56bec294de
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>
147 lines
3.6 KiB
C
147 lines
3.6 KiB
C
/*
|
|
* Copyright (C) 2007 Oracle. All rights reserved.
|
|
*
|
|
* This program is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU General Public
|
|
* License v2 as published by the Free Software Foundation.
|
|
*
|
|
* This program is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
* General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public
|
|
* License along with this program; if not, write to the
|
|
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
* Boston, MA 021110-1307, USA.
|
|
*/
|
|
|
|
#include <linux/sched.h>
|
|
#include "ctree.h"
|
|
#include "disk-io.h"
|
|
#include "print-tree.h"
|
|
#include "transaction.h"
|
|
#include "locking.h"
|
|
|
|
/* defrag all the leaves in a given btree. If cache_only == 1, don't read
|
|
* things from disk, otherwise read all the leaves and try to get key order to
|
|
* better reflect disk order
|
|
*/
|
|
|
|
int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
|
|
struct btrfs_root *root, int cache_only)
|
|
{
|
|
struct btrfs_path *path = NULL;
|
|
struct btrfs_key key;
|
|
int ret = 0;
|
|
int wret;
|
|
int level;
|
|
int orig_level;
|
|
int is_extent = 0;
|
|
int next_key_ret = 0;
|
|
u64 last_ret = 0;
|
|
u64 min_trans = 0;
|
|
|
|
if (cache_only)
|
|
goto out;
|
|
|
|
if (root->fs_info->extent_root == root) {
|
|
/*
|
|
* there's recursion here right now in the tree locking,
|
|
* we can't defrag the extent root without deadlock
|
|
*/
|
|
goto out;
|
|
}
|
|
|
|
if (root->ref_cows == 0 && !is_extent)
|
|
goto out;
|
|
|
|
if (btrfs_test_opt(root, SSD))
|
|
goto out;
|
|
|
|
path = btrfs_alloc_path();
|
|
if (!path)
|
|
return -ENOMEM;
|
|
|
|
level = btrfs_header_level(root->node);
|
|
orig_level = level;
|
|
|
|
if (level == 0)
|
|
goto out;
|
|
|
|
if (root->defrag_progress.objectid == 0) {
|
|
struct extent_buffer *root_node;
|
|
u32 nritems;
|
|
|
|
root_node = btrfs_lock_root_node(root);
|
|
btrfs_set_lock_blocking(root_node);
|
|
nritems = btrfs_header_nritems(root_node);
|
|
root->defrag_max.objectid = 0;
|
|
/* from above we know this is not a leaf */
|
|
btrfs_node_key_to_cpu(root_node, &root->defrag_max,
|
|
nritems - 1);
|
|
btrfs_tree_unlock(root_node);
|
|
free_extent_buffer(root_node);
|
|
memset(&key, 0, sizeof(key));
|
|
} else {
|
|
memcpy(&key, &root->defrag_progress, sizeof(key));
|
|
}
|
|
|
|
path->keep_locks = 1;
|
|
if (cache_only)
|
|
min_trans = root->defrag_trans_start;
|
|
|
|
ret = btrfs_search_forward(root, &key, NULL, path,
|
|
cache_only, min_trans);
|
|
if (ret < 0)
|
|
goto out;
|
|
if (ret > 0) {
|
|
ret = 0;
|
|
goto out;
|
|
}
|
|
btrfs_release_path(root, path);
|
|
wret = btrfs_search_slot(trans, root, &key, path, 0, 1);
|
|
|
|
if (wret < 0) {
|
|
ret = wret;
|
|
goto out;
|
|
}
|
|
if (!path->nodes[1]) {
|
|
ret = 0;
|
|
goto out;
|
|
}
|
|
path->slots[1] = btrfs_header_nritems(path->nodes[1]);
|
|
next_key_ret = btrfs_find_next_key(root, path, &key, 1, cache_only,
|
|
min_trans);
|
|
ret = btrfs_realloc_node(trans, root,
|
|
path->nodes[1], 0,
|
|
cache_only, &last_ret,
|
|
&root->defrag_progress);
|
|
WARN_ON(ret && ret != -EAGAIN);
|
|
if (next_key_ret == 0) {
|
|
memcpy(&root->defrag_progress, &key, sizeof(key));
|
|
ret = -EAGAIN;
|
|
}
|
|
|
|
btrfs_release_path(root, path);
|
|
out:
|
|
if (path)
|
|
btrfs_free_path(path);
|
|
if (ret == -EAGAIN) {
|
|
if (root->defrag_max.objectid > root->defrag_progress.objectid)
|
|
goto done;
|
|
if (root->defrag_max.type > root->defrag_progress.type)
|
|
goto done;
|
|
if (root->defrag_max.offset > root->defrag_progress.offset)
|
|
goto done;
|
|
ret = 0;
|
|
}
|
|
done:
|
|
if (ret != -EAGAIN) {
|
|
memset(&root->defrag_progress, 0,
|
|
sizeof(root->defrag_progress));
|
|
root->defrag_trans_start = trans->transid;
|
|
}
|
|
return ret;
|
|
}
|