Merge tag 'nfs-for-4.10-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

   - further attribute cache improvements to make revalidation more fine
     grained

   - NFSv4 locking improvements

  Bugfixes:

   - nfs4_fl_prepare_ds must be careful about reporting success in files
     layout

   - pNFS/flexfiles: Instead of marking a device inactive, remove it
     from the cache"

* tag 'nfs-for-4.10-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFSv4: Retry the DELEGRETURN if the embedded GETATTR is rejected with EACCES
  NFS: Retry the CLOSE if the embedded GETATTR is rejected with EACCES
  NFSv4: Place the GETATTR operation before the CLOSE
  NFSv4: Also ask for attributes when downgrading to a READ-only state
  NFS: Don't abuse NFS_INO_REVAL_FORCED in nfs_post_op_update_inode_locked()
  pNFS: Return RW layouts on OPEN_DOWNGRADE
  NFSv4: Add encode/decode of the layoutreturn op in OPEN_DOWNGRADE
  NFS: Don't disconnect open-owner on NFS4ERR_BAD_SEQID
  NFSv4: ensure __nfs4_find_lock_state returns consistent result.
  NFSv4.1: nfs4_fl_prepare_ds must be careful about reporting success.
  pNFS/flexfiles: delete deviceid, don't mark inactive
  NFS: Clean up nfs_attribute_timeout()
  NFS: Remove unused function nfs_revalidate_inode_rcu()
  NFS: Fix and clean up the access cache validity checking
  NFS: Only look at the change attribute cache state in nfs_weak_revalidate()
  NFS: Clean up cache validity checking
  NFS: Don't revalidate the file on close if we hold a delegation
  NFSv4: Don't discard the attributes returned by asynchronous DELEGRETURN
  NFSv4: Update the attribute cache info in update_changeattr

11 files changed, 166 insertions, 109 deletions

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index cb22a9f..fad8104 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1273,8 +1273,8 @@ out_error:
  */
 static int nfs_weak_revalidate(struct dentry *dentry, unsigned int flags)
 {
-	int error;
 	struct inode *inode = d_inode(dentry);
+	int error = 0;
 
 	/*
 	 * I believe we can only get a negative dentry here in the case of a
@@ -1293,7 +1293,8 @@ static int nfs_weak_revalidate(struct dentry *dentry, unsigned int flags)
 		return 0;
 	}
 
-	error = nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	if (nfs_mapping_need_revalidate_inode(inode))
+		error = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
 	dfprintk(LOOKUPCACHE, "NFS: %s: inode %lu is %s\n",
 			__func__, inode->i_ino, error ? "invalid" : "valid");
 	return !error;
@@ -2285,8 +2286,7 @@ static int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, str
 		if (cache == NULL)
 			goto out;
 		/* Found an entry, is our attribute cache valid? */
-		if (!nfs_attribute_cache_expired(inode) &&
-		    !(nfsi->cache_validity & NFS_INO_INVALID_ATTR))
+		if (!nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS))
 			break;
 		err = -ECHILD;
 		if (!may_block)
@@ -2334,12 +2334,12 @@ static int nfs_access_get_cached_rcu(struct inode *inode, struct rpc_cred *cred,
 		cache = NULL;
 	if (cache == NULL)
 		goto out;
-	err = nfs_revalidate_inode_rcu(NFS_SERVER(inode), inode);
-	if (err)
+	if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS))
 		goto out;
 	res->jiffies = cache->jiffies;
 	res->cred = cache->cred;
 	res->mask = cache->mask;
+	err = 0;
 out:
 	rcu_read_unlock();
 	return err;
@@ -2491,12 +2491,13 @@ EXPORT_SYMBOL_GPL(nfs_may_open);
 static int nfs_execute_ok(struct inode *inode, int mask)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
-	int ret;
+	int ret = 0;
 
