Merge tag 'nfs-for-4.9-1' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client updates from Anna Schumaker:
 "Highlights include:

  Stable bugfixes:
   - sunrpc: fix writ espace race causing stalls
   - NFS: Fix inode corruption in nfs_prime_dcache()
   - NFSv4: Don't report revoked delegations as valid in nfs_have_delegation()
   - NFSv4: nfs4_copy_delegation_stateid() must fail if the delegation is invalid
   - NFSv4: Open state recovery must account for file permission changes
   - NFSv4.2: Fix a reference leak in nfs42_proc_layoutstats_generic

  Features:
   - Add support for tracking multiple layout types with an ordered list
   - Add support for using multiple backchannel threads on the client
   - Add support for pNFS file layout session trunking
   - Delay xprtrdma use of DMA API (for device driver removal)
   - Add support for xprtrdma remote invalidation
   - Add support for larger xprtrdma inline thresholds
   - Use a scatter/gather list for sending xprtrdma RPC calls
   - Add support for the CB_NOTIFY_LOCK callback
   - Improve hashing sunrpc auth_creds by using both uid and gid

  Bugfixes:
   - Fix xprtrdma use of DMA API
   - Validate filenames before adding to the dcache
   - Fix corruption of xdr->nwords in xdr_copy_to_scratch
   - Fix setting buffer length in xdr_set_next_buffer()
   - Don't deadlock the state manager on the SEQUENCE status flags
   - Various delegation and stateid related fixes
   - Retry operations if an interrupted slot receives EREMOTEIO
   - Make nfs boot time y2038 safe"

* tag 'nfs-for-4.9-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (100 commits)
  NFSv4.2: Fix a reference leak in nfs42_proc_layoutstats_generic
  fs: nfs: Make nfs boot time y2038 safe
  sunrpc: replace generic auth_cred hash with auth-specific function
  sunrpc: add RPCSEC_GSS hash_cred() function
  sunrpc: add auth_unix hash_cred() function
  sunrpc: add generic_auth hash_cred() function
  sunrpc: add hash_cred() function to rpc_authops struct
  Retry operation on EREMOTEIO on an interrupted slot
  pNFS: Fix atime updates on pNFS clients
  sunrpc: queue work on system_power_efficient_wq
  NFSv4.1: Even if the stateid is OK, we may need to recover the open modes
  NFSv4: If recovery failed for a specific open stateid, then don't retry
  NFSv4: Fix retry issues with nfs41_test/free_stateid
  NFSv4: Open state recovery must account for file permission changes
  NFSv4: Mark the lock and open stateids as invalid after freeing them
  NFSv4: Don't test open_stateid unless it is set
  NFSv4: nfs4_do_handle_exception() handle revoke/expiry of a single stateid
  NFS: Always call nfs_inode_find_state_and_recover() when revoking a delegation
  NFSv4: Fix a race when updating an open_stateid
  NFSv4: Fix a race in nfs_inode_reclaim_delegation()
  ...

Author: torvalds

Documentation/kernel-parameters.txt

@@ -2470,6 +2470,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 	nfsrootdebug	[NFS] enable nfsroot debugging messages.
 			See Documentation/filesystems/nfs/nfsroot.txt.
 
+	nfs.callback_nr_threads=
+			[NFSv4] set the total number of threads that the
+			NFS client will assign to service NFSv4 callback
+			requests.
+
 	nfs.callback_tcpport=
 			[NFS] set the TCP port on which the NFSv4 callback
 			channel should listen.
@@ -2493,6 +2498,13 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			of returning the full 64-bit number.
 			The default is to return 64-bit inode numbers.
 
+	nfs.max_session_cb_slots=
+			[NFSv4.1] Sets the maximum number of session
+			slots the client will assign to the callback
+			channel. This determines the maximum number of
+			callbacks the client will process in parallel for
+			a particular server.
+
 	nfs.max_session_slots=
 			[NFSv4.1] Sets the maximum number of session slots
 			the client will attempt to negotiate with the server.

fs/nfs/cache_lib.c

@@ -76,7 +76,7 @@ static void nfs_dns_cache_revisit(struct cache_deferred_req *d, int toomany)
 
 	dreq = container_of(d, struct nfs_cache_defer_req, deferred_req);
 
-	complete_all(&dreq->completion);
+	complete(&dreq->completion);
 	nfs_cache_defer_req_put(dreq);
 }
 

fs/nfs/callback.c

@@ -31,8 +31,6 @@
 struct nfs_callback_data {
 	unsigned int users;
 	struct svc_serv *serv;
-	struct svc_rqst *rqst;
-	struct task_struct *task;
 };
 
 static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1];
@@ -89,15 +87,6 @@ nfs4_callback_svc(void *vrqstp)
 	return 0;
 }
 
