Merge branch 'nfs-for-2.6.35' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'nfs-for-2.6.35' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (78 commits)
  SUNRPC: Don't spam gssd with upcall requests when the kerberos key expired
  SUNRPC: Reorder the struct rpc_task fields
  SUNRPC: Remove the 'tk_magic' debugging field
  SUNRPC: Move the task->tk_bytes_sent and tk_rtt to struct rpc_rqst
  NFS: Don't call iput() in nfs_access_cache_shrinker
  NFS: Clean up nfs_access_zap_cache()
  NFS: Don't run nfs_access_cache_shrinker() when the mask is GFP_NOFS
  SUNRPC: Ensure rpcauth_prune_expired() respects the nr_to_scan parameter
  SUNRPC: Ensure memory shrinker doesn't waste time in rpcauth_prune_expired()
  SUNRPC: Dont run rpcauth_cache_shrinker() when gfp_mask is GFP_NOFS
  NFS: Read requests can use GFP_KERNEL.
  NFS: Clean up nfs_create_request()
  NFS: Don't use GFP_KERNEL in rpcsec_gss downcalls
  NFSv4: Don't use GFP_KERNEL allocations in state recovery
  SUNRPC: Fix xs_setup_bc_tcp()
  SUNRPC: Replace jiffies-based metrics with ktime-based metrics
  ktime: introduce ktime_to_ms()
  SUNRPC: RPC metrics and RTT estimator should use same RTT value
  NFS: Calldata for nfs4_renew_done()
  NFS: Squelch compiler warning in nfs_add_server_stats()
  ...

fs/nfs/client.c

@@ -934,7 +934,6 @@ static int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, str
 	}
 
 	fsinfo.fattr = fattr;
-	nfs_fattr_init(fattr);
 	error = clp->rpc_ops->fsinfo(server, mntfh, &fsinfo);
 	if (error < 0)
 		goto out_error;
@@ -1047,13 +1046,18 @@ struct nfs_server *nfs_create_server(const struct nfs_parsed_mount_data *data,
 				     struct nfs_fh *mntfh)
 {
 	struct nfs_server *server;
-	struct nfs_fattr fattr;
+	struct nfs_fattr *fattr;
 	int error;
 
 	server = nfs_alloc_server();
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
+	error = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto error;
+
 	/* Get a client representation */
 	error = nfs_init_server(server, data);
 	if (error < 0)
@@ -1064,7 +1068,7 @@ struct nfs_server *nfs_create_server(const struct nfs_parsed_mount_data *data,
 	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
 	/* Probe the root fh to retrieve its FSID */
-	error = nfs_probe_fsinfo(server, mntfh, &fattr);
+	error = nfs_probe_fsinfo(server, mntfh, fattr);
 	if (error < 0)
 		goto error;
 	if (server->nfs_client->rpc_ops->version == 3) {
@@ -1077,14 +1081,14 @@ struct nfs_server *nfs_create_server(const struct nfs_parsed_mount_data *data,
 			server->namelen = NFS2_MAXNAMLEN;
 	}
 
-	if (!(fattr.valid & NFS_ATTR_FATTR)) {
-		error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr);
+	if (!(fattr->valid & NFS_ATTR_FATTR)) {
+		error = server->nfs_client->rpc_ops->getattr(server, mntfh, fattr);
 		if (error < 0) {
 			dprintk("nfs_create_server: getattr error = %d\n", -error);
 			goto error;
 		}
 	}
-	memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid));
+	memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid));
 
 	dprintk("Server FSID: %llx:%llx\n",
 		(unsigned long long) server->fsid.major,
@@ -1096,9 +1100,11 @@ struct nfs_server *nfs_create_server(const struct nfs_parsed_mount_data *data,
 	spin_unlock(&nfs_client_lock);
 
 	server->mount_time = jiffies;
+	nfs_free_fattr(fattr);
 	return server;
 
 error:
+	nfs_free_fattr(fattr);
 	nfs_free_server(server);
 	return ERR_PTR(error);
 }
@@ -1340,7 +1346,7 @@ error:
 struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 				      struct nfs_fh *mntfh)
 {
-	struct nfs_fattr fattr;
+	struct nfs_fattr *fattr;
 	struct nfs_server *server;
 	int error;
 
@@ -1350,6 +1356,11 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
+	error = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto error;
+
 	/* set up the general RPC client */
 	error = nfs4_init_server(server, data);
 	if (error < 0)
@@ -1364,7 +1375,7 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 		goto error;
 
 	/* Probe the root fh to retrieve its FSID */
-	error = nfs4_path_walk(server, mntfh, data->nfs_server.export_path);
+	error = nfs4_get_rootfh(server, mntfh);
 	if (error < 0)
 		goto error;
 
@@ -1375,7 +1386,7 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 
 	nfs4_session_set_rwsize(server);
 
-	error = nfs_probe_fsinfo(server, mntfh, &fattr);
+	error = nfs_probe_fsinfo(server, mntfh, fattr);
 	if (error < 0)
 		goto error;
 
@@ -1389,9 +1400,11 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 
 	server->mount_time = jiffies;
 	dprintk("<-- nfs4_create_server() = %p\n", server);
+	nfs_free_fattr(fattr);
 	return server;
 
 error:
+	nfs_free_fattr(fattr);
 	nfs_free_server(server);
 	dprintk("<-- nfs4_create_server() = error %d\n", error);
 	return ERR_PTR(error);
@@ -1405,7 +1418,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 {
 	struct nfs_client *parent_client;
 	struct nfs_server *server, *parent_server;
-	struct nfs_fattr fattr;
+	struct nfs_fattr *fattr;
 	int error;
 
 	dprintk("--> nfs4_create_referral_server()\n");
@@ -1414,6 +1427,11 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
+	error = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto error;
+
 	parent_server = NFS_SB(data->sb);
 	parent_client = parent_server->nfs_client;
 
@@ -1443,12 +1461,12 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
 	/* Probe the root fh to retrieve its FSID and filehandle */
-	error = nfs4_path_walk(server, mntfh, data->mnt_path);
+	error = nfs4_get_rootfh(server, mntfh);
 	if (error < 0)
 		goto error;
 
 	/* probe the filesystem info for this server filesystem */
-	error = nfs_probe_fsinfo(server, mntfh, &fattr);
+	error = nfs_probe_fsinfo(server, mntfh, fattr);
 	if (error < 0)
 		goto error;
 
@@ -1466,10 +1484,12 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 
 	server->mount_time = jiffies;
 
+	nfs_free_fattr(fattr);
 	dprintk("<-- nfs_create_referral_server() = %p\n", server);
 	return server;
 
 error:
+	nfs_free_fattr(fattr);
 	nfs_free_server(server);
 	dprintk("<-- nfs4_create_referral_server() = error %d\n", error);
 	return ERR_PTR(error);
@@ -1485,7 +1505,7 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source,
 				    struct nfs_fattr *fattr)
 {
 	struct nfs_server *server;
-	struct nfs_fattr fattr_fsinfo;
+	struct nfs_fattr *fattr_fsinfo;
 	int error;
 
 	dprintk("--> nfs_clone_server(,%llx:%llx,)\n",
@@ -1496,6 +1516,11 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
+	error = -ENOMEM;
+	fattr_fsinfo = nfs_alloc_fattr();
+	if (fattr_fsinfo == NULL)
+		goto out_free_server;
+
 	/* Copy data from the source */
 	server->nfs_client = source->nfs_client;
 	atomic_inc(&server->nfs_client->cl_count);
@@ -1512,7 +1537,7 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source,
 		nfs_init_server_aclclient(server);
 
 	/* probe the filesystem info for this server filesystem */
-	error = nfs_probe_fsinfo(server, fh, &fattr_fsinfo);
+	error = nfs_probe_fsinfo(server, fh, fattr_fsinfo);
 	if (error < 0)
 		goto out_free_server;
 
@@ -1534,10 +1559,12 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source,
 
 	server->mount_time = jiffies;
 
+	nfs_free_fattr(fattr_fsinfo);
 	dprintk("<-- nfs_clone_server() = %p\n", server);
 	return server;
 
 out_free_server:
+	nfs_free_fattr(fattr_fsinfo);
 	nfs_free_server(server);
 	dprintk("<-- nfs_clone_server() = error %d\n", error);
 	return ERR_PTR(error);

fs/nfs/delegation.c

@@ -213,7 +213,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 	struct nfs_delegation *freeme = NULL;
 	int status = 0;
 
-	delegation = kmalloc(sizeof(*delegation), GFP_KERNEL);
+	delegation = kmalloc(sizeof(*delegation), GFP_NOFS);
 	if (delegation == NULL)
 		return -ENOMEM;
 	memcpy(delegation->stateid.data, res->delegation.data,

fs/nfs/dir.c

@@ -530,9 +530,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	nfs_readdir_descriptor_t my_desc,
 			*desc = &my_desc;
 	struct nfs_entry my_entry;
-	struct nfs_fh	 fh;
-	struct nfs_fattr fattr;
-	long		res;
+	int res = -ENOMEM;
 
 	dfprintk(FILE, "NFS: readdir(%s/%s) starting at cookie %llu\n",
 			dentry->d_parent->d_name.name, dentry->d_name.name,
@@ -554,9 +552,11 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 
 	my_entry.cookie = my_entry.prev_cookie = 0;
 	my_entry.eof = 0;
-	my_entry.fh = &fh;
-	my_entry.fattr = &fattr;
-	nfs_fattr_init(&fattr);
+	my_entry.fh = nfs_alloc_fhandle();
+	my_entry.fattr = nfs_alloc_fattr();
+	if (my_entry.fh == NULL || my_entry.fattr == NULL)
+		goto out_alloc_failed;
+
 	desc->entry = &my_entry;
 
 	nfs_block_sillyrename(dentry);
@@ -598,7 +598,10 @@ out:
 	nfs_unblock_sillyrename(dentry);
 	if (res > 0)
 		res = 0;
-	dfprintk(FILE, "NFS: readdir(%s/%s) returns %ld\n",
+out_alloc_failed:
+	nfs_free_fattr(my_entry.fattr);
+	nfs_free_fhandle(my_entry.fh);
+	dfprintk(FILE, "NFS: readdir(%s/%s) returns %d\n",
 			dentry->d_parent->d_name.name, dentry->d_name.name,
 			res);
 	return res;
@@ -776,9 +779,9 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
 	struct inode *dir;
 	struct inode *inode;
 	struct dentry *parent;
+	struct nfs_fh *fhandle = NULL;
+	struct nfs_fattr *fattr = NULL;
 	int error;
-	struct nfs_fh fhandle;
-	struct nfs_fattr fattr;
 
 	parent = dget_parent(dentry);
 	dir = parent->d_inode;
@@ -811,14 +814,22 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
 	if (NFS_STALE(inode))
 		goto out_bad;
 
-	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, &fhandle, &fattr);
+	error = -ENOMEM;
+	fhandle = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	if (fhandle == NULL || fattr == NULL)
+		goto out_error;
+
+	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
 	if (error)
 		goto out_bad;
-	if (nfs_compare_fh(NFS_FH(inode), &fhandle))
+	if (nfs_compare_fh(NFS_FH(inode), fhandle))
 		goto out_bad;
-	if ((error = nfs_refresh_inode(inode, &fattr)) != 0)
+	if ((error = nfs_refresh_inode(inode, fattr)) != 0)
 		goto out_bad;
 
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fhandle);
 out_set_verifier:
 	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
  out_valid:
@@ -842,11 +853,21 @@ out_zap_parent:
 		shrink_dcache_parent(dentry);
 	}
 	d_drop(dentry);
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fhandle);
 	dput(parent);
 	dfprintk(LOOKUPCACHE, "NFS: %s(%s/%s) is invalid\n",
 			__func__, dentry->d_parent->d_name.name,
 			dentry->d_name.name);
 	return 0;
+out_error:
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fhandle);
+	dput(parent);
+	dfprintk(LOOKUPCACHE, "NFS: %s(%s/%s) lookup returned error %d\n",
+			__func__, dentry->d_parent->d_name.name,
+			dentry->d_name.name, error);
+	return error;
 }
 
 /*
@@ -911,9 +932,9 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
 	struct dentry *res;
 	struct dentry *parent;
 	struct inode *inode = NULL;
+	struct nfs_fh *fhandle = NULL;
+	struct nfs_fattr *fattr = NULL;
 	int error;
-	struct nfs_fh fhandle;
-	struct nfs_fattr fattr;
 
 	dfprintk(VFS, "NFS: lookup(%s/%s)\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
@@ -923,7 +944,6 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
 	if (dentry->d_name.len > NFS_SERVER(dir)->namelen)
 		goto out;
 
-	res = ERR_PTR(-ENOMEM);
 	dentry->d_op = NFS_PROTO(dir)->dentry_ops;
 
 	/*
@@ -936,17 +956,23 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
 		goto out;
 	}
 
+	res = ERR_PTR(-ENOMEM);
+	fhandle = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	if (fhandle == NULL || fattr == NULL)
+		goto out;
+
 	parent = dentry->d_parent;
 	/* Protect against concurrent sillydeletes */
 	nfs_block_sillyrename(parent);
-	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, &fhandle, &fattr);
+	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
 	if (error == -ENOENT)
 		goto no_entry;
 	if (error < 0) {
 		res = ERR_PTR(error);
 		goto out_unblock_sillyrename;
 	}
-	inode = nfs_fhget(dentry->d_sb, &fhandle, &fattr);
+	inode = nfs_fhget(dentry->d_sb, fhandle, fattr);
 	res = (struct dentry *)inode;
 	if (IS_ERR(res))
 		goto out_unblock_sillyrename;
@@ -962,6 +988,8 @@ no_entry:
 out_unblock_sillyrename:
 	nfs_unblock_sillyrename(parent);
 out:
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fhandle);
 	return res;
 }
 
@@ -1669,28 +1697,33 @@ static void nfs_access_free_entry(struct nfs_access_entry *entry)
 	smp_mb__after_atomic_dec();
 }
 
+static void nfs_access_free_list(struct list_head *head)
+{
+	struct nfs_access_entry *cache;
+
+	while (!list_empty(head)) {
+		cache = list_entry(head->next, struct nfs_access_entry, lru);
+		list_del(&cache->lru);
+		nfs_access_free_entry(cache);
+	}
+}
+
 int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask)
 {
 	LIST_HEAD(head);
 	struct nfs_inode *nfsi;
 	struct nfs_access_entry *cache;
 
-restart:
+	if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL)
+		return (nr_to_scan == 0) ? 0 : -1;
+
 	spin_lock(&nfs_access_lru_lock);
 	list_for_each_entry(nfsi, &nfs_access_lru_list, access_cache_inode_lru) {
-		struct rw_semaphore *s_umount;
 		struct inode *inode;
 
 		if (nr_to_scan-- == 0)
 			break;
-		s_umount = &nfsi->vfs_inode.i_sb->s_umount;
-		if (!down_read_trylock(s_umount))
-			continue;
-		inode = igrab(&nfsi->vfs_inode);
-		if (inode == NULL) {
-			up_read(s_umount);
-			continue;
-		}
+		inode = &nfsi->vfs_inode;
 		spin_lock(&inode->i_lock);
 		if (list_empty(&nfsi->access_cache_entry_lru))
 			goto remove_lru_entry;
@@ -1704,61 +1737,47 @@ restart:
 		else {
 remove_lru_entry:
 			list_del_init(&nfsi->access_cache_inode_lru);
+			smp_mb__before_clear_bit();
 			clear_bit(NFS_INO_ACL_LRU_SET, &nfsi->flags);
+			smp_mb__after_clear_bit();
 		}
-		spin_unlock(&inode->i_lock);
-		spin_unlock(&nfs_access_lru_lock);
-		iput(inode);
-		up_read(s_umount);
-		goto restart;
 	}
 	spin_unlock(&nfs_access_lru_lock);
-	while (!list_empty(&head)) {
-		cache = list_entry(head.next, struct nfs_access_entry, lru);
-		list_del(&cache->lru);
-		nfs_access_free_entry(cache);
-	}
+	nfs_access_free_list(&head);
 	return (atomic_long_read(&nfs_access_nr_entries) / 100) * sysctl_vfs_cache_pressure;
 }
 
-static void __nfs_access_zap_cache(struct inode *inode)
+static void __nfs_access_zap_cache(struct nfs_inode *nfsi, struct list_head *head)
 {
-	struct nfs_inode *nfsi = NFS_I(inode);
 	struct rb_root *root_node = &nfsi->access_cache;
-	struct rb_node *n, *dispose = NULL;
+	struct rb_node *n;
 	struct nfs_access_entry *entry;
 
 	/* Unhook entries from the cache */
 	while ((n = rb_first(root_node)) != NULL) {
 		entry = rb_entry(n, struct nfs_access_entry, rb_node);
 		rb_erase(n, root_node);
-		list_del(&entry->lru);
-		n->rb_left = dispose;
-		dispose = n;
+		list_move(&entry->lru, head);
 	}
 	nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS;
-	spin_unlock(&inode->i_lock);
-
-	/* Now kill them all! */
-	while (dispose != NULL) {
-		n = dispose;
-		dispose = n->rb_left;
-		nfs_access_free_entry(rb_entry(n, struct nfs_access_entry, rb_node));
-	}
 }
 
 void nfs_access_zap_cache(struct inode *inode)
 {
+	LIST_HEAD(head);
+
+	if (test_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags) == 0)
+		return;
 	/* Remove from global LRU init */
-	if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) {
-		spin_lock(&nfs_access_lru_lock);
+	spin_lock(&nfs_access_lru_lock);
+	if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags))
 		list_del_init(&NFS_I(inode)->access_cache_inode_lru);
-		spin_unlock(&nfs_access_lru_lock);
-	}
 
 	spin_lock(&inode->i_lock);
-	/* This will release the spinlock */
-	__nfs_access_zap_cache(inode);
+	__nfs_access_zap_cache(NFS_I(inode), &head);
+	spin_unlock(&inode->i_lock);
+	spin_unlock(&nfs_access_lru_lock);
+	nfs_access_free_list(&head);
 }
 
 static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, struct rpc_cred *cred)
@@ -1809,8 +1828,8 @@ out_stale:
 	nfs_access_free_entry(cache);
 	return -ENOENT;
 out_zap:
-	/* This will release the spinlock */
-	__nfs_access_zap_cache(inode);
+	spin_unlock(&inode->i_lock);
+	nfs_access_zap_cache(inode);
 	return -ENOENT;
 }
 
@@ -1865,9 +1884,11 @@ static void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *s
 	smp_mb__after_atomic_inc();
 
 	/* Add inode to global LRU list */
-	if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) {
+	if (!test_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) {
 		spin_lock(&nfs_access_lru_lock);
-		list_add_tail(&NFS_I(inode)->access_cache_inode_lru, &nfs_access_lru_list);
+		if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags))
+			list_add_tail(&NFS_I(inode)->access_cache_inode_lru,
+					&nfs_access_lru_list);
 		spin_unlock(&nfs_access_lru_lock);
 	}
 }

fs/nfs/file.c

@@ -161,14 +161,17 @@ 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);
 
-	if (server->flags & NFS_MOUNT_NOAC)
-		goto force_reval;
+	if (nfs_have_delegated_attributes(inode))
+		goto out_noreval;
+
 	if (filp->f_flags & O_DIRECT)
 		goto force_reval;
-	if (nfsi->npages != 0)
-		return 0;
-	if (!(nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) && !nfs_attribute_timeout(inode))
-		return 0;
+	if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE)
+		goto force_reval;
+	if (nfs_attribute_timeout(inode))
+		goto force_reval;
+out_noreval:
+	return 0;
 force_reval:
 	return __nfs_revalidate_inode(server, inode);
 }

fs/nfs/fscache.c

@@ -467,7 +467,8 @@ int __nfs_readpages_from_fscache(struct nfs_open_context *ctx,
 				 struct list_head *pages,
 				 unsigned *nr_pages)
 {
-	int ret, npages = *nr_pages;
+	unsigned npages = *nr_pages;
+	int ret;
 
 	dfprintk(FSCACHE, "NFS: nfs_getpages_from_fscache (0x%p/%u/0x%p)\n",
 		 NFS_I(inode)->fscache, npages, inode);

fs/nfs/getroot.c

@@ -78,159 +78,94 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh)
 {
 	struct nfs_server *server = NFS_SB(sb);
 	struct nfs_fsinfo fsinfo;
-	struct nfs_fattr fattr;
-	struct dentry *mntroot;
+	struct dentry *ret;
 	struct inode *inode;
 	int error;
 
 	/* get the actual root for this mount */
-	fsinfo.fattr = &fattr;
+	fsinfo.fattr = nfs_alloc_fattr();
+	if (fsinfo.fattr == NULL)
+		return ERR_PTR(-ENOMEM);
 
 	error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
 	if (error < 0) {
 		dprintk("nfs_get_root: getattr error = %d\n", -error);
-		return ERR_PTR(error);
+		ret = ERR_PTR(error);
+		goto out;
 	}
 
 	inode = nfs_fhget(sb, mntfh, fsinfo.fattr);
 	if (IS_ERR(inode)) {
 		dprintk("nfs_get_root: get root inode failed\n");
-		return ERR_CAST(inode);
+		ret = ERR_CAST(inode);
+		goto out;
 	}
 
 	error = nfs_superblock_set_dummy_root(sb, inode);
-	if (error != 0)
-		return ERR_PTR(error);
+	if (error != 0) {
+		ret = ERR_PTR(error);
+		goto out;
+	}
 
 	/* root dentries normally start off anonymous and get spliced in later
 	 * if the dentry tree reaches them; however if the dentry already
 	 * exists, we'll pick it up at this point and use it as the root
 	 */
-	mntroot = d_obtain_alias(inode);
-	if (IS_ERR(mntroot)) {
+	ret = d_obtain_alias(inode);
+	if (IS_ERR(ret)) {
 		dprintk("nfs_get_root: get root dentry failed\n");
-		return mntroot;
+		goto out;
 	}
 
-	security_d_instantiate(mntroot, inode);
-
-	if (!mntroot->d_op)
-		mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops;
+	security_d_instantiate(ret, inode);
 
-	return mntroot;
+	if (ret->d_op == NULL)
+		ret->d_op = server->nfs_client->rpc_ops->dentry_ops;
+out:
+	nfs_free_fattr(fsinfo.fattr);
+	return ret;
 }
 
 #ifdef CONFIG_NFS_V4
 
-/*
- * Do a simple pathwalk from the root FH of the server to the nominated target
- * of the mountpoint
- * - give error on symlinks
- * - give error on ".." occurring in the path
- * - follow traversals
- */
-int nfs4_path_walk(struct nfs_server *server,
-		   struct nfs_fh *mntfh,
-		   const char *path)
+int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh)
 {
 	struct nfs_fsinfo fsinfo;
-	struct nfs_fattr fattr;
-	struct nfs_fh lastfh;
-	struct qstr name;
-	int ret;
+	int ret = -ENOMEM;
 
-	dprintk("--> nfs4_path_walk(,,%s)\n", path);
+	dprintk("--> nfs4_get_rootfh()\n");
 
-	fsinfo.fattr = &fattr;
-	nfs_fattr_init(&fattr);
-
-	/* Eat leading slashes */
-	while (*path == '/')
-		path++;
+	fsinfo.fattr = nfs_alloc_fattr();
+	if (fsinfo.fattr == NULL)
+		goto out;
 
 	/* Start by getting the root filehandle from the server */
 	ret = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
 	if (ret < 0) {
-		dprintk("nfs4_get_root: getroot error = %d\n", -ret);
-		return ret;
+		dprintk("nfs4_get_rootfh: getroot error = %d\n", -ret);
+		goto out;
 	}
 
-	if (!S_ISDIR(fattr.mode)) {
-		printk(KERN_ERR "nfs4_get_root:"
+	if (!(fsinfo.fattr->valid & NFS_ATTR_FATTR_MODE)
+			|| !S_ISDIR(fsinfo.fattr->mode)) {
+		printk(KERN_ERR "nfs4_get_rootfh:"
 		       " getroot encountered non-directory\n");
-		return -ENOTDIR;
+		ret = -ENOTDIR;
+		goto out;
 	}
 
-	/* FIXME: It is quite valid for the server to return a referral here */
-	if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) {
-		printk(KERN_ERR "nfs4_get_root:"
+	if (fsinfo.fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		printk(KERN_ERR "nfs4_get_rootfh:"
 		       " getroot obtained referral\n");
-		return -EREMOTE;
-	}
-
-next_component:
-	dprintk("Next: %s\n", path);
-
-	/* extract the next bit of the path */
-	if (!*path)
-		goto path_walk_complete;
-
-	name.name = path;
-	while (*path && *path != '/')
-		path++;
-	name.len = path - (const char *) name.name;
-
-	if (name.len > NFS4_MAXNAMLEN)
-		return -ENAMETOOLONG;
-
-eat_dot_dir:
-	while (*path == '/')
-		path++;
-
-	if (path[0] == '.' && (path[1] == '/' || !path[1])) {
-		path += 2;
-		goto eat_dot_dir;
-	}
-
-	/* FIXME: Why shouldn't the user be able to use ".." in the path? */
-	if (path[0] == '.' && path[1] == '.' && (path[2] == '/' || !path[2])
-	    ) {
-		printk(KERN_ERR "nfs4_get_root:"
-		       " Mount path contains reference to \"..\"\n");
-		return -EINVAL;
+		ret = -EREMOTE;
+		goto out;
 	}
 
-	/* lookup the next FH in the sequence */
-	memcpy(&lastfh, mntfh, sizeof(lastfh));
-
-	dprintk("LookupFH: %*.*s [%s]\n", name.len, name.len, name.name, path);
-
-	ret = server->nfs_client->rpc_ops->lookupfh(server, &lastfh, &name,
-						    mntfh, &fattr);
-	if (ret < 0) {
-		dprintk("nfs4_get_root: getroot error = %d\n", -ret);
-		return ret;
-	}
-
-	if (!S_ISDIR(fattr.mode)) {
-		printk(KERN_ERR "nfs4_get_root:"
-		       " lookupfh encountered non-directory\n");
-		return -ENOTDIR;
-	}
-
-	/* FIXME: Referrals are quite valid here too */
-	if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) {
-		printk(KERN_ERR "nfs4_get_root:"
-		       " lookupfh obtained referral\n");
-		return -EREMOTE;
-	}
-
-	goto next_component;
-
-path_walk_complete:
-	memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid));
-	dprintk("<-- nfs4_path_walk() = 0\n");
-	return 0;
+	memcpy(&server->fsid, &fsinfo.fattr->fsid, sizeof(server->fsid));
+out:
+	nfs_free_fattr(fsinfo.fattr);
+	dprintk("<-- nfs4_get_rootfh() = %d\n", ret);
+	return ret;
 }
 
 /*
@@ -239,8 +174,8 @@ path_walk_complete:
 struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
 {
 	struct nfs_server *server = NFS_SB(sb);
-	struct nfs_fattr fattr;
-	struct dentry *mntroot;
+	struct nfs_fattr *fattr = NULL;
+	struct dentry *ret;
 	struct inode *inode;
 	int error;
 
@@ -254,40 +189,50 @@ struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
 		return ERR_PTR(error);
 	}
 
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		return ERR_PTR(-ENOMEM);;
+
 	/* get the actual root for this mount */
-	error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr);
+	error = server->nfs_client->rpc_ops->getattr(server, mntfh, fattr);
 	if (error < 0) {
 		dprintk("nfs_get_root: getattr error = %d\n", -error);
-		return ERR_PTR(error);
+		ret = ERR_PTR(error);
+		goto out;
 	}
 
-	inode = nfs_fhget(sb, mntfh, &fattr);
+	inode = nfs_fhget(sb, mntfh, fattr);
 	if (IS_ERR(inode)) {
 		dprintk("nfs_get_root: get root inode failed\n");
-		return ERR_CAST(inode);
+		ret = ERR_CAST(inode);
+		goto out;
 	}
 
 	error = nfs_superblock_set_dummy_root(sb, inode);
-	if (error != 0)
-		return ERR_PTR(error);
+	if (error != 0) {
+		ret = ERR_PTR(error);
+		goto out;
+	}
 
 	/* root dentries normally start off anonymous and get spliced in later
 	 * if the dentry tree reaches them; however if the dentry already
 	 * exists, we'll pick it up at this point and use it as the root
 	 */
-	mntroot = d_obtain_alias(inode);
-	if (IS_ERR(mntroot)) {
+	ret = d_obtain_alias(inode);
+	if (IS_ERR(ret)) {
 		dprintk("nfs_get_root: get root dentry failed\n");
-		return mntroot;
+		goto out;
 	}
 
-	security_d_instantiate(mntroot, inode);
+	security_d_instantiate(ret, inode);
 
-	if (!mntroot->d_op)
-		mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops;
+	if (ret->d_op == NULL)
+		ret->d_op = server->nfs_client->rpc_ops->dentry_ops;
 
+out:
+	nfs_free_fattr(fattr);
 	dprintk("<-- nfs4_get_root()\n");
-	return mntroot;
+	return ret;
 }
 
 #endif /* CONFIG_NFS_V4 */

fs/nfs/inode.c

@@ -393,8 +393,8 @@ int
 nfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
-	struct nfs_fattr fattr;
-	int error;
+	struct nfs_fattr *fattr;
+	int error = -ENOMEM;
 
 	nfs_inc_stats(inode, NFSIOS_VFSSETATTR);
 
@@ -417,14 +417,20 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
 		filemap_write_and_wait(inode->i_mapping);
 		nfs_wb_all(inode);
 	}
+
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out;
 	/*
 	 * Return any delegations if we're going to change ACLs
 	 */
 	if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0)
 		nfs_inode_return_delegation(inode);
-	error = NFS_PROTO(inode)->setattr(dentry, &fattr, attr);
+	error = NFS_PROTO(inode)->setattr(dentry, fattr, attr);
 	if (error == 0)
-		nfs_refresh_inode(inode, &fattr);
+		nfs_refresh_inode(inode, fattr);
+	nfs_free_fattr(fattr);
+out:
 	return error;
 }
 
@@ -682,7 +688,7 @@ int
 __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 {
 	int		 status = -ESTALE;
-	struct nfs_fattr fattr;
+	struct nfs_fattr *fattr = NULL;
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n",
@@ -693,8 +699,13 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 	if (NFS_STALE(inode))
 		goto out;
 
+	status = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out;
+
 	nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE);
-	status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), &fattr);
+	status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr);
 	if (status != 0) {
 		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) getattr failed, error=%d\n",
 			 inode->i_sb->s_id,
@@ -707,7 +718,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		goto out;
 	}
 
-	status = nfs_refresh_inode(inode, &fattr);
+	status = nfs_refresh_inode(inode, fattr);
 	if (status) {
 		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n",
 			 inode->i_sb->s_id,
@@ -723,6 +734,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		(long long)NFS_FILEID(inode));
 
  out:
+	nfs_free_fattr(fattr);
 	return status;
 }
 
@@ -730,9 +742,14 @@ 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 int nfs_attribute_cache_expired(struct inode *inode)
+{
 	if (nfs_have_delegated_attributes(inode))
 		return 0;
-	return !time_in_range_open(jiffies, nfsi->read_cache_jiffies, nfsi->read_cache_jiffies + nfsi->attrtimeo);
+	return nfs_attribute_timeout(inode);
 }
 
 /**
@@ -745,7 +762,7 @@ int nfs_attribute_timeout(struct inode *inode)
 int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 {
 	if (!(NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATTR)
-			&& !nfs_attribute_timeout(inode))
+			&& !nfs_attribute_cache_expired(inode))
 		return NFS_STALE(inode) ? -ESTALE : 0;
 	return __nfs_revalidate_inode(server, inode);
 }
@@ -782,7 +799,8 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
 	int ret = 0;
 
 	if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE)
-			|| nfs_attribute_timeout(inode) || NFS_STALE(inode)) {
+			|| nfs_attribute_cache_expired(inode)
+			|| NFS_STALE(inode)) {
 		ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
 		if (ret < 0)
 			goto out;
@@ -916,6 +934,26 @@ void nfs_fattr_init(struct nfs_fattr *fattr)
 	fattr->gencount = nfs_inc_attr_generation_counter();
 }
 
