Commit ffcec0e1 authored by Zach Brown's avatar Zach Brown Committed by Andy Grover

RDS: don't call rds_conn_shutdown() from rds_conn_destroy()

rds_conn_shutdown() can return before the connection is shut down when
it encounters an existing state that it doesn't understand.  This lets
rds_conn_destroy() then start tearing down the conn from under paths
that are still using it.

It's more reliable the shutdown work and wait for krdsd to complete the
shutdown callback.  This stopped some hangs I was seeing where krdsd was
trying to shut down a freed conn.
Signed-off-by: default avatarZach Brown <zach.brown@oracle.com>
parent 5adb5bc6
...@@ -312,6 +312,10 @@ void rds_conn_shutdown(struct rds_connection *conn) ...@@ -312,6 +312,10 @@ void rds_conn_shutdown(struct rds_connection *conn)
/* /*
* Stop and free a connection. * Stop and free a connection.
*
* This can only be used in very limited circumstances. It assumes that once
* the conn has been shutdown that no one else is referencing the connection.
* We can only ensure this in the rmmod path in the current code.
*/ */
void rds_conn_destroy(struct rds_connection *conn) void rds_conn_destroy(struct rds_connection *conn)
{ {
...@@ -326,10 +330,11 @@ void rds_conn_destroy(struct rds_connection *conn) ...@@ -326,10 +330,11 @@ void rds_conn_destroy(struct rds_connection *conn)
spin_lock_irq(&rds_conn_lock); spin_lock_irq(&rds_conn_lock);
hlist_del_init_rcu(&conn->c_hash_node); hlist_del_init_rcu(&conn->c_hash_node);
spin_unlock_irq(&rds_conn_lock); spin_unlock_irq(&rds_conn_lock);
synchronize_rcu(); synchronize_rcu();
rds_conn_shutdown(conn); /* shut the connection down */
rds_conn_drop(conn);
flush_work(&conn->c_down_w);
/* tear down queued messages */ /* tear down queued messages */
list_for_each_entry_safe(rm, rtmp, list_for_each_entry_safe(rm, rtmp,
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment