Merge branch 'for-3.7' of git://linux-nfs.org/~bfields/linux

Pull nfsd update from J Bruce Fields:
 "Another relatively quiet cycle.  There was some progress on my
  remaining 4.1 todo's, but a couple of them were just of the form
  "check that we do X correctly", so didn't have much affect on the
  code.

  Other than that, a bunch of cleanup and some bugfixes (including an
  annoying NFSv4.0 state leak and a busy-loop in the server that could
  cause it to peg the CPU without making progress)."

* 'for-3.7' of git://linux-nfs.org/~bfields/linux: (46 commits)
  UAPI: (Scripted) Disintegrate include/linux/sunrpc
  UAPI: (Scripted) Disintegrate include/linux/nfsd
  nfsd4: don't allow reclaims of expired clients
  nfsd4: remove redundant callback probe
  nfsd4: expire old client earlier
  nfsd4: separate session allocation and initialization
  nfsd4: clean up session allocation
  nfsd4: minor free_session cleanup
  nfsd4: new_conn_from_crses should only allocate
  nfsd4: separate connection allocation and initialization
  nfsd4: reject bad forechannel attrs earlier
  nfsd4: enforce per-client sessions/no-sessions distinction
  nfsd4: set cl_minorversion at create time
  nfsd4: don't pin clientids to pseudoflavors
  nfsd4: fix bind_conn_to_session xdr comment
  nfsd4: cast readlink() bug argument
  NFSD: pass null terminated buf to kstrtouint()
  nfsd: remove duplicate init in nfsd4_cb_recall
  nfsd4: eliminate redundant nfs4_free_stateid
  fs/nfsd/nfs4idmap.c: adjust inconsistent IS_ERR and PTR_ERR
  ...

Documentation/filesystems/nfs/nfsd-admin-interfaces.txt

@@ -0,0 +1,41 @@
+Administrative interfaces for nfsd
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Note that normally these interfaces are used only by the utilities in
+nfs-utils.
+
+nfsd is controlled mainly by pseudofiles under the "nfsd" filesystem,
+which is normally mounted at /proc/fs/nfsd/.
+
+The server is always started by the first write of a nonzero value to
+nfsd/threads.
+
+Before doing that, NFSD can be told which sockets to listen on by
+writing to nfsd/portlist; that write may be:
+
+	- an ascii-encoded file descriptor, which should refer to a
+	  bound (and listening, for tcp) socket, or
+	- "transportname port", where transportname is currently either
+	  "udp", "tcp", or "rdma".
+
+If nfsd is started without doing any of these, then it will create one
+udp and one tcp listener at port 2049 (see nfsd_init_socks).
+
+On startup, nfsd and lockd grace periods start.
+
+nfsd is shut down by a write of 0 to nfsd/threads.  All locks and state
+are thrown away at that point.
+
+Between startup and shutdown, the number of threads may be adjusted up
+or down by additional writes to nfsd/threads or by writes to
+nfsd/pool_threads.
+
+For more detail about files under nfsd/ and what they control, see
+fs/nfsd/nfsctl.c; most of them have detailed comments.
+
+Implementation notes
+^^^^^^^^^^^^^^^^^^^^
+
+Note that the rpc server requires the caller to serialize addition and
+removal of listening sockets, and startup and shutdown of the server.
+For nfsd this is done using nfsd_mutex.

fs/lockd/svc.c

@@ -126,7 +126,7 @@ static void restart_grace(void)
 static int
 lockd(void *vrqstp)
 {
-	int		err = 0, preverr = 0;
+	int		err = 0;
 	struct svc_rqst *rqstp = vrqstp;
 
 	/* try_to_freeze() is called from svc_recv() */
@@ -165,21 +165,8 @@ lockd(void *vrqstp)
 		 * recvfrom routine.
 		 */
 		err = svc_recv(rqstp, timeout);
-		if (err == -EAGAIN || err == -EINTR) {
-			preverr = err;
+		if (err == -EAGAIN || err == -EINTR)
 			continue;
-		}
-		if (err < 0) {
-			if (err != preverr) {
-				printk(KERN_WARNING "%s: unexpected error "
-					"from svc_recv (%d)\n", __func__, err);
-				preverr = err;
-			}
-			schedule_timeout_interruptible(HZ);
-			continue;
-		}
-		preverr = err;
-
 		dprintk("lockd: request from %s\n",
 				svc_print_addr(rqstp, buf, sizeof(buf)));
 

fs/locks.c

@@ -1289,7 +1289,7 @@ EXPORT_SYMBOL(__break_lease);
 void lease_get_mtime(struct inode *inode, struct timespec *time)
 {
 	struct file_lock *flock = inode->i_flock;
-	if (flock && IS_LEASE(flock) && (flock->fl_type & F_WRLCK))
+	if (flock && IS_LEASE(flock) && (flock->fl_type == F_WRLCK))
 		*time = current_fs_time(inode->i_sb);
 	else
 		*time = inode->i_mtime;
@@ -2185,8 +2185,8 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 	} else {
 		seq_printf(f, "%s ",
 			       (lease_breaking(fl))
-			       ? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ "
-			       : (fl->fl_type & F_WRLCK) ? "WRITE" : "READ ");
+			       ? (fl->fl_type == F_UNLCK) ? "UNLCK" : "READ "
+			       : (fl->fl_type == F_WRLCK) ? "WRITE" : "READ ");
 	}
 	if (inode) {
 #ifdef WE_CAN_BREAK_LSLK_NOW

fs/nfs/callback.c

@@ -72,7 +72,7 @@ out_err:
 static int
 nfs4_callback_svc(void *vrqstp)
 {
-	int err, preverr = 0;
+	int err;
 	struct svc_rqst *rqstp = vrqstp;
 
 	set_freezable();
@@ -82,20 +82,8 @@ nfs4_callback_svc(void *vrqstp)
 		 * Listen for a request on the socket
 		 */
 		err = svc_recv(rqstp, MAX_SCHEDULE_TIMEOUT);
-		if (err == -EAGAIN || err == -EINTR) {
-			preverr = err;
+		if (err == -EAGAIN || err == -EINTR)
 			continue;
-		}
-		if (err < 0) {
-			if (err != preverr) {
-				printk(KERN_WARNING "NFS: %s: unexpected error "
-					"from svc_recv (%d)\n", __func__, err);
-				preverr = err;
-			}
-			schedule_timeout_uninterruptible(HZ);
-			continue;
-		}
-		preverr = err;
 		svc_process(rqstp);
 	}
 	return 0;

fs/nfsd/nfs2acl.c

@@ -218,8 +218,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p,
  * There must be an encoding function for void results so svc_process
  * will work properly.
  */
-int
-nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
+static int nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
 {
 	return xdr_ressize_check(rqstp, p);
 }

fs/nfsd/nfs3proc.c

@@ -247,7 +247,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp,
 	/* Now create the file and set attributes */
 	nfserr = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len,
 				attr, newfhp,
-				argp->createmode, argp->verf, NULL, NULL);
+				argp->createmode, (u32 *)argp->verf, NULL, NULL);
 
 	RETURN_STATUS(nfserr);
 }

fs/nfsd/nfs4callback.c

@@ -1028,7 +1028,6 @@ void nfsd4_cb_recall(struct nfs4_delegation *dp)
 	cb->cb_msg.rpc_cred = callback_cred;
 
 	cb->cb_ops = &nfsd4_cb_recall_ops;
-	dp->dl_retries = 1;
 
 	INIT_LIST_HEAD(&cb->cb_per_client);
 	cb->cb_done = true;

fs/nfsd/nfs4idmap.c

@@ -478,7 +478,7 @@ nfsd_idmap_init(struct net *net)
 		goto destroy_idtoname_cache;
 	nn->nametoid_cache = cache_create_net(&nametoid_cache_template, net);
 	if (IS_ERR(nn->nametoid_cache)) {
-		rv = PTR_ERR(nn->idtoname_cache);
+		rv = PTR_ERR(nn->nametoid_cache);
 		goto unregister_idtoname_cache;
 	}
 	rv = cache_register_net(nn->nametoid_cache, net);
@@ -598,7 +598,7 @@ numeric_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namel
 	/* Just to make sure it's null-terminated: */
 	memcpy(buf, name, namelen);
 	buf[namelen] = '\0';
-	ret = kstrtouint(name, 10, id);
+	ret = kstrtouint(buf, 10, id);
 	return ret == 0;
 }
 

fs/nfsd/nfs4proc.c

@@ -370,7 +370,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 			break;
 		case NFS4_OPEN_CLAIM_PREVIOUS:
 			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
-			status = nfs4_check_open_reclaim(&open->op_clientid);
+			status = nfs4_check_open_reclaim(&open->op_clientid, cstate->minorversion);
 			if (status)
 				goto out;
 		case NFS4_OPEN_CLAIM_FH:
@@ -1054,8 +1054,8 @@ struct nfsd4_operation {
 	char *op_name;
 	/* Try to get response size before operation */
 	nfsd4op_rsize op_rsize_bop;
-	stateid_setter op_get_currentstateid;
-	stateid_getter op_set_currentstateid;
+	stateid_getter op_get_currentstateid;
+	stateid_setter op_set_currentstateid;
 };
 
 static struct nfsd4_operation nfsd4_ops[];

fs/nfsd/nfs4state.c

@@ -758,7 +758,7 @@ static void nfsd4_put_drc_mem(int slotsize, int num)
 	spin_unlock(&nfsd_drc_lock);
 }
 
-static struct nfsd4_session *alloc_session(int slotsize, int numslots)
+static struct nfsd4_session *__alloc_session(int slotsize, int numslots)
 {
 	struct nfsd4_session *new;
 	int mem, i;
@@ -852,35 +852,28 @@ static int nfsd4_register_conn(struct nfsd4_conn *conn)
 	return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user);
 }
 
-static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses, u32 dir)
+static void nfsd4_init_conn(struct svc_rqst *rqstp, struct nfsd4_conn *conn, struct nfsd4_session *ses)
 {
-	struct nfsd4_conn *conn;
 	int ret;
 
-	conn = alloc_conn(rqstp, dir);
-	if (!conn)
-		return nfserr_jukebox;
 	nfsd4_hash_conn(conn, ses);
 	ret = nfsd4_register_conn(conn);
 	if (ret)
 		/* oops; xprt is already down: */
 		nfsd4_conn_lost(&conn->cn_xpt_user);
-	if (ses->se_client->cl_cb_state == NFSD4_CB_DOWN &&
-		dir & NFS4_CDFC4_BACK) {
+	if (conn->cn_flags & NFS4_CDFC4_BACK) {
 		/* callback channel may be back up */
 		nfsd4_probe_callback(ses->se_client);
 	}
-	return nfs_ok;
 }
 
-static __be32 nfsd4_new_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_session *ses)
+static struct nfsd4_conn *alloc_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_create_session *cses)
 {
 	u32 dir = NFS4_CDFC4_FORE;
 
-	if (ses->se_flags & SESSION4_BACK_CHAN)
+	if (cses->flags & SESSION4_BACK_CHAN)
 		dir |= NFS4_CDFC4_BACK;
-
-	return nfsd4_new_conn(rqstp, ses, dir);
+	return alloc_conn(rqstp, dir);
 }
 
 /* must be called under client_lock */
@@ -903,20 +896,21 @@ static void nfsd4_del_conns(struct nfsd4_session *s)
 	spin_unlock(&clp->cl_lock);
 }
 
+static void __free_session(struct nfsd4_session *ses)
+{
+	nfsd4_put_drc_mem(slot_bytes(&ses->se_fchannel), ses->se_fchannel.maxreqs);
+	free_session_slots(ses);
+	kfree(ses);
+}
+
 static void free_session(struct kref *kref)
 {
 	struct nfsd4_session *ses;
-	int mem;
 
 	lockdep_assert_held(&client_lock);
 	ses = container_of(kref, struct nfsd4_session, se_ref);
 	nfsd4_del_conns(ses);
-	spin_lock(&nfsd_drc_lock);
-	mem = ses->se_fchannel.maxreqs * slot_bytes(&ses->se_fchannel);
-	nfsd_drc_mem_used -= mem;
-	spin_unlock(&nfsd_drc_lock);
-	free_session_slots(ses);
-	kfree(ses);
+	__free_session(ses);
 }
 
 void nfsd4_put_session(struct nfsd4_session *ses)
@@ -926,14 +920,10 @@ void nfsd4_put_session(struct nfsd4_session *ses)
 	spin_unlock(&client_lock);
 }
 
-static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp, struct nfsd4_create_session *cses)
+static struct nfsd4_session *alloc_session(struct nfsd4_channel_attrs *fchan)
 {
 	struct nfsd4_session *new;
-	struct nfsd4_channel_attrs *fchan = &cses->fore_channel;
 	int numslots, slotsize;
-	__be32 status;
-	int idx;
-
 	/*
 	 * Note decreasing slot size below client's request may
 	 * make it difficult for client to function correctly, whereas
@@ -946,12 +936,18 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n
 	if (numslots < 1)
 		return NULL;
 
