From 125c4cf9f37c98fed2c08229b31358cfec63dcf6 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Sep 2014 21:22:14 +0100 Subject: [PATCH 1/2] Btrfs: set inode's logged_trans/last_log_commit after ranged fsync When a ranged fsync finishes if there are still extent maps in the modified list, still set the inode's logged_trans and last_log_commit. This is important in case an inode is fsync'ed and unlinked in the same transaction, to ensure its inode ref gets deleted from the log and the respective dentries in its parent are deleted too from the log (if the parent directory was fsync'ed in the same transaction). Instead make btrfs_inode_in_log() return false if the list of modified extent maps isn't empty. This is an incremental on top of the v4 version of the patch: "Btrfs: fix fsync data loss after a ranged fsync" which was added to its v5, but didn't make it on time. Signed-off-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/btrfs_inode.h | 13 +++++++++++-- fs/btrfs/tree-log.c | 14 ++------------ 2 files changed, 13 insertions(+), 14 deletions(-) diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 43527fd78825..56b8522d5767 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -234,8 +234,17 @@ static inline int btrfs_inode_in_log(struct inode *inode, u64 generation) BTRFS_I(inode)->last_sub_trans <= BTRFS_I(inode)->last_log_commit && BTRFS_I(inode)->last_sub_trans <= - BTRFS_I(inode)->root->last_log_commit) - return 1; + BTRFS_I(inode)->root->last_log_commit) { + /* + * After a ranged fsync we might have left some extent maps + * (that fall outside the fsync's range). So return false + * here if the list isn't empty, to make sure btrfs_log_inode() + * will be called and process those extent maps. + */ + smp_mb(); + if (list_empty(&BTRFS_I(inode)->extent_tree.modified_extents)) + return 1; + } return 0; } diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index d296efe2d3e7..1d1ba083ca6e 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -4093,18 +4093,8 @@ log_extents: } } - write_lock(&em_tree->lock); - /* - * If we're doing a ranged fsync and there are still modified extents - * in the list, we must run on the next fsync call as it might cover - * those extents (a full fsync or an fsync for other range). - */ - if (list_empty(&em_tree->modified_extents)) { - BTRFS_I(inode)->logged_trans = trans->transid; - BTRFS_I(inode)->last_log_commit = - BTRFS_I(inode)->last_sub_trans; - } - write_unlock(&em_tree->lock); + BTRFS_I(inode)->logged_trans = trans->transid; + BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans; out_unlock: if (unlikely(err)) btrfs_put_logged_extents(&logged_list); From 0f23ae74f589304bf33233f85737f4fd368549eb Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Thu, 18 Sep 2014 07:49:05 -0700 Subject: [PATCH 2/2] Revert "Btrfs: device_list_add() should not update list when mounted" This reverts commit b96de000bc8bc9688b3a2abea4332bd57648a49f. This commit is triggering failures to mount by subvolume id in some configurations. The main problem is how many different ways this scanning function is used, both for scanning while mounted and unmounted. A proper cleanup is too big for late rcs. For now, just revert the commit and we'll put a better fix into a later merge window. Signed-off-by: Chris Mason --- fs/btrfs/volumes.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 340a92d08e84..2c2d6d1d8eee 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -529,12 +529,12 @@ static noinline int device_list_add(const char *path, */ /* - * As of now don't allow update to btrfs_fs_device through - * the btrfs dev scan cli, after FS has been mounted. + * For now, we do allow update to btrfs_fs_device through the + * btrfs dev scan cli after FS has been mounted. We're still + * tracking a problem where systems fail mount by subvolume id + * when we reject replacement on a mounted FS. */ - if (fs_devices->opened) { - return -EBUSY; - } else { + if (!fs_devices->opened && found_transid < device->generation) { /* * That is if the FS is _not_ mounted and if you * are here, that means there is more than one @@ -542,8 +542,7 @@ static noinline int device_list_add(const char *path, * with larger generation number or the last-in if * generation are equal. */ - if (found_transid < device->generation) - return -EEXIST; + return -EEXIST; } name = rcu_string_strdup(path, GFP_NOFS);