Merge git://git.linux-nfs.org/pub/linux/nfs-2.6

* git://git.linux-nfs.org/pub/linux/nfs-2.6: (131 commits)
  NFSv4: Fix a typo in nfs_inode_reclaim_delegation
  NFS: Add a boot parameter to disable 64 bit inode numbers
  NFS: nfs_refresh_inode should clear cache_validity flags on success
  NFS: Fix a connectathon regression in NFSv3 and NFSv4
  NFS: Use nfs_refresh_inode() in ops that aren't expected to change the inode
  SUNRPC: Don't call xprt_release in call refresh
  SUNRPC: Don't call xprt_release() if call_allocate fails
  SUNRPC: Fix buggy UDP transmission
  [23/37] Clean up duplicate includes in
  [2.6 patch] net/sunrpc/rpcb_clnt.c: make struct rpcb_program static
  SUNRPC: Use correct type in buffer length calculations
  SUNRPC: Fix default hostname created in rpc_create()
  nfs: add server port to rpc_pipe info file
  NFS: Get rid of some obsolete macros
  NFS: Simplify filehandle revalidation
  NFS: Ensure that nfs_link() returns a hashed dentry
  NFS: Be strict about dentry revalidation when doing exclusive create
  NFS: Don't zap the readdir caches upon error
  NFS: Remove the redundant nfs_reval_fsid()
  NFSv3: Always use directory post-op attributes in nfs3_proc_lookup
  ...

Fix up trivial conflict due to sock_owned_by_user() cleanup manually in
net/sunrpc/xprtsock.c

Documentation/kernel-parameters.txt

@@ -1083,6 +1083,13 @@ and is between 256 and 4096 characters. It is defined in the file
 			[NFS] set the maximum lifetime for idmapper cache
 			entries.
 
+	nfs.enable_ino64=
+			[NFS] enable 64-bit inode numbers.
+			If zero, the NFS client will fake up a 32-bit inode
+			number for the readdir() and stat() syscalls instead
+			of returning the full 64-bit number.
+			The default is to return 64-bit inode numbers.
+
 	nmi_watchdog=	[KNL,BUGS=X86-32] Debugging features for SMP kernels
 
 	no387		[BUGS=X86-32] Tells the kernel to use the 387 maths

fs/Kconfig

@@ -1755,6 +1755,14 @@ config SUNRPC
 config SUNRPC_GSS
 	tristate
 
+config SUNRPC_XPRT_RDMA
+	tristate "RDMA transport for sunrpc (EXPERIMENTAL)"
+	depends on SUNRPC && INFINIBAND && EXPERIMENTAL
+	default m
+	help
+	  Adds a client RPC transport for supporting kernel NFS over RDMA
+	  mounts, including Infiniband and iWARP. Experimental.
+
 config SUNRPC_BIND34
 	bool "Support for rpcbind versions 3 & 4 (EXPERIMENTAL)"
 	depends on SUNRPC && EXPERIMENTAL

fs/lockd/mon.c

@@ -10,6 +10,7 @@
 #include <linux/utsname.h>
 #include <linux/kernel.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/xprtsock.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/sm_inter.h>
@@ -132,7 +133,7 @@ nsm_create(void)
 		.sin_port	= 0,
 	};
 	struct rpc_create_args args = {
-		.protocol	= IPPROTO_UDP,
+		.protocol	= XPRT_TRANSPORT_UDP,
 		.address	= (struct sockaddr *)&sin,
 		.addrsize	= sizeof(sin),
 		.servername	= "localhost",

fs/lockd/xdr.c

@@ -62,8 +62,9 @@ static __be32 *nlm_decode_cookie(__be32 *p, struct nlm_cookie *c)
 	}
 	else 
 	{
-		printk(KERN_NOTICE
-			"lockd: bad cookie size %d (only cookies under %d bytes are supported.)\n", len, NLM_MAXCOOKIELEN);
+		dprintk("lockd: bad cookie size %d (only cookies under "
+			"%d bytes are supported.)\n",
+				len, NLM_MAXCOOKIELEN);
 		return NULL;
 	}
 	return p;
@@ -84,8 +85,7 @@ nlm_decode_fh(__be32 *p, struct nfs_fh *f)
 	unsigned int	len;
 
 	if ((len = ntohl(*p++)) != NFS2_FHSIZE) {
-		printk(KERN_NOTICE
-			"lockd: bad fhandle size %d (should be %d)\n",
+		dprintk("lockd: bad fhandle size %d (should be %d)\n",
 			len, NFS2_FHSIZE);
 		return NULL;
 	}

fs/lockd/xdr4.c

@@ -64,8 +64,9 @@ nlm4_decode_cookie(__be32 *p, struct nlm_cookie *c)
 	}
 	else 
 	{
-		printk(KERN_NOTICE
-			"lockd: bad cookie size %d (only cookies under %d bytes are supported.)\n", len, NLM_MAXCOOKIELEN);
+		dprintk("lockd: bad cookie size %d (only cookies under "
+			"%d bytes are supported.)\n",
+				len, NLM_MAXCOOKIELEN);
 		return NULL;
 	}
 	return p;
@@ -86,8 +87,7 @@ nlm4_decode_fh(__be32 *p, struct nfs_fh *f)
 	memset(f->data, 0, sizeof(f->data));
 	f->size = ntohl(*p++);
 	if (f->size > NFS_MAXFHSIZE) {
-		printk(KERN_NOTICE
-			"lockd: bad fhandle size %d (should be <=%d)\n",
+		dprintk("lockd: bad fhandle size %d (should be <=%d)\n",
 			f->size, NFS_MAXFHSIZE);
 		return NULL;
 	}

fs/nfs/Makefile

@@ -16,4 +16,3 @@ nfs-$(CONFIG_NFS_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o \
 			   nfs4namespace.o
 nfs-$(CONFIG_NFS_DIRECTIO) += direct.o
 nfs-$(CONFIG_SYSCTL) += sysctl.o
-nfs-objs		:= $(nfs-y)

fs/nfs/client.c

@@ -23,6 +23,8 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/metrics.h>
+#include <linux/sunrpc/xprtsock.h>
+#include <linux/sunrpc/xprtrdma.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_mount.h>
 #include <linux/nfs4_mount.h>
@@ -340,7 +342,8 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 		to->to_retries = 2;
 
 	switch (proto) {
-	case IPPROTO_TCP:
+	case XPRT_TRANSPORT_TCP:
+	case XPRT_TRANSPORT_RDMA:
 		if (!to->to_initval)
 			to->to_initval = 60 * HZ;
 		if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
@@ -349,7 +352,7 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 		to->to_maxval = to->to_initval + (to->to_increment * to->to_retries);
 		to->to_exponential = 0;
 		break;
-	case IPPROTO_UDP:
+	case XPRT_TRANSPORT_UDP:
 	default:
 		if (!to->to_initval)
 			to->to_initval = 11 * HZ / 10;
@@ -501,9 +504,9 @@ static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t
 /*
  * Initialise an NFS2 or NFS3 client
  */
-static int nfs_init_client(struct nfs_client *clp, const struct nfs_mount_data *data)
+static int nfs_init_client(struct nfs_client *clp,
+			   const struct nfs_parsed_mount_data *data)
 {
-	int proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
 	int error;
 
 	if (clp->cl_cons_state == NFS_CS_READY) {
@@ -522,8 +525,8 @@ static int nfs_init_client(struct nfs_client *clp, const struct nfs_mount_data *
 	 * Create a client RPC handle for doing FSSTAT with UNIX auth only
 	 * - RFC 2623, sec 2.3.2
 	 */
-	error = nfs_create_rpc_client(clp, proto, data->timeo, data->retrans,
-					RPC_AUTH_UNIX, 0);
+	error = nfs_create_rpc_client(clp, data->nfs_server.protocol,
+				data->timeo, data->retrans, RPC_AUTH_UNIX, 0);
 	if (error < 0)
 		goto error;
 	nfs_mark_client_ready(clp, NFS_CS_READY);
@@ -538,7 +541,8 @@ static int nfs_init_client(struct nfs_client *clp, const struct nfs_mount_data *
 /*
  * Create a version 2 or 3 client
  */
-static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_data *data)
+static int nfs_init_server(struct nfs_server *server,
+			   const struct nfs_parsed_mount_data *data)
 {
 	struct nfs_client *clp;
 	int error, nfsvers = 2;
@@ -551,7 +555,8 @@ static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_dat
 #endif
 
 	/* Allocate or find a client reference we can use */
-	clp = nfs_get_client(data->hostname, &data->addr, nfsvers);
+	clp = nfs_get_client(data->nfs_server.hostname,
+				&data->nfs_server.address, nfsvers);
 	if (IS_ERR(clp)) {
 		dprintk("<-- nfs_init_server() = error %ld\n", PTR_ERR(clp));
 		return PTR_ERR(clp);
@@ -581,7 +586,7 @@ static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_dat
 	if (error < 0)
 		goto error;
 
-	error = nfs_init_server_rpcclient(server, data->pseudoflavor);
+	error = nfs_init_server_rpcclient(server, data->auth_flavors[0]);
 	if (error < 0)
 		goto error;
 
@@ -760,7 +765,7 @@ void nfs_free_server(struct nfs_server *server)
  * Create a version 2 or 3 volume record
  * - keyed on server and FSID
  */
-struct nfs_server *nfs_create_server(const struct nfs_mount_data *data,
+struct nfs_server *nfs_create_server(const struct nfs_parsed_mount_data *data,
 				     struct nfs_fh *mntfh)
 {
 	struct nfs_server *server;
@@ -906,7 +911,7 @@ static int nfs4_set_client(struct nfs_server *server,
  * Create a version 4 volume record
  */
 static int nfs4_init_server(struct nfs_server *server,
-		const struct nfs4_mount_data *data, rpc_authflavor_t authflavour)
+		const struct nfs_parsed_mount_data *data)
 {
 	int error;
 
@@ -926,7 +931,7 @@ static int nfs4_init_server(struct nfs_server *server,
 	server->acdirmin = data->acdirmin * HZ;
 	server->acdirmax = data->acdirmax * HZ;
 
-	error = nfs_init_server_rpcclient(server, authflavour);
+	error = nfs_init_server_rpcclient(server, data->auth_flavors[0]);
 
 	/* Done */
 	dprintk("<-- nfs4_init_server() = %d\n", error);
@@ -937,12 +942,7 @@ static int nfs4_init_server(struct nfs_server *server,
  * Create a version 4 volume record
  * - keyed on server and FSID
  */
-struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data,
-				      const char *hostname,
-				      const struct sockaddr_in *addr,
-				      const char *mntpath,
-				      const char *ip_addr,
-				      rpc_authflavor_t authflavour,
+struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 				      struct nfs_fh *mntfh)
 {
 	struct nfs_fattr fattr;
@@ -956,13 +956,18 @@ struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data,
 		return ERR_PTR(-ENOMEM);
 
 	/* Get a client record */
-	error = nfs4_set_client(server, hostname, addr, ip_addr, authflavour,
-			data->proto, data->timeo, data->retrans);
+	error = nfs4_set_client(server,
+			data->nfs_server.hostname,
+			&data->nfs_server.address,
+			data->client_address,
+			data->auth_flavors[0],
+			data->nfs_server.protocol,
+			data->timeo, data->retrans);
 	if (error < 0)
 		goto error;
 
 	/* set up the general RPC client */
-	error = nfs4_init_server(server, data, authflavour);
+	error = nfs4_init_server(server, data);
 	if (error < 0)
 		goto error;
 
@@ -971,7 +976,7 @@ struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data,
 	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
 	/* Probe the root fh to retrieve its FSID */
-	error = nfs4_path_walk(server, mntfh, mntpath);
+	error = nfs4_path_walk(server, mntfh, data->nfs_server.export_path);
 	if (error < 0)
 		goto error;
 

fs/nfs/delegation.c

@@ -52,7 +52,7 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 	for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
-		if ((struct nfs_open_context *)fl->fl_file->private_data != ctx)
+		if (nfs_file_open_context(fl->fl_file) != ctx)
 			continue;
 		status = nfs4_lock_delegation_recall(state, fl);
 		if (status >= 0)
@@ -109,6 +109,7 @@ static void nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *
 void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
 {
 	struct nfs_delegation *delegation = NFS_I(inode)->delegation;
+	struct rpc_cred *oldcred;
 
 	if (delegation == NULL)
 		return;
@@ -116,11 +117,12 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st
 			sizeof(delegation->stateid.data));
 	delegation->type = res->delegation_type;
 	delegation->maxsize = res->maxsize;
-	put_rpccred(cred);
+	oldcred = delegation->cred;
 	delegation->cred = get_rpccred(cred);
 	delegation->flags &= ~NFS_DELEGATION_NEED_RECLAIM;
 	NFS_I(inode)->delegation_state = delegation->type;
 	smp_wmb();
+	put_rpccred(oldcred);
 }
 
 /*

fs/nfs/direct.c

@@ -368,7 +368,7 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size
 		return -ENOMEM;
 
 	dreq->inode = inode;
-	dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data);
+	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));
 	if (!is_sync_kiocb(iocb))
 		dreq->iocb = iocb;
 
@@ -510,7 +510,6 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode
 			nfs_direct_write_reschedule(dreq);
 			break;
 		default:
-			nfs_end_data_update(inode);
 			if (dreq->commit_data != NULL)
 				nfs_commit_free(dreq->commit_data);
 			nfs_direct_free_writedata(dreq);
@@ -533,7 +532,6 @@ static inline void nfs_alloc_commit_data(struct nfs_direct_req *dreq)
 
 static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode)
 {
-	nfs_end_data_update(inode);
 	nfs_direct_free_writedata(dreq);
 	nfs_zap_mapping(inode, inode->i_mapping);
 	nfs_direct_complete(dreq);
@@ -718,14 +716,12 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz
 		sync = FLUSH_STABLE;
 
 	dreq->inode = inode;
-	dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data);
+	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));
 	if (!is_sync_kiocb(iocb))
 		dreq->iocb = iocb;
 
 	nfs_add_stats(inode, NFSIOS_DIRECTWRITTENBYTES, count);
 
-	nfs_begin_data_update(inode);
-
 	rpc_clnt_sigmask(clnt, &oldset);
 	result = nfs_direct_write_schedule(dreq, user_addr, count, pos, sync);
 	if (!result)

fs/nfs/file.c

@@ -33,6 +33,7 @@
 #include <asm/system.h>
 
 #include "delegation.h"
+#include "internal.h"
 #include "iostat.h"
 
