Merge tag 'nfs-for-4.2-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client updates from Trond Myklebust:
 "Highlights include:

  Stable patches:
   - Fix a crash in the NFSv4 file locking code.
   - Fix an fsync() regression, where we were failing to retry I/O in
     some circumstances.
   - Fix an infinite loop in NFSv4.0 OPEN stateid recovery
   - Fix a memory leak when an attempted pnfs fails.
   - Fix a memory leak in the backchannel code
   - Large hostnames were not supported correctly in NFSv4.1
   - Fix a pNFS/flexfiles bug that was impeding error reporting on I/O.
   - Fix a couple of credential issues in pNFS/flexfiles

  Bugfixes + cleanups:
   - Open flag sanity checks in the NFSv4 atomic open codepath
   - More NFSv4 delegation related bugfixes
   - Various NFSv4.1 backchannel bugfixes and cleanups
   - Fix the NFS swap socket code
   - Various cleanups of the NFSv4 SETCLIENTID and EXCHANGE_ID code
   - Fix a UDP transport deadlock issue

  Features:
   - More RDMA client transport improvements
   - NFSv4.2 LAYOUTSTATS functionality for pnfs flexfiles"

* tag 'nfs-for-4.2-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (87 commits)
  nfs: Remove invalid tk_pid from debug message
  nfs: Remove invalid NFS_ATTR_FATTR_V4_REFERRAL checking in nfs4_get_rootfh
  nfs: Drop bad comment in nfs41_walk_client_list()
  nfs: Remove unneeded micro checking of CONFIG_PROC_FS
  nfs: Don't setting FILE_CREATED flags always
  nfs: Use remove_proc_subtree() instead remove_proc_entry()
  nfs: Remove unused argument in nfs_server_set_fsinfo()
  nfs: Fix a memory leak when meeting an unsupported state protect
  nfs: take extra reference to fl->fl_file when running a LOCKU operation
  NFSv4: When returning a delegation, don't reclaim an incompatible open mode.
  NFSv4.2: LAYOUTSTATS is optional to implement
  NFSv4.2: Fix up a decoding error in layoutstats
  pNFS/flexfiles: Fix the reset of struct pgio_header when resending
  pNFS/flexfiles: Turn off layoutcommit for servers that don't need it
  pnfs/flexfiles: protect ktime manipulation with mirror lock
  nfs: provide pnfs_report_layoutstat when NFS42 is disabled
  nfs: verify open flags before allowing open
  nfs: always update creds in mirror, even when we have an already connected ds
  nfs: fix potential credential leak in ff_layout_update_mirror_cred
  pnfs/flexfiles: report layoutstat regularly
  ...

fs/nfs/callback.c

@@ -458,7 +458,7 @@ check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp)
  * pg_authenticate method for nfsv4 callback threads.
  *
  * The authflavor has been negotiated, so an incorrect flavor is a server
- * bug. Drop packets with incorrect authflavor.
+ * bug. Deny packets with incorrect authflavor.
  *
  * All other checking done after NFS decoding where the nfs_client can be
  * found in nfs4_callback_compound
@@ -468,12 +468,12 @@ static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 	switch (rqstp->rq_authop->flavour) {
 	case RPC_AUTH_NULL:
 		if (rqstp->rq_proc != CB_NULL)
-			return SVC_DROP;
+			return SVC_DENIED;
 		break;
 	case RPC_AUTH_GSS:
 		/* No RPC_AUTH_GSS support yet in NFSv4.1 */
 		 if (svc_is_backchannel(rqstp))
-			return SVC_DROP;
+			return SVC_DENIED;
 	}
 	return SVC_OK;
 }

fs/nfs/callback_proc.c

@@ -327,10 +327,8 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
 	dprintk("%s slot table seqid: %u\n", __func__, slot->seq_nr);
 
 	/* Normal */
-	if (likely(args->csa_sequenceid == slot->seq_nr + 1)) {
-		slot->seq_nr++;
+	if (likely(args->csa_sequenceid == slot->seq_nr + 1))
 		goto out_ok;
-	}
 
 	/* Replay */
 	if (args->csa_sequenceid == slot->seq_nr) {
@@ -418,6 +416,7 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 			      struct cb_process_state *cps)
 {
 	struct nfs4_slot_table *tbl;
+	struct nfs4_slot *slot;
 	struct nfs_client *clp;
 	int i;
 	__be32 status = htonl(NFS4ERR_BADSESSION);
@@ -429,25 +428,32 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 
 	if (!(clp->cl_session->flags & SESSION4_BACK_CHAN))
 		goto out;
+
 	tbl = &clp->cl_session->bc_slot_table;
+	slot = tbl->slots + args->csa_slotid;
 
 	spin_lock(&tbl->slot_tbl_lock);
 	/* state manager is resetting the session */
 	if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) {
-		spin_unlock(&tbl->slot_tbl_lock);
 		status = htonl(NFS4ERR_DELAY);
 		/* Return NFS4ERR_BADSESSION if we're draining the session
 		 * in order to reset it.
 		 */
 		if (test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state))
 			status = htonl(NFS4ERR_BADSESSION);
-		goto out;
+		goto out_unlock;
 	}
 
-	status = validate_seqid(&clp->cl_session->bc_slot_table, args);
-	spin_unlock(&tbl->slot_tbl_lock);
+	memcpy(&res->csr_sessionid, &args->csa_sessionid,
+	       sizeof(res->csr_sessionid));
+	res->csr_sequenceid = args->csa_sequenceid;
+	res->csr_slotid = args->csa_slotid;
+	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+
+	status = validate_seqid(tbl, args);
 	if (status)
-		goto out;
+		goto out_unlock;
 
 	cps->slotid = args->csa_slotid;
 
@@ -458,15 +464,17 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
 	 */
 	if (referring_call_exists(clp, args->csa_nrclists, args->csa_rclists)) {
 		status = htonl(NFS4ERR_DELAY);
-		goto out;
+		goto out_unlock;
 	}
 
-	memcpy(&res->csr_sessionid, &args->csa_sessionid,
-	       sizeof(res->csr_sessionid));
-	res->csr_sequenceid = args->csa_sequenceid;
-	res->csr_slotid = args->csa_slotid;
-	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
-	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+	/*
+	 * RFC5661 20.9.3
+	 * If CB_SEQUENCE returns an error, then the state of the slot
+	 * (sequence ID, cached reply) MUST NOT change.
+	 */
+	slot->seq_nr++;
+out_unlock:
+	spin_unlock(&tbl->slot_tbl_lock);
 
 out:
 	cps->clp = clp; /* put in nfs4_callback_compound */

fs/nfs/callback_xdr.c

@@ -909,7 +909,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 	xdr_init_encode(&xdr_out, &rqstp->rq_res, p);
 
 	status = decode_compound_hdr_arg(&xdr_in, &hdr_arg);
-	if (status == __constant_htonl(NFS4ERR_RESOURCE))
+	if (status == htonl(NFS4ERR_RESOURCE))
 		return rpc_garbage_args;
 
 	if (hdr_arg.minorversion == 0) {

fs/nfs/client.c

@@ -825,7 +825,6 @@ error:
  * Load up the server record from information gained in an fsinfo record
  */
 static void nfs_server_set_fsinfo(struct nfs_server *server,
-				  struct nfs_fh *mntfh,
 				  struct nfs_fsinfo *fsinfo)
 {
 	unsigned long max_rpc_payload;
@@ -901,7 +900,7 @@ int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs
 	if (error < 0)
 		goto out_error;
 
-	nfs_server_set_fsinfo(server, mntfh, &fsinfo);
+	nfs_server_set_fsinfo(server, &fsinfo);
 
 	/* Get some general file system info */
 	if (server->namelen == 0) {
@@ -1193,8 +1192,6 @@ void nfs_clients_init(struct net *net)
 }
 
 #ifdef CONFIG_PROC_FS
-static struct proc_dir_entry *proc_fs_nfs;
-
 static int nfs_server_list_open(struct inode *inode, struct file *file);
 static void *nfs_server_list_start(struct seq_file *p, loff_t *pos);
 static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos);
@@ -1364,27 +1361,29 @@ static int nfs_volume_list_show(struct seq_file *m, void *v)
 {
 	struct nfs_server *server;
 	struct nfs_client *clp;
-	char dev[8], fsid[17];
+	char dev[13];	// 8 for 2^24, 1 for ':', 3 for 2^8, 1 for '\0'
+	char fsid[34];	// 2 * 16 for %llx, 1 for ':', 1 for '\0'
 	struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id);
 
 	/* display header on line 1 */
 	if (v == &nn->nfs_volume_list) {
-		seq_puts(m, "NV SERVER   PORT DEV     FSID              FSC\n");
+		seq_puts(m, "NV SERVER   PORT DEV          FSID"
+			    "                              FSC\n");
 		return 0;
 	}
 	/* display one transport per line on subsequent lines */
 	server = list_entry(v, struct nfs_server, master_link);
 	clp = server->nfs_client;
 
-	snprintf(dev, 8, "%u:%u",
+	snprintf(dev, sizeof(dev), "%u:%u",
 		 MAJOR(server->s_dev), MINOR(server->s_dev));
 
-	snprintf(fsid, 17, "%llx:%llx",
+	snprintf(fsid, sizeof(fsid), "%llx:%llx",
 		 (unsigned long long) server->fsid.major,
 		 (unsigned long long) server->fsid.minor);
 
 	rcu_read_lock();
-	seq_printf(m, "v%u %s %s %-7s %-17s %s\n",
+	seq_printf(m, "v%u %s %s %-12s %-33s %s\n",
 		   clp->rpc_ops->version,
 		   rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_ADDR),
 		   rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_PORT),
@@ -1434,27 +1433,20 @@ void nfs_fs_proc_net_exit(struct net *net)
  */
 int __init nfs_fs_proc_init(void)
 {
-	struct proc_dir_entry *p;
-
-	proc_fs_nfs = proc_mkdir("fs/nfsfs", NULL);
-	if (!proc_fs_nfs)
+	if (!proc_mkdir("fs/nfsfs", NULL))
 		goto error_0;
 
 	/* a file of servers with which we're dealing */
-	p = proc_symlink("servers", proc_fs_nfs, "../../net/nfsfs/servers");
-	if (!p)
+	if (!proc_symlink("fs/nfsfs/servers", NULL, "../../net/nfsfs/servers"))
 		goto error_1;
 
