Fixed MTP to work with TWRP

fs/nfs/Kconfig

@@ -0,0 +1,197 @@
+config NFS_FS
+	tristate "NFS client support"
+	depends on INET && FILE_LOCKING
+	select LOCKD
+	select SUNRPC
+	select NFS_ACL_SUPPORT if NFS_V3_ACL
+	help
+	  Choose Y here if you want to access files residing on other
+	  computers using Sun's Network File System protocol.  To compile
+	  this file system support as a module, choose M here: the module
+	  will be called nfs.
+
+	  To mount file systems exported by NFS servers, you also need to
+	  install the user space mount.nfs command which can be found in
+	  the Linux nfs-utils package, available from http://linux-nfs.org/.
+	  Information about using the mount command is available in the
+	  mount(8) man page.  More detail about the Linux NFS client
+	  implementation is available via the nfs(5) man page.
+
+	  Below you can choose which versions of the NFS protocol are
+	  available in the kernel to mount NFS servers.  Support for NFS
+	  version 2 (RFC 1094) is always available when NFS_FS is selected.
+
+	  To configure a system which mounts its root file system via NFS
+	  at boot time, say Y here, select "Kernel level IP
+	  autoconfiguration" in the NETWORK menu, and select "Root file
+	  system on NFS" below.  You cannot compile this file system as a
+	  module in this case.
+
+	  If unsure, say N.
+
+config NFS_V2
+	tristate "NFS client support for NFS version 2"
+	depends on NFS_FS
+	default y
+	help
+	  This option enables support for version 2 of the NFS protocol
+	  (RFC 1094) in the kernel's NFS client.
+
+	  If unsure, say Y.
+
+config NFS_V3
+	tristate "NFS client support for NFS version 3"
+	depends on NFS_FS
+	default y
+	help
+	  This option enables support for version 3 of the NFS protocol
+	  (RFC 1813) in the kernel's NFS client.
+
+	  If unsure, say Y.
+
+config NFS_V3_ACL
+	bool "NFS client support for the NFSv3 ACL protocol extension"
+	depends on NFS_V3
+	help
+	  Some NFS servers support an auxiliary NFSv3 ACL protocol that
+	  Sun added to Solaris but never became an official part of the
+	  NFS version 3 protocol.  This protocol extension allows
+	  applications on NFS clients to manipulate POSIX Access Control
+	  Lists on files residing on NFS servers.  NFS servers enforce
+	  ACLs on local files whether this protocol is available or not.
+
+	  Choose Y here if your NFS server supports the Solaris NFSv3 ACL
+	  protocol extension and you want your NFS client to allow
+	  applications to access and modify ACLs on files on the server.
+
+	  Most NFS servers don't support the Solaris NFSv3 ACL protocol
+	  extension.  You can choose N here or specify the "noacl" mount
+	  option to prevent your NFS client from trying to use the NFSv3
+	  ACL protocol.
+
+	  If unsure, say N.
+
+config NFS_V4
+	tristate "NFS client support for NFS version 4"
+	depends on NFS_FS
+	select SUNRPC_GSS
+	select KEYS
+	help
+	  This option enables support for version 4 of the NFS protocol
+	  (RFC 3530) in the kernel's NFS client.
+
+	  To mount NFS servers using NFSv4, you also need to install user
+	  space programs which can be found in the Linux nfs-utils package,
+	  available from http://linux-nfs.org/.
+
+	  If unsure, say Y.
+
+config NFS_SWAP
+	bool "Provide swap over NFS support"
+	default n
+	depends on NFS_FS
+	select SUNRPC_SWAP
+	help
+	  This option enables swapon to work on files located on NFS mounts.
+
+config NFS_V4_1
+	bool "NFS client support for NFSv4.1"
+	depends on NFS_V4
+	select SUNRPC_BACKCHANNEL
+	help
+	  This option enables support for minor version 1 of the NFSv4 protocol
+	  (RFC 5661) in the kernel's NFS client.
+
+	  If unsure, say N.
+
+config NFS_V4_2
+	bool "NFS client support for NFSv4.2"
+	depends on NFS_V4_1
+	help
+	  This option enables support for minor version 2 of the NFSv4 protocol
+	  in the kernel's NFS client.
+
+	  If unsure, say N.
+
+config PNFS_FILE_LAYOUT
+	tristate
+	depends on NFS_V4_1
+	default NFS_V4
+
+config PNFS_BLOCK
+	tristate
+	depends on NFS_V4_1 && BLK_DEV_DM
+	default NFS_V4
+
+config PNFS_OBJLAYOUT
+	tristate
+	depends on NFS_V4_1 && SCSI_OSD_ULD
+	default NFS_V4
+
+config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN
+	string "NFSv4.1 Implementation ID Domain"
+	depends on NFS_V4_1
+	default "kernel.org"
+	help
+	  This option defines the domain portion of the implementation ID that
+	  may be sent in the NFS exchange_id operation.  The value must be in
+	  the format of a DNS domain name and should be set to the DNS domain
+	  name of the distribution.
+	  If the NFS client is unchanged from the upstream kernel, this
+	  option should be set to the default "kernel.org".
+
+config NFS_V4_1_MIGRATION
+	bool "NFSv4.1 client support for migration"
+	depends on NFS_V4_1
+	default n
+	help
+	  This option makes the NFS client advertise to NFSv4.1 servers that
+          it can support NFSv4 migration.
+
+          The NFSv4.1 pieces of the Linux NFSv4 migration implementation are
+          still experimental.  If you are not an NFSv4 developer, say N here.
+
+config NFS_V4_SECURITY_LABEL
+	bool
+	depends on NFS_V4_2 && SECURITY
+	default y
+
+config ROOT_NFS
+	bool "Root file system on NFS"
+	depends on NFS_FS=y && IP_PNP
+	help
+	  If you want your system to mount its root file system via NFS,
+	  choose Y here.  This is common practice for managing systems
+	  without local permanent storage.  For details, read
+	  <file:Documentation/filesystems/nfs/nfsroot.txt>.
+
+	  Most people say N here.
+
+config NFS_FSCACHE
+	bool "Provide NFS client caching support"
+	depends on NFS_FS=m && FSCACHE || NFS_FS=y && FSCACHE=y
+	help
+	  Say Y here if you want NFS data to be cached locally on disc through
+	  the general filesystem cache manager
+
+config NFS_USE_LEGACY_DNS
+	bool "Use the legacy NFS DNS resolver"
+	depends on NFS_V4
+	help
+	  The kernel now provides a method for translating a host name into an
+	  IP address.  Select Y here if you would rather use your own DNS
+	  resolver script.
+
+	  If unsure, say N
+
+config NFS_USE_KERNEL_DNS
+	bool
+	depends on NFS_V4 && !NFS_USE_LEGACY_DNS
+	select DNS_RESOLVER
+	default y
+
+config NFS_DEBUG
+	bool
+	depends on NFS_FS && SUNRPC_DEBUG
+	select CRC32
+	default y

fs/nfs/Makefile

@@ -0,0 +1,35 @@
+#
+# Makefile for the Linux nfs filesystem routines.
+#
+
+obj-$(CONFIG_NFS_FS) += nfs.o
+
+CFLAGS_nfstrace.o += -I$(src)
+nfs-y 			:= client.o dir.o file.o getroot.o inode.o super.o \
+			   direct.o pagelist.o read.o symlink.o unlink.o \
+			   write.o namespace.o mount_clnt.o nfstrace.o
+nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o
+nfs-$(CONFIG_SYSCTL)	+= sysctl.o
+nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o
+
+obj-$(CONFIG_NFS_V2) += nfsv2.o
+nfsv2-y := nfs2super.o proc.o nfs2xdr.o
+
+obj-$(CONFIG_NFS_V3) += nfsv3.o
+nfsv3-y := nfs3super.o nfs3client.o nfs3proc.o nfs3xdr.o
+nfsv3-$(CONFIG_NFS_V3_ACL) += nfs3acl.o
+
+obj-$(CONFIG_NFS_V4) += nfsv4.o
+CFLAGS_nfs4trace.o += -I$(src)
+nfsv4-y := nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o nfs4super.o nfs4file.o \
+	  delegation.o idmap.o callback.o callback_xdr.o callback_proc.o \
+	  nfs4namespace.o nfs4getroot.o nfs4client.o nfs4session.o \
+	  dns_resolve.o nfs4trace.o
+nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o
+nfsv4-$(CONFIG_SYSCTL)	+= nfs4sysctl.o
+nfsv4-$(CONFIG_NFS_V4_1)	+= pnfs.o pnfs_dev.o
+nfsv4-$(CONFIG_NFS_V4_2)	+= nfs42proc.o
+
+obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/
+obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayout/
+obj-$(CONFIG_PNFS_BLOCK) += blocklayout/

fs/nfs/blocklayout/Makefile

@@ -0,0 +1,6 @@
+#
+# Makefile for the pNFS block layout driver kernel module
+#
+obj-$(CONFIG_PNFS_BLOCK) += blocklayoutdriver.o
+
+blocklayoutdriver-y += blocklayout.o dev.o extent_tree.o rpc_pipefs.o

fs/nfs/blocklayout/blocklayout.c

@@ -0,0 +1,925 @@
+/*
+ *  linux/fs/nfs/blocklayout/blocklayout.c
+ *
+ *  Module for the NFSv4.1 pNFS block layout driver.
+ *
+ *  Copyright (c) 2006 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/bio.h>		/* struct bio */
+#include <linux/prefetch.h>
+#include <linux/pagevec.h>
+
+#include "../pnfs.h"
+#include "../nfs4session.h"
+#include "../internal.h"
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY	NFSDBG_PNFS_LD
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Andy Adamson <andros@citi.umich.edu>");
+MODULE_DESCRIPTION("The NFSv4.1 pNFS Block layout driver");
+
+static bool is_hole(struct pnfs_block_extent *be)
+{
+	switch (be->be_state) {
+	case PNFS_BLOCK_NONE_DATA:
+		return true;
+	case PNFS_BLOCK_INVALID_DATA:
+		return be->be_tag ? false : true;
+	default:
+		return false;
+	}
+}
+
+/* The data we are handed might be spread across several bios.  We need
+ * to track when the last one is finished.
+ */
+struct parallel_io {
+	struct kref refcnt;
+	void (*pnfs_callback) (void *data);
+	void *data;
+};
+
+static inline struct parallel_io *alloc_parallel(void *data)
+{
+	struct parallel_io *rv;
+
+	rv  = kmalloc(sizeof(*rv), GFP_NOFS);
+	if (rv) {
+		rv->data = data;
+		kref_init(&rv->refcnt);
+	}
+	return rv;
+}
+
+static inline void get_parallel(struct parallel_io *p)
+{
+	kref_get(&p->refcnt);
+}
+
+static void destroy_parallel(struct kref *kref)
+{
+	struct parallel_io *p = container_of(kref, struct parallel_io, refcnt);
+
+	dprintk("%s enter\n", __func__);
+	p->pnfs_callback(p->data);
+	kfree(p);
+}
+
+static inline void put_parallel(struct parallel_io *p)
+{
+	kref_put(&p->refcnt, destroy_parallel);
+}
+
+static struct bio *
+bl_submit_bio(int rw, struct bio *bio)
+{
+	if (bio) {
+		get_parallel(bio->bi_private);
+		dprintk("%s submitting %s bio %u@%llu\n", __func__,
+			rw == READ ? "read" : "write", bio->bi_iter.bi_size,
+			(unsigned long long)bio->bi_iter.bi_sector);
+		submit_bio(rw, bio);
+	}
+	return NULL;
+}
+
+static struct bio *
+bl_alloc_init_bio(int npg, struct block_device *bdev, sector_t disk_sector,
+		void (*end_io)(struct bio *, int err), struct parallel_io *par)
+{
+	struct bio *bio;
+
+	npg = min(npg, BIO_MAX_PAGES);
+	bio = bio_alloc(GFP_NOIO, npg);
+	if (!bio && (current->flags & PF_MEMALLOC)) {
+		while (!bio && (npg /= 2))
+			bio = bio_alloc(GFP_NOIO, npg);
+	}
+
+	if (bio) {
+		bio->bi_iter.bi_sector = disk_sector;
+		bio->bi_bdev = bdev;
+		bio->bi_end_io = end_io;
+		bio->bi_private = par;
+	}
+	return bio;
+}
+
+static struct bio *
+do_add_page_to_bio(struct bio *bio, int npg, int rw, sector_t isect,
+		struct page *page, struct pnfs_block_dev_map *map,
+		struct pnfs_block_extent *be,
+		void (*end_io)(struct bio *, int err),
+		struct parallel_io *par, unsigned int offset, int *len)
+{
+	struct pnfs_block_dev *dev =
+		container_of(be->be_device, struct pnfs_block_dev, node);
+	u64 disk_addr, end;
+
+	dprintk("%s: npg %d rw %d isect %llu offset %u len %d\n", __func__,
+		npg, rw, (unsigned long long)isect, offset, *len);
+
+	/* translate to device offset */
+	isect += be->be_v_offset;
+	isect -= be->be_f_offset;
+
+	/* translate to physical disk offset */
+	disk_addr = (u64)isect << SECTOR_SHIFT;
+	if (disk_addr < map->start || disk_addr >= map->start + map->len) {
+		if (!dev->map(dev, disk_addr, map))
+			return ERR_PTR(-EIO);
+		bio = bl_submit_bio(rw, bio);
+	}
+	disk_addr += map->disk_offset;
+	disk_addr -= map->start;
+
+	/* limit length to what the device mapping allows */
+	end = disk_addr + *len;
+	if (end >= map->start + map->len)
+		*len = map->start + map->len - disk_addr;
+
+retry:
+	if (!bio) {
+		bio = bl_alloc_init_bio(npg, map->bdev,
+				disk_addr >> SECTOR_SHIFT, end_io, par);
+		if (!bio)
+			return ERR_PTR(-ENOMEM);
+	}
+	if (bio_add_page(bio, page, *len, offset) < *len) {
+		bio = bl_submit_bio(rw, bio);
+		goto retry;
+	}
+	return bio;
+}
+
+static void bl_end_io_read(struct bio *bio, int err)
+{
+	struct parallel_io *par = bio->bi_private;
+
+	if (err) {
+		struct nfs_pgio_header *header = par->data;
+
+		if (!header->pnfs_error)
+			header->pnfs_error = -EIO;
+		pnfs_set_lo_fail(header->lseg);
+	}
+
+	bio_put(bio);
+	put_parallel(par);
+}
+
+static void bl_read_cleanup(struct work_struct *work)
+{
+	struct rpc_task *task;
+	struct nfs_pgio_header *hdr;
+	dprintk("%s enter\n", __func__);
+	task = container_of(work, struct rpc_task, u.tk_work);
+	hdr = container_of(task, struct nfs_pgio_header, task);
+	pnfs_ld_read_done(hdr);
+}
+
+static void
+bl_end_par_io_read(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	hdr->task.tk_status = hdr->pnfs_error;
+	INIT_WORK(&hdr->task.u.tk_work, bl_read_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
+}
+
+static enum pnfs_try_status
+bl_read_pagelist(struct nfs_pgio_header *header)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 };
+	struct bio *bio = NULL;
+	struct pnfs_block_extent be;
+	sector_t isect, extent_length = 0;
+	struct parallel_io *par;
+	loff_t f_offset = header->args.offset;
+	size_t bytes_left = header->args.count;
+	unsigned int pg_offset, pg_len;
+	struct page **pages = header->args.pages;
+	int pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT;
+	const bool is_dio = (header->dreq != NULL);
+	struct blk_plug plug;
+	int i;
+
+	dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__,
+		header->page_array.npages, f_offset,
+		(unsigned int)header->args.count);
+
+	par = alloc_parallel(header);
+	if (!par)
+		return PNFS_NOT_ATTEMPTED;
+	par->pnfs_callback = bl_end_par_io_read;
+
+	blk_start_plug(&plug);
+
+	isect = (sector_t) (f_offset >> SECTOR_SHIFT);
+	/* Code assumes extents are page-aligned */
+	for (i = pg_index; i < header->page_array.npages; i++) {
+		if (extent_length <= 0) {
+			/* We've used up the previous extent */
+			bio = bl_submit_bio(READ, bio);
+
+			/* Get the next one */
+			if (!ext_tree_lookup(bl, isect, &be, false)) {
+				header->pnfs_error = -EIO;
+				goto out;
+			}
+			extent_length = be.be_length - (isect - be.be_f_offset);
+		}
+
+		pg_offset = f_offset & ~PAGE_CACHE_MASK;
+		if (is_dio) {
+			if (pg_offset + bytes_left > PAGE_CACHE_SIZE)
+				pg_len = PAGE_CACHE_SIZE - pg_offset;
+			else
+				pg_len = bytes_left;
+		} else {
+			BUG_ON(pg_offset != 0);
+			pg_len = PAGE_CACHE_SIZE;
+		}
+
+		isect += (pg_offset >> SECTOR_SHIFT);
+		extent_length -= (pg_offset >> SECTOR_SHIFT);
+
+		if (is_hole(&be)) {
+			bio = bl_submit_bio(READ, bio);
+			/* Fill hole w/ zeroes w/o accessing device */
+			dprintk("%s Zeroing page for hole\n", __func__);
+			zero_user_segment(pages[i], pg_offset, pg_len);
+
+			/* invalidate map */
+			map.start = NFS4_MAX_UINT64;
+		} else {
+			bio = do_add_page_to_bio(bio,
+						 header->page_array.npages - i,
+						 READ,
+						 isect, pages[i], &map, &be,
+						 bl_end_io_read, par,
+						 pg_offset, &pg_len);
+			if (IS_ERR(bio)) {
+				header->pnfs_error = PTR_ERR(bio);
+				bio = NULL;
+				goto out;
+			}
+		}
+		isect += (pg_len >> SECTOR_SHIFT);
+		extent_length -= (pg_len >> SECTOR_SHIFT);
+		f_offset += pg_len;
+		bytes_left -= pg_len;
+	}
+	if ((isect << SECTOR_SHIFT) >= header->inode->i_size) {
+		header->res.eof = 1;
+		header->res.count = header->inode->i_size - header->args.offset;
+	} else {
+		header->res.count = (isect << SECTOR_SHIFT) - header->args.offset;
+	}
+out:
+	bl_submit_bio(READ, bio);
+	blk_finish_plug(&plug);
+	put_parallel(par);
+	return PNFS_ATTEMPTED;
+}
+
+static void bl_end_io_write(struct bio *bio, int err)
+{
+	struct parallel_io *par = bio->bi_private;
+	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
+	struct nfs_pgio_header *header = par->data;
+
+	if (!uptodate) {
+		if (!header->pnfs_error)
+			header->pnfs_error = -EIO;
+		pnfs_set_lo_fail(header->lseg);
+	}
+	bio_put(bio);
+	put_parallel(par);
+}
+
+/* Function scheduled for call during bl_end_par_io_write,
+ * it marks sectors as written and extends the commitlist.
+ */
+static void bl_write_cleanup(struct work_struct *work)
+{
+	struct rpc_task *task = container_of(work, struct rpc_task, u.tk_work);
+	struct nfs_pgio_header *hdr =
+			container_of(task, struct nfs_pgio_header, task);
+
+	dprintk("%s enter\n", __func__);
+
+	if (likely(!hdr->pnfs_error)) {
+		struct pnfs_block_layout *bl = BLK_LSEG2EXT(hdr->lseg);
+		u64 start = hdr->args.offset & (loff_t)PAGE_CACHE_MASK;
+		u64 end = (hdr->args.offset + hdr->args.count +
+			PAGE_CACHE_SIZE - 1) & (loff_t)PAGE_CACHE_MASK;
+
+		ext_tree_mark_written(bl, start >> SECTOR_SHIFT,
+					(end - start) >> SECTOR_SHIFT);
+	}
+
+	pnfs_ld_write_done(hdr);
+}
+
+/* Called when last of bios associated with a bl_write_pagelist call finishes */
+static void bl_end_par_io_write(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	hdr->task.tk_status = hdr->pnfs_error;
+	hdr->verf.committed = NFS_FILE_SYNC;
+	INIT_WORK(&hdr->task.u.tk_work, bl_write_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
+}
+
+static enum pnfs_try_status
+bl_write_pagelist(struct nfs_pgio_header *header, int sync)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 };
+	struct bio *bio = NULL;
+	struct pnfs_block_extent be;
+	sector_t isect, extent_length = 0;
+	struct parallel_io *par = NULL;
+	loff_t offset = header->args.offset;
+	size_t count = header->args.count;
+	struct page **pages = header->args.pages;
+	int pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT;
+	unsigned int pg_len;
+	struct blk_plug plug;
+	int i;
+
+	dprintk("%s enter, %Zu@%lld\n", __func__, count, offset);
+
+	/* At this point, header->page_aray is a (sequential) list of nfs_pages.
+	 * We want to write each, and if there is an error set pnfs_error
+	 * to have it redone using nfs.
+	 */
+	par = alloc_parallel(header);
+	if (!par)
+		return PNFS_NOT_ATTEMPTED;
+	par->pnfs_callback = bl_end_par_io_write;
+
+	blk_start_plug(&plug);
+
+	/* we always write out the whole page */
+	offset = offset & (loff_t)PAGE_CACHE_MASK;
+	isect = offset >> SECTOR_SHIFT;
+
+	for (i = pg_index; i < header->page_array.npages; i++) {
+		if (extent_length <= 0) {
+			/* We've used up the previous extent */
+			bio = bl_submit_bio(WRITE, bio);
+			/* Get the next one */
+			if (!ext_tree_lookup(bl, isect, &be, true)) {
+				header->pnfs_error = -EINVAL;
+				goto out;
+			}
+
+			extent_length = be.be_length - (isect - be.be_f_offset);
+		}
+
+		pg_len = PAGE_CACHE_SIZE;
+		bio = do_add_page_to_bio(bio, header->page_array.npages - i,
+					 WRITE, isect, pages[i], &map, &be,
+					 bl_end_io_write, par,
+					 0, &pg_len);
+		if (IS_ERR(bio)) {
+			header->pnfs_error = PTR_ERR(bio);
+			bio = NULL;
+			goto out;
+		}
+
+		offset += pg_len;
+		count -= pg_len;
+		isect += (pg_len >> SECTOR_SHIFT);
+		extent_length -= (pg_len >> SECTOR_SHIFT);
+	}
+
+	header->res.count = header->args.count;
+out:
+	bl_submit_bio(WRITE, bio);
+	blk_finish_plug(&plug);
+	put_parallel(par);
+	return PNFS_ATTEMPTED;
+}
+
+static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	int err;
+
+	dprintk("%s enter\n", __func__);
+
+	err = ext_tree_remove(bl, true, 0, LLONG_MAX);
+	WARN_ON(err);
+
+	kfree(bl);
+}
+
+static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode,
+						   gfp_t gfp_flags)
+{
+	struct pnfs_block_layout *bl;
+
+	dprintk("%s enter\n", __func__);
+	bl = kzalloc(sizeof(*bl), gfp_flags);
+	if (!bl)
+		return NULL;
+
+	bl->bl_ext_rw = RB_ROOT;
+	bl->bl_ext_ro = RB_ROOT;
+	spin_lock_init(&bl->bl_ext_lock);
+
+	return &bl->bl_layout;
+}
+
+static void bl_free_lseg(struct pnfs_layout_segment *lseg)
+{
+	dprintk("%s enter\n", __func__);
+	kfree(lseg);
+}
+
+/* Tracks info needed to ensure extents in layout obey constraints of spec */
+struct layout_verification {
+	u32 mode;	/* R or RW */
+	u64 start;	/* Expected start of next non-COW extent */
+	u64 inval;	/* Start of INVAL coverage */
+	u64 cowread;	/* End of COW read coverage */
+};
+
+/* Verify the extent meets the layout requirements of the pnfs-block draft,
+ * section 2.3.1.
+ */
+static int verify_extent(struct pnfs_block_extent *be,
+			 struct layout_verification *lv)
+{
+	if (lv->mode == IOMODE_READ) {
+		if (be->be_state == PNFS_BLOCK_READWRITE_DATA ||
+		    be->be_state == PNFS_BLOCK_INVALID_DATA)
+			return -EIO;
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		return 0;
+	}
+	/* lv->mode == IOMODE_RW */
+	if (be->be_state == PNFS_BLOCK_READWRITE_DATA) {
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		if (lv->cowread > lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		lv->inval = lv->start;
+		return 0;
+	} else if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		return 0;
+	} else if (be->be_state == PNFS_BLOCK_READ_DATA) {
+		if (be->be_f_offset > lv->start)
+			return -EIO;
+		if (be->be_f_offset < lv->inval)
+			return -EIO;
+		if (be->be_f_offset < lv->cowread)
+			return -EIO;
+		/* It looks like you might want to min this with lv->start,
+		 * but you really don't.
+		 */
+		lv->inval = lv->inval + be->be_length;
+		lv->cowread = be->be_f_offset + be->be_length;
+		return 0;
+	} else
+		return -EIO;
+}
+
+static int decode_sector_number(__be32 **rp, sector_t *sp)
+{
+	uint64_t s;
+
+	*rp = xdr_decode_hyper(*rp, &s);
+	if (s & 0x1ff) {
+		printk(KERN_WARNING "NFS: %s: sector not aligned\n", __func__);
+		return -1;
+	}
+	*sp = s >> SECTOR_SHIFT;
+	return 0;
+}
+
+static int
+bl_alloc_extent(struct xdr_stream *xdr, struct pnfs_layout_hdr *lo,
+		struct layout_verification *lv, struct list_head *extents,
+		gfp_t gfp_mask)
+{
+	struct pnfs_block_extent *be;
+	struct nfs4_deviceid id;
+	int error;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, 28 + NFS4_DEVICEID4_SIZE);
+	if (!p)
+		return -EIO;
+
+	be = kzalloc(sizeof(*be), GFP_NOFS);
+	if (!be)
+		return -ENOMEM;
+
+	memcpy(&id, p, NFS4_DEVICEID4_SIZE);
+	p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE);
+
+	error = -EIO;
+	be->be_device = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), &id,
+						lo->plh_lc_cred, gfp_mask);
+	if (!be->be_device)
+		goto out_free_be;
+
+	/*
+	 * The next three values are read in as bytes, but stored in the
+	 * extent structure in 512-byte granularity.
+	 */
+	if (decode_sector_number(&p, &be->be_f_offset) < 0)
+		goto out_put_deviceid;
+	if (decode_sector_number(&p, &be->be_length) < 0)
+		goto out_put_deviceid;
+	if (decode_sector_number(&p, &be->be_v_offset) < 0)
+		goto out_put_deviceid;
+	be->be_state = be32_to_cpup(p++);
+
+	error = verify_extent(be, lv);
+	if (error) {
+		dprintk("%s: extent verification failed\n", __func__);
+		goto out_put_deviceid;
+	}
+
+	list_add_tail(&be->be_list, extents);
+	return 0;
+
+out_put_deviceid:
+	nfs4_put_deviceid_node(be->be_device);
+out_free_be:
+	kfree(be);
+	return error;
+}
+
+static struct pnfs_layout_segment *
+bl_alloc_lseg(struct pnfs_layout_hdr *lo, struct nfs4_layoutget_res *lgr,
+		gfp_t gfp_mask)
+{
+	struct layout_verification lv = {
+		.mode = lgr->range.iomode,
+		.start = lgr->range.offset >> SECTOR_SHIFT,
+		.inval = lgr->range.offset >> SECTOR_SHIFT,
+		.cowread = lgr->range.offset >> SECTOR_SHIFT,
+	};
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	struct pnfs_layout_segment *lseg;
+	struct xdr_buf buf;
+	struct xdr_stream xdr;
+	struct page *scratch;
+	int status, i;
+	uint32_t count;
+	__be32 *p;
+	LIST_HEAD(extents);
+
+	dprintk("---> %s\n", __func__);
+
+	lseg = kzalloc(sizeof(*lseg), gfp_mask);
+	if (!lseg)
+		return ERR_PTR(-ENOMEM);
+
+	status = -ENOMEM;
+	scratch = alloc_page(gfp_mask);
+	if (!scratch)
+		goto out;
+
+	xdr_init_decode_pages(&xdr, &buf,
+			lgr->layoutp->pages, lgr->layoutp->len);
+	xdr_set_scratch_buffer(&xdr, page_address(scratch), PAGE_SIZE);
+
+	status = -EIO;
+	p = xdr_inline_decode(&xdr, 4);
+	if (unlikely(!p))
+		goto out_free_scratch;
+
+	count = be32_to_cpup(p++);
+	dprintk("%s: number of extents %d\n", __func__, count);
+
+	/*
+	 * Decode individual extents, putting them in temporary staging area
+	 * until whole layout is decoded to make error recovery easier.
+	 */
+	for (i = 0; i < count; i++) {
+		status = bl_alloc_extent(&xdr, lo, &lv, &extents, gfp_mask);
+		if (status)
+			goto process_extents;
+	}
+
+	if (lgr->range.offset + lgr->range.length !=
+			lv.start << SECTOR_SHIFT) {
+		dprintk("%s Final length mismatch\n", __func__);
+		status = -EIO;
+		goto process_extents;
+	}
+
+	if (lv.start < lv.cowread) {
+		dprintk("%s Final uncovered COW extent\n", __func__);
+		status = -EIO;
+	}
+
+process_extents:
+	while (!list_empty(&extents)) {
+		struct pnfs_block_extent *be =
+			list_first_entry(&extents, struct pnfs_block_extent,
+					 be_list);
+		list_del(&be->be_list);
+
+		if (!status)
+			status = ext_tree_insert(bl, be);
+
+		if (status) {
+			nfs4_put_deviceid_node(be->be_device);
+			kfree(be);
+		}
+	}
+
+out_free_scratch:
+	__free_page(scratch);
+out:
+	dprintk("%s returns %d\n", __func__, status);
+	if (status) {
+		kfree(lseg);
+		return ERR_PTR(status);
+	}
+	return lseg;
+}
+
+static void
+bl_return_range(struct pnfs_layout_hdr *lo,
+		struct pnfs_layout_range *range)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	sector_t offset = range->offset >> SECTOR_SHIFT, end;
+
+	if (range->offset % 8) {
+		dprintk("%s: offset %lld not block size aligned\n",
+			__func__, range->offset);
+		return;
+	}
+
+	if (range->length != NFS4_MAX_UINT64) {
+		if (range->length % 8) {
+			dprintk("%s: length %lld not block size aligned\n",
+				__func__, range->length);
+			return;
+		}
+
+		end = offset + (range->length >> SECTOR_SHIFT);
+	} else {
+		end = round_down(NFS4_MAX_UINT64, PAGE_SIZE);
+	}
+
+	ext_tree_remove(bl, range->iomode & IOMODE_RW, offset, end);
+}
+
+static int
+bl_prepare_layoutcommit(struct nfs4_layoutcommit_args *arg)
+{
+	return ext_tree_prepare_commit(arg);
+}
+
+static void
+bl_cleanup_layoutcommit(struct nfs4_layoutcommit_data *lcdata)
+{
+	ext_tree_mark_committed(&lcdata->args, lcdata->res.status);
+}
+
+static int
+bl_set_layoutdriver(struct nfs_server *server, const struct nfs_fh *fh)
+{
+	dprintk("%s enter\n", __func__);
+
+	if (server->pnfs_blksize == 0) {
+		dprintk("%s Server did not return blksize\n", __func__);
+		return -EINVAL;
+	}
+	if (server->pnfs_blksize > PAGE_SIZE) {
+		printk(KERN_ERR "%s: pNFS blksize %d not supported.\n",
+			__func__, server->pnfs_blksize);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static bool
+is_aligned_req(struct nfs_pageio_descriptor *pgio,
+		struct nfs_page *req, unsigned int alignment)
+{
+	/*
+	 * Always accept buffered writes, higher layers take care of the
+	 * right alignment.
+	 */
+	if (pgio->pg_dreq == NULL)
+		return true;
+
+	if (!IS_ALIGNED(req->wb_offset, alignment))
+		return false;
+
+	if (IS_ALIGNED(req->wb_bytes, alignment))
+		return true;
+
+	if (req_offset(req) + req->wb_bytes == i_size_read(pgio->pg_inode)) {
+		/*
+		 * If the write goes up to the inode size, just write
+		 * the full page.  Data past the inode size is
+		 * guaranteed to be zeroed by the higher level client
+		 * code, and this behaviour is mandated by RFC 5663
+		 * section 2.3.2.
+		 */
+		return true;
+	}
+
+	return false;
+}
+
+static void
+bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE)) {
+		nfs_pageio_reset_read_mds(pgio);
+		return;
+	}
+
+	pnfs_generic_pg_init_read(pgio, req);
+}
+
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t
+bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+		struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE))
+		return 0;
+	return pnfs_generic_pg_test(pgio, prev, req);
+}
+
+/*
+ * Return the number of contiguous bytes for a given inode
+ * starting at page frame idx.
+ */
+static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx)
+{
+	struct address_space *mapping = inode->i_mapping;
+	pgoff_t end;
+
+	/* Optimize common case that writes from 0 to end of file */
+	end = DIV_ROUND_UP(i_size_read(inode), PAGE_CACHE_SIZE);
+	if (end != NFS_I(inode)->npages) {
+		rcu_read_lock();
+		end = page_cache_next_hole(mapping, idx + 1, ULONG_MAX);
+		rcu_read_unlock();
+	}
+
+	if (!end)
+		return i_size_read(inode) - (idx << PAGE_CACHE_SHIFT);
+	else
+		return (end - idx) << PAGE_CACHE_SHIFT;
+}
+
+static void
+bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	u64 wb_size;
+
+	if (!is_aligned_req(pgio, req, PAGE_SIZE)) {
+		nfs_pageio_reset_write_mds(pgio);
+		return;
+	}
+
+	if (pgio->pg_dreq == NULL)
+		wb_size = pnfs_num_cont_bytes(pgio->pg_inode,
+					      req->wb_index);
+	else
+		wb_size = nfs_dreq_bytes_left(pgio->pg_dreq);
+
+	pnfs_generic_pg_init_write(pgio, req, wb_size);
+}
+
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t
+bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+		 struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, PAGE_SIZE))
+		return 0;
+	return pnfs_generic_pg_test(pgio, prev, req);
+}
+
+static const struct nfs_pageio_ops bl_pg_read_ops = {
+	.pg_init = bl_pg_init_read,
+	.pg_test = bl_pg_test_read,
+	.pg_doio = pnfs_generic_pg_readpages,
+};
+
+static const struct nfs_pageio_ops bl_pg_write_ops = {
+	.pg_init = bl_pg_init_write,
+	.pg_test = bl_pg_test_write,
+	.pg_doio = pnfs_generic_pg_writepages,
+};
+
+static struct pnfs_layoutdriver_type blocklayout_type = {
+	.id				= LAYOUT_BLOCK_VOLUME,
+	.name				= "LAYOUT_BLOCK_VOLUME",
+	.owner				= THIS_MODULE,
+	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_READ_WHOLE_PAGE,
+	.read_pagelist			= bl_read_pagelist,
+	.write_pagelist			= bl_write_pagelist,
+	.alloc_layout_hdr		= bl_alloc_layout_hdr,
+	.free_layout_hdr		= bl_free_layout_hdr,
+	.alloc_lseg			= bl_alloc_lseg,
+	.free_lseg			= bl_free_lseg,
+	.return_range			= bl_return_range,
+	.prepare_layoutcommit		= bl_prepare_layoutcommit,
+	.cleanup_layoutcommit		= bl_cleanup_layoutcommit,
+	.set_layoutdriver		= bl_set_layoutdriver,
+	.alloc_deviceid_node		= bl_alloc_deviceid_node,
+	.free_deviceid_node		= bl_free_deviceid_node,
+	.pg_read_ops			= &bl_pg_read_ops,
+	.pg_write_ops			= &bl_pg_write_ops,
+};
+
+static int __init nfs4blocklayout_init(void)
+{
+	int ret;
+
+	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
+
+	ret = pnfs_register_layoutdriver(&blocklayout_type);
+	if (ret)
+		goto out;
+	ret = bl_init_pipefs();
+	if (ret)
+		goto out_unregister;
+	return 0;
+
+out_unregister:
+	pnfs_unregister_layoutdriver(&blocklayout_type);
+out:
+	return ret;
+}
+
+static void __exit nfs4blocklayout_exit(void)
+{
+	dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n",
+	       __func__);
+
+	bl_cleanup_pipefs();
+	pnfs_unregister_layoutdriver(&blocklayout_type);
+}
+
+MODULE_ALIAS("nfs-layouttype4-3");
+
+module_init(nfs4blocklayout_init);
+module_exit(nfs4blocklayout_exit);