+struct nfs_fattr *nfs_alloc_fattr(void)
+{
+	struct nfs_fattr *fattr;
+
+	fattr = kmalloc(sizeof(*fattr), GFP_NOFS);
+	if (fattr != NULL)
+		nfs_fattr_init(fattr);
+	return fattr;
+}
+
+struct nfs_fh *nfs_alloc_fhandle(void)
+{
+	struct nfs_fh *fh;
+
+	fh = kmalloc(sizeof(struct nfs_fh), GFP_NOFS);
+	if (fh != NULL)
+		fh->size = 0;
+	return fh;
+}
+
 /**
  * nfs_inode_attrs_need_update - check if the inode attributes need updating
  * @inode - pointer to inode

fs/nfs/internal.h

@@ -244,9 +244,7 @@ extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *);
 #ifdef CONFIG_NFS_V4
 extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *);
 
-extern int nfs4_path_walk(struct nfs_server *server,
-			  struct nfs_fh *mntfh,
-			  const char *path);
+extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh);
 #endif
 
 /* read.c */

fs/nfs/iostat.h

@@ -36,14 +36,14 @@ static inline void nfs_inc_stats(const struct inode *inode,
 
 static inline void nfs_add_server_stats(const struct nfs_server *server,
 					enum nfs_stat_bytecounters stat,
-					unsigned long addend)
+					long addend)
 {
 	this_cpu_add(server->io_stats->bytes[stat], addend);
 }
 
 static inline void nfs_add_stats(const struct inode *inode,
 				 enum nfs_stat_bytecounters stat,
-				 unsigned long addend)
+				 long addend)
 {
 	nfs_add_server_stats(NFS_SERVER(inode), stat, addend);
 }
@@ -51,7 +51,7 @@ static inline void nfs_add_stats(const struct inode *inode,
 #ifdef CONFIG_NFS_FSCACHE
 static inline void nfs_add_fscache_stats(struct inode *inode,
 					 enum nfs_stat_fscachecounters stat,
-					 unsigned long addend)
+					 long addend)
 {
 	this_cpu_add(NFS_SERVER(inode)->io_stats->fscache[stat], addend);
 }

fs/nfs/namespace.c

@@ -105,8 +105,8 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	struct vfsmount *mnt;
 	struct nfs_server *server = NFS_SERVER(dentry->d_inode);
 	struct dentry *parent;
-	struct nfs_fh fh;
-	struct nfs_fattr fattr;
+	struct nfs_fh *fh = NULL;
+	struct nfs_fattr *fattr = NULL;
 	int err;
 
 	dprintk("--> nfs_follow_mountpoint()\n");
@@ -115,6 +115,12 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	if (IS_ROOT(dentry))
 		goto out_err;
 
+	err = -ENOMEM;
+	fh = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	if (fh == NULL || fattr == NULL)
+		goto out_err;
+
 	dprintk("%s: enter\n", __func__);
 	dput(nd->path.dentry);
 	nd->path.dentry = dget(dentry);
@@ -123,16 +129,16 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	parent = dget_parent(nd->path.dentry);
 	err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
 						  &nd->path.dentry->d_name,
-						  &fh, &fattr);
+						  fh, fattr);
 	dput(parent);
 	if (err != 0)
 		goto out_err;
 
-	if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL)
+	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
 		mnt = nfs_do_refmount(nd->path.mnt, nd->path.dentry);
 	else
-		mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, &fh,
-				      &fattr);
+		mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, fh,
+				      fattr);
 	err = PTR_ERR(mnt);
 	if (IS_ERR(mnt))
 		goto out_err;
@@ -151,6 +157,8 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	nd->path.dentry = dget(mnt->mnt_root);
 	schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
 out:
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fh);
 	dprintk("%s: done, returned %d\n", __func__, err);
 
 	dprintk("<-- nfs_follow_mountpoint() = %d\n", err);

fs/nfs/nfs3acl.c

@@ -185,7 +185,6 @@ static void nfs3_cache_acls(struct inode *inode, struct posix_acl *acl,
 struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_fattr fattr;
 	struct page *pages[NFSACL_MAXPAGES] = { };
 	struct nfs3_getaclargs args = {
 		.fh = NFS_FH(inode),
@@ -193,7 +192,7 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 		.pages = pages,
 	};
 	struct nfs3_getaclres res = {
-		.fattr =	&fattr,
+		0
 	};
 	struct rpc_message msg = {
 		.rpc_argp	= &args,
@@ -228,7 +227,10 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 
 	dprintk("NFS call getacl\n");
 	msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_GETACL];
-	nfs_fattr_init(&fattr);
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		return ERR_PTR(-ENOMEM);
+
 	status = rpc_call_sync(server->client_acl, &msg, 0);
 	dprintk("NFS reply getacl: %d\n", status);
 
@@ -238,7 +240,7 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 
 	switch (status) {
 		case 0:
-			status = nfs_refresh_inode(inode, &fattr);
+			status = nfs_refresh_inode(inode, res.fattr);
 			break;
 		case -EPFNOSUPPORT:
 		case -EPROTONOSUPPORT:
@@ -278,6 +280,7 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 getout:
 	posix_acl_release(res.acl_access);
 	posix_acl_release(res.acl_default);
+	nfs_free_fattr(res.fattr);
 
 	if (status != 0) {
 		posix_acl_release(acl);
@@ -290,7 +293,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 		  struct posix_acl *dfacl)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_fattr fattr;
+	struct nfs_fattr *fattr;
 	struct page *pages[NFSACL_MAXPAGES];
 	struct nfs3_setaclargs args = {
 		.inode = inode,
@@ -335,8 +338,13 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 	}
 
 	dprintk("NFS call setacl\n");
+	status = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out_freepages;
+
 	msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_SETACL];
-	nfs_fattr_init(&fattr);
+	msg.rpc_resp = fattr;
 	status = rpc_call_sync(server->client_acl, &msg, 0);
 	nfs_access_zap_cache(inode);
 	nfs_zap_acl_cache(inode);
@@ -344,7 +352,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 
 	switch (status) {
 		case 0:
-			status = nfs_refresh_inode(inode, &fattr);
+			status = nfs_refresh_inode(inode, fattr);
 			nfs3_cache_acls(inode, acl, dfacl);
 			break;
 		case -EPFNOSUPPORT:
@@ -355,6 +363,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 		case -ENOTSUPP:
 			status = -EOPNOTSUPP;
 	}
+	nfs_free_fattr(fattr);
 out_freepages:
 	while (args.npages != 0) {
 		args.npages--;

fs/nfs/nfs3proc.c

@@ -144,14 +144,12 @@ static int
 nfs3_proc_lookup(struct inode *dir, struct qstr *name,
 		 struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
-	struct nfs_fattr	dir_attr;
 	struct nfs3_diropargs	arg = {
 		.fh		= NFS_FH(dir),
 		.name		= name->name,
 		.len		= name->len
 	};
 	struct nfs3_diropres	res = {
-		.dir_attr	= &dir_attr,
 		.fh		= fhandle,
 		.fattr		= fattr
 	};
@@ -163,29 +161,30 @@ nfs3_proc_lookup(struct inode *dir, struct qstr *name,
 	int			status;
 
 	dprintk("NFS call  lookup %s\n", name->name);
-	nfs_fattr_init(&dir_attr);
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		return -ENOMEM;
+
 	nfs_fattr_init(fattr);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-	nfs_refresh_inode(dir, &dir_attr);
+	nfs_refresh_inode(dir, res.dir_attr);
 	if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) {
 		msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR];
 		msg.rpc_argp = fhandle;
 		msg.rpc_resp = fattr;
 		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	}
+	nfs_free_fattr(res.dir_attr);
 	dprintk("NFS reply lookup: %d\n", status);
 	return status;
 }
 
 static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 {
-	struct nfs_fattr	fattr;
 	struct nfs3_accessargs	arg = {
 		.fh		= NFS_FH(inode),
 	};
-	struct nfs3_accessres	res = {
-		.fattr		= &fattr,
-	};
+	struct nfs3_accessres	res;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_ACCESS],
 		.rpc_argp	= &arg,
@@ -193,7 +192,7 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 		.rpc_cred	= entry->cred,
 	};
 	int mode = entry->mask;
-	int status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  access\n");
 
@@ -210,9 +209,13 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 		if (mode & MAY_EXEC)
 			arg.access |= NFS3_ACCESS_EXECUTE;
 	}
-	nfs_fattr_init(&fattr);
+
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		goto out;
+
 	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
-	nfs_refresh_inode(inode, &fattr);
+	nfs_refresh_inode(inode, res.fattr);
 	if (status == 0) {
 		entry->mask = 0;
 		if (res.access & NFS3_ACCESS_READ)
@@ -222,6 +225,8 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 		if (res.access & (NFS3_ACCESS_LOOKUP|NFS3_ACCESS_EXECUTE))
 			entry->mask |= MAY_EXEC;
 	}
+	nfs_free_fattr(res.fattr);
+out:
 	dprintk("NFS reply access: %d\n", status);
 	return status;
 }
@@ -229,7 +234,7 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 static int nfs3_proc_readlink(struct inode *inode, struct page *page,
 		unsigned int pgbase, unsigned int pglen)
 {
-	struct nfs_fattr	fattr;
+	struct nfs_fattr	*fattr;
 	struct nfs3_readlinkargs args = {
 		.fh		= NFS_FH(inode),
 		.pgbase		= pgbase,
@@ -239,14 +244,19 @@ static int nfs3_proc_readlink(struct inode *inode, struct page *page,
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_READLINK],
 		.rpc_argp	= &args,
-		.rpc_resp	= &fattr,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  readlink\n");
-	nfs_fattr_init(&fattr);
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out;
+	msg.rpc_resp = fattr;
+
 	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
-	nfs_refresh_inode(inode, &fattr);
+	nfs_refresh_inode(inode, fattr);
+	nfs_free_fattr(fattr);
+out:
 	dprintk("NFS reply readlink: %d\n", status);
 	return status;
 }
@@ -396,12 +406,17 @@ nfs3_proc_remove(struct inode *dir, struct qstr *name)
 		.rpc_argp = &arg,
 		.rpc_resp = &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  remove %s\n", name->name);
-	nfs_fattr_init(&res.dir_attr);
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		goto out;
+
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-	nfs_post_op_update_inode(dir, &res.dir_attr);
+	nfs_post_op_update_inode(dir, res.dir_attr);
+	nfs_free_fattr(res.dir_attr);
+out:
 	dprintk("NFS reply remove: %d\n", status);
 	return status;
 }
@@ -419,7 +434,7 @@ nfs3_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 	if (nfs3_async_handle_jukebox(task, dir))
 		return 0;
 	res = task->tk_msg.rpc_resp;
-	nfs_post_op_update_inode(dir, &res->dir_attr);
+	nfs_post_op_update_inode(dir, res->dir_attr);
 	return 1;
 }
 
@@ -427,7 +442,6 @@ static int
 nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name,
 		 struct inode *new_dir, struct qstr *new_name)
 {
-	struct nfs_fattr	old_dir_attr, new_dir_attr;
 	struct nfs3_renameargs	arg = {
 		.fromfh		= NFS_FH(old_dir),
 		.fromname	= old_name->name,
@@ -436,23 +450,27 @@ nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name,
 		.toname		= new_name->name,
 		.tolen		= new_name->len
 	};
-	struct nfs3_renameres	res = {
-		.fromattr	= &old_dir_attr,
-		.toattr		= &new_dir_attr
-	};
+	struct nfs3_renameres res;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_RENAME],
 		.rpc_argp	= &arg,
 		.rpc_resp	= &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  rename %s -> %s\n", old_name->name, new_name->name);
-	nfs_fattr_init(&old_dir_attr);
-	nfs_fattr_init(&new_dir_attr);
+
+	res.fromattr = nfs_alloc_fattr();
+	res.toattr = nfs_alloc_fattr();
+	if (res.fromattr == NULL || res.toattr == NULL)
+		goto out;
+
 	status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0);
-	nfs_post_op_update_inode(old_dir, &old_dir_attr);
-	nfs_post_op_update_inode(new_dir, &new_dir_attr);
+	nfs_post_op_update_inode(old_dir, res.fromattr);
+	nfs_post_op_update_inode(new_dir, res.toattr);
+out:
+	nfs_free_fattr(res.toattr);
+	nfs_free_fattr(res.fromattr);
 	dprintk("NFS reply rename: %d\n", status);
 	return status;
 }
@@ -460,30 +478,32 @@ nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name,
 static int
 nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 {
-	struct nfs_fattr	dir_attr, fattr;
 	struct nfs3_linkargs	arg = {
 		.fromfh		= NFS_FH(inode),
 		.tofh		= NFS_FH(dir),
 		.toname		= name->name,
 		.tolen		= name->len
 	};
-	struct nfs3_linkres	res = {
-		.dir_attr	= &dir_attr,
-		.fattr		= &fattr
-	};
+	struct nfs3_linkres	res;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_LINK],
 		.rpc_argp	= &arg,
 		.rpc_resp	= &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  link %s\n", name->name);
-	nfs_fattr_init(&dir_attr);
-	nfs_fattr_init(&fattr);
+	res.fattr = nfs_alloc_fattr();
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.fattr == NULL || res.dir_attr == NULL)
+		goto out;
+
 	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
-	nfs_post_op_update_inode(dir, &dir_attr);
-	nfs_post_op_update_inode(inode, &fattr);
+	nfs_post_op_update_inode(dir, res.dir_attr);
+	nfs_post_op_update_inode(inode, res.fattr);
+out:
+	nfs_free_fattr(res.dir_attr);
+	nfs_free_fattr(res.fattr);
 	dprintk("NFS reply link: %d\n", status);
 	return status;
 }
@@ -554,7 +574,7 @@ out:
 static int
 nfs3_proc_rmdir(struct inode *dir, struct qstr *name)
 {
-	struct nfs_fattr	dir_attr;
+	struct nfs_fattr	*dir_attr;
 	struct nfs3_diropargs	arg = {
 		.fh		= NFS_FH(dir),
 		.name		= name->name,
@@ -563,14 +583,19 @@ nfs3_proc_rmdir(struct inode *dir, struct qstr *name)
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_RMDIR],
 		.rpc_argp	= &arg,
-		.rpc_resp	= &dir_attr,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  rmdir %s\n", name->name);
-	nfs_fattr_init(&dir_attr);
+	dir_attr = nfs_alloc_fattr();
+	if (dir_attr == NULL)
+		goto out;
+
+	msg.rpc_resp = dir_attr;
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-	nfs_post_op_update_inode(dir, &dir_attr);
+	nfs_post_op_update_inode(dir, dir_attr);
+	nfs_free_fattr(dir_attr);
+out:
 	dprintk("NFS reply rmdir: %d\n", status);
 	return status;
 }
@@ -589,7 +614,6 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 		  u64 cookie, struct page *page, unsigned int count, int plus)
 {
 	struct inode		*dir = dentry->d_inode;
-	struct nfs_fattr	dir_attr;
 	__be32			*verf = NFS_COOKIEVERF(dir);
 	struct nfs3_readdirargs	arg = {
 		.fh		= NFS_FH(dir),
@@ -600,7 +624,6 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 		.pages		= &page
 	};
 	struct nfs3_readdirres	res = {
-		.dir_attr	= &dir_attr,
 		.verf		= verf,
 		.plus		= plus
 	};
@@ -610,7 +633,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 		.rpc_resp	= &res,
 		.rpc_cred	= cred
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	if (plus)
 		msg.rpc_proc = &nfs3_procedures[NFS3PROC_READDIRPLUS];
@@ -618,12 +641,17 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 	dprintk("NFS call  readdir%s %d\n",
 			plus? "plus" : "", (unsigned int) cookie);
 
-	nfs_fattr_init(&dir_attr);
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		goto out;
+
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 
 	nfs_invalidate_atime(dir);
+	nfs_refresh_inode(dir, res.dir_attr);
 
-	nfs_refresh_inode(dir, &dir_attr);
+	nfs_free_fattr(res.dir_attr);
+out:
 	dprintk("NFS reply readdir: %d\n", status);
 	return status;
 }

fs/nfs/nfs3xdr.c

@@ -762,7 +762,7 @@ nfs3_xdr_wccstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr)
 static int
 nfs3_xdr_removeres(struct rpc_rqst *req, __be32 *p, struct nfs_removeres *res)
 {
-	return nfs3_xdr_wccstat(req, p, &res->dir_attr);
+	return nfs3_xdr_wccstat(req, p, res->dir_attr);
 }
 
 /*

fs/nfs/nfs4_fs.h

@@ -206,14 +206,14 @@ extern ssize_t nfs4_listxattr(struct dentry *, char *, size_t);
 
 
 /* nfs4proc.c */
-extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *);
-extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *);
+extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *);
 extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
 extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
 extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *);
-extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait);
+extern int nfs4_do_close(struct path *path, struct nfs4_state *state, gfp_t gfp_mask, int wait);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
 extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle);
@@ -286,7 +286,7 @@ extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
 
-extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter);
+extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask);
 extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
 extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
 extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);

fs/nfs/nfs4namespace.c

@@ -115,6 +115,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
 				     char *page, char *page2,
 				     const struct nfs4_fs_location *location)
 {
+	const size_t addr_bufsize = sizeof(struct sockaddr_storage);
 	struct vfsmount *mnt = ERR_PTR(-ENOENT);
 	char *mnt_path;
 	unsigned int maxbuflen;
@@ -126,9 +127,12 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
 	mountdata->mnt_path = mnt_path;
 	maxbuflen = mnt_path - 1 - page2;
 
+	mountdata->addr = kmalloc(addr_bufsize, GFP_KERNEL);
+	if (mountdata->addr == NULL)
+		return ERR_PTR(-ENOMEM);
+
 	for (s = 0; s < location->nservers; s++) {
 		const struct nfs4_string *buf = &location->servers[s];
-		struct sockaddr_storage addr;
 
 		if (buf->len <= 0 || buf->len >= maxbuflen)
 			continue;
@@ -137,11 +141,10 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
 			continue;
 
 		mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len,
-				(struct sockaddr *)&addr, sizeof(addr));
+				mountdata->addr, addr_bufsize);
 		if (mountdata->addrlen == 0)
 			continue;
 
-		mountdata->addr = (struct sockaddr *)&addr;
 		rpc_set_port(mountdata->addr, NFS_PORT);
 
 		memcpy(page2, buf->data, buf->len);
@@ -156,6 +159,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
 		if (!IS_ERR(mnt))
 			break;
 	}
+	kfree(mountdata->addr);
 	return mnt;
 }
 
@@ -221,8 +225,8 @@ out:
 
 /*
  * nfs_do_refmount - handle crossing a referral on server
+ * @mnt_parent - mountpoint of referral
  * @dentry - dentry of referral
- * @nd - nameidata info
  *
  */
 struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry)

fs/nfs/nfs4proc.c

@@ -70,6 +70,9 @@ static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinf
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *, struct nfs4_state *);
 static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
 static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
+static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
+			    struct nfs_fattr *fattr, struct iattr *sattr,
+			    struct nfs4_state *state);
 
 /* Prevent leaks of NFSv4 errors into userland */
 static int nfs4_map_errors(int err)
@@ -714,17 +717,18 @@ static void nfs4_init_opendata_res(struct nfs4_opendata *p)
 
 static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 		struct nfs4_state_owner *sp, fmode_t fmode, int flags,
-		const struct iattr *attrs)
+		const struct iattr *attrs,
+		gfp_t gfp_mask)
 {
 	struct dentry *parent = dget_parent(path->dentry);
 	struct inode *dir = parent->d_inode;
 	struct nfs_server *server = NFS_SERVER(dir);
 	struct nfs4_opendata *p;
 
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc(sizeof(*p), gfp_mask);
 	if (p == NULL)
 		goto err;
-	p->o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
+	p->o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid, gfp_mask);
 	if (p->o_arg.seqid == NULL)
 		goto err_free;
 	path_get(path);
@@ -1060,7 +1064,7 @@ static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context
 {
 	struct nfs4_opendata *opendata;
 
-	opendata = nfs4_opendata_alloc(&ctx->path, state->owner, 0, 0, NULL);
+	opendata = nfs4_opendata_alloc(&ctx->path, state->owner, 0, 0, NULL, GFP_NOFS);
 	if (opendata == NULL)
 		return ERR_PTR(-ENOMEM);
 	opendata->state = state;
@@ -1648,7 +1652,7 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in
 	if (path->dentry->d_inode != NULL)
 		nfs4_return_incompatible_delegation(path->dentry->d_inode, fmode);
 	status = -ENOMEM;
-	opendata = nfs4_opendata_alloc(path, sp, fmode, flags, sattr);
+	opendata = nfs4_opendata_alloc(path, sp, fmode, flags, sattr, GFP_KERNEL);
 	if (opendata == NULL)
 		goto err_put_state_owner;
 
@@ -1659,15 +1663,24 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in
 	if (status != 0)
 		goto err_opendata_put;
 
-	if (opendata->o_arg.open_flags & O_EXCL)
-		nfs4_exclusive_attrset(opendata, sattr);
-
 	state = nfs4_opendata_to_nfs4_state(opendata);
 	status = PTR_ERR(state);
 	if (IS_ERR(state))
 		goto err_opendata_put;
 	if (server->caps & NFS_CAP_POSIX_LOCK)
 		set_bit(NFS_STATE_POSIX_LOCKS, &state->flags);
+
+	if (opendata->o_arg.open_flags & O_EXCL) {
+		nfs4_exclusive_attrset(opendata, sattr);
+
+		nfs_fattr_init(opendata->o_res.f_attr);
+		status = nfs4_do_setattr(state->inode, cred,
+				opendata->o_res.f_attr, sattr,
+				state);
+		if (status == 0)
+			nfs_setattr_update_inode(state->inode, sattr);
+		nfs_post_op_update_inode(state->inode, opendata->o_res.f_attr);
+	}
 	nfs4_opendata_put(opendata);
 	nfs4_put_state_owner(sp);
 	*res = state;
@@ -1914,7 +1927,7 @@ static const struct rpc_call_ops nfs4_close_ops = {
  *
  * NOTE: Caller must be holding the sp->so_owner semaphore!
  */
-int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
+int nfs4_do_close(struct path *path, struct nfs4_state *state, gfp_t gfp_mask, int wait)
 {
 	struct nfs_server *server = NFS_SERVER(state->inode);
 	struct nfs4_closedata *calldata;
@@ -1933,7 +1946,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	};
 	int status = -ENOMEM;
 
-	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
+	calldata = kzalloc(sizeof(*calldata), gfp_mask);
 	if (calldata == NULL)
 		goto out;
 	calldata->inode = state->inode;
@@ -1941,7 +1954,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->arg.fh = NFS_FH(state->inode);
 	calldata->arg.stateid = &state->open_stateid;
 	/* Serialization for the sequence id */
-	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid);
+	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid, gfp_mask);
 	if (calldata->arg.seqid == NULL)
 		goto out_free_calldata;
 	calldata->arg.fmode = 0;
@@ -2404,14 +2417,12 @@ static int nfs4_proc_lookup(struct inode *dir, struct qstr *name, struct nfs_fh
 static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry)
 {
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_fattr fattr;
 	struct nfs4_accessargs args = {
 		.fh = NFS_FH(inode),
 		.bitmask = server->attr_bitmask,
 	};
 	struct nfs4_accessres res = {
 		.server = server,
-		.fattr = &fattr,
 	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ACCESS],
@@ -2438,7 +2449,11 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry
 		if (mode & MAY_EXEC)
 			args.access |= NFS4_ACCESS_EXECUTE;
 	}
-	nfs_fattr_init(&fattr);
+
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		return -ENOMEM;
+
 	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	if (!status) {
 		entry->mask = 0;
@@ -2448,8 +2463,9 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry
 			entry->mask |= MAY_WRITE;
 		if (res.access & (NFS4_ACCESS_LOOKUP|NFS4_ACCESS_EXECUTE))
 			entry->mask |= MAY_EXEC;
-		nfs_refresh_inode(inode, &fattr);
+		nfs_refresh_inode(inode, res.fattr);
 	}
+	nfs_free_fattr(res.fattr);
 	return status;
 }
 
@@ -2562,13 +2578,6 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 	}
 	d_add(dentry, igrab(state->inode));
 	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
-	if (flags & O_EXCL) {
-		struct nfs_fattr fattr;
-		status = nfs4_do_setattr(state->inode, cred, &fattr, sattr, state);
-		if (status == 0)
-			nfs_setattr_update_inode(state->inode, sattr);
-		nfs_post_op_update_inode(state->inode, &fattr);
-	}
 	if (status == 0 && (nd->flags & LOOKUP_OPEN) != 0)
 		status = nfs4_intent_set_file(nd, &path, state, fmode);
 	else
@@ -2596,14 +2605,19 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
-	int			status;
+	int status = -ENOMEM;
+
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		goto out;
 
-	nfs_fattr_init(&res.dir_attr);
 	status = nfs4_call_sync(server, &msg, &args, &res, 1);
 	if (status == 0) {
 		update_changeattr(dir, &res.cinfo);
-		nfs_post_op_update_inode(dir, &res.dir_attr);
+		nfs_post_op_update_inode(dir, res.dir_attr);
 	}
+	nfs_free_fattr(res.dir_attr);
+out:
 	return status;
 }
 
@@ -2638,7 +2652,7 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 	if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN)
 		return 0;
 	update_changeattr(dir, &res->cinfo);
-	nfs_post_op_update_inode(dir, &res->dir_attr);
+	nfs_post_op_update_inode(dir, res->dir_attr);
 	return 1;
 }
 
@@ -2653,29 +2667,31 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
 		.new_name = new_name,
 		.bitmask = server->attr_bitmask,
 	};
-	struct nfs_fattr old_fattr, new_fattr;
 	struct nfs4_rename_res res = {
 		.server = server,
-		.old_fattr = &old_fattr,
-		.new_fattr = &new_fattr,
 	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME],
 		.rpc_argp = &arg,
 		.rpc_resp = &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 	
-	nfs_fattr_init(res.old_fattr);
-	nfs_fattr_init(res.new_fattr);
-	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
+	res.old_fattr = nfs_alloc_fattr();
+	res.new_fattr = nfs_alloc_fattr();
+	if (res.old_fattr == NULL || res.new_fattr == NULL)
+		goto out;
 
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (!status) {
 		update_changeattr(old_dir, &res.old_cinfo);
 		nfs_post_op_update_inode(old_dir, res.old_fattr);
 		update_changeattr(new_dir, &res.new_cinfo);
 		nfs_post_op_update_inode(new_dir, res.new_fattr);
 	}
+out:
+	nfs_free_fattr(res.new_fattr);
+	nfs_free_fattr(res.old_fattr);
 	return status;
 }
 
@@ -2702,28 +2718,30 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *
 		.name   = name,
 		.bitmask = server->attr_bitmask,
 	};
-	struct nfs_fattr fattr, dir_attr;
 	struct nfs4_link_res res = {
 		.server = server,
-		.fattr = &fattr,
-		.dir_attr = &dir_attr,
 	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK],
 		.rpc_argp = &arg,
 		.rpc_resp = &res,
 	};
-	int			status;
+	int status = -ENOMEM;
+
+	res.fattr = nfs_alloc_fattr();
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.fattr == NULL || res.dir_attr == NULL)
+		goto out;
 
-	nfs_fattr_init(res.fattr);
-	nfs_fattr_init(res.dir_attr);
 	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (!status) {
 		update_changeattr(dir, &res.cinfo);
 		nfs_post_op_update_inode(dir, res.dir_attr);
 		nfs_post_op_update_inode(inode, res.fattr);
 	}
-
+out:
+	nfs_free_fattr(res.dir_attr);
+	nfs_free_fattr(res.fattr);
 	return status;
 }
 
@@ -3146,23 +3164,31 @@ static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_messa
 	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT];
 }
 
+struct nfs4_renewdata {
+	struct nfs_client	*client;
+	unsigned long		timestamp;
+};
+
 /*
  * nfs4_proc_async_renew(): This is not one of the nfs_rpc_ops; it is a special
  * standalone procedure for queueing an asynchronous RENEW.
  */
-static void nfs4_renew_release(void *data)
+static void nfs4_renew_release(void *calldata)
 {
-	struct nfs_client *clp = data;
+	struct nfs4_renewdata *data = calldata;
+	struct nfs_client *clp = data->client;
 
 	if (atomic_read(&clp->cl_count) > 1)
 		nfs4_schedule_state_renewal(clp);
 	nfs_put_client(clp);
+	kfree(data);
 }
 
-static void nfs4_renew_done(struct rpc_task *task, void *data)
+static void nfs4_renew_done(struct rpc_task *task, void *calldata)
 {
-	struct nfs_client *clp = data;
-	unsigned long timestamp = task->tk_start;
+	struct nfs4_renewdata *data = calldata;
+	struct nfs_client *clp = data->client;
+	unsigned long timestamp = data->timestamp;
 
 	if (task->tk_status < 0) {
 		/* Unless we're shutting down, schedule state recovery! */
@@ -3188,11 +3214,17 @@ int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred)
 		.rpc_argp	= clp,
 		.rpc_cred	= cred,
 	};
+	struct nfs4_renewdata *data;
 
 	if (!atomic_inc_not_zero(&clp->cl_count))
 		return -EIO;
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	if (data == NULL)
+		return -ENOMEM;
+	data->client = clp;
+	data->timestamp = jiffies;
 	return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT,
-			&nfs4_renew_ops, clp);
+			&nfs4_renew_ops, data);
 }
 
 int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred)
@@ -3494,7 +3526,9 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 	return _nfs4_async_handle_error(task, server, server->nfs_client, state);
 }
 
-int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
+int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
+		unsigned short port, struct rpc_cred *cred,
+		struct nfs4_setclientid_res *res)
 {
 	nfs4_verifier sc_verifier;
 	struct nfs4_setclientid setclientid = {
@@ -3504,7 +3538,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID],
 		.rpc_argp = &setclientid,
-		.rpc_resp = clp,
+		.rpc_resp = res,
 		.rpc_cred = cred,
 	};
 	__be32 *p;
@@ -3547,12 +3581,14 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po
 	return status;
 }
 
-static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
+static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
+		struct nfs4_setclientid_res *arg,
+		struct rpc_cred *cred)
 {
 	struct nfs_fsinfo fsinfo;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID_CONFIRM],
-		.rpc_argp = clp,
+		.rpc_argp = arg,
 		.rpc_resp = &fsinfo,
 		.rpc_cred = cred,
 	};
@@ -3570,12 +3606,14 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cre
 	return status;
 }
 
-int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
+int nfs4_proc_setclientid_confirm(struct nfs_client *clp,
+		struct nfs4_setclientid_res *arg,
+		struct rpc_cred *cred)
 {
 	long timeout = 0;
 	int err;
 	do {
-		err = _nfs4_proc_setclientid_confirm(clp, cred);
+		err = _nfs4_proc_setclientid_confirm(clp, arg, cred);
 		switch (err) {
 			case 0:
 				return err;
@@ -3667,7 +3705,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	};
 	int status = 0;
 
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	data = kzalloc(sizeof(*data), GFP_NOFS);
 	if (data == NULL)
 		return -ENOMEM;
 	data->args.fhandle = &data->fh;
@@ -3823,7 +3861,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	struct nfs4_unlockdata *p;
 	struct inode *inode = lsp->ls_state->inode;
 
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc(sizeof(*p), GFP_NOFS);
 	if (p == NULL)
 		return NULL;
 	p->arg.fh = NFS_FH(inode);
@@ -3961,7 +3999,7 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *
 	if (test_bit(NFS_DELEGATED_STATE, &state->flags))
 		goto out;
 	lsp = request->fl_u.nfs4_fl.owner;
-	seqid = nfs_alloc_seqid(&lsp->ls_seqid);
+	seqid = nfs_alloc_seqid(&lsp->ls_seqid, GFP_KERNEL);
 	status = -ENOMEM;
 	if (seqid == NULL)
 		goto out;
@@ -3989,22 +4027,23 @@ struct nfs4_lockdata {
 };
 
 static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
-		struct nfs_open_context *ctx, struct nfs4_lock_state *lsp)
+		struct nfs_open_context *ctx, struct nfs4_lock_state *lsp,
+		gfp_t gfp_mask)
 {
 	struct nfs4_lockdata *p;
 	struct inode *inode = lsp->ls_state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
 
