exit(spa, SCL_CONFIG, FTAG);
}
-static void
-zdb_leak_fini(spa_t *spa)
+static boolean_t
+zdb_check_for_obsolete_leaks(vdev_t *vd, zdb_cb_t *zcb)
{
- int c, m;
+ boolean_t leaks = B_FALSE;
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+ uint64_t total_leaked = 0;
+
+ ASSERT(vim != NULL);
+
+ for (uint64_t i = 0; i < vdev_indirect_mapping_num_entries(vim); i++) {
+ vdev_indirect_mapping_entry_phys_t *vimep =
+ &vim->vim_entries[i];
+ uint64_t obsolete_bytes = 0;
+ uint64_t offset = DVA_MAPPING_GET_SRC_OFFSET(vimep);
+ metaslab_t *msp = vd->vdev_ms[offset >> vd->vdev_ms_shift];
+
+ /*
+ * This is not very efficient but it's easy to
+ * verify correctness.
+ */
+ for (uint64_t inner_offset = 0;
+ inner_offset < DVA_GET_ASIZE(&vimep->vimep_dst);
+ inner_offset += 1 << vd->vdev_ashift) {
+ if (range_tree_contains(msp->ms_tree,
+ offset + inner_offset, 1 << vd->vdev_ashift)) {
+ obsolete_bytes += 1 << vd->vdev_ashift;
+ }
+ }
+
+ int64_t bytes_leaked = obsolete_bytes -
+ zcb->zcb_vd_obsolete_counts[vd->vdev_id][i];
+ ASSERT3U(DVA_GET_ASIZE(&vimep->vimep_dst), >=,
+ zcb->zcb_vd_obsolete_counts[vd->vdev_id][i]);
+ if (bytes_leaked != 0 &&
+ (vdev_obsolete_counts_are_precise(vd) ||
+ dump_opt['d'] >= 5)) {
+ (void) printf("obsolete indirect mapping count "
+ "mismatch on %llu:%llx:%llx : %llx bytes leaked\n",
+ (u_longlong_t)vd->vdev_id,
+ (u_longlong_t)DVA_MAPPING_GET_SRC_OFFSET(vimep),
+ (u_longlong_t)DVA_GET_ASIZE(&vimep->vimep_dst),
+ (u_longlong_t)bytes_leaked);
+ }
+ total_leaked += ABS(bytes_leaked);
+ }
+
+ if (!vdev_obsolete_counts_are_precise(vd) && total_leaked > 0) {
+ int pct_leaked = total_leaked * 100 /
+ vdev_indirect_mapping_bytes_mapped(vim);
+ (void) printf("cannot verify obsolete indirect mapping "
+ "counts of vdev %llu because precise feature was not "
+ "enabled when it was removed: %d%% (%llx bytes) of mapping"
+ "unreferenced\n",
+ (u_longlong_t)vd->vdev_id, pct_leaked,
+ (u_longlong_t)total_leaked);
+ } else if (total_leaked > 0) {
+ (void) printf("obsolete indirect mapping count mismatch "
+ "for vdev %llu -- %llx total bytes mismatched\n",
+ (u_longlong_t)vd->vdev_id,
+ (u_longlong_t)total_leaked);
+ leaks |= B_TRUE;
+ }
+
+ vdev_indirect_mapping_free_obsolete_counts(vim,
+ zcb->zcb_vd_obsolete_counts[vd->vdev_id]);
+ zcb->zcb_vd_obsolete_counts[vd->vdev_id] = NULL;
+
+ return (leaks);
+}
+static boolean_t
+zdb_leak_fini(spa_t *spa, zdb_cb_t *zcb)
+{
+ boolean_t leaks = B_FALSE;
if (!dump_opt['L']) {
vdev_t *rvd = spa->spa_root_vdev;
- for (c = 0; c < rvd->vdev_children; c++) {
+ for (unsigned c = 0; c < rvd->vdev_children; c++) {
vdev_t *vd = rvd->vdev_child[c];
- for (m = 0; m < vd->vdev_ms_count; m++) {
+ ASSERTV(metaslab_group_t *mg = vd->vdev_mg);
+
+ if (zcb->zcb_vd_obsolete_counts[c] != NULL) {
+ leaks |= zdb_check_for_obsolete_leaks(vd, zcb);
+ }
+
+ for (uint64_t m = 0; m < vd->vdev_ms_count; m++) {
metaslab_t *msp = vd->vdev_ms[m];
- mutex_enter(&msp->ms_lock);
+ ASSERT3P(mg, ==, msp->ms_group);
/*
* The ms_tree has been overloaded to
* represents an allocated block that we
* did not claim during the traversal.
* Claimed blocks would have been removed
- * from the ms_tree.
+ * from the ms_tree. For indirect vdevs,
+ * space remaining in the tree represents
+ * parts of the mapping that are not
+ * referenced, which is not a bug.