-/*
- * Prepare to bring up the NFSv4 callback service
- */
-static struct svc_rqst *
-nfs4_callback_up(struct svc_serv *serv)
-{
-	return svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
-}
-
 #if defined(CONFIG_NFS_V4_1)
 /*
  * The callback service for NFSv4.1 callbacks
@@ -139,29 +128,6 @@ nfs41_callback_svc(void *vrqstp)
 	return 0;
 }
 
-/*
- * Bring up the NFSv4.1 callback service
- */
-static struct svc_rqst *
-nfs41_callback_up(struct svc_serv *serv)
-{
-	struct svc_rqst *rqstp;
-
-	INIT_LIST_HEAD(&serv->sv_cb_list);
-	spin_lock_init(&serv->sv_cb_lock);
-	init_waitqueue_head(&serv->sv_cb_waitq);
-	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
-	dprintk("--> %s return %d\n", __func__, PTR_ERR_OR_ZERO(rqstp));
-	return rqstp;
-}
-
-static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv,
-		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
-{
-	*rqstpp = nfs41_callback_up(serv);
-	*callback_svc = nfs41_callback_svc;
-}
-
 static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
 		struct svc_serv *serv)
 {
@@ -173,13 +139,6 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
 		xprt->bc_serv = serv;
 }
 #else
-static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv,
-		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
-{
-	*rqstpp = ERR_PTR(-ENOTSUPP);
-	*callback_svc = ERR_PTR(-ENOTSUPP);
-}
-
 static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
 		struct svc_serv *serv)
 {
@@ -189,45 +148,22 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
 static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt,
 				  struct svc_serv *serv)
 {
-	struct svc_rqst *rqstp;
-	int (*callback_svc)(void *vrqstp);
-	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+	int nrservs = nfs_callback_nr_threads;
 	int ret;
 
 	nfs_callback_bc_serv(minorversion, xprt, serv);
 
-	if (cb_info->task)
-		return 0;
+	if (nrservs < NFS4_MIN_NR_CALLBACK_THREADS)
+		nrservs = NFS4_MIN_NR_CALLBACK_THREADS;
 
-	switch (minorversion) {
-	case 0:
-		/* v4.0 callback setup */
-		rqstp = nfs4_callback_up(serv);
-		callback_svc = nfs4_callback_svc;
-		break;
-	default:
-		nfs_minorversion_callback_svc_setup(serv,
-				&rqstp, &callback_svc);
-	}
-
-	if (IS_ERR(rqstp))
-		return PTR_ERR(rqstp);
-
-	svc_sock_update_bufs(serv);
+	if (serv->sv_nrthreads-1 == nrservs)
+		return 0;
 
-	cb_info->serv = serv;
-	cb_info->rqst = rqstp;
-	cb_info->task = kthread_create(callback_svc, cb_info->rqst,
-				    "nfsv4.%u-svc", minorversion);
-	if (IS_ERR(cb_info->task)) {
-		ret = PTR_ERR(cb_info->task);
-		svc_exit_thread(cb_info->rqst);
-		cb_info->rqst = NULL;
-		cb_info->task = NULL;
+	ret = serv->sv_ops->svo_setup(serv, NULL, nrservs);
+	if (ret) {
+		serv->sv_ops->svo_setup(serv, NULL, 0);
 		return ret;
 	}
-	rqstp->rq_task = cb_info->task;
-	wake_up_process(cb_info->task);
 	dprintk("nfs_callback_up: service started\n");
 	return 0;
 }
@@ -281,19 +217,41 @@ static int nfs_callback_up_net(int minorversion, struct svc_serv *serv,
 	return ret;
 }
 
-static struct svc_serv_ops nfs_cb_sv_ops = {
+static struct svc_serv_ops nfs40_cb_sv_ops = {
+	.svo_function		= nfs4_callback_svc,
 	.svo_enqueue_xprt	= svc_xprt_do_enqueue,
+	.svo_setup		= svc_set_num_threads,
+	.svo_module		= THIS_MODULE,
+};
+#if defined(CONFIG_NFS_V4_1)
+static struct svc_serv_ops nfs41_cb_sv_ops = {
+	.svo_function		= nfs41_callback_svc,
+	.svo_enqueue_xprt	= svc_xprt_do_enqueue,
+	.svo_setup		= svc_set_num_threads,
+	.svo_module		= THIS_MODULE,
+};
+
+struct svc_serv_ops *nfs4_cb_sv_ops[] = {
+	[0] = &nfs40_cb_sv_ops,
+	[1] = &nfs41_cb_sv_ops,
+};
+#else
+struct svc_serv_ops *nfs4_cb_sv_ops[] = {
+	[0] = &nfs40_cb_sv_ops,
+	[1] = NULL,
 };
+#endif
 
 static struct svc_serv *nfs_callback_create_svc(int minorversion)
 {
 	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
 	struct svc_serv *serv;
+	struct svc_serv_ops *sv_ops;
 