-	if (mask & MAY_NOT_BLOCK)
-		ret = nfs_revalidate_inode_rcu(server, inode);
-	else
-		ret = nfs_revalidate_inode(server, inode);
+	if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) {
+		if (mask & MAY_NOT_BLOCK)
+			return -ECHILD;
+		ret = __nfs_revalidate_inode(server, inode);
+	}
 	if (ret == 0 && !execute_ok(inode))
 		ret = -EACCES;
 	return ret;
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 55208b9..157cb43 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -101,21 +101,11 @@ EXPORT_SYMBOL_GPL(nfs_file_release);
 static int nfs_revalidate_file_size(struct inode *inode, struct file *filp)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_inode *nfsi = NFS_I(inode);
-	const unsigned long force_reval = NFS_INO_REVAL_PAGECACHE|NFS_INO_REVAL_FORCED;
-	unsigned long cache_validity = nfsi->cache_validity;
-
-	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ) &&
-	    (cache_validity & force_reval) != force_reval)
-		goto out_noreval;
 
 	if (filp->f_flags & O_DIRECT)
 		goto force_reval;
-	if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE)
-		goto force_reval;
-	if (nfs_attribute_timeout(inode))
+	if (nfs_check_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE))
 		goto force_reval;
-out_noreval:
 	return 0;
 force_reval:
 	return __nfs_revalidate_inode(server, inode);
diff --git a/fs/nfs/filelayout/filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c
index a5589b7..f956ca2 100644
--- a/fs/nfs/filelayout/filelayoutdev.c
+++ b/fs/nfs/filelayout/filelayoutdev.c
@@ -282,7 +282,8 @@ nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx)
 			     s->nfs_client->cl_minorversion);
 
 out_test_devid:
-	if (filelayout_test_devid_unavailable(devid))
+	if (ret->ds_clp == NULL ||
+	    filelayout_test_devid_unavailable(devid))
 		ret = NULL;
 out:
 	return ret;
diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index 9e111d0..45962fe 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -1126,7 +1126,8 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task,
 	case -EPIPE:
 		dprintk("%s DS connection error %d\n", __func__,
 			task->tk_status);
-		nfs4_mark_deviceid_unavailable(devid);
+		nfs4_delete_deviceid(devid->ld, devid->nfs_client,
+				&devid->deviceid);
 		rpc_wake_up(&tbl->slot_tbl_waitq);
 		/* fall through */
 	default:
@@ -1175,7 +1176,8 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task,
 	default:
 		dprintk("%s DS connection error %d\n", __func__,
 			task->tk_status);
-		nfs4_mark_deviceid_unavailable(devid);
+		nfs4_delete_deviceid(devid->ld, devid->nfs_client,
+				&devid->deviceid);
 	}
 	/* FIXME: Need to prevent infinite looping here. */
 	return -NFS4ERR_RESET_TO_PNFS;
diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c
index 3cc39d1..e5a6f24 100644
--- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c
+++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c
@@ -177,7 +177,7 @@ out_err:
 static void ff_layout_mark_devid_invalid(struct pnfs_layout_segment *lseg,
 		struct nfs4_deviceid_node *devid)
 {
-	nfs4_mark_deviceid_unavailable(devid);
+	nfs4_delete_deviceid(devid->ld, devid->nfs_client, &devid->deviceid);
 	if (!ff_layout_has_available_ds(lseg))
 		pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode,
 				lseg);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 5864146..011e4f8 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -160,6 +160,43 @@ int nfs_sync_mapping(struct address_space *mapping)
 	return ret;
 }
 
+static int nfs_attribute_timeout(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	return !time_in_range_open(jiffies, nfsi->read_cache_jiffies, nfsi->read_cache_jiffies + nfsi->attrtimeo);
+}
+
+static bool nfs_check_cache_invalid_delegated(struct inode *inode, unsigned long flags)
+{
+	unsigned long cache_validity = READ_ONCE(NFS_I(inode)->cache_validity);
+
+	/* Special case for the pagecache or access cache */
+	if (flags == NFS_INO_REVAL_PAGECACHE &&
+	    !(cache_validity & NFS_INO_REVAL_FORCED))
+		return false;
+	return (cache_validity & flags) != 0;
+}
+
+static bool nfs_check_cache_invalid_not_delegated(struct inode *inode, unsigned long flags)
+{
+	unsigned long cache_validity = READ_ONCE(NFS_I(inode)->cache_validity);
+
+	if ((cache_validity & flags) != 0)
+		return true;
+	if (nfs_attribute_timeout(inode))
+		return true;
+	return false;
+}
+
+bool nfs_check_cache_invalid(struct inode *inode, unsigned long flags)
+{
+	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+		return nfs_check_cache_invalid_delegated(inode, flags);
+
+	return nfs_check_cache_invalid_not_delegated(inode, flags);
+}
+
 static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
@@ -795,6 +832,8 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
 	if (!is_sync)
 		return;
 	inode = d_inode(ctx->dentry);
+	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+		return;
 	nfsi = NFS_I(inode);
 	if (inode->i_mapping->nrpages == 0)
 		return;
@@ -1044,13 +1083,6 @@ out:
 	return status;
 }
 
