btrfs: loop only once over data sizes array when inserting an item batch
When inserting a batch of items into a btree, we end up looping over the
data sizes array 3 times:
1) Once in the caller of btrfs_insert_empty_items(), when it populates the
array with the data sizes for each item;
2) Once at btrfs_insert_empty_items() to sum the elements of the data
sizes array and compute the total data size;
3) And then once again at setup_items_for_insert(), where we do exactly
the same as what we do at btrfs_insert_empty_items(), to compute the
total data size.
That is not bad for small arrays, but when the arrays have hundreds of
elements, the time spent on looping is not negligible. For example when
doing batch inserts of delayed items for dir index items or when logging
a directory, it's common to have 200 to 260 dir index items in a single
batch when using a leaf size of 16K and using file names between 8 and 12
characters. For a 64K leaf size, multiply that by 4. Taking into account
that during directory logging or when flushing delayed dir index items we
can have many of those large batches, the time spent on the looping adds
up quickly.
It's also more important to avoid it at setup_items_for_insert(), since
we are holding a write lock on a leaf and, in some cases, on upper nodes
of the btree, which causes us to block other tasks that want to access
the leaf and nodes for longer than necessary.
So change the code so that setup_items_for_insert() and
btrfs_insert_empty_items() no longer compute the total data size, and
instead rely on the caller to supply it. This makes us loop over the
array only once, where we can both populate the data size array and
compute the total data size, taking advantage of spatial and temporal
locality. To make this more manageable, use a structure to contain
all the relevant details for a batch of items (keys array, data sizes
array, total data size, number of items), and use it as an argument
for btrfs_insert_empty_items() and setup_items_for_insert().
This patch is part of a small patchset that is comprised of the following
patches:
btrfs: loop only once over data sizes array when inserting an item batch
btrfs: unexport setup_items_for_insert()
btrfs: use single bulk copy operations when logging directories
This is patch 1/3 and performance results, and the specific tests, are
included in the changelog of patch 3/3.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b394ba4..140fe98 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2897,16 +2897,56 @@ static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
return btrfs_del_items(trans, root, path, path->slots[0], 1);
}
+/*
+ * Describes a batch of items to insert in a btree. This is used by
+ * btrfs_insert_empty_items() and setup_items_for_insert().
+ */
+struct btrfs_item_batch {
+ /*
+ * Pointer to an array containing the keys of the items to insert (in
+ * sorted order).
+ */
+ const struct btrfs_key *keys;
+ /* Pointer to an array containing the data size for each item to insert. */
+ const u32 *data_sizes;
+ /*
+ * The sum of data sizes for all items. The caller can compute this while
+ * setting up the data_sizes array, so it ends up being more efficient
+ * than having btrfs_insert_empty_items() or setup_item_for_insert()
+ * doing it, as it would avoid an extra loop over a potentially large
+ * array, and in the case of setup_item_for_insert(), we would be doing
+ * it while holding a write lock on a leaf and often on upper level nodes
+ * too, unnecessarily increasing the size of a critical section.
+ */
+ u32 total_data_size;
+ /* Size of the keys and data_sizes arrays (number of items in the batch). */
+ int nr;
+};
+
void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
- const struct btrfs_key *cpu_key, u32 *data_size,
- int nr);
+ const struct btrfs_item_batch *batch);
+
+static inline void setup_item_for_insert(struct btrfs_root *root,
+ struct btrfs_path *path,
+ const struct btrfs_key *key,
+ u32 data_size)
+{
+ struct btrfs_item_batch batch;
+
+ batch.keys = key;
+ batch.data_sizes = &data_size;
+ batch.total_data_size = data_size;
+ batch.nr = 1;
+
+ setup_items_for_insert(root, path, &batch);
+}
+
int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
const struct btrfs_key *key, void *data, u32 data_size);
int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- const struct btrfs_key *cpu_key, u32 *data_size,
- int nr);
+ const struct btrfs_item_batch *batch);
static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -2914,7 +2954,14 @@ static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
const struct btrfs_key *key,
u32 data_size)
{
- return btrfs_insert_empty_items(trans, root, path, key, &data_size, 1);
+ struct btrfs_item_batch batch;
+
+ batch.keys = key;
+ batch.data_sizes = &data_size;
+ batch.total_data_size = data_size;
+ batch.nr = 1;
+
+ return btrfs_insert_empty_items(trans, root, path, &batch);
}
int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path);