 	/* a file of volumes that we have mounted */
-	p = proc_symlink("volumes", proc_fs_nfs, "../../net/nfsfs/volumes");
-	if (!p)
-		goto error_2;
-	return 0;
+	if (!proc_symlink("fs/nfsfs/volumes", NULL, "../../net/nfsfs/volumes"))
+		goto error_1;
 
-error_2:
-	remove_proc_entry("servers", proc_fs_nfs);
+	return 0;
 error_1:
-	remove_proc_entry("fs/nfsfs", NULL);
+	remove_proc_subtree("fs/nfsfs", NULL);
 error_0:
 	return -ENOMEM;
 }
@@ -1464,9 +1456,7 @@ error_0:
  */
 void nfs_fs_proc_exit(void)
 {
-	remove_proc_entry("volumes", proc_fs_nfs);
-	remove_proc_entry("servers", proc_fs_nfs);
-	remove_proc_entry("fs/nfsfs", NULL);
+	remove_proc_subtree("fs/nfsfs", NULL);
 }
 
 #endif /* CONFIG_PROC_FS */

fs/nfs/dir.c

@@ -1470,9 +1470,6 @@ static int nfs_finish_open(struct nfs_open_context *ctx,
 {
 	int err;
 
-	if ((open_flags & (O_CREAT | O_EXCL)) == (O_CREAT | O_EXCL))
-		*opened |= FILE_CREATED;
-
 	err = finish_open(file, dentry, do_open, opened);
 	if (err)
 		goto out;

fs/nfs/file.c

@@ -555,31 +555,22 @@ static int nfs_launder_page(struct page *page)
 	return nfs_wb_page(inode, page);
 }
 
-#ifdef CONFIG_NFS_SWAP
 static int nfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 						sector_t *span)
 {
-	int ret;
 	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
 
 	*span = sis->pages;
 
-	rcu_read_lock();
-	ret = xs_swapper(rcu_dereference(clnt->cl_xprt), 1);
-	rcu_read_unlock();
-
-	return ret;
+	return rpc_clnt_swap_activate(clnt);
 }
 
 static void nfs_swap_deactivate(struct file *file)
 {
 	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
 
-	rcu_read_lock();
-	xs_swapper(rcu_dereference(clnt->cl_xprt), 0);
-	rcu_read_unlock();
+	rpc_clnt_swap_deactivate(clnt);
 }
-#endif
 
 const struct address_space_operations nfs_file_aops = {
 	.readpage = nfs_readpage,
@@ -596,10 +587,8 @@ const struct address_space_operations nfs_file_aops = {
 	.launder_page = nfs_launder_page,
 	.is_dirty_writeback = nfs_check_dirty_writeback,
 	.error_remove_page = generic_error_remove_page,
-#ifdef CONFIG_NFS_SWAP
 	.swap_activate = nfs_swap_activate,
 	.swap_deactivate = nfs_swap_deactivate,
-#endif
 };
 
 /*

fs/nfs/flexfilelayout/flexfilelayout.c

@@ -20,6 +20,7 @@
 #include "../nfs4trace.h"
 #include "../iostat.h"
 #include "../nfs.h"
+#include "../nfs42.h"
 
 #define NFSDBG_FACILITY         NFSDBG_PNFS_LD
 
@@ -182,17 +183,14 @@ static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls)
 
 static void ff_layout_sort_mirrors(struct nfs4_ff_layout_segment *fls)
 {
-	struct nfs4_ff_layout_mirror *tmp;
 	int i, j;
 
 	for (i = 0; i < fls->mirror_array_cnt - 1; i++) {
 		for (j = i + 1; j < fls->mirror_array_cnt; j++)
 			if (fls->mirror_array[i]->efficiency <
-			    fls->mirror_array[j]->efficiency) {
-				tmp = fls->mirror_array[i];
-				fls->mirror_array[i] = fls->mirror_array[j];
-				fls->mirror_array[j] = tmp;
-			}
+			    fls->mirror_array[j]->efficiency)
+				swap(fls->mirror_array[i],
+				     fls->mirror_array[j]);
 	}
 }
 
@@ -274,6 +272,7 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 
 		spin_lock_init(&fls->mirror_array[i]->lock);
 		fls->mirror_array[i]->ds_count = ds_count;
+		fls->mirror_array[i]->lseg = &fls->generic_hdr;
 
 		/* deviceid */
 		rc = decode_deviceid(&stream, &devid);
@@ -344,6 +343,10 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 			fls->mirror_array[i]->gid);
 	}
 
+	p = xdr_inline_decode(&stream, 4);
+	if (p)
+		fls->flags = be32_to_cpup(p);
+
 	ff_layout_sort_mirrors(fls);
 	rc = ff_layout_check_layout(lgr);
 	if (rc)
@@ -415,6 +418,146 @@ ff_layout_get_lseg_count(struct nfs4_ff_layout_segment *fls)
 	return 1;
 }
 
+static void
+nfs4_ff_start_busy_timer(struct nfs4_ff_busy_timer *timer)
+{
+	/* first IO request? */
+	if (atomic_inc_return(&timer->n_ops) == 1) {
+		timer->start_time = ktime_get();
+	}
+}
+
+static ktime_t
+nfs4_ff_end_busy_timer(struct nfs4_ff_busy_timer *timer)
+{
+	ktime_t start, now;
+
+	if (atomic_dec_return(&timer->n_ops) < 0)
+		WARN_ON_ONCE(1);
+
+	now = ktime_get();
+	start = timer->start_time;
+	timer->start_time = now;
+	return ktime_sub(now, start);
+}
+
+static ktime_t
+nfs4_ff_layout_calc_completion_time(struct rpc_task *task)
+{
+	return ktime_sub(ktime_get(), task->tk_start);
+}
+
+static bool
+nfs4_ff_layoutstat_start_io(struct nfs4_ff_layout_mirror *mirror,
+			    struct nfs4_ff_layoutstat *layoutstat)
+{
+	static const ktime_t notime = {0};
+	ktime_t now = ktime_get();
+
+	nfs4_ff_start_busy_timer(&layoutstat->busy_timer);
+	if (ktime_equal(mirror->start_time, notime))
+		mirror->start_time = now;
+	if (ktime_equal(mirror->last_report_time, notime))
+		mirror->last_report_time = now;
+	if (ktime_to_ms(ktime_sub(now, mirror->last_report_time)) >=
+			FF_LAYOUTSTATS_REPORT_INTERVAL) {
+		mirror->last_report_time = now;
+		return true;
+	}
+
+	return false;
+}
+
+static void
+nfs4_ff_layout_stat_io_update_requested(struct nfs4_ff_layoutstat *layoutstat,
+		__u64 requested)
+{
+	struct nfs4_ff_io_stat *iostat = &layoutstat->io_stat;
+
+	iostat->ops_requested++;
+	iostat->bytes_requested += requested;
+}
+
+static void
+nfs4_ff_layout_stat_io_update_completed(struct nfs4_ff_layoutstat *layoutstat,
+		__u64 requested,
+		__u64 completed,
+		ktime_t time_completed)
+{
+	struct nfs4_ff_io_stat *iostat = &layoutstat->io_stat;
+	ktime_t timer;
+
+	iostat->ops_completed++;
+	iostat->bytes_completed += completed;
+	iostat->bytes_not_delivered += requested - completed;
+
+	timer = nfs4_ff_end_busy_timer(&layoutstat->busy_timer);
+	iostat->total_busy_time =
+			ktime_add(iostat->total_busy_time, timer);
+	iostat->aggregate_completion_time =
+			ktime_add(iostat->aggregate_completion_time, time_completed);
+}
+
+static void
+nfs4_ff_layout_stat_io_start_read(struct nfs4_ff_layout_mirror *mirror,
+		__u64 requested)
+{
+	bool report;
+
+	spin_lock(&mirror->lock);
+	report = nfs4_ff_layoutstat_start_io(mirror, &mirror->read_stat);
+	nfs4_ff_layout_stat_io_update_requested(&mirror->read_stat, requested);
+	spin_unlock(&mirror->lock);
+
+	if (report)
+		pnfs_report_layoutstat(mirror->lseg->pls_layout->plh_inode);
+}
+
+static void
+nfs4_ff_layout_stat_io_end_read(struct rpc_task *task,
+		struct nfs4_ff_layout_mirror *mirror,
+		__u64 requested,
+		__u64 completed)
+{
+	spin_lock(&mirror->lock);
+	nfs4_ff_layout_stat_io_update_completed(&mirror->read_stat,
+			requested, completed,
+			nfs4_ff_layout_calc_completion_time(task));
+	spin_unlock(&mirror->lock);
+}
+
+static void
+nfs4_ff_layout_stat_io_start_write(struct nfs4_ff_layout_mirror *mirror,
+		__u64 requested)
+{
+	bool report;
+
+	spin_lock(&mirror->lock);
+	report = nfs4_ff_layoutstat_start_io(mirror , &mirror->write_stat);
+	nfs4_ff_layout_stat_io_update_requested(&mirror->write_stat, requested);
+	spin_unlock(&mirror->lock);
+
+	if (report)
+		pnfs_report_layoutstat(mirror->lseg->pls_layout->plh_inode);
+}
+
+static void
+nfs4_ff_layout_stat_io_end_write(struct rpc_task *task,
+		struct nfs4_ff_layout_mirror *mirror,
+		__u64 requested,
+		__u64 completed,
+		enum nfs3_stable_how committed)
+{
+	if (committed == NFS_UNSTABLE)
+		requested = completed = 0;
+
+	spin_lock(&mirror->lock);
+	nfs4_ff_layout_stat_io_update_completed(&mirror->write_stat,
+			requested, completed,
+			nfs4_ff_layout_calc_completion_time(task));
+	spin_unlock(&mirror->lock);
+}
+
 static int
 ff_layout_alloc_commit_info(struct pnfs_layout_segment *lseg,
 			    struct nfs_commit_info *cinfo,
@@ -631,7 +774,7 @@ static void ff_layout_reset_write(struct nfs_pgio_header *hdr, bool retry_pnfs)
 			nfs_direct_set_resched_writes(hdr->dreq);
 			/* fake unstable write to let common nfs resend pages */
 			hdr->verf.committed = NFS_UNSTABLE;
-			hdr->good_bytes = 0;
+			hdr->good_bytes = hdr->args.count;
 		}
 		return;
 	}
@@ -879,6 +1022,12 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 	return 0;
 }
 