fs/nfs/blocklayout/blocklayout.h

@@ -0,0 +1,204 @@
+/*
+ *  linux/fs/nfs/blocklayout/blocklayout.h
+ *
+ *  Module for the NFSv4.1 pNFS block layout driver.
+ *
+ *  Copyright (c) 2006 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+#ifndef FS_NFS_NFS4BLOCKLAYOUT_H
+#define FS_NFS_NFS4BLOCKLAYOUT_H
+
+#include <linux/device-mapper.h>
+#include <linux/nfs_fs.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+
+#include "../nfs4_fs.h"
+#include "../pnfs.h"
+#include "../netns.h"
+
+#define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT)
+#define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)
+#define SECTOR_SIZE (1 << SECTOR_SHIFT)
+
+struct pnfs_block_dev;
+
+enum pnfs_block_volume_type {
+	PNFS_BLOCK_VOLUME_SIMPLE	= 0,
+	PNFS_BLOCK_VOLUME_SLICE		= 1,
+	PNFS_BLOCK_VOLUME_CONCAT	= 2,
+	PNFS_BLOCK_VOLUME_STRIPE	= 3,
+};
+
+#define PNFS_BLOCK_MAX_UUIDS	4
+#define PNFS_BLOCK_MAX_DEVICES	64
+
+/*
+ * Random upper cap for the uuid length to avoid unbounded allocation.
+ * Not actually limited by the protocol.
+ */
+#define PNFS_BLOCK_UUID_LEN	128
+
+
+struct pnfs_block_volume {
+	enum pnfs_block_volume_type	type;
+	union {
+		struct {
+			int		len;
+			int		nr_sigs;
+			struct {
+				u64		offset;
+				u32		sig_len;
+				u8		sig[PNFS_BLOCK_UUID_LEN];
+			} sigs[PNFS_BLOCK_MAX_UUIDS];
+		} simple;
+		struct {
+			u64		start;
+			u64		len;
+			u32		volume;
+		} slice;
+		struct {
+			u32		volumes_count;
+			u32		volumes[PNFS_BLOCK_MAX_DEVICES];
+		} concat;
+		struct {
+			u64		chunk_size;
+			u32		volumes_count;
+			u32		volumes[PNFS_BLOCK_MAX_DEVICES];
+		} stripe;
+	};
+};
+
+struct pnfs_block_dev_map {
+	sector_t			start;
+	sector_t			len;
+
+	sector_t			disk_offset;
+	struct block_device		*bdev;
+};
+
+struct pnfs_block_dev {
+	struct nfs4_deviceid_node	node;
+
+	u64				start;
+	u64				len;
+
+	u32				nr_children;
+	struct pnfs_block_dev		*children;
+	u64				chunk_size;
+
+	struct block_device		*bdev;
+	u64				disk_offset;
+
+	bool (*map)(struct pnfs_block_dev *dev, u64 offset,
+			struct pnfs_block_dev_map *map);
+};
+
+enum exstate4 {
+	PNFS_BLOCK_READWRITE_DATA	= 0,
+	PNFS_BLOCK_READ_DATA		= 1,
+	PNFS_BLOCK_INVALID_DATA		= 2, /* mapped, but data is invalid */
+	PNFS_BLOCK_NONE_DATA		= 3  /* unmapped, it's a hole */
+};
+
+/* sector_t fields are all in 512-byte sectors */
+struct pnfs_block_extent {
+	union {
+		struct rb_node	be_node;
+		struct list_head be_list;
+	};
+	struct nfs4_deviceid_node *be_device;
+	sector_t	be_f_offset;	/* the starting offset in the file */
+	sector_t	be_length;	/* the size of the extent */
+	sector_t	be_v_offset;	/* the starting offset in the volume */
+	enum exstate4	be_state;	/* the state of this extent */
+#define EXTENT_WRITTEN		1
+#define EXTENT_COMMITTING	2
+	unsigned int	be_tag;
+};
+
+/* on the wire size of the extent */
+#define BL_EXTENT_SIZE	(7 * sizeof(__be32) + NFS4_DEVICEID4_SIZE)
+
+struct pnfs_block_layout {
+	struct pnfs_layout_hdr	bl_layout;
+	struct rb_root		bl_ext_rw;
+	struct rb_root		bl_ext_ro;
+	spinlock_t		bl_ext_lock;   /* Protects list manipulation */
+};
+
+static inline struct pnfs_block_layout *
+BLK_LO2EXT(struct pnfs_layout_hdr *lo)
+{
+	return container_of(lo, struct pnfs_block_layout, bl_layout);
+}
+
+static inline struct pnfs_block_layout *
+BLK_LSEG2EXT(struct pnfs_layout_segment *lseg)
+{
+	return BLK_LO2EXT(lseg->pls_layout);
+}
+
+struct bl_pipe_msg {
+	struct rpc_pipe_msg msg;
+	wait_queue_head_t *bl_wq;
+};
+
+struct bl_msg_hdr {
+	u8  type;
+	u16 totallen; /* length of entire message, including hdr itself */
+};
+
+#define BL_DEVICE_UMOUNT               0x0 /* Umount--delete devices */
+#define BL_DEVICE_MOUNT                0x1 /* Mount--create devices*/
+#define BL_DEVICE_REQUEST_INIT         0x0 /* Start request */
+#define BL_DEVICE_REQUEST_PROC         0x1 /* User level process succeeds */
+#define BL_DEVICE_REQUEST_ERR          0x2 /* User level process fails */
+
+/* dev.c */
+struct nfs4_deviceid_node *bl_alloc_deviceid_node(struct nfs_server *server,
+		struct pnfs_device *pdev, gfp_t gfp_mask);
+void bl_free_deviceid_node(struct nfs4_deviceid_node *d);
+
+/* extent_tree.c */
+int ext_tree_insert(struct pnfs_block_layout *bl,
+		struct pnfs_block_extent *new);
+int ext_tree_remove(struct pnfs_block_layout *bl, bool rw, sector_t start,
+		sector_t end);
+int ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
+		sector_t len);
+bool ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect,
+		struct pnfs_block_extent *ret, bool rw);
+int ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg);
+void ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status);
+
+/* rpc_pipefs.c */
+dev_t bl_resolve_deviceid(struct nfs_server *server,
+		struct pnfs_block_volume *b, gfp_t gfp_mask);
+int __init bl_init_pipefs(void);
+void __exit bl_cleanup_pipefs(void);
+
+#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */

fs/nfs/blocklayout/dev.c

@@ -0,0 +1,363 @@
+/*
+ * Copyright (c) 2014 Christoph Hellwig.
+ */
+#include <linux/sunrpc/svc.h>
+#include <linux/blkdev.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_xdr.h>
+
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS_LD
+
+static void
+bl_free_device(struct pnfs_block_dev *dev)
+{
+	if (dev->nr_children) {
+		int i;
+
+		for (i = 0; i < dev->nr_children; i++)
+			bl_free_device(&dev->children[i]);
+		kfree(dev->children);
+	} else {
+		if (dev->bdev)
+			blkdev_put(dev->bdev, FMODE_READ);
+	}
+}
+
+void
+bl_free_deviceid_node(struct nfs4_deviceid_node *d)
+{
+	struct pnfs_block_dev *dev =
+		container_of(d, struct pnfs_block_dev, node);
+
+	bl_free_device(dev);
+	kfree(dev);
+}
+
+static int
+nfs4_block_decode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b)
+{
+	__be32 *p;
+	int i;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (!p)
+		return -EIO;
+	b->type = be32_to_cpup(p++);
+
+	switch (b->type) {
+	case PNFS_BLOCK_VOLUME_SIMPLE:
+		p = xdr_inline_decode(xdr, 4);
+		if (!p)
+			return -EIO;
+		b->simple.nr_sigs = be32_to_cpup(p++);
+		if (!b->simple.nr_sigs) {
+			dprintk("no signature\n");
+			return -EIO;
+		}
+
+		b->simple.len = 4 + 4;
+		for (i = 0; i < b->simple.nr_sigs; i++) {
+			p = xdr_inline_decode(xdr, 8 + 4);
+			if (!p)
+				return -EIO;
+			p = xdr_decode_hyper(p, &b->simple.sigs[i].offset);
+			b->simple.sigs[i].sig_len = be32_to_cpup(p++);
+
+			p = xdr_inline_decode(xdr, b->simple.sigs[i].sig_len);
+			if (!p)
+				return -EIO;
+			memcpy(&b->simple.sigs[i].sig, p,
+				b->simple.sigs[i].sig_len);
+
+			b->simple.len += 8 + 4 + b->simple.sigs[i].sig_len;
+		}
+		break;
+	case PNFS_BLOCK_VOLUME_SLICE:
+		p = xdr_inline_decode(xdr, 8 + 8 + 4);
+		if (!p)
+			return -EIO;
+		p = xdr_decode_hyper(p, &b->slice.start);
+		p = xdr_decode_hyper(p, &b->slice.len);
+		b->slice.volume = be32_to_cpup(p++);
+		break;
+	case PNFS_BLOCK_VOLUME_CONCAT:
+		p = xdr_inline_decode(xdr, 4);
+		if (!p)
+			return -EIO;
+		b->concat.volumes_count = be32_to_cpup(p++);
+
+		p = xdr_inline_decode(xdr, b->concat.volumes_count * 4);
+		if (!p)
+			return -EIO;
+		for (i = 0; i < b->concat.volumes_count; i++)
+			b->concat.volumes[i] = be32_to_cpup(p++);
+		break;
+	case PNFS_BLOCK_VOLUME_STRIPE:
+		p = xdr_inline_decode(xdr, 8 + 4);
+		if (!p)
+			return -EIO;
+		p = xdr_decode_hyper(p, &b->stripe.chunk_size);
+		b->stripe.volumes_count = be32_to_cpup(p++);
+
+		p = xdr_inline_decode(xdr, b->stripe.volumes_count * 4);
+		if (!p)
+			return -EIO;
+		for (i = 0; i < b->stripe.volumes_count; i++)
+			b->stripe.volumes[i] = be32_to_cpup(p++);
+		break;
+	default:
+		dprintk("unknown volume type!\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static bool bl_map_simple(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	map->start = dev->start;
+	map->len = dev->len;
+	map->disk_offset = dev->disk_offset;
+	map->bdev = dev->bdev;
+	return true;
+}
+
+static bool bl_map_concat(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	int i;
+
+	for (i = 0; i < dev->nr_children; i++) {
+		struct pnfs_block_dev *child = &dev->children[i];
+
+		if (child->start > offset ||
+		    child->start + child->len <= offset)
+			continue;
+
+		child->map(child, offset - child->start, map);
+		return true;
+	}
+
+	dprintk("%s: ran off loop!\n", __func__);
+	return false;
+}
+
+static bool bl_map_stripe(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	struct pnfs_block_dev *child;
+	u64 chunk;
+	u32 chunk_idx;
+	u64 disk_offset;
+
+	chunk = div_u64(offset, dev->chunk_size);
+	div_u64_rem(chunk, dev->nr_children, &chunk_idx);
+
+	if (chunk_idx > dev->nr_children) {
+		dprintk("%s: invalid chunk idx %d (%lld/%lld)\n",
+			__func__, chunk_idx, offset, dev->chunk_size);
+		/* error, should not happen */
+		return false;
+	}
+
+	/* truncate offset to the beginning of the stripe */
+	offset = chunk * dev->chunk_size;
+
+	/* disk offset of the stripe */
+	disk_offset = div_u64(offset, dev->nr_children);
+
+	child = &dev->children[chunk_idx];
+	child->map(child, disk_offset, map);
+
+	map->start += offset;
+	map->disk_offset += disk_offset;
+	map->len = dev->chunk_size;
+	return true;
+}
+
+static int
+bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask);
+
+
+static int
+bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	dev_t dev;
+
+	dev = bl_resolve_deviceid(server, v, gfp_mask);
+	if (!dev)
+		return -EIO;
+
+	d->bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL);
+	if (IS_ERR(d->bdev)) {
+		printk(KERN_WARNING "pNFS: failed to open device %d:%d (%ld)\n",
+			MAJOR(dev), MINOR(dev), PTR_ERR(d->bdev));
+		return PTR_ERR(d->bdev);
+	}
+
+
+	d->len = i_size_read(d->bdev->bd_inode);
+	d->map = bl_map_simple;
+
+	printk(KERN_INFO "pNFS: using block device %s\n",
+		d->bdev->bd_disk->disk_name);
+	return 0;
+}
+
+static int
+bl_parse_slice(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	int ret;
+
+	ret = bl_parse_deviceid(server, d, volumes, v->slice.volume, gfp_mask);
+	if (ret)
+		return ret;
+
+	d->disk_offset = v->slice.start;
+	d->len = v->slice.len;
+	return 0;
+}
+
+static int
+bl_parse_concat(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	u64 len = 0;
+	int ret, i;
+
+	d->children = kcalloc(v->concat.volumes_count,
+			sizeof(struct pnfs_block_dev), GFP_KERNEL);
+	if (!d->children)
+		return -ENOMEM;
+
+	for (i = 0; i < v->concat.volumes_count; i++) {
+		ret = bl_parse_deviceid(server, &d->children[i],
+				volumes, v->concat.volumes[i], gfp_mask);
+		if (ret)
+			return ret;
+
+		d->nr_children++;
+		d->children[i].start += len;
+		len += d->children[i].len;
+	}
+
+	d->len = len;
+	d->map = bl_map_concat;
+	return 0;
+}
+
+static int
+bl_parse_stripe(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	u64 len = 0;
+	int ret, i;
+
+	d->children = kcalloc(v->stripe.volumes_count,
+			sizeof(struct pnfs_block_dev), GFP_KERNEL);
+	if (!d->children)
+		return -ENOMEM;
+
+	for (i = 0; i < v->stripe.volumes_count; i++) {
+		ret = bl_parse_deviceid(server, &d->children[i],
+				volumes, v->stripe.volumes[i], gfp_mask);
+		if (ret)
+			return ret;
+
+		d->nr_children++;
+		len += d->children[i].len;
+	}
+
+	d->len = len;
+	d->chunk_size = v->stripe.chunk_size;
+	d->map = bl_map_stripe;
+	return 0;
+}
+
+static int
+bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	switch (volumes[idx].type) {
+	case PNFS_BLOCK_VOLUME_SIMPLE:
+		return bl_parse_simple(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_SLICE:
+		return bl_parse_slice(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_CONCAT:
+		return bl_parse_concat(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_STRIPE:
+		return bl_parse_stripe(server, d, volumes, idx, gfp_mask);
+	default:
+		dprintk("unsupported volume type: %d\n", volumes[idx].type);
+		return -EIO;
+	}
+}
+
+struct nfs4_deviceid_node *
+bl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
+		gfp_t gfp_mask)
+{
+	struct nfs4_deviceid_node *node = NULL;
+	struct pnfs_block_volume *volumes;
+	struct pnfs_block_dev *top;
+	struct xdr_stream xdr;
+	struct xdr_buf buf;
+	struct page *scratch;
+	int nr_volumes, ret, i;
+	__be32 *p;
+
+	scratch = alloc_page(gfp_mask);
+	if (!scratch)
+		goto out;
+
+	xdr_init_decode_pages(&xdr, &buf, pdev->pages, pdev->pglen);
+	xdr_set_scratch_buffer(&xdr, page_address(scratch), PAGE_SIZE);
+
+	p = xdr_inline_decode(&xdr, sizeof(__be32));
+	if (!p)
+		goto out_free_scratch;
+	nr_volumes = be32_to_cpup(p++);
+
+	volumes = kcalloc(nr_volumes, sizeof(struct pnfs_block_volume),
+			  gfp_mask);
+	if (!volumes)
+		goto out_free_scratch;
+
+	for (i = 0; i < nr_volumes; i++) {
+		ret = nfs4_block_decode_volume(&xdr, &volumes[i]);
+		if (ret < 0)
+			goto out_free_volumes;
+	}
+
+	top = kzalloc(sizeof(*top), gfp_mask);
+	if (!top)
+		goto out_free_volumes;
+
+	ret = bl_parse_deviceid(server, top, volumes, nr_volumes - 1, gfp_mask);
+	if (ret) {
+		bl_free_device(top);
+		kfree(top);
+		goto out_free_volumes;
+	}
+
+	node = &top->node;
+	nfs4_init_deviceid_node(node, server, &pdev->dev_id);
+
+out_free_volumes:
+	kfree(volumes);
+out_free_scratch:
+	__free_page(scratch);
+out:
+	return node;
+}

fs/nfs/blocklayout/extent_tree.c

@@ -0,0 +1,602 @@
+/*
+ * Copyright (c) 2014 Christoph Hellwig.
+ */
+
+#include <linux/vmalloc.h>
+
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS_LD
+
+static inline struct pnfs_block_extent *
+ext_node(struct rb_node *node)
+{
+	return rb_entry(node, struct pnfs_block_extent, be_node);
+}
+
+static struct pnfs_block_extent *
+ext_tree_first(struct rb_root *root)
+{
+	struct rb_node *node = rb_first(root);
+	return node ? ext_node(node) : NULL;
+}
+
+static struct pnfs_block_extent *
+ext_tree_prev(struct pnfs_block_extent *be)
+{
+	struct rb_node *node = rb_prev(&be->be_node);
+	return node ? ext_node(node) : NULL;
+}
+
+static struct pnfs_block_extent *
+ext_tree_next(struct pnfs_block_extent *be)
+{
+	struct rb_node *node = rb_next(&be->be_node);
+	return node ? ext_node(node) : NULL;
+}
+
+static inline sector_t
+ext_f_end(struct pnfs_block_extent *be)
+{
+	return be->be_f_offset + be->be_length;
+}
+
+static struct pnfs_block_extent *
+__ext_tree_search(struct rb_root *root, sector_t start)
+{
+	struct rb_node *node = root->rb_node;
+	struct pnfs_block_extent *be = NULL;
+
+	while (node) {
+		be = ext_node(node);
+		if (start < be->be_f_offset)
+			node = node->rb_left;
+		else if (start >= ext_f_end(be))
+			node = node->rb_right;
+		else
+			return be;
+	}
+
+	if (be) {
+		if (start < be->be_f_offset)
+			return be;
+
+		if (start >= ext_f_end(be))
+			return ext_tree_next(be);
+	}
+
+	return NULL;
+}
+
+static bool
+ext_can_merge(struct pnfs_block_extent *be1, struct pnfs_block_extent *be2)
+{
+	if (be1->be_state != be2->be_state)
+		return false;
+	if (be1->be_device != be2->be_device)
+		return false;
+
+	if (be1->be_f_offset + be1->be_length != be2->be_f_offset)
+		return false;
+
+	if (be1->be_state != PNFS_BLOCK_NONE_DATA &&
+	    (be1->be_v_offset + be1->be_length != be2->be_v_offset))
+		return false;
+
+	if (be1->be_state == PNFS_BLOCK_INVALID_DATA &&
+	    be1->be_tag != be2->be_tag)
+		return false;
+
+	return true;
+}
+
+static struct pnfs_block_extent *
+ext_try_to_merge_left(struct rb_root *root, struct pnfs_block_extent *be)
+{
+	struct pnfs_block_extent *left = ext_tree_prev(be);
+
+	if (left && ext_can_merge(left, be)) {
+		left->be_length += be->be_length;
+		rb_erase(&be->be_node, root);
+		nfs4_put_deviceid_node(be->be_device);
+		kfree(be);
+		return left;
+	}
+
+	return be;
+}
+
+static struct pnfs_block_extent *
+ext_try_to_merge_right(struct rb_root *root, struct pnfs_block_extent *be)
+{
+	struct pnfs_block_extent *right = ext_tree_next(be);
+
+	if (right && ext_can_merge(be, right)) {
+		be->be_length += right->be_length;
+		rb_erase(&right->be_node, root);
+		nfs4_put_deviceid_node(right->be_device);
+		kfree(right);
+	}
+
+	return be;
+}
+
+static void
+__ext_tree_insert(struct rb_root *root,
+		struct pnfs_block_extent *new, bool merge_ok)
+{
+	struct rb_node **p = &root->rb_node, *parent = NULL;
+	struct pnfs_block_extent *be;
+
+	while (*p) {
+		parent = *p;
+		be = ext_node(parent);
+
+		if (new->be_f_offset < be->be_f_offset) {
+			if (merge_ok && ext_can_merge(new, be)) {
+				be->be_f_offset = new->be_f_offset;
+				if (be->be_state != PNFS_BLOCK_NONE_DATA)
+					be->be_v_offset = new->be_v_offset;
+				be->be_length += new->be_length;
+				be = ext_try_to_merge_left(root, be);
+				goto free_new;
+			}
+			p = &(*p)->rb_left;
+		} else if (new->be_f_offset >= ext_f_end(be)) {
+			if (merge_ok && ext_can_merge(be, new)) {
+				be->be_length += new->be_length;
+				be = ext_try_to_merge_right(root, be);
+				goto free_new;
+			}
+			p = &(*p)->rb_right;
+		} else {
+			BUG();
+		}
+	}
+
+	rb_link_node(&new->be_node, parent, p);
+	rb_insert_color(&new->be_node, root);
+	return;
+free_new:
+	nfs4_put_deviceid_node(new->be_device);
+	kfree(new);
+}
+
+static int
+__ext_tree_remove(struct rb_root *root, sector_t start, sector_t end)
+{
+	struct pnfs_block_extent *be;
+	sector_t len1 = 0, len2 = 0;
+	sector_t orig_v_offset;
+	sector_t orig_len;
+
+	be = __ext_tree_search(root, start);
+	if (!be)
+		return 0;
+	if (be->be_f_offset >= end)
+		return 0;
+
+	orig_v_offset = be->be_v_offset;
+	orig_len = be->be_length;
+
+	if (start > be->be_f_offset)
+		len1 = start - be->be_f_offset;
+	if (ext_f_end(be) > end)
+		len2 = ext_f_end(be) - end;
+
+	if (len2 > 0) {
+		if (len1 > 0) {
+			struct pnfs_block_extent *new;
+
+			new = kzalloc(sizeof(*new), GFP_ATOMIC);
+			if (!new)
+				return -ENOMEM;
+
+			be->be_length = len1;
+
+			new->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA) {
+				new->be_v_offset =
+					orig_v_offset + orig_len - len2;
+			}
+			new->be_length = len2;
+			new->be_state = be->be_state;
+			new->be_tag = be->be_tag;
+			new->be_device = nfs4_get_deviceid(be->be_device);
+
+			__ext_tree_insert(root, new, true);
+		} else {
+			be->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA) {
+				be->be_v_offset =
+					orig_v_offset + orig_len - len2;
+			}
+			be->be_length = len2;
+		}
+	} else {
+		if (len1 > 0) {
+			be->be_length = len1;
+			be = ext_tree_next(be);
+		}
+
+		while (be && ext_f_end(be) <= end) {
+			struct pnfs_block_extent *next = ext_tree_next(be);
+
+			rb_erase(&be->be_node, root);
+			nfs4_put_deviceid_node(be->be_device);
+			kfree(be);
+			be = next;
+		}
+
+		if (be && be->be_f_offset < end) {
+			len1 = ext_f_end(be) - end;
+			be->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA)
+				be->be_v_offset += be->be_length - len1;
+			be->be_length = len1;
+		}
+	}
+
+	return 0;
+}
+
+int
+ext_tree_insert(struct pnfs_block_layout *bl, struct pnfs_block_extent *new)
+{
+	struct pnfs_block_extent *be;
+	struct rb_root *root;
+	int err = 0;
+
+	switch (new->be_state) {
+	case PNFS_BLOCK_READWRITE_DATA:
+	case PNFS_BLOCK_INVALID_DATA:
+		root = &bl->bl_ext_rw;
+		break;
+	case PNFS_BLOCK_READ_DATA:
+	case PNFS_BLOCK_NONE_DATA:
+		root = &bl->bl_ext_ro;
+		break;
+	default:
+		dprintk("invalid extent type\n");
+		return -EINVAL;
+	}
+
+	spin_lock(&bl->bl_ext_lock);
+retry:
+	be = __ext_tree_search(root, new->be_f_offset);
+	if (!be || be->be_f_offset >= ext_f_end(new)) {
+		__ext_tree_insert(root, new, true);
+	} else if (new->be_f_offset >= be->be_f_offset) {
+		if (ext_f_end(new) <= ext_f_end(be)) {
+			nfs4_put_deviceid_node(new->be_device);
+			kfree(new);
+		} else {
+			sector_t new_len = ext_f_end(new) - ext_f_end(be);
+			sector_t diff = new->be_length - new_len;
+
+			new->be_f_offset += diff;
+			new->be_v_offset += diff;
+			new->be_length = new_len;
+			goto retry;
+		}
+	} else if (ext_f_end(new) <= ext_f_end(be)) {
+		new->be_length = be->be_f_offset - new->be_f_offset;
+		__ext_tree_insert(root, new, true);
+	} else {
+		struct pnfs_block_extent *split;
+		sector_t new_len = ext_f_end(new) - ext_f_end(be);
+		sector_t diff = new->be_length - new_len;
+
+		split = kmemdup(new, sizeof(*new), GFP_ATOMIC);
+		if (!split) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		split->be_length = be->be_f_offset - split->be_f_offset;
+		split->be_device = nfs4_get_deviceid(new->be_device);
+		__ext_tree_insert(root, split, true);
+
+		new->be_f_offset += diff;
+		new->be_v_offset += diff;
+		new->be_length = new_len;
+		goto retry;
+	}
+out:
+	spin_unlock(&bl->bl_ext_lock);
+	return err;
+}
+
+static bool
+__ext_tree_lookup(struct rb_root *root, sector_t isect,
+		struct pnfs_block_extent *ret)
+{
+	struct rb_node *node;
+	struct pnfs_block_extent *be;
+
+	node = root->rb_node;
+	while (node) {
+		be = ext_node(node);
+		if (isect < be->be_f_offset)
+			node = node->rb_left;
+		else if (isect >= ext_f_end(be))
+			node = node->rb_right;
+		else {
+			*ret = *be;
+			return true;
+		}
+	}
+
+	return false;
+}
+
+bool
+ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect,
+	    struct pnfs_block_extent *ret, bool rw)
+{
+	bool found = false;
+
+	spin_lock(&bl->bl_ext_lock);
+	if (!rw)
+		found = __ext_tree_lookup(&bl->bl_ext_ro, isect, ret);
+	if (!found)
+		found = __ext_tree_lookup(&bl->bl_ext_rw, isect, ret);
+	spin_unlock(&bl->bl_ext_lock);
+
+	return found;
+}
+
+int ext_tree_remove(struct pnfs_block_layout *bl, bool rw,
+		sector_t start, sector_t end)
+{
+	int err, err2;
+
+	spin_lock(&bl->bl_ext_lock);
+	err = __ext_tree_remove(&bl->bl_ext_ro, start, end);
+	if (rw) {
+		err2 = __ext_tree_remove(&bl->bl_ext_rw, start, end);
+		if (!err)
+			err = err2;
+	}
+	spin_unlock(&bl->bl_ext_lock);
+
+	return err;
+}
+
+static int
+ext_tree_split(struct rb_root *root, struct pnfs_block_extent *be,
+		sector_t split)
+{
+	struct pnfs_block_extent *new;
+	sector_t orig_len = be->be_length;
+
+	new = kzalloc(sizeof(*new), GFP_ATOMIC);
+	if (!new)
+		return -ENOMEM;
+
+	be->be_length = split - be->be_f_offset;
+
+	new->be_f_offset = split;
+	if (be->be_state != PNFS_BLOCK_NONE_DATA)
+		new->be_v_offset = be->be_v_offset + be->be_length;
+	new->be_length = orig_len - be->be_length;
+	new->be_state = be->be_state;
+	new->be_tag = be->be_tag;
+	new->be_device = nfs4_get_deviceid(be->be_device);
+
+	__ext_tree_insert(root, new, false);
+	return 0;
+}
+
+int
+ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
+		sector_t len)
+{
+	struct rb_root *root = &bl->bl_ext_rw;
+	sector_t end = start + len;
+	struct pnfs_block_extent *be;
+	int err = 0;
+
+	spin_lock(&bl->bl_ext_lock);
+	/*
+	 * First remove all COW extents or holes from written to range.
+	 */
+	err = __ext_tree_remove(&bl->bl_ext_ro, start, end);
+	if (err)
+		goto out;
+
+	/*
+	 * Then mark all invalid extents in the range as written to.
+	 */
+	for (be = __ext_tree_search(root, start); be; be = ext_tree_next(be)) {
+		if (be->be_f_offset >= end)
+			break;
+
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA || be->be_tag)
+			continue;
+
+		if (be->be_f_offset < start) {
+			struct pnfs_block_extent *left = ext_tree_prev(be);
+
+			if (left && ext_can_merge(left, be)) {
+				sector_t diff = start - be->be_f_offset;
+
+				left->be_length += diff;
+
+				be->be_f_offset += diff;
+				be->be_v_offset += diff;
+				be->be_length -= diff;
+			} else {
+				err = ext_tree_split(root, be, start);
+				if (err)
+					goto out;
+			}
+		}
+
+		if (ext_f_end(be) > end) {
+			struct pnfs_block_extent *right = ext_tree_next(be);
+
+			if (right && ext_can_merge(be, right)) {
+				sector_t diff = end - be->be_f_offset;
+
+				be->be_length -= diff;
+
+				right->be_f_offset -= diff;
+				right->be_v_offset -= diff;
+				right->be_length += diff;
+			} else {
+				err = ext_tree_split(root, be, end);
+				if (err)
+					goto out;
+			}
+		}
+
+		if (be->be_f_offset >= start && ext_f_end(be) <= end) {
+			be->be_tag = EXTENT_WRITTEN;
+			be = ext_try_to_merge_left(root, be);
+			be = ext_try_to_merge_right(root, be);
+		}
+	}
+out:
+	spin_unlock(&bl->bl_ext_lock);
+	return err;
+}
+
+static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
+		size_t buffer_size)
+{
+	if (arg->layoutupdate_pages != &arg->layoutupdate_page) {
+		int nr_pages = DIV_ROUND_UP(buffer_size, PAGE_SIZE), i;
+
+		for (i = 0; i < nr_pages; i++)
+			put_page(arg->layoutupdate_pages[i]);
+		kfree(arg->layoutupdate_pages);
+	} else {
+		put_page(arg->layoutupdate_page);
+	}
+}
+
+static int ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
+		size_t buffer_size, size_t *count)
+{
+	struct pnfs_block_extent *be;
+	int ret = 0;
+
+	spin_lock(&bl->bl_ext_lock);
+	for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_WRITTEN)
+			continue;
+
+		(*count)++;
+		if (*count * BL_EXTENT_SIZE > buffer_size) {
+			/* keep counting.. */
+			ret = -ENOSPC;
+			continue;
+		}
+
+		p = xdr_encode_opaque_fixed(p, be->be_device->deviceid.data,
+				NFS4_DEVICEID4_SIZE);
+		p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
+		p = xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
+		p = xdr_encode_hyper(p, 0LL);
+		*p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA);
+
+		be->be_tag = EXTENT_COMMITTING;
+	}
+	spin_unlock(&bl->bl_ext_lock);
+
+	return ret;
+}
+
+int
+ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout);
+	size_t count = 0, buffer_size = PAGE_SIZE;
+	__be32 *start_p;
+	int ret;
+
+	dprintk("%s enter\n", __func__);
+
+	arg->layoutupdate_page = alloc_page(GFP_NOFS);
+	if (!arg->layoutupdate_page)
+		return -ENOMEM;
+	start_p = page_address(arg->layoutupdate_page);
+	arg->layoutupdate_pages = &arg->layoutupdate_page;
+
+retry:
+	ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size, &count);
+	if (unlikely(ret)) {
+		ext_tree_free_commitdata(arg, buffer_size);
+
+		buffer_size = sizeof(__be32) + BL_EXTENT_SIZE * count;
+		count = 0;
+
+		arg->layoutupdate_pages =
+			kcalloc(DIV_ROUND_UP(buffer_size, PAGE_SIZE),
+				sizeof(struct page *), GFP_NOFS);
+		if (!arg->layoutupdate_pages)
+			return -ENOMEM;
+
+		start_p = __vmalloc(buffer_size, GFP_NOFS, PAGE_KERNEL);
+		if (!start_p) {
+			kfree(arg->layoutupdate_pages);
+			return -ENOMEM;
+		}
+
+		goto retry;
+	}
+
+	*start_p = cpu_to_be32(count);
+	arg->layoutupdate_len = sizeof(__be32) + BL_EXTENT_SIZE * count;
+
+	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
+		__be32 *p = start_p;
+		int i = 0;
+
+		for (p = start_p;
+		     p < start_p + arg->layoutupdate_len;
+		     p += PAGE_SIZE) {
+			arg->layoutupdate_pages[i++] = vmalloc_to_page(p);
+		}
+	}
+
+	dprintk("%s found %zu ranges\n", __func__, count);
+	return 0;
+}
+
+void
+ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout);
+	struct rb_root *root = &bl->bl_ext_rw;
+	struct pnfs_block_extent *be;
+
+	dprintk("%s status %d\n", __func__, status);
+
+	ext_tree_free_commitdata(arg, arg->layoutupdate_len);
+
+	spin_lock(&bl->bl_ext_lock);
+	for (be = ext_tree_first(root); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_COMMITTING)
+			continue;
+
+		if (status) {
+			/*
+			 * Mark as written and try again.
+			 *
+			 * XXX: some real error handling here wouldn't hurt..
+			 */
+			be->be_tag = EXTENT_WRITTEN;
+		} else {
+			be->be_state = PNFS_BLOCK_READWRITE_DATA;
+			be->be_tag = 0;
+		}
+
+		be = ext_try_to_merge_left(root, be);
+		be = ext_try_to_merge_right(root, be);
+	}
+	spin_unlock(&bl->bl_ext_lock);
+}

fs/nfs/blocklayout/rpc_pipefs.c

@@ -0,0 +1,288 @@
+/*
+ *  Copyright (c) 2006,2007 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+
+#include <linux/module.h>
+#include <linux/genhd.h>
+#include <linux/blkdev.h>
+
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
+
+static void
+nfs4_encode_simple(__be32 *p, struct pnfs_block_volume *b)
+{
+	int i;
+
+	*p++ = cpu_to_be32(1);
+	*p++ = cpu_to_be32(b->type);
+	*p++ = cpu_to_be32(b->simple.nr_sigs);
+	for (i = 0; i < b->simple.nr_sigs; i++) {
+		p = xdr_encode_hyper(p, b->simple.sigs[i].offset);
+		p = xdr_encode_opaque(p, b->simple.sigs[i].sig,
+					 b->simple.sigs[i].sig_len);
+	}
+}
+
+dev_t
+bl_resolve_deviceid(struct nfs_server *server, struct pnfs_block_volume *b,
+		gfp_t gfp_mask)
+{
+	struct net *net = server->nfs_client->cl_net;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct bl_dev_msg *reply = &nn->bl_mount_reply;
+	struct bl_pipe_msg bl_pipe_msg;
+	struct rpc_pipe_msg *msg = &bl_pipe_msg.msg;
+	struct bl_msg_hdr *bl_msg;
+	DECLARE_WAITQUEUE(wq, current);
+	dev_t dev = 0;
+	int rc;
+
+	dprintk("%s CREATING PIPEFS MESSAGE\n", __func__);
+
+	mutex_lock(&nn->bl_mutex);
+	bl_pipe_msg.bl_wq = &nn->bl_wq;
+
+	b->simple.len += 4;	/* single volume */
+	if (b->simple.len > PAGE_SIZE)
+		goto out_unlock;
+
+	memset(msg, 0, sizeof(*msg));
+	msg->len = sizeof(*bl_msg) + b->simple.len;
+	msg->data = kzalloc(msg->len, gfp_mask);
+	if (!msg->data)
+		goto out_free_data;
+
+	bl_msg = msg->data;
+	bl_msg->type = BL_DEVICE_MOUNT,
+	bl_msg->totallen = b->simple.len;
+	nfs4_encode_simple(msg->data + sizeof(*bl_msg), b);
+
+	dprintk("%s CALLING USERSPACE DAEMON\n", __func__);
+	add_wait_queue(&nn->bl_wq, &wq);
+	rc = rpc_queue_upcall(nn->bl_device_pipe, msg);
+	if (rc < 0) {
+		remove_wait_queue(&nn->bl_wq, &wq);
+		goto out_free_data;
+	}
+
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	schedule();
+	remove_wait_queue(&nn->bl_wq, &wq);
+
+	if (reply->status != BL_DEVICE_REQUEST_PROC) {
+		printk(KERN_WARNING "%s failed to decode device: %d\n",
+			__func__, reply->status);
+		goto out_free_data;
+	}
+
+	dev = MKDEV(reply->major, reply->minor);
+out_free_data:
+	kfree(msg->data);
+out_unlock:
+	mutex_unlock(&nn->bl_mutex);
+	return dev;
+}
+
+static ssize_t bl_pipe_downcall(struct file *filp, const char __user *src,
+			 size_t mlen)
+{
+	struct nfs_net *nn = net_generic(filp->f_dentry->d_sb->s_fs_info,
+					 nfs_net_id);
+
+	if (mlen != sizeof (struct bl_dev_msg))
+		return -EINVAL;
+
+	if (copy_from_user(&nn->bl_mount_reply, src, mlen) != 0)
+		return -EFAULT;
+
+	wake_up(&nn->bl_wq);
+
+	return mlen;
+}
+
+static void bl_pipe_destroy_msg(struct rpc_pipe_msg *msg)
+{
+	struct bl_pipe_msg *bl_pipe_msg =
+		container_of(msg, struct bl_pipe_msg, msg);
+
+	if (msg->errno >= 0)
+		return;
+	wake_up(bl_pipe_msg->bl_wq);
+}
+
+static const struct rpc_pipe_ops bl_upcall_ops = {
+	.upcall		= rpc_pipe_generic_upcall,
+	.downcall	= bl_pipe_downcall,
+	.destroy_msg	= bl_pipe_destroy_msg,
+};
+
+static struct dentry *nfs4blocklayout_register_sb(struct super_block *sb,
+					    struct rpc_pipe *pipe)
+{
+	struct dentry *dir, *dentry;
+
+	dir = rpc_d_lookup_sb(sb, NFS_PIPE_DIRNAME);
+	if (dir == NULL)
+		return ERR_PTR(-ENOENT);
+	dentry = rpc_mkpipe_dentry(dir, "blocklayout", NULL, pipe);
+	dput(dir);
+	return dentry;
+}
+
+static void nfs4blocklayout_unregister_sb(struct super_block *sb,
+					  struct rpc_pipe *pipe)
+{
+	if (pipe->dentry)
+		rpc_unlink(pipe->dentry);
+}
+
+static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,
+			   void *ptr)
+{
+	struct super_block *sb = ptr;
+	struct net *net = sb->s_fs_info;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct dentry *dentry;
+	int ret = 0;
+
+	if (!try_module_get(THIS_MODULE))
+		return 0;
+
+	if (nn->bl_device_pipe == NULL) {
+		module_put(THIS_MODULE);
+		return 0;
+	}
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		dentry = nfs4blocklayout_register_sb(sb, nn->bl_device_pipe);
+		if (IS_ERR(dentry)) {
+			ret = PTR_ERR(dentry);
+			break;
+		}
+		nn->bl_device_pipe->dentry = dentry;
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		if (nn->bl_device_pipe->dentry)
+			nfs4blocklayout_unregister_sb(sb, nn->bl_device_pipe);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+static struct notifier_block nfs4blocklayout_block = {
+	.notifier_call = rpc_pipefs_event,
+};
+
+static struct dentry *nfs4blocklayout_register_net(struct net *net,
+						   struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+	struct dentry *dentry;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (!pipefs_sb)
+		return NULL;
+	dentry = nfs4blocklayout_register_sb(pipefs_sb, pipe);
+	rpc_put_sb_net(net);
+	return dentry;
+}
+
+static void nfs4blocklayout_unregister_net(struct net *net,
+					   struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (pipefs_sb) {
+		nfs4blocklayout_unregister_sb(pipefs_sb, pipe);
+		rpc_put_sb_net(net);
+	}
+}
+
+static int nfs4blocklayout_net_init(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct dentry *dentry;
+
+	mutex_init(&nn->bl_mutex);
+	init_waitqueue_head(&nn->bl_wq);
+	nn->bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0);
+	if (IS_ERR(nn->bl_device_pipe))
+		return PTR_ERR(nn->bl_device_pipe);
+	dentry = nfs4blocklayout_register_net(net, nn->bl_device_pipe);
+	if (IS_ERR(dentry)) {
+		rpc_destroy_pipe_data(nn->bl_device_pipe);
+		return PTR_ERR(dentry);
+	}
+	nn->bl_device_pipe->dentry = dentry;
+	return 0;
+}
+
+static void nfs4blocklayout_net_exit(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	nfs4blocklayout_unregister_net(net, nn->bl_device_pipe);
+	rpc_destroy_pipe_data(nn->bl_device_pipe);
+	nn->bl_device_pipe = NULL;
+}
+
+static struct pernet_operations nfs4blocklayout_net_ops = {
+	.init = nfs4blocklayout_net_init,
+	.exit = nfs4blocklayout_net_exit,
+};
+
+int __init bl_init_pipefs(void)
+{
+	int ret;
+
+	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
+	if (ret)
+		goto out;
+	ret = register_pernet_subsys(&nfs4blocklayout_net_ops);
+	if (ret)
+		goto out_unregister_notifier;
+	return 0;
+
+out_unregister_notifier:
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
+out:
+	return ret;
+}
+
+void __exit bl_cleanup_pipefs(void)
+{
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
+	unregister_pernet_subsys(&nfs4blocklayout_net_ops);
+}

