Merge tag 'nfsd-4.4' of git://linux-nfs.org/~bfields/linux

Pull nfsd updates from Bruce Fields:
 "Apologies for coming a little late in the merge window.  Fortunately
  this is another fairly quiet one:

  Mainly smaller bugfixes and cleanup.  We're still finding some bugs
  from the breakup of the big NFSv4 state lock in 3.17 -- thanks
  especially to Andrew Elble and Jeff Layton for tracking down some of
  the remaining races"

* tag 'nfsd-4.4' of git://linux-nfs.org/~bfields/linux:
  svcrpc: document lack of some memory barriers
  nfsd: fix race with open / open upgrade stateids
  nfsd: eliminate sending duplicate and repeated delegations
  nfsd: remove recurring workqueue job to clean DRC
  SUNRPC: drop stale comment in svc_setup_socket()
  nfsd: ensure that seqid morphing operations are atomic wrt to copies
  nfsd: serialize layout stateid morphing operations
  nfsd: improve client_has_state to check for unused openowners
  nfsd: fix clid_inuse on mount with security change
  sunrpc/cache: make cache flushing more reliable.
  nfsd: move include of state.h from trace.c to trace.h
  sunrpc: avoid warning in gss_key_timeout
  lockd: get rid of reference-counted NSM RPC clients
  SUNRPC: Use MSG_SENDPAGE_NOTLAST when calling sendpage()
  lockd: create NSM handles per net namespace
  nfsd: switch unsigned char flags in svc_fh to bools
  nfsd: move svc_fh->fh_maxsize to just after fh_handle
  nfsd: drop null test before destroy functions
  nfsd: serialize state seqid morphing operations

fs/lockd/host.c

@@ -116,7 +116,7 @@ static struct nlm_host *nlm_alloc_host(struct nlm_lookup_host_info *ni,
 		atomic_inc(&nsm->sm_count);
 	else {
 		host = NULL;
-		nsm = nsm_get_handle(ni->sap, ni->salen,
+		nsm = nsm_get_handle(ni->net, ni->sap, ni->salen,
 					ni->hostname, ni->hostname_len);
 		if (unlikely(nsm == NULL)) {
 			dprintk("lockd: %s failed; no nsm handle\n",
@@ -161,6 +161,7 @@ static struct nlm_host *nlm_alloc_host(struct nlm_lookup_host_info *ni,
 	host->h_nsmhandle  = nsm;
 	host->h_addrbuf    = nsm->sm_addrbuf;
 	host->net	   = ni->net;
+	strlcpy(host->nodename, utsname()->nodename, sizeof(host->nodename));
 
 out:
 	return host;
@@ -534,17 +535,18 @@ static struct nlm_host *next_host_state(struct hlist_head *cache,
 
 /**
  * nlm_host_rebooted - Release all resources held by rebooted host
+ * @net:  network namespace
  * @info: pointer to decoded results of NLM_SM_NOTIFY call
  *
  * We were notified that the specified host has rebooted.  Release
  * all resources held by that peer.
  */
-void nlm_host_rebooted(const struct nlm_reboot *info)
+void nlm_host_rebooted(const struct net *net, const struct nlm_reboot *info)
 {
 	struct nsm_handle *nsm;
 	struct nlm_host	*host;
 
-	nsm = nsm_reboot_lookup(info);
+	nsm = nsm_reboot_lookup(net, info);
 	if (unlikely(nsm == NULL))
 		return;
 

fs/lockd/mon.c

@@ -42,7 +42,7 @@ struct nsm_args {
 	u32			proc;
 
 	char			*mon_name;
-	char			*nodename;
+	const char		*nodename;
 };
 
 struct nsm_res {
@@ -51,7 +51,6 @@ struct nsm_res {
 };
 
 static const struct rpc_program	nsm_program;
-static				LIST_HEAD(nsm_handles);
 static				DEFINE_SPINLOCK(nsm_lock);
 
 /*
@@ -87,69 +86,18 @@ static struct rpc_clnt *nsm_create(struct net *net, const char *nodename)
 	return rpc_create(&args);
 }
 
-static struct rpc_clnt *nsm_client_set(struct lockd_net *ln,
-		struct rpc_clnt *clnt)
-{
-	spin_lock(&ln->nsm_clnt_lock);
-	if (ln->nsm_users == 0) {
-		if (clnt == NULL)
-			goto out;
-		ln->nsm_clnt = clnt;
-	}
-	clnt = ln->nsm_clnt;
-	ln->nsm_users++;
-out:
-	spin_unlock(&ln->nsm_clnt_lock);
-	return clnt;
-}
-
-static struct rpc_clnt *nsm_client_get(struct net *net, const char *nodename)
-{
-	struct rpc_clnt	*clnt, *new;
-	struct lockd_net *ln = net_generic(net, lockd_net_id);
-
-	clnt = nsm_client_set(ln, NULL);
-	if (clnt != NULL)
-		goto out;
-
-	clnt = new = nsm_create(net, nodename);
-	if (IS_ERR(clnt))
-		goto out;
-
-	clnt = nsm_client_set(ln, new);
-	if (clnt != new)
-		rpc_shutdown_client(new);
-out:
-	return clnt;
-}
-
-static void nsm_client_put(struct net *net)
-{
-	struct lockd_net *ln = net_generic(net, lockd_net_id);
-	struct rpc_clnt	*clnt = NULL;
-
-	spin_lock(&ln->nsm_clnt_lock);
-	ln->nsm_users--;
-	if (ln->nsm_users == 0) {
-		clnt = ln->nsm_clnt;
-		ln->nsm_clnt = NULL;
-	}
-	spin_unlock(&ln->nsm_clnt_lock);
-	if (clnt != NULL)
-		rpc_shutdown_client(clnt);
-}
-
 static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res,
-			 struct rpc_clnt *clnt)
+			 const struct nlm_host *host)
 {
 	int		status;
+	struct rpc_clnt *clnt;
 	struct nsm_args args = {
 		.priv		= &nsm->sm_priv,
 		.prog		= NLM_PROGRAM,
 		.vers		= 3,
 		.proc		= NLMPROC_NSM_NOTIFY,
 		.mon_name	= nsm->sm_mon_name,
-		.nodename	= clnt->cl_nodename,
+		.nodename	= host->nodename,
 	};
 	struct rpc_message msg = {
 		.rpc_argp	= &args,
@@ -158,6 +106,13 @@ static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res,
 
 	memset(res, 0, sizeof(*res));
 