-int nfs_attribute_timeout(struct inode *inode)
-{
-	struct nfs_inode *nfsi = NFS_I(inode);
-
-	return !time_in_range_open(jiffies, nfsi->read_cache_jiffies, nfsi->read_cache_jiffies + nfsi->attrtimeo);
-}
-
 int nfs_attribute_cache_expired(struct inode *inode)
 {
 	if (nfs_have_delegated_attributes(inode))
@@ -1073,15 +1105,6 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 }
 EXPORT_SYMBOL_GPL(nfs_revalidate_inode);
 
-int nfs_revalidate_inode_rcu(struct nfs_server *server, struct inode *inode)
-{
-	if (!(NFS_I(inode)->cache_validity &
-			(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL))
-			&& !nfs_attribute_cache_expired(inode))
-		return NFS_STALE(inode) ? -ESTALE : 0;
-	return -ECHILD;
-}
-
 static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
@@ -1114,17 +1137,8 @@ static int nfs_invalidate_mapping(struct inode *inode, struct address_space *map
 
 bool nfs_mapping_need_revalidate_inode(struct inode *inode)
 {
-	unsigned long cache_validity = NFS_I(inode)->cache_validity;
-
-	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) {
-		const unsigned long force_reval =
-			NFS_INO_REVAL_PAGECACHE|NFS_INO_REVAL_FORCED;
-		return (cache_validity & force_reval) == force_reval;
-	}
-
-	return (cache_validity & NFS_INO_REVAL_PAGECACHE)
-		|| nfs_attribute_timeout(inode)
-		|| NFS_STALE(inode);
+	return nfs_check_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE) ||
+		NFS_STALE(inode);
 }
 
 int nfs_revalidate_mapping_rcu(struct inode *inode)
@@ -1536,13 +1550,6 @@ static int nfs_post_op_update_inode_locked(struct inode *inode, struct nfs_fattr
 {
 	unsigned long invalid = NFS_INO_INVALID_ATTR;
 
-	/*
-	 * Don't revalidate the pagecache if we hold a delegation, but do
-	 * force an attribute update
-	 */
-	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
-		invalid = NFS_INO_INVALID_ATTR|NFS_INO_REVAL_FORCED;
-
 	if (S_ISDIR(inode->i_mode))
 		invalid |= NFS_INO_INVALID_DATA;
 	nfs_set_cache_invalid(inode, invalid);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 6b79c2c..09ca5095 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -381,6 +381,7 @@ extern int nfs_drop_inode(struct inode *);
 extern void nfs_clear_inode(struct inode *);
 extern void nfs_evict_inode(struct inode *);
 void nfs_zap_acl_cache(struct inode *inode);
+extern bool nfs_check_cache_invalid(struct inode *, unsigned long);
 extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode);
 extern int nfs_wait_atomic_killable(atomic_t *p);
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index d33242c..6dcbc5d 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1089,8 +1089,15 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 
 	spin_lock(&dir->i_lock);
 	nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
-	if (!cinfo->atomic || cinfo->before != dir->i_version)
+	if (cinfo->atomic && cinfo->before == dir->i_version) {
+		nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
+		nfsi->attrtimeo_timestamp = jiffies;
+	} else {
 		nfs_force_lookup_revalidate(dir);
+		if (cinfo->before != dir->i_version)
+			nfsi->cache_validity |= NFS_INO_INVALID_ACCESS |
+				NFS_INO_INVALID_ACL;
+	}
 	dir->i_version = cinfo->after;
 	nfsi->attr_gencount = nfs_inc_attr_generation_counter();
 	nfs_fscache_invalidate(dir);
@@ -3115,6 +3122,16 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 			res_stateid = &calldata->res.stateid;
 			renew_lease(server, calldata->timestamp);
 			break;
+		case -NFS4ERR_ACCESS:
+			if (calldata->arg.bitmask != NULL) {
+				calldata->arg.bitmask = NULL;
+				calldata->res.fattr = NULL;
+				task->tk_status = 0;
+				rpc_restart_call_prepare(task);
+				goto out_release;
+
+			}
+			break;
 		case -NFS4ERR_ADMIN_REVOKED:
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
@@ -3140,7 +3157,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 			res_stateid, calldata->arg.fmode);
 out_release:
 	nfs_release_seqid(calldata->arg.seqid);
-	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
+	nfs_refresh_inode(calldata->inode, &calldata->fattr);
 	dprintk("%s: done, ret = %d!\n", __func__, task->tk_status);
 }
 