+static bool
+ff_layout_need_layoutcommit(struct pnfs_layout_segment *lseg)
+{
+	return !(FF_LAYOUT_LSEG(lseg)->flags & FF_FLAGS_NO_LAYOUTCOMMIT);
+}
+
 /*
  * We reference the rpc_cred of the first WRITE that triggers the need for
  * a LAYOUTCOMMIT, and use it to send the layoutcommit compound.
@@ -891,6 +1040,9 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 static void
 ff_layout_set_layoutcommit(struct nfs_pgio_header *hdr)
 {
+	if (!ff_layout_need_layoutcommit(hdr->lseg))
+		return;
+
 	pnfs_set_layoutcommit(hdr->inode, hdr->lseg,
 			hdr->mds_offset + hdr->res.count);
 	dprintk("%s inode %lu pls_end_pos %lu\n", __func__, hdr->inode->i_ino,
@@ -909,6 +1061,10 @@ ff_layout_reset_to_mds(struct pnfs_layout_segment *lseg, int idx)
 static int ff_layout_read_prepare_common(struct rpc_task *task,
 					 struct nfs_pgio_header *hdr)
 {
+	nfs4_ff_layout_stat_io_start_read(
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count);
+
 	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return -EIO;
@@ -962,15 +1118,15 @@ static void ff_layout_read_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
-	if (ff_layout_read_prepare_common(task, hdr))
-		return;
-
 	if (ff_layout_setup_sequence(hdr->ds_clp,
 				     &hdr->args.seq_args,
 				     &hdr->res.seq_res,
 				     task))
 		return;
 
+	if (ff_layout_read_prepare_common(task, hdr))
+		return;
+
 	if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context,
 			hdr->args.lock_context, FMODE_READ) == -EIO)
 		rpc_exit(task, -EIO); /* lost lock, terminate I/O */
@@ -982,6 +1138,10 @@ static void ff_layout_read_call_done(struct rpc_task *task, void *data)
 
 	dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status);
 
+	nfs4_ff_layout_stat_io_end_read(task,
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count, hdr->res.count);
+
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
 		nfs4_sequence_done(task, &hdr->res.seq_res);
@@ -1074,7 +1234,8 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 		return -EAGAIN;
 	}
 
-	if (data->verf.committed == NFS_UNSTABLE)
+	if (data->verf.committed == NFS_UNSTABLE
+	    && ff_layout_need_layoutcommit(data->lseg))
 		pnfs_set_layoutcommit(data->inode, data->lseg, data->lwb);
 
 	return 0;
@@ -1083,6 +1244,10 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 static int ff_layout_write_prepare_common(struct rpc_task *task,
 					  struct nfs_pgio_header *hdr)
 {
+	nfs4_ff_layout_stat_io_start_write(
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count);
+
 	if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) {
 		rpc_exit(task, -EIO);
 		return -EIO;
@@ -1116,15 +1281,15 @@ static void ff_layout_write_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
-	if (ff_layout_write_prepare_common(task, hdr))
-		return;
-
 	if (ff_layout_setup_sequence(hdr->ds_clp,
 				     &hdr->args.seq_args,
 				     &hdr->res.seq_res,
 				     task))
 		return;
 
+	if (ff_layout_write_prepare_common(task, hdr))
+		return;
+
 	if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context,
 			hdr->args.lock_context, FMODE_WRITE) == -EIO)
 		rpc_exit(task, -EIO); /* lost lock, terminate I/O */
@@ -1134,6 +1299,11 @@ static void ff_layout_write_call_done(struct rpc_task *task, void *data)
 {
 	struct nfs_pgio_header *hdr = data;
 
+	nfs4_ff_layout_stat_io_end_write(task,
+			FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx),
+			hdr->args.count, hdr->res.count,
+			hdr->res.verf->committed);
+
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags) &&
 	    task->tk_status == 0) {
 		nfs4_sequence_done(task, &hdr->res.seq_res);
@@ -1152,8 +1322,17 @@ static void ff_layout_write_count_stats(struct rpc_task *task, void *data)
 	    &NFS_CLIENT(hdr->inode)->cl_metrics[NFSPROC4_CLNT_WRITE]);
 }
 
+static void ff_layout_commit_prepare_common(struct rpc_task *task,
+		struct nfs_commit_data *cdata)
+{
+	nfs4_ff_layout_stat_io_start_write(
+			FF_LAYOUT_COMP(cdata->lseg, cdata->ds_commit_index),
+			0);
+}
+
 static void ff_layout_commit_prepare_v3(struct rpc_task *task, void *data)
 {
+	ff_layout_commit_prepare_common(task, data);
 	rpc_call_start(task);
 }
 
@@ -1161,10 +1340,30 @@ static void ff_layout_commit_prepare_v4(struct rpc_task *task, void *data)
 {
 	struct nfs_commit_data *wdata = data;
 
-	ff_layout_setup_sequence(wdata->ds_clp,
+	if (ff_layout_setup_sequence(wdata->ds_clp,
 				 &wdata->args.seq_args,
 				 &wdata->res.seq_res,
-				 task);
+				 task))
+		return;
+	ff_layout_commit_prepare_common(task, data);
+}
+
+static void ff_layout_commit_done(struct rpc_task *task, void *data)
+{
+	struct nfs_commit_data *cdata = data;
+	struct nfs_page *req;
+	__u64 count = 0;
+
+	if (task->tk_status == 0) {
+		list_for_each_entry(req, &cdata->pages, wb_list)
+			count += req->wb_bytes;
+	}
+
+	nfs4_ff_layout_stat_io_end_write(task,
+			FF_LAYOUT_COMP(cdata->lseg, cdata->ds_commit_index),
+			count, count, NFS_FILE_SYNC);
+
+	pnfs_generic_write_commit_done(task, data);
 }
 
 static void ff_layout_commit_count_stats(struct rpc_task *task, void *data)
@@ -1205,14 +1404,14 @@ static const struct rpc_call_ops ff_layout_write_call_ops_v4 = {
 
 static const struct rpc_call_ops ff_layout_commit_call_ops_v3 = {
 	.rpc_call_prepare = ff_layout_commit_prepare_v3,
-	.rpc_call_done = pnfs_generic_write_commit_done,
+	.rpc_call_done = ff_layout_commit_done,
 	.rpc_count_stats = ff_layout_commit_count_stats,
 	.rpc_release = pnfs_generic_commit_release,
 };
 
 static const struct rpc_call_ops ff_layout_commit_call_ops_v4 = {
 	.rpc_call_prepare = ff_layout_commit_prepare_v4,
-	.rpc_call_done = pnfs_generic_write_commit_done,
+	.rpc_call_done = ff_layout_commit_done,
 	.rpc_count_stats = ff_layout_commit_count_stats,
 	.rpc_release = pnfs_generic_commit_release,
 };
@@ -1256,7 +1455,6 @@ ff_layout_read_pagelist(struct nfs_pgio_header *hdr)
 	fh = nfs4_ff_layout_select_ds_fh(lseg, idx);
 	if (fh)
 		hdr->args.fh = fh;
-
 	/*
 	 * Note that if we ever decide to split across DSes,
 	 * then we may need to handle dense-like offsets.
@@ -1385,6 +1583,7 @@ static int ff_layout_initiate_commit(struct nfs_commit_data *data, int how)
 	fh = select_ds_fh_from_commit(lseg, data->ds_commit_index);
 	if (fh)
 		data->args.fh = fh;
+
 	return nfs_initiate_commit(ds_clnt, data, ds->ds_clp->rpc_ops,
 				   vers == 3 ? &ff_layout_commit_call_ops_v3 :
 					       &ff_layout_commit_call_ops_v4,
@@ -1488,6 +1687,247 @@ out:
 	dprintk("%s: Return\n", __func__);
 }
 