+	clnt = nsm_create(host->net, host->nodename);
+	if (IS_ERR(clnt)) {
+		dprintk("lockd: failed to create NSM upcall transport, "
+			"status=%ld, net=%p\n", PTR_ERR(clnt), host->net);
+		return PTR_ERR(clnt);
+	}
+
 	msg.rpc_proc = &clnt->cl_procinfo[proc];
 	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFTCONN);
 	if (status == -ECONNREFUSED) {
@@ -171,6 +126,8 @@ static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res,
 				status);
 	else
 		status = 0;
+
+	rpc_shutdown_client(clnt);
 	return status;
 }
 
@@ -190,32 +147,19 @@ int nsm_monitor(const struct nlm_host *host)
 	struct nsm_handle *nsm = host->h_nsmhandle;
 	struct nsm_res	res;
 	int		status;
-	struct rpc_clnt *clnt;
-	const char *nodename = NULL;
 
 	dprintk("lockd: nsm_monitor(%s)\n", nsm->sm_name);
 
 	if (nsm->sm_monitored)
 		return 0;
 
-	if (host->h_rpcclnt)
-		nodename = host->h_rpcclnt->cl_nodename;
-
 	/*
 	 * Choose whether to record the caller_name or IP address of
 	 * this peer in the local rpc.statd's database.
 	 */
 	nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf;
 
-	clnt = nsm_client_get(host->net, nodename);
-	if (IS_ERR(clnt)) {
-		status = PTR_ERR(clnt);
-		dprintk("lockd: failed to create NSM upcall transport, "
-				"status=%d, net=%p\n", status, host->net);
-		return status;
-	}
-
-	status = nsm_mon_unmon(nsm, NSMPROC_MON, &res, clnt);
+	status = nsm_mon_unmon(nsm, NSMPROC_MON, &res, host);
 	if (unlikely(res.status != 0))
 		status = -EIO;
 	if (unlikely(status < 0)) {
@@ -247,11 +191,9 @@ void nsm_unmonitor(const struct nlm_host *host)
 
 	if (atomic_read(&nsm->sm_count) == 1
 	 && nsm->sm_monitored && !nsm->sm_sticky) {
-		struct lockd_net *ln = net_generic(host->net, lockd_net_id);
-
 		dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name);
 
-		status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res, ln->nsm_clnt);
+		status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res, host);
 		if (res.status != 0)
 			status = -EIO;
 		if (status < 0)
@@ -259,38 +201,38 @@ void nsm_unmonitor(const struct nlm_host *host)
 					nsm->sm_name);
 		else
 			nsm->sm_monitored = 0;
-
-		nsm_client_put(host->net);
 	}
 }
 
-static struct nsm_handle *nsm_lookup_hostname(const char *hostname,
-					      const size_t len)
+static struct nsm_handle *nsm_lookup_hostname(const struct list_head *nsm_handles,
+					const char *hostname, const size_t len)
 {
 	struct nsm_handle *nsm;
 
-	list_for_each_entry(nsm, &nsm_handles, sm_link)
+	list_for_each_entry(nsm, nsm_handles, sm_link)
 		if (strlen(nsm->sm_name) == len &&
 		    memcmp(nsm->sm_name, hostname, len) == 0)
 			return nsm;
 	return NULL;
 }
 
-static struct nsm_handle *nsm_lookup_addr(const struct sockaddr *sap)
+static struct nsm_handle *nsm_lookup_addr(const struct list_head *nsm_handles,
+					const struct sockaddr *sap)
 {
 	struct nsm_handle *nsm;
 
-	list_for_each_entry(nsm, &nsm_handles, sm_link)
+	list_for_each_entry(nsm, nsm_handles, sm_link)
 		if (rpc_cmp_addr(nsm_addr(nsm), sap))
 			return nsm;
 	return NULL;
 }
 
-static struct nsm_handle *nsm_lookup_priv(const struct nsm_private *priv)
+static struct nsm_handle *nsm_lookup_priv(const struct list_head *nsm_handles,
+					const struct nsm_private *priv)
 {
 	struct nsm_handle *nsm;
 
-	list_for_each_entry(nsm, &nsm_handles, sm_link)
+	list_for_each_entry(nsm, nsm_handles, sm_link)
 		if (memcmp(nsm->sm_priv.data, priv->data,
 					sizeof(priv->data)) == 0)
 			return nsm;
@@ -353,6 +295,7 @@ static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap,
 