-	new = alloc_session(slotsize, numslots);
+	new = __alloc_session(slotsize, numslots);
 	if (!new) {
 		nfsd4_put_drc_mem(slotsize, fchan->maxreqs);
 		return NULL;
 	}
 	init_forechannel_attrs(&new->se_fchannel, fchan, numslots, slotsize);
+	return new;
+}
+
+static struct nfsd4_session *init_session(struct svc_rqst *rqstp, struct nfsd4_session *new, struct nfs4_client *clp, struct nfsd4_create_session *cses)
+{
+	int idx;
 
 	new->se_client = clp;
 	gen_sessionid(new);
@@ -970,14 +966,6 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n
 	spin_unlock(&clp->cl_lock);
 	spin_unlock(&client_lock);
 
-	status = nfsd4_new_conn_from_crses(rqstp, new);
-	/* whoops: benny points out, status is ignored! (err, or bogus) */
-	if (status) {
-		spin_lock(&client_lock);
-		free_session(&new->se_ref);
-		spin_unlock(&client_lock);
-		return NULL;
-	}
 	if (cses->flags & SESSION4_BACK_CHAN) {
 		struct sockaddr *sa = svc_addr(rqstp);
 		/*
@@ -990,7 +978,6 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n
 		rpc_copy_addr((struct sockaddr *)&clp->cl_cb_conn.cb_addr, sa);
 		clp->cl_cb_conn.cb_addrlen = svc_addr_len(sa);
 	}
-	nfsd4_probe_callback(clp);
 	return new;
 }
 
@@ -1131,7 +1118,7 @@ unhash_client_locked(struct nfs4_client *clp)
 }
 
 static void
-expire_client(struct nfs4_client *clp)
+destroy_client(struct nfs4_client *clp)
 {
 	struct nfs4_openowner *oo;
 	struct nfs4_delegation *dp;
@@ -1165,6 +1152,12 @@ expire_client(struct nfs4_client *clp)
 	spin_unlock(&client_lock);
 }
 
+static void expire_client(struct nfs4_client *clp)
+{
+	nfsd4_client_record_remove(clp);
+	destroy_client(clp);
+}
+
 static void copy_verf(struct nfs4_client *target, nfs4_verifier *source)
 {
 	memcpy(target->cl_verifier.data, source->data,
@@ -1223,10 +1216,26 @@ static bool groups_equal(struct group_info *g1, struct group_info *g2)
 	return true;
 }
 
+/*
+ * RFC 3530 language requires clid_inuse be returned when the
+ * "principal" associated with a requests differs from that previously
+ * used.  We use uid, gid's, and gss principal string as our best
+ * approximation.  We also don't want to allow non-gss use of a client
+ * established using gss: in theory cr_principal should catch that
+ * change, but in practice cr_principal can be null even in the gss case
+ * since gssd doesn't always pass down a principal string.
+ */
+static bool is_gss_cred(struct svc_cred *cr)
+{
+	/* Is cr_flavor one of the gss "pseudoflavors"?: */
+	return (cr->cr_flavor > RPC_AUTH_MAXFLAVOR);
+}
+
+
 static bool
 same_creds(struct svc_cred *cr1, struct svc_cred *cr2)
 {
-	if ((cr1->cr_flavor != cr2->cr_flavor)
+	if ((is_gss_cred(cr1) != is_gss_cred(cr2))
 		|| (cr1->cr_uid != cr2->cr_uid)
 		|| (cr1->cr_gid != cr2->cr_gid)
 		|| !groups_equal(cr1->cr_group_info, cr2->cr_group_info))
@@ -1340,13 +1349,15 @@ move_to_confirmed(struct nfs4_client *clp)
 }
 
 static struct nfs4_client *
-find_confirmed_client(clientid_t *clid)
+find_confirmed_client(clientid_t *clid, bool sessions)
 {
 	struct nfs4_client *clp;
 	unsigned int idhashval = clientid_hashval(clid->cl_id);
 
 	list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) {
 		if (same_clid(&clp->cl_clientid, clid)) {
+			if ((bool)clp->cl_minorversion != sessions)
+				return NULL;
 			renew_client(clp);
 			return clp;
 		}
@@ -1355,14 +1366,17 @@ find_confirmed_client(clientid_t *clid)
 }
 
 static struct nfs4_client *
-find_unconfirmed_client(clientid_t *clid)
+find_unconfirmed_client(clientid_t *clid, bool sessions)
 {
 	struct nfs4_client *clp;
 	unsigned int idhashval = clientid_hashval(clid->cl_id);
 
 	list_for_each_entry(clp, &unconf_id_hashtbl[idhashval], cl_idhash) {
-		if (same_clid(&clp->cl_clientid, clid))
+		if (same_clid(&clp->cl_clientid, clid)) {
+			if ((bool)clp->cl_minorversion != sessions)
+				return NULL;
 			return clp;
+		}
 	}
 	return NULL;
 }
@@ -1651,6 +1665,7 @@ out_new:
 		status = nfserr_jukebox;
 		goto out;
 	}
+	new->cl_minorversion = 1;
 
 	gen_clid(new);
 	add_to_unconfirmed(new, strhashval);
@@ -1743,67 +1758,71 @@ nfsd4_create_session(struct svc_rqst *rqstp,
 	struct sockaddr *sa = svc_addr(rqstp);
 	struct nfs4_client *conf, *unconf;
 	struct nfsd4_session *new;
+	struct nfsd4_conn *conn;
 	struct nfsd4_clid_slot *cs_slot = NULL;
-	bool confirm_me = false;
 	__be32 status = 0;
 
 	if (cr_ses->flags & ~SESSION4_FLAG_MASK_A)
 		return nfserr_inval;
+	if (check_forechannel_attrs(cr_ses->fore_channel))
+		return nfserr_toosmall;
+	new = alloc_session(&cr_ses->fore_channel);
+	if (!new)
+		return nfserr_jukebox;
+	status = nfserr_jukebox;
+	conn = alloc_conn_from_crses(rqstp, cr_ses);
+	if (!conn)
+		goto out_free_session;
 
 	nfs4_lock_state();
-	unconf = find_unconfirmed_client(&cr_ses->clientid);
-	conf = find_confirmed_client(&cr_ses->clientid);
+	unconf = find_unconfirmed_client(&cr_ses->clientid, true);
+	conf = find_confirmed_client(&cr_ses->clientid, true);
 
 	if (conf) {
 		cs_slot = &conf->cl_cs_slot;
 		status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0);
 		if (status == nfserr_replay_cache) {
 			status = nfsd4_replay_create_session(cr_ses, cs_slot);
-			goto out;
+			goto out_free_conn;
 		} else if (cr_ses->seqid != cs_slot->sl_seqid + 1) {
 			status = nfserr_seq_misordered;
-			goto out;
+			goto out_free_conn;
 		}
 	} else if (unconf) {
+		unsigned int hash;
+		struct nfs4_client *old;
 		if (!same_creds(&unconf->cl_cred, &rqstp->rq_cred) ||
 		    !rpc_cmp_addr(sa, (struct sockaddr *) &unconf->cl_addr)) {
 			status = nfserr_clid_inuse;
-			goto out;
+			goto out_free_conn;
 		}
 		cs_slot = &unconf->cl_cs_slot;
 		status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0);
 		if (status) {
 			/* an unconfirmed replay returns misordered */
 			status = nfserr_seq_misordered;
-			goto out;
+			goto out_free_conn;
 		}
-		confirm_me = true;
+		hash = clientstr_hashval(unconf->cl_recdir);
+		old = find_confirmed_client_by_str(unconf->cl_recdir, hash);
+		if (old)
+			expire_client(old);
+		move_to_confirmed(unconf);
 		conf = unconf;
 	} else {
 		status = nfserr_stale_clientid;
-		goto out;
+		goto out_free_conn;
 	}
-
-	/*
-	 * XXX: we should probably set this at creation time, and check
-	 * for consistent minorversion use throughout:
-	 */
-	conf->cl_minorversion = 1;
+	status = nfs_ok;
 	/*
 	 * We do not support RDMA or persistent sessions
 	 */
 	cr_ses->flags &= ~SESSION4_PERSIST;
 	cr_ses->flags &= ~SESSION4_RDMA;
 
-	status = nfserr_toosmall;
-	if (check_forechannel_attrs(cr_ses->fore_channel))
-		goto out;
+	init_session(rqstp, new, conf, cr_ses);
+	nfsd4_init_conn(rqstp, conn, new);
 
-	status = nfserr_jukebox;
-	new = alloc_init_session(rqstp, conf, cr_ses);
-	if (!new)
-		goto out;
-	status = nfs_ok;
 	memcpy(cr_ses->sessionid.data, new->se_sessionid.data,
 	       NFS4_MAX_SESSIONID_LEN);
 	memcpy(&cr_ses->fore_channel, &new->se_fchannel,
@@ -1813,18 +1832,15 @@ nfsd4_create_session(struct svc_rqst *rqstp,
 
 	/* cache solo and embedded create sessions under the state lock */
 	nfsd4_cache_create_session(cr_ses, cs_slot, status);
-	if (confirm_me) {
-		unsigned int hash = clientstr_hashval(unconf->cl_recdir);
-		struct nfs4_client *old =
-			find_confirmed_client_by_str(conf->cl_recdir, hash);
-		if (old)
-			expire_client(old);
-		move_to_confirmed(conf);
-	}
 out:
 	nfs4_unlock_state();
 	dprintk("%s returns %d\n", __func__, ntohl(status));
 	return status;
+out_free_conn:
+	free_conn(conn);
+out_free_session:
+	__free_session(new);
+	goto out;
 }
 
 static bool nfsd4_last_compound_op(struct svc_rqst *rqstp)
@@ -1854,6 +1870,7 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
 		     struct nfsd4_bind_conn_to_session *bcts)
 {
 	__be32 status;
+	struct nfsd4_conn *conn;
 
 	if (!nfsd4_last_compound_op(rqstp))
 		return nfserr_not_only_op;
@@ -1870,9 +1887,13 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
 		return nfserr_badsession;
 
 	status = nfsd4_map_bcts_dir(&bcts->dir);
-	if (!status)
-		nfsd4_new_conn(rqstp, cstate->session, bcts->dir);
-	return status;
+	if (status)
+		return status;
+	conn = alloc_conn(rqstp, bcts->dir);
+	if (!conn)
+		return nfserr_jukebox;
+	nfsd4_init_conn(rqstp, conn, cstate->session);
+	return nfs_ok;
 }
 
 static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid)
@@ -2085,8 +2106,8 @@ nfsd4_destroy_clientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta
 	__be32 status = 0;
 
 	nfs4_lock_state();
-	unconf = find_unconfirmed_client(&dc->clientid);
-	conf = find_confirmed_client(&dc->clientid);
+	unconf = find_unconfirmed_client(&dc->clientid, true);
+	conf = find_confirmed_client(&dc->clientid, true);
 
 	if (conf) {
 		clp = conf;
@@ -2200,10 +2221,6 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		copy_clid(new, conf);
 	else /* case 4 (new client) or cases 2, 3 (client reboot): */
 		gen_clid(new);
-	/*
-	 * XXX: we should probably set this at creation time, and check
-	 * for consistent minorversion use throughout:
-	 */
 	new->cl_minorversion = 0;
 	gen_callback(new, setclid, rqstp);
 	add_to_unconfirmed(new, strhashval);
@@ -2232,8 +2249,8 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 		return nfserr_stale_clientid;
 	nfs4_lock_state();
 
-	conf = find_confirmed_client(clid);
-	unconf = find_unconfirmed_client(clid);
+	conf = find_confirmed_client(clid, false);
+	unconf = find_unconfirmed_client(clid, false);
 	/*
 	 * We try hard to give out unique clientid's, so if we get an
 	 * attempt to confirm the same clientid with a different cred,
@@ -2262,10 +2279,8 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 		unsigned int hash = clientstr_hashval(unconf->cl_recdir);
 
 		conf = find_confirmed_client_by_str(unconf->cl_recdir, hash);
-		if (conf) {
-			nfsd4_client_record_remove(conf);
+		if (conf)
 			expire_client(conf);
-		}
 		move_to_confirmed(unconf);
 		nfsd4_probe_callback(unconf);
 	}
@@ -2447,16 +2462,20 @@ same_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner,
 }
 
 static struct nfs4_openowner *
-find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open)
+find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open, bool sessions)
 {
 	struct nfs4_stateowner *so;
 	struct nfs4_openowner *oo;
+	struct nfs4_client *clp;
 
 	list_for_each_entry(so, &ownerstr_hashtbl[hashval], so_strhash) {
 		if (!so->so_is_open_owner)
 			continue;
 		if (same_owner_str(so, &open->op_owner, &open->op_clientid)) {
 			oo = openowner(so);
+			clp = oo->oo_owner.so_client;
+			if ((bool)clp->cl_minorversion != sessions)
+				return NULL;
 			renew_client(oo->oo_owner.so_client);
 			return oo;
 		}
@@ -2600,10 +2619,10 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate,
 		return nfserr_jukebox;
 
 	strhashval = ownerstr_hashval(clientid->cl_id, &open->op_owner);
-	oo = find_openstateowner_str(strhashval, open);
+	oo = find_openstateowner_str(strhashval, open, cstate->minorversion);
 	open->op_openowner = oo;
 	if (!oo) {
-		clp = find_confirmed_client(clientid);
+		clp = find_confirmed_client(clientid, cstate->minorversion);
 		if (clp == NULL)
 			return nfserr_expired;
 		goto new_owner;
@@ -2705,11 +2724,6 @@ nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_ol_st
 	return nfs_ok;
 }
 
