diff options
Diffstat (limited to 'sys')
24 files changed, 218 insertions, 96 deletions
diff --git a/sys/contrib/openzfs/META b/sys/contrib/openzfs/META index e4b476aff112..5f834d5cc7c4 100644 --- a/sys/contrib/openzfs/META +++ b/sys/contrib/openzfs/META @@ -1,8 +1,8 @@ Meta: 1 Name: zfs Branch: 1.0 -Version: 2.1.99 -Release: 1 +Version: 2.2.0 +Release: rc1 Release-Tags: relext License: CDDL Author: OpenZFS diff --git a/sys/contrib/openzfs/cmd/zdb/zdb.c b/sys/contrib/openzfs/cmd/zdb/zdb.c index 04a10c4eedd7..9568d2bbfe38 100644 --- a/sys/contrib/openzfs/cmd/zdb/zdb.c +++ b/sys/contrib/openzfs/cmd/zdb/zdb.c @@ -794,7 +794,7 @@ usage(void) "\t\t[-o <var>=<value>]... [-t <txg>] [-U <cache>] [-x <dumpdir>]\n" "\t\t[-K <key>] <poolname>/<objset id> [<backupflags>]\n" "\t%s [-v] <bookmark>\n" - "\t%s -C [-A] [-U <cache>]\n" + "\t%s -C [-A] [-U <cache>] [<poolname>]\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" diff --git a/sys/contrib/openzfs/cmd/zfs/zfs_main.c b/sys/contrib/openzfs/cmd/zfs/zfs_main.c index f26adeae70d5..afd330630f69 100644 --- a/sys/contrib/openzfs/cmd/zfs/zfs_main.c +++ b/sys/contrib/openzfs/cmd/zfs/zfs_main.c @@ -6057,8 +6057,8 @@ construct_fsacl_list(boolean_t un, struct allow_opts *opts, nvlist_t **nvlp) if (p != NULL) rid = p->pw_uid; else if (*endch != '\0') { - (void) snprintf(errbuf, 256, gettext( - "invalid user %s\n"), curr); + (void) snprintf(errbuf, sizeof (errbuf), + gettext("invalid user %s\n"), curr); allow_usage(un, B_TRUE, errbuf); } } else if (opts->group) { @@ -6071,8 +6071,9 @@ construct_fsacl_list(boolean_t un, struct allow_opts *opts, nvlist_t **nvlp) if (g != NULL) rid = g->gr_gid; else if (*endch != '\0') { - (void) snprintf(errbuf, 256, gettext( - "invalid group %s\n"), curr); + (void) snprintf(errbuf, sizeof (errbuf), + gettext("invalid group %s\n"), + curr); allow_usage(un, B_TRUE, errbuf); } } else { @@ -6097,8 +6098,9 @@ construct_fsacl_list(boolean_t un, struct allow_opts *opts, nvlist_t **nvlp) who_type = ZFS_DELEG_GROUP; rid = g->gr_gid; } else { - (void) snprintf(errbuf, 256, gettext( - "invalid user/group %s\n"), curr); + (void) snprintf(errbuf, sizeof (errbuf), + gettext("invalid user/group %s\n"), + curr); allow_usage(un, B_TRUE, errbuf); } } diff --git a/sys/contrib/openzfs/cmd/zpool/zpool_main.c b/sys/contrib/openzfs/cmd/zpool/zpool_main.c index 013dd4a23380..10a3b5b14fc9 100644 --- a/sys/contrib/openzfs/cmd/zpool/zpool_main.c +++ b/sys/contrib/openzfs/cmd/zpool/zpool_main.c @@ -7662,11 +7662,11 @@ static void print_scan_scrub_resilver_status(pool_scan_stat_t *ps) { time_t start, end, pause; - uint64_t pass_scanned, scanned, pass_issued, issued, total; + uint64_t pass_scanned, scanned, pass_issued, issued, total_s, total_i; uint64_t elapsed, scan_rate, issue_rate; double fraction_done; - char processed_buf[7], scanned_buf[7], issued_buf[7], total_buf[7]; - char srate_buf[7], irate_buf[7], time_buf[32]; + char processed_buf[7], scanned_buf[7], issued_buf[7], total_s_buf[7]; + char total_i_buf[7], srate_buf[7], irate_buf[7], time_buf[32]; printf(" "); printf_color(ANSI_BOLD, gettext("scan:")); @@ -7738,10 +7738,11 @@ print_scan_scrub_resilver_status(pool_scan_stat_t *ps) pass_scanned = ps->pss_pass_exam; issued = ps->pss_issued; pass_issued = ps->pss_pass_issued; - total = ps->pss_to_examine; + total_s = ps->pss_to_examine; + total_i = ps->pss_to_examine - ps->pss_skipped; /* we are only done with a block once we have issued the IO for it */ - fraction_done = (double)issued / total; + fraction_done = (double)issued / total_i; /* elapsed time for this pass, rounding up to 1 if it's 0 */ elapsed = time(NULL) - ps->pss_pass_start; @@ -7750,26 +7751,25 @@ print_scan_scrub_resilver_status(pool_scan_stat_t *ps) scan_rate = pass_scanned / elapsed; issue_rate = pass_issued / elapsed; - uint64_t total_secs_left = (issue_rate != 0 && total >= issued) ? - ((total - issued) / issue_rate) : UINT64_MAX; - secs_to_dhms(total_secs_left, time_buf); /* format all of the numbers we will be reporting */ zfs_nicebytes(scanned, scanned_buf, sizeof (scanned_buf)); zfs_nicebytes(issued, issued_buf, sizeof (issued_buf)); - zfs_nicebytes(total, total_buf, sizeof (total_buf)); - zfs_nicebytes(scan_rate, srate_buf, sizeof (srate_buf)); - zfs_nicebytes(issue_rate, irate_buf, sizeof (irate_buf)); + zfs_nicebytes(total_s, total_s_buf, sizeof (total_s_buf)); + zfs_nicebytes(total_i, total_i_buf, sizeof (total_i_buf)); /* do not print estimated time if we have a paused scrub */ - if (pause == 0) { - (void) printf(gettext("\t%s scanned at %s/s, " - "%s issued at %s/s, %s total\n"), - scanned_buf, srate_buf, issued_buf, irate_buf, total_buf); - } else { - (void) printf(gettext("\t%s scanned, %s issued, %s total\n"), - scanned_buf, issued_buf, total_buf); + (void) printf(gettext("\t%s / %s scanned"), scanned_buf, total_s_buf); + if (pause == 0 && scan_rate > 0) { + zfs_nicebytes(scan_rate, srate_buf, sizeof (srate_buf)); + (void) printf(gettext(" at %s/s"), srate_buf); } + (void) printf(gettext(", %s / %s issued"), issued_buf, total_i_buf); + if (pause == 0 && issue_rate > 0) { + zfs_nicebytes(issue_rate, irate_buf, sizeof (irate_buf)); + (void) printf(gettext(" at %s/s"), irate_buf); + } + (void) printf(gettext("\n")); if (is_resilver) { (void) printf(gettext("\t%s resilvered, %.2f%% done"), @@ -7782,16 +7782,16 @@ print_scan_scrub_resilver_status(pool_scan_stat_t *ps) if (pause == 0) { /* * Only provide an estimate iff: - * 1) the time remaining is valid, and + * 1) we haven't yet issued all we expected, and * 2) the issue rate exceeds 10 MB/s, and * 3) it's either: * a) a resilver which has started repairs, or * b) a scrub which has entered the issue phase. */ - if (total_secs_left != UINT64_MAX && - issue_rate >= 10 * 1024 * 1024 && + if (total_i >= issued && issue_rate >= 10 * 1024 * 1024 && ((is_resilver && ps->pss_processed > 0) || (is_scrub && issued > 0))) { + secs_to_dhms((total_i - issued) / issue_rate, time_buf); (void) printf(gettext(", %s to go\n"), time_buf); } else { (void) printf(gettext(", no estimated " @@ -7803,7 +7803,7 @@ print_scan_scrub_resilver_status(pool_scan_stat_t *ps) } static void -print_rebuild_status_impl(vdev_rebuild_stat_t *vrs, char *vdev_name) +print_rebuild_status_impl(vdev_rebuild_stat_t *vrs, uint_t c, char *vdev_name) { if (vrs == NULL || vrs->vrs_state == VDEV_REBUILD_NONE) return; @@ -7815,17 +7815,20 @@ print_rebuild_status_impl(vdev_rebuild_stat_t *vrs, char *vdev_name) uint64_t bytes_scanned = vrs->vrs_bytes_scanned; uint64_t bytes_issued = vrs->vrs_bytes_issued; uint64_t bytes_rebuilt = vrs->vrs_bytes_rebuilt; - uint64_t bytes_est = vrs->vrs_bytes_est; + uint64_t bytes_est_s = vrs->vrs_bytes_est; + uint64_t bytes_est_i = vrs->vrs_bytes_est; + if (c > offsetof(vdev_rebuild_stat_t, vrs_pass_bytes_skipped) / 8) + bytes_est_i -= vrs->vrs_pass_bytes_skipped; uint64_t scan_rate = (vrs->vrs_pass_bytes_scanned / (vrs->vrs_pass_time_ms + 1)) * 1000; uint64_t issue_rate = (vrs->vrs_pass_bytes_issued / (vrs->vrs_pass_time_ms + 1)) * 1000; double scan_pct = MIN((double)bytes_scanned * 100 / - (bytes_est + 1), 100); + (bytes_est_s + 1), 100); /* Format all of the numbers we will be reporting */ char bytes_scanned_buf[7], bytes_issued_buf[7]; - char bytes_rebuilt_buf[7], bytes_est_buf[7]; + char bytes_rebuilt_buf[7], bytes_est_s_buf[7], bytes_est_i_buf[7]; char scan_rate_buf[7], issue_rate_buf[7], time_buf[32]; zfs_nicebytes(bytes_scanned, bytes_scanned_buf, sizeof (bytes_scanned_buf)); @@ -7833,9 +7836,8 @@ print_rebuild_status_impl(vdev_rebuild_stat_t *vrs, char *vdev_name) sizeof (bytes_issued_buf)); zfs_nicebytes(bytes_rebuilt, bytes_rebuilt_buf, sizeof (bytes_rebuilt_buf)); - zfs_nicebytes(bytes_est, bytes_est_buf, sizeof (bytes_est_buf)); - zfs_nicebytes(scan_rate, scan_rate_buf, sizeof (scan_rate_buf)); - zfs_nicebytes(issue_rate, issue_rate_buf, sizeof (issue_rate_buf)); + zfs_nicebytes(bytes_est_s, bytes_est_s_buf, sizeof (bytes_est_s_buf)); + zfs_nicebytes(bytes_est_i, bytes_est_i_buf, sizeof (bytes_est_i_buf)); time_t start = vrs->vrs_start_time; time_t end = vrs->vrs_end_time; @@ -7858,17 +7860,29 @@ print_rebuild_status_impl(vdev_rebuild_stat_t *vrs, char *vdev_name) assert(vrs->vrs_state == VDEV_REBUILD_ACTIVE); - secs_to_dhms(MAX((int64_t)bytes_est - (int64_t)bytes_scanned, 0) / - MAX(scan_rate, 1), time_buf); + (void) printf(gettext("\t%s / %s scanned"), bytes_scanned_buf, + bytes_est_s_buf); + if (scan_rate > 0) { + zfs_nicebytes(scan_rate, scan_rate_buf, sizeof (scan_rate_buf)); + (void) printf(gettext(" at %s/s"), scan_rate_buf); + } + (void) printf(gettext(", %s / %s issued"), bytes_issued_buf, + bytes_est_i_buf); + if (issue_rate > 0) { + zfs_nicebytes(issue_rate, issue_rate_buf, + sizeof (issue_rate_buf)); + (void) printf(gettext(" at %s/s"), issue_rate_buf); + } + (void) printf(gettext("\n")); - (void) printf(gettext("\t%s scanned at %s/s, %s issued %s/s, " - "%s total\n"), bytes_scanned_buf, scan_rate_buf, - bytes_issued_buf, issue_rate_buf, bytes_est_buf); (void) printf(gettext("\t%s resilvered, %.2f%% done"), bytes_rebuilt_buf, scan_pct); if (vrs->vrs_state == VDEV_REBUILD_ACTIVE) { - if (scan_rate >= 10 * 1024 * 1024) { + if (bytes_est_s >= bytes_scanned && + scan_rate >= 10 * 1024 * 1024) { + secs_to_dhms((bytes_est_s - bytes_scanned) / scan_rate, + time_buf); (void) printf(gettext(", %s to go\n"), time_buf); } else { (void) printf(gettext(", no estimated " @@ -7900,7 +7914,7 @@ print_rebuild_status(zpool_handle_t *zhp, nvlist_t *nvroot) ZPOOL_CONFIG_REBUILD_STATS, (uint64_t **)&vrs, &i) == 0) { char *name = zpool_vdev_name(g_zfs, zhp, child[c], VDEV_NAME_TYPE_ID); - print_rebuild_status_impl(vrs, name); + print_rebuild_status_impl(vrs, i, name); free(name); } } @@ -8005,13 +8019,15 @@ print_scan_status(zpool_handle_t *zhp, nvlist_t *nvroot) active_resilver = (ps->pss_state == DSS_SCANNING); } - have_resilver = (ps->pss_func == POOL_SCAN_RESILVER); have_scrub = (ps->pss_func == POOL_SCAN_SCRUB); scrub_start = ps->pss_start_time; - have_errorscrub = (ps->pss_error_scrub_func == - POOL_SCAN_ERRORSCRUB); - errorscrub_start = ps->pss_error_scrub_start; + if (c > offsetof(pool_scan_stat_t, + pss_pass_error_scrub_pause) / 8) { + have_errorscrub = (ps->pss_error_scrub_func == + POOL_SCAN_ERRORSCRUB); + errorscrub_start = ps->pss_error_scrub_start; + } } boolean_t active_rebuild = check_rebuilding(nvroot, &rebuild_end_time); diff --git a/sys/contrib/openzfs/cmd/zpool_influxdb/zpool_influxdb.c b/sys/contrib/openzfs/cmd/zpool_influxdb/zpool_influxdb.c index 80d08485891e..520e56926905 100644 --- a/sys/contrib/openzfs/cmd/zpool_influxdb/zpool_influxdb.c +++ b/sys/contrib/openzfs/cmd/zpool_influxdb/zpool_influxdb.c @@ -238,6 +238,7 @@ print_scan_status(nvlist_t *nvroot, const char *pool_name) print_kv("end_ts", ps->pss_end_time); print_kv(",errors", ps->pss_errors); print_kv(",examined", examined); + print_kv(",skipped", ps->pss_skipped); print_kv(",issued", ps->pss_issued); print_kv(",pass_examined", pass_exam); print_kv(",pass_issued", ps->pss_pass_issued); @@ -249,7 +250,6 @@ print_scan_status(nvlist_t *nvroot, const char *pool_name) print_kv(",remaining_t", remaining_time); print_kv(",start_ts", ps->pss_start_time); print_kv(",to_examine", ps->pss_to_examine); - print_kv(",to_process", ps->pss_to_process); printf(" %llu\n", (u_longlong_t)timestamp); return (0); } diff --git a/sys/contrib/openzfs/contrib/dracut/90zfs/module-setup.sh.in b/sys/contrib/openzfs/contrib/dracut/90zfs/module-setup.sh.in index e55cb60e1612..acad468edfd1 100755 --- a/sys/contrib/openzfs/contrib/dracut/90zfs/module-setup.sh.in +++ b/sys/contrib/openzfs/contrib/dracut/90zfs/module-setup.sh.in @@ -36,7 +36,7 @@ install() { { dfatal "Failed to install essential binaries"; exit 1; } # Adapted from https://github.com/zbm-dev/zfsbootmenu - if ! ldd "$(command -v zpool)" | grep -qF 'libgcc_s.so'; then + if ! ldd "$(command -v zpool)" | grep -qF 'libgcc_s.so' && ldconfig -p 2> /dev/null | grep -qF 'libc.so.6' ; then # On systems with gcc-config (Gentoo, Funtoo, etc.), use it to find libgcc_s if command -v gcc-config >/dev/null; then inst_simple "/usr/lib/gcc/$(s=$(gcc-config -c); echo "${s%-*}/${s##*-}")/libgcc_s.so.1" || diff --git a/sys/contrib/openzfs/include/sys/dsl_scan.h b/sys/contrib/openzfs/include/sys/dsl_scan.h index 6753b4a8f359..2e3452e5ebaa 100644 --- a/sys/contrib/openzfs/include/sys/dsl_scan.h +++ b/sys/contrib/openzfs/include/sys/dsl_scan.h @@ -61,7 +61,7 @@ typedef struct dsl_scan_phys { uint64_t scn_end_time; uint64_t scn_to_examine; /* total bytes to be scanned */ uint64_t scn_examined; /* bytes scanned so far */ - uint64_t scn_to_process; + uint64_t scn_skipped; /* bytes skipped by scanner */ uint64_t scn_processed; uint64_t scn_errors; /* scan I/O error count */ uint64_t scn_ddt_class_max; diff --git a/sys/contrib/openzfs/include/sys/fs/zfs.h b/sys/contrib/openzfs/include/sys/fs/zfs.h index 93193fa142da..bc940e8a7929 100644 --- a/sys/contrib/openzfs/include/sys/fs/zfs.h +++ b/sys/contrib/openzfs/include/sys/fs/zfs.h @@ -1088,7 +1088,7 @@ typedef struct pool_scan_stat { uint64_t pss_end_time; /* scan end time */ uint64_t pss_to_examine; /* total bytes to scan */ uint64_t pss_examined; /* total bytes located by scanner */ - uint64_t pss_to_process; /* total bytes to process */ + uint64_t pss_skipped; /* total bytes skipped by scanner */ uint64_t pss_processed; /* total processed bytes */ uint64_t pss_errors; /* scan errors */ @@ -1152,6 +1152,7 @@ typedef struct vdev_rebuild_stat { uint64_t vrs_pass_time_ms; /* pass run time (millisecs) */ uint64_t vrs_pass_bytes_scanned; /* bytes scanned since start/resume */ uint64_t vrs_pass_bytes_issued; /* bytes rebuilt since start/resume */ + uint64_t vrs_pass_bytes_skipped; /* bytes skipped since start/resume */ } vdev_rebuild_stat_t; /* diff --git a/sys/contrib/openzfs/include/sys/vdev_rebuild.h b/sys/contrib/openzfs/include/sys/vdev_rebuild.h index c4cfe0c56762..55ec6c570316 100644 --- a/sys/contrib/openzfs/include/sys/vdev_rebuild.h +++ b/sys/contrib/openzfs/include/sys/vdev_rebuild.h @@ -79,6 +79,7 @@ typedef struct vdev_rebuild { uint64_t vr_pass_start_time; uint64_t vr_pass_bytes_scanned; uint64_t vr_pass_bytes_issued; + uint64_t vr_pass_bytes_skipped; /* On-disk state updated by vdev_rebuild_zap_update_sync() */ vdev_rebuild_phys_t vr_rebuild_phys; diff --git a/sys/contrib/openzfs/include/sys/zio.h b/sys/contrib/openzfs/include/sys/zio.h index 85217b873dc8..f4da80783e56 100644 --- a/sys/contrib/openzfs/include/sys/zio.h +++ b/sys/contrib/openzfs/include/sys/zio.h @@ -341,9 +341,9 @@ typedef struct zio_prop { enum zio_checksum zp_checksum; enum zio_compress zp_compress; uint8_t zp_complevel; - dmu_object_type_t zp_type; uint8_t zp_level; uint8_t zp_copies; + dmu_object_type_t zp_type; boolean_t zp_dedup; boolean_t zp_dedup_verify; boolean_t zp_nopwrite; @@ -611,6 +611,7 @@ extern zio_t *zio_walk_parents(zio_t *cio, zio_link_t **); extern zio_t *zio_walk_children(zio_t *pio, zio_link_t **); extern zio_t *zio_unique_parent(zio_t *cio); extern void zio_add_child(zio_t *pio, zio_t *cio); +extern void zio_add_child_first(zio_t *pio, zio_t *cio); extern void *zio_buf_alloc(size_t size); extern void zio_buf_free(void *buf, size_t size); diff --git a/sys/contrib/openzfs/lib/libzfs/libzfs_dataset.c b/sys/contrib/openzfs/lib/libzfs/libzfs_dataset.c index fe9f3268d338..11d3eb6a3c60 100644 --- a/sys/contrib/openzfs/lib/libzfs/libzfs_dataset.c +++ b/sys/contrib/openzfs/lib/libzfs/libzfs_dataset.c @@ -1789,7 +1789,8 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props) nvlist_t *nvl; int nvl_len = 0; int added_resv = 0; - zfs_prop_t prop = 0; + zfs_prop_t prop; + boolean_t nsprop = B_FALSE; nvpair_t *elem; (void) snprintf(errbuf, sizeof (errbuf), @@ -1836,6 +1837,7 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props) elem = nvlist_next_nvpair(nvl, elem)) { prop = zfs_name_to_prop(nvpair_name(elem)); + nsprop |= zfs_is_namespace_prop(prop); assert(cl_idx < nvl_len); /* @@ -1934,8 +1936,7 @@ zfs_prop_set_list(zfs_handle_t *zhp, nvlist_t *props) * if one of the options handled by the generic * Linux namespace layer has been modified. */ - if (zfs_is_namespace_prop(prop) && - zfs_is_mounted(zhp, NULL)) + if (nsprop && zfs_is_mounted(zhp, NULL)) ret = zfs_mount(zhp, MNTOPT_REMOUNT, 0); } } diff --git a/sys/contrib/openzfs/man/man4/zfs.4 b/sys/contrib/openzfs/man/man4/zfs.4 index 04bbbc5fdf59..271b02b6ee42 100644 --- a/sys/contrib/openzfs/man/man4/zfs.4 +++ b/sys/contrib/openzfs/man/man4/zfs.4 @@ -239,6 +239,16 @@ relative to the pool. Make some blocks above a certain size be gang blocks. This option is used by the test suite to facilitate testing. . +.It Sy zfs_ddt_zap_default_bs Ns = Ns Sy 15 Po 32 KiB Pc Pq int +Default DDT ZAP data block size as a power of 2. Note that changing this after +creating a DDT on the pool will not affect existing DDTs, only newly created +ones. +. +.It Sy zfs_ddt_zap_default_ibs Ns = Ns Sy 15 Po 32 KiB Pc Pq int +Default DDT ZAP indirect block size as a power of 2. Note that changing this +after creating a DDT on the pool will not affect existing DDTs, only newly +created ones. +. .It Sy zfs_default_bs Ns = Ns Sy 9 Po 512 B Pc Pq int Default dnode block size as a power of 2. . diff --git a/sys/contrib/openzfs/man/man8/zdb.8 b/sys/contrib/openzfs/man/man8/zdb.8 index 031953c543a1..52c8e452fa7c 100644 --- a/sys/contrib/openzfs/man/man8/zdb.8 +++ b/sys/contrib/openzfs/man/man8/zdb.8 @@ -14,7 +14,7 @@ .