diff options
| author |   Filipe David Borba Manana <fdmanana@gmail.com> | 2013-09-02 12:19:13 +0100 |
|---|---|---|
| committer |   Chris Mason <chris.mason@fusionio.com> | 2013-09-21 10:58:55 -0400 |
| commit | cef2193729595686f787c5858aad69b4369de757 (patch) | |
| tree | d6df5311a9d0bff68309cfc05cc742e86cc720c7 /fs/btrfs/inode.c | |
| parent | Btrfs: do not add replace target to the alloc_list (diff) | |
| download | linux-dev-cef2193729595686f787c5858aad69b4369de757.tar.xz linux-dev-cef2193729595686f787c5858aad69b4369de757.zip | |
Btrfs: more efficient inode tree replace operation
Instead of removing the current inode from the red black tree
and then add the new one, just use the red black tree replace
operation, which is more efficient.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Zach Brown <zab@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Diffstat (limited to 'fs/btrfs/inode.c')
| -rw-r--r-- | fs/btrfs/inode.c | 10 |
1 files changed, 5 insertions, 5 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6091ba9d2494..abed81d93687 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4688,11 +4688,11 @@ static void inode_tree_add(struct inode *inode) struct btrfs_inode *entry; struct rb_node **p; struct rb_node *parent; + struct rb_node *new = &BTRFS_I(inode)->rb_node; u64 ino = btrfs_ino(inode); if (inode_unhashed(inode)) return; -again: parent = NULL; spin_lock(&root->inode_lock); p = &root->inode_tree.rb_node; @@ -4707,14 +4707,14 @@ again: else { WARN_ON(!(entry->vfs_inode.i_state & (I_WILL_FREE | I_FREEING))); - rb_erase(parent, &root->inode_tree); + rb_replace_node(parent, new, &root->inode_tree); RB_CLEAR_NODE(parent); spin_unlock(&root->inode_lock); - goto again; + return; } } - rb_link_node(&BTRFS_I(inode)->rb_node, parent, p); - rb_insert_color(&BTRFS_I(inode)->rb_node, &root->inode_tree); + rb_link_node(new, parent, p); + rb_insert_color(new, &root->inode_tree); spin_unlock(&root->inode_lock); } |