NFSv4: Fix a race in the net namespace mount notification
Since the struct nfs_client gets added to the global nfs_client_list
before it is initialised, it is possible that rpc_pipefs_event can
end up trying to create idmapper entries on such a thing.
The solution is to have the mount notification wait for the
initialisation of each nfs_client to complete, and then to
skip any entries for which the it failed.
Reported-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Acked-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 25dde07..d356642 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -507,6 +507,17 @@
return NULL;
}
+static bool nfs_client_init_is_complete(const struct nfs_client *clp)
+{
+ return clp->cl_cons_state != NFS_CS_INITING;
+}
+
+int nfs_wait_client_init_complete(const struct nfs_client *clp)
+{
+ return wait_event_killable(nfs_client_active_wq,
+ nfs_client_init_is_complete(clp));
+}
+
/*
* Found an existing client. Make sure it's ready before returning.
*/
@@ -516,8 +527,7 @@
{
int error;
- error = wait_event_killable(nfs_client_active_wq,
- clp->cl_cons_state < NFS_CS_INITING);
+ error = nfs_wait_client_init_complete(clp);
if (error < 0) {
nfs_put_client(clp);
return ERR_PTR(-ERESTARTSYS);
@@ -1333,7 +1343,7 @@
* so that the client back channel can find the
* nfs_client struct
*/
- clp->cl_cons_state = NFS_CS_SESSION_INITING;
+ nfs_mark_client_ready(clp, NFS_CS_SESSION_INITING);
}
#endif /* CONFIG_NFS_V4_1 */