diff --git a/include/sys/vdev_impl.h b/include/sys/vdev_impl.h index 16cf4d77071a..16fcaa042278 100644 --- a/include/sys/vdev_impl.h +++ b/include/sys/vdev_impl.h @@ -1,351 +1,352 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2013 by Delphix. All rights reserved. */ #ifndef _SYS_VDEV_IMPL_H #define _SYS_VDEV_IMPL_H #include #include #include #include #include #include #include #include #ifdef __cplusplus extern "C" { #endif /* * Virtual device descriptors. * * All storage pool operations go through the virtual device framework, * which provides data replication and I/O scheduling. */ /* * Forward declarations that lots of things need. */ typedef struct vdev_queue vdev_queue_t; typedef struct vdev_io vdev_io_t; typedef struct vdev_cache vdev_cache_t; typedef struct vdev_cache_entry vdev_cache_entry_t; /* * Virtual device operations */ typedef int vdev_open_func_t(vdev_t *vd, uint64_t *size, uint64_t *max_size, uint64_t *ashift); typedef void vdev_close_func_t(vdev_t *vd); typedef uint64_t vdev_asize_func_t(vdev_t *vd, uint64_t psize); typedef int vdev_io_start_func_t(zio_t *zio); typedef void vdev_io_done_func_t(zio_t *zio); typedef void vdev_state_change_func_t(vdev_t *vd, int, int); typedef void vdev_hold_func_t(vdev_t *vd); typedef void vdev_rele_func_t(vdev_t *vd); typedef const struct vdev_ops { vdev_open_func_t *vdev_op_open; vdev_close_func_t *vdev_op_close; vdev_asize_func_t *vdev_op_asize; vdev_io_start_func_t *vdev_op_io_start; vdev_io_done_func_t *vdev_op_io_done; vdev_state_change_func_t *vdev_op_state_change; vdev_hold_func_t *vdev_op_hold; vdev_rele_func_t *vdev_op_rele; char vdev_op_type[16]; boolean_t vdev_op_leaf; } vdev_ops_t; /* * Virtual device properties */ struct vdev_cache_entry { char *ve_data; uint64_t ve_offset; clock_t ve_lastused; avl_node_t ve_offset_node; avl_node_t ve_lastused_node; uint32_t ve_hits; uint16_t ve_missed_update; zio_t *ve_fill_io; }; struct vdev_cache { avl_tree_t vc_offset_tree; avl_tree_t vc_lastused_tree; kmutex_t vc_lock; }; typedef struct vdev_queue_class { uint32_t vqc_active; /* * Sorted by offset or timestamp, depending on if the queue is * LBA-ordered vs FIFO. */ avl_tree_t vqc_queued_tree; } vdev_queue_class_t; struct vdev_queue { vdev_t *vq_vdev; vdev_queue_class_t vq_class[ZIO_PRIORITY_NUM_QUEUEABLE]; avl_tree_t vq_active_tree; uint64_t vq_last_offset; hrtime_t vq_io_complete_ts; /* time last i/o completed */ hrtime_t vq_io_delta_ts; list_t vq_io_list; + zio_t vq_io_search; /* used as local for stack reduction */ kmutex_t vq_lock; }; struct vdev_io { char vi_buffer[SPA_MAXBLOCKSIZE]; /* Must be first */ list_node_t vi_node; }; /* * Virtual device descriptor */ struct vdev { /* * Common to all vdev types. */ uint64_t vdev_id; /* child number in vdev parent */ uint64_t vdev_guid; /* unique ID for this vdev */ uint64_t vdev_guid_sum; /* self guid + all child guids */ uint64_t vdev_orig_guid; /* orig. guid prior to remove */ uint64_t vdev_asize; /* allocatable device capacity */ uint64_t vdev_min_asize; /* min acceptable asize */ uint64_t vdev_max_asize; /* max acceptable asize */ uint64_t vdev_ashift; /* block alignment shift */ uint64_t vdev_state; /* see VDEV_STATE_* #defines */ uint64_t vdev_prevstate; /* used when reopening a vdev */ vdev_ops_t *vdev_ops; /* vdev operations */ spa_t *vdev_spa; /* spa for this vdev */ void *vdev_tsd; /* type-specific data */ vnode_t *vdev_name_vp; /* vnode for pathname */ vnode_t *vdev_devid_vp; /* vnode for devid */ vdev_t *vdev_top; /* top-level vdev */ vdev_t *vdev_parent; /* parent vdev */ vdev_t **vdev_child; /* array of children */ uint64_t vdev_children; /* number of children */ vdev_stat_t vdev_stat; /* virtual device statistics */ boolean_t vdev_expanding; /* expand the vdev? */ boolean_t vdev_reopening; /* reopen in progress? */ int vdev_open_error; /* error on last open */ kthread_t *vdev_open_thread; /* thread opening children */ uint64_t vdev_crtxg; /* txg when top-level was added */ /* * Top-level vdev state. */ uint64_t vdev_ms_array; /* metaslab array object */ uint64_t vdev_ms_shift; /* metaslab size shift */ uint64_t vdev_ms_count; /* number of metaslabs */ metaslab_group_t *vdev_mg; /* metaslab group */ metaslab_t **vdev_ms; /* metaslab array */ uint64_t vdev_pending_fastwrite; /* allocated fastwrites */ txg_list_t vdev_ms_list; /* per-txg dirty metaslab lists */ txg_list_t vdev_dtl_list; /* per-txg dirty DTL lists */ txg_node_t vdev_txg_node; /* per-txg