+static int
+ff_layout_ntop4(const struct sockaddr *sap, char *buf, const size_t buflen)
+{
+	const struct sockaddr_in *sin = (struct sockaddr_in *)sap;
+
+	return snprintf(buf, buflen, "%pI4", &sin->sin_addr);
+}
+
+static size_t
+ff_layout_ntop6_noscopeid(const struct sockaddr *sap, char *buf,
+			  const int buflen)
+{
+	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
+	const struct in6_addr *addr = &sin6->sin6_addr;
+
+	/*
+	 * RFC 4291, Section 2.2.2
+	 *
+	 * Shorthanded ANY address
+	 */
+	if (ipv6_addr_any(addr))
+		return snprintf(buf, buflen, "::");
+
+	/*
+	 * RFC 4291, Section 2.2.2
+	 *
+	 * Shorthanded loopback address
+	 */
+	if (ipv6_addr_loopback(addr))
+		return snprintf(buf, buflen, "::1");
+
+	/*
+	 * RFC 4291, Section 2.2.3
+	 *
+	 * Special presentation address format for mapped v4
+	 * addresses.
+	 */
+	if (ipv6_addr_v4mapped(addr))
+		return snprintf(buf, buflen, "::ffff:%pI4",
+					&addr->s6_addr32[3]);
+
+	/*
+	 * RFC 4291, Section 2.2.1
+	 */
+	return snprintf(buf, buflen, "%pI6c", addr);
+}
+
+/* Derived from rpc_sockaddr2uaddr */
+static void
+ff_layout_encode_netaddr(struct xdr_stream *xdr, struct nfs4_pnfs_ds_addr *da)
+{
+	struct sockaddr *sap = (struct sockaddr *)&da->da_addr;
+	char portbuf[RPCBIND_MAXUADDRPLEN];
+	char addrbuf[RPCBIND_MAXUADDRLEN];
+	char *netid;
+	unsigned short port;
+	int len, netid_len;
+	__be32 *p;
+
+	switch (sap->sa_family) {
+	case AF_INET:
+		if (ff_layout_ntop4(sap, addrbuf, sizeof(addrbuf)) == 0)
+			return;
+		port = ntohs(((struct sockaddr_in *)sap)->sin_port);
+		netid = "tcp";
+		netid_len = 3;
+		break;
+	case AF_INET6:
+		if (ff_layout_ntop6_noscopeid(sap, addrbuf, sizeof(addrbuf)) == 0)
+			return;
+		port = ntohs(((struct sockaddr_in6 *)sap)->sin6_port);
+		netid = "tcp6";
+		netid_len = 4;
+		break;
+	default:
+		/* we only support tcp and tcp6 */
+		WARN_ON_ONCE(1);
+		return;
+	}
+
+	snprintf(portbuf, sizeof(portbuf), ".%u.%u", port >> 8, port & 0xff);
+	len = strlcat(addrbuf, portbuf, sizeof(addrbuf));
+
+	p = xdr_reserve_space(xdr, 4 + netid_len);
+	xdr_encode_opaque(p, netid, netid_len);
+
+	p = xdr_reserve_space(xdr, 4 + len);
+	xdr_encode_opaque(p, addrbuf, len);
+}
+
+static void
+ff_layout_encode_nfstime(struct xdr_stream *xdr,
+			 ktime_t t)
+{
+	struct timespec64 ts;
+	__be32 *p;
+
+	p = xdr_reserve_space(xdr, 12);
+	ts = ktime_to_timespec64(t);
+	p = xdr_encode_hyper(p, ts.tv_sec);
+	*p++ = cpu_to_be32(ts.tv_nsec);
+}
+
+static void
+ff_layout_encode_io_latency(struct xdr_stream *xdr,
+			    struct nfs4_ff_io_stat *stat)
+{
+	__be32 *p;
+
+	p = xdr_reserve_space(xdr, 5 * 8);
+	p = xdr_encode_hyper(p, stat->ops_requested);
+	p = xdr_encode_hyper(p, stat->bytes_requested);
+	p = xdr_encode_hyper(p, stat->ops_completed);
+	p = xdr_encode_hyper(p, stat->bytes_completed);
+	p = xdr_encode_hyper(p, stat->bytes_not_delivered);
+	ff_layout_encode_nfstime(xdr, stat->total_busy_time);
+	ff_layout_encode_nfstime(xdr, stat->aggregate_completion_time);
+}
+
+static void
+ff_layout_encode_layoutstats(struct xdr_stream *xdr,
+			     struct nfs42_layoutstat_args *args,
+			     struct nfs42_layoutstat_devinfo *devinfo)
+{
+	struct nfs4_ff_layout_mirror *mirror = devinfo->layout_private;
+	struct nfs4_pnfs_ds_addr *da;
+	struct nfs4_pnfs_ds *ds = mirror->mirror_ds->ds;
+	struct nfs_fh *fh = &mirror->fh_versions[0];
+	__be32 *p, *start;
+
+	da = list_first_entry(&ds->ds_addrs, struct nfs4_pnfs_ds_addr, da_node);
+	dprintk("%s: DS %s: encoding address %s\n",
+		__func__, ds->ds_remotestr, da->da_remotestr);
+	/* layoutupdate length */
+	start = xdr_reserve_space(xdr, 4);
+	/* netaddr4 */
+	ff_layout_encode_netaddr(xdr, da);
+	/* nfs_fh4 */
+	p = xdr_reserve_space(xdr, 4 + fh->size);
+	xdr_encode_opaque(p, fh->data, fh->size);
+	/* ff_io_latency4 read */
+	spin_lock(&mirror->lock);
+	ff_layout_encode_io_latency(xdr, &mirror->read_stat.io_stat);
+	/* ff_io_latency4 write */
+	ff_layout_encode_io_latency(xdr, &mirror->write_stat.io_stat);
+	spin_unlock(&mirror->lock);
+	/* nfstime4 */
+	ff_layout_encode_nfstime(xdr, ktime_sub(ktime_get(), mirror->start_time));
+	/* bool */
+	p = xdr_reserve_space(xdr, 4);
+	*p = cpu_to_be32(false);
+
+	*start = cpu_to_be32((xdr->p - start - 1) * 4);
+}
+
+static bool
+ff_layout_mirror_prepare_stats(struct nfs42_layoutstat_args *args,
+			       struct pnfs_layout_segment *pls,
+			       int *dev_count, int dev_limit)
+{
+	struct nfs4_ff_layout_mirror *mirror;
+	struct nfs4_deviceid_node *dev;
+	struct nfs42_layoutstat_devinfo *devinfo;
+	int i;
+
+	for (i = 0; i <= FF_LAYOUT_MIRROR_COUNT(pls); i++) {
+		if (*dev_count >= dev_limit)
+			break;
+		mirror = FF_LAYOUT_COMP(pls, i);
+		if (!mirror || !mirror->mirror_ds)
+			continue;
+		dev = FF_LAYOUT_DEVID_NODE(pls, i);
+		devinfo = &args->devinfo[*dev_count];
+		memcpy(&devinfo->dev_id, &dev->deviceid, NFS4_DEVICEID4_SIZE);
+		devinfo->offset = pls->pls_range.offset;
+		devinfo->length = pls->pls_range.length;
+		/* well, we don't really know if IO is continuous or not! */
+		devinfo->read_count = mirror->read_stat.io_stat.bytes_completed;
+		devinfo->read_bytes = mirror->read_stat.io_stat.bytes_completed;
+		devinfo->write_count = mirror->write_stat.io_stat.bytes_completed;
+		devinfo->write_bytes = mirror->write_stat.io_stat.bytes_completed;
+		devinfo->layout_type = LAYOUT_FLEX_FILES;
+		devinfo->layoutstats_encode = ff_layout_encode_layoutstats;
+		devinfo->layout_private = mirror;
+		/* lseg refcount put in cleanup_layoutstats */
+		pnfs_get_lseg(pls);
+
+		++(*dev_count);
+	}
+
+	return *dev_count < dev_limit;
+}
+
+static int
+ff_layout_prepare_layoutstats(struct nfs42_layoutstat_args *args)
+{
+	struct pnfs_layout_segment *pls;
+	int dev_count = 0;
+
+	spin_lock(&args->inode->i_lock);
+	list_for_each_entry(pls, &NFS_I(args->inode)->layout->plh_segs, pls_list) {
+		dev_count += FF_LAYOUT_MIRROR_COUNT(pls);
+	}
+	spin_unlock(&args->inode->i_lock);
+	/* For now, send at most PNFS_LAYOUTSTATS_MAXDEV statistics */
+	if (dev_count > PNFS_LAYOUTSTATS_MAXDEV) {
+		dprintk("%s: truncating devinfo to limit (%d:%d)\n",
+			__func__, dev_count, PNFS_LAYOUTSTATS_MAXDEV);
+		dev_count = PNFS_LAYOUTSTATS_MAXDEV;
+	}
+	args->devinfo = kmalloc(dev_count * sizeof(*args->devinfo), GFP_KERNEL);
+	if (!args->devinfo)
+		return -ENOMEM;
+
+	dev_count = 0;
+	spin_lock(&args->inode->i_lock);
+	list_for_each_entry(pls, &NFS_I(args->inode)->layout->plh_segs, pls_list) {
+		if (!ff_layout_mirror_prepare_stats(args, pls, &dev_count,
+						    PNFS_LAYOUTSTATS_MAXDEV)) {
+			break;
+		}
+	}
+	spin_unlock(&args->inode->i_lock);
+	args->num_dev = dev_count;
+
+	return 0;
+}
+
+static void
+ff_layout_cleanup_layoutstats(struct nfs42_layoutstat_data *data)
+{
+	struct nfs4_ff_layout_mirror *mirror;
+	int i;
+
+	for (i = 0; i < data->args.num_dev; i++) {
+		mirror = data->args.devinfo[i].layout_private;
+		data->args.devinfo[i].layout_private = NULL;
+		pnfs_put_lseg(mirror->lseg);
+	}
+}
+
 static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.id			= LAYOUT_FLEX_FILES,
 	.name			= "LAYOUT_FLEX_FILES",
@@ -1510,6 +1950,8 @@ static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.alloc_deviceid_node    = ff_layout_alloc_deviceid_node,
 	.encode_layoutreturn    = ff_layout_encode_layoutreturn,
 	.sync			= pnfs_nfs_generic_sync,
+	.prepare_layoutstats	= ff_layout_prepare_layoutstats,
+	.cleanup_layoutstats	= ff_layout_cleanup_layoutstats,
 };
 
 static int __init nfs4flexfilelayout_init(void)

fs/nfs/flexfilelayout/flexfilelayout.h

@@ -9,12 +9,17 @@
 #ifndef FS_NFS_NFS4FLEXFILELAYOUT_H
 #define FS_NFS_NFS4FLEXFILELAYOUT_H
 
+#define FF_FLAGS_NO_LAYOUTCOMMIT 1
+
 #include "../pnfs.h"
 
 /* XXX: Let's filter out insanely large mirror count for now to avoid oom
  * due to network error etc. */
 #define NFS4_FLEXFILE_LAYOUT_MAX_MIRROR_CNT 4096
 
+/* LAYOUTSTATS report interval in ms */
+#define FF_LAYOUTSTATS_REPORT_INTERVAL (60000L)
+
 struct nfs4_ff_ds_version {
 	u32				version;
 	u32				minor_version;
@@ -41,24 +46,48 @@ struct nfs4_ff_layout_ds_err {
 	struct nfs4_deviceid		deviceid;
 };
 
