Merge tag 'nfs-for-3.14-4' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client bugfixes from Trond Myklebust:
 "Highlights include stable fixes for the following bugs:

   - General performance regression due to NFS_INO_INVALID_LABEL being
     set when the server doesn't support labeled NFS
   - Hang in the RPC code due to a socket out-of-buffer race
   - Infinite loop when trying to establish the NFSv4 lease
   - Use-after-free bug in the RPCSEC gss code.
   - nfs4_select_rw_stateid is returning with a non-zero error value on
     success

  Other bug fixes:

  - Potential memory scribble in the RPC bi-directional RPC code
  - Pipe version reference leak
  - Use the correct net namespace in the new NFSv4 migration code"

* tag 'nfs-for-3.14-4' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFS fix error return in nfs4_select_rw_stateid
  NFSv4: Use the correct net namespace in nfs4_update_server
  SUNRPC: Fix a pipe_version reference leak
  SUNRPC: Ensure that gss_auth isn't freed before its upcall messages
  SUNRPC: Fix potential memory scribble in xprt_free_bc_request()
  SUNRPC: Fix races in xs_nospace()
  SUNRPC: Don't create a gss auth cache unless rpc.gssd is running
  NFS: Do not set NFS_INO_INVALID_LABEL unless server supports labeled NFS

8 files changed, 60 insertions, 21 deletions

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 28a0a3cbd3b7..360114ae8b82 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -164,17 +164,16 @@ static void nfs_zap_caches_locked(struct inode *inode)
         if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) {
                 nfs_fscache_invalidate(inode);
                 nfsi->cache_validity |= NFS_INO_INVALID_ATTR
-                                        | NFS_INO_INVALID_LABEL
                                         | NFS_INO_INVALID_DATA
                                         | NFS_INO_INVALID_ACCESS
                                         | NFS_INO_INVALID_ACL
                                         | NFS_INO_REVAL_PAGECACHE;
         } else
                 nfsi->cache_validity |= NFS_INO_INVALID_ATTR
-                                        | NFS_INO_INVALID_LABEL
                                         | NFS_INO_INVALID_ACCESS
                                         | NFS_INO_INVALID_ACL
                                         | NFS_INO_REVAL_PAGECACHE;
+        nfs_zap_label_cache_locked(nfsi);
 }
 
 void nfs_zap_caches(struct inode *inode)
@@ -266,6 +265,13 @@ nfs_init_locked(struct inode *inode, void *opaque)
 }
 
 #ifdef CONFIG_NFS_V4_SECURITY_LABEL
+static void nfs_clear_label_invalid(struct inode *inode)
+{
+        spin_lock(&inode->i_lock);
+        NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_LABEL;
+        spin_unlock(&inode->i_lock);
+}
+
 void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr,
                                         struct nfs4_label *label)
 {
@@ -283,6 +289,7 @@ void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr,
                                         __func__,
                                         (char *)label->label,
                                         label->len, error);
+                nfs_clear_label_invalid(inode);
         }
 }
 
@@ -1648,7 +1655,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
                 inode->i_blocks = fattr->du.nfs2.blocks;
 
         /* Update attrtimeo value if we're out of the unstable period */
-        if (invalid & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL)) {
+        if (invalid & NFS_INO_INVALID_ATTR) {
                 nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE);
                 nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
                 nfsi->attrtimeo_timestamp = now;
@@ -1661,7 +1668,6 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
                 }
         }
         invalid &= ~NFS_INO_INVALID_ATTR;
-        invalid &= ~NFS_INO_INVALID_LABEL;
         /* Don't invalidate the data if we were to blame */
         if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
                                 || S_ISLNK(inode->i_mode)))
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 8b5cc04a8611..b46cf5a67329 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -176,7 +176,8 @@ extern struct nfs_server *nfs4_create_server(
 extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *,
                                                       struct nfs_fh *);
 extern int nfs4_update_server(struct nfs_server *server, const char *hostname,
-                                        struct sockaddr *sap, size_t salen);
+                                        struct sockaddr *sap, size_t salen,
+                                        struct net *net);
 extern void nfs_free_server(struct nfs_server *server);
 extern struct nfs_server *nfs_clone_server(struct nfs_server *,
                                            struct nfs_fh *,
@@ -279,9 +280,18 @@ static inline void nfs4_label_free(struct nfs4_label *label)
         }
         return;
 }
+
+static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi)
+{
+        if (nfs_server_capable(&nfsi->vfs_inode, NFS_CAP_SECURITY_LABEL))
+                nfsi->cache_validity |= NFS_INO_INVALID_LABEL;
+}
 #else
 static inline struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { return NULL; }
 static inline void nfs4_label_free(void *label) {}
+static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi)
+{
+}
 #endif /* CONFIG_NFS_V4_SECURITY_LABEL */
 