 #define NFSDBG_FACILITY		NFSDBG_FILE
@@ -55,6 +56,8 @@ static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl);
 static int nfs_flock(struct file *filp, int cmd, struct file_lock *fl);
 static int nfs_setlease(struct file *file, long arg, struct file_lock **fl);
 
+static struct vm_operations_struct nfs_file_vm_ops;
+
 const struct file_operations nfs_file_operations = {
 	.llseek		= nfs_file_llseek,
 	.read		= do_sync_read,
@@ -173,6 +176,31 @@ static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin)
 	return remote_llseek(filp, offset, origin);
 }
 
+/*
+ * Helper for nfs_file_flush() and nfs_fsync()
+ *
+ * Notice that it clears the NFS_CONTEXT_ERROR_WRITE before synching to
+ * disk, but it retrieves and clears ctx->error after synching, despite
+ * the two being set at the same time in nfs_context_set_write_error().
+ * This is because the former is used to notify the _next_ call to
+ * nfs_file_write() that a write error occured, and hence cause it to
+ * fall back to doing a synchronous write.
+ */
+static int nfs_do_fsync(struct nfs_open_context *ctx, struct inode *inode)
+{
+	int have_error, status;
+	int ret = 0;
+
+	have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
+	status = nfs_wb_all(inode);
+	have_error |= test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
+	if (have_error)
+		ret = xchg(&ctx->error, 0);
+	if (!ret)
+		ret = status;
+	return ret;
+}
+
 /*
  * Flush all dirty pages, and check for write errors.
  *
@@ -180,7 +208,7 @@ static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin)
 static int
 nfs_file_flush(struct file *file, fl_owner_t id)
 {
-	struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data;
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode	*inode = file->f_path.dentry->d_inode;
 	int		status;
 
@@ -189,16 +217,11 @@ nfs_file_flush(struct file *file, fl_owner_t id)
 	if ((file->f_mode & FMODE_WRITE) == 0)
 		return 0;
 	nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
-	lock_kernel();
+
 	/* Ensure that data+attribute caches are up to date after close() */
-	status = nfs_wb_all(inode);
-	if (!status) {
-		status = ctx->error;
-		ctx->error = 0;
+	status = nfs_do_fsync(ctx, inode);
 	if (!status)
 		nfs_revalidate_inode(NFS_SERVER(inode), inode);
-	}
-	unlock_kernel();
 	return status;
 }
 
@@ -257,8 +280,11 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
 	status = nfs_revalidate_mapping(inode, file->f_mapping);
-	if (!status)
-		status = generic_file_mmap(file, vma);
+	if (!status) {
+		vma->vm_ops = &nfs_file_vm_ops;
+		vma->vm_flags |= VM_CAN_NONLINEAR;
+		file_accessed(file);
+	}
 	return status;
 }
 
@@ -270,21 +296,13 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
 static int
 nfs_fsync(struct file *file, struct dentry *dentry, int datasync)
 {
-	struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data;
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode *inode = dentry->d_inode;
-	int status;
 
 	dfprintk(VFS, "nfs: fsync(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino);
 
 	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
-	lock_kernel();
-	status = nfs_wb_all(inode);
-	if (!status) {
-		status = ctx->error;
-		ctx->error = 0;
-	}
-	unlock_kernel();
-	return status;
+	return nfs_do_fsync(ctx, inode);
 }
 
 /*
@@ -333,7 +351,7 @@ static int nfs_launder_page(struct page *page)
 const struct address_space_operations nfs_file_aops = {
 	.readpage = nfs_readpage,
 	.readpages = nfs_readpages,
-	.set_page_dirty = nfs_set_page_dirty,
+	.set_page_dirty = __set_page_dirty_nobuffers,
 	.writepage = nfs_writepage,
 	.writepages = nfs_writepages,
 	.prepare_write = nfs_prepare_write,
@@ -346,6 +364,43 @@ const struct address_space_operations nfs_file_aops = {
 	.launder_page = nfs_launder_page,
 };
 
+static int nfs_vm_page_mkwrite(struct vm_area_struct *vma, struct page *page)
+{
+	struct file *filp = vma->vm_file;
+	unsigned pagelen;
+	int ret = -EINVAL;
+
+	lock_page(page);
+	if (page->mapping != vma->vm_file->f_path.dentry->d_inode->i_mapping)
+		goto out_unlock;
+	pagelen = nfs_page_length(page);
+	if (pagelen == 0)
+		goto out_unlock;
+	ret = nfs_prepare_write(filp, page, 0, pagelen);
+	if (!ret)
+		ret = nfs_commit_write(filp, page, 0, pagelen);
+out_unlock:
+	unlock_page(page);
+	return ret;
+}
+
+static struct vm_operations_struct nfs_file_vm_ops = {
+	.fault = filemap_fault,
+	.page_mkwrite = nfs_vm_page_mkwrite,
+};
+
+static int nfs_need_sync_write(struct file *filp, struct inode *inode)
+{
+	struct nfs_open_context *ctx;
+
+	if (IS_SYNC(inode) || (filp->f_flags & O_SYNC))
+		return 1;
+	ctx = nfs_file_open_context(filp);
+	if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags))
+		return 1;
+	return 0;
+}
+
 static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos)
 {
@@ -382,8 +437,8 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
 	result = generic_file_aio_write(iocb, iov, nr_segs, pos);
 	/* Return error values for O_SYNC and IS_SYNC() */
-	if (result >= 0 && (IS_SYNC(inode) || (iocb->ki_filp->f_flags & O_SYNC))) {
-		int err = nfs_fsync(iocb->ki_filp, dentry, 1);
+	if (result >= 0 && nfs_need_sync_write(iocb->ki_filp, inode)) {
+		int err = nfs_do_fsync(nfs_file_open_context(iocb->ki_filp), inode);
 		if (err < 0)
 			result = err;
 	}

fs/nfs/internal.h

@@ -5,8 +5,6 @@
 #include <linux/mount.h>
 
 struct nfs_string;
-struct nfs_mount_data;
-struct nfs4_mount_data;
 
 /* Maximum number of readahead requests
  * FIXME: this should really be a sysctl so that users may tune it to suit
@@ -27,19 +25,49 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
+/*
+ * In-kernel mount arguments
+ */
+struct nfs_parsed_mount_data {
+	int			flags;
+	int			rsize, wsize;
+	int			timeo, retrans;
+	int			acregmin, acregmax,
+				acdirmin, acdirmax;
+	int			namlen;
+	unsigned int		bsize;
+	unsigned int		auth_flavor_len;
+	rpc_authflavor_t	auth_flavors[1];
+	char			*client_address;
+
+	struct {
+		struct sockaddr_in	address;
+		char			*hostname;
+		unsigned int		program;
+		unsigned int		version;
+		unsigned short		port;
+		int			protocol;
+	} mount_server;
+
+	struct {
+		struct sockaddr_in	address;
+		char			*hostname;
+		char			*export_path;
+		unsigned int		program;
+		int			protocol;
+	} nfs_server;
+};
+
 /* client.c */
 extern struct rpc_program nfs_program;
 
 extern void nfs_put_client(struct nfs_client *);
 extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int);
-extern struct nfs_server *nfs_create_server(const struct nfs_mount_data *,
+extern struct nfs_server *nfs_create_server(
+					const struct nfs_parsed_mount_data *,
 					struct nfs_fh *);
-extern struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *,
-					     const char *,
-					     const struct sockaddr_in *,
-					     const char *,
-					     const char *,
-					     rpc_authflavor_t,
+extern struct nfs_server *nfs4_create_server(
+					const struct nfs_parsed_mount_data *,
 					struct nfs_fh *);
 extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *,
 						      struct nfs_fh *);

fs/nfs/nfs2xdr.c

@@ -251,6 +251,7 @@ nfs_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args)
 	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS_readres_sz) << 2;
 	xdr_inline_pages(&req->rq_rcv_buf, replen,
 			 args->pages, args->pgbase, count);
+	req->rq_rcv_buf.flags |= XDRBUF_READ;
 	return 0;
 }
 
@@ -271,7 +272,7 @@ nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res)
 	res->eof = 0;
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READ reply header overflowed:"
+		dprintk("NFS: READ reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
 		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
@@ -281,7 +282,7 @@ nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res)
 
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (count > recvd) {
-		printk(KERN_WARNING "NFS: server cheating in read reply: "
+		dprintk("NFS: server cheating in read reply: "
 			"count %d > recvd %d\n", count, recvd);
 		count = recvd;
 	}
@@ -313,6 +314,7 @@ nfs_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args)
 
 	/* Copy the page array */
 	xdr_encode_pages(sndbuf, args->pages, args->pgbase, count);
+	sndbuf->flags |= XDRBUF_WRITE;
 	return 0;
 }
 
@@ -431,7 +433,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy)
 
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READDIR reply header overflowed:"
+		dprintk("NFS: READDIR reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
 		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
@@ -454,7 +456,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy)
 		len = ntohl(*p++);
 		p += XDR_QUADLEN(len) + 1;	/* name plus cookie */
 		if (len > NFS2_MAXNAMLEN) {
-			printk(KERN_WARNING "NFS: giant filename in readdir (len 0x%x)!\n",
+			dprintk("NFS: giant filename in readdir (len 0x%x)!\n",
 						len);
 			goto err_unmap;
 		}
@@ -471,7 +473,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy)
 	entry[0] = entry[1] = 0;
 	/* truncate listing ? */
 	if (!nr) {
-		printk(KERN_NOTICE "NFS: readdir reply truncated!\n");
+		dprintk("NFS: readdir reply truncated!\n");
 		entry[1] = 1;
 	}
 	goto out;
@@ -583,12 +585,12 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy)
 	/* Convert length of symlink */
 	len = ntohl(*p++);
 	if (len >= rcvbuf->page_len || len <= 0) {
-		dprintk(KERN_WARNING "nfs: server returned giant symlink!\n");
+		dprintk("nfs: server returned giant symlink!\n");
 		return -ENAMETOOLONG;
 	}
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READLINK reply header overflowed:"
+		dprintk("NFS: READLINK reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
 		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
@@ -597,7 +599,7 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy)
 	}
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (recvd < len) {
-		printk(KERN_WARNING "NFS: server cheating in readlink reply: "
+		dprintk("NFS: server cheating in readlink reply: "
 				"count %u > recvd %u\n", len, recvd);
 		return -EIO;
 	}
@@ -695,7 +697,7 @@ nfs_stat_to_errno(int stat)
 		if (nfs_errtbl[i].stat == stat)
 			return nfs_errtbl[i].errno;
 	}
-	printk(KERN_ERR "nfs_stat_to_errno: bad nfs status return value: %d\n", stat);
+	dprintk("nfs_stat_to_errno: bad nfs status return value: %d\n", stat);
 	return nfs_errtbl[i].errno;
 }
 

fs/nfs/nfs3acl.c

@@ -317,13 +317,11 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 	}
 
 	dprintk("NFS call setacl\n");
-	nfs_begin_data_update(inode);
 	msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_SETACL];
 	status = rpc_call_sync(server->client_acl, &msg, 0);
 	spin_lock(&inode->i_lock);
 	NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS;
 	spin_unlock(&inode->i_lock);
-	nfs_end_data_update(inode);
 	dprintk("NFS reply setacl: %d\n", status);
 
 	/* pages may have been allocated at the xdr layer. */

fs/nfs/nfs3proc.c

@@ -166,6 +166,7 @@ nfs3_proc_lookup(struct inode *dir, struct qstr *name,
 	nfs_fattr_init(&dir_attr);
 	nfs_fattr_init(fattr);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_refresh_inode(dir, &dir_attr);
 	if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) {
 		msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR];
 		msg.rpc_argp = fhandle;
@@ -173,8 +174,6 @@ nfs3_proc_lookup(struct inode *dir, struct qstr *name,
 		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	}
 	dprintk("NFS reply lookup: %d\n", status);
-	if (status >= 0)
-		status = nfs_refresh_inode(dir, &dir_attr);
 	return status;
 }
 
@@ -607,6 +606,9 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 
 	nfs_fattr_init(&dir_attr);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+
+	nfs_invalidate_atime(dir);
+
 	nfs_refresh_inode(dir, &dir_attr);
 	dprintk("NFS reply readdir: %d\n", status);
 	return status;
@@ -724,8 +726,8 @@ static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data)
 {
 	if (nfs3_async_handle_jukebox(task, data->inode))
 		return -EAGAIN;
-	/* Call back common NFS readpage processing */
-	if (task->tk_status >= 0)
+
+	nfs_invalidate_atime(data->inode);
 	nfs_refresh_inode(data->inode, &data->fattr);
 	return 0;
 }
@@ -747,7 +749,7 @@ static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data)
 	if (nfs3_async_handle_jukebox(task, data->inode))
 		return -EAGAIN;
 	if (task->tk_status >= 0)
-		nfs_post_op_update_inode(data->inode, data->res.fattr);
+		nfs_post_op_update_inode_force_wcc(data->inode, data->res.fattr);
 	return 0;
 }
 
@@ -775,8 +777,7 @@ static int nfs3_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	if (nfs3_async_handle_jukebox(task, data->inode))
 		return -EAGAIN;
-	if (task->tk_status >= 0)
-		nfs_post_op_update_inode(data->inode, data->res.fattr);
+	nfs_refresh_inode(data->inode, data->res.fattr);
 	return 0;
 }
 

fs/nfs/nfs3xdr.c

@@ -346,6 +346,7 @@ nfs3_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args)
 	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS3_readres_sz) << 2;
 	xdr_inline_pages(&req->rq_rcv_buf, replen,
 			 args->pages, args->pgbase, count);
+	req->rq_rcv_buf.flags |= XDRBUF_READ;
 	return 0;
 }
 
@@ -367,6 +368,7 @@ nfs3_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args)
 
 	/* Copy the page array */
 	xdr_encode_pages(sndbuf, args->pages, args->pgbase, count);
+	sndbuf->flags |= XDRBUF_WRITE;
 	return 0;
 }
 