 /**
  * nsm_get_handle - Find or create a cached nsm_handle
+ * @net: network namespace
  * @sap: pointer to socket address of handle to find
  * @salen: length of socket address
  * @hostname: pointer to C string containing hostname to find
@@ -365,11 +308,13 @@ static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap,
  * @hostname cannot be found in the handle cache.  Returns NULL if
  * an error occurs.
  */
-struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
+struct nsm_handle *nsm_get_handle(const struct net *net,
+				  const struct sockaddr *sap,
 				  const size_t salen, const char *hostname,
 				  const size_t hostname_len)
 {
 	struct nsm_handle *cached, *new = NULL;
+	struct lockd_net *ln = net_generic(net, lockd_net_id);
 
 	if (hostname && memchr(hostname, '/', hostname_len) != NULL) {
 		if (printk_ratelimit()) {
@@ -384,9 +329,10 @@ struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
 	spin_lock(&nsm_lock);
 
 	if (nsm_use_hostnames && hostname != NULL)
-		cached = nsm_lookup_hostname(hostname, hostname_len);
+		cached = nsm_lookup_hostname(&ln->nsm_handles,
+					hostname, hostname_len);
 	else
-		cached = nsm_lookup_addr(sap);
+		cached = nsm_lookup_addr(&ln->nsm_handles, sap);
 
 	if (cached != NULL) {
 		atomic_inc(&cached->sm_count);
@@ -400,7 +346,7 @@ struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
 	}
 
 	if (new != NULL) {
-		list_add(&new->sm_link, &nsm_handles);
+		list_add(&new->sm_link, &ln->nsm_handles);
 		spin_unlock(&nsm_lock);
 		dprintk("lockd: created nsm_handle for %s (%s)\n",
 				new->sm_name, new->sm_addrbuf);
@@ -417,19 +363,22 @@ struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
 
 /**
  * nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle
+ * @net:  network namespace
  * @info: pointer to NLMPROC_SM_NOTIFY arguments
  *
  * Returns a matching nsm_handle if found in the nsm cache. The returned
  * nsm_handle's reference count is bumped. Otherwise returns NULL if some
  * error occurred.
  */
-struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info)
+struct nsm_handle *nsm_reboot_lookup(const struct net *net,
+				const struct nlm_reboot *info)
 {
 	struct nsm_handle *cached;
+	struct lockd_net *ln = net_generic(net, lockd_net_id);
 
 	spin_lock(&nsm_lock);
 
-	cached = nsm_lookup_priv(&info->priv);
+	cached = nsm_lookup_priv(&ln->nsm_handles, &info->priv);
 	if (unlikely(cached == NULL)) {
 		spin_unlock(&nsm_lock);
 		dprintk("lockd: never saw rebooted peer '%.*s' before\n",

fs/lockd/netns.h

@@ -12,9 +12,7 @@ struct lockd_net {
 	struct delayed_work grace_period_end;
 	struct lock_manager lockd_manager;
 
-	spinlock_t nsm_clnt_lock;
-	unsigned int nsm_users;
-	struct rpc_clnt *nsm_clnt;
+	struct list_head nsm_handles;
 };
 
 extern int lockd_net_id;

fs/lockd/svc.c

@@ -592,7 +592,7 @@ static int lockd_init_net(struct net *net)
 	INIT_DELAYED_WORK(&ln->grace_period_end, grace_ender);
 	INIT_LIST_HEAD(&ln->lockd_manager.list);
 	ln->lockd_manager.block_opens = false;
-	spin_lock_init(&ln->nsm_clnt_lock);
+	INIT_LIST_HEAD(&ln->nsm_handles);
 	return 0;
 }
 

fs/lockd/svc4proc.c

@@ -421,7 +421,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 		return rpc_system_err;
 	}
 
-	nlm_host_rebooted(argp);
+	nlm_host_rebooted(SVC_NET(rqstp), argp);
 	return rpc_success;
 }
 

fs/lockd/svcproc.c

@@ -464,7 +464,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 		return rpc_system_err;
 	}
 
-	nlm_host_rebooted(argp);
+	nlm_host_rebooted(SVC_NET(rqstp), argp);
 	return rpc_success;
 }
 

fs/nfsd/nfs3xdr.c

@@ -262,11 +262,11 @@ void fill_post_wcc(struct svc_fh *fhp)
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
 	fhp->fh_post_change = d_inode(fhp->fh_dentry)->i_version;
 	if (err) {
-		fhp->fh_post_saved = 0;
+		fhp->fh_post_saved = false;
 		/* Grab the ctime anyway - set_change_info might use it */
 		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
 	} else
-		fhp->fh_post_saved = 1;
+		fhp->fh_post_saved = true;
 }
 
 /*

fs/nfsd/nfs4layouts.c

@@ -201,6 +201,7 @@ nfsd4_alloc_layout_stateid(struct nfsd4_compound_state *cstate,
 	INIT_LIST_HEAD(&ls->ls_perfile);
 	spin_lock_init(&ls->ls_lock);
 	INIT_LIST_HEAD(&ls->ls_layouts);
+	mutex_init(&ls->ls_mutex);
 	ls->ls_layout_type = layout_type;
 	nfsd4_init_cb(&ls->ls_recall, clp, &nfsd4_cb_layout_ops,
 			NFSPROC4_CLNT_CB_LAYOUT);
@@ -262,19 +263,23 @@ nfsd4_preprocess_layout_stateid(struct svc_rqst *rqstp,
 		status = nfserr_jukebox;
 		if (!ls)
 			goto out;
+		mutex_lock(&ls->ls_mutex);
 	} else {
 		ls = container_of(stid, struct nfs4_layout_stateid, ls_stid);
 
 		status = nfserr_bad_stateid;
+		mutex_lock(&ls->ls_mutex);
 		if (stateid->si_generation > stid->sc_stateid.si_generation)
-			goto out_put_stid;
+			goto out_unlock_stid;
 		if (layout_type != ls->ls_layout_type)
-			goto out_put_stid;
+			goto out_unlock_stid;
 	}
 
 	*lsp = ls;
 	return 0;
 
