Merge tag 'nfs-for-5.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Stable fixes:
   - Fix a page leak in nfs_destroy_unlinked_subrequests()

   - Fix use-after-free issues in nfs_pageio_add_request()

   - Fix new mount code constant_table array definitions

   - finish_automount() requires us to hold 2 refs to the mount record

  Features:
   - Improve the accuracy of telldir/seekdir by using 64-bit cookies
     when possible.

   - Allow one RDMA active connection and several zombie connections to
     prevent blocking if the remote server is unresponsive.

   - Limit the size of the NFS access cache by default

   - Reduce the number of references to credentials that are taken by
     NFS

   - pNFS files and flexfiles drivers now support per-layout segment
     COMMIT lists.

   - Enable partial-file layout segments in the pNFS/flexfiles driver.

   - Add support for CB_RECALL_ANY to the pNFS flexfiles layout type

   - pNFS/flexfiles Report NFS4ERR_DELAY and NFS4ERR_GRACE errors from
     the DS using the layouterror mechanism.

  Bugfixes and cleanups:
   - SUNRPC: Fix krb5p regressions

   - Don't specify NFS version in "UDP not supported" error

   - nfsroot: set tcp as the default transport protocol

   - pnfs: Return valid stateids in nfs_layout_find_inode_by_stateid()

   - alloc_nfs_open_context() must use the file cred when available

   - Fix locking when dereferencing the delegation cred

   - Fix memory leaks in O_DIRECT when nfs_get_lock_context() fails

   - Various clean ups of the NFS O_DIRECT commit code

   - Clean up RDMA connect/disconnect

   - Replace zero-length arrays with C99-style flexible arrays"

* tag 'nfs-for-5.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (86 commits)
  NFS: Clean up process of marking inode stale.
  SUNRPC: Don't start a timer on an already queued rpc task
  NFS/pnfs: Reference the layout cred in pnfs_prepare_layoutreturn()
  NFS/pnfs: Fix dereference of layout cred in pnfs_layoutcommit_inode()
  NFS: Beware when dereferencing the delegation cred
  NFS: Add a module parameter to set nfs_mountpoint_expiry_timeout
  NFS: finish_automount() requires us to hold 2 refs to the mount record
  NFS: Fix a few constant_table array definitions
  NFS: Try to join page groups before an O_DIRECT retransmission
  NFS: Refactor nfs_lock_and_join_requests()
  NFS: Reverse the submission order of requests in __nfs_pageio_add_request()
  NFS: Clean up nfs_lock_and_join_requests()
  NFS: Remove the redundant function nfs_pgio_has_mirroring()
  NFS: Fix memory leaks in nfs_pageio_stop_mirroring()
  NFS: Fix a request reference leak in nfs_direct_write_clear_reqs()
  NFS: Fix use-after-free issues in nfs_pageio_add_request()
  NFS: Fix races nfs_page_group_destroy() vs nfs_destroy_unlinked_subrequests()
  NFS: Fix a page leak in nfs_destroy_unlinked_subrequests()
  NFS: Remove unused FLUSH_SYNC support in nfs_initiate_pgio()
  pNFS/flexfiles: Specify the layout segment range in LAYOUTGET
  ...

fs/nfs/blocklayout/blocklayout.c

@@ -476,7 +476,7 @@ static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo)
 	err = ext_tree_remove(bl, true, 0, LLONG_MAX);
 	WARN_ON(err);
 
-	kfree(bl);
+	kfree_rcu(bl, bl_layout.plh_rcu);
 }
 
 static struct pnfs_layout_hdr *__bl_alloc_layout_hdr(struct inode *inode,

fs/nfs/callback.h

@@ -127,7 +127,9 @@ extern __be32 nfs4_callback_sequence(void *argp, void *resp,
 #define RCA4_TYPE_MASK_OBJ_LAYOUT_MAX  9
 #define RCA4_TYPE_MASK_OTHER_LAYOUT_MIN 12
 #define RCA4_TYPE_MASK_OTHER_LAYOUT_MAX 15
-#define RCA4_TYPE_MASK_ALL 0xf31f
+#define PNFS_FF_RCA4_TYPE_MASK_READ 16
+#define PNFS_FF_RCA4_TYPE_MASK_RW 17
+#define RCA4_TYPE_MASK_ALL 0x3f31f
 
 struct cb_recallanyargs {
 	uint32_t	craa_objs_to_keep;

fs/nfs/callback_proc.c

@@ -121,31 +121,31 @@ __be32 nfs4_callback_recall(void *argp, void *resp,
  */
 static struct inode *nfs_layout_find_inode_by_stateid(struct nfs_client *clp,
 		const nfs4_stateid *stateid)
+	__must_hold(RCU)
 {
 	struct nfs_server *server;
 	struct inode *inode;
 	struct pnfs_layout_hdr *lo;
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+		list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
+			if (!pnfs_layout_is_valid(lo))
+				continue;
 			if (stateid != NULL &&
 			    !nfs4_stateid_match_other(stateid, &lo->plh_stateid))
 				continue;
+			if (!nfs_sb_active(server->super))
+				continue;
 			inode = igrab(lo->plh_inode);
-			if (!inode)
-				return ERR_PTR(-EAGAIN);
-			if (!nfs_sb_active(inode->i_sb)) {
-				rcu_read_unlock();
-				spin_unlock(&clp->cl_lock);
-				iput(inode);
-				spin_lock(&clp->cl_lock);
-				rcu_read_lock();
-				return ERR_PTR(-EAGAIN);
-			}
-			return inode;
+			rcu_read_unlock();
+			if (inode)
+				return inode;
+			nfs_sb_deactive(server->super);
+			return ERR_PTR(-EAGAIN);
 		}
 	}
-
+	rcu_read_unlock();
 	return ERR_PTR(-ENOENT);
 }
 
@@ -163,28 +163,25 @@ static struct inode *nfs_layout_find_inode_by_fh(struct nfs_client *clp,
 	struct inode *inode;
 	struct pnfs_layout_hdr *lo;
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+		list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
 			nfsi = NFS_I(lo->plh_inode);
 			if (nfs_compare_fh(fh, &nfsi->fh))
 				continue;
 			if (nfsi->layout != lo)
 				continue;
+			if (!nfs_sb_active(server->super))
+				continue;
 			inode = igrab(lo->plh_inode);
-			if (!inode)
-				return ERR_PTR(-EAGAIN);
-			if (!nfs_sb_active(inode->i_sb)) {
-				rcu_read_unlock();
-				spin_unlock(&clp->cl_lock);
-				iput(inode);
-				spin_lock(&clp->cl_lock);
-				rcu_read_lock();
-				return ERR_PTR(-EAGAIN);
-			}
-			return inode;
+			rcu_read_unlock();
+			if (inode)
+				return inode;
+			nfs_sb_deactive(server->super);
+			return ERR_PTR(-EAGAIN);
 		}
 	}
-
+	rcu_read_unlock();
 	return ERR_PTR(-ENOENT);
 }
 
@@ -194,14 +191,9 @@ static struct inode *nfs_layout_find_inode(struct nfs_client *clp,
 {
 	struct inode *inode;
 
-	spin_lock(&clp->cl_lock);
-	rcu_read_lock();
 	inode = nfs_layout_find_inode_by_stateid(clp, stateid);
 	if (inode == ERR_PTR(-ENOENT))
 		inode = nfs_layout_find_inode_by_fh(clp, fh);
-	rcu_read_unlock();
-	spin_unlock(&clp->cl_lock);
-
 	return inode;
 }
 
@@ -280,7 +272,7 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 		goto unlock;
 	}
 
-	pnfs_set_layout_stateid(lo, &args->cbl_stateid, true);
+	pnfs_set_layout_stateid(lo, &args->cbl_stateid, NULL, true);
 	switch (pnfs_mark_matching_lsegs_return(lo, &free_me_list,
 				&args->cbl_range,
 				be32_to_cpu(args->cbl_stateid.seqid))) {
@@ -605,6 +597,7 @@ __be32 nfs4_callback_recallany(void *argp, void *resp,
 	struct cb_recallanyargs *args = argp;
 	__be32 status;
 	fmode_t flags = 0;
+	bool schedule_manager = false;
 
 	status = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION);
 	if (!cps->clp) /* set in cb_sequence */
@@ -627,6 +620,18 @@ __be32 nfs4_callback_recallany(void *argp, void *resp,
 
 	if (args->craa_type_mask & BIT(RCA4_TYPE_MASK_FILE_LAYOUT))
 		pnfs_recall_all_layouts(cps->clp);
+
+	if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_READ)) {
+		set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &cps->clp->cl_state);
+		schedule_manager = true;
+	}
+	if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_RW)) {
+		set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &cps->clp->cl_state);
+		schedule_manager = true;
+	}
+	if (schedule_manager)
+		nfs4_schedule_state_manager(cps->clp);
+
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	return status;

fs/nfs/delegation.c

@@ -377,6 +377,18 @@ nfs_inode_detach_delegation(struct inode *inode)
 	return delegation;
 }
 
+static void
+nfs_update_delegation_cred(struct nfs_delegation *delegation,
+		const struct cred *cred)
+{
+	const struct cred *old;
+
+	if (cred_fscmp(delegation->cred, cred) != 0) {
+		old = xchg(&delegation->cred, get_cred(cred));
+		put_cred(old);
+	}
+}
+
 static void
 nfs_update_inplace_delegation(struct nfs_delegation *delegation,
 		const struct nfs_delegation *update)
@@ -385,8 +397,14 @@ nfs_update_inplace_delegation(struct nfs_delegation *delegation,
 		delegation->stateid.seqid = update->stateid.seqid;
 		smp_wmb();
 		delegation->type = update->type;
-		if (test_and_clear_bit(NFS_DELEGATION_REVOKED, &delegation->flags))
+		delegation->pagemod_limit = update->pagemod_limit;
+		if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
+			delegation->change_attr = update->change_attr;
+			nfs_update_delegation_cred(delegation, update->cred);
+			/* smp_mb__before_atomic() is implicit due to xchg() */
+			clear_bit(NFS_DELEGATION_REVOKED, &delegation->flags);
 			atomic_long_inc(&nfs_active_delegations);
+		}
 	}
 }
 
