Another set of vdev queue optimizations.

Switch FIFO queues (SYNC/TRIM) and active queue of vdev queue from
time-sorted AVL-trees to simple lists.  AVL-trees are too expensive
for such a simple task.  To change I/O priority without searching
through the trees, add io_queue_state field to struct zio.

To not check number of queued I/Os for each priority add vq_cqueued
bitmap to struct vdev_queue.  Update it when adding/removing I/Os.
Make vq_cactive a separate array instead of struct vdev_queue_class
member.  Together those allow to avoid lots of cache misses when
looking for work in vdev_queue_class_to_issue().

Introduce deadline of ~0.5s for LBA-sorted queues.  Before this I
saw some I/Os waiting in a queue for up to 8 seconds and possibly
more due to starvation.  With this change I no longer see it.  I
had to slightly more complicate the comparison function, but since
it uses all the same cache lines the difference is minimal.  For a
sequential I/Os the new code in vdev_queue_io_to_issue() actually
often uses more simple avl_first(), falling back to avl_find() and
avl_nearest() only when needed.

Arrange members in struct zio to access only one cache line when
searching through vdev queues.  While there, remove io_alloc_node,
reusing the io_queue_node instead.  Those two are never used same
time.

Remove zfs_vdev_aggregate_trim parameter.  It was disabled for 4
years since implemented, while still wasted time maintaining the
offset-sorted tree of TRIM requests.  Just remove the tree.

Remove locking from txg_all_lists_empty().  It is racy by design,
while 2 pair of locks/unlocks take noticeable time under the vdev
queue lock.

With these changes in my tests with volblocksize=4KB I measure vdev
queue lock spin time reduction by 50% on read and 75% on write.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by:	Alexander Motin <mav@FreeBSD.org>
Sponsored by:	iXsystems, Inc.
Closes #14925

include/sys/vdev.h

@@ -164,8 +164,9 @@ extern zio_t *vdev_queue_io(zio_t *zio);
 extern void vdev_queue_io_done(zio_t *zio);
 extern void vdev_queue_change_io_priority(zio_t *zio, zio_priority_t priority);
 
-extern int vdev_queue_length(vdev_t *vd);
+extern uint32_t vdev_queue_length(vdev_t *vd);
 extern uint64_t vdev_queue_last_offset(vdev_t *vd);
+extern uint64_t vdev_queue_class_length(vdev_t *vq, zio_priority_t p);
 
 extern void vdev_config_dirty(vdev_t *vd);
 extern void vdev_config_clean(vdev_t *vd);

include/sys/vdev_impl.h

@@ -130,27 +130,24 @@ typedef const struct vdev_ops {
 /*
  * Virtual device properties
  */
-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;
+typedef union vdev_queue_class {
+	list_t		vqc_list;
+	avl_tree_t	vqc_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;
 	avl_tree_t	vq_read_offset_tree;
 	avl_tree_t	vq_write_offset_tree;
-	avl_tree_t	vq_trim_offset_tree;
 	uint64_t	vq_last_offset;
 	zio_priority_t	vq_last_prio;	/* Last sent I/O priority. */
+	uint32_t	vq_cqueued;	/* Classes with queued I/Os. */
+	uint32_t	vq_cactive[ZIO_PRIORITY_NUM_QUEUEABLE];
+	uint32_t	vq_active;	/* Number of active I/Os. */
 	uint32_t	vq_ia_active;	/* Active interactive I/Os. */
 	uint32_t	vq_nia_credit;	/* Non-interactive I/Os credit. */
+	list_t		vq_active_list;	/* List of active I/Os. */
 	hrtime_t	vq_io_complete_ts; /* time last i/o completed */
 	hrtime_t	vq_io_delta_ts;
 	zio_t		vq_io_search; /* used as local for stack reduction */

include/sys/zio.h

@@ -436,6 +436,12 @@ typedef struct zio_link {
 	list_node_t	zl_child_node;
 } zio_link_t;
 
+enum zio_qstate {
+	ZIO_QS_NONE = 0,
+	ZIO_QS_QUEUED,
+	ZIO_QS_ACTIVE,
+};
+
 struct zio {
 	/* Core information about this I/O */
 	zbookmark_phys_t	io_bookmark;
@@ -479,6 +485,12 @@ struct zio {
 	const zio_vsd_ops_t *io_vsd_ops;
 	metaslab_class_t *io_metaslab_class;	/* dva throttle class */
 
+	enum zio_qstate	io_queue_state;	/* vdev queue state */
+	union {
+		list_node_t l;
+		avl_node_t a;
+	} io_queue_node ____cacheline_aligned;	/* allocator and vdev queues */
+	avl_node_t	io_offset_node;	/* vdev offset queues */
 	uint64_t	io_offset;
 	hrtime_t	io_timestamp;	/* submitted at */
 	hrtime_t	io_queued_timestamp;
@@ -486,9 +498,6 @@ struct zio {
 	hrtime_t	io_delta;	/* vdev queue service delta */
 	hrtime_t	io_delay;	/* Device access time (disk or */
 					/* file). */
-	avl_node_t	io_queue_node;
-	avl_node_t	io_offset_node;
-	avl_node_t	io_alloc_node;
 	zio_alloc_list_t 	io_alloc_list;
 
 	/* Internal pipeline state */