Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index fb6eeef..6216dfb 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -31,7 +31,7 @@
struct extent_map_tree {
struct rb_root map;
- spinlock_t lock;
+ rwlock_t lock;
};
static inline u64 extent_map_end(struct extent_map *em)