os/bluestore/bluefs: Fix false collision with lockdep module

Usually sequence of locking is 1) FileWriter 2) File.
In _compact_log_async_LD_NF_D it was in reversed order.
No real deadlock was possible, but lockdep complained.

Bonus: Improved lock dependency graph.

Fixes: https://tracker.ceph.com/issues/52939
Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>

src/os/bluestore/BlueFS.cc

@@ -2437,16 +2437,16 @@ void BlueFS::_compact_log_async_LD_NF_D() //also locks FW for new_writer
   new_log_writer = _create_writer(new_log);
 
   new_log_writer->append(bl);
-  new_log->lock.lock();
   new_log_writer->lock.lock();
+  new_log->lock.lock();
   // 3. flush
   r = _flush_special(new_log_writer);
   ceph_assert(r == 0);
 
   // 4. wait
   _flush_bdev(new_log_writer);
-  new_log_writer->lock.unlock();
   new_log->lock.unlock();
+  new_log_writer->lock.unlock();
   // 5. update our log fnode
   // discard first old_log_jump_to extents
 

src/os/bluestore/BlueFS.h

@@ -703,13 +703,13 @@ class FitToFastVolumeSelector : public OriginalVolumeSelector {
  * Column represents last lock taken.
  * Row represents next lock taken.
  *
- *     <        | L | D | N | F | W
+ *     >        | W | L | D | N | F
  * -------------|---|---|---|---|---
- * log        L |     <   <         
- * dirty      D |                   
- * nodes      N |             <     
- * File       F |                  
- * FileWriter W | <   <       <      
+ * FileWriter W |   | > | > |   | >
+ * log        L |       | > | > | >
+ * dirty      D |           |   | >
+ * nodes      N |               | >
+ * File       F |
  * 
  * Claim: Deadlock is possible IFF graph contains cycles.
  */