From: Zheng Liu <[email protected]>

This commit tries to fix a livelock in bcache.  This livelock might
happen when we causes a huge number of cache misses simultaneously.

When we get a cache miss, bcache will execute the following path.

->cached_dev_make_request()
  ->cached_dev_read()
    ->cached_lookup()
      ->bch->btree_map_keys()
        ->btree_root()  <------------------------
          ->bch_btree_map_keys_recurse()        |
            ->cache_lookup_fn()                 |
              ->cached_dev_cache_miss()         |
                ->bch_btree_insert_check_key() -|
                  [If btree->seq is not equal to seq + 1, we should return
                   EINTR and traverse btree again.]

In bch_btree_insert_check_key() function we first need to check upgrade
flag (op->lock == -1), and when this flag is true we need to release
read btree->lock and try to take write btree->lock.  During taking and
releasing this write lock, btree->seq will be monotone increased in
order to prevent other threads modify this in cache miss (see btree.h:74).
But if there are some cache misses caused by some requested, we could
meet a livelock because btree->seq is always changed by others.  Thus no
one can make progress.

This commit will try to take write btree->lock if it encounters a race
when we traverse btree.  Although it sacrifice the scalability but we
can ensure that only one can modify the btree.

Signed-off-by: Zheng Liu <[email protected]>
Tested-by: Joshua Schmid <[email protected]>
Cc: Joshua Schmid <[email protected]>
Cc: Zhu Yanhai <[email protected]>
Cc: Kent Overstreet <[email protected]>
---
changelog:
v2: fix a bug that stops all concurrency writes unconditionally.

 drivers/md/bcache/btree.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 00cde40..2a74635 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -2162,8 +2162,10 @@ int bch_btree_insert_check_key(struct btree *b, struct 
btree_op *op,
                rw_lock(true, b, b->level);
 
                if (b->key.ptr[0] != btree_ptr ||
-                   b->seq != seq + 1)
+                   b->seq != seq + 1) {
+                       op->lock = b->level;
                        goto out;
+               }
        }
 
        SET_KEY_PTRS(check_key, 1);
-- 
1.7.9.7

--
To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to