diff options
author | Chris Mason <chris.mason@oracle.com> | 2009-03-13 10:10:06 -0400 |
---|---|---|
committer | Chris Mason <chris.mason@oracle.com> | 2009-03-24 16:14:25 -0400 |
commit | 56bec294dea971335d4466b30f2d959f28f6e36d (patch) | |
tree | fc0b5bbf4bb6ab35582a4c7f58f5ac88f71c38bf /fs/btrfs/disk-io.c | |
parent | 9fa8cfe706f9c20067c042a064999d5825a35330 (diff) | |
download | op-kernel-dev-56bec294dea971335d4466b30f2d959f28f6e36d.zip op-kernel-dev-56bec294dea971335d4466b30f2d959f28f6e36d.tar.gz |
Btrfs: do extent allocation and reference count updates in the background
The extent allocation tree maintains a reference count and full
back reference information for every extent allocated in the
filesystem. For subvolume and snapshot trees, every time
a block goes through COW, the new copy of the block adds a reference
on every block it points to.
If a btree node points to 150 leaves, then the COW code needs to go
and add backrefs on 150 different extents, which might be spread all
over the extent allocation tree.
These updates currently happen during btrfs_cow_block, and most COWs
happen during btrfs_search_slot. btrfs_search_slot has locks held
on both the parent and the node we are COWing, and so we really want
to avoid IO during the COW if we can.
This commit adds an rbtree of pending reference count updates and extent
allocations. The tree is ordered by byte number of the extent and byte number
of the parent for the back reference. The tree allows us to:
1) Modify back references in something close to disk order, reducing seeks
2) Significantly reduce the number of modifications made as block pointers
are balanced around
3) Do all of the extent insertion and back reference modifications outside
of the performance critical btrfs_search_slot code.
#3 has the added benefit of greatly reducing the btrfs stack footprint.
The extent allocation tree modifications are done without the deep
(and somewhat recursive) call chains used in the past.
These delayed back reference updates must be done before the transaction
commits, and so the rbtree is tied to the transaction. Throttling is
implemented to help keep the queue of backrefs at a reasonable size.
Since there was a similar mechanism in place for the extent tree
extents, that is removed and replaced by the delayed reference tree.
Yan Zheng <yan.zheng@oracle.com> helped review and fixup this code.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/disk-io.c')
-rw-r--r-- | fs/btrfs/disk-io.c | 29 |
1 files changed, 18 insertions, 11 deletions
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3e18175..4f43e22 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1458,6 +1458,7 @@ static int transaction_kthread(void *arg) struct btrfs_root *root = arg; struct btrfs_trans_handle *trans; struct btrfs_transaction *cur; + struct btrfs_fs_info *info = root->fs_info; unsigned long now; unsigned long delay; int ret; @@ -1471,12 +1472,6 @@ static int transaction_kthread(void *arg) vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE); mutex_lock(&root->fs_info->transaction_kthread_mutex); - if (root->fs_info->total_ref_cache_size > 20 * 1024 * 1024) { - printk(KERN_INFO "btrfs: total reference cache " - "size %llu\n", - root->fs_info->total_ref_cache_size); - } - mutex_lock(&root->fs_info->trans_mutex); cur = root->fs_info->running_transaction; if (!cur) { @@ -1486,13 +1481,30 @@ static int transaction_kthread(void *arg) now = get_seconds(); if (now < cur->start_time || now - cur->start_time < 30) { + unsigned long num_delayed; + num_delayed = cur->delayed_refs.num_entries; mutex_unlock(&root->fs_info->trans_mutex); delay = HZ * 5; + + /* + * we may have been woken up early to start + * processing the delayed extent ref updates + * If so, run some of them and then loop around again + * to see if we need to force a commit + */ + if (num_delayed > 64) { + mutex_unlock(&info->transaction_kthread_mutex); + trans = btrfs_start_transaction(root, 1); + btrfs_run_delayed_refs(trans, root, 256); + btrfs_end_transaction(trans, root); + continue; + } goto sleep; } mutex_unlock(&root->fs_info->trans_mutex); trans = btrfs_start_transaction(root, 1); ret = btrfs_commit_transaction(trans, root); + sleep: wake_up_process(root->fs_info->cleaner_kthread); mutex_unlock(&root->fs_info->transaction_kthread_mutex); @@ -1611,10 +1623,6 @@ struct btrfs_root *open_ctree(struct super_block *sb, extent_io_tree_init(&fs_info->pinned_extents, fs_info->btree_inode->i_mapping, GFP_NOFS); - extent_io_tree_init(&fs_info->pending_del, - fs_info->btree_inode->i_mapping, GFP_NOFS); - extent_io_tree_init(&fs_info->extent_ins, - fs_info->btree_inode->i_mapping, GFP_NOFS); fs_info->do_barriers = 1; INIT_LIST_HEAD(&fs_info->dead_reloc_roots); @@ -1629,7 +1637,6 @@ struct btrfs_root *open_ctree(struct super_block *sb, mutex_init(&fs_info->trans_mutex); mutex_init(&fs_info->tree_log_mutex); mutex_init(&fs_info->drop_mutex); - mutex_init(&fs_info->extent_ins_mutex); mutex_init(&fs_info->pinned_mutex); mutex_init(&fs_info->chunk_mutex); mutex_init(&fs_info->transaction_kthread_mutex); |