btrfs: switch extent_buffer blocking_writers from atomic to int

The blocking_writers is either 0 or 1 and always updated under the lock,
so we don't need the atomic_t semantics.

Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This commit is contained in:
David Sterba
2019-05-02 16:47:23 +02:00
parent 38e9372e39
commit 06297d8cef
4 changed files with 23 additions and 29 deletions

View File

@@ -167,7 +167,7 @@ struct extent_buffer {
struct rcu_head rcu_head;
pid_t lock_owner;
atomic_t blocking_writers;
int blocking_writers;
atomic_t blocking_readers;
bool lock_nested;
/* >= 0 if eb belongs to a log tree, -1 otherwise */