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc(sizeof(*p), gfp_mask);
 	if (p == NULL)
 		return NULL;
 
 	p->arg.fh = NFS_FH(inode);
 	p->arg.fl = &p->fl;
-	p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid);
+	p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid, gfp_mask);
 	if (p->arg.open_seqid == NULL)
 		goto out_free;
-	p->arg.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid);
+	p->arg.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid, gfp_mask);
 	if (p->arg.lock_seqid == NULL)
 		goto out_free_seqid;
 	p->arg.lock_stateid = &lsp->ls_stateid;
@@ -4158,7 +4197,8 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 
 	dprintk("%s: begin!\n", __func__);
 	data = nfs4_alloc_lockdata(fl, nfs_file_open_context(fl->fl_file),
-			fl->fl_u.nfs4_fl.owner);
+			fl->fl_u.nfs4_fl.owner,
+			recovery_type == NFS_LOCK_NEW ? GFP_KERNEL : GFP_NOFS);
 	if (data == NULL)
 		return -ENOMEM;
 	if (IS_SETLKW(cmd))
@@ -4647,7 +4687,7 @@ static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs,
 	if (max_reqs != tbl->max_slots) {
 		ret = -ENOMEM;
 		new = kmalloc(max_reqs * sizeof(struct nfs4_slot),
-			      GFP_KERNEL);
+			      GFP_NOFS);
 		if (!new)
 			goto out;
 		ret = 0;
@@ -4712,7 +4752,7 @@ static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 
 	dprintk("--> %s: max_reqs=%u\n", __func__, max_slots);
 
-	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
+	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_NOFS);
 	if (!slot)
 		goto out;
 	ret = 0;
@@ -4761,7 +4801,7 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	struct nfs4_session *session;
 	struct nfs4_slot_table *tbl;
 
-	session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL);
+	session = kzalloc(sizeof(struct nfs4_session), GFP_NOFS);
 	if (!session)
 		return NULL;
 
@@ -5105,8 +5145,8 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 
 	if (!atomic_inc_not_zero(&clp->cl_count))
 		return -EIO;
-	args = kzalloc(sizeof(*args), GFP_KERNEL);
-	res = kzalloc(sizeof(*res), GFP_KERNEL);
+	args = kzalloc(sizeof(*args), GFP_NOFS);
+	res = kzalloc(sizeof(*res), GFP_NOFS);
 	if (!args || !res) {
 		kfree(args);
 		kfree(res);
@@ -5207,7 +5247,7 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp)
 	int status = -ENOMEM;
 
 	dprintk("--> %s\n", __func__);
-	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
+	calldata = kzalloc(sizeof(*calldata), GFP_NOFS);
 	if (calldata == NULL)
 		goto out;
 	calldata->clp = clp;

fs/nfs/nfs4state.c

@@ -62,6 +62,7 @@ static LIST_HEAD(nfs4_clientid_list);
 
 int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
+	struct nfs4_setclientid_res clid;
 	unsigned short port;
 	int status;
 
@@ -69,11 +70,15 @@ int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 	if (clp->cl_addr.ss_family == AF_INET6)
 		port = nfs_callback_tcpport6;
 
-	status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred);
-	if (status == 0)
-		status = nfs4_proc_setclientid_confirm(clp, cred);
-	if (status == 0)
-		nfs4_schedule_state_renewal(clp);
+	status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid);
+	if (status != 0)
+		goto out;
+	status = nfs4_proc_setclientid_confirm(clp, &clid, cred);
+	if (status != 0)
+		goto out;
+	clp->cl_clientid = clid.clientid;
+	nfs4_schedule_state_renewal(clp);
+out:
 	return status;
 }
 
@@ -361,7 +366,7 @@ nfs4_alloc_state_owner(void)
 {
 	struct nfs4_state_owner *sp;
 
-	sp = kzalloc(sizeof(*sp),GFP_KERNEL);
+	sp = kzalloc(sizeof(*sp),GFP_NOFS);
 	if (!sp)
 		return NULL;
 	spin_lock_init(&sp->so_lock);
@@ -435,7 +440,7 @@ nfs4_alloc_open_state(void)
 {
 	struct nfs4_state *state;
 
-	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	state = kzalloc(sizeof(*state), GFP_NOFS);
 	if (!state)
 		return NULL;
 	atomic_set(&state->count, 1);
@@ -537,7 +542,8 @@ void nfs4_put_open_state(struct nfs4_state *state)
 /*
  * Close the current file.
  */
-static void __nfs4_close(struct path *path, struct nfs4_state *state, fmode_t fmode, int wait)
+static void __nfs4_close(struct path *path, struct nfs4_state *state,
+		fmode_t fmode, gfp_t gfp_mask, int wait)
 {
 	struct nfs4_state_owner *owner = state->owner;
 	int call_close = 0;
@@ -578,17 +584,17 @@ static void __nfs4_close(struct path *path, struct nfs4_state *state, fmode_t fm
 		nfs4_put_open_state(state);
 		nfs4_put_state_owner(owner);
 	} else
-		nfs4_do_close(path, state, wait);
+		nfs4_do_close(path, state, gfp_mask, wait);
 }
 
 void nfs4_close_state(struct path *path, struct nfs4_state *state, fmode_t fmode)
 {
-	__nfs4_close(path, state, fmode, 0);
+	__nfs4_close(path, state, fmode, GFP_NOFS, 0);
 }
 
 void nfs4_close_sync(struct path *path, struct nfs4_state *state, fmode_t fmode)
 {
-	__nfs4_close(path, state, fmode, 1);
+	__nfs4_close(path, state, fmode, GFP_KERNEL, 1);
 }
 
 /*
@@ -618,7 +624,7 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f
 	struct nfs4_lock_state *lsp;
 	struct nfs_client *clp = state->owner->so_client;
 
-	lsp = kzalloc(sizeof(*lsp), GFP_KERNEL);
+	lsp = kzalloc(sizeof(*lsp), GFP_NOFS);
 	if (lsp == NULL)
 		return NULL;
 	rpc_init_wait_queue(&lsp->ls_sequence.wait, "lock_seqid_waitqueue");
@@ -754,11 +760,11 @@ void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t f
 	nfs4_put_lock_state(lsp);
 }
 
-struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter)
+struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask)
 {
 	struct nfs_seqid *new;
 
-	new = kmalloc(sizeof(*new), GFP_KERNEL);
+	new = kmalloc(sizeof(*new), gfp_mask);
 	if (new != NULL) {
 		new->sequence = counter;
 		INIT_LIST_HEAD(&new->list);
@@ -1347,7 +1353,7 @@ static int nfs4_recall_slot(struct nfs_client *clp)
 
 	nfs4_begin_drain_session(clp);
 	new = kmalloc(fc_tbl->target_max_slots * sizeof(struct nfs4_slot),
-		      GFP_KERNEL);
+		      GFP_NOFS);
         if (!new)
 		return -ENOMEM;
 

fs/nfs/nfs4xdr.c

@@ -1504,14 +1504,14 @@ static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclie
 	hdr->replen += decode_setclientid_maxsz;
 }
 
-static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state, struct compound_hdr *hdr)
+static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_setclientid_res *arg, struct compound_hdr *hdr)
 {
 	__be32 *p;
 
 	p = reserve_space(xdr, 12 + NFS4_VERIFIER_SIZE);
 	*p++ = cpu_to_be32(OP_SETCLIENTID_CONFIRM);
-	p = xdr_encode_hyper(p, client_state->cl_clientid);
-	xdr_encode_opaque_fixed(p, client_state->cl_confirm.data, NFS4_VERIFIER_SIZE);
+	p = xdr_encode_hyper(p, arg->clientid);
+	xdr_encode_opaque_fixed(p, arg->confirm.data, NFS4_VERIFIER_SIZE);
 	hdr->nops++;
 	hdr->replen += decode_setclientid_confirm_maxsz;
 }
@@ -2324,7 +2324,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4
 /*
  * a SETCLIENTID_CONFIRM request
  */
-static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp)
+static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs4_setclientid_res *arg)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -2334,7 +2334,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
-	encode_setclientid_confirm(&xdr, clp, &hdr);
+	encode_setclientid_confirm(&xdr, arg, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_fsinfo(&xdr, lease_bitmap, &hdr);
 	encode_nops(&hdr);
@@ -4397,7 +4397,7 @@ out_overflow:
 	return -EIO;
 }
 
-static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp)
+static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_setclientid_res *res)
 {
 	__be32 *p;
 	uint32_t opnum;
@@ -4417,8 +4417,8 @@ static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp)
 		p = xdr_inline_decode(xdr, 8 + NFS4_VERIFIER_SIZE);
 		if (unlikely(!p))
 			goto out_overflow;
-		p = xdr_decode_hyper(p, &clp->cl_clientid);
-		memcpy(clp->cl_confirm.data, p, NFS4_VERIFIER_SIZE);
+		p = xdr_decode_hyper(p, &res->clientid);
+		memcpy(res->confirm.data, p, NFS4_VERIFIER_SIZE);
 	} else if (nfserr == NFSERR_CLID_INUSE) {
 		uint32_t len;
 
@@ -4815,7 +4815,7 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs_rem
 		goto out;
 	if ((status = decode_remove(&xdr, &res->cinfo)) != 0)
 		goto out;
-	decode_getfattr(&xdr, &res->dir_attr, res->server,
+	decode_getfattr(&xdr, res->dir_attr, res->server,
 			!RPC_IS_ASYNC(rqstp->rq_task));
 out:
 	return status;
@@ -5498,7 +5498,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, __be32 *p, void *dummy)
  * Decode SETCLIENTID response
  */
 static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
-		struct nfs_client *clp)
+		struct nfs4_setclientid_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -5507,7 +5507,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (!status)
-		status = decode_setclientid(&xdr, clp);
+		status = decode_setclientid(&xdr, res);
 	return status;
 }
 

fs/nfs/nfsroot.c

@@ -488,7 +488,6 @@ static int __init root_nfs_ports(void)
  */
 static int __init root_nfs_get_handle(void)
 {
-	struct nfs_fh fh;
 	struct sockaddr_in sin;
 	unsigned int auth_flav_len = 0;
 	struct nfs_mount_request request = {
@@ -499,21 +498,24 @@ static int __init root_nfs_get_handle(void)
 					NFS_MNT3_VERSION : NFS_MNT_VERSION,
 		.protocol	= (nfs_data.flags & NFS_MOUNT_TCP) ?
 					XPRT_TRANSPORT_TCP : XPRT_TRANSPORT_UDP,
-		.fh		= &fh,
 		.auth_flav_len	= &auth_flav_len,
 	};
-	int status;
+	int status = -ENOMEM;
 
+	request.fh = nfs_alloc_fhandle();
+	if (!request.fh)
+		goto out;
 	set_sockaddr(&sin, servaddr, htons(mount_port));
 	status = nfs_mount(&request);
 	if (status < 0)
 		printk(KERN_ERR "Root-NFS: Server returned error %d "
 				"while mounting %s\n", status, nfs_export_path);
 	else {
-		nfs_data.root.size = fh.size;
-		memcpy(nfs_data.root.data, fh.data, fh.size);
+		nfs_data.root.size = request.fh->size;
+		memcpy(&nfs_data.root.data, request.fh->data, request.fh->size);
 	}
-
+	nfs_free_fhandle(request.fh);
+out:
 	return status;
 }
 

fs/nfs/pagelist.c

@@ -60,16 +60,10 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode,
 {
 	struct nfs_page		*req;
 
-	for (;;) {
-		/* try to allocate the request struct */
-		req = nfs_page_alloc();
-		if (req != NULL)
-			break;
-
-		if (fatal_signal_pending(current))
-			return ERR_PTR(-ERESTARTSYS);
-		yield();
-	}
+	/* try to allocate the request struct */
+	req = nfs_page_alloc();
+	if (req == NULL)
+		return ERR_PTR(-ENOMEM);
 
 	/* Initialize the request struct. Initially, we assume a
 	 * long write-back delay. This will be adjusted in

fs/nfs/proc.c

@@ -224,35 +224,60 @@ static int nfs_proc_readlink(struct inode *inode, struct page *page,
 	return status;
 }
 
+struct nfs_createdata {
+	struct nfs_createargs arg;
+	struct nfs_diropok res;
+	struct nfs_fh fhandle;
+	struct nfs_fattr fattr;
+};
+
+static struct nfs_createdata *nfs_alloc_createdata(struct inode *dir,
+		struct dentry *dentry, struct iattr *sattr)
+{
+	struct nfs_createdata *data;
+
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+
+	if (data != NULL) {
+		data->arg.fh = NFS_FH(dir);
+		data->arg.name = dentry->d_name.name;
+		data->arg.len = dentry->d_name.len;
+		data->arg.sattr = sattr;
+		nfs_fattr_init(&data->fattr);
+		data->fhandle.size = 0;
+		data->res.fh = &data->fhandle;
+		data->res.fattr = &data->fattr;
+	}
+	return data;
+};
+
+static void nfs_free_createdata(const struct nfs_createdata *data)
+{
+	kfree(data);
+}
+
 static int
 nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 		int flags, struct nameidata *nd)
 {
-	struct nfs_fh		fhandle;
-	struct nfs_fattr	fattr;
-	struct nfs_createargs	arg = {
-		.fh		= NFS_FH(dir),
-		.name		= dentry->d_name.name,
-		.len		= dentry->d_name.len,
-		.sattr		= sattr
-	};
-	struct nfs_diropok	res = {
-		.fh		= &fhandle,
-		.fattr		= &fattr
-	};
+	struct nfs_createdata *data;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs_procedures[NFSPROC_CREATE],
-		.rpc_argp	= &arg,
-		.rpc_resp	= &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 
-	nfs_fattr_init(&fattr);
 	dprintk("NFS call  create %s\n", dentry->d_name.name);
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
 	if (status == 0)
-		status = nfs_instantiate(dentry, &fhandle, &fattr);
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr);
+	nfs_free_createdata(data);
+out:
 	dprintk("NFS reply create: %d\n", status);
 	return status;
 }
@@ -264,24 +289,12 @@ static int
 nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 	       dev_t rdev)
 {
-	struct nfs_fh fhandle;
-	struct nfs_fattr fattr;
-	struct nfs_createargs	arg = {
-		.fh		= NFS_FH(dir),
-		.name		= dentry->d_name.name,
-		.len		= dentry->d_name.len,
-		.sattr		= sattr
-	};
-	struct nfs_diropok	res = {
-		.fh		= &fhandle,
-		.fattr		= &fattr
-	};
+	struct nfs_createdata *data;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs_procedures[NFSPROC_CREATE],
-		.rpc_argp	= &arg,
-		.rpc_resp	= &res,
 	};
-	int status, mode;
+	umode_t mode;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  mknod %s\n", dentry->d_name.name);
 
@@ -294,17 +307,24 @@ nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 		sattr->ia_size = new_encode_dev(rdev);/* get out your barf bag */
 	}
 
-	nfs_fattr_init(&fattr);
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
+
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
 
 	if (status == -EINVAL && S_ISFIFO(mode)) {
 		sattr->ia_mode = mode;
-		nfs_fattr_init(&fattr);
+		nfs_fattr_init(data->res.fattr);
 		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	}
 	if (status == 0)
-		status = nfs_instantiate(dentry, &fhandle, &fattr);
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr);
+	nfs_free_createdata(data);
+out:
 	dprintk("NFS reply mknod: %d\n", status);
 	return status;
 }
@@ -398,8 +418,8 @@ static int
 nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
 		 unsigned int len, struct iattr *sattr)
 {
-	struct nfs_fh fhandle;
-	struct nfs_fattr fattr;
+	struct nfs_fh *fh;
+	struct nfs_fattr *fattr;
 	struct nfs_symlinkargs	arg = {
 		.fromfh		= NFS_FH(dir),
 		.fromname	= dentry->d_name.name,
@@ -412,12 +432,18 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
 		.rpc_proc	= &nfs_procedures[NFSPROC_SYMLINK],
 		.rpc_argp	= &arg,
 	};
-	int			status;
+	int status = -ENAMETOOLONG;
+
+	dprintk("NFS call  symlink %s\n", dentry->d_name.name);
 
 	if (len > NFS2_MAXPATHLEN)
-		return -ENAMETOOLONG;
+		goto out;
 
-	dprintk("NFS call  symlink %s\n", dentry->d_name.name);
+	fh = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	status = -ENOMEM;
+	if (fh == NULL || fattr == NULL)
+		goto out;
 
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
@@ -427,12 +453,12 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
 	 * filehandle size to zero indicates to nfs_instantiate that it
 	 * should fill in the data with a LOOKUP call on the wire.
 	 */
-	if (status == 0) {
-		nfs_fattr_init(&fattr);
-		fhandle.size = 0;
-		status = nfs_instantiate(dentry, &fhandle, &fattr);
-	}
+	if (status == 0)
+		status = nfs_instantiate(dentry, fh, fattr);
 
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fh);
+out:
 	dprintk("NFS reply symlink: %d\n", status);
 	return status;
 }
@@ -440,31 +466,25 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
 static int
 nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
 {
-	struct nfs_fh fhandle;
-	struct nfs_fattr fattr;
-	struct nfs_createargs	arg = {
-		.fh		= NFS_FH(dir),
-		.name		= dentry->d_name.name,
-		.len		= dentry->d_name.len,
-		.sattr		= sattr
-	};
-	struct nfs_diropok	res = {
-		.fh		= &fhandle,
-		.fattr		= &fattr
-	};
+	struct nfs_createdata *data;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs_procedures[NFSPROC_MKDIR],
-		.rpc_argp	= &arg,
-		.rpc_resp	= &res,
 	};
-	int			status;
+	int status = -ENOMEM;
 
 	dprintk("NFS call  mkdir %s\n", dentry->d_name.name);
-	nfs_fattr_init(&fattr);
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
+
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
 	if (status == 0)
-		status = nfs_instantiate(dentry, &fhandle, &fattr);
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr);
+	nfs_free_createdata(data);
+out:
 	dprintk("NFS reply mkdir: %d\n", status);
 	return status;
 }

fs/nfs/read.c

@@ -40,7 +40,7 @@ static mempool_t *nfs_rdata_mempool;
 
 struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 {
-	struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, GFP_NOFS);
+	struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, GFP_KERNEL);
 
 	if (p) {
 		memset(p, 0, sizeof(*p));
@@ -50,7 +50,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 		if (pagecount <= ARRAY_SIZE(p->page_array))
 			p->pagevec = p->page_array;
 		else {
-			p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS);
+			p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_KERNEL);
 			if (!p->pagevec) {
 				mempool_free(p, nfs_rdata_mempool);
 				p = NULL;

fs/nfs/super.c

@@ -141,7 +141,6 @@ static const match_table_t nfs_mount_option_tokens = {
 	{ Opt_resvport, "resvport" },
 	{ Opt_noresvport, "noresvport" },
 	{ Opt_fscache, "fsc" },
-	{ Opt_fscache_uniq, "fsc=%s" },
 	{ Opt_nofscache, "nofsc" },
 
 	{ Opt_port, "port=%s" },
@@ -171,6 +170,7 @@ static const match_table_t nfs_mount_option_tokens = {
 	{ Opt_mountaddr, "mountaddr=%s" },
 
 	{ Opt_lookupcache, "lookupcache=%s" },
+	{ Opt_fscache_uniq, "fsc=%s" },
 
 	{ Opt_err, NULL }
 };
@@ -423,15 +423,19 @@ static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	unsigned char blockbits;
 	unsigned long blockres;
 	struct nfs_fh *fh = NFS_FH(dentry->d_inode);
-	struct nfs_fattr fattr;
-	struct nfs_fsstat res = {
-			.fattr = &fattr,
-	};
-	int error;
+	struct nfs_fsstat res;
+	int error = -ENOMEM;
+
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		goto out_err;
 
 	error = server->nfs_client->rpc_ops->statfs(server, fh, &res);
+
+	nfs_free_fattr(res.fattr);
 	if (error < 0)
 		goto out_err;
+
 	buf->f_type = NFS_SUPER_MAGIC;
 
 	/*
@@ -1046,14 +1050,6 @@ static int nfs_parse_mount_options(char *raw,
 			kfree(mnt->fscache_uniq);
 			mnt->fscache_uniq = NULL;
 			break;
-		case Opt_fscache_uniq:
-			string = match_strdup(args);
-			if (!string)
-				goto out_nomem;
-			kfree(mnt->fscache_uniq);
-			mnt->fscache_uniq = string;
-			mnt->options |= NFS_OPTION_FSCACHE;
-			break;
 
 		/*
 		 * options that take numeric values
@@ -1384,6 +1380,14 @@ static int nfs_parse_mount_options(char *raw,
 					return 0;
 			};
 			break;
+		case Opt_fscache_uniq:
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			kfree(mnt->fscache_uniq);
+			mnt->fscache_uniq = string;
+			mnt->options |= NFS_OPTION_FSCACHE;
+			break;
 
 		/*
 		 * Special options
@@ -2172,7 +2176,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 	int error = -ENOMEM;
 
 	data = nfs_alloc_parsed_mount_data(3);
-	mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
+	mntfh = nfs_alloc_fhandle();
 	if (data == NULL || mntfh == NULL)
 		goto out_free_fh;
 
@@ -2247,7 +2251,7 @@ out:
 	kfree(data->fscache_uniq);
 	security_free_mnt_opts(&data->lsm_opts);
 out_free_fh:
-	kfree(mntfh);
+	nfs_free_fhandle(mntfh);
 	kfree(data);
 	return error;
 
@@ -2556,7 +2560,7 @@ static int nfs4_remote_get_sb(struct file_system_type *fs_type,
 	};
 	int error = -ENOMEM;
 
-	mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
+	mntfh = nfs_alloc_fhandle();
 	if (data == NULL || mntfh == NULL)
 		goto out_free_fh;
 
@@ -2614,7 +2618,7 @@ static int nfs4_remote_get_sb(struct file_system_type *fs_type,
 out:
 	security_free_mnt_opts(&data->lsm_opts);
 out_free_fh:
-	kfree(mntfh);
+	nfs_free_fhandle(mntfh);
 	return error;
 
 out_free:
@@ -2669,41 +2673,120 @@ out_freepage:
 	free_page((unsigned long)page);
 }
 
+struct nfs_referral_count {
+	struct list_head list;
+	const struct task_struct *task;
+	unsigned int referral_count;
+};
+
+static LIST_HEAD(nfs_referral_count_list);
+static DEFINE_SPINLOCK(nfs_referral_count_list_lock);
+
+static struct nfs_referral_count *nfs_find_referral_count(void)
+{
+	struct nfs_referral_count *p;
+
+	list_for_each_entry(p, &nfs_referral_count_list, list) {
+		if (p->task == current)
+			return p;
+	}
+	return NULL;
+}
+
+#define NFS_MAX_NESTED_REFERRALS 2
+
+static int nfs_referral_loop_protect(void)
+{
+	struct nfs_referral_count *p, *new;
+	int ret = -ENOMEM;
+
+	new = kmalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		goto out;
+	new->task = current;
+	new->referral_count = 1;
+
+	ret = 0;
+	spin_lock(&nfs_referral_count_list_lock);
+	p = nfs_find_referral_count();
+	if (p != NULL) {
+		if (p->referral_count >= NFS_MAX_NESTED_REFERRALS)
+			ret = -ELOOP;
+		else
+			p->referral_count++;
+	} else {
+		list_add(&new->list, &nfs_referral_count_list);
+		new = NULL;
+	}
+	spin_unlock(&nfs_referral_count_list_lock);
+	kfree(new);
+out:
+	return ret;
+}
+
+static void nfs_referral_loop_unprotect(void)
+{
+	struct nfs_referral_count *p;
+
+	spin_lock(&nfs_referral_count_list_lock);
+	p = nfs_find_referral_count();
+	p->referral_count--;
+	if (p->referral_count == 0)
+		list_del(&p->list);
+	else
+		p = NULL;
+	spin_unlock(&nfs_referral_count_list_lock);
+	kfree(p);
+}
+
 static int nfs_follow_remote_path(struct vfsmount *root_mnt,
 		const char *export_path, struct vfsmount *mnt_target)
 {
+	struct nameidata *nd = NULL;
 	struct mnt_namespace *ns_private;
-	struct nameidata nd;
 	struct super_block *s;
 	int ret;
 
+	nd = kmalloc(sizeof(*nd), GFP_KERNEL);
+	if (nd == NULL)
+		return -ENOMEM;
+
 	ns_private = create_mnt_ns(root_mnt);
 	ret = PTR_ERR(ns_private);
 	if (IS_ERR(ns_private))
 		goto out_mntput;
 
+	ret = nfs_referral_loop_protect();
+	if (ret != 0)
+		goto out_put_mnt_ns;
+
 	ret = vfs_path_lookup(root_mnt->mnt_root, root_mnt,
-			export_path, LOOKUP_FOLLOW, &nd);
+			export_path, LOOKUP_FOLLOW, nd);
 
+	nfs_referral_loop_unprotect();
 	put_mnt_ns(ns_private);
 
 	if (ret != 0)
 		goto out_err;
 
-	s = nd.path.mnt->mnt_sb;
+	s = nd->path.mnt->mnt_sb;
 	atomic_inc(&s->s_active);
 	mnt_target->mnt_sb = s;
-	mnt_target->mnt_root = dget(nd.path.dentry);
+	mnt_target->mnt_root = dget(nd->path.dentry);
 
 	/* Correct the device pathname */
-	nfs_fix_devname(&nd.path, mnt_target);
+	nfs_fix_devname(&nd->path, mnt_target);
 
-	path_put(&nd.path);
+	path_put(&nd->path);
+	kfree(nd);
 	down_write(&s->s_umount);
 	return 0;
+out_put_mnt_ns:
+	put_mnt_ns(ns_private);
 out_mntput:
 	mntput(root_mnt);
 out_err:
+	kfree(nd);
 	return ret;
 }
 
@@ -2874,17 +2957,21 @@ static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
 	struct super_block *s;
 	struct nfs_server *server;
 	struct dentry *mntroot;
-	struct nfs_fh mntfh;
+	struct nfs_fh *mntfh;
 	int (*compare_super)(struct super_block *, void *) = nfs_compare_super;
 	struct nfs_sb_mountdata sb_mntdata = {
 		.mntflags = flags,
 	};
-	int error;
+	int error = -ENOMEM;
 
 	dprintk("--> nfs4_referral_get_sb()\n");
 
+	mntfh = nfs_alloc_fhandle();
+	if (mntfh == NULL)
+		goto out_err_nofh;
+
 	/* create a new volume representation */
-	server = nfs4_create_referral_server(data, &mntfh);
+	server = nfs4_create_referral_server(data, mntfh);
 	if (IS_ERR(server)) {
 		error = PTR_ERR(server);
 		goto out_err_noserver;
@@ -2916,7 +3003,7 @@ static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
 		nfs_fscache_get_super_cookie(s, NULL, data);
 	}
 
-	mntroot = nfs4_get_root(s, &mntfh);
+	mntroot = nfs4_get_root(s, mntfh);
 	if (IS_ERR(mntroot)) {
 		error = PTR_ERR(mntroot);
 		goto error_splat_super;
@@ -2933,12 +3020,15 @@ static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
 
 	security_sb_clone_mnt_opts(data->sb, s);
 
+	nfs_free_fhandle(mntfh);
 	dprintk("<-- nfs4_referral_get_sb() = 0\n");
 	return 0;
 
 out_err_nosb:
 	nfs_free_server(server);
 out_err_noserver:
+	nfs_free_fhandle(mntfh);
+out_err_nofh:
 	dprintk("<-- nfs4_referral_get_sb() = %d [error]\n", error);
 	return error;
 
@@ -2947,6 +3037,7 @@ error_splat_super:
 		bdi_unregister(&server->backing_dev_info);
 error_splat_bdi:
 	deactivate_locked_super(s);
+	nfs_free_fhandle(mntfh);
 	dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error);
 	return error;
 }

fs/nfs/unlink.c

@@ -23,6 +23,7 @@ struct nfs_unlinkdata {
 	struct nfs_removeres res;
 	struct inode *dir;
 	struct rpc_cred	*cred;
+	struct nfs_fattr dir_attr;
 };
 
 /**
@@ -169,7 +170,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
 	}
 	nfs_sb_active(dir->i_sb);
 	data->args.fh = NFS_FH(dir);
-	nfs_fattr_init(&data->res.dir_attr);
+	nfs_fattr_init(data->res.dir_attr);
 
 	NFS_PROTO(dir)->unlink_setup(&msg, dir);
 
@@ -259,6 +260,7 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry)
 		goto out_free;
 	}
 	data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
+	data->res.dir_attr = &data->dir_attr;
 
 	status = -EBUSY;
 	spin_lock(&dentry->d_lock);

include/linux/ktime.h

@@ -130,7 +130,7 @@ static inline ktime_t timeval_to_ktime(struct timeval tv)
 /* Convert ktime_t to nanoseconds - NOP in the scalar storage format: */
 #define ktime_to_ns(kt)			((kt).tv64)
 
-#else
+#else	/* !((BITS_PER_LONG == 64) || defined(CONFIG_KTIME_SCALAR)) */
 
 /*
  * Helper macros/inlines to get the ktime_t math right in the timespec
@@ -275,7 +275,7 @@ static inline s64 ktime_to_ns(const ktime_t kt)
 	return (s64) kt.tv.sec * NSEC_PER_SEC + kt.tv.nsec;
 }
 
-#endif
+#endif	/* !((BITS_PER_LONG == 64) || defined(CONFIG_KTIME_SCALAR)) */
 
 /**
  * ktime_equal - Compares two ktime_t variables to see if they are equal
@@ -295,6 +295,12 @@ static inline s64 ktime_to_us(const ktime_t kt)
 	return (s64) tv.tv_sec * USEC_PER_SEC + tv.tv_usec;
 }
 
+static inline s64 ktime_to_ms(const ktime_t kt)
+{
+	struct timeval tv = ktime_to_timeval(kt);
+	return (s64) tv.tv_sec * MSEC_PER_SEC + tv.tv_usec / USEC_PER_MSEC;
+}
+
 static inline s64 ktime_us_delta(const ktime_t later, const ktime_t earlier)
 {
        return ktime_to_us(ktime_sub(later, earlier));

include/linux/nfs_fs.h

@@ -356,6 +356,20 @@ extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struc
 extern u64 nfs_compat_user_ino64(u64 fileid);
 extern void nfs_fattr_init(struct nfs_fattr *fattr);
 
+extern struct nfs_fattr *nfs_alloc_fattr(void);
+
+static inline void nfs_free_fattr(const struct nfs_fattr *fattr)
+{
+	kfree(fattr);
+}
+
+extern struct nfs_fh *nfs_alloc_fhandle(void);
+
+static inline void nfs_free_fhandle(const struct nfs_fh *fh)
+{
+	kfree(fh);
+}
+
 /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */
 extern __be32 root_nfs_parse_addr(char *name); /*__init*/
 extern unsigned long nfs_inc_attr_generation_counter(void);

include/linux/nfs_fs_sb.h

@@ -44,7 +44,6 @@ struct nfs_client {
 
 #ifdef CONFIG_NFS_V4
 	u64			cl_clientid;	/* constant */
-	nfs4_verifier		cl_confirm;
 	unsigned long		cl_state;
 
 	struct rb_root		cl_openowner_id;

include/linux/nfs_xdr.h

@@ -386,8 +386,8 @@ struct nfs_removeargs {
 
 struct nfs_removeres {
 	const struct nfs_server *server;
+	struct nfs_fattr	*dir_attr;
 	struct nfs4_change_info	cinfo;
-	struct nfs_fattr	dir_attr;
 	struct nfs4_sequence_res 	seq_res;
 };
 
@@ -824,6 +824,11 @@ struct nfs4_setclientid {
 	u32				sc_cb_ident;
 };
 
+struct nfs4_setclientid_res {
+	u64				clientid;
+	nfs4_verifier			confirm;
+};
+
 struct nfs4_statfs_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;

include/linux/sunrpc/auth.h

@@ -54,6 +54,7 @@ struct rpc_cred {
 #define RPCAUTH_CRED_NEW	0
 #define RPCAUTH_CRED_UPTODATE	1
 #define RPCAUTH_CRED_HASHED	2
+#define RPCAUTH_CRED_NEGATIVE	3
 