\" Copyright (c) 2017 Lawrence Livermore National Security, LLC. .\" Copyright (c) 2017 Intel Corporation. .\" -.Dd June 4, 2023 +.Dd June 27, 2023 .Dt ZDB 8 .Os . @@ -51,6 +51,7 @@ .Fl C .Op Fl A .Op Fl U Ar cache +.Op Ar poolname .Nm .Fl E .Op Fl A diff --git a/sys/contrib/openzfs/man/man8/zpool-scrub.8 b/sys/contrib/openzfs/man/man8/zpool-scrub.8 index 138226e4562c..03f3ad4991f9 100644 --- a/sys/contrib/openzfs/man/man8/zpool-scrub.8 +++ b/sys/contrib/openzfs/man/man8/zpool-scrub.8 @@ -26,7 +26,7 @@ .\" Copyright 2017 Nexenta Systems, Inc. .\" Copyright (c) 2017 Open-E, Inc. All Rights Reserved. .\" -.Dd July 25, 2021 +.Dd June 22, 2023 .Dt ZPOOL-SCRUB 8 .Os . @@ -123,7 +123,7 @@ Status of pool with ongoing scrub: .No # Nm zpool Cm status ... scan: scrub in progress since Sun Jul 25 16:07:49 2021 - 403M scanned at 100M/s, 68.4M issued at 10.0M/s, 405M total + 403M / 405M scanned at 100M/s, 68.4M / 405M issued at 10.0M/s 0B repaired, 16.91% done, 00:00:04 to go ... .Ed diff --git a/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c b/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c index cdf32c78b4fe..38bc8e2c4eeb 100644 --- a/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c +++ b/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c @@ -54,7 +54,7 @@ static unsigned int zvol_prefetch_bytes = (128 * 1024); static unsigned long zvol_max_discard_blocks = 16384; #ifndef HAVE_BLKDEV_GET_ERESTARTSYS -static const unsigned int zvol_open_timeout_ms = 1000; +static unsigned int zvol_open_timeout_ms = 1000; #endif static unsigned int zvol_threads = 0; @@ -1612,4 +1612,9 @@ MODULE_PARM_DESC(zvol_blk_mq_blocks_per_thread, "Process volblocksize blocks per thread"); #endif +#ifndef HAVE_BLKDEV_GET_ERESTARTSYS +module_param(zvol_open_timeout_ms, uint, 0644); +MODULE_PARM_DESC(zvol_open_timeout_ms, "Timeout for ZVOL open retries"); +#endif + /* END CSTYLED */ diff --git a/sys/contrib/openzfs/module/zfs/ddt.c b/sys/contrib/openzfs/module/zfs/ddt.c index 33fea0ba3d3c..1fb198219904 100644 --- a/sys/contrib/openzfs/module/zfs/ddt.c +++ b/sys/contrib/openzfs/module/zfs/ddt.c @@ -1209,10 +1209,19 @@ ddt_addref(spa_t *spa, const blkptr_t *bp) ASSERT3S(dde->dde_class, <, DDT_CLASSES); ddp = &dde->dde_phys[BP_GET_NDVAS(bp)]; - if (ddp->ddp_refcnt == 0) { - /* This should never happen? */ - ddt_phys_fill(ddp, bp); - } + + /* + * This entry already existed (dde_type is real), so it must + * have refcnt >0 at the start of this txg. We are called from + * brt_pending_apply(), before frees are issued, so the refcnt + * can't be lowered yet. Therefore, it must be >0. We assert + * this because if the order of BRT and DDT interactions were + * ever to change and the refcnt was ever zero here, then + * likely further action is required to fill out the DDT entry, + * and this is a place that is likely to be missed in testing. + */ + ASSERT3U(ddp->ddp_refcnt, >, 0); + ddt_phys_addref(ddp); result = B_TRUE; } else { diff --git a/sys/contrib/openzfs/module/zfs/ddt_zap.c b/sys/contrib/openzfs/module/zfs/ddt_zap.c index 27dbbc55f121..8f6397a6d108 100644 --- a/sys/contrib/openzfs/module/zfs/ddt_zap.c +++ b/sys/contrib/openzfs/module/zfs/ddt_zap.c @@ -31,8 +31,8 @@ #include <sys/zap.h> #include <sys/dmu_tx.h> -static const int ddt_zap_leaf_blockshift = 12; -static const int ddt_zap_indirect_blockshift = 12; +static unsigned int ddt_zap_default_bs = 15; +static unsigned int ddt_zap_default_ibs = 15; static int ddt_zap_create(objset_t *os, uint64_t *objectp, dmu_tx_t *tx, boolean_t prehash) @@ -43,7 +43,7 @@ ddt_zap_create(objset_t *os, uint64_t *objectp, dmu_tx_t *tx, boolean_t prehash) flags |= ZAP_FLAG_PRE_HASHED_KEY; *objectp = zap_create_flags(os, 0, flags, DMU_OT_DDT_ZAP, - ddt_zap_leaf_blockshift, ddt_zap_indirect_blockshift, + ddt_zap_default_bs, ddt_zap_default_ibs, DMU_OT_NONE, 0, tx); return (*objectp == 0 ? SET_ERROR(ENOTSUP) : 0); @@ -166,3 +166,10 @@ const ddt_ops_t ddt_zap_ops = { ddt_zap_walk, ddt_zap_count, }; + +/* BEGIN CSTYLED */ +ZFS_MODULE_PARAM(zfs_dedup, , ddt_zap_default_bs, UINT, ZMOD_RW, + "DDT ZAP leaf blockshift"); +ZFS_MODULE_PARAM(zfs_dedup, , ddt_zap_default_ibs, UINT, ZMOD_RW, + "DDT ZAP indirect