*/
- range_tree_vacate(msp->ms_tree, zdb_leak, vd);
- msp->ms_loaded = B_FALSE;
+ if (vd->vdev_ops == &vdev_indirect_ops) {
+ range_tree_vacate(msp->ms_tree,
+ NULL, NULL);
+ } else {
+ range_tree_vacate(msp->ms_tree,
+ zdb_leak, vd);
+ }
- mutex_exit(&msp->ms_lock);
+ if (msp->ms_loaded)
+ msp->ms_loaded = B_FALSE;
}
}
+
+ umem_free(zcb->zcb_vd_obsolete_counts,
+ rvd->vdev_children * sizeof (uint32_t *));
+ zcb->zcb_vd_obsolete_counts = NULL;
}
+ return (leaks);
}
/* ARGSUSED */
zdb_cb_t zcb;
zdb_blkstats_t *zb, *tzb;
uint64_t norm_alloc, norm_space, total_alloc, total_found;
- int flags = TRAVERSE_PRE | TRAVERSE_PREFETCH_METADATA | TRAVERSE_HARD;
+ int flags = TRAVERSE_PRE | TRAVERSE_PREFETCH_METADATA |
+ TRAVERSE_NO_DECRYPT | TRAVERSE_HARD;
boolean_t leaks = B_FALSE;
- int e, c;
+ int e, c, err;
bp_embedded_type_t i;
+ bzero(&zcb, sizeof (zcb));
(void) printf("\nTraversing all blocks %s%s%s%s%s...\n\n",
(dump_opt['c'] || !dump_opt['L']) ? "to verify " : "",
(dump_opt['c'] == 1) ? "metadata " : "",
*/
(void) bpobj_iterate_nofree(&spa->spa_deferred_bpobj,
count_block_cb, &zcb, NULL);
+
if (spa_version(spa) >= SPA_VERSION_DEADLISTS) {
(void) bpobj_iterate_nofree(&spa->spa_dsl_pool->dp_free_bpobj,
count_block_cb, &zcb, NULL);
}
+
+ zdb_claim_removing(spa, &zcb);
+
if (spa_feature_is_active(spa, SPA_FEATURE_ASYNC_DESTROY)) {
VERIFY3U(0, ==, bptree_iterate(spa->spa_meta_objset,
spa->spa_dsl_pool->dp_bptree_obj, B_FALSE, count_block_cb,
zcb.zcb_totalasize = metaslab_class_get_alloc(spa_normal_class(spa));
zcb.zcb_start = zcb.zcb_lastprint = gethrtime();
- zcb.zcb_haderrors |= traverse_pool(spa, 0, flags, zdb_blkptr_cb, &zcb);
+ err = traverse_pool(spa, 0, flags, zdb_blkptr_cb, &zcb);
/*
* If we've traversed the data blocks then we need to wait for those
}
}
+ /*
+ * Done after zio_wait() since zcb_haderrors is modified in
+ * zdb_blkptr_done()
+ */
+ zcb.zcb_haderrors |= err;
+
if (zcb.zcb_haderrors) {
(void) printf("\nError counts:\n\n");
(void) printf("\t%5s %s\n", "errno", "count");
/*
* Report any leaked segments.
*/
- zdb_leak_fini(spa);
+ leaks |= zdb_leak_fini(spa, &zcb);
tzb = &zcb.zcb_type[ZB_TOTAL][ZDB_OT_TOTAL];
norm_space = metaslab_class_get_space(spa_normal_class(spa));
total_alloc = norm_alloc + metaslab_class_get_alloc(spa_log_class(spa));
- total_found = tzb->zb_asize - zcb.zcb_dedup_asize;
+ total_found = tzb->zb_asize - zcb.zcb_dedup_asize +
+ zcb.zcb_removing_size;
if (total_found == total_alloc) {
if (!dump_opt['L'])
(longlong_t)tzb->zb_ditto_samevdev);
}
+ for (uint64_t v = 0; v < spa->spa_root_vdev->vdev_children; v++) {
+ vdev_t *vd = spa->spa_root_vdev->vdev_child[v];
+ vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping;
+
+ if (vim == NULL) {
+ continue;
+ }
+
+ char mem[32];
+ zdb_nicenum(vdev_indirect_mapping_num_entries(vim),
+ mem, vdev_indirect_mapping_size(vim));
+
+ (void) printf("\tindirect vdev id %llu has %llu segments "
+ "(%s in memory)\n",
+ (longlong_t)vd->vdev_id,
+ (longlong_t)vdev_indirect_mapping_num_entries(vim), mem);
+ }
+
if (dump_opt['b'] >= 2) {
int l, t, level;
(void) printf("\nBlocks\tLSIZE\tPSIZE\tASIZE"
for (t = 0; t <= ZDB_OT_TOTAL; t++) {
char csize[32], lsize[32], psize[32], asize[32];
char avg[32], gang[32];
- char *typename;
+ const char *typename;
+
+ /* make sure nicenum has enough space */
+ CTASSERT(sizeof (csize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (lsize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (psize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (asize) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (avg) >= NN_NUMBUF_SZ);
+ CTASSERT(sizeof (gang) >= NN_NUMBUF_SZ);
if (t < DMU_OT_NUMTYPES)
typename = dmu_ot[t].ot_name;
zcb.zcb_type[ZB_TOTAL][t].zb_asize)
continue;
- zdb_nicenum(zb->zb_count, csize);
- zdb_nicenum(zb->zb_lsize, lsize);
- zdb_nicenum(zb->zb_psize, psize);
- zdb_nicenum(zb->zb_asize, asize);
- zdb_nicenum(zb->zb_asize / zb->zb_count, avg);
- zdb_nicenum(zb->zb_gangs, gang);
+ zdb_nicenum(zb->zb_count, csize,
+ sizeof (csize));
+ zdb_nicenum(zb->zb_lsize, lsize,
+ sizeof (lsize));
+ zdb_nicenum(zb->zb_psize, psize,
+ sizeof (psize));