@@ -524,7 +526,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res
 
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READDIR reply header overflowed:"
+		dprintk("NFS: READDIR reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
 		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
@@ -547,7 +549,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res
 		len = ntohl(*p++);		/* string length */
 		p += XDR_QUADLEN(len) + 2;	/* name + cookie */
 		if (len > NFS3_MAXNAMLEN) {
-			printk(KERN_WARNING "NFS: giant filename in readdir (len %x)!\n",
+			dprintk("NFS: giant filename in readdir (len %x)!\n",
 						len);
 			goto err_unmap;
 		}
@@ -567,7 +569,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res
 					goto short_pkt;
 				len = ntohl(*p++);
 				if (len > NFS3_FHSIZE) {
-					printk(KERN_WARNING "NFS: giant filehandle in "
+					dprintk("NFS: giant filehandle in "
 						"readdir (len %x)!\n", len);
 					goto err_unmap;
 				}
@@ -588,7 +590,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res
 	entry[0] = entry[1] = 0;
 	/* truncate listing ? */
 	if (!nr) {
-		printk(KERN_NOTICE "NFS: readdir reply truncated!\n");
+		dprintk("NFS: readdir reply truncated!\n");
 		entry[1] = 1;
 	}
 	goto out;
@@ -826,22 +828,23 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr)
 	/* Convert length of symlink */
 	len = ntohl(*p++);
 	if (len >= rcvbuf->page_len || len <= 0) {
-		dprintk(KERN_WARNING "nfs: server returned giant symlink!\n");
+		dprintk("nfs: server returned giant symlink!\n");
 		return -ENAMETOOLONG;
 	}
 
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READLINK reply header overflowed:"
+		dprintk("NFS: READLINK reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
 		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
-		dprintk("NFS: READLINK header is short. iovec will be shifted.\n");
+		dprintk("NFS: READLINK header is short. "
+			"iovec will be shifted.\n");
 		xdr_shift_buf(rcvbuf, iov->iov_len - hdrlen);
 	}
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (recvd < len) {
-		printk(KERN_WARNING "NFS: server cheating in readlink reply: "
+		dprintk("NFS: server cheating in readlink reply: "
 				"count %u > recvd %u\n", len, recvd);
 		return -EIO;
 	}
@@ -876,13 +879,13 @@ nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res)
 	ocount   = ntohl(*p++);
 
 	if (ocount != count) {
-		printk(KERN_WARNING "NFS: READ count doesn't match RPC opaque count.\n");
+		dprintk("NFS: READ count doesn't match RPC opaque count.\n");
 		return -errno_NFSERR_IO;
 	}
 
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	if (iov->iov_len < hdrlen) {
-		printk(KERN_WARNING "NFS: READ reply header overflowed:"
+		dprintk("NFS: READ reply header overflowed:"
 				"length %d > %Zu\n", hdrlen, iov->iov_len);
        		return -errno_NFSERR_IO;
 	} else if (iov->iov_len != hdrlen) {
@@ -892,7 +895,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res)
 
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (count > recvd) {
-		printk(KERN_WARNING "NFS: server cheating in read reply: "
+		dprintk("NFS: server cheating in read reply: "
 			"count %d > recvd %d\n", count, recvd);
 		count = recvd;
 		res->eof = 0;

fs/nfs/nfs4proc.c

@@ -62,10 +62,8 @@ struct nfs4_opendata;
 static int _nfs4_proc_open(struct nfs4_opendata *data);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *);
-static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry);
 static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
 static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp);
-static int _nfs4_do_access(struct inode *inode, struct rpc_cred *cred, int openflags);
 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);
 
@@ -177,7 +175,7 @@ static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dent
 		*p++ = xdr_one;                         /* bitmap length */
 		*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
 		*p++ = htonl(8);              /* attribute buffer length */
-		p = xdr_encode_hyper(p, dentry->d_inode->i_ino);
+		p = xdr_encode_hyper(p, NFS_FILEID(dentry->d_inode));
 	}
 	
 	*p++ = xdr_one;                                  /* next */
@@ -189,7 +187,7 @@ static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dent
 	*p++ = xdr_one;                         /* bitmap length */
 	*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
 	*p++ = htonl(8);              /* attribute buffer length */
-	p = xdr_encode_hyper(p, dentry->d_parent->d_inode->i_ino);
+	p = xdr_encode_hyper(p, NFS_FILEID(dentry->d_parent->d_inode));
 
 	readdir->pgbase = (char *)p - (char *)start;
 	readdir->count -= readdir->pgbase;
@@ -211,7 +209,8 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 
 	spin_lock(&dir->i_lock);
 	nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA;
-	if (cinfo->before == nfsi->change_attr && cinfo->atomic)
+	if (!cinfo->atomic || cinfo->before != nfsi->change_attr)
+		nfsi->cache_change_attribute = jiffies;
 	nfsi->change_attr = cinfo->after;
 	spin_unlock(&dir->i_lock);
 }
@@ -454,7 +453,7 @@ static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata)
 		memcpy(stateid.data, delegation->stateid.data, sizeof(stateid.data));
 		rcu_read_unlock();
 		lock_kernel();
-		ret = _nfs4_do_access(state->inode, state->owner->so_cred, open_mode);
+		ret = nfs_may_open(state->inode, state->owner->so_cred, open_mode);
 		unlock_kernel();
 		if (ret != 0)
 			goto out;
@@ -948,36 +947,6 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 	return 0;
 }
 
-static int _nfs4_do_access(struct inode *inode, struct rpc_cred *cred, int openflags)
-{
-	struct nfs_access_entry cache;
-	int mask = 0;
-	int status;
-
-	if (openflags & FMODE_READ)
-		mask |= MAY_READ;
-	if (openflags & FMODE_WRITE)
-		mask |= MAY_WRITE;
-	if (openflags & FMODE_EXEC)
-		mask |= MAY_EXEC;
-	status = nfs_access_get_cached(inode, cred, &cache);
-	if (status == 0)
-		goto out;
-
-	/* Be clever: ask server to check for all possible rights */
-	cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ;
-	cache.cred = cred;
-	cache.jiffies = jiffies;
-	status = _nfs4_proc_access(inode, &cache);
-	if (status != 0)
-		return status;
-	nfs_access_add_cache(inode, &cache);
-out:
-	if ((cache.mask & mask) == mask)
-		return 0;
-	return -EACCES;
-}
-
 static int nfs4_recover_expired_lease(struct nfs_server *server)
 {
 	struct nfs_client *clp = server->nfs_client;
@@ -1381,7 +1350,7 @@ static int nfs4_intent_set_file(struct nameidata *nd, struct path *path, struct
 
 	/* If the open_intent is for execute, we have an extra check to make */
 	if (nd->intent.open.flags & FMODE_EXEC) {
-		ret = _nfs4_do_access(state->inode,
+		ret = nfs_may_open(state->inode,
 				state->owner->so_cred,
 				nd->intent.open.flags);
 		if (ret < 0)
@@ -1390,7 +1359,7 @@ static int nfs4_intent_set_file(struct nameidata *nd, struct path *path, struct
 	filp = lookup_instantiate_filp(nd, path->dentry, NULL);
 	if (!IS_ERR(filp)) {
 		struct nfs_open_context *ctx;
-		ctx = (struct nfs_open_context *)filp->private_data;
+		ctx = nfs_file_open_context(filp);
 		ctx->state = state;
 		return 0;
 	}
@@ -1428,13 +1397,16 @@ nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	state = nfs4_do_open(dir, &path, nd->intent.open.flags, &attr, cred);
 	put_rpccred(cred);
 	if (IS_ERR(state)) {
-		if (PTR_ERR(state) == -ENOENT)
+		if (PTR_ERR(state) == -ENOENT) {
 			d_add(dentry, NULL);
+			nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+		}
 		return (struct dentry *)state;
 	}
 	res = d_add_unique(dentry, igrab(state->inode));
 	if (res != NULL)
 		path.dentry = res;
+	nfs_set_verifier(path.dentry, nfs_save_change_attribute(dir));
 	nfs4_intent_set_file(nd, &path, state);
 	return res;
 }
@@ -1468,6 +1440,7 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, st
 		}
 	}
 	if (state->inode == dentry->d_inode) {
+		nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 		nfs4_intent_set_file(nd, &path, state);
 		return 1;
 	}
@@ -1757,10 +1730,16 @@ 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 nfs4_accessres res = { 0 };
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ACCESS],
 		.rpc_argp = &args,
@@ -1786,6 +1765,7 @@ 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);
 	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
 	if (!status) {
 		entry->mask = 0;
@@ -1795,6 +1775,7 @@ 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);
 	}
 	return status;
 }
@@ -1900,11 +1881,13 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 	}
 	state = nfs4_do_open(dir, &path, flags, sattr, cred);
 	put_rpccred(cred);
+	d_drop(dentry);
 	if (IS_ERR(state)) {
 		status = PTR_ERR(state);
 		goto out;
 	}
-	d_instantiate(dentry, igrab(state->inode));
+	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, &fattr, sattr, state);
@@ -2218,6 +2201,9 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	if (status == 0)
 		memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE);
+
+	nfs_invalidate_atime(dir);
+
 	dprintk("%s: returns %d\n", __FUNCTION__, status);
 	return status;
 }
@@ -2414,6 +2400,8 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 		rpc_restart_call(task);
 		return -EAGAIN;
 	}
+
+	nfs_invalidate_atime(data->inode);
 	if (task->tk_status > 0)
 		renew_lease(server, data->timestamp);
 	return 0;
@@ -2443,7 +2431,7 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 	}
 	if (task->tk_status >= 0) {
 		renew_lease(NFS_SERVER(inode), data->timestamp);
-		nfs_post_op_update_inode(inode, data->res.fattr);
+		nfs_post_op_update_inode_force_wcc(inode, data->res.fattr);
 	}
 	return 0;
 }
@@ -2485,8 +2473,7 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 		rpc_restart_call(task);
 		return -EAGAIN;
 	}
-	if (task->tk_status >= 0)
-		nfs_post_op_update_inode(inode, data->res.fattr);
+	nfs_refresh_inode(inode, data->res.fattr);
 	return 0;
 }
 
@@ -3056,7 +3043,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	if (status == 0) {
 		status = data->rpc_status;
 		if (status == 0)
-			nfs_post_op_update_inode(inode, &data->fattr);
+			nfs_refresh_inode(inode, &data->fattr);
 	}
 	rpc_put_task(task);
 	return status;
@@ -3303,7 +3290,7 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *
 	status = -ENOMEM;
 	if (seqid == NULL)
 		goto out;
-	task = nfs4_do_unlck(request, request->fl_file->private_data, lsp, seqid);
+	task = nfs4_do_unlck(request, nfs_file_open_context(request->fl_file), lsp, seqid);
 	status = PTR_ERR(task);
 	if (IS_ERR(task))
 		goto out;
@@ -3447,7 +3434,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 	int ret;
 
 	dprintk("%s: begin!\n", __FUNCTION__);
-	data = nfs4_alloc_lockdata(fl, fl->fl_file->private_data,
+	data = nfs4_alloc_lockdata(fl, nfs_file_open_context(fl->fl_file),
 			fl->fl_u.nfs4_fl.owner);
 	if (data == NULL)
 		return -ENOMEM;
@@ -3573,7 +3560,7 @@ nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request)
 	int status;
 
 	/* verify open state */
-	ctx = (struct nfs_open_context *)filp->private_data;
+	ctx = nfs_file_open_context(filp);
 	state = ctx->state;
 
 	if (request->fl_start < 0 || request->fl_end < 0)

fs/nfs/nfs4state.c

@@ -774,7 +774,7 @@ static int nfs4_reclaim_locks(struct nfs4_state_recovery_ops *ops, struct nfs4_s
 	for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
-		if (((struct nfs_open_context *)fl->fl_file->private_data)->state != state)
+		if (nfs_file_open_context(fl->fl_file)->state != state)
 			continue;
 		status = ops->recover_lock(state, fl);
 		if (status >= 0)

fs/nfs/nfs4xdr.c

@@ -376,10 +376,12 @@ static int nfs4_stat_to_errno(int);
 				decode_locku_maxsz)
 #define NFS4_enc_access_sz	(compound_encode_hdr_maxsz + \
 				encode_putfh_maxsz + \
-				encode_access_maxsz)
+				encode_access_maxsz + \
+				encode_getattr_maxsz)
 #define NFS4_dec_access_sz	(compound_decode_hdr_maxsz + \
 				decode_putfh_maxsz + \
-				decode_access_maxsz)
+				decode_access_maxsz + \
+				decode_getattr_maxsz)
 #define NFS4_enc_getattr_sz	(compound_encode_hdr_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
@@ -562,7 +564,6 @@ struct compound_hdr {
 
 #define RESERVE_SPACE(nbytes)	do {				\
 	p = xdr_reserve_space(xdr, nbytes);			\
-	if (!p) printk("RESERVE_SPACE(%d) failed in function %s\n", (int) (nbytes), __FUNCTION__); \
 	BUG_ON(!p);						\
 } while (0)
 
@@ -628,7 +629,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 	if (iap->ia_valid & ATTR_UID) {
 		owner_namelen = nfs_map_uid_to_name(server->nfs_client, iap->ia_uid, owner_name);
 		if (owner_namelen < 0) {
-			printk(KERN_WARNING "nfs: couldn't resolve uid %d to string\n",
+			dprintk("nfs: couldn't resolve uid %d to string\n",
 					iap->ia_uid);
 			/* XXX */
 			strcpy(owner_name, "nobody");
@@ -640,7 +641,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 	if (iap->ia_valid & ATTR_GID) {
 		owner_grouplen = nfs_map_gid_to_group(server->nfs_client, iap->ia_gid, owner_group);
 		if (owner_grouplen < 0) {
-			printk(KERN_WARNING "nfs4: couldn't resolve gid %d to string\n",
+			dprintk("nfs: couldn't resolve gid %d to string\n",
 					iap->ia_gid);
 			strcpy(owner_group, "nobody");
 			owner_grouplen = sizeof("nobody") - 1;
@@ -711,7 +712,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 	 * Now we backfill the bitmap and the attribute buffer length.
 	 */
 	if (len != ((char *)p - (char *)q) + 4) {
-		printk ("encode_attr: Attr length calculation error! %u != %Zu\n",
+		printk(KERN_ERR "nfs: Attr length error, %u != %Zu\n",
 				len, ((char *)p - (char *)q) + 4);
 		BUG();
 	}
@@ -1376,14 +1377,20 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 2,
+		.nops = 3,
 	};
 	int status;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, &hdr);
-	if ((status = encode_putfh(&xdr, args->fh)) == 0)
+	status = encode_putfh(&xdr, args->fh);
+	if (status != 0)
+		goto out;
 	status = encode_access(&xdr, args->access);
+	if (status != 0)
+		goto out;
+	status = encode_getfattr(&xdr, args->bitmask);
+out:
 	return status;
 }
 
