Cap metaslab memory usage

On systems with large amounts of storage and high fragmentation, a huge 
amount of space can be used by storing metaslab range trees. Since 
metaslabs are only unloaded during a txg sync, and only if they have 
been inactive for 8 txgs, it is possible to get into a state where all 
of the system's memory is consumed by range trees and metaslabs, and 
txgs cannot sync. While ZFS knows how to evict ARC data when needed, 
it has no such mechanism for range tree data. This can result in boot 
hangs for some system configurations.

First, we add the ability to unload metaslabs outside of syncing 
context. Second, we store a multilist of all loaded metaslabs, sorted 
by their selection txg, so we can quickly identify the oldest 
metaslabs.  We use a multilist to reduce lock contention during heavy 
write workloads. Finally, we add logic that will unload a metaslab 
when we're loading a new metaslab, if we're using more than a certain 
fraction of the available memory on range trees.

Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Sebastien Roy <sebastien.roy@delphix.com>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes #9128

include/sys/metaslab_impl.h

@@ -36,6 +36,7 @@
 #include <sys/vdev.h>
 #include <sys/txg.h>
 #include <sys/avl.h>
+#include <sys/multilist.h>
 
 #ifdef	__cplusplus
 extern "C" {
@@ -194,6 +195,12 @@ struct metaslab_class {
 	uint64_t		mc_space;	/* total space (alloc + free) */
 	uint64_t		mc_dspace;	/* total deflated space */
 	uint64_t		mc_histogram[RANGE_TREE_HISTOGRAM_SIZE];
+
+	/*
+	 * List of all loaded metaslabs in the class, sorted in order of most
+	 * recent use.
+	 */
+	multilist_t		*mc_metaslab_txg_list;
 };
 
 /*
@@ -378,6 +385,7 @@ struct metaslab {
 	range_tree_t	*ms_allocating[TXG_SIZE];
 	range_tree_t	*ms_allocatable;
 	uint64_t	ms_allocated_this_txg;
+	uint64_t	ms_allocating_total;
 
 	/*
 	 * The following range trees are accessed only from syncing context.
@@ -508,6 +516,10 @@ struct metaslab {
 	avl_node_t	ms_group_node;	/* node in metaslab group tree	*/
 	txg_node_t	ms_txg_node;	/* per-txg dirty metaslab links	*/
 	avl_node_t	ms_spa_txg_node; /* node in spa_metaslabs_by_txg */
+	/*
+	 * Node in metaslab class's selected txg list
+	 */
+	multilist_node_t	ms_class_txg_node;
 
 	/*
 	 * Allocs and frees that are committed to the vdev log spacemap but