+ zdb_nicenum(zb->zb_asize, asize,
+ sizeof (asize));
+ zdb_nicenum(zb->zb_asize / zb->zb_count, avg,
+ sizeof (avg));
+ zdb_nicenum(zb->zb_gangs, gang, sizeof (gang));
(void) printf("%6s\t%5s\t%5s\t%5s\t%5s"
"\t%5.2f\t%6.2f\t",
avl_index_t where;
zdb_ddt_entry_t *zdde, zdde_search;
- if (BP_IS_HOLE(bp) || BP_IS_EMBEDDED(bp))
+ if (bp == NULL || BP_IS_HOLE(bp) || BP_IS_EMBEDDED(bp))
return (0);
if (dump_opt['S'] > 1 && zb->zb_level == ZB_ROOT_LEVEL) {
ddt_histogram_t ddh_total;
ddt_stat_t dds_total;
- bzero(&ddh_total, sizeof (ddt_histogram_t));
- bzero(&dds_total, sizeof (ddt_stat_t));
-
+ bzero(&ddh_total, sizeof (ddh_total));
+ bzero(&dds_total, sizeof (dds_total));
avl_create(&t, ddt_entry_compare,
sizeof (zdb_ddt_entry_t), offsetof(zdb_ddt_entry_t, zdde_node));
spa_config_enter(spa, SCL_CONFIG, FTAG, RW_READER);
- (void) traverse_pool(spa, 0, TRAVERSE_PRE | TRAVERSE_PREFETCH_METADATA,
- zdb_ddt_add_cb, &t);
+ (void) traverse_pool(spa, 0, TRAVERSE_PRE | TRAVERSE_PREFETCH_METADATA |
+ TRAVERSE_NO_DECRYPT, zdb_ddt_add_cb, &t);
spa_config_exit(spa, SCL_CONFIG, FTAG);
dump_dedup_ratio(&dds_total);
}
+static int
+verify_device_removal_feature_counts(spa_t *spa)
+{
+ uint64_t dr_feature_refcount = 0;
+ uint64_t oc_feature_refcount = 0;
+ uint64_t indirect_vdev_count = 0;
+ uint64_t precise_vdev_count = 0;
+ uint64_t obsolete_counts_object_count = 0;
+ uint64_t obsolete_sm_count = 0;
+ uint64_t obsolete_counts_count = 0;
+ uint64_t scip_count = 0;
+ uint64_t obsolete_bpobj_count = 0;
+ int ret = 0;
+
+ spa_condensing_indirect_phys_t *scip =
+ &spa->spa_condensing_indirect_phys;
+ if (scip->scip_next_mapping_object != 0) {
+ vdev_t *vd = spa->spa_root_vdev->vdev_child[scip->scip_vdev];
+ ASSERT(scip->scip_prev_obsolete_sm_object != 0);
+ ASSERT3P(vd->vdev_ops, ==, &vdev_indirect_ops);
+
+ (void) printf("Condensing indirect vdev %llu: new mapping "
+ "object %llu, prev obsolete sm %llu\n",
+ (u_longlong_t)scip->scip_vdev,
+ (u_longlong_t)scip->scip_next_mapping_object,
+ (u_longlong_t)scip->scip_prev_obsolete_sm_object);
+ if (scip->scip_prev_obsolete_sm_object != 0) {
+ space_map_t *prev_obsolete_sm = NULL;
+ VERIFY0(space_map_open(&prev_obsolete_sm,
+ spa->spa_meta_objset,
+ scip->scip_prev_obsolete_sm_object,
+ 0, vd->vdev_asize, 0));
+ space_map_update(prev_obsolete_sm);
+ dump_spacemap(spa->spa_meta_objset, prev_obsolete_sm);
+ (void) printf("\n");
+ space_map_close(prev_obsolete_sm);
+ }
+
+ scip_count += 2;
+ }
+
+ for (uint64_t i = 0; i < spa->spa_root_vdev->vdev_children; i++) {
+ vdev_t *vd = spa->spa_root_vdev->vdev_child[i];
+ vdev_indirect_config_t *vic = &vd->vdev_indirect_config;
+
+ if (vic->vic_mapping_object != 0) {
+ ASSERT(vd->vdev_ops == &vdev_indirect_ops ||
+ vd->vdev_removing);
+ indirect_vdev_count++;
+
+ if (vd->vdev_indirect_mapping->vim_havecounts) {
+ obsolete_counts_count++;
+ }
+ }
+ if (vdev_obsolete_counts_are_precise(vd)) {
+ ASSERT(vic->vic_mapping_object != 0);
+ precise_vdev_count++;
+ }
+ if (vdev_obsolete_sm_object(vd) != 0) {
+ ASSERT(vic->vic_mapping_object != 0);
+ obsolete_sm_count++;
+ }
+ }
+
+ (void) feature_get_refcount(spa,
+ &spa_feature_table[SPA_FEATURE_DEVICE_REMOVAL],
+ &dr_feature_refcount);
+ (void) feature_get_refcount(spa,
+ &spa_feature_table[SPA_FEATURE_OBSOLETE_COUNTS],
+ &oc_feature_refcount);
+
+ if (dr_feature_refcount != indirect_vdev_count) {
+ ret = 1;
+ (void) printf("Number of indirect vdevs (%llu) " \
+ "does not match feature count (%llu)\n",
+ (u_longlong_t)indirect_vdev_count,
+ (u_longlong_t)dr_feature_refcount);
+ } else {
+ (void) printf("Verified device_removal feature refcount " \
+ "of %llu is correct\n",
+ (u_longlong_t)dr_feature_refcount);
+ }
+
+ if (zap_contains(spa_meta_objset(spa), DMU_POOL_DIRECTORY_OBJECT,
+ DMU_POOL_OBSOLETE_BPOBJ) == 0) {
+ obsolete_bpobj_count++;
+ }
+
+
+ obsolete_counts_object_count = precise_vdev_count;
+ obsolete_counts_object_count += obsolete_sm_count;
+ obsolete_counts_object_count += obsolete_counts_count;
+ obsolete_counts_object_count += scip_count;
+ obsolete_counts_object_count += obsolete_bpobj_count;
+ obsolete_counts_object_count += remap_deadlist_count;