blockshift"); +/* END CSTYLED */ diff --git a/sys/contrib/openzfs/module/zfs/dsl_scan.c b/sys/contrib/openzfs/module/zfs/dsl_scan.c index 1dd44171c10e..50428bff3ef4 100644 --- a/sys/contrib/openzfs/module/zfs/dsl_scan.c +++ b/sys/contrib/openzfs/module/zfs/dsl_scan.c @@ -573,7 +573,8 @@ dsl_scan_init(dsl_pool_t *dp, uint64_t txg) * counter to how far we've scanned. We know we're consistent * up to here. */ - scn->scn_issued_before_pass = scn->scn_phys.scn_examined; + scn->scn_issued_before_pass = scn->scn_phys.scn_examined - + scn->scn_phys.scn_skipped; if (dsl_scan_is_running(scn) && spa_prev_software_version(dp->dp_spa) < SPA_VERSION_SCAN) { @@ -4362,7 +4363,7 @@ dsl_scan_sync(dsl_pool_t *dp, dmu_tx_t *tx) * Disabled by default, set zfs_scan_report_txgs to report * average performance over the last zfs_scan_report_txgs TXGs. */ - if (!dsl_scan_is_paused_scrub(scn) && zfs_scan_report_txgs != 0 && + if (zfs_scan_report_txgs != 0 && tx->tx_txg % zfs_scan_report_txgs == 0) { scn->scn_issued_before_pass += spa->spa_scan_pass_issued; spa_scan_stat_init(spa); @@ -4565,6 +4566,15 @@ count_block_issued(spa_t *spa, const blkptr_t *bp, boolean_t all) } static void +count_block_skipped(dsl_scan_t *scn, const blkptr_t *bp, boolean_t all) +{ + if (BP_IS_EMBEDDED(bp)) + return; + atomic_add_64(&scn->scn_phys.scn_skipped, + all ? BP_GET_ASIZE(bp) : DVA_GET_ASIZE(&bp->blk_dva[0])); +} + +static void count_block(zfs_all_blkstats_t *zab, const blkptr_t *bp) { /* @@ -4709,7 +4719,7 @@ dsl_scan_scrub_cb(dsl_pool_t *dp, count_block(dp->dp_blkstats, bp); if (phys_birth <= scn->scn_phys.scn_min_txg || phys_birth >= scn->scn_phys.scn_max_txg) { - count_block_issued(spa, bp, B_TRUE); + count_block_skipped(scn, bp, B_TRUE); return (0); } @@ -4750,7 +4760,7 @@ dsl_scan_scrub_cb(dsl_pool_t *dp, if (needs_io && !zfs_no_scrub_io) { dsl_scan_enqueue(dp, bp, zio_flags, zb); } else { - count_block_issued(spa, bp, B_TRUE); + count_block_skipped(scn, bp, B_TRUE); } /* do not relocate this block */ @@ -5119,9 +5129,9 @@ dsl_scan_freed_dva(spa_t *spa, const blkptr_t *bp, int dva_i) ASSERT(range_tree_contains(queue->q_exts_by_addr, start, size)); range_tree_remove_fill(queue->q_exts_by_addr, start, size); - /* count the block as though we issued it */ + /* count the block as though we skipped it */ sio2bp(sio, &tmpbp); - count_block_issued(spa, &tmpbp, B_FALSE); + count_block_skipped(scn, &tmpbp, B_FALSE); sio_free(sio); } diff --git a/sys/contrib/openzfs/module/zfs/spa_misc.c b/sys/contrib/openzfs/module/zfs/spa_misc.c index 8dc83445e198..06f640769043 100644 --- a/sys/contrib/openzfs/module/zfs/spa_misc.c +++ b/sys/contrib/openzfs/module/zfs/spa_misc.c @@ -2611,7 +2611,7 @@ spa_scan_get_stats(spa_t *spa, pool_scan_stat_t *ps) ps->pss_end_time = scn->scn_phys.scn_end_time; ps->pss_to_examine = scn->scn_phys.scn_to_examine; ps->pss_examined = scn->scn_phys.scn_examined; - ps->pss_to_process = scn->scn_phys.scn_to_process; + ps->pss_skipped = scn->scn_phys.scn_skipped; ps->pss_processed = scn->scn_phys.scn_processed; ps->pss_errors = scn->scn_phys.scn_errors; diff --git a/sys/contrib/openzfs/module/zfs/vdev_rebuild.c b/sys/contrib/openzfs/module/zfs/vdev_rebuild.c index 62aa61b3b9e7..75c3900cbb0c 100644 --- a/sys/contrib/openzfs/module/zfs/vdev_rebuild.c +++ b/sys/contrib/openzfs/module/zfs/vdev_rebuild.c @@ -571,8 +571,10 @@ vdev_rebuild_range(vdev_rebuild_t *vr, uint64_t start, uint64_t size) vdev_rebuild_blkptr_init(&blk, vd, start, size); uint64_t psize = BP_GET_PSIZE(&blk); - if (!vdev_dtl_need_resilver(vd, &blk.blk_dva[0], psize, TXG_UNKNOWN)) + if (!vdev_dtl_need_resilver(vd, &blk.blk_dva[0], psize, TXG_UNKNOWN)) { + vr->vr_pass_bytes_skipped += size; return (0); + } mutex_enter(&vr->vr_io_lock); @@ -786,6 +788,7 @@ vdev_rebuild_thread(void *arg) vr->vr_pass_start_time = gethrtime(); vr->vr_pass_bytes_scanned = 0; vr->vr_pass_bytes_issued = 0; + vr->vr_pass_bytes_skipped = 0; uint64_t update_est_time = gethrtime(); vdev_rebuild_update_bytes_est(vd, 0); @@ -1153,6 +1156,7 @@ vdev_rebuild_get_stats(vdev_t *tvd, vdev_rebuild_stat_t *vrs) vr->vr_pass_start_time); vrs->vrs_pass_bytes_scanned = vr->vr_pass_bytes_scanned; vrs->vrs_pass_bytes_issued = vr->vr_pass_bytes_issued; + vrs->vrs_pass_bytes_skipped = vr->vr_pass_bytes_skipped; mutex_exit(&tvd->vdev_rebuild_lock); } diff --git