summaryrefslogtreecommitdiffstats
path: root/fs/btrfs/print-tree.c
diff options
context:
space:
mode:
authorYan Zheng <zheng.yan@oracle.com>2009-06-10 10:45:14 -0400
committerChris Mason <chris.mason@oracle.com>2009-06-10 11:29:46 -0400
commit5d4f98a28c7d334091c1b7744f48a1acdd2a4ae0 (patch)
treec611d7d824cbcdb777dd2d8e33e2ed1c5df8a9c6 /fs/btrfs/print-tree.c
parent5c939df56c3ea018b58e5aa76181284c2053d699 (diff)
downloadop-kernel-dev-5d4f98a28c7d334091c1b7744f48a1acdd2a4ae0.zip
op-kernel-dev-5d4f98a28c7d334091c1b7744f48a1acdd2a4ae0.tar.gz
Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE)
This commit introduces a new kind of back reference for btrfs metadata. Once a filesystem has been mounted with this commit, IT WILL NO LONGER BE MOUNTABLE BY OLDER KERNELS. When a tree block in subvolume tree is cow'd, the reference counts of all extents it points to are increased by one. At transaction commit time, the old root of the subvolume is recorded in a "dead root" data structure, and the btree it points to is later walked, dropping reference counts and freeing any blocks where the reference count goes to 0. The increments done during cow and decrements done after commit cancel out, and the walk is a very expensive way to go about freeing the blocks that are no longer referenced by the new btree root. This commit reduces the transaction overhead by avoiding the need for dead root records. When a non-shared tree block is cow'd, we free the old block at once, and the new block inherits old block's references. When a tree block with reference count > 1 is cow'd, we increase the reference counts of all extents the new block points to by one, and decrease the old block's reference count by one. This dead tree avoidance code removes the need to modify the reference counts of lower level extents when a non-shared tree block is cow'd. But we still need to update back ref for all pointers in the block. This is because the location of the block is recorded in the back ref item. We can solve this by introducing a new type of back ref. The new back ref provides information about pointer's key, level and in which tree the pointer lives. This information allow us to find the pointer by searching the tree. The shortcoming of the new back ref is that it only works for pointers in tree blocks referenced by their owner trees. This is mostly a problem for snapshots, where resolving one of these fuzzy back references would be O(number_of_snapshots) and quite slow. The solution used here is to use the fuzzy back references in the common case where a given tree block is only referenced by one root, and use the full back references when multiple roots have a reference on a given block. This commit adds per subvolume red-black tree to keep trace of cached inodes. The red-black tree helps the balancing code to find cached inodes whose inode numbers within a given range. This commit improves the balancing code by introducing several data structures to keep the state of balancing. The most important one is the back ref cache. It caches how the upper level tree blocks are referenced. This greatly reduce the overhead of checking back ref. The improved balancing code scales significantly better with a large number of snapshots. This is a very large commit and was written in a number of pieces. But, they depend heavily on the disk format change and were squashed together to make sure git bisect didn't end up in a bad state wrt space balancing or the format change. Signed-off-by: Yan Zheng <zheng.yan@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/print-tree.c')
-rw-r--r--fs/btrfs/print-tree.c155
1 files changed, 138 insertions, 17 deletions
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 5f8f218..6d6523d 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -45,22 +45,132 @@ static void print_dev_item(struct extent_buffer *eb,
(unsigned long long)btrfs_device_total_bytes(eb, dev_item),
(unsigned long long)btrfs_device_bytes_used(eb, dev_item));
}
+static void print_extent_data_ref(struct extent_buffer *eb,
+ struct btrfs_extent_data_ref *ref)
+{
+ printk(KERN_INFO "\t\textent data backref root %llu "
+ "objectid %llu offset %llu count %u\n",
+ (unsigned long long)btrfs_extent_data_ref_root(eb, ref),
+ (unsigned long long)btrfs_extent_data_ref_objectid(eb, ref),
+ (unsigned long long)btrfs_extent_data_ref_offset(eb, ref),
+ btrfs_extent_data_ref_count(eb, ref));
+}
+
+static void print_extent_item(struct extent_buffer *eb, int slot)
+{
+ struct btrfs_extent_item *ei;
+ struct btrfs_extent_inline_ref *iref;
+ struct btrfs_extent_data_ref *dref;
+ struct btrfs_shared_data_ref *sref;
+ struct btrfs_disk_key key;
+ unsigned long end;
+ unsigned long ptr;
+ int type;
+ u32 item_size = btrfs_item_size_nr(eb, slot);
+ u64 flags;
+ u64 offset;
+
+ if (item_size < sizeof(*ei)) {
+#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
+ struct btrfs_extent_item_v0 *ei0;