fs/nfs/cache_lib.c

@@ -0,0 +1,158 @@
+/*
+ * linux/fs/nfs/cache_lib.c
+ *
+ * Helper routines for the NFS client caches
+ *
+ * Copyright (c) 2009 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/kmod.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/slab.h>
+#include <linux/sunrpc/cache.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+#include <net/net_namespace.h>
+
+#include "cache_lib.h"
+
+#define NFS_CACHE_UPCALL_PATHLEN 256
+#define NFS_CACHE_UPCALL_TIMEOUT 15
+
+static char nfs_cache_getent_prog[NFS_CACHE_UPCALL_PATHLEN] =
+				"/sbin/nfs_cache_getent";
+static unsigned long nfs_cache_getent_timeout = NFS_CACHE_UPCALL_TIMEOUT;
+
+module_param_string(cache_getent, nfs_cache_getent_prog,
+		sizeof(nfs_cache_getent_prog), 0600);
+MODULE_PARM_DESC(cache_getent, "Path to the client cache upcall program");
+module_param_named(cache_getent_timeout, nfs_cache_getent_timeout, ulong, 0600);
+MODULE_PARM_DESC(cache_getent_timeout, "Timeout (in seconds) after which "
+		"the cache upcall is assumed to have failed");
+
+int nfs_cache_upcall(struct cache_detail *cd, char *entry_name)
+{
+	static char *envp[] = { "HOME=/",
+		"TERM=linux",
+		"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
+		NULL
+	};
+	char *argv[] = {
+		nfs_cache_getent_prog,
+		cd->name,
+		entry_name,
+		NULL
+	};
+	int ret = -EACCES;
+
+	if (nfs_cache_getent_prog[0] == '\0')
+		goto out;
+	ret = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC);
+	/*
+	 * Disable the upcall mechanism if we're getting an ENOENT or
+	 * EACCES error. The admin can re-enable it on the fly by using
+	 * sysfs to set the 'cache_getent' parameter once the problem
+	 * has been fixed.
+	 */
+	if (ret == -ENOENT || ret == -EACCES)
+		nfs_cache_getent_prog[0] = '\0';
+out:
+	return ret > 0 ? 0 : ret;
+}
+
+/*
+ * Deferred request handling
+ */
+void nfs_cache_defer_req_put(struct nfs_cache_defer_req *dreq)
+{
+	if (atomic_dec_and_test(&dreq->count))
+		kfree(dreq);
+}
+
+static void nfs_dns_cache_revisit(struct cache_deferred_req *d, int toomany)
+{
+	struct nfs_cache_defer_req *dreq;
+
+	dreq = container_of(d, struct nfs_cache_defer_req, deferred_req);
+
+	complete_all(&dreq->completion);
+	nfs_cache_defer_req_put(dreq);
+}
+
+static struct cache_deferred_req *nfs_dns_cache_defer(struct cache_req *req)
+{
+	struct nfs_cache_defer_req *dreq;
+
+	dreq = container_of(req, struct nfs_cache_defer_req, req);
+	dreq->deferred_req.revisit = nfs_dns_cache_revisit;
+	atomic_inc(&dreq->count);
+
+	return &dreq->deferred_req;
+}
+
+struct nfs_cache_defer_req *nfs_cache_defer_req_alloc(void)
+{
+	struct nfs_cache_defer_req *dreq;
+
+	dreq = kzalloc(sizeof(*dreq), GFP_KERNEL);
+	if (dreq) {
+		init_completion(&dreq->completion);
+		atomic_set(&dreq->count, 1);
+		dreq->req.defer = nfs_dns_cache_defer;
+	}
+	return dreq;
+}
+
+int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq)
+{
+	if (wait_for_completion_timeout(&dreq->completion,
+			nfs_cache_getent_timeout * HZ) == 0)
+		return -ETIMEDOUT;
+	return 0;
+}
+
+int nfs_cache_register_sb(struct super_block *sb, struct cache_detail *cd)
+{
+	int ret;
+	struct dentry *dir;
+
+	dir = rpc_d_lookup_sb(sb, "cache");
+	ret = sunrpc_cache_register_pipefs(dir, cd->name, 0600, cd);
+	dput(dir);
+	return ret;
+}
+
+int nfs_cache_register_net(struct net *net, struct cache_detail *cd)
+{
+	struct super_block *pipefs_sb;
+	int ret = 0;
+
+	sunrpc_init_cache_detail(cd);
+	pipefs_sb = rpc_get_sb_net(net);
+	if (pipefs_sb) {
+		ret = nfs_cache_register_sb(pipefs_sb, cd);
+		rpc_put_sb_net(net);
+		if (ret)
+			sunrpc_destroy_cache_detail(cd);
+	}
+	return ret;
+}
+
+void nfs_cache_unregister_sb(struct super_block *sb, struct cache_detail *cd)
+{
+	if (cd->u.pipefs.dir)
+		sunrpc_cache_unregister_pipefs(cd);
+}
+
+void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd)
+{
+	struct super_block *pipefs_sb;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (pipefs_sb) {
+		nfs_cache_unregister_sb(pipefs_sb, cd);
+		rpc_put_sb_net(net);
+	}
+	sunrpc_destroy_cache_detail(cd);
+}

fs/nfs/cache_lib.h

@@ -0,0 +1,31 @@
+/*
+ * Helper routines for the NFS client caches
+ *
+ * Copyright (c) 2009 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+
+#include <linux/completion.h>
+#include <linux/sunrpc/cache.h>
+#include <linux/atomic.h>
+
+/*
+ * Deferred request handling
+ */
+struct nfs_cache_defer_req {
+	struct cache_req req;
+	struct cache_deferred_req deferred_req;
+	struct completion completion;
+	atomic_t count;
+};
+
+extern int nfs_cache_upcall(struct cache_detail *cd, char *entry_name);
+extern struct nfs_cache_defer_req *nfs_cache_defer_req_alloc(void);
+extern void nfs_cache_defer_req_put(struct nfs_cache_defer_req *dreq);
+extern int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq);
+
+extern int nfs_cache_register_net(struct net *net, struct cache_detail *cd);
+extern void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd);
+extern int nfs_cache_register_sb(struct super_block *sb,
+				 struct cache_detail *cd);
+extern void nfs_cache_unregister_sb(struct super_block *sb,
+				    struct cache_detail *cd);

fs/nfs/callback.c

@@ -0,0 +1,499 @@
+/*
+ * linux/fs/nfs/callback.c
+ *
+ * Copyright (C) 2004 Trond Myklebust
+ *
+ * NFSv4 callback handling
+ */
+
+#include <linux/completion.h>
+#include <linux/ip.h>
+#include <linux/module.h>
+#include <linux/sunrpc/svc.h>
+#include <linux/sunrpc/svcsock.h>
+#include <linux/nfs_fs.h>
+#include <linux/errno.h>
+#include <linux/mutex.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
+#include <linux/sunrpc/svcauth_gss.h>
+#include <linux/sunrpc/bc_xprt.h>
+
+#include <net/inet_sock.h>
+
+#include "nfs4_fs.h"
+#include "callback.h"
+#include "internal.h"
+#include "netns.h"
+
+#define NFSDBG_FACILITY NFSDBG_CALLBACK
+
+struct nfs_callback_data {
+	unsigned int users;
+	struct svc_serv *serv;
+	struct svc_rqst *rqst;
+	struct task_struct *task;
+};
+
+static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1];
+static DEFINE_MUTEX(nfs_callback_mutex);
+static struct svc_program nfs4_callback_program;
+
+static int nfs4_callback_up_net(struct svc_serv *serv, struct net *net)
+{
+	int ret;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	ret = svc_create_xprt(serv, "tcp", net, PF_INET,
+				nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS);
+	if (ret <= 0)
+		goto out_err;
+	nn->nfs_callback_tcpport = ret;
+	dprintk("NFS: Callback listener port = %u (af %u, net %p)\n",
+			nn->nfs_callback_tcpport, PF_INET, net);
+
+	ret = svc_create_xprt(serv, "tcp", net, PF_INET6,
+				nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS);
+	if (ret > 0) {
+		nn->nfs_callback_tcpport6 = ret;
+		dprintk("NFS: Callback listener port = %u (af %u, net %p)\n",
+				nn->nfs_callback_tcpport6, PF_INET6, net);
+	} else if (ret != -EAFNOSUPPORT)
+		goto out_err;
+	return 0;
+
+out_err:
+	return (ret) ? ret : -ENOMEM;
+}
+
+/*
+ * This is the NFSv4 callback kernel thread.
+ */
+static int
+nfs4_callback_svc(void *vrqstp)
+{
+	int err;
+	struct svc_rqst *rqstp = vrqstp;
+
+	set_freezable();
+
+	while (!kthread_should_stop()) {
+		/*
+		 * Listen for a request on the socket
+		 */
+		err = svc_recv(rqstp, MAX_SCHEDULE_TIMEOUT);
+		if (err == -EAGAIN || err == -EINTR)
+			continue;
+		svc_process(rqstp);
+	}
+	return 0;
+}
+
+/*
+ * Prepare to bring up the NFSv4 callback service
+ */
+static struct svc_rqst *
+nfs4_callback_up(struct svc_serv *serv)
+{
+	return svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
+}
+
+#if defined(CONFIG_NFS_V4_1)
+static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net)
+{
+	/*
+	 * Create an svc_sock for the back channel service that shares the
+	 * fore channel connection.
+	 * Returns the input port (0) and sets the svc_serv bc_xprt on success
+	 */
+	return svc_create_xprt(serv, "tcp-bc", net, PF_INET, 0,
+			      SVC_SOCK_ANONYMOUS);
+}
+
+/*
+ * The callback service for NFSv4.1 callbacks
+ */
+static int
+nfs41_callback_svc(void *vrqstp)
+{
+	struct svc_rqst *rqstp = vrqstp;
+	struct svc_serv *serv = rqstp->rq_server;
+	struct rpc_rqst *req;
+	int error;
+	DEFINE_WAIT(wq);
+
+	set_freezable();
+
+	while (!kthread_should_stop()) {
+		if (try_to_freeze())
+			continue;
+
+		prepare_to_wait(&serv->sv_cb_waitq, &wq, TASK_UNINTERRUPTIBLE);
+		spin_lock_bh(&serv->sv_cb_lock);
+		if (!list_empty(&serv->sv_cb_list)) {
+			req = list_first_entry(&serv->sv_cb_list,
+					struct rpc_rqst, rq_bc_list);
+			list_del(&req->rq_bc_list);
+			spin_unlock_bh(&serv->sv_cb_lock);
+			finish_wait(&serv->sv_cb_waitq, &wq);
+			dprintk("Invoking bc_svc_process()\n");
+			error = bc_svc_process(serv, req, rqstp);
+			dprintk("bc_svc_process() returned w/ error code= %d\n",
+				error);
+		} else {
+			spin_unlock_bh(&serv->sv_cb_lock);
+			/* schedule_timeout to game the hung task watchdog */
+			schedule_timeout(60 * HZ);
+			finish_wait(&serv->sv_cb_waitq, &wq);
+		}
+	}
+	return 0;
+}
+
+/*
+ * Bring up the NFSv4.1 callback service
+ */
+static struct svc_rqst *
+nfs41_callback_up(struct svc_serv *serv)
+{
+	struct svc_rqst *rqstp;
+
+	INIT_LIST_HEAD(&serv->sv_cb_list);
+	spin_lock_init(&serv->sv_cb_lock);
+	init_waitqueue_head(&serv->sv_cb_waitq);
+	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE);
+	if (IS_ERR(rqstp)) {
+		svc_xprt_put(serv->sv_bc_xprt);
+		serv->sv_bc_xprt = NULL;
+	}
+	dprintk("--> %s return %d\n", __func__, PTR_ERR_OR_ZERO(rqstp));
+	return rqstp;
+}
+
+static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv,
+		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
+{
+	*rqstpp = nfs41_callback_up(serv);
+	*callback_svc = nfs41_callback_svc;
+}
+
+static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
+		struct svc_serv *serv)
+{
+	if (minorversion)
+		/*
+		 * Save the svc_serv in the transport so that it can
+		 * be referenced when the session backchannel is initialized
+		 */
+		xprt->bc_serv = serv;
+}
+#else
+static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net)
+{
+	return 0;
+}
+
+static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv,
+		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
+{
+	*rqstpp = ERR_PTR(-ENOTSUPP);
+	*callback_svc = ERR_PTR(-ENOTSUPP);
+}
+
+static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
+		struct svc_serv *serv)
+{
+}
+#endif /* CONFIG_NFS_V4_1 */
+
+static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt,
+				  struct svc_serv *serv)
+{
+	struct svc_rqst *rqstp;
+	int (*callback_svc)(void *vrqstp);
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+	int ret;
+
+	nfs_callback_bc_serv(minorversion, xprt, serv);
+
+	if (cb_info->task)
+		return 0;
+
+	switch (minorversion) {
+	case 0:
+		/* v4.0 callback setup */
+		rqstp = nfs4_callback_up(serv);
+		callback_svc = nfs4_callback_svc;
+		break;
+	default:
+		nfs_minorversion_callback_svc_setup(serv,
+				&rqstp, &callback_svc);
+	}
+
+	if (IS_ERR(rqstp))
+		return PTR_ERR(rqstp);
+
+	svc_sock_update_bufs(serv);
+
+	cb_info->serv = serv;
+	cb_info->rqst = rqstp;
+	cb_info->task = kthread_create(callback_svc, cb_info->rqst,
+				    "nfsv4.%u-svc", minorversion);
+	if (IS_ERR(cb_info->task)) {
+		ret = PTR_ERR(cb_info->task);
+		svc_exit_thread(cb_info->rqst);
+		cb_info->rqst = NULL;
+		cb_info->task = NULL;
+		return ret;
+	}
+	rqstp->rq_task = cb_info->task;
+	wake_up_process(cb_info->task);
+	dprintk("nfs_callback_up: service started\n");
+	return 0;
+}
+
+static void nfs_callback_down_net(u32 minorversion, struct svc_serv *serv, struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	if (--nn->cb_users[minorversion])
+		return;
+
+	dprintk("NFS: destroy per-net callback data; net=%p\n", net);
+	svc_shutdown_net(serv, net);
+}
+
+static int nfs_callback_up_net(int minorversion, struct svc_serv *serv, struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	int ret;
+
+	if (nn->cb_users[minorversion]++)
+		return 0;
+
+	dprintk("NFS: create per-net callback data; net=%p\n", net);
+
+	ret = svc_bind(serv, net);
+	if (ret < 0) {
+		printk(KERN_WARNING "NFS: bind callback service failed\n");
+		goto err_bind;
+	}
+
+	switch (minorversion) {
+		case 0:
+			ret = nfs4_callback_up_net(serv, net);
+			break;
+		case 1:
+		case 2:
+			ret = nfs41_callback_up_net(serv, net);
+			break;
+		default:
+			printk(KERN_ERR "NFS: unknown callback version: %d\n",
+					minorversion);
+			ret = -EINVAL;
+			break;
+	}
+
+	if (ret < 0) {
+		printk(KERN_ERR "NFS: callback service start failed\n");
+		goto err_socks;
+	}
+	return 0;
+
+err_socks:
+	svc_rpcb_cleanup(serv, net);
+err_bind:
+	dprintk("NFS: Couldn't create callback socket: err = %d; "
+			"net = %p\n", ret, net);
+	return ret;
+}
+
+static struct svc_serv *nfs_callback_create_svc(int minorversion)
+{
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+	struct svc_serv *serv;
+
+	/*
+	 * Check whether we're already up and running.
+	 */
+	if (cb_info->task) {
+		/*
+		 * Note: increase service usage, because later in case of error
+		 * svc_destroy() will be called.
+		 */
+		svc_get(cb_info->serv);
+		return cb_info->serv;
+	}
+
+	/*
+	 * Sanity check: if there's no task,
+	 * we should be the first user ...
+	 */
+	if (cb_info->users)
+		printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n",
+			cb_info->users);
+
+	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL);
+	if (!serv) {
+		printk(KERN_ERR "nfs_callback_create_svc: create service failed\n");
+		return ERR_PTR(-ENOMEM);
+	}
+	/* As there is only one thread we need to over-ride the
+	 * default maximum of 80 connections
+	 */
+	serv->sv_maxconn = 1024;
+	dprintk("nfs_callback_create_svc: service created\n");
+	return serv;
+}
+
+/*
+ * Bring up the callback thread if it is not already up.
+ */
+int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
+{
+	struct svc_serv *serv;
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+	int ret;
+	struct net *net = xprt->xprt_net;
+
+	mutex_lock(&nfs_callback_mutex);
+
+	serv = nfs_callback_create_svc(minorversion);
+	if (IS_ERR(serv)) {
+		ret = PTR_ERR(serv);
+		goto err_create;
+	}
+
+	ret = nfs_callback_up_net(minorversion, serv, net);
+	if (ret < 0)
+		goto err_net;
+
+	ret = nfs_callback_start_svc(minorversion, xprt, serv);
+	if (ret < 0)
+		goto err_start;
+
+	cb_info->users++;
+	/*
+	 * svc_create creates the svc_serv with sv_nrthreads == 1, and then
+	 * svc_prepare_thread increments that. So we need to call svc_destroy
+	 * on both success and failure so that the refcount is 1 when the
+	 * thread exits.
+	 */
+err_net:
+	svc_destroy(serv);
+err_create:
+	mutex_unlock(&nfs_callback_mutex);
+	return ret;
+
+err_start:
+	nfs_callback_down_net(minorversion, serv, net);
+	dprintk("NFS: Couldn't create server thread; err = %d\n", ret);
+	goto err_net;
+}
+
+/*
+ * Kill the callback thread if it's no longer being used.
+ */
+void nfs_callback_down(int minorversion, struct net *net)
+{
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+
+	mutex_lock(&nfs_callback_mutex);
+	nfs_callback_down_net(minorversion, cb_info->serv, net);
+	cb_info->users--;
+	if (cb_info->users == 0 && cb_info->task != NULL) {
+		kthread_stop(cb_info->task);
+		dprintk("nfs_callback_down: service stopped\n");
+		svc_exit_thread(cb_info->rqst);
+		dprintk("nfs_callback_down: service destroyed\n");
+		cb_info->serv = NULL;
+		cb_info->rqst = NULL;
+		cb_info->task = NULL;
+	}
+	mutex_unlock(&nfs_callback_mutex);
+}
+
+/* Boolean check of RPC_AUTH_GSS principal */
+int
+check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp)
+{
+	char *p = rqstp->rq_cred.cr_principal;
+
+	if (rqstp->rq_authop->flavour != RPC_AUTH_GSS)
+		return 1;
+
+	/* No RPC_AUTH_GSS on NFSv4.1 back channel yet */
+	if (clp->cl_minorversion != 0)
+		return 0;
+	/*
+	 * It might just be a normal user principal, in which case
+	 * userspace won't bother to tell us the name at all.
+	 */
+	if (p == NULL)
+		return 0;
+
+	/*
+	 * Did we get the acceptor from userland during the SETCLIENID
+	 * negotiation?
+	 */
+	if (clp->cl_acceptor)
+		return !strcmp(p, clp->cl_acceptor);
+
+	/*
+	 * Otherwise try to verify it using the cl_hostname. Note that this
+	 * doesn't work if a non-canonical hostname was used in the devname.
+	 */
+
+	/* Expect a GSS_C_NT_HOSTBASED_NAME like "nfs@serverhostname" */
+
+	if (memcmp(p, "nfs@", 4) != 0)
+		return 0;
+	p += 4;
+	if (strcmp(p, clp->cl_hostname) != 0)
+		return 0;
+	return 1;
+}
+
+/*
+ * pg_authenticate method for nfsv4 callback threads.
+ *
+ * The authflavor has been negotiated, so an incorrect flavor is a server
+ * bug. Drop packets with incorrect authflavor.
+ *
+ * All other checking done after NFS decoding where the nfs_client can be
+ * found in nfs4_callback_compound
+ */
+static int nfs_callback_authenticate(struct svc_rqst *rqstp)
+{
+	switch (rqstp->rq_authop->flavour) {
+	case RPC_AUTH_NULL:
+		if (rqstp->rq_proc != CB_NULL)
+			return SVC_DROP;
+		break;
+	case RPC_AUTH_GSS:
+		/* No RPC_AUTH_GSS support yet in NFSv4.1 */
+		 if (svc_is_backchannel(rqstp))
+			return SVC_DROP;
+	}
+	return SVC_OK;
+}
+
+/*
+ * Define NFS4 callback program
+ */
+static struct svc_version *nfs4_callback_version[] = {
+	[1] = &nfs4_callback_version1,
+	[4] = &nfs4_callback_version4,
+};
+
+static struct svc_stat nfs4_callback_stats;
+
+static struct svc_program nfs4_callback_program = {
+	.pg_prog = NFS4_CALLBACK,			/* RPC service number */
+	.pg_nvers = ARRAY_SIZE(nfs4_callback_version),	/* Number of entries */
+	.pg_vers = nfs4_callback_version,		/* version table */
+	.pg_name = "NFSv4 callback",			/* service name */
+	.pg_class = "nfs",				/* authentication class */
+	.pg_stats = &nfs4_callback_stats,
+	.pg_authenticate = nfs_callback_authenticate,
+};

fs/nfs/callback.h

@@ -0,0 +1,214 @@
+/*
+ * linux/fs/nfs/callback.h
+ *
+ * Copyright (C) 2004 Trond Myklebust
+ *
+ * NFSv4 callback definitions
+ */
+#ifndef __LINUX_FS_NFS_CALLBACK_H
+#define __LINUX_FS_NFS_CALLBACK_H
+#include <linux/sunrpc/svc.h>
+
+#define NFS4_CALLBACK 0x40000000
+#define NFS4_CALLBACK_XDRSIZE 2048
+#define NFS4_CALLBACK_BUFSIZE (1024 + NFS4_CALLBACK_XDRSIZE)
+
+enum nfs4_callback_procnum {
+	CB_NULL = 0,
+	CB_COMPOUND = 1,
+};
+
+enum nfs4_callback_opnum {
+	OP_CB_GETATTR = 3,
+	OP_CB_RECALL  = 4,
+/* Callback operations new to NFSv4.1 */
+	OP_CB_LAYOUTRECALL  = 5,
+	OP_CB_NOTIFY        = 6,
+	OP_CB_PUSH_DELEG    = 7,
+	OP_CB_RECALL_ANY    = 8,
+	OP_CB_RECALLABLE_OBJ_AVAIL = 9,
+	OP_CB_RECALL_SLOT   = 10,
+	OP_CB_SEQUENCE      = 11,
+	OP_CB_WANTS_CANCELLED = 12,
+	OP_CB_NOTIFY_LOCK   = 13,
+	OP_CB_NOTIFY_DEVICEID = 14,
+/* Callback operations new to NFSv4.2 */
+	OP_CB_OFFLOAD = 15,
+	OP_CB_ILLEGAL = 10044,
+};
+
+struct cb_process_state {
+	__be32			drc_status;
+	struct nfs_client	*clp;
+	u32			slotid;
+	u32			minorversion;
+	struct net		*net;
+};
+
+struct cb_compound_hdr_arg {
+	unsigned int taglen;
+	const char *tag;
+	unsigned int minorversion;
+	unsigned int cb_ident; /* v4.0 callback identifier */
+	unsigned nops;
+};
+
+struct cb_compound_hdr_res {
+	__be32 *status;
+	unsigned int taglen;
+	const char *tag;
+	__be32 *nops;
+};
+
+struct cb_getattrargs {
+	struct sockaddr *addr;
+	struct nfs_fh fh;
+	uint32_t bitmap[2];
+};
+
+struct cb_getattrres {
+	__be32 status;
+	uint32_t bitmap[2];
+	uint64_t size;
+	uint64_t change_attr;
+	struct timespec ctime;
+	struct timespec mtime;
+};
+
+struct cb_recallargs {
+	struct sockaddr *addr;
+	struct nfs_fh fh;
+	nfs4_stateid stateid;
+	uint32_t truncate;
+};
+
+#if defined(CONFIG_NFS_V4_1)
+
+struct referring_call {
+	uint32_t			rc_sequenceid;
+	uint32_t			rc_slotid;
+};
+
+struct referring_call_list {
+	struct nfs4_sessionid		rcl_sessionid;
+	uint32_t			rcl_nrefcalls;
+	struct referring_call 		*rcl_refcalls;
+};
+
+struct cb_sequenceargs {
+	struct sockaddr			*csa_addr;
+	struct nfs4_sessionid		csa_sessionid;
+	uint32_t			csa_sequenceid;
+	uint32_t			csa_slotid;
+	uint32_t			csa_highestslotid;
+	uint32_t			csa_cachethis;
+	uint32_t			csa_nrclists;
+	struct referring_call_list	*csa_rclists;
+};
+
+struct cb_sequenceres {
+	__be32				csr_status;
+	struct nfs4_sessionid		csr_sessionid;
+	uint32_t			csr_sequenceid;
+	uint32_t			csr_slotid;
+	uint32_t			csr_highestslotid;
+	uint32_t			csr_target_highestslotid;
+};
+
+extern __be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
+				       struct cb_sequenceres *res,
+				       struct cb_process_state *cps);
+
+extern int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation,
+					     const nfs4_stateid *stateid);
+
+#define RCA4_TYPE_MASK_RDATA_DLG	0
+#define RCA4_TYPE_MASK_WDATA_DLG	1
+#define RCA4_TYPE_MASK_DIR_DLG         2
+#define RCA4_TYPE_MASK_FILE_LAYOUT     3
+#define RCA4_TYPE_MASK_BLK_LAYOUT      4
+#define RCA4_TYPE_MASK_OBJ_LAYOUT_MIN  8
+#define RCA4_TYPE_MASK_OBJ_LAYOUT_MAX  9
+#define RCA4_TYPE_MASK_OTHER_LAYOUT_MIN 12
+#define RCA4_TYPE_MASK_OTHER_LAYOUT_MAX 15
+#define RCA4_TYPE_MASK_ALL 0xf31f
+
+struct cb_recallanyargs {
+	struct sockaddr	*craa_addr;
+	uint32_t	craa_objs_to_keep;
+	uint32_t	craa_type_mask;
+};
+
+extern __be32 nfs4_callback_recallany(struct cb_recallanyargs *args,
+					void *dummy,
+					struct cb_process_state *cps);
+
+struct cb_recallslotargs {
+	struct sockaddr	*crsa_addr;
+	uint32_t	crsa_target_highest_slotid;
+};
+extern __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args,
+					 void *dummy,
+					 struct cb_process_state *cps);
+
+struct cb_layoutrecallargs {
+	struct sockaddr		*cbl_addr;
+	uint32_t		cbl_recall_type;
+	uint32_t		cbl_layout_type;
+	uint32_t		cbl_layoutchanged;
+	union {
+		struct {
+			struct nfs_fh		cbl_fh;
+			struct pnfs_layout_range cbl_range;
+			nfs4_stateid		cbl_stateid;
+		};
+		struct nfs_fsid		cbl_fsid;
+	};
+};
+
+extern __be32 nfs4_callback_layoutrecall(
+	struct cb_layoutrecallargs *args,
+	void *dummy, struct cb_process_state *cps);
+
+struct cb_devicenotifyitem {
+	uint32_t		cbd_notify_type;
+	uint32_t		cbd_layout_type;
+	struct nfs4_deviceid	cbd_dev_id;
+	uint32_t		cbd_immediate;
+};
+
+struct cb_devicenotifyargs {
+	int				 ndevs;
+	struct cb_devicenotifyitem	 *devs;
+};
+
+extern __be32 nfs4_callback_devicenotify(
+	struct cb_devicenotifyargs *args,
+	void *dummy, struct cb_process_state *cps);
+
+#endif /* CONFIG_NFS_V4_1 */
+extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *);
+extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args,
+				    struct cb_getattrres *res,
+				    struct cb_process_state *cps);
+extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
+				   struct cb_process_state *cps);
+#if IS_ENABLED(CONFIG_NFS_V4)
+extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
+extern void nfs_callback_down(int minorversion, struct net *net);
+extern int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation,
+					    const nfs4_stateid *stateid);
+extern int nfs4_set_callback_sessionid(struct nfs_client *clp);
+#endif /* CONFIG_NFS_V4 */
+/*
+ * nfs41: Callbacks are expected to not cause substantial latency,
+ * so we limit their concurrency to 1 by setting up the maximum number
+ * of slots for the backchannel.
+ */
+#define NFS41_BC_MIN_CALLBACKS 1
+#define NFS41_BC_MAX_CALLBACKS 1
+
+extern unsigned int nfs_callback_set_tcpport;
+extern unsigned short nfs_callback_tcpport;
+
+#endif /* __LINUX_FS_NFS_CALLBACK_H */

fs/nfs/callback_proc.c

@@ -0,0 +1,552 @@
+/*
+ * linux/fs/nfs/callback_proc.c
+ *
+ * Copyright (C) 2004 Trond Myklebust
+ *
+ * NFSv4 callback procedures
+ */
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include <linux/slab.h>
+#include <linux/rcupdate.h>
+#include "nfs4_fs.h"
+#include "callback.h"
+#include "delegation.h"
+#include "internal.h"
+#include "pnfs.h"
+#include "nfs4session.h"
+#include "nfs4trace.h"
+
+#ifdef NFS_DEBUG
+#define NFSDBG_FACILITY NFSDBG_CALLBACK
+#endif
+
+__be32 nfs4_callback_getattr(struct cb_getattrargs *args,
+			     struct cb_getattrres *res,
+			     struct cb_process_state *cps)
+{
+	struct nfs_delegation *delegation;
+	struct nfs_inode *nfsi;
+	struct inode *inode;
+
+	res->status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	if (!cps->clp) /* Always set for v4.0. Set in cb_sequence for v4.1 */
+		goto out;
+
+	res->bitmap[0] = res->bitmap[1] = 0;
+	res->status = htonl(NFS4ERR_BADHANDLE);
+
+	dprintk_rcu("NFS: GETATTR callback request from %s\n",
+		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	inode = nfs_delegation_find_inode(cps->clp, &args->fh);
+	if (inode == NULL)
+		goto out;
+	nfsi = NFS_I(inode);
+	rcu_read_lock();
+	delegation = rcu_dereference(nfsi->delegation);
+	if (delegation == NULL || (delegation->type & FMODE_WRITE) == 0)
+		goto out_iput;
+	res->size = i_size_read(inode);
+	res->change_attr = delegation->change_attr;
+	if (nfsi->npages != 0)
+		res->change_attr++;
+	res->ctime = inode->i_ctime;
+	res->mtime = inode->i_mtime;
+	res->bitmap[0] = (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) &
+		args->bitmap[0];
+	res->bitmap[1] = (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) &
+		args->bitmap[1];
+	res->status = 0;
+out_iput:
+	rcu_read_unlock();
+	iput(inode);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(res->status));
+	return res->status;
+}
+
+__be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy,
+			    struct cb_process_state *cps)
+{
+	struct inode *inode;
+	__be32 res;
+	
+	res = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	if (!cps->clp) /* Always set for v4.0. Set in cb_sequence for v4.1 */
+		goto out;
+
+	dprintk_rcu("NFS: RECALL callback request from %s\n",
+		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	res = htonl(NFS4ERR_BADHANDLE);
+	inode = nfs_delegation_find_inode(cps->clp, &args->fh);
+	if (inode == NULL)
+		goto out;
+	/* Set up a helper thread to actually return the delegation */
+	switch (nfs_async_inode_return_delegation(inode, &args->stateid)) {
+	case 0:
+		res = 0;
+		break;
+	case -ENOENT:
+		res = htonl(NFS4ERR_BAD_STATEID);
+		break;
+	default:
+		res = htonl(NFS4ERR_RESOURCE);
+	}
+	trace_nfs4_recall_delegation(inode, -ntohl(res));
+	iput(inode);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(res));
+	return res;
+}
+
+#if defined(CONFIG_NFS_V4_1)
+
+/*
+ * Lookup a layout by filehandle.
+ *
+ * Note: gets a refcount on the layout hdr and on its respective inode.
+ * Caller must put the layout hdr and the inode.
+ *
+ * TODO: keep track of all layouts (and delegations) in a hash table
+ * hashed by filehandle.
+ */
+static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp,
+		struct nfs_fh *fh, nfs4_stateid *stateid)
+{
+	struct nfs_server *server;
+	struct inode *ino;
+	struct pnfs_layout_hdr *lo;
+
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+			if (!nfs4_stateid_match_other(&lo->plh_stateid, stateid))
+				continue;
+			if (nfs_compare_fh(fh, &NFS_I(lo->plh_inode)->fh))
+				continue;
+			ino = igrab(lo->plh_inode);
+			if (!ino)
+				break;
+			spin_lock(&ino->i_lock);
+			/* Is this layout in the process of being freed? */
+			if (NFS_I(ino)->layout != lo) {
+				spin_unlock(&ino->i_lock);
+				iput(ino);
+				break;
+			}
+			pnfs_get_layout_hdr(lo);
+			spin_unlock(&ino->i_lock);
+			return lo;
+		}
+	}
+
+	return NULL;
+}
+
+static struct pnfs_layout_hdr * get_layout_by_fh(struct nfs_client *clp,
+		struct nfs_fh *fh, nfs4_stateid *stateid)
+{
+	struct pnfs_layout_hdr *lo;
+
+	spin_lock(&clp->cl_lock);
+	rcu_read_lock();
+	lo = get_layout_by_fh_locked(clp, fh, stateid);
+	rcu_read_unlock();
+	spin_unlock(&clp->cl_lock);
+
+	return lo;
+}
+
+static u32 initiate_file_draining(struct nfs_client *clp,
+				  struct cb_layoutrecallargs *args)
+{
+	struct inode *ino;
+	struct pnfs_layout_hdr *lo;
+	u32 rv = NFS4ERR_NOMATCHING_LAYOUT;
+	LIST_HEAD(free_me_list);
+
+	lo = get_layout_by_fh(clp, &args->cbl_fh, &args->cbl_stateid);
+	if (!lo)
+		goto out;
+
+	ino = lo->plh_inode;
+
+	spin_lock(&ino->i_lock);
+	pnfs_set_layout_stateid(lo, &args->cbl_stateid, true);
+	spin_unlock(&ino->i_lock);
+
+	pnfs_layoutcommit_inode(ino, false);
+
+	spin_lock(&ino->i_lock);
+	if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) ||
+	    pnfs_mark_matching_lsegs_invalid(lo, &free_me_list,
+					&args->cbl_range)) {
+		rv = NFS4ERR_DELAY;
+		goto unlock;
+	}
+
+	if (NFS_SERVER(ino)->pnfs_curr_ld->return_range) {
+		NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo,
+			&args->cbl_range);
+	}
+unlock:
+	spin_unlock(&ino->i_lock);
+	pnfs_free_lseg_list(&free_me_list);
+	pnfs_put_layout_hdr(lo);
+	iput(ino);
+out:
+	return rv;
+}
+
+static u32 initiate_bulk_draining(struct nfs_client *clp,
+				  struct cb_layoutrecallargs *args)
+{
+	int stat;
+
+	if (args->cbl_recall_type == RETURN_FSID)
+		stat = pnfs_destroy_layouts_byfsid(clp, &args->cbl_fsid, true);
+	else
+		stat = pnfs_destroy_layouts_byclid(clp, true);
+	if (stat != 0)
+		return NFS4ERR_DELAY;
+	return NFS4ERR_NOMATCHING_LAYOUT;
+}
+
+static u32 do_callback_layoutrecall(struct nfs_client *clp,
+				    struct cb_layoutrecallargs *args)
+{
+	u32 res;
+
+	dprintk("%s enter, type=%i\n", __func__, args->cbl_recall_type);
+	if (args->cbl_recall_type == RETURN_FILE)
+		res = initiate_file_draining(clp, args);
+	else
+		res = initiate_bulk_draining(clp, args);
+	dprintk("%s returning %i\n", __func__, res);
+	return res;
+
+}
+
+__be32 nfs4_callback_layoutrecall(struct cb_layoutrecallargs *args,
+				  void *dummy, struct cb_process_state *cps)
+{
+	u32 res;
+
+	dprintk("%s: -->\n", __func__);
+
+	if (cps->clp)
+		res = do_callback_layoutrecall(cps->clp, args);
+	else
+		res = NFS4ERR_OP_NOT_IN_SESSION;
+
+	dprintk("%s: exit with status = %d\n", __func__, res);
+	return cpu_to_be32(res);
+}
+
+static void pnfs_recall_all_layouts(struct nfs_client *clp)
+{
+	struct cb_layoutrecallargs args;
+
+	/* Pretend we got a CB_LAYOUTRECALL(ALL) */
+	memset(&args, 0, sizeof(args));
+	args.cbl_recall_type = RETURN_ALL;
+	/* FIXME we ignore errors, what should we do? */
+	do_callback_layoutrecall(clp, &args);
+}
+
+__be32 nfs4_callback_devicenotify(struct cb_devicenotifyargs *args,
+				  void *dummy, struct cb_process_state *cps)
+{
+	int i;
+	__be32 res = 0;
+	struct nfs_client *clp = cps->clp;
+	struct nfs_server *server = NULL;
+
+	dprintk("%s: -->\n", __func__);
+
+	if (!clp) {
+		res = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION);
+		goto out;
+	}
+
+	for (i = 0; i < args->ndevs; i++) {
+		struct cb_devicenotifyitem *dev = &args->devs[i];
+
+		if (!server ||
+		    server->pnfs_curr_ld->id != dev->cbd_layout_type) {
+			rcu_read_lock();
+			list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+				if (server->pnfs_curr_ld &&
+				    server->pnfs_curr_ld->id == dev->cbd_layout_type) {
+					rcu_read_unlock();
+					goto found;
+				}
+			rcu_read_unlock();
+			dprintk("%s: layout type %u not found\n",
+				__func__, dev->cbd_layout_type);
+			continue;
+		}
+
+	found:
+		nfs4_delete_deviceid(server->pnfs_curr_ld, clp, &dev->cbd_dev_id);
+	}
+
+out:
+	kfree(args->devs);
+	dprintk("%s: exit with status = %u\n",
+		__func__, be32_to_cpu(res));
+	return res;
+}
+
+/*
+ * Validate the sequenceID sent by the server.
+ * Return success if the sequenceID is one more than what we last saw on
+ * this slot, accounting for wraparound.  Increments the slot's sequence.
+ *
+ * We don't yet implement a duplicate request cache, instead we set the
+ * back channel ca_maxresponsesize_cached to zero. This is OK for now
+ * since we only currently implement idempotent callbacks anyway.
+ *
+ * We have a single slot backchannel at this time, so we don't bother
+ * checking the used_slots bit array on the table.  The lower layer guarantees
+ * a single outstanding callback request at a time.
+ */
+static __be32
+validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args)
+{
+	struct nfs4_slot *slot;
+
+	dprintk("%s enter. slotid %u seqid %u\n",
+		__func__, args->csa_slotid, args->csa_sequenceid);
+
+	if (args->csa_slotid >= NFS41_BC_MAX_CALLBACKS)
+		return htonl(NFS4ERR_BADSLOT);
+
+	slot = tbl->slots + args->csa_slotid;
+	dprintk("%s slot table seqid: %u\n", __func__, slot->seq_nr);
+
+	/* Normal */
+	if (likely(args->csa_sequenceid == slot->seq_nr + 1)) {
+		slot->seq_nr++;
+		goto out_ok;
+	}
+
+	/* Replay */
+	if (args->csa_sequenceid == slot->seq_nr) {
+		dprintk("%s seqid %u is a replay\n",
+			__func__, args->csa_sequenceid);
+		/* Signal process_op to set this error on next op */
+		if (args->csa_cachethis == 0)
+			return htonl(NFS4ERR_RETRY_UNCACHED_REP);
+
+		/* The ca_maxresponsesize_cached is 0 with no DRC */
+		else if (args->csa_cachethis == 1)
+			return htonl(NFS4ERR_REP_TOO_BIG_TO_CACHE);
+	}
+
+	/* Wraparound */
+	if (args->csa_sequenceid == 1 && (slot->seq_nr + 1) == 0) {
+		slot->seq_nr = 1;
+		goto out_ok;
+	}
+
+	/* Misordered request */
+	return htonl(NFS4ERR_SEQ_MISORDERED);
+out_ok:
+	tbl->highest_used_slotid = args->csa_slotid;
+	return htonl(NFS4_OK);
+}
+
+/*
+ * For each referring call triple, check the session's slot table for
+ * a match.  If the slot is in use and the sequence numbers match, the
+ * client is still waiting for a response to the original request.
+ */
+static bool referring_call_exists(struct nfs_client *clp,
+				  uint32_t nrclists,
+				  struct referring_call_list *rclists)
+{
+	bool status = 0;
+	int i, j;
+	struct nfs4_session *session;
+	struct nfs4_slot_table *tbl;
+	struct referring_call_list *rclist;
+	struct referring_call *ref;
+
+	/*
+	 * XXX When client trunking is implemented, this becomes
+	 * a session lookup from within the loop
+	 */
+	session = clp->cl_session;
+	tbl = &session->fc_slot_table;
+
+	for (i = 0; i < nrclists; i++) {
+		rclist = &rclists[i];
+		if (memcmp(session->sess_id.data,
+			   rclist->rcl_sessionid.data,
+			   NFS4_MAX_SESSIONID_LEN) != 0)
+			continue;
+
+		for (j = 0; j < rclist->rcl_nrefcalls; j++) {
+			ref = &rclist->rcl_refcalls[j];
+
+			dprintk("%s: sessionid %x:%x:%x:%x sequenceid %u "
+				"slotid %u\n", __func__,
+				((u32 *)&rclist->rcl_sessionid.data)[0],
+				((u32 *)&rclist->rcl_sessionid.data)[1],
+				((u32 *)&rclist->rcl_sessionid.data)[2],
+				((u32 *)&rclist->rcl_sessionid.data)[3],
+				ref->rc_sequenceid, ref->rc_slotid);
+
+			spin_lock(&tbl->slot_tbl_lock);
+			status = (test_bit(ref->rc_slotid, tbl->used_slots) &&
+				  tbl->slots[ref->rc_slotid].seq_nr ==
+					ref->rc_sequenceid);
+			spin_unlock(&tbl->slot_tbl_lock);
+			if (status)
+				goto out;
+		}
+	}
+
+out:
+	return status;
+}
+
+__be32 nfs4_callback_sequence(struct cb_sequenceargs *args,
+			      struct cb_sequenceres *res,
+			      struct cb_process_state *cps)
+{
+	struct nfs4_slot_table *tbl;
+	struct nfs_client *clp;
+	int i;
+	__be32 status = htonl(NFS4ERR_BADSESSION);
+
+	clp = nfs4_find_client_sessionid(cps->net, args->csa_addr,
+					 &args->csa_sessionid, cps->minorversion);
+	if (clp == NULL)
+		goto out;
+
+	tbl = &clp->cl_session->bc_slot_table;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	/* state manager is resetting the session */
+	if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) {
+		spin_unlock(&tbl->slot_tbl_lock);
+		status = htonl(NFS4ERR_DELAY);
+		/* Return NFS4ERR_BADSESSION if we're draining the session
+		 * in order to reset it.
+		 */
+		if (test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state))
+			status = htonl(NFS4ERR_BADSESSION);
+		goto out;
+	}
+
+	status = validate_seqid(&clp->cl_session->bc_slot_table, args);
+	spin_unlock(&tbl->slot_tbl_lock);
+	if (status)
+		goto out;
+
+	cps->slotid = args->csa_slotid;
+
+	/*
+	 * Check for pending referring calls.  If a match is found, a
+	 * related callback was received before the response to the original
+	 * call.
+	 */
+	if (referring_call_exists(clp, args->csa_nrclists, args->csa_rclists)) {
+		status = htonl(NFS4ERR_DELAY);
+		goto out;
+	}
+
+	memcpy(&res->csr_sessionid, &args->csa_sessionid,
+	       sizeof(res->csr_sessionid));
+	res->csr_sequenceid = args->csa_sequenceid;
+	res->csr_slotid = args->csa_slotid;
+	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+
+out:
+	cps->clp = clp; /* put in nfs4_callback_compound */
+	for (i = 0; i < args->csa_nrclists; i++)
+		kfree(args->csa_rclists[i].rcl_refcalls);
+	kfree(args->csa_rclists);
+
+	if (status == htonl(NFS4ERR_RETRY_UNCACHED_REP)) {
+		cps->drc_status = status;
+		status = 0;
+	} else
+		res->csr_status = status;
+
+	trace_nfs4_cb_sequence(args, res, status);
+	dprintk("%s: exit with status = %d res->csr_status %d\n", __func__,
+		ntohl(status), ntohl(res->csr_status));
+	return status;
+}
+
+static bool
+validate_bitmap_values(unsigned long mask)
+{
+	return (mask & ~RCA4_TYPE_MASK_ALL) == 0;
+}
+
+__be32 nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy,
+			       struct cb_process_state *cps)
+{
+	__be32 status;
+	fmode_t flags = 0;
+
+	status = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION);
+	if (!cps->clp) /* set in cb_sequence */
+		goto out;
+
+	dprintk_rcu("NFS: RECALL_ANY callback request from %s\n",
+		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	status = cpu_to_be32(NFS4ERR_INVAL);
+	if (!validate_bitmap_values(args->craa_type_mask))
+		goto out;
+
+	status = cpu_to_be32(NFS4_OK);
+	if (test_bit(RCA4_TYPE_MASK_RDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags = FMODE_READ;
+	if (test_bit(RCA4_TYPE_MASK_WDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags |= FMODE_WRITE;
+	if (test_bit(RCA4_TYPE_MASK_FILE_LAYOUT, (const unsigned long *)
+		     &args->craa_type_mask))
+		pnfs_recall_all_layouts(cps->clp);
+	if (flags)
+		nfs_expire_unused_delegation_types(cps->clp, flags);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
+
+/* Reduce the fore channel's max_slots to the target value */
+__be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy,
+				struct cb_process_state *cps)
+{
+	struct nfs4_slot_table *fc_tbl;
+	__be32 status;
+
+	status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	if (!cps->clp) /* set in cb_sequence */
+		goto out;
+
+	dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %u\n",
+		rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR),
+		args->crsa_target_highest_slotid);
+
+	fc_tbl = &cps->clp->cl_session->fc_slot_table;
+
+	status = htonl(NFS4_OK);
+
+	nfs41_set_target_slotid(fc_tbl, args->crsa_target_highest_slotid);
+	nfs41_server_notify_target_slotid_update(cps->clp);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
+#endif /* CONFIG_NFS_V4_1 */

fs/nfs/delegation.c

@@ -0,0 +1,864 @@
+/*
+ * linux/fs/nfs/delegation.c
+ *
+ * Copyright (C) 2004 Trond Myklebust
+ *
+ * NFS file delegation management
+ *
+ */
+#include <linux/completion.h>
+#include <linux/kthread.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_xdr.h>
+
+#include "nfs4_fs.h"
+#include "delegation.h"
+#include "internal.h"
+#include "nfs4trace.h"
+
+static void nfs_free_delegation(struct nfs_delegation *delegation)
+{
+	if (delegation->cred) {
+		put_rpccred(delegation->cred);
+		delegation->cred = NULL;
+	}
+	kfree_rcu(delegation, rcu);
+}
+
+/**
+ * nfs_mark_delegation_referenced - set delegation's REFERENCED flag
+ * @delegation: delegation to process
+ *
+ */
+void nfs_mark_delegation_referenced(struct nfs_delegation *delegation)
+{
+	set_bit(NFS_DELEGATION_REFERENCED, &delegation->flags);
+}
+
+static int
+nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark)
+{
+	struct nfs_delegation *delegation;
+	int ret = 0;
+
+	flags &= FMODE_READ|FMODE_WRITE;
+	rcu_read_lock();
+	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	if (delegation != NULL && (delegation->type & flags) == flags &&
+	    !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) {
+		if (mark)
+			nfs_mark_delegation_referenced(delegation);
+		ret = 1;
+	}
+	rcu_read_unlock();
+	return ret;
+}
+/**
+ * nfs_have_delegation - check if inode has a delegation, mark it
+ * NFS_DELEGATION_REFERENCED if there is one.
+ * @inode: inode to check
+ * @flags: delegation types to check for
+ *
+ * Returns one if inode has the indicated delegation, otherwise zero.
+ */
+int nfs4_have_delegation(struct inode *inode, fmode_t flags)
+{
+	return nfs4_do_check_delegation(inode, flags, true);
+}
+
+/*
+ * nfs4_check_delegation - check if inode has a delegation, do not mark
+ * NFS_DELEGATION_REFERENCED if it has one.
+ */
+int nfs4_check_delegation(struct inode *inode, fmode_t flags)
+{
+	return nfs4_do_check_delegation(inode, flags, false);
+}
+
+static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid)
+{
+	struct inode *inode = state->inode;
+	struct file_lock *fl;
+	int status = 0;
+
+	if (inode->i_flock == NULL)
+		goto out;
+
+	/* Protect inode->i_flock using the i_lock */
+	spin_lock(&inode->i_lock);
+	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
+		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
+			continue;
+		if (nfs_file_open_context(fl->fl_file) != ctx)
+			continue;
+		spin_unlock(&inode->i_lock);
+		status = nfs4_lock_delegation_recall(fl, state, stateid);
+		if (status < 0)
+			goto out;
+		spin_lock(&inode->i_lock);
+	}
+	spin_unlock(&inode->i_lock);
+out:
+	return status;
+}
+
+static int nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_open_context *ctx;
+	struct nfs4_state_owner *sp;
+	struct nfs4_state *state;
+	unsigned int seq;
+	int err;
+
+again:
+	spin_lock(&inode->i_lock);
+	list_for_each_entry(ctx, &nfsi->open_files, list) {
+		state = ctx->state;
+		if (state == NULL)
+			continue;
+		if (!test_bit(NFS_DELEGATED_STATE, &state->flags))
+			continue;
+		if (!nfs4_valid_open_stateid(state))
+			continue;
+		if (!nfs4_stateid_match(&state->stateid, stateid))
+			continue;
+		get_nfs_open_context(ctx);
+		spin_unlock(&inode->i_lock);
+		sp = state->owner;
+		/* Block nfs4_proc_unlck */
+		mutex_lock(&sp->so_delegreturn_mutex);
+		seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
+		err = nfs4_open_delegation_recall(ctx, state, stateid);
+		if (!err)
+			err = nfs_delegation_claim_locks(ctx, state, stateid);
+		if (!err && read_seqcount_retry(&sp->so_reclaim_seqcount, seq))
+			err = -EAGAIN;
+		mutex_unlock(&sp->so_delegreturn_mutex);
+		put_nfs_open_context(ctx);
+		if (err != 0)
+			return err;
+		goto again;
+	}
+	spin_unlock(&inode->i_lock);
+	return 0;
+}
+
+/**
+ * nfs_inode_reclaim_delegation - process a delegation reclaim request
+ * @inode: inode to process
+ * @cred: credential to use for request
+ * @res: new delegation state from server
+ *
+ */
+void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred,
+				  struct nfs_openres *res)
+{
+	struct nfs_delegation *delegation;
+	struct rpc_cred *oldcred = NULL;
+
+	rcu_read_lock();
+	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	if (delegation != NULL) {
+		spin_lock(&delegation->lock);
+		if (delegation->inode != NULL) {
+			nfs4_stateid_copy(&delegation->stateid, &res->delegation);
+			delegation->type = res->delegation_type;
+			delegation->maxsize = res->maxsize;
+			oldcred = delegation->cred;
+			delegation->cred = get_rpccred(cred);
+			clear_bit(NFS_DELEGATION_NEED_RECLAIM,
+				  &delegation->flags);
+			NFS_I(inode)->delegation_state = delegation->type;
+			spin_unlock(&delegation->lock);
+			rcu_read_unlock();
+			put_rpccred(oldcred);
+			trace_nfs4_reclaim_delegation(inode, res->delegation_type);
+		} else {
+			/* We appear to have raced with a delegation return. */
+			spin_unlock(&delegation->lock);
+			rcu_read_unlock();
+			nfs_inode_set_delegation(inode, cred, res);
+		}
+	} else {
+		rcu_read_unlock();
+	}
+}
+
+static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync)
+{
+	int res = 0;
+
+	if (!test_bit(NFS_DELEGATION_REVOKED, &delegation->flags))
+		res = nfs4_proc_delegreturn(inode,
+				delegation->cred,
+				&delegation->stateid,
+				issync);
+	nfs_free_delegation(delegation);
+	return res;
+}
+
+static struct inode *nfs_delegation_grab_inode(struct nfs_delegation *delegation)
+{
+	struct inode *inode = NULL;
+
+	spin_lock(&delegation->lock);
+	if (delegation->inode != NULL)
+		inode = igrab(delegation->inode);
+	spin_unlock(&delegation->lock);
+	return inode;
+}
+
+static struct nfs_delegation *
+nfs_start_delegation_return_locked(struct nfs_inode *nfsi)
+{
+	struct nfs_delegation *ret = NULL;
+	struct nfs_delegation *delegation = rcu_dereference(nfsi->delegation);
+
+	if (delegation == NULL)
+		goto out;
+	spin_lock(&delegation->lock);
+	if (!test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags))
+		ret = delegation;
+	spin_unlock(&delegation->lock);
+out:
+	return ret;
+}
+
+static struct nfs_delegation *
+nfs_start_delegation_return(struct nfs_inode *nfsi)
+{
+	struct nfs_delegation *delegation;
+
+	rcu_read_lock();
+	delegation = nfs_start_delegation_return_locked(nfsi);
+	rcu_read_unlock();
+	return delegation;
+}
+
+static void
+nfs_abort_delegation_return(struct nfs_delegation *delegation,
+		struct nfs_client *clp)
+{
+
+	spin_lock(&delegation->lock);
+	clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+	set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
+	spin_unlock(&delegation->lock);
+	set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+}
+
+static struct nfs_delegation *
+nfs_detach_delegation_locked(struct nfs_inode *nfsi,
+		struct nfs_delegation *delegation,
+		struct nfs_client *clp)
+{
+	struct nfs_delegation *deleg_cur =
+		rcu_dereference_protected(nfsi->delegation,
+				lockdep_is_held(&clp->cl_lock));
+
+	if (deleg_cur == NULL || delegation != deleg_cur)
+		return NULL;
+
+	spin_lock(&delegation->lock);
+	set_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+	list_del_rcu(&delegation->super_list);
+	delegation->inode = NULL;
+	nfsi->delegation_state = 0;
+	rcu_assign_pointer(nfsi->delegation, NULL);
+	spin_unlock(&delegation->lock);
+	return delegation;
+}
+
+static struct nfs_delegation *nfs_detach_delegation(struct nfs_inode *nfsi,
+		struct nfs_delegation *delegation,
+		struct nfs_server *server)
+{
+	struct nfs_client *clp = server->nfs_client;
+
+	spin_lock(&clp->cl_lock);
+	delegation = nfs_detach_delegation_locked(nfsi, delegation, clp);
+	spin_unlock(&clp->cl_lock);
+	return delegation;
+}
+
+static struct nfs_delegation *
+nfs_inode_detach_delegation(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_delegation *delegation;
+
+	delegation = nfs_start_delegation_return(nfsi);
+	if (delegation == NULL)
+		return NULL;
+	return nfs_detach_delegation(nfsi, delegation, server);
+}
+
+/**
+ * nfs_inode_set_delegation - set up a delegation on an inode
+ * @inode: inode to which delegation applies
+ * @cred: cred to use for subsequent delegation processing
+ * @res: new delegation state from server
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_client *clp = server->nfs_client;
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_delegation *delegation, *old_delegation;
+	struct nfs_delegation *freeme = NULL;
+	int status = 0;
+
+	delegation = kmalloc(sizeof(*delegation), GFP_NOFS);
+	if (delegation == NULL)
+		return -ENOMEM;
+	nfs4_stateid_copy(&delegation->stateid, &res->delegation);
+	delegation->type = res->delegation_type;
+	delegation->maxsize = res->maxsize;
+	delegation->change_attr = inode->i_version;
+	delegation->cred = get_rpccred(cred);
+	delegation->inode = inode;
+	delegation->flags = 1<<NFS_DELEGATION_REFERENCED;
+	spin_lock_init(&delegation->lock);
+
+	spin_lock(&clp->cl_lock);
+	old_delegation = rcu_dereference_protected(nfsi->delegation,
+					lockdep_is_held(&clp->cl_lock));
+	if (old_delegation != NULL) {
+		if (nfs4_stateid_match(&delegation->stateid,
+					&old_delegation->stateid) &&
+				delegation->type == old_delegation->type) {
+			goto out;
+		}
+		/*
+		 * Deal with broken servers that hand out two
+		 * delegations for the same file.
+		 * Allow for upgrades to a WRITE delegation, but
+		 * nothing else.
+		 */
+		dfprintk(FILE, "%s: server %s handed out "
+				"a duplicate delegation!\n",
+				__func__, clp->cl_hostname);
+		if (delegation->type == old_delegation->type ||
+		    !(delegation->type & FMODE_WRITE)) {
+			freeme = delegation;
+			delegation = NULL;
+			goto out;
+		}
+		freeme = nfs_detach_delegation_locked(nfsi, 
+				old_delegation, clp);
+		if (freeme == NULL)
+			goto out;
+	}
+	list_add_rcu(&delegation->super_list, &server->delegations);
+	nfsi->delegation_state = delegation->type;
+	rcu_assign_pointer(nfsi->delegation, delegation);
+	delegation = NULL;
+
+	/* Ensure we revalidate the attributes and page cache! */
+	spin_lock(&inode->i_lock);
+	nfsi->cache_validity |= NFS_INO_REVAL_FORCED;
+	spin_unlock(&inode->i_lock);
+	trace_nfs4_set_delegation(inode, res->delegation_type);
+
+out:
+	spin_unlock(&clp->cl_lock);
+	if (delegation != NULL)
+		nfs_free_delegation(delegation);
+	if (freeme != NULL)
+		nfs_do_return_delegation(inode, freeme, 0);
+	return status;
+}
+
+/*
+ * Basic procedure for returning a delegation to the server
+ */
+static int nfs_end_delegation_return(struct inode *inode, struct nfs_delegation *delegation, int issync)
+{
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
+	struct nfs_inode *nfsi = NFS_I(inode);
+	int err = 0;
+
+	if (delegation == NULL)
+		return 0;
+	do {
+		if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags))
+			break;
+		err = nfs_delegation_claim_opens(inode, &delegation->stateid);
+		if (!issync || err != -EAGAIN)
+			break;
+		/*
+		 * Guard against state recovery
+		 */
+		err = nfs4_wait_clnt_recover(clp);
+	} while (err == 0);
+
+	if (err) {
+		nfs_abort_delegation_return(delegation, clp);
+		goto out;
+	}
+	if (!nfs_detach_delegation(nfsi, delegation, NFS_SERVER(inode)))
+		goto out;
+
+	err = nfs_do_return_delegation(inode, delegation, issync);
+out:
+	return err;
+}
+
+static bool nfs_delegation_need_return(struct nfs_delegation *delegation)
+{
+	bool ret = false;
+
+	if (test_and_clear_bit(NFS_DELEGATION_RETURN, &delegation->flags))
+		ret = true;
+	if (test_and_clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags) && !ret) {
+		struct inode *inode;
+
+		spin_lock(&delegation->lock);
+		inode = delegation->inode;
+		if (inode && list_empty(&NFS_I(inode)->open_files))
+			ret = true;
+		spin_unlock(&delegation->lock);
+	}
+	return ret;
+}
+
+/**
+ * nfs_client_return_marked_delegations - return previously marked delegations
+ * @clp: nfs_client to process
+ *
+ * Note that this function is designed to be called by the state
+ * manager thread. For this reason, it cannot flush the dirty data,
+ * since that could deadlock in case of a state recovery error.
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs_client_return_marked_delegations(struct nfs_client *clp)
+{
+	struct nfs_delegation *delegation;
+	struct nfs_server *server;
+	struct inode *inode;
+	int err = 0;
+
+restart:
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		list_for_each_entry_rcu(delegation, &server->delegations,
+								super_list) {
+			if (!nfs_delegation_need_return(delegation))
+				continue;
+			inode = nfs_delegation_grab_inode(delegation);
+			if (inode == NULL)
+				continue;
+			delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+			rcu_read_unlock();
+
+			err = nfs_end_delegation_return(inode, delegation, 0);
+			iput(inode);
+			if (!err)
+				goto restart;
+			set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+			return err;
+		}
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
+/**
+ * nfs_inode_return_delegation_noreclaim - return delegation, don't reclaim opens
+ * @inode: inode to process
+ *
+ * Does not protect against delegation reclaims, therefore really only safe
+ * to be called from nfs4_clear_inode().
+ */
+void nfs_inode_return_delegation_noreclaim(struct inode *inode)
+{
+	struct nfs_delegation *delegation;
+
+	delegation = nfs_inode_detach_delegation(inode);
+	if (delegation != NULL)
+		nfs_do_return_delegation(inode, delegation, 0);
+}
+
+/**
+ * nfs_inode_return_delegation - synchronously return a delegation
+ * @inode: inode to process
+ *
+ * This routine will always flush any dirty data to disk on the
+ * assumption that if we need to return the delegation, then
+ * we should stop caching.
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs4_inode_return_delegation(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_delegation *delegation;
+	int err = 0;
+
+	nfs_wb_all(inode);
+	delegation = nfs_start_delegation_return(nfsi);
+	if (delegation != NULL)
+		err = nfs_end_delegation_return(inode, delegation, 1);
+	return err;
+}
+
+static void nfs_mark_return_if_closed_delegation(struct nfs_server *server,
+		struct nfs_delegation *delegation)
+{
+	set_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags);
+	set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state);
+}
+
+static void nfs_mark_return_delegation(struct nfs_server *server,
+		struct nfs_delegation *delegation)
+{
+	set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
+	set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state);
+}
+
+static bool nfs_server_mark_return_all_delegations(struct nfs_server *server)
+{
+	struct nfs_delegation *delegation;
+	bool ret = false;
+
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		nfs_mark_return_delegation(server, delegation);
+		ret = true;
+	}
+	return ret;
+}
+
+static void nfs_client_mark_return_all_delegations(struct nfs_client *clp)
+{
+	struct nfs_server *server;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+		nfs_server_mark_return_all_delegations(server);
+	rcu_read_unlock();
+}
+
+static void nfs_delegation_run_state_manager(struct nfs_client *clp)
+{
+	if (test_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state))
+		nfs4_schedule_state_manager(clp);
+}
+
+/**
+ * nfs_expire_all_delegations
+ * @clp: client to process
+ *
+ */
+void nfs_expire_all_delegations(struct nfs_client *clp)
+{
+	nfs_client_mark_return_all_delegations(clp);
+	nfs_delegation_run_state_manager(clp);
+}
+
+/**
+ * nfs_super_return_all_delegations - return delegations for one superblock
+ * @sb: sb to process
+ *
+ */
+void nfs_server_return_all_delegations(struct nfs_server *server)
+{
+	struct nfs_client *clp = server->nfs_client;
+	bool need_wait;
+
+	if (clp == NULL)
+		return;
+
+	rcu_read_lock();
+	need_wait = nfs_server_mark_return_all_delegations(server);
+	rcu_read_unlock();
+
+	if (need_wait) {
+		nfs4_schedule_state_manager(clp);
+		nfs4_wait_clnt_recover(clp);
+	}
+}
+
+static void nfs_mark_return_unused_delegation_types(struct nfs_server *server,
+						 fmode_t flags)
+{
+	struct nfs_delegation *delegation;
+
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if ((delegation->type == (FMODE_READ|FMODE_WRITE)) && !(flags & FMODE_WRITE))
+			continue;
+		if (delegation->type & flags)
+			nfs_mark_return_if_closed_delegation(server, delegation);
+	}
+}
+
+static void nfs_client_mark_return_unused_delegation_types(struct nfs_client *clp,
+							fmode_t flags)
+{
+	struct nfs_server *server;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+		nfs_mark_return_unused_delegation_types(server, flags);
+	rcu_read_unlock();
+}
+
+static void nfs_revoke_delegation(struct inode *inode)
+{
+	struct nfs_delegation *delegation;
+	rcu_read_lock();
+	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	if (delegation != NULL) {
+		set_bit(NFS_DELEGATION_REVOKED, &delegation->flags);
+		nfs_mark_return_delegation(NFS_SERVER(inode), delegation);
+	}
+	rcu_read_unlock();
+}
+
+void nfs_remove_bad_delegation(struct inode *inode)
+{
+	struct nfs_delegation *delegation;
+
+	nfs_revoke_delegation(inode);
+	delegation = nfs_inode_detach_delegation(inode);
+	if (delegation) {
+		nfs_inode_find_state_and_recover(inode, &delegation->stateid);
+		nfs_free_delegation(delegation);
+	}
+}
+EXPORT_SYMBOL_GPL(nfs_remove_bad_delegation);
+
+/**
+ * nfs_expire_unused_delegation_types
+ * @clp: client to process
+ * @flags: delegation types to expire
+ *
+ */
+void nfs_expire_unused_delegation_types(struct nfs_client *clp, fmode_t flags)
+{
+	nfs_client_mark_return_unused_delegation_types(clp, flags);
+	nfs_delegation_run_state_manager(clp);
+}
+
+static void nfs_mark_return_unreferenced_delegations(struct nfs_server *server)
+{
+	struct nfs_delegation *delegation;
+
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if (test_and_clear_bit(NFS_DELEGATION_REFERENCED, &delegation->flags))
+			continue;
+		nfs_mark_return_if_closed_delegation(server, delegation);
+	}
+}
+
+/**
+ * nfs_expire_unreferenced_delegations - Eliminate unused delegations
+ * @clp: nfs_client to process
+ *
+ */
+void nfs_expire_unreferenced_delegations(struct nfs_client *clp)
+{
+	struct nfs_server *server;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+		nfs_mark_return_unreferenced_delegations(server);
+	rcu_read_unlock();
+
+	nfs_delegation_run_state_manager(clp);
+}
+
+/**
+ * nfs_async_inode_return_delegation - asynchronously return a delegation
+ * @inode: inode to process
+ * @stateid: state ID information
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs_async_inode_return_delegation(struct inode *inode,
+				      const nfs4_stateid *stateid)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_client *clp = server->nfs_client;
+	struct nfs_delegation *delegation;
+
+	filemap_flush(inode->i_mapping);
+
+	rcu_read_lock();
+	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	if (delegation == NULL)
+		goto out_enoent;
+
+	if (!clp->cl_mvops->match_stateid(&delegation->stateid, stateid))
+		goto out_enoent;
+	nfs_mark_return_delegation(server, delegation);
+	rcu_read_unlock();
+
+	nfs_delegation_run_state_manager(clp);
+	return 0;
+out_enoent:
+	rcu_read_unlock();
+	return -ENOENT;
+}
+
+static struct inode *
+nfs_delegation_find_inode_server(struct nfs_server *server,
+				 const struct nfs_fh *fhandle)
+{
+	struct nfs_delegation *delegation;
+	struct inode *res = NULL;
+
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		spin_lock(&delegation->lock);
+		if (delegation->inode != NULL &&
+		    nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) {
+			res = igrab(delegation->inode);
+		}
+		spin_unlock(&delegation->lock);
+		if (res != NULL)
+			break;
+	}
+	return res;
+}
+
+/**
+ * nfs_delegation_find_inode - retrieve the inode associated with a delegation
+ * @clp: client state handle
+ * @fhandle: filehandle from a delegation recall
+ *
+ * Returns pointer to inode matching "fhandle," or NULL if a matching inode
+ * cannot be found.
+ */
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp,
+					const struct nfs_fh *fhandle)
+{
+	struct nfs_server *server;
+	struct inode *res = NULL;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		res = nfs_delegation_find_inode_server(server, fhandle);
+		if (res != NULL)
+			break;
+	}
+	rcu_read_unlock();
+	return res;
+}
+
+static void nfs_delegation_mark_reclaim_server(struct nfs_server *server)
+{
+	struct nfs_delegation *delegation;
+
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list)
+		set_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags);
+}
+
+/**
+ * nfs_delegation_mark_reclaim - mark all delegations as needing to be reclaimed
+ * @clp: nfs_client to process
+ *
+ */
+void nfs_delegation_mark_reclaim(struct nfs_client *clp)
+{
+	struct nfs_server *server;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+		nfs_delegation_mark_reclaim_server(server);
+	rcu_read_unlock();
+}
+
+/**
+ * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done
+ * @clp: nfs_client to process
+ *
+ */
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
+{
+	struct nfs_delegation *delegation;
+	struct nfs_server *server;
+	struct inode *inode;
+
+restart:
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		list_for_each_entry_rcu(delegation, &server->delegations,
+								super_list) {
+			if (test_bit(NFS_DELEGATION_NEED_RECLAIM,
+						&delegation->flags) == 0)
+				continue;
+			inode = nfs_delegation_grab_inode(delegation);
+			if (inode == NULL)
+				continue;
+			delegation = nfs_detach_delegation(NFS_I(inode),
+					delegation, server);
+			rcu_read_unlock();
+
+			if (delegation != NULL)
+				nfs_free_delegation(delegation);
+			iput(inode);
+			goto restart;
+		}
+	}
+	rcu_read_unlock();
+}
+
+/**
+ * nfs_delegations_present - check for existence of delegations
+ * @clp: client state handle
+ *
+ * Returns one if there are any nfs_delegation structures attached
+ * to this nfs_client.
+ */
+int nfs_delegations_present(struct nfs_client *clp)
+{
+	struct nfs_server *server;
+	int ret = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
+		if (!list_empty(&server->delegations)) {
+			ret = 1;
+			break;
+		}
+	rcu_read_unlock();
+	return ret;
+}
+
+/**
+ * nfs4_copy_delegation_stateid - Copy inode's state ID information
+ * @dst: stateid data structure to fill in
+ * @inode: inode to check
+ * @flags: delegation type requirement
+ *
+ * Returns "true" and fills in "dst->data" * if inode had a delegation,
+ * otherwise "false" is returned.
+ */
+bool nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode,
+		fmode_t flags)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_delegation *delegation;
+	bool ret;
+
+	flags &= FMODE_READ|FMODE_WRITE;
+	rcu_read_lock();
+	delegation = rcu_dereference(nfsi->delegation);
+	ret = (delegation != NULL && (delegation->type & flags) == flags);
+	if (ret) {
+		nfs4_stateid_copy(dst, &delegation->stateid);
+		nfs_mark_delegation_referenced(delegation);
+	}
+	rcu_read_unlock();
+	return ret;
+}

fs/nfs/delegation.h

@@ -0,0 +1,73 @@
+/*
+ * linux/fs/nfs/delegation.h
+ *
+ * Copyright (c) Trond Myklebust
+ *
+ * Definitions pertaining to NFS delegated files
+ */
+#ifndef FS_NFS_DELEGATION_H
+#define FS_NFS_DELEGATION_H
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+/*
+ * NFSv4 delegation
+ */
+struct nfs_delegation {
+	struct list_head super_list;
+	struct rpc_cred *cred;
+	struct inode *inode;
+	nfs4_stateid stateid;
+	fmode_t type;
+	loff_t maxsize;
+	__u64 change_attr;
+	unsigned long flags;
+	spinlock_t lock;
+	struct rcu_head rcu;
+};
+
+enum {
+	NFS_DELEGATION_NEED_RECLAIM = 0,
+	NFS_DELEGATION_RETURN,
+	NFS_DELEGATION_RETURN_IF_CLOSED,
+	NFS_DELEGATION_REFERENCED,
+	NFS_DELEGATION_RETURNING,
+	NFS_DELEGATION_REVOKED,
+};
+
+int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
+void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
+int nfs4_inode_return_delegation(struct inode *inode);
+int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
+void nfs_inode_return_delegation_noreclaim(struct inode *inode);
+
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
+void nfs_server_return_all_delegations(struct nfs_server *);
+void nfs_expire_all_delegations(struct nfs_client *clp);
+void nfs_expire_unused_delegation_types(struct nfs_client *clp, fmode_t flags);
+void nfs_expire_unreferenced_delegations(struct nfs_client *clp);
+int nfs_client_return_marked_delegations(struct nfs_client *clp);
+int nfs_delegations_present(struct nfs_client *clp);
+void nfs_remove_bad_delegation(struct inode *inode);
+
+void nfs_delegation_mark_reclaim(struct nfs_client *clp);
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
+
+/* NFSv4 delegation-related procedures */
+int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync);
+int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid);
+int nfs4_lock_delegation_recall(struct file_lock *fl, struct nfs4_state *state, const nfs4_stateid *stateid);
+bool nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode, fmode_t flags);
+
+void nfs_mark_delegation_referenced(struct nfs_delegation *delegation);
+int nfs4_have_delegation(struct inode *inode, fmode_t flags);
+int nfs4_check_delegation(struct inode *inode, fmode_t flags);
+
+#endif
+
+static inline int nfs_have_delegated_attributes(struct inode *inode)
+{
+	return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ) &&
+		!(NFS_I(inode)->cache_validity & NFS_INO_REVAL_FORCED);
+}
+
+#endif

