[lkml]   [2018]   [Jan]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[PATCH 3.18 16/52] dm btree: fix serious bug in btree_split_beneath()
3.18-stable review patch.  If anyone has any objections, please let me know.


From: Joe Thornber <>

commit bc68d0a43560e950850fc69b58f0f8254b28f6d6 upstream.

When inserting a new key/value pair into a btree we walk down the spine of
btree nodes performing the following 2 operations:

i) space for a new entry
ii) adjusting the first key entry if the new key is lower than any in the node.

If the _root_ node is full, the function btree_split_beneath() allocates 2 new
nodes, and redistibutes the root nodes entries between them. The root node is
left with 2 entries corresponding to the 2 new nodes.

btree_split_beneath() then adjusts the spine to point to one of the two new
children. This means the first key is never adjusted if the new key was lower,
ie. operation (ii) gets missed out. This can result in the new key being
'lost' for a period; until another low valued key is inserted that will uncover

This is a serious bug, and quite hard to make trigger in normal use. A
reproducing test case ("thin create devices-in-reverse-order") is
available as part of the thin-provision-tools project:

Fix the issue by changing btree_split_beneath() so it no longer adjusts
the spine. Instead it unlocks both the new nodes, and lets the main
loop in btree_insert_raw() relock the appropriate one and make any
neccessary adjustments.

Reported-by: Monty Pavel <>
Signed-off-by: Joe Thornber <>
Signed-off-by: Mike Snitzer <>
Signed-off-by: Greg Kroah-Hartman <>

drivers/md/persistent-data/dm-btree.c | 19 ++-----------------
1 file changed, 2 insertions(+), 17 deletions(-)

--- a/drivers/md/persistent-data/dm-btree.c
+++ b/drivers/md/persistent-data/dm-btree.c
@@ -572,23 +572,8 @@ static int btree_split_beneath(struct sh
pn->keys[1] = rn->keys[0];
memcpy_disk(value_ptr(pn, 1), &val, sizeof(__le64));

- /*
- * rejig the spine. This is ugly, since it knows too
- * much about the spine
- */
- if (s->nodes[0] != new_parent) {
- unlock_block(s->info, s->nodes[0]);
- s->nodes[0] = new_parent;
- }
- if (key < le64_to_cpu(rn->keys[0])) {
- unlock_block(s->info, right);
- s->nodes[1] = left;
- } else {
- unlock_block(s->info, left);
- s->nodes[1] = right;
- }
- s->count = 2;
+ unlock_block(s->info, left);
+ unlock_block(s->info, right);
return 0;

 \ /
  Last update: 2018-01-29 21:18    [W:0.478 / U:0.252 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site