This is a test patch being supplied for a trial run on syzkaller.

Explicitly cancel the workq before releasing resources that
will allow netns deletion, so that the connect request does
not trip up on a use-after free of the netns afterward.

Signed-off-by: Sowmini Varadhan <sowmini.varad...@oracle.com>
Reported-by: Dmitry Vyukov <dvyu...@google.com>
---
 net/rds/tcp.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index 57bb523..2568eb1 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -509,6 +509,8 @@ static void rds_tcp_conn_paths_destroy(struct 
rds_connection *conn)
 
        for (i = 0; i < RDS_MPATH_WORKERS; i++) {
                cp = &conn->c_path[i];
+               if (cancel_delayed_work_sync(&cp->cp_conn_w))
+                       pr_info("cancelled on path %d\n", i);
                tc = cp->cp_transport_data;
                if (!tc->t_sock)
                        continue;
-- 
1.7.1

Reply via email to