dirty vdev linkage */ boolean_t vdev_remove_wanted; /* async remove wanted? */ boolean_t vdev_probe_wanted; /* async probe wanted? */ list_node_t vdev_config_dirty_node; /* config dirty list */ list_node_t vdev_state_dirty_node; /* state dirty list */ uint64_t vdev_deflate_ratio; /* deflation ratio (x512) */ uint64_t vdev_islog; /* is an intent log device */ uint64_t vdev_removing; /* device is being removed? */ boolean_t vdev_ishole; /* is a hole in the namespace */ /* * Leaf vdev state. */ range_tree_t *vdev_dtl[DTL_TYPES]; /* dirty time logs */ space_map_t *vdev_dtl_sm; /* dirty time log space map */ txg_node_t vdev_dtl_node; /* per-txg dirty DTL linkage */ uint64_t vdev_dtl_object; /* DTL object */ uint64_t vdev_psize; /* physical device capacity */ uint64_t vdev_wholedisk; /* true if this is a whole disk */ uint64_t vdev_offline; /* persistent offline state */ uint64_t vdev_faulted; /* persistent faulted state */ uint64_t vdev_degraded; /* persistent degraded state */ uint64_t vdev_removed; /* persistent removed state */ uint64_t vdev_resilver_txg; /* persistent resilvering state */ uint64_t vdev_nparity; /* number of parity devices for raidz */ char *vdev_path; /* vdev path (if any) */ char *vdev_devid; /* vdev devid (if any) */ char *vdev_physpath; /* vdev device path (if any) */ char *vdev_fru; /* physical FRU location */ uint64_t vdev_not_present; /* not present during import */ uint64_t vdev_unspare; /* unspare when resilvering done */ boolean_t vdev_nowritecache; /* true if flushwritecache failed */ boolean_t vdev_checkremove; /* temporary online test */ boolean_t vdev_forcefault; /* force online fault */ boolean_t vdev_splitting; /* split or repair in progress */ boolean_t vdev_delayed_close; /* delayed device close? */ boolean_t vdev_tmpoffline; /* device taken offline temporarily? */ boolean_t vdev_detached; /* device detached? */ boolean_t vdev_cant_read; /* vdev is failing all reads */ boolean_t vdev_cant_write; /* vdev is failing all writes */ boolean_t vdev_isspare; /* was a hot spare */ boolean_t vdev_isl2cache; /* was a l2cache device */ vdev_queue_t vdev_queue; /* I/O deadline schedule queue */ vdev_cache_t vdev_cache; /* physical block cache */ spa_aux_vdev_t *vdev_aux; /* for l2cache vdevs */ zio_t *vdev_probe_zio; /* root of current probe */ vdev_aux_t vdev_label_aux; /* on-disk aux state */ /* * For DTrace to work in userland (libzpool) context, these fields must * remain at the end of the structure. DTrace will use the kernel's * CTF definition for 'struct vdev', and since the size of a kmutex_t is * larger in userland, the offsets for the rest of the fields would be * incorrect. */ kmutex_t vdev_dtl_lock; /* vdev_dtl_{map,resilver} */ kmutex_t vdev_stat_lock; /* vdev_stat */ kmutex_t vdev_probe_lock; /* protects vdev_probe_zio */ }; #define VDEV_RAIDZ_MAXPARITY 3 #define VDEV_PAD_SIZE (8 << 10) /* 2 padding areas (vl_pad1 and vl_pad2) to skip */ #define VDEV_SKIP_SIZE VDEV_PAD_SIZE * 2 #define VDEV_PHYS_SIZE (112 << 10) #define VDEV_UBERBLOCK_RING (128 << 10) #define VDEV_UBERBLOCK_SHIFT(vd) \ MAX((vd)->vdev_top->vdev_ashift, UBERBLOCK_SHIFT) #define VDEV_UBERBLOCK_COUNT(vd) \ (VDEV_UBERBLOCK_RING >> VDEV_UBERBLOCK_SHIFT(vd)) #define VDEV_UBERBLOCK_OFFSET(vd, n) \ offsetof(vdev_label_t, vl_uberblock[(n) << VDEV_UBERBLOCK_SHIFT(vd)]) #define VDEV_UBERBLOCK_SIZE(vd) (1ULL << VDEV_UBERBLOCK_SHIFT(vd)) typedef struct vdev_phys { char vp_nvlist[VDEV_PHYS_SIZE - sizeof (zio_eck_t)]; zio_eck_t vp_zbt; } vdev_phys_t; typedef struct vdev_label { char vl_pad1[VDEV_PAD_SIZE]; /* 8K */ char vl_pad2[VDEV_PAD_SIZE]; /* 8K */ vdev_phys_t vl_vdev_phys; /* 112K */ char vl_uberblock[VDEV_UBERBLOCK_RING]; /* 128K */ } vdev_label_t; /* 256K total */ /* * vdev_dirty() flags */ #define VDD_METASLAB 0x01 #define VDD_DTL 0x02 /* Offset of embedded boot loader region on each label */ #define VDEV_BOOT_OFFSET (2 * sizeof (vdev_label_t)) /* * Size of embedded boot loader region on each label. * The total size of the first two labels plus the boot area is 4MB. */ #define VDEV_BOOT_SIZE (7ULL << 19) /* 3.5M */ /* * Size of label regions at the start and end of each leaf device. */ #define VDEV_LABEL_START_SIZE (2 * sizeof (vdev_label_t) + VDEV_BOOT_SIZE) #define VDEV_LABEL_END_SIZE (2 * sizeof (vdev_label_t)) #define VDEV_LABELS 4 #define VDEV_BEST_LABEL VDEV_LABELS #define VDEV_ALLOC_LOAD 0 #define VDEV_ALLOC_ADD 1 #define VDEV_ALLOC_SPARE 2 #define VDEV_ALLOC_L2CACHE 3 #define VDEV_ALLOC_ROOTPOOL 4 #define VDEV_ALLOC_SPLIT 5 #define VDEV_ALLOC_ATTACH 6 /* * Allocate or free a vdev */ extern vdev_t *vdev_alloc_common(spa_t *spa, uint_t id, uint64_t guid, vdev_ops_t *ops); extern int vdev_alloc(spa_t *spa, vdev_t **vdp, nvlist_t *config, vdev_t *parent, uint_t id, int alloctype); extern void vdev_free(vdev_t *vd); /* * Add or remove children and parents */ extern void vdev_add_child(vdev_t *pvd, vdev_t *cvd); extern void vdev_remove_child(vdev_t *pvd, vdev_t *cvd); extern void vdev_compact_children(vdev_t *pvd); extern vdev_t *vdev_add_parent(vdev_t *cvd, vdev_ops_t *ops); extern void vdev_remove_parent(vdev_t *cvd); /* * vdev sync load and sync */ extern void vdev_load_log_state(vdev_t *nvd, vdev_t *ovd); extern boolean_t vdev_log_state_valid(vdev_t *vd); extern void vdev_load(vdev_t *vd); extern int vdev_dtl_load(vdev_t *vd); extern void vdev_sync(vdev_t *vd, uint64_t txg); extern void vdev_sync_done(vdev_t *vd, uint64_t txg); extern void vdev_dirty(vdev_t *vd, int flags, void *arg, uint64_t txg); extern void vdev_dirty_leaves(vdev_t *vd, int flags, uint64_t txg); /* * Available vdev types. */ extern vdev_ops_t vdev_root_ops; extern vdev_ops_t vdev_mirror_ops; extern vdev_ops_t vdev_replacing_ops; extern vdev_ops_t vdev_raidz_ops; extern vdev_ops_t vdev_disk_ops; extern vdev_ops_t vdev_file_ops; extern vdev_ops_t vdev_missing_ops; extern vdev_ops_t vdev_hole_ops; extern vdev_ops_t vdev_spare_ops; /* * Common size functions */ extern uint64_t vdev_default_asize(vdev_t *vd, uint64_t psize); extern uint64_t vdev_get_min_asize(vdev_t *vd); extern void vdev_set_min_asize(vdev_t *vd); /* * Global variables */ /* zdb uses this tunable, so it must be declared here to make lint happy. */ extern int zfs_vdev_cache_size; #ifdef __cplusplus } #endif #endif /* _SYS_VDEV_IMPL_H */ diff --git a/module/zfs/vdev_queue.c b/module/zfs/vdev_queue.c index 0dc733efc3e4..e2758d1c41f5 100644 --- a/module/zfs/vdev_queue.c +++ b/module/zfs/vdev_queue.c @@ -1,850 +1,848 @@ /* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright 2009 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* * Copyright (c) 2013 by Delphix. All rights reserved. */ #include #include #include #include #include #include #include #include #include /* * ZFS I/O Scheduler * --------------- * * ZFS issues I/O operations to leaf vdevs to satisfy and complete zios. The * I/O scheduler determines when and in what order those operations are * issued. The I/O scheduler divides operations into five I/O classes * prioritized in the following order: sync read, sync write, async read, * async write, and scrub/resilver. Each queue defines the minimum and * maximum number of concurrent operations that may be issued to the device. * In addition, the device has an aggregate maximum. Note that the sum of the * per-queue minimums must not exceed the aggregate maximum. If the * sum of the per-queue maximums exceeds the aggregate maximum, then the * number of active i/os may reach zfs_vdev_max_active, in which case no * further i/os will be issued regardless of whether all per-queue * minimums have been met. * * For many physical devices, throughput increases with the number of * concurrent operations, but latency typically suffers. Further, physical * devices typically have a limit at which more concurrent operations have no * effect on throughput or can actually cause it to decrease. * * The scheduler selects the next operation to issue by first looking for an * I/O class whose minimum has not been satisfied. Once all are satisfied and * the aggregate maximum has not been hit, the scheduler looks for classes * whose maximum has not been satisfied. Iteration through the I/O classes is * done in the order specified above. No further operations are issued if the * aggregate maximum number of concurrent operations has been hit or if there * are no operations queued for an I/O class that has not hit its maximum. * Every time an i/o is queued or an operation completes, the I/O scheduler * looks for new operations to issue. * * All I/O classes have a fixed maximum number of outstanding operations * except for the async write class. Asynchronous writes represent the data * that is committed to stable storage during the syncing stage for * transaction groups (see txg.c). Transaction groups enter the syncing state * periodically so the number of queued async writes will