+
+ if (oc_feature_refcount != obsolete_counts_object_count) {
+ ret = 1;
+ (void) printf("Number of obsolete counts objects (%llu) " \
+ "does not match feature count (%llu)\n",
+ (u_longlong_t)obsolete_counts_object_count,
+ (u_longlong_t)oc_feature_refcount);
+ (void) printf("pv:%llu os:%llu oc:%llu sc:%llu "
+ "ob:%llu rd:%llu\n",
+ (u_longlong_t)precise_vdev_count,
+ (u_longlong_t)obsolete_sm_count,
+ (u_longlong_t)obsolete_counts_count,
+ (u_longlong_t)scip_count,
+ (u_longlong_t)obsolete_bpobj_count,
+ (u_longlong_t)remap_deadlist_count);
+ } else {
+ (void) printf("Verified indirect_refcount feature refcount " \
+ "of %llu is correct\n",
+ (u_longlong_t)oc_feature_refcount);
+ }
+ return (ret);
+}
+
static void
dump_zpool(spa_t *spa)
{
dump_metaslab_groups(spa);
if (dump_opt['d'] || dump_opt['i']) {
- uint64_t refcount;
+ spa_feature_t f;
+
dump_dir(dp->dp_meta_objset);
if (dump_opt['d'] >= 3) {
+ dsl_pool_t *dp = spa->spa_dsl_pool;
dump_full_bpobj(&spa->spa_deferred_bpobj,
"Deferred frees", 0);
if (spa_version(spa) >= SPA_VERSION_DEADLISTS) {
- dump_full_bpobj(
- &spa->spa_dsl_pool->dp_free_bpobj,
+ dump_full_bpobj(&dp->dp_free_bpobj,
"Pool snapshot frees", 0);
}
+ if (bpobj_is_open(&dp->dp_obsolete_bpobj)) {
+ ASSERT(spa_feature_is_enabled(spa,
+ SPA_FEATURE_DEVICE_REMOVAL));
+ dump_full_bpobj(&dp->dp_obsolete_bpobj,
+ "Pool obsolete blocks", 0);
+ }
if (spa_feature_is_active(spa,
SPA_FEATURE_ASYNC_DESTROY)) {
dump_bptree(spa->spa_meta_objset,
- spa->spa_dsl_pool->dp_bptree_obj,
+ dp->dp_bptree_obj,
"Pool dataset frees");
}
dump_dtl(spa->spa_root_vdev, 0);
(void) dmu_objset_find(spa_name(spa), dump_one_dir,
NULL, DS_FIND_SNAPSHOTS | DS_FIND_CHILDREN);
- (void) feature_get_refcount(spa,
- &spa_feature_table[SPA_FEATURE_LARGE_BLOCKS], &refcount);
- if (num_large_blocks != refcount) {
- (void) printf("large_blocks feature refcount mismatch: "
- "expected %lld != actual %lld\n",
- (longlong_t)num_large_blocks,
- (longlong_t)refcount);
- rc = 2;
- } else {
- (void) printf("Verified large_blocks feature refcount "
- "is correct (%llu)\n", (longlong_t)refcount);
+ for (f = 0; f < SPA_FEATURES; f++) {
+ uint64_t refcount;
+
+ if (!(spa_feature_table[f].fi_flags &
+ ZFEATURE_FLAG_PER_DATASET) ||
+ !spa_feature_is_enabled(spa, f)) {
+ ASSERT0(dataset_feature_count[f]);
+ continue;
+ }
+ if (feature_get_refcount(spa, &spa_feature_table[f],
+ &refcount) == ENOTSUP)
+ continue;
+ if (dataset_feature_count[f] != refcount) {
+ (void) printf("%s feature refcount mismatch: "
+ "%lld datasets != %lld refcount\n",
+ spa_feature_table[f].fi_uname,
+ (longlong_t)dataset_feature_count[f],
+ (longlong_t)refcount);
+ rc = 2;
+ } else {
+ (void) printf("Verified %s feature refcount "
+ "of %llu is correct\n",
+ spa_feature_table[f].fi_uname,
+ (longlong_t)refcount);
+ }
+ }
+
+ if (rc == 0) {
+ rc = verify_device_removal_feature_counts(spa);
}
}
if (rc == 0 && (dump_opt['b'] || dump_opt['c']))
if (dump_opt['h'])
dump_history(spa);
- if (rc != 0)
+ if (rc != 0) {
+ dump_debug_buffer();
exit(rc);
+ }
}
#define ZDB_FLAG_CHECKSUM 0x0001
#define ZDB_FLAG_RAW 0x0040
#define ZDB_FLAG_PRINT_BLKPTR 0x0080
-int flagbits[256];
+static int flagbits[256];
static void
zdb_print_blkptr(blkptr_t *bp, int flags)
zdb_dump_block(char *label, void *buf, uint64_t size, int flags)
{
uint64_t *d = (uint64_t *)buf;
- int nwords = size / sizeof (uint64_t);
+ unsigned nwords = size / sizeof (uint64_t);
int do_bswap = !!(flags & ZDB_FLAG_BSWAP);
- int i, j;
- char *hdr, *c;
+ unsigned i, j;
+ const char *hdr;
+ char *c;
if (do_bswap)
(void) printf("\n%s\n%6s %s 0123456789abcdef\n", label, "", hdr);
+#ifdef _LITTLE_ENDIAN
+ /* correct the endianness */
+ do_bswap = !do_bswap;
+#endif
for (i = 0; i < nwords; i += 2) {
(void) printf("%06llx: %016llx %016llx ",
(u_longlong_t)(i * sizeof (uint64_t)),
* child[.child]* - For example: 0.1.1
*
* The second form can be used to specify arbitrary vdevs anywhere
- * in the heirarchy. For example, in a pool with a mirror of
+ * in the hierarchy. For example, in a pool with a mirror of
* RAID-Zs, you can specify either RAID-Z vdev with 0.0 or 0.1 .