fs/nfs/dns_resolve.c

@@ -0,0 +1,470 @@
+/*
+ * linux/fs/nfs/dns_resolve.c
+ *
+ * Copyright (c) 2009 Trond Myklebust <Trond.Myklebust@netapp.com>
+ *
+ * Resolves DNS hostnames into valid ip addresses
+ */
+
+#ifdef CONFIG_NFS_USE_KERNEL_DNS
+
+#include <linux/module.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/dns_resolver.h>
+#include "dns_resolve.h"
+
+ssize_t nfs_dns_resolve_name(struct net *net, char *name, size_t namelen,
+		struct sockaddr *sa, size_t salen)
+{
+	ssize_t ret;
+	char *ip_addr = NULL;
+	int ip_len;
+
+	ip_len = dns_query(NULL, name, namelen, NULL, &ip_addr, NULL);
+	if (ip_len > 0)
+		ret = rpc_pton(net, ip_addr, ip_len, sa, salen);
+	else
+		ret = -ESRCH;
+	kfree(ip_addr);
+	return ret;
+}
+
+#else
+
+#include <linux/module.h>
+#include <linux/hash.h>
+#include <linux/string.h>
+#include <linux/kmod.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/socket.h>
+#include <linux/seq_file.h>
+#include <linux/inet.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/sunrpc/cache.h>
+#include <linux/sunrpc/svcauth.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+#include <linux/nfs_fs.h>
+
+#include "nfs4_fs.h"
+#include "dns_resolve.h"
+#include "cache_lib.h"
+#include "netns.h"
+
+#define NFS_DNS_HASHBITS 4
+#define NFS_DNS_HASHTBL_SIZE (1 << NFS_DNS_HASHBITS)
+
+struct nfs_dns_ent {
+	struct cache_head h;
+
+	char *hostname;
+	size_t namelen;
+
+	struct sockaddr_storage addr;
+	size_t addrlen;
+};
+
+
+static void nfs_dns_ent_update(struct cache_head *cnew,
+		struct cache_head *ckey)
+{
+	struct nfs_dns_ent *new;
+	struct nfs_dns_ent *key;
+
+	new = container_of(cnew, struct nfs_dns_ent, h);
+	key = container_of(ckey, struct nfs_dns_ent, h);
+
+	memcpy(&new->addr, &key->addr, key->addrlen);
+	new->addrlen = key->addrlen;
+}
+
+static void nfs_dns_ent_init(struct cache_head *cnew,
+		struct cache_head *ckey)
+{
+	struct nfs_dns_ent *new;
+	struct nfs_dns_ent *key;
+
+	new = container_of(cnew, struct nfs_dns_ent, h);
+	key = container_of(ckey, struct nfs_dns_ent, h);
+
+	kfree(new->hostname);
+	new->hostname = kstrndup(key->hostname, key->namelen, GFP_KERNEL);
+	if (new->hostname) {
+		new->namelen = key->namelen;
+		nfs_dns_ent_update(cnew, ckey);
+	} else {
+		new->namelen = 0;
+		new->addrlen = 0;
+	}
+}
+
+static void nfs_dns_ent_put(struct kref *ref)
+{
+	struct nfs_dns_ent *item;
+
+	item = container_of(ref, struct nfs_dns_ent, h.ref);
+	kfree(item->hostname);
+	kfree(item);
+}
+
+static struct cache_head *nfs_dns_ent_alloc(void)
+{
+	struct nfs_dns_ent *item = kmalloc(sizeof(*item), GFP_KERNEL);
+
+	if (item != NULL) {
+		item->hostname = NULL;
+		item->namelen = 0;
+		item->addrlen = 0;
+		return &item->h;
+	}
+	return NULL;
+};
+
+static unsigned int nfs_dns_hash(const struct nfs_dns_ent *key)
+{
+	return hash_str(key->hostname, NFS_DNS_HASHBITS);
+}
+
+static void nfs_dns_request(struct cache_detail *cd,
+		struct cache_head *ch,
+		char **bpp, int *blen)
+{
+	struct nfs_dns_ent *key = container_of(ch, struct nfs_dns_ent, h);
+
+	qword_add(bpp, blen, key->hostname);
+	(*bpp)[-1] = '\n';
+}
+
+static int nfs_dns_upcall(struct cache_detail *cd,
+		struct cache_head *ch)
+{
+	struct nfs_dns_ent *key = container_of(ch, struct nfs_dns_ent, h);
+	int ret;
+
+	ret = nfs_cache_upcall(cd, key->hostname);
+	if (ret)
+		ret = sunrpc_cache_pipe_upcall(cd, ch);
+	return ret;
+}
+
+static int nfs_dns_match(struct cache_head *ca,
+		struct cache_head *cb)
+{
+	struct nfs_dns_ent *a;
+	struct nfs_dns_ent *b;
+
+	a = container_of(ca, struct nfs_dns_ent, h);
+	b = container_of(cb, struct nfs_dns_ent, h);
+
+	if (a->namelen == 0 || a->namelen != b->namelen)
+		return 0;
+	return memcmp(a->hostname, b->hostname, a->namelen) == 0;
+}
+
+static int nfs_dns_show(struct seq_file *m, struct cache_detail *cd,
+		struct cache_head *h)
+{
+	struct nfs_dns_ent *item;
+	long ttl;
+
+	if (h == NULL) {
+		seq_puts(m, "# ip address      hostname        ttl\n");
+		return 0;
+	}
+	item = container_of(h, struct nfs_dns_ent, h);
+	ttl = item->h.expiry_time - seconds_since_boot();
+	if (ttl < 0)
+		ttl = 0;
+
+	if (!test_bit(CACHE_NEGATIVE, &h->flags)) {
+		char buf[INET6_ADDRSTRLEN+IPV6_SCOPE_ID_LEN+1];
+
+		rpc_ntop((struct sockaddr *)&item->addr, buf, sizeof(buf));
+		seq_printf(m, "%15s ", buf);
+	} else
+		seq_puts(m, "<none>          ");
+	seq_printf(m, "%15s %ld\n", item->hostname, ttl);
+	return 0;
+}
+
+static struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
+		struct nfs_dns_ent *key)
+{
+	struct cache_head *ch;
+
+	ch = sunrpc_cache_lookup(cd,
+			&key->h,
+			nfs_dns_hash(key));
+	if (!ch)
+		return NULL;
+	return container_of(ch, struct nfs_dns_ent, h);
+}
+
+static struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd,
+		struct nfs_dns_ent *new,
+		struct nfs_dns_ent *key)
+{
+	struct cache_head *ch;
+
+	ch = sunrpc_cache_update(cd,
+			&new->h, &key->h,
+			nfs_dns_hash(key));
+	if (!ch)
+		return NULL;
+	return container_of(ch, struct nfs_dns_ent, h);
+}
+
+static int nfs_dns_parse(struct cache_detail *cd, char *buf, int buflen)
+{
+	char buf1[NFS_DNS_HOSTNAME_MAXLEN+1];
+	struct nfs_dns_ent key, *item;
+	unsigned int ttl;
+	ssize_t len;
+	int ret = -EINVAL;
+
+	if (buf[buflen-1] != '\n')
+		goto out;
+	buf[buflen-1] = '\0';
+
+	len = qword_get(&buf, buf1, sizeof(buf1));
+	if (len <= 0)
+		goto out;
+	key.addrlen = rpc_pton(cd->net, buf1, len,
+			(struct sockaddr *)&key.addr,
+			sizeof(key.addr));
+
+	len = qword_get(&buf, buf1, sizeof(buf1));
+	if (len <= 0)
+		goto out;
+
+	key.hostname = buf1;
+	key.namelen = len;
+	memset(&key.h, 0, sizeof(key.h));
+
+	if (get_uint(&buf, &ttl) < 0)
+		goto out;
+	if (ttl == 0)
+		goto out;
+	key.h.expiry_time = ttl + seconds_since_boot();
+
+	ret = -ENOMEM;
+	item = nfs_dns_lookup(cd, &key);
+	if (item == NULL)
+		goto out;
+
+	if (key.addrlen == 0)
+		set_bit(CACHE_NEGATIVE, &key.h.flags);
+
+	item = nfs_dns_update(cd, &key, item);
+	if (item == NULL)
+		goto out;
+
+	ret = 0;
+	cache_put(&item->h, cd);
+out:
+	return ret;
+}
+
+static int do_cache_lookup(struct cache_detail *cd,
+		struct nfs_dns_ent *key,
+		struct nfs_dns_ent **item,
+		struct nfs_cache_defer_req *dreq)
+{
+	int ret = -ENOMEM;
+
+	*item = nfs_dns_lookup(cd, key);
+	if (*item) {
+		ret = cache_check(cd, &(*item)->h, &dreq->req);
+		if (ret)
+			*item = NULL;
+	}
+	return ret;
+}
+
+static int do_cache_lookup_nowait(struct cache_detail *cd,
+		struct nfs_dns_ent *key,
+		struct nfs_dns_ent **item)
+{
+	int ret = -ENOMEM;
+
+	*item = nfs_dns_lookup(cd, key);
+	if (!*item)
+		goto out_err;
+	ret = -ETIMEDOUT;
+	if (!test_bit(CACHE_VALID, &(*item)->h.flags)
+			|| (*item)->h.expiry_time < seconds_since_boot()
+			|| cd->flush_time > (*item)->h.last_refresh)
+		goto out_put;
+	ret = -ENOENT;
+	if (test_bit(CACHE_NEGATIVE, &(*item)->h.flags))
+		goto out_put;
+	return 0;
+out_put:
+	cache_put(&(*item)->h, cd);
+out_err:
+	*item = NULL;
+	return ret;
+}
+
+static int do_cache_lookup_wait(struct cache_detail *cd,
+		struct nfs_dns_ent *key,
+		struct nfs_dns_ent **item)
+{
+	struct nfs_cache_defer_req *dreq;
+	int ret = -ENOMEM;
+
+	dreq = nfs_cache_defer_req_alloc();
+	if (!dreq)
+		goto out;
+	ret = do_cache_lookup(cd, key, item, dreq);
+	if (ret == -EAGAIN) {
+		ret = nfs_cache_wait_for_upcall(dreq);
+		if (!ret)
+			ret = do_cache_lookup_nowait(cd, key, item);
+	}
+	nfs_cache_defer_req_put(dreq);
+out:
+	return ret;
+}
+
+ssize_t nfs_dns_resolve_name(struct net *net, char *name,
+		size_t namelen, struct sockaddr *sa, size_t salen)
+{
+	struct nfs_dns_ent key = {
+		.hostname = name,
+		.namelen = namelen,
+	};
+	struct nfs_dns_ent *item = NULL;
+	ssize_t ret;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	ret = do_cache_lookup_wait(nn->nfs_dns_resolve, &key, &item);
+	if (ret == 0) {
+		if (salen >= item->addrlen) {
+			memcpy(sa, &item->addr, item->addrlen);
+			ret = item->addrlen;
+		} else
+			ret = -EOVERFLOW;
+		cache_put(&item->h, nn->nfs_dns_resolve);
+	} else if (ret == -ENOENT)
+		ret = -ESRCH;
+	return ret;
+}
+
+static struct cache_detail nfs_dns_resolve_template = {
+	.owner		= THIS_MODULE,
+	.hash_size	= NFS_DNS_HASHTBL_SIZE,
+	.name		= "dns_resolve",
+	.cache_put	= nfs_dns_ent_put,
+	.cache_upcall	= nfs_dns_upcall,
+	.cache_request	= nfs_dns_request,
+	.cache_parse	= nfs_dns_parse,
+	.cache_show	= nfs_dns_show,
+	.match		= nfs_dns_match,
+	.init		= nfs_dns_ent_init,
+	.update		= nfs_dns_ent_update,
+	.alloc		= nfs_dns_ent_alloc,
+};
+
+
+int nfs_dns_resolver_cache_init(struct net *net)
+{
+	int err;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	nn->nfs_dns_resolve = cache_create_net(&nfs_dns_resolve_template, net);
+	if (IS_ERR(nn->nfs_dns_resolve))
+		return PTR_ERR(nn->nfs_dns_resolve);
+
+	err = nfs_cache_register_net(net, nn->nfs_dns_resolve);
+	if (err)
+		goto err_reg;
+	return 0;
+
+err_reg:
+	cache_destroy_net(nn->nfs_dns_resolve, net);
+	return err;
+}
+
+void nfs_dns_resolver_cache_destroy(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	nfs_cache_unregister_net(net, nn->nfs_dns_resolve);
+	cache_destroy_net(nn->nfs_dns_resolve, net);
+}
+
+static int nfs4_dns_net_init(struct net *net)
+{
+	return nfs_dns_resolver_cache_init(net);
+}
+
+static void nfs4_dns_net_exit(struct net *net)
+{
+	nfs_dns_resolver_cache_destroy(net);
+}
+
+static struct pernet_operations nfs4_dns_resolver_ops = {
+	.init = nfs4_dns_net_init,
+	.exit = nfs4_dns_net_exit,
+};
+
+static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,
+			   void *ptr)
+{
+	struct super_block *sb = ptr;
+	struct net *net = sb->s_fs_info;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct cache_detail *cd = nn->nfs_dns_resolve;
+	int ret = 0;
+
+	if (cd == NULL)
+		return 0;
+
+	if (!try_module_get(THIS_MODULE))
+		return 0;
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		ret = nfs_cache_register_sb(sb, cd);
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		nfs_cache_unregister_sb(sb, cd);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+static struct notifier_block nfs_dns_resolver_block = {
+	.notifier_call	= rpc_pipefs_event,
+};
+
+int nfs_dns_resolver_init(void)
+{
+	int err;
+
+	err = register_pernet_subsys(&nfs4_dns_resolver_ops);
+	if (err < 0)
+		goto out;
+	err = rpc_pipefs_notifier_register(&nfs_dns_resolver_block);
+	if (err < 0)
+		goto out1;
+	return 0;
+out1:
+	unregister_pernet_subsys(&nfs4_dns_resolver_ops);
+out:
+	return err;
+}
+
+void nfs_dns_resolver_destroy(void)
+{
+	rpc_pipefs_notifier_unregister(&nfs_dns_resolver_block);
+	unregister_pernet_subsys(&nfs4_dns_resolver_ops);
+}
+#endif

fs/nfs/dns_resolve.h

@@ -0,0 +1,36 @@
+/*
+ * Resolve DNS hostnames into valid ip addresses
+ */
+#ifndef __LINUX_FS_NFS_DNS_RESOLVE_H
+#define __LINUX_FS_NFS_DNS_RESOLVE_H
+
+#define NFS_DNS_HOSTNAME_MAXLEN	(128)
+
+
+#ifdef CONFIG_NFS_USE_KERNEL_DNS
+static inline int nfs_dns_resolver_init(void)
+{
+	return 0;
+}
+
+static inline void nfs_dns_resolver_destroy(void)
+{}
+
+static inline int nfs_dns_resolver_cache_init(struct net *net)
+{
+	return 0;
+}
+
+static inline void nfs_dns_resolver_cache_destroy(struct net *net)
+{}
+#else
+extern int nfs_dns_resolver_init(void);
+extern void nfs_dns_resolver_destroy(void);
+extern int nfs_dns_resolver_cache_init(struct net *net);
+extern void nfs_dns_resolver_cache_destroy(struct net *net);
+#endif
+
+extern ssize_t nfs_dns_resolve_name(struct net *net, char *name,
+		size_t namelen,	struct sockaddr *sa, size_t salen);
+
+#endif

fs/nfs/file.c

@@ -0,0 +1,940 @@
+/*
+ *  linux/fs/nfs/file.c
+ *
+ *  Copyright (C) 1992  Rick Sladkey
+ *
+ *  Changes Copyright (C) 1994 by Florian La Roche
+ *   - Do not copy data too often around in the kernel.
+ *   - In nfs_file_read the return value of kmalloc wasn't checked.
+ *   - Put in a better version of read look-ahead buffering. Original idea
+ *     and implementation by Wai S Kok elekokws@ee.nus.sg.
+ *
+ *  Expire cache on write to a file by Wai S Kok (Oct 1994).
+ *
+ *  Total rewrite of read side for new NFS buffer cache.. Linus.
+ *
+ *  nfs regular file handling functions
+ */
+
+#include <linux/module.h>
+#include <linux/time.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fcntl.h>
+#include <linux/stat.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/aio.h>
+#include <linux/gfp.h>
+#include <linux/swap.h>
+
+#include <asm/uaccess.h>
+
+#include "delegation.h"
+#include "internal.h"
+#include "iostat.h"
+#include "fscache.h"
+#include "pnfs.h"
+
+#include "nfstrace.h"
+
+#define NFSDBG_FACILITY		NFSDBG_FILE
+
+static const struct vm_operations_struct nfs_file_vm_ops;
+
+/* Hack for future NFS swap support */
+#ifndef IS_SWAPFILE
+# define IS_SWAPFILE(inode)	(0)
+#endif
+
+int nfs_check_flags(int flags)
+{
+	if ((flags & (O_APPEND | O_DIRECT)) == (O_APPEND | O_DIRECT))
+		return -EINVAL;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(nfs_check_flags);
+
+/*
+ * Open file
+ */
+static int
+nfs_file_open(struct inode *inode, struct file *filp)
+{
+	int res;
+
+	dprintk("NFS: open file(%pD2)\n", filp);
+
+	nfs_inc_stats(inode, NFSIOS_VFSOPEN);
+	res = nfs_check_flags(filp->f_flags);
+	if (res)
+		return res;
+
+	res = nfs_open(inode, filp);
+	return res;
+}
+
+int
+nfs_file_release(struct inode *inode, struct file *filp)
+{
+	dprintk("NFS: release(%pD2)\n", filp);
+
+	nfs_inc_stats(inode, NFSIOS_VFSRELEASE);
+	return nfs_release(inode, filp);
+}
+EXPORT_SYMBOL_GPL(nfs_file_release);
+
+/**
+ * nfs_revalidate_size - Revalidate the file size
+ * @inode - pointer to inode struct
+ * @file - pointer to struct file
+ *
+ * Revalidates the file length. This is basically a wrapper around
+ * nfs_revalidate_inode() that takes into account the fact that we may
+ * have cached writes (in which case we don't care about the server's
+ * idea of what the file length is), or O_DIRECT (in which case we
+ * shouldn't trust the cache).
+ */
+static int nfs_revalidate_file_size(struct inode *inode, struct file *filp)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	if (nfs_have_delegated_attributes(inode))
+		goto out_noreval;
+
+	if (filp->f_flags & O_DIRECT)
+		goto force_reval;
+	if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE)
+		goto force_reval;
+	if (nfs_attribute_timeout(inode))
+		goto force_reval;
+out_noreval:
+	return 0;
+force_reval:
+	return __nfs_revalidate_inode(server, inode);
+}
+
+loff_t nfs_file_llseek(struct file *filp, loff_t offset, int whence)
+{
+	dprintk("NFS: llseek file(%pD2, %lld, %d)\n",
+			filp, offset, whence);
+
+	/*
+	 * whence == SEEK_END || SEEK_DATA || SEEK_HOLE => we must revalidate
+	 * the cached file length
+	 */
+	if (whence != SEEK_SET && whence != SEEK_CUR) {
+		struct inode *inode = filp->f_mapping->host;
+
+		int retval = nfs_revalidate_file_size(inode, filp);
+		if (retval < 0)
+			return (loff_t)retval;
+	}
+
+	return generic_file_llseek(filp, offset, whence);
+}
+EXPORT_SYMBOL_GPL(nfs_file_llseek);
+
+/*
+ * Flush all dirty pages, and check for write errors.
+ */
+int
+nfs_file_flush(struct file *file, fl_owner_t id)
+{
+	struct inode	*inode = file_inode(file);
+
+	dprintk("NFS: flush(%pD2)\n", file);
+
+	nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
+	if ((file->f_mode & FMODE_WRITE) == 0)
+		return 0;
+
+	/*
+	 * If we're holding a write delegation, then just start the i/o
+	 * but don't wait for completion (or send a commit).
+	 */
+	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE))
+		return filemap_fdatawrite(file->f_mapping);
+
+	/* Flush writes to the server and return any errors */
+	return vfs_fsync(file, 0);
+}
+EXPORT_SYMBOL_GPL(nfs_file_flush);
+
+ssize_t
+nfs_file_read(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t result;
+
+	if (iocb->ki_filp->f_flags & O_DIRECT)
+		return nfs_file_direct_read(iocb, to, iocb->ki_pos);
+
+	dprintk("NFS: read(%pD2, %zu@%lu)\n",
+		iocb->ki_filp,
+		iov_iter_count(to), (unsigned long) iocb->ki_pos);
+
+	result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
+	if (!result) {
+		result = generic_file_read_iter(iocb, to);
+		if (result > 0)
+			nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, result);
+	}
+	return result;
+}
+EXPORT_SYMBOL_GPL(nfs_file_read);
+
+ssize_t
+nfs_file_splice_read(struct file *filp, loff_t *ppos,
+		     struct pipe_inode_info *pipe, size_t count,
+		     unsigned int flags)
+{
+	struct inode *inode = file_inode(filp);
+	ssize_t res;
+
+	dprintk("NFS: splice_read(%pD2, %lu@%Lu)\n",
+		filp, (unsigned long) count, (unsigned long long) *ppos);
+
+	res = nfs_revalidate_mapping(inode, filp->f_mapping);
+	if (!res) {
+		res = generic_file_splice_read(filp, ppos, pipe, count, flags);
+		if (res > 0)
+			nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, res);
+	}
+	return res;
+}
+EXPORT_SYMBOL_GPL(nfs_file_splice_read);
+
+int
+nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
+{
+	struct inode *inode = file_inode(file);
+	int	status;
+
+	dprintk("NFS: mmap(%pD2)\n", file);
+
+	/* Note: generic_file_mmap() returns ENOSYS on nommu systems
+	 *       so we call that before revalidating the mapping
+	 */
+	status = generic_file_mmap(file, vma);
+	if (!status) {
+		vma->vm_ops = &nfs_file_vm_ops;
+		status = nfs_revalidate_mapping(inode, file->f_mapping);
+	}
+	return status;
+}
+EXPORT_SYMBOL_GPL(nfs_file_mmap);
+
+/*
+ * Flush any dirty pages for this process, and check for write errors.
+ * The return status from this call provides a reliable indication of
+ * whether any write errors occurred for this process.
+ *
+ * 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 occurred, and hence cause it to
+ * fall back to doing a synchronous write.
+ */
+int
+nfs_file_fsync_commit(struct file *file, loff_t start, loff_t end, int datasync)
+{
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
+	struct inode *inode = file_inode(file);
+	int have_error, do_resend, status;
+	int ret = 0;
+
+	dprintk("NFS: fsync file(%pD2) datasync %d\n", file, datasync);
+
+	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
+	do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
+	have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
+	status = nfs_commit_inode(inode, FLUSH_SYNC);
+	have_error |= test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
+	if (have_error) {
+		ret = xchg(&ctx->error, 0);
+		if (ret)
+			goto out;
+	}
+	if (status < 0) {
+		ret = status;
+		goto out;
+	}
+	do_resend |= test_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
+	if (do_resend)
+		ret = -EAGAIN;
+out:
+	return ret;
+}
+EXPORT_SYMBOL_GPL(nfs_file_fsync_commit);
+
+static int
+nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
+{
+	int ret;
+	struct inode *inode = file_inode(file);
+
+	trace_nfs_fsync_enter(inode);
+
+	do {
+		ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+		if (ret != 0)
+			break;
+		mutex_lock(&inode->i_mutex);
+		ret = nfs_file_fsync_commit(file, start, end, datasync);
+		mutex_unlock(&inode->i_mutex);
+		/*
+		 * If nfs_file_fsync_commit detected a server reboot, then
+		 * resend all dirty pages that might have been covered by
+		 * the NFS_CONTEXT_RESEND_WRITES flag
+		 */
+		start = 0;
+		end = LLONG_MAX;
+	} while (ret == -EAGAIN);
+
+	trace_nfs_fsync_exit(inode, ret);
+	return ret;
+}
+
+/*
+ * Decide whether a read/modify/write cycle may be more efficient
+ * then a modify/write/read cycle when writing to a page in the
+ * page cache.
+ *
+ * The modify/write/read cycle may occur if a page is read before
+ * being completely filled by the writer.  In this situation, the
+ * page must be completely written to stable storage on the server
+ * before it can be refilled by reading in the page from the server.
+ * This can lead to expensive, small, FILE_SYNC mode writes being
+ * done.
+ *
+ * It may be more efficient to read the page first if the file is
+ * open for reading in addition to writing, the page is not marked
+ * as Uptodate, it is not dirty or waiting to be committed,
+ * indicating that it was previously allocated and then modified,
+ * that there were valid bytes of data in that range of the file,
+ * and that the new data won't completely replace the old data in
+ * that range of the file.
+ */
+static int nfs_want_read_modify_write(struct file *file, struct page *page,
+			loff_t pos, unsigned len)
+{
+	unsigned int pglen = nfs_page_length(page);
+	unsigned int offset = pos & (PAGE_CACHE_SIZE - 1);
+	unsigned int end = offset + len;
+
+	if (pnfs_ld_read_whole_page(file->f_mapping->host)) {
+		if (!PageUptodate(page))
+			return 1;
+		return 0;
+	}
+
+	if ((file->f_mode & FMODE_READ) &&	/* open for read? */
+	    !PageUptodate(page) &&		/* Uptodate? */
+	    !PagePrivate(page) &&		/* i/o request already? */
+	    pglen &&				/* valid bytes of file? */
+	    (end < pglen || offset))		/* replace all valid bytes? */
+		return 1;
+	return 0;
+}
+
+/*
+ * This does the "real" work of the write. We must allocate and lock the
+ * page to be sent back to the generic routine, which then copies the
+ * data from user space.
+ *
+ * If the writer ends up delaying the write, the writer needs to
+ * increment the page use counts until he is done with the page.
+ */
+static int nfs_write_begin(struct file *file, struct address_space *mapping,
+			loff_t pos, unsigned len, unsigned flags,
+			struct page **pagep, void **fsdata)
+{
+	int ret;
+	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
+	struct page *page;
+	int once_thru = 0;
+
+	dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%lu), %u@%lld)\n",
+		file, mapping->host->i_ino, len, (long long) pos);
+
+start:
+	/*
+	 * Prevent starvation issues if someone is doing a consistency
+	 * sync-to-disk
+	 */
+	ret = wait_on_bit_action(&NFS_I(mapping->host)->flags, NFS_INO_FLUSHING,
+				 nfs_wait_bit_killable, TASK_KILLABLE);
+	if (ret)
+		return ret;
+
+	page = grab_cache_page_write_begin(mapping, index, flags);
+	if (!page)
+		return -ENOMEM;
+	*pagep = page;
+
+	ret = nfs_flush_incompatible(file, page);
+	if (ret) {
+		unlock_page(page);
+		page_cache_release(page);
+	} else if (!once_thru &&
+		   nfs_want_read_modify_write(file, page, pos, len)) {
+		once_thru = 1;
+		ret = nfs_readpage(file, page);
+		page_cache_release(page);
+		if (!ret)
+			goto start;
+	}
+	return ret;
+}
+
+static int nfs_write_end(struct file *file, struct address_space *mapping,
+			loff_t pos, unsigned len, unsigned copied,
+			struct page *page, void *fsdata)
+{
+	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
+	int status;
+
+	dfprintk(PAGECACHE, "NFS: write_end(%pD2(%lu), %u@%lld)\n",
+		file, mapping->host->i_ino, len, (long long) pos);
+
+	/*
+	 * Zero any uninitialised parts of the page, and then mark the page
+	 * as up to date if it turns out that we're extending the file.
+	 */
+	if (!PageUptodate(page)) {
+		unsigned pglen = nfs_page_length(page);
+		unsigned end = offset + len;
+
+		if (pglen == 0) {
+			zero_user_segments(page, 0, offset,
+					end, PAGE_CACHE_SIZE);
+			SetPageUptodate(page);
+		} else if (end >= pglen) {
+			zero_user_segment(page, end, PAGE_CACHE_SIZE);
+			if (offset == 0)
+				SetPageUptodate(page);
+		} else
+			zero_user_segment(page, pglen, PAGE_CACHE_SIZE);
+	}
+
+	status = nfs_updatepage(file, page, offset, copied);
+
+	unlock_page(page);
+	page_cache_release(page);
+
+	if (status < 0)
+		return status;
+	NFS_I(mapping->host)->write_io += copied;
+
+	if (nfs_ctx_key_to_expire(ctx)) {
+		status = nfs_wb_all(mapping->host);
+		if (status < 0)
+			return status;
+	}
+
+	return copied;
+}
+
+/*
+ * Partially or wholly invalidate a page
+ * - Release the private state associated with a page if undergoing complete
+ *   page invalidation
+ * - Called if either PG_private or PG_fscache is set on the page
+ * - Caller holds page lock
+ */
+static void nfs_invalidate_page(struct page *page, unsigned int offset,
+				unsigned int length)
+{
+	dfprintk(PAGECACHE, "NFS: invalidate_page(%p, %u, %u)\n",
+		 page, offset, length);
+
+	if (offset != 0 || length < PAGE_CACHE_SIZE)
+		return;
+	/* Cancel any unstarted writes on this page */
+	nfs_wb_page_cancel(page_file_mapping(page)->host, page);
+
+	nfs_fscache_invalidate_page(page, page->mapping->host);
+}
+
+/*
+ * Attempt to release the private state associated with a page
+ * - Called if either PG_private or PG_fscache is set on the page
+ * - Caller holds page lock
+ * - Return true (may release page) or false (may not)
+ */
+static int nfs_release_page(struct page *page, gfp_t gfp)
+{
+	struct address_space *mapping = page->mapping;
+
+	dfprintk(PAGECACHE, "NFS: release_page(%p)\n", page);
+
+	/* Always try to initiate a 'commit' if relevant, but only
+	 * wait for it if __GFP_WAIT is set.  Even then, only wait 1
+	 * second and only if the 'bdi' is not congested.
+	 * Waiting indefinitely can cause deadlocks when the NFS
+	 * server is on this machine, when a new TCP connection is
+	 * needed and in other rare cases.  There is no particular
+	 * need to wait extensively here.  A short wait has the
+	 * benefit that someone else can worry about the freezer.
+	 */
+	if (mapping) {
+		struct nfs_server *nfss = NFS_SERVER(mapping->host);
+		nfs_commit_inode(mapping->host, 0);
+		if ((gfp & __GFP_WAIT) &&
+		    !bdi_write_congested(&nfss->backing_dev_info)) {
+			wait_on_page_bit_killable_timeout(page, PG_private,
+							  HZ);
+			if (PagePrivate(page))
+				set_bdi_congested(&nfss->backing_dev_info,
+						  BLK_RW_ASYNC);
+		}
+	}
+	/* If PagePrivate() is set, then the page is not freeable */
+	if (PagePrivate(page))
+		return 0;
+	return nfs_fscache_release_page(page, gfp);
+}
+
+static void nfs_check_dirty_writeback(struct page *page,
+				bool *dirty, bool *writeback)
+{
+	struct nfs_inode *nfsi;
+	struct address_space *mapping = page_file_mapping(page);
+
+	if (!mapping || PageSwapCache(page))
+		return;
+
+	/*
+	 * Check if an unstable page is currently being committed and
+	 * if so, have the VM treat it as if the page is under writeback
+	 * so it will not block due to pages that will shortly be freeable.
+	 */
+	nfsi = NFS_I(mapping->host);
+	if (test_bit(NFS_INO_COMMIT, &nfsi->flags)) {
+		*writeback = true;
+		return;
+	}
+
+	/*
+	 * If PagePrivate() is set, then the page is not freeable and as the
+	 * inode is not being committed, it's not going to be cleaned in the
+	 * near future so treat it as dirty
+	 */
+	if (PagePrivate(page))
+		*dirty = true;
+}
+
+/*
+ * Attempt to clear the private state associated with a page when an error
+ * occurs that requires the cached contents of an inode to be written back or
+ * destroyed
+ * - Called if either PG_private or fscache is set on the page
+ * - Caller holds page lock
+ * - Return 0 if successful, -error otherwise
+ */
+static int nfs_launder_page(struct page *page)
+{
+	struct inode *inode = page_file_mapping(page)->host;
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	dfprintk(PAGECACHE, "NFS: launder_page(%ld, %llu)\n",
+		inode->i_ino, (long long)page_offset(page));
+
+	nfs_fscache_wait_on_page_write(nfsi, page);
+	return nfs_wb_page(inode, page);
+}
+
+#ifdef CONFIG_NFS_SWAP
+static int nfs_swap_activate(struct swap_info_struct *sis, struct file *file,
+						sector_t *span)
+{
+	int ret;
+	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
+
+	*span = sis->pages;
+
+	rcu_read_lock();
+	ret = xs_swapper(rcu_dereference(clnt->cl_xprt), 1);
+	rcu_read_unlock();
+
+	return ret;
+}
+
+static void nfs_swap_deactivate(struct file *file)
+{
+	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
+
+	rcu_read_lock();
+	xs_swapper(rcu_dereference(clnt->cl_xprt), 0);
+	rcu_read_unlock();
+}
+#endif
+
+const struct address_space_operations nfs_file_aops = {
+	.readpage = nfs_readpage,
+	.readpages = nfs_readpages,
+	.set_page_dirty = __set_page_dirty_nobuffers,
+	.writepage = nfs_writepage,
+	.writepages = nfs_writepages,
+	.write_begin = nfs_write_begin,
+	.write_end = nfs_write_end,
+	.invalidatepage = nfs_invalidate_page,
+	.releasepage = nfs_release_page,
+	.direct_IO = nfs_direct_IO,
+	.migratepage = nfs_migrate_page,
+	.launder_page = nfs_launder_page,
+	.is_dirty_writeback = nfs_check_dirty_writeback,
+	.error_remove_page = generic_error_remove_page,
+#ifdef CONFIG_NFS_SWAP
+	.swap_activate = nfs_swap_activate,
+	.swap_deactivate = nfs_swap_deactivate,
+#endif
+};
+
+/*
+ * Notification that a PTE pointing to an NFS page is about to be made
+ * writable, implying that someone is about to modify the page through a
+ * shared-writable mapping
+ */
+static int nfs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	struct page *page = vmf->page;
+	struct file *filp = vma->vm_file;
+	struct inode *inode = file_inode(filp);
+	unsigned pagelen;
+	int ret = VM_FAULT_NOPAGE;
+	struct address_space *mapping;
+
+	dfprintk(PAGECACHE, "NFS: vm_page_mkwrite(%pD2(%lu), offset %lld)\n",
+		filp, filp->f_mapping->host->i_ino,
+		(long long)page_offset(page));
+
+	/* make sure the cache has finished storing the page */
+	nfs_fscache_wait_on_page_write(NFS_I(inode), page);
+
+	lock_page(page);
+	mapping = page_file_mapping(page);
+	if (mapping != inode->i_mapping)
+		goto out_unlock;
+
+	wait_on_page_writeback(page);
+
+	pagelen = nfs_page_length(page);
+	if (pagelen == 0)
+		goto out_unlock;
+
+	ret = VM_FAULT_LOCKED;
+	if (nfs_flush_incompatible(filp, page) == 0 &&
+	    nfs_updatepage(filp, page, 0, pagelen) == 0)
+		goto out;
+
+	ret = VM_FAULT_SIGBUS;
+out_unlock:
+	unlock_page(page);
+out:
+	return ret;
+}
+
+static const struct vm_operations_struct nfs_file_vm_ops = {
+	.fault = filemap_fault,
+	.map_pages = filemap_map_pages,
+	.page_mkwrite = nfs_vm_page_mkwrite,
+	.remap_pages = generic_file_remap_pages,
+};
+
+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_DSYNC))
+		return 1;
+	ctx = nfs_file_open_context(filp);
+	if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags) ||
+	    nfs_ctx_key_to_expire(ctx))
+		return 1;
+	return 0;
+}
+
+ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	unsigned long written = 0;
+	ssize_t result;
+	size_t count = iov_iter_count(from);
+	loff_t pos = iocb->ki_pos;
+
+	result = nfs_key_timeout_notify(file, inode);
+	if (result)
+		return result;
+
+	if (file->f_flags & O_DIRECT)
+		return nfs_file_direct_write(iocb, from, pos);
+
+	dprintk("NFS: write(%pD2, %zu@%Ld)\n",
+		file, count, (long long) pos);
+
+	result = -EBUSY;
+	if (IS_SWAPFILE(inode))
+		goto out_swapfile;
+	/*
+	 * O_APPEND implies that we must revalidate the file length.
+	 */
+	if (file->f_flags & O_APPEND) {
+		result = nfs_revalidate_file_size(inode, file);
+		if (result)
+			goto out;
+	}
+
+	result = count;
+	if (!count)
+		goto out;
+
+	result = generic_file_write_iter(iocb, from);
+	if (result > 0)
+		written = result;
+
+	/* Return error values for O_DSYNC and IS_SYNC() */
+	if (result >= 0 && nfs_need_sync_write(file, inode)) {
+		int err = vfs_fsync(file, 0);
+		if (err < 0)
+			result = err;
+	}
+	if (result > 0)
+		nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, written);
+out:
+	return result;
+
+out_swapfile:
+	printk(KERN_INFO "NFS: attempt to write to active swap file!\n");
+	goto out;
+}
+EXPORT_SYMBOL_GPL(nfs_file_write);
+
+static int
+do_getlk(struct file *filp, int cmd, struct file_lock *fl, int is_local)
+{
+	struct inode *inode = filp->f_mapping->host;
+	int status = 0;
+	unsigned int saved_type = fl->fl_type;
+
+	/* Try local locking first */
+	posix_test_lock(filp, fl);
+	if (fl->fl_type != F_UNLCK) {
+		/* found a conflict */
+		goto out;
+	}
+	fl->fl_type = saved_type;
+
+	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+		goto out_noconflict;
+
+	if (is_local)
+		goto out_noconflict;
+
+	status = NFS_PROTO(inode)->lock(filp, cmd, fl);
+out:
+	return status;
+out_noconflict:
+	fl->fl_type = F_UNLCK;
+	goto out;
+}
+
+static int do_vfs_lock(struct file *file, struct file_lock *fl)
+{
+	int res = 0;
+	switch (fl->fl_flags & (FL_POSIX|FL_FLOCK)) {
+		case FL_POSIX:
+			res = posix_lock_file_wait(file, fl);
+			break;
+		case FL_FLOCK:
+			res = flock_lock_file_wait(file, fl);
+			break;
+		default:
+			BUG();
+	}
+	return res;
+}
+
+static int
+do_unlk(struct file *filp, int cmd, struct file_lock *fl, int is_local)
+{
+	struct inode *inode = filp->f_mapping->host;
+	struct nfs_lock_context *l_ctx;
+	int status;
+
+	/*
+	 * Flush all pending writes before doing anything
+	 * with locks..
+	 */
+	nfs_sync_mapping(filp->f_mapping);
+
+	l_ctx = nfs_get_lock_context(nfs_file_open_context(filp));
+	if (!IS_ERR(l_ctx)) {
+		status = nfs_iocounter_wait(&l_ctx->io_count);
+		nfs_put_lock_context(l_ctx);
+		if (status < 0)
+			return status;
+	}
+
+	/* NOTE: special case
+	 * 	If we're signalled while cleaning up locks on process exit, we
+	 * 	still need to complete the unlock.
+	 */
+	/*
+	 * Use local locking if mounted with "-onolock" or with appropriate
+	 * "-olocal_lock="
+	 */
+	if (!is_local)
+		status = NFS_PROTO(inode)->lock(filp, cmd, fl);
+	else
+		status = do_vfs_lock(filp, fl);
+	return status;
+}
+
+static int
+is_time_granular(struct timespec *ts) {
+	return ((ts->tv_sec == 0) && (ts->tv_nsec <= 1000));
+}
+
+static int
+do_setlk(struct file *filp, int cmd, struct file_lock *fl, int is_local)
+{
+	struct inode *inode = filp->f_mapping->host;
+	int status;
+
+	/*
+	 * Flush all pending writes before doing anything
+	 * with locks..
+	 */
+	status = nfs_sync_mapping(filp->f_mapping);
+	if (status != 0)
+		goto out;
+
+	/*
+	 * Use local locking if mounted with "-onolock" or with appropriate
+	 * "-olocal_lock="
+	 */
+	if (!is_local)
+		status = NFS_PROTO(inode)->lock(filp, cmd, fl);
+	else
+		status = do_vfs_lock(filp, fl);
+	if (status < 0)
+		goto out;
+
+	/*
+	 * Revalidate the cache if the server has time stamps granular
+	 * enough to detect subsecond changes.  Otherwise, clear the
+	 * cache to prevent missing any changes.
+	 *
+	 * This makes locking act as a cache coherency point.
+	 */
+	nfs_sync_mapping(filp->f_mapping);
+	if (!NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) {
+		if (is_time_granular(&NFS_SERVER(inode)->time_delta))
+			__nfs_revalidate_inode(NFS_SERVER(inode), inode);
+		else
+			nfs_zap_caches(inode);
+	}
+out:
+	return status;
+}
+
+/*
+ * Lock a (portion of) a file
+ */
+int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
+{
+	struct inode *inode = filp->f_mapping->host;
+	int ret = -ENOLCK;
+	int is_local = 0;
+
+	dprintk("NFS: lock(%pD2, t=%x, fl=%x, r=%lld:%lld)\n",
+			filp, fl->fl_type, fl->fl_flags,
+			(long long)fl->fl_start, (long long)fl->fl_end);
+
+	nfs_inc_stats(inode, NFSIOS_VFSLOCK);
+
+	/* No mandatory locks over NFS */
+	if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
+		goto out_err;
+
+	if (NFS_SERVER(inode)->flags & NFS_MOUNT_LOCAL_FCNTL)
+		is_local = 1;
+
+	if (NFS_PROTO(inode)->lock_check_bounds != NULL) {
+		ret = NFS_PROTO(inode)->lock_check_bounds(fl);
+		if (ret < 0)
+			goto out_err;
+	}
+
+	if (IS_GETLK(cmd))
+		ret = do_getlk(filp, cmd, fl, is_local);
+	else if (fl->fl_type == F_UNLCK)
+		ret = do_unlk(filp, cmd, fl, is_local);
+	else
+		ret = do_setlk(filp, cmd, fl, is_local);
+out_err:
+	return ret;
+}
+EXPORT_SYMBOL_GPL(nfs_lock);
+
+/*
+ * Lock a (portion of) a file
+ */
+int nfs_flock(struct file *filp, int cmd, struct file_lock *fl)
+{
+	struct inode *inode = filp->f_mapping->host;
+	int is_local = 0;
+
+	dprintk("NFS: flock(%pD2, t=%x, fl=%x)\n",
+			filp, fl->fl_type, fl->fl_flags);
+
+	if (!(fl->fl_flags & FL_FLOCK))
+		return -ENOLCK;
+
+	/*
+	 * The NFSv4 protocol doesn't support LOCK_MAND, which is not part of
+	 * any standard. In principle we might be able to support LOCK_MAND
+	 * on NFSv2/3 since NLMv3/4 support DOS share modes, but for now the
+	 * NFS code is not set up for it.
+	 */
+	if (fl->fl_type & LOCK_MAND)
+		return -EINVAL;
+
+	if (NFS_SERVER(inode)->flags & NFS_MOUNT_LOCAL_FLOCK)
+		is_local = 1;
+
+	/* We're simulating flock() locks using posix locks on the server */
+	if (fl->fl_type == F_UNLCK)
+		return do_unlk(filp, cmd, fl, is_local);
+	return do_setlk(filp, cmd, fl, is_local);
+}
+EXPORT_SYMBOL_GPL(nfs_flock);
+
+const struct file_operations nfs_file_operations = {
+	.llseek		= nfs_file_llseek,
+	.read		= new_sync_read,
+	.write		= new_sync_write,
+	.read_iter	= nfs_file_read,
+	.write_iter	= nfs_file_write,
+	.mmap		= nfs_file_mmap,
+	.open		= nfs_file_open,
+	.flush		= nfs_file_flush,
+	.release	= nfs_file_release,
+	.fsync		= nfs_file_fsync,
+	.lock		= nfs_lock,
+	.flock		= nfs_flock,
+	.splice_read	= nfs_file_splice_read,
+	.splice_write	= iter_file_splice_write,
+	.check_flags	= nfs_check_flags,
+	.setlease	= simple_nosetlease,
+};
+EXPORT_SYMBOL_GPL(nfs_file_operations);