quickly burst up and * then bleed down to zero. Rather than servicing them as quickly as possible, * the I/O scheduler changes the maximum number of active async write i/os * according to the amount of dirty data in the pool (see dsl_pool.c). Since * both throughput and latency typically increase with the number of * concurrent operations issued to physical devices, reducing the burstiness * in the number of concurrent operations also stabilizes the response time of * operations from other -- and in particular synchronous -- queues. In broad * strokes, the I/O scheduler will issue more concurrent operations from the * async write queue as there's more dirty data in the pool. * * Async Writes * * The number of concurrent operations issued for the async write I/O class * follows a piece-wise linear function defined by a few adjustable points. * * | o---------| <-- zfs_vdev_async_write_max_active * ^ | /^ | * | | / | | * active | / | | * I/O | / | | * count | / | | * | / | | * |------------o | | <-- zfs_vdev_async_write_min_active * 0|____________^______|_________| * 0% | | 100% of zfs_dirty_data_max * | | * | `-- zfs_vdev_async_write_active_max_dirty_percent * `--------- zfs_vdev_async_write_active_min_dirty_percent * * Until the amount of dirty data exceeds a minimum percentage of the dirty * data allowed in the pool, the I/O scheduler will limit the number of * concurrent operations to the minimum. As that threshold is crossed, the * number of concurrent operations issued increases linearly to the maximum at * the specified maximum percentage of the dirty data allowed in the pool. * * Ideally, the amount of dirty data on a busy pool will stay in the sloped * part of the function between zfs_vdev_async_write_active_min_dirty_percent * and zfs_vdev_async_write_active_max_dirty_percent. If it exceeds the * maximum percentage, this indicates that the rate of incoming data is * greater than the rate that the backend storage can handle. In this case, we * must further throttle incoming writes (see dmu_tx_delay() for details). */ /* * The maximum number of i/os active to each device. Ideally, this will be >= * the sum of each queue's max_active. It must be at least the sum of each * queue's min_active. */ uint32_t zfs_vdev_max_active = 1000; /* * Per-queue limits on the number of i/os active to each device. If the * number of active i/os is < zfs_vdev_max_active, then the min_active comes * into play. We will send min_active from each queue, and then select from * queues in the order defined by zio_priority_t. * * In general, smaller max_active's will lead to lower latency of synchronous * operations. Larger max_active's may lead to higher overall throughput, * depending on underlying storage. * * The ratio of the queues' max_actives determines the balance of performance * between reads, writes, and scrubs. E.g., increasing * zfs_vdev_scrub_max_active will cause the scrub or resilver to complete * more quickly, but reads and writes to have higher latency and lower * throughput. */ uint32_t zfs_vdev_sync_read_min_active = 10; uint32_t zfs_vdev_sync_read_max_active = 10; uint32_t zfs_vdev_sync_write_min_active = 10; uint32_t zfs_vdev_sync_write_max_active = 10; uint32_t zfs_vdev_async_read_min_active = 1; uint32_t zfs_vdev_async_read_max_active = 3; uint32_t zfs_vdev_async_write_min_active = 1; uint32_t zfs_vdev_async_write_max_active = 10; uint32_t zfs_vdev_scrub_min_active = 1; uint32_t zfs_vdev_scrub_max_active = 2; /* * When the pool has less than zfs_vdev_async_write_active_min_dirty_percent * dirty data, use zfs_vdev_async_write_min_active. When it has more than * zfs_vdev_async_write_active_max_dirty_percent, use * zfs_vdev_async_write_max_active. The value is linearly interpolated * between min and max. */ int zfs_vdev_async_write_active_min_dirty_percent = 30; int zfs_vdev_async_write_active_max_dirty_percent = 60; /* * To reduce IOPs, we aggregate small adjacent I/Os into one large I/O. * For read I/Os, we also aggregate across small adjacency gaps; for writes * we include spans of optional I/Os to aid aggregation at the disk even when * they aren't able to help us aggregate at this level. */ int zfs_vdev_aggregation_limit = SPA_MAXBLOCKSIZE; int zfs_vdev_read_gap_limit = 32 << 10; int zfs_vdev_write_gap_limit = 4 << 10; int vdev_queue_offset_compare(const void *x1, const void *x2) { const zio_t *z1 = x1; const zio_t *z2 = x2; if (z1->io_offset < z2->io_offset) return (-1); if (z1->io_offset > z2->io_offset) return (1); if (z1 < z2) return (-1); if (z1 > z2) return (1); return (0); } int vdev_queue_timestamp_compare(const void *x1, const void *x2) { const zio_t *z1 = x1; const zio_t *z2 = x2; if (z1->io_timestamp < z2->io_timestamp) return (-1); if (z1->io_timestamp > z2->io_timestamp) return (1); if (z1 < z2) return (-1); if (z1 > z2) return (1); return (0); } static int vdev_queue_class_min_active(zio_priority_t p) { switch (p) { case ZIO_PRIORITY_SYNC_READ: return (zfs_vdev_sync_read_min_active); case ZIO_PRIORITY_SYNC_WRITE: return (zfs_vdev_sync_write_min_active); case ZIO_PRIORITY_ASYNC_READ: return (zfs_vdev_async_read_min_active); case ZIO_PRIORITY_ASYNC_WRITE: return (zfs_vdev_async_write_min_active); case ZIO_PRIORITY_SCRUB: return (zfs_vdev_scrub_min_active); default: panic("invalid priority %u", p); return (0); } } static int vdev_queue_max_async_writes(uint64_t dirty) { int writes; uint64_t min_bytes = zfs_dirty_data_max * zfs_vdev_async_write_active_min_dirty_percent / 100; uint64_t max_bytes = zfs_dirty_data_max * zfs_vdev_async_write_active_max_dirty_percent / 100; if (dirty < min_bytes) return (zfs_vdev_async_write_min_active); if (dirty > max_bytes) return (zfs_vdev_async_write_max_active); /* * linear interpolation: * slope = (max_writes - min_writes) / (max_bytes - min_bytes) * move right by min_bytes * move up by min_writes */ writes = (dirty - min_bytes) * (zfs_vdev_async_write_max_active - zfs_vdev_async_write_min_active) / (max_bytes - min_bytes) + zfs_vdev_async_write_min_active; ASSERT3U(writes, >=, zfs_vdev_async_write_min_active); ASSERT3U(writes, <=, zfs_vdev_async_write_max_active); return (writes); } static int vdev_queue_class_max_active(spa_t *spa, zio_priority_t p) { switch (p) { case ZIO_PRIORITY_SYNC_READ: return (zfs_vdev_sync_read_max_active); case ZIO_PRIORITY_SYNC_WRITE: return (zfs_vdev_sync_write_max_active); case ZIO_PRIORITY_ASYNC_READ: return (zfs_vdev_async_read_max_active); case ZIO_PRIORITY_ASYNC_WRITE: return (vdev_queue_max_async_writes( spa->spa_dsl_pool->dp_dirty_total)); case ZIO_PRIORITY_SCRUB: return (zfs_vdev_scrub_max_active); default: panic("invalid priority %u", p); return (0); } } /* * Return the i/o class to issue from, or ZIO_PRIORITY_MAX_QUEUEABLE if * there is no eligible class. */ static zio_priority_t vdev_queue_class_to_issue(vdev_queue_t *vq) { spa_t *spa = vq->vq_vdev->vdev_spa; zio_priority_t p; if (avl_numnodes(&vq->vq_active_tree) >= zfs_vdev_max_active) return (ZIO_PRIORITY_NUM_QUEUEABLE); /* find a queue that has not reached its minimum # outstanding i/os */ for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) { if (avl_numnodes(&vq->vq_class[p].vqc_queued_tree) > 0 && vq->vq_class[p].vqc_active < vdev_queue_class_min_active(p)) return (p); } /* * If we haven't found a queue, look for one that hasn't reached its * maximum # outstanding i/os. */ for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) { if (avl_numnodes(&vq->vq_class[p].vqc_queued_tree) > 0 && vq->vq_class[p].vqc_active < vdev_queue_class_max_active(spa, p)) return (p); } /* No eligible queued i/os */ return (ZIO_PRIORITY_NUM_QUEUEABLE); } void vdev_queue_init(vdev_t *vd) { vdev_queue_t *vq = &vd->vdev_queue; int max_active_sum; zio_priority_t p; int i; mutex_init(&vq->vq_lock, NULL, MUTEX_DEFAULT, NULL); vq->vq_vdev = vd; avl_create(&vq->vq_active_tree, vdev_queue_offset_compare, sizeof (zio_t), offsetof(struct zio, io_queue_node)); for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) { /* * The synchronous i/o queues are FIFO rather than LBA ordered. * This provides more consistent latency for these i/os, and * they tend to not be tightly clustered anyway so there is * little to no throughput loss. */ boolean_t fifo = (p == ZIO_PRIORITY_SYNC_READ || p == ZIO_PRIORITY_SYNC_WRITE); avl_create(&vq->vq_class[p].vqc_queued_tree, fifo ? vdev_queue_timestamp_compare : vdev_queue_offset_compare, sizeof (zio_t), offsetof(struct zio, io_queue_node)); } /* * A list of buffers which can be used for aggregate I/O, this * avoids the need to allocate them on demand when memory is low. */ list_create(&vq->vq_io_list, sizeof (vdev_io_t), offsetof(vdev_io_t, vi_node)); max_active_sum = zfs_vdev_sync_read_max_active + zfs_vdev_sync_write_max_active + zfs_vdev_async_read_max_active + zfs_vdev_async_write_max_active + zfs_vdev_scrub_max_active; for (i = 0; i < max_active_sum; i++) list_insert_tail(&vq->vq_io_list, zio_vdev_alloc()); } void vdev_queue_fini(vdev_t *vd) { vdev_queue_t *vq = &vd->vdev_queue; vdev_io_t *vi; zio_priority_t p; for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) avl_destroy(&vq->vq_class[p].vqc_queued_tree); avl_destroy(&vq->vq_active_tree); while ((vi = list_head(&vq->vq_io_list)) != NULL) { list_remove(&vq->vq_io_list, vi); zio_vdev_free(vi); } list_destroy(&vq->vq_io_list); mutex_destroy(&vq->vq_lock); } static void vdev_queue_io_add(vdev_queue_t *vq, zio_t *zio) { spa_t *spa = zio->io_spa; spa_stats_history_t *ssh = &spa->spa_stats.io_history; ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE); avl_add(&vq->vq_class[zio->io_priority].vqc_queued_tree, zio); if (ssh->kstat != NULL) { mutex_enter(&ssh->lock); kstat_waitq_enter(ssh->kstat->ks_data); mutex_exit(&ssh->lock); } } static void vdev_queue_io_remove(vdev_queue_t *vq, zio_t *zio) { spa_t *spa = zio->io_spa; spa_stats_history_t *ssh = &spa->spa_stats.io_history; ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE); avl_remove(&vq->vq_class[zio->io_priority].vqc_queued_tree, zio); if (ssh->kstat != NULL) { mutex_enter(&ssh->lock); kstat_waitq_exit(ssh->kstat->ks_data); mutex_exit(&ssh->lock); } } static void vdev_queue_pending_add(vdev_queue_t *vq, zio_t *zio) { spa_t *spa = zio->io_spa; spa_stats_history_t *ssh = &spa->spa_stats.io_history; ASSERT(MUTEX_HELD(&vq->vq_lock)); ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE); vq->vq_class[zio->io_priority].vqc_active++; avl_add(&vq->vq_active_tree, zio); if (ssh->kstat != NULL) { mutex_enter(&ssh->lock); kstat_runq_enter(ssh->kstat->ks_data); mutex_exit(&ssh->lock); } } static void vdev_queue_pending_remove(vdev_queue_t *vq, zio_t *zio) { spa_t *spa = zio->io_spa; spa_stats_history_t *ssh = &spa->spa_stats.io_history; ASSERT(MUTEX_HELD(&vq->vq_lock)); ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE); vq->vq_class[zio->io_priority].vqc_active--; avl_remove(&vq->vq_active_tree, zio); if (ssh->kstat != NULL) { kstat_io_t *ksio = ssh->kstat->ks_data; mutex_enter(&ssh->lock); kstat_runq_exit(ksio); if (zio->io_type == ZIO_TYPE_READ) { ksio->reads++; ksio->nread += zio->io_size; } else if (zio->io_type == ZIO_TYPE_WRITE) { ksio->writes++; ksio->nwritten += zio->io_size; } mutex_exit(&ssh->lock); } } static void vdev_queue_agg_io_done(zio_t *aio) { vdev_queue_t *vq = &aio->io_vd->vdev_queue; vdev_io_t *vi = aio->io_data; if (aio->io_type == ZIO_TYPE_READ) { zio_t *pio; while ((pio = zio_walk_parents(aio)) != NULL) { bcopy((char *)aio->io_data + (pio->io_offset - aio->io_offset), pio->io_data, pio->io_size); } } mutex_enter(&vq->vq_lock); list_insert_tail(&vq->vq_io_list, vi); mutex_exit(&vq->vq_lock); } /* * Compute the range spanned by two i/os, which is the endpoint of the last * (lio->io_offset + lio->io_size) minus start of the first (fio->io_offset). * Conveniently, the gap between fio and lio is given by -IO_SPAN(lio, fio); * thus fio and lio are adjacent if and only if IO_SPAN(lio, fio) == 0. */ #define IO_SPAN(fio, lio) ((lio)->io_offset + (lio)->io_size - (fio)->io_offset) #define IO_GAP(fio, lio) (-IO_SPAN(lio, fio)) static zio_t * vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio) { vdev_io_t *vi; zio_t *first, *last, *aio, *dio, *mandatory, *nio; uint64_t maxgap = 0; uint64_t size; boolean_t stretch = B_FALSE; vdev_queue_class_t *vqc = &vq->vq_class[zio->io_priority]; avl_tree_t *t = &vqc->vqc_queued_tree; enum zio_flag flags = zio->io_flags & ZIO_FLAG_AGG_INHERIT; if (zio->io_flags & ZIO_FLAG_DONT_AGGREGATE) return (NULL); /* * Prevent users from setting the zfs_vdev_aggregation_limit * tuning larger than SPA_MAXBLOCKSIZE. */ zfs_vdev_aggregation_limit = MIN(zfs_vdev_aggregation_limit, SPA_MAXBLOCKSIZE); /* * The synchronous i/o queues are not sorted by LBA, so we can't * find adjacent i/os. These i/os tend to not be tightly clustered, * or too large to aggregate, so this has little impact on performance. */ if (zio->io_priority == ZIO_PRIORITY_SYNC_READ || zio->io_priority == ZIO_PRIORITY_SYNC_WRITE) return (NULL); first = last = zio; if (zio->io_type == ZIO_TYPE_READ) maxgap = zfs_vdev_read_gap_limit; vi = list_head(&vq->vq_io_list); if (vi == NULL) { vi = zio_vdev_alloc(); list_insert_head(&vq->vq_io_list, vi); } /* * We can aggregate I/Os that are sufficiently adjacent and of * the same flavor, as expressed by the AGG_INHERIT flags. * The latter requirement is