@@ -3193,9 +3210,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 		goto out_wait;
 	}
 
-	if (calldata->arg.fmode == 0) {
+	if (calldata->arg.fmode == 0)
 		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE];
 
+	if (calldata->arg.fmode == 0 || calldata->arg.fmode == FMODE_READ) {
 		/* Close-to-open cache consistency revalidation */
 		if (!nfs4_have_delegation(inode, FMODE_READ))
 			calldata->arg.bitmask = NFS_SERVER(inode)->cache_consistency_bitmask;
@@ -3207,7 +3225,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 		nfs4_map_atomic_open_share(NFS_SERVER(inode),
 				calldata->arg.fmode, 0);
 
-	nfs_fattr_init(calldata->res.fattr);
+	if (calldata->res.fattr == NULL)
+		calldata->arg.bitmask = NULL;
+	else if (calldata->arg.bitmask == NULL)
+		calldata->res.fattr = NULL;
 	calldata->timestamp = jiffies;
 	if (nfs4_setup_sequence(NFS_SERVER(inode),
 				&calldata->arg.seq_args,
@@ -3274,6 +3295,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 	calldata->arg.seqid = alloc_seqid(&state->owner->so_seqid, gfp_mask);
 	if (IS_ERR(calldata->arg.seqid))
 		goto out_free_calldata;
+	nfs_fattr_init(&calldata->fattr);
 	calldata->arg.fmode = 0;
 	calldata->lr.arg.ld_private = &calldata->lr.ld_private;
 	calldata->res.fattr = &calldata->fattr;
@@ -5673,6 +5695,14 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 	case -NFS4ERR_STALE_STATEID:
 		task->tk_status = 0;
 		break;
+	case -NFS4ERR_ACCESS:
+		if (data->args.bitmask) {
+			data->args.bitmask = NULL;
+			data->res.fattr = NULL;
+			task->tk_status = 0;
+			rpc_restart_call_prepare(task);
+			return;
+		}
 	default:
 		if (nfs4_async_handle_error(task, data->res.server,
 					    NULL, NULL) == -EAGAIN) {
@@ -5692,6 +5722,7 @@ static void nfs4_delegreturn_release(void *calldata)
 		if (data->lr.roc)
 			pnfs_roc_release(&data->lr.arg, &data->lr.res,
 					data->res.lr_ret);
+		nfs_post_op_update_inode_force_wcc(inode, &data->fattr);
 		nfs_iput_and_deactive(inode);
 	}
 	kfree(calldata);
@@ -5780,10 +5811,6 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	if (status != 0)
 		goto out;
 	status = data->rpc_status;
-	if (status == 0)
-		nfs_post_op_update_inode_force_wcc(inode, &data->fattr);
-	else
-		nfs_refresh_inode(inode, &data->fattr);
 out:
 	rpc_put_task(task);
 	return status;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 95baf7d..1d152f4 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -494,21 +494,18 @@ nfs4_alloc_state_owner(struct nfs_server *server,
 }
 