+struct nfs4_ff_io_stat {
+	__u64				ops_requested;
+	__u64				bytes_requested;
+	__u64				ops_completed;
+	__u64				bytes_completed;
+	__u64				bytes_not_delivered;
+	ktime_t				total_busy_time;
+	ktime_t				aggregate_completion_time;
+};
+
+struct nfs4_ff_busy_timer {
+	ktime_t start_time;
+	atomic_t n_ops;
+};
+
+struct nfs4_ff_layoutstat {
+	struct nfs4_ff_io_stat io_stat;
+	struct nfs4_ff_busy_timer busy_timer;
+};
+
 struct nfs4_ff_layout_mirror {
+	struct pnfs_layout_segment	*lseg; /* back pointer */
 	u32				ds_count;
 	u32				efficiency;
 	struct nfs4_ff_layout_ds	*mirror_ds;
 	u32				fh_versions_cnt;
 	struct nfs_fh			*fh_versions;
 	nfs4_stateid			stateid;
-	struct nfs4_string		user_name;
-	struct nfs4_string		group_name;
 	u32				uid;
 	u32				gid;
 	struct rpc_cred			*cred;
 	spinlock_t			lock;
+	struct nfs4_ff_layoutstat	read_stat;
+	struct nfs4_ff_layoutstat	write_stat;
+	ktime_t				start_time;
+	ktime_t				last_report_time;
 };
 
 struct nfs4_ff_layout_segment {
 	struct pnfs_layout_segment	generic_hdr;
 	u64				stripe_unit;
+	u32				flags;
 	u32				mirror_array_cnt;
 	struct nfs4_ff_layout_mirror	**mirror_array;
 };

fs/nfs/flexfilelayout/flexfilelayoutdev.c

@@ -324,7 +324,8 @@ static int ff_layout_update_mirror_cred(struct nfs4_ff_layout_mirror *mirror,
 				__func__, PTR_ERR(cred));
 			return PTR_ERR(cred);
 		} else {
-			mirror->cred = cred;
+			if (cmpxchg(&mirror->cred, NULL, cred))
+				put_rpccred(cred);
 		}
 	}
 	return 0;
@@ -386,7 +387,7 @@ nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx,
 	/* matching smp_wmb() in _nfs4_pnfs_v3/4_ds_connect */
 	smp_rmb();
 	if (ds->ds_clp)
-		goto out;
+		goto out_update_creds;
 
 	flavor = nfs4_ff_layout_choose_authflavor(mirror);
 
@@ -430,7 +431,7 @@ nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx,
 			}
 		}
 	}
-
+out_update_creds:
 	if (ff_layout_update_mirror_cred(mirror, ds))
 		ds = NULL;
 out:

fs/nfs/inode.c

@@ -678,6 +678,8 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
 	if (!err) {
 		generic_fillattr(inode, stat);
 		stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode));
+		if (S_ISDIR(inode->i_mode))
+			stat->blksize = NFS_SERVER(inode)->dtsize;
 	}
 out:
 	trace_nfs_getattr_exit(inode, err);
@@ -2008,17 +2010,15 @@ static int __init init_nfs_fs(void)
 	if (err)
 		goto out1;
 
-#ifdef CONFIG_PROC_FS
 	rpc_proc_register(&init_net, &nfs_rpcstat);
-#endif
-	if ((err = register_nfs_fs()) != 0)
+
+	err = register_nfs_fs();
+	if (err)
 		goto out0;
 
 	return 0;
 out0:
-#ifdef CONFIG_PROC_FS
 	rpc_proc_unregister(&init_net, "nfs");
-#endif
 	nfs_destroy_directcache();
 out1:
 	nfs_destroy_writepagecache();
@@ -2049,9 +2049,7 @@ static void __exit exit_nfs_fs(void)
 	nfs_destroy_nfspagecache();
 	nfs_fscache_unregister();
 	unregister_pernet_subsys(&nfs_net_ops);
-#ifdef CONFIG_PROC_FS
 	rpc_proc_unregister(&init_net, "nfs");
-#endif
 	unregister_nfs_fs();
 	nfs_fs_proc_exit();
 	nfsiod_stop();

fs/nfs/nfs3xdr.c

@@ -1342,7 +1342,7 @@ static void nfs3_xdr_enc_setacl3args(struct rpc_rqst *req,
 	if (args->npages != 0)
 		xdr_write_pages(xdr, args->pages, 0, args->len);
 	else
-		xdr_reserve_space(xdr, NFS_ACL_INLINE_BUFSIZE);
+		xdr_reserve_space(xdr, args->len);
 
 	error = nfsacl_encode(xdr->buf, base, args->inode,
 			    (args->mask & NFS_ACL) ?

fs/nfs/nfs42.h

@@ -5,11 +5,18 @@
 #ifndef __LINUX_FS_NFS_NFS4_2_H
 #define __LINUX_FS_NFS_NFS4_2_H
 
+/*
+ * FIXME:  four LAYOUTSTATS calls per compound at most! Do we need to support
+ * more? Need to consider not to pre-alloc too much for a compound.
+ */
+#define PNFS_LAYOUTSTATS_MAXDEV (4)
+
 /* nfs4.2proc.c */
 int nfs42_proc_allocate(struct file *, loff_t, loff_t);
 int nfs42_proc_deallocate(struct file *, loff_t, loff_t);
 loff_t nfs42_proc_llseek(struct file *, loff_t, int);
-
+int nfs42_proc_layoutstats_generic(struct nfs_server *,
+				   struct nfs42_layoutstat_data *);
 /* nfs4.2xdr.h */
 extern struct rpc_procinfo nfs4_2_procedures[];
 

fs/nfs/nfs42proc.c

@@ -10,6 +10,11 @@
 #include <linux/nfs_fs.h>
 #include "nfs4_fs.h"
 #include "nfs42.h"
+#include "iostat.h"
+#include "pnfs.h"
+#include "internal.h"
+
+#define NFSDBG_FACILITY NFSDBG_PNFS
 
 static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
 				fmode_t fmode)
@@ -165,3 +170,85 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 
 	return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes);
 }
+
+static void
+nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs42_layoutstat_data *data = calldata;
+	struct nfs_server *server = NFS_SERVER(data->args.inode);
+
+	nfs41_setup_sequence(nfs4_get_session(server), &data->args.seq_args,
+			     &data->res.seq_res, task);
+}
+
+static void
+nfs42_layoutstat_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs42_layoutstat_data *data = calldata;
+
+	if (!nfs4_sequence_done(task, &data->res.seq_res))
+		return;
+
+	switch (task->tk_status) {
+	case 0:
+		break;
+	case -ENOTSUPP:
+	case -EOPNOTSUPP:
+		NFS_SERVER(data->inode)->caps &= ~NFS_CAP_LAYOUTSTATS;
+	default:
+		dprintk("%s server returns %d\n", __func__, task->tk_status);
+	}
+}
+
+static void
+nfs42_layoutstat_release(void *calldata)
+{
+	struct nfs42_layoutstat_data *data = calldata;
+	struct nfs_server *nfss = NFS_SERVER(data->args.inode);
+
+	if (nfss->pnfs_curr_ld->cleanup_layoutstats)
+		nfss->pnfs_curr_ld->cleanup_layoutstats(data);
+
+	pnfs_put_layout_hdr(NFS_I(data->args.inode)->layout);
+	smp_mb__before_atomic();
+	clear_bit(NFS_INO_LAYOUTSTATS, &NFS_I(data->args.inode)->flags);
+	smp_mb__after_atomic();
+	nfs_iput_and_deactive(data->inode);
+	kfree(data->args.devinfo);
+	kfree(data);
+}
+
+static const struct rpc_call_ops nfs42_layoutstat_ops = {
+	.rpc_call_prepare = nfs42_layoutstat_prepare,
+	.rpc_call_done = nfs42_layoutstat_done,
+	.rpc_release = nfs42_layoutstat_release,
+};
+
+int nfs42_proc_layoutstats_generic(struct nfs_server *server,
+				   struct nfs42_layoutstat_data *data)
+{
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTSTATS],
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+	};
+	struct rpc_task_setup task_setup = {
+		.rpc_client = server->client,
+		.rpc_message = &msg,
+		.callback_ops = &nfs42_layoutstat_ops,
+		.callback_data = data,
+		.flags = RPC_TASK_ASYNC,
+	};
+	struct rpc_task *task;
+
+	data->inode = nfs_igrab_and_active(data->args.inode);
+	if (!data->inode) {
+		nfs42_layoutstat_release(data);
+		return -EAGAIN;
+	}
+	nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0);
+	task = rpc_run_task(&task_setup);
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+	return 0;
+}

fs/nfs/nfs42xdr.c

@@ -4,6 +4,8 @@
 #ifndef __LINUX_FS_NFS_NFS4_2XDR_H
 #define __LINUX_FS_NFS_NFS4_2XDR_H
 
+#include "nfs42.h"
+
 #define encode_fallocate_maxsz		(encode_stateid_maxsz + \
 					 2 /* offset */ + \
 					 2 /* length */)
@@ -22,6 +24,16 @@
 					 1 /* whence */ + \
 					 2 /* offset */ + \
 					 2 /* length */)
+#define encode_io_info_maxsz		4
+#define encode_layoutstats_maxsz	(op_decode_hdr_maxsz + \
+					2 /* offset */ + \
+					2 /* length */ + \
+					encode_stateid_maxsz + \
+					encode_io_info_maxsz + \
+					encode_io_info_maxsz + \
+					1 /* opaque devaddr4 length */ + \
+					XDR_QUADLEN(PNFS_LAYOUTSTATS_MAXSIZE))
+#define decode_layoutstats_maxsz	(op_decode_hdr_maxsz)
 
 #define NFS4_enc_allocate_sz		(compound_encode_hdr_maxsz + \
 					 encode_putfh_maxsz + \
@@ -45,6 +57,14 @@
 #define NFS4_dec_seek_sz		(compound_decode_hdr_maxsz + \
 					 decode_putfh_maxsz + \
 					 decode_seek_maxsz)
+#define NFS4_enc_layoutstats_sz		(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_putfh_maxsz + \
+					 PNFS_LAYOUTSTATS_MAXDEV * encode_layoutstats_maxsz)
+#define NFS4_dec_layoutstats_sz		(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_putfh_maxsz + \
+					 PNFS_LAYOUTSTATS_MAXDEV * decode_layoutstats_maxsz)
 
 
 static void encode_fallocate(struct xdr_stream *xdr,
@@ -81,6 +101,33 @@ static void encode_seek(struct xdr_stream *xdr,
 	encode_uint32(xdr, args->sa_what);
 }
 