+ BUG_ON(item_size != sizeof(*ei0));
+ ei0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_item_v0);
+ printk(KERN_INFO "\t\textent refs %u\n",
+ btrfs_extent_refs_v0(eb, ei0));
+ return;
+#else
+ BUG();
+#endif
+ }
+
+ ei = btrfs_item_ptr(eb, slot, struct btrfs_extent_item);
+ flags = btrfs_extent_flags(eb, ei);
+
+ printk(KERN_INFO "\t\textent refs %llu gen %llu flags %llu\n",
+ (unsigned long long)btrfs_extent_refs(eb, ei),
+ (unsigned long long)btrfs_extent_generation(eb, ei),
+ (unsigned long long)flags);
+
+ if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
+ struct btrfs_tree_block_info *info;
+ info = (struct btrfs_tree_block_info *)(ei + 1);
+ btrfs_tree_block_key(eb, info, &key);
+ printk(KERN_INFO "\t\ttree block key (%llu %x %llu) "
+ "level %d\n",
+ (unsigned long long)btrfs_disk_key_objectid(&key),
+ key.type,
+ (unsigned long long)btrfs_disk_key_offset(&key),
+ btrfs_tree_block_level(eb, info));
+ iref = (struct btrfs_extent_inline_ref *)(info + 1);
+ } else {
+ iref = (struct btrfs_extent_inline_ref *)(ei + 1);
+ }
+
+ ptr = (unsigned long)iref;
+ end = (unsigned long)ei + item_size;
+ while (ptr < end) {
+ iref = (struct btrfs_extent_inline_ref *)ptr;
+ type = btrfs_extent_inline_ref_type(eb, iref);
+ offset = btrfs_extent_inline_ref_offset(eb, iref);
+ switch (type) {
+ case BTRFS_TREE_BLOCK_REF_KEY:
+ printk(KERN_INFO "\t\ttree block backref "
+ "root %llu\n", (unsigned long long)offset);
+ break;
+ case BTRFS_SHARED_BLOCK_REF_KEY:
+ printk(KERN_INFO "\t\tshared block backref "
+ "parent %llu\n", (unsigned long long)offset);
+ break;
+ case BTRFS_EXTENT_DATA_REF_KEY:
+ dref = (struct btrfs_extent_data_ref *)(&iref->offset);
+ print_extent_data_ref(eb, dref);
+ break;
+ case BTRFS_SHARED_DATA_REF_KEY:
+ sref = (struct btrfs_shared_data_ref *)(iref + 1);
+ printk(KERN_INFO "\t\tshared data backref "
+ "parent %llu count %u\n",
+ (unsigned long long)offset,
+ btrfs_shared_data_ref_count(eb, sref));
+ break;
+ default:
+ BUG();
+ }
+ ptr += btrfs_extent_inline_ref_size(type);
+ }
+ WARN_ON(ptr > end);
+}
+
+#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
+static void print_extent_ref_v0(struct extent_buffer *eb, int slot)
+{
+ struct btrfs_extent_ref_v0 *ref0;
+
+ ref0 = btrfs_item_ptr(eb, slot, struct btrfs_extent_ref_v0);
+ printk("\t\textent back ref root %llu gen %llu "
+ "owner %llu num_refs %lu\n",
+ (unsigned long long)btrfs_ref_root_v0(eb, ref0),
+ (unsigned long long)btrfs_ref_generation_v0(eb, ref0),
+ (unsigned long long)btrfs_ref_objectid_v0(eb, ref0),
+ (unsigned long)btrfs_ref_count_v0(eb, ref0));
+}
+#endif
+
void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
{
int i;
+ u32 type;
u32 nr = btrfs_header_nritems(l);
struct btrfs_item *item;
- struct btrfs_extent_item *ei;
struct btrfs_root_item *ri;
struct btrfs_dir_item *di;
struct btrfs_inode_item *ii;
struct btrfs_block_group_item *bi;
struct btrfs_file_extent_item *fi;
+ struct btrfs_extent_data_ref *dref;
+ struct btrfs_shared_data_ref *sref;
+ struct btrfs_dev_extent *dev_extent;
struct btrfs_key key;
struct btrfs_key found_key;
- struct btrfs_extent_ref *ref;
- struct btrfs_dev_extent *dev_extent;
- u32 type;
printk(KERN_INFO "leaf %llu total ptrs %d free space %d\n",
(unsigned long long)btrfs_header_bytenr(l), nr,
@@ -100,20 +210,25 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
btrfs_disk_root_refs(l, ri));
break;
case BTRFS_EXTENT_ITEM_KEY:
- ei = btrfs_item_ptr(l, i, struct btrfs_extent_item);
- printk(KERN_INFO "\t\textent data refs %u\n",
- btrfs_extent_refs(l, ei));
- break;
- case BTRFS_EXTENT_REF_KEY:
- ref = btrfs_item_ptr(l, i, struct btrfs_extent_ref);
- printk(KERN_INFO "\t\textent back ref root %llu "
- "gen %llu owner %llu num_refs %lu\n",
- (unsigned long long)btrfs_ref_root(l, ref),
- (unsigned long long)btrfs_ref_generation(l, ref),
- (unsigned long long)btrfs_ref_objectid(l, ref),
- (unsigned long)btrfs_ref_num_refs(l, ref));
+ print_extent_item(l, i);
+ break;
+ case BTRFS_TREE_BLOCK_REF_KEY:
+ printk(KERN_INFO "\t\ttree block backref\n");
+ break;
+ case BTRFS_SHARED_BLOCK_REF_KEY:
+ printk(KERN_INFO "\t\tshared block backref\n");
+ break;
+ case BTRFS_EXTENT_DATA_REF_KEY:
+ dref = btrfs_item_ptr(l, i,
+ struct btrfs_extent_data_ref);
+ print_extent_data_ref(l, dref);
+ break;
+ case BTRFS_SHARED_DATA_REF_KEY:
+ sref = btrfs_item_ptr(l, i,
+ struct btrfs_shared_data_ref);
+ printk(KERN_INFO "\t\tshared data backref count %u\n",
+ btrfs_shared_data_ref_count(l, sref));
break;
-
case BTRFS_EXTENT_DATA_KEY:
fi = btrfs_item_ptr(l, i,
struct btrfs_file_extent_item);
@@ -139,6 +254,12 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
(unsigned long long)
btrfs_file_extent_ram_bytes(l, fi));
break;
+ case BTRFS_EXTENT_REF_V0_KEY:
+#ifdef BTRFS_COMPAT_EXTENT_TREE_V0
+ print_extent_ref_v0(l, i);
+#else
+ BUG();
+#endif
case BTRFS_BLOCK_GROUP_ITEM_KEY:
bi = btrfs_item_ptr(l, i,
struct btrfs_block_group_item);
OpenPOWER on IntegriCloud