fs/nfs/filelayout/Makefile

@@ -0,0 +1,5 @@
+#
+# Makefile for the pNFS Files Layout Driver kernel module
+#
+obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o
+nfs_layout_nfsv41_files-y := filelayout.o filelayoutdev.o

fs/nfs/filelayout/filelayout.h

@@ -0,0 +1,157 @@
+/*
+ *  NFSv4 file layout driver data structures.
+ *
+ *  Copyright (c) 2002
+ *  The Regents of the University of Michigan
+ *  All Rights Reserved
+ *
+ *  Dean Hildebrand <dhildebz@umich.edu>
+ *
+ *  Permission is granted to use, copy, create derivative works, and
+ *  redistribute this software and such derivative works for any purpose,
+ *  so long as the name of the University of Michigan is not used in
+ *  any advertising or publicity pertaining to the use or distribution
+ *  of this software without specific, written prior authorization. If
+ *  the above copyright notice or any other identification of the
+ *  University of Michigan is included in any copy of any portion of
+ *  this software, then the disclaimer below must also be included.
+ *
+ *  This software is provided as is, without representation or warranty
+ *  of any kind either express or implied, including without limitation
+ *  the implied warranties of merchantability, fitness for a particular
+ *  purpose, or noninfringement.  The Regents of the University of
+ *  Michigan shall not be liable for any damages, including special,
+ *  indirect, incidental, or consequential damages, with respect to any
+ *  claim arising out of or in connection with the use of the software,
+ *  even if it has been or is hereafter advised of the possibility of
+ *  such damages.
+ */
+
+#ifndef FS_NFS_NFS4FILELAYOUT_H
+#define FS_NFS_NFS4FILELAYOUT_H
+
+#include "../pnfs.h"
+
+/*
+ * Default data server connection timeout and retrans vaules.
+ * Set by module paramters dataserver_timeo and dataserver_retrans.
+ */
+#define NFS4_DEF_DS_TIMEO   600 /* in tenths of a second */
+#define NFS4_DEF_DS_RETRANS 5
+
+/*
+ * Field testing shows we need to support up to 4096 stripe indices.
+ * We store each index as a u8 (u32 on the wire) to keep the memory footprint
+ * reasonable. This in turn means we support a maximum of 256
+ * RFC 5661 multipath_list4 structures.
+ */
+#define NFS4_PNFS_MAX_STRIPE_CNT 4096
+#define NFS4_PNFS_MAX_MULTI_CNT  256 /* 256 fit into a u8 stripe_index */
+
+/* error codes for internal use */
+#define NFS4ERR_RESET_TO_MDS   12001
+
+enum stripetype4 {
+	STRIPE_SPARSE = 1,
+	STRIPE_DENSE = 2
+};
+
+/* Individual ip address */
+struct nfs4_pnfs_ds_addr {
+	struct sockaddr_storage	da_addr;
+	size_t			da_addrlen;
+	struct list_head	da_node;  /* nfs4_pnfs_dev_hlist dev_dslist */
+	char			*da_remotestr;	/* human readable addr+port */
+};
+
+struct nfs4_pnfs_ds {
+	struct list_head	ds_node;  /* nfs4_pnfs_dev_hlist dev_dslist */
+	char			*ds_remotestr;	/* comma sep list of addrs */
+	struct list_head	ds_addrs;
+	struct nfs_client	*ds_clp;
+	atomic_t		ds_count;
+	unsigned long		ds_state;
+#define NFS4DS_CONNECTING	0	/* ds is establishing connection */
+};
+
+struct nfs4_file_layout_dsaddr {
+	struct nfs4_deviceid_node	id_node;
+	u32				stripe_count;
+	u8				*stripe_indices;
+	u32				ds_num;
+	struct nfs4_pnfs_ds		*ds_list[1];
+};
+
+struct nfs4_filelayout_segment {
+	struct pnfs_layout_segment generic_hdr;
+	u32 stripe_type;
+	u32 commit_through_mds;
+	u32 stripe_unit;
+	u32 first_stripe_index;
+	u64 pattern_offset;
+	struct nfs4_file_layout_dsaddr *dsaddr; /* Point to GETDEVINFO data */
+	unsigned int num_fh;
+	struct nfs_fh **fh_array;
+};
+
+struct nfs4_filelayout {
+	struct pnfs_layout_hdr generic_hdr;
+	struct pnfs_ds_commit_info commit_info;
+};
+
+static inline struct nfs4_filelayout *
+FILELAYOUT_FROM_HDR(struct pnfs_layout_hdr *lo)
+{
+	return container_of(lo, struct nfs4_filelayout, generic_hdr);
+}
+
+static inline struct nfs4_filelayout_segment *
+FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg)
+{
+	return container_of(lseg,
+			    struct nfs4_filelayout_segment,
+			    generic_hdr);
+}
+
+static inline struct nfs4_deviceid_node *
+FILELAYOUT_DEVID_NODE(struct pnfs_layout_segment *lseg)
+{
+	return &FILELAYOUT_LSEG(lseg)->dsaddr->id_node;
+}
+
+static inline void
+filelayout_mark_devid_invalid(struct nfs4_deviceid_node *node)
+{
+	u32 *p = (u32 *)&node->deviceid;
+
+	printk(KERN_WARNING "NFS: Deviceid [%x%x%x%x] marked out of use.\n",
+		p[0], p[1], p[2], p[3]);
+
+	set_bit(NFS_DEVICEID_INVALID, &node->flags);
+}
+
+static inline bool
+filelayout_test_devid_invalid(struct nfs4_deviceid_node *node)
+{
+	return test_bit(NFS_DEVICEID_INVALID, &node->flags);
+}
+
+extern bool
+filelayout_test_devid_unavailable(struct nfs4_deviceid_node *node);
+
+extern struct nfs_fh *
+nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j);
+
+extern void print_ds(struct nfs4_pnfs_ds *ds);
+u32 nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset);
+u32 nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j);
+struct nfs4_pnfs_ds *nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg,
+					u32 ds_idx);
+
+extern struct nfs4_file_layout_dsaddr *
+nfs4_fl_alloc_deviceid_node(struct nfs_server *server,
+	struct pnfs_device *pdev, gfp_t gfp_flags);
+extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr);
+extern void nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr);
+
+#endif /* FS_NFS_NFS4FILELAYOUT_H */

fs/nfs/filelayout/filelayoutdev.c

@@ -0,0 +1,745 @@
+/*
+ *  Device operations for the pnfs nfs4 file layout driver.
+ *
+ *  Copyright (c) 2002
+ *  The Regents of the University of Michigan
+ *  All Rights Reserved
+ *
+ *  Dean Hildebrand <dhildebz@umich.edu>
+ *  Garth Goodson   <Garth.Goodson@netapp.com>
+ *
+ *  Permission is granted to use, copy, create derivative works, and
+ *  redistribute this software and such derivative works for any purpose,
+ *  so long as the name of the University of Michigan is not used in
+ *  any advertising or publicity pertaining to the use or distribution
+ *  of this software without specific, written prior authorization. If
+ *  the above copyright notice or any other identification of the
+ *  University of Michigan is included in any copy of any portion of
+ *  this software, then the disclaimer below must also be included.
+ *
+ *  This software is provided as is, without representation or warranty
+ *  of any kind either express or implied, including without limitation
+ *  the implied warranties of merchantability, fitness for a particular
+ *  purpose, or noninfringement.  The Regents of the University of
+ *  Michigan shall not be liable for any damages, including special,
+ *  indirect, incidental, or consequential damages, with respect to any
+ *  claim arising out of or in connection with the use of the software,
+ *  even if it has been or is hereafter advised of the possibility of
+ *  such damages.
+ */
+
+#include <linux/nfs_fs.h>
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+#include <linux/sunrpc/addr.h>
+
+#include "../internal.h"
+#include "../nfs4session.h"
+#include "filelayout.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS_LD
+
+static unsigned int dataserver_timeo = NFS4_DEF_DS_TIMEO;
+static unsigned int dataserver_retrans = NFS4_DEF_DS_RETRANS;
+
+/*
+ * Data server cache
+ *
+ * Data servers can be mapped to different device ids.
+ * nfs4_pnfs_ds reference counting
+ *   - set to 1 on allocation
+ *   - incremented when a device id maps a data server already in the cache.
+ *   - decremented when deviceid is removed from the cache.
+ */
+static DEFINE_SPINLOCK(nfs4_ds_cache_lock);
+static LIST_HEAD(nfs4_data_server_cache);
+
+/* Debug routines */
+void
+print_ds(struct nfs4_pnfs_ds *ds)
+{
+	if (ds == NULL) {
+		printk("%s NULL device\n", __func__);
+		return;
+	}
+	printk("        ds %s\n"
+		"        ref count %d\n"
+		"        client %p\n"
+		"        cl_exchange_flags %x\n",
+		ds->ds_remotestr,
+		atomic_read(&ds->ds_count), ds->ds_clp,
+		ds->ds_clp ? ds->ds_clp->cl_exchange_flags : 0);
+}
+
+static bool
+same_sockaddr(struct sockaddr *addr1, struct sockaddr *addr2)
+{
+	struct sockaddr_in *a, *b;
+	struct sockaddr_in6 *a6, *b6;
+
+	if (addr1->sa_family != addr2->sa_family)
+		return false;
+
+	switch (addr1->sa_family) {
+	case AF_INET:
+		a = (struct sockaddr_in *)addr1;
+		b = (struct sockaddr_in *)addr2;
+
+		if (a->sin_addr.s_addr == b->sin_addr.s_addr &&
+		    a->sin_port == b->sin_port)
+			return true;
+		break;
+
+	case AF_INET6:
+		a6 = (struct sockaddr_in6 *)addr1;
+		b6 = (struct sockaddr_in6 *)addr2;
+
+		/* LINKLOCAL addresses must have matching scope_id */
+		if (ipv6_addr_src_scope(&a6->sin6_addr) ==
+		    IPV6_ADDR_SCOPE_LINKLOCAL &&
+		    a6->sin6_scope_id != b6->sin6_scope_id)
+			return false;
+
+		if (ipv6_addr_equal(&a6->sin6_addr, &b6->sin6_addr) &&
+		    a6->sin6_port == b6->sin6_port)
+			return true;
+		break;
+
+	default:
+		dprintk("%s: unhandled address family: %u\n",
+			__func__, addr1->sa_family);
+		return false;
+	}
+
+	return false;
+}
+
+static bool
+_same_data_server_addrs_locked(const struct list_head *dsaddrs1,
+			       const struct list_head *dsaddrs2)
+{
+	struct nfs4_pnfs_ds_addr *da1, *da2;
+
+	/* step through both lists, comparing as we go */
+	for (da1 = list_first_entry(dsaddrs1, typeof(*da1), da_node),
+	     da2 = list_first_entry(dsaddrs2, typeof(*da2), da_node);
+	     da1 != NULL && da2 != NULL;
+	     da1 = list_entry(da1->da_node.next, typeof(*da1), da_node),
+	     da2 = list_entry(da2->da_node.next, typeof(*da2), da_node)) {
+		if (!same_sockaddr((struct sockaddr *)&da1->da_addr,
+				   (struct sockaddr *)&da2->da_addr))
+			return false;
+	}
+	if (da1 == NULL && da2 == NULL)
+		return true;
+
+	return false;
+}
+
+/*
+ * Lookup DS by addresses.  nfs4_ds_cache_lock is held
+ */
+static struct nfs4_pnfs_ds *
+_data_server_lookup_locked(const struct list_head *dsaddrs)
+{
+	struct nfs4_pnfs_ds *ds;
+
+	list_for_each_entry(ds, &nfs4_data_server_cache, ds_node)
+		if (_same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs))
+			return ds;
+	return NULL;
+}
+
+/*
+ * Create an rpc connection to the nfs4_pnfs_ds data server
+ * Currently only supports IPv4 and IPv6 addresses
+ */
+static int
+nfs4_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds)
+{
+	struct nfs_client *clp = ERR_PTR(-EIO);
+	struct nfs4_pnfs_ds_addr *da;
+	int status = 0;
+
+	dprintk("--> %s DS %s au_flavor %d\n", __func__, ds->ds_remotestr,
+		mds_srv->nfs_client->cl_rpcclient->cl_auth->au_flavor);
+
+	list_for_each_entry(da, &ds->ds_addrs, da_node) {
+		dprintk("%s: DS %s: trying address %s\n",
+			__func__, ds->ds_remotestr, da->da_remotestr);
+
+		clp = nfs4_set_ds_client(mds_srv->nfs_client,
+					(struct sockaddr *)&da->da_addr,
+					da->da_addrlen, IPPROTO_TCP,
+					dataserver_timeo, dataserver_retrans);
+		if (!IS_ERR(clp))
+			break;
+	}
+
+	if (IS_ERR(clp)) {
+		status = PTR_ERR(clp);
+		goto out;
+	}
+
+	status = nfs4_init_ds_session(clp, mds_srv->nfs_client->cl_lease_time);
+	if (status)
+		goto out_put;
+
+	smp_wmb();
+	ds->ds_clp = clp;
+	dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr);
+out:
+	return status;
+out_put:
+	nfs_put_client(clp);
+	goto out;
+}
+
+static void
+destroy_ds(struct nfs4_pnfs_ds *ds)
+{
+	struct nfs4_pnfs_ds_addr *da;
+
+	dprintk("--> %s\n", __func__);
+	ifdebug(FACILITY)
+		print_ds(ds);
+
+	if (ds->ds_clp)
+		nfs_put_client(ds->ds_clp);
+
+	while (!list_empty(&ds->ds_addrs)) {
+		da = list_first_entry(&ds->ds_addrs,
+				      struct nfs4_pnfs_ds_addr,
+				      da_node);
+		list_del_init(&da->da_node);
+		kfree(da->da_remotestr);
+		kfree(da);
+	}
+
+	kfree(ds->ds_remotestr);
+	kfree(ds);
+}
+
+void
+nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
+{
+	struct nfs4_pnfs_ds *ds;
+	int i;
+
+	nfs4_print_deviceid(&dsaddr->id_node.deviceid);
+
+	for (i = 0; i < dsaddr->ds_num; i++) {
+		ds = dsaddr->ds_list[i];
+		if (ds != NULL) {
+			if (atomic_dec_and_lock(&ds->ds_count,
+						&nfs4_ds_cache_lock)) {
+				list_del_init(&ds->ds_node);
+				spin_unlock(&nfs4_ds_cache_lock);
+				destroy_ds(ds);
+			}
+		}
+	}
+	kfree(dsaddr->stripe_indices);
+	kfree(dsaddr);
+}
+
+/*
+ * Create a string with a human readable address and port to avoid
+ * complicated setup around many dprinks.
+ */
+static char *
+nfs4_pnfs_remotestr(struct list_head *dsaddrs, gfp_t gfp_flags)
+{
+	struct nfs4_pnfs_ds_addr *da;
+	char *remotestr;
+	size_t len;
+	char *p;
+
+	len = 3;        /* '{', '}' and eol */
+	list_for_each_entry(da, dsaddrs, da_node) {
+		len += strlen(da->da_remotestr) + 1;    /* string plus comma */
+	}
+
+	remotestr = kzalloc(len, gfp_flags);
+	if (!remotestr)
+		return NULL;
+
+	p = remotestr;
+	*(p++) = '{';
+	len--;
+	list_for_each_entry(da, dsaddrs, da_node) {
+		size_t ll = strlen(da->da_remotestr);
+
+		if (ll > len)
+			goto out_err;
+
+		memcpy(p, da->da_remotestr, ll);
+		p += ll;
+		len -= ll;
+
+		if (len < 1)
+			goto out_err;
+		(*p++) = ',';
+		len--;
+	}
+	if (len < 2)
+		goto out_err;
+	*(p++) = '}';
+	*p = '\0';
+	return remotestr;
+out_err:
+	kfree(remotestr);
+	return NULL;
+}
+
+static struct nfs4_pnfs_ds *
+nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags)
+{
+	struct nfs4_pnfs_ds *tmp_ds, *ds = NULL;
+	char *remotestr;
+
+	if (list_empty(dsaddrs)) {
+		dprintk("%s: no addresses defined\n", __func__);
+		goto out;
+	}
+
+	ds = kzalloc(sizeof(*ds), gfp_flags);
+	if (!ds)
+		goto out;
+
+	/* this is only used for debugging, so it's ok if its NULL */
+	remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags);
+
+	spin_lock(&nfs4_ds_cache_lock);
+	tmp_ds = _data_server_lookup_locked(dsaddrs);
+	if (tmp_ds == NULL) {
+		INIT_LIST_HEAD(&ds->ds_addrs);
+		list_splice_init(dsaddrs, &ds->ds_addrs);
+		ds->ds_remotestr = remotestr;
+		atomic_set(&ds->ds_count, 1);
+		INIT_LIST_HEAD(&ds->ds_node);
+		ds->ds_clp = NULL;
+		list_add(&ds->ds_node, &nfs4_data_server_cache);
+		dprintk("%s add new data server %s\n", __func__,
+			ds->ds_remotestr);
+	} else {
+		kfree(remotestr);
+		kfree(ds);
+		atomic_inc(&tmp_ds->ds_count);
+		dprintk("%s data server %s found, inc'ed ds_count to %d\n",
+			__func__, tmp_ds->ds_remotestr,
+			atomic_read(&tmp_ds->ds_count));
+		ds = tmp_ds;
+	}
+	spin_unlock(&nfs4_ds_cache_lock);
+out:
+	return ds;
+}
+
+/*
+ * Currently only supports ipv4, ipv6 and one multi-path address.
+ */
+static struct nfs4_pnfs_ds_addr *
+decode_ds_addr(struct net *net, struct xdr_stream *streamp, gfp_t gfp_flags)
+{
+	struct nfs4_pnfs_ds_addr *da = NULL;
+	char *buf, *portstr;
+	__be16 port;
+	int nlen, rlen;
+	int tmp[2];
+	__be32 *p;
+	char *netid, *match_netid;
+	size_t len, match_netid_len;
+	char *startsep = "";
+	char *endsep = "";
+
+
+	/* r_netid */
+	p = xdr_inline_decode(streamp, 4);
+	if (unlikely(!p))
+		goto out_err;
+	nlen = be32_to_cpup(p++);
+
+	p = xdr_inline_decode(streamp, nlen);
+	if (unlikely(!p))
+		goto out_err;
+
+	netid = kmalloc(nlen+1, gfp_flags);
+	if (unlikely(!netid))
+		goto out_err;
+
+	netid[nlen] = '\0';
+	memcpy(netid, p, nlen);
+
+	/* r_addr: ip/ip6addr with port in dec octets - see RFC 5665 */
+	p = xdr_inline_decode(streamp, 4);
+	if (unlikely(!p))
+		goto out_free_netid;
+	rlen = be32_to_cpup(p);
+
+	p = xdr_inline_decode(streamp, rlen);
+	if (unlikely(!p))
+		goto out_free_netid;
+
+	/* port is ".ABC.DEF", 8 chars max */
+	if (rlen > INET6_ADDRSTRLEN + IPV6_SCOPE_ID_LEN + 8) {
+		dprintk("%s: Invalid address, length %d\n", __func__,
+			rlen);
+		goto out_free_netid;
+	}
+	buf = kmalloc(rlen + 1, gfp_flags);
+	if (!buf) {
+		dprintk("%s: Not enough memory\n", __func__);
+		goto out_free_netid;
+	}
+	buf[rlen] = '\0';
+	memcpy(buf, p, rlen);
+
+	/* replace port '.' with '-' */
+	portstr = strrchr(buf, '.');
+	if (!portstr) {
+		dprintk("%s: Failed finding expected dot in port\n",
+			__func__);
+		goto out_free_buf;
+	}
+	*portstr = '-';
+
+	/* find '.' between address and port */
+	portstr = strrchr(buf, '.');
+	if (!portstr) {
+		dprintk("%s: Failed finding expected dot between address and "
+			"port\n", __func__);
+		goto out_free_buf;
+	}
+	*portstr = '\0';
+
+	da = kzalloc(sizeof(*da), gfp_flags);
+	if (unlikely(!da))
+		goto out_free_buf;
+
+	INIT_LIST_HEAD(&da->da_node);
+
+	if (!rpc_pton(net, buf, portstr-buf, (struct sockaddr *)&da->da_addr,
+		      sizeof(da->da_addr))) {
+		dprintk("%s: error parsing address %s\n", __func__, buf);
+		goto out_free_da;
+	}
+
+	portstr++;
+	sscanf(portstr, "%d-%d", &tmp[0], &tmp[1]);
+	port = htons((tmp[0] << 8) | (tmp[1]));
+
+	switch (da->da_addr.ss_family) {
+	case AF_INET:
+		((struct sockaddr_in *)&da->da_addr)->sin_port = port;
+		da->da_addrlen = sizeof(struct sockaddr_in);
+		match_netid = "tcp";
+		match_netid_len = 3;
+		break;
+
+	case AF_INET6:
+		((struct sockaddr_in6 *)&da->da_addr)->sin6_port = port;
+		da->da_addrlen = sizeof(struct sockaddr_in6);
+		match_netid = "tcp6";
+		match_netid_len = 4;
+		startsep = "[";
+		endsep = "]";
+		break;
+
+	default:
+		dprintk("%s: unsupported address family: %u\n",
+			__func__, da->da_addr.ss_family);
+		goto out_free_da;
+	}
+
+	if (nlen != match_netid_len || strncmp(netid, match_netid, nlen)) {
+		dprintk("%s: ERROR: r_netid \"%s\" != \"%s\"\n",
+			__func__, netid, match_netid);
+		goto out_free_da;
+	}
+
+	/* save human readable address */
+	len = strlen(startsep) + strlen(buf) + strlen(endsep) + 7;
+	da->da_remotestr = kzalloc(len, gfp_flags);
+
+	/* NULL is ok, only used for dprintk */
+	if (da->da_remotestr)
+		snprintf(da->da_remotestr, len, "%s%s%s:%u", startsep,
+			 buf, endsep, ntohs(port));
+
+	dprintk("%s: Parsed DS addr %s\n", __func__, da->da_remotestr);
+	kfree(buf);
+	kfree(netid);
+	return da;
+
+out_free_da:
+	kfree(da);
+out_free_buf:
+	dprintk("%s: Error parsing DS addr: %s\n", __func__, buf);
+	kfree(buf);
+out_free_netid:
+	kfree(netid);
+out_err:
+	return NULL;
+}
+
+/* Decode opaque device data and return the result */
+struct nfs4_file_layout_dsaddr *
+nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
+		gfp_t gfp_flags)
+{
+	int i;
+	u32 cnt, num;
+	u8 *indexp;
+	__be32 *p;
+	u8 *stripe_indices;
+	u8 max_stripe_index;
+	struct nfs4_file_layout_dsaddr *dsaddr = NULL;
+	struct xdr_stream stream;
+	struct xdr_buf buf;
+	struct page *scratch;
+	struct list_head dsaddrs;
+	struct nfs4_pnfs_ds_addr *da;
+
+	/* set up xdr stream */
+	scratch = alloc_page(gfp_flags);
+	if (!scratch)
+		goto out_err;
+
+	xdr_init_decode_pages(&stream, &buf, pdev->pages, pdev->pglen);
+	xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);
+
+	/* Get the stripe count (number of stripe index) */
+	p = xdr_inline_decode(&stream, 4);
+	if (unlikely(!p))
+		goto out_err_free_scratch;
+
+	cnt = be32_to_cpup(p);
+	dprintk("%s stripe count  %d\n", __func__, cnt);
+	if (cnt > NFS4_PNFS_MAX_STRIPE_CNT) {
+		printk(KERN_WARNING "NFS: %s: stripe count %d greater than "
+		       "supported maximum %d\n", __func__,
+			cnt, NFS4_PNFS_MAX_STRIPE_CNT);
+		goto out_err_free_scratch;
+	}
+
+	/* read stripe indices */
+	stripe_indices = kcalloc(cnt, sizeof(u8), gfp_flags);
+	if (!stripe_indices)
+		goto out_err_free_scratch;
+
+	p = xdr_inline_decode(&stream, cnt << 2);
+	if (unlikely(!p))
+		goto out_err_free_stripe_indices;
+
+	indexp = &stripe_indices[0];
+	max_stripe_index = 0;
+	for (i = 0; i < cnt; i++) {
+		*indexp = be32_to_cpup(p++);
+		max_stripe_index = max(max_stripe_index, *indexp);
+		indexp++;
+	}
+
+	/* Check the multipath list count */
+	p = xdr_inline_decode(&stream, 4);
+	if (unlikely(!p))
+		goto out_err_free_stripe_indices;
+
+	num = be32_to_cpup(p);
+	dprintk("%s ds_num %u\n", __func__, num);
+	if (num > NFS4_PNFS_MAX_MULTI_CNT) {
+		printk(KERN_WARNING "NFS: %s: multipath count %d greater than "
+			"supported maximum %d\n", __func__,
+			num, NFS4_PNFS_MAX_MULTI_CNT);
+		goto out_err_free_stripe_indices;
+	}
+
+	/* validate stripe indices are all < num */
+	if (max_stripe_index >= num) {
+		printk(KERN_WARNING "NFS: %s: stripe index %u >= num ds %u\n",
+			__func__, max_stripe_index, num);
+		goto out_err_free_stripe_indices;
+	}
+
+	dsaddr = kzalloc(sizeof(*dsaddr) +
+			(sizeof(struct nfs4_pnfs_ds *) * (num - 1)),
+			gfp_flags);
+	if (!dsaddr)
+		goto out_err_free_stripe_indices;
+
+	dsaddr->stripe_count = cnt;
+	dsaddr->stripe_indices = stripe_indices;
+	stripe_indices = NULL;
+	dsaddr->ds_num = num;
+	nfs4_init_deviceid_node(&dsaddr->id_node, server, &pdev->dev_id);
+
+	INIT_LIST_HEAD(&dsaddrs);
+
+	for (i = 0; i < dsaddr->ds_num; i++) {
+		int j;
+		u32 mp_count;
+
+		p = xdr_inline_decode(&stream, 4);
+		if (unlikely(!p))
+			goto out_err_free_deviceid;
+
+		mp_count = be32_to_cpup(p); /* multipath count */
+		for (j = 0; j < mp_count; j++) {
+			da = decode_ds_addr(server->nfs_client->cl_net,
+					    &stream, gfp_flags);
+			if (da)
+				list_add_tail(&da->da_node, &dsaddrs);
+		}
+		if (list_empty(&dsaddrs)) {
+			dprintk("%s: no suitable DS addresses found\n",
+				__func__);
+			goto out_err_free_deviceid;
+		}
+
+		dsaddr->ds_list[i] = nfs4_pnfs_ds_add(&dsaddrs, gfp_flags);
+		if (!dsaddr->ds_list[i])
+			goto out_err_drain_dsaddrs;
+
+		/* If DS was already in cache, free ds addrs */
+		while (!list_empty(&dsaddrs)) {
+			da = list_first_entry(&dsaddrs,
+					      struct nfs4_pnfs_ds_addr,
+					      da_node);
+			list_del_init(&da->da_node);
+			kfree(da->da_remotestr);
+			kfree(da);
+		}
+	}
+
+	__free_page(scratch);
+	return dsaddr;
+
+out_err_drain_dsaddrs:
+	while (!list_empty(&dsaddrs)) {
+		da = list_first_entry(&dsaddrs, struct nfs4_pnfs_ds_addr,
+				      da_node);
+		list_del_init(&da->da_node);
+		kfree(da->da_remotestr);
+		kfree(da);
+	}
+out_err_free_deviceid:
+	nfs4_fl_free_deviceid(dsaddr);
+	/* stripe_indicies was part of dsaddr */
+	goto out_err_free_scratch;
+out_err_free_stripe_indices:
+	kfree(stripe_indices);
+out_err_free_scratch:
+	__free_page(scratch);
+out_err:
+	dprintk("%s ERROR: returning NULL\n", __func__);
+	return NULL;
+}
+
+void
+nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
+{
+	nfs4_put_deviceid_node(&dsaddr->id_node);
+}
+
+/*
+ * Want res = (offset - layout->pattern_offset)/ layout->stripe_unit
+ * Then: ((res + fsi) % dsaddr->stripe_count)
+ */
+u32
+nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset)
+{
+	struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+	u64 tmp;
+
+	tmp = offset - flseg->pattern_offset;
+	do_div(tmp, flseg->stripe_unit);
+	tmp += flseg->first_stripe_index;
+	return do_div(tmp, flseg->dsaddr->stripe_count);
+}
+
+u32
+nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j)
+{
+	return FILELAYOUT_LSEG(lseg)->dsaddr->stripe_indices[j];
+}
+
+struct nfs_fh *
+nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j)
+{
+	struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+	u32 i;
+
+	if (flseg->stripe_type == STRIPE_SPARSE) {
+		if (flseg->num_fh == 1)
+			i = 0;
+		else if (flseg->num_fh == 0)
+			/* Use the MDS OPEN fh set in nfs_read_rpcsetup */
+			return NULL;
+		else
+			i = nfs4_fl_calc_ds_index(lseg, j);
+	} else
+		i = j;
+	return flseg->fh_array[i];
+}
+
+static void nfs4_wait_ds_connect(struct nfs4_pnfs_ds *ds)
+{
+	might_sleep();
+	wait_on_bit_action(&ds->ds_state, NFS4DS_CONNECTING,
+			   nfs_wait_bit_killable, TASK_KILLABLE);
+}
+
+static void nfs4_clear_ds_conn_bit(struct nfs4_pnfs_ds *ds)
+{
+	smp_mb__before_atomic();
+	clear_bit(NFS4DS_CONNECTING, &ds->ds_state);
+	smp_mb__after_atomic();
+	wake_up_bit(&ds->ds_state, NFS4DS_CONNECTING);
+}
+
+
+struct nfs4_pnfs_ds *
+nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx)
+{
+	struct nfs4_file_layout_dsaddr *dsaddr = FILELAYOUT_LSEG(lseg)->dsaddr;
+	struct nfs4_pnfs_ds *ds = dsaddr->ds_list[ds_idx];
+	struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(lseg);
+	struct nfs4_pnfs_ds *ret = ds;
+
+	if (ds == NULL) {
+		printk(KERN_ERR "NFS: %s: No data server for offset index %d\n",
+			__func__, ds_idx);
+		filelayout_mark_devid_invalid(devid);
+		goto out;
+	}
+	smp_rmb();
+	if (ds->ds_clp)
+		goto out_test_devid;
+
+	if (test_and_set_bit(NFS4DS_CONNECTING, &ds->ds_state) == 0) {
+		struct nfs_server *s = NFS_SERVER(lseg->pls_layout->plh_inode);
+		int err;
+
+		err = nfs4_ds_connect(s, ds);
+		if (err)
+			nfs4_mark_deviceid_unavailable(devid);
+		nfs4_clear_ds_conn_bit(ds);
+	} else {
+		/* Either ds is connected, or ds is NULL */
+		nfs4_wait_ds_connect(ds);
+	}
+out_test_devid:
+	if (filelayout_test_devid_unavailable(devid))
+		ret = NULL;
+out:
+	return ret;
+}
+
+module_param(dataserver_retrans, uint, 0644);
+MODULE_PARM_DESC(dataserver_retrans, "The  number of times the NFSv4.1 client "
+			"retries a request before it attempts further "
+			" recovery  action.");
+module_param(dataserver_timeo, uint, 0644);
+MODULE_PARM_DESC(dataserver_timeo, "The time (in tenths of a second) the "
+			"NFSv4.1  client  waits for a response from a "
+			" data server before it retries an NFS request.");

fs/nfs/fscache-index.c

@@ -0,0 +1,336 @@
+/* NFS FS-Cache index structure definition
+ *
+ * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_fs_sb.h>
+#include <linux/in6.h>
+
+#include "internal.h"
+#include "fscache.h"
+
+#define NFSDBG_FACILITY		NFSDBG_FSCACHE
+
+/*
+ * Define the NFS filesystem for FS-Cache.  Upon registration FS-Cache sticks
+ * the cookie for the top-level index object for NFS into here.  The top-level
+ * index can than have other cache objects inserted into it.
+ */
+struct fscache_netfs nfs_fscache_netfs = {
+	.name		= "nfs",
+	.version	= 0,
+};
+
+/*
+ * Register NFS for caching
+ */
+int nfs_fscache_register(void)
+{
+	return fscache_register_netfs(&nfs_fscache_netfs);
+}
+
+/*
+ * Unregister NFS for caching
+ */
+void nfs_fscache_unregister(void)
+{
+	fscache_unregister_netfs(&nfs_fscache_netfs);
+}
+
+/*
+ * Layout of the key for an NFS server cache object.
+ */
+struct nfs_server_key {
+	uint16_t	nfsversion;		/* NFS protocol version */
+	uint16_t	family;			/* address family */
+	uint16_t	port;			/* IP port */
+	union {
+		struct in_addr	ipv4_addr;	/* IPv4 address */
+		struct in6_addr ipv6_addr;	/* IPv6 address */
+	} addr[0];
+};
+
+/*
+ * Generate a key to describe a server in the main NFS index
+ * - We return the length of the key, or 0 if we can't generate one
+ */
+static uint16_t nfs_server_get_key(const void *cookie_netfs_data,
+				   void *buffer, uint16_t bufmax)
+{
+	const struct nfs_client *clp = cookie_netfs_data;
+	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) &clp->cl_addr;
+	const struct sockaddr_in *sin = (struct sockaddr_in *) &clp->cl_addr;
+	struct nfs_server_key *key = buffer;
+	uint16_t len = sizeof(struct nfs_server_key);
+
+	memset(key, 0, len);
+	key->nfsversion = clp->rpc_ops->version;
+	key->family = clp->cl_addr.ss_family;
+
+	switch (clp->cl_addr.ss_family) {
+	case AF_INET:
+		key->port = sin->sin_port;
+		key->addr[0].ipv4_addr = sin->sin_addr;
+		len += sizeof(key->addr[0].ipv4_addr);
+		break;
+
+	case AF_INET6:
+		key->port = sin6->sin6_port;
+		key->addr[0].ipv6_addr = sin6->sin6_addr;
+		len += sizeof(key->addr[0].ipv6_addr);
+		break;
+
+	default:
+		printk(KERN_WARNING "NFS: Unknown network family '%d'\n",
+		       clp->cl_addr.ss_family);
+		len = 0;
+		break;
+	}
+
+	return len;
+}
+
+/*
+ * Define the server object for FS-Cache.  This is used to describe a server
+ * object to fscache_acquire_cookie().  It is keyed by the NFS protocol and
+ * server address parameters.
+ */
+const struct fscache_cookie_def nfs_fscache_server_index_def = {
+	.name		= "NFS.server",
+	.type 		= FSCACHE_COOKIE_TYPE_INDEX,
+	.get_key	= nfs_server_get_key,
+};
+
+/*
+ * Generate a key to describe a superblock key in the main NFS index
+ */
+static uint16_t nfs_super_get_key(const void *cookie_netfs_data,
+				  void *buffer, uint16_t bufmax)
+{
+	const struct nfs_fscache_key *key;
+	const struct nfs_server *nfss = cookie_netfs_data;
+	uint16_t len;
+
+	key = nfss->fscache_key;
+	len = sizeof(key->key) + key->key.uniq_len;
+	if (len > bufmax) {
+		len = 0;
+	} else {
+		memcpy(buffer, &key->key, sizeof(key->key));
+		memcpy(buffer + sizeof(key->key),
+		       key->key.uniquifier, key->key.uniq_len);
+	}
+
+	return len;
+}
+
+/*
+ * Define the superblock object for FS-Cache.  This is used to describe a
+ * superblock object to fscache_acquire_cookie().  It is keyed by all the NFS
+ * parameters that might cause a separate superblock.
+ */
+const struct fscache_cookie_def nfs_fscache_super_index_def = {
+	.name		= "NFS.super",
+	.type 		= FSCACHE_COOKIE_TYPE_INDEX,
+	.get_key	= nfs_super_get_key,
+};
+
+/*
+ * Definition of the auxiliary data attached to NFS inode storage objects
+ * within the cache.
+ *
+ * The contents of this struct are recorded in the on-disk local cache in the
+ * auxiliary data attached to the data storage object backing an inode.  This
+ * permits coherency to be managed when a new inode binds to an already extant
+ * cache object.
+ */
+struct nfs_fscache_inode_auxdata {
+	struct timespec	mtime;
+	struct timespec	ctime;
+	loff_t		size;
+	u64		change_attr;
+};
+
+/*
+ * Generate a key to describe an NFS inode in an NFS server's index
+ */
+static uint16_t nfs_fscache_inode_get_key(const void *cookie_netfs_data,
+					  void *buffer, uint16_t bufmax)
+{
+	const struct nfs_inode *nfsi = cookie_netfs_data;
+	uint16_t nsize;
+
+	/* use the inode's NFS filehandle as the key */
+	nsize = nfsi->fh.size;
+	memcpy(buffer, nfsi->fh.data, nsize);
+	return nsize;
+}
+
+/*
+ * Get certain file attributes from the netfs data
+ * - This function can be absent for an index
+ * - Not permitted to return an error
+ * - The netfs data from the cookie being used as the source is presented
+ */
+static void nfs_fscache_inode_get_attr(const void *cookie_netfs_data,
+				       uint64_t *size)
+{
+	const struct nfs_inode *nfsi = cookie_netfs_data;
+
+	*size = nfsi->vfs_inode.i_size;
+}
+
+/*
+ * Get the auxiliary data from netfs data
+ * - This function can be absent if the index carries no state data
+ * - Should store the auxiliary data in the buffer
+ * - Should return the amount of amount stored
+ * - Not permitted to return an error
+ * - The netfs data from the cookie being used as the source is presented
+ */
+static uint16_t nfs_fscache_inode_get_aux(const void *cookie_netfs_data,
+					  void *buffer, uint16_t bufmax)
+{
+	struct nfs_fscache_inode_auxdata auxdata;
+	const struct nfs_inode *nfsi = cookie_netfs_data;
+
+	memset(&auxdata, 0, sizeof(auxdata));
+	auxdata.size = nfsi->vfs_inode.i_size;
+	auxdata.mtime = nfsi->vfs_inode.i_mtime;
+	auxdata.ctime = nfsi->vfs_inode.i_ctime;
+
+	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
+		auxdata.change_attr = nfsi->vfs_inode.i_version;
+
+	if (bufmax > sizeof(auxdata))
+		bufmax = sizeof(auxdata);
+
+	memcpy(buffer, &auxdata, bufmax);
+	return bufmax;
+}
+
+/*
+ * Consult the netfs about the state of an object
+ * - This function can be absent if the index carries no state data
+ * - The netfs data from the cookie being used as the target is
+ *   presented, as is the auxiliary data
+ */
+static
+enum fscache_checkaux nfs_fscache_inode_check_aux(void *cookie_netfs_data,
+						  const void *data,
+						  uint16_t datalen)
+{
+	struct nfs_fscache_inode_auxdata auxdata;
+	struct nfs_inode *nfsi = cookie_netfs_data;
+
+	if (datalen != sizeof(auxdata))
+		return FSCACHE_CHECKAUX_OBSOLETE;
+
+	memset(&auxdata, 0, sizeof(auxdata));
+	auxdata.size = nfsi->vfs_inode.i_size;
+	auxdata.mtime = nfsi->vfs_inode.i_mtime;
+	auxdata.ctime = nfsi->vfs_inode.i_ctime;
+
+	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
+		auxdata.change_attr = nfsi->vfs_inode.i_version;
+
+	if (memcmp(data, &auxdata, datalen) != 0)
+		return FSCACHE_CHECKAUX_OBSOLETE;
+
+	return FSCACHE_CHECKAUX_OKAY;
+}
+
+/*
+ * Indication from FS-Cache that the cookie is no longer cached
+ * - This function is called when the backing store currently caching a cookie
+ *   is removed
+ * - The netfs should use this to clean up any markers indicating cached pages
+ * - This is mandatory for any object that may have data
+ */
+static void nfs_fscache_inode_now_uncached(void *cookie_netfs_data)
+{
+	struct nfs_inode *nfsi = cookie_netfs_data;
+	struct pagevec pvec;
+	pgoff_t first;
+	int loop, nr_pages;
+
+	pagevec_init(&pvec, 0);
+	first = 0;
+
+	dprintk("NFS: nfs_inode_now_uncached: nfs_inode 0x%p\n", nfsi);
+
+	for (;;) {
+		/* grab a bunch of pages to unmark */
+		nr_pages = pagevec_lookup(&pvec,
+					  nfsi->vfs_inode.i_mapping,
+					  first,
+					  PAGEVEC_SIZE - pagevec_count(&pvec));
+		if (!nr_pages)
+			break;
+
+		for (loop = 0; loop < nr_pages; loop++)
+			ClearPageFsCache(pvec.pages[loop]);
+
+		first = pvec.pages[nr_pages - 1]->index + 1;
+
+		pvec.nr = nr_pages;
+		pagevec_release(&pvec);
+		cond_resched();
+	}
+}
+
+/*
+ * Get an extra reference on a read context.
+ * - This function can be absent if the completion function doesn't require a
+ *   context.
+ * - The read context is passed back to NFS in the event that a data read on the
+ *   cache fails with EIO - in which case the server must be contacted to
+ *   retrieve the data, which requires the read context for security.
+ */
+static void nfs_fh_get_context(void *cookie_netfs_data, void *context)
+{
+	get_nfs_open_context(context);
+}
+
+/*
+ * Release an extra reference on a read context.
+ * - This function can be absent if the completion function doesn't require a
+ *   context.
+ */
+static void nfs_fh_put_context(void *cookie_netfs_data, void *context)
+{
+	if (context)
+		put_nfs_open_context(context);
+}
+
+/*
+ * Define the inode object for FS-Cache.  This is used to describe an inode
+ * object to fscache_acquire_cookie().  It is keyed by the NFS file handle for
+ * an inode.
+ *
+ * Coherency is managed by comparing the copies of i_size, i_mtime and i_ctime
+ * held in the cache auxiliary data for the data storage object with those in
+ * the inode struct in memory.
+ */
+const struct fscache_cookie_def nfs_fscache_inode_object_def = {
+	.name		= "NFS.fh",
+	.type		= FSCACHE_COOKIE_TYPE_DATAFILE,
+	.get_key	= nfs_fscache_inode_get_key,
+	.get_attr	= nfs_fscache_inode_get_attr,
+	.get_aux	= nfs_fscache_inode_get_aux,
+	.check_aux	= nfs_fscache_inode_check_aux,
+	.now_uncached	= nfs_fscache_inode_now_uncached,
+	.get_context	= nfs_fh_get_context,
+	.put_context	= nfs_fh_put_context,
+};