necessary so that certain * attributes of the I/O, such as whether it's a normal I/O * or a scrub/resilver, can be preserved in the aggregate. * We can include optional I/Os, but don't allow them * to begin a range as they add no benefit in that situation. */ /* * We keep track of the last non-optional I/O. */ mandatory = (first->io_flags & ZIO_FLAG_OPTIONAL) ? NULL : first; /* * Walk backwards through sufficiently contiguous I/Os * recording the last non-option I/O. */ while ((dio = AVL_PREV(t, first)) != NULL && (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags && IO_SPAN(dio, last) <= zfs_vdev_aggregation_limit && IO_GAP(dio, first) <= maxgap) { first = dio; if (mandatory == NULL && !(first->io_flags & ZIO_FLAG_OPTIONAL)) mandatory = first; } /* * Skip any initial optional I/Os. */ while ((first->io_flags & ZIO_FLAG_OPTIONAL) && first != last) { first = AVL_NEXT(t, first); ASSERT(first != NULL); } /* * Walk forward through sufficiently contiguous I/Os. */ while ((dio = AVL_NEXT(t, last)) != NULL && (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags && IO_SPAN(first, dio) <= zfs_vdev_aggregation_limit && IO_GAP(last, dio) <= maxgap) { last = dio; if (!(last->io_flags & ZIO_FLAG_OPTIONAL)) mandatory = last; } /* * Now that we've established the range of the I/O aggregation * we must decide what to do with trailing optional I/Os. * For reads, there's nothing to do. While we are unable to * aggregate further, it's possible that a trailing optional * I/O would allow the underlying device to aggregate with * subsequent I/Os. We must therefore determine if the next * non-optional I/O is close enough to make aggregation * worthwhile. */ if (zio->io_type == ZIO_TYPE_WRITE && mandatory != NULL) { zio_t *nio = last; while ((dio = AVL_NEXT(t, nio)) != NULL && IO_GAP(nio, dio) == 0 && IO_GAP(mandatory, dio) <= zfs_vdev_write_gap_limit) { nio = dio; if (!(nio->io_flags & ZIO_FLAG_OPTIONAL)) { stretch = B_TRUE; break; } } } if (stretch) { /* This may be a no-op. */ dio = AVL_NEXT(t, last); dio->io_flags &= ~ZIO_FLAG_OPTIONAL; } else { while (last != mandatory && last != first) { ASSERT(last->io_flags & ZIO_FLAG_OPTIONAL); last = AVL_PREV(t, last); ASSERT(last != NULL); } } if (first == last) return (NULL); ASSERT(vi != NULL); size = IO_SPAN(first, last); ASSERT3U(size, <=, zfs_vdev_aggregation_limit); aio = zio_vdev_delegated_io(first->io_vd, first->io_offset, vi, size, first->io_type, zio->io_priority, flags | ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_QUEUE, vdev_queue_agg_io_done, NULL); aio->io_timestamp = first->io_timestamp; nio = first; do { dio = nio; nio = AVL_NEXT(t, dio); ASSERT3U(dio->io_type, ==, aio->io_type); if (dio->io_flags & ZIO_FLAG_NODATA) { ASSERT3U(dio->io_type, ==, ZIO_TYPE_WRITE); bzero((char *)aio->io_data + (dio->io_offset - aio->io_offset), dio->io_size); } else if (dio->io_type == ZIO_TYPE_WRITE) { bcopy(dio->io_data, (char *)aio->io_data + (dio->io_offset - aio->io_offset), dio->io_size); } zio_add_child(dio, aio); vdev_queue_io_remove(vq, dio); zio_vdev_io_bypass(dio); zio_execute(dio); } while (dio != last); list_remove(&vq->vq_io_list, vi); return (aio); } static zio_t * vdev_queue_io_to_issue(vdev_queue_t *vq) { zio_t *zio, *aio; zio_priority_t p; avl_index_t idx; vdev_queue_class_t *vqc; - zio_t *search; again: ASSERT(MUTEX_HELD(&vq->vq_lock)); p = vdev_queue_class_to_issue(vq); if (p == ZIO_PRIORITY_NUM_QUEUEABLE) { /* No eligible queued i/os */ return (NULL); } /* * For LBA-ordered queues (async / scrub), issue the i/o which follows * the most recently issued i/o in LBA (offset) order. * * For FIFO queues (sync), issue the i/o with the lowest timestamp. */ vqc = &vq->vq_class[p]; - search = zio_buf_alloc(sizeof (*search)); - search->io_timestamp = 0; - search->io_offset = vq->vq_last_offset + 1; - VERIFY3P(avl_find(&vqc->vqc_queued_tree, search, &idx), ==, NULL); - zio_buf_free(search, sizeof (*search)); + vq->vq_io_search.io_timestamp = 0; + vq->vq_io_search.io_offset = vq->vq_last_offset + 1; + VERIFY3P(avl_find(&vqc->vqc_queued_tree, &vq->vq_io_search, + &idx), ==, NULL); zio = avl_nearest(&vqc->vqc_queued_tree, idx, AVL_AFTER); if (zio == NULL) zio = avl_first(&vqc->vqc_queued_tree); ASSERT3U(zio->io_priority, ==, p); aio = vdev_queue_aggregate(vq, zio); if (aio != NULL) zio = aio; else vdev_queue_io_remove(vq, zio); /* * If the I/O is or was optional and therefore has no data, we need to * simply discard