-static void nfs4_free_stateid(struct nfs4_ol_stateid *s)
-{
-	kmem_cache_free(stateid_slab, s);
-}
-
 static inline int nfs4_access_to_access(u32 nfs4_access)
 {
 	int flags = 0;
@@ -3087,7 +3101,7 @@ void nfsd4_cleanup_open_state(struct nfsd4_open *open, __be32 status)
 	if (open->op_file)
 		nfsd4_free_file(open->op_file);
 	if (open->op_stp)
-		nfs4_free_stateid(open->op_stp);
+		free_generic_stateid(open->op_stp);
 }
 
 __be32
@@ -3104,7 +3118,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	status = nfserr_stale_clientid;
 	if (STALE_CLIENTID(clid, nn))
 		goto out;
-	clp = find_confirmed_client(clid);
+	clp = find_confirmed_client(clid, cstate->minorversion);
 	status = nfserr_expired;
 	if (clp == NULL) {
 		/* We assume the client took too long to RENEW. */
@@ -3180,7 +3194,6 @@ nfs4_laundromat(void)
 		clp = list_entry(pos, struct nfs4_client, cl_lru);
 		dprintk("NFSD: purging unused client (clientid %08x)\n",
 			clp->cl_clientid.cl_id);
-		nfsd4_client_record_remove(clp);
 		expire_client(clp);
 	}
 	spin_lock(&recall_lock);
@@ -3372,7 +3385,7 @@ static __be32 nfsd4_validate_stateid(struct nfs4_client *cl, stateid_t *stateid)
 	return nfs_ok;
 }
 
-static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s)
+static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s, bool sessions)
 {
 	struct nfs4_client *cl;
 	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
@@ -3381,7 +3394,7 @@ static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, s
 		return nfserr_bad_stateid;
 	if (STALE_STATEID(stateid, nn))
 		return nfserr_stale_stateid;
-	cl = find_confirmed_client(&stateid->si_opaque.so_clid);
+	cl = find_confirmed_client(&stateid->si_opaque.so_clid, sessions);
 	if (!cl)
 		return nfserr_expired;
 	*s = find_stateid_by_type(cl, stateid, typemask);
@@ -3414,7 +3427,7 @@ nfs4_preprocess_stateid_op(struct net *net, struct nfsd4_compound_state *cstate,
 	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid))
 		return check_special_stateids(net, current_fh, stateid, flags);
 
-	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, &s);
+	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, &s, cstate->minorversion);
 	if (status)
 		return status;
 	status = check_stateid_generation(stateid, &s->sc_stateid, nfsd4_has_session(cstate));
@@ -3564,7 +3577,7 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid,
 		seqid, STATEID_VAL(stateid));
 
 	*stpp = NULL;
-	status = nfsd4_lookup_stateid(stateid, typemask, &s);
+	status = nfsd4_lookup_stateid(stateid, typemask, &s, cstate->minorversion);
 	if (status)
 		return status;
 	*stpp = openlockstateid(s);
@@ -3765,6 +3778,7 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	memcpy(&close->cl_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 
 	nfsd4_close_open_stateid(stp);
+	release_last_closed_stateid(oo);
 	oo->oo_last_closed_stid = stp;
 
 	if (list_empty(&oo->oo_owner.so_stateids)) {
@@ -3801,7 +3815,7 @@ nfsd4_delegreturn(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	inode = cstate->current_fh.fh_dentry->d_inode;
 
 	nfs4_lock_state();
-	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID, &s);
+	status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID, &s, cstate->minorversion);
 	if (status)
 		goto out;
 	dp = delegstateid(s);
@@ -4045,8 +4059,8 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	struct nfs4_lockowner *lock_sop = NULL;
 	struct nfs4_ol_stateid *lock_stp;
 	struct file *filp = NULL;
-	struct file_lock file_lock;
-	struct file_lock conflock;
+	struct file_lock *file_lock = NULL;
+	struct file_lock *conflock = NULL;
 	__be32 status = 0;
 	bool new_state = false;
 	int lkflg;
@@ -4116,21 +4130,28 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (!locks_in_grace(SVC_NET(rqstp)) && lock->lk_reclaim)
 		goto out;
 
-	locks_init_lock(&file_lock);
+	file_lock = locks_alloc_lock();
+	if (!file_lock) {
+		dprintk("NFSD: %s: unable to allocate lock!\n", __func__);
+		status = nfserr_jukebox;
+		goto out;
+	}
+
+	locks_init_lock(file_lock);
 	switch (lock->lk_type) {
 		case NFS4_READ_LT:
 		case NFS4_READW_LT:
 			filp = find_readable_file(lock_stp->st_file);
 			if (filp)
 				get_lock_access(lock_stp, NFS4_SHARE_ACCESS_READ);
-			file_lock.fl_type = F_RDLCK;
+			file_lock->fl_type = F_RDLCK;
 			break;
 		case NFS4_WRITE_LT:
 		case NFS4_WRITEW_LT:
 			filp = find_writeable_file(lock_stp->st_file);
 			if (filp)
 				get_lock_access(lock_stp, NFS4_SHARE_ACCESS_WRITE);
-			file_lock.fl_type = F_WRLCK;
+			file_lock->fl_type = F_WRLCK;
 			break;
 		default:
 			status = nfserr_inval;
@@ -4140,22 +4161,23 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		status = nfserr_openmode;
 		goto out;
 	}
-	file_lock.fl_owner = (fl_owner_t)lock_sop;
-	file_lock.fl_pid = current->tgid;
-	file_lock.fl_file = filp;
-	file_lock.fl_flags = FL_POSIX;
-	file_lock.fl_lmops = &nfsd_posix_mng_ops;
-
-	file_lock.fl_start = lock->lk_offset;
-	file_lock.fl_end = last_byte_offset(lock->lk_offset, lock->lk_length);
-	nfs4_transform_lock_offset(&file_lock);
-
-	/*
-	* Try to lock the file in the VFS.
-	* Note: locks.c uses the BKL to protect the inode's lock list.
-	*/
+	file_lock->fl_owner = (fl_owner_t)lock_sop;
+	file_lock->fl_pid = current->tgid;
+	file_lock->fl_file = filp;
+	file_lock->fl_flags = FL_POSIX;
+	file_lock->fl_lmops = &nfsd_posix_mng_ops;
+	file_lock->fl_start = lock->lk_offset;
+	file_lock->fl_end = last_byte_offset(lock->lk_offset, lock->lk_length);
+	nfs4_transform_lock_offset(file_lock);
+
+	conflock = locks_alloc_lock();
+	if (!conflock) {
+		dprintk("NFSD: %s: unable to allocate lock!\n", __func__);
+		status = nfserr_jukebox;
+		goto out;
+	}
 
-	err = vfs_lock_file(filp, F_SETLK, &file_lock, &conflock);
+	err = vfs_lock_file(filp, F_SETLK, file_lock, conflock);
 	switch (-err) {
 	case 0: /* success! */
 		update_stateid(&lock_stp->st_stid.sc_stateid);
@@ -4166,7 +4188,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	case (EAGAIN):		/* conflock holds conflicting lock */
 		status = nfserr_denied;
 		dprintk("NFSD: nfsd4_lock: conflicting lock found!\n");
-		nfs4_set_lock_denied(&conflock, &lock->lk_denied);
+		nfs4_set_lock_denied(conflock, &lock->lk_denied);
 		break;
 	case (EDEADLK):
 		status = nfserr_deadlock;
@@ -4181,6 +4203,10 @@ out:
 		release_lockowner(lock_sop);
 	if (!cstate->replay_owner)
 		nfs4_unlock_state();
+	if (file_lock)
+		locks_free_lock(file_lock);
+	if (conflock)
+		locks_free_lock(conflock);
 	return status;
 }
 
@@ -4209,7 +4235,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    struct nfsd4_lockt *lockt)
 {
 	struct inode *inode;
-	struct file_lock file_lock;
+	struct file_lock *file_lock = NULL;
 	struct nfs4_lockowner *lo;
 	__be32 status;
 	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
@@ -4230,15 +4256,21 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		goto out;
 
 	inode = cstate->current_fh.fh_dentry->d_inode;
-	locks_init_lock(&file_lock);
+	file_lock = locks_alloc_lock();
+	if (!file_lock) {
+		dprintk("NFSD: %s: unable to allocate lock!\n", __func__);
+		status = nfserr_jukebox;
+		goto out;
+	}
+	locks_init_lock(file_lock);
 	switch (lockt->lt_type) {
 		case NFS4_READ_LT:
 		case NFS4_READW_LT:
-			file_lock.fl_type = F_RDLCK;
+			file_lock->fl_type = F_RDLCK;
 		break;
 		case NFS4_WRITE_LT:
 		case NFS4_WRITEW_LT:
-			file_lock.fl_type = F_WRLCK;
+			file_lock->fl_type = F_WRLCK;
 		break;
 		default:
 			dprintk("NFSD: nfs4_lockt: bad lock type!\n");
@@ -4248,25 +4280,27 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	lo = find_lockowner_str(inode, &lockt->lt_clientid, &lockt->lt_owner);
 	if (lo)
-		file_lock.fl_owner = (fl_owner_t)lo;
-	file_lock.fl_pid = current->tgid;
-	file_lock.fl_flags = FL_POSIX;
+		file_lock->fl_owner = (fl_owner_t)lo;
+	file_lock->fl_pid = current->tgid;
+	file_lock->fl_flags = FL_POSIX;
 
-	file_lock.fl_start = lockt->lt_offset;
-	file_lock.fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length);
+	file_lock->fl_start = lockt->lt_offset;
+	file_lock->fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length);
 
-	nfs4_transform_lock_offset(&file_lock);
+	nfs4_transform_lock_offset(file_lock);
 
-	status = nfsd_test_lock(rqstp, &cstate->current_fh, &file_lock);
+	status = nfsd_test_lock(rqstp, &cstate->current_fh, file_lock);
 	if (status)
 		goto out;
 
-	if (file_lock.fl_type != F_UNLCK) {
+	if (file_lock->fl_type != F_UNLCK) {
 		status = nfserr_denied;
-		nfs4_set_lock_denied(&file_lock, &lockt->lt_denied);
+		nfs4_set_lock_denied(file_lock, &lockt->lt_denied);
 	}
 out:
 	nfs4_unlock_state();
+	if (file_lock)
+		locks_free_lock(file_lock);
 	return status;
 }
 
@@ -4276,7 +4310,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	struct nfs4_ol_stateid *stp;
 	struct file *filp = NULL;
-	struct file_lock file_lock;
+	struct file_lock *file_lock = NULL;
 	__be32 status;
 	int err;
 						        
@@ -4298,23 +4332,29 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		status = nfserr_lock_range;
 		goto out;
 	}
-	BUG_ON(!filp);
-	locks_init_lock(&file_lock);
-	file_lock.fl_type = F_UNLCK;
-	file_lock.fl_owner = (fl_owner_t)lockowner(stp->st_stateowner);
-	file_lock.fl_pid = current->tgid;
-	file_lock.fl_file = filp;
-	file_lock.fl_flags = FL_POSIX; 
-	file_lock.fl_lmops = &nfsd_posix_mng_ops;
-	file_lock.fl_start = locku->lu_offset;
-
-	file_lock.fl_end = last_byte_offset(locku->lu_offset, locku->lu_length);
-	nfs4_transform_lock_offset(&file_lock);
+	file_lock = locks_alloc_lock();
+	if (!file_lock) {
+		dprintk("NFSD: %s: unable to allocate lock!\n", __func__);
+		status = nfserr_jukebox;
+		goto out;
+	}
+	locks_init_lock(file_lock);
+	file_lock->fl_type = F_UNLCK;
+	file_lock->fl_owner = (fl_owner_t)lockowner(stp->st_stateowner);
+	file_lock->fl_pid = current->tgid;
+	file_lock->fl_file = filp;
+	file_lock->fl_flags = FL_POSIX;
+	file_lock->fl_lmops = &nfsd_posix_mng_ops;
+	file_lock->fl_start = locku->lu_offset;
+
+	file_lock->fl_end = last_byte_offset(locku->lu_offset,
+						locku->lu_length);
+	nfs4_transform_lock_offset(file_lock);
 