fs/nfs/fscache.c

@@ -0,0 +1,439 @@
+/* NFS filesystem cache interface
+ *
+ * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_fs_sb.h>
+#include <linux/in6.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+
+#include "internal.h"
+#include "iostat.h"
+#include "fscache.h"
+
+#define NFSDBG_FACILITY		NFSDBG_FSCACHE
+
+static struct rb_root nfs_fscache_keys = RB_ROOT;
+static DEFINE_SPINLOCK(nfs_fscache_keys_lock);
+
+/*
+ * Get the per-client index cookie for an NFS client if the appropriate mount
+ * flag was set
+ * - We always try and get an index cookie for the client, but get filehandle
+ *   cookies on a per-superblock basis, depending on the mount flags
+ */
+void nfs_fscache_get_client_cookie(struct nfs_client *clp)
+{
+	/* create a cache index for looking up filehandles */
+	clp->fscache = fscache_acquire_cookie(nfs_fscache_netfs.primary_index,
+					      &nfs_fscache_server_index_def,
+					      clp, true);
+	dfprintk(FSCACHE, "NFS: get client cookie (0x%p/0x%p)\n",
+		 clp, clp->fscache);
+}
+
+/*
+ * Dispose of a per-client cookie
+ */
+void nfs_fscache_release_client_cookie(struct nfs_client *clp)
+{
+	dfprintk(FSCACHE, "NFS: releasing client cookie (0x%p/0x%p)\n",
+		 clp, clp->fscache);
+
+	fscache_relinquish_cookie(clp->fscache, 0);
+	clp->fscache = NULL;
+}
+
+/*
+ * Get the cache cookie for an NFS superblock.  We have to handle
+ * uniquification here because the cache doesn't do it for us.
+ *
+ * The default uniquifier is just an empty string, but it may be overridden
+ * either by the 'fsc=xxx' option to mount, or by inheriting it from the parent
+ * superblock across an automount point of some nature.
+ */
+void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int ulen)
+{
+	struct nfs_fscache_key *key, *xkey;
+	struct nfs_server *nfss = NFS_SB(sb);
+	struct rb_node **p, *parent;
+	int diff;
+
+	if (!uniq) {
+		uniq = "";
+		ulen = 1;
+	}
+
+	key = kzalloc(sizeof(*key) + ulen, GFP_KERNEL);
+	if (!key)
+		return;
+
+	key->nfs_client = nfss->nfs_client;
+	key->key.super.s_flags = sb->s_flags & NFS_MS_MASK;
+	key->key.nfs_server.flags = nfss->flags;
+	key->key.nfs_server.rsize = nfss->rsize;
+	key->key.nfs_server.wsize = nfss->wsize;
+	key->key.nfs_server.acregmin = nfss->acregmin;
+	key->key.nfs_server.acregmax = nfss->acregmax;
+	key->key.nfs_server.acdirmin = nfss->acdirmin;
+	key->key.nfs_server.acdirmax = nfss->acdirmax;
+	key->key.nfs_server.fsid = nfss->fsid;
+	key->key.rpc_auth.au_flavor = nfss->client->cl_auth->au_flavor;
+
+	key->key.uniq_len = ulen;
+	memcpy(key->key.uniquifier, uniq, ulen);
+
+	spin_lock(&nfs_fscache_keys_lock);
+	p = &nfs_fscache_keys.rb_node;
+	parent = NULL;
+	while (*p) {
+		parent = *p;
+		xkey = rb_entry(parent, struct nfs_fscache_key, node);
+
+		if (key->nfs_client < xkey->nfs_client)
+			goto go_left;
+		if (key->nfs_client > xkey->nfs_client)
+			goto go_right;
+
+		diff = memcmp(&key->key, &xkey->key, sizeof(key->key));
+		if (diff < 0)
+			goto go_left;
+		if (diff > 0)
+			goto go_right;
+
+		if (key->key.uniq_len == 0)
+			goto non_unique;
+		diff = memcmp(key->key.uniquifier,
+			      xkey->key.uniquifier,
+			      key->key.uniq_len);
+		if (diff < 0)
+			goto go_left;
+		if (diff > 0)
+			goto go_right;
+		goto non_unique;
+
+	go_left:
+		p = &(*p)->rb_left;
+		continue;
+	go_right:
+		p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&key->node, parent, p);
+	rb_insert_color(&key->node, &nfs_fscache_keys);
+	spin_unlock(&nfs_fscache_keys_lock);
+	nfss->fscache_key = key;
+
+	/* create a cache index for looking up filehandles */
+	nfss->fscache = fscache_acquire_cookie(nfss->nfs_client->fscache,
+					       &nfs_fscache_super_index_def,
+					       nfss, true);
+	dfprintk(FSCACHE, "NFS: get superblock cookie (0x%p/0x%p)\n",
+		 nfss, nfss->fscache);
+	return;
+
+non_unique:
+	spin_unlock(&nfs_fscache_keys_lock);
+	kfree(key);
+	nfss->fscache_key = NULL;
+	nfss->fscache = NULL;
+	printk(KERN_WARNING "NFS:"
+	       " Cache request denied due to non-unique superblock keys\n");
+}
+
+/*
+ * release a per-superblock cookie
+ */
+void nfs_fscache_release_super_cookie(struct super_block *sb)
+{
+	struct nfs_server *nfss = NFS_SB(sb);
+
+	dfprintk(FSCACHE, "NFS: releasing superblock cookie (0x%p/0x%p)\n",
+		 nfss, nfss->fscache);
+
+	fscache_relinquish_cookie(nfss->fscache, 0);
+	nfss->fscache = NULL;
+
+	if (nfss->fscache_key) {
+		spin_lock(&nfs_fscache_keys_lock);
+		rb_erase(&nfss->fscache_key->node, &nfs_fscache_keys);
+		spin_unlock(&nfs_fscache_keys_lock);
+		kfree(nfss->fscache_key);
+		nfss->fscache_key = NULL;
+	}
+}
+
+/*
+ * Initialise the per-inode cache cookie pointer for an NFS inode.
+ */
+void nfs_fscache_init_inode(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	nfsi->fscache = NULL;
+	if (!S_ISREG(inode->i_mode))
+		return;
+	nfsi->fscache = fscache_acquire_cookie(NFS_SB(inode->i_sb)->fscache,
+					       &nfs_fscache_inode_object_def,
+					       nfsi, false);
+}
+
+/*
+ * Release a per-inode cookie.
+ */
+void nfs_fscache_clear_inode(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct fscache_cookie *cookie = nfs_i_fscache(inode);
+
+	dfprintk(FSCACHE, "NFS: clear cookie (0x%p/0x%p)\n", nfsi, cookie);
+
+	fscache_relinquish_cookie(cookie, false);
+	nfsi->fscache = NULL;
+}
+
+static bool nfs_fscache_can_enable(void *data)
+{
+	struct inode *inode = data;
+
+	return !inode_is_open_for_write(inode);
+}
+
+/*
+ * Enable or disable caching for a file that is being opened as appropriate.
+ * The cookie is allocated when the inode is initialised, but is not enabled at
+ * that time.  Enablement is deferred to file-open time to avoid stat() and
+ * access() thrashing the cache.
+ *
+ * For now, with NFS, only regular files that are open read-only will be able
+ * to use the cache.
+ *
+ * We enable the cache for an inode if we open it read-only and it isn't
+ * currently open for writing.  We disable the cache if the inode is open
+ * write-only.
+ *
+ * The caller uses the file struct to pin i_writecount on the inode before
+ * calling us when a file is opened for writing, so we can make use of that.
+ *
+ * Note that this may be invoked multiple times in parallel by parallel
+ * nfs_open() functions.
+ */
+void nfs_fscache_open_file(struct inode *inode, struct file *filp)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct fscache_cookie *cookie = nfs_i_fscache(inode);
+
+	if (!fscache_cookie_valid(cookie))
+		return;
+
+	if (inode_is_open_for_write(inode)) {
+		dfprintk(FSCACHE, "NFS: nfsi 0x%p disabling cache\n", nfsi);
+		clear_bit(NFS_INO_FSCACHE, &nfsi->flags);
+		fscache_disable_cookie(cookie, true);
+		fscache_uncache_all_inode_pages(cookie, inode);
+	} else {
+		dfprintk(FSCACHE, "NFS: nfsi 0x%p enabling cache\n", nfsi);
+		fscache_enable_cookie(cookie, nfs_fscache_can_enable, inode);
+		if (fscache_cookie_enabled(cookie))
+			set_bit(NFS_INO_FSCACHE, &NFS_I(inode)->flags);
+	}
+}
+EXPORT_SYMBOL_GPL(nfs_fscache_open_file);
+
+/*
+ * Release the caching state associated with a page, if the page isn't busy
+ * interacting with the cache.
+ * - Returns true (can release page) or false (page busy).
+ */
+int nfs_fscache_release_page(struct page *page, gfp_t gfp)
+{
+	if (PageFsCache(page)) {
+		struct fscache_cookie *cookie = nfs_i_fscache(page->mapping->host);
+
+		BUG_ON(!cookie);
+		dfprintk(FSCACHE, "NFS: fscache releasepage (0x%p/0x%p/0x%p)\n",
+			 cookie, page, NFS_I(page->mapping->host));
+
+		if (!fscache_maybe_release_page(cookie, page, gfp))
+			return 0;
+
+		nfs_add_fscache_stats(page->mapping->host,
+				      NFSIOS_FSCACHE_PAGES_UNCACHED, 1);
+	}
+
+	return 1;
+}
+
+/*
+ * Release the caching state associated with a page if undergoing complete page
+ * invalidation.
+ */
+void __nfs_fscache_invalidate_page(struct page *page, struct inode *inode)
+{
+	struct fscache_cookie *cookie = nfs_i_fscache(inode);
+
+	BUG_ON(!cookie);
+
+	dfprintk(FSCACHE, "NFS: fscache invalidatepage (0x%p/0x%p/0x%p)\n",
+		 cookie, page, NFS_I(inode));
+
+	fscache_wait_on_page_write(cookie, page);
+
+	BUG_ON(!PageLocked(page));
+	fscache_uncache_page(cookie, page);
+	nfs_add_fscache_stats(page->mapping->host,
+			      NFSIOS_FSCACHE_PAGES_UNCACHED, 1);
+}
+
+/*
+ * Handle completion of a page being read from the cache.
+ * - Called in process (keventd) context.
+ */
+static void nfs_readpage_from_fscache_complete(struct page *page,
+					       void *context,
+					       int error)
+{
+	dfprintk(FSCACHE,
+		 "NFS: readpage_from_fscache_complete (0x%p/0x%p/%d)\n",
+		 page, context, error);
+
+	/* if the read completes with an error, we just unlock the page and let
+	 * the VM reissue the readpage */
+	if (!error) {
+		SetPageUptodate(page);
+		unlock_page(page);
+	} else {
+		error = nfs_readpage_async(context, page->mapping->host, page);
+		if (error)
+			unlock_page(page);
+	}
+}
+
+/*
+ * Retrieve a page from fscache
+ */
+int __nfs_readpage_from_fscache(struct nfs_open_context *ctx,
+				struct inode *inode, struct page *page)
+{
+	int ret;
+
+	dfprintk(FSCACHE,
+		 "NFS: readpage_from_fscache(fsc:%p/p:%p(i:%lx f:%lx)/0x%p)\n",
+		 nfs_i_fscache(inode), page, page->index, page->flags, inode);
+
+	ret = fscache_read_or_alloc_page(nfs_i_fscache(inode),
+					 page,
+					 nfs_readpage_from_fscache_complete,
+					 ctx,
+					 GFP_KERNEL);
+
+	switch (ret) {
+	case 0: /* read BIO submitted (page in fscache) */
+		dfprintk(FSCACHE,
+			 "NFS:    readpage_from_fscache: BIO submitted\n");
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_OK, 1);
+		return ret;
+
+	case -ENOBUFS: /* inode not in cache */
+	case -ENODATA: /* page not in cache */
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_FAIL, 1);
+		dfprintk(FSCACHE,
+			 "NFS:    readpage_from_fscache %d\n", ret);
+		return 1;
+
+	default:
+		dfprintk(FSCACHE, "NFS:    readpage_from_fscache %d\n", ret);
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_FAIL, 1);
+	}
+	return ret;
+}
+
+/*
+ * Retrieve a set of pages from fscache
+ */
+int __nfs_readpages_from_fscache(struct nfs_open_context *ctx,
+				 struct inode *inode,
+				 struct address_space *mapping,
+				 struct list_head *pages,
+				 unsigned *nr_pages)
+{
+	unsigned npages = *nr_pages;
+	int ret;
+
+	dfprintk(FSCACHE, "NFS: nfs_getpages_from_fscache (0x%p/%u/0x%p)\n",
+		 nfs_i_fscache(inode), npages, inode);
+
+	ret = fscache_read_or_alloc_pages(nfs_i_fscache(inode),
+					  mapping, pages, nr_pages,
+					  nfs_readpage_from_fscache_complete,
+					  ctx,
+					  mapping_gfp_mask(mapping));
+	if (*nr_pages < npages)
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_OK,
+				      npages);
+	if (*nr_pages > 0)
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_FAIL,
+				      *nr_pages);
+
+	switch (ret) {
+	case 0: /* read submitted to the cache for all pages */
+		BUG_ON(!list_empty(pages));
+		BUG_ON(*nr_pages != 0);
+		dfprintk(FSCACHE,
+			 "NFS: nfs_getpages_from_fscache: submitted\n");
+
+		return ret;
+
+	case -ENOBUFS: /* some pages aren't cached and can't be */
+	case -ENODATA: /* some pages aren't cached */
+		dfprintk(FSCACHE,
+			 "NFS: nfs_getpages_from_fscache: no page: %d\n", ret);
+		return 1;
+
+	default:
+		dfprintk(FSCACHE,
+			 "NFS: nfs_getpages_from_fscache: ret  %d\n", ret);
+	}
+
+	return ret;
+}
+
+/*
+ * Store a newly fetched page in fscache
+ * - PG_fscache must be set on the page
+ */
+void __nfs_readpage_to_fscache(struct inode *inode, struct page *page, int sync)
+{
+	int ret;
+
+	dfprintk(FSCACHE,
+		 "NFS: readpage_to_fscache(fsc:%p/p:%p(i:%lx f:%lx)/%d)\n",
+		 nfs_i_fscache(inode), page, page->index, page->flags, sync);
+
+	ret = fscache_write_page(nfs_i_fscache(inode), page, GFP_KERNEL);
+	dfprintk(FSCACHE,
+		 "NFS:     readpage_to_fscache: p:%p(i:%lu f:%lx) ret %d\n",
+		 page, page->index, page->flags, ret);
+
+	if (ret != 0) {
+		fscache_uncache_page(nfs_i_fscache(inode), page);
+		nfs_add_fscache_stats(inode,
+				      NFSIOS_FSCACHE_PAGES_WRITTEN_FAIL, 1);
+		nfs_add_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_UNCACHED, 1);
+	} else {
+		nfs_add_fscache_stats(inode,
+				      NFSIOS_FSCACHE_PAGES_WRITTEN_OK, 1);
+	}
+}

fs/nfs/fscache.h

@@ -0,0 +1,229 @@
+/* NFS filesystem cache interface definitions
+ *
+ * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#ifndef _NFS_FSCACHE_H
+#define _NFS_FSCACHE_H
+
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/nfs4_mount.h>
+#include <linux/fscache.h>
+
+#ifdef CONFIG_NFS_FSCACHE
+
+/*
+ * set of NFS FS-Cache objects that form a superblock key
+ */
+struct nfs_fscache_key {
+	struct rb_node		node;
+	struct nfs_client	*nfs_client;	/* the server */
+
+	/* the elements of the unique key - as used by nfs_compare_super() and
+	 * nfs_compare_mount_options() to distinguish superblocks */
+	struct {
+		struct {
+			unsigned long	s_flags;	/* various flags
+							 * (& NFS_MS_MASK) */
+		} super;
+
+		struct {
+			struct nfs_fsid fsid;
+			int		flags;
+			unsigned int	rsize;		/* read size */
+			unsigned int	wsize;		/* write size */
+			unsigned int	acregmin;	/* attr cache timeouts */
+			unsigned int	acregmax;
+			unsigned int	acdirmin;
+			unsigned int	acdirmax;
+		} nfs_server;
+
+		struct {
+			rpc_authflavor_t au_flavor;
+		} rpc_auth;
+
+		/* uniquifier - can be used if nfs_server.flags includes
+		 * NFS_MOUNT_UNSHARED  */
+		u8 uniq_len;
+		char uniquifier[0];
+	} key;
+};
+
+/*
+ * fscache-index.c
+ */
+extern struct fscache_netfs nfs_fscache_netfs;
+extern const struct fscache_cookie_def nfs_fscache_server_index_def;
+extern const struct fscache_cookie_def nfs_fscache_super_index_def;
+extern const struct fscache_cookie_def nfs_fscache_inode_object_def;
+
+extern int nfs_fscache_register(void);
+extern void nfs_fscache_unregister(void);
+
+/*
+ * fscache.c
+ */
+extern void nfs_fscache_get_client_cookie(struct nfs_client *);
+extern void nfs_fscache_release_client_cookie(struct nfs_client *);
+
+extern void nfs_fscache_get_super_cookie(struct super_block *, const char *, int);
+extern void nfs_fscache_release_super_cookie(struct super_block *);
+
+extern void nfs_fscache_init_inode(struct inode *);
+extern void nfs_fscache_clear_inode(struct inode *);
+extern void nfs_fscache_open_file(struct inode *, struct file *);
+
+extern void __nfs_fscache_invalidate_page(struct page *, struct inode *);
+extern int nfs_fscache_release_page(struct page *, gfp_t);
+
+extern int __nfs_readpage_from_fscache(struct nfs_open_context *,
+				       struct inode *, struct page *);
+extern int __nfs_readpages_from_fscache(struct nfs_open_context *,
+					struct inode *, struct address_space *,
+					struct list_head *, unsigned *);
+extern void __nfs_readpage_to_fscache(struct inode *, struct page *, int);
+
+/*
+ * wait for a page to complete writing to the cache
+ */
+static inline void nfs_fscache_wait_on_page_write(struct nfs_inode *nfsi,
+						  struct page *page)
+{
+	if (PageFsCache(page))
+		fscache_wait_on_page_write(nfsi->fscache, page);
+}
+
+/*
+ * release the caching state associated with a page if undergoing complete page
+ * invalidation
+ */
+static inline void nfs_fscache_invalidate_page(struct page *page,
+					       struct inode *inode)
+{
+	if (PageFsCache(page))
+		__nfs_fscache_invalidate_page(page, inode);
+}
+
+/*
+ * Retrieve a page from an inode data storage object.
+ */
+static inline int nfs_readpage_from_fscache(struct nfs_open_context *ctx,
+					    struct inode *inode,
+					    struct page *page)
+{
+	if (NFS_I(inode)->fscache)
+		return __nfs_readpage_from_fscache(ctx, inode, page);
+	return -ENOBUFS;
+}
+
+/*
+ * Retrieve a set of pages from an inode data storage object.
+ */
+static inline int nfs_readpages_from_fscache(struct nfs_open_context *ctx,
+					     struct inode *inode,
+					     struct address_space *mapping,
+					     struct list_head *pages,
+					     unsigned *nr_pages)
+{
+	if (NFS_I(inode)->fscache)
+		return __nfs_readpages_from_fscache(ctx, inode, mapping, pages,
+						    nr_pages);
+	return -ENOBUFS;
+}
+
+/*
+ * Store a page newly fetched from the server in an inode data storage object
+ * in the cache.
+ */
+static inline void nfs_readpage_to_fscache(struct inode *inode,
+					   struct page *page,
+					   int sync)
+{
+	if (PageFsCache(page))
+		__nfs_readpage_to_fscache(inode, page, sync);
+}
+
+/*
+ * Invalidate the contents of fscache for this inode.  This will not sleep.
+ */
+static inline void nfs_fscache_invalidate(struct inode *inode)
+{
+	fscache_invalidate(NFS_I(inode)->fscache);
+}
+
+/*
+ * Wait for an object to finish being invalidated.
+ */
+static inline void nfs_fscache_wait_on_invalidate(struct inode *inode)
+{
+	fscache_wait_on_invalidate(NFS_I(inode)->fscache);
+}
+
+/*
+ * indicate the client caching state as readable text
+ */
+static inline const char *nfs_server_fscache_state(struct nfs_server *server)
+{
+	if (server->fscache && (server->options & NFS_OPTION_FSCACHE))
+		return "yes";
+	return "no ";
+}
+
+#else /* CONFIG_NFS_FSCACHE */
+static inline int nfs_fscache_register(void) { return 0; }
+static inline void nfs_fscache_unregister(void) {}
+
+static inline void nfs_fscache_get_client_cookie(struct nfs_client *clp) {}
+static inline void nfs_fscache_release_client_cookie(struct nfs_client *clp) {}
+
+static inline void nfs_fscache_release_super_cookie(struct super_block *sb) {}
+
+static inline void nfs_fscache_init_inode(struct inode *inode) {}
+static inline void nfs_fscache_clear_inode(struct inode *inode) {}
+static inline void nfs_fscache_open_file(struct inode *inode,
+					 struct file *filp) {}
+
+static inline int nfs_fscache_release_page(struct page *page, gfp_t gfp)
+{
+	return 1; /* True: may release page */
+}
+static inline void nfs_fscache_invalidate_page(struct page *page,
+					       struct inode *inode) {}
+static inline void nfs_fscache_wait_on_page_write(struct nfs_inode *nfsi,
+						  struct page *page) {}
+
+static inline int nfs_readpage_from_fscache(struct nfs_open_context *ctx,
+					    struct inode *inode,
+					    struct page *page)
+{
+	return -ENOBUFS;
+}
+static inline int nfs_readpages_from_fscache(struct nfs_open_context *ctx,
+					     struct inode *inode,
+					     struct address_space *mapping,
+					     struct list_head *pages,
+					     unsigned *nr_pages)
+{
+	return -ENOBUFS;
+}
+static inline void nfs_readpage_to_fscache(struct inode *inode,
+					   struct page *page, int sync) {}
+
+
+static inline void nfs_fscache_invalidate(struct inode *inode) {}
+static inline void nfs_fscache_wait_on_invalidate(struct inode *inode) {}
+
+static inline const char *nfs_server_fscache_state(struct nfs_server *server)
+{
+	return "no ";
+}
+
+#endif /* CONFIG_NFS_FSCACHE */
+#endif /* _NFS_FSCACHE_H */

fs/nfs/getroot.c

@@ -0,0 +1,133 @@
+/* getroot.c: get the root dentry for an NFS mount
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <linux/time.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/stat.h>
+#include <linux/errno.h>
+#include <linux/unistd.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/stats.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/lockd/bind.h>
+#include <linux/seq_file.h>
+#include <linux/mount.h>
+#include <linux/vfs.h>
+#include <linux/namei.h>
+#include <linux/security.h>
+
+#include <asm/uaccess.h>
+
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_CLIENT
+
+/*
+ * Set the superblock root dentry.
+ * Note that this function frees the inode in case of error.
+ */
+static int nfs_superblock_set_dummy_root(struct super_block *sb, struct inode *inode)
+{
+	/* The mntroot acts as the dummy root dentry for this superblock */
+	if (sb->s_root == NULL) {
+		sb->s_root = d_make_root(inode);
+		if (sb->s_root == NULL)
+			return -ENOMEM;
+		ihold(inode);
+		/*
+		 * Ensure that this dentry is invisible to d_find_alias().
+		 * Otherwise, it may be spliced into the tree by
+		 * d_materialise_unique if a parent directory from the same
+		 * filesystem gets mounted at a later time.
+		 * This again causes shrink_dcache_for_umount_subtree() to
+		 * Oops, since the test for IS_ROOT() will fail.
+		 */
+		spin_lock(&sb->s_root->d_inode->i_lock);
+		spin_lock(&sb->s_root->d_lock);
+		hlist_del_init(&sb->s_root->d_u.d_alias);
+		spin_unlock(&sb->s_root->d_lock);
+		spin_unlock(&sb->s_root->d_inode->i_lock);
+	}
+	return 0;
+}
+
+/*
+ * get an NFS2/NFS3 root dentry from the root filehandle
+ */
+struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh,
+			    const char *devname)
+{
+	struct nfs_server *server = NFS_SB(sb);
+	struct nfs_fsinfo fsinfo;
+	struct dentry *ret;
+	struct inode *inode;
+	void *name = kstrdup(devname, GFP_KERNEL);
+	int error;
+
+	if (!name)
+		return ERR_PTR(-ENOMEM);
+
+	/* get the actual root for this mount */
+	fsinfo.fattr = nfs_alloc_fattr();
+	if (fsinfo.fattr == NULL) {
+		kfree(name);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
+	if (error < 0) {
+		dprintk("nfs_get_root: getattr error = %d\n", -error);
+		ret = ERR_PTR(error);
+		goto out;
+	}
+
+	inode = nfs_fhget(sb, mntfh, fsinfo.fattr, NULL);
+	if (IS_ERR(inode)) {
+		dprintk("nfs_get_root: get root inode failed\n");
+		ret = ERR_CAST(inode);
+		goto out;
+	}
+
+	error = nfs_superblock_set_dummy_root(sb, inode);
+	if (error != 0) {
+		ret = ERR_PTR(error);
+		goto out;
+	}
+
+	/* root dentries normally start off anonymous and get spliced in later
+	 * if the dentry tree reaches them; however if the dentry already
+	 * exists, we'll pick it up at this point and use it as the root
+	 */
+	ret = d_obtain_root(inode);
+	if (IS_ERR(ret)) {
+		dprintk("nfs_get_root: get root dentry failed\n");
+		goto out;
+	}
+
+	security_d_instantiate(ret, inode);
+	spin_lock(&ret->d_lock);
+	if (IS_ROOT(ret) && !ret->d_fsdata &&
+	    !(ret->d_flags & DCACHE_NFSFS_RENAMED)) {
+		ret->d_fsdata = name;
+		name = NULL;
+	}
+	spin_unlock(&ret->d_lock);
+out:
+	kfree(name);
+	nfs_free_fattr(fsinfo.fattr);
+	return ret;
+}

