Lines Matching refs:node
13 * When UBIFS is mounted, it checks a flag on the master node to determine if
31 * Hence, if UBIFS finds a corrupted node at offset X, it expects only the min.
81 * get_master_node - get the last valid master node allowing for corruption.
85 * @mst: master node, if found, is returned here
89 * returns the last valid master node allowing for one area of corruption.
91 * that it is the result of an unclean unmount while the master node was being
93 * master node.
112 /* Find the first position that is definitely not a node */
125 /* See if there was a valid master node before that */
143 * trying to write a master node.
152 dbg_rcvry("found a master node at %d:%d", lnum, offs);
186 * write_rcvrd_mst_node - write recovered master node.
188 * @mst: master node
219 * ubifs_recover_master_node - recover the master node.
222 * This function recovers the master node from corruption that may occur due to
277 * there must be only one master node in the first LEB
297 ubifs_msg(c, "recovered master node from LEB %d",
312 * We had to recover the master node, which means there was an
313 * unclean reboot. However, it is possible that the master node
317 * 1. UBIFS was cleanly unmounted, so the master node is clean
319 * node in the first (%UBIFS_MST_LNUM). A power cut happens,
323 * recover the master node from the second LEB
327 * 4. However, this master node (@c->mst_node) is marked as
330 * node when it is re-mounter R/W later.
332 * Thus, to force the recovery by marking the master node as
337 /* Write the recovered master node */
352 ubifs_err(c, "failed to recover master node");
354 ubifs_err(c, "dumping first master node");
358 ubifs_err(c, "dumping second master node");
367 * ubifs_write_rcvrd_mst_node - write the recovered master node.
370 * This function writes the master node that was recovered during mounting in
454 * This function ensures that the corrupted node at @offs is the last thing
464 /* Check for empty space after the corrupt node's common header */
476 /* Now we know the corrupt node's length we can skip over it */
496 /* Get the end offset of the last node we are keeping */
570 ch = snod->node;
574 dbg_rcvry("dropping grouped node at %d:%d",
584 * drop_last_node - drop the last node.
589 * node of the scanned LEB.
599 dbg_rcvry("dropping last node at %d:%d",
649 /* A valid node, and not a padding node */
661 /* Padding bytes or a valid padding node */
721 * contains valid nodes or the end of a valid node, and the
731 * padding node there? We can, and this works fine with one
751 * because of the padding node which we added to LEB Y. The
793 * @lnum: LEB number of commit start node
794 * @offs: offset of commit start node
817 ubifs_err(c, "Not a valid node");
821 ubifs_err(c, "Not a CS node, type is %d", cs_node->ch.node_type);
825 ubifs_err(c, "CS node cmt_no %llu != current cmt_no %llu",
1008 /* A valid node, and not a padding node */
1020 /* Padding bytes or a valid padding node */
1146 * written to the master node on unmounting. In the case of an unclean unmount
1147 * the value of gc_lnum recorded in the master node is out of date and cannot
1332 * @key: node key
1333 * @deletion: node is for a deletion
1341 * the size from the inode node, the maximum size of any data node (also altered
1342 * by truncations) and a flag indicating a inode number for which no inode node
1411 /* Locate the inode node LEB number and offset */
1417 * If the size recorded on the inode node is greater than the size that