@@ -1857,6 +1864,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_read_sz) << 2;
 	xdr_inline_pages(&req->rq_rcv_buf, replen,
 			 args->pages, args->pgbase, args->count);
+	req->rq_rcv_buf.flags |= XDRBUF_READ;
 out:
 	return status;
 }
@@ -1933,6 +1941,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 	status = encode_write(&xdr, args);
 	if (status)
 		goto out;
+	req->rq_snd_buf.flags |= XDRBUF_WRITE;
 	status = encode_getfattr(&xdr, args->bitmask);
 out:
 	return status;
@@ -2180,9 +2189,9 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 #define READ_BUF(nbytes)  do { \
 	p = xdr_inline_decode(xdr, nbytes); \
 	if (unlikely(!p)) { \
-		printk(KERN_INFO "%s: prematurely hit end of receive" \
+		dprintk("nfs: %s: prematurely hit end of receive" \
 				" buffer\n", __FUNCTION__); \
-		printk(KERN_INFO "%s: xdr->p=%p, bytes=%u, xdr->end=%p\n", \
+		dprintk("nfs: %s: xdr->p=%p, bytes=%u, xdr->end=%p\n", \
 				__FUNCTION__, xdr->p, nbytes, xdr->end); \
 		return -EIO; \
 	} \
@@ -2223,8 +2232,7 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected)
 	READ_BUF(8);
 	READ32(opnum);
 	if (opnum != expected) {
-		printk(KERN_NOTICE
-				"nfs4_decode_op_hdr: Server returned operation"
+		dprintk("nfs: Server returned operation"
 			" %d but we issued a request for %d\n",
 				opnum, expected);
 		return -EIO;
@@ -2758,7 +2766,7 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf
 				dprintk("%s: nfs_map_name_to_uid failed!\n",
 						__FUNCTION__);
 		} else
-			printk(KERN_WARNING "%s: name too long (%u)!\n",
+			dprintk("%s: name too long (%u)!\n",
 					__FUNCTION__, len);
 		bitmap[1] &= ~FATTR4_WORD1_OWNER;
 	}
@@ -2783,7 +2791,7 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nf
 				dprintk("%s: nfs_map_group_to_gid failed!\n",
 						__FUNCTION__);
 		} else
-			printk(KERN_WARNING "%s: name too long (%u)!\n",
+			dprintk("%s: name too long (%u)!\n",
 					__FUNCTION__, len);
 		bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP;
 	}
@@ -2950,7 +2958,8 @@ static int verify_attr_len(struct xdr_stream *xdr, __be32 *savep, uint32_t attrl
 	unsigned int nwords = xdr->p - savep;
 
 	if (unlikely(attrwords != nwords)) {
-		printk(KERN_WARNING "%s: server returned incorrect attribute length: %u %c %u\n",
+		dprintk("%s: server returned incorrect attribute length: "
+			"%u %c %u\n",
 				__FUNCTION__,
 				attrwords << 2,
 				(attrwords < nwords) ? '<' : '>',
@@ -3451,7 +3460,7 @@ static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs_
 	hdrlen = (u8 *) p - (u8 *) iov->iov_base;
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (count > recvd) {
-		printk(KERN_WARNING "NFS: server cheating in read reply: "
+		dprintk("NFS: server cheating in read reply: "
 				"count %u > recvd %u\n", count, recvd);
 		count = recvd;
 		eof = 0;
@@ -3500,7 +3509,8 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n
 		p += 2;			/* cookie */
 		len = ntohl(*p++);	/* filename length */
 		if (len > NFS4_MAXNAMLEN) {
-			printk(KERN_WARNING "NFS: giant filename in readdir (len 0x%x)\n", len);
+			dprintk("NFS: giant filename in readdir (len 0x%x)\n",
+					len);
 			goto err_unmap;
 		}
 		xlen = XDR_QUADLEN(len);
@@ -3528,7 +3538,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n
 	entry[0] = entry[1] = 0;
 	/* truncate listing ? */
 	if (!nr) {
-		printk(KERN_NOTICE "NFS: readdir reply truncated!\n");
+		dprintk("NFS: readdir reply truncated!\n");
 		entry[1] = 1;
 	}
 	goto out;
@@ -3554,13 +3564,13 @@ static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req)
 	READ_BUF(4);
 	READ32(len);
 	if (len >= rcvbuf->page_len || len <= 0) {
-		dprintk(KERN_WARNING "nfs: server returned giant symlink!\n");
+		dprintk("nfs: server returned giant symlink!\n");
 		return -ENAMETOOLONG;
 	}
 	hdrlen = (char *) xdr->p - (char *) iov->iov_base;
 	recvd = req->rq_rcv_buf.len - hdrlen;
 	if (recvd < len) {
-		printk(KERN_WARNING "NFS: server cheating in readlink reply: "
+		dprintk("NFS: server cheating in readlink reply: "
 				"count %u > recvd %u\n", len, recvd);
 		return -EIO;
 	}
@@ -3643,7 +3653,7 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req,
 		hdrlen = (u8 *)xdr->p - (u8 *)iov->iov_base;
 		recvd = req->rq_rcv_buf.len - hdrlen;
 		if (attrlen > recvd) {
-			printk(KERN_WARNING "NFS: server cheating in getattr"
+			dprintk("NFS: server cheating in getattr"
 					" acl reply: attrlen %u > recvd %u\n",
 					attrlen, recvd);
 			return -EINVAL;
@@ -3688,8 +3698,7 @@ static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp)
 	READ_BUF(8);
 	READ32(opnum);
 	if (opnum != OP_SETCLIENTID) {
-		printk(KERN_NOTICE
-				"nfs4_decode_setclientid: Server returned operation"
+		dprintk("nfs: decode_setclientid: Server returned operation"
 			       	" %d\n", opnum);
 		return -EIO;
 	}
@@ -3783,8 +3792,13 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_ac
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
 		goto out;
-	if ((status = decode_putfh(&xdr)) == 0)
+	status = decode_putfh(&xdr);
+	if (status != 0)
+		goto out;
 	status = decode_access(&xdr, res);
+	if (status != 0)
+		goto out;
+	decode_getfattr(&xdr, res->fattr, res->server);
 out:
 	return status;
 }

fs/nfs/nfsroot.c

@@ -76,6 +76,7 @@
 #include <linux/fs.h>
 #include <linux/init.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/xprtsock.h>
 #include <linux/nfs.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_mount.h>
@@ -491,7 +492,7 @@ static int __init root_nfs_get_handle(void)
 	struct sockaddr_in sin;
 	int status;
 	int protocol = (nfs_data.flags & NFS_MOUNT_TCP) ?
-					IPPROTO_TCP : IPPROTO_UDP;
+					XPRT_TRANSPORT_TCP : XPRT_TRANSPORT_UDP;
 	int version = (nfs_data.flags & NFS_MOUNT_VER3) ?
 					NFS_MNT3_VERSION : NFS_MNT_VERSION;
 

fs/nfs/proc.c

@@ -476,6 +476,8 @@ nfs_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 	dprintk("NFS call  readdir %d\n", (unsigned int)cookie);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 
+	nfs_invalidate_atime(dir);
+
 	dprintk("NFS reply readdir: %d\n", status);
 	return status;
 }
@@ -550,6 +552,7 @@ nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
 
 static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data)
 {
+	nfs_invalidate_atime(data->inode);
 	if (task->tk_status >= 0) {
 		nfs_refresh_inode(data->inode, data->res.fattr);
 		/* Emulate the eof flag, which isn't normally needed in NFSv2
@@ -576,7 +579,7 @@ static void nfs_proc_read_setup(struct nfs_read_data *data)
 static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	if (task->tk_status >= 0)
-		nfs_post_op_update_inode(data->inode, data->res.fattr);
+		nfs_post_op_update_inode_force_wcc(data->inode, data->res.fattr);
 	return 0;
 }
 

fs/nfs/read.c

@@ -341,9 +341,6 @@ int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data)
 		set_bit(NFS_INO_STALE, &NFS_FLAGS(data->inode));
 		nfs_mark_for_revalidate(data->inode);
 	}
-	spin_lock(&data->inode->i_lock);
-	NFS_I(data->inode)->cache_validity |= NFS_INO_INVALID_ATIME;
-	spin_unlock(&data->inode->i_lock);
 	return 0;
 }
 
@@ -497,8 +494,7 @@ int nfs_readpage(struct file *file, struct page *page)
 		if (ctx == NULL)
 			goto out_unlock;
 	} else
-		ctx = get_nfs_open_context((struct nfs_open_context *)
-				file->private_data);
+		ctx = get_nfs_open_context(nfs_file_open_context(file));
 
 	error = nfs_readpage_async(ctx, inode, page);
 
@@ -576,8 +572,7 @@ int nfs_readpages(struct file *filp, struct address_space *mapping,
 		if (desc.ctx == NULL)
 			return -EBADF;
 	} else
-		desc.ctx = get_nfs_open_context((struct nfs_open_context *)
-				filp->private_data);
+		desc.ctx = get_nfs_open_context(nfs_file_open_context(filp));
 	if (rsize < PAGE_CACHE_SIZE)
 		nfs_pageio_init(&pgio, inode, nfs_pagein_multi, rsize, 0);
 	else

fs/nfs/unlink.c

@@ -66,7 +66,6 @@ static void nfs_async_unlink_init(struct rpc_task *task, void *calldata)
 		.rpc_cred = data->cred,
 	};
 
-	nfs_begin_data_update(dir);
 	NFS_PROTO(dir)->unlink_setup(&msg, dir);
 	rpc_call_setup(task, &msg, 0);
 }
@@ -84,8 +83,6 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
 		rpc_restart_call(task);
-	else
-		nfs_end_data_update(dir);
 }
 
 /**

fs/nfsd/nfs4xdr.c

@@ -102,7 +102,8 @@ check_filename(char *str, int len, __be32 err)
 out:						\
 	return status;				\
 xdr_error:					\
-	printk(KERN_NOTICE "xdr error! (%s:%d)\n", __FILE__, __LINE__);	\
+	dprintk("NFSD: xdr error (%s:%d)\n",	\
+			__FILE__, __LINE__);	\
 	status = nfserr_bad_xdr;		\
 	goto out
 
@@ -124,7 +125,8 @@ xdr_error:					\
 	if (!(x = (p==argp->tmp || p == argp->tmpp) ? \
  		savemem(argp, p, nbytes) :	\
  		(char *)p)) {			\
-		printk(KERN_NOTICE "xdr error! (%s:%d)\n", __FILE__, __LINE__); \
+		dprintk("NFSD: xdr error (%s:%d)\n", \
+				__FILE__, __LINE__); \
 		goto xdr_error;			\
 		}				\
 	p += XDR_QUADLEN(nbytes);		\
@@ -140,7 +142,8 @@ xdr_error:					\
 		p = argp->p;			\
 		argp->p += XDR_QUADLEN(nbytes);	\
 	} else if (!(p = read_buf(argp, nbytes))) { \
-		printk(KERN_NOTICE "xdr error! (%s:%d)\n", __FILE__, __LINE__); \
+		dprintk("NFSD: xdr error (%s:%d)\n", \
+				__FILE__, __LINE__); \
 		goto xdr_error;			\
 	}					\
 } while (0)
@@ -948,7 +951,8 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write)
 	 */
 	avail = (char*)argp->end - (char*)argp->p;
 	if (avail + argp->pagelen < write->wr_buflen) {
-		printk(KERN_NOTICE "xdr error! (%s:%d)\n", __FILE__, __LINE__); 
+		dprintk("NFSD: xdr error (%s:%d)\n",
+				__FILE__, __LINE__);
 		goto xdr_error;
 	}
 	argp->rqstp->rq_vec[0].iov_base = p;
@@ -1019,7 +1023,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 		argp->ops = kmalloc(argp->opcnt * sizeof(*argp->ops), GFP_KERNEL);
 		if (!argp->ops) {
 			argp->ops = argp->iops;
-			printk(KERN_INFO "nfsd: couldn't allocate room for COMPOUND\n");
+			dprintk("nfsd: couldn't allocate room for COMPOUND\n");
 			goto xdr_error;
 		}
 	}
@@ -1326,7 +1330,7 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *
 	path = exp->ex_path;
 
 	if (strncmp(path, rootpath, strlen(rootpath))) {
-		printk("nfsd: fs_locations failed;"
+		dprintk("nfsd: fs_locations failed;"
 			"%s is not contained in %s\n", path, rootpath);
 		*stat = nfserr_notsupp;
 		return NULL;

include/linux/jiffies.h

@@ -109,6 +109,10 @@ static inline u64 get_jiffies_64(void)
 	 ((long)(a) - (long)(b) >= 0))
 #define time_before_eq(a,b)	time_after_eq(b,a)
 
+#define time_in_range(a,b,c) \
+	(time_after_eq(a,b) && \
+	 time_before_eq(a,c))
+
 /* Same as above, but does so with platform independent 64bit types.
  * These must be used when utilizing jiffies_64 (i.e. return value of
  * get_jiffies_64() */

include/linux/nfs_fs.h

@@ -47,10 +47,8 @@
 #include <linux/nfs3.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_xdr.h>
-
 #include <linux/nfs_fs_sb.h>
 
-#include <linux/rwsem.h>
 #include <linux/mempool.h>
 
 /*
@@ -77,6 +75,9 @@ struct nfs_open_context {
 	struct nfs4_state *state;
 	fl_owner_t lockowner;
 	int mode;
+
+	unsigned long flags;
+#define NFS_CONTEXT_ERROR_WRITE		(0)
 	int error;
 
 	struct list_head list;
@@ -133,11 +134,6 @@ struct nfs_inode {
 	 * server.
 	 */
 	unsigned long		cache_change_attribute;
-	/*
-	 * Counter indicating the number of outstanding requests that
-	 * will cause a file data update.
-	 */
-	atomic_t		data_updates;
 
 	struct rb_root		access_cache;
 	struct list_head	access_cache_entry_lru;
@@ -205,27 +201,18 @@ static inline struct nfs_inode *NFS_I(struct inode *inode)
 #define NFS_CLIENT(inode)		(NFS_SERVER(inode)->client)
 #define NFS_PROTO(inode)		(NFS_SERVER(inode)->nfs_client->rpc_ops)
 #define NFS_COOKIEVERF(inode)		(NFS_I(inode)->cookieverf)
-#define NFS_READTIME(inode)		(NFS_I(inode)->read_cache_jiffies)
-#define NFS_CHANGE_ATTR(inode)		(NFS_I(inode)->change_attr)
-#define NFS_ATTRTIMEO(inode)		(NFS_I(inode)->attrtimeo)
 #define NFS_MINATTRTIMEO(inode) \
 	(S_ISDIR(inode->i_mode)? NFS_SERVER(inode)->acdirmin \
 			       : NFS_SERVER(inode)->acregmin)
 #define NFS_MAXATTRTIMEO(inode) \
 	(S_ISDIR(inode->i_mode)? NFS_SERVER(inode)->acdirmax \
 			       : NFS_SERVER(inode)->acregmax)
-#define NFS_ATTRTIMEO_UPDATE(inode)	(NFS_I(inode)->attrtimeo_timestamp)
 