+out_unlock_stid:
+	mutex_unlock(&ls->ls_mutex);
 out_put_stid:
 	nfs4_put_stid(stid);
 out:
@@ -296,8 +301,6 @@ nfsd4_recall_file_layout(struct nfs4_layout_stateid *ls)
 	trace_layout_recall(&ls->ls_stid.sc_stateid);
 
 	atomic_inc(&ls->ls_stid.sc_count);
-	update_stateid(&ls->ls_stid.sc_stateid);
-	memcpy(&ls->ls_recall_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
 	nfsd4_run_cb(&ls->ls_recall);
 
 out_unlock:
@@ -406,8 +409,7 @@ nfsd4_insert_layout(struct nfsd4_layoutget *lgp, struct nfs4_layout_stateid *ls)
 	list_add_tail(&new->lo_perstate, &ls->ls_layouts);
 	new = NULL;
 done:
-	update_stateid(&ls->ls_stid.sc_stateid);
-	memcpy(&lgp->lg_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
+	nfs4_inc_and_copy_stateid(&lgp->lg_sid, &ls->ls_stid);
 	spin_unlock(&ls->ls_lock);
 out:
 	spin_unlock(&fp->fi_lock);
@@ -481,11 +483,8 @@ nfsd4_return_file_layouts(struct svc_rqst *rqstp,
 		}
 	}
 	if (!list_empty(&ls->ls_layouts)) {
-		if (found) {
-			update_stateid(&ls->ls_stid.sc_stateid);
-			memcpy(&lrp->lr_sid, &ls->ls_stid.sc_stateid,
-				sizeof(stateid_t));
-		}
+		if (found)
+			nfs4_inc_and_copy_stateid(&lrp->lr_sid, &ls->ls_stid);
 		lrp->lrs_present = 1;
 	} else {
 		trace_layoutstate_unhash(&ls->ls_stid.sc_stateid);
@@ -494,6 +493,7 @@ nfsd4_return_file_layouts(struct svc_rqst *rqstp,
 	}
 	spin_unlock(&ls->ls_lock);
 
+	mutex_unlock(&ls->ls_mutex);
 	nfs4_put_stid(&ls->ls_stid);
 	nfsd4_free_layouts(&reaplist);
 	return nfs_ok;
@@ -608,6 +608,16 @@ nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
 	}
 }
 
+static void
+nfsd4_cb_layout_prepare(struct nfsd4_callback *cb)
+{
+	struct nfs4_layout_stateid *ls =
+		container_of(cb, struct nfs4_layout_stateid, ls_recall);
+
+	mutex_lock(&ls->ls_mutex);
+	nfs4_inc_and_copy_stateid(&ls->ls_recall_sid, &ls->ls_stid);
+}
+
 static int
 nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 {
@@ -649,12 +659,14 @@ nfsd4_cb_layout_release(struct nfsd4_callback *cb)
 
 	trace_layout_recall_release(&ls->ls_stid.sc_stateid);
 
+	mutex_unlock(&ls->ls_mutex);
 	nfsd4_return_all_layouts(ls, &reaplist);
 	nfsd4_free_layouts(&reaplist);
 	nfs4_put_stid(&ls->ls_stid);
 }
 
 static struct nfsd4_callback_ops nfsd4_cb_layout_ops = {
+	.prepare	= nfsd4_cb_layout_prepare,
 	.done		= nfsd4_cb_layout_done,
 	.release	= nfsd4_cb_layout_release,
 };

fs/nfsd/nfs4proc.c

@@ -1309,6 +1309,7 @@ nfsd4_layoutget(struct svc_rqst *rqstp,
 	nfserr = nfsd4_insert_layout(lgp, ls);
 
 out_put_stid:
+	mutex_unlock(&ls->ls_mutex);
 	nfs4_put_stid(&ls->ls_stid);
 out:
 	return nfserr;
@@ -1362,6 +1363,9 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp,
 		goto out;
 	}
 
+	/* LAYOUTCOMMIT does not require any serialization */
+	mutex_unlock(&ls->ls_mutex);
+
 	if (new_size > i_size_read(inode)) {
 		lcp->lc_size_chg = 1;
 		lcp->lc_newsize = new_size;

fs/nfsd/nfscache.c

@@ -63,7 +63,6 @@ static unsigned int		longest_chain;
 static unsigned int		longest_chain_cachesize;
 
 static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
-static void	cache_cleaner_func(struct work_struct *unused);
 static unsigned long nfsd_reply_cache_count(struct shrinker *shrink,
 					    struct shrink_control *sc);
 static unsigned long nfsd_reply_cache_scan(struct shrinker *shrink,
@@ -75,13 +74,6 @@ static struct shrinker nfsd_reply_cache_shrinker = {
 	.seeks	= 1,
 };
 
-/*
- * locking for the reply cache:
- * A cache entry is "single use" if c_state == RC_INPROG
- * Otherwise, it when accessing _prev or _next, the lock must be held.
- */
-static DECLARE_DELAYED_WORK(cache_cleaner, cache_cleaner_func);
-
 /*
  * Put a cap on the size of the DRC based on the amount of available
  * low memory in the machine.
@@ -203,7 +195,6 @@ void nfsd_reply_cache_shutdown(void)
 	unsigned int i;
 
 	unregister_shrinker(&nfsd_reply_cache_shrinker);
-	cancel_delayed_work_sync(&cache_cleaner);
 
 	for (i = 0; i < drc_hashsize; i++) {
 		struct list_head *head = &drc_hashtbl[i].lru_head;
@@ -217,10 +208,8 @@ void nfsd_reply_cache_shutdown(void)
 	drc_hashtbl = NULL;
 	drc_hashsize = 0;
 