@@ -545,21 +563,11 @@ static bool nfs_delegation_need_return(struct nfs_delegation *delegation)
 	return ret;
 }
 
-/**
- * nfs_client_return_marked_delegations - return previously marked delegations
- * @clp: nfs_client to process
- *
- * Note that this function is designed to be called by the state
- * manager thread. For this reason, it cannot flush the dirty data,
- * since that could deadlock in case of a state recovery error.
- *
- * Returns zero on success, or a negative errno value.
- */
-int nfs_client_return_marked_delegations(struct nfs_client *clp)
+static int nfs_server_return_marked_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
 	struct nfs_delegation *prev;
-	struct nfs_server *server;
 	struct inode *inode;
 	struct inode *place_holder = NULL;
 	struct nfs_delegation *place_holder_deleg = NULL;
@@ -569,78 +577,79 @@ int nfs_client_return_marked_delegations(struct nfs_client *clp)
 	/*
 	 * To avoid quadratic looping we hold a reference
 	 * to an inode place_holder.  Each time we restart, we
-	 * list nfs_servers from the server of that inode, and
-	 * delegation in the server from the delegations of that
-	 * inode.
+	 * list delegation in the server from the delegations
+	 * of that inode.
 	 * prev is an RCU-protected pointer to a delegation which
 	 * wasn't marked for return and might be a good choice for
 	 * the next place_holder.
 	 */
-	rcu_read_lock();
 	prev = NULL;
+	delegation = NULL;
+	rcu_read_lock();
 	if (place_holder)
-		server = NFS_SERVER(place_holder);
-	else
-		server = list_entry_rcu(clp->cl_superblocks.next,
-					struct nfs_server, client_link);
-	list_for_each_entry_from_rcu(server, &clp->cl_superblocks, client_link) {
-		delegation = NULL;
-		if (place_holder && server == NFS_SERVER(place_holder))
-			delegation = rcu_dereference(NFS_I(place_holder)->delegation);
-		if (!delegation || delegation != place_holder_deleg)
-			delegation = list_entry_rcu(server->delegations.next,
-						    struct nfs_delegation, super_list);
-		list_for_each_entry_from_rcu(delegation, &server->delegations, super_list) {
-			struct inode *to_put = NULL;
-
-			if (!nfs_delegation_need_return(delegation)) {
+		delegation = rcu_dereference(NFS_I(place_holder)->delegation);
+	if (!delegation || delegation != place_holder_deleg)
+		delegation = list_entry_rcu(server->delegations.next,
+					    struct nfs_delegation, super_list);
+	list_for_each_entry_from_rcu(delegation, &server->delegations, super_list) {
+		struct inode *to_put = NULL;
+
+		if (test_bit(NFS_DELEGATION_INODE_FREEING, &delegation->flags))
+			continue;
+		if (!nfs_delegation_need_return(delegation)) {
+			if (nfs4_is_valid_delegation(delegation, 0))
 				prev = delegation;
-				continue;
-			}
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-
-			if (prev) {
-				struct inode *tmp;
-
-				tmp = nfs_delegation_grab_inode(prev);
-				if (tmp) {
-					to_put = place_holder;
-					place_holder = tmp;
-					place_holder_deleg = prev;
-				}
-			}
+			continue;
+		}
 
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				if (to_put)
-					iput(to_put);
-				nfs_sb_deactive(server->super);
-				goto restart;
+		if (prev) {
+			struct inode *tmp = nfs_delegation_grab_inode(prev);
+			if (tmp) {
+				to_put = place_holder;
+				place_holder = tmp;
+				place_holder_deleg = prev;
 			}
-			delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		}
+
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL) {
 			rcu_read_unlock();
+			iput(to_put);
+			goto restart;
+		}
+		delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		rcu_read_unlock();
 
-			if (to_put)
-				iput(to_put);
+		iput(to_put);
 
-			err = nfs_end_delegation_return(inode, delegation, 0);
-			iput(inode);
-			nfs_sb_deactive(server->super);
-			cond_resched();
-			if (!err)
-				goto restart;
-			set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
-			if (place_holder)
-				iput(place_holder);
-			return err;
-		}
+		err = nfs_end_delegation_return(inode, delegation, 0);
+		iput(inode);
+		cond_resched();
+		if (!err)
+			goto restart;
+		set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state);
+		goto out;
 	}
 	rcu_read_unlock();
-	if (place_holder)
-		iput(place_holder);
-	return 0;
+out:
+	iput(place_holder);
+	return err;
+}
+
+/**
+ * nfs_client_return_marked_delegations - return previously marked delegations
+ * @clp: nfs_client to process
+ *
+ * Note that this function is designed to be called by the state
+ * manager thread. For this reason, it cannot flush the dirty data,
+ * since that could deadlock in case of a state recovery error.
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs_client_return_marked_delegations(struct nfs_client *clp)
+{
+	return nfs_client_for_each_server(clp,
+			nfs_server_return_marked_delegations, NULL);
 }
 
 /**
@@ -1083,53 +1092,51 @@ void nfs_delegation_mark_reclaim(struct nfs_client *clp)
 	rcu_read_unlock();
 }
 
-/**
- * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done
- * @clp: nfs_client to process
- *
- */
-void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
+static int nfs_server_reap_unclaimed_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
-	struct nfs_server *server;
 	struct inode *inode;
-
 restart:
 	rcu_read_lock();
-	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry_rcu(delegation, &server->delegations,
-								super_list) {
-			if (test_bit(NFS_DELEGATION_INODE_FREEING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_RETURNING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_NEED_RECLAIM,
-						&delegation->flags) == 0)
-				continue;
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				nfs_sb_deactive(server->super);
-				goto restart;
-			}
-			delegation = nfs_start_delegation_return_locked(NFS_I(inode));
-			rcu_read_unlock();
-			if (delegation != NULL) {
-				if (nfs_detach_delegation(NFS_I(inode), delegation,
-							server) != NULL)
-					nfs_free_delegation(delegation);
-				/* Match nfs_start_delegation_return_locked */
-				nfs_put_delegation(delegation);
-			}
-			iput(inode);
-			nfs_sb_deactive(server->super);
-			cond_resched();
-			goto restart;
+restart_locked:
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if (test_bit(NFS_DELEGATION_INODE_FREEING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_RETURNING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_NEED_RECLAIM,
+					&delegation->flags) == 0)
+			continue;
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL)
+			goto restart_locked;
+		delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		rcu_read_unlock();
+		if (delegation != NULL) {
+			if (nfs_detach_delegation(NFS_I(inode), delegation,
+						server) != NULL)
+				nfs_free_delegation(delegation);
+			/* Match nfs_start_delegation_return_locked */
+			nfs_put_delegation(delegation);
 		}
+		iput(inode);
+		cond_resched();
+		goto restart;
 	}
 	rcu_read_unlock();
+	return 0;
+}
+
+/**
+ * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done
+ * @clp: nfs_client to process
+ *
+ */
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
+{
+	nfs_client_for_each_server(clp, nfs_server_reap_unclaimed_delegations,
+			NULL);
 }
 
 static inline bool nfs4_server_rebooted(const struct nfs_client *clp)
@@ -1215,62 +1222,61 @@ nfs_delegation_test_free_expired(struct inode *inode,
 		nfs_remove_bad_delegation(inode, stateid);
 }
 
-/**
- * nfs_reap_expired_delegations - reap expired delegations
- * @clp: nfs_client to process
- *
- * Iterates through all the delegations associated with this server and
- * checks if they have may have been revoked. This function is usually
- * expected to be called in cases where the server may have lost its
- * lease.
- */
-void nfs_reap_expired_delegations(struct nfs_client *clp)
+static int nfs_server_reap_expired_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
-	struct nfs_server *server;
 	struct inode *inode;
 	const struct cred *cred;
 	nfs4_stateid stateid;
-
 restart:
 	rcu_read_lock();
-	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry_rcu(delegation, &server->delegations,
-								super_list) {
-			if (test_bit(NFS_DELEGATION_INODE_FREEING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_RETURNING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_TEST_EXPIRED,
-						&delegation->flags) == 0)
-				continue;
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				nfs_sb_deactive(server->super);
-				goto restart;
-			}
-			cred = get_cred_rcu(delegation->cred);
-			nfs4_stateid_copy(&stateid, &delegation->stateid);
-			clear_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags);
-			rcu_read_unlock();
-			nfs_delegation_test_free_expired(inode, &stateid, cred);
-			put_cred(cred);
-			if (nfs4_server_rebooted(clp)) {
-				nfs_inode_mark_test_expired_delegation(server,inode);
-				iput(inode);
-				nfs_sb_deactive(server->super);
-				return;
-			}
+restart_locked:
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if (test_bit(NFS_DELEGATION_INODE_FREEING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_RETURNING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_TEST_EXPIRED,
+					&delegation->flags) == 0)
+			continue;
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL)
+			goto restart_locked;
+		spin_lock(&delegation->lock);
+		cred = get_cred_rcu(delegation->cred);
+		nfs4_stateid_copy(&stateid, &delegation->stateid);
+		spin_unlock(&delegation->lock);
+		clear_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags);
+		rcu_read_unlock();
+		nfs_delegation_test_free_expired(inode, &stateid, cred);
+		put_cred(cred);
+		if (!nfs4_server_rebooted(server->nfs_client)) {
 			iput(inode);
-			nfs_sb_deactive(server->super);
 			cond_resched();
 			goto restart;
 		}
+		nfs_inode_mark_test_expired_delegation(server,inode);
+		iput(inode);
+		return -EAGAIN;
 	}
 	rcu_read_unlock();
+	return 0;
+}
+
+/**
+ * nfs_reap_expired_delegations - reap expired delegations
+ * @clp: nfs_client to process
+ *
+ * Iterates through all the delegations associated with this server and
+ * checks if they have may have been revoked. This function is usually
+ * expected to be called in cases where the server may have lost its
+ * lease.
+ */
+void nfs_reap_expired_delegations(struct nfs_client *clp)
+{
+	nfs_client_for_each_server(clp, nfs_server_reap_expired_delegations,
+			NULL);
 }
 
 void nfs_inode_find_delegation_state_and_recover(struct inode *inode,
@@ -1359,11 +1365,14 @@ bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags,
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
-	bool ret;
+	bool ret = false;
 
 	flags &= FMODE_READ|FMODE_WRITE;
 	rcu_read_lock();
 	delegation = rcu_dereference(nfsi->delegation);
+	if (!delegation)
+		goto out;
+	spin_lock(&delegation->lock);
 	ret = nfs4_is_valid_delegation(delegation, flags);
 	if (ret) {
 		nfs4_stateid_copy(dst, &delegation->stateid);
@@ -1371,6 +1380,8 @@ bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags,
 		if (cred)
 			*cred = get_cred(delegation->cred);
 	}
+	spin_unlock(&delegation->lock);
+out:
 	rcu_read_unlock();
 	return ret;
 }

fs/nfs/dir.c

@@ -141,10 +141,9 @@ struct nfs_cache_array {
 	int size;
 	int eof_index;
 	u64 last_cookie;
-	struct nfs_cache_array_entry array[0];
+	struct nfs_cache_array_entry array[];
 };
 
-typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool);
 typedef struct {
 	struct file	*file;
 	struct page	*page;
@@ -153,7 +152,7 @@ typedef struct {
 	u64		*dir_cookie;
 	u64		last_cookie;
 	loff_t		current_index;
-	decode_dirent_t	decode;
+	loff_t		prev_index;
 
 	unsigned long	dir_verifier;
 	unsigned long	timestamp;
@@ -240,6 +239,25 @@ int nfs_readdir_add_to_array(struct nfs_entry *entry, struct page *page)
 	return ret;
 }
 
+static inline
+int is_32bit_api(void)
+{
+#ifdef CONFIG_COMPAT
+	return in_compat_syscall();
+#else
+	return (BITS_PER_LONG == 32);
+#endif
+}
+
+static
+bool nfs_readdir_use_cookie(const struct file *filp)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return false;
+	return true;
+}
+
 static
 int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc)
 {
@@ -289,7 +307,7 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des
 			    !nfs_readdir_inode_mapping_valid(nfsi)) {
 				ctx->duped = 0;
 				ctx->attr_gencount = nfsi->attr_gencount;
-			} else if (new_pos < desc->ctx->pos) {
+			} else if (new_pos < desc->prev_index) {
 				if (ctx->duped > 0
 				    && ctx->dup_cookie == *desc->dir_cookie) {
 					if (printk_ratelimit()) {
@@ -305,7 +323,11 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des
 				ctx->dup_cookie = *desc->dir_cookie;
 				ctx->duped = -1;
 			}
-			desc->ctx->pos = new_pos;
+			if (nfs_readdir_use_cookie(desc->file))
+				desc->ctx->pos = *desc->dir_cookie;
+			else
+				desc->ctx->pos = new_pos;
+			desc->prev_index = new_pos;
 			desc->cache_entry_index = i;
 			return 0;
 		}
@@ -376,9 +398,10 @@ int nfs_readdir_xdr_filler(struct page **pages, nfs_readdir_descriptor_t *desc,
 static int xdr_decode(nfs_readdir_descriptor_t *desc,
 		      struct nfs_entry *entry, struct xdr_stream *xdr)
 {
+	struct inode *inode = file_inode(desc->file);
 	int error;
 
-	error = desc->decode(xdr, entry, desc->plus);
+	error = NFS_PROTO(inode)->decode_dirent(xdr, entry, desc->plus);
 	if (error)
 		return error;
 	entry->fattr->time_start = desc->timestamp;
@@ -756,6 +779,7 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
 
 	if (desc->page_index == 0) {
 		desc->current_index = 0;
+		desc->prev_index = 0;
 		desc->last_cookie = 0;
 	}
 	do {
@@ -786,11 +810,14 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc)
 			desc->eof = true;
 			break;
 		}
-		desc->ctx->pos++;
 		if (i < (array->size-1))
 			*desc->dir_cookie = array->array[i+1].cookie;
 		else
 			*desc->dir_cookie = array->last_cookie;
+		if (nfs_readdir_use_cookie(file))
+			desc->ctx->pos = *desc->dir_cookie;
+		else
+			desc->ctx->pos++;
 		if (ctx->duped != 0)
 			ctx->duped = 1;
 	}
@@ -860,9 +887,14 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 {
 	struct dentry	*dentry = file_dentry(file);
 	struct inode	*inode = d_inode(dentry);
-	nfs_readdir_descriptor_t my_desc,
-			*desc = &my_desc;
 	struct nfs_open_dir_context *dir_ctx = file->private_data;
+	nfs_readdir_descriptor_t my_desc = {
+		.file = file,
+		.ctx = ctx,
+		.dir_cookie = &dir_ctx->dir_cookie,
+		.plus = nfs_use_readdirplus(inode, ctx),
+	},
+			*desc = &my_desc;
 	int res = 0;
 
 	dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n",
@@ -875,14 +907,6 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 	 * to either find the entry with the appropriate number or
 	 * revalidate the cookie.
 	 */
-	memset(desc, 0, sizeof(*desc));
-
-	desc->file = file;
-	desc->ctx = ctx;
-	desc->dir_cookie = &dir_ctx->dir_cookie;
-	desc->decode = NFS_PROTO(inode)->decode_dirent;
-	desc->plus = nfs_use_readdirplus(inode, ctx);
-
 	if (ctx->pos == 0 || nfs_attribute_cache_expired(inode))
 		res = nfs_revalidate_mapping(inode, file->f_mapping);
 	if (res < 0)
@@ -954,7 +978,10 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence)
 	}
 	if (offset != filp->f_pos) {
 		filp->f_pos = offset;
-		dir_ctx->dir_cookie = 0;
+		if (nfs_readdir_use_cookie(filp))
+			dir_ctx->dir_cookie = offset;
+		else
+			dir_ctx->dir_cookie = 0;
 		dir_ctx->duped = 0;
 	}
 	inode_unlock(inode);
@@ -2282,7 +2309,7 @@ static DEFINE_SPINLOCK(nfs_access_lru_lock);
 static LIST_HEAD(nfs_access_lru_list);
 static atomic_long_t nfs_access_nr_entries;
 
-static unsigned long nfs_access_max_cachesize = ULONG_MAX;
+static unsigned long nfs_access_max_cachesize = 4*1024*1024;
 module_param(nfs_access_max_cachesize, ulong, 0644);
 MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length");
 
@@ -2642,9 +2669,10 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask)
 	status = NFS_PROTO(inode)->access(inode, &cache);
 	if (status != 0) {
 		if (status == -ESTALE) {
-			nfs_zap_caches(inode);
 			if (!S_ISDIR(inode->i_mode))
-				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+				nfs_set_inode_stale(inode);
+			else
+				nfs_zap_caches(inode);
 		}
 		goto out;
 	}
@@ -2732,14 +2760,7 @@ int nfs_permission(struct inode *inode, int mask)
 	if (!NFS_PROTO(inode)->access)
 		goto out_notsup;
 
-	/* Always try fast lookups first */
-	rcu_read_lock();
-	res = nfs_do_access(inode, cred, mask|MAY_NOT_BLOCK);
-	rcu_read_unlock();
-	if (res == -ECHILD && !(mask & MAY_NOT_BLOCK)) {
-		/* Fast lookup failed, try the slow way */
-		res = nfs_do_access(inode, cred, mask);
-	}
+	res = nfs_do_access(inode, cred, mask);
 out:
 	if (!res && (mask & MAY_EXEC))
 		res = nfs_execute_ok(inode, mask);

fs/nfs/direct.c

@@ -94,7 +94,7 @@ struct nfs_direct_req {
 #define NFS_ODIRECT_RESCHED_WRITES	(2)	/* write verification failed */
 	/* for read */
 #define NFS_ODIRECT_SHOULD_DIRTY	(3)	/* dirty user-space page after read */
-	struct nfs_writeverf	verf;		/* unstable write verifier */
+#define NFS_ODIRECT_DONE		INT_MAX	/* write verification failed */
 };
 
 static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops;
@@ -151,106 +151,6 @@ nfs_direct_count_bytes(struct nfs_direct_req *dreq,
 		dreq->count = dreq_len;
 }
 
-/*
- * nfs_direct_select_verf - select the right verifier
- * @dreq - direct request possibly spanning multiple servers
- * @ds_clp - nfs_client of data server or NULL if MDS / non-pnfs
- * @commit_idx - commit bucket index for the DS
- *
- * returns the correct verifier to use given the role of the server
- */
-static struct nfs_writeverf *
-nfs_direct_select_verf(struct nfs_direct_req *dreq,
-		       struct nfs_client *ds_clp,
-		       int commit_idx)
-{
-	struct nfs_writeverf *verfp = &dreq->verf;
-
-#ifdef CONFIG_NFS_V4_1
-	/*
-	 * pNFS is in use, use the DS verf except commit_through_mds is set
-	 * for layout segment where nbuckets is zero.
-	 */
-	if (ds_clp && dreq->ds_cinfo.nbuckets > 0) {
-		if (commit_idx >= 0 && commit_idx < dreq->ds_cinfo.nbuckets)
-			verfp = &dreq->ds_cinfo.buckets[commit_idx].direct_verf;
-		else
-			WARN_ON_ONCE(1);
-	}
-#endif
-	return verfp;
-}
-
-
-/*
- * nfs_direct_set_hdr_verf - set the write/commit verifier
- * @dreq - direct request possibly spanning multiple servers
- * @hdr - pageio header to validate against previously seen verfs
- *
- * Set the server's (MDS or DS) "seen" verifier
- */
-static void nfs_direct_set_hdr_verf(struct nfs_direct_req *dreq,
-				    struct nfs_pgio_header *hdr)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);
-	WARN_ON_ONCE(verfp->committed >= 0);
-	memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf));
-	WARN_ON_ONCE(verfp->committed < 0);
-}
-
-static int nfs_direct_cmp_verf(const struct nfs_writeverf *v1,
-		const struct nfs_writeverf *v2)
-{
-	return nfs_write_verifier_cmp(&v1->verifier, &v2->verifier);
-}
-
-/*
- * nfs_direct_cmp_hdr_verf - compare verifier for pgio header
- * @dreq - direct request possibly spanning multiple servers
- * @hdr - pageio header to validate against previously seen verf
- *
- * set the server's "seen" verf if not initialized.
- * returns result of comparison between @hdr->verf and the "seen"
- * verf of the server used by @hdr (DS or MDS)
- */
-static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq,
-					  struct nfs_pgio_header *hdr)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);
-	if (verfp->committed < 0) {
-		nfs_direct_set_hdr_verf(dreq, hdr);
-		return 0;
-	}
-	return nfs_direct_cmp_verf(verfp, &hdr->verf);
-}
-
-/*
- * nfs_direct_cmp_commit_data_verf - compare verifier for commit data
- * @dreq - direct request possibly spanning multiple servers
- * @data - commit data to validate against previously seen verf
- *
- * returns result of comparison between @data->verf and the verf of
- * the server used by @data (DS or MDS)
- */
-static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq,
-					   struct nfs_commit_data *data)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, data->ds_clp,
-					 data->ds_commit_index);
-
-	/* verifier not set so always fail */
-	if (verfp->committed < 0 || data->res.verf->committed <= NFS_UNSTABLE)
-		return 1;
-
-	return nfs_direct_cmp_verf(verfp, data->res.verf);
-}
-
 /**
  * nfs_direct_IO - NFS address space operation for direct I/O
  * @iocb: target I/O control block
@@ -305,7 +205,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void)
 	kref_get(&dreq->kref);
 	init_completion(&dreq->completion);
 	INIT_LIST_HEAD(&dreq->mds_cinfo.list);
-	dreq->verf.committed = NFS_INVALID_STABLE_HOW;	/* not set yet */
+	pnfs_init_ds_commit_info(&dreq->ds_cinfo);
 	INIT_WORK(&dreq->work, nfs_direct_write_schedule_work);
 	spin_lock_init(&dreq->lock);
 