 	/*
 	*  Try to unlock the file in the VFS.
 	*/
-	err = vfs_lock_file(filp, F_SETLK, &file_lock, NULL);
+	err = vfs_lock_file(filp, F_SETLK, file_lock, NULL);
 	if (err) {
 		dprintk("NFSD: nfs4_locku: vfs_lock_file failed!\n");
 		goto out_nfserr;
@@ -4328,6 +4368,8 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 out:
 	if (!cstate->replay_owner)
 		nfs4_unlock_state();
+	if (file_lock)
+		locks_free_lock(file_lock);
 	return status;
 
 out_nfserr:
@@ -4501,12 +4543,12 @@ nfsd4_find_reclaim_client(struct nfs4_client *clp)
 * Called from OPEN. Look for clientid in reclaim list.
 */
 __be32
-nfs4_check_open_reclaim(clientid_t *clid)
+nfs4_check_open_reclaim(clientid_t *clid, bool sessions)
 {
 	struct nfs4_client *clp;
 
 	/* find clientid in conf_id_hashtbl */
-	clp = find_confirmed_client(clid);
+	clp = find_confirmed_client(clid, sessions);
 	if (clp == NULL)
 		return nfserr_reclaim_bad;
 
@@ -4522,7 +4564,6 @@ void nfsd_forget_clients(u64 num)
 
 	nfs4_lock_state();
 	list_for_each_entry_safe(clp, next, &client_lru, cl_lru) {
-		nfsd4_client_record_remove(clp);
 		expire_client(clp);
 		if (++count == num)
 			break;
@@ -4582,7 +4623,7 @@ void nfsd_forget_openowners(u64 num)
 	printk(KERN_INFO "NFSD: Forgot %d open owners", count);
 }
 
-int nfsd_process_n_delegations(u64 num, struct list_head *list)
+static int nfsd_process_n_delegations(u64 num, struct list_head *list)
 {
 	int i, count = 0;
 	struct nfs4_file *fp, *fnext;
@@ -4747,11 +4788,11 @@ __nfs4_state_shutdown(void)
 	for (i = 0; i < CLIENT_HASH_SIZE; i++) {
 		while (!list_empty(&conf_id_hashtbl[i])) {
 			clp = list_entry(conf_id_hashtbl[i].next, struct nfs4_client, cl_idhash);
-			expire_client(clp);
+			destroy_client(clp);
 		}
 		while (!list_empty(&unconf_str_hashtbl[i])) {
 			clp = list_entry(unconf_str_hashtbl[i].next, struct nfs4_client, cl_strhash);
-			expire_client(clp);
+			destroy_client(clp);
 		}
 	}
 	INIT_LIST_HEAD(&reaplist);

fs/nfsd/nfs4xdr.c

@@ -2659,7 +2659,7 @@ static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp,
 		RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 8);
 		WRITEMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN);
 		WRITE32(bcts->dir);
-		/* XXX: ? */
+		/* Sorry, we do not yet support RDMA over 4.1: */
 		WRITE32(0);
 		ADJUST_ARGS();
 	}

fs/nfsd/nfsctl.c

@@ -406,7 +406,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
 			return rv;
 		if (newthreads < 0)
 			return -EINVAL;
-		rv = nfsd_svc(NFS_PORT, newthreads);
+		rv = nfsd_svc(newthreads);
 		if (rv < 0)
 			return rv;
 	} else
@@ -682,25 +682,6 @@ static ssize_t __write_ports_addfd(char *buf)
 	return err;
 }
 
-/*
- * A '-' followed by the 'name' of a socket means we close the socket.
- */
-static ssize_t __write_ports_delfd(char *buf)
-{
-	char *toclose;
-	int len = 0;
-
-	toclose = kstrdup(buf + 1, GFP_KERNEL);
-	if (toclose == NULL)
-		return -ENOMEM;
-
-	if (nfsd_serv != NULL)
-		len = svc_sock_names(nfsd_serv, buf,
-					SIMPLE_TRANSACTION_LIMIT, toclose);
-	kfree(toclose);
-	return len;
-}
-
 /*
  * A transport listener is added by writing it's transport name and
  * a port number.
@@ -712,7 +693,7 @@ static ssize_t __write_ports_addxprt(char *buf)
 	int port, err;
 	struct net *net = &init_net;
 
-	if (sscanf(buf, "%15s %4u", transport, &port) != 2)
+	if (sscanf(buf, "%15s %5u", transport, &port) != 2)
 		return -EINVAL;
 
 	if (port < 1 || port > USHRT_MAX)
@@ -746,31 +727,6 @@ out_err:
 	return err;
 }
 
-/*
- * A transport listener is removed by writing a "-", it's transport
- * name, and it's port number.
- */
-static ssize_t __write_ports_delxprt(char *buf)
-{
-	struct svc_xprt *xprt;
-	char transport[16];
-	int port;
-
-	if (sscanf(&buf[1], "%15s %4u", transport, &port) != 2)
-		return -EINVAL;
-
-	if (port < 1 || port > USHRT_MAX || nfsd_serv == NULL)
-		return -EINVAL;
-
-	xprt = svc_find_xprt(nfsd_serv, transport, &init_net, AF_UNSPEC, port);
-	if (xprt == NULL)
-		return -ENOTCONN;
-
-	svc_close_xprt(xprt);
-	svc_xprt_put(xprt);
-	return 0;
-}
-
 static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 {
 	if (size == 0)
@@ -779,15 +735,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 	if (isdigit(buf[0]))
 		return __write_ports_addfd(buf);
 
-	if (buf[0] == '-' && isdigit(buf[1]))
-		return __write_ports_delfd(buf);
-
 	if (isalpha(buf[0]))
 		return __write_ports_addxprt(buf);
 
-	if (buf[0] == '-' && isalpha(buf[1]))
-		return __write_ports_delxprt(buf);
-
 	return -EINVAL;
 }
 
@@ -825,21 +775,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
  * OR
  *
  * Input:
- *			buf:		C string containing a "-" followed
- *					by an integer value representing a
- *					previously passed in socket file
- *					descriptor
- *			size:		non-zero length of C string in @buf
- * Output:
- *	On success:	NFS service no longer listens on that socket;
- *			passed-in buffer filled with a '\n'-terminated C
- *			string containing a unique name of the listener;
- *			return code is the size in bytes of the string
- *	On error:	return code is a negative errno value
- *
- * OR
- *
- * Input:
  *			buf:		C string containing a transport
  *					name and an unsigned integer value
  *					representing the port to listen on,
@@ -848,19 +783,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
  * Output:
  *	On success:	returns zero; NFS service is started
  *	On error:	return code is a negative errno value
- *
- * OR
- *
- * Input:
- *			buf:		C string containing a "-" followed
- *					by a transport name and an unsigned
- *					integer value representing the port
- *					to listen on, separated by whitespace
- *			size:		non-zero length of C string in @buf
- * Output:
- *	On success:	returns zero; NFS service no longer listens
- *			on that transport
- *	On error:	return code is a negative errno value
  */
 static ssize_t write_ports(struct file *file, char *buf, size_t size)
 {
@@ -1008,8 +930,6 @@ static ssize_t write_gracetime(struct file *file, char *buf, size_t size)
 	return nfsd4_write_time(file, buf, size, &nfsd4_grace);
 }
 
-extern char *nfs4_recoverydir(void);
-
 static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size)
 {
 	char *mesg = buf;

fs/nfsd/nfsd.h

@@ -65,7 +65,7 @@ extern const struct seq_operations nfs_exports_op;
 /*
  * Function prototypes.
  */
-int		nfsd_svc(unsigned short port, int nrservs);
+int		nfsd_svc(int nrservs);
 int		nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp);
 
 int		nfsd_nrthreads(void);
@@ -124,6 +124,7 @@ int nfs4_state_start(void);
 void nfs4_state_shutdown(void);
 void nfs4_reset_lease(time_t leasetime);
 int nfs4_reset_recoverydir(char *recdir);
+char * nfs4_recoverydir(void);
 #else
 static inline void nfs4_state_init(void) { }
 static inline int nfsd4_init_slabs(void) { return 0; }
@@ -132,6 +133,7 @@ static inline int nfs4_state_start(void) { return 0; }
 static inline void nfs4_state_shutdown(void) { }
 static inline void nfs4_reset_lease(time_t leasetime) { }
 static inline int nfs4_reset_recoverydir(char *recdir) { return 0; }
+static inline char * nfs4_recoverydir(void) {return NULL; }
 #endif
 
 /*

fs/nfsd/nfssvc.c

@@ -183,18 +183,18 @@ int nfsd_nrthreads(void)
 	return rv;
 }
 
-static int nfsd_init_socks(int port)
+static int nfsd_init_socks(void)
 {
 	int error;
 	if (!list_empty(&nfsd_serv->sv_permsocks))
 		return 0;
 
-	error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, port,
+	error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, NFS_PORT,
 					SVC_SOCK_DEFAULTS);
 	if (error < 0)
 		return error;
 
-	error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, port,
+	error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, NFS_PORT,
 					SVC_SOCK_DEFAULTS);
 	if (error < 0)
 		return error;
@@ -204,7 +204,7 @@ static int nfsd_init_socks(int port)
 
 static bool nfsd_up = false;
 
-static int nfsd_startup(unsigned short port, int nrservs)
+static int nfsd_startup(int nrservs)
 {
 	int ret;
 
@@ -218,7 +218,7 @@ static int nfsd_startup(unsigned short port, int nrservs)
 	ret = nfsd_racache_init(2*nrservs);
 	if (ret)
 		return ret;
-	ret = nfsd_init_socks(port);
+	ret = nfsd_init_socks();
 	if (ret)
 		goto out_racache;
 	ret = lockd_up(&init_net);
@@ -436,7 +436,7 @@ int nfsd_set_nrthreads(int n, int *nthreads)
  * this is the first time nrservs is nonzero.
  */
 int
-nfsd_svc(unsigned short port, int nrservs)
+nfsd_svc(int nrservs)
 {
 	int	error;
 	bool	nfsd_up_before;
@@ -458,7 +458,7 @@ nfsd_svc(unsigned short port, int nrservs)
 
 	nfsd_up_before = nfsd_up;
 
-	error = nfsd_startup(port, nrservs);
+	error = nfsd_startup(nrservs);
 	if (error)
 		goto out_destroy;
 	error = svc_set_num_threads(nfsd_serv, NULL, nrservs);
@@ -487,7 +487,7 @@ static int
 nfsd(void *vrqstp)
 {
 	struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
-	int err, preverr = 0;
+	int err;
 
 	/* Lock module and set up kernel thread */
 	mutex_lock(&nfsd_mutex);
@@ -534,16 +534,6 @@ nfsd(void *vrqstp)
 			;
 		if (err == -EINTR)
 			break;
-		else if (err < 0) {
-			if (err != preverr) {
-				printk(KERN_WARNING "%s: unexpected error "
-					"from svc_recv (%d)\n", __func__, -err);
-				preverr = err;
-			}
-			schedule_timeout_uninterruptible(HZ);
-			continue;
-		}
-
 		validate_process_creds();
 		svc_process(rqstp);
 		validate_process_creds();

fs/nfsd/state.h

@@ -373,11 +373,7 @@ static inline struct nfs4_lockowner * lockowner(struct nfs4_stateowner *so)
 	return container_of(so, struct nfs4_lockowner, lo_owner);
 }
 
-/*
-*  nfs4_file: a file opened by some number of (open) nfs4_stateowners.
-*    o fi_perfile list is used to search for conflicting 
-*      share_acces, share_deny on the file.
-*/
+/* nfs4_file: a file opened by some number of (open) nfs4_stateowners. */
 struct nfs4_file {
 	atomic_t		fi_ref;
 	struct list_head        fi_hash;    /* hash by "struct inode *" */
@@ -459,7 +455,7 @@ extern void nfs4_unlock_state(void);
 extern int nfs4_in_grace(void);
 extern void nfs4_release_reclaim(void);
 extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(struct nfs4_client *crp);
-extern __be32 nfs4_check_open_reclaim(clientid_t *clid);
+extern __be32 nfs4_check_open_reclaim(clientid_t *clid, bool sessions);
 extern void nfs4_free_openowner(struct nfs4_openowner *);
 extern void nfs4_free_lockowner(struct nfs4_lockowner *);
 extern int set_callback_cred(void);

fs/nfsd/vfs.c

@@ -1581,7 +1581,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp)
 	 */
 
 	oldfs = get_fs(); set_fs(KERNEL_DS);
-	host_err = inode->i_op->readlink(path.dentry, buf, *lenp);
+	host_err = inode->i_op->readlink(path.dentry, (char __user *)buf, *lenp);
 	set_fs(oldfs);
 
 	if (host_err < 0)

include/linux/nfsd/Kbuild

@@ -1,5 +0,0 @@
-header-y += cld.h
-header-y += debug.h
-header-y += export.h
-header-y += nfsfh.h
-header-y += stats.h

include/linux/nfsd/debug.h

@@ -5,44 +5,15 @@
  *
  * Copyright (C) 1995 Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef LINUX_NFSD_DEBUG_H
 #define LINUX_NFSD_DEBUG_H
 