-	if (drc_slab) {
 	kmem_cache_destroy(drc_slab);
 	drc_slab = NULL;
-	}
 }
 
 /*
@@ -232,7 +221,6 @@ lru_put_end(struct nfsd_drc_bucket *b, struct svc_cacherep *rp)
 {
 	rp->c_timestamp = jiffies;
 	list_move_tail(&rp->c_lru, &b->lru_head);
-	schedule_delayed_work(&cache_cleaner, RC_EXPIRE);
 }
 
 static long
@@ -266,7 +254,6 @@ prune_cache_entries(void)
 {
 	unsigned int i;
 	long freed = 0;
-	bool cancel = true;
 
 	for (i = 0; i < drc_hashsize; i++) {
 		struct nfsd_drc_bucket *b = &drc_hashtbl[i];
@@ -275,26 +262,11 @@ prune_cache_entries(void)
 			continue;
 		spin_lock(&b->cache_lock);
 		freed += prune_bucket(b);
-		if (!list_empty(&b->lru_head))
-			cancel = false;
 		spin_unlock(&b->cache_lock);
 	}
-
-	/*
-	 * Conditionally rearm the job to run in RC_EXPIRE since we just
-	 * ran the pruner.
-	 */
-	if (!cancel)
-		mod_delayed_work(system_wq, &cache_cleaner, RC_EXPIRE);
 	return freed;
 }
 
-static void
-cache_cleaner_func(struct work_struct *unused)
-{
-	prune_cache_entries();
-}
-
 static unsigned long
 nfsd_reply_cache_count(struct shrinker *shrink, struct shrink_control *sc)
 {

fs/nfsd/nfsfh.c

@@ -631,10 +631,7 @@ fh_put(struct svc_fh *fhp)
 		fh_unlock(fhp);
 		fhp->fh_dentry = NULL;
 		dput(dentry);
-#ifdef CONFIG_NFSD_V3
-		fhp->fh_pre_saved = 0;
-		fhp->fh_post_saved = 0;
-#endif
+		fh_clear_wcc(fhp);
 	}
 	fh_drop_write(fhp);
 	if (exp) {

fs/nfsd/nfsfh.h

@@ -26,16 +26,16 @@ static inline ino_t u32_to_ino_t(__u32 uino)
  */
 typedef struct svc_fh {
 	struct knfsd_fh		fh_handle;	/* FH data */
+	int			fh_maxsize;	/* max size for fh_handle */
 	struct dentry *		fh_dentry;	/* validated dentry */
 	struct svc_export *	fh_export;	/* export pointer */
-	int			fh_maxsize;	/* max size for fh_handle */
 
-	unsigned char		fh_locked;	/* inode locked by us */
-	unsigned char		fh_want_write;	/* remount protection taken */
+	bool			fh_locked;	/* inode locked by us */
+	bool			fh_want_write;	/* remount protection taken */
 
 #ifdef CONFIG_NFSD_V3
-	unsigned char		fh_post_saved;	/* post-op attrs saved */
-	unsigned char		fh_pre_saved;	/* pre-op attrs saved */
+	bool			fh_post_saved;	/* post-op attrs saved */
+	bool			fh_pre_saved;	/* pre-op attrs saved */
 
 	/* Pre-op attributes saved during fh_lock */
 	__u64			fh_pre_size;	/* size before operation */
@@ -213,8 +213,8 @@ static inline bool fh_fsid_match(struct knfsd_fh *fh1, struct knfsd_fh *fh2)
 static inline void
 fh_clear_wcc(struct svc_fh *fhp)
 {
-	fhp->fh_post_saved = 0;
-	fhp->fh_pre_saved = 0;
+	fhp->fh_post_saved = false;
+	fhp->fh_pre_saved = false;
 }
 
 /*
@@ -231,7 +231,7 @@ fill_pre_wcc(struct svc_fh *fhp)
 		fhp->fh_pre_ctime = inode->i_ctime;
 		fhp->fh_pre_size  = inode->i_size;
 		fhp->fh_pre_change = inode->i_version;
-		fhp->fh_pre_saved = 1;
+		fhp->fh_pre_saved = true;
 	}
 }
 
@@ -267,7 +267,7 @@ fh_lock_nested(struct svc_fh *fhp, unsigned int subclass)
 	inode = d_inode(dentry);
 	mutex_lock_nested(&inode->i_mutex, subclass);
 	fill_pre_wcc(fhp);
-	fhp->fh_locked = 1;
+	fhp->fh_locked = true;
 }
 
 static inline void
@@ -285,7 +285,7 @@ fh_unlock(struct svc_fh *fhp)
 	if (fhp->fh_locked) {
 		fill_post_wcc(fhp);
 		mutex_unlock(&d_inode(fhp->fh_dentry)->i_mutex);
-		fhp->fh_locked = 0;
+		fhp->fh_locked = false;
 	}
 }
 

fs/nfsd/state.h

@@ -96,6 +96,7 @@ struct nfs4_stid {
 #define NFS4_LAYOUT_STID 64
 	unsigned char		sc_type;
 	stateid_t		sc_stateid;
+	spinlock_t		sc_lock;
 	struct nfs4_client	*sc_client;
 	struct nfs4_file	*sc_file;
 	void			(*sc_free)(struct nfs4_stid *);
@@ -364,15 +365,6 @@ struct nfs4_client_reclaim {
 	char			cr_recdir[HEXDIR_LEN]; /* recover dir */
 };
 