 /* proc.c */
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index 860ad26a5590..0e46d3d1b6cc 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -1135,6 +1135,7 @@ static int nfs_probe_destination(struct nfs_server *server)
  * @hostname: new end-point's hostname
  * @sap: new end-point's socket address
  * @salen: size of "sap"
+ * @net: net namespace
  *
  * The nfs_server must be quiescent before this function is invoked.
  * Either its session is drained (NFSv4.1+), or its transport is
@@ -1143,13 +1144,13 @@ static int nfs_probe_destination(struct nfs_server *server)
  * Returns zero on success, or a negative errno value.
  */
 int nfs4_update_server(struct nfs_server *server, const char *hostname,
-                       struct sockaddr *sap, size_t salen)
+                       struct sockaddr *sap, size_t salen, struct net *net)
 {
         struct nfs_client *clp = server->nfs_client;
         struct rpc_clnt *clnt = server->client;
         struct xprt_create xargs = {
                 .ident          = clp->cl_proto,
-                .net            = &init_net,
+                .net            = net,
                 .dstaddr        = sap,
                 .addrlen        = salen,
                 .servername     = hostname,
@@ -1189,7 +1190,7 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname,
         error = nfs4_set_client(server, hostname, sap, salen, buf,
                                 clp->cl_rpcclient->cl_auth->au_flavor,
                                 clp->cl_proto, clnt->cl_timeout,
-                                clp->cl_minorversion, clp->cl_net);
+                                clp->cl_minorversion, net);
         nfs_put_client(clp);
         if (error != 0) {
                 nfs_server_insert_lists(server);
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index 4e7f05d3e9db..3d5dbf80d46a 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -121,9 +121,8 @@ static int nfs4_validate_fspath(struct dentry *dentry,
 }
 
 static size_t nfs_parse_server_name(char *string, size_t len,
-                struct sockaddr *sa, size_t salen, struct nfs_server *server)
+                struct sockaddr *sa, size_t salen, struct net *net)
 {
-        struct net *net = rpc_net_ns(server->client);
         ssize_t ret;
 
         ret = rpc_pton(net, string, len, sa, salen);
@@ -223,6 +222,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
                                      const struct nfs4_fs_location *location)
 {
         const size_t addr_bufsize = sizeof(struct sockaddr_storage);
+        struct net *net = rpc_net_ns(NFS_SB(mountdata->sb)->client);
         struct vfsmount *mnt = ERR_PTR(-ENOENT);
         char *mnt_path;
         unsigned int maxbuflen;
@@ -248,8 +248,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
                         continue;
 
                 mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len,
-                                mountdata->addr, addr_bufsize,
-                                NFS_SB(mountdata->sb));
+                                mountdata->addr, addr_bufsize, net);
                 if (mountdata->addrlen == 0)
                         continue;
 
@@ -419,6 +418,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server,
                 const struct nfs4_fs_location *location)
 {
         const size_t addr_bufsize = sizeof(struct sockaddr_storage);
+        struct net *net = rpc_net_ns(server->client);
         struct sockaddr *sap;
         unsigned int s;
         size_t salen;
@@ -440,7 +440,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server,
                         continue;
 
                 salen = nfs_parse_server_name(buf->data, buf->len,
-                                                sap, addr_bufsize, server);
+                                                sap, addr_bufsize, net);
                 if (salen == 0)
                         continue;
                 rpc_set_port(sap, NFS_PORT);
@@ -450,7 +450,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server,
                 if (hostname == NULL)
                         break;
 
-                error = nfs4_update_server(server, hostname, sap, salen);
+                error = nfs4_update_server(server, hostname, sap, salen, net);
                 kfree(hostname);
                 if (error == 0)
                         break;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e5be72518bd7..e1a47217c05e 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1015,8 +1015,11 @@ int nfs4_select_rw_stateid(nfs4_stateid *dst, struct nfs4_state *state,
         if (ret == -EIO)
                 /* A lost lock - don't even consider delegations */
                 goto out;
-        if (nfs4_copy_delegation_stateid(dst, state->inode, fmode))
+        /* returns true if delegation stateid found and copied */
+        if (nfs4_copy_delegation_stateid(dst, state->inode, fmode)) {
+                ret = 0;
                 goto out;
+        }
         if (ret != -ENOENT)
                 /* nfs4_copy_delegation_stateid() didn't over-write
                  * dst, so it still has the lock stateid which we now
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 6c0513a7f992..36e431ee1c90 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -108,6 +108,7 @@ struct gss_auth {
 static DEFINE_SPINLOCK(pipe_version_lock);
 static struct rpc_wait_queue pipe_version_rpc_waitqueue;
 static DECLARE_WAIT_QUEUE_HEAD(pipe_version_waitqueue);
+static void gss_put_auth(struct gss_auth *gss_auth);
 
 static void gss_free_ctx(struct gss_cl_ctx *);
 static const struct rpc_pipe_ops gss_upcall_ops_v0;
@@ -320,6 +321,7 @@ gss_release_msg(struct gss_upcall_msg *gss_msg)
         if (gss_msg->ctx != NULL)
                 gss_put_ctx(gss_msg->ctx);
         rpc_destroy_wait_queue(&gss_msg->rpc_waitqueue);
+        gss_put_auth(gss_msg->auth);
         kfree(gss_msg);
 }
 