 #define RPCAUTH_CRED_MAGIC	0x0f4aa4f0
 

include/linux/sunrpc/auth_gss.h

@@ -82,6 +82,7 @@ struct gss_cred {
 	enum rpc_gss_svc	gc_service;
 	struct gss_cl_ctx	*gc_ctx;
 	struct gss_upcall_msg	*gc_upcall;
+	unsigned long		gc_upcall_timestamp;
 	unsigned char		gc_machine_cred : 1;
 };
 

include/linux/sunrpc/gss_api.h

@@ -35,7 +35,8 @@ int gss_import_sec_context(
 		const void*		input_token,
 		size_t			bufsize,
 		struct gss_api_mech	*mech,
-		struct gss_ctx		**ctx_id);
+		struct gss_ctx		**ctx_id,
+		gfp_t			gfp_mask);
 u32 gss_get_mic(
 		struct gss_ctx		*ctx_id,
 		struct xdr_buf		*message,
@@ -80,6 +81,8 @@ struct gss_api_mech {
 	/* pseudoflavors supported by this mechanism: */
 	int			gm_pf_num;
 	struct pf_desc *	gm_pfs;
+	/* Should the following be a callback operation instead? */
+	const char		*gm_upcall_enctypes;
 };
 
 /* and must provide the following operations: */
@@ -87,7 +90,8 @@ struct gss_api_ops {
 	int (*gss_import_sec_context)(
 			const void		*input_token,
 			size_t			bufsize,
-			struct gss_ctx		*ctx_id);
+			struct gss_ctx		*ctx_id,
+			gfp_t			gfp_mask);
 	u32 (*gss_get_mic)(
 			struct gss_ctx		*ctx_id,
 			struct xdr_buf		*message,

include/linux/sunrpc/gss_krb5.h

@@ -4,7 +4,7 @@
  *  Adapted from MIT Kerberos 5-1.2.1 lib/include/krb5.h,
  *  lib/gssapi/krb5/gssapiP_krb5.h, and others
  *
- *  Copyright (c) 2000 The Regents of the University of Michigan.
+ *  Copyright (c) 2000-2008 The Regents of the University of Michigan.
  *  All rights reserved.
  *
  *  Andy Adamson   <andros@umich.edu>
@@ -36,17 +36,86 @@
  *
  */
 
+#include <linux/crypto.h>
 #include <linux/sunrpc/auth_gss.h>
 #include <linux/sunrpc/gss_err.h>
 #include <linux/sunrpc/gss_asn1.h>
 
+/* Length of constant used in key derivation */
+#define GSS_KRB5_K5CLENGTH (5)
+
+/* Maximum key length (in bytes) for the supported crypto algorithms*/
+#define GSS_KRB5_MAX_KEYLEN (32)
+
+/* Maximum checksum function output for the supported crypto algorithms */
+#define GSS_KRB5_MAX_CKSUM_LEN  (20)
+
+/* Maximum blocksize for the supported crypto algorithms */
+#define GSS_KRB5_MAX_BLOCKSIZE  (16)
+
+struct krb5_ctx;
+
+struct gss_krb5_enctype {
+	const u32		etype;		/* encryption (key) type */
+	const u32		ctype;		/* checksum type */
+	const char		*name;		/* "friendly" name */
+	const char		*encrypt_name;	/* crypto encrypt name */
+	const char		*cksum_name;	/* crypto checksum name */
+	const u16		signalg;	/* signing algorithm */
+	const u16		sealalg;	/* sealing algorithm */
+	const u32		blocksize;	/* encryption blocksize */
+	const u32		conflen;	/* confounder length
+						   (normally the same as
+						   the blocksize) */
+	const u32		cksumlength;	/* checksum length */
+	const u32		keyed_cksum;	/* is it a keyed cksum? */
+	const u32		keybytes;	/* raw key len, in bytes */
+	const u32		keylength;	/* final key len, in bytes */
+	u32 (*encrypt) (struct crypto_blkcipher *tfm,
+			void *iv, void *in, void *out,
+			int length);		/* encryption function */
+	u32 (*decrypt) (struct crypto_blkcipher *tfm,
+			void *iv, void *in, void *out,
+			int length);		/* decryption function */
+	u32 (*mk_key) (const struct gss_krb5_enctype *gk5e,
+		       struct xdr_netobj *in,
+		       struct xdr_netobj *out);	/* complete key generation */
+	u32 (*encrypt_v2) (struct krb5_ctx *kctx, u32 offset,
+			   struct xdr_buf *buf, int ec,
+			   struct page **pages); /* v2 encryption function */
+	u32 (*decrypt_v2) (struct krb5_ctx *kctx, u32 offset,
+			   struct xdr_buf *buf, u32 *headskip,
+			   u32 *tailskip);	/* v2 decryption function */
+};
+
+/* krb5_ctx flags definitions */
+#define KRB5_CTX_FLAG_INITIATOR         0x00000001
+#define KRB5_CTX_FLAG_CFX               0x00000002
+#define KRB5_CTX_FLAG_ACCEPTOR_SUBKEY   0x00000004
+
 struct krb5_ctx {
 	int			initiate; /* 1 = initiating, 0 = accepting */
+	u32			enctype;
+	u32			flags;
+	const struct gss_krb5_enctype *gk5e; /* enctype-specific info */
 	struct crypto_blkcipher	*enc;
 	struct crypto_blkcipher	*seq;
+	struct crypto_blkcipher *acceptor_enc;
+	struct crypto_blkcipher *initiator_enc;
+	struct crypto_blkcipher *acceptor_enc_aux;
+	struct crypto_blkcipher *initiator_enc_aux;
+	u8			Ksess[GSS_KRB5_MAX_KEYLEN]; /* session key */
+	u8			cksum[GSS_KRB5_MAX_KEYLEN];
 	s32			endtime;
 	u32			seq_send;
+	u64			seq_send64;
 	struct xdr_netobj	mech_used;
+	u8			initiator_sign[GSS_KRB5_MAX_KEYLEN];
+	u8			acceptor_sign[GSS_KRB5_MAX_KEYLEN];
+	u8			initiator_seal[GSS_KRB5_MAX_KEYLEN];
+	u8			acceptor_seal[GSS_KRB5_MAX_KEYLEN];
+	u8			initiator_integ[GSS_KRB5_MAX_KEYLEN];
+	u8			acceptor_integ[GSS_KRB5_MAX_KEYLEN];
 };
 
 extern spinlock_t krb5_seq_lock;
@@ -57,6 +126,18 @@ extern spinlock_t krb5_seq_lock;
 #define KG_TOK_MIC_MSG    0x0101
 #define KG_TOK_WRAP_MSG   0x0201
 
+#define KG2_TOK_INITIAL     0x0101
+#define KG2_TOK_RESPONSE    0x0202
+#define KG2_TOK_MIC         0x0404
+#define KG2_TOK_WRAP        0x0504
+
+#define KG2_TOKEN_FLAG_SENTBYACCEPTOR   0x01
+#define KG2_TOKEN_FLAG_SEALED           0x02
+#define KG2_TOKEN_FLAG_ACCEPTORSUBKEY   0x04
+
+#define KG2_RESP_FLAG_ERROR             0x0001
+#define KG2_RESP_FLAG_DELEG_OK          0x0002
+
 enum sgn_alg {
 	SGN_ALG_DES_MAC_MD5 = 0x0000,
 	SGN_ALG_MD2_5 = 0x0001,
@@ -81,6 +162,9 @@ enum seal_alg {
 #define CKSUMTYPE_RSA_MD5_DES		0x0008
 #define CKSUMTYPE_NIST_SHA		0x0009
 #define CKSUMTYPE_HMAC_SHA1_DES3	0x000c
+#define CKSUMTYPE_HMAC_SHA1_96_AES128   0x000f
+#define CKSUMTYPE_HMAC_SHA1_96_AES256   0x0010
+#define CKSUMTYPE_HMAC_MD5_ARCFOUR      -138 /* Microsoft md5 hmac cksumtype */
 
 /* from gssapi_err_krb5.h */
 #define KG_CCACHE_NOMATCH                        (39756032L)
@@ -111,11 +195,56 @@ enum seal_alg {
 #define ENCTYPE_DES3_CBC_RAW    0x0006	/* DES-3 cbc mode raw */
 #define ENCTYPE_DES_HMAC_SHA1   0x0008
 #define ENCTYPE_DES3_CBC_SHA1   0x0010
+#define ENCTYPE_AES128_CTS_HMAC_SHA1_96 0x0011
+#define ENCTYPE_AES256_CTS_HMAC_SHA1_96 0x0012
+#define ENCTYPE_ARCFOUR_HMAC            0x0017
+#define ENCTYPE_ARCFOUR_HMAC_EXP        0x0018
 #define ENCTYPE_UNKNOWN         0x01ff
 
-s32
-make_checksum(char *, char *header, int hdrlen, struct xdr_buf *body,
-		   int body_offset, struct xdr_netobj *cksum);
+/*
+ * Constants used for key derivation
+ */
+/* for 3DES */
+#define KG_USAGE_SEAL (22)
+#define KG_USAGE_SIGN (23)
+#define KG_USAGE_SEQ  (24)
+
+/* from rfc3961 */
+#define KEY_USAGE_SEED_CHECKSUM         (0x99)
+#define KEY_USAGE_SEED_ENCRYPTION       (0xAA)
+#define KEY_USAGE_SEED_INTEGRITY        (0x55)
+
+/* from rfc4121 */
+#define KG_USAGE_ACCEPTOR_SEAL  (22)
+#define KG_USAGE_ACCEPTOR_SIGN  (23)
+#define KG_USAGE_INITIATOR_SEAL (24)
+#define KG_USAGE_INITIATOR_SIGN (25)
+
+/*
+ * This compile-time check verifies that we will not exceed the
+ * slack space allotted by the client and server auth_gss code
+ * before they call gss_wrap().
+ */
+#define GSS_KRB5_MAX_SLACK_NEEDED \
+	(GSS_KRB5_TOK_HDR_LEN     /* gss token header */         \
+	+ GSS_KRB5_MAX_CKSUM_LEN  /* gss token checksum */       \
+	+ GSS_KRB5_MAX_BLOCKSIZE  /* confounder */               \
+	+ GSS_KRB5_MAX_BLOCKSIZE  /* possible padding */         \
+	+ GSS_KRB5_TOK_HDR_LEN    /* encrypted hdr in v2 token */\
+	+ GSS_KRB5_MAX_CKSUM_LEN  /* encryption hmac */          \
+	+ 4 + 4                   /* RPC verifier */             \
+	+ GSS_KRB5_TOK_HDR_LEN                                   \
+	+ GSS_KRB5_MAX_CKSUM_LEN)
+
+u32
+make_checksum(struct krb5_ctx *kctx, char *header, int hdrlen,
+		struct xdr_buf *body, int body_offset, u8 *cksumkey,
+		unsigned int usage, struct xdr_netobj *cksumout);
+
+u32
+make_checksum_v2(struct krb5_ctx *, char *header, int hdrlen,
+		 struct xdr_buf *body, int body_offset, u8 *key,
+		 unsigned int usage, struct xdr_netobj *cksum);
 
 u32 gss_get_mic_kerberos(struct gss_ctx *, struct xdr_buf *,
 		struct xdr_netobj *);
@@ -149,11 +278,54 @@ gss_decrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *inbuf,
 		    int offset);
 
 s32
-krb5_make_seq_num(struct crypto_blkcipher *key,
+krb5_make_seq_num(struct krb5_ctx *kctx,
+		struct crypto_blkcipher *key,
 		int direction,
 		u32 seqnum, unsigned char *cksum, unsigned char *buf);
 
 s32
-krb5_get_seq_num(struct crypto_blkcipher *key,
+krb5_get_seq_num(struct krb5_ctx *kctx,
 	       unsigned char *cksum,
 	       unsigned char *buf, int *direction, u32 *seqnum);
+
+int
+xdr_extend_head(struct xdr_buf *buf, unsigned int base, unsigned int shiftlen);
+
+u32
+krb5_derive_key(const struct gss_krb5_enctype *gk5e,
+		const struct xdr_netobj *inkey,
+		struct xdr_netobj *outkey,
+		const struct xdr_netobj *in_constant,
+		gfp_t gfp_mask);
+
+u32
+gss_krb5_des3_make_key(const struct gss_krb5_enctype *gk5e,
+		       struct xdr_netobj *randombits,
+		       struct xdr_netobj *key);
+
+u32
+gss_krb5_aes_make_key(const struct gss_krb5_enctype *gk5e,
+		      struct xdr_netobj *randombits,
+		      struct xdr_netobj *key);
+
+u32
+gss_krb5_aes_encrypt(struct krb5_ctx *kctx, u32 offset,
+		     struct xdr_buf *buf, int ec,
+		     struct page **pages);
+
+u32
+gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset,
+		     struct xdr_buf *buf, u32 *plainoffset,
+		     u32 *plainlen);
+
+int
+krb5_rc4_setup_seq_key(struct krb5_ctx *kctx,
+		       struct crypto_blkcipher *cipher,
+		       unsigned char *cksum);
+
+int
+krb5_rc4_setup_enc_key(struct krb5_ctx *kctx,
+		       struct crypto_blkcipher *cipher,
+		       s32 seqnum);
+void
+gss_krb5_make_confounder(char *p, u32 conflen);

include/linux/sunrpc/metrics.h

@@ -26,6 +26,7 @@
 #define _LINUX_SUNRPC_METRICS_H
 
 #include <linux/seq_file.h>
+#include <linux/ktime.h>
 
 #define RPC_IOSTATS_VERS	"1.0"
 
@@ -58,9 +59,9 @@ struct rpc_iostats {
 	 * and the total time the request spent from init to release
 	 * are measured.
 	 */
-	unsigned long long	om_queue,	/* jiffies queued for xmit */
-				om_rtt,		/* jiffies for RPC RTT */
-				om_execute;	/* jiffies for RPC execution */
+	ktime_t			om_queue,	/* queued for xmit */
+				om_rtt,		/* RPC RTT */
+				om_execute;	/* RPC execution */
 } ____cacheline_aligned;
 
 struct rpc_task;

include/linux/sunrpc/sched.h

@@ -10,6 +10,7 @@
 #define _LINUX_SUNRPC_SCHED_H_
 
 #include <linux/timer.h>
+#include <linux/ktime.h>
 #include <linux/sunrpc/types.h>
 #include <linux/spinlock.h>
 #include <linux/wait.h>
@@ -40,21 +41,15 @@ struct rpc_wait {
  * This is the RPC task struct
  */
 struct rpc_task {
-#ifdef RPC_DEBUG
-	unsigned long		tk_magic;	/* 0xf00baa */
-#endif
 	atomic_t		tk_count;	/* Reference count */
 	struct list_head	tk_task;	/* global list of tasks */
 	struct rpc_clnt *	tk_client;	/* RPC client */
 	struct rpc_rqst *	tk_rqstp;	/* RPC request */
-	int			tk_status;	/* result of last operation */
 
 	/*
 	 * RPC call state
 	 */
 	struct rpc_message	tk_msg;		/* RPC call info */
-	__u8			tk_garb_retry;
-	__u8			tk_cred_retry;
 
 	/*
 	 * callback	to be executed after waking up
@@ -67,7 +62,6 @@ struct rpc_task {
 	void *			tk_calldata;
 
 	unsigned long		tk_timeout;	/* timeout for rpc_sleep() */
-	unsigned short		tk_flags;	/* misc flags */
 	unsigned long		tk_runstate;	/* Task run status */
 	struct workqueue_struct	*tk_workqueue;	/* Normally rpciod, but could
 						 * be any workqueue
@@ -78,17 +72,19 @@ struct rpc_task {
 		struct rpc_wait		tk_wait;	/* RPC wait */
 	} u;
 
-	unsigned short		tk_timeouts;	/* maj timeouts */
-	size_t			tk_bytes_sent;	/* total bytes sent */
-	unsigned long		tk_start;	/* RPC task init timestamp */
-	long			tk_rtt;		/* round-trip time (jiffies) */
+	ktime_t			tk_start;	/* RPC task init timestamp */
 
 	pid_t			tk_owner;	/* Process id for batching tasks */
-	unsigned char		tk_priority : 2;/* Task priority */
+	int			tk_status;	/* result of last operation */
+	unsigned short		tk_flags;	/* misc flags */
+	unsigned short		tk_timeouts;	/* maj timeouts */
 
 #ifdef RPC_DEBUG
 	unsigned short		tk_pid;		/* debugging aid */
 #endif
+	unsigned char		tk_priority : 2,/* Task priority */
+				tk_garb_retry : 2,
+				tk_cred_retry : 2;
 };
 #define tk_xprt			tk_client->cl_xprt
 

include/linux/sunrpc/xdr.h

@@ -1,7 +1,10 @@
 /*
- * include/linux/sunrpc/xdr.h
+ * XDR standard data types and function declarations
  *
  * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
+ *
+ * Based on:
+ *   RFC 4506 "XDR: External Data Representation Standard", May 2006
  */
 
 #ifndef _SUNRPC_XDR_H_
@@ -62,7 +65,6 @@ struct xdr_buf {
 
 	unsigned int	buflen,		/* Total length of storage buffer */
 			len;		/* Length of XDR encoded message */
-
 };
 
 /*
@@ -178,7 +180,7 @@ struct xdr_array2_desc {
 };
 
 extern int xdr_decode_array2(struct xdr_buf *buf, unsigned int base,
-                             struct xdr_array2_desc *desc);
+			     struct xdr_array2_desc *desc);
 extern int xdr_encode_array2(struct xdr_buf *buf, unsigned int base,
 			     struct xdr_array2_desc *desc);
 

include/linux/sunrpc/xprt.h

@@ -13,6 +13,7 @@
 #include <linux/socket.h>
 #include <linux/in.h>
 #include <linux/kref.h>
+#include <linux/ktime.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/xdr.h>
 #include <linux/sunrpc/msg_prot.h>
@@ -65,8 +66,6 @@ struct rpc_rqst {
 	struct rpc_task *	rq_task;	/* RPC task data */
 	__be32			rq_xid;		/* request XID */
 	int			rq_cong;	/* has incremented xprt->cong */
-	int			rq_reply_bytes_recvd;	/* number of reply */
-							/* bytes received */
 	u32			rq_seqno;	/* gss seq no. used on req. */
 	int			rq_enc_pages_num;
 	struct page		**rq_enc_pages;	/* scratch pages for use by
@@ -77,12 +76,16 @@ struct rpc_rqst {
 	__u32 *			rq_buffer;	/* XDR encode buffer */
 	size_t			rq_callsize,
 				rq_rcvsize;
+	size_t			rq_xmit_bytes_sent;	/* total bytes sent */
+	size_t			rq_reply_bytes_recvd;	/* total reply bytes */
+							/* received */
 
 	struct xdr_buf		rq_private_buf;		/* The receive buffer
 							 * used in the softirq.
 							 */
 	unsigned long		rq_majortimeo;	/* major timeout alarm */
 	unsigned long		rq_timeout;	/* Current timeout value */
+	ktime_t			rq_rtt;		/* round-trip time */
 	unsigned int		rq_retries;	/* # of retries */
 	unsigned int		rq_connect_cookie;
 						/* A cookie used to track the
@@ -94,7 +97,7 @@ struct rpc_rqst {
 	 */
 	u32			rq_bytes_sent;	/* Bytes we have sent */
 
-	unsigned long		rq_xtime;	/* when transmitted */
+	ktime_t			rq_xtime;	/* transmit time stamp */
 	int			rq_ntrans;
 
 #if defined(CONFIG_NFS_V4_1)
@@ -174,8 +177,7 @@ struct rpc_xprt {
 	/*
 	 * Connection of transports
 	 */
-	unsigned long		connect_timeout,
-				bind_timeout,
+	unsigned long		bind_timeout,
 				reestablish_timeout;
 	unsigned int		connect_cookie;	/* A cookie that gets bumped
 						   every time the transport
@@ -294,7 +296,6 @@ void			xprt_set_retrans_timeout_rtt(struct rpc_task *task);
 void			xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status);
 void			xprt_wait_for_buffer_space(struct rpc_task *task, rpc_action action);
 void			xprt_write_space(struct rpc_xprt *xprt);
-void			xprt_update_rtt(struct rpc_task *task);
 void			xprt_adjust_cwnd(struct rpc_task *task, int result);
 struct rpc_rqst *	xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid);
 void			xprt_complete_rqst(struct rpc_task *task, int copied);

net/sunrpc/auth.c

@@ -236,10 +236,15 @@ rpcauth_prune_expired(struct list_head *free, int nr_to_scan)
 
 	list_for_each_entry_safe(cred, next, &cred_unused, cr_lru) {
 
-		/* Enforce a 60 second garbage collection moratorium */
+		if (nr_to_scan-- == 0)
+			break;
+		/*
+		 * Enforce a 60 second garbage collection moratorium
+		 * Note that the cred_unused list must be time-ordered.
+		 */
 		if (time_in_range(cred->cr_expire, expired, jiffies) &&
 		    test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) != 0)
-			continue;
+			return 0;
 
 		list_del_init(&cred->cr_lru);
 		number_cred_unused--;
@@ -252,13 +257,10 @@ rpcauth_prune_expired(struct list_head *free, int nr_to_scan)
 			get_rpccred(cred);
 			list_add_tail(&cred->cr_lru, free);
 			rpcauth_unhash_cred_locked(cred);
-			nr_to_scan--;
 		}
 		spin_unlock(cache_lock);
-		if (nr_to_scan == 0)
-			break;
 	}
-	return nr_to_scan;
+	return (number_cred_unused / 100) * sysctl_vfs_cache_pressure;
 }
 
 /*
@@ -270,11 +272,12 @@ rpcauth_cache_shrinker(int nr_to_scan, gfp_t gfp_mask)
 	LIST_HEAD(free);
 	int res;
 
+	if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL)
+		return (nr_to_scan == 0) ? 0 : -1;
 	if (list_empty(&cred_unused))
 		return 0;
 	spin_lock(&rpc_credcache_lock);
-	nr_to_scan = rpcauth_prune_expired(&free, nr_to_scan);
-	res = (number_cred_unused / 100) * sysctl_vfs_cache_pressure;
+	res = rpcauth_prune_expired(&free, nr_to_scan);
 	spin_unlock(&rpc_credcache_lock);
 	rpcauth_destroy_credlist(&free);
 	return res;

net/sunrpc/auth_gss/Makefile

@@ -10,7 +10,7 @@ auth_rpcgss-objs := auth_gss.o gss_generic_token.o \
 obj-$(CONFIG_RPCSEC_GSS_KRB5) += rpcsec_gss_krb5.o
 
 rpcsec_gss_krb5-objs := gss_krb5_mech.o gss_krb5_seal.o gss_krb5_unseal.o \
-	gss_krb5_seqnum.o gss_krb5_wrap.o gss_krb5_crypto.o
+	gss_krb5_seqnum.o gss_krb5_wrap.o gss_krb5_crypto.o gss_krb5_keys.o
 
 obj-$(CONFIG_RPCSEC_GSS_SPKM3) += rpcsec_gss_spkm3.o
 

net/sunrpc/auth_gss/auth_gss.c

@@ -57,11 +57,14 @@ static const struct rpc_authops authgss_ops;
 static const struct rpc_credops gss_credops;
 static const struct rpc_credops gss_nullops;
 
+#define GSS_RETRY_EXPIRED 5
+static unsigned int gss_expired_cred_retry_delay = GSS_RETRY_EXPIRED;
+
 #ifdef RPC_DEBUG
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
-#define GSS_CRED_SLACK		1024
+#define GSS_CRED_SLACK		(RPC_MAX_AUTH_SIZE * 2)
 /* length of a krb5 verifier (48), plus data added before arguments when
  * using integrity (two 4-byte integers): */
 #define GSS_VERF_SLACK		100
@@ -229,7 +232,7 @@ gss_fill_context(const void *p, const void *end, struct gss_cl_ctx *ctx, struct
 		p = ERR_PTR(-EFAULT);
 		goto err;
 	}
-	ret = gss_import_sec_context(p, seclen, gm, &ctx->gc_gss_ctx);
+	ret = gss_import_sec_context(p, seclen, gm, &ctx->gc_gss_ctx, GFP_NOFS);
 	if (ret < 0) {
 		p = ERR_PTR(ret);
 		goto err;
@@ -349,6 +352,24 @@ gss_unhash_msg(struct gss_upcall_msg *gss_msg)
 	spin_unlock(&inode->i_lock);
 }
 
+static void
+gss_handle_downcall_result(struct gss_cred *gss_cred, struct gss_upcall_msg *gss_msg)
+{
+	switch (gss_msg->msg.errno) {
+	case 0:
+		if (gss_msg->ctx == NULL)
+			break;
+		clear_bit(RPCAUTH_CRED_NEGATIVE, &gss_cred->gc_base.cr_flags);
+		gss_cred_set_ctx(&gss_cred->gc_base, gss_msg->ctx);
+		break;
+	case -EKEYEXPIRED:
+		set_bit(RPCAUTH_CRED_NEGATIVE, &gss_cred->gc_base.cr_flags);
+	}
+	gss_cred->gc_upcall_timestamp = jiffies;
+	gss_cred->gc_upcall = NULL;
+	rpc_wake_up_status(&gss_msg->rpc_waitqueue, gss_msg->msg.errno);
+}
+
 static void
 gss_upcall_callback(struct rpc_task *task)
 {
@@ -358,13 +379,9 @@ gss_upcall_callback(struct rpc_task *task)
 	struct inode *inode = &gss_msg->inode->vfs_inode;
 
 	spin_lock(&inode->i_lock);
-	if (gss_msg->ctx)
-		gss_cred_set_ctx(task->tk_msg.rpc_cred, gss_msg->ctx);
-	else
-		task->tk_status = gss_msg->msg.errno;
-	gss_cred->gc_upcall = NULL;
-	rpc_wake_up_status(&gss_msg->rpc_waitqueue, gss_msg->msg.errno);
+	gss_handle_downcall_result(gss_cred, gss_msg);
 	spin_unlock(&inode->i_lock);
+	task->tk_status = gss_msg->msg.errno;
 	gss_release_msg(gss_msg);
 }
 
@@ -377,11 +394,12 @@ static void gss_encode_v0_msg(struct gss_upcall_msg *gss_msg)
 static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg,
 				struct rpc_clnt *clnt, int machine_cred)
 {
+	struct gss_api_mech *mech = gss_msg->auth->mech;
 	char *p = gss_msg->databuf;
 	int len = 0;
 
 	gss_msg->msg.len = sprintf(gss_msg->databuf, "mech=%s uid=%d ",
-				   gss_msg->auth->mech->gm_name,
+				   mech->gm_name,
 				   gss_msg->uid);
 	p += gss_msg->msg.len;
 	if (clnt->cl_principal) {
@@ -398,6 +416,11 @@ static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg,
 		p += len;
 		gss_msg->msg.len += len;
 	}
+	if (mech->gm_upcall_enctypes) {
+		len = sprintf(p, mech->gm_upcall_enctypes);
+		p += len;
+		gss_msg->msg.len += len;
+	}
 	len = sprintf(p, "\n");
 	gss_msg->msg.len += len;
 
@@ -507,18 +530,16 @@ gss_refresh_upcall(struct rpc_task *task)
 	spin_lock(&inode->i_lock);
 	if (gss_cred->gc_upcall != NULL)
 		rpc_sleep_on(&gss_cred->gc_upcall->rpc_waitqueue, task, NULL);
-	else if (gss_msg->ctx != NULL) {
-		gss_cred_set_ctx(task->tk_msg.rpc_cred, gss_msg->ctx);
-		gss_cred->gc_upcall = NULL;
-		rpc_wake_up_status(&gss_msg->rpc_waitqueue, gss_msg->msg.errno);
-	} else if (gss_msg->msg.errno >= 0) {
+	else if (gss_msg->ctx == NULL && gss_msg->msg.errno >= 0) {
 		task->tk_timeout = 0;
 		gss_cred->gc_upcall = gss_msg;
 		/* gss_upcall_callback will release the reference to gss_upcall_msg */
 		atomic_inc(&gss_msg->count);
 		rpc_sleep_on(&gss_msg->rpc_waitqueue, task, gss_upcall_callback);
-	} else
+	} else {
+		gss_handle_downcall_result(gss_cred, gss_msg);
 		err = gss_msg->msg.errno;
+	}
 	spin_unlock(&inode->i_lock);
 	gss_release_msg(gss_msg);
 out:
@@ -1117,6 +1138,23 @@ static int gss_renew_cred(struct rpc_task *task)
 	return 0;
 }
 
+static int gss_cred_is_negative_entry(struct rpc_cred *cred)
+{
+	if (test_bit(RPCAUTH_CRED_NEGATIVE, &cred->cr_flags)) {
+		unsigned long now = jiffies;
+		unsigned long begin, expire;
+		struct gss_cred *gss_cred; 
+
+		gss_cred = container_of(cred, struct gss_cred, gc_base);
+		begin = gss_cred->gc_upcall_timestamp;
+		expire = begin + gss_expired_cred_retry_delay * HZ;
+
+		if (time_in_range_open(now, begin, expire))
+			return 1;
+	}
+	return 0;
+}
+
 /*
 * Refresh credentials. XXX - finish
 */
@@ -1126,6 +1164,9 @@ gss_refresh(struct rpc_task *task)
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 	int ret = 0;
 
+	if (gss_cred_is_negative_entry(cred))
+		return -EKEYEXPIRED;
+
 	if (!test_bit(RPCAUTH_CRED_NEW, &cred->cr_flags) &&
 			!test_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags)) {
 		ret = gss_renew_cred(task);
@@ -1316,15 +1357,21 @@ gss_wrap_req_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
 	inpages = snd_buf->pages + first;
 	snd_buf->pages = rqstp->rq_enc_pages;
 	snd_buf->page_base -= first << PAGE_CACHE_SHIFT;
-	/* Give the tail its own page, in case we need extra space in the
-	 * head when wrapping: */
+	/*
+	 * Give the tail its own page, in case we need extra space in the
+	 * head when wrapping:
+	 *
+	 * call_allocate() allocates twice the slack space required
+	 * by the authentication flavor to rq_callsize.
+	 * For GSS, slack is GSS_CRED_SLACK.
+	 */
 	if (snd_buf->page_len || snd_buf->tail[0].iov_len) {
 		tmp = page_address(rqstp->rq_enc_pages[rqstp->rq_enc_pages_num - 1]);
 		memcpy(tmp, snd_buf->tail[0].iov_base, snd_buf->tail[0].iov_len);
 		snd_buf->tail[0].iov_base = tmp;
 	}
 	maj_stat = gss_wrap(ctx->gc_gss_ctx, offset, snd_buf, inpages);
-	/* RPC_SLACK_SPACE should prevent this ever happening: */
+	/* slack space should prevent this ever happening: */
 	BUG_ON(snd_buf->len > snd_buf->buflen);
 	status = -EIO;
 	/* We're assuming that when GSS_S_CONTEXT_EXPIRED, the encryption was
@@ -1573,5 +1620,11 @@ static void __exit exit_rpcsec_gss(void)
 }
 
 MODULE_LICENSE("GPL");
+module_param_named(expired_cred_retry_delay,
+		   gss_expired_cred_retry_delay,
+		   uint, 0644);
+MODULE_PARM_DESC(expired_cred_retry_delay, "Timeout (in seconds) until "
+		"the RPC engine retries an expired credential");
+
 module_init(init_rpcsec_gss)
 module_exit(exit_rpcsec_gss)

net/sunrpc/auth_gss/gss_krb5_crypto.c

@@ -1,7 +1,7 @@
 /*
  *  linux/net/sunrpc/gss_krb5_crypto.c
  *
- *  Copyright (c) 2000 The Regents of the University of Michigan.
+ *  Copyright (c) 2000-2008 The Regents of the University of Michigan.
  *  All rights reserved.
  *
  *  Andy Adamson   <andros@umich.edu>
@@ -41,6 +41,7 @@
 #include <linux/crypto.h>
 #include <linux/highmem.h>
 #include <linux/pagemap.h>
+#include <linux/random.h>
 #include <linux/sunrpc/gss_krb5.h>
 #include <linux/sunrpc/xdr.h>
 
@@ -58,13 +59,13 @@ krb5_encrypt(
 {
 	u32 ret = -EINVAL;
 	struct scatterlist sg[1];
-	u8 local_iv[16] = {0};
+	u8 local_iv[GSS_KRB5_MAX_BLOCKSIZE] = {0};
 	struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv };
 
 	if (length % crypto_blkcipher_blocksize(tfm) != 0)
 		goto out;
 