@@ -316,7 +216,7 @@ static void nfs_direct_req_free(struct kref *kref)
 {
 	struct nfs_direct_req *dreq = container_of(kref, struct nfs_direct_req, kref);
 
-	nfs_free_pnfs_ds_cinfo(&dreq->ds_cinfo);
+	pnfs_release_ds_info(&dreq->ds_cinfo, dreq->inode);
 	if (dreq->l_ctx != NULL)
 		nfs_put_lock_context(dreq->l_ctx);
 	if (dreq->ctx != NULL)
@@ -571,6 +471,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter)
 	l_ctx = nfs_get_lock_context(dreq->ctx);
 	if (IS_ERR(l_ctx)) {
 		result = PTR_ERR(l_ctx);
+		nfs_direct_req_release(dreq);
 		goto out_release;
 	}
 	dreq->l_ctx = l_ctx;
@@ -604,16 +505,31 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter)
 	return result;
 }
 
+static void
+nfs_direct_join_group(struct list_head *list, struct inode *inode)
+{
+	struct nfs_page *req, *next;
+
+	list_for_each_entry(req, list, wb_list) {
+		if (req->wb_head != req || req->wb_this_page == req)
+			continue;
+		for (next = req->wb_this_page;
+				next != req->wb_head;
+				next = next->wb_this_page) {
+			nfs_list_remove_request(next);
+			nfs_release_request(next);
+		}
+		nfs_join_page_group(req, inode);
+	}
+}
+
 static void
 nfs_direct_write_scan_commit_list(struct inode *inode,
 				  struct list_head *list,
 				  struct nfs_commit_info *cinfo)
 {
 	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
-#ifdef CONFIG_NFS_V4_1
-	if (cinfo->ds != NULL && cinfo->ds->nwritten != 0)
-		NFS_SERVER(inode)->pnfs_curr_ld->recover_commit_reqs(list, cinfo);
-#endif
+	pnfs_recover_commit_reqs(list, cinfo);
 	nfs_scan_commit_list(&cinfo->mds->list, list, cinfo, 0);
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 }
@@ -629,11 +545,12 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
 	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
 
+	nfs_direct_join_group(&reqs, dreq->inode);
+
 	dreq->count = 0;
 	dreq->max_count = 0;
 	list_for_each_entry(req, &reqs, wb_list)
 		dreq->max_count += req->wb_bytes;
-	dreq->verf.committed = NFS_INVALID_STABLE_HOW;
 	nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo);
 	get_dreq(dreq);
 
@@ -670,27 +587,35 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 
 static void nfs_direct_commit_complete(struct nfs_commit_data *data)
 {
+	const struct nfs_writeverf *verf = data->res.verf;
 	struct nfs_direct_req *dreq = data->dreq;
 	struct nfs_commit_info cinfo;
 	struct nfs_page *req;
 	int status = data->task.tk_status;
 
+	if (status < 0) {
+		/* Errors in commit are fatal */
+		dreq->error = status;
+		dreq->max_count = 0;
+		dreq->count = 0;
+		dreq->flags = NFS_ODIRECT_DONE;
+	} else if (dreq->flags == NFS_ODIRECT_DONE)
+		status = dreq->error;
+
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
-	if (status < 0 || nfs_direct_cmp_commit_data_verf(dreq, data))
-		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 
 	while (!list_empty(&data->pages)) {
 		req = nfs_list_entry(data->pages.next);
 		nfs_list_remove_request(req);
-		if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) {
+		if (status >= 0 && !nfs_write_match_verf(verf, req)) {
+			dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 			/*
 			 * Despite the reboot, the write was successful,
 			 * so reset wb_nio.
 			 */
 			req->wb_nio = 0;
-			/* Note the rewrite will go through mds */
 			nfs_mark_request_commit(req, NULL, &cinfo, 0);
-		} else
+		} else /* Error or match */
 			nfs_release_request(req);
 		nfs_unlock_and_release_request(req);
 	}
@@ -705,7 +630,8 @@ static void nfs_direct_resched_write(struct nfs_commit_info *cinfo,
 	struct nfs_direct_req *dreq = cinfo->dreq;
 
 	spin_lock(&dreq->lock);
-	dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
+	if (dreq->flags != NFS_ODIRECT_DONE)
+		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 	spin_unlock(&dreq->lock);
 	nfs_mark_request_commit(req, NULL, cinfo, 0);
 }
@@ -728,6 +654,23 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 		nfs_direct_write_reschedule(dreq);
 }
 
+static void nfs_direct_write_clear_reqs(struct nfs_direct_req *dreq)
+{
+	struct nfs_commit_info cinfo;
+	struct nfs_page *req;
+	LIST_HEAD(reqs);
+
+	nfs_init_cinfo_from_dreq(&cinfo, dreq);
+	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
+
+	while (!list_empty(&reqs)) {
+		req = nfs_list_entry(reqs.next);
+		nfs_list_remove_request(req);
+		nfs_release_request(req);
+		nfs_unlock_and_release_request(req);
+	}
+}
+
 static void nfs_direct_write_schedule_work(struct work_struct *work)
 {
 	struct nfs_direct_req *dreq = container_of(work, struct nfs_direct_req, work);
@@ -742,6 +685,7 @@ static void nfs_direct_write_schedule_work(struct work_struct *work)
 			nfs_direct_write_reschedule(dreq);
 			break;
 		default:
+			nfs_direct_write_clear_reqs(dreq);
 			nfs_zap_mapping(dreq->inode, dreq->inode->i_mapping);
 			nfs_direct_complete(dreq);
 	}
@@ -768,20 +712,15 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 	}
 
 	nfs_direct_count_bytes(dreq, hdr);
-	if (hdr->good_bytes != 0) {
-		if (nfs_write_need_commit(hdr)) {
-			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)
-				request_commit = true;
-			else if (dreq->flags == 0) {
-				nfs_direct_set_hdr_verf(dreq, hdr);
-				request_commit = true;
-				dreq->flags = NFS_ODIRECT_DO_COMMIT;
-			} else if (dreq->flags == NFS_ODIRECT_DO_COMMIT) {
-				request_commit = true;
-				if (nfs_direct_set_or_cmp_hdr_verf(dreq, hdr))
-					dreq->flags =
-						NFS_ODIRECT_RESCHED_WRITES;
-			}
+	if (hdr->good_bytes != 0 && nfs_write_need_commit(hdr)) {
+		switch (dreq->flags) {
+		case 0:
+			dreq->flags = NFS_ODIRECT_DO_COMMIT;
+			request_commit = true;
+			break;
+		case NFS_ODIRECT_RESCHED_WRITES:
+		case NFS_ODIRECT_DO_COMMIT:
+			request_commit = true;
 		}
 	}
 	spin_unlock(&dreq->lock);
@@ -990,11 +929,13 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 	l_ctx = nfs_get_lock_context(dreq->ctx);
 	if (IS_ERR(l_ctx)) {
 		result = PTR_ERR(l_ctx);
+		nfs_direct_req_release(dreq);
 		goto out_release;
 	}
 	dreq->l_ctx = l_ctx;
 	if (!is_sync_kiocb(iocb))
 		dreq->iocb = iocb;
+	pnfs_init_ds_commit_info_ops(&dreq->ds_cinfo, inode);
 
 	nfs_start_io_direct(inode);
 

fs/nfs/filelayout/filelayout.c

@@ -49,6 +49,7 @@ MODULE_AUTHOR("Dean Hildebrand <dhildebz@umich.edu>");
 MODULE_DESCRIPTION("The NFSv4 file layout driver");
 
 #define FILELAYOUT_POLL_RETRY_MAX     (15*HZ)
+static const struct pnfs_commit_ops filelayout_commit_ops;
 
 static loff_t
 filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg,
@@ -750,72 +751,17 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg)
 	/* This assumes a single RW lseg */
 	if (lseg->pls_range.iomode == IOMODE_RW) {
 		struct nfs4_filelayout *flo;
+		struct inode *inode;
 
 		flo = FILELAYOUT_FROM_HDR(lseg->pls_layout);
-		flo->commit_info.nbuckets = 0;
-		kfree(flo->commit_info.buckets);
-		flo->commit_info.buckets = NULL;
+		inode = flo->generic_hdr.plh_inode;
+		spin_lock(&inode->i_lock);
+		pnfs_generic_ds_cinfo_release_lseg(&flo->commit_info, lseg);
+		spin_unlock(&inode->i_lock);
 	}
 	_filelayout_free_lseg(fl);
 }
 