-#include <linux/sunrpc/debug.h>
+#include <uapi/linux/nfsd/debug.h>
 
-/*
- * Enable debugging for nfsd.
- * Requires RPC_DEBUG.
- */
-#ifdef RPC_DEBUG
-# define NFSD_DEBUG		1
-#endif
-
-/*
- * knfsd debug flags
- */
-#define NFSDDBG_SOCK		0x0001
-#define NFSDDBG_FH		0x0002
-#define NFSDDBG_EXPORT		0x0004
-#define NFSDDBG_SVC		0x0008
-#define NFSDDBG_PROC		0x0010
-#define NFSDDBG_FILEOP		0x0020
-#define NFSDDBG_AUTH		0x0040
-#define NFSDDBG_REPCACHE	0x0080
-#define NFSDDBG_XDR		0x0100
-#define NFSDDBG_LOCKD		0x0200
-#define NFSDDBG_ALL		0x7FFF
-#define NFSDDBG_NOCHANGE	0xFFFF
-
-
-#ifdef __KERNEL__
 # undef ifdebug
 # ifdef NFSD_DEBUG
 #  define ifdebug(flag)		if (nfsd_debug & NFSDDBG_##flag)
 # else
 #  define ifdebug(flag)		if (0)
 # endif
-#endif /* __KERNEL__ */
-
 #endif /* LINUX_NFSD_DEBUG_H */

include/linux/nfsd/export.h

@@ -6,58 +6,11 @@
  *
  * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef NFSD_EXPORT_H
 #define NFSD_EXPORT_H
 
-# include <linux/types.h>
-#ifdef __KERNEL__
 # include <linux/nfsd/nfsfh.h>
-#endif
-
-/*
- * Important limits for the exports stuff.
- */
-#define NFSCLNT_IDMAX		1024
-#define NFSCLNT_ADDRMAX		16
-#define NFSCLNT_KEYMAX		32
-
-/*
- * Export flags.
- */
-#define NFSEXP_READONLY		0x0001
-#define NFSEXP_INSECURE_PORT	0x0002
-#define NFSEXP_ROOTSQUASH	0x0004
-#define NFSEXP_ALLSQUASH	0x0008
-#define NFSEXP_ASYNC		0x0010
-#define NFSEXP_GATHERED_WRITES	0x0020
-/* 40 80 100 currently unused */
-#define NFSEXP_NOHIDE		0x0200
-#define NFSEXP_NOSUBTREECHECK	0x0400
-#define	NFSEXP_NOAUTHNLM	0x0800		/* Don't authenticate NLM requests - just trust */
-#define NFSEXP_MSNFS		0x1000	/* do silly things that MS clients expect; no longer supported */
-#define NFSEXP_FSID		0x2000
-#define	NFSEXP_CROSSMOUNT	0x4000
-#define	NFSEXP_NOACL		0x8000	/* reserved for possible ACL related use */
-/*
- * The NFSEXP_V4ROOT flag causes the kernel to give access only to NFSv4
- * clients, and only to the single directory that is the root of the
- * export; further lookup and readdir operations are treated as if every
- * subdirectory was a mountpoint, and ignored if they are not themselves
- * exported.  This is used by nfsd and mountd to construct the NFSv4
- * pseudofilesystem, which provides access only to paths leading to each
- * exported filesystem.
- */
-#define	NFSEXP_V4ROOT		0x10000
-/* All flags that we claim to support.  (Note we don't support NOACL.) */
-#define NFSEXP_ALLFLAGS		0x17E3F
-
-/* The flags that may vary depending on security flavor: */
-#define NFSEXP_SECINFO_FLAGS	(NFSEXP_READONLY | NFSEXP_ROOTSQUASH \
-					| NFSEXP_ALLSQUASH \
-					| NFSEXP_INSECURE_PORT)
-
-#ifdef __KERNEL__
+#include <uapi/linux/nfsd/export.h>
 
 /*
  * FS Locations
@@ -154,7 +107,4 @@ static inline void exp_get(struct svc_export *exp)
 }
 struct svc_export * rqst_exp_find(struct svc_rqst *, int, u32 *);
 
-#endif /* __KERNEL__ */
-
 #endif /* NFSD_EXPORT_H */
-

include/linux/nfsd/nfsfh.h

@@ -10,117 +10,11 @@
  *
  * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef _LINUX_NFSD_FH_H
 #define _LINUX_NFSD_FH_H
 
-#include <linux/types.h>
-#include <linux/nfs.h>
-#include <linux/nfs2.h>
-#include <linux/nfs3.h>
-#include <linux/nfs4.h>
-#ifdef __KERNEL__
 # include <linux/sunrpc/svc.h>
-#endif
-
-/*
- * This is the old "dentry style" Linux NFSv2 file handle.
- *
- * The xino and xdev fields are currently used to transport the
- * ino/dev of the exported inode.
- */
-struct nfs_fhbase_old {
-	__u32		fb_dcookie;	/* dentry cookie - always 0xfeebbaca */
-	__u32		fb_ino;		/* our inode number */
-	__u32		fb_dirino;	/* dir inode number, 0 for directories */
-	__u32		fb_dev;		/* our device */
-	__u32		fb_xdev;
-	__u32		fb_xino;
-	__u32		fb_generation;
-};
-
-/*
- * This is the new flexible, extensible style NFSv2/v3 file handle.
- * by Neil Brown <neilb@cse.unsw.edu.au> - March 2000
- *
- * The file handle starts with a sequence of four-byte words.
- * The first word contains a version number (1) and three descriptor bytes
- * that tell how the remaining 3 variable length fields should be handled.
- * These three bytes are auth_type, fsid_type and fileid_type.
- *
- * All four-byte values are in host-byte-order.
- *
- * The auth_type field specifies how the filehandle can be authenticated
- * This might allow a file to be confirmed to be in a writable part of a
- * filetree without checking the path from it up to the root.
- * Current values:
- *     0  - No authentication.  fb_auth is 0 bytes long
- * Possible future values:
- *     1  - 4 bytes taken from MD5 hash of the remainer of the file handle
- *          prefixed by a secret and with the important export flags.
- *
- * The fsid_type identifies how the filesystem (or export point) is
- *    encoded.
- *  Current values:
- *     0  - 4 byte device id (ms-2-bytes major, ls-2-bytes minor), 4byte inode number
- *        NOTE: we cannot use the kdev_t device id value, because kdev_t.h
- *              says we mustn't.  We must break it up and reassemble.
- *     1  - 4 byte user specified identifier
- *     2  - 4 byte major, 4 byte minor, 4 byte inode number - DEPRECATED
- *     3  - 4 byte device id, encoded for user-space, 4 byte inode number
- *     4  - 4 byte inode number and 4 byte uuid
- *     5  - 8 byte uuid
- *     6  - 16 byte uuid
- *     7  - 8 byte inode number and 16 byte uuid
- *
- * The fileid_type identified how the file within the filesystem is encoded.
- * This is (will be) passed to, and set by, the underlying filesystem if it supports
- * filehandle operations.  The filesystem must not use the value '0' or '0xff' and may
- * only use the values 1 and 2 as defined below:
- *  Current values:
- *    0   - The root, or export point, of the filesystem.  fb_fileid is 0 bytes.
- *    1   - 32bit inode number, 32 bit generation number.
- *    2   - 32bit inode number, 32 bit generation number, 32 bit parent directory inode number.
- *
- */
-struct nfs_fhbase_new {
-	__u8		fb_version;	/* == 1, even => nfs_fhbase_old */
-	__u8		fb_auth_type;
-	__u8		fb_fsid_type;
-	__u8		fb_fileid_type;
-	__u32		fb_auth[1];
-/*	__u32		fb_fsid[0]; floating */
-/*	__u32		fb_fileid[0]; floating */
-};
-
-struct knfsd_fh {
-	unsigned int	fh_size;	/* significant for NFSv3.
-					 * Points to the current size while building
-					 * a new file handle
-					 */
-	union {
-		struct nfs_fhbase_old	fh_old;
-		__u32			fh_pad[NFS4_FHSIZE/4];
-		struct nfs_fhbase_new	fh_new;
-	} fh_base;
-};
-
-#define ofh_dcookie		fh_base.fh_old.fb_dcookie
-#define ofh_ino			fh_base.fh_old.fb_ino
-#define ofh_dirino		fh_base.fh_old.fb_dirino
-#define ofh_dev			fh_base.fh_old.fb_dev
-#define ofh_xdev		fh_base.fh_old.fb_xdev
-#define ofh_xino		fh_base.fh_old.fb_xino
-#define ofh_generation		fh_base.fh_old.fb_generation
-
-#define	fh_version		fh_base.fh_new.fb_version
-#define	fh_fsid_type		fh_base.fh_new.fb_fsid_type
-#define	fh_auth_type		fh_base.fh_new.fb_auth_type
-#define	fh_fileid_type		fh_base.fh_new.fb_fileid_type
-#define	fh_auth			fh_base.fh_new.fb_auth
-#define	fh_fsid			fh_base.fh_new.fb_auth
-
-#ifdef __KERNEL__
+#include <uapi/linux/nfsd/nfsfh.h>
 
 static inline __u32 ino_t_to_u32(ino_t ino)
 {
@@ -166,7 +60,4 @@ typedef struct svc_fh {
 
 } svc_fh;
 
-#endif /* __KERNEL__ */
-
-
 #endif /* _LINUX_NFSD_FH_H */

include/linux/nfsd/stats.h

@@ -5,16 +5,11 @@
  *
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef LINUX_NFSD_STATS_H
 #define LINUX_NFSD_STATS_H
 
-#include <linux/nfs4.h>
-
-/* thread usage wraps very million seconds (approx one fortnight) */
-#define	NFSD_USAGE_WRAP	(HZ*1000000)
+#include <uapi/linux/nfsd/stats.h>
 
-#ifdef __KERNEL__
 
 struct nfsd_stats {
 	unsigned int	rchits;		/* repcache hits */
@@ -47,5 +42,4 @@ extern struct svc_stat		nfsd_svcstats;
 void	nfsd_stat_init(void);
 void	nfsd_stat_shutdown(void);
 
-#endif /* __KERNEL__ */
 #endif /* LINUX_NFSD_STATS_H */

include/linux/sunrpc/Kbuild

@@ -1 +0,0 @@
-header-y += debug.h

include/linux/sunrpc/debug.h

@@ -5,28 +5,11 @@
  *
  * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
  */
-
 #ifndef _LINUX_SUNRPC_DEBUG_H_
 #define _LINUX_SUNRPC_DEBUG_H_
 
-/*
- * RPC debug facilities
- */
-#define RPCDBG_XPRT		0x0001
-#define RPCDBG_CALL		0x0002
-#define RPCDBG_DEBUG		0x0004
-#define RPCDBG_NFS		0x0008
-#define RPCDBG_AUTH		0x0010
-#define RPCDBG_BIND		0x0020
-#define RPCDBG_SCHED		0x0040
-#define RPCDBG_TRANS		0x0080
-#define RPCDBG_SVCXPRT		0x0100
-#define RPCDBG_SVCDSP		0x0200
-#define RPCDBG_MISC		0x0400
-#define RPCDBG_CACHE		0x0800
-#define RPCDBG_ALL		0x7fff
+#include <uapi/linux/sunrpc/debug.h>
 
-#ifdef __KERNEL__
 
 /*
  * Enable RPC debugging/profiling.
@@ -87,24 +70,4 @@ void		rpc_register_sysctl(void);
 void		rpc_unregister_sysctl(void);
 #endif
 