 #define NFS_FLAGS(inode)		(NFS_I(inode)->flags)
 #define NFS_STALE(inode)		(test_bit(NFS_INO_STALE, &NFS_FLAGS(inode)))
 
 #define NFS_FILEID(inode)		(NFS_I(inode)->fileid)
 
-static inline int nfs_caches_unstable(struct inode *inode)
-{
-	return atomic_read(&NFS_I(inode)->data_updates) != 0;
-}
-
 static inline void nfs_mark_for_revalidate(struct inode *inode)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
@@ -237,12 +224,6 @@ static inline void nfs_mark_for_revalidate(struct inode *inode)
 	spin_unlock(&inode->i_lock);
 }
 
-static inline void NFS_CACHEINV(struct inode *inode)
-{
-	if (!nfs_caches_unstable(inode))
-		nfs_mark_for_revalidate(inode);
-}
-
 static inline int nfs_server_capable(struct inode *inode, int cap)
 {
 	return NFS_SERVER(inode)->caps & cap;
@@ -253,28 +234,33 @@ static inline int NFS_USE_READDIRPLUS(struct inode *inode)
 	return test_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
 }
 
+static inline void nfs_set_verifier(struct dentry * dentry, unsigned long verf)
+{
+	dentry->d_time = verf;
+}
+
 /**
  * nfs_save_change_attribute - Returns the inode attribute change cookie
- * @inode - pointer to inode
+ * @dir - pointer to parent directory inode
  * The "change attribute" is updated every time we finish an operation
  * that will result in a metadata change on the server.
  */
-static inline long nfs_save_change_attribute(struct inode *inode)
+static inline unsigned long nfs_save_change_attribute(struct inode *dir)
 {
-	return NFS_I(inode)->cache_change_attribute;
+	return NFS_I(dir)->cache_change_attribute;
 }
 
 /**
- * nfs_verify_change_attribute - Detects NFS inode cache updates
- * @inode - pointer to inode
+ * nfs_verify_change_attribute - Detects NFS remote directory changes
+ * @dir - pointer to parent directory inode
  * @chattr - previously saved change attribute
- * Return "false" if metadata has been updated (or is in the process of
- * being updated) since the change attribute was saved.
+ * Return "false" if the verifiers doesn't match the change attribute.
+ * This would usually indicate that the directory contents have changed on
+ * the server, and that any dentries need revalidating.
  */
-static inline int nfs_verify_change_attribute(struct inode *inode, unsigned long chattr)
+static inline int nfs_verify_change_attribute(struct inode *dir, unsigned long chattr)
 {
-	return !nfs_caches_unstable(inode)
-		&& time_after_eq(chattr, NFS_I(inode)->cache_change_attribute);
+	return chattr == NFS_I(dir)->cache_change_attribute;
 }
 
 /*
@@ -283,15 +269,14 @@ static inline int nfs_verify_change_attribute(struct inode *inode, unsigned long
 extern int nfs_sync_mapping(struct address_space *mapping);
 extern void nfs_zap_mapping(struct inode *inode, struct address_space *mapping);
 extern void nfs_zap_caches(struct inode *);
+extern void nfs_invalidate_atime(struct inode *);
 extern struct inode *nfs_fhget(struct super_block *, struct nfs_fh *,
 				struct nfs_fattr *);
 extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *);
 extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr);
+extern int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fattr);
 extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 extern int nfs_permission(struct inode *, int, struct nameidata *);
-extern int nfs_access_get_cached(struct inode *, struct rpc_cred *, struct nfs_access_entry *);
-extern void nfs_access_add_cache(struct inode *, struct nfs_access_entry *);
-extern void nfs_access_zap_cache(struct inode *inode);
 extern int nfs_open(struct inode *, struct file *);
 extern int nfs_release(struct inode *, struct file *);
 extern int nfs_attribute_timeout(struct inode *inode);
@@ -301,13 +286,10 @@ extern int nfs_revalidate_mapping(struct inode *inode, struct address_space *map
 extern int nfs_revalidate_mapping_nolock(struct inode *inode, struct address_space *mapping);
 extern int nfs_setattr(struct dentry *, struct iattr *);
 extern void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr);
-extern void nfs_begin_attr_update(struct inode *);
-extern void nfs_end_attr_update(struct inode *);
-extern void nfs_begin_data_update(struct inode *);
-extern void nfs_end_data_update(struct inode *);
 extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx);
 extern void put_nfs_open_context(struct nfs_open_context *ctx);
 extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode);
+extern u64 nfs_compat_user_ino64(u64 fileid);
 
 /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */
 extern __be32 root_nfs_parse_addr(char *name); /*__init*/
@@ -328,14 +310,15 @@ extern const struct inode_operations nfs3_file_inode_operations;
 extern const struct file_operations nfs_file_operations;
 extern const struct address_space_operations nfs_file_aops;
 
-static inline struct rpc_cred *nfs_file_cred(struct file *file)
+static inline struct nfs_open_context *nfs_file_open_context(struct file *filp)
 {
-	if (file != NULL) {
-		struct nfs_open_context *ctx;
+	return filp->private_data;
+}
 
-		ctx = (struct nfs_open_context*)file->private_data;
-		return ctx->cred;
-	}
+static inline struct rpc_cred *nfs_file_cred(struct file *file)
+{
+	if (file != NULL)
+		return nfs_file_open_context(file)->cred;
 	return NULL;
 }
 
@@ -378,6 +361,8 @@ extern const struct file_operations nfs_dir_operations;
 extern struct dentry_operations nfs_dentry_operations;
 
 extern int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fh, struct nfs_fattr *fattr);
+extern int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags);
+extern void nfs_access_zap_cache(struct inode *inode);
 
 /*
  * linux/fs/nfs/symlink.c
@@ -420,15 +405,14 @@ extern int  nfs_flush_incompatible(struct file *file, struct page *page);
 extern int  nfs_updatepage(struct file *, struct page *, unsigned int, unsigned int);
 extern int nfs_writeback_done(struct rpc_task *, struct nfs_write_data *);
 extern void nfs_writedata_release(void *);
-extern int nfs_set_page_dirty(struct page *);
 
 /*
  * Try to write back everything synchronously (but check the
  * return value!)
  */
 extern long nfs_sync_mapping_wait(struct address_space *, struct writeback_control *, int);
-extern int nfs_sync_mapping_range(struct address_space *, loff_t, loff_t, int);
 extern int nfs_wb_all(struct inode *inode);
+extern int nfs_wb_nocommit(struct inode *inode);
 extern int nfs_wb_page(struct inode *inode, struct page* page);
 extern int nfs_wb_page_priority(struct inode *inode, struct page* page, int how);
 extern int nfs_wb_page_cancel(struct inode *inode, struct page* page);

include/linux/nfs_page.h

@@ -30,7 +30,6 @@
 #define PG_BUSY			0
 #define PG_NEED_COMMIT		1
 #define PG_NEED_RESCHED		2
-#define PG_NEED_FLUSH		3
 
 struct nfs_inode;
 struct nfs_page {

include/linux/nfs_xdr.h

@@ -62,7 +62,8 @@ struct nfs_fattr {
 #define NFS_ATTR_FATTR		0x0002		/* post-op attributes */
 #define NFS_ATTR_FATTR_V3	0x0004		/* NFSv3 attributes */
 #define NFS_ATTR_FATTR_V4	0x0008		/* NFSv4 change attribute */
-#define NFS_ATTR_FATTR_V4_REFERRAL	0x0010		/* NFSv4 referral */
+#define NFS_ATTR_WCC_V4		0x0010		/* pre-op change attribute */
+#define NFS_ATTR_FATTR_V4_REFERRAL	0x0020		/* NFSv4 referral */
 
 /*
  * Info on the file system
@@ -538,10 +539,13 @@ typedef u64 clientid4;
 
 struct nfs4_accessargs {
 	const struct nfs_fh *		fh;
+	const u32 *			bitmask;
 	u32				access;
 };
 
 struct nfs4_accessres {
+	const struct nfs_server *	server;
+	struct nfs_fattr *		fattr;
 	u32				supported;
 	u32				access;
 };

include/linux/sunrpc/clnt.h

@@ -117,7 +117,7 @@ struct rpc_create_args {
 
 struct rpc_clnt *rpc_create(struct rpc_create_args *args);
 struct rpc_clnt	*rpc_bind_new_program(struct rpc_clnt *,
-				struct rpc_program *, int);
+				struct rpc_program *, u32);
 struct rpc_clnt *rpc_clone_client(struct rpc_clnt *);
 void		rpc_shutdown_client(struct rpc_clnt *);
 void		rpc_release_client(struct rpc_clnt *);

include/linux/sunrpc/debug.h

@@ -88,6 +88,11 @@ enum {
 	CTL_SLOTTABLE_TCP,
 	CTL_MIN_RESVPORT,
 	CTL_MAX_RESVPORT,
+	CTL_SLOTTABLE_RDMA,
+	CTL_RDMA_MAXINLINEREAD,
+	CTL_RDMA_MAXINLINEWRITE,
+	CTL_RDMA_WRITEPADDING,
+	CTL_RDMA_MEMREG,
 };
 
 #endif /* _LINUX_SUNRPC_DEBUG_H_ */

include/linux/sunrpc/msg_prot.h

@@ -138,6 +138,19 @@ typedef __be32	rpc_fraghdr;
 #define RPC_MAX_HEADER_WITH_AUTH \
 	(RPC_CALLHDRSIZE + 2*(2+RPC_MAX_AUTH_SIZE/4))
 
+/*
+ * RFC1833/RFC3530 rpcbind (v3+) well-known netid's.
+ */
+#define RPCBIND_NETID_UDP	"udp"
+#define RPCBIND_NETID_TCP	"tcp"
+#define RPCBIND_NETID_UDP6	"udp6"
+#define RPCBIND_NETID_TCP6	"tcp6"
+
+/*
+ * Note that RFC 1833 does not put any size restrictions on the
+ * netid string, but all currently defined netid's fit in 4 bytes.
+ */
+#define RPCBIND_MAXNETIDLEN	(4u)
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_SUNRPC_MSGPROT_H_ */

include/linux/sunrpc/rpc_rdma.h

+/*
+ * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the BSD-type
+ * license below:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *      Redistributions of source code must retain the above copyright
+ *      notice, this list of conditions and the following disclaimer.
+ *
+ *      Redistributions in binary form must reproduce the above
+ *      copyright notice, this list of conditions and the following
+ *      disclaimer in the documentation and/or other materials provided
+ *      with the distribution.
+ *
+ *      Neither the name of the Network Appliance, Inc. nor the names of
+ *      its contributors may be used to endorse or promote products
+ *      derived from this software without specific prior written
+ *      permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _LINUX_SUNRPC_RPC_RDMA_H
+#define _LINUX_SUNRPC_RPC_RDMA_H
+
+struct rpcrdma_segment {
+	uint32_t rs_handle;	/* Registered memory handle */
+	uint32_t rs_length;	/* Length of the chunk in bytes */
+	uint64_t rs_offset;	/* Chunk virtual address or offset */
+};
+
+/*
+ * read chunk(s), encoded as a linked list.
+ */
+struct rpcrdma_read_chunk {
+	uint32_t rc_discrim;	/* 1 indicates presence */
+	uint32_t rc_position;	/* Position in XDR stream */
+	struct rpcrdma_segment rc_target;
+};
+
+/*
+ * write chunk, and reply chunk.
+ */
+struct rpcrdma_write_chunk {
+	struct rpcrdma_segment wc_target;
+};
+
+/*
+ * write chunk(s), encoded as a counted array.
+ */
+struct rpcrdma_write_array {
+	uint32_t wc_discrim;	/* 1 indicates presence */
+	uint32_t wc_nchunks;	/* Array count */
+	struct rpcrdma_write_chunk wc_array[0];
+};
+
+struct rpcrdma_msg {
+	uint32_t rm_xid;	/* Mirrors the RPC header xid */
+	uint32_t rm_vers;	/* Version of this protocol */
+	uint32_t rm_credit;	/* Buffers requested/granted */
+	uint32_t rm_type;	/* Type of message (enum rpcrdma_proc) */
+	union {
+
+		struct {			/* no chunks */
+			uint32_t rm_empty[3];	/* 3 empty chunk lists */
+		} rm_nochunks;
+
+		struct {			/* no chunks and padded */
+			uint32_t rm_align;	/* Padding alignment */
+			uint32_t rm_thresh;	/* Padding threshold */
+			uint32_t rm_pempty[3];	/* 3 empty chunk lists */
+		} rm_padded;
+
+		uint32_t rm_chunks[0];	/* read, write and reply chunks */
+
+	} rm_body;
+};
+
+#define RPCRDMA_HDRLEN_MIN	28
+
+enum rpcrdma_errcode {
+	ERR_VERS = 1,
+	ERR_CHUNK = 2
+};
+
+struct rpcrdma_err_vers {
+	uint32_t rdma_vers_low;	/* Version range supported by peer */
+	uint32_t rdma_vers_high;
+};
+
+enum rpcrdma_proc {
+	RDMA_MSG = 0,		/* An RPC call or reply msg */
+	RDMA_NOMSG = 1,		/* An RPC call or reply msg - separate body */
+	RDMA_MSGP = 2,		/* An RPC call or reply msg with padding */
+	RDMA_DONE = 3,		/* Client signals reply completion */
+	RDMA_ERROR = 4		/* An RPC RDMA encoding error */
+};
+
+#endif				/* _LINUX_SUNRPC_RPC_RDMA_H */

include/linux/sunrpc/xdr.h

@@ -70,7 +70,10 @@ struct xdr_buf {
 
 	struct page **	pages;		/* Array of contiguous pages */
 	unsigned int	page_base,	/* Start of page data */
-			page_len;	/* Length of page data */
+			page_len,	/* Length of page data */
+			flags;		/* Flags for data disposition */
+#define XDRBUF_READ		0x01		/* target of file read */
+#define XDRBUF_WRITE		0x02		/* source of file write */
 
 	unsigned int	buflen,		/* Total length of storage buffer */
 			len;		/* Length of XDR encoded message */

include/linux/sunrpc/xprt.h

@@ -19,24 +19,10 @@
 
 #ifdef __KERNEL__
 
-extern unsigned int xprt_udp_slot_table_entries;
-extern unsigned int xprt_tcp_slot_table_entries;
-
 #define RPC_MIN_SLOT_TABLE	(2U)
 #define RPC_DEF_SLOT_TABLE	(16U)
 #define RPC_MAX_SLOT_TABLE	(128U)
 