-static int
-filelayout_alloc_commit_info(struct pnfs_layout_segment *lseg,
-			     struct nfs_commit_info *cinfo,
-			     gfp_t gfp_flags)
-{
-	struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
-	struct pnfs_commit_bucket *buckets;
-	int size, i;
-
-	if (fl->commit_through_mds)
-		return 0;
-
-	size = (fl->stripe_type == STRIPE_SPARSE) ?
-		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
-
-	if (cinfo->ds->nbuckets >= size) {
-		/* This assumes there is only one IOMODE_RW lseg.  What
-		 * we really want to do is have a layout_hdr level
-		 * dictionary of <multipath_list4, fh> keys, each
-		 * associated with a struct list_head, populated by calls
-		 * to filelayout_write_pagelist().
-		 * */
-		return 0;
-	}
-
-	buckets = kcalloc(size, sizeof(struct pnfs_commit_bucket),
-			  gfp_flags);
-	if (!buckets)
-		return -ENOMEM;
-	for (i = 0; i < size; i++) {
-		INIT_LIST_HEAD(&buckets[i].written);
-		INIT_LIST_HEAD(&buckets[i].committing);
-		/* mark direct verifier as unset */
-		buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
-	}
-
-	spin_lock(&cinfo->inode->i_lock);
-	if (cinfo->ds->nbuckets >= size)
-		goto out;
-	for (i = 0; i < cinfo->ds->nbuckets; i++) {
-		list_splice(&cinfo->ds->buckets[i].written,
-			    &buckets[i].written);
-		list_splice(&cinfo->ds->buckets[i].committing,
-			    &buckets[i].committing);
-		buckets[i].direct_verf.committed =
-			cinfo->ds->buckets[i].direct_verf.committed;
-		buckets[i].wlseg = cinfo->ds->buckets[i].wlseg;
-		buckets[i].clseg = cinfo->ds->buckets[i].clseg;
-	}
-	swap(cinfo->ds->buckets, buckets);
-	cinfo->ds->nbuckets = size;
-out:
-	spin_unlock(&cinfo->inode->i_lock);
-	kfree(buckets);
-	return 0;
-}
-
 static struct pnfs_layout_segment *
 filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 		      struct nfs4_layoutget_res *lgr,
@@ -938,9 +884,6 @@ static void
 filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 			 struct nfs_page *req)
 {
-	struct nfs_commit_info cinfo;
-	int status;
-
 	pnfs_generic_pg_check_layout(pgio);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode,
@@ -959,17 +902,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 
 	/* If no lseg, fall back to write through mds */
 	if (pgio->pg_lseg == NULL)
-		goto out_mds;
-	nfs_init_cinfo(&cinfo, pgio->pg_inode, pgio->pg_dreq);
-	status = filelayout_alloc_commit_info(pgio->pg_lseg, &cinfo, GFP_NOFS);
-	if (status < 0) {
-		pnfs_put_lseg(pgio->pg_lseg);
-		pgio->pg_lseg = NULL;
-		goto out_mds;
-	}
-	return;
-out_mds:
-	nfs_pageio_reset_write_mds(pgio);
+		nfs_pageio_reset_write_mds(pgio);
 }
 
 static const struct nfs_pageio_ops filelayout_pg_read_ops = {
@@ -1078,36 +1011,6 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how)
 	return -EAGAIN;
 }
 
-/* filelayout_search_commit_reqs - Search lists in @cinfo for the head reqest
- *				   for @page
- * @cinfo - commit info for current inode
- * @page - page to search for matching head request
- *
- * Returns a the head request if one is found, otherwise returns NULL.
- */
-static struct nfs_page *
-filelayout_search_commit_reqs(struct nfs_commit_info *cinfo, struct page *page)
-{
-	struct nfs_page *freq, *t;
-	struct pnfs_commit_bucket *b;
-	int i;
-
-	/* Linearly search the commit lists for each bucket until a matching
-	 * request is found */
-	for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) {
-		list_for_each_entry_safe(freq, t, &b->written, wb_list) {
-			if (freq->wb_page == page)
-				return freq->wb_head;
-		}
-		list_for_each_entry_safe(freq, t, &b->committing, wb_list) {
-			if (freq->wb_page == page)
-				return freq->wb_head;
-		}
-	}
-
-	return NULL;
-}
-
 static int
 filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 			   int how, struct nfs_commit_info *cinfo)
@@ -1140,13 +1043,17 @@ filelayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 	struct nfs4_filelayout *flo;
 
 	flo = kzalloc(sizeof(*flo), gfp_flags);
-	return flo != NULL ? &flo->generic_hdr : NULL;
+	if (flo == NULL)
+		return NULL;
+	pnfs_init_ds_commit_info(&flo->commit_info);
+	flo->commit_info.ops = &filelayout_commit_ops;
+	return &flo->generic_hdr;
 }
 
 static void
 filelayout_free_layout_hdr(struct pnfs_layout_hdr *lo)
 {
-	kfree(FILELAYOUT_FROM_HDR(lo));
+	kfree_rcu(FILELAYOUT_FROM_HDR(lo), generic_hdr.plh_rcu);
 }
 
 static struct pnfs_ds_commit_info *
@@ -1160,6 +1067,46 @@ filelayout_get_ds_info(struct inode *inode)
 		return &FILELAYOUT_FROM_HDR(layout)->commit_info;
 }
 
+static void
+filelayout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
+	struct inode *inode = lseg->pls_layout->plh_inode;
+	struct pnfs_commit_array *array, *new;
+	unsigned int size = (fl->stripe_type == STRIPE_SPARSE) ?
+		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
+
+	new = pnfs_alloc_commit_array(size, GFP_NOIO);
+	if (new) {
+		spin_lock(&inode->i_lock);
+		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
+		spin_unlock(&inode->i_lock);
+		if (array != new)
+			pnfs_free_commit_array(new);
+	}
+}
+
+static void
+filelayout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	pnfs_generic_ds_cinfo_destroy(fl_cinfo);
+	spin_unlock(&inode->i_lock);
+}
+
+static const struct pnfs_commit_ops filelayout_commit_ops = {
+	.setup_ds_info		= filelayout_setup_ds_info,
+	.release_ds_info	= filelayout_release_ds_info,
+	.mark_request_commit	= filelayout_mark_request_commit,
+	.clear_request_commit	= pnfs_generic_clear_request_commit,
+	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
+	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
+	.search_commit_reqs	= pnfs_generic_search_commit_reqs,
+	.commit_pagelist	= filelayout_commit_pagelist,
+};
+
 static struct pnfs_layoutdriver_type filelayout_type = {
 	.id			= LAYOUT_NFSV4_1_FILES,
 	.name			= "LAYOUT_NFSV4_1_FILES",
@@ -1173,12 +1120,6 @@ static struct pnfs_layoutdriver_type filelayout_type = {
 	.pg_read_ops		= &filelayout_pg_read_ops,
 	.pg_write_ops		= &filelayout_pg_write_ops,
 	.get_ds_info		= &filelayout_get_ds_info,
-	.mark_request_commit	= filelayout_mark_request_commit,
-	.clear_request_commit	= pnfs_generic_clear_request_commit,
-	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
-	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
-	.search_commit_reqs	= filelayout_search_commit_reqs,
-	.commit_pagelist	= filelayout_commit_pagelist,
 	.read_pagelist		= filelayout_read_pagelist,
 	.write_pagelist		= filelayout_write_pagelist,
 	.alloc_deviceid_node	= filelayout_alloc_deviceid_node,

fs/nfs/flexfilelayout/flexfilelayout.c

@@ -32,6 +32,7 @@
 
 static unsigned short io_maxretrans;
 
+static const struct pnfs_commit_ops ff_layout_commit_ops;
 static void ff_layout_read_record_layoutstats_done(struct rpc_task *task,
 		struct nfs_pgio_header *hdr);
 static int ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo,
@@ -48,9 +49,11 @@ ff_layout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 
 	ffl = kzalloc(sizeof(*ffl), gfp_flags);
 	if (ffl) {
+		pnfs_init_ds_commit_info(&ffl->commit_info);
 		INIT_LIST_HEAD(&ffl->error_list);
 		INIT_LIST_HEAD(&ffl->mirrors);
 		ffl->last_report_time = ktime_get();
+		ffl->commit_info.ops = &ff_layout_commit_ops;
 		return &ffl->generic_hdr;
 	} else
 		return NULL;
@@ -59,14 +62,14 @@ ff_layout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 static void
 ff_layout_free_layout_hdr(struct pnfs_layout_hdr *lo)
 {
+	struct nfs4_flexfile_layout *ffl = FF_LAYOUT_FROM_HDR(lo);
 	struct nfs4_ff_layout_ds_err *err, *n;
 
-	list_for_each_entry_safe(err, n, &FF_LAYOUT_FROM_HDR(lo)->error_list,
-				 list) {
+	list_for_each_entry_safe(err, n, &ffl->error_list, list) {
 		list_del(&err->list);
 		kfree(err);
 	}
-	kfree(FF_LAYOUT_FROM_HDR(lo));
+	kfree_rcu(ffl, generic_hdr.plh_rcu);
 }
 
 static int decode_pnfs_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid)
@@ -248,36 +251,10 @@ static void ff_layout_put_mirror(struct nfs4_ff_layout_mirror *mirror)
 
 static void ff_layout_free_mirror_array(struct nfs4_ff_layout_segment *fls)
 {
-	int i;
-
-	if (fls->mirror_array) {
-		for (i = 0; i < fls->mirror_array_cnt; i++) {
-			/* normally mirror_ds is freed in
-			 * .free_deviceid_node but we still do it here
-			 * for .alloc_lseg error path */
-			ff_layout_put_mirror(fls->mirror_array[i]);
-		}
-		kfree(fls->mirror_array);
-		fls->mirror_array = NULL;
-	}
-}
-
-static int ff_layout_check_layout(struct nfs4_layoutget_res *lgr)
-{
-	int ret = 0;
+	u32 i;
 
-	dprintk("--> %s\n", __func__);
-
-	/* FIXME: remove this check when layout segment support is added */
-	if (lgr->range.offset != 0 ||
-	    lgr->range.length != NFS4_MAX_UINT64) {
-		dprintk("%s Only whole file layouts supported. Use MDS i/o\n",
-			__func__);
-		ret = -EINVAL;
-	}
-
-	dprintk("--> %s returns %d\n", __func__, ret);
-	return ret;
+	for (i = 0; i < fls->mirror_array_cnt; i++)
+		ff_layout_put_mirror(fls->mirror_array[i]);
 }
 
 static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls)
@@ -288,6 +265,23 @@ static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls)
 	}
 }
 
