race of nfsd inetaddr notifiers vs nn->nfsd_serv change

nfsd_inet[6]addr_event uses nn->nfsd_serv without taking nfsd_mutex,
which can be changed during execution of notifiers and crash the host.

Moreover if notifiers were enabled in one net namespace they are enabled
in all other net namespaces, from creation until destruction.

This patch allows notifiers to access nn->nfsd_serv only after the
pointer is correctly initialized and delays cleanup until notifiers are
no longer in use.

Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Tested-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>

3 files changed, 17 insertions, 3 deletions

diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h
index 1c91391f4805..36358d435cb0 100644
--- a/fs/nfsd/netns.h
+++ b/fs/nfsd/netns.h
@@ -119,6 +119,9 @@ struct nfsd_net {
         u32 clverifier_counter;
 
         struct svc_serv *nfsd_serv;
+
+        wait_queue_head_t ntf_wq;
+        atomic_t ntf_refcnt;
 };
 
 /* Simple check to find out if a given net was properly initialized */
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 6493df6b1bd5..d107b4426f7e 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1241,6 +1241,9 @@ static __net_init int nfsd_init_net(struct net *net)
         nn->nfsd4_grace = 90;
         nn->clverifier_counter = prandom_u32();
         nn->clientid_counter = prandom_u32();
+
+        atomic_set(&nn->ntf_refcnt, 0);
+        init_waitqueue_head(&nn->ntf_wq);
         return 0;
 
 out_idmap_error:
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 33117d4ffce0..89cb484f1cfb 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -335,7 +335,8 @@ static int nfsd_inetaddr_event(struct notifier_block *this, unsigned long event,
         struct nfsd_net *nn = net_generic(net, nfsd_net_id);
         struct sockaddr_in sin;
 
-        if (event != NETDEV_DOWN)
+        if ((event != NETDEV_DOWN) ||
+            !atomic_inc_not_zero(&nn->ntf_refcnt))
                 goto out;
 
         if (nn->nfsd_serv) {
@@ -344,6 +345,8 @@ static int nfsd_inetaddr_event(struct notifier_block *this, unsigned long event,
                 sin.sin_addr.s_addr = ifa->ifa_local;
                 svc_age_temp_xprts_now(nn->nfsd_serv, (struct sockaddr *)&sin);
         }
+        atomic_dec(&nn->ntf_refcnt);
+        wake_up(&nn->ntf_wq);
 
 out:
         return NOTIFY_DONE;
@@ -363,7 +366,8 @@ static int nfsd_inet6addr_event(struct notifier_block *this,
         struct nfsd_net *nn = net_generic(net, nfsd_net_id);
         struct sockaddr_in6 sin6;
 
-        if (event != NETDEV_DOWN)
+        if ((event != NETDEV_DOWN) ||
+            !atomic_inc_not_zero(&nn->ntf_refcnt))
                 goto out;
 
         if (nn->nfsd_serv) {
@@ -374,7 +378,8 @@ static int nfsd_inet6addr_event(struct notifier_block *this,
                         sin6.sin6_scope_id = ifa->idev->dev->ifindex;
                 svc_age_temp_xprts_now(nn->nfsd_serv, (struct sockaddr *)&sin6);
         }
-
+        atomic_dec(&nn->ntf_refcnt);
+        wake_up(&nn->ntf_wq);
 out:
         return NOTIFY_DONE;
 }
@@ -391,6 +396,7 @@ static void nfsd_last_thread(struct svc_serv *serv, struct net *net)
 {
         struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
+        atomic_dec(&nn->ntf_refcnt);
         /* check if the notifier still has clients */
         if (atomic_dec_return(&nfsd_notifier_refcount) == 0) {
                 unregister_inetaddr_notifier(&nfsd_inetaddr_notifier);
@@ -398,6 +404,7 @@ static void nfsd_last_thread(struct svc_serv *serv, struct net *net)
                 unregister_inet6addr_notifier(&nfsd_inet6addr_notifier);
 #endif
         }
+        wait_event(nn->ntf_wq, atomic_read(&nn->ntf_refcnt) == 0);
 
         /*
          * write_ports can create the server without actually starting
@@ -517,6 +524,7 @@ int nfsd_create_serv(struct net *net)
                 register_inet6addr_notifier(&nfsd_inet6addr_notifier);
 #endif
         }
+        atomic_inc(&nn->ntf_refcnt);
         ktime_get_real_ts64(&nn->nfssvc_boot); /* record boot time */
         return 0;
 }