-/*
- * Parameters for choosing a free port
- */
-extern unsigned int xprt_min_resvport;
-extern unsigned int xprt_max_resvport;
-
-#define RPC_MIN_RESVPORT	(1U)
-#define RPC_MAX_RESVPORT	(65535U)
-#define RPC_DEF_MIN_RESVPORT	(665U)
-#define RPC_DEF_MAX_RESVPORT	(1023U)
-
 /*
  * This describes a timeout strategy
  */
@@ -53,6 +39,10 @@ enum rpc_display_format_t {
 	RPC_DISPLAY_PORT,
 	RPC_DISPLAY_PROTO,
 	RPC_DISPLAY_ALL,
+	RPC_DISPLAY_HEX_ADDR,
+	RPC_DISPLAY_HEX_PORT,
+	RPC_DISPLAY_UNIVERSAL_ADDR,
+	RPC_DISPLAY_NETID,
 	RPC_DISPLAY_MAX,
 };
 
@@ -196,14 +186,22 @@ struct rpc_xprt {
 	char *			address_strings[RPC_DISPLAY_MAX];
 };
 
-struct rpc_xprtsock_create {
-	int			proto;		/* IPPROTO_UDP or IPPROTO_TCP */
+struct xprt_create {
+	int			ident;		/* XPRT_TRANSPORT identifier */
 	struct sockaddr *	srcaddr;	/* optional local address */
 	struct sockaddr *	dstaddr;	/* remote peer address */
 	size_t			addrlen;
 	struct rpc_timeout *	timeout;	/* optional timeout parameters */
 };
 
+struct xprt_class {
+	struct list_head	list;
+	int			ident;		/* XPRT_TRANSPORT identifier */
+	struct rpc_xprt *	(*setup)(struct xprt_create *);
+	struct module		*owner;
+	char			name[32];
+};
+
 /*
  * Transport operations used by ULPs
  */
@@ -212,7 +210,7 @@ void			xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long
 /*
  * Generic internal transport functions
  */
-struct rpc_xprt *	xprt_create_transport(struct rpc_xprtsock_create *args);
+struct rpc_xprt		*xprt_create_transport(struct xprt_create *args);
 void			xprt_connect(struct rpc_task *task);
 void			xprt_reserve(struct rpc_task *task);
 int			xprt_reserve_xprt(struct rpc_task *task);
@@ -235,6 +233,8 @@ static inline __be32 *xprt_skip_transport_header(struct rpc_xprt *xprt, __be32 *
 /*
  * Transport switch helper functions
  */
+int			xprt_register_transport(struct xprt_class *type);
+int			xprt_unregister_transport(struct xprt_class *type);
 void			xprt_set_retrans_timeout_def(struct rpc_task *task);
 void			xprt_set_retrans_timeout_rtt(struct rpc_task *task);
 void			xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status);
@@ -247,14 +247,6 @@ void			xprt_complete_rqst(struct rpc_task *task, int copied);
 void			xprt_release_rqst_cong(struct rpc_task *task);
 void			xprt_disconnect(struct rpc_xprt *xprt);
 
-/*
- * Socket transport setup operations
- */
-struct rpc_xprt *	xs_setup_udp(struct rpc_xprtsock_create *args);
-struct rpc_xprt *	xs_setup_tcp(struct rpc_xprtsock_create *args);
-int			init_socket_xprt(void);
-void			cleanup_socket_xprt(void);
-
 /*
  * Reserved bit positions in xprt->state
  */

include/linux/sunrpc/xprtrdma.h

+/*
+ * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the BSD-type
+ * license below:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *      Redistributions of source code must retain the above copyright
+ *      notice, this list of conditions and the following disclaimer.
+ *
+ *      Redistributions in binary form must reproduce the above
+ *      copyright notice, this list of conditions and the following
+ *      disclaimer in the documentation and/or other materials provided
+ *      with the distribution.
+ *
+ *      Neither the name of the Network Appliance, Inc. nor the names of
+ *      its contributors may be used to endorse or promote products
+ *      derived from this software without specific prior written
+ *      permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _LINUX_SUNRPC_XPRTRDMA_H
+#define _LINUX_SUNRPC_XPRTRDMA_H
+
+/*
+ * RPC transport identifier for RDMA
+ */
+#define XPRT_TRANSPORT_RDMA	256
+
+/*
+ * rpcbind (v3+) RDMA netid.
+ */
+#define RPCBIND_NETID_RDMA	"rdma"
+
+/*
+ * Constants. Max RPC/NFS header is big enough to account for
+ * additional marshaling buffers passed down by Linux client.
+ *
+ * RDMA header is currently fixed max size, and is big enough for a
+ * fully-chunked NFS message (read chunks are the largest). Note only
+ * a single chunk type per message is supported currently.
+ */
+#define RPCRDMA_MIN_SLOT_TABLE	(2U)
+#define RPCRDMA_DEF_SLOT_TABLE	(32U)
+#define RPCRDMA_MAX_SLOT_TABLE	(256U)
+
+#define RPCRDMA_DEF_INLINE  (1024)	/* default inline max */
+
+#define RPCRDMA_INLINE_PAD_THRESH  (512)/* payload threshold to pad (bytes) */
+
+#define RDMA_RESOLVE_TIMEOUT	(5*HZ)	/* TBD 5 seconds */
+#define RDMA_CONNECT_RETRY_MAX	(2)	/* retries if no listener backlog */
+
+/* memory registration strategies */
+#define RPCRDMA_PERSISTENT_REGISTRATION (1)
+
+enum rpcrdma_memreg {
+	RPCRDMA_BOUNCEBUFFERS = 0,
+	RPCRDMA_REGISTER,
+	RPCRDMA_MEMWINDOWS,
+	RPCRDMA_MEMWINDOWS_ASYNC,
+	RPCRDMA_MTHCAFMR,
+	RPCRDMA_ALLPHYSICAL,
+	RPCRDMA_LAST
+};
+
+#endif /* _LINUX_SUNRPC_XPRTRDMA_H */

include/linux/sunrpc/xprtsock.h

+/*
+ *  linux/include/linux/sunrpc/xprtsock.h
+ *
+ *  Declarations for the RPC transport socket provider.
+ */
+
+#ifndef _LINUX_SUNRPC_XPRTSOCK_H
+#define _LINUX_SUNRPC_XPRTSOCK_H
+
+#ifdef __KERNEL__
+
+/*
+ * Socket transport setup operations
+ */
+struct rpc_xprt *xs_setup_udp(struct xprt_create *args);
+struct rpc_xprt *xs_setup_tcp(struct xprt_create *args);
+
+int		init_socket_xprt(void);
+void		cleanup_socket_xprt(void);
+
+/*
+ * RPC transport identifiers for UDP, TCP
+ *
+ * To preserve compatibility with the historical use of raw IP protocol
+ * id's for transport selection, these are specified with the previous
+ * values. No such restriction exists for new transports, except that
+ * they may not collide with these values (17 and 6, respectively).
+ */
+#define XPRT_TRANSPORT_UDP	IPPROTO_UDP
+#define XPRT_TRANSPORT_TCP	IPPROTO_TCP
+
+/*
+ * RPC slot table sizes for UDP, TCP transports
+ */
+extern unsigned int xprt_udp_slot_table_entries;
+extern unsigned int xprt_tcp_slot_table_entries;
+
+/*
+ * Parameters for choosing a free port
+ */
+extern unsigned int xprt_min_resvport;
+extern unsigned int xprt_max_resvport;
+
+#define RPC_MIN_RESVPORT	(1U)
+#define RPC_MAX_RESVPORT	(65535U)
+#define RPC_DEF_MIN_RESVPORT	(665U)
+#define RPC_DEF_MAX_RESVPORT	(1023U)
+
+#endif /* __KERNEL__ */
+
+#endif /* _LINUX_SUNRPC_XPRTSOCK_H */

include/linux/writeback.h

@@ -62,8 +62,6 @@ struct writeback_control {
 	unsigned for_reclaim:1;		/* Invoked from the page allocator */
 	unsigned for_writepages:1;	/* This is a writepages() call */
 	unsigned range_cyclic:1;	/* range_start is cyclic */
-
-	void *fs_private;		/* For use by ->writepages() */
 };
 
 /*

kernel/auditsc.c

@@ -1525,6 +1525,7 @@ void __audit_inode_child(const char *dname, const struct inode *inode,
 			context->names[idx].ino = (unsigned long)-1;
 	}
 }
+EXPORT_SYMBOL_GPL(__audit_inode_child);
 
 /**
  * auditsc_get_stamp - get local copies of audit_context values

net/sunrpc/Makefile

@@ -5,6 +5,7 @@
 
 obj-$(CONFIG_SUNRPC) += sunrpc.o
 obj-$(CONFIG_SUNRPC_GSS) += auth_gss/
+obj-$(CONFIG_SUNRPC_XPRT_RDMA) += xprtrdma/
 
 sunrpc-y := clnt.o xprt.o socklib.o xprtsock.o sched.o \
 	    auth.o auth_null.o auth_unix.o \

net/sunrpc/auth_gss/gss_krb5_wrap.c

@@ -42,7 +42,7 @@ gss_krb5_remove_padding(struct xdr_buf *buf, int blocksize)
 {
 	u8 *ptr;
 	u8 pad;
-	int len = buf->len;
+	size_t len = buf->len;
 
 	if (len <= buf->head[0].iov_len) {
 		pad = *(u8 *)(buf->head[0].iov_base + len - 1);
@@ -53,9 +53,9 @@ gss_krb5_remove_padding(struct xdr_buf *buf, int blocksize)
 	} else
 		len -= buf->head[0].iov_len;
 	if (len <= buf->page_len) {
-		int last = (buf->page_base + len - 1)
+		unsigned int last = (buf->page_base + len - 1)
 					>>PAGE_CACHE_SHIFT;
-		int offset = (buf->page_base + len - 1)
+		unsigned int offset = (buf->page_base + len - 1)
 					& (PAGE_CACHE_SIZE - 1);
 		ptr = kmap_atomic(buf->pages[last], KM_USER0);
 		pad = *(ptr + offset);

net/sunrpc/clnt.c

@@ -127,7 +127,14 @@ static struct rpc_clnt * rpc_new_client(struct rpc_xprt *xprt, char *servname, s
 	struct rpc_clnt		*clnt = NULL;
 	struct rpc_auth		*auth;
 	int err;
-	int len;
+	size_t len;
+
+	/* sanity check the name before trying to print it */
+	err = -EINVAL;
+	len = strlen(servname);
+	if (len > RPC_MAXNETNAMELEN)
+		goto out_no_rpciod;
+	len++;
 
 	dprintk("RPC:       creating %s client for %s (xprt %p)\n",
 			program->name, servname, xprt);
@@ -148,7 +155,6 @@ static struct rpc_clnt * rpc_new_client(struct rpc_xprt *xprt, char *servname, s
 	clnt->cl_parent = clnt;
 
 	clnt->cl_server = clnt->cl_inline_name;
-	len = strlen(servname) + 1;
 	if (len > sizeof(clnt->cl_inline_name)) {
 		char *buf = kmalloc(len, GFP_KERNEL);
 		if (buf != 0)
@@ -234,8 +240,8 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 {
 	struct rpc_xprt *xprt;
 	struct rpc_clnt *clnt;
-	struct rpc_xprtsock_create xprtargs = {
-		.proto = args->protocol,
+	struct xprt_create xprtargs = {
+		.ident = args->protocol,
 		.srcaddr = args->saddress,
 		.dstaddr = args->address,
 		.addrlen = args->addrsize,
@@ -253,7 +259,7 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 	 */
 	if (args->servername == NULL) {
 		struct sockaddr_in *addr =
-					(struct sockaddr_in *) &args->address;
+					(struct sockaddr_in *) args->address;
 		snprintf(servername, sizeof(servername), NIPQUAD_FMT,
 			NIPQUAD(addr->sin_addr.s_addr));
 		args->servername = servername;
@@ -269,9 +275,6 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 	if (args->flags & RPC_CLNT_CREATE_NONPRIVPORT)
 		xprt->resvport = 0;
 
-	dprintk("RPC:       creating %s client for %s (xprt %p)\n",
-			args->program->name, args->servername, xprt);
-
 	clnt = rpc_new_client(xprt, args->servername, args->program,
 				args->version, args->authflavor);
 	if (IS_ERR(clnt))
@@ -439,7 +442,7 @@ rpc_release_client(struct rpc_clnt *clnt)
  */
 struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old,
 				      struct rpc_program *program,
-				      int vers)
+				      u32 vers)
 {
 	struct rpc_clnt *clnt;
 	struct rpc_version *version;
@@ -843,8 +846,7 @@ call_allocate(struct rpc_task *task)
 	dprintk("RPC: %5u rpc_buffer allocation failed\n", task->tk_pid);
 
 	if (RPC_IS_ASYNC(task) || !signalled()) {
-		xprt_release(task);
-		task->tk_action = call_reserve;
+		task->tk_action = call_allocate;
 		rpc_delay(task, HZ>>4);
 		return;
 	}
@@ -871,6 +873,7 @@ rpc_xdr_buf_init(struct xdr_buf *buf, void *start, size_t len)
 	buf->head[0].iov_len = len;
 	buf->tail[0].iov_len = 0;
 	buf->page_len = 0;
+	buf->flags = 0;
 	buf->len = 0;
 	buf->buflen = len;
 }
@@ -937,7 +940,7 @@ call_bind(struct rpc_task *task)
 static void
 call_bind_status(struct rpc_task *task)
 {
-	int status = -EACCES;
+	int status = -EIO;
 
 	if (task->tk_status >= 0) {
 		dprint_status(task);
@@ -947,9 +950,20 @@ call_bind_status(struct rpc_task *task)
 	}
 
 	switch (task->tk_status) {
+	case -EAGAIN:
+		dprintk("RPC: %5u rpcbind waiting for another request "
+				"to finish\n", task->tk_pid);
+		/* avoid busy-waiting here -- could be a network outage. */
+		rpc_delay(task, 5*HZ);
+		goto retry_timeout;
 	case -EACCES:
 		dprintk("RPC: %5u remote rpcbind: RPC program/version "
 				"unavailable\n", task->tk_pid);
+		/* fail immediately if this is an RPC ping */
+		if (task->tk_msg.rpc_proc->p_proc == 0) {
+			status = -EOPNOTSUPP;
+			break;
+		}
 		rpc_delay(task, 3*HZ);
 		goto retry_timeout;
 	case -ETIMEDOUT:
@@ -957,6 +971,7 @@ call_bind_status(struct rpc_task *task)
 				task->tk_pid);
 		goto retry_timeout;
 	case -EPFNOSUPPORT:
+		/* server doesn't support any rpcbind version we know of */
 		dprintk("RPC: %5u remote rpcbind service unavailable\n",
 				task->tk_pid);
 		break;
@@ -969,7 +984,6 @@ call_bind_status(struct rpc_task *task)
 	default:
 		dprintk("RPC: %5u unrecognized rpcbind error (%d)\n",
 				task->tk_pid, -task->tk_status);
-		status = -EIO;
 	}
 
 	rpc_exit(task, status);
@@ -1257,7 +1271,6 @@ call_refresh(struct rpc_task *task)
 {
 	dprint_status(task);
 