 	/*
 	 * Check whether we're already up and running.
 	 */
-	if (cb_info->task) {
+	if (cb_info->serv) {
 		/*
 		 * Note: increase service usage, because later in case of error
 		 * svc_destroy() will be called.
@@ -302,6 +260,17 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion)
 		return cb_info->serv;
 	}
 
+	switch (minorversion) {
+	case 0:
+		sv_ops = nfs4_cb_sv_ops[0];
+		break;
+	default:
+		sv_ops = nfs4_cb_sv_ops[1];
+	}
+
+	if (sv_ops == NULL)
+		return ERR_PTR(-ENOTSUPP);
+
 	/*
 	 * Sanity check: if there's no task,
 	 * we should be the first user ...
@@ -310,11 +279,12 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion)
 		printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n",
 			cb_info->users);
 
-	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, &nfs_cb_sv_ops);
+	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, sv_ops);
 	if (!serv) {
 		printk(KERN_ERR "nfs_callback_create_svc: create service failed\n");
 		return ERR_PTR(-ENOMEM);
 	}
+	cb_info->serv = serv;
 	/* As there is only one thread we need to over-ride the
 	 * default maximum of 80 connections
 	 */
@@ -357,6 +327,8 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	 * thread exits.
 	 */
 err_net:
+	if (!cb_info->users)
+		cb_info->serv = NULL;
 	svc_destroy(serv);
 err_create:
 	mutex_unlock(&nfs_callback_mutex);
@@ -374,18 +346,18 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 void nfs_callback_down(int minorversion, struct net *net)
 {
 	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+	struct svc_serv *serv;
 
 	mutex_lock(&nfs_callback_mutex);
-	nfs_callback_down_net(minorversion, cb_info->serv, net);
+	serv = cb_info->serv;
+	nfs_callback_down_net(minorversion, serv, net);
 	cb_info->users--;
-	if (cb_info->users == 0 && cb_info->task != NULL) {
-		kthread_stop(cb_info->task);
-		dprintk("nfs_callback_down: service stopped\n");
-		svc_exit_thread(cb_info->rqst);
+	if (cb_info->users == 0) {
+		svc_get(serv);
+		serv->sv_ops->svo_setup(serv, NULL, 0);
+		svc_destroy(serv);
 		dprintk("nfs_callback_down: service destroyed\n");
 		cb_info->serv = NULL;
-		cb_info->rqst = NULL;
-		cb_info->task = NULL;
 	}
 	mutex_unlock(&nfs_callback_mutex);
 }

fs/nfs/callback.h

@@ -179,6 +179,15 @@ extern __be32 nfs4_callback_devicenotify(
 	struct cb_devicenotifyargs *args,
 	void *dummy, struct cb_process_state *cps);
 
+struct cb_notify_lock_args {
+	struct nfs_fh			cbnl_fh;
+	struct nfs_lowner		cbnl_owner;
+	bool				cbnl_valid;
+};
+
+extern __be32 nfs4_callback_notify_lock(struct cb_notify_lock_args *args,
+					 void *dummy,
+					 struct cb_process_state *cps);
 #endif /* CONFIG_NFS_V4_1 */
 extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *);
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args,
@@ -198,6 +207,9 @@ extern void nfs_callback_down(int minorversion, struct net *net);
 #define NFS41_BC_MIN_CALLBACKS 1
 #define NFS41_BC_MAX_CALLBACKS 1
 
+#define NFS4_MIN_NR_CALLBACK_THREADS 1
+
 extern unsigned int nfs_callback_set_tcpport;
+extern unsigned short nfs_callback_nr_threads;
 
 #endif /* __LINUX_FS_NFS_CALLBACK_H */

fs/nfs/callback_proc.c

@@ -628,4 +628,20 @@ __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy,
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	return status;
 }
+
+__be32 nfs4_callback_notify_lock(struct cb_notify_lock_args *args, void *dummy,
+				 struct cb_process_state *cps)
+{
+	if (!cps->clp) /* set in cb_sequence */
+		return htonl(NFS4ERR_OP_NOT_IN_SESSION);
+
+	dprintk_rcu("NFS: CB_NOTIFY_LOCK request from %s\n",
+		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	/* Don't wake anybody if the string looked bogus */
+	if (args->cbnl_valid)
+		__wake_up(&cps->clp->cl_lock_waitq, TASK_NORMAL, 0, args);
+
+	return htonl(NFS4_OK);
+}
 #endif /* CONFIG_NFS_V4_1 */