*/
static vdev_t *
-zdb_vdev_lookup(vdev_t *vdev, char *path)
+zdb_vdev_lookup(vdev_t *vdev, const char *path)
{
char *s, *p, *q;
- int i;
+ unsigned i;
if (vdev == NULL)
return (NULL);
/* First, assume the x.x.x.x format */
- i = (int)strtoul(path, &s, 10);
+ i = strtoul(path, &s, 10);
if (s == path || (s && *s != '.' && *s != '\0'))
goto name;
- if (i < 0 || i >= vdev->vdev_children)
+ if (i >= vdev->vdev_children)
return (NULL);
vdev = vdev->vdev_child[i];
- if (*s == '\0')
+ if (s && *s == '\0')
return (vdev);
return (zdb_vdev_lookup(vdev, s+1));
uint64_t offset = 0, size = 0, psize = 0, lsize = 0, blkptr_offset = 0;
zio_t *zio;
vdev_t *vd;
- void *pbuf, *lbuf, *buf;
- char *s, *p, *dup, *vdev, *flagstr;
+ abd_t *pabd;
+ void *lbuf, *buf;
+ const char *s, *vdev;
+ char *p, *dup, *flagstr;
int i, error;
+ boolean_t borrowed = B_FALSE;
dup = strdup(thing);
s = strtok(dup, ":");
s = strtok(NULL, ":");
size = strtoull(s ? s : "", NULL, 16);
s = strtok(NULL, ":");
- flagstr = s ? s : "";
+ if (s)
+ flagstr = strdup(s);
+ else
+ flagstr = strdup("");
s = NULL;
if (size == 0)
s = "offset must be a multiple of sector size";
if (s) {
(void) printf("Invalid block specifier: %s - %s\n", thing, s);
+ free(flagstr);
free(dup);
return;
}
continue;
p = &flagstr[i + 1];
- if (bit == ZDB_FLAG_PRINT_BLKPTR)
+ if (bit == ZDB_FLAG_PRINT_BLKPTR) {
blkptr_offset = strtoull(p, &p, 16);
+ i = p - &flagstr[i + 1];
+ }
if (*p != ':' && *p != '\0') {
(void) printf("***Invalid flag arg: '%s'\n", s);
+ free(flagstr);
free(dup);
return;
}
}
}
+ free(flagstr);
vd = zdb_vdev_lookup(spa->spa_root_vdev, vdev);
if (vd == NULL) {
psize = size;
lsize = size;
- pbuf = umem_alloc_aligned(SPA_MAXBLOCKSIZE, 512, UMEM_NOFAIL);
+ pabd = abd_alloc_for_io(SPA_MAXBLOCKSIZE, B_FALSE);
lbuf = umem_alloc(SPA_MAXBLOCKSIZE, UMEM_NOFAIL);
BP_ZERO(bp);
/*
* Treat this as a normal block read.
*/
- zio_nowait(zio_read(zio, spa, bp, pbuf, psize, NULL, NULL,
+ zio_nowait(zio_read(zio, spa, bp, pabd, psize, NULL, NULL,
ZIO_PRIORITY_SYNC_READ,
ZIO_FLAG_CANFAIL | ZIO_FLAG_RAW, NULL));
} else {
/*
* Treat this as a vdev child I/O.
*/
- zio_nowait(zio_vdev_child_io(zio, bp, vd, offset, pbuf, psize,
- ZIO_TYPE_READ, ZIO_PRIORITY_SYNC_READ,
+ zio_nowait(zio_vdev_child_io(zio, bp, vd, offset, pabd,
+ psize, ZIO_TYPE_READ, ZIO_PRIORITY_SYNC_READ,
ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_QUEUE |
ZIO_FLAG_DONT_PROPAGATE | ZIO_FLAG_DONT_RETRY |
- ZIO_FLAG_CANFAIL | ZIO_FLAG_RAW, NULL, NULL));
+ ZIO_FLAG_CANFAIL | ZIO_FLAG_RAW | ZIO_FLAG_OPTIONAL,
+ NULL, NULL));
}
error = zio_wait(zio);
* every decompress function at every inflated blocksize.