it. We need to drop the vdev queue's lock to avoid a * deadlock that we could encounter since this I/O will complete * immediately. */ if (zio->io_flags & ZIO_FLAG_NODATA) { mutex_exit(&vq->vq_lock); zio_vdev_io_bypass(zio); zio_execute(zio); mutex_enter(&vq->vq_lock); goto again; } vdev_queue_pending_add(vq, zio); vq->vq_last_offset = zio->io_offset; return (zio); } zio_t * vdev_queue_io(zio_t *zio) { vdev_queue_t *vq = &zio->io_vd->vdev_queue; zio_t *nio; if (zio->io_flags & ZIO_FLAG_DONT_QUEUE) return (zio); /* * Children i/os inherent their parent's priority, which might * not match the child's i/o type. Fix it up here. */ if (zio->io_type == ZIO_TYPE_READ) { if (zio->io_priority != ZIO_PRIORITY_SYNC_READ && zio->io_priority != ZIO_PRIORITY_ASYNC_READ && zio->io_priority != ZIO_PRIORITY_SCRUB) zio->io_priority = ZIO_PRIORITY_ASYNC_READ; } else { ASSERT(zio->io_type == ZIO_TYPE_WRITE); if (zio->io_priority != ZIO_PRIORITY_SYNC_WRITE && zio->io_priority != ZIO_PRIORITY_ASYNC_WRITE) zio->io_priority = ZIO_PRIORITY_ASYNC_WRITE; } zio->io_flags |= ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_QUEUE; mutex_enter(&vq->vq_lock); zio->io_timestamp = gethrtime(); vdev_queue_io_add(vq, zio); nio = vdev_queue_io_to_issue(vq); mutex_exit(&vq->vq_lock); if (nio == NULL) return (NULL); if (nio->io_done == vdev_queue_agg_io_done) { zio_nowait(nio); return (NULL); } return (nio); } void vdev_queue_io_done(zio_t *zio) { vdev_queue_t *vq = &zio->io_vd->vdev_queue; zio_t *nio; if (zio_injection_enabled) delay(SEC_TO_TICK(zio_handle_io_delay(zio))); mutex_enter(&vq->vq_lock); vdev_queue_pending_remove(vq, zio); zio->io_delta = gethrtime() - zio->io_timestamp; vq->vq_io_complete_ts = gethrtime(); vq->vq_io_delta_ts = vq->vq_io_complete_ts - zio->io_timestamp; while ((nio = vdev_queue_io_to_issue(vq)) != NULL) { mutex_exit(&vq->vq_lock); if (nio->io_done == vdev_queue_agg_io_done) { zio_nowait(nio); } else { zio_vdev_io_reissue(nio); zio_execute(nio); } mutex_enter(&vq->vq_lock); } mutex_exit(&vq->vq_lock); } #if defined(_KERNEL) && defined(HAVE_SPL) module_param(zfs_vdev_aggregation_limit, int, 0644); MODULE_PARM_DESC(zfs_vdev_aggregation_limit, "Max vdev I/O aggregation size"); module_param(zfs_vdev_read_gap_limit, int, 0644); MODULE_PARM_DESC(zfs_vdev_read_gap_limit, "Aggregate read I/O over gap"); module_param(zfs_vdev_write_gap_limit, int, 0644); MODULE_PARM_DESC(zfs_vdev_write_gap_limit, "Aggregate write I/O over gap"); module_param(zfs_vdev_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_max_active, "Maximum number of active I/Os per vdev"); module_param(zfs_vdev_async_write_active_max_dirty_percent, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_write_active_max_dirty_percent, "Async write concurrency max threshold"); module_param(zfs_vdev_async_write_active_min_dirty_percent, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_write_active_min_dirty_percent, "Async write concurrency min threshold"); module_param(zfs_vdev_async_read_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_read_max_active, "Max active async read I/Os per vdev"); module_param(zfs_vdev_async_read_min_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_read_min_active, "Min active async read I/Os per vdev"); module_param(zfs_vdev_async_write_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_write_max_active, "Max active async write I/Os per vdev"); module_param(zfs_vdev_async_write_min_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_async_write_min_active, "Min active async write I/Os per vdev"); module_param(zfs_vdev_scrub_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_scrub_max_active, "Max active scrub I/Os per vdev"); module_param(zfs_vdev_scrub_min_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_scrub_min_active, "Min active scrub I/Os per vdev"); module_param(zfs_vdev_sync_read_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_sync_read_max_active, "Max active sync read I/Os per vdev"); module_param(zfs_vdev_sync_read_min_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_sync_read_min_active, "Min active sync read I/Os per vdev"); module_param(zfs_vdev_sync_write_max_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_sync_write_max_active, "Max active sync write I/Os per vdev"); module_param(zfs_vdev_sync_write_min_active, int, 0644); MODULE_PARM_DESC(zfs_vdev_sync_write_min_active, "Min active sync write I/Osper vdev"); #endif