-	if (crypto_blkcipher_ivsize(tfm) > 16) {
+	if (crypto_blkcipher_ivsize(tfm) > GSS_KRB5_MAX_BLOCKSIZE) {
 		dprintk("RPC:       gss_k5encrypt: tfm iv size too large %d\n",
 			crypto_blkcipher_ivsize(tfm));
 		goto out;
@@ -92,13 +93,13 @@ krb5_decrypt(
 {
 	u32 ret = -EINVAL;
 	struct scatterlist sg[1];
-	u8 local_iv[16] = {0};
+	u8 local_iv[GSS_KRB5_MAX_BLOCKSIZE] = {0};
 	struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv };
 
 	if (length % crypto_blkcipher_blocksize(tfm) != 0)
 		goto out;
 
-	if (crypto_blkcipher_ivsize(tfm) > 16) {
+	if (crypto_blkcipher_ivsize(tfm) > GSS_KRB5_MAX_BLOCKSIZE) {
 		dprintk("RPC:       gss_k5decrypt: tfm iv size too large %d\n",
 			crypto_blkcipher_ivsize(tfm));
 		goto out;
@@ -123,21 +124,155 @@ checksummer(struct scatterlist *sg, void *data)
 	return crypto_hash_update(desc, sg, sg->length);
 }
 
-/* checksum the plaintext data and hdrlen bytes of the token header */
-s32
-make_checksum(char *cksumname, char *header, int hdrlen, struct xdr_buf *body,
-		   int body_offset, struct xdr_netobj *cksum)
+static int
+arcfour_hmac_md5_usage_to_salt(unsigned int usage, u8 salt[4])
+{
+	unsigned int ms_usage;
+
+	switch (usage) {
+	case KG_USAGE_SIGN:
+		ms_usage = 15;
+		break;
+	case KG_USAGE_SEAL:
+		ms_usage = 13;
+		break;
+	default:
+		return EINVAL;;
+	}
+	salt[0] = (ms_usage >> 0) & 0xff;
+	salt[1] = (ms_usage >> 8) & 0xff;
+	salt[2] = (ms_usage >> 16) & 0xff;
+	salt[3] = (ms_usage >> 24) & 0xff;
+
+	return 0;
+}
+
+static u32
+make_checksum_hmac_md5(struct krb5_ctx *kctx, char *header, int hdrlen,
+		       struct xdr_buf *body, int body_offset, u8 *cksumkey,
+		       unsigned int usage, struct xdr_netobj *cksumout)
 {
-	struct hash_desc                desc; /* XXX add to ctx? */
+	struct hash_desc                desc;
 	struct scatterlist              sg[1];
 	int err;
+	u8 checksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	u8 rc4salt[4];
+	struct crypto_hash *md5;
+	struct crypto_hash *hmac_md5;
+
+	if (cksumkey == NULL)
+		return GSS_S_FAILURE;
+
+	if (cksumout->len < kctx->gk5e->cksumlength) {
+		dprintk("%s: checksum buffer length, %u, too small for %s\n",
+			__func__, cksumout->len, kctx->gk5e->name);
+		return GSS_S_FAILURE;
+	}
+
+	if (arcfour_hmac_md5_usage_to_salt(usage, rc4salt)) {
+		dprintk("%s: invalid usage value %u\n", __func__, usage);
+		return GSS_S_FAILURE;
+	}
+
+	md5 = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(md5))
+		return GSS_S_FAILURE;
+
+	hmac_md5 = crypto_alloc_hash(kctx->gk5e->cksum_name, 0,
+				     CRYPTO_ALG_ASYNC);
+	if (IS_ERR(hmac_md5)) {
+		crypto_free_hash(md5);
+		return GSS_S_FAILURE;
+	}
+
+	desc.tfm = md5;
+	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out;
+	sg_init_one(sg, rc4salt, 4);
+	err = crypto_hash_update(&desc, sg, 4);
+	if (err)
+		goto out;
+
+	sg_init_one(sg, header, hdrlen);
+	err = crypto_hash_update(&desc, sg, hdrlen);
+	if (err)
+		goto out;
+	err = xdr_process_buf(body, body_offset, body->len - body_offset,
+			      checksummer, &desc);
+	if (err)
+		goto out;
+	err = crypto_hash_final(&desc, checksumdata);
+	if (err)
+		goto out;
+
+	desc.tfm = hmac_md5;
+	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out;
+	err = crypto_hash_setkey(hmac_md5, cksumkey, kctx->gk5e->keylength);
+	if (err)
+		goto out;
+
+	sg_init_one(sg, checksumdata, crypto_hash_digestsize(md5));
+	err = crypto_hash_digest(&desc, sg, crypto_hash_digestsize(md5),
+				 checksumdata);
+	if (err)
+		goto out;
+
+	memcpy(cksumout->data, checksumdata, kctx->gk5e->cksumlength);
+	cksumout->len = kctx->gk5e->cksumlength;
+out:
+	crypto_free_hash(md5);
+	crypto_free_hash(hmac_md5);
+	return err ? GSS_S_FAILURE : 0;
+}
+
+/*
+ * checksum the plaintext data and hdrlen bytes of the token header
+ * The checksum is performed over the first 8 bytes of the
+ * gss token header and then over the data body
+ */
+u32
+make_checksum(struct krb5_ctx *kctx, char *header, int hdrlen,
+	      struct xdr_buf *body, int body_offset, u8 *cksumkey,
+	      unsigned int usage, struct xdr_netobj *cksumout)
+{
+	struct hash_desc                desc;
+	struct scatterlist              sg[1];
+	int err;
+	u8 checksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	unsigned int checksumlen;
+
+	if (kctx->gk5e->ctype == CKSUMTYPE_HMAC_MD5_ARCFOUR)
+		return make_checksum_hmac_md5(kctx, header, hdrlen,
+					      body, body_offset,
+					      cksumkey, usage, cksumout);
+
+	if (cksumout->len < kctx->gk5e->cksumlength) {
+		dprintk("%s: checksum buffer length, %u, too small for %s\n",
+			__func__, cksumout->len, kctx->gk5e->name);
+		return GSS_S_FAILURE;
+	}
 
-	desc.tfm = crypto_alloc_hash(cksumname, 0, CRYPTO_ALG_ASYNC);
+	desc.tfm = crypto_alloc_hash(kctx->gk5e->cksum_name, 0, CRYPTO_ALG_ASYNC);
 	if (IS_ERR(desc.tfm))
 		return GSS_S_FAILURE;
-	cksum->len = crypto_hash_digestsize(desc.tfm);
 	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
 
+	checksumlen = crypto_hash_digestsize(desc.tfm);
+
+	if (cksumkey != NULL) {
+		err = crypto_hash_setkey(desc.tfm, cksumkey,
+					 kctx->gk5e->keylength);
+		if (err)
+			goto out;
+	}
+
 	err = crypto_hash_init(&desc);
 	if (err)
 		goto out;
@@ -149,15 +284,109 @@ make_checksum(char *cksumname, char *header, int hdrlen, struct xdr_buf *body,
 			      checksummer, &desc);
 	if (err)
 		goto out;
-	err = crypto_hash_final(&desc, cksum->data);
+	err = crypto_hash_final(&desc, checksumdata);
+	if (err)
+		goto out;
 
+	switch (kctx->gk5e->ctype) {
+	case CKSUMTYPE_RSA_MD5:
+		err = kctx->gk5e->encrypt(kctx->seq, NULL, checksumdata,
+					  checksumdata, checksumlen);
+		if (err)
+			goto out;
+		memcpy(cksumout->data,
+		       checksumdata + checksumlen - kctx->gk5e->cksumlength,
+		       kctx->gk5e->cksumlength);
+		break;
+	case CKSUMTYPE_HMAC_SHA1_DES3:
+		memcpy(cksumout->data, checksumdata, kctx->gk5e->cksumlength);
+		break;
+	default:
+		BUG();
+		break;
+	}
+	cksumout->len = kctx->gk5e->cksumlength;
+out:
+	crypto_free_hash(desc.tfm);
+	return err ? GSS_S_FAILURE : 0;
+}
+
+/*
+ * checksum the plaintext data and hdrlen bytes of the token header
+ * Per rfc4121, sec. 4.2.4, the checksum is performed over the data
+ * body then over the first 16 octets of the MIC token
+ * Inclusion of the header data in the calculation of the
+ * checksum is optional.
+ */
+u32
+make_checksum_v2(struct krb5_ctx *kctx, char *header, int hdrlen,
+		 struct xdr_buf *body, int body_offset, u8 *cksumkey,
+		 unsigned int usage, struct xdr_netobj *cksumout)
+{
+	struct hash_desc desc;
+	struct scatterlist sg[1];
+	int err;
+	u8 checksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	unsigned int checksumlen;
+
+	if (kctx->gk5e->keyed_cksum == 0) {
+		dprintk("%s: expected keyed hash for %s\n",
+			__func__, kctx->gk5e->name);
+		return GSS_S_FAILURE;
+	}
+	if (cksumkey == NULL) {
+		dprintk("%s: no key supplied for %s\n",
+			__func__, kctx->gk5e->name);
+		return GSS_S_FAILURE;
+	}
+
+	desc.tfm = crypto_alloc_hash(kctx->gk5e->cksum_name, 0,
+							CRYPTO_ALG_ASYNC);
+	if (IS_ERR(desc.tfm))
+		return GSS_S_FAILURE;
+	checksumlen = crypto_hash_digestsize(desc.tfm);
+	desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	err = crypto_hash_setkey(desc.tfm, cksumkey, kctx->gk5e->keylength);
+	if (err)
+		goto out;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out;
+	err = xdr_process_buf(body, body_offset, body->len - body_offset,
+			      checksummer, &desc);
+	if (err)
+		goto out;
+	if (header != NULL) {
+		sg_init_one(sg, header, hdrlen);
+		err = crypto_hash_update(&desc, sg, hdrlen);
+		if (err)
+			goto out;
+	}
+	err = crypto_hash_final(&desc, checksumdata);
+	if (err)
+		goto out;
+
+	cksumout->len = kctx->gk5e->cksumlength;
+
+	switch (kctx->gk5e->ctype) {
+	case CKSUMTYPE_HMAC_SHA1_96_AES128:
+	case CKSUMTYPE_HMAC_SHA1_96_AES256:
+		/* note that this truncates the hash */
+		memcpy(cksumout->data, checksumdata, kctx->gk5e->cksumlength);
+		break;
+	default:
+		BUG();
+		break;
+	}
 out:
 	crypto_free_hash(desc.tfm);
 	return err ? GSS_S_FAILURE : 0;
 }
 
 struct encryptor_desc {
-	u8 iv[8]; /* XXX hard-coded blocksize */
+	u8 iv[GSS_KRB5_MAX_BLOCKSIZE];
 	struct blkcipher_desc desc;
 	int pos;
 	struct xdr_buf *outbuf;
@@ -198,7 +427,7 @@ encryptor(struct scatterlist *sg, void *data)
 	desc->fraglen += sg->length;
 	desc->pos += sg->length;
 
-	fraglen = thislen & 7; /* XXX hardcoded blocksize */
+	fraglen = thislen & (crypto_blkcipher_blocksize(desc->desc.tfm) - 1);
 	thislen -= fraglen;
 
 	if (thislen == 0)
@@ -256,7 +485,7 @@ gss_encrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf,
 }
 
 struct decryptor_desc {
-	u8 iv[8]; /* XXX hard-coded blocksize */
+	u8 iv[GSS_KRB5_MAX_BLOCKSIZE];
 	struct blkcipher_desc desc;
 	struct scatterlist frags[4];
 	int fragno;
@@ -278,7 +507,7 @@ decryptor(struct scatterlist *sg, void *data)
 	desc->fragno++;
 	desc->fraglen += sg->length;
 
-	fraglen = thislen & 7; /* XXX hardcoded blocksize */
+	fraglen = thislen & (crypto_blkcipher_blocksize(desc->desc.tfm) - 1);
 	thislen -= fraglen;
 
 	if (thislen == 0)
@@ -325,3 +554,437 @@ gss_decrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf,
 
 	return xdr_process_buf(buf, offset, buf->len - offset, decryptor, &desc);
 }
+
+/*
+ * This function makes the assumption that it was ultimately called
+ * from gss_wrap().
+ *
+ * The client auth_gss code moves any existing tail data into a
+ * separate page before calling gss_wrap.
+ * The server svcauth_gss code ensures that both the head and the
+ * tail have slack space of RPC_MAX_AUTH_SIZE before calling gss_wrap.
+ *
+ * Even with that guarantee, this function may be called more than
+ * once in the processing of gss_wrap().  The best we can do is
+ * verify at compile-time (see GSS_KRB5_SLACK_CHECK) that the
+ * largest expected shift will fit within RPC_MAX_AUTH_SIZE.
+ * At run-time we can verify that a single invocation of this
+ * function doesn't attempt to use more the RPC_MAX_AUTH_SIZE.
+ */
+
+int
+xdr_extend_head(struct xdr_buf *buf, unsigned int base, unsigned int shiftlen)
+{
+	u8 *p;
+
+	if (shiftlen == 0)
+		return 0;
+
+	BUILD_BUG_ON(GSS_KRB5_MAX_SLACK_NEEDED > RPC_MAX_AUTH_SIZE);
+	BUG_ON(shiftlen > RPC_MAX_AUTH_SIZE);
+
+	p = buf->head[0].iov_base + base;
+
+	memmove(p + shiftlen, p, buf->head[0].iov_len - base);
+
+	buf->head[0].iov_len += shiftlen;
+	buf->len += shiftlen;
+
+	return 0;
+}
+
+static u32
+gss_krb5_cts_crypt(struct crypto_blkcipher *cipher, struct xdr_buf *buf,
+		   u32 offset, u8 *iv, struct page **pages, int encrypt)
+{
+	u32 ret;
+	struct scatterlist sg[1];
+	struct blkcipher_desc desc = { .tfm = cipher, .info = iv };
+	u8 data[crypto_blkcipher_blocksize(cipher) * 2];
+	struct page **save_pages;
+	u32 len = buf->len - offset;
+
+	BUG_ON(len > crypto_blkcipher_blocksize(cipher) * 2);
+
+	/*
+	 * For encryption, we want to read from the cleartext
+	 * page cache pages, and write the encrypted data to
+	 * the supplied xdr_buf pages.
+	 */
+	save_pages = buf->pages;
+	if (encrypt)
+		buf->pages = pages;
+
+	ret = read_bytes_from_xdr_buf(buf, offset, data, len);
+	buf->pages = save_pages;
+	if (ret)
+		goto out;
+
+	sg_init_one(sg, data, len);
+
+	if (encrypt)
+		ret = crypto_blkcipher_encrypt_iv(&desc, sg, sg, len);
+	else
+		ret = crypto_blkcipher_decrypt_iv(&desc, sg, sg, len);
+
+	if (ret)
+		goto out;
+
+	ret = write_bytes_to_xdr_buf(buf, offset, data, len);
+
+out:
+	return ret;
+}
+
+u32
+gss_krb5_aes_encrypt(struct krb5_ctx *kctx, u32 offset,
+		     struct xdr_buf *buf, int ec, struct page **pages)
+{
+	u32 err;
+	struct xdr_netobj hmac;
+	u8 *cksumkey;
+	u8 *ecptr;
+	struct crypto_blkcipher *cipher, *aux_cipher;
+	int blocksize;
+	struct page **save_pages;
+	int nblocks, nbytes;
+	struct encryptor_desc desc;
+	u32 cbcbytes;
+	unsigned int usage;
+
+	if (kctx->initiate) {
+		cipher = kctx->initiator_enc;
+		aux_cipher = kctx->initiator_enc_aux;
+		cksumkey = kctx->initiator_integ;
+		usage = KG_USAGE_INITIATOR_SEAL;
+	} else {
+		cipher = kctx->acceptor_enc;
+		aux_cipher = kctx->acceptor_enc_aux;
+		cksumkey = kctx->acceptor_integ;
+		usage = KG_USAGE_ACCEPTOR_SEAL;
+	}
+	blocksize = crypto_blkcipher_blocksize(cipher);
+
+	/* hide the gss token header and insert the confounder */
+	offset += GSS_KRB5_TOK_HDR_LEN;
+	if (xdr_extend_head(buf, offset, kctx->gk5e->conflen))
+		return GSS_S_FAILURE;
+	gss_krb5_make_confounder(buf->head[0].iov_base + offset, kctx->gk5e->conflen);
+	offset -= GSS_KRB5_TOK_HDR_LEN;
+
+	if (buf->tail[0].iov_base != NULL) {
+		ecptr = buf->tail[0].iov_base + buf->tail[0].iov_len;
+	} else {
+		buf->tail[0].iov_base = buf->head[0].iov_base
+							+ buf->head[0].iov_len;
+		buf->tail[0].iov_len = 0;
+		ecptr = buf->tail[0].iov_base;
+	}
+
+	memset(ecptr, 'X', ec);
+	buf->tail[0].iov_len += ec;
+	buf->len += ec;
+
+	/* copy plaintext gss token header after filler (if any) */
+	memcpy(ecptr + ec, buf->head[0].iov_base + offset,
+						GSS_KRB5_TOK_HDR_LEN);
+	buf->tail[0].iov_len += GSS_KRB5_TOK_HDR_LEN;
+	buf->len += GSS_KRB5_TOK_HDR_LEN;
+
+	/* Do the HMAC */
+	hmac.len = GSS_KRB5_MAX_CKSUM_LEN;
+	hmac.data = buf->tail[0].iov_base + buf->tail[0].iov_len;
+
+	/*
+	 * When we are called, pages points to the real page cache
+	 * data -- which we can't go and encrypt!  buf->pages points
+	 * to scratch pages which we are going to send off to the
+	 * client/server.  Swap in the plaintext pages to calculate
+	 * the hmac.
+	 */
+	save_pages = buf->pages;
+	buf->pages = pages;
+
+	err = make_checksum_v2(kctx, NULL, 0, buf,
+			       offset + GSS_KRB5_TOK_HDR_LEN,
+			       cksumkey, usage, &hmac);
+	buf->pages = save_pages;
+	if (err)
+		return GSS_S_FAILURE;
+
+	nbytes = buf->len - offset - GSS_KRB5_TOK_HDR_LEN;
+	nblocks = (nbytes + blocksize - 1) / blocksize;
+	cbcbytes = 0;
+	if (nblocks > 2)
+		cbcbytes = (nblocks - 2) * blocksize;
+
+	memset(desc.iv, 0, sizeof(desc.iv));
+
+	if (cbcbytes) {
+		desc.pos = offset + GSS_KRB5_TOK_HDR_LEN;
+		desc.fragno = 0;
+		desc.fraglen = 0;
+		desc.pages = pages;
+		desc.outbuf = buf;
+		desc.desc.info = desc.iv;
+		desc.desc.flags = 0;
+		desc.desc.tfm = aux_cipher;
+
+		sg_init_table(desc.infrags, 4);
+		sg_init_table(desc.outfrags, 4);
+
+		err = xdr_process_buf(buf, offset + GSS_KRB5_TOK_HDR_LEN,
+				      cbcbytes, encryptor, &desc);
+		if (err)
+			goto out_err;
+	}
+
+	/* Make sure IV carries forward from any CBC results. */
+	err = gss_krb5_cts_crypt(cipher, buf,
+				 offset + GSS_KRB5_TOK_HDR_LEN + cbcbytes,
+				 desc.iv, pages, 1);
+	if (err) {
+		err = GSS_S_FAILURE;
+		goto out_err;
+	}
+
+	/* Now update buf to account for HMAC */
+	buf->tail[0].iov_len += kctx->gk5e->cksumlength;
+	buf->len += kctx->gk5e->cksumlength;
+
+out_err:
+	if (err)
+		err = GSS_S_FAILURE;
+	return err;
+}
+
+u32
+gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, struct xdr_buf *buf,
+		     u32 *headskip, u32 *tailskip)
+{
+	struct xdr_buf subbuf;
+	u32 ret = 0;
+	u8 *cksum_key;
+	struct crypto_blkcipher *cipher, *aux_cipher;
+	struct xdr_netobj our_hmac_obj;
+	u8 our_hmac[GSS_KRB5_MAX_CKSUM_LEN];
+	u8 pkt_hmac[GSS_KRB5_MAX_CKSUM_LEN];
+	int nblocks, blocksize, cbcbytes;
+	struct decryptor_desc desc;
+	unsigned int usage;
+
+	if (kctx->initiate) {
+		cipher = kctx->acceptor_enc;
+		aux_cipher = kctx->acceptor_enc_aux;
+		cksum_key = kctx->acceptor_integ;
+		usage = KG_USAGE_ACCEPTOR_SEAL;
+	} else {
+		cipher = kctx->initiator_enc;
+		aux_cipher = kctx->initiator_enc_aux;
+		cksum_key = kctx->initiator_integ;
+		usage = KG_USAGE_INITIATOR_SEAL;
+	}
+	blocksize = crypto_blkcipher_blocksize(cipher);
+
+
+	/* create a segment skipping the header and leaving out the checksum */
+	xdr_buf_subsegment(buf, &subbuf, offset + GSS_KRB5_TOK_HDR_LEN,
+				    (buf->len - offset - GSS_KRB5_TOK_HDR_LEN -
+				     kctx->gk5e->cksumlength));
+
+	nblocks = (subbuf.len + blocksize - 1) / blocksize;
+
+	cbcbytes = 0;
+	if (nblocks > 2)
+		cbcbytes = (nblocks - 2) * blocksize;
+
+	memset(desc.iv, 0, sizeof(desc.iv));
+
+	if (cbcbytes) {
+		desc.fragno = 0;
+		desc.fraglen = 0;
+		desc.desc.info = desc.iv;
+		desc.desc.flags = 0;
+		desc.desc.tfm = aux_cipher;
+
+		sg_init_table(desc.frags, 4);
+
+		ret = xdr_process_buf(&subbuf, 0, cbcbytes, decryptor, &desc);
+		if (ret)
+			goto out_err;
+	}
+
+	/* Make sure IV carries forward from any CBC results. */
+	ret = gss_krb5_cts_crypt(cipher, &subbuf, cbcbytes, desc.iv, NULL, 0);
+	if (ret)
+		goto out_err;
+
+
+	/* Calculate our hmac over the plaintext data */
+	our_hmac_obj.len = sizeof(our_hmac);
+	our_hmac_obj.data = our_hmac;
+
+	ret = make_checksum_v2(kctx, NULL, 0, &subbuf, 0,
+			       cksum_key, usage, &our_hmac_obj);
+	if (ret)
+		goto out_err;
+
+	/* Get the packet's hmac value */
+	ret = read_bytes_from_xdr_buf(buf, buf->len - kctx->gk5e->cksumlength,
+				      pkt_hmac, kctx->gk5e->cksumlength);
+	if (ret)
+		goto out_err;
+
+	if (memcmp(pkt_hmac, our_hmac, kctx->gk5e->cksumlength) != 0) {
+		ret = GSS_S_BAD_SIG;
+		goto out_err;
+	}
+	*headskip = kctx->gk5e->conflen;
+	*tailskip = kctx->gk5e->cksumlength;
+out_err:
+	if (ret && ret != GSS_S_BAD_SIG)
+		ret = GSS_S_FAILURE;
+	return ret;
+}
+
+/*
+ * Compute Kseq given the initial session key and the checksum.
+ * Set the key of the given cipher.
+ */
+int
+krb5_rc4_setup_seq_key(struct krb5_ctx *kctx, struct crypto_blkcipher *cipher,
+		       unsigned char *cksum)
+{
+	struct crypto_hash *hmac;
+	struct hash_desc desc;
+	struct scatterlist sg[1];
+	u8 Kseq[GSS_KRB5_MAX_KEYLEN];
+	u32 zeroconstant = 0;
+	int err;
+
+	dprintk("%s: entered\n", __func__);
+
+	hmac = crypto_alloc_hash(kctx->gk5e->cksum_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(hmac)) {
+		dprintk("%s: error %ld, allocating hash '%s'\n",
+			__func__, PTR_ERR(hmac), kctx->gk5e->cksum_name);
+		return PTR_ERR(hmac);
+	}
+
+	desc.tfm = hmac;
+	desc.flags = 0;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out_err;
+
+	/* Compute intermediate Kseq from session key */
+	err = crypto_hash_setkey(hmac, kctx->Ksess, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	sg_init_table(sg, 1);
+	sg_set_buf(sg, &zeroconstant, 4);
+
+	err = crypto_hash_digest(&desc, sg, 4, Kseq);
+	if (err)
+		goto out_err;
+
+	/* Compute final Kseq from the checksum and intermediate Kseq */
+	err = crypto_hash_setkey(hmac, Kseq, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	sg_set_buf(sg, cksum, 8);
+
+	err = crypto_hash_digest(&desc, sg, 8, Kseq);
+	if (err)
+		goto out_err;
+
+	err = crypto_blkcipher_setkey(cipher, Kseq, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	err = 0;
+
+out_err:
+	crypto_free_hash(hmac);
+	dprintk("%s: returning %d\n", __func__, err);
+	return err;
+}
+
+/*
+ * Compute Kcrypt given the initial session key and the plaintext seqnum.
+ * Set the key of cipher kctx->enc.
+ */
+int
+krb5_rc4_setup_enc_key(struct krb5_ctx *kctx, struct crypto_blkcipher *cipher,
+		       s32 seqnum)
+{
+	struct crypto_hash *hmac;
+	struct hash_desc desc;
+	struct scatterlist sg[1];
+	u8 Kcrypt[GSS_KRB5_MAX_KEYLEN];
+	u8 zeroconstant[4] = {0};
+	u8 seqnumarray[4];
+	int err, i;
+
+	dprintk("%s: entered, seqnum %u\n", __func__, seqnum);
+
+	hmac = crypto_alloc_hash(kctx->gk5e->cksum_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(hmac)) {
+		dprintk("%s: error %ld, allocating hash '%s'\n",
+			__func__, PTR_ERR(hmac), kctx->gk5e->cksum_name);
+		return PTR_ERR(hmac);
+	}
+
+	desc.tfm = hmac;
+	desc.flags = 0;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out_err;
+
+	/* Compute intermediate Kcrypt from session key */
+	for (i = 0; i < kctx->gk5e->keylength; i++)
+		Kcrypt[i] = kctx->Ksess[i] ^ 0xf0;
+
+	err = crypto_hash_setkey(hmac, Kcrypt, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	sg_init_table(sg, 1);
+	sg_set_buf(sg, zeroconstant, 4);
+
+	err = crypto_hash_digest(&desc, sg, 4, Kcrypt);
+	if (err)
+		goto out_err;
+
+	/* Compute final Kcrypt from the seqnum and intermediate Kcrypt */
+	err = crypto_hash_setkey(hmac, Kcrypt, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	seqnumarray[0] = (unsigned char) ((seqnum >> 24) & 0xff);
+	seqnumarray[1] = (unsigned char) ((seqnum >> 16) & 0xff);
+	seqnumarray[2] = (unsigned char) ((seqnum >> 8) & 0xff);
+	seqnumarray[3] = (unsigned char) ((seqnum >> 0) & 0xff);
+
+	sg_set_buf(sg, seqnumarray, 4);
+
+	err = crypto_hash_digest(&desc, sg, 4, Kcrypt);
+	if (err)
+		goto out_err;
+
+	err = crypto_blkcipher_setkey(cipher, Kcrypt, kctx->gk5e->keylength);
+	if (err)
+		goto out_err;
+
+	err = 0;
+
+out_err:
+	crypto_free_hash(hmac);
+	dprintk("%s: returning %d\n", __func__, err);
+	return err;
+}
+