+static void encode_layoutstats(struct xdr_stream *xdr,
+			       struct nfs42_layoutstat_args *args,
+			       struct nfs42_layoutstat_devinfo *devinfo,
+			       struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	encode_op_hdr(xdr, OP_LAYOUTSTATS, decode_layoutstats_maxsz, hdr);
+	p = reserve_space(xdr, 8 + 8);
+	p = xdr_encode_hyper(p, devinfo->offset);
+	p = xdr_encode_hyper(p, devinfo->length);
+	encode_nfs4_stateid(xdr, &args->stateid);
+	p = reserve_space(xdr, 4*8 + NFS4_DEVICEID4_SIZE + 4);
+	p = xdr_encode_hyper(p, devinfo->read_count);
+	p = xdr_encode_hyper(p, devinfo->read_bytes);
+	p = xdr_encode_hyper(p, devinfo->write_count);
+	p = xdr_encode_hyper(p, devinfo->write_bytes);
+	p = xdr_encode_opaque_fixed(p, devinfo->dev_id.data,
+			NFS4_DEVICEID4_SIZE);
+	/* Encode layoutupdate4 */
+	*p++ = cpu_to_be32(devinfo->layout_type);
+	if (devinfo->layoutstats_encode != NULL)
+		devinfo->layoutstats_encode(xdr, args, devinfo);
+	else
+		encode_uint32(xdr, 0);
+}
+
 /*
  * Encode ALLOCATE request
  */
@@ -137,6 +184,28 @@ static void nfs4_xdr_enc_seek(struct rpc_rqst *req,
 	encode_nops(&hdr);
 }
 
+/*
+ * Encode LAYOUTSTATS request
+ */
+static void nfs4_xdr_enc_layoutstats(struct rpc_rqst *req,
+				     struct xdr_stream *xdr,
+				     struct nfs42_layoutstat_args *args)
+{
+	int i;
+
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_putfh(xdr, args->fh, &hdr);
+	WARN_ON(args->num_dev > PNFS_LAYOUTSTATS_MAXDEV);
+	for (i = 0; i < args->num_dev; i++)
+		encode_layoutstats(xdr, args, &args->devinfo[i], &hdr);
+	encode_nops(&hdr);
+}
+
 static int decode_allocate(struct xdr_stream *xdr, struct nfs42_falloc_res *res)
 {
 	return decode_op_hdr(xdr, OP_ALLOCATE);
@@ -169,6 +238,12 @@ out_overflow:
 	return -EIO;
 }
 
+static int decode_layoutstats(struct xdr_stream *xdr,
+			      struct nfs42_layoutstat_res *res)
+{
+	return decode_op_hdr(xdr, OP_LAYOUTSTATS);
+}
+
 /*
  * Decode ALLOCATE request
  */
@@ -246,4 +321,35 @@ static int nfs4_xdr_dec_seek(struct rpc_rqst *rqstp,
 out:
 	return status;
 }
+
+/*
+ * Decode LAYOUTSTATS request
+ */
+static int nfs4_xdr_dec_layoutstats(struct rpc_rqst *rqstp,
+				    struct xdr_stream *xdr,
+				    struct nfs42_layoutstat_res *res)
+{
+	struct compound_hdr hdr;
+	int status, i;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_putfh(xdr);
+	if (status)
+		goto out;
+	WARN_ON(res->num_dev > PNFS_LAYOUTSTATS_MAXDEV);
+	for (i = 0; i < res->num_dev; i++) {
+		status = decode_layoutstats(xdr, res);
+		if (status)
+			goto out;
+	}
+out:
+	res->rpc_status = status;
+	return status;
+}
+
 #endif /* __LINUX_FS_NFS_NFS4_2XDR_H */

fs/nfs/nfs4_fs.h

@@ -233,6 +233,7 @@ extern int nfs4_handle_exception(struct nfs_server *, int, struct nfs4_exception
 extern int nfs4_call_sync(struct rpc_clnt *, struct nfs_server *,
 			  struct rpc_message *, struct nfs4_sequence_args *,
 			  struct nfs4_sequence_res *, int);
+extern void nfs4_init_sequence(struct nfs4_sequence_args *, struct nfs4_sequence_res *, int);
 extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *);
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *);
 extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *, bool);

fs/nfs/nfs4client.c

@@ -676,7 +676,6 @@ found:
 		break;
 	}
 
-	/* No matching nfs_client found. */
 	spin_unlock(&nn->nfs_client_lock);
 	dprintk("NFS: <-- %s status = %d\n", __func__, status);
 	nfs_put_client(prev);

fs/nfs/nfs4file.c

@@ -41,6 +41,10 @@ nfs4_file_open(struct inode *inode, struct file *filp)
 
 	dprintk("NFS: open file(%pd2)\n", dentry);
 
+	err = nfs_check_flags(openflags);
+	if (err)
+		return err;
+
 	if ((openflags & O_ACCMODE) == 3)
 		openflags--;
 

fs/nfs/nfs4getroot.c

@@ -35,13 +35,6 @@ int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_p
 		goto out;
 	}
 
-	if (fsinfo.fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
-		printk(KERN_ERR "nfs4_get_rootfh:"
-		       " getroot obtained referral\n");
-		ret = -EREMOTE;
-		goto out;
-	}
-
 	memcpy(&server->fsid, &fsinfo.fattr->fsid, sizeof(server->fsid));
 out:
 	nfs_free_fattr(fsinfo.fattr);

fs/nfs/nfs4idmap.c

@@ -494,12 +494,7 @@ nfs_idmap_delete(struct nfs_client *clp)
 
 int nfs_idmap_init(void)
 {
-	int ret;
-	ret = nfs_idmap_init_keyring();
-	if (ret != 0)
-		goto out;
-out:
-	return ret;
+	return nfs_idmap_init_keyring();
 }
 
 void nfs_idmap_quit(void)

fs/nfs/nfs4proc.c

@@ -356,6 +356,9 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_
 		case 0:
 			return 0;
 		case -NFS4ERR_OPENMODE:
+		case -NFS4ERR_DELEG_REVOKED:
+		case -NFS4ERR_ADMIN_REVOKED:
+		case -NFS4ERR_BAD_STATEID:
 			if (inode && nfs4_have_delegation(inode, FMODE_READ)) {
 				nfs4_inode_return_delegation(inode);
 				exception->retry = 1;
@@ -367,15 +370,6 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_
 			if (ret < 0)
 				break;
 			goto wait_on_recovery;
-		case -NFS4ERR_DELEG_REVOKED:
-		case -NFS4ERR_ADMIN_REVOKED:
-		case -NFS4ERR_BAD_STATEID:
-			if (state == NULL)
-				break;
-			ret = nfs4_schedule_stateid_recovery(server, state);
-			if (ret < 0)
-				break;
-			goto wait_on_recovery;
 		case -NFS4ERR_EXPIRED:
 			if (state != NULL) {
 				ret = nfs4_schedule_stateid_recovery(server, state);
@@ -482,8 +476,8 @@ struct nfs4_call_sync_data {
 	struct nfs4_sequence_res *seq_res;
 };
 
-static void nfs4_init_sequence(struct nfs4_sequence_args *args,
-			       struct nfs4_sequence_res *res, int cache_reply)
+void nfs4_init_sequence(struct nfs4_sequence_args *args,
+			struct nfs4_sequence_res *res, int cache_reply)
 {
 	args->sa_slot = NULL;
 	args->sa_cache_this = cache_reply;
@@ -1553,6 +1547,13 @@ static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, fmode_t fmod
 	struct nfs4_state *newstate;
 	int ret;
 
+	if ((opendata->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR ||
+	     opendata->o_arg.claim == NFS4_OPEN_CLAIM_DELEG_CUR_FH) &&
+	    (opendata->o_arg.u.delegation_type & fmode) != fmode)
+		/* This mode can't have been delegated, so we must have
+		 * a valid open_stateid to cover it - not need to reclaim.
+		 */
+		return 0;
 	opendata->o_arg.open_flags = 0;
 	opendata->o_arg.fmode = fmode;
 	opendata->o_arg.share_access = nfs4_map_atomic_open_share(
@@ -1684,6 +1685,7 @@ static int nfs4_handle_delegation_recall_error(struct nfs_server *server, struct
 					"%d.\n", __func__, err);
 		case 0:
 		case -ENOENT:
+		case -EAGAIN:
 		case -ESTALE:
 			break;
 		case -NFS4ERR_BADSESSION:
@@ -3355,6 +3357,8 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir,
 			goto out;
 		case -NFS4ERR_MOVED:
 			err = nfs4_get_referral(client, dir, name, fattr, fhandle);
+			if (err == -NFS4ERR_MOVED)
+				err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception);
 			goto out;
 		case -NFS4ERR_WRONGSEC:
 			err = -EPERM;
@@ -4955,49 +4959,128 @@ static void nfs4_init_boot_verifier(const struct nfs_client *clp,
 	memcpy(bootverf->data, verf, sizeof(bootverf->data));
 }
 
-static unsigned int
-nfs4_init_nonuniform_client_string(struct nfs_client *clp,
-				   char *buf, size_t len)
+static int
+nfs4_init_nonuniform_client_string(struct nfs_client *clp)
 {
-	unsigned int result;
+	int result;
+	size_t len;
+	char *str;
+	bool retried = false;
 
 	if (clp->cl_owner_id != NULL)
-		return strlcpy(buf, clp->cl_owner_id, len);
+		return 0;
+retry:
+	rcu_read_lock();
+	len = 10 + strlen(clp->cl_ipaddr) + 1 +
+		strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)) +
+		1 +
+		strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_PROTO)) +
+		1;
+	rcu_read_unlock();
+
+	if (len > NFS4_OPAQUE_LIMIT + 1)
+		return -EINVAL;
+
+	/*
+	 * Since this string is allocated at mount time, and held until the
+	 * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying
+	 * about a memory-reclaim deadlock.
+	 */
+	str = kmalloc(len, GFP_KERNEL);
+	if (!str)
+		return -ENOMEM;
 
 	rcu_read_lock();
-	result = scnprintf(buf, len, "Linux NFSv4.0 %s/%s %s",
-				clp->cl_ipaddr,
-				rpc_peeraddr2str(clp->cl_rpcclient,
-							RPC_DISPLAY_ADDR),
-				rpc_peeraddr2str(clp->cl_rpcclient,
-							RPC_DISPLAY_PROTO));
+	result = scnprintf(str, len, "Linux NFSv4.0 %s/%s %s",
+			clp->cl_ipaddr,
+			rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR),
+			rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_PROTO));
 	rcu_read_unlock();
-	clp->cl_owner_id = kstrdup(buf, GFP_KERNEL);
-	return result;
+
+	/* Did something change? */
+	if (result >= len) {
+		kfree(str);
+		if (retried)
+			return -EINVAL;
+		retried = true;
+		goto retry;
+	}
+	clp->cl_owner_id = str;
+	return 0;
 }
 