+static bool
+ff_lseg_match_mirrors(struct pnfs_layout_segment *l1,
+		struct pnfs_layout_segment *l2)
+{
+	const struct nfs4_ff_layout_segment *fl1 = FF_LAYOUT_LSEG(l1);
+	const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l1);
+	u32 i;
+
+	if (fl1->mirror_array_cnt != fl2->mirror_array_cnt)
+		return false;
+	for (i = 0; i < fl1->mirror_array_cnt; i++) {
+		if (fl1->mirror_array[i] != fl2->mirror_array[i])
+			return false;
+	}
+	return true;
+}
+
 static bool
 ff_lseg_range_is_after(const struct pnfs_layout_range *l1,
 		const struct pnfs_layout_range *l2)
@@ -323,6 +317,8 @@ ff_lseg_merge(struct pnfs_layout_segment *new,
 			new->pls_range.length);
 	if (new_end < old->pls_range.offset)
 		return false;
+	if (!ff_lseg_match_mirrors(new, old))
+		return false;
 
 	/* Mergeable: copy info from 'old' to 'new' */
 	if (new_end < old_end)
@@ -400,16 +396,13 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 		goto out_err_free;
 
 	rc = -ENOMEM;
-	fls = kzalloc(sizeof(*fls), gfp_flags);
+	fls = kzalloc(struct_size(fls, mirror_array, mirror_array_cnt),
+			gfp_flags);
 	if (!fls)
 		goto out_err_free;
 
 	fls->mirror_array_cnt = mirror_array_cnt;
 	fls->stripe_unit = stripe_unit;
-	fls->mirror_array = kcalloc(fls->mirror_array_cnt,
-				    sizeof(fls->mirror_array[0]), gfp_flags);
-	if (fls->mirror_array == NULL)
-		goto out_err_free;
 
 	for (i = 0; i < fls->mirror_array_cnt; i++) {
 		struct nfs4_ff_layout_mirror *mirror;
@@ -545,9 +538,6 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 
 out_sort_mirrors:
 	ff_layout_sort_mirrors(fls);
-	rc = ff_layout_check_layout(lgr);
-	if (rc)
-		goto out_err_free;
 	ret = &fls->generic_hdr;
 	dprintk("<-- %s (success)\n", __func__);
 out_free_page:
@@ -560,17 +550,6 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 	goto out_free_page;
 }
 
-static bool ff_layout_has_rw_segments(struct pnfs_layout_hdr *layout)
-{
-	struct pnfs_layout_segment *lseg;
-
-	list_for_each_entry(lseg, &layout->plh_segs, pls_list)
-		if (lseg->pls_range.iomode == IOMODE_RW)
-			return true;
-
-	return false;
-}
-
 static void
 ff_layout_free_lseg(struct pnfs_layout_segment *lseg)
 {
@@ -585,23 +564,12 @@ ff_layout_free_lseg(struct pnfs_layout_segment *lseg)
 		ffl = FF_LAYOUT_FROM_HDR(lseg->pls_layout);
 		inode = ffl->generic_hdr.plh_inode;
 		spin_lock(&inode->i_lock);
-		if (!ff_layout_has_rw_segments(lseg->pls_layout)) {
-			ffl->commit_info.nbuckets = 0;
-			kfree(ffl->commit_info.buckets);
-			ffl->commit_info.buckets = NULL;
-		}
+		pnfs_generic_ds_cinfo_release_lseg(&ffl->commit_info, lseg);
 		spin_unlock(&inode->i_lock);
 	}
 	_ff_layout_free_lseg(fls);
 }
 
-/* Return 1 until we have multiple lsegs support */
-static int
-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, ktime_t now)
 {
@@ -746,52 +714,6 @@ nfs4_ff_layout_stat_io_end_write(struct rpc_task *task,
 	spin_unlock(&mirror->lock);
 }
 
-static int
-ff_layout_alloc_commit_info(struct pnfs_layout_segment *lseg,
-			    struct nfs_commit_info *cinfo,
-			    gfp_t gfp_flags)
-{
-	struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg);
-	struct pnfs_commit_bucket *buckets;
-	int size;
-
-	if (cinfo->ds->nbuckets != 0) {
-		/* This assumes there is only one RW lseg per file.
-		 * To support multiple lseg per file, we need to
-		 * change struct pnfs_commit_bucket to allow dynamic
-		 * increasing nbuckets.
-		 */
-		return 0;
-	}
-
-	size = ff_layout_get_lseg_count(fls) * FF_LAYOUT_MIRROR_COUNT(lseg);
-
-	buckets = kcalloc(size, sizeof(struct pnfs_commit_bucket),
-			  gfp_flags);
-	if (!buckets)
-		return -ENOMEM;
-	else {
-		int i;
-
-		spin_lock(&cinfo->inode->i_lock);
-		if (cinfo->ds->nbuckets != 0)
-			kfree(buckets);
-		else {
-			cinfo->ds->buckets = buckets;
-			cinfo->ds->nbuckets = size;
-			for (i = 0; i < size; i++) {
-				INIT_LIST_HEAD(&buckets[i].written);
-				INIT_LIST_HEAD(&buckets[i].committing);
-				/* mark direct verifier as unset */
-				buckets[i].direct_verf.committed =
-					NFS_INVALID_STABLE_HOW;
-			}
-		}
-		spin_unlock(&cinfo->inode->i_lock);
-		return 0;
-	}
-}
-
 static void
 ff_layout_mark_ds_unreachable(struct pnfs_layout_segment *lseg, int idx)
 {
@@ -876,8 +798,8 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio,
 	pnfs_put_lseg(pgio->pg_lseg);
 	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 					   nfs_req_openctx(req),
-					   0,
-					   NFS4_MAX_UINT64,
+					   req_offset(req),
+					   req->wb_bytes,
 					   IOMODE_READ,
 					   strict_iomode,
 					   GFP_KERNEL);
@@ -887,6 +809,14 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio,
 	}
 }
 
+static void
+ff_layout_pg_check_layout(struct nfs_pageio_descriptor *pgio,
+			  struct nfs_page *req)
+{
+	pnfs_generic_pg_check_layout(pgio);
+	pnfs_generic_pg_check_range(pgio, req);
+}
+
 static void
 ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 			struct nfs_page *req)
@@ -897,7 +827,7 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	int ds_idx;
 
 retry:
-	pnfs_generic_pg_check_layout(pgio);
+	ff_layout_pg_check_layout(pgio, req);
 	/* Use full layout for now */
 	if (!pgio->pg_lseg) {
 		ff_layout_pg_get_read(pgio, req, false);
@@ -953,18 +883,16 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 {
 	struct nfs4_ff_layout_mirror *mirror;
 	struct nfs_pgio_mirror *pgm;
-	struct nfs_commit_info cinfo;
 	struct nfs4_pnfs_ds *ds;
 	int i;
-	int status;
 
 retry:
-	pnfs_generic_pg_check_layout(pgio);
+	ff_layout_pg_check_layout(pgio, req);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   nfs_req_openctx(req),
-						   0,
-						   NFS4_MAX_UINT64,
+						   req_offset(req),
+						   req->wb_bytes,
 						   IOMODE_RW,
 						   false,
 						   GFP_NOFS);
@@ -978,11 +906,6 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 	if (pgio->pg_lseg == NULL)
 		goto out_mds;
 
-	nfs_init_cinfo(&cinfo, pgio->pg_inode, pgio->pg_dreq);
-	status = ff_layout_alloc_commit_info(pgio->pg_lseg, &cinfo, GFP_NOFS);
-	if (status < 0)
-		goto out_mds;
-
 	/* Use a direct mapping of ds_idx to pgio mirror_idx */
 	if (WARN_ON_ONCE(pgio->pg_mirror_count !=
 	    FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg)))
@@ -1297,21 +1220,23 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 		}
 	}
 
+	mirror = FF_LAYOUT_COMP(lseg, idx);
+	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
+				       mirror, offset, length, status, opnum,
+				       GFP_NOIO);
+
 	switch (status) {
 	case NFS4ERR_DELAY:
 	case NFS4ERR_GRACE:
-		return;
-	default:
 		break;
+	case NFS4ERR_NXIO:
+		ff_layout_mark_ds_unreachable(lseg, idx);
+		/* Fallthrough */
+	default:
+		pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode,
+						  lseg);
 	}
 
-	mirror = FF_LAYOUT_COMP(lseg, idx);
-	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
-				       mirror, offset, length, status, opnum,
-				       GFP_NOIO);
-	if (status == NFS4ERR_NXIO)
-		ff_layout_mark_ds_unreachable(lseg, idx);
-	pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode, lseg);
 	dprintk("%s: err %d op %d status %u\n", __func__, err, opnum, status);
 }
 
@@ -2011,6 +1936,33 @@ ff_layout_get_ds_info(struct inode *inode)
 	return &FF_LAYOUT_FROM_HDR(layout)->commit_info;
 }
 
+static void
+ff_layout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg);
+	struct inode *inode = lseg->pls_layout->plh_inode;
+	struct pnfs_commit_array *array, *new;
+
+	new = pnfs_alloc_commit_array(flseg->mirror_array_cnt, GFP_NOIO);
+	if (new) {
+		spin_lock(&inode->i_lock);
+		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
+		spin_unlock(&inode->i_lock);
+		if (array != new)
+			pnfs_free_commit_array(new);
+	}
+}
+
+static void
+ff_layout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	pnfs_generic_ds_cinfo_destroy(fl_cinfo);
+	spin_unlock(&inode->i_lock);
+}
+
 static void
 ff_layout_free_deviceid_node(struct nfs4_deviceid_node *d)
 {
@@ -2496,6 +2448,16 @@ ff_layout_set_layoutdriver(struct nfs_server *server,
 	return 0;
 }
 