net/sunrpc/auth_gss/gss_krb5_keys.c

@@ -0,0 +1,336 @@
+/*
+ * COPYRIGHT (c) 2008
+ * The Regents of the University of Michigan
+ * ALL RIGHTS RESERVED
+ *
+ * Permission is granted to use, copy, create derivative works
+ * and redistribute this software and such derivative works
+ * for any purpose, so long as the name of The University of
+ * Michigan is not used in any advertising or publicity
+ * pertaining to the use of distribution of this software
+ * without specific, written prior authorization.  If the
+ * above copyright notice or any other identification of the
+ * University of Michigan is included in any copy of any
+ * portion of this software, then the disclaimer below must
+ * also be included.
+ *
+ * THIS SOFTWARE IS PROVIDED AS IS, WITHOUT REPRESENTATION
+ * FROM THE UNIVERSITY OF MICHIGAN AS TO ITS FITNESS FOR ANY
+ * PURPOSE, AND WITHOUT WARRANTY BY THE UNIVERSITY OF
+ * MICHIGAN OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING
+ * WITHOUT LIMITATION THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE
+ * REGENTS OF THE UNIVERSITY OF MICHIGAN SHALL NOT BE LIABLE
+ * FOR ANY DAMAGES, INCLUDING SPECIAL, INDIRECT, INCIDENTAL, OR
+ * CONSEQUENTIAL DAMAGES, WITH RESPECT TO ANY CLAIM ARISING
+ * OUT OF OR IN CONNECTION WITH THE USE OF THE SOFTWARE, EVEN
+ * IF IT HAS BEEN OR IS HEREAFTER ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGES.
+ */
+
+/*
+ * Copyright (C) 1998 by the FundsXpress, INC.
+ *
+ * All rights reserved.
+ *
+ * Export of this software from the United States of America may require
+ * a specific license from the United States Government.  It is the
+ * responsibility of any person or organization contemplating export to
+ * obtain such a license before exporting.
+ *
+ * WITHIN THAT CONSTRAINT, permission to use, copy, modify, and
+ * distribute this software and its documentation for any purpose and
+ * without fee is hereby granted, provided that the above copyright
+ * notice appear in all copies and that both that copyright notice and
+ * this permission notice appear in supporting documentation, and that
+ * the name of FundsXpress. not be used in advertising or publicity pertaining
+ * to distribution of the software without specific, written prior
+ * permission.  FundsXpress makes no representations about the suitability of
+ * this software for any purpose.  It is provided "as is" without express
+ * or implied warranty.
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
+ * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
+ * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
+ */
+
+#include <linux/err.h>
+#include <linux/types.h>
+#include <linux/crypto.h>
+#include <linux/sunrpc/gss_krb5.h>
+#include <linux/sunrpc/xdr.h>
+
+#ifdef RPC_DEBUG
+# define RPCDBG_FACILITY        RPCDBG_AUTH
+#endif
+
+/*
+ * This is the n-fold function as described in rfc3961, sec 5.1
+ * Taken from MIT Kerberos and modified.
+ */
+
+static void krb5_nfold(u32 inbits, const u8 *in,
+		       u32 outbits, u8 *out)
+{
+	int a, b, c, lcm;
+	int byte, i, msbit;
+
+	/* the code below is more readable if I make these bytes
+	   instead of bits */
+
+	inbits >>= 3;
+	outbits >>= 3;
+
+	/* first compute lcm(n,k) */
+
+	a = outbits;
+	b = inbits;
+
+	while (b != 0) {
+		c = b;
+		b = a%b;
+		a = c;
+	}
+
+	lcm = outbits*inbits/a;
+
+	/* now do the real work */
+
+	memset(out, 0, outbits);
+	byte = 0;
+
+	/* this will end up cycling through k lcm(k,n)/k times, which
+	   is correct */
+	for (i = lcm-1; i >= 0; i--) {
+		/* compute the msbit in k which gets added into this byte */
+		msbit = (
+			/* first, start with the msbit in the first,
+			 * unrotated byte */
+			 ((inbits << 3) - 1)
+			 /* then, for each byte, shift to the right
+			  * for each repetition */
+			 + (((inbits << 3) + 13) * (i/inbits))
+			 /* last, pick out the correct byte within
+			  * that shifted repetition */
+			 + ((inbits - (i % inbits)) << 3)
+			 ) % (inbits << 3);
+
+		/* pull out the byte value itself */
+		byte += (((in[((inbits - 1) - (msbit >> 3)) % inbits] << 8)|
+				  (in[((inbits) - (msbit >> 3)) % inbits]))
+				 >> ((msbit & 7) + 1)) & 0xff;
+
+		/* do the addition */
+		byte += out[i % outbits];
+		out[i % outbits] = byte & 0xff;
+
+		/* keep around the carry bit, if any */
+		byte >>= 8;
+
+	}
+
+	/* if there's a carry bit left over, add it back in */
+	if (byte) {
+		for (i = outbits - 1; i >= 0; i--) {
+			/* do the addition */
+			byte += out[i];
+			out[i] = byte & 0xff;
+
+			/* keep around the carry bit, if any */
+			byte >>= 8;
+		}
+	}
+}
+
+/*
+ * This is the DK (derive_key) function as described in rfc3961, sec 5.1
+ * Taken from MIT Kerberos and modified.
+ */
+
+u32 krb5_derive_key(const struct gss_krb5_enctype *gk5e,
+		    const struct xdr_netobj *inkey,
+		    struct xdr_netobj *outkey,
+		    const struct xdr_netobj *in_constant,
+		    gfp_t gfp_mask)
+{
+	size_t blocksize, keybytes, keylength, n;
+	unsigned char *inblockdata, *outblockdata, *rawkey;
+	struct xdr_netobj inblock, outblock;
+	struct crypto_blkcipher *cipher;
+	u32 ret = EINVAL;
+
+	blocksize = gk5e->blocksize;
+	keybytes = gk5e->keybytes;
+	keylength = gk5e->keylength;
+
+	if ((inkey->len != keylength) || (outkey->len != keylength))
+		goto err_return;
+
+	cipher = crypto_alloc_blkcipher(gk5e->encrypt_name, 0,
+					CRYPTO_ALG_ASYNC);
+	if (IS_ERR(cipher))
+		goto err_return;
+	if (crypto_blkcipher_setkey(cipher, inkey->data, inkey->len))
+		goto err_return;
+
+	/* allocate and set up buffers */
+
+	ret = ENOMEM;
+	inblockdata = kmalloc(blocksize, gfp_mask);
+	if (inblockdata == NULL)
+		goto err_free_cipher;
+
+	outblockdata = kmalloc(blocksize, gfp_mask);
+	if (outblockdata == NULL)
+		goto err_free_in;
+
+	rawkey = kmalloc(keybytes, gfp_mask);
+	if (rawkey == NULL)
+		goto err_free_out;
+
+	inblock.data = (char *) inblockdata;
+	inblock.len = blocksize;
+
+	outblock.data = (char *) outblockdata;
+	outblock.len = blocksize;
+
+	/* initialize the input block */
+
+	if (in_constant->len == inblock.len) {
+		memcpy(inblock.data, in_constant->data, inblock.len);
+	} else {
+		krb5_nfold(in_constant->len * 8, in_constant->data,
+			   inblock.len * 8, inblock.data);
+	}
+
+	/* loop encrypting the blocks until enough key bytes are generated */
+
+	n = 0;
+	while (n < keybytes) {
+		(*(gk5e->encrypt))(cipher, NULL, inblock.data,
+				   outblock.data, inblock.len);
+
+		if ((keybytes - n) <= outblock.len) {
+			memcpy(rawkey + n, outblock.data, (keybytes - n));
+			break;
+		}
+
+		memcpy(rawkey + n, outblock.data, outblock.len);
+		memcpy(inblock.data, outblock.data, outblock.len);
+		n += outblock.len;
+	}
+
+	/* postprocess the key */
+
+	inblock.data = (char *) rawkey;
+	inblock.len = keybytes;
+
+	BUG_ON(gk5e->mk_key == NULL);
+	ret = (*(gk5e->mk_key))(gk5e, &inblock, outkey);
+	if (ret) {
+		dprintk("%s: got %d from mk_key function for '%s'\n",
+			__func__, ret, gk5e->encrypt_name);
+		goto err_free_raw;
+	}
+
+	/* clean memory, free resources and exit */
+
+	ret = 0;
+
+err_free_raw:
+	memset(rawkey, 0, keybytes);
+	kfree(rawkey);
+err_free_out:
+	memset(outblockdata, 0, blocksize);
+	kfree(outblockdata);
+err_free_in:
+	memset(inblockdata, 0, blocksize);
+	kfree(inblockdata);
+err_free_cipher:
+	crypto_free_blkcipher(cipher);
+err_return:
+	return ret;
+}
+
+#define smask(step) ((1<<step)-1)
+#define pstep(x, step) (((x)&smask(step))^(((x)>>step)&smask(step)))
+#define parity_char(x) pstep(pstep(pstep((x), 4), 2), 1)
+
+static void mit_des_fixup_key_parity(u8 key[8])
+{
+	int i;
+	for (i = 0; i < 8; i++) {
+		key[i] &= 0xfe;
+		key[i] |= 1^parity_char(key[i]);
+	}
+}
+
+/*
+ * This is the des3 key derivation postprocess function
+ */
+u32 gss_krb5_des3_make_key(const struct gss_krb5_enctype *gk5e,
+			   struct xdr_netobj *randombits,
+			   struct xdr_netobj *key)
+{
+	int i;
+	u32 ret = EINVAL;
+
+	if (key->len != 24) {
+		dprintk("%s: key->len is %d\n", __func__, key->len);
+		goto err_out;
+	}
+	if (randombits->len != 21) {
+		dprintk("%s: randombits->len is %d\n",
+			__func__, randombits->len);
+		goto err_out;
+	}
+
+	/* take the seven bytes, move them around into the top 7 bits of the
+	   8 key bytes, then compute the parity bits.  Do this three times. */
+
+	for (i = 0; i < 3; i++) {
+		memcpy(key->data + i*8, randombits->data + i*7, 7);
+		key->data[i*8+7] = (((key->data[i*8]&1)<<1) |
+				    ((key->data[i*8+1]&1)<<2) |
+				    ((key->data[i*8+2]&1)<<3) |
+				    ((key->data[i*8+3]&1)<<4) |
+				    ((key->data[i*8+4]&1)<<5) |
+				    ((key->data[i*8+5]&1)<<6) |
+				    ((key->data[i*8+6]&1)<<7));
+
+		mit_des_fixup_key_parity(key->data + i*8);
+	}
+	ret = 0;
+err_out:
+	return ret;
+}
+
+/*
+ * This is the aes key derivation postprocess function
+ */
+u32 gss_krb5_aes_make_key(const struct gss_krb5_enctype *gk5e,
+			  struct xdr_netobj *randombits,
+			  struct xdr_netobj *key)
+{
+	u32 ret = EINVAL;
+
+	if (key->len != 16 && key->len != 32) {
+		dprintk("%s: key->len is %d\n", __func__, key->len);
+		goto err_out;
+	}
+	if (randombits->len != 16 && randombits->len != 32) {
+		dprintk("%s: randombits->len is %d\n",
+			__func__, randombits->len);
+		goto err_out;
+	}
+	if (randombits->len != key->len) {
+		dprintk("%s: randombits->len is %d, key->len is %d\n",
+			__func__, randombits->len, key->len);
+		goto err_out;
+	}
+	memcpy(key->data, randombits->data, key->len);
+	ret = 0;
+err_out:
+	return ret;
+}
+

net/sunrpc/auth_gss/gss_krb5_mech.c

@@ -1,7 +1,7 @@
 /*
  *  linux/net/sunrpc/gss_krb5_mech.c
  *
- *  Copyright (c) 2001 The Regents of the University of Michigan.
+ *  Copyright (c) 2001-2008 The Regents of the University of Michigan.
  *  All rights reserved.
  *
  *  Andy Adamson <andros@umich.edu>
@@ -48,6 +48,143 @@
 # define RPCDBG_FACILITY	RPCDBG_AUTH
 #endif
 
+static struct gss_api_mech gss_kerberos_mech;	/* forward declaration */
+
+static const struct gss_krb5_enctype supported_gss_krb5_enctypes[] = {
+	/*
+	 * DES (All DES enctypes are mapped to the same gss functionality)
+	 */
+	{
+	  .etype = ENCTYPE_DES_CBC_RAW,
+	  .ctype = CKSUMTYPE_RSA_MD5,
+	  .name = "des-cbc-crc",
+	  .encrypt_name = "cbc(des)",
+	  .cksum_name = "md5",
+	  .encrypt = krb5_encrypt,
+	  .decrypt = krb5_decrypt,
+	  .mk_key = NULL,
+	  .signalg = SGN_ALG_DES_MAC_MD5,
+	  .sealalg = SEAL_ALG_DES,
+	  .keybytes = 7,
+	  .keylength = 8,
+	  .blocksize = 8,
+	  .conflen = 8,
+	  .cksumlength = 8,
+	  .keyed_cksum = 0,
+	},
+	/*
+	 * RC4-HMAC
+	 */
+	{
+	  .etype = ENCTYPE_ARCFOUR_HMAC,
+	  .ctype = CKSUMTYPE_HMAC_MD5_ARCFOUR,
+	  .name = "rc4-hmac",
+	  .encrypt_name = "ecb(arc4)",
+	  .cksum_name = "hmac(md5)",
+	  .encrypt = krb5_encrypt,
+	  .decrypt = krb5_decrypt,
+	  .mk_key = NULL,
+	  .signalg = SGN_ALG_HMAC_MD5,
+	  .sealalg = SEAL_ALG_MICROSOFT_RC4,
+	  .keybytes = 16,
+	  .keylength = 16,
+	  .blocksize = 1,
+	  .conflen = 8,
+	  .cksumlength = 8,
+	  .keyed_cksum = 1,
+	},
+	/*
+	 * 3DES
+	 */
+	{
+	  .etype = ENCTYPE_DES3_CBC_RAW,
+	  .ctype = CKSUMTYPE_HMAC_SHA1_DES3,
+	  .name = "des3-hmac-sha1",
+	  .encrypt_name = "cbc(des3_ede)",
+	  .cksum_name = "hmac(sha1)",
+	  .encrypt = krb5_encrypt,
+	  .decrypt = krb5_decrypt,
+	  .mk_key = gss_krb5_des3_make_key,
+	  .signalg = SGN_ALG_HMAC_SHA1_DES3_KD,
+	  .sealalg = SEAL_ALG_DES3KD,
+	  .keybytes = 21,
+	  .keylength = 24,
+	  .blocksize = 8,
+	  .conflen = 8,
+	  .cksumlength = 20,
+	  .keyed_cksum = 1,
+	},
+	/*
+	 * AES128
+	 */
+	{
+	  .etype = ENCTYPE_AES128_CTS_HMAC_SHA1_96,
+	  .ctype = CKSUMTYPE_HMAC_SHA1_96_AES128,
+	  .name = "aes128-cts",
+	  .encrypt_name = "cts(cbc(aes))",
+	  .cksum_name = "hmac(sha1)",
+	  .encrypt = krb5_encrypt,
+	  .decrypt = krb5_decrypt,
+	  .mk_key = gss_krb5_aes_make_key,
+	  .encrypt_v2 = gss_krb5_aes_encrypt,
+	  .decrypt_v2 = gss_krb5_aes_decrypt,
+	  .signalg = -1,
+	  .sealalg = -1,
+	  .keybytes = 16,
+	  .keylength = 16,
+	  .blocksize = 16,
+	  .conflen = 16,
+	  .cksumlength = 12,
+	  .keyed_cksum = 1,
+	},
+	/*
+	 * AES256
+	 */
+	{
+	  .etype = ENCTYPE_AES256_CTS_HMAC_SHA1_96,
+	  .ctype = CKSUMTYPE_HMAC_SHA1_96_AES256,
+	  .name = "aes256-cts",
+	  .encrypt_name = "cts(cbc(aes))",
+	  .cksum_name = "hmac(sha1)",
+	  .encrypt = krb5_encrypt,
+	  .decrypt = krb5_decrypt,
+	  .mk_key = gss_krb5_aes_make_key,
+	  .encrypt_v2 = gss_krb5_aes_encrypt,
+	  .decrypt_v2 = gss_krb5_aes_decrypt,
+	  .signalg = -1,
+	  .sealalg = -1,
+	  .keybytes = 32,
+	  .keylength = 32,
+	  .blocksize = 16,
+	  .conflen = 16,
+	  .cksumlength = 12,
+	  .keyed_cksum = 1,
+	},
+};
+
+static const int num_supported_enctypes =
+	ARRAY_SIZE(supported_gss_krb5_enctypes);
+
+static int
+supported_gss_krb5_enctype(int etype)
+{
+	int i;
+	for (i = 0; i < num_supported_enctypes; i++)
+		if (supported_gss_krb5_enctypes[i].etype == etype)
+			return 1;
+	return 0;
+}
+
+static const struct gss_krb5_enctype *
+get_gss_krb5_enctype(int etype)
+{
+	int i;
+	for (i = 0; i < num_supported_enctypes; i++)
+		if (supported_gss_krb5_enctypes[i].etype == etype)
+			return &supported_gss_krb5_enctypes[i];
+	return NULL;
+}
+
 static const void *
 simple_get_bytes(const void *p, const void *end, void *res, int len)
 {
@@ -78,35 +215,45 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *res)
 }
 
 static inline const void *
-get_key(const void *p, const void *end, struct crypto_blkcipher **res)
+get_key(const void *p, const void *end,
+	struct krb5_ctx *ctx, struct crypto_blkcipher **res)
 {
 	struct xdr_netobj	key;
 	int			alg;
-	char			*alg_name;
 
 	p = simple_get_bytes(p, end, &alg, sizeof(alg));
 	if (IS_ERR(p))
 		goto out_err;
+
+	switch (alg) {
+	case ENCTYPE_DES_CBC_CRC:
+	case ENCTYPE_DES_CBC_MD4:
+	case ENCTYPE_DES_CBC_MD5:
+		/* Map all these key types to ENCTYPE_DES_CBC_RAW */
+		alg = ENCTYPE_DES_CBC_RAW;
+		break;
+	}
+
+	if (!supported_gss_krb5_enctype(alg)) {
+		printk(KERN_WARNING "gss_kerberos_mech: unsupported "
+			"encryption key algorithm %d\n", alg);
+		goto out_err;
+	}
 	p = simple_get_netobj(p, end, &key);
 	if (IS_ERR(p))
 		goto out_err;
 
-	switch (alg) {
-		case ENCTYPE_DES_CBC_RAW:
-			alg_name = "cbc(des)";
-			break;
-		default:
-			printk("gss_kerberos_mech: unsupported algorithm %d\n", alg);
-			goto out_err_free_key;
-	}
-	*res = crypto_alloc_blkcipher(alg_name, 0, CRYPTO_ALG_ASYNC);
+	*res = crypto_alloc_blkcipher(ctx->gk5e->encrypt_name, 0,
+							CRYPTO_ALG_ASYNC);
 	if (IS_ERR(*res)) {
-		printk("gss_kerberos_mech: unable to initialize crypto algorithm %s\n", alg_name);
+		printk(KERN_WARNING "gss_kerberos_mech: unable to initialize "
+			"crypto algorithm %s\n", ctx->gk5e->encrypt_name);
 		*res = NULL;
 		goto out_err_free_key;
 	}
 	if (crypto_blkcipher_setkey(*res, key.data, key.len)) {
-		printk("gss_kerberos_mech: error setting key for crypto algorithm %s\n", alg_name);
+		printk(KERN_WARNING "gss_kerberos_mech: error setting key for "
+			"crypto algorithm %s\n", ctx->gk5e->encrypt_name);
 		goto out_err_free_tfm;
 	}
 
@@ -123,56 +270,55 @@ out_err:
 }
 
 static int
-gss_import_sec_context_kerberos(const void *p,
-				size_t len,
-				struct gss_ctx *ctx_id)
+gss_import_v1_context(const void *p, const void *end, struct krb5_ctx *ctx)
 {
-	const void *end = (const void *)((const char *)p + len);
-	struct	krb5_ctx *ctx;
 	int tmp;
 
-	if (!(ctx = kzalloc(sizeof(*ctx), GFP_NOFS))) {
-		p = ERR_PTR(-ENOMEM);
-		goto out_err;
-	}
-
 	p = simple_get_bytes(p, end, &ctx->initiate, sizeof(ctx->initiate));
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
+		goto out_err;
+
+	/* Old format supports only DES!  Any other enctype uses new format */
+	ctx->enctype = ENCTYPE_DES_CBC_RAW;
+
+	ctx->gk5e = get_gss_krb5_enctype(ctx->enctype);
+	if (ctx->gk5e == NULL)
+		goto out_err;
+
 	/* The downcall format was designed before we completely understood
 	 * the uses of the context fields; so it includes some stuff we
 	 * just give some minimal sanity-checking, and some we ignore
 	 * completely (like the next twenty bytes): */
 	if (unlikely(p + 20 > end || p + 20 < p))
-		goto out_err_free_ctx;
+		goto out_err;
 	p += 20;
 	p = simple_get_bytes(p, end, &tmp, sizeof(tmp));
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
+		goto out_err;
 	if (tmp != SGN_ALG_DES_MAC_MD5) {
 		p = ERR_PTR(-ENOSYS);
-		goto out_err_free_ctx;
+		goto out_err;
 	}
 	p = simple_get_bytes(p, end, &tmp, sizeof(tmp));
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
+		goto out_err;
 	if (tmp != SEAL_ALG_DES) {
 		p = ERR_PTR(-ENOSYS);
-		goto out_err_free_ctx;
+		goto out_err;
 	}
 	p = simple_get_bytes(p, end, &ctx->endtime, sizeof(ctx->endtime));
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
+		goto out_err;
 	p = simple_get_bytes(p, end, &ctx->seq_send, sizeof(ctx->seq_send));
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
+		goto out_err;
 	p = simple_get_netobj(p, end, &ctx->mech_used);
 	if (IS_ERR(p))
-		goto out_err_free_ctx;
-	p = get_key(p, end, &ctx->enc);
+		goto out_err;
+	p = get_key(p, end, ctx, &ctx->enc);
 	if (IS_ERR(p))
 		goto out_err_free_mech;
-	p = get_key(p, end, &ctx->seq);
+	p = get_key(p, end, ctx, &ctx->seq);
 	if (IS_ERR(p))
 		goto out_err_free_key1;
 	if (p != end) {
@@ -180,9 +326,6 @@ gss_import_sec_context_kerberos(const void *p,
 		goto out_err_free_key2;
 	}
 
-	ctx_id->internal_ctx_id = ctx;
-
-	dprintk("RPC:       Successfully imported new context.\n");
 	return 0;
 
 out_err_free_key2:
@@ -191,18 +334,378 @@ out_err_free_key1:
 	crypto_free_blkcipher(ctx->enc);
 out_err_free_mech:
 	kfree(ctx->mech_used.data);
-out_err_free_ctx:
-	kfree(ctx);
 out_err:
 	return PTR_ERR(p);
 }
 
+struct crypto_blkcipher *
+context_v2_alloc_cipher(struct krb5_ctx *ctx, const char *cname, u8 *key)
+{
+	struct crypto_blkcipher *cp;
+
+	cp = crypto_alloc_blkcipher(cname, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(cp)) {
+		dprintk("gss_kerberos_mech: unable to initialize "
+			"crypto algorithm %s\n", cname);
+		return NULL;
+	}
+	if (crypto_blkcipher_setkey(cp, key, ctx->gk5e->keylength)) {
+		dprintk("gss_kerberos_mech: error setting key for "
+			"crypto algorithm %s\n", cname);
+		crypto_free_blkcipher(cp);
+		return NULL;
+	}
+	return cp;
+}
+
+static inline void
+set_cdata(u8 cdata[GSS_KRB5_K5CLENGTH], u32 usage, u8 seed)
+{
+	cdata[0] = (usage>>24)&0xff;
+	cdata[1] = (usage>>16)&0xff;
+	cdata[2] = (usage>>8)&0xff;
+	cdata[3] = usage&0xff;
+	cdata[4] = seed;
+}
+
+static int
+context_derive_keys_des3(struct krb5_ctx *ctx, gfp_t gfp_mask)
+{
+	struct xdr_netobj c, keyin, keyout;
+	u8 cdata[GSS_KRB5_K5CLENGTH];
+	u32 err;
+
+	c.len = GSS_KRB5_K5CLENGTH;
+	c.data = cdata;
+
+	keyin.data = ctx->Ksess;
+	keyin.len = ctx->gk5e->keylength;
+	keyout.len = ctx->gk5e->keylength;
+
+	/* seq uses the raw key */
+	ctx->seq = context_v2_alloc_cipher(ctx, ctx->gk5e->encrypt_name,
+					   ctx->Ksess);
+	if (ctx->seq == NULL)
+		goto out_err;
+
+	ctx->enc = context_v2_alloc_cipher(ctx, ctx->gk5e->encrypt_name,
+					   ctx->Ksess);
+	if (ctx->enc == NULL)
+		goto out_free_seq;
+
+	/* derive cksum */
+	set_cdata(cdata, KG_USAGE_SIGN, KEY_USAGE_SEED_CHECKSUM);
+	keyout.data = ctx->cksum;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving cksum key\n",
+			__func__, err);
+		goto out_free_enc;
+	}
+
+	return 0;
+
+out_free_enc:
+	crypto_free_blkcipher(ctx->enc);
+out_free_seq:
+	crypto_free_blkcipher(ctx->seq);
+out_err:
+	return -EINVAL;
+}
+
+/*
+ * Note that RC4 depends on deriving keys using the sequence
+ * number or the checksum of a token.  Therefore, the final keys
+ * cannot be calculated until the token is being constructed!
+ */
+static int
+context_derive_keys_rc4(struct krb5_ctx *ctx)
+{
+	struct crypto_hash *hmac;
+	char sigkeyconstant[] = "signaturekey";
+	int slen = strlen(sigkeyconstant) + 1;	/* include null terminator */
+	struct hash_desc desc;
+	struct scatterlist sg[1];
+	int err;
+
+	dprintk("RPC:       %s: entered\n", __func__);
+	/*
+	 * derive cksum (aka Ksign) key
+	 */
+	hmac = crypto_alloc_hash(ctx->gk5e->cksum_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(hmac)) {
+		dprintk("%s: error %ld allocating hash '%s'\n",
+			__func__, PTR_ERR(hmac), ctx->gk5e->cksum_name);
+		err = PTR_ERR(hmac);
+		goto out_err;
+	}
+
+	err = crypto_hash_setkey(hmac, ctx->Ksess, ctx->gk5e->keylength);
+	if (err)
+		goto out_err_free_hmac;
+
+	sg_init_table(sg, 1);
+	sg_set_buf(sg, sigkeyconstant, slen);
+
+	desc.tfm = hmac;
+	desc.flags = 0;
+
+	err = crypto_hash_init(&desc);
+	if (err)
+		goto out_err_free_hmac;
+
+	err = crypto_hash_digest(&desc, sg, slen, ctx->cksum);
+	if (err)
+		goto out_err_free_hmac;
+	/*
+	 * allocate hash, and blkciphers for data and seqnum encryption
+	 */
+	ctx->enc = crypto_alloc_blkcipher(ctx->gk5e->encrypt_name, 0,
+					  CRYPTO_ALG_ASYNC);
+	if (IS_ERR(ctx->enc)) {
+		err = PTR_ERR(ctx->enc);
+		goto out_err_free_hmac;
+	}
+
+	ctx->seq = crypto_alloc_blkcipher(ctx->gk5e->encrypt_name, 0,
+					  CRYPTO_ALG_ASYNC);
+	if (IS_ERR(ctx->seq)) {
+		crypto_free_blkcipher(ctx->enc);
+		err = PTR_ERR(ctx->seq);
+		goto out_err_free_hmac;
+	}
+
+	dprintk("RPC:       %s: returning success\n", __func__);
+
+	err = 0;
+
+out_err_free_hmac:
+	crypto_free_hash(hmac);
+out_err:
+	dprintk("RPC:       %s: returning %d\n", __func__, err);
+	return err;
+}
+
+static int
+context_derive_keys_new(struct krb5_ctx *ctx, gfp_t gfp_mask)
+{
+	struct xdr_netobj c, keyin, keyout;
+	u8 cdata[GSS_KRB5_K5CLENGTH];
+	u32 err;
+
+	c.len = GSS_KRB5_K5CLENGTH;
+	c.data = cdata;
+
+	keyin.data = ctx->Ksess;
+	keyin.len = ctx->gk5e->keylength;
+	keyout.len = ctx->gk5e->keylength;
+
+	/* initiator seal encryption */
+	set_cdata(cdata, KG_USAGE_INITIATOR_SEAL, KEY_USAGE_SEED_ENCRYPTION);
+	keyout.data = ctx->initiator_seal;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving initiator_seal key\n",
+			__func__, err);
+		goto out_err;
+	}
+	ctx->initiator_enc = context_v2_alloc_cipher(ctx,
+						     ctx->gk5e->encrypt_name,
+						     ctx->initiator_seal);
+	if (ctx->initiator_enc == NULL)
+		goto out_err;
+
+	/* acceptor seal encryption */
+	set_cdata(cdata, KG_USAGE_ACCEPTOR_SEAL, KEY_USAGE_SEED_ENCRYPTION);
+	keyout.data = ctx->acceptor_seal;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving acceptor_seal key\n",
+			__func__, err);
+		goto out_free_initiator_enc;
+	}
+	ctx->acceptor_enc = context_v2_alloc_cipher(ctx,
+						    ctx->gk5e->encrypt_name,
+						    ctx->acceptor_seal);
+	if (ctx->acceptor_enc == NULL)
+		goto out_free_initiator_enc;
+
+	/* initiator sign checksum */
+	set_cdata(cdata, KG_USAGE_INITIATOR_SIGN, KEY_USAGE_SEED_CHECKSUM);
+	keyout.data = ctx->initiator_sign;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving initiator_sign key\n",
+			__func__, err);
+		goto out_free_acceptor_enc;
+	}
+
+	/* acceptor sign checksum */
+	set_cdata(cdata, KG_USAGE_ACCEPTOR_SIGN, KEY_USAGE_SEED_CHECKSUM);
+	keyout.data = ctx->acceptor_sign;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving acceptor_sign key\n",
+			__func__, err);
+		goto out_free_acceptor_enc;
+	}
+
+	/* initiator seal integrity */
+	set_cdata(cdata, KG_USAGE_INITIATOR_SEAL, KEY_USAGE_SEED_INTEGRITY);
+	keyout.data = ctx->initiator_integ;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving initiator_integ key\n",
+			__func__, err);
+		goto out_free_acceptor_enc;
+	}
+
+	/* acceptor seal integrity */
+	set_cdata(cdata, KG_USAGE_ACCEPTOR_SEAL, KEY_USAGE_SEED_INTEGRITY);
+	keyout.data = ctx->acceptor_integ;
+	err = krb5_derive_key(ctx->gk5e, &keyin, &keyout, &c, gfp_mask);
+	if (err) {
+		dprintk("%s: Error %d deriving acceptor_integ key\n",
+			__func__, err);
+		goto out_free_acceptor_enc;
+	}
+
+	switch (ctx->enctype) {
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		ctx->initiator_enc_aux =
+			context_v2_alloc_cipher(ctx, "cbc(aes)",
+						ctx->initiator_seal);
+		if (ctx->initiator_enc_aux == NULL)
+			goto out_free_acceptor_enc;
+		ctx->acceptor_enc_aux =
+			context_v2_alloc_cipher(ctx, "cbc(aes)",
+						ctx->acceptor_seal);
+		if (ctx->acceptor_enc_aux == NULL) {
+			crypto_free_blkcipher(ctx->initiator_enc_aux);
+			goto out_free_acceptor_enc;
+		}
+	}
+
+	return 0;
+
+out_free_acceptor_enc:
+	crypto_free_blkcipher(ctx->acceptor_enc);
+out_free_initiator_enc:
+	crypto_free_blkcipher(ctx->initiator_enc);
+out_err:
+	return -EINVAL;
+}
+
+static int
+gss_import_v2_context(const void *p, const void *end, struct krb5_ctx *ctx,
+		gfp_t gfp_mask)
+{
+	int keylen;
+
+	p = simple_get_bytes(p, end, &ctx->flags, sizeof(ctx->flags));
+	if (IS_ERR(p))
+		goto out_err;
+	ctx->initiate = ctx->flags & KRB5_CTX_FLAG_INITIATOR;
+
+	p = simple_get_bytes(p, end, &ctx->endtime, sizeof(ctx->endtime));
+	if (IS_ERR(p))
+		goto out_err;
+	p = simple_get_bytes(p, end, &ctx->seq_send64, sizeof(ctx->seq_send64));
+	if (IS_ERR(p))
+		goto out_err;
+	/* set seq_send for use by "older" enctypes */
+	ctx->seq_send = ctx->seq_send64;
+	if (ctx->seq_send64 != ctx->seq_send) {
+		dprintk("%s: seq_send64 %lx, seq_send %x overflow?\n", __func__,
+			(long unsigned)ctx->seq_send64, ctx->seq_send);
+		goto out_err;
+	}
+	p = simple_get_bytes(p, end, &ctx->enctype, sizeof(ctx->enctype));
+	if (IS_ERR(p))
+		goto out_err;
+	/* Map ENCTYPE_DES3_CBC_SHA1 to ENCTYPE_DES3_CBC_RAW */
+	if (ctx->enctype == ENCTYPE_DES3_CBC_SHA1)
+		ctx->enctype = ENCTYPE_DES3_CBC_RAW;
+	ctx->gk5e = get_gss_krb5_enctype(ctx->enctype);
+	if (ctx->gk5e == NULL) {
+		dprintk("gss_kerberos_mech: unsupported krb5 enctype %u\n",
+			ctx->enctype);
+		p = ERR_PTR(-EINVAL);
+		goto out_err;
+	}
+	keylen = ctx->gk5e->keylength;
+
+	p = simple_get_bytes(p, end, ctx->Ksess, keylen);
+	if (IS_ERR(p))
+		goto out_err;
+
+	if (p != end) {
+		p = ERR_PTR(-EINVAL);
+		goto out_err;
+	}
+
+	ctx->mech_used.data = kmemdup(gss_kerberos_mech.gm_oid.data,
+				      gss_kerberos_mech.gm_oid.len, gfp_mask);
+	if (unlikely(ctx->mech_used.data == NULL)) {
+		p = ERR_PTR(-ENOMEM);
+		goto out_err;
+	}
+	ctx->mech_used.len = gss_kerberos_mech.gm_oid.len;
+
+	switch (ctx->enctype) {
+	case ENCTYPE_DES3_CBC_RAW:
+		return context_derive_keys_des3(ctx, gfp_mask);
+	case ENCTYPE_ARCFOUR_HMAC:
+		return context_derive_keys_rc4(ctx);
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		return context_derive_keys_new(ctx, gfp_mask);
+	default:
+		return -EINVAL;
+	}
+
+out_err:
+	return PTR_ERR(p);
+}
+
+static int
+gss_import_sec_context_kerberos(const void *p, size_t len,
+				struct gss_ctx *ctx_id,
+				gfp_t gfp_mask)
+{
+	const void *end = (const void *)((const char *)p + len);
+	struct  krb5_ctx *ctx;
+	int ret;
+
+	ctx = kzalloc(sizeof(*ctx), gfp_mask);
+	if (ctx == NULL)
+		return -ENOMEM;
+
+	if (len == 85)
+		ret = gss_import_v1_context(p, end, ctx);
+	else
+		ret = gss_import_v2_context(p, end, ctx, gfp_mask);
+
+	if (ret == 0)
+		ctx_id->internal_ctx_id = ctx;
+	else
+		kfree(ctx);
+
+	dprintk("RPC:       %s: returning %d\n", __func__, ret);
+	return ret;
+}
+
 static void
 gss_delete_sec_context_kerberos(void *internal_ctx) {
 	struct krb5_ctx *kctx = internal_ctx;
 
 	crypto_free_blkcipher(kctx->seq);
 	crypto_free_blkcipher(kctx->enc);
+	crypto_free_blkcipher(kctx->acceptor_enc);
+	crypto_free_blkcipher(kctx->initiator_enc);
+	crypto_free_blkcipher(kctx->acceptor_enc_aux);
+	crypto_free_blkcipher(kctx->initiator_enc_aux);
 	kfree(kctx->mech_used.data);
 	kfree(kctx);
 }
@@ -241,6 +744,7 @@ static struct gss_api_mech gss_kerberos_mech = {
 	.gm_ops		= &gss_kerberos_ops,
 	.gm_pf_num	= ARRAY_SIZE(gss_kerberos_pfs),
 	.gm_pfs		= gss_kerberos_pfs,
+	.gm_upcall_enctypes = "enctypes=18,17,16,23,3,1,2 ",
 };
 
 static int __init init_kerberos_module(void)

net/sunrpc/auth_gss/gss_krb5_seal.c

@@ -3,7 +3,7 @@
  *
  *  Adapted from MIT Kerberos 5-1.2.1 lib/gssapi/krb5/k5seal.c
  *
- *  Copyright (c) 2000 The Regents of the University of Michigan.
+ *  Copyright (c) 2000-2008 The Regents of the University of Michigan.
  *  All rights reserved.
  *
  *  Andy Adamson	<andros@umich.edu>
@@ -70,53 +70,154 @@
 
 DEFINE_SPINLOCK(krb5_seq_lock);
 