@@ -498,9 +500,12 @@ gss_alloc_msg(struct gss_auth *gss_auth,
         default:
                 err = gss_encode_v1_msg(gss_msg, service_name, gss_auth->target_name);
                 if (err)
-                        goto err_free_msg;
+                        goto err_put_pipe_version;
         };
+        kref_get(&gss_auth->kref);
         return gss_msg;
+err_put_pipe_version:
+        put_pipe_version(gss_auth->net);
 err_free_msg:
         kfree(gss_msg);
 err:
@@ -991,6 +996,8 @@ gss_create_new(struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
         gss_auth->service = gss_pseudoflavor_to_service(gss_auth->mech, flavor);
         if (gss_auth->service == 0)
                 goto err_put_mech;
+        if (!gssd_running(gss_auth->net))
+                goto err_put_mech;
         auth = &gss_auth->rpc_auth;
         auth->au_cslack = GSS_CRED_SLACK >> 2;
         auth->au_rslack = GSS_VERF_SLACK >> 2;
@@ -1062,6 +1069,12 @@ gss_free_callback(struct kref *kref)
 }
 
 static void
+gss_put_auth(struct gss_auth *gss_auth)
+{
+        kref_put(&gss_auth->kref, gss_free_callback);
+}
+
+static void
 gss_destroy(struct rpc_auth *auth)
 {
         struct gss_auth *gss_auth = container_of(auth,
@@ -1082,7 +1095,7 @@ gss_destroy(struct rpc_auth *auth)
         gss_auth->gss_pipe[1] = NULL;
         rpcauth_destroy_credcache(auth);
 
-        kref_put(&gss_auth->kref, gss_free_callback);
+        gss_put_auth(gss_auth);
 }
 
 /*
@@ -1253,7 +1266,7 @@ gss_destroy_nullcred(struct rpc_cred *cred)
         call_rcu(&cred->cr_rcu, gss_free_cred_callback);
         if (ctx)
                 gss_put_ctx(ctx);
-        kref_put(&gss_auth->kref, gss_free_callback);
+        gss_put_auth(gss_auth);
 }
 
 static void
diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
index 890a29912d5a..e860d4f7ed2a 100644
--- a/net/sunrpc/backchannel_rqst.c
+++ b/net/sunrpc/backchannel_rqst.c
@@ -64,7 +64,6 @@ static void xprt_free_allocation(struct rpc_rqst *req)
         free_page((unsigned long)xbufp->head[0].iov_base);
         xbufp = &req->rq_snd_buf;
         free_page((unsigned long)xbufp->head[0].iov_base);
-        list_del(&req->rq_bc_pa_list);
         kfree(req);
 }
 
@@ -168,8 +167,10 @@ out_free:
         /*
          * Memory allocation failed, free the temporary list
          */
-        list_for_each_entry_safe(req, tmp, &tmp_list, rq_bc_pa_list)
+        list_for_each_entry_safe(req, tmp, &tmp_list, rq_bc_pa_list) {
+                list_del(&req->rq_bc_pa_list);
                 xprt_free_allocation(req);
+        }
 
         dprintk("RPC:       setup backchannel transport failed\n");
         return -ENOMEM;
@@ -198,6 +199,7 @@ void xprt_destroy_backchannel(struct rpc_xprt *xprt, unsigned int max_reqs)
         xprt_dec_alloc_count(xprt, max_reqs);
         list_for_each_entry_safe(req, tmp, &xprt->bc_pa_list, rq_bc_pa_list) {
                 dprintk("RPC:        req=%p\n", req);
+                list_del(&req->rq_bc_pa_list);
                 xprt_free_allocation(req);
                 if (--max_reqs == 0)
                         break;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 817a1e523969..0addefca8e77 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -510,6 +510,7 @@ static int xs_nospace(struct rpc_task *task)
         struct rpc_rqst *req = task->tk_rqstp;
         struct rpc_xprt *xprt = req->rq_xprt;
         struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
+        struct sock *sk = transport->inet;
         int ret = -EAGAIN;
 
         dprintk("RPC: %5u xmit incomplete (%u left of %u)\n",
@@ -527,7 +528,7 @@ static int xs_nospace(struct rpc_task *task)
                          * window size
                          */
                         set_bit(SOCK_NOSPACE, &transport->sock->flags);
-                        transport->inet->sk_write_pending++;
+                        sk->sk_write_pending++;
                         /* ...and wait for more buffer space */
                         xprt_wait_for_buffer_space(task, xs_nospace_callback);
                 }
@@ -537,6 +538,9 @@ static int xs_nospace(struct rpc_task *task)
         }
 
         spin_unlock_bh(&xprt->transport_lock);
+
+        /* Race breaker in case memory is freed before above code is called */
+        sk->sk_write_space(sk);
         return ret;
 }