a/sys/contrib/openzfs/module/zfs/zil.c b/sys/contrib/openzfs/module/zfs/zil.c index ef6f52542ded..00d66a2481d7 100644 --- a/sys/contrib/openzfs/module/zfs/zil.c +++ b/sys/contrib/openzfs/module/zfs/zil.c @@ -798,8 +798,8 @@ zil_free_lwb(zilog_t *zilog, lwb_t *lwb) { ASSERT(MUTEX_HELD(&zilog->zl_lock)); ASSERT(!MUTEX_HELD(&lwb->lwb_vdev_lock)); - ASSERT(list_is_empty(&lwb->lwb_waiters)); - ASSERT(list_is_empty(&lwb->lwb_itxs)); + VERIFY(list_is_empty(&lwb->lwb_waiters)); + VERIFY(list_is_empty(&lwb->lwb_itxs)); ASSERT(avl_is_empty(&lwb->lwb_vdev_tree)); ASSERT3P(lwb->lwb_write_zio, ==, NULL); ASSERT3P(lwb->lwb_root_zio, ==, NULL); @@ -2525,10 +2525,10 @@ zil_clean(zilog_t *zilog, uint64_t synced_txg) * This function will traverse the queue of itxs that need to be * committed, and move them onto the ZIL's zl_itx_commit_list. */ -static void +static uint64_t zil_get_commit_list(zilog_t *zilog) { - uint64_t otxg, txg; + uint64_t otxg, txg, wtxg = 0; list_t *commit_list = &zilog->zl_itx_commit_list; ASSERT(MUTEX_HELD(&zilog->zl_issuer_lock)); @@ -2562,10 +2562,22 @@ zil_get_commit_list(zilog_t *zilog) */ ASSERT(zilog_is_dirty_in_txg(zilog, txg) || spa_freeze_txg(zilog->zl_spa) != UINT64_MAX); - list_move_tail(commit_list, &itxg->itxg_itxs->i_sync_list); + list_t *sync_list = &itxg->itxg_itxs->i_sync_list; + if (unlikely(zilog->zl_suspend > 0)) { + /* + * ZIL was just suspended, but we lost the race. + * Allow all earlier itxs to be committed, but ask + * caller to do txg_wait_synced(txg) for any new. + */ + if (!list_is_empty(sync_list)) + wtxg = MAX(wtxg, txg); + } else { + list_move_tail(commit_list, sync_list); + } mutex_exit(&itxg->itxg_lock); } + return (wtxg); } /* @@ -2953,11 +2965,12 @@ zil_process_commit_list(zilog_t *zilog, zil_commit_waiter_t *zcw, list_t *ilwbs) * not issued, we rely on future calls to zil_commit_writer() to issue * the lwb, or the timeout mechanism found in zil_commit_waiter(). */ -static void +static uint64_t zil_commit_writer(zilog_t *zilog, zil_commit_waiter_t *zcw) { list_t ilwbs; lwb_t *lwb; + uint64_t wtxg = 0; ASSERT(!MUTEX_HELD(&zilog->zl_lock)); ASSERT(spa_writeable(zilog->zl_spa)); @@ -2987,7 +3000,7 @@ zil_commit_writer(zilog_t *zilog, zil_commit_waiter_t *zcw) ZIL_STAT_BUMP(zilog, zil_commit_writer_count); - zil_get_commit_list(zilog); + wtxg = zil_get_commit_list(zilog); zil_prune_commit_list(zilog); zil_process_commit_list(zilog, zcw, &ilwbs); @@ -2996,6 +3009,7 @@ out: while ((lwb = list_remove_head(&ilwbs)) != NULL) zil_lwb_write_issue(zilog, lwb); list_destroy(&ilwbs); + return (wtxg); } static void @@ -3511,7 +3525,7 @@ zil_commit_impl(zilog_t *zilog, uint64_t foid) zil_commit_waiter_t *zcw = zil_alloc_commit_waiter(); zil_commit_itx_assign(zilog, zcw); - zil_commit_writer(zilog, zcw); + uint64_t wtxg = zil_commit_writer(zilog, zcw); zil_commit_waiter(zilog, zcw); if (zcw->zcw_zio_error != 0) { @@ -3526,6 +3540,8 @@ zil_commit_impl(zilog_t *zilog, uint64_t foid) DTRACE_PROBE2(zil__commit__io__error, zilog_t *, zilog, zil_commit_waiter_t *, zcw); txg_wait_synced(zilog->zl_dmu_pool, 0); + } else if (wtxg != 0) { + txg_wait_synced(zilog->zl_dmu_pool, wtxg); } zil_free_commit_waiter(zcw); @@ -3905,11 +3921,13 @@ zil_suspend(const char *osname, void **cookiep) return (error); zilog = dmu_objset_zil(os); + mutex_enter(&zilog->zl_issuer_lock); mutex_enter(&zilog->zl_lock); zh = zilog->zl_header; if (zh->zh_flags & ZIL_REPLAY_NEEDED) { /* unplayed log */ mutex_exit(&zilog->zl_lock); + mutex_exit(&zilog->zl_issuer_lock); dmu_objset_rele(os, suspend_tag); return (SET_ERROR(EBUSY)); } @@ -3923,6 +3941,7 @@ zil_suspend(const char *osname, void **cookiep) if (cookiep == NULL && !zilog->zl_suspending && (zilog->zl_suspend > 0 || BP_IS_HOLE(&zh->zh_log))) { mutex_exit(&zilog->zl_lock); + mutex_exit(&zilog->zl_issuer_lock); dmu_objset_rele(os, suspend_tag); return (0); } @@ -3931,6 +3950,7 @@ zil_suspend(const char *osname, void **cookiep) dsl_pool_rele(dmu_objset_pool(os), suspend_tag); zilog->zl_suspend++; + mutex_exit(&zilog->zl_issuer_lock); if (zilog->zl_suspend > 1) { /* diff --git a/sys/contrib/openzfs/module/zfs/zio.c b/sys/contrib/openzfs/module/zfs/zio.c index fb8164f0aea9..10279fde89df 100644 --- a/sys/contrib/openzfs/module/zfs/zio.c +++ b/sys/contrib/openzfs/module/zfs/zio.c @@ -626,8 +626,6 @@ zio_unique_parent(zio_t *cio) void