*/
enum zio_compress c;
- void *pbuf2 = umem_alloc(SPA_MAXBLOCKSIZE, UMEM_NOFAIL);
void *lbuf2 = umem_alloc(SPA_MAXBLOCKSIZE, UMEM_NOFAIL);
- bcopy(pbuf, pbuf2, psize);
+ /*
+ * XXX - On the one hand, with SPA_MAXBLOCKSIZE at 16MB,
+ * this could take a while and we should let the user know
+ * we are not stuck. On the other hand, printing progress
+ * info gets old after a while. What to do?
+ */
+ for (lsize = psize + SPA_MINBLOCKSIZE;
+ lsize <= SPA_MAXBLOCKSIZE; lsize += SPA_MINBLOCKSIZE) {
+ for (c = 0; c < ZIO_COMPRESS_FUNCTIONS; c++) {
+ /*
+ * ZLE can easily decompress non zle stream.
+ * So have an option to disable it.
+ */
+ if (c == ZIO_COMPRESS_ZLE &&
+ getenv("ZDB_NO_ZLE"))
+ continue;
- VERIFY(random_get_pseudo_bytes((uint8_t *)pbuf + psize,
- SPA_MAXBLOCKSIZE - psize) == 0);
+ (void) fprintf(stderr,
+ "Trying %05llx -> %05llx (%s)\n",
+ (u_longlong_t)psize, (u_longlong_t)lsize,
+ zio_compress_table[c].ci_name);
- VERIFY(random_get_pseudo_bytes((uint8_t *)pbuf2 + psize,
- SPA_MAXBLOCKSIZE - psize) == 0);
+ /*
+ * We randomize lbuf2, and decompress to both
+ * lbuf and lbuf2. This way, we will know if
+ * decompression fill exactly to lsize.
+ */
+ VERIFY0(random_get_pseudo_bytes(lbuf2, lsize));
- for (lsize = SPA_MAXBLOCKSIZE; lsize > psize;
- lsize -= SPA_MINBLOCKSIZE) {
- for (c = 0; c < ZIO_COMPRESS_FUNCTIONS; c++) {
- if (zio_decompress_data(c, pbuf, lbuf,
- psize, lsize) == 0 &&
- zio_decompress_data(c, pbuf2, lbuf2,
- psize, lsize) == 0 &&
+ if (zio_decompress_data(c, pabd,
+ lbuf, psize, lsize) == 0 &&
+ zio_decompress_data(c, pabd,
+ lbuf2, psize, lsize) == 0 &&
bcmp(lbuf, lbuf2, lsize) == 0)
break;
}
if (c != ZIO_COMPRESS_FUNCTIONS)
break;
- lsize -= SPA_MINBLOCKSIZE;
}
-
- umem_free(pbuf2, SPA_MAXBLOCKSIZE);
umem_free(lbuf2, SPA_MAXBLOCKSIZE);
- if (lsize <= psize) {
+ if (lsize > SPA_MAXBLOCKSIZE) {
(void) printf("Decompress of %s failed\n", thing);
goto out;
}
buf = lbuf;
size = lsize;
} else {
- buf = pbuf;
size = psize;
+ buf = abd_borrow_buf_copy(pabd, size);
+ borrowed = B_TRUE;
}
if (flags & ZDB_FLAG_PRINT_BLKPTR)
else
zdb_dump_block(thing, buf, size, flags);
+ if (borrowed)
+ abd_return_buf_copy(pabd, buf, size);
+
out:
- umem_free(pbuf, SPA_MAXBLOCKSIZE);
+ abd_free(pabd);
umem_free(lbuf, SPA_MAXBLOCKSIZE);
free(dup);
}
-static boolean_t
-pool_match(nvlist_t *cfg, char *tgt)
+static void
+zdb_embedded_block(char *thing)
{
- uint64_t v, guid = strtoull(tgt, NULL, 0);
- char *s;
-
- if (guid != 0) {
- if (nvlist_lookup_uint64(cfg, ZPOOL_CONFIG_POOL_GUID, &v) == 0)
- return (v == guid);
- } else {
- if (nvlist_lookup_string(cfg, ZPOOL_CONFIG_POOL_NAME, &s) == 0)
- return (strcmp(s, tgt) == 0);
- }
- return (B_FALSE);
-}
-
-static char *
-find_zpool(char **target, nvlist_t **configp, int dirc, char **dirv)
-{
- nvlist_t *pools;
- nvlist_t *match = NULL;
- char *name = NULL;
- char *sepp = NULL;
- char sep = 0;
- int count = 0;
- importargs_t args = { 0 };
-
- args.paths = dirc;
- args.path = dirv;
- args.can_be_active = B_TRUE;
-
- if ((sepp = strpbrk(*target, "/@")) != NULL) {
- sep = *sepp;
- *sepp = '\0';
- }
-
- pools = zpool_search_import(g_zfs, &args);
-
- if (pools != NULL) {
- nvpair_t *elem = NULL;
- while ((elem = nvlist_next_nvpair(pools, elem)) != NULL) {
- verify(nvpair_value_nvlist(elem, configp) == 0);
- if (pool_match(*configp, *target)) {
- count++;
- if (match != NULL) {
- /* print previously found config */
- if (name != NULL) {
- (void) printf("%s\n", name);
- dump_nvlist(match, 8);
- name = NULL;
- }
- (void) printf("%s\n",
- nvpair_name(elem));
- dump_nvlist(*configp, 8);
- } else {