-static unsigned int
-nfs4_init_uniform_client_string(struct nfs_client *clp,
-				char *buf, size_t len)
+static int
+nfs4_init_uniquifier_client_string(struct nfs_client *clp)
 {
-	const char *nodename = clp->cl_rpcclient->cl_nodename;
-	unsigned int result;
+	int result;
+	size_t len;
+	char *str;
+
+	len = 10 + 10 + 1 + 10 + 1 +
+		strlen(nfs4_client_id_uniquifier) + 1 +
+		strlen(clp->cl_rpcclient->cl_nodename) + 1;
+
+	if (len > NFS4_OPAQUE_LIMIT + 1)
+		return -EINVAL;
+
+	/*
+	 * Since this string is allocated at mount time, and held until the
+	 * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying
+	 * about a memory-reclaim deadlock.
+	 */
+	str = kmalloc(len, GFP_KERNEL);
+	if (!str)
+		return -ENOMEM;
+
+	result = scnprintf(str, len, "Linux NFSv%u.%u %s/%s",
+			clp->rpc_ops->version, clp->cl_minorversion,
+			nfs4_client_id_uniquifier,
+			clp->cl_rpcclient->cl_nodename);
+	if (result >= len) {
+		kfree(str);
+		return -EINVAL;
+	}
+	clp->cl_owner_id = str;
+	return 0;
+}
+
+static int
+nfs4_init_uniform_client_string(struct nfs_client *clp)
+{
+	int result;
+	size_t len;
+	char *str;
 
 	if (clp->cl_owner_id != NULL)
-		return strlcpy(buf, clp->cl_owner_id, len);
+		return 0;
 
 	if (nfs4_client_id_uniquifier[0] != '\0')
-		result = scnprintf(buf, len, "Linux NFSv%u.%u %s/%s",
-				clp->rpc_ops->version,
-				clp->cl_minorversion,
-				nfs4_client_id_uniquifier,
-				nodename);
-	else
-		result = scnprintf(buf, len, "Linux NFSv%u.%u %s",
-				clp->rpc_ops->version, clp->cl_minorversion,
-				nodename);
-	clp->cl_owner_id = kstrdup(buf, GFP_KERNEL);
-	return result;
+		return nfs4_init_uniquifier_client_string(clp);
+
+	len = 10 + 10 + 1 + 10 + 1 +
+		strlen(clp->cl_rpcclient->cl_nodename) + 1;
+
+	if (len > NFS4_OPAQUE_LIMIT + 1)
+		return -EINVAL;
+
+	/*
+	 * Since this string is allocated at mount time, and held until the
+	 * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying
+	 * about a memory-reclaim deadlock.
+	 */
+	str = kmalloc(len, GFP_KERNEL);
+	if (!str)
+		return -ENOMEM;
+
+	result = scnprintf(str, len, "Linux NFSv%u.%u %s",
+			clp->rpc_ops->version, clp->cl_minorversion,
+			clp->cl_rpcclient->cl_nodename);
+	if (result >= len) {
+		kfree(str);
+		return -EINVAL;
+	}
+	clp->cl_owner_id = str;
+	return 0;
 }
 
 /*
@@ -5044,7 +5127,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 	struct nfs4_setclientid setclientid = {
 		.sc_verifier = &sc_verifier,
 		.sc_prog = program,
-		.sc_cb_ident = clp->cl_cb_ident,
+		.sc_clnt = clp,
 	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID],
@@ -5064,16 +5147,15 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 
 	/* nfs_client_id4 */
 	nfs4_init_boot_verifier(clp, &sc_verifier);
+
 	if (test_bit(NFS_CS_MIGRATION, &clp->cl_flags))
-		setclientid.sc_name_len =
-				nfs4_init_uniform_client_string(clp,
-						setclientid.sc_name,
-						sizeof(setclientid.sc_name));
+		status = nfs4_init_uniform_client_string(clp);
 	else
-		setclientid.sc_name_len =
-				nfs4_init_nonuniform_client_string(clp,
-						setclientid.sc_name,
-						sizeof(setclientid.sc_name));
+		status = nfs4_init_nonuniform_client_string(clp);
+
+	if (status)
+		goto out;
+
 	/* cb_client4 */
 	setclientid.sc_netid_len =
 				nfs4_init_callback_netid(clp,
@@ -5083,9 +5165,9 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 				sizeof(setclientid.sc_uaddr), "%s.%u.%u",
 				clp->cl_ipaddr, port >> 8, port & 255);
 
-	dprintk("NFS call  setclientid auth=%s, '%.*s'\n",
+	dprintk("NFS call  setclientid auth=%s, '%s'\n",
 		clp->cl_rpcclient->cl_auth->au_ops->au_name,
-		setclientid.sc_name_len, setclientid.sc_name);
+		clp->cl_owner_id);
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task)) {
 		status = PTR_ERR(task);
@@ -5402,6 +5484,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	atomic_inc(&lsp->ls_count);
 	/* Ensure we don't close file until we're done freeing locks! */
 	p->ctx = get_nfs_open_context(ctx);
+	get_file(fl->fl_file);
 	memcpy(&p->fl, fl, sizeof(p->fl));
 	p->server = NFS_SERVER(inode);
 	return p;
@@ -5413,6 +5496,7 @@ static void nfs4_locku_release_calldata(void *data)
 	nfs_free_seqid(calldata->arg.seqid);
 	nfs4_put_lock_state(calldata->lsp);
 	put_nfs_open_context(calldata->ctx);
+	fput(calldata->fl.fl_file);
 	kfree(calldata);
 }
 
@@ -6846,11 +6930,14 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred,
 	};
 
 	nfs4_init_boot_verifier(clp, &verifier);
-	args.id_len = nfs4_init_uniform_client_string(clp, args.id,
-							sizeof(args.id));
-	dprintk("NFS call  exchange_id auth=%s, '%.*s'\n",
+
+	status = nfs4_init_uniform_client_string(clp);
+	if (status)
+		goto out;
+
+	dprintk("NFS call  exchange_id auth=%s, '%s'\n",
 		clp->cl_rpcclient->cl_auth->au_ops->au_name,
-		args.id_len, args.id);
+		clp->cl_owner_id);
 
 	res.server_owner = kzalloc(sizeof(struct nfs41_server_owner),
 					GFP_NOFS);
@@ -6885,7 +6972,7 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred,
 		/* unsupported! */
 		WARN_ON_ONCE(1);
 		status = -EINVAL;
-		goto out_server_scope;
+		goto out_impl_id;
 	}
 
 	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
@@ -6913,6 +7000,7 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred,
 		/* use the most recent implementation id */
 		kfree(clp->cl_implid);
 		clp->cl_implid = res.impl_id;
+		res.impl_id = NULL;
 
 		if (clp->cl_serverscope != NULL &&
 		    !nfs41_same_server_scope(clp->cl_serverscope,
@@ -6926,15 +7014,16 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred,
 
 		if (clp->cl_serverscope == NULL) {
 			clp->cl_serverscope = res.server_scope;
-			goto out;
+			res.server_scope = NULL;
 		}
-	} else
-		kfree(res.impl_id);
+	}
 
-out_server_owner:
-	kfree(res.server_owner);
+out_impl_id:
+	kfree(res.impl_id);
 out_server_scope:
 	kfree(res.server_scope);
+out_server_owner:
+	kfree(res.server_owner);
 out:
 	if (clp->cl_implid != NULL)
 		dprintk("NFS reply exchange_id: Server Implementation ID: "
@@ -8061,9 +8150,8 @@ nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync)
 	struct rpc_task *task;
 	int status = 0;
 
-	dprintk("NFS: %4d initiating layoutcommit call. sync %d "
-		"lbw: %llu inode %lu\n",
-		data->task.tk_pid, sync,
+	dprintk("NFS: initiating layoutcommit call. sync %d "
+		"lbw: %llu inode %lu\n", sync,
 		data->args.lastbytewritten,
 		data->args.inode->i_ino);
 
@@ -8557,7 +8645,8 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = {
 		| NFS_CAP_ATOMIC_OPEN_V1
 		| NFS_CAP_ALLOCATE
 		| NFS_CAP_DEALLOCATE
-		| NFS_CAP_SEEK,
+		| NFS_CAP_SEEK
+		| NFS_CAP_LAYOUTSTATS,
 	.init_client = nfs41_init_client,
 	.shutdown_client = nfs41_shutdown_client,
 	.match_stateid = nfs41_match_stateid,

fs/nfs/nfs4state.c

@@ -309,7 +309,6 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 
 	if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
 		goto do_confirm;
-	nfs4_begin_drain_session(clp);
 	status = nfs4_proc_exchange_id(clp, cred);
 	if (status != 0)
 		goto out;
@@ -1482,6 +1481,8 @@ restart:
 					spin_unlock(&state->state_lock);
 				}
 				nfs4_put_open_state(state);
+				clear_bit(NFS4CLNT_RECLAIM_NOGRACE,
+					&state->flags);
 				spin_lock(&sp->so_lock);
 				goto restart;
 			}
@@ -1830,6 +1831,7 @@ static int nfs4_establish_lease(struct nfs_client *clp)
 		clp->cl_mvops->reboot_recovery_ops;
 	int status;
 
+	nfs4_begin_drain_session(clp);
 	cred = nfs4_get_clid_cred(clp);
 	if (cred == NULL)
 		return -ENOENT;

fs/nfs/nfs4xdr.c

@@ -139,7 +139,8 @@ static int nfs4_stat_to_errno(int);
 #define encode_setclientid_maxsz \
 				(op_encode_hdr_maxsz + \
 				XDR_QUADLEN(NFS4_VERIFIER_SIZE) + \
-				XDR_QUADLEN(NFS4_SETCLIENTID_NAMELEN) + \
+				/* client name */ \
+				1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \
 				1 /* sc_prog */ + \
 				1 + XDR_QUADLEN(RPCBIND_MAXNETIDLEN) + \
 				1 + XDR_QUADLEN(RPCBIND_MAXUADDRLEN) + \
@@ -288,7 +289,8 @@ static int nfs4_stat_to_errno(int);
 #define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \
 				encode_verifier_maxsz + \
 				1 /* co_ownerid.len */ + \
-				XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \
+				/* eia_clientowner */ \
+				1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \
 				1 /* flags */ + \
 				1 /* spa_how */ + \
 				/* max is SP4_MACH_CRED (for now) */ + \
@@ -1667,13 +1669,14 @@ static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclie
 	encode_op_hdr(xdr, OP_SETCLIENTID, decode_setclientid_maxsz, hdr);
 	encode_nfs4_verifier(xdr, setclientid->sc_verifier);
 
