Introduce hash bucket locks

There is one RW lock per hash bucket. Write lock is required
to map cacheline, read lock is sufficient for traversing.
Hash bucket locks are always acquired under global metadata
read lock. This assures mutual exclusion with eviction and
management paths, where global metadata write lock is held.

Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>

src/ocf_request.c

@@ -311,3 +311,14 @@ void ocf_req_clear_map(struct ocf_request *req)
 		ENV_BUG_ON(env_memset(req->map,
 			   sizeof(req->map[0]) * req->core_line_count, 0));
 }
+
+void ocf_req_hash(struct ocf_request *req)
+{
+	int i;
+
+	for (i = 0; i < req->core_line_count; i++) {
+		req->map[i].hash = ocf_metadata_hash_func(req->cache,
+				req->core_line_first + i,
+				ocf_core_get_id(req->core));
+	}
+}