-static inline void
-update_stateid(stateid_t *stateid)
-{
-	stateid->si_generation++;
-	/* Wraparound recommendation from 3530bis-13 9.1.3.2: */
-	if (stateid->si_generation == 0)
-		stateid->si_generation = 1;
-}
-
 /* A reasonable value for REPLAY_ISIZE was estimated as follows:  
  * The OPEN response, typically the largest, requires 
  *   4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) +  8(verifier) + 
@@ -534,15 +526,16 @@ struct nfs4_file {
  * Better suggestions welcome.
  */
 struct nfs4_ol_stateid {
-	struct nfs4_stid    st_stid; /* must be first field */
+	struct nfs4_stid		st_stid;
 	struct list_head		st_perfile;
 	struct list_head		st_perstateowner;
 	struct list_head		st_locks;
-	struct nfs4_stateowner      * st_stateowner;
-	struct nfs4_clnt_odstate    * st_clnt_odstate;
+	struct nfs4_stateowner		*st_stateowner;
+	struct nfs4_clnt_odstate	*st_clnt_odstate;
 	unsigned char			st_access_bmap;
 	unsigned char			st_deny_bmap;
-	struct nfs4_ol_stateid         * st_openstp;
+	struct nfs4_ol_stateid		*st_openstp;
+	struct rw_semaphore		st_rwsem;
 };
 
 static inline struct nfs4_ol_stateid *openlockstateid(struct nfs4_stid *s)
@@ -561,6 +554,7 @@ struct nfs4_layout_stateid {
 	struct nfsd4_callback		ls_recall;
 	stateid_t			ls_recall_sid;
 	bool				ls_recalled;
+	struct mutex			ls_mutex;
 };
 
 static inline struct nfs4_layout_stateid *layoutstateid(struct nfs4_stid *s)
@@ -593,6 +587,7 @@ struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl,
 		struct kmem_cache *slab);
 void nfs4_unhash_stid(struct nfs4_stid *s);
 void nfs4_put_stid(struct nfs4_stid *s);
+void nfs4_inc_and_copy_stateid(stateid_t *dst, struct nfs4_stid *stid);
 void nfs4_remove_reclaim_record(struct nfs4_client_reclaim *, struct nfsd_net *);
 extern void nfs4_release_reclaim(struct nfsd_net *);
 extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(const char *recdir,

fs/nfsd/trace.c

 
-#include "state.h"
-
 #define CREATE_TRACE_POINTS
 #include "trace.h"

fs/nfsd/trace.h

@@ -9,6 +9,8 @@
 
 #include <linux/tracepoint.h>
 
+#include "state.h"
+
 DECLARE_EVENT_CLASS(nfsd_stateid_class,
 	TP_PROTO(stateid_t *stp),
 	TP_ARGS(stp),

fs/nfsd/vfs.c

@@ -1631,7 +1631,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
 	/* cannot use fh_lock as we need deadlock protective ordering
 	 * so do it by hand */
 	trap = lock_rename(tdentry, fdentry);
-	ffhp->fh_locked = tfhp->fh_locked = 1;
+	ffhp->fh_locked = tfhp->fh_locked = true;
 	fill_pre_wcc(ffhp);
 	fill_pre_wcc(tfhp);
 
@@ -1681,7 +1681,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
 	fill_post_wcc(ffhp);
 	fill_post_wcc(tfhp);
 	unlock_rename(tdentry, fdentry);
-	ffhp->fh_locked = tfhp->fh_locked = 0;
+	ffhp->fh_locked = tfhp->fh_locked = false;
 	fh_drop_write(ffhp);
 
 out:

fs/nfsd/vfs.h

@@ -112,14 +112,14 @@ static inline int fh_want_write(struct svc_fh *fh)
 	int ret = mnt_want_write(fh->fh_export->ex_path.mnt);
 
 	if (!ret)
-		fh->fh_want_write = 1;
+		fh->fh_want_write = true;
 	return ret;
 }
 
 static inline void fh_drop_write(struct svc_fh *fh)
 {
 	if (fh->fh_want_write) {
-		fh->fh_want_write = 0;
+		fh->fh_want_write = false;
 		mnt_drop_write(fh->fh_export->ex_path.mnt);
 	}
 }

fs/nfsd/xdr4.h

@@ -632,7 +632,7 @@ static inline void
 set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp)
 {
 	BUG_ON(!fhp->fh_pre_saved);
-	cinfo->atomic = fhp->fh_post_saved;
+	cinfo->atomic = (u32)fhp->fh_post_saved;
 	cinfo->change_supported = IS_I_VERSION(d_inode(fhp->fh_dentry));
 
 	cinfo->before_change = fhp->fh_pre_change;

include/linux/lockd/lockd.h

@@ -68,6 +68,7 @@ struct nlm_host {
 	struct nsm_handle	*h_nsmhandle;	/* NSM status handle */
 	char			*h_addrbuf;	/* address eyecatcher */
 	struct net		*net;		/* host net */
+	char			nodename[UNX_MAXNODENAME + 1];
 };
 
 /*
@@ -235,7 +236,8 @@ void		  nlm_rebind_host(struct nlm_host *);
 struct nlm_host * nlm_get_host(struct nlm_host *);
 void		  nlm_shutdown_hosts(void);
 void		  nlm_shutdown_hosts_net(struct net *net);
-void		  nlm_host_rebooted(const struct nlm_reboot *);
+void		  nlm_host_rebooted(const struct net *net,
+					const struct nlm_reboot *);
 
 /*
  * Host monitoring
@@ -243,11 +245,13 @@ void		  nlm_host_rebooted(const struct nlm_reboot *);
 int		  nsm_monitor(const struct nlm_host *host);
 void		  nsm_unmonitor(const struct nlm_host *host);
 
-struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
+struct nsm_handle *nsm_get_handle(const struct net *net,
+					const struct sockaddr *sap,
 					const size_t salen,
 					const char *hostname,
 					const size_t hostname_len);
-struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info);
+struct nsm_handle *nsm_reboot_lookup(const struct net *net,
+					const struct nlm_reboot *info);
 void		  nsm_release(struct nsm_handle *nsm);
 