- match = *configp;
- name = nvpair_name(elem);
- }
- }
- }
+ blkptr_t bp;
+ unsigned long long *words = (void *)&bp;
+ char *buf;
+ int err;
+
+ buf = umem_alloc(SPA_MAXBLOCKSIZE, UMEM_NOFAIL);
+
+ bzero(&bp, sizeof (bp));
+ err = sscanf(thing, "%llx:%llx:%llx:%llx:%llx:%llx:%llx:%llx:"
+ "%llx:%llx:%llx:%llx:%llx:%llx:%llx:%llx",
+ words + 0, words + 1, words + 2, words + 3,
+ words + 4, words + 5, words + 6, words + 7,
+ words + 8, words + 9, words + 10, words + 11,
+ words + 12, words + 13, words + 14, words + 15);
+ if (err != 16) {
+ (void) printf("invalid input format\n");
+ exit(1);
}
- if (count > 1)
- (void) fatal("\tMatched %d pools - use pool GUID "
- "instead of pool name or \n"
- "\tpool name part of a dataset name to select pool", count);
-
- if (sepp)
- *sepp = sep;
- /*
- * If pool GUID was specified for pool id, replace it with pool name
- */
- if (name && (strstr(*target, name) != *target)) {
- int sz = 1 + strlen(name) + ((sepp) ? strlen(sepp) : 0);
-
- *target = umem_alloc(sz, UMEM_NOFAIL);
- (void) snprintf(*target, sz, "%s%s", name, sepp ? sepp : "");
+ ASSERT3U(BPE_GET_LSIZE(&bp), <=, SPA_MAXBLOCKSIZE);
+ err = decode_embedded_bp(&bp, buf, BPE_GET_LSIZE(&bp));
+ if (err != 0) {
+ (void) printf("decode failed: %u\n", err);
+ exit(1);
}
-
- *configp = name ? match : NULL;
-
- return (name);
+ zdb_dump_block_raw(buf, BPE_GET_LSIZE(&bp), 0);
+ umem_free(buf, SPA_MAXBLOCKSIZE);
}
int
main(int argc, char **argv)
{
- int i, c;
+ int c;
struct rlimit rl = { 1024, 1024 };
spa_t *spa = NULL;
objset_t *os = NULL;
int error = 0;
char **searchdirs = NULL;
int nsearch = 0;
- char *target;
+ char *target, *target_pool;
nvlist_t *policy = NULL;
uint64_t max_txg = UINT64_MAX;
int flags = ZFS_IMPORT_MISSING_LOG;
int rewind = ZPOOL_NEVER_REWIND;
char *spa_config_path_env;
- const char *opts = "bcdhilmMI:suCDRSAFLXevp:t:U:P";
boolean_t target_is_spa = B_TRUE;
(void) setrlimit(RLIMIT_NOFILE, &rl);
if (spa_config_path_env != NULL)
spa_config_path = spa_config_path_env;
- while ((c = getopt(argc, argv, opts)) != -1) {
+ while ((c = getopt(argc, argv,
+ "AbcCdDeEFGhiI:lLmMo:Op:PqRsSt:uU:vVx:X")) != -1) {
switch (c) {
case 'b':
case 'c':
+ case 'C':
case 'd':
+ case 'D':
+ case 'E':
+ case 'G':
case 'h':
case 'i':
case 'l':
case 'm':
- case 's':
- case 'u':
- case 'C':
- case 'D':
case 'M':
+ case 'O':
case 'R':
+ case 's':
case 'S':
+ case 'u':
dump_opt[c]++;
dump_all = 0;
break;
case 'A':
+ case 'e':
case 'F':
case 'L':
- case 'X':
- case 'e':
case 'P':
+ case 'q':
+ case 'X':
dump_opt[c]++;
break;
- case 'V':
- flags = ZFS_IMPORT_VERBATIM;
- break;
+ /* NB: Sort single match options below. */
case 'I':
max_inflight = strtoull(optarg, NULL, 0);
if (max_inflight == 0) {
usage();
}
break;
+ case 'o':
+ error = set_global_var(optarg);
+ if (error != 0)
+ usage();
+ break;
case 'p':
if (searchdirs == NULL) {
searchdirs = umem_alloc(sizeof (char *),
break;
case 'U':
spa_config_path = optarg;
+ if (spa_config_path[0] != '/') {
+ (void) fprintf(stderr,
+ "cachefile must be an absolute path "
+ "(i.e. start with a slash)\n");
+ usage();
+ }
break;
case 'v':
verbose++;
break;
+ case 'V':
+ flags = ZFS_IMPORT_VERBATIM;
+ break;
+ case 'x':
+ vn_dumpdir = optarg;
+ break;
default:
usage();
break;
zfs_arc_max = zfs_arc_meta_limit = 256 * 1024 * 1024;
#endif
+ /*
+ * "zdb -c" uses checksum-verifying scrub i/os which are async reads.
+ * "zdb -b" uses traversal prefetch which uses async reads.
+ * For good performance, let several of them be active at once.
+ */
+ zfs_vdev_async_read_max_active = 10;
+
+ /*
+ * Disable reference tracking for better performance.