-	encode_string(xdr, setclientid->sc_name_len, setclientid->sc_name);
+	encode_string(xdr, strlen(setclientid->sc_clnt->cl_owner_id),
+			setclientid->sc_clnt->cl_owner_id);
 	p = reserve_space(xdr, 4);
 	*p = cpu_to_be32(setclientid->sc_prog);
 	encode_string(xdr, setclientid->sc_netid_len, setclientid->sc_netid);
 	encode_string(xdr, setclientid->sc_uaddr_len, setclientid->sc_uaddr);
 	p = reserve_space(xdr, 4);
-	*p = cpu_to_be32(setclientid->sc_cb_ident);
+	*p = cpu_to_be32(setclientid->sc_clnt->cl_cb_ident);
 }
 
 static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_setclientid_res *arg, struct compound_hdr *hdr)
@@ -1747,7 +1750,8 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 	encode_op_hdr(xdr, OP_EXCHANGE_ID, decode_exchange_id_maxsz, hdr);
 	encode_nfs4_verifier(xdr, args->verifier);
 
-	encode_string(xdr, args->id_len, args->id);
+	encode_string(xdr, strlen(args->client->cl_owner_id),
+			args->client->cl_owner_id);
 
 	encode_uint32(xdr, args->flags);
 	encode_uint32(xdr, args->state_protect.how);
@@ -7427,6 +7431,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
 	PROC(SEEK,		enc_seek,		dec_seek),
 	PROC(ALLOCATE,		enc_allocate,		dec_allocate),
 	PROC(DEALLOCATE,	enc_deallocate,		dec_deallocate),
+	PROC(LAYOUTSTATS,	enc_layoutstats,	dec_layoutstats),
 #endif /* CONFIG_NFS_V4_2 */
 };
 

fs/nfs/pagelist.c

@@ -636,9 +636,8 @@ int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr,
 
 	hdr->rw_ops->rw_initiate(hdr, &msg, rpc_ops, &task_setup_data, how);
 
-	dprintk("NFS: %5u initiated pgio call "
+	dprintk("NFS: initiated pgio call "
 		"(req %s/%llu, %u bytes @ offset %llu)\n",
-		hdr->task.tk_pid,
 		hdr->inode->i_sb->s_id,
 		(unsigned long long)NFS_FILEID(hdr->inode),
 		hdr->args.count,
@@ -690,8 +689,6 @@ static int nfs_pgio_error(struct nfs_pageio_descriptor *desc,
 static void nfs_pgio_release(void *calldata)
 {
 	struct nfs_pgio_header *hdr = calldata;
-	if (hdr->rw_ops->rw_release)
-		hdr->rw_ops->rw_release(hdr);
 	nfs_pgio_data_destroy(hdr);
 	hdr->completion_ops->completion(hdr);
 }
@@ -711,7 +708,9 @@ static void nfs_pageio_mirror_init(struct nfs_pgio_mirror *mirror,
  * nfs_pageio_init - initialise a page io descriptor
  * @desc: pointer to descriptor
  * @inode: pointer to inode
- * @doio: pointer to io function
+ * @pg_ops: pointer to pageio operations
+ * @compl_ops: pointer to pageio completion operations
+ * @rw_ops: pointer to nfs read/write operations
  * @bsize: io block size
  * @io_flags: extra parameters for the io function
  */
@@ -1186,6 +1185,7 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
  * nfs_pageio_complete_mirror - Complete I/O on the current mirror of an
  *				nfs_pageio_descriptor
  * @desc: pointer to io descriptor
+ * @mirror_idx: pointer to mirror index
  */
 static void nfs_pageio_complete_mirror(struct nfs_pageio_descriptor *desc,
 				       u32 mirror_idx)

fs/nfs/pnfs.c

@@ -35,6 +35,7 @@
 #include "iostat.h"
 #include "nfs4trace.h"
 #include "delegation.h"
+#include "nfs42.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PNFS
 #define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ)
@@ -1821,6 +1822,7 @@ int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *hdr)
 	/* Resend all requests through the MDS */
 	nfs_pageio_init_write(&pgio, hdr->inode, FLUSH_STABLE, true,
 			      hdr->completion_ops);
+	set_bit(NFS_CONTEXT_RESEND_WRITES, &hdr->args.context->flags);
 	return nfs_pageio_resend(&pgio, hdr);
 }
 EXPORT_SYMBOL_GPL(pnfs_write_done_resend_to_mds);
@@ -1865,6 +1867,7 @@ pnfs_write_through_mds(struct nfs_pageio_descriptor *desc,
 		mirror->pg_recoalesce = 1;
 	}
 	nfs_pgio_data_destroy(hdr);
+	hdr->release(hdr);
 }
 
 static enum pnfs_try_status
@@ -1979,6 +1982,7 @@ pnfs_read_through_mds(struct nfs_pageio_descriptor *desc,
 		mirror->pg_recoalesce = 1;
 	}
 	nfs_pgio_data_destroy(hdr);
+	hdr->release(hdr);
 }
 
 /*
@@ -2247,3 +2251,63 @@ struct nfs4_threshold *pnfs_mdsthreshold_alloc(void)
 	}
 	return thp;
 }
+
+#if IS_ENABLED(CONFIG_NFS_V4_2)
+int
+pnfs_report_layoutstat(struct inode *inode)
+{
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs42_layoutstat_data *data;
+	struct pnfs_layout_hdr *hdr;
+	int status = 0;
+
+	if (!pnfs_enabled_sb(server) || !ld->prepare_layoutstats)
+		goto out;
+
+	if (!nfs_server_capable(inode, NFS_CAP_LAYOUTSTATS))
+		goto out;
+
+	if (test_and_set_bit(NFS_INO_LAYOUTSTATS, &nfsi->flags))
+		goto out;
+
+	spin_lock(&inode->i_lock);
+	if (!NFS_I(inode)->layout) {
+		spin_unlock(&inode->i_lock);
+		goto out;
+	}
+	hdr = NFS_I(inode)->layout;
+	pnfs_get_layout_hdr(hdr);
+	spin_unlock(&inode->i_lock);
+
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (!data) {
+		status = -ENOMEM;
+		goto out_put;
+	}
+
+	data->args.fh = NFS_FH(inode);
+	data->args.inode = inode;
+	nfs4_stateid_copy(&data->args.stateid, &hdr->plh_stateid);
+	status = ld->prepare_layoutstats(&data->args);
+	if (status)
+		goto out_free;
+
+	status = nfs42_proc_layoutstats_generic(NFS_SERVER(inode), data);
+
+out:
+	dprintk("%s returns %d\n", __func__, status);
+	return status;
+
+out_free:
+	kfree(data);
+out_put:
+	pnfs_put_layout_hdr(hdr);
+	smp_mb__before_atomic();
+	clear_bit(NFS_INO_LAYOUTSTATS, &nfsi->flags);
+	smp_mb__after_atomic();
+	goto out;
+}
+EXPORT_SYMBOL_GPL(pnfs_report_layoutstat);
+#endif

fs/nfs/pnfs.h

@@ -178,6 +178,8 @@ struct pnfs_layoutdriver_type {
 	void (*encode_layoutcommit) (struct pnfs_layout_hdr *lo,
 				     struct xdr_stream *xdr,
 				     const struct nfs4_layoutcommit_args *args);
+	int (*prepare_layoutstats) (struct nfs42_layoutstat_args *args);
+	void (*cleanup_layoutstats) (struct nfs42_layoutstat_data *data);
 };
 
 struct pnfs_layout_hdr {
@@ -290,7 +292,6 @@ int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *);
 struct nfs4_threshold *pnfs_mdsthreshold_alloc(void);
 void pnfs_error_mark_layout_for_return(struct inode *inode,
 				       struct pnfs_layout_segment *lseg);
-
 /* nfs4_deviceid_flags */
 enum {
 	NFS_DEVICEID_INVALID = 0,       /* set when MDS clientid recalled */
@@ -689,4 +690,14 @@ static inline void nfs4_pnfs_v3_ds_connect_unload(void)
 
 #endif /* CONFIG_NFS_V4_1 */
 
+#if IS_ENABLED(CONFIG_NFS_V4_2)
+int pnfs_report_layoutstat(struct inode *inode);
+#else
+static inline int
+pnfs_report_layoutstat(struct inode *inode)
+{
+	return 0;
+}
+#endif
+
 #endif /* FS_NFS_PNFS_H */

fs/nfs/write.c

@@ -1290,6 +1290,7 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr,
 static void nfs_redirty_request(struct nfs_page *req)
 {
 	nfs_mark_request_dirty(req);
+	set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags);
 	nfs_unlock_request(req);
 	nfs_end_page_writeback(req);
 	nfs_release_request(req);
@@ -1348,11 +1349,6 @@ void nfs_commit_prepare(struct rpc_task *task, void *calldata)
 	NFS_PROTO(data->inode)->commit_rpc_prepare(task, data);
 }
 
-static void nfs_writeback_release_common(struct nfs_pgio_header *hdr)
-{
-	/* do nothing! */
-}
-
 /*
  * Special version of should_remove_suid() that ignores capabilities.
  */
@@ -1556,7 +1552,7 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data,
 	/* Set up the initial task struct.  */
 	nfs_ops->commit_setup(data, &msg);
 
-	dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
+	dprintk("NFS: initiated commit call\n");
 
 	nfs4_state_protect(NFS_SERVER(data->inode)->nfs_client,
 		NFS_SP4_MACH_CRED_COMMIT, &task_setup_data.rpc_client, &msg);
@@ -2013,7 +2009,6 @@ static const struct nfs_rw_ops nfs_rw_write_ops = {
 	.rw_mode		= FMODE_WRITE,
 	.rw_alloc_header	= nfs_writehdr_alloc,
 	.rw_free_header		= nfs_writehdr_free,
-	.rw_release		= nfs_writeback_release_common,
 	.rw_done		= nfs_writeback_done,
 	.rw_result		= nfs_writeback_result,
 	.rw_initiate		= nfs_initiate_write,