 static void
-nfs4_drop_state_owner(struct nfs4_state_owner *sp)
-{
-	struct rb_node *rb_node = &sp->so_server_node;
-
-	if (!RB_EMPTY_NODE(rb_node)) {
-		struct nfs_server *server = sp->so_server;
-		struct nfs_client *clp = server->nfs_client;
-
-		spin_lock(&clp->cl_lock);
-		if (!RB_EMPTY_NODE(rb_node)) {
-			rb_erase(rb_node, &server->state_owners);
-			RB_CLEAR_NODE(rb_node);
-		}
-		spin_unlock(&clp->cl_lock);
-	}
+nfs4_reset_state_owner(struct nfs4_state_owner *sp)
+{
+	/* This state_owner is no longer usable, but must
+	 * remain in place so that state recovery can find it
+	 * and the opens associated with it.
+	 * It may also be used for new 'open' request to
+	 * return a delegation to the server.
+	 * So update the 'create_time' so that it looks like
+	 * a new state_owner.  This will cause the server to
+	 * request an OPEN_CONFIRM to start a new sequence.
+	 */
+	sp->so_seqid.create_time = ktime_get();
 }
 
 static void nfs4_free_state_owner(struct nfs4_state_owner *sp)
@@ -797,21 +794,33 @@ void nfs4_close_sync(struct nfs4_state *state, fmode_t fmode)
 
 /*
  * Search the state->lock_states for an existing lock_owner
- * that is compatible with current->files
+ * that is compatible with either of the given owners.
+ * If the second is non-zero, then the first refers to a Posix-lock
+ * owner (current->files) and the second refers to a flock/OFD
+ * owner (struct file*).  In that case, prefer a match for the first
+ * owner.
+ * If both sorts of locks are held on the one file we cannot know
+ * which stateid was intended to be used, so a "correct" choice cannot
+ * be made.  Failing that, a "consistent" choice is preferable.  The
+ * consistent choice we make is to prefer the first owner, that of a
+ * Posix lock.
  */
 static struct nfs4_lock_state *
 __nfs4_find_lock_state(struct nfs4_state *state,
 		       fl_owner_t fl_owner, fl_owner_t fl_owner2)
 {
-	struct nfs4_lock_state *pos;
+	struct nfs4_lock_state *pos, *ret = NULL;
 	list_for_each_entry(pos, &state->lock_states, ls_locks) {
-		if (pos->ls_owner != fl_owner &&
-		    pos->ls_owner != fl_owner2)
-			continue;
-		atomic_inc(&pos->ls_count);
-		return pos;
+		if (pos->ls_owner == fl_owner) {
+			ret = pos;
+			break;
+		}
+		if (pos->ls_owner == fl_owner2)
+			ret = pos;
 	}
-	return NULL;
+	if (ret)
+		atomic_inc(&ret->ls_count);
+	return ret;
 }
 
 /*
@@ -1101,7 +1110,7 @@ void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid)
 
 	sp = container_of(seqid->sequence, struct nfs4_state_owner, so_seqid);
 	if (status == -NFS4ERR_BAD_SEQID)
-		nfs4_drop_state_owner(sp);
+		nfs4_reset_state_owner(sp);
 	if (!nfs4_has_session(sp->so_server->nfs_client))
 		nfs_increment_seqid(status, seqid);
 }
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1af6268..e9255cb 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -502,11 +502,13 @@ static int nfs4_stat_to_errno(int);
 				(compound_encode_hdr_maxsz + \
 				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
+				 encode_layoutreturn_maxsz + \
 				 encode_open_downgrade_maxsz)
 #define NFS4_dec_open_downgrade_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
+				 decode_layoutreturn_maxsz + \
 				 decode_open_downgrade_maxsz)
 #define NFS4_enc_close_sz	(compound_encode_hdr_maxsz + \
 				 encode_sequence_maxsz + \
@@ -2277,9 +2279,9 @@ static void nfs4_xdr_enc_close(struct rpc_rqst *req, struct xdr_stream *xdr,
 	encode_putfh(xdr, args->fh, &hdr);
 	if (args->lr_args)
 		encode_layoutreturn(xdr, args->lr_args, &hdr);
-	encode_close(xdr, args, &hdr);
 	if (args->bitmask != NULL)
 		encode_getfattr(xdr, args->bitmask, &hdr);
+	encode_close(xdr, args, &hdr);
 	encode_nops(&hdr);
 }
 
