Message ID | 1415217364-32108-1-git-send-email-fdmanana@suse.com (mailing list archive) |
---|---|
State | Changes Requested |
Headers | show |
On 11/05/2014 02:56 PM, Filipe Manana wrote: > We have a race while deleting unused block groups that causes extents written > by past generations/transactions to be rewritten by the current transaction > before that transaction is committed. The steps that lead to this issue: > > 1) At transaction N one or more block groups became unused and we added them > to the list fs_info->unused_bgs; > > 2) While still at transaction N we write btree extents to block group X and the > transaction is committed; > > 3) The cleaner kthread is awaken and calls btrfs_delete_unused_bgs() to go through > the list fs_info->unused_bgs and remove unused block groups; > > 4) Transaction N + 1 starts; > > 5) At transaction N + 1, block group X becomes unused and is added to the list > fs_info->unused_bgs - this implies delayed refs were run, so we had the > following function calls: btrfs_run_delayed_refs() -> __btrfs_free_extent() > -> update_block_group(). The update_block_group() function grabs the lock > fs_info->unused_bgs_lock, adds block group X to fs_info->unused_bgs and > releases that lock; > > 6) The cleaner kthread, while at btrfs_delete_unused_bgs(), sees block group X > added by transaction N + 1 because it's doing a loop that finishes only when > the list fs_info->unused_bgs is empty and locks and unlocks the spinlock > fs_info->unused_bgs_lock on each iteration. So it deletes the block group > and its respective chunk is released. Even if it didn't do the lock/unlock > per iteration, it could still see block group X in the list, because the > cleaner kthread might call btrfs_delete_unused_bgs() multiple times (for > example if there are several snapshots to delete); > > 7) A new block group X' is created for data, and it's associated to the same chunk > that block group X was associated to; > > 8) Extents from block group X' are allocated for file data and for example an fsync > makes the file data be effectively written to disk; > > 9) A crash/reboot happens before transaction N + 1 is committed; > > 10) On the next mount, we will read extents from block group/chunk X but they no > longer have valid btree nodes/leafs - they have instead file data, and therefore > all sorts of errors will happen. > > So fix this by ensuring the cleaner kthread can never delete a block group that > became unused in the current transaction, that is, only delete block groups that > were added to the unused_bgs list by past transactions. > > Signed-off-by: Filipe Manana <fdmanana@suse.com> > --- > fs/btrfs/ctree.h | 1 + > fs/btrfs/disk-io.c | 1 + > fs/btrfs/extent-tree.c | 5 +++-- > fs/btrfs/transaction.c | 5 +++++ > 4 files changed, 10 insertions(+), 2 deletions(-) > > diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h > index 36f82ba..a5e471a 100644 > --- a/fs/btrfs/ctree.h > +++ b/fs/btrfs/ctree.h > @@ -1726,6 +1726,7 @@ struct btrfs_fs_info { > > spinlock_t unused_bgs_lock; > struct list_head unused_bgs; > + struct list_head unused_bgs_to_clean; > > /* For btrfs to record security options */ > struct security_mnt_opts security_opts; > diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c > index 2409718..702bbdf 100644 > --- a/fs/btrfs/disk-io.c > +++ b/fs/btrfs/disk-io.c > @@ -2243,6 +2243,7 @@ int open_ctree(struct super_block *sb, > INIT_LIST_HEAD(&fs_info->space_info); > INIT_LIST_HEAD(&fs_info->tree_mod_seq_list); > INIT_LIST_HEAD(&fs_info->unused_bgs); > + INIT_LIST_HEAD(&fs_info->unused_bgs_to_clean); > btrfs_mapping_init(&fs_info->mapping_tree); > btrfs_init_block_rsv(&fs_info->global_block_rsv, > BTRFS_BLOCK_RSV_GLOBAL); > diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c > index 744b580..bc1c0b7 100644 > --- a/fs/btrfs/extent-tree.c > +++ b/fs/btrfs/extent-tree.c > @@ -8858,6 +8858,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info) > up_write(&info->commit_root_sem); > > spin_lock(&info->unused_bgs_lock); > + list_splice_init(&info->unused_bgs_to_clean, &info->unused_bgs); > while (!list_empty(&info->unused_bgs)) { > block_group = list_first_entry(&info->unused_bgs, > struct btrfs_block_group_cache, > @@ -9466,10 +9467,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) > return; > > spin_lock(&fs_info->unused_bgs_lock); > - while (!list_empty(&fs_info->unused_bgs)) { > + while (!list_empty(&fs_info->unused_bgs_to_clean)) { > u64 start, end; > > - block_group = list_first_entry(&fs_info->unused_bgs, > + block_group = list_first_entry(&fs_info->unused_bgs_to_clean, > struct btrfs_block_group_cache, > bg_list); > space_info = block_group->space_info; > diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c > index 396ae8b..86d7cf5 100644 > --- a/fs/btrfs/transaction.c > +++ b/fs/btrfs/transaction.c > @@ -1937,6 +1937,11 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, > > btrfs_prepare_extent_commit(trans, root); > > + spin_lock(&root->fs_info->unused_bgs_lock); > + list_splice_init(&root->fs_info->unused_bgs, > + &root->fs_info->unused_bgs_to_clean); > + spin_unlock(&root->fs_info->unused_bgs_lock); > + > cur_trans = root->fs_info->running_transaction; > > btrfs_set_root_node(&root->fs_info->tree_root->root_item, > Eesh that's horrible, thanks for catching it. Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 11/05/2014 02:56 PM, Filipe Manana wrote: > We have a race while deleting unused block groups that causes extents written > by past generations/transactions to be rewritten by the current transaction > before that transaction is committed. The steps that lead to this issue: > > 1) At transaction N one or more block groups became unused and we added them > to the list fs_info->unused_bgs; > > 2) While still at transaction N we write btree extents to block group X and the > transaction is committed; > > 3) The cleaner kthread is awaken and calls btrfs_delete_unused_bgs() to go through > the list fs_info->unused_bgs and remove unused block groups; > > 4) Transaction N + 1 starts; > > 5) At transaction N + 1, block group X becomes unused and is added to the list > fs_info->unused_bgs - this implies delayed refs were run, so we had the > following function calls: btrfs_run_delayed_refs() -> __btrfs_free_extent() > -> update_block_group(). The update_block_group() function grabs the lock > fs_info->unused_bgs_lock, adds block group X to fs_info->unused_bgs and > releases that lock; > > 6) The cleaner kthread, while at btrfs_delete_unused_bgs(), sees block group X > added by transaction N + 1 because it's doing a loop that finishes only when > the list fs_info->unused_bgs is empty and locks and unlocks the spinlock > fs_info->unused_bgs_lock on each iteration. So it deletes the block group > and its respective chunk is released. Even if it didn't do the lock/unlock > per iteration, it could still see block group X in the list, because the > cleaner kthread might call btrfs_delete_unused_bgs() multiple times (for > example if there are several snapshots to delete); > > 7) A new block group X' is created for data, and it's associated to the same chunk > that block group X was associated to; > Actually this can't happen, we search the commit root for a free dev extent, so if block group X` get's mapped to a dev extent that was deleted in the same transaction as it was free'd in then that is a different problem. > 8) Extents from block group X' are allocated for file data and for example an fsync > makes the file data be effectively written to disk; > Also if a new block group is allocated fsync() will trigger a full transaction commit. So thinking about this more I'm not entirely sure there is actually a problem here. Did you observe this issue? Are you sure it's because of this change and not just exacerbated by it? Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Nov 5, 2014 at 8:33 PM, Josef Bacik <jbacik@fb.com> wrote: > On 11/05/2014 02:56 PM, Filipe Manana wrote: >> >> We have a race while deleting unused block groups that causes extents >> written >> by past generations/transactions to be rewritten by the current >> transaction >> before that transaction is committed. The steps that lead to this issue: >> >> 1) At transaction N one or more block groups became unused and we added >> them >> to the list fs_info->unused_bgs; >> >> 2) While still at transaction N we write btree extents to block group X >> and the >> transaction is committed; >> >> 3) The cleaner kthread is awaken and calls btrfs_delete_unused_bgs() to go >> through >> the list fs_info->unused_bgs and remove unused block groups; >> >> 4) Transaction N + 1 starts; >> >> 5) At transaction N + 1, block group X becomes unused and is added to the >> list >> fs_info->unused_bgs - this implies delayed refs were run, so we had >> the >> following function calls: btrfs_run_delayed_refs() -> >> __btrfs_free_extent() >> -> update_block_group(). The update_block_group() function grabs the >> lock >> fs_info->unused_bgs_lock, adds block group X to fs_info->unused_bgs >> and >> releases that lock; >> >> 6) The