-u32
-gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text,
+static char *
+setup_token(struct krb5_ctx *ctx, struct xdr_netobj *token)
+{
+	__be16 *ptr, *krb5_hdr;
+	int body_size = GSS_KRB5_TOK_HDR_LEN + ctx->gk5e->cksumlength;
+
+	token->len = g_token_size(&ctx->mech_used, body_size);
+
+	ptr = (__be16 *)token->data;
+	g_make_token_header(&ctx->mech_used, body_size, (unsigned char **)&ptr);
+
+	/* ptr now at start of header described in rfc 1964, section 1.2.1: */
+	krb5_hdr = ptr;
+	*ptr++ = KG_TOK_MIC_MSG;
+	*ptr++ = cpu_to_le16(ctx->gk5e->signalg);
+	*ptr++ = SEAL_ALG_NONE;
+	*ptr++ = 0xffff;
+
+	return (char *)krb5_hdr;
+}
+
+static void *
+setup_token_v2(struct krb5_ctx *ctx, struct xdr_netobj *token)
+{
+	__be16 *ptr, *krb5_hdr;
+	u8 *p, flags = 0x00;
+
+	if ((ctx->flags & KRB5_CTX_FLAG_INITIATOR) == 0)
+		flags |= 0x01;
+	if (ctx->flags & KRB5_CTX_FLAG_ACCEPTOR_SUBKEY)
+		flags |= 0x04;
+
+	/* Per rfc 4121, sec 4.2.6.1, there is no header,
+	 * just start the token */
+	krb5_hdr = ptr = (__be16 *)token->data;
+
+	*ptr++ = KG2_TOK_MIC;
+	p = (u8 *)ptr;
+	*p++ = flags;
+	*p++ = 0xff;
+	ptr = (__be16 *)p;
+	*ptr++ = 0xffff;
+	*ptr++ = 0xffff;
+
+	token->len = GSS_KRB5_TOK_HDR_LEN + ctx->gk5e->cksumlength;
+	return krb5_hdr;
+}
+
+static u32
+gss_get_mic_v1(struct krb5_ctx *ctx, struct xdr_buf *text,
 		struct xdr_netobj *token)
 {
-	struct krb5_ctx		*ctx = gss_ctx->internal_ctx_id;
-	char			cksumdata[16];
-	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
-	unsigned char		*ptr, *msg_start;
+	char			cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj	md5cksum = {.len = sizeof(cksumdata),
+					    .data = cksumdata};
+	void			*ptr;
 	s32			now;
 	u32			seq_send;
+	u8			*cksumkey;
 
-	dprintk("RPC:       gss_krb5_seal\n");
+	dprintk("RPC:       %s\n", __func__);
 	BUG_ON(ctx == NULL);
 
 	now = get_seconds();
 
-	token->len = g_token_size(&ctx->mech_used, GSS_KRB5_TOK_HDR_LEN + 8);
+	ptr = setup_token(ctx, token);
 
-	ptr = token->data;
-	g_make_token_header(&ctx->mech_used, GSS_KRB5_TOK_HDR_LEN + 8, &ptr);
+	if (ctx->gk5e->keyed_cksum)
+		cksumkey = ctx->cksum;
+	else
+		cksumkey = NULL;
 
-	/* ptr now at header described in rfc 1964, section 1.2.1: */
-	ptr[0] = (unsigned char) ((KG_TOK_MIC_MSG >> 8) & 0xff);
-	ptr[1] = (unsigned char) (KG_TOK_MIC_MSG & 0xff);
+	if (make_checksum(ctx, ptr, 8, text, 0, cksumkey,
+			  KG_USAGE_SIGN, &md5cksum))
+		return GSS_S_FAILURE;
 
-	msg_start = ptr + GSS_KRB5_TOK_HDR_LEN + 8;
+	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data, md5cksum.len);
 
-	*(__be16 *)(ptr + 2) = htons(SGN_ALG_DES_MAC_MD5);
-	memset(ptr + 4, 0xff, 4);
+	spin_lock(&krb5_seq_lock);
+	seq_send = ctx->seq_send++;
+	spin_unlock(&krb5_seq_lock);
 
-	if (make_checksum("md5", ptr, 8, text, 0, &md5cksum))
+	if (krb5_make_seq_num(ctx, ctx->seq, ctx->initiate ? 0 : 0xff,
+			      seq_send, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8))
 		return GSS_S_FAILURE;
 
-	if (krb5_encrypt(ctx->seq, NULL, md5cksum.data,
-			  md5cksum.data, md5cksum.len))
-		return GSS_S_FAILURE;
+	return (ctx->endtime < now) ? GSS_S_CONTEXT_EXPIRED : GSS_S_COMPLETE;
+}
+
+u32
+gss_get_mic_v2(struct krb5_ctx *ctx, struct xdr_buf *text,
+		struct xdr_netobj *token)
+{
+	char cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj cksumobj = { .len = sizeof(cksumdata),
+				       .data = cksumdata};
+	void *krb5_hdr;
+	s32 now;
+	u64 seq_send;
+	u8 *cksumkey;
+	unsigned int cksum_usage;
 
-	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data + md5cksum.len - 8, 8);
+	dprintk("RPC:       %s\n", __func__);
 
+	krb5_hdr = setup_token_v2(ctx, token);
+
+	/* Set up the sequence number. Now 64-bits in clear
+	 * text and w/o direction indicator */
 	spin_lock(&krb5_seq_lock);
-	seq_send = ctx->seq_send++;
+	seq_send = ctx->seq_send64++;
 	spin_unlock(&krb5_seq_lock);
-
-	if (krb5_make_seq_num(ctx->seq, ctx->initiate ? 0 : 0xff,
-			      seq_send, ptr + GSS_KRB5_TOK_HDR_LEN,
-			      ptr + 8))
+	*((u64 *)(krb5_hdr + 8)) = cpu_to_be64(seq_send);
+
+	if (ctx->initiate) {
+		cksumkey = ctx->initiator_sign;
+		cksum_usage = KG_USAGE_INITIATOR_SIGN;
+	} else {
+		cksumkey = ctx->acceptor_sign;
+		cksum_usage = KG_USAGE_ACCEPTOR_SIGN;
+	}
+
+	if (make_checksum_v2(ctx, krb5_hdr, GSS_KRB5_TOK_HDR_LEN,
+			     text, 0, cksumkey, cksum_usage, &cksumobj))
 		return GSS_S_FAILURE;
 
+	memcpy(krb5_hdr + GSS_KRB5_TOK_HDR_LEN, cksumobj.data, cksumobj.len);
+
+	now = get_seconds();
+
 	return (ctx->endtime < now) ? GSS_S_CONTEXT_EXPIRED : GSS_S_COMPLETE;
 }
+
+u32
+gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text,
+		     struct xdr_netobj *token)
+{
+	struct krb5_ctx		*ctx = gss_ctx->internal_ctx_id;
+
+	switch (ctx->enctype) {
+	default:
+		BUG();
+	case ENCTYPE_DES_CBC_RAW:
+	case ENCTYPE_DES3_CBC_RAW:
+	case ENCTYPE_ARCFOUR_HMAC:
+		return gss_get_mic_v1(ctx, text, token);
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		return gss_get_mic_v2(ctx, text, token);
+	}
+}
+

net/sunrpc/auth_gss/gss_krb5_seqnum.c

@@ -39,14 +39,51 @@
 # define RPCDBG_FACILITY        RPCDBG_AUTH
 #endif
 
+static s32
+krb5_make_rc4_seq_num(struct krb5_ctx *kctx, int direction, s32 seqnum,
+		      unsigned char *cksum, unsigned char *buf)
+{
+	struct crypto_blkcipher *cipher;
+	unsigned char plain[8];
+	s32 code;
+
+	dprintk("RPC:       %s:\n", __func__);
+	cipher = crypto_alloc_blkcipher(kctx->gk5e->encrypt_name, 0,
+					CRYPTO_ALG_ASYNC);
+	if (IS_ERR(cipher))
+		return PTR_ERR(cipher);
+
+	plain[0] = (unsigned char) ((seqnum >> 24) & 0xff);
+	plain[1] = (unsigned char) ((seqnum >> 16) & 0xff);
+	plain[2] = (unsigned char) ((seqnum >> 8) & 0xff);
+	plain[3] = (unsigned char) ((seqnum >> 0) & 0xff);
+	plain[4] = direction;
+	plain[5] = direction;
+	plain[6] = direction;
+	plain[7] = direction;
+
+	code = krb5_rc4_setup_seq_key(kctx, cipher, cksum);
+	if (code)
+		goto out;
+
+	code = krb5_encrypt(cipher, cksum, plain, buf, 8);
+out:
+	crypto_free_blkcipher(cipher);
+	return code;
+}
 s32
-krb5_make_seq_num(struct crypto_blkcipher *key,
+krb5_make_seq_num(struct krb5_ctx *kctx,
+		struct crypto_blkcipher *key,
 		int direction,
 		u32 seqnum,
 		unsigned char *cksum, unsigned char *buf)
 {
 	unsigned char plain[8];
 
+	if (kctx->enctype == ENCTYPE_ARCFOUR_HMAC)
+		return krb5_make_rc4_seq_num(kctx, direction, seqnum,
+					     cksum, buf);
+
 	plain[0] = (unsigned char) (seqnum & 0xff);
 	plain[1] = (unsigned char) ((seqnum >> 8) & 0xff);
 	plain[2] = (unsigned char) ((seqnum >> 16) & 0xff);
@@ -60,17 +97,59 @@ krb5_make_seq_num(struct crypto_blkcipher *key,
 	return krb5_encrypt(key, cksum, plain, buf, 8);
 }
 
+static s32
+krb5_get_rc4_seq_num(struct krb5_ctx *kctx, unsigned char *cksum,
+		     unsigned char *buf, int *direction, s32 *seqnum)
+{
+	struct crypto_blkcipher *cipher;
+	unsigned char plain[8];
+	s32 code;
+
+	dprintk("RPC:       %s:\n", __func__);
+	cipher = crypto_alloc_blkcipher(kctx->gk5e->encrypt_name, 0,
+					CRYPTO_ALG_ASYNC);
+	if (IS_ERR(cipher))
+		return PTR_ERR(cipher);
+
+	code = krb5_rc4_setup_seq_key(kctx, cipher, cksum);
+	if (code)
+		goto out;
+
+	code = krb5_decrypt(cipher, cksum, buf, plain, 8);
+	if (code)
+		goto out;
+
+	if ((plain[4] != plain[5]) || (plain[4] != plain[6])
+				   || (plain[4] != plain[7])) {
+		code = (s32)KG_BAD_SEQ;
+		goto out;
+	}
+
+	*direction = plain[4];
+
+	*seqnum = ((plain[0] << 24) | (plain[1] << 16) |
+					(plain[2] << 8) | (plain[3]));
+out:
+	crypto_free_blkcipher(cipher);
+	return code;
+}
+
 s32
-krb5_get_seq_num(struct crypto_blkcipher *key,
+krb5_get_seq_num(struct krb5_ctx *kctx,
 	       unsigned char *cksum,
 	       unsigned char *buf,
 	       int *direction, u32 *seqnum)
 {
 	s32 code;
 	unsigned char plain[8];
+	struct crypto_blkcipher *key = kctx->seq;
 
 	dprintk("RPC:       krb5_get_seq_num:\n");
 
+	if (kctx->enctype == ENCTYPE_ARCFOUR_HMAC)
+		return krb5_get_rc4_seq_num(kctx, cksum, buf,
+					    direction, seqnum);
+
 	if ((code = krb5_decrypt(key, cksum, buf, plain, 8)))
 		return code;
 

net/sunrpc/auth_gss/gss_krb5_unseal.c

@@ -3,7 +3,7 @@
  *
  *  Adapted from MIT Kerberos 5-1.2.1 lib/gssapi/krb5/k5unseal.c
  *
- *  Copyright (c) 2000 The Regents of the University of Michigan.
+ *  Copyright (c) 2000-2008 The Regents of the University of Michigan.
  *  All rights reserved.
  *
  *  Andy Adamson   <andros@umich.edu>
@@ -70,20 +70,21 @@
 /* read_token is a mic token, and message_buffer is the data that the mic was
  * supposedly taken over. */
 
-u32
-gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
+static u32
+gss_verify_mic_v1(struct krb5_ctx *ctx,
 		struct xdr_buf *message_buffer, struct xdr_netobj *read_token)
 {
-	struct krb5_ctx		*ctx = gss_ctx->internal_ctx_id;
 	int			signalg;
 	int			sealalg;
-	char			cksumdata[16];
-	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
+	char			cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj	md5cksum = {.len = sizeof(cksumdata),
+					    .data = cksumdata};
 	s32			now;
 	int			direction;
 	u32			seqnum;
 	unsigned char		*ptr = (unsigned char *)read_token->data;
 	int			bodysize;
+	u8			*cksumkey;
 
 	dprintk("RPC:       krb5_read_token\n");
 
@@ -98,7 +99,7 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 	/* XXX sanity-check bodysize?? */
 
 	signalg = ptr[2] + (ptr[3] << 8);
-	if (signalg != SGN_ALG_DES_MAC_MD5)
+	if (signalg != ctx->gk5e->signalg)
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	sealalg = ptr[4] + (ptr[5] << 8);
@@ -108,13 +109,17 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 	if ((ptr[6] != 0xff) || (ptr[7] != 0xff))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if (make_checksum("md5", ptr, 8, message_buffer, 0, &md5cksum))
-		return GSS_S_FAILURE;
+	if (ctx->gk5e->keyed_cksum)
+		cksumkey = ctx->cksum;
+	else
+		cksumkey = NULL;
 
-	if (krb5_encrypt(ctx->seq, NULL, md5cksum.data, md5cksum.data, 16))
+	if (make_checksum(ctx, ptr, 8, message_buffer, 0,
+			  cksumkey, KG_USAGE_SIGN, &md5cksum))
 		return GSS_S_FAILURE;
 
-	if (memcmp(md5cksum.data + 8, ptr + GSS_KRB5_TOK_HDR_LEN, 8))
+	if (memcmp(md5cksum.data, ptr + GSS_KRB5_TOK_HDR_LEN,
+					ctx->gk5e->cksumlength))
 		return GSS_S_BAD_SIG;
 
 	/* it got through unscathed.  Make sure the context is unexpired */
@@ -126,7 +131,8 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 
 	/* do sequencing checks */
 
-	if (krb5_get_seq_num(ctx->seq, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8, &direction, &seqnum))
+	if (krb5_get_seq_num(ctx, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8,
+			     &direction, &seqnum))
 		return GSS_S_FAILURE;
 
 	if ((ctx->initiate && direction != 0xff) ||
@@ -135,3 +141,86 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 
 	return GSS_S_COMPLETE;
 }
+
+static u32
+gss_verify_mic_v2(struct krb5_ctx *ctx,
+		struct xdr_buf *message_buffer, struct xdr_netobj *read_token)
+{
+	char cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj cksumobj = {.len = sizeof(cksumdata),
+				      .data = cksumdata};
+	s32 now;
+	u64 seqnum;
+	u8 *ptr = read_token->data;
+	u8 *cksumkey;
+	u8 flags;
+	int i;
+	unsigned int cksum_usage;
+
+	dprintk("RPC:       %s\n", __func__);
+
+	if (be16_to_cpu(*((__be16 *)ptr)) != KG2_TOK_MIC)
+		return GSS_S_DEFECTIVE_TOKEN;
+
+	flags = ptr[2];
+	if ((!ctx->initiate && (flags & KG2_TOKEN_FLAG_SENTBYACCEPTOR)) ||
+	    (ctx->initiate && !(flags & KG2_TOKEN_FLAG_SENTBYACCEPTOR)))
+		return GSS_S_BAD_SIG;
+
+	if (flags & KG2_TOKEN_FLAG_SEALED) {
+		dprintk("%s: token has unexpected sealed flag\n", __func__);
+		return GSS_S_FAILURE;
+	}
+
+	for (i = 3; i < 8; i++)
+		if (ptr[i] != 0xff)
+			return GSS_S_DEFECTIVE_TOKEN;
+
+	if (ctx->initiate) {
+		cksumkey = ctx->acceptor_sign;
+		cksum_usage = KG_USAGE_ACCEPTOR_SIGN;
+	} else {
+		cksumkey = ctx->initiator_sign;
+		cksum_usage = KG_USAGE_INITIATOR_SIGN;
+	}
+
+	if (make_checksum_v2(ctx, ptr, GSS_KRB5_TOK_HDR_LEN, message_buffer, 0,
+			     cksumkey, cksum_usage, &cksumobj))
+		return GSS_S_FAILURE;
+
+	if (memcmp(cksumobj.data, ptr + GSS_KRB5_TOK_HDR_LEN,
+				ctx->gk5e->cksumlength))
+		return GSS_S_BAD_SIG;
+
+	/* it got through unscathed.  Make sure the context is unexpired */
+	now = get_seconds();
+	if (now > ctx->endtime)
+		return GSS_S_CONTEXT_EXPIRED;
+
+	/* do sequencing checks */
+
+	seqnum = be64_to_cpup((__be64 *)ptr + 8);
+
+	return GSS_S_COMPLETE;
+}
+
+u32
+gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
+			struct xdr_buf *message_buffer,
+			struct xdr_netobj *read_token)
+{
+	struct krb5_ctx *ctx = gss_ctx->internal_ctx_id;
+
+	switch (ctx->enctype) {
+	default:
+		BUG();
+	case ENCTYPE_DES_CBC_RAW:
+	case ENCTYPE_DES3_CBC_RAW:
+	case ENCTYPE_ARCFOUR_HMAC:
+		return gss_verify_mic_v1(ctx, message_buffer, read_token);
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		return gss_verify_mic_v2(ctx, message_buffer, read_token);
+	}
+}
+

net/sunrpc/auth_gss/gss_krb5_wrap.c

@@ -1,3 +1,33 @@
+/*
+ * COPYRIGHT (c) 2008
+ * The Regents of the University of Michigan
+ * ALL RIGHTS RESERVED
+ *
+ * Permission is granted to use, copy, create derivative works
+ * and redistribute this software and such derivative works
+ * for any purpose, so long as the name of The University of
+ * Michigan is not used in any advertising or publicity
+ * pertaining to the use of distribution of this software
+ * without specific, written prior authorization.  If the
+ * above copyright notice or any other identification of the
+ * University of Michigan is included in any copy of any
+ * portion of this software, then the disclaimer below must
+ * also be included.
+ *
+ * THIS SOFTWARE IS PROVIDED AS IS, WITHOUT REPRESENTATION
+ * FROM THE UNIVERSITY OF MICHIGAN AS TO ITS FITNESS FOR ANY
+ * PURPOSE, AND WITHOUT WARRANTY BY THE UNIVERSITY OF
+ * MICHIGAN OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING
+ * WITHOUT LIMITATION THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE
+ * REGENTS OF THE UNIVERSITY OF MICHIGAN SHALL NOT BE LIABLE
+ * FOR ANY DAMAGES, INCLUDING SPECIAL, INDIRECT, INCIDENTAL, OR
+ * CONSEQUENTIAL DAMAGES, WITH RESPECT TO ANY CLAIM ARISING
+ * OUT OF OR IN CONNECTION WITH THE USE OF THE SOFTWARE, EVEN
+ * IF IT HAS BEEN OR IS HEREAFTER ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGES.
+ */
+
 #include <linux/types.h>
 #include <linux/jiffies.h>
 #include <linux/sunrpc/gss_krb5.h>
@@ -12,10 +42,7 @@
 static inline int
 gss_krb5_padding(int blocksize, int length)
 {
-	/* Most of the code is block-size independent but currently we
-	 * use only 8: */
-	BUG_ON(blocksize != 8);
-	return 8 - (length & 7);
+	return blocksize - (length % blocksize);
 }
 
 static inline void
@@ -86,8 +113,8 @@ out:
 	return 0;
 }
 
-static void
-make_confounder(char *p, u32 conflen)
+void
+gss_krb5_make_confounder(char *p, u32 conflen)
 {
 	static u64 i = 0;
 	u64 *q = (u64 *)p;
@@ -127,69 +154,73 @@ make_confounder(char *p, u32 conflen)
 
 /* XXX factor out common code with seal/unseal. */
 
-u32
-gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
+static u32
+gss_wrap_kerberos_v1(struct krb5_ctx *kctx, int offset,
 		struct xdr_buf *buf, struct page **pages)
 {
-	struct krb5_ctx		*kctx = ctx->internal_ctx_id;
-	char			cksumdata[16];
-	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
+	char			cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj	md5cksum = {.len = sizeof(cksumdata),
+					    .data = cksumdata};
 	int			blocksize = 0, plainlen;
 	unsigned char		*ptr, *msg_start;
 	s32			now;
 	int			headlen;
 	struct page		**tmp_pages;
 	u32			seq_send;
+	u8			*cksumkey;
+	u32			conflen = kctx->gk5e->conflen;
 
-	dprintk("RPC:       gss_wrap_kerberos\n");
+	dprintk("RPC:       %s\n", __func__);
 
 	now = get_seconds();
 
 	blocksize = crypto_blkcipher_blocksize(kctx->enc);
 	gss_krb5_add_padding(buf, offset, blocksize);
 	BUG_ON((buf->len - offset) % blocksize);
-	plainlen = blocksize + buf->len - offset;
+	plainlen = conflen + buf->len - offset;
 
-	headlen = g_token_size(&kctx->mech_used, 24 + plainlen) -
-						(buf->len - offset);
+	headlen = g_token_size(&kctx->mech_used,
+		GSS_KRB5_TOK_HDR_LEN + kctx->gk5e->cksumlength + plainlen) -
+		(buf->len - offset);
 
 	ptr = buf->head[0].iov_base + offset;
 	/* shift data to make room for header. */
+	xdr_extend_head(buf, offset, headlen);
+
 	/* XXX Would be cleverer to encrypt while copying. */
-	/* XXX bounds checking, slack, etc. */
-	memmove(ptr + headlen, ptr, buf->head[0].iov_len - offset);
-	buf->head[0].iov_len += headlen;
-	buf->len += headlen;
 	BUG_ON((buf->len - offset - headlen) % blocksize);
 
 	g_make_token_header(&kctx->mech_used,
-				GSS_KRB5_TOK_HDR_LEN + 8 + plainlen, &ptr);
+				GSS_KRB5_TOK_HDR_LEN +
+				kctx->gk5e->cksumlength + plainlen, &ptr);
 
 
 	/* ptr now at header described in rfc 1964, section 1.2.1: */
 	ptr[0] = (unsigned char) ((KG_TOK_WRAP_MSG >> 8) & 0xff);
 	ptr[1] = (unsigned char) (KG_TOK_WRAP_MSG & 0xff);
 
-	msg_start = ptr + 24;
+	msg_start = ptr + GSS_KRB5_TOK_HDR_LEN + kctx->gk5e->cksumlength;
 
-	*(__be16 *)(ptr + 2) = htons(SGN_ALG_DES_MAC_MD5);
+	*(__be16 *)(ptr + 2) = cpu_to_le16(kctx->gk5e->signalg);
 	memset(ptr + 4, 0xff, 4);
-	*(__be16 *)(ptr + 4) = htons(SEAL_ALG_DES);
+	*(__be16 *)(ptr + 4) = cpu_to_le16(kctx->gk5e->sealalg);
 
-	make_confounder(msg_start, blocksize);
+	gss_krb5_make_confounder(msg_start, conflen);
+
+	if (kctx->gk5e->keyed_cksum)
+		cksumkey = kctx->cksum;
+	else
+		cksumkey = NULL;
 
 	/* XXXJBF: UGH!: */
 	tmp_pages = buf->pages;
 	buf->pages = pages;
-	if (make_checksum("md5", ptr, 8, buf,
-				offset + headlen - blocksize, &md5cksum))
+	if (make_checksum(kctx, ptr, 8, buf, offset + headlen - conflen,
+					cksumkey, KG_USAGE_SEAL, &md5cksum))
 		return GSS_S_FAILURE;
 	buf->pages = tmp_pages;
 
-	if (krb5_encrypt(kctx->seq, NULL, md5cksum.data,
-			  md5cksum.data, md5cksum.len))
-		return GSS_S_FAILURE;
-	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data + md5cksum.len - 8, 8);
+	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data, md5cksum.len);
 
 	spin_lock(&krb5_seq_lock);
 	seq_send = kctx->seq_send++;
@@ -197,25 +228,42 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 
 	/* XXX would probably be more efficient to compute checksum
 	 * and encrypt at the same time: */
-	if ((krb5_make_seq_num(kctx->seq, kctx->initiate ? 0 : 0xff,
+	if ((krb5_make_seq_num(kctx, kctx->seq, kctx->initiate ? 0 : 0xff,
 			       seq_send, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8)))
 		return GSS_S_FAILURE;
 
-	if (gss_encrypt_xdr_buf(kctx->enc, buf, offset + headlen - blocksize,
-									pages))
-		return GSS_S_FAILURE;
+	if (kctx->enctype == ENCTYPE_ARCFOUR_HMAC) {
+		struct crypto_blkcipher *cipher;
+		int err;
+		cipher = crypto_alloc_blkcipher(kctx->gk5e->encrypt_name, 0,
+						CRYPTO_ALG_ASYNC);
+		if (IS_ERR(cipher))
+			return GSS_S_FAILURE;
+
+		krb5_rc4_setup_enc_key(kctx, cipher, seq_send);
+
+		err = gss_encrypt_xdr_buf(cipher, buf,
+					  offset + headlen - conflen, pages);
+		crypto_free_blkcipher(cipher);
+		if (err)
+			return GSS_S_FAILURE;
+	} else {
+		if (gss_encrypt_xdr_buf(kctx->enc, buf,
+					offset + headlen - conflen, pages))
+			return GSS_S_FAILURE;
+	}
 
 	return (kctx->endtime < now) ? GSS_S_CONTEXT_EXPIRED : GSS_S_COMPLETE;
 }
 
-u32
-gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
+static u32
+gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 {
-	struct krb5_ctx		*kctx = ctx->internal_ctx_id;
 	int			signalg;
 	int			sealalg;
-	char			cksumdata[16];
-	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
+	char			cksumdata[GSS_KRB5_MAX_CKSUM_LEN];
+	struct xdr_netobj	md5cksum = {.len = sizeof(cksumdata),
+					    .data = cksumdata};
 	s32			now;
 	int			direction;
 	s32			seqnum;
@@ -224,6 +272,9 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 	void			*data_start, *orig_start;
 	int			data_len;
 	int			blocksize;
+	u32			conflen = kctx->gk5e->conflen;
+	int			crypt_offset;
+	u8			*cksumkey;
 
 	dprintk("RPC:       gss_unwrap_kerberos\n");
 
@@ -241,29 +292,65 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 	/* get the sign and seal algorithms */
 
 	signalg = ptr[2] + (ptr[3] << 8);
-	if (signalg != SGN_ALG_DES_MAC_MD5)
+	if (signalg != kctx->gk5e->signalg)
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	sealalg = ptr[4] + (ptr[5] << 8);
-	if (sealalg != SEAL_ALG_DES)
+	if (sealalg != kctx->gk5e->sealalg)
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	if ((ptr[6] != 0xff) || (ptr[7] != 0xff))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if (gss_decrypt_xdr_buf(kctx->enc, buf,
-			ptr + GSS_KRB5_TOK_HDR_LEN + 8 - (unsigned char *)buf->head[0].iov_base))
-		return GSS_S_DEFECTIVE_TOKEN;
+	/*
+	 * Data starts after token header and checksum.  ptr points
+	 * to the beginning of the token header
+	 */
+	crypt_offset = ptr + (GSS_KRB5_TOK_HDR_LEN + kctx->gk5e->cksumlength) -
+					(unsigned char *)buf->head[0].iov_base;
+
+	/*
+	 * Need plaintext seqnum to derive encryption key for arcfour-hmac
+	 */
+	if (krb5_get_seq_num(kctx, ptr + GSS_KRB5_TOK_HDR_LEN,
+			     ptr + 8, &direction, &seqnum))
+		return GSS_S_BAD_SIG;
 
-	if (make_checksum("md5", ptr, 8, buf,
-		 ptr + GSS_KRB5_TOK_HDR_LEN + 8 - (unsigned char *)buf->head[0].iov_base, &md5cksum))
-		return GSS_S_FAILURE;
+	if ((kctx->initiate && direction != 0xff) ||
+	    (!kctx->initiate && direction != 0))
+		return GSS_S_BAD_SIG;
+
+	if (kctx->enctype == ENCTYPE_ARCFOUR_HMAC) {
+		struct crypto_blkcipher *cipher;
+		int err;
+
+		cipher = crypto_alloc_blkcipher(kctx->gk5e->encrypt_name, 0,
+						CRYPTO_ALG_ASYNC);
+		if (IS_ERR(cipher))
+			return GSS_S_FAILURE;
+
+		krb5_rc4_setup_enc_key(kctx, cipher, seqnum);
 
-	if (krb5_encrypt(kctx->seq, NULL, md5cksum.data,
-			   md5cksum.data, md5cksum.len))
+		err = gss_decrypt_xdr_buf(cipher, buf, crypt_offset);
+		crypto_free_blkcipher(cipher);
+		if (err)
+			return GSS_S_DEFECTIVE_TOKEN;
+	} else {
+		if (gss_decrypt_xdr_buf(kctx->enc, buf, crypt_offset))
+			return GSS_S_DEFECTIVE_TOKEN;
+	}
+
+	if (kctx->gk5e->keyed_cksum)
+		cksumkey = kctx->cksum;
+	else
+		cksumkey = NULL;
+
+	if (make_checksum(kctx, ptr, 8, buf, crypt_offset,
+					cksumkey, KG_USAGE_SEAL, &md5cksum))
 		return GSS_S_FAILURE;
 
-	if (memcmp(md5cksum.data + 8, ptr + GSS_KRB5_TOK_HDR_LEN, 8))
+	if (memcmp(md5cksum.data, ptr + GSS_KRB5_TOK_HDR_LEN,
+						kctx->gk5e->cksumlength))
 		return GSS_S_BAD_SIG;
 
 	/* it got through unscathed.  Make sure the context is unexpired */
@@ -275,19 +362,12 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 
 	/* do sequencing checks */
 