@@ -2356,6 +2358,8 @@ static void nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req,
 	encode_compound_hdr(xdr, req, &hdr);
 	encode_sequence(xdr, &args->seq_args, &hdr);
 	encode_putfh(xdr, args->fh, &hdr);
+	if (args->lr_args)
+		encode_layoutreturn(xdr, args->lr_args, &hdr);
 	encode_open_downgrade(xdr, args, &hdr);
 	encode_nops(&hdr);
 }
@@ -2701,7 +2705,8 @@ static void nfs4_xdr_enc_delegreturn(struct rpc_rqst *req,
 	encode_putfh(xdr, args->fhandle, &hdr);
 	if (args->lr_args)
 		encode_layoutreturn(xdr, args->lr_args, &hdr);
-	encode_getfattr(xdr, args->bitmask, &hdr);
+	if (args->bitmask)
+		encode_getfattr(xdr, args->bitmask, &hdr);
 	encode_delegreturn(xdr, args->stateid, &hdr);
 	encode_nops(&hdr);
 }
@@ -6151,6 +6156,12 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp,
 	status = decode_putfh(xdr);
 	if (status)
 		goto out;
+	if (res->lr_res) {
+		status = decode_layoutreturn(xdr, res->lr_res);
+		res->lr_ret = status;
+		if (status)
+			goto out;
+	}
 	status = decode_open_downgrade(xdr, res);
 out:
 	return status;
@@ -6484,16 +6495,12 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, struct xdr_stream *xdr,
 		if (status)
 			goto out;
 	}
+	if (res->fattr != NULL) {
+		status = decode_getfattr(xdr, res->fattr, res->server);
+		if (status != 0)
+			goto out;
+	}
 	status = decode_close(xdr, res);
-	if (status != 0)
-		goto out;
-	/*
-	 * Note: Server may do delete on close for this file
-	 * 	in which case the getattr call will fail with
-	 * 	an ESTALE error. Shouldn't be a problem,
-	 * 	though, since fattr->valid will remain unset.
-	 */
-	decode_getfattr(xdr, res->fattr, res->server);
 out:
 	return status;
 }
@@ -6966,9 +6973,11 @@ static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp,
 		if (status)
 			goto out;
 	}
-	status = decode_getfattr(xdr, res->fattr, res->server);
-	if (status != 0)
-		goto out;
+	if (res->fattr) {
+		status = decode_getfattr(xdr, res->fattr, res->server);
+		if (status != 0)
+			goto out;
+	}
 	status = decode_delegreturn(xdr);
 out:
 	return status;
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 896df7b..59554f3 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -1251,6 +1251,7 @@ bool pnfs_roc(struct inode *ino,
 	nfs4_stateid stateid;
 	enum pnfs_iomode iomode = 0;
 	bool layoutreturn = false, roc = false;
+	bool skip_read = false;
 
 	if (!nfs_have_layout(ino))
 		return false;
@@ -1270,18 +1271,27 @@ retry:
 	}
 
 	/* no roc if we hold a delegation */
-	if (nfs4_check_delegation(ino, FMODE_READ))
-		goto out_noroc;
+	if (nfs4_check_delegation(ino, FMODE_READ)) {
+		if (nfs4_check_delegation(ino, FMODE_WRITE))
+			goto out_noroc;
+		skip_read = true;
+	}
 
 	list_for_each_entry(ctx, &nfsi->open_files, list) {
 		state = ctx->state;
+		if (state == NULL)
+			continue;
 		/* Don't return layout if there is open file state */
-		if (state != NULL && state->state != 0)
+		if (state->state & FMODE_WRITE)
 			goto out_noroc;
+		if (state->state & FMODE_READ)
+			skip_read = true;
 	}
 
 
 	list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) {
+		if (skip_read && lseg->pls_range.iomode == IOMODE_READ)
+			continue;
 		/* If we are sending layoutreturn, invalidate all valid lsegs */
 		if (!test_and_clear_bit(NFS_LSEG_ROC, &lseg->pls_flags))
 			continue;