Lines Matching full:nodes

35  * The end result is that we have one or more new nodes being written -
37  * an interior node) which will make all these new nodes visible.
39  * Additionally, as we split/rewrite nodes we free the old nodes - but the old
40  * nodes can't be freed (their space on disk can't be reclaimed) until the
42  * until then, the old nodes are still reachable on disk.
74 	 * The update that made the new nodes visible was a regular update to an
76 	 * until the new nodes we created are finished writing, so we block @b
84 	 * We may be freeing nodes that were dirty, and thus had journal entries
91 	/* Preallocated nodes we reserve when we start the update: */
97 	/* Nodes being freed: */
102 	/* Nodes being added: */
107 	/* New nodes, that will be made reachable by this update: */
207 	 * Number of nodes we might have to allocate in a worst case btree  in btree_update_reserve_required()