-#endif /* __KERNEL__ */
-
-/*
- * Declarations for the sysctl debug interface, which allows to read or
- * change the debug flags for rpc, nfs, nfsd, and lockd. Since the sunrpc
- * module currently registers its sysctl table dynamically, the sysctl path
- * for module FOO is <CTL_SUNRPC, CTL_FOODEBUG>.
- */
-
-enum {
-	CTL_RPCDEBUG = 1,
-	CTL_NFSDEBUG,
-	CTL_NFSDDEBUG,
-	CTL_NLMDEBUG,
-	CTL_SLOTTABLE_UDP,
-	CTL_SLOTTABLE_TCP,
-	CTL_MIN_RESVPORT,
-	CTL_MAX_RESVPORT,
-};
-
 #endif /* _LINUX_SUNRPC_DEBUG_H_ */

include/linux/sunrpc/svc_xprt.h

@@ -114,7 +114,6 @@ void	svc_xprt_init(struct net *, struct svc_xprt_class *, struct svc_xprt *,
 int	svc_create_xprt(struct svc_serv *, const char *, struct net *,
 			const int, const unsigned short, int);
 void	svc_xprt_enqueue(struct svc_xprt *xprt);
-void	svc_xprt_received(struct svc_xprt *);
 void	svc_xprt_put(struct svc_xprt *xprt);
 void	svc_xprt_copy_addrs(struct svc_rqst *rqstp, struct svc_xprt *xprt);
 void	svc_close_xprt(struct svc_xprt *xprt);
@@ -124,6 +123,7 @@ struct	svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name,
 			struct net *net, const sa_family_t af,
 			const unsigned short port);
 int	svc_xprt_names(struct svc_serv *serv, char *buf, const int buflen);
+void	svc_add_new_perm_xprt(struct svc_serv *serv, struct svc_xprt *xprt);
 
 static inline void svc_xprt_get(struct svc_xprt *xprt)
 {
@@ -166,8 +166,7 @@ static inline size_t svc_addr_len(const struct sockaddr *sa)
 	case AF_INET6:
 		return sizeof(struct sockaddr_in6);
 	}
-
-	return 0;
+	BUG();
 }
 
 static inline unsigned short svc_xprt_local_port(const struct svc_xprt *xprt)

include/linux/sunrpc/svcsock.h

@@ -39,9 +39,6 @@ int		svc_recv(struct svc_rqst *, long);
 int		svc_send(struct svc_rqst *);
 void		svc_drop(struct svc_rqst *);
 void		svc_sock_update_bufs(struct svc_serv *serv);
-int		svc_sock_names(struct svc_serv *serv, char *buf,
-					const size_t buflen,
-					const char *toclose);
 int		svc_addsock(struct svc_serv *serv, const int fd,
 					char *name_return, const size_t len);
 void		svc_init_xprt_sock(void);

include/uapi/linux/nfsd/Kbuild

@@ -1 +1,6 @@
 # UAPI Header export list
+header-y += cld.h
+header-y += debug.h
+header-y += export.h
+header-y += nfsfh.h
+header-y += stats.h

include/uapi/linux/nfsd/debug.h

@@ -0,0 +1,40 @@
+/*
+ * linux/include/linux/nfsd/debug.h
+ *
+ * Debugging-related stuff for nfsd
+ *
+ * Copyright (C) 1995 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPILINUX_NFSD_DEBUG_H
+#define _UAPILINUX_NFSD_DEBUG_H
+
+#include <linux/sunrpc/debug.h>
+
+/*
+ * Enable debugging for nfsd.
+ * Requires RPC_DEBUG.
+ */
+#ifdef RPC_DEBUG
+# define NFSD_DEBUG		1
+#endif
+
+/*
+ * knfsd debug flags
+ */
+#define NFSDDBG_SOCK		0x0001
+#define NFSDDBG_FH		0x0002
+#define NFSDDBG_EXPORT		0x0004
+#define NFSDDBG_SVC		0x0008
+#define NFSDDBG_PROC		0x0010
+#define NFSDDBG_FILEOP		0x0020
+#define NFSDDBG_AUTH		0x0040
+#define NFSDDBG_REPCACHE	0x0080
+#define NFSDDBG_XDR		0x0100
+#define NFSDDBG_LOCKD		0x0200
+#define NFSDDBG_ALL		0x7FFF
+#define NFSDDBG_NOCHANGE	0xFFFF
+
+
+
+#endif /* _UAPILINUX_NFSD_DEBUG_H */

include/uapi/linux/nfsd/export.h

@@ -0,0 +1,58 @@
+/*
+ * include/linux/nfsd/export.h
+ * 
+ * Public declarations for NFS exports. The definitions for the
+ * syscall interface are in nfsctl.h
+ *
+ * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPINFSD_EXPORT_H
+#define _UAPINFSD_EXPORT_H
+
+# include <linux/types.h>
+
+/*
+ * Important limits for the exports stuff.
+ */
+#define NFSCLNT_IDMAX		1024
+#define NFSCLNT_ADDRMAX		16
+#define NFSCLNT_KEYMAX		32
+
+/*
+ * Export flags.
+ */
+#define NFSEXP_READONLY		0x0001
+#define NFSEXP_INSECURE_PORT	0x0002
+#define NFSEXP_ROOTSQUASH	0x0004
+#define NFSEXP_ALLSQUASH	0x0008
+#define NFSEXP_ASYNC		0x0010
+#define NFSEXP_GATHERED_WRITES	0x0020
+/* 40 80 100 currently unused */
+#define NFSEXP_NOHIDE		0x0200
+#define NFSEXP_NOSUBTREECHECK	0x0400
+#define	NFSEXP_NOAUTHNLM	0x0800		/* Don't authenticate NLM requests - just trust */
+#define NFSEXP_MSNFS		0x1000	/* do silly things that MS clients expect; no longer supported */
+#define NFSEXP_FSID		0x2000
+#define	NFSEXP_CROSSMOUNT	0x4000
+#define	NFSEXP_NOACL		0x8000	/* reserved for possible ACL related use */
+/*
+ * The NFSEXP_V4ROOT flag causes the kernel to give access only to NFSv4
+ * clients, and only to the single directory that is the root of the
+ * export; further lookup and readdir operations are treated as if every
+ * subdirectory was a mountpoint, and ignored if they are not themselves
+ * exported.  This is used by nfsd and mountd to construct the NFSv4
+ * pseudofilesystem, which provides access only to paths leading to each
+ * exported filesystem.
+ */
+#define	NFSEXP_V4ROOT		0x10000
+/* All flags that we claim to support.  (Note we don't support NOACL.) */
+#define NFSEXP_ALLFLAGS		0x17E3F
+
+/* The flags that may vary depending on security flavor: */
+#define NFSEXP_SECINFO_FLAGS	(NFSEXP_READONLY | NFSEXP_ROOTSQUASH \
+					| NFSEXP_ALLSQUASH \
+					| NFSEXP_INSECURE_PORT)
+
+
+#endif /* _UAPINFSD_EXPORT_H */

include/uapi/linux/nfsd/nfsfh.h

@@ -0,0 +1,122 @@
+/*
+ * include/linux/nfsd/nfsfh.h
+ *
+ * This file describes the layout of the file handles as passed
+ * over the wire.
+ *
+ * Earlier versions of knfsd used to sign file handles using keyed MD5
+ * or SHA. I've removed this code, because it doesn't give you more
+ * security than blocking external access to port 2049 on your firewall.
+ *
+ * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPI_LINUX_NFSD_FH_H
+#define _UAPI_LINUX_NFSD_FH_H
+
+#include <linux/types.h>
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs3.h>
+#include <linux/nfs4.h>
+
+/*
+ * This is the old "dentry style" Linux NFSv2 file handle.
+ *
+ * The xino and xdev fields are currently used to transport the
+ * ino/dev of the exported inode.
+ */
+struct nfs_fhbase_old {
+	__u32		fb_dcookie;	/* dentry cookie - always 0xfeebbaca */
+	__u32		fb_ino;		/* our inode number */
+	__u32		fb_dirino;	/* dir inode number, 0 for directories */
+	__u32		fb_dev;		/* our device */
+	__u32		fb_xdev;
+	__u32		fb_xino;
+	__u32		fb_generation;
+};
+
+/*
+ * This is the new flexible, extensible style NFSv2/v3 file handle.
+ * by Neil Brown <neilb@cse.unsw.edu.au> - March 2000
+ *
+ * The file handle starts with a sequence of four-byte words.
+ * The first word contains a version number (1) and three descriptor bytes
+ * that tell how the remaining 3 variable length fields should be handled.
+ * These three bytes are auth_type, fsid_type and fileid_type.
+ *
+ * All four-byte values are in host-byte-order.
+ *
+ * The auth_type field specifies how the filehandle can be authenticated
+ * This might allow a file to be confirmed to be in a writable part of a
+ * filetree without checking the path from it up to the root.
+ * Current values:
+ *     0  - No authentication.  fb_auth is 0 bytes long
+ * Possible future values:
+ *     1  - 4 bytes taken from MD5 hash of the remainer of the file handle
+ *          prefixed by a secret and with the important export flags.
+ *
+ * The fsid_type identifies how the filesystem (or export point) is
+ *    encoded.
+ *  Current values:
+ *     0  - 4 byte device id (ms-2-bytes major, ls-2-bytes minor), 4byte inode number
+ *        NOTE: we cannot use the kdev_t device id value, because kdev_t.h
+ *              says we mustn't.  We must break it up and reassemble.
+ *     1  - 4 byte user specified identifier
+ *     2  - 4 byte major, 4 byte minor, 4 byte inode number - DEPRECATED
+ *     3  - 4 byte device id, encoded for user-space, 4 byte inode number
+ *     4  - 4 byte inode number and 4 byte uuid
+ *     5  - 8 byte uuid
+ *     6  - 16 byte uuid
+ *     7  - 8 byte inode number and 16 byte uuid
+ *
+ * The fileid_type identified how the file within the filesystem is encoded.
+ * This is (will be) passed to, and set by, the underlying filesystem if it supports
+ * filehandle operations.  The filesystem must not use the value '0' or '0xff' and may
+ * only use the values 1 and 2 as defined below:
+ *  Current values:
+ *    0   - The root, or export point, of the filesystem.  fb_fileid is 0 bytes.
+ *    1   - 32bit inode number, 32 bit generation number.
+ *    2   - 32bit inode number, 32 bit generation number, 32 bit parent directory inode number.
+ *
+ */
+struct nfs_fhbase_new {
+	__u8		fb_version;	/* == 1, even => nfs_fhbase_old */
+	__u8		fb_auth_type;
+	__u8		fb_fsid_type;
+	__u8		fb_fileid_type;
+	__u32		fb_auth[1];
+/*	__u32		fb_fsid[0]; floating */
+/*	__u32		fb_fileid[0]; floating */
+};
+
+struct knfsd_fh {
+	unsigned int	fh_size;	/* significant for NFSv3.
+					 * Points to the current size while building
+					 * a new file handle
+					 */
+	union {
+		struct nfs_fhbase_old	fh_old;
+		__u32			fh_pad[NFS4_FHSIZE/4];
+		struct nfs_fhbase_new	fh_new;
+	} fh_base;
+};
+
+#define ofh_dcookie		fh_base.fh_old.fb_dcookie
+#define ofh_ino			fh_base.fh_old.fb_ino
+#define ofh_dirino		fh_base.fh_old.fb_dirino
+#define ofh_dev			fh_base.fh_old.fb_dev
+#define ofh_xdev		fh_base.fh_old.fb_xdev
+#define ofh_xino		fh_base.fh_old.fb_xino
+#define ofh_generation		fh_base.fh_old.fb_generation
+
+#define	fh_version		fh_base.fh_new.fb_version
+#define	fh_fsid_type		fh_base.fh_new.fb_fsid_type
+#define	fh_auth_type		fh_base.fh_new.fb_auth_type
+#define	fh_fileid_type		fh_base.fh_new.fb_fileid_type
+#define	fh_auth			fh_base.fh_new.fb_auth
+#define	fh_fsid			fh_base.fh_new.fb_auth
+
+
+
+#endif /* _UAPI_LINUX_NFSD_FH_H */

include/uapi/linux/nfsd/stats.h

@@ -0,0 +1,17 @@
+/*
+ * linux/include/linux/nfsd/stats.h
+ *
+ * Statistics for NFS server.
+ *
+ * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPILINUX_NFSD_STATS_H
+#define _UAPILINUX_NFSD_STATS_H
+
+#include <linux/nfs4.h>
+
+/* thread usage wraps very million seconds (approx one fortnight) */
+#define	NFSD_USAGE_WRAP	(HZ*1000000)
+
+#endif /* _UAPILINUX_NFSD_STATS_H */

include/uapi/linux/sunrpc/Kbuild

@@ -1 +1,2 @@
 # UAPI Header export list
+header-y += debug.h

include/uapi/linux/sunrpc/debug.h

@@ -0,0 +1,48 @@
+/*
+ * linux/include/linux/sunrpc/debug.h
+ *
+ * Debugging support for sunrpc module
+ *
+ * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _UAPI_LINUX_SUNRPC_DEBUG_H_
+#define _UAPI_LINUX_SUNRPC_DEBUG_H_
+
+/*
+ * RPC debug facilities
+ */
+#define RPCDBG_XPRT		0x0001
+#define RPCDBG_CALL		0x0002
+#define RPCDBG_DEBUG		0x0004
+#define RPCDBG_NFS		0x0008
+#define RPCDBG_AUTH		0x0010
+#define RPCDBG_BIND		0x0020
+#define RPCDBG_SCHED		0x0040
+#define RPCDBG_TRANS		0x0080
+#define RPCDBG_SVCXPRT		0x0100
+#define RPCDBG_SVCDSP		0x0200
+#define RPCDBG_MISC		0x0400
+#define RPCDBG_CACHE		0x0800
+#define RPCDBG_ALL		0x7fff
+
+
+/*
+ * Declarations for the sysctl debug interface, which allows to read or
+ * change the debug flags for rpc, nfs, nfsd, and lockd. Since the sunrpc
+ * module currently registers its sysctl table dynamically, the sysctl path
+ * for module FOO is <CTL_SUNRPC, CTL_FOODEBUG>.
+ */
+
+enum {
+	CTL_RPCDEBUG = 1,
+	CTL_NFSDEBUG,
+	CTL_NFSDDEBUG,
+	CTL_NLMDEBUG,
+	CTL_SLOTTABLE_UDP,
+	CTL_SLOTTABLE_TCP,
+	CTL_MIN_RESVPORT,
+	CTL_MAX_RESVPORT,
+};
+
+#endif /* _UAPI_LINUX_SUNRPC_DEBUG_H_ */

net/sunrpc/svc_xprt.c

@@ -208,6 +208,35 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl,
 	return xcl->xcl_ops->xpo_create(serv, net, sap, len, flags);
 }
 