-	xprt_release(task);	/* Must do to obtain new XID */
 	task->tk_action = call_refreshresult;
 	task->tk_status = 0;
 	task->tk_client->cl_stats->rpcauthrefresh++;
@@ -1375,6 +1388,8 @@ call_verify(struct rpc_task *task)
 			dprintk("RPC: %5u %s: retry stale creds\n",
 					task->tk_pid, __FUNCTION__);
 			rpcauth_invalcred(task);
+			/* Ensure we obtain a new XID! */
+			xprt_release(task);
 			task->tk_action = call_refresh;
 			goto out_retry;
 		case RPC_AUTH_BADCRED:
@@ -1523,13 +1538,18 @@ void rpc_show_tasks(void)
 		spin_lock(&clnt->cl_lock);
 		list_for_each_entry(t, &clnt->cl_tasks, tk_task) {
 			const char *rpc_waitq = "none";
+			int proc;
+
+			if (t->tk_msg.rpc_proc)
+				proc = t->tk_msg.rpc_proc->p_proc;
+			else
+				proc = -1;
 
 			if (RPC_IS_QUEUED(t))
 				rpc_waitq = rpc_qname(t->u.tk_wait.rpc_waitq);
 
 			printk("%5u %04d %04x %6d %8p %6d %8p %8ld %8s %8p %8p\n",
-				t->tk_pid,
-				(t->tk_msg.rpc_proc ? t->tk_msg.rpc_proc->p_proc : -1),
+				t->tk_pid, proc,
 				t->tk_flags, t->tk_status,
 				t->tk_client,
 				(t->tk_client ? t->tk_client->cl_prog : 0),

net/sunrpc/rpc_pipe.c

@@ -14,7 +14,7 @@
 #include <linux/pagemap.h>
 #include <linux/mount.h>
 #include <linux/namei.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/kernel.h>
 
 #include <asm/ioctls.h>
@@ -329,6 +329,7 @@ rpc_show_info(struct seq_file *m, void *v)
 			clnt->cl_prog, clnt->cl_vers);
 	seq_printf(m, "address: %s\n", rpc_peeraddr2str(clnt, RPC_DISPLAY_ADDR));
 	seq_printf(m, "protocol: %s\n", rpc_peeraddr2str(clnt, RPC_DISPLAY_PROTO));
+	seq_printf(m, "port: %s\n", rpc_peeraddr2str(clnt, RPC_DISPLAY_PORT));
 	return 0;
 }
 
@@ -585,6 +586,7 @@ rpc_populate(struct dentry *parent,
 		if (S_ISDIR(mode))
 			inc_nlink(dir);
 		d_add(dentry, inode);
+		fsnotify_create(dir, dentry);
 	}
 	mutex_unlock(&dir->i_mutex);
 	return 0;
@@ -606,7 +608,7 @@ __rpc_mkdir(struct inode *dir, struct dentry *dentry)
 	inode->i_ino = iunique(dir->i_sb, 100);
 	d_instantiate(dentry, inode);
 	inc_nlink(dir);
-	inode_dir_notify(dir, DN_CREATE);
+	fsnotify_mkdir(dir, dentry);
 	return 0;
 out_err:
 	printk(KERN_WARNING "%s: %s failed to allocate inode for dentry %s\n",
@@ -748,7 +750,7 @@ rpc_mkpipe(struct dentry *parent, const char *name, void *private, struct rpc_pi
 	rpci->flags = flags;
 	rpci->ops = ops;
 	rpci->nkern_readwriters = 1;
-	inode_dir_notify(dir, DN_CREATE);
+	fsnotify_create(dir, dentry);
 	dget(dentry);
 out:
 	mutex_unlock(&dir->i_mutex);

net/sunrpc/rpcb_clnt.c

@@ -16,11 +16,14 @@
 
 #include <linux/types.h>
 #include <linux/socket.h>
+#include <linux/in.h>
+#include <linux/in6.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
 
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/xprtsock.h>
 
 #ifdef RPC_DEBUG
 # define RPCDBG_FACILITY	RPCDBG_BIND
@@ -90,26 +93,6 @@ enum {
  */
 #define RPCB_MAXADDRLEN		(128u)
 
-/*
- * r_netid
- *
- * Quoting RFC 3530, section 2.2:
- *
- * For TCP over IPv4 the value of r_netid is the string "tcp".  For UDP
- * over IPv4 the value of r_netid is the string "udp".
- *
- * ...
- *
- * For TCP over IPv6 the value of r_netid is the string "tcp6".  For UDP
- * over IPv6 the value of r_netid is the string "udp6".
- */
-#define RPCB_NETID_UDP	"\165\144\160"		/* "udp" */
-#define RPCB_NETID_TCP	"\164\143\160"		/* "tcp" */
-#define RPCB_NETID_UDP6	"\165\144\160\066"	/* "udp6" */
-#define RPCB_NETID_TCP6	"\164\143\160\066"	/* "tcp6" */
-
-#define RPCB_MAXNETIDLEN	(4u)
-
 /*
  * r_owner
  *
@@ -120,7 +103,7 @@ enum {
 #define RPCB_MAXOWNERLEN	sizeof(RPCB_OWNER_STRING)
 
 static void			rpcb_getport_done(struct rpc_task *, void *);
-extern struct rpc_program	rpcb_program;
+static struct rpc_program	rpcb_program;
 
 struct rpcbind_args {
 	struct rpc_xprt *	r_xprt;
@@ -137,10 +120,13 @@ struct rpcbind_args {
 static struct rpc_procinfo rpcb_procedures2[];
 static struct rpc_procinfo rpcb_procedures3[];
 
-static struct rpcb_info {
+struct rpcb_info {
 	int			rpc_vers;
 	struct rpc_procinfo *	rpc_proc;
-} rpcb_next_version[];
+};
+
+static struct rpcb_info rpcb_next_version[];
+static struct rpcb_info rpcb_next_version6[];
 
 static void rpcb_getport_prepare(struct rpc_task *task, void *calldata)
 {
@@ -190,7 +176,17 @@ static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr,
 				   RPC_CLNT_CREATE_INTR),
 	};
 
+	switch (srvaddr->sa_family) {
+	case AF_INET:
 		((struct sockaddr_in *)srvaddr)->sin_port = htons(RPCBIND_PORT);
+		break;
+	case AF_INET6:
+		((struct sockaddr_in6 *)srvaddr)->sin6_port = htons(RPCBIND_PORT);
+		break;
+	default:
+		return NULL;
+	}
+
 	if (!privileged)
 		args.flags |= RPC_CLNT_CREATE_NONPRIVPORT;
 	return rpc_create(&args);
@@ -234,7 +230,7 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay)
 			prog, vers, prot, port);
 
 	rpcb_clnt = rpcb_create("localhost", (struct sockaddr *) &sin,
-					IPPROTO_UDP, 2, 1);
+					XPRT_TRANSPORT_UDP, 2, 1);
 	if (IS_ERR(rpcb_clnt))
 		return PTR_ERR(rpcb_clnt);
 
@@ -316,6 +312,7 @@ void rpcb_getport_async(struct rpc_task *task)
 	struct rpc_task	*child;
 	struct sockaddr addr;
 	int status;
+	struct rpcb_info *info;
 
 	dprintk("RPC: %5u %s(%s, %u, %u, %d)\n",
 		task->tk_pid, __FUNCTION__,
@@ -325,7 +322,7 @@ void rpcb_getport_async(struct rpc_task *task)
 	BUG_ON(clnt->cl_parent != clnt);
 
 	if (xprt_test_and_set_binding(xprt)) {
-		status = -EACCES;		/* tell caller to check again */
+		status = -EAGAIN;	/* tell caller to check again */
 		dprintk("RPC: %5u %s: waiting for another binder\n",
 			task->tk_pid, __FUNCTION__);
 		goto bailout_nowake;
@@ -343,18 +340,43 @@ void rpcb_getport_async(struct rpc_task *task)
 		goto bailout_nofree;
 	}
 
-	if (rpcb_next_version[xprt->bind_index].rpc_proc == NULL) {
+	rpc_peeraddr(clnt, (void *)&addr, sizeof(addr));
+
+	/* Don't ever use rpcbind v2 for AF_INET6 requests */
+	switch (addr.sa_family) {
+	case AF_INET:
+		info = rpcb_next_version;
+		break;
+	case AF_INET6:
+		info = rpcb_next_version6;
+		break;
+	default:
+		status = -EAFNOSUPPORT;
+		dprintk("RPC: %5u %s: bad address family\n",
+				task->tk_pid, __FUNCTION__);
+		goto bailout_nofree;
+	}
+	if (info[xprt->bind_index].rpc_proc == NULL) {
 		xprt->bind_index = 0;
-		status = -EACCES;	/* tell caller to try again later */
+		status = -EPFNOSUPPORT;
 		dprintk("RPC: %5u %s: no more getport versions available\n",
 			task->tk_pid, __FUNCTION__);
 		goto bailout_nofree;
 	}
-	bind_version = rpcb_next_version[xprt->bind_index].rpc_vers;
+	bind_version = info[xprt->bind_index].rpc_vers;
 
 	dprintk("RPC: %5u %s: trying rpcbind version %u\n",
 		task->tk_pid, __FUNCTION__, bind_version);
 
+	rpcb_clnt = rpcb_create(clnt->cl_server, &addr, xprt->prot,
+				bind_version, 0);
+	if (IS_ERR(rpcb_clnt)) {
+		status = PTR_ERR(rpcb_clnt);
+		dprintk("RPC: %5u %s: rpcb_create failed, error %ld\n",
+			task->tk_pid, __FUNCTION__, PTR_ERR(rpcb_clnt));
+		goto bailout_nofree;
+	}
+
 	map = kzalloc(sizeof(struct rpcbind_args), GFP_ATOMIC);
 	if (!map) {
 		status = -ENOMEM;
@@ -367,28 +389,19 @@ void rpcb_getport_async(struct rpc_task *task)
 	map->r_prot = xprt->prot;
 	map->r_port = 0;
 	map->r_xprt = xprt_get(xprt);
-	map->r_netid = (xprt->prot == IPPROTO_TCP) ? RPCB_NETID_TCP :
-						   RPCB_NETID_UDP;
-	memcpy(&map->r_addr, rpc_peeraddr2str(clnt, RPC_DISPLAY_ADDR),
+	map->r_netid = rpc_peeraddr2str(clnt, RPC_DISPLAY_NETID);
+	memcpy(map->r_addr,
+	       rpc_peeraddr2str(rpcb_clnt, RPC_DISPLAY_UNIVERSAL_ADDR),
 	       sizeof(map->r_addr));
 	map->r_owner = RPCB_OWNER_STRING;	/* ignored for GETADDR */
 
-	rpc_peeraddr(clnt, (void *)&addr, sizeof(addr));
-	rpcb_clnt = rpcb_create(clnt->cl_server, &addr, xprt->prot, bind_version, 0);
-	if (IS_ERR(rpcb_clnt)) {
-		status = PTR_ERR(rpcb_clnt);
-		dprintk("RPC: %5u %s: rpcb_create failed, error %ld\n",
-			task->tk_pid, __FUNCTION__, PTR_ERR(rpcb_clnt));
-		goto bailout;
-	}
-
 	child = rpc_run_task(rpcb_clnt, RPC_TASK_ASYNC, &rpcb_getport_ops, map);
 	rpc_release_client(rpcb_clnt);
 	if (IS_ERR(child)) {
 		status = -EIO;
 		dprintk("RPC: %5u %s: rpc_run_task failed\n",
 			task->tk_pid, __FUNCTION__);
-		goto bailout_nofree;
+		goto bailout;
 	}
 	rpc_put_task(child);
 
@@ -403,6 +416,7 @@ void rpcb_getport_async(struct rpc_task *task)
 bailout_nowake:
 	task->tk_status = status;
 }
+EXPORT_SYMBOL_GPL(rpcb_getport_async);
 
 /*
  * Rpcbind child task calls this callback via tk_exit.
@@ -413,6 +427,10 @@ static void rpcb_getport_done(struct rpc_task *child, void *data)
 	struct rpc_xprt *xprt = map->r_xprt;
 	int status = child->tk_status;
 
+	/* Garbage reply: retry with a lesser rpcbind version */
+	if (status == -EIO)
+		status = -EPROTONOSUPPORT;
+
 	/* rpcbind server doesn't support this rpcbind protocol version */
 	if (status == -EPROTONOSUPPORT)
 		xprt->bind_index++;
@@ -490,16 +508,24 @@ static int rpcb_decode_getaddr(struct rpc_rqst *req, __be32 *p,
 			       unsigned short *portp)
 {
 	char *addr;
-	int addr_len, c, i, f, first, val;
+	u32 addr_len;
+	int c, i, f, first, val;
 
 	*portp = 0;
-	addr_len = (unsigned int) ntohl(*p++);
-	if (addr_len > RPCB_MAXADDRLEN)			/* sanity */
-		return -EINVAL;
+	addr_len = ntohl(*p++);
 
-	dprintk("RPC:       rpcb_decode_getaddr returned string: '%s'\n",
-			(char *) p);
+	/*
+	 * Simple sanity check.  The smallest possible universal
+	 * address is an IPv4 address string containing 11 bytes.
+	 */
+	if (addr_len < 11 || addr_len > RPCB_MAXADDRLEN)
+		goto out_err;
 