+static const struct pnfs_commit_ops ff_layout_commit_ops = {
+	.setup_ds_info		= ff_layout_setup_ds_info,
+	.release_ds_info	= ff_layout_release_ds_info,
+	.mark_request_commit	= pnfs_layout_mark_request_commit,
+	.clear_request_commit	= pnfs_generic_clear_request_commit,
+	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
+	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
+	.commit_pagelist	= ff_layout_commit_pagelist,
+};
+
 static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.id			= LAYOUT_FLEX_FILES,
 	.name			= "LAYOUT_FLEX_FILES",
@@ -2512,11 +2474,6 @@ static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.pg_write_ops		= &ff_layout_pg_write_ops,
 	.get_ds_info		= ff_layout_get_ds_info,
 	.free_deviceid_node	= ff_layout_free_deviceid_node,
-	.mark_request_commit	= pnfs_layout_mark_request_commit,
-	.clear_request_commit	= pnfs_generic_clear_request_commit,
-	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
-	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
-	.commit_pagelist	= ff_layout_commit_pagelist,
 	.read_pagelist		= ff_layout_read_pagelist,
 	.write_pagelist		= ff_layout_write_pagelist,
 	.alloc_deviceid_node    = ff_layout_alloc_deviceid_node,

fs/nfs/flexfilelayout/flexfilelayout.h

@@ -99,7 +99,7 @@ struct nfs4_ff_layout_segment {
 	u64				stripe_unit;
 	u32				flags;
 	u32				mirror_array_cnt;
-	struct nfs4_ff_layout_mirror	**mirror_array;
+	struct nfs4_ff_layout_mirror	*mirror_array[];
 };
 
 struct nfs4_flexfile_layout {

fs/nfs/fs_context.c

@@ -190,6 +190,7 @@ static const struct constant_table nfs_vers_tokens[] = {
 	{ "4.0",	Opt_vers_4_0 },
 	{ "4.1",	Opt_vers_4_1 },
 	{ "4.2",	Opt_vers_4_2 },
+	{}
 };
 
 enum {
@@ -202,13 +203,14 @@ enum {
 	nr__Opt_xprt
 };
 
-static const struct constant_table nfs_xprt_protocol_tokens[nr__Opt_xprt] = {
+static const struct constant_table nfs_xprt_protocol_tokens[] = {
 	{ "rdma",	Opt_xprt_rdma },
 	{ "rdma6",	Opt_xprt_rdma6 },
 	{ "tcp",	Opt_xprt_tcp },
 	{ "tcp6",	Opt_xprt_tcp6 },
 	{ "udp",	Opt_xprt_udp },
 	{ "udp6",	Opt_xprt_udp6 },
+	{}
 };
 
 enum {
@@ -239,6 +241,7 @@ static const struct constant_table nfs_secflavor_tokens[] = {
 	{ "spkm3i",	Opt_sec_spkmi },
 	{ "spkm3p",	Opt_sec_spkmp },
 	{ "sys",	Opt_sec_sys },
+	{}
 };
 
 /*
@@ -1135,7 +1138,7 @@ static int nfs4_parse_monolithic(struct fs_context *fc,
 	return nfs_invalf(fc, "NFS4: mount program didn't pass remote address");
 
 out_invalid_transport_udp:
-	return nfs_invalf(fc, "NFSv4: Unsupported transport protocol udp");
+	return nfs_invalf(fc, "NFS: Unsupported transport protocol udp");
 }
 #endif
 
@@ -1257,7 +1260,7 @@ static int nfs_fs_context_validate(struct fs_context *fc)
 	nfs_errorf(fc, "NFS: NFSv4 is not compiled into kernel");
 	return -EPROTONOSUPPORT;
 out_invalid_transport_udp:
-	return nfs_invalf(fc, "NFSv4: Unsupported transport protocol udp");
+	return nfs_invalf(fc, "NFS: Unsupported transport protocol udp");
 out_no_address:
 	return nfs_invalf(fc, "NFS: mount program didn't pass remote address");
 out_mountproto_mismatch:

fs/nfs/inode.c

@@ -62,7 +62,6 @@
 /* Default is to see 64-bit inode numbers */
 static bool enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED;
 
-static void nfs_invalidate_inode(struct inode *);
 static int nfs_update_inode(struct inode *, struct nfs_fattr *);
 
 static struct kmem_cache * nfs_inode_cachep;
@@ -284,10 +283,18 @@ EXPORT_SYMBOL_GPL(nfs_invalidate_atime);
  * Invalidate, but do not unhash, the inode.
  * NB: must be called with inode->i_lock held!
  */
-static void nfs_invalidate_inode(struct inode *inode)
+static void nfs_set_inode_stale_locked(struct inode *inode)
 {
 	set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
 	nfs_zap_caches_locked(inode);
+	trace_nfs_set_inode_stale(inode);
+}
+
+void nfs_set_inode_stale(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	nfs_set_inode_stale_locked(inode);
+	spin_unlock(&inode->i_lock);
 }
 
 struct nfs_find_desc {
@@ -959,16 +966,16 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry,
 						struct file *filp)
 {
 	struct nfs_open_context *ctx;
-	const struct cred *cred = get_current_cred();
 
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
-	if (!ctx) {
-		put_cred(cred);
+	if (!ctx)
 		return ERR_PTR(-ENOMEM);
-	}
 	nfs_sb_active(dentry->d_sb);
 	ctx->dentry = dget(dentry);
-	ctx->cred = cred;
+	if (filp)
+		ctx->cred = get_cred(filp->f_cred);
+	else
+		ctx->cred = get_current_cred();
 	ctx->ll_cred = NULL;
 	ctx->state = NULL;
 	ctx->mode = f_mode;
@@ -1163,9 +1170,10 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 				status = 0;
 			break;
 		case -ESTALE:
-			nfs_zap_caches(inode);
 			if (!S_ISDIR(inode->i_mode))
-				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+				nfs_set_inode_stale(inode);
+			else
+				nfs_zap_caches(inode);
 		}
 		goto err_out;
 	}
@@ -2064,7 +2072,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	 * lookup validation will know that the inode is bad.
 	 * (But we fall through to invalidate the caches.)
 	 */
-	nfs_invalidate_inode(inode);
+	nfs_set_inode_stale_locked(inode);
 	return -ESTALE;
 }
 

fs/nfs/internal.h

@@ -274,12 +274,6 @@ void nfs_free_request(struct nfs_page *req);
 struct nfs_pgio_mirror *
 nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc);
 
-static inline bool nfs_pgio_has_mirroring(struct nfs_pageio_descriptor *desc)
-{
-	WARN_ON_ONCE(desc->pg_mirror_count < 1);
-	return desc->pg_mirror_count > 1;
-}
-
 static inline bool nfs_match_open_context(const struct nfs_open_context *ctx1,
 		const struct nfs_open_context *ctx2)
 {
@@ -417,7 +411,9 @@ extern int __init register_nfs_fs(void);
 extern void __exit unregister_nfs_fs(void);
 extern bool nfs_sb_active(struct super_block *sb);
 extern void nfs_sb_deactive(struct super_block *sb);
-
+extern int nfs_client_for_each_server(struct nfs_client *clp,
+				      int (*fn)(struct nfs_server *, void *),
+				      void *data);
 /* io.c */
 extern void nfs_start_io_read(struct inode *inode);
 extern void nfs_end_io_read(struct inode *inode);
@@ -515,13 +511,25 @@ int nfs_filemap_write_and_wait_range(struct address_space *mapping,
 		loff_t lstart, loff_t lend);
 
 #ifdef CONFIG_NFS_V4_1
+static inline void
+pnfs_bucket_clear_pnfs_ds_commit_verifiers(struct pnfs_commit_bucket *buckets,
+		unsigned int nbuckets)
+{
+	unsigned int i;
+
+	for (i = 0; i < nbuckets; i++)
+		buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
+}
 static inline
 void nfs_clear_pnfs_ds_commit_verifiers(struct pnfs_ds_commit_info *cinfo)
 {
-	int i;
+	struct pnfs_commit_array *array;
 
-	for (i = 0; i < cinfo->nbuckets; i++)
-		cinfo->buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
+	rcu_read_lock();
+	list_for_each_entry_rcu(array, &cinfo->commits, cinfo_list)
+		pnfs_bucket_clear_pnfs_ds_commit_verifiers(array->buckets,
+				array->nbuckets);
+	rcu_read_unlock();
 }
 #else
 static inline
@@ -542,6 +550,14 @@ nfs_write_verifier_cmp(const struct nfs_write_verifier *v1,
 	return memcmp(v1->data, v2->data, sizeof(v1->data));
 }
 
+static inline bool
+nfs_write_match_verf(const struct nfs_writeverf *verf,
+		struct nfs_page *req)
+{
+	return verf->committed > NFS_UNSTABLE &&
+		!nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier);
+}
+
 /* unlink.c */
 extern struct rpc_task *
 nfs_async_rename(struct inode *old_dir, struct inode *new_dir,

fs/nfs/namespace.c

@@ -145,6 +145,7 @@ struct vfsmount *nfs_d_automount(struct path *path)
 	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
 	struct nfs_server *server = NFS_SERVER(d_inode(path->dentry));
 	struct nfs_client *client = server->nfs_client;
+	int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout);
 	int ret;
 
 	if (IS_ROOT(path->dentry))
@@ -190,12 +191,12 @@ struct vfsmount *nfs_d_automount(struct path *path)
 	if (IS_ERR(mnt))
 		goto out_fc;
 
-	if (nfs_mountpoint_expiry_timeout < 0)
+	mntget(mnt); /* prevent immediate expiration */
+	if (timeout <= 0)
 		goto out_fc;
 
-	mntget(mnt); /* prevent immediate expiration */
 	mnt_set_expiry(mnt, &nfs_automount_list);
-	schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+	schedule_delayed_work(&nfs_automount_task, timeout);
 
 out_fc:
 	put_fs_context(fc);
@@ -233,10 +234,11 @@ const struct inode_operations nfs_referral_inode_operations = {
 static void nfs_expire_automounts(struct work_struct *work)
 {
 	struct list_head *list = &nfs_automount_list;
+	int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout);
 
 	mark_mounts_for_expiry(list);
-	if (!list_empty(list))
-		schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+	if (!list_empty(list) && timeout > 0)
+		schedule_delayed_work(&nfs_automount_task, timeout);
 }
 
 void nfs_release_automount_timer(void)
@@ -247,10 +249,7 @@ void nfs_release_automount_timer(void)
 
