Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs update from Chris Mason:
 "From a feature point of view, most of the code here comes from Miao
  Xie and others at Fujitsu to implement scrubbing and replacing devices
  on raid56.  This has been in development for a while, and it's a big
  improvement.

  Filipe and Josef have a great assortment of fixes, many of which solve
  problems corruptions either after a crash or in error conditions.  I
  still have a round two from Filipe for next week that solves
  corruptions with discard and block group removal"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (62 commits)
  Btrfs: make get_caching_control unconditionally return the ctl
  Btrfs: fix unprotected deletion from pending_chunks list
  Btrfs: fix fs mapping extent map leak
  Btrfs: fix memory leak after block remove + trimming
  Btrfs: make btrfs_abort_transaction consider existence of new block groups
  Btrfs: fix race between writing free space cache and trimming
  Btrfs: fix race between fs trimming and block group remove/allocation
  Btrfs, replace: enable dev-replace for raid56
  Btrfs: fix freeing used extents after removing empty block group
  Btrfs: fix crash caused by block group removal
  Btrfs: fix invalid block group rbtree access after bg is removed
  Btrfs, raid56: fix use-after-free problem in the final device replace procedure on raid56
  Btrfs, replace: write raid56 parity into the replace target device
  Btrfs, replace: write dirty pages into the replace target device
  Btrfs, raid56: support parity scrub on raid56
  Btrfs, raid56: use a variant to record the operation type
  Btrfs, scrub: repair the common data on RAID5/6 if it is corrupted
  Btrfs, raid56: don't change bbio and raid_map
  Btrfs: remove unnecessary code of stripe_index assignment in __btrfs_map_block
  Btrfs: remove noused bbio_ret in __btrfs_map_block in condition
  ...

1 files changed, 40 insertions, 10 deletions

diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 286213cec861..9a02da16f2be 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -2599,12 +2599,14 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	index2 = root_log_ctx.log_transid % 2;
 	if (atomic_read(&log_root_tree->log_commit[index2])) {
 		blk_finish_plug(&plug);
-		btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
+		ret = btrfs_wait_marked_extents(log, &log->dirty_log_pages,
+						mark);
+		btrfs_wait_logged_extents(trans, log, log_transid);
 		wait_log_commit(trans, log_root_tree,
 				root_log_ctx.log_transid);
-		btrfs_free_logged_extents(log, log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
-		ret = root_log_ctx.log_ret;
+		if (!ret)
+			ret = root_log_ctx.log_ret;
 		goto out;
 	}
 	ASSERT(root_log_ctx.log_transid == log_root_tree->log_transid);
@@ -2641,11 +2643,18 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 		mutex_unlock(&log_root_tree->log_mutex);
 		goto out_wake_log_root;
 	}
-	btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
-	btrfs_wait_marked_extents(log_root_tree,
-				  &log_root_tree->dirty_log_pages,
-				  EXTENT_NEW | EXTENT_DIRTY);
-	btrfs_wait_logged_extents(log, log_transid);
+	ret = btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
+	if (!ret)
+		ret = btrfs_wait_marked_extents(log_root_tree,
+						&log_root_tree->dirty_log_pages,
+						EXTENT_NEW | EXTENT_DIRTY);
+	if (ret) {
+		btrfs_set_log_full_commit(root->fs_info, trans);
+		btrfs_free_logged_extents(log, log_transid);
+		mutex_unlock(&log_root_tree->log_mutex);
+		goto out_wake_log_root;
+	}
+	btrfs_wait_logged_extents(trans, log, log_transid);
 
 	btrfs_set_super_log_root(root->fs_info->super_for_commit,
 				log_root_tree->node->start);
@@ -3626,6 +3635,12 @@ static int wait_ordered_extents(struct btrfs_trans_handle *trans,
 			    test_bit(BTRFS_ORDERED_IOERR, &ordered->flags)));
 
 		if (test_bit(BTRFS_ORDERED_IOERR, &ordered->flags)) {
+			/*
+			 * Clear the AS_EIO/AS_ENOSPC flags from the inode's
+			 * i_mapping flags, so that the next fsync won't get
+			 * an outdated io error too.
+			 */
+			btrfs_inode_check_errors(inode);
 			*ordered_io_error = true;
 			break;
 		}
@@ -3766,7 +3781,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
 
-	btrfs_set_token_file_extent_generation(leaf, fi, em->generation,
+	btrfs_set_token_file_extent_generation(leaf, fi, trans->transid,
 					       &token);
 	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
 		btrfs_set_token_file_extent_type(leaf, fi,
@@ -3963,7 +3978,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&BTRFS_I(inode)->log_mutex);
 
-	btrfs_get_logged_extents(inode, &logged_list);
+	btrfs_get_logged_extents(inode, &logged_list, start, end);
 
 	/*
 	 * a brute force approach to making sure we get the most uptodate
@@ -4089,6 +4104,21 @@ log_extents:
 	btrfs_release_path(path);
 	btrfs_release_path(dst_path);
 	if (fast_search) {
+		/*
+		 * Some ordered extents started by fsync might have completed
+		 * before we collected the ordered extents in logged_list, which
+		 * means they're gone, not in our logged_list nor in the inode's
+		 * ordered tree. We want the application/user space to know an
+		 * error happened while attempting to persist file data so that
+		 * it can take proper action. If such error happened, we leave
+		 * without writing to the log tree and the fsync must report the
+		 * file data write error and not commit the current transaction.
+		 */
+		err = btrfs_inode_check_errors(inode);
+		if (err) {
+			ctx->io_err = err;
+			goto out_unlock;
+		}
 		ret = btrfs_log_changed_extents(trans, root, inode, dst_path,
 						&logged_list, ctx);
 		if (ret) {