-	if (krb5_get_seq_num(kctx->seq, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8,
-				    &direction, &seqnum))
-		return GSS_S_BAD_SIG;
-
-	if ((kctx->initiate && direction != 0xff) ||
-	    (!kctx->initiate && direction != 0))
-		return GSS_S_BAD_SIG;
-
 	/* Copy the data back to the right position.  XXX: Would probably be
 	 * better to copy and encrypt at the same time. */
 
 	blocksize = crypto_blkcipher_blocksize(kctx->enc);
-	data_start = ptr + GSS_KRB5_TOK_HDR_LEN + 8 + blocksize;
+	data_start = ptr + (GSS_KRB5_TOK_HDR_LEN + kctx->gk5e->cksumlength) +
+					conflen;
 	orig_start = buf->head[0].iov_base + offset;
 	data_len = (buf->head[0].iov_base + buf->head[0].iov_len) - data_start;
 	memmove(orig_start, data_start, data_len);
@@ -299,3 +379,209 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 
 	return GSS_S_COMPLETE;
 }
+
+/*
+ * We cannot currently handle tokens with rotated data.  We need a
+ * generalized routine to rotate the data in place.  It is anticipated
+ * that we won't encounter rotated data in the general case.
+ */
+static u32
+rotate_left(struct krb5_ctx *kctx, u32 offset, struct xdr_buf *buf, u16 rrc)
+{
+	unsigned int realrrc = rrc % (buf->len - offset - GSS_KRB5_TOK_HDR_LEN);
+
+	if (realrrc == 0)
+		return 0;
+
+	dprintk("%s: cannot process token with rotated data: "
+		"rrc %u, realrrc %u\n", __func__, rrc, realrrc);
+	return 1;
+}
+
+static u32
+gss_wrap_kerberos_v2(struct krb5_ctx *kctx, u32 offset,
+		     struct xdr_buf *buf, struct page **pages)
+{
+	int		blocksize;
+	u8		*ptr, *plainhdr;
+	s32		now;
+	u8		flags = 0x00;
+	__be16		*be16ptr, ec = 0;
+	__be64		*be64ptr;
+	u32		err;
+
+	dprintk("RPC:       %s\n", __func__);
+
+	if (kctx->gk5e->encrypt_v2 == NULL)
+		return GSS_S_FAILURE;
+
+	/* make room for gss token header */
+	if (xdr_extend_head(buf, offset, GSS_KRB5_TOK_HDR_LEN))
+		return GSS_S_FAILURE;
+
+	/* construct gss token header */
+	ptr = plainhdr = buf->head[0].iov_base + offset;
+	*ptr++ = (unsigned char) ((KG2_TOK_WRAP>>8) & 0xff);
+	*ptr++ = (unsigned char) (KG2_TOK_WRAP & 0xff);
+
+	if ((kctx->flags & KRB5_CTX_FLAG_INITIATOR) == 0)
+		flags |= KG2_TOKEN_FLAG_SENTBYACCEPTOR;
+	if ((kctx->flags & KRB5_CTX_FLAG_ACCEPTOR_SUBKEY) != 0)
+		flags |= KG2_TOKEN_FLAG_ACCEPTORSUBKEY;
+	/* We always do confidentiality in wrap tokens */
+	flags |= KG2_TOKEN_FLAG_SEALED;
+
+	*ptr++ = flags;
+	*ptr++ = 0xff;
+	be16ptr = (__be16 *)ptr;
+
+	blocksize = crypto_blkcipher_blocksize(kctx->acceptor_enc);
+	*be16ptr++ = cpu_to_be16(ec);
+	/* "inner" token header always uses 0 for RRC */
+	*be16ptr++ = cpu_to_be16(0);
+
+	be64ptr = (__be64 *)be16ptr;
+	spin_lock(&krb5_seq_lock);
+	*be64ptr = cpu_to_be64(kctx->seq_send64++);
+	spin_unlock(&krb5_seq_lock);
+
+	err = (*kctx->gk5e->encrypt_v2)(kctx, offset, buf, ec, pages);
+	if (err)
+		return err;
+
+	now = get_seconds();
+	return (kctx->endtime < now) ? GSS_S_CONTEXT_EXPIRED : GSS_S_COMPLETE;
+}
+
+static u32
+gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
+{
+	s32		now;
+	u64		seqnum;
+	u8		*ptr;
+	u8		flags = 0x00;
+	u16		ec, rrc;
+	int		err;
+	u32		headskip, tailskip;
+	u8		decrypted_hdr[GSS_KRB5_TOK_HDR_LEN];
+	unsigned int	movelen;
+
+
+	dprintk("RPC:       %s\n", __func__);
+
+	if (kctx->gk5e->decrypt_v2 == NULL)
+		return GSS_S_FAILURE;
+
+	ptr = buf->head[0].iov_base + offset;
+
+	if (be16_to_cpu(*((__be16 *)ptr)) != KG2_TOK_WRAP)
+		return GSS_S_DEFECTIVE_TOKEN;
+
+	flags = ptr[2];
+	if ((!kctx->initiate && (flags & KG2_TOKEN_FLAG_SENTBYACCEPTOR)) ||
+	    (kctx->initiate && !(flags & KG2_TOKEN_FLAG_SENTBYACCEPTOR)))
+		return GSS_S_BAD_SIG;
+
+	if ((flags & KG2_TOKEN_FLAG_SEALED) == 0) {
+		dprintk("%s: token missing expected sealed flag\n", __func__);
+		return GSS_S_DEFECTIVE_TOKEN;
+	}
+
+	if (ptr[3] != 0xff)
+		return GSS_S_DEFECTIVE_TOKEN;
+
+	ec = be16_to_cpup((__be16 *)(ptr + 4));
+	rrc = be16_to_cpup((__be16 *)(ptr + 6));
+
+	seqnum = be64_to_cpup((__be64 *)(ptr + 8));
+
+	if (rrc != 0) {
+		err = rotate_left(kctx, offset, buf, rrc);
+		if (err)
+			return GSS_S_FAILURE;
+	}
+
+	err = (*kctx->gk5e->decrypt_v2)(kctx, offset, buf,
+					&headskip, &tailskip);
+	if (err)
+		return GSS_S_FAILURE;
+
+	/*
+	 * Retrieve the decrypted gss token header and verify
+	 * it against the original
+	 */
+	err = read_bytes_from_xdr_buf(buf,
+				buf->len - GSS_KRB5_TOK_HDR_LEN - tailskip,
+				decrypted_hdr, GSS_KRB5_TOK_HDR_LEN);
+	if (err) {
+		dprintk("%s: error %u getting decrypted_hdr\n", __func__, err);
+		return GSS_S_FAILURE;
+	}
+	if (memcmp(ptr, decrypted_hdr, 6)
+				|| memcmp(ptr + 8, decrypted_hdr + 8, 8)) {
+		dprintk("%s: token hdr, plaintext hdr mismatch!\n", __func__);
+		return GSS_S_FAILURE;
+	}
+
+	/* do sequencing checks */
+
+	/* it got through unscathed.  Make sure the context is unexpired */
+	now = get_seconds();
+	if (now > kctx->endtime)
+		return GSS_S_CONTEXT_EXPIRED;
+
+	/*
+	 * Move the head data back to the right position in xdr_buf.
+	 * We ignore any "ec" data since it might be in the head or
+	 * the tail, and we really don't need to deal with it.
+	 * Note that buf->head[0].iov_len may indicate the available
+	 * head buffer space rather than that actually occupied.
+	 */
+	movelen = min_t(unsigned int, buf->head[0].iov_len, buf->len);
+	movelen -= offset + GSS_KRB5_TOK_HDR_LEN + headskip;
+	BUG_ON(offset + GSS_KRB5_TOK_HDR_LEN + headskip + movelen >
+							buf->head[0].iov_len);
+	memmove(ptr, ptr + GSS_KRB5_TOK_HDR_LEN + headskip, movelen);
+	buf->head[0].iov_len -= GSS_KRB5_TOK_HDR_LEN + headskip;
+	buf->len -= GSS_KRB5_TOK_HDR_LEN + headskip;
+
+	return GSS_S_COMPLETE;
+}
+
+u32
+gss_wrap_kerberos(struct gss_ctx *gctx, int offset,
+		  struct xdr_buf *buf, struct page **pages)
+{
+	struct krb5_ctx	*kctx = gctx->internal_ctx_id;
+
+	switch (kctx->enctype) {
+	default:
+		BUG();
+	case ENCTYPE_DES_CBC_RAW:
+	case ENCTYPE_DES3_CBC_RAW:
+	case ENCTYPE_ARCFOUR_HMAC:
+		return gss_wrap_kerberos_v1(kctx, offset, buf, pages);
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		return gss_wrap_kerberos_v2(kctx, offset, buf, pages);
+	}
+}
+
+u32
+gss_unwrap_kerberos(struct gss_ctx *gctx, int offset, struct xdr_buf *buf)
+{
+	struct krb5_ctx	*kctx = gctx->internal_ctx_id;
+
+	switch (kctx->enctype) {
+	default:
+		BUG();
+	case ENCTYPE_DES_CBC_RAW:
+	case ENCTYPE_DES3_CBC_RAW:
+	case ENCTYPE_ARCFOUR_HMAC:
+		return gss_unwrap_kerberos_v1(kctx, offset, buf);
+	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
+	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
+		return gss_unwrap_kerberos_v2(kctx, offset, buf);
+	}
+}
+

net/sunrpc/auth_gss/gss_mech_switch.c

@@ -249,14 +249,15 @@ EXPORT_SYMBOL_GPL(gss_mech_put);
 int
 gss_import_sec_context(const void *input_token, size_t bufsize,
 		       struct gss_api_mech	*mech,
-		       struct gss_ctx		**ctx_id)
+		       struct gss_ctx		**ctx_id,
+		       gfp_t gfp_mask)
 {
-	if (!(*ctx_id = kzalloc(sizeof(**ctx_id), GFP_KERNEL)))
+	if (!(*ctx_id = kzalloc(sizeof(**ctx_id), gfp_mask)))
 		return -ENOMEM;
 	(*ctx_id)->mech_type = gss_mech_get(mech);
 
 	return mech->gm_ops
-		->gss_import_sec_context(input_token, bufsize, *ctx_id);
+		->gss_import_sec_context(input_token, bufsize, *ctx_id, gfp_mask);
 }
 
 /* gss_get_mic: compute a mic over message and return mic_token. */
@@ -285,6 +286,20 @@ gss_verify_mic(struct gss_ctx		*context_handle,
 				 mic_token);
 }
 
+/*
+ * This function is called from both the client and server code.
+ * Each makes guarantees about how much "slack" space is available
+ * for the underlying function in "buf"'s head and tail while
+ * performing the wrap.
+ *
+ * The client and server code allocate RPC_MAX_AUTH_SIZE extra
+ * space in both the head and tail which is available for use by
+ * the wrap function.
+ *
+ * Underlying functions should verify they do not use more than
+ * RPC_MAX_AUTH_SIZE of extra space in either the head or tail
+ * when performing the wrap.
+ */
 u32
 gss_wrap(struct gss_ctx	*ctx_id,
 	 int		offset,

net/sunrpc/auth_gss/gss_spkm3_mech.c

@@ -84,13 +84,14 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *res)
 
 static int
 gss_import_sec_context_spkm3(const void *p, size_t len,
-				struct gss_ctx *ctx_id)
+				struct gss_ctx *ctx_id,
+				gfp_t gfp_mask)
 {
 	const void *end = (const void *)((const char *)p + len);
 	struct	spkm3_ctx *ctx;
 	int	version;
 
-	if (!(ctx = kzalloc(sizeof(*ctx), GFP_NOFS)))
+	if (!(ctx = kzalloc(sizeof(*ctx), gfp_mask)))
 		goto out_err;
 
 	p = simple_get_bytes(p, end, &version, sizeof(version));

net/sunrpc/auth_gss/svcauth_gss.c

@@ -494,7 +494,7 @@ static int rsc_parse(struct cache_detail *cd,
 		len = qword_get(&mesg, buf, mlen);
 		if (len < 0)
 			goto out;
-		status = gss_import_sec_context(buf, len, gm, &rsci.mechctx);
+		status = gss_import_sec_context(buf, len, gm, &rsci.mechctx, GFP_KERNEL);
 		if (status)
 			goto out;
 
@@ -1315,6 +1315,14 @@ svcauth_gss_wrap_resp_priv(struct svc_rqst *rqstp)
 	inpages = resbuf->pages;
 	/* XXX: Would be better to write some xdr helper functions for
 	 * nfs{2,3,4}xdr.c that place the data right, instead of copying: */
+
+	/*
+	 * If there is currently tail data, make sure there is
+	 * room for the head, tail, and 2 * RPC_MAX_AUTH_SIZE in
+	 * the page, and move the current tail data such that
+	 * there is RPC_MAX_AUTH_SIZE slack space available in
+	 * both the head and tail.
+	 */
 	if (resbuf->tail[0].iov_base) {
 		BUG_ON(resbuf->tail[0].iov_base >= resbuf->head[0].iov_base
 							+ PAGE_SIZE);
@@ -1327,6 +1335,13 @@ svcauth_gss_wrap_resp_priv(struct svc_rqst *rqstp)
 			resbuf->tail[0].iov_len);
 		resbuf->tail[0].iov_base += RPC_MAX_AUTH_SIZE;
 	}
+	/*
+	 * If there is no current tail data, make sure there is
+	 * room for the head data, and 2 * RPC_MAX_AUTH_SIZE in the
+	 * allotted page, and set up tail information such that there
+	 * is RPC_MAX_AUTH_SIZE slack space available in both the
+	 * head and tail.
+	 */
 	if (resbuf->tail[0].iov_base == NULL) {
 		if (resbuf->head[0].iov_len + 2*RPC_MAX_AUTH_SIZE > PAGE_SIZE)
 			return -ENOMEM;

net/sunrpc/clnt.c

@@ -556,26 +556,16 @@ static const struct rpc_call_ops rpc_default_ops = {
  */
 struct rpc_task *rpc_run_task(const struct rpc_task_setup *task_setup_data)
 {
-	struct rpc_task *task, *ret;
+	struct rpc_task *task;
 
 	task = rpc_new_task(task_setup_data);
-	if (task == NULL) {
-		rpc_release_calldata(task_setup_data->callback_ops,
-				task_setup_data->callback_data);
-		ret = ERR_PTR(-ENOMEM);
+	if (IS_ERR(task))
 		goto out;
-	}
 
-	if (task->tk_status != 0) {
-		ret = ERR_PTR(task->tk_status);
-		rpc_put_task(task);
-		goto out;
-	}
 	atomic_inc(&task->tk_count);
 	rpc_execute(task);
-	ret = task;
 out:
-	return ret;
+	return task;
 }
 EXPORT_SYMBOL_GPL(rpc_run_task);
 
@@ -657,9 +647,8 @@ struct rpc_task *rpc_run_bc_task(struct rpc_rqst *req,
 	 * Create an rpc_task to send the data
 	 */
 	task = rpc_new_task(&task_setup_data);
-	if (!task) {
+	if (IS_ERR(task)) {
 		xprt_free_bc_request(req);
-		task = ERR_PTR(-ENOMEM);
 		goto out;
 	}
 	task->tk_rqstp = req;

net/sunrpc/sched.c

@@ -25,7 +25,6 @@
 
 #ifdef RPC_DEBUG
 #define RPCDBG_FACILITY		RPCDBG_SCHED
-#define RPC_TASK_MAGIC_ID	0xf00baa
 #endif
 
 /*
@@ -237,7 +236,6 @@ static void rpc_task_set_debuginfo(struct rpc_task *task)
 {
 	static atomic_t rpc_pid;
 
-	task->tk_magic = RPC_TASK_MAGIC_ID;
 	task->tk_pid = atomic_inc_return(&rpc_pid);
 }
 #else
@@ -360,9 +358,6 @@ static void __rpc_do_wake_up_task(struct rpc_wait_queue *queue, struct rpc_task
 	dprintk("RPC: %5u __rpc_wake_up_task (now %lu)\n",
 			task->tk_pid, jiffies);
 
-#ifdef RPC_DEBUG
-	BUG_ON(task->tk_magic != RPC_TASK_MAGIC_ID);
-#endif
 	/* Has the task been executed yet? If not, we cannot wake it up! */
 	if (!RPC_IS_ACTIVATED(task)) {
 		printk(KERN_ERR "RPC: Inactive task (%p) being woken up!\n", task);
@@ -834,7 +829,7 @@ static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *ta
 	}
 
 	/* starting timestamp */
-	task->tk_start = jiffies;
+	task->tk_start = ktime_get();
 
 	dprintk("RPC:       new task initialized, procpid %u\n",
 				task_pid_nr(current));
@@ -856,16 +851,23 @@ struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data)
 
 	if (task == NULL) {
 		task = rpc_alloc_task();
-		if (task == NULL)
-			goto out;
+		if (task == NULL) {
+			rpc_release_calldata(setup_data->callback_ops,
+					setup_data->callback_data);
+			return ERR_PTR(-ENOMEM);
+		}
 		flags = RPC_TASK_DYNAMIC;
 	}
 
 	rpc_init_task(task, setup_data);
+	if (task->tk_status < 0) {
+		int err = task->tk_status;
+		rpc_put_task(task);
+		return ERR_PTR(err);
+	}
 
 	task->tk_flags |= flags;
 	dprintk("RPC:       allocated task %p\n", task);
-out:
 	return task;
 }
 
@@ -909,9 +911,6 @@ EXPORT_SYMBOL_GPL(rpc_put_task);
 
 static void rpc_release_task(struct rpc_task *task)
 {
-#ifdef RPC_DEBUG
-	BUG_ON(task->tk_magic != RPC_TASK_MAGIC_ID);
-#endif
 	dprintk("RPC: %5u release task\n", task->tk_pid);
 
 	if (!list_empty(&task->tk_task)) {
@@ -923,9 +922,6 @@ static void rpc_release_task(struct rpc_task *task)
 	}
 	BUG_ON (RPC_IS_QUEUED(task));
 
-#ifdef RPC_DEBUG
-	task->tk_magic = 0;
-#endif
 	/* Wake up anyone who is waiting for task completion */
 	rpc_mark_complete_task(task);
 

net/sunrpc/stats.c

@@ -144,7 +144,7 @@ void rpc_count_iostats(struct rpc_task *task)
 	struct rpc_rqst *req = task->tk_rqstp;
 	struct rpc_iostats *stats;
 	struct rpc_iostats *op_metrics;
-	long rtt, execute, queue;
+	ktime_t delta;
 
 	if (!task->tk_client || !task->tk_client->cl_metrics || !req)
 		return;
@@ -156,23 +156,16 @@ void rpc_count_iostats(struct rpc_task *task)
 	op_metrics->om_ntrans += req->rq_ntrans;
 	op_metrics->om_timeouts += task->tk_timeouts;
 
-	op_metrics->om_bytes_sent += task->tk_bytes_sent;
+	op_metrics->om_bytes_sent += req->rq_xmit_bytes_sent;
 	op_metrics->om_bytes_recv += req->rq_reply_bytes_recvd;
 
-	queue = (long)req->rq_xtime - task->tk_start;
-	if (queue < 0)
-		queue = -queue;
-	op_metrics->om_queue += queue;
+	delta = ktime_sub(req->rq_xtime, task->tk_start);
+	op_metrics->om_queue = ktime_add(op_metrics->om_queue, delta);
 
-	rtt = task->tk_rtt;
-	if (rtt < 0)
-		rtt = -rtt;
-	op_metrics->om_rtt += rtt;
+	op_metrics->om_rtt = ktime_add(op_metrics->om_rtt, req->rq_rtt);
 
-	execute = (long)jiffies - task->tk_start;
-	if (execute < 0)
-		execute = -execute;
-	op_metrics->om_execute += execute;
+	delta = ktime_sub(ktime_get(), task->tk_start);
+	op_metrics->om_execute = ktime_add(op_metrics->om_execute, delta);
 }
 
 static void _print_name(struct seq_file *seq, unsigned int op,
@@ -186,8 +179,6 @@ static void _print_name(struct seq_file *seq, unsigned int op,
 		seq_printf(seq, "\t%12u: ", op);
 }
 
-#define MILLISECS_PER_JIFFY	(1000 / HZ)
-
 void rpc_print_iostats(struct seq_file *seq, struct rpc_clnt *clnt)
 {
 	struct rpc_iostats *stats = clnt->cl_metrics;
@@ -214,9 +205,9 @@ void rpc_print_iostats(struct seq_file *seq, struct rpc_clnt *clnt)
 				metrics->om_timeouts,
 				metrics->om_bytes_sent,
 				metrics->om_bytes_recv,
-				metrics->om_queue * MILLISECS_PER_JIFFY,
-				metrics->om_rtt * MILLISECS_PER_JIFFY,
-				metrics->om_execute * MILLISECS_PER_JIFFY);
+				ktime_to_ms(metrics->om_queue),
+				ktime_to_ms(metrics->om_rtt),
+				ktime_to_ms(metrics->om_execute));
 	}
 }
 EXPORT_SYMBOL_GPL(rpc_print_iostats);

net/sunrpc/xdr.c

@@ -762,6 +762,7 @@ int write_bytes_to_xdr_buf(struct xdr_buf *buf, unsigned int base, void *obj, un
 	__write_bytes_to_xdr_buf(&subbuf, obj, len);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(write_bytes_to_xdr_buf);
 
 int
 xdr_decode_word(struct xdr_buf *buf, unsigned int base, u32 *obj)

net/sunrpc/xprt.c

@@ -43,6 +43,7 @@
 #include <linux/interrupt.h>
 #include <linux/workqueue.h>
 #include <linux/net.h>
+#include <linux/ktime.h>
 
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/metrics.h>
@@ -62,7 +63,6 @@
  * Local functions
  */
 static void	xprt_request_init(struct rpc_task *, struct rpc_xprt *);
-static inline void	do_xprt_reserve(struct rpc_task *);
 static void	xprt_connect_status(struct rpc_task *task);
 static int      __xprt_get_cong(struct rpc_xprt *, struct rpc_task *);
 
@@ -711,12 +711,16 @@ void xprt_connect(struct rpc_task *task)
 		if (task->tk_rqstp)
 			task->tk_rqstp->rq_bytes_sent = 0;
 
-		task->tk_timeout = xprt->connect_timeout;
+		task->tk_timeout = task->tk_rqstp->rq_timeout;
 		rpc_sleep_on(&xprt->pending, task, xprt_connect_status);
+
+		if (test_bit(XPRT_CLOSING, &xprt->state))
+			return;
+		if (xprt_test_and_set_connecting(xprt))
+			return;
 		xprt->stat.connect_start = jiffies;
 		xprt->ops->connect(task);
 	}
-	return;
 }
 
 static void xprt_connect_status(struct rpc_task *task)
@@ -771,25 +775,19 @@ struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 }
 EXPORT_SYMBOL_GPL(xprt_lookup_rqst);
 
-/**
- * xprt_update_rtt - update an RPC client's RTT state after receiving a reply
- * @task: RPC request that recently completed
- *
- */
-void xprt_update_rtt(struct rpc_task *task)
+static void xprt_update_rtt(struct rpc_task *task)
 {
 	struct rpc_rqst *req = task->tk_rqstp;
 	struct rpc_rtt *rtt = task->tk_client->cl_rtt;
 	unsigned timer = task->tk_msg.rpc_proc->p_timer;
+	long m = usecs_to_jiffies(ktime_to_us(req->rq_rtt));
 
 	if (timer) {
 		if (req->rq_ntrans == 1)
-			rpc_update_rtt(rtt, timer,
-					(long)jiffies - req->rq_xtime);
+			rpc_update_rtt(rtt, timer, m);
 		rpc_set_timeo(rtt, timer, req->rq_ntrans - 1);
 	}
 }
-EXPORT_SYMBOL_GPL(xprt_update_rtt);
 
 /**
  * xprt_complete_rqst - called when reply processing is complete
@@ -807,7 +805,9 @@ void xprt_complete_rqst(struct rpc_task *task, int copied)
 			task->tk_pid, ntohl(req->rq_xid), copied);
 
 	xprt->stat.recvs++;
-	task->tk_rtt = (long)jiffies - req->rq_xtime;
+	req->rq_rtt = ktime_sub(ktime_get(), req->rq_xtime);
+	if (xprt->ops->timer != NULL)
+		xprt_update_rtt(task);
 
 	list_del_init(&req->rq_list);
 	req->rq_private_buf.len = copied;
@@ -906,7 +906,7 @@ void xprt_transmit(struct rpc_task *task)
 		return;
 
 	req->rq_connect_cookie = xprt->connect_cookie;
-	req->rq_xtime = jiffies;
+	req->rq_xtime = ktime_get();
 	status = xprt->ops->send_request(task);
 	if (status != 0) {
 		task->tk_status = status;
@@ -935,7 +935,7 @@ void xprt_transmit(struct rpc_task *task)
 	spin_unlock_bh(&xprt->transport_lock);
 }
 
-static inline void do_xprt_reserve(struct rpc_task *task)
+static void xprt_alloc_slot(struct rpc_task *task)
 {
 	struct rpc_xprt	*xprt = task->tk_xprt;
 
@@ -955,6 +955,16 @@ static inline void do_xprt_reserve(struct rpc_task *task)
 	rpc_sleep_on(&xprt->backlog, task, NULL);
 }
 
+static void xprt_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
+{
+	memset(req, 0, sizeof(*req));	/* mark unused */
+
+	spin_lock(&xprt->reserve_lock);
+	list_add(&req->rq_list, &xprt->free);
+	rpc_wake_up_next(&xprt->backlog);
+	spin_unlock(&xprt->reserve_lock);
+}
+
 /**
  * xprt_reserve - allocate an RPC request slot
  * @task: RPC task requesting a slot allocation
@@ -968,7 +978,7 @@ void xprt_reserve(struct rpc_task *task)
 
 	task->tk_status = -EIO;
 	spin_lock(&xprt->reserve_lock);
-	do_xprt_reserve(task);
+	xprt_alloc_slot(task);
 	spin_unlock(&xprt->reserve_lock);
 }
 
@@ -1006,14 +1016,10 @@ void xprt_release(struct rpc_task *task)
 {
 	struct rpc_xprt	*xprt;
 	struct rpc_rqst	*req;
-	int is_bc_request;
 
 	if (!(req = task->tk_rqstp))
 		return;
 
-	/* Preallocated backchannel request? */
-	is_bc_request = bc_prealloc(req);
-
 	xprt = req->rq_xprt;
 	rpc_count_iostats(task);
 	spin_lock_bh(&xprt->transport_lock);
@@ -1027,21 +1033,16 @@ void xprt_release(struct rpc_task *task)
 		mod_timer(&xprt->timer,
 				xprt->last_used + xprt->idle_timeout);
 	spin_unlock_bh(&xprt->transport_lock);
-	if (!bc_prealloc(req))
+	if (req->rq_buffer)
 		xprt->ops->buf_free(req->rq_buffer);
 	task->tk_rqstp = NULL;
 	if (req->rq_release_snd_buf)
 		req->rq_release_snd_buf(req);
 
 	dprintk("RPC: %5u release request %p\n", task->tk_pid, req);
-	if (likely(!is_bc_request)) {
-		memset(req, 0, sizeof(*req));	/* mark unused */
-
-		spin_lock(&xprt->reserve_lock);
-		list_add(&req->rq_list, &xprt->free);
-		rpc_wake_up_next(&xprt->backlog);
-		spin_unlock(&xprt->reserve_lock);
-	} else
+	if (likely(!bc_prealloc(req)))
+		xprt_free_slot(xprt, req);
+	else
 		xprt_free_bc_request(req);
 }
 

net/sunrpc/xprtrdma/transport.c

@@ -305,7 +305,6 @@ xprt_setup_rdma(struct xprt_create *args)
 	/* 60 second timeout, no retries */
 	xprt->timeout = &xprt_rdma_default_timeout;
 	xprt->bind_timeout = (60U * HZ);
-	xprt->connect_timeout = (60U * HZ);
 	xprt->reestablish_timeout = (5U * HZ);
 	xprt->idle_timeout = (5U * 60 * HZ);
 
@@ -449,21 +448,19 @@ xprt_rdma_connect(struct rpc_task *task)
 	struct rpc_xprt *xprt = (struct rpc_xprt *)task->tk_xprt;
 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 
-	if (!xprt_test_and_set_connecting(xprt)) {
-		if (r_xprt->rx_ep.rep_connected != 0) {
-			/* Reconnect */
-			schedule_delayed_work(&r_xprt->rdma_connect,
-				xprt->reestablish_timeout);
-			xprt->reestablish_timeout <<= 1;
-			if (xprt->reestablish_timeout > (30 * HZ))
-				xprt->reestablish_timeout = (30 * HZ);
-			else if (xprt->reestablish_timeout < (5 * HZ))
-				xprt->reestablish_timeout = (5 * HZ);
-		} else {
-			schedule_delayed_work(&r_xprt->rdma_connect, 0);
-			if (!RPC_IS_ASYNC(task))
-				flush_scheduled_work();
-		}
+	if (r_xprt->rx_ep.rep_connected != 0) {
+		/* Reconnect */
+		schedule_delayed_work(&r_xprt->rdma_connect,
+			xprt->reestablish_timeout);
+		xprt->reestablish_timeout <<= 1;
+		if (xprt->reestablish_timeout > (30 * HZ))
+			xprt->reestablish_timeout = (30 * HZ);
+		else if (xprt->reestablish_timeout < (5 * HZ))
+			xprt->reestablish_timeout = (5 * HZ);
+	} else {
+		schedule_delayed_work(&r_xprt->rdma_connect, 0);
+		if (!RPC_IS_ASYNC(task))
+			flush_scheduled_work();
 	}
 }
 
@@ -677,7 +674,7 @@ xprt_rdma_send_request(struct rpc_task *task)
 	if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req))
 		goto drop_connection;
 
-	task->tk_bytes_sent += rqst->rq_snd_buf.len;
+	rqst->rq_xmit_bytes_sent += rqst->rq_snd_buf.len;
 	rqst->rq_bytes_sent = 0;
 	return 0;
 

net/sunrpc/xprtsock.c

@@ -137,20 +137,6 @@ static ctl_table sunrpc_table[] = {
 
 #endif
 
-/*
- * Time out for an RPC UDP socket connect.  UDP socket connects are
- * synchronous, but we set a timeout anyway in case of resource
- * exhaustion on the local host.
- */
-#define XS_UDP_CONN_TO		(5U * HZ)
-
-/*
- * Wait duration for an RPC TCP connection to be established.  Solaris
- * NFS over TCP uses 60 seconds, for example, which is in line with how
- * long a server takes to reboot.
- */
-#define XS_TCP_CONN_TO		(60U * HZ)
-
 /*
  * Wait duration for a reply from the RPC portmapper.
  */
@@ -542,7 +528,7 @@ static int xs_udp_send_request(struct rpc_task *task)
 			xdr->len - req->rq_bytes_sent, status);
 
 	if (status >= 0) {
-		task->tk_bytes_sent += status;
+		req->rq_xmit_bytes_sent += status;
 		if (status >= req->rq_slen)
 			return 0;
 		/* Still some bytes left; set up for a retry later. */
@@ -638,7 +624,7 @@ static int xs_tcp_send_request(struct rpc_task *task)
 		/* If we've sent the entire packet, immediately
 		 * reset the count of bytes sent. */
 		req->rq_bytes_sent += status;
-		task->tk_bytes_sent += status;
+		req->rq_xmit_bytes_sent += status;
 		if (likely(req->rq_bytes_sent >= req->rq_slen)) {
 			req->rq_bytes_sent = 0;
 			return 0;
@@ -858,7 +844,6 @@ static void xs_udp_data_ready(struct sock *sk, int len)
 	dst_confirm(skb_dst(skb));
 
 	xprt_adjust_cwnd(task, copied);
-	xprt_update_rtt(task);
 	xprt_complete_rqst(task, copied);
 
  out_unlock:
@@ -2016,9 +2001,6 @@ static void xs_connect(struct rpc_task *task)
 	struct rpc_xprt *xprt = task->tk_xprt;
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
 
-	if (xprt_test_and_set_connecting(xprt))
-		return;
-
 	if (transport->sock != NULL && !RPC_IS_SOFTCONN(task)) {
 		dprintk("RPC:       xs_connect delayed xprt %p for %lu "
 				"seconds\n",
@@ -2038,16 +2020,6 @@ static void xs_connect(struct rpc_task *task)
 	}
 }
 
-static void xs_tcp_connect(struct rpc_task *task)
-{
-	struct rpc_xprt *xprt = task->tk_xprt;
-
-	/* Exit if we need to wait for socket shutdown to complete */
-	if (test_bit(XPRT_CLOSING, &xprt->state))
-		return;
-	xs_connect(task);
-}
-
 /**
  * xs_udp_print_stats - display UDP socket-specifc stats
  * @xprt: rpc_xprt struct containing statistics
@@ -2246,7 +2218,7 @@ static struct rpc_xprt_ops xs_tcp_ops = {
 	.release_xprt		= xs_tcp_release_xprt,
 	.rpcbind		= rpcb_getport_async,
 	.set_port		= xs_set_port,
-	.connect		= xs_tcp_connect,
+	.connect		= xs_connect,
 	.buf_alloc		= rpc_malloc,
 	.buf_free		= rpc_free,
 	.send_request		= xs_tcp_send_request,
@@ -2337,7 +2309,6 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)
 	xprt->max_payload = (1U << 16) - (MAX_HEADER << 3);
 
 	xprt->bind_timeout = XS_BIND_TO;
-	xprt->connect_timeout = XS_UDP_CONN_TO;
 	xprt->reestablish_timeout = XS_UDP_REEST_TO;
 	xprt->idle_timeout = XS_IDLE_DISC_TO;
 
@@ -2412,7 +2383,6 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)
 	xprt->max_payload = RPC_MAX_FRAGMENT_SIZE;
 
 	xprt->bind_timeout = XS_BIND_TO;
-	xprt->connect_timeout = XS_TCP_CONN_TO;
 	xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO;
 	xprt->idle_timeout = XS_IDLE_DISC_TO;
 
@@ -2472,9 +2442,6 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args)
 	struct sock_xprt *transport;
 	struct svc_sock *bc_sock;
 
-	if (!args->bc_xprt)
-		ERR_PTR(-EINVAL);
-
 	xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries);
 	if (IS_ERR(xprt))
 		return xprt;
@@ -2488,7 +2455,6 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args)
 	/* backchannel */
 	xprt_set_bound(xprt);
 	xprt->bind_timeout = 0;
-	xprt->connect_timeout = 0;
 	xprt->reestablish_timeout = 0;
 	xprt->idle_timeout = 0;