 /*

include/linux/sunrpc/cache.h

@@ -48,8 +48,10 @@
 struct cache_head {
 	struct hlist_node	cache_list;
 	time_t		expiry_time;	/* After time time, don't use the data */
-	time_t		last_refresh;   /* If CACHE_PENDING, this is when upcall 
-					 * was sent, else this is when update was received
+	time_t		last_refresh;   /* If CACHE_PENDING, this is when upcall was
+					 * sent, else this is when update was
+					 * received, though it is alway set to
+					 * be *after* ->flush_time.
 					 */
 	struct kref	ref;
 	unsigned long	flags;
@@ -105,8 +107,12 @@ struct cache_detail {
 	/* fields below this comment are for internal use
 	 * and should not be touched by cache owners
 	 */
-	time_t			flush_time;		/* flush all cache items with last_refresh
-							 * earlier than this */
+	time_t			flush_time;		/* flush all cache items with
+							 * last_refresh at or earlier
+							 * than this.  last_refresh
+							 * is never set at or earlier
+							 * than this.
+							 */
 	struct list_head	others;
 	time_t			nextcheck;
 	int			entries;
@@ -203,7 +209,7 @@ static inline void cache_put(struct cache_head *h, struct cache_detail *cd)
 static inline int cache_is_expired(struct cache_detail *detail, struct cache_head *h)
 {
 	return  (h->expiry_time < seconds_since_boot()) ||
-		(detail->flush_time > h->last_refresh);
+		(detail->flush_time >= h->last_refresh);
 }
 
 extern int cache_check(struct cache_detail *detail,

net/sunrpc/auth_gss/auth_gss.c

@@ -1411,17 +1411,16 @@ gss_key_timeout(struct rpc_cred *rc)
 {
 	struct gss_cred *gss_cred = container_of(rc, struct gss_cred, gc_base);
 	struct gss_cl_ctx *ctx;
-	unsigned long now = jiffies;
-	unsigned long expire;
+	unsigned long timeout = jiffies + (gss_key_expire_timeo * HZ);
+	int ret = 0;
 
 	rcu_read_lock();
 	ctx = rcu_dereference(gss_cred->gc_ctx);
-	if (ctx)
-		expire = ctx->gc_expiry - (gss_key_expire_timeo * HZ);
+	if (!ctx || time_after(timeout, ctx->gc_expiry))
+		ret = -EACCES;
 	rcu_read_unlock();
-	if (!ctx || time_after(now, expire))
-		return -EACCES;
-	return 0;
+
+	return ret;
 }
 
 static int

net/sunrpc/cache.c

@@ -41,13 +41,16 @@
 static bool cache_defer_req(struct cache_req *req, struct cache_head *item);
 static void cache_revisit_request(struct cache_head *item);
 
-static void cache_init(struct cache_head *h)
+static void cache_init(struct cache_head *h, struct cache_detail *detail)
 {
 	time_t now = seconds_since_boot();
 	INIT_HLIST_NODE(&h->cache_list);
 	h->flags = 0;
 	kref_init(&h->ref);
 	h->expiry_time = now + CACHE_NEW_EXPIRY;
+	if (now <= detail->flush_time)
+		/* ensure it isn't already expired */
+		now = detail->flush_time + 1;
 	h->last_refresh = now;
 }
 
@@ -81,7 +84,7 @@ struct cache_head *sunrpc_cache_lookup(struct cache_detail *detail,
 	 * we might get lose if we need to
 	 * cache_put it soon.
 	 */
-	cache_init(new);
+	cache_init(new, detail);
 	detail->init(new, key);
 
 	write_lock(&detail->hash_lock);
@@ -116,10 +119,15 @@ EXPORT_SYMBOL_GPL(sunrpc_cache_lookup);
 
 static void cache_dequeue(struct cache_detail *detail, struct cache_head *ch);
 
-static void cache_fresh_locked(struct cache_head *head, time_t expiry)
+static void cache_fresh_locked(struct cache_head *head, time_t expiry,
+			       struct cache_detail *detail)
 {
+	time_t now = seconds_since_boot();
+	if (now <= detail->flush_time)
+		/* ensure it isn't immediately treated as expired */
+		now = detail->flush_time + 1;
 	head->expiry_time = expiry;
-	head->last_refresh = seconds_since_boot();
+	head->last_refresh = now;
 	smp_wmb(); /* paired with smp_rmb() in cache_is_valid() */
 	set_bit(CACHE_VALID, &head->flags);
 }
@@ -149,7 +157,7 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
 				set_bit(CACHE_NEGATIVE, &old->flags);
 			else
 				detail->update(old, new);
-			cache_fresh_locked(old, new->expiry_time);
+			cache_fresh_locked(old, new->expiry_time, detail);
 			write_unlock(&detail->hash_lock);
 			cache_fresh_unlocked(old, detail);
 			return old;
@@ -162,7 +170,7 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
 		cache_put(old, detail);
 		return NULL;
 	}
-	cache_init(tmp);
+	cache_init(tmp, detail);
 	detail->init(tmp, old);
 
 	write_lock(&detail->hash_lock);
@@ -173,8 +181,8 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
 	hlist_add_head(&tmp->cache_list, &detail->hash_table[hash]);
 	detail->entries++;
 	cache_get(tmp);
-	cache_fresh_locked(tmp, new->expiry_time);
-	cache_fresh_locked(old, 0);
+	cache_fresh_locked(tmp, new->expiry_time, detail);
+	cache_fresh_locked(old, 0, detail);
 	write_unlock(&detail->hash_lock);
 	cache_fresh_unlocked(tmp, detail);
 	cache_fresh_unlocked(old, detail);
@@ -219,7 +227,8 @@ static int try_to_negate_entry(struct cache_detail *detail, struct cache_head *h
 	rv = cache_is_valid(h);
 	if (rv == -EAGAIN) {
 		set_bit(CACHE_NEGATIVE, &h->flags);
-		cache_fresh_locked(h, seconds_since_boot()+CACHE_NEW_EXPIRY);
+		cache_fresh_locked(h, seconds_since_boot()+CACHE_NEW_EXPIRY,
+				   detail);
 		rv = -ENOENT;
 	}
 	write_unlock(&detail->hash_lock);
@@ -487,10 +496,13 @@ EXPORT_SYMBOL_GPL(cache_flush);
 
 void cache_purge(struct cache_detail *detail)
 {
-	detail->flush_time = LONG_MAX;
+	time_t now = seconds_since_boot();
+	if (detail->flush_time >= now)
+		now = detail->flush_time + 1;
+	/* 'now' is the maximum value any 'last_refresh' can have */
+	detail->flush_time = now;
 	detail->nextcheck = seconds_since_boot();
 	cache_flush();
-	detail->flush_time = 1;
 }
 EXPORT_SYMBOL_GPL(cache_purge);
 
