Reduce dbuf_find() lock contention

Holding a dbuf is a common operation which can become highly contended
in dbuf_find() when acquiring the dbuf hash mutex.  This is particularly
true on Linux when reading/writing volumes since by default up to 32
threads from the zvol_taskq may be taking a hold of the same dbuf.
This should also be observable on FreeBSD as long as there are enough
processes accessing the volume concurrently.

This is further aggregrated by the fact that only the block id will
be unique when calculating the dbuf hash for a single volume.  The
objset id, object id, and level will be the same for data blocks.
This has been observed to result in a somehwat less than uniform hash
distribution and a longer than expected max hash chain depth (~20)
on a large memory system (256 GB) using volumes.

This commit improves the siutation by switching the hash mutex to
an rwlock to allow concurrent lookups, and increasing DBUF_RWLOCKS
from 2048 to 8192 to further reduce the odds of a hash collision.

Reviewed-by: Tony Hutter <[email protected]>
Reviewed-by: Alexander Motin <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #13405 

1 files changed, 3 insertions, 4 deletions

diff --git a/include/sys/dbuf.h b/include/sys/dbuf.h
index 9017cf345..deaab82b7 100644
--- a/include/sys/dbuf.h
+++ b/include/sys/dbuf.h
@@ -321,13 +321,12 @@ typedef struct dmu_buf_impl {
 	uint8_t db_dirtycnt;
 } dmu_buf_impl_t;
 
-/* Note: the dbuf hash table is exposed only for the mdb module */
-#define	DBUF_MUTEXES 2048
-#define	DBUF_HASH_MUTEX(h, idx) (&(h)->hash_mutexes[(idx) & (DBUF_MUTEXES-1)])
+#define	DBUF_RWLOCKS 8192
+#define	DBUF_HASH_RWLOCK(h, idx) (&(h)->hash_rwlocks[(idx) & (DBUF_RWLOCKS-1)])
 typedef struct dbuf_hash_table {
 	uint64_t hash_table_mask;
 	dmu_buf_impl_t **hash_table;
-	kmutex_t hash_mutexes[DBUF_MUTEXES] ____cacheline_aligned;
+	krwlock_t hash_rwlocks[DBUF_RWLOCKS] ____cacheline_aligned;
 } dbuf_hash_table_t;
 
 typedef void (*dbuf_prefetch_fn)(void *, boolean_t);