 /**
  * nfs_do_submount - set up mountpoint when crossing a filesystem boundary
- * @dentry: parent directory
- * @fh: filehandle for new root dentry
- * @fattr: attributes for new root inode
- * @authflavor: security flavor to use when performing the mount
+ * @fc: pointer to struct nfs_fs_context
  *
  */
 int nfs_do_submount(struct fs_context *fc)
@@ -312,3 +311,53 @@ int nfs_submount(struct fs_context *fc, struct nfs_server *server)
 	return nfs_do_submount(fc);
 }
 EXPORT_SYMBOL_GPL(nfs_submount);
+
+static int param_set_nfs_timeout(const char *val, const struct kernel_param *kp)
+{
+	long num;
+	int ret;
+
+	if (!val)
+		return -EINVAL;
+	ret = kstrtol(val, 0, &num);
+	if (ret)
+		return -EINVAL;
+	if (num > 0) {
+		if (num >= INT_MAX / HZ)
+			num = INT_MAX;
+		else
+			num *= HZ;
+		*((int *)kp->arg) = num;
+		if (!list_empty(&nfs_automount_list))
+			mod_delayed_work(system_wq, &nfs_automount_task, num);
+	} else {
+		*((int *)kp->arg) = -1*HZ;
+		cancel_delayed_work(&nfs_automount_task);
+	}
+	return 0;
+}
+
+static int param_get_nfs_timeout(char *buffer, const struct kernel_param *kp)
+{
+	long num = *((int *)kp->arg);
+
+	if (num > 0) {
+		if (num >= INT_MAX - (HZ - 1))
+			num = INT_MAX / HZ;
+		else
+			num = (num + (HZ - 1)) / HZ;
+	} else
+		num = -1;
+	return scnprintf(buffer, PAGE_SIZE, "%li\n", num);
+}
+
+static const struct kernel_param_ops param_ops_nfs_timeout = {
+	.set = param_set_nfs_timeout,
+	.get = param_get_nfs_timeout,
+};
+#define param_check_nfs_timeout(name, p) __param_check(name, p, int);
+
+module_param(nfs_mountpoint_expiry_timeout, nfs_timeout, 0644);
+MODULE_PARM_DESC(nfs_mountpoint_expiry_timeout,
+		"Set the NFS automounted mountpoint timeout value (seconds)."
+		"Values <= 0 turn expiration off.");

fs/nfs/nfs4_fs.h

@@ -42,7 +42,9 @@ enum nfs4_client_state {
 	NFS4CLNT_LEASE_MOVED,
 	NFS4CLNT_DELEGATION_EXPIRED,
 	NFS4CLNT_RUN_MANAGER,
-	NFS4CLNT_DELEGRETURN_RUNNING,
+	NFS4CLNT_RECALL_RUNNING,
+	NFS4CLNT_RECALL_ANY_LAYOUT_READ,
+	NFS4CLNT_RECALL_ANY_LAYOUT_RW,
 };
 
 #define NFS4_RENEW_TIMEOUT		0x01

fs/nfs/nfs4file.c

@@ -252,6 +252,9 @@ static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off,
 	if (remap_flags & ~REMAP_FILE_ADVISORY)
 		return -EINVAL;
 
+	if (IS_SWAPFILE(dst_inode) || IS_SWAPFILE(src_inode))
+		return -ETXTBSY;
+
 	/* check alignment w.r.t. clone_blksize */
 	ret = -EINVAL;
 	if (bs) {

fs/nfs/nfs4namespace.c

@@ -354,7 +354,7 @@ static int try_location(struct fs_context *fc,
 
 /**
  * nfs_follow_referral - set up mountpoint when hitting a referral on moved error
- * @dentry: parent directory
+ * @fc: pointer to struct nfs_fs_context
  * @locations: array of NFSv4 server location information
  *
  */

fs/nfs/nfs4proc.c

@@ -2346,7 +2346,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data)
 		.callback_ops = &nfs4_open_confirm_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status;
 
@@ -2511,7 +2511,7 @@ static int nfs4_run_open_task(struct nfs4_opendata *data,
 		.callback_ops = &nfs4_open_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status;
 
@@ -2790,16 +2790,19 @@ static int nfs41_check_delegation_stateid(struct nfs4_state *state)
 		return NFS_OK;
 	}
 
+	spin_lock(&delegation->lock);
 	nfs4_stateid_copy(&stateid, &delegation->stateid);
 
 	if (!test_and_clear_bit(NFS_DELEGATION_TEST_EXPIRED,
 				&delegation->flags)) {
+		spin_unlock(&delegation->lock);
 		rcu_read_unlock();
 		return NFS_OK;
 	}
 
 	if (delegation->cred)
 		cred = get_cred(delegation->cred);
+	spin_unlock(&delegation->lock);
 	rcu_read_unlock();
 	status = nfs41_test_and_free_expired_stateid(server, &stateid, cred);
 	trace_nfs4_test_delegation_stateid(state, NULL, status);
@@ -3651,7 +3654,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_close_ops,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status = -ENOMEM;
 
@@ -5544,7 +5547,7 @@ static int buf_to_pages_noslab(const void *buf, size_t buflen,
 struct nfs4_cached_acl {
 	int cached;
 	size_t len;
-	char data[0];
+	char data[];
 };
 
 static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl)
@@ -6253,6 +6256,7 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 		/* Fallthrough */
 	case -NFS4ERR_BAD_STATEID:
 	case -NFS4ERR_STALE_STATEID:
+	case -ETIMEDOUT:
 		task->tk_status = 0;
 		break;
 	case -NFS4ERR_OLD_STATEID:
@@ -6343,7 +6347,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred,
 		.rpc_client = server->client,
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_delegreturn_ops,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF | RPC_TASK_TIMEOUT,
 	};
 	int status = 0;
 
@@ -6926,7 +6930,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_lock_ops,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int ret;
 
@@ -9170,7 +9174,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_layoutget_call_ops,
 		.callback_data = lgp,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	struct pnfs_layout_segment *lseg = NULL;
 	struct nfs4_exception exception = {
@@ -9287,6 +9291,7 @@ static void nfs4_layoutreturn_release(void *calldata)
 		lrp->ld_private.ops->free(&lrp->ld_private);
 	pnfs_put_layout_hdr(lrp->args.layout);
 	nfs_iput_and_deactive(lrp->inode);
+	put_cred(lrp->cred);
 	kfree(calldata);
 	dprintk("<-- %s\n", __func__);
 }

fs/nfs/nfs4state.c

@@ -2524,6 +2524,21 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 	}
 	return 0;
 }
+
+static void nfs4_layoutreturn_any_run(struct nfs_client *clp)
+{
+	int iomode = 0;
+
+	if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &clp->cl_state))
+		iomode += IOMODE_READ;
+	if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &clp->cl_state))
+		iomode += IOMODE_RW;
+	/* Note: IOMODE_READ + IOMODE_RW == IOMODE_ANY */
+	if (iomode) {
+		pnfs_layout_return_unused_byclid(clp, iomode);
+		set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
+	}
+}
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
 
@@ -2531,6 +2546,10 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 {
 	return 0;
 }
+
+static void nfs4_layoutreturn_any_run(struct nfs_client *clp)
+{
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void nfs4_state_manager(struct nfs_client *clp)
@@ -2635,12 +2654,13 @@ static void nfs4_state_manager(struct nfs_client *clp)
 		nfs4_end_drain_session(clp);
 		nfs4_clear_state_manager_bit(clp);
 
-		if (!test_and_set_bit(NFS4CLNT_DELEGRETURN_RUNNING, &clp->cl_state)) {
+		if (!test_and_set_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state)) {
 			if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) {
 				nfs_client_return_marked_delegations(clp);
 				set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
 			}
-			clear_bit(NFS4CLNT_DELEGRETURN_RUNNING, &clp->cl_state);
+			nfs4_layoutreturn_any_run(clp);
+			clear_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state);
 		}
 
 		/* Did we race with an attempt to give us more work? */

fs/nfs/nfs4trace.h

@@ -584,7 +584,9 @@ TRACE_DEFINE_ENUM(NFS4CLNT_MOVED);
 TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_MOVED);
 TRACE_DEFINE_ENUM(NFS4CLNT_DELEGATION_EXPIRED);
 TRACE_DEFINE_ENUM(NFS4CLNT_RUN_MANAGER);
-TRACE_DEFINE_ENUM(NFS4CLNT_DELEGRETURN_RUNNING);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_RUNNING);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_READ);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_RW);
 
 #define show_nfs4_clp_state(state) \
 	__print_flags(state, "|", \
@@ -605,7 +607,9 @@ TRACE_DEFINE_ENUM(NFS4CLNT_DELEGRETURN_RUNNING);
 		{ NFS4CLNT_LEASE_MOVED,		"LEASE_MOVED" }, \
 		{ NFS4CLNT_DELEGATION_EXPIRED,	"DELEGATION_EXPIRED" }, \
 		{ NFS4CLNT_RUN_MANAGER,		"RUN_MANAGER" }, \
-		{ NFS4CLNT_DELEGRETURN_RUNNING,	"DELEGRETURN_RUNNING" })
+		{ NFS4CLNT_RECALL_RUNNING,	"RECALL_RUNNING" }, \
+		{ NFS4CLNT_RECALL_ANY_LAYOUT_READ, "RECALL_ANY_LAYOUT_READ" }, \
+		{ NFS4CLNT_RECALL_ANY_LAYOUT_RW, "RECALL_ANY_LAYOUT_RW" })
 
 TRACE_EVENT(nfs4_state_mgr,
 		TP_PROTO(

fs/nfs/nfsroot.c

@@ -88,7 +88,7 @@
 #define NFS_ROOT		"/tftpboot/%s"
 
 /* Default NFSROOT mount options. */
-#define NFS_DEF_OPTIONS		"vers=2,udp,rsize=4096,wsize=4096"
+#define NFS_DEF_OPTIONS		"vers=2,tcp,rsize=4096,wsize=4096"
 
 /* Parameters passed from the kernel command line */
 static char nfs_root_parms[NFS_MAXPATHLEN + 1] __initdata = "";