+	/*
+	 * Start at the end and walk backwards until the first dot
+	 * is encountered.  When the second dot is found, we have
+	 * both parts of the port number.
+	 */
 	addr = (char *)p;
 	val = 0;
 	first = 1;
@@ -521,8 +547,19 @@ static int rpcb_decode_getaddr(struct rpc_rqst *req, __be32 *p,
 		}
 	}
 
+	/*
+	 * Simple sanity check.  If we never saw a dot in the reply,
+	 * then this was probably just garbage.
+	 */
+	if (first)
+		goto out_err;
+
 	dprintk("RPC:       rpcb_decode_getaddr port=%u\n", *portp);
 	return 0;
+
+out_err:
+	dprintk("RPC:       rpcbind server returned malformed reply\n");
+	return -EIO;
 }
 
 #define RPCB_program_sz		(1u)
@@ -531,7 +568,7 @@ static int rpcb_decode_getaddr(struct rpc_rqst *req, __be32 *p,
 #define RPCB_port_sz		(1u)
 #define RPCB_boolean_sz		(1u)
 
-#define RPCB_netid_sz		(1+XDR_QUADLEN(RPCB_MAXNETIDLEN))
+#define RPCB_netid_sz		(1+XDR_QUADLEN(RPCBIND_MAXNETIDLEN))
 #define RPCB_addr_sz		(1+XDR_QUADLEN(RPCB_MAXADDRLEN))
 #define RPCB_ownerstring_sz	(1+XDR_QUADLEN(RPCB_MAXOWNERLEN))
 
@@ -593,6 +630,14 @@ static struct rpcb_info rpcb_next_version[] = {
 	{ 0, NULL },
 };
 
+static struct rpcb_info rpcb_next_version6[] = {
+#ifdef CONFIG_SUNRPC_BIND34
+	{ 4, &rpcb_procedures4[RPCBPROC_GETVERSADDR] },
+	{ 3, &rpcb_procedures3[RPCBPROC_GETADDR] },
+#endif
+	{ 0, NULL },
+};
+
 static struct rpc_version rpcb_version2 = {
 	.number		= 2,
 	.nrprocs	= RPCB_HIGHPROC_2,
@@ -621,7 +666,7 @@ static struct rpc_version *rpcb_version[] = {
 
 static struct rpc_stat rpcb_stats;
 
-struct rpc_program rpcb_program = {
+static struct rpc_program rpcb_program = {
 	.name		= "rpcbind",
 	.number		= RPCBIND_PROGRAM,
 	.nrvers		= ARRAY_SIZE(rpcb_version),

net/sunrpc/sched.c

@@ -777,6 +777,7 @@ void *rpc_malloc(struct rpc_task *task, size_t size)
 			task->tk_pid, size, buf);
 	return &buf->data;
 }
+EXPORT_SYMBOL_GPL(rpc_malloc);
 
 /**
  * rpc_free - free buffer allocated via rpc_malloc
@@ -802,6 +803,7 @@ void rpc_free(void *buffer)
 	else
 		kfree(buf);
 }
+EXPORT_SYMBOL_GPL(rpc_free);
 
 /*
  * Creation and deletion of RPC task structures

net/sunrpc/socklib.c

@@ -34,6 +34,7 @@ size_t xdr_skb_read_bits(struct xdr_skb_reader *desc, void *to, size_t len)
 	desc->offset += len;
 	return len;
 }
+EXPORT_SYMBOL_GPL(xdr_skb_read_bits);
 
 /**
  * xdr_skb_read_and_csum_bits - copy and checksum from skb to buffer
@@ -137,6 +138,7 @@ ssize_t xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, struct
 out:
 	return copied;
 }
+EXPORT_SYMBOL_GPL(xdr_partial_copy_from_skb);
 
 /**
  * csum_partial_copy_to_xdr - checksum and copy data
@@ -179,3 +181,4 @@ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
 		return -1;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(csum_partial_copy_to_xdr);

net/sunrpc/sunrpc_syms.c

@@ -20,7 +20,7 @@
 #include <linux/sunrpc/auth.h>
 #include <linux/workqueue.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
-
+#include <linux/sunrpc/xprtsock.h>
 
 /* RPC scheduler */
 EXPORT_SYMBOL(rpc_execute);

net/sunrpc/timer.c

@@ -17,6 +17,7 @@
 
 #include <linux/types.h>
 #include <linux/unistd.h>
+#include <linux/module.h>
 
 #include <linux/sunrpc/clnt.h>
 
@@ -40,6 +41,7 @@ rpc_init_rtt(struct rpc_rtt *rt, unsigned long timeo)
 		rt->ntimeouts[i] = 0;
 	}
 }
+EXPORT_SYMBOL_GPL(rpc_init_rtt);
 
 /*
  * NB: When computing the smoothed RTT and standard deviation,
@@ -75,6 +77,7 @@ rpc_update_rtt(struct rpc_rtt *rt, unsigned timer, long m)
 	if (*sdrtt < RPC_RTO_MIN)
 		*sdrtt = RPC_RTO_MIN;
 }
+EXPORT_SYMBOL_GPL(rpc_update_rtt);
 
 /*
  * Estimate rto for an nfs rpc sent via. an unreliable datagram.
@@ -103,3 +106,4 @@ rpc_calc_rto(struct rpc_rtt *rt, unsigned timer)
 
 	return res;
 }
+EXPORT_SYMBOL_GPL(rpc_calc_rto);

net/sunrpc/xprt.c

@@ -62,6 +62,9 @@ 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 *);
 
+static spinlock_t xprt_list_lock = SPIN_LOCK_UNLOCKED;
+static LIST_HEAD(xprt_list);
+
 /*
  * The transport code maintains an estimate on the maximum number of out-
  * standing RPC requests, using a smoothed version of the congestion
@@ -80,6 +83,78 @@ static int      __xprt_get_cong(struct rpc_xprt *, struct rpc_task *);
 
 #define RPCXPRT_CONGESTED(xprt) ((xprt)->cong >= (xprt)->cwnd)
 
+/**
+ * xprt_register_transport - register a transport implementation
+ * @transport: transport to register
+ *
+ * If a transport implementation is loaded as a kernel module, it can
+ * call this interface to make itself known to the RPC client.
+ *
+ * Returns:
+ * 0:		transport successfully registered
+ * -EEXIST:	transport already registered
+ * -EINVAL:	transport module being unloaded
+ */
+int xprt_register_transport(struct xprt_class *transport)
+{
+	struct xprt_class *t;
+	int result;
+
+	result = -EEXIST;
+	spin_lock(&xprt_list_lock);
+	list_for_each_entry(t, &xprt_list, list) {
+		/* don't register the same transport class twice */
+		if (t->ident == transport->ident)
+			goto out;
+	}
+
+	result = -EINVAL;
+	if (try_module_get(THIS_MODULE)) {
+		list_add_tail(&transport->list, &xprt_list);
+		printk(KERN_INFO "RPC: Registered %s transport module.\n",
+			transport->name);
+		result = 0;
+	}
+
+out:
+	spin_unlock(&xprt_list_lock);
+	return result;
+}
+EXPORT_SYMBOL_GPL(xprt_register_transport);
+
+/**
+ * xprt_unregister_transport - unregister a transport implementation
+ * transport: transport to unregister
+ *
+ * Returns:
+ * 0:		transport successfully unregistered
+ * -ENOENT:	transport never registered
+ */
+int xprt_unregister_transport(struct xprt_class *transport)
+{
+	struct xprt_class *t;
+	int result;
+
+	result = 0;
+	spin_lock(&xprt_list_lock);
+	list_for_each_entry(t, &xprt_list, list) {
+		if (t == transport) {
+			printk(KERN_INFO
+				"RPC: Unregistered %s transport module.\n",
+				transport->name);
+			list_del_init(&transport->list);
+			module_put(THIS_MODULE);
+			goto out;
+		}
+	}
+	result = -ENOENT;
+
+out:
+	spin_unlock(&xprt_list_lock);
+	return result;
+}
+EXPORT_SYMBOL_GPL(xprt_unregister_transport);
+
 /**
  * xprt_reserve_xprt - serialize write access to transports
  * @task: task that is requesting access to the transport
@@ -118,6 +193,7 @@ int xprt_reserve_xprt(struct rpc_task *task)
 		rpc_sleep_on(&xprt->sending, task, NULL, NULL);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xprt_reserve_xprt);
 
 static void xprt_clear_locked(struct rpc_xprt *xprt)
 {
@@ -167,6 +243,7 @@ int xprt_reserve_xprt_cong(struct rpc_task *task)
 		rpc_sleep_on(&xprt->sending, task, NULL, NULL);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xprt_reserve_xprt_cong);
 
 static inline int xprt_lock_write(struct rpc_xprt *xprt, struct rpc_task *task)
 {
@@ -246,6 +323,7 @@ void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 		__xprt_lock_write_next(xprt);
 	}
 }
+EXPORT_SYMBOL_GPL(xprt_release_xprt);
 
 /**
  * xprt_release_xprt_cong - allow other requests to use a transport
@@ -262,6 +340,7 @@ void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 		__xprt_lock_write_next_cong(xprt);
 	}
 }
+EXPORT_SYMBOL_GPL(xprt_release_xprt_cong);
 
 static inline void xprt_release_write(struct rpc_xprt *xprt, struct rpc_task *task)
 {
@@ -314,6 +393,7 @@ void xprt_release_rqst_cong(struct rpc_task *task)
 {
 	__xprt_put_cong(task->tk_xprt, task->tk_rqstp);
 }
+EXPORT_SYMBOL_GPL(xprt_release_rqst_cong);
 
 /**
  * xprt_adjust_cwnd - adjust transport congestion window
@@ -345,6 +425,7 @@ void xprt_adjust_cwnd(struct rpc_task *task, int result)
 	xprt->cwnd = cwnd;
 	__xprt_put_cong(xprt, req);
 }
+EXPORT_SYMBOL_GPL(xprt_adjust_cwnd);
 
 /**
  * xprt_wake_pending_tasks - wake all tasks on a transport's pending queue
@@ -359,6 +440,7 @@ void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status)
 	else
 		rpc_wake_up(&xprt->pending);
 }
+EXPORT_SYMBOL_GPL(xprt_wake_pending_tasks);
 
 /**
  * xprt_wait_for_buffer_space - wait for transport output buffer to clear
@@ -373,6 +455,7 @@ void xprt_wait_for_buffer_space(struct rpc_task *task)
 	task->tk_timeout = req->rq_timeout;
 	rpc_sleep_on(&xprt->pending, task, NULL, NULL);
 }
+EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space);
 
 /**
  * xprt_write_space - wake the task waiting for transport output buffer space
@@ -393,6 +476,7 @@ void xprt_write_space(struct rpc_xprt *xprt)
 	}
 	spin_unlock_bh(&xprt->transport_lock);
 }
+EXPORT_SYMBOL_GPL(xprt_write_space);
 
 /**
  * xprt_set_retrans_timeout_def - set a request's retransmit timeout
@@ -406,6 +490,7 @@ void xprt_set_retrans_timeout_def(struct rpc_task *task)
 {
 	task->tk_timeout = task->tk_rqstp->rq_timeout;
 }
+EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_def);
 
 /*
  * xprt_set_retrans_timeout_rtt - set a request's retransmit timeout
@@ -425,6 +510,7 @@ void xprt_set_retrans_timeout_rtt(struct rpc_task *task)
 	if (task->tk_timeout > max_timeout || task->tk_timeout == 0)
 		task->tk_timeout = max_timeout;
 }
+EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_rtt);
 
 static void xprt_reset_majortimeo(struct rpc_rqst *req)
 {
@@ -500,6 +586,7 @@ void xprt_disconnect(struct rpc_xprt *xprt)
 	xprt_wake_pending_tasks(xprt, -ENOTCONN);
 	spin_unlock_bh(&xprt->transport_lock);
 }
+EXPORT_SYMBOL_GPL(xprt_disconnect);
 
 static void
 xprt_init_autodisconnect(unsigned long data)
@@ -610,6 +697,7 @@ struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 	xprt->stat.bad_xids++;
 	return NULL;
 }
+EXPORT_SYMBOL_GPL(xprt_lookup_rqst);
 
 /**
  * xprt_update_rtt - update an RPC client's RTT state after receiving a reply
@@ -629,6 +717,7 @@ void xprt_update_rtt(struct rpc_task *task)
 		rpc_set_timeo(rtt, timer, req->rq_ntrans - 1);
 	}
 }
+EXPORT_SYMBOL_GPL(xprt_update_rtt);
 
 /**
  * xprt_complete_rqst - called when reply processing is complete
@@ -653,6 +742,7 @@ void xprt_complete_rqst(struct rpc_task *task, int copied)
 	req->rq_received = req->rq_private_buf.len = copied;
 	rpc_wake_up_task(task);
 }
+EXPORT_SYMBOL_GPL(xprt_complete_rqst);
 
 static void xprt_timer(struct rpc_task *task)
 {
@@ -889,23 +979,25 @@ void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long i
  * @args: rpc transport creation arguments
  *
  */
-struct rpc_xprt *xprt_create_transport(struct rpc_xprtsock_create *args)
+struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
 {
 	struct rpc_xprt	*xprt;
 	struct rpc_rqst	*req;
+	struct xprt_class *t;
 
-	switch (args->proto) {
-	case IPPROTO_UDP:
-		xprt = xs_setup_udp(args);
-		break;
-	case IPPROTO_TCP:
-		xprt = xs_setup_tcp(args);
-		break;
-	default:
-		printk(KERN_ERR "RPC: unrecognized transport protocol: %d\n",
-				args->proto);
-		return ERR_PTR(-EIO);
+	spin_lock(&xprt_list_lock);
+	list_for_each_entry(t, &xprt_list, list) {
+		if (t->ident == args->ident) {
+			spin_unlock(&xprt_list_lock);
+			goto found;
 		}
+	}
+	spin_unlock(&xprt_list_lock);
+	printk(KERN_ERR "RPC: transport (%d) not supported\n", args->ident);
+	return ERR_PTR(-EIO);
+
+found:
+	xprt = t->setup(args);
 	if (IS_ERR(xprt)) {
 		dprintk("RPC:       xprt_create_transport: failed, %ld\n",
 				-PTR_ERR(xprt));

net/sunrpc/xprtrdma/Makefile

+obj-$(CONFIG_SUNRPC_XPRT_RDMA) += xprtrdma.o
+
+xprtrdma-y := transport.o rpc_rdma.o verbs.o