zio_add_child(zio_t *pio, zio_t *cio) { - zio_link_t *zl = kmem_cache_alloc(zio_link_cache, KM_SLEEP); - /* * Logical I/Os can have logical, gang, or vdev children. * Gang I/Os can have gang or vdev children. @@ -636,6 +634,7 @@ zio_add_child(zio_t *pio, zio_t *cio) */ ASSERT3S(cio->io_child_type, <=, pio->io_child_type); + zio_link_t *zl = kmem_cache_alloc(zio_link_cache, KM_SLEEP); zl->zl_parent = pio; zl->zl_child = cio; @@ -644,8 +643,9 @@ zio_add_child(zio_t *pio, zio_t *cio) ASSERT(pio->io_state[ZIO_WAIT_DONE] == 0); + uint64_t *countp = pio->io_children[cio->io_child_type]; for (int w = 0; w < ZIO_WAIT_TYPES; w++) - pio->io_children[cio->io_child_type][w] += !cio->io_state[w]; + countp[w] += !cio->io_state[w]; list_insert_head(&pio->io_child_list, zl); list_insert_head(&cio->io_parent_list, zl); @@ -654,6 +654,37 @@ zio_add_child(zio_t *pio, zio_t *cio) mutex_exit(&pio->io_lock); } +void +zio_add_child_first(zio_t *pio, zio_t *cio) +{ + /* + * Logical I/Os can have logical, gang, or vdev children. + * Gang I/Os can have gang or vdev children. + * Vdev I/Os can only have vdev children. + * The following ASSERT captures all of these constraints. + */ + ASSERT3S(cio->io_child_type, <=, pio->io_child_type); + + zio_link_t *zl = kmem_cache_alloc(zio_link_cache, KM_SLEEP); + zl->zl_parent = pio; + zl->zl_child = cio; + + ASSERT(list_is_empty(&cio->io_parent_list)); + list_insert_head(&cio->io_parent_list, zl); + + mutex_enter(&pio->io_lock); + + ASSERT(pio->io_state[ZIO_WAIT_DONE] == 0); + + uint64_t *countp = pio->io_children[cio->io_child_type]; + for (int w = 0; w < ZIO_WAIT_TYPES; w++) + countp[w] += !cio->io_state[w]; + + list_insert_head(&pio->io_child_list, zl); + + mutex_exit(&pio->io_lock); +} + static void zio_remove_child(zio_t *pio, zio_t *cio, zio_link_t *zl) { @@ -840,12 +871,14 @@ zio_create(zio_t *pio, spa_t *spa, uint64_t txg, const blkptr_t *bp, zio->io_child_type = ZIO_CHILD_LOGICAL; if (bp != NULL) { - zio->io_bp = (blkptr_t *)bp; - zio->io_bp_copy = *bp; - zio->io_bp_orig = *bp; if (type != ZIO_TYPE_WRITE || - zio->io_child_type == ZIO_CHILD_DDT) + zio->io_child_type == ZIO_CHILD_DDT) { + zio->io_bp_copy = *bp; zio->io_bp = &zio->io_bp_copy; /* so caller can free */ + } else { + zio->io_bp = (blkptr_t *)bp; + } + zio->io_bp_orig = *bp; if (zio->io_child_type == ZIO_CHILD_LOGICAL) zio->io_logical = zio; if (zio->io_child_type > ZIO_CHILD_GANG && BP_IS_GANG(bp)) @@ -880,7 +913,7 @@ zio_create(zio_t *pio, spa_t *spa, uint64_t txg, const blkptr_t *bp, zio->io_logical = pio->io_logical; if (zio->io_child_type == ZIO_CHILD_GANG) zio->io_gang_leader = pio->io_gang_leader; - zio_add_child(pio, zio); + zio_add_child_first(pio, zio); } taskq_init_ent(&zio->io_tqent); @@ -1601,7 +1634,6 @@ zio_read_bp_init(zio_t *zio) abd_return_buf_copy(zio->io_abd, data, psize); } else { ASSERT(!BP_IS_EMBEDDED(bp)); - ASSERT3P(zio->io_bp, ==, &zio->io_bp_copy); } if (BP_GET_DEDUP(bp) && zio->io_child_type == ZIO_CHILD_LOGICAL) @@ -4442,8 +4474,10 @@ zio_ready(zio_t *zio) zio->io_ready(zio); } +#ifdef ZFS_DEBUG if (bp != NULL && bp != &zio->io_bp_copy) zio->io_bp_copy = *bp; +#endif if (zio->io_error != 0) { zio->io_pipeline = ZIO_INTERLOCK_PIPELINE; diff --git a/sys/modules/zfs/zfs_config.h b/sys/modules/zfs/zfs_config.h index ddddc1ef9504..d448d8dcc96f 100644 --- a/sys/modules/zfs/zfs_config.h +++ b/sys/modules/zfs/zfs_config.h @@ -1051,7 +1051,7 @@ /* #undef ZFS_IS_GPL_COMPATIBLE */ /* Define the project alias string. */ -#define ZFS_META_ALIAS "zfs-2.1.99-FreeBSD_ga9d6b0690" +#define ZFS_META_ALIAS "zfs-2.2.0-FreeBSD_g009d3288" /* Define the project author. */ #define ZFS_META_AUTHOR "OpenZFS" @@ -1081,10 +1081,10 @@ #define ZFS_META_NAME "zfs" /* Define the project release. */ -#define ZFS_META_RELEASE "FreeBSD_ga9d6b0690" +#define ZFS_META_RELEASE "FreeBSD_g009d3288" /* Define the project version. */ -#define ZFS_META_VERSION "2.1.99" +#define ZFS_META_VERSION "2.2.0" /* count is located in percpu_ref.data */ /* #undef ZFS_PERCPU_REF_COUNT_IN_DATA */ diff --git a/sys/modules/zfs/zfs_gitrev.h b/sys/modules/zfs/zfs_gitrev.h index 50f01248623d..afe3b29c865b 100644 --- a/sys/modules/zfs/zfs_gitrev.h +++ b/sys/modules/zfs/zfs_gitrev.h @@ -1 +1 @@ -#define ZFS_META_GITREV "zfs-2.1.99-2004-ga9d6b0690" +#define ZFS_META_GITREV "zfs-2.2.0-rc1-0-g009d3288d" |