cleaner kthread, while at btrfs_delete_unused_bgs(), sees block >> group X >> added by transaction N + 1 because it's doing a loop that finishes >> only when >> the list fs_info->unused_bgs is empty and locks and unlocks the >> spinlock >> fs_info->unused_bgs_lock on each iteration. So it deletes the block >> group >> and its respective chunk is released. Even if it didn't do the >> lock/unlock >> per iteration, it could still see block group X in the list, because >> the >> cleaner kthread might call btrfs_delete_unused_bgs() multiple times >> (for >> example if there are several snapshots to delete); >> >> 7) A new block group X' is created for data, and it's associated to the >> same chunk >> that block group X was associated to; >> > > Actually this can't happen, we search the commit root for a free dev extent, > so if block group X` get's mapped to a dev extent that was deleted in the > same transaction as it was free'd in then that is a different problem. Hum, yep I missed the detail that when looking for free device extents we use the commit root. In that case I don't think anymore that there's a problem. Thanks for looking at it. > >> 8) Extents from block group X' are allocated for file data and for example >> an fsync >> makes the file data be effectively written to disk; >> > > Also if a new block group is allocated fsync() will trigger a full > transaction commit. So thinking about this more I'm not entirely sure there > is actually a problem here. Did you observe this issue? Are you sure it's > because of this change and not just exacerbated by it? Thanks, > > Josef > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html
On 11/05/2014 04:03 PM, Filipe David Manana wrote: > On Wed, Nov 5, 2014 at 8:33 PM, Josef Bacik <jbacik@fb.com> wrote: >> On 11/05/2014 02:56 PM, Filipe Manana wrote: >>> >>> We have a race while deleting unused block groups that causes extents >>> written >>> by past generations/transactions to be rewritten by the current >>> transaction >>> before that transaction is committed. The steps that lead to this issue: >>> >>> 1) At transaction N one or more block groups became unused and we added >>> them >>> to the list fs_info->unused_bgs; >>> >>> 2) While still at transaction N we write btree extents to block group X >>> and the >>> transaction is committed; >>> >>> 3) The cleaner kthread is awaken and calls btrfs_delete_unused_bgs() to go >>> through >>> the list fs_info->unused_bgs and remove unused block groups; >>> >>> 4) Transaction N + 1 starts; >>> >>> 5) At transaction N + 1, block group X becomes unused and is added to the >>> list >>> fs_info->unused_bgs - this implies delayed refs were run, so we had >>> the >>> following function calls: btrfs_run_delayed_refs() -> >>> __btrfs_free_extent() >>> -> update_block_group(). The update_block_group() function grabs the >>> lock >>> fs_info->unused_bgs_lock, adds block group X to fs_info->unused_bgs >>> and >>> releases that lock; >>> >>> 6) The cleaner kthread, while at btrfs_delete_unused_bgs(), sees block >>> group X >>> added by transaction N + 1 because it's doing a loop that finishes >>> only when >>> the list fs_info->unused_bgs is empty and locks and unlocks the >>> spinlock >>> fs_info->unused_bgs_lock on each iteration. So it deletes the block >>> group >>> and its respective chunk is released. Even if it didn't do the >>> lock/unlock >>> per iteration, it could still see block group X in the list, because >>> the >>> cleaner kthread might call btrfs_delete_unused_bgs() multiple times >>> (for >>> example if there are several snapshots to delete); >>> >>> 7) A new block group X' is created for data, and it's associated to the >>> same chunk >>> that block group X was associated to; >>> >> >> Actually this can't happen, we search the commit root for a free dev extent, >> so if block group X` get's mapped to a dev extent that was deleted in the >> same transaction as it was free'd in then that is a different problem. > > Hum, yep I missed the detail that when looking for free device extents > we use the commit root. > In that case I don't think anymore that there's a problem. > > Thanks for looking at it. > I still think its a good idea just so we don't have a lot of churn, but change up the commit message to make it sound less like the original stuff would eat children. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 36f82ba..a5e471a 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1726,6 +1726,7 @@ struct btrfs_fs_info { spinlock_t unused_bgs_lock; struct list_head unused_bgs; + struct list_head unused_bgs_to_clean; /* For btrfs to record security options */ struct security_mnt_opts security_opts; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 2409718..702bbdf 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2243,6 +2243,7 @@ int open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->space_info); INIT_LIST_HEAD(&fs_info->tree_mod_seq_list); INIT_LIST_HEAD(&fs_info->unused_bgs); + INIT_LIST_HEAD(&fs_info->unused_bgs_to_clean); btrfs_mapping_init(&fs_info->mapping_tree); btrfs_init_block_rsv(&fs_info->global_block_rsv, BTRFS_BLOCK_RSV_GLOBAL); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 744b580..bc1c0b7 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -8858,6 +8858,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info) up_write(&info->commit_root_sem); spin_lock(&info->unused_bgs_lock); + list_splice_init(&info->unused_bgs_to_clean, &info->unused_bgs); while (!list_empty(&info->unused_bgs)) { block_group = list_first_entry(&info->unused_bgs, struct btrfs_block_group_cache, @@ -9466,10 +9467,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) return; spin_lock(&fs_info->unused_bgs_lock); - while (!list_empty(&fs_info->unused_bgs)) { + while (!list_empty(&fs_info->unused_bgs_to_clean)) { u64 start, end; - block_group = list_first_entry(&fs_info->unused_bgs, + block_group = list_first_entry(&fs_info->unused_bgs_to_clean, struct btrfs_block_group_cache, bg_list); space_info = block_group->space_info; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 396ae8b..86d7cf5 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1937,6 +1937,11 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, btrfs_prepare_extent_commit(trans, root); + spin_lock(&root->fs_info->unused_bgs_lock); + list_splice_init(&root->fs_info->unused_bgs, + &root->fs_info->unused_bgs_to_clean); + spin_unlock(&root->fs_info->unused_bgs_lock); + cur_trans = root->fs_info->running_transaction; btrfs_set_root_node(&root->fs_info->tree_root->root_item,
We have a race while deleting unused block groups that causes extents written by past generations/transactions to be rewritten by the current transaction before that transaction is committed. The steps that lead to this issue: 1) At transaction N one or more block groups became unused and we added them to the list fs_info->unused_bgs; 2) While still at transaction N we write btree extents to block group X and the transaction is committed; 3) The cleaner kthread is awaken and calls btrfs_delete_unused_bgs() to go through the list fs_info->unused_bgs and remove unused block groups; 4) Transaction N + 1 starts; 5) At transaction N + 1, block group X becomes unused and is added to the list fs_info->unused_bgs - this implies delayed refs were run, so we had the following function calls: btrfs_run_delayed_refs() -> __btrfs_free_extent() -> update_block_group(). The update_block_group() function grabs the lock fs_info->unused_bgs_lock, adds block group X to fs_info->unused_bgs and releases that lock; 6) The cleaner kthread, while at btrfs_delete_unused_bgs(), sees block group X added by transaction N + 1 because it's doing a loop that finishes only when the list fs_info->unused_bgs is empty and locks and unlocks the spinlock fs_info->unused_bgs_lock on each iteration. So it deletes the block group and its respective chunk is released. Even if it didn't do the lock/unlock per iteration, it could still see block group X in the list, because the cleaner kthread might call btrfs_delete_unused_bgs() multiple times (for example if there are several snapshots to delete); 7) A new block group X' is created for data, and it's associated to the same chunk that block group X was associated to; 8) Extents from block group X' are allocated for file data and for example an fsync makes the file data be effectively written to disk; 9) A crash/reboot happens before transaction N + 1 is committed; 10) On the next mount, we will read extents from block group/chunk X but they no longer have valid btree nodes/leafs - they have instead file data, and therefore all sorts of errors will happen. So fix this by ensuring the cleaner kthread can never delete a block group that became unused in the current transaction, that is, only delete block groups that were added to the unused_bgs list by past transactions. Signed-off-by: Filipe Manana <fdmanana@suse.com> --- fs/btrfs/ctree.h | 1 + fs/btrfs/disk-io.c | 1 + fs/btrfs/extent-tree.c | 5 +++-- fs/btrfs/transaction.c | 5 +++++ 4 files changed, 10 insertions(+), 2 deletions(-)