+/*
+ * svc_xprt_received conditionally queues the transport for processing
+ * by another thread. The caller must hold the XPT_BUSY bit and must
+ * not thereafter touch transport data.
+ *
+ * Note: XPT_DATA only gets cleared when a read-attempt finds no (or
+ * insufficient) data.
+ */
+static void svc_xprt_received(struct svc_xprt *xprt)
+{
+	BUG_ON(!test_bit(XPT_BUSY, &xprt->xpt_flags));
+	/* As soon as we clear busy, the xprt could be closed and
+	 * 'put', so we need a reference to call svc_xprt_enqueue with:
+	 */
+	svc_xprt_get(xprt);
+	clear_bit(XPT_BUSY, &xprt->xpt_flags);
+	svc_xprt_enqueue(xprt);
+	svc_xprt_put(xprt);
+}
+
+void svc_add_new_perm_xprt(struct svc_serv *serv, struct svc_xprt *new)
+{
+	clear_bit(XPT_TEMP, &new->xpt_flags);
+	spin_lock_bh(&serv->sv_lock);
+	list_add(&new->xpt_list, &serv->sv_permsocks);
+	spin_unlock_bh(&serv->sv_lock);
+	svc_xprt_received(new);
+}
+
 int svc_create_xprt(struct svc_serv *serv, const char *xprt_name,
 		    struct net *net, const int family,
 		    const unsigned short port, int flags)
@@ -232,13 +261,8 @@ int svc_create_xprt(struct svc_serv *serv, const char *xprt_name,
 			module_put(xcl->xcl_owner);
 			return PTR_ERR(newxprt);
 		}
-
-		clear_bit(XPT_TEMP, &newxprt->xpt_flags);
-		spin_lock_bh(&serv->sv_lock);
-		list_add(&newxprt->xpt_list, &serv->sv_permsocks);
-		spin_unlock_bh(&serv->sv_lock);
+		svc_add_new_perm_xprt(serv, newxprt);
 		newport = svc_xprt_local_port(newxprt);
-		clear_bit(XPT_BUSY, &newxprt->xpt_flags);
 		return newport;
 	}
  err:
@@ -394,27 +418,6 @@ static struct svc_xprt *svc_xprt_dequeue(struct svc_pool *pool)
 	return xprt;
 }
 
-/*
- * svc_xprt_received conditionally queues the transport for processing
- * by another thread. The caller must hold the XPT_BUSY bit and must
- * not thereafter touch transport data.
- *
- * Note: XPT_DATA only gets cleared when a read-attempt finds no (or
- * insufficient) data.
- */
-void svc_xprt_received(struct svc_xprt *xprt)
-{
-	BUG_ON(!test_bit(XPT_BUSY, &xprt->xpt_flags));
-	/* As soon as we clear busy, the xprt could be closed and
-	 * 'put', so we need a reference to call svc_xprt_enqueue with:
-	 */
-	svc_xprt_get(xprt);
-	clear_bit(XPT_BUSY, &xprt->xpt_flags);
-	svc_xprt_enqueue(xprt);
-	svc_xprt_put(xprt);
-}
-EXPORT_SYMBOL_GPL(svc_xprt_received);
-
 /**
  * svc_reserve - change the space reserved for the reply to a request.
  * @rqstp:  The request in question
@@ -565,33 +568,12 @@ static void svc_check_conn_limits(struct svc_serv *serv)
 	}
 }
 
-/*
- * Receive the next request on any transport.  This code is carefully
- * organised not to touch any cachelines in the shared svc_serv
- * structure, only cachelines in the local svc_pool.
- */
-int svc_recv(struct svc_rqst *rqstp, long timeout)
+int svc_alloc_arg(struct svc_rqst *rqstp)
 {
-	struct svc_xprt		*xprt = NULL;
-	struct svc_serv		*serv = rqstp->rq_server;
-	struct svc_pool		*pool = rqstp->rq_pool;
-	int			len, i;
-	int			pages;
-	struct xdr_buf		*arg;
-	DECLARE_WAITQUEUE(wait, current);
-	long			time_left;
-
-	dprintk("svc: server %p waiting for data (to = %ld)\n",
-		rqstp, timeout);
-
-	if (rqstp->rq_xprt)
-		printk(KERN_ERR
-			"svc_recv: service %p, transport not NULL!\n",
-			 rqstp);
-	if (waitqueue_active(&rqstp->rq_wait))
-		printk(KERN_ERR
-			"svc_recv: service %p, wait queue active!\n",
-			 rqstp);
+	struct svc_serv *serv = rqstp->rq_server;
+	struct xdr_buf *arg;
+	int pages;
+	int i;
 
 	/* now allocate needed pages.  If we get a failure, sleep briefly */
 	pages = (serv->sv_max_mesg + PAGE_SIZE) / PAGE_SIZE;
@@ -621,11 +603,15 @@ int svc_recv(struct svc_rqst *rqstp, long timeout)
 	arg->page_len = (pages-2)*PAGE_SIZE;
 	arg->len = (pages-1)*PAGE_SIZE;
 	arg->tail[0].iov_len = 0;
+	return 0;
+}
 
-	try_to_freeze();
-	cond_resched();
-	if (signalled() || kthread_should_stop())
-		return -EINTR;
+struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout)
+{
+	struct svc_xprt *xprt;
+	struct svc_pool		*pool = rqstp->rq_pool;
+	DECLARE_WAITQUEUE(wait, current);
+	long			time_left;
 
 	/* Normally we will wait up to 5 seconds for any required
 	 * cache information to be provided.
@@ -663,7 +649,7 @@ int svc_recv(struct svc_rqst *rqstp, long timeout)
 		if (kthread_should_stop()) {
 			set_current_state(TASK_RUNNING);
 			spin_unlock_bh(&pool->sp_lock);
-			return -EINTR;
+			return ERR_PTR(-EINTR);
 		}
 
 		add_wait_queue(&rqstp->rq_wait, &wait);
@@ -684,48 +670,58 @@ int svc_recv(struct svc_rqst *rqstp, long timeout)
 			spin_unlock_bh(&pool->sp_lock);
 			dprintk("svc: server %p, no data yet\n", rqstp);
 			if (signalled() || kthread_should_stop())
-				return -EINTR;
+				return ERR_PTR(-EINTR);
 			else
-				return -EAGAIN;
+				return ERR_PTR(-EAGAIN);
 		}
 	}
 	spin_unlock_bh(&pool->sp_lock);
+	return xprt;
+}
+
+void svc_add_new_temp_xprt(struct svc_serv *serv, struct svc_xprt *newxpt)
+{
+	spin_lock_bh(&serv->sv_lock);
+	set_bit(XPT_TEMP, &newxpt->xpt_flags);
+	list_add(&newxpt->xpt_list, &serv->sv_tempsocks);
+	serv->sv_tmpcnt++;
+	if (serv->sv_temptimer.function == NULL) {
+		/* setup timer to age temp transports */
+		setup_timer(&serv->sv_temptimer, svc_age_temp_xprts,
+			    (unsigned long)serv);
+		mod_timer(&serv->sv_temptimer,
+			  jiffies + svc_conn_age_period * HZ);
+	}
+	spin_unlock_bh(&serv->sv_lock);
+	svc_xprt_received(newxpt);
+}
+
+static int svc_handle_xprt(struct svc_rqst *rqstp, struct svc_xprt *xprt)
+{
+	struct svc_serv *serv = rqstp->rq_server;
+	int len = 0;
 
-	len = 0;
 	if (test_bit(XPT_CLOSE, &xprt->xpt_flags)) {
 		dprintk("svc_recv: found XPT_CLOSE\n");
 		svc_delete_xprt(xprt);
 		/* Leave XPT_BUSY set on the dead xprt: */
-		goto out;
+		return 0;
 	}
 	if (test_bit(XPT_LISTENER, &xprt->xpt_flags)) {
 		struct svc_xprt *newxpt;
+		/*
+		 * We know this module_get will succeed because the
+		 * listener holds a reference too
+		 */
+		__module_get(xprt->xpt_class->xcl_owner);
+		svc_check_conn_limits(xprt->xpt_server);
 		newxpt = xprt->xpt_ops->xpo_accept(xprt);
-		if (newxpt) {
-			/*
-			 * We know this module_get will succeed because the
-			 * listener holds a reference too
-			 */
-			__module_get(newxpt->xpt_class->xcl_owner);
-			svc_check_conn_limits(xprt->xpt_server);
-			spin_lock_bh(&serv->sv_lock);
-			set_bit(XPT_TEMP, &newxpt->xpt_flags);
-			list_add(&newxpt->xpt_list, &serv->sv_tempsocks);
-			serv->sv_tmpcnt++;
-			if (serv->sv_temptimer.function == NULL) {
-				/* setup timer to age temp transports */
-				setup_timer(&serv->sv_temptimer,
-					    svc_age_temp_xprts,
-					    (unsigned long)serv);
-				mod_timer(&serv->sv_temptimer,
-					  jiffies + svc_conn_age_period * HZ);
-			}
-			spin_unlock_bh(&serv->sv_lock);
-			svc_xprt_received(newxpt);
-		}
+		if (newxpt)
+			svc_add_new_temp_xprt(serv, newxpt);
 	} else if (xprt->xpt_ops->xpo_has_wspace(xprt)) {
+		/* XPT_DATA|XPT_DEFERRED case: */
 		dprintk("svc: server %p, pool %u, transport %p, inuse=%d\n",
-			rqstp, pool->sp_id, xprt,
+			rqstp, rqstp->rq_pool->sp_id, xprt,
 			atomic_read(&xprt->xpt_ref.refcount));
 		rqstp->rq_deferred = svc_deferred_dequeue(xprt);
 		if (rqstp->rq_deferred)
@@ -736,10 +732,51 @@ int svc_recv(struct svc_rqst *rqstp, long timeout)
 		rqstp->rq_reserved = serv->sv_max_mesg;
 		atomic_add(rqstp->rq_reserved, &xprt->xpt_reserved);
 	}
+	/* clear XPT_BUSY: */
 	svc_xprt_received(xprt);
+	return len;
+}
+
+/*
+ * Receive the next request on any transport.  This code is carefully
+ * organised not to touch any cachelines in the shared svc_serv
+ * structure, only cachelines in the local svc_pool.
+ */
+int svc_recv(struct svc_rqst *rqstp, long timeout)
+{
+	struct svc_xprt		*xprt = NULL;
+	struct svc_serv		*serv = rqstp->rq_server;
+	int			len, err;
+
+	dprintk("svc: server %p waiting for data (to = %ld)\n",
+		rqstp, timeout);
+
+	if (rqstp->rq_xprt)
+		printk(KERN_ERR
+			"svc_recv: service %p, transport not NULL!\n",
+			 rqstp);
+	if (waitqueue_active(&rqstp->rq_wait))
+		printk(KERN_ERR
+			"svc_recv: service %p, wait queue active!\n",
+			 rqstp);
+
+	err = svc_alloc_arg(rqstp);
+	if (err)
+		return err;
+
+	try_to_freeze();
+	cond_resched();
+	if (signalled() || kthread_should_stop())
+		return -EINTR;
+
+	xprt = svc_get_next_xprt(rqstp, timeout);
+	if (IS_ERR(xprt))
+		return PTR_ERR(xprt);
+
+	len = svc_handle_xprt(rqstp, xprt);
 
 	/* No data, incomplete (TCP) read, or accept() */
-	if (len == 0 || len == -EAGAIN)
+	if (len <= 0)
 		goto out;
 
 	clear_bit(XPT_OLD, &xprt->xpt_flags);
@@ -917,16 +954,18 @@ void svc_close_xprt(struct svc_xprt *xprt)
 }
 EXPORT_SYMBOL_GPL(svc_close_xprt);
 
-static void svc_close_list(struct list_head *xprt_list, struct net *net)
+static void svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net)
 {
 	struct svc_xprt *xprt;
 
+	spin_lock(&serv->sv_lock);
 	list_for_each_entry(xprt, xprt_list, xpt_list) {
 		if (xprt->xpt_net != net)
 			continue;
 		set_bit(XPT_CLOSE, &xprt->xpt_flags);
 		set_bit(XPT_BUSY, &xprt->xpt_flags);
 	}
+	spin_unlock(&serv->sv_lock);
 }
 
 static void svc_clear_pools(struct svc_serv *serv, struct net *net)
@@ -949,24 +988,28 @@ static void svc_clear_pools(struct svc_serv *serv, struct net *net)
 	}
 }
 
-static void svc_clear_list(struct list_head *xprt_list, struct net *net)
+static void svc_clear_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net)
 {
 	struct svc_xprt *xprt;
 	struct svc_xprt *tmp;
+	LIST_HEAD(victims);
 
+	spin_lock(&serv->sv_lock);
 	list_for_each_entry_safe(xprt, tmp, xprt_list, xpt_list) {
 		if (xprt->xpt_net != net)
 			continue;
-		svc_delete_xprt(xprt);
+		list_move(&xprt->xpt_list, &victims);
 	}
-	list_for_each_entry(xprt, xprt_list, xpt_list)
-		BUG_ON(xprt->xpt_net == net);
+	spin_unlock(&serv->sv_lock);
+
+	list_for_each_entry_safe(xprt, tmp, &victims, xpt_list)
+		svc_delete_xprt(xprt);
 }
 
 void svc_close_net(struct svc_serv *serv, struct net *net)
 {
-	svc_close_list(&serv->sv_tempsocks, net);
-	svc_close_list(&serv->sv_permsocks, net);
+	svc_close_list(serv, &serv->sv_tempsocks, net);
+	svc_close_list(serv, &serv->sv_permsocks, net);
 
 	svc_clear_pools(serv, net);
 	/*
@@ -974,8 +1017,8 @@ void svc_close_net(struct svc_serv *serv, struct net *net)
 	 * svc_xprt_enqueue will not add new entries without taking the
 	 * sp_lock and checking XPT_BUSY.
 	 */
-	svc_clear_list(&serv->sv_tempsocks, net);
-	svc_clear_list(&serv->sv_permsocks, net);
+	svc_clear_list(serv, &serv->sv_tempsocks, net);
+	svc_clear_list(serv, &serv->sv_permsocks, net);
 }
 
 /*

net/sunrpc/svcsock.c

@@ -59,7 +59,7 @@
 
 
 static struct svc_sock *svc_setup_socket(struct svc_serv *, struct socket *,
-					 int *errp, int flags);
+					 int flags);
 static void		svc_udp_data_ready(struct sock *, int);
 static int		svc_udp_recvfrom(struct svc_rqst *);
 static int		svc_udp_sendto(struct svc_rqst *);
@@ -305,57 +305,6 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 	return len;
 }
 
-/**
- * svc_sock_names - construct a list of listener names in a string
- * @serv: pointer to RPC service
- * @buf: pointer to a buffer to fill in with socket names
- * @buflen: size of the buffer to be filled
- * @toclose: pointer to '\0'-terminated C string containing the name
- *		of a listener to be closed
- *
- * Fills in @buf with a '\n'-separated list of names of listener
- * sockets.  If @toclose is not NULL, the socket named by @toclose
- * is closed, and is not included in the output list.
- *
- * Returns positive length of the socket name string, or a negative
- * errno value on error.
- */
-int svc_sock_names(struct svc_serv *serv, char *buf, const size_t buflen,
-		   const char *toclose)
-{
-	struct svc_sock *svsk, *closesk = NULL;
-	int len = 0;
-
-	if (!serv)
-		return 0;
-
-	spin_lock_bh(&serv->sv_lock);
-	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list) {
-		int onelen = svc_one_sock_name(svsk, buf + len, buflen - len);
-		if (onelen < 0) {
-			len = onelen;
-			break;
-		}
-		if (toclose && strcmp(toclose, buf + len) == 0) {
-			closesk = svsk;
-			svc_xprt_get(&closesk->sk_xprt);
-		} else
-			len += onelen;
-	}
-	spin_unlock_bh(&serv->sv_lock);
-
-	if (closesk) {
-		/* Should unregister with portmap, but you cannot
-		 * unregister just one protocol...
-		 */
-		svc_close_xprt(&closesk->sk_xprt);
-		svc_xprt_put(&closesk->sk_xprt);
-	} else if (toclose)
-		return -ENOENT;
-	return len;
-}
-EXPORT_SYMBOL_GPL(svc_sock_names);
-
 /*
  * Check input queue length
  */
@@ -598,11 +547,9 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 			dprintk("svc: recvfrom returned error %d\n", -err);
 			set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		}
-		return -EAGAIN;
+		return 0;
 	}
 	len = svc_addr_len(svc_addr(rqstp));
-	if (len == 0)
-		return -EAFNOSUPPORT;
 	rqstp->rq_addrlen = len;
 	if (skb->tstamp.tv64 == 0) {
 		skb->tstamp = ktime_get_real();
@@ -620,10 +567,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 	if (!svc_udp_get_dest_address(rqstp, cmh)) {
 		net_warn_ratelimited("svc: received unknown control message %d/%d; dropping RPC reply datagram\n",
 				     cmh->cmsg_level, cmh->cmsg_type);
-out_free:
-		trace_kfree_skb(skb, svc_udp_recvfrom);
-		skb_free_datagram_locked(svsk->sk_sk, skb);
-		return 0;
+		goto out_free;
 	}
 	rqstp->rq_daddrlen = svc_addr_len(svc_daddr(rqstp));
 
@@ -662,6 +606,10 @@ out_free:
 		serv->sv_stats->netudpcnt++;
 
 	return len;
+out_free:
+	trace_kfree_skb(skb, svc_udp_recvfrom);
+	skb_free_datagram_locked(svsk->sk_sk, skb);
+	return 0;
 }
 
 static int
@@ -900,8 +848,9 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	 */
 	newsock->sk->sk_sndtimeo = HZ*30;
 
-	if (!(newsvsk = svc_setup_socket(serv, newsock, &err,
-				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY))))
+	newsvsk = svc_setup_socket(serv, newsock,
+				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY));
+	if (IS_ERR(newsvsk))
 		goto failed;
 	svc_xprt_set_remote(&newsvsk->sk_xprt, sin, slen);
 	err = kernel_getsockname(newsock, sin, &slen);
@@ -1174,13 +1123,13 @@ error:
 	if (len != -EAGAIN)
 		goto err_other;
 	dprintk("RPC: TCP recvfrom got EAGAIN\n");
-	return -EAGAIN;
+	return 0;
 err_other:
 	printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
 	       svsk->sk_xprt.xpt_server->sv_name, -len);
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
 err_noclose:
-	return -EAGAIN;	/* record not complete */
+	return 0;	/* record not complete */
 }
 
 /*
@@ -1383,29 +1332,29 @@ EXPORT_SYMBOL_GPL(svc_sock_update_bufs);
  */
 static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 						struct socket *sock,
-						int *errp, int flags)
+						int flags)
 {
 	struct svc_sock	*svsk;
 	struct sock	*inet;
 	int		pmap_register = !(flags & SVC_SOCK_ANONYMOUS);
+	int		err = 0;
 
 	dprintk("svc: svc_setup_socket %p\n", sock);
-	if (!(svsk = kzalloc(sizeof(*svsk), GFP_KERNEL))) {
-		*errp = -ENOMEM;
-		return NULL;
-	}
+	svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
+	if (!svsk)
+		return ERR_PTR(-ENOMEM);
 
 	inet = sock->sk;
 
 	/* Register socket with portmapper */
-	if (*errp >= 0 && pmap_register)
-		*errp = svc_register(serv, sock_net(sock->sk), inet->sk_family,
+	if (pmap_register)
+		err = svc_register(serv, sock_net(sock->sk), inet->sk_family,
 				     inet->sk_protocol,
 				     ntohs(inet_sk(inet)->inet_sport));
 
-	if (*errp < 0) {
+	if (err < 0) {
 		kfree(svsk);
-		return NULL;
+		return ERR_PTR(err);
 	}
 
 	inet->sk_user_data = svsk;
@@ -1450,42 +1399,38 @@ int svc_addsock(struct svc_serv *serv, const int fd, char *name_return,
 	int err = 0;
 	struct socket *so = sockfd_lookup(fd, &err);
 	struct svc_sock *svsk = NULL;
+	struct sockaddr_storage addr;
+	struct sockaddr *sin = (struct sockaddr *)&addr;
+	int salen;
 
 	if (!so)
 		return err;
+	err = -EAFNOSUPPORT;
 	if ((so->sk->sk_family != PF_INET) && (so->sk->sk_family != PF_INET6))
-		err =  -EAFNOSUPPORT;
-	else if (so->sk->sk_protocol != IPPROTO_TCP &&
+		goto out;
+	err =  -EPROTONOSUPPORT;
+	if (so->sk->sk_protocol != IPPROTO_TCP &&
 	    so->sk->sk_protocol != IPPROTO_UDP)
-		err =  -EPROTONOSUPPORT;
-	else if (so->state > SS_UNCONNECTED)
-		err = -EISCONN;
-	else {
-		if (!try_module_get(THIS_MODULE))
-			err = -ENOENT;
-		else
-			svsk = svc_setup_socket(serv, so, &err,
-						SVC_SOCK_DEFAULTS);
-		if (svsk) {
-			struct sockaddr_storage addr;
-			struct sockaddr *sin = (struct sockaddr *)&addr;
-			int salen;
-			if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
-				svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
-			clear_bit(XPT_TEMP, &svsk->sk_xprt.xpt_flags);
-			spin_lock_bh(&serv->sv_lock);
-			list_add(&svsk->sk_xprt.xpt_list, &serv->sv_permsocks);
-			spin_unlock_bh(&serv->sv_lock);
-			svc_xprt_received(&svsk->sk_xprt);
-			err = 0;
-		} else
-			module_put(THIS_MODULE);
-	}
-	if (err) {
-		sockfd_put(so);
-		return err;
+		goto out;
+	err = -EISCONN;
+	if (so->state > SS_UNCONNECTED)
+		goto out;
+	err = -ENOENT;
+	if (!try_module_get(THIS_MODULE))
+		goto out;
+	svsk = svc_setup_socket(serv, so, SVC_SOCK_DEFAULTS);
+	if (IS_ERR(svsk)) {
+		module_put(THIS_MODULE);
+		err = PTR_ERR(svsk);
+		goto out;
 	}
+	if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
+		svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
+	svc_add_new_perm_xprt(serv, &svsk->sk_xprt);
 	return svc_one_sock_name(svsk, name_return, len);
+out:
+	sockfd_put(so);
+	return err;
 }
 EXPORT_SYMBOL_GPL(svc_addsock);
 
@@ -1563,11 +1508,13 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 			goto bummer;
 	}
 
-	if ((svsk = svc_setup_socket(serv, sock, &error, flags)) != NULL) {
-		svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
-		return (struct svc_xprt *)svsk;
+	svsk = svc_setup_socket(serv, sock, flags);
+	if (IS_ERR(svsk)) {
+		error = PTR_ERR(svsk);
+		goto bummer;
 	}
-
+	svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
+	return (struct svc_xprt *)svsk;
 bummer:
 	dprintk("svc: svc_create_socket error = %d\n", -error);
 	sock_release(sock);

net/sunrpc/xprtrdma/svc_rdma_transport.c

@@ -578,10 +578,6 @@ static void handle_connect_req(struct rdma_cm_id *new_cma_id, size_t client_ird)
 	list_add_tail(&newxprt->sc_accept_q, &listen_xprt->sc_accept_q);
 	spin_unlock_bh(&listen_xprt->sc_lock);
 
-	/*
-	 * Can't use svc_xprt_received here because we are not on a
-	 * rqstp thread
-	*/
 	set_bit(XPT_CONN, &listen_xprt->sc_xprt.xpt_flags);
 	svc_xprt_enqueue(&listen_xprt->sc_xprt);
 }