+ */
+ reference_tracking_enable = B_FALSE;
+
kernel_init(FREAD);
- if ((g_zfs = libzfs_init()) == NULL)
+ if ((g_zfs = libzfs_init()) == NULL) {
+ (void) fprintf(stderr, "%s", libzfs_error_init(errno));
return (1);
+ }
if (dump_all)
verbose = MAX(verbose, 1);
for (c = 0; c < 256; c++) {
- if (dump_all && !strchr("elAFLRSXP", c))
+ if (dump_all && strchr("AeEFlLOPRSX", c) == NULL)
dump_opt[c] = 1;
if (dump_opt[c])
dump_opt[c] += verbose;
if (argc < 2 && dump_opt['R'])
usage();
+
+ if (dump_opt['E']) {
+ if (argc != 1)
+ usage();
+ zdb_embedded_block(argv[0]);
+ return (0);
+ }
+
if (argc < 1) {
if (!dump_opt['e'] && dump_opt['C']) {
dump_cachefile(spa_config_path);
usage();
}
- if (dump_opt['l']) {
- dump_label(argv[0]);
- return (0);
+ if (dump_opt['l'])
+ return (dump_label(argv[0]));
+
+ if (dump_opt['O']) {
+ if (argc != 2)
+ usage();
+ dump_opt['v'] = verbose + 3;
+ return (dump_path(argv[0], argv[1]));
}
if (dump_opt['X'] || dump_opt['F'])
error = 0;
target = argv[0];
+ if (strpbrk(target, "/@") != NULL) {
+ size_t targetlen;
+
+ target_pool = strdup(target);
+ *strpbrk(target_pool, "/@") = '\0';
+
+ target_is_spa = B_FALSE;
+ targetlen = strlen(target);
+ if (targetlen && target[targetlen - 1] == '/')
+ target[targetlen - 1] = '\0';
+ } else {
+ target_pool = target;
+ }
+
if (dump_opt['e']) {
+ importargs_t args = { 0 };
nvlist_t *cfg = NULL;
- char *name = find_zpool(&target, &cfg, nsearch, searchdirs);
- error = ENOENT;
- if (name) {
- if (dump_opt['C'] > 1) {
- (void) printf("\nConfiguration for import:\n");
- dump_nvlist(cfg, 8);
- }
+ args.paths = nsearch;
+ args.path = searchdirs;
+ args.can_be_active = B_TRUE;
+
+ error = zpool_tryimport(g_zfs, target_pool, &cfg, &args);
+
+ if (error == 0) {
+
if (nvlist_add_nvlist(cfg,
ZPOOL_REWIND_POLICY, policy) != 0) {
fatal("can't open '%s': %s",
target, strerror(ENOMEM));
}
- error = spa_import(name, cfg, NULL, flags);
+
+ /*
+ * Disable the activity check to allow examination of
+ * active pools.
+ */
+ if (dump_opt['C'] > 1) {
+ (void) printf("\nConfiguration for import:\n");
+ dump_nvlist(cfg, 8);
+ }
+ error = spa_import(target_pool, cfg, NULL,
+ flags | ZFS_IMPORT_SKIP_MMP);
}
}
- if (strpbrk(target, "/@") != NULL) {
- size_t targetlen;
-
- target_is_spa = B_FALSE;
- targetlen = strlen(target);
- if (targetlen && target[targetlen - 1] == '/')
- target[targetlen - 1] = '\0';
- }
+ if (target_pool != target)
+ free(target_pool);
if (error == 0) {
if (target_is_spa || dump_opt['R']) {
+ /*
+ * Disable the activity check to allow examination of
+ * active pools.
+ */
+ mutex_enter(&spa_namespace_lock);
+ if ((spa = spa_lookup(target)) != NULL) {
+ spa->spa_import_flags |= ZFS_IMPORT_SKIP_MMP;
+ }
+ mutex_exit(&spa_namespace_lock);
+
error = spa_open_rewind(target, &spa, FTAG, policy,
NULL);
if (error) {
}
}
} else {
- error = dmu_objset_own(target, DMU_OST_ANY,
- B_TRUE, FTAG, &os);
+ error = open_objset(target, DMU_OST_ANY, FTAG, &os);
+ if (error == 0)
+ spa = dmu_objset_spa(os);
}
}
nvlist_free(policy);
if (error)
fatal("can't open '%s': %s", target, strerror(error));
+ /*
+ * Set the pool failure mode to panic in order to prevent the pool
+ * from suspending. A suspended I/O will have no way to resume and
+ * can prevent the zdb(8) command from terminating as expected.
+ */
+ if (spa != NULL)
+ spa->spa_failmode = ZIO_FAILURE_MODE_PANIC;
+
argv++;
argc--;
if (!dump_opt['R']) {
if (argc > 0) {
zopt_objects = argc;
zopt_object = calloc(zopt_objects, sizeof (uint64_t));
- for (i = 0; i < zopt_objects; i++) {
+ for (unsigned i = 0; i < zopt_objects; i++) {
errno = 0;
zopt_object[i] = strtoull(argv[i], NULL, 0);
if (zopt_object[i] == 0 && errno != 0)
flagbits['p'] = ZDB_FLAG_PHYS;
flagbits['r'] = ZDB_FLAG_RAW;
- for (i = 0; i < argc; i++)
+ for (int i = 0; i < argc; i++)
zdb_read_block(argv[i], spa);
}
- (os != NULL) ? dmu_objset_disown(os, FTAG) : spa_close(spa, FTAG);
+ if (os != NULL)
+ close_objset(os, FTAG);
+ else
+ spa_close(spa, FTAG);
fuid_table_destroy();
- sa_loaded = B_FALSE;
+
+ dump_debug_buffer();
libzfs_fini(g_zfs);
kernel_fini();
projects / mirror_zfs.git / blobdiff