@@ -1436,6 +1448,7 @@ static ssize_t write_flush(struct file *file, const char __user *buf,
 {
 	char tbuf[20];
 	char *bp, *ep;
+	time_t then, now;
 
 	if (*ppos || count > sizeof(tbuf)-1)
 		return -EINVAL;
@@ -1447,8 +1460,22 @@ static ssize_t write_flush(struct file *file, const char __user *buf,
 		return -EINVAL;
 
 	bp = tbuf;
-	cd->flush_time = get_expiry(&bp);
-	cd->nextcheck = seconds_since_boot();
+	then = get_expiry(&bp);
+	now = seconds_since_boot();
+	cd->nextcheck = now;
+	/* Can only set flush_time to 1 second beyond "now", or
+	 * possibly 1 second beyond flushtime.  This is because
+	 * flush_time never goes backwards so it mustn't get too far
+	 * ahead of time.
+	 */
+	if (then >= now) {
+		/* Want to flush everything, so behave like cache_purge() */
+		if (cd->flush_time >= now)
+			now = cd->flush_time + 1;
+		then = now;
+	}
+
+	cd->flush_time = then;
 	cache_flush();
 
 	*ppos += count;

net/sunrpc/svcsock.c

@@ -181,7 +181,7 @@ int svc_send_common(struct socket *sock, struct xdr_buf *xdr,
 	struct page	**ppage = xdr->pages;
 	size_t		base = xdr->page_base;
 	unsigned int	pglen = xdr->page_len;
-	unsigned int	flags = MSG_MORE;
+	unsigned int	flags = MSG_MORE | MSG_SENDPAGE_NOTLAST;
 	int		slen;
 	int		len = 0;
 
@@ -399,6 +399,31 @@ static int svc_sock_secure_port(struct svc_rqst *rqstp)
 	return svc_port_is_privileged(svc_addr(rqstp));
 }
 
+static bool sunrpc_waitqueue_active(wait_queue_head_t *wq)
+{
+	if (!wq)
+		return false;
+	/*
+	 * There should normally be a memory * barrier here--see
+	 * wq_has_sleeper().
+	 *
+	 * It appears that isn't currently necessary, though, basically
+	 * because callers all appear to have sufficient memory barriers
+	 * between the time the relevant change is made and the
+	 * time they call these callbacks.
+	 *
+	 * The nfsd code itself doesn't actually explicitly wait on
+	 * these waitqueues, but it may wait on them for example in
+	 * sendpage() or sendmsg() calls.  (And those may be the only
+	 * places, since it it uses nonblocking reads.)
+	 *
+	 * Maybe we should add the memory barriers anyway, but these are
+	 * hot paths so we'd need to be convinced there's no sigificant
+	 * penalty.
+	 */
+	return waitqueue_active(wq);
+}
+
 /*
  * INET callback when data has been received on the socket.
  */
@@ -414,7 +439,7 @@ static void svc_udp_data_ready(struct sock *sk)
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (wq && waitqueue_active(wq))
+	if (sunrpc_waitqueue_active(wq))
 		wake_up_interruptible(wq);
 }
 
@@ -432,7 +457,7 @@ static void svc_write_space(struct sock *sk)
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
 
-	if (wq && waitqueue_active(wq)) {
+	if (sunrpc_waitqueue_active(wq)) {
 		dprintk("RPC svc_write_space: someone sleeping on %p\n",
 		       svsk);
 		wake_up_interruptible(wq);
@@ -787,7 +812,7 @@ static void svc_tcp_listen_data_ready(struct sock *sk)
 	}
 
 	wq = sk_sleep(sk);
-	if (wq && waitqueue_active(wq))
+	if (sunrpc_waitqueue_active(wq))
 		wake_up_interruptible_all(wq);
 }
 
@@ -808,7 +833,7 @@ static void svc_tcp_state_change(struct sock *sk)
 		set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (wq && waitqueue_active(wq))
+	if (sunrpc_waitqueue_active(wq))
 		wake_up_interruptible_all(wq);
 }
 
@@ -823,7 +848,7 @@ static void svc_tcp_data_ready(struct sock *sk)
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (wq && waitqueue_active(wq))
+	if (sunrpc_waitqueue_active(wq))
 		wake_up_interruptible(wq);
 }
 
@@ -1367,7 +1392,6 @@ EXPORT_SYMBOL_GPL(svc_sock_update_bufs);
 
 /*
  * Initialize socket for RPC use and create svc_sock struct
- * XXX: May want to setsockopt SO_SNDBUF and SO_RCVBUF.
  */
 static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 						struct socket *sock,
@@ -1594,7 +1618,7 @@ static void svc_sock_detach(struct svc_xprt *xprt)
 	sk->sk_write_space = svsk->sk_owspace;
 
 	wq = sk_sleep(sk);
-	if (wq && waitqueue_active(wq))
+	if (sunrpc_waitqueue_active(wq))
 		wake_up_interruptible(wq);
 }