fs/nfs/idmap.c

@@ -0,0 +1,791 @@
+/*
+ * fs/nfs/idmap.c
+ *
+ *  UID and GID to name mapping for clients.
+ *
+ *  Copyright (c) 2002 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Marius Aamodt Eriksen <marius@umich.edu>
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the University 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 ``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 REGENTS 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.
+ */
+#include <linux/types.h>
+#include <linux/parser.h>
+#include <linux/fs.h>
+#include <linux/nfs_idmap.h>
+#include <net/net_namespace.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_fs_sb.h>
+#include <linux/key.h>
+#include <linux/keyctl.h>
+#include <linux/key-type.h>
+#include <keys/user-type.h>
+#include <linux/module.h>
+
+#include "internal.h"
+#include "netns.h"
+#include "nfs4trace.h"
+
+#define NFS_UINT_MAXLEN 11
+
+static const struct cred *id_resolver_cache;
+static struct key_type key_type_id_resolver_legacy;
+
+struct idmap_legacy_upcalldata {
+	struct rpc_pipe_msg pipe_msg;
+	struct idmap_msg idmap_msg;
+	struct key_construction	*key_cons;
+	struct idmap *idmap;
+};
+
+struct idmap {
+	struct rpc_pipe_dir_object idmap_pdo;
+	struct rpc_pipe		*idmap_pipe;
+	struct idmap_legacy_upcalldata *idmap_upcall_data;
+	struct mutex		idmap_mutex;
+};
+
+/**
+ * nfs_fattr_init_names - initialise the nfs_fattr owner_name/group_name fields
+ * @fattr: fully initialised struct nfs_fattr
+ * @owner_name: owner name string cache
+ * @group_name: group name string cache
+ */
+void nfs_fattr_init_names(struct nfs_fattr *fattr,
+		struct nfs4_string *owner_name,
+		struct nfs4_string *group_name)
+{
+	fattr->owner_name = owner_name;
+	fattr->group_name = group_name;
+}
+
+static void nfs_fattr_free_owner_name(struct nfs_fattr *fattr)
+{
+	fattr->valid &= ~NFS_ATTR_FATTR_OWNER_NAME;
+	kfree(fattr->owner_name->data);
+}
+
+static void nfs_fattr_free_group_name(struct nfs_fattr *fattr)
+{
+	fattr->valid &= ~NFS_ATTR_FATTR_GROUP_NAME;
+	kfree(fattr->group_name->data);
+}
+
+static bool nfs_fattr_map_owner_name(struct nfs_server *server, struct nfs_fattr *fattr)
+{
+	struct nfs4_string *owner = fattr->owner_name;
+	kuid_t uid;
+
+	if (!(fattr->valid & NFS_ATTR_FATTR_OWNER_NAME))
+		return false;
+	if (nfs_map_name_to_uid(server, owner->data, owner->len, &uid) == 0) {
+		fattr->uid = uid;
+		fattr->valid |= NFS_ATTR_FATTR_OWNER;
+	}
+	return true;
+}
+
+static bool nfs_fattr_map_group_name(struct nfs_server *server, struct nfs_fattr *fattr)
+{
+	struct nfs4_string *group = fattr->group_name;
+	kgid_t gid;
+
+	if (!(fattr->valid & NFS_ATTR_FATTR_GROUP_NAME))
+		return false;
+	if (nfs_map_group_to_gid(server, group->data, group->len, &gid) == 0) {
+		fattr->gid = gid;
+		fattr->valid |= NFS_ATTR_FATTR_GROUP;
+	}
+	return true;
+}
+
+/**
+ * nfs_fattr_free_names - free up the NFSv4 owner and group strings
+ * @fattr: a fully initialised nfs_fattr structure
+ */
+void nfs_fattr_free_names(struct nfs_fattr *fattr)
+{
+	if (fattr->valid & NFS_ATTR_FATTR_OWNER_NAME)
+		nfs_fattr_free_owner_name(fattr);
+	if (fattr->valid & NFS_ATTR_FATTR_GROUP_NAME)
+		nfs_fattr_free_group_name(fattr);
+}
+
+/**
+ * nfs_fattr_map_and_free_names - map owner/group strings into uid/gid and free
+ * @server: pointer to the filesystem nfs_server structure
+ * @fattr: a fully initialised nfs_fattr structure
+ *
+ * This helper maps the cached NFSv4 owner/group strings in fattr into
+ * their numeric uid/gid equivalents, and then frees the cached strings.
+ */
+void nfs_fattr_map_and_free_names(struct nfs_server *server, struct nfs_fattr *fattr)
+{
+	if (nfs_fattr_map_owner_name(server, fattr))
+		nfs_fattr_free_owner_name(fattr);
+	if (nfs_fattr_map_group_name(server, fattr))
+		nfs_fattr_free_group_name(fattr);
+}
+
+static int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *res)
+{
+	unsigned long val;
+	char buf[16];
+
+	if (memchr(name, '@', namelen) != NULL || namelen >= sizeof(buf))
+		return 0;
+	memcpy(buf, name, namelen);
+	buf[namelen] = '\0';
+	if (kstrtoul(buf, 0, &val) != 0)
+		return 0;
+	*res = val;
+	return 1;
+}
+
+static int nfs_map_numeric_to_string(__u32 id, char *buf, size_t buflen)
+{
+	return snprintf(buf, buflen, "%u", id);
+}
+
+static struct key_type key_type_id_resolver = {
+	.name		= "id_resolver",
+	.preparse	= user_preparse,
+	.free_preparse	= user_free_preparse,
+	.instantiate	= generic_key_instantiate,
+	.revoke		= user_revoke,
+	.destroy	= user_destroy,
+	.describe	= user_describe,
+	.read		= user_read,
+};
+
+static int nfs_idmap_init_keyring(void)
+{
+	struct cred *cred;
+	struct key *keyring;
+	int ret = 0;
+
+	printk(KERN_NOTICE "NFS: Registering the %s key type\n",
+		key_type_id_resolver.name);
+
+	cred = prepare_kernel_cred(NULL);
+	if (!cred)
+		return -ENOMEM;
+
+	keyring = keyring_alloc(".id_resolver",
+				GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred,
+				(KEY_POS_ALL & ~KEY_POS_SETATTR) |
+				KEY_USR_VIEW | KEY_USR_READ,
+				KEY_ALLOC_NOT_IN_QUOTA, NULL);
+	if (IS_ERR(keyring)) {
+		ret = PTR_ERR(keyring);
+		goto failed_put_cred;
+	}
+
+	ret = register_key_type(&key_type_id_resolver);
+	if (ret < 0)
+		goto failed_put_key;
+
+	ret = register_key_type(&key_type_id_resolver_legacy);
+	if (ret < 0)
+		goto failed_reg_legacy;
+
+	set_bit(KEY_FLAG_ROOT_CAN_CLEAR, &keyring->flags);
+	cred->thread_keyring = keyring;
+	cred->jit_keyring = KEY_REQKEY_DEFL_THREAD_KEYRING;
+	id_resolver_cache = cred;
+	return 0;
+
+failed_reg_legacy:
+	unregister_key_type(&key_type_id_resolver);
+failed_put_key:
+	key_put(keyring);
+failed_put_cred:
+	put_cred(cred);
+	return ret;
+}
+
+static void nfs_idmap_quit_keyring(void)
+{
+	key_revoke(id_resolver_cache->thread_keyring);
+	unregister_key_type(&key_type_id_resolver);
+	unregister_key_type(&key_type_id_resolver_legacy);
+	put_cred(id_resolver_cache);
+}
+
+/*
+ * Assemble the description to pass to request_key()
+ * This function will allocate a new string and update dest to point
+ * at it.  The caller is responsible for freeing dest.
+ *
+ * On error 0 is returned.  Otherwise, the length of dest is returned.
+ */
+static ssize_t nfs_idmap_get_desc(const char *name, size_t namelen,
+				const char *type, size_t typelen, char **desc)
+{
+	char *cp;
+	size_t desclen = typelen + namelen + 2;
+
+	*desc = kmalloc(desclen, GFP_KERNEL);
+	if (!*desc)
+		return -ENOMEM;
+
+	cp = *desc;
+	memcpy(cp, type, typelen);
+	cp += typelen;
+	*cp++ = ':';
+
+	memcpy(cp, name, namelen);
+	cp += namelen;
+	*cp = '\0';
+	return desclen;
+}
+
+static struct key *nfs_idmap_request_key(const char *name, size_t namelen,
+					 const char *type, struct idmap *idmap)
+{
+	char *desc;
+	struct key *rkey;
+	ssize_t ret;
+
+	ret = nfs_idmap_get_desc(name, namelen, type, strlen(type), &desc);
+	if (ret <= 0)
+		return ERR_PTR(ret);
+
+	rkey = request_key(&key_type_id_resolver, desc, "");
+	if (IS_ERR(rkey)) {
+		mutex_lock(&idmap->idmap_mutex);
+		rkey = request_key_with_auxdata(&key_type_id_resolver_legacy,
+						desc, "", 0, idmap);
+		mutex_unlock(&idmap->idmap_mutex);
+	}
+	if (!IS_ERR(rkey))
+		set_bit(KEY_FLAG_ROOT_CAN_INVAL, &rkey->flags);
+
+	kfree(desc);
+	return rkey;
+}
+
+static ssize_t nfs_idmap_get_key(const char *name, size_t namelen,
+				 const char *type, void *data,
+				 size_t data_size, struct idmap *idmap)
+{
+	const struct cred *saved_cred;
+	struct key *rkey;
+	struct user_key_payload *payload;
+	ssize_t ret;
+
+	saved_cred = override_creds(id_resolver_cache);
+	rkey = nfs_idmap_request_key(name, namelen, type, idmap);
+	revert_creds(saved_cred);
+
+	if (IS_ERR(rkey)) {
+		ret = PTR_ERR(rkey);
+		goto out;
+	}
+
+	rcu_read_lock();
+	rkey->perm |= KEY_USR_VIEW;
+
+	ret = key_validate(rkey);
+	if (ret < 0)
+		goto out_up;
+
+	payload = rcu_dereference(rkey->payload.rcudata);
+	if (IS_ERR_OR_NULL(payload)) {
+		ret = PTR_ERR(payload);
+		goto out_up;
+	}
+
+	ret = payload->datalen;
+	if (ret > 0 && ret <= data_size)
+		memcpy(data, payload->data, ret);
+	else
+		ret = -EINVAL;
+
+out_up:
+	rcu_read_unlock();
+	key_put(rkey);
+out:
+	return ret;
+}
+
+/* ID -> Name */
+static ssize_t nfs_idmap_lookup_name(__u32 id, const char *type, char *buf,
+				     size_t buflen, struct idmap *idmap)
+{
+	char id_str[NFS_UINT_MAXLEN];
+	int id_len;
+	ssize_t ret;
+
+	id_len = snprintf(id_str, sizeof(id_str), "%u", id);
+	ret = nfs_idmap_get_key(id_str, id_len, type, buf, buflen, idmap);
+	if (ret < 0)
+		return -EINVAL;
+	return ret;
+}
+
+/* Name -> ID */
+static int nfs_idmap_lookup_id(const char *name, size_t namelen, const char *type,
+			       __u32 *id, struct idmap *idmap)
+{
+	char id_str[NFS_UINT_MAXLEN];
+	long id_long;
+	ssize_t data_size;
+	int ret = 0;
+
+	data_size = nfs_idmap_get_key(name, namelen, type, id_str, NFS_UINT_MAXLEN, idmap);
+	if (data_size <= 0) {
+		ret = -EINVAL;
+	} else {
+		ret = kstrtol(id_str, 10, &id_long);
+		*id = (__u32)id_long;
+	}
+	return ret;
+}
+
+/* idmap classic begins here */
+
+enum {
+	Opt_find_uid, Opt_find_gid, Opt_find_user, Opt_find_group, Opt_find_err
+};
+
+static const match_table_t nfs_idmap_tokens = {
+	{ Opt_find_uid, "uid:%s" },
+	{ Opt_find_gid, "gid:%s" },
+	{ Opt_find_user, "user:%s" },
+	{ Opt_find_group, "group:%s" },
+	{ Opt_find_err, NULL }
+};
+
+static int nfs_idmap_legacy_upcall(struct key_construction *, const char *, void *);
+static ssize_t idmap_pipe_downcall(struct file *, const char __user *,
+				   size_t);
+static void idmap_release_pipe(struct inode *);
+static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *);
+
+static const struct rpc_pipe_ops idmap_upcall_ops = {
+	.upcall		= rpc_pipe_generic_upcall,
+	.downcall	= idmap_pipe_downcall,
+	.release_pipe	= idmap_release_pipe,
+	.destroy_msg	= idmap_pipe_destroy_msg,
+};
+
+static struct key_type key_type_id_resolver_legacy = {
+	.name		= "id_legacy",
+	.preparse	= user_preparse,
+	.free_preparse	= user_free_preparse,
+	.instantiate	= generic_key_instantiate,
+	.revoke		= user_revoke,
+	.destroy	= user_destroy,
+	.describe	= user_describe,
+	.read		= user_read,
+	.request_key	= nfs_idmap_legacy_upcall,
+};
+
+static void nfs_idmap_pipe_destroy(struct dentry *dir,
+		struct rpc_pipe_dir_object *pdo)
+{
+	struct idmap *idmap = pdo->pdo_data;
+	struct rpc_pipe *pipe = idmap->idmap_pipe;
+
+	if (pipe->dentry) {
+		rpc_unlink(pipe->dentry);
+		pipe->dentry = NULL;
+	}
+}
+
+static int nfs_idmap_pipe_create(struct dentry *dir,
+		struct rpc_pipe_dir_object *pdo)
+{
+	struct idmap *idmap = pdo->pdo_data;
+	struct rpc_pipe *pipe = idmap->idmap_pipe;
+	struct dentry *dentry;
+
+	dentry = rpc_mkpipe_dentry(dir, "idmap", idmap, pipe);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
+	pipe->dentry = dentry;
+	return 0;
+}
+
+static const struct rpc_pipe_dir_object_ops nfs_idmap_pipe_dir_object_ops = {
+	.create = nfs_idmap_pipe_create,
+	.destroy = nfs_idmap_pipe_destroy,
+};
+
+int
+nfs_idmap_new(struct nfs_client *clp)
+{
+	struct idmap *idmap;
+	struct rpc_pipe *pipe;
+	int error;
+
+	idmap = kzalloc(sizeof(*idmap), GFP_KERNEL);
+	if (idmap == NULL)
+		return -ENOMEM;
+
+	rpc_init_pipe_dir_object(&idmap->idmap_pdo,
+			&nfs_idmap_pipe_dir_object_ops,
+			idmap);
+
+	pipe = rpc_mkpipe_data(&idmap_upcall_ops, 0);
+	if (IS_ERR(pipe)) {
+		error = PTR_ERR(pipe);
+		goto err;
+	}
+	idmap->idmap_pipe = pipe;
+	mutex_init(&idmap->idmap_mutex);
+
+	error = rpc_add_pipe_dir_object(clp->cl_net,
+			&clp->cl_rpcclient->cl_pipedir_objects,
+			&idmap->idmap_pdo);
+	if (error)
+		goto err_destroy_pipe;
+
+	clp->cl_idmap = idmap;
+	return 0;
+err_destroy_pipe:
+	rpc_destroy_pipe_data(idmap->idmap_pipe);
+err:
+	kfree(idmap);
+	return error;
+}
+
+void
+nfs_idmap_delete(struct nfs_client *clp)
+{
+	struct idmap *idmap = clp->cl_idmap;
+
+	if (!idmap)
+		return;
+	clp->cl_idmap = NULL;
+	rpc_remove_pipe_dir_object(clp->cl_net,
+			&clp->cl_rpcclient->cl_pipedir_objects,
+			&idmap->idmap_pdo);
+	rpc_destroy_pipe_data(idmap->idmap_pipe);
+	kfree(idmap);
+}
+
+int nfs_idmap_init(void)
+{
+	int ret;
+	ret = nfs_idmap_init_keyring();
+	if (ret != 0)
+		goto out;
+out:
+	return ret;
+}
+
+void nfs_idmap_quit(void)
+{
+	nfs_idmap_quit_keyring();
+}
+
+static int nfs_idmap_prepare_message(char *desc, struct idmap *idmap,
+				     struct idmap_msg *im,
+				     struct rpc_pipe_msg *msg)
+{
+	substring_t substr;
+	int token, ret;
+
+	im->im_type = IDMAP_TYPE_GROUP;
+	token = match_token(desc, nfs_idmap_tokens, &substr);
+
+	switch (token) {
+	case Opt_find_uid:
+		im->im_type = IDMAP_TYPE_USER;
+	case Opt_find_gid:
+		im->im_conv = IDMAP_CONV_NAMETOID;
+		ret = match_strlcpy(im->im_name, &substr, IDMAP_NAMESZ);
+		break;
+
+	case Opt_find_user:
+		im->im_type = IDMAP_TYPE_USER;
+	case Opt_find_group:
+		im->im_conv = IDMAP_CONV_IDTONAME;
+		ret = match_int(&substr, &im->im_id);
+		break;
+
+	default:
+		ret = -EINVAL;
+		goto out;
+	}
+
+	msg->data = im;
+	msg->len  = sizeof(struct idmap_msg);
+
+out:
+	return ret;
+}
+
+static bool
+nfs_idmap_prepare_pipe_upcall(struct idmap *idmap,
+		struct idmap_legacy_upcalldata *data)
+{
+	if (idmap->idmap_upcall_data != NULL) {
+		WARN_ON_ONCE(1);
+		return false;
+	}
+	idmap->idmap_upcall_data = data;
+	return true;
+}
+
+static void
+nfs_idmap_complete_pipe_upcall_locked(struct idmap *idmap, int ret)
+{
+	struct key_construction *cons = idmap->idmap_upcall_data->key_cons;
+
+	kfree(idmap->idmap_upcall_data);
+	idmap->idmap_upcall_data = NULL;
+	complete_request_key(cons, ret);
+}
+
+static void
+nfs_idmap_abort_pipe_upcall(struct idmap *idmap, int ret)
+{
+	if (idmap->idmap_upcall_data != NULL)
+		nfs_idmap_complete_pipe_upcall_locked(idmap, ret);
+}
+
+static int nfs_idmap_legacy_upcall(struct key_construction *cons,
+				   const char *op,
+				   void *aux)
+{
+	struct idmap_legacy_upcalldata *data;
+	struct rpc_pipe_msg *msg;
+	struct idmap_msg *im;
+	struct idmap *idmap = (struct idmap *)aux;
+	struct key *key = cons->key;
+	int ret = -ENOMEM;
+
+	/* msg and im are freed in idmap_pipe_destroy_msg */
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		goto out1;
+
+	msg = &data->pipe_msg;
+	im = &data->idmap_msg;
+	data->idmap = idmap;
+	data->key_cons = cons;
+
+	ret = nfs_idmap_prepare_message(key->description, idmap, im, msg);
+	if (ret < 0)
+		goto out2;
+
+	ret = -EAGAIN;
+	if (!nfs_idmap_prepare_pipe_upcall(idmap, data))
+		goto out2;
+
+	ret = rpc_queue_upcall(idmap->idmap_pipe, msg);
+	if (ret < 0)
+		nfs_idmap_abort_pipe_upcall(idmap, ret);
+
+	return ret;
+out2:
+	kfree(data);
+out1:
+	complete_request_key(cons, ret);
+	return ret;
+}
+
+static int nfs_idmap_instantiate(struct key *key, struct key *authkey, char *data, size_t datalen)
+{
+	return key_instantiate_and_link(key, data, datalen,
+					id_resolver_cache->thread_keyring,
+					authkey);
+}
+
+static int nfs_idmap_read_and_verify_message(struct idmap_msg *im,
+		struct idmap_msg *upcall,
+		struct key *key, struct key *authkey)
+{
+	char id_str[NFS_UINT_MAXLEN];
+	size_t len;
+	int ret = -ENOKEY;
+
+	/* ret = -ENOKEY */
+	if (upcall->im_type != im->im_type || upcall->im_conv != im->im_conv)
+		goto out;
+	switch (im->im_conv) {
+	case IDMAP_CONV_NAMETOID:
+		if (strcmp(upcall->im_name, im->im_name) != 0)
+			break;
+		/* Note: here we store the NUL terminator too */
+		len = sprintf(id_str, "%d", im->im_id) + 1;
+		ret = nfs_idmap_instantiate(key, authkey, id_str, len);
+		break;
+	case IDMAP_CONV_IDTONAME:
+		if (upcall->im_id != im->im_id)
+			break;
+		len = strlen(im->im_name);
+		ret = nfs_idmap_instantiate(key, authkey, im->im_name, len);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+out:
+	return ret;
+}
+
+static ssize_t
+idmap_pipe_downcall(struct file *filp, const char __user *src, size_t mlen)
+{
+	struct rpc_inode *rpci = RPC_I(file_inode(filp));
+	struct idmap *idmap = (struct idmap *)rpci->private;
+	struct key_construction *cons;
+	struct idmap_msg im;
+	size_t namelen_in;
+	int ret = -ENOKEY;
+
+	/* If instantiation is successful, anyone waiting for key construction
+	 * will have been woken up and someone else may now have used
+	 * idmap_key_cons - so after this point we may no longer touch it.
+	 */
+	if (idmap->idmap_upcall_data == NULL)
+		goto out_noupcall;
+
+	cons = idmap->idmap_upcall_data->key_cons;
+
+	if (mlen != sizeof(im)) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	if (copy_from_user(&im, src, mlen) != 0) {
+		ret = -EFAULT;
+		goto out;
+	}
+
+	if (!(im.im_status & IDMAP_STATUS_SUCCESS)) {
+		ret = -ENOKEY;
+		goto out;
+	}
+
+	namelen_in = strnlen(im.im_name, IDMAP_NAMESZ);
+	if (namelen_in == 0 || namelen_in == IDMAP_NAMESZ) {
+		ret = -EINVAL;
+		goto out;
+}
+
+	ret = nfs_idmap_read_and_verify_message(&im,
+			&idmap->idmap_upcall_data->idmap_msg,
+			cons->key, cons->authkey);
+	if (ret >= 0) {
+		key_set_timeout(cons->key, nfs_idmap_cache_timeout);
+		ret = mlen;
+	}
+
+out:
+	nfs_idmap_complete_pipe_upcall_locked(idmap, ret);
+out_noupcall:
+	return ret;
+}
+
+static void
+idmap_pipe_destroy_msg(struct rpc_pipe_msg *msg)
+{
+	struct idmap_legacy_upcalldata *data = container_of(msg,
+			struct idmap_legacy_upcalldata,
+			pipe_msg);
+	struct idmap *idmap = data->idmap;
+
+	if (msg->errno)
+		nfs_idmap_abort_pipe_upcall(idmap, msg->errno);
+}
+
+static void
+idmap_release_pipe(struct inode *inode)
+{
+	struct rpc_inode *rpci = RPC_I(inode);
+	struct idmap *idmap = (struct idmap *)rpci->private;
+
+	nfs_idmap_abort_pipe_upcall(idmap, -EPIPE);
+}
+
+int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_t namelen, kuid_t *uid)
+{
+	struct idmap *idmap = server->nfs_client->cl_idmap;
+	__u32 id = -1;
+	int ret = 0;
+
+	if (!nfs_map_string_to_numeric(name, namelen, &id))
+		ret = nfs_idmap_lookup_id(name, namelen, "uid", &id, idmap);
+	if (ret == 0) {
+		*uid = make_kuid(&init_user_ns, id);
+		if (!uid_valid(*uid))
+			ret = -ERANGE;
+	}
+	trace_nfs4_map_name_to_uid(name, namelen, id, ret);
+	return ret;
+}
+
+int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size_t namelen, kgid_t *gid)
+{
+	struct idmap *idmap = server->nfs_client->cl_idmap;
+	__u32 id = -1;
+	int ret = 0;
+
+	if (!nfs_map_string_to_numeric(name, namelen, &id))
+		ret = nfs_idmap_lookup_id(name, namelen, "gid", &id, idmap);
+	if (ret == 0) {
+		*gid = make_kgid(&init_user_ns, id);
+		if (!gid_valid(*gid))
+			ret = -ERANGE;
+	}
+	trace_nfs4_map_group_to_gid(name, namelen, id, ret);
+	return ret;
+}
+
+int nfs_map_uid_to_name(const struct nfs_server *server, kuid_t uid, char *buf, size_t buflen)
+{
+	struct idmap *idmap = server->nfs_client->cl_idmap;
+	int ret = -EINVAL;
+	__u32 id;
+
+	id = from_kuid(&init_user_ns, uid);
+	if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+		ret = nfs_idmap_lookup_name(id, "user", buf, buflen, idmap);
+	if (ret < 0)
+		ret = nfs_map_numeric_to_string(id, buf, buflen);
+	trace_nfs4_map_uid_to_name(buf, ret, id, ret);
+	return ret;
+}
+int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, size_t buflen)
+{
+	struct idmap *idmap = server->nfs_client->cl_idmap;
+	int ret = -EINVAL;
+	__u32 id;
+
+	id = from_kgid(&init_user_ns, gid);
+	if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+		ret = nfs_idmap_lookup_name(id, "group", buf, buflen, idmap);
+	if (ret < 0)
+		ret = nfs_map_numeric_to_string(id, buf, buflen);
+	trace_nfs4_map_gid_to_group(buf, ret, id, ret);
+	return ret;
+}

fs/nfs/internal.h

@@ -0,0 +1,651 @@
+/*
+ * NFS internal definitions
+ */
+
+#include "nfs4_fs.h"
+#include <linux/mount.h>
+#include <linux/security.h>
+#include <linux/crc32.h>
+
+#define NFS_MS_MASK (MS_RDONLY|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_SYNCHRONOUS)
+
+struct nfs_string;
+
+/* Maximum number of readahead requests
+ * FIXME: this should really be a sysctl so that users may tune it to suit
+ *        their needs. People that do NFS over a slow network, might for
+ *        instance want to reduce it to something closer to 1 for improved
+ *        interactive response.
+ */
+#define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)
+
+static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct nfs_fattr *fattr)
+{
+	if (!nfs_fsid_equal(&NFS_SB(parent)->fsid, &fattr->fsid))
+		fattr->valid |= NFS_ATTR_FATTR_MOUNTPOINT;
+}
+
+static inline int nfs_attr_use_mounted_on_fileid(struct nfs_fattr *fattr)
+{
+	if (((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) == 0) ||
+	    (((fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) &&
+	     ((fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) == 0)))
+		return 0;
+
+	fattr->fileid = fattr->mounted_on_fileid;
+	return 1;
+}
+
+struct nfs_clone_mount {
+	const struct super_block *sb;
+	const struct dentry *dentry;
+	struct nfs_fh *fh;
+	struct nfs_fattr *fattr;
+	char *hostname;
+	char *mnt_path;
+	struct sockaddr *addr;
+	size_t addrlen;
+	rpc_authflavor_t authflavor;
+};
+
+/*
+ * Note: RFC 1813 doesn't limit the number of auth flavors that
+ * a server can return, so make something up.
+ */
+#define NFS_MAX_SECFLAVORS	(12)
+
+/*
+ * Value used if the user did not specify a port value.
+ */
+#define NFS_UNSPEC_PORT		(-1)
+
+/*
+ * Maximum number of pages that readdir can use for creating
+ * a vmapped array of pages.
+ */
+#define NFS_MAX_READDIR_PAGES 8
+
+struct nfs_client_initdata {
+	unsigned long init_flags;
+	const char *hostname;
+	const struct sockaddr *addr;
+	size_t addrlen;
+	struct nfs_subversion *nfs_mod;
+	int proto;
+	u32 minorversion;
+	struct net *net;
+};
+
+/*
+ * In-kernel mount arguments
+ */
+struct nfs_parsed_mount_data {
+	int			flags;
+	unsigned int		rsize, wsize;
+	unsigned int		timeo, retrans;
+	unsigned int		acregmin, acregmax,
+				acdirmin, acdirmax;
+	unsigned int		namlen;
+	unsigned int		options;
+	unsigned int		bsize;
+	struct nfs_auth_info	auth_info;
+	rpc_authflavor_t	selected_flavor;
+	char			*client_address;
+	unsigned int		version;
+	unsigned int		minorversion;
+	char			*fscache_uniq;
+	bool			need_mount;
+
+	struct {
+		struct sockaddr_storage	address;
+		size_t			addrlen;
+		char			*hostname;
+		u32			version;
+		int			port;
+		unsigned short		protocol;
+	} mount_server;
+
+	struct {
+		struct sockaddr_storage	address;
+		size_t			addrlen;
+		char			*hostname;
+		char			*export_path;
+		int			port;
+		unsigned short		protocol;
+	} nfs_server;
+
+	struct security_mnt_opts lsm_opts;
+	struct net		*net;
+};
+
+/* mount_clnt.c */
+struct nfs_mount_request {
+	struct sockaddr		*sap;
+	size_t			salen;
+	char			*hostname;
+	char			*dirpath;
+	u32			version;
+	unsigned short		protocol;
+	struct nfs_fh		*fh;
+	int			noresvport;
+	unsigned int		*auth_flav_len;
+	rpc_authflavor_t	*auth_flavs;
+	struct net		*net;
+};
+
+struct nfs_mount_info {
+	void (*fill_super)(struct super_block *, struct nfs_mount_info *);
+	int (*set_security)(struct super_block *, struct dentry *, struct nfs_mount_info *);
+	struct nfs_parsed_mount_data *parsed;
+	struct nfs_clone_mount *cloned;
+	struct nfs_fh *mntfh;
+};
+
+extern int nfs_mount(struct nfs_mount_request *info);
+extern void nfs_umount(const struct nfs_mount_request *info);
+
+/* client.c */
+extern const struct rpc_program nfs_program;
+extern void nfs_clients_init(struct net *net);
+extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *);
+int nfs_create_rpc_client(struct nfs_client *, const struct rpc_timeout *, rpc_authflavor_t);
+struct nfs_client *nfs_get_client(const struct nfs_client_initdata *,
+				  const struct rpc_timeout *, const char *,
+				  rpc_authflavor_t);
+int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *, struct nfs_fattr *);
+void nfs_server_insert_lists(struct nfs_server *);
+void nfs_server_remove_lists(struct nfs_server *);
+void nfs_init_timeout_values(struct rpc_timeout *, int, unsigned int, unsigned int);
+int nfs_init_server_rpcclient(struct nfs_server *, const struct rpc_timeout *t,
+		rpc_authflavor_t);
+struct nfs_server *nfs_alloc_server(void);
+void nfs_server_copy_userdata(struct nfs_server *, struct nfs_server *);
+
+extern void nfs_cleanup_cb_ident_idr(struct net *);
+extern void nfs_put_client(struct nfs_client *);
+extern void nfs_free_client(struct nfs_client *);
+extern struct nfs_client *nfs4_find_client_ident(struct net *, int);
+extern struct nfs_client *
+nfs4_find_client_sessionid(struct net *, const struct sockaddr *,
+				struct nfs4_sessionid *, u32);
+extern struct nfs_server *nfs_create_server(struct nfs_mount_info *,
+					struct nfs_subversion *);
+extern struct nfs_server *nfs4_create_server(
+					struct nfs_mount_info *,
+					struct nfs_subversion *);
+extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *,
+						      struct nfs_fh *);
+extern int nfs4_update_server(struct nfs_server *server, const char *hostname,
+					struct sockaddr *sap, size_t salen,
+					struct net *net);
+extern void nfs_free_server(struct nfs_server *server);
+extern struct nfs_server *nfs_clone_server(struct nfs_server *,
+					   struct nfs_fh *,
+					   struct nfs_fattr *,
+					   rpc_authflavor_t);
+extern int nfs_wait_client_init_complete(const struct nfs_client *clp);
+extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
+extern struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp,
+					     const struct sockaddr *ds_addr,
+					     int ds_addrlen, int ds_proto,
+					     unsigned int ds_timeo,
+					     unsigned int ds_retrans);
+extern struct rpc_clnt *nfs4_find_or_create_ds_client(struct nfs_client *,
+						struct inode *);
+#ifdef CONFIG_PROC_FS
+extern int __init nfs_fs_proc_init(void);
+extern void nfs_fs_proc_exit(void);
+extern int nfs_fs_proc_net_init(struct net *net);
+extern void nfs_fs_proc_net_exit(struct net *net);
+#else
+static inline int nfs_fs_proc_net_init(struct net *net)
+{
+	return 0;
+}
+static inline void nfs_fs_proc_net_exit(struct net *net)
+{
+}
+static inline int nfs_fs_proc_init(void)
+{
+	return 0;
+}
+static inline void nfs_fs_proc_exit(void)
+{
+}
+#endif
+
+#ifdef CONFIG_NFS_V4_1
+int nfs_sockaddr_match_ipaddr(const struct sockaddr *, const struct sockaddr *);
+#endif
+
+/* callback_xdr.c */
+extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
+
+struct nfs_pageio_descriptor;
+/* pagelist.c */
+extern int __init nfs_init_nfspagecache(void);
+extern void nfs_destroy_nfspagecache(void);
+extern int __init nfs_init_readpagecache(void);
+extern void nfs_destroy_readpagecache(void);
+extern int __init nfs_init_writepagecache(void);
+extern void nfs_destroy_writepagecache(void);
+
+extern int __init nfs_init_directcache(void);
+extern void nfs_destroy_directcache(void);
+extern void nfs_pgheader_init(struct nfs_pageio_descriptor *desc,
+			      struct nfs_pgio_header *hdr,
+			      void (*release)(struct nfs_pgio_header *hdr));
+void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos);
+int nfs_iocounter_wait(struct nfs_io_counter *c);
+
+extern const struct nfs_pageio_ops nfs_pgio_rw_ops;
+struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *);
+void nfs_pgio_header_free(struct nfs_pgio_header *);
+void nfs_pgio_data_destroy(struct nfs_pgio_header *);
+int nfs_generic_pgio(struct nfs_pageio_descriptor *, struct nfs_pgio_header *);
+int nfs_initiate_pgio(struct rpc_clnt *, struct nfs_pgio_header *,
+		      const struct rpc_call_ops *, int, int);
+void nfs_free_request(struct nfs_page *req);
+
+static inline void nfs_iocounter_init(struct nfs_io_counter *c)
+{
+	c->flags = 0;
+	atomic_set(&c->io_count, 0);
+}
+
+/* nfs2xdr.c */
+extern struct rpc_procinfo nfs_procedures[];
+extern int nfs2_decode_dirent(struct xdr_stream *,
+				struct nfs_entry *, int);
+
+/* nfs3xdr.c */
+extern struct rpc_procinfo nfs3_procedures[];
+extern int nfs3_decode_dirent(struct xdr_stream *,
+				struct nfs_entry *, int);
+
+/* nfs4xdr.c */
+#if IS_ENABLED(CONFIG_NFS_V4)
+extern int nfs4_decode_dirent(struct xdr_stream *,
+				struct nfs_entry *, int);
+#endif
+#ifdef CONFIG_NFS_V4_1
+extern const u32 nfs41_maxread_overhead;
+extern const u32 nfs41_maxwrite_overhead;
+extern const u32 nfs41_maxgetdevinfo_overhead;
+#endif
+
+/* nfs4proc.c */
+#if IS_ENABLED(CONFIG_NFS_V4)
+extern struct rpc_procinfo nfs4_procedures[];
+#endif
+
+#ifdef CONFIG_NFS_V4_SECURITY_LABEL
+extern struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags);
+static inline void nfs4_label_free(struct nfs4_label *label)
+{
+	if (label) {
+		kfree(label->label);
+		kfree(label);
+	}
+	return;
+}
+
+static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi)
+{
+	if (nfs_server_capable(&nfsi->vfs_inode, NFS_CAP_SECURITY_LABEL))
+		nfsi->cache_validity |= NFS_INO_INVALID_LABEL;
+}
+#else
+static inline struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { return NULL; }
+static inline void nfs4_label_free(void *label) {}
+static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi)
+{
+}
+#endif /* CONFIG_NFS_V4_SECURITY_LABEL */
+
+/* proc.c */
+void nfs_close_context(struct nfs_open_context *ctx, int is_sync);
+extern struct nfs_client *nfs_init_client(struct nfs_client *clp,
+			   const struct rpc_timeout *timeparms,
+			   const char *ip_addr);
+
+/* dir.c */
+extern void nfs_force_use_readdirplus(struct inode *dir);
+extern unsigned long nfs_access_cache_count(struct shrinker *shrink,
+					    struct shrink_control *sc);
+extern unsigned long nfs_access_cache_scan(struct shrinker *shrink,
+					   struct shrink_control *sc);
+struct dentry *nfs_lookup(struct inode *, struct dentry *, unsigned int);
+int nfs_create(struct inode *, struct dentry *, umode_t, bool);
+int nfs_mkdir(struct inode *, struct dentry *, umode_t);
+int nfs_rmdir(struct inode *, struct dentry *);
+int nfs_unlink(struct inode *, struct dentry *);
+int nfs_symlink(struct inode *, struct dentry *, const char *);
+int nfs_link(struct dentry *, struct inode *, struct dentry *);
+int nfs_mknod(struct inode *, struct dentry *, umode_t, dev_t);
+int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *);
+
+/* file.c */
+int nfs_file_fsync_commit(struct file *, loff_t, loff_t, int);
+loff_t nfs_file_llseek(struct file *, loff_t, int);
+int nfs_file_flush(struct file *, fl_owner_t);
+ssize_t nfs_file_read(struct kiocb *, struct iov_iter *);
+ssize_t nfs_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *,
+			     size_t, unsigned int);
+int nfs_file_mmap(struct file *, struct vm_area_struct *);
+ssize_t nfs_file_write(struct kiocb *, struct iov_iter *);
+int nfs_file_release(struct inode *, struct file *);
+int nfs_lock(struct file *, int, struct file_lock *);
+int nfs_flock(struct file *, int, struct file_lock *);
+int nfs_check_flags(int);
+
+/* inode.c */
+extern struct workqueue_struct *nfsiod_workqueue;
+extern struct inode *nfs_alloc_inode(struct super_block *sb);
+extern void nfs_destroy_inode(struct inode *);
+extern int nfs_write_inode(struct inode *, struct writeback_control *);
+extern int nfs_drop_inode(struct inode *);
+extern void nfs_clear_inode(struct inode *);
+extern void nfs_evict_inode(struct inode *);
+void nfs_zap_acl_cache(struct inode *inode);
+extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode);
+
+/* super.c */
+extern const struct super_operations nfs_sops;
+extern struct file_system_type nfs_fs_type;
+extern struct file_system_type nfs_xdev_fs_type;
+#if IS_ENABLED(CONFIG_NFS_V4)
+extern struct file_system_type nfs4_xdev_fs_type;
+extern struct file_system_type nfs4_referral_fs_type;
+#endif
+bool nfs_auth_info_match(const struct nfs_auth_info *, rpc_authflavor_t);
+struct dentry *nfs_try_mount(int, const char *, struct nfs_mount_info *,
+			struct nfs_subversion *);
+void nfs_initialise_sb(struct super_block *);
+int nfs_set_sb_security(struct super_block *, struct dentry *, struct nfs_mount_info *);
+int nfs_clone_sb_security(struct super_block *, struct dentry *, struct nfs_mount_info *);
+struct dentry *nfs_fs_mount_common(struct nfs_server *, int, const char *,
+				   struct nfs_mount_info *, struct nfs_subversion *);
+struct dentry *nfs_fs_mount(struct file_system_type *, int, const char *, void *);
+struct dentry * nfs_xdev_mount_common(struct file_system_type *, int,
+		const char *, struct nfs_mount_info *);
+void nfs_kill_super(struct super_block *);
+void nfs_fill_super(struct super_block *, struct nfs_mount_info *);
+
+extern struct rpc_stat nfs_rpcstat;
+
+extern int __init register_nfs_fs(void);
+extern void __exit unregister_nfs_fs(void);
+extern bool nfs_sb_active(struct super_block *sb);
+extern void nfs_sb_deactive(struct super_block *sb);
+
+/* namespace.c */
+#define NFS_PATH_CANONICAL 1
+extern char *nfs_path(char **p, struct dentry *dentry,
+		      char *buffer, ssize_t buflen, unsigned flags);
+extern struct vfsmount *nfs_d_automount(struct path *path);
+struct vfsmount *nfs_submount(struct nfs_server *, struct dentry *,
+			      struct nfs_fh *, struct nfs_fattr *);
+struct vfsmount *nfs_do_submount(struct dentry *, struct nfs_fh *,
+				 struct nfs_fattr *, rpc_authflavor_t);
+
+/* getroot.c */
+extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *,
+				   const char *);
+#if IS_ENABLED(CONFIG_NFS_V4)
+extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *,
+				    const char *);
+
+extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool);
+#endif
+
+struct nfs_pgio_completion_ops;
+/* read.c */
+extern void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio,
+			struct inode *inode, bool force_mds,
+			const struct nfs_pgio_completion_ops *compl_ops);
+extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
+extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio);
+
+/* super.c */
+void nfs_clone_super(struct super_block *, struct nfs_mount_info *);
+void nfs_umount_begin(struct super_block *);
+int  nfs_statfs(struct dentry *, struct kstatfs *);
+int  nfs_show_options(struct seq_file *, struct dentry *);
+int  nfs_show_devname(struct seq_file *, struct dentry *);
+int  nfs_show_path(struct seq_file *, struct dentry *);
+int  nfs_show_stats(struct seq_file *, struct dentry *);
+void nfs_put_super(struct super_block *);
+int nfs_remount(struct super_block *sb, int *flags, char *raw_data);
+
+/* write.c */
+extern void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
+			struct inode *inode, int ioflags, bool force_mds,
+			const struct nfs_pgio_completion_ops *compl_ops);
+extern void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio);
+extern void nfs_commit_free(struct nfs_commit_data *p);
+extern void nfs_write_prepare(struct rpc_task *task, void *calldata);
+extern void nfs_commit_prepare(struct rpc_task *task, void *calldata);
+extern int nfs_initiate_commit(struct rpc_clnt *clnt,
+			       struct nfs_commit_data *data,
+			       const struct rpc_call_ops *call_ops,
+			       int how, int flags);
+extern void nfs_init_commit(struct nfs_commit_data *data,
+			    struct list_head *head,
+			    struct pnfs_layout_segment *lseg,
+			    struct nfs_commit_info *cinfo);
+int nfs_scan_commit_list(struct list_head *src, struct list_head *dst,
+			 struct nfs_commit_info *cinfo, int max);
+unsigned long nfs_reqs_to_commit(struct nfs_commit_info *);
+int nfs_scan_commit(struct inode *inode, struct list_head *dst,
+		    struct nfs_commit_info *cinfo);
+void nfs_mark_request_commit(struct nfs_page *req,
+			     struct pnfs_layout_segment *lseg,
+			     struct nfs_commit_info *cinfo);
+int nfs_write_need_commit(struct nfs_pgio_header *);
+int nfs_generic_commit_list(struct inode *inode, struct list_head *head,
+			    int how, struct nfs_commit_info *cinfo);
+void nfs_retry_commit(struct list_head *page_list,
+		      struct pnfs_layout_segment *lseg,
+		      struct nfs_commit_info *cinfo);
+void nfs_commitdata_release(struct nfs_commit_data *data);
+void nfs_request_add_commit_list(struct nfs_page *req, struct list_head *dst,
+				 struct nfs_commit_info *cinfo);
+void nfs_request_remove_commit_list(struct nfs_page *req,
+				    struct nfs_commit_info *cinfo);
+void nfs_init_cinfo(struct nfs_commit_info *cinfo,
+		    struct inode *inode,
+		    struct nfs_direct_req *dreq);
+int nfs_key_timeout_notify(struct file *filp, struct inode *inode);
+bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx);
+
+#ifdef CONFIG_MIGRATION
+extern int nfs_migrate_page(struct address_space *,
+		struct page *, struct page *, enum migrate_mode);
+#else
+#define nfs_migrate_page NULL
+#endif
+
+/* unlink.c */
+extern struct rpc_task *
+nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
+		 struct dentry *old_dentry, struct dentry *new_dentry,
+		 void (*complete)(struct rpc_task *, struct nfs_renamedata *));
+extern int nfs_sillyrename(struct inode *dir, struct dentry *dentry);
+
+/* direct.c */
+void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
+			      struct nfs_direct_req *dreq);
+static inline void nfs_inode_dio_wait(struct inode *inode)
+{
+	inode_dio_wait(inode);
+}
+extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq);
+
+/* nfs4proc.c */
+extern void __nfs4_read_done_cb(struct nfs_pgio_header *);
+extern struct nfs_client *nfs4_init_client(struct nfs_client *clp,
+			    const struct rpc_timeout *timeparms,
+			    const char *ip_addr);
+extern int nfs40_walk_client_list(struct nfs_client *clp,
+				struct nfs_client **result,
+				struct rpc_cred *cred);
+extern int nfs41_walk_client_list(struct nfs_client *clp,
+				struct nfs_client **result,
+				struct rpc_cred *cred);
+
+static inline struct inode *nfs_igrab_and_active(struct inode *inode)
+{
+	inode = igrab(inode);
+	if (inode != NULL && !nfs_sb_active(inode->i_sb)) {
+		iput(inode);
+		inode = NULL;
+	}
+	return inode;
+}
+
+static inline void nfs_iput_and_deactive(struct inode *inode)
+{
+	if (inode != NULL) {
+		struct super_block *sb = inode->i_sb;
+
+		iput(inode);
+		nfs_sb_deactive(sb);
+	}
+}
+
+/*
+ * Determine the device name as a string
+ */
+static inline char *nfs_devname(struct dentry *dentry,
+				char *buffer, ssize_t buflen)
+{
+	char *dummy;
+	return nfs_path(&dummy, dentry, buffer, buflen, NFS_PATH_CANONICAL);
+}
+
+/*
+ * Determine the actual block size (and log2 thereof)
+ */
+static inline
+unsigned long nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp)
+{
+	/* make sure blocksize is a power of two */
+	if ((bsize & (bsize - 1)) || nrbitsp) {
+		unsigned char	nrbits;
+
+		for (nrbits = 31; nrbits && !(bsize & (1 << nrbits)); nrbits--)
+			;
+		bsize = 1 << nrbits;
+		if (nrbitsp)
+			*nrbitsp = nrbits;
+	}
+
+	return bsize;
+}
+
+/*
+ * Calculate the number of 512byte blocks used.
+ */
+static inline blkcnt_t nfs_calc_block_size(u64 tsize)
+{
+	blkcnt_t used = (tsize + 511) >> 9;
+	return (used > ULONG_MAX) ? ULONG_MAX : used;
+}
+
+/*
+ * Compute and set NFS server blocksize
+ */
+static inline
+unsigned long nfs_block_size(unsigned long bsize, unsigned char *nrbitsp)
+{
+	if (bsize < NFS_MIN_FILE_IO_SIZE)
+		bsize = NFS_DEF_FILE_IO_SIZE;
+	else if (bsize >= NFS_MAX_FILE_IO_SIZE)
+		bsize = NFS_MAX_FILE_IO_SIZE;
+
+	return nfs_block_bits(bsize, nrbitsp);
+}
+
+/*
+ * Determine the maximum file size for a superblock
+ */
+static inline
+void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize)
+{
+	sb->s_maxbytes = (loff_t)maxfilesize;
+	if (sb->s_maxbytes > MAX_LFS_FILESIZE || sb->s_maxbytes <= 0)
+		sb->s_maxbytes = MAX_LFS_FILESIZE;
+}
+
+/*
+ * Determine the number of bytes of data the page contains
+ */
+static inline
+unsigned int nfs_page_length(struct page *page)
+{
+	loff_t i_size = i_size_read(page_file_mapping(page)->host);
+
+	if (i_size > 0) {
+		pgoff_t page_index = page_file_index(page);
+		pgoff_t end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
+		if (page_index < end_index)
+			return PAGE_CACHE_SIZE;
+		if (page_index == end_index)
+			return ((i_size - 1) & ~PAGE_CACHE_MASK) + 1;
+	}
+	return 0;
+}
+
+/*
+ * Convert a umode to a dirent->d_type
+ */
+static inline
+unsigned char nfs_umode_to_dtype(umode_t mode)
+{
+	return (mode >> 12) & 15;
+}
+
+/*
+ * Determine the number of pages in an array of length 'len' and
+ * with a base offset of 'base'
+ */
+static inline
+unsigned int nfs_page_array_len(unsigned int base, size_t len)
+{
+	return ((unsigned long)len + (unsigned long)base +
+		PAGE_SIZE - 1) >> PAGE_SHIFT;
+}
+
+/*
+ * Convert a struct timespec into a 64-bit change attribute
+ *
+ * This does approximately the same thing as timespec_to_ns(),
+ * but for calculation efficiency, we multiply the seconds by
+ * 1024*1024*1024.
+ */
+static inline
+u64 nfs_timespec_to_change_attr(const struct timespec *ts)
+{
+	return ((u64)ts->tv_sec << 30) + