Retire z_nr_znodes

Added in ab26409db7 ("Linux 3.1 compat, super_block->s_shrink"), with
the only consumer which needed the count getting retired in 066e825221
("Linux compat: Minimum kernel version 3.10").

The counter gets in the way of not maintaining the list to begin with.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Closes #15274

module/os/linux/zfs/zfs_ctldir.c

@@ -537,7 +537,6 @@ zfsctl_inode_alloc(zfsvfs_t *zfsvfs, uint64_t id,
 
 	mutex_enter(&zfsvfs->z_znodes_lock);
 	list_insert_tail(&zfsvfs->z_all_znodes, zp);
-	zfsvfs->z_nr_znodes++;
 	membar_producer();
 	mutex_exit(&zfsvfs->z_znodes_lock);
 

module/os/linux/zfs/zfs_vfsops.c

@@ -1330,12 +1330,11 @@ zfsvfs_teardown(zfsvfs_t *zfsvfs, boolean_t unmounting)
 		 * may add the parents of dir-based xattrs to the taskq
 		 * so we want to wait for these.
 		 *
-		 * We can safely read z_nr_znodes without locking because the
-		 * VFS has already blocked operations which add to the
-		 * z_all_znodes list and thus increment z_nr_znodes.
+		 * We can safely check z_all_znodes for being empty because the
+		 * VFS has already blocked operations which add to it.
 		 */
 		int round = 0;
-		while (zfsvfs->z_nr_znodes > 0) {
+		while (!list_is_empty(&zfsvfs->z_all_znodes)) {
 			taskq_wait_outstanding(dsl_pool_zrele_taskq(
 			    dmu_objset_pool(zfsvfs->z_os)), 0);
 			if (++round > 1 && !unmounting)

module/os/linux/zfs/zfs_znode.c

@@ -390,7 +390,6 @@ zfs_inode_destroy(struct inode *ip)
 	mutex_enter(&zfsvfs->z_znodes_lock);
 	if (list_link_active(&zp->z_link_node)) {
 		list_remove(&zfsvfs->z_all_znodes, zp);
-		zfsvfs->z_nr_znodes--;
 	}
 	mutex_exit(&zfsvfs->z_znodes_lock);
 
@@ -641,7 +640,6 @@ zfs_znode_alloc(zfsvfs_t *zfsvfs, dmu_buf_t *db, int blksz,
 
 	mutex_enter(&zfsvfs->z_znodes_lock);
 	list_insert_tail(&zfsvfs->z_all_znodes, zp);
-	zfsvfs->z_nr_znodes++;
 	mutex_exit(&zfsvfs->z_znodes_lock);
 
 	if (links > 0)