docs/RCU: Bring smp_wmb() back

Commit Message

Alan Huang June 17, 2023, 11:07 a.m. UTC

There are two memory ordering required in the insertion algorithm,
we need wo make obj->key is updated before obj->obj_node.next
and obj->refcnt, atomic_set_release is not enough to provide the
required memory barrier.

Signed-off-by: Alan Huang <[e-mail]>
---
 Documentation/RCU/rculist_nulls.rst | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

Patch

diff --git a/Documentation/RCU/rculist_nulls.rst b/Documentation/RCU/rculist_nulls.rst
index e06ed40bb6..77244adbdf 100644
--- a/Documentation/RCU/rculist_nulls.rst
+++ b/Documentation/RCU/rculist_nulls.rst
@@ -98,7 +98,7 @@  Quoting Corey Minyard::
 ----------------------
 
 We need to make sure a reader cannot read the new 'obj->obj_node.next' value
-and previous value of 'obj->key'. Otherwise, an item could be deleted
+and previous value of 'obj->key' at the same time. Otherwise, an item could be deleted
 from a chain, and inserted into another chain. If new chain was empty
 before the move, 'next' pointer is NULL, and lockless reader can not
 detect the fact that it missed following items in original chain.
@@ -112,7 +112,12 @@  detect the fact that it missed following items in original chain.
   obj = kmem_cache_alloc(...);
   lock_chain(); // typically a spin_lock()
   obj->key = key;
-  atomic_set_release(&obj->refcnt, 1); // key before refcnt
+  /*
+  * we need to make sure obj->key is updated before obj->obj_node.next
+  * and obj->refcnt
+  */
+  smp_wmb();
+  atomic_set(&obj->refcnt, 1);
   hlist_add_head_rcu(&obj->obj_node, list);
   unlock_chain(); // typically a spin_unlock()