mirrors/scoutfs
1
0
Fork 0
mirror of https://github.com/versity/scoutfs.git synced 2026-02-07 11:10:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

scoutfs: split on btree deletion

Browse Source 
We were getting asserts during deletion that insertion while updating a
parent item didn't have enough room in the block.

Our btree has variable length keys.  During a merge it's possible that
the items moved between the blocks can result in the final key of a
block changing from a small key to a large key.  To update the parent
ref the parent block must have as much free space as the difference in
the key sizes.

We ensure free space in parents during descent by trying to split the
block.  Deletion wasn't doing that, it was only trying to merge blocks.
We need to try to split as well as merge during deletion.

And we have to update the merge threshold so that we don't just split
the resulting block again if it doesn't have the min free space for a
new parent item.

Signed-off-by: Zach Brown <zab@versity.com>
This commit is contained in:
Zach Brown
2017-09-19 09:54:20 -07:00
committed by Zach Brown
parent 1012ee5e8f
commit 0b15cfe7f8
2 changed files with 37 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
kmod/src/btree.c
View File

@@ -169,6 +169,36 @@ static inline unsigned int all_len_bytes(unsigned key_len, unsigned val_len)
len_bytes(key_len, val_len);
}
/* number of bytes needed to insert potentially max size child item */
static inline unsigned int parent_min_free_bytes(void)
{
return all_len_bytes(SCOUTFS_BTREE_MAX_KEY_LEN,
sizeof(struct scoutfs_btree_ref));
}
/*
* The minimum number of bytes we allow in a block. During descent to
* modify if we see a block with fewer used bytes then we'll try to
* merge items from neighbours. If the neighbour also has less than the
* min bytes then the two blocks are merged.
*
* This is carefully calculated so that if two blocks are merged the
* resulting block will have at least parent min free bytes free so
* that it's not immediately split again.
*
* new_used = min_used + min_used - hdr
* new_used <= (bs - parent_min_free)
*
* min_used + min_used - hdr <= (bs - parent_min_free)
* 2 * min_used <= (bs - parent_min_free - hdr)
* min_used <= (bs - parent_min_free - hdr) / 2
*/
static inline unsigned int min_used_bytes(void)
{
return (SCOUTFS_BLOCK_SIZE - sizeof(struct scoutfs_btree_block) -
parent_min_free_bytes()) / 2;
}
/* total block bytes used by an existing item */
static inline unsigned int all_item_bytes(struct scoutfs_btree_item *item)
{
@@ -953,8 +983,7 @@ static int try_split(struct super_block *sb, struct scoutfs_btree_root *root,
int ret;
if (right->level)
all_bytes = all_len_bytes(SCOUTFS_BTREE_MAX_KEY_LEN,
sizeof(struct scoutfs_btree_ref));
all_bytes = parent_min_free_bytes();
else
all_bytes = all_len_bytes(key_len, val_len);
@@ -1019,7 +1048,7 @@ static int try_merge(struct super_block *sb, struct scoutfs_btree_root *root,
int to_move;
int ret;
if (reclaimable_free(bt) <= SCOUTFS_BTREE_FREE_LIMIT)
if (used_total(bt) >= min_used_bytes())
return 0;
/* move items right into our block if we have a left sibling */
@@ -1035,10 +1064,10 @@ static int try_merge(struct super_block *sb, struct scoutfs_btree_root *root,
if (ret)
return ret;
if (used_total(sib) <= reclaimable_free(bt))
if (used_total(sib) < min_used_bytes())
to_move = used_total(sib);
else
to_move = reclaimable_free(bt) - SCOUTFS_BTREE_FREE_LIMIT;
to_move = min_used_bytes() - used_total(bt);
move_items(bt, sib, move_right, to_move);
@@ -1316,13 +1345,12 @@ restart:
* than try and special case modifying the path to
* reflect the tree changes.
*/
if (flags & BTW_INSERT)
ret = 0;
if (flags & (BTW_INSERT | BTW_DELETE))
ret = try_split(sb, root, key, key_len, val_len,
parent, pos, bt);
else if ((flags & BTW_DELETE) && parent)
if (ret == 0 && (flags & BTW_DELETE) && parent)
ret = try_merge(sb, root, parent, pos, bt);
else
ret = 0;
if (ret > 0)
goto restart;
else if (ret < 0)

5
kmod/src/format.h
View File

@@ -65,11 +65,6 @@ struct scoutfs_block_header {
* generous.
*/
#define SCOUTFS_BTREE_MAX_HEIGHT 20
/* btree blocks (beyond the first) need to be at least half full */
#define SCOUTFS_BTREE_FREE_LIMIT \
((SCOUTFS_BLOCK_SIZE - sizeof(struct scoutfs_btree_block)) / 2)
#define SCOUTFS_BTREE_BITS 8
/*