list: add "list_del_init_careful()" to go with "list_empty_careful()"

That gives us ordering guarantees around the pair.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

3 files changed, 21 insertions, 8 deletions

diff --git a/include/linux/list.h b/include/linux/list.h
index aff44d34f4e4..0d0d17a10d25 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -283,6 +283,24 @@ static inline int list_empty(const struct list_head *head)
 }
 
 /**
+ * list_del_init_careful - deletes entry from list and reinitialize it.
+ * @entry: the element to delete from the list.
+ *
+ * This is the same as list_del_init(), except designed to be used
+ * together with list_empty_careful() in a way to guarantee ordering
+ * of other memory operations.
+ *
+ * Any memory operations done before a list_del_init_careful() are
+ * guaranteed to be visible after a list_empty_careful() test.
+ */
+static inline void list_del_init_careful(struct list_head *entry)
+{
+	__list_del_entry(entry);
+	entry->prev = entry;
+	smp_store_release(&entry->next, entry);
+}
+
+/**
  * list_empty_careful - tests whether a list is empty and not being modified
  * @head: the list to test
  *
@@ -297,7 +315,7 @@ static inline int list_empty(const struct list_head *head)
  */
 static inline int list_empty_careful(const struct list_head *head)
 {
-	struct list_head *next = head->next;
+	struct list_head *next = smp_load_acquire(&head->next);
 	return (next == head) && (next == head->prev);
 }
 
diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c
index ba059fbfc53a..01f5d3020589 100644
--- a/kernel/sched/wait.c
+++ b/kernel/sched/wait.c
@@ -389,7 +389,7 @@ int autoremove_wake_function(struct wait_queue_entry *wq_entry, unsigned mode, i
 	int ret = default_wake_function(wq_entry, mode, sync, key);
 
 	if (ret)
-		list_del_init(&wq_entry->entry);
+		list_del_init_careful(&wq_entry->entry);
 
 	return ret;
 }
diff --git a/mm/filemap.c b/mm/filemap.c
index 8c3d3e233d37..991503bbf922 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1041,13 +1041,8 @@ static int wake_page_function(wait_queue_entry_t *wait, unsigned mode, int sync,
 	 * since after list_del_init(&wait->entry) the wait entry
 	 * might be de-allocated and the process might even have
 	 * exited.
-	 *
-	 * We _really_ should have a "list_del_init_careful()" to
-	 * properly pair with the unlocked "list_empty_careful()"
-	 * in finish_wait().
 	 */
-	smp_mb();
-	list_del_init(&wait->entry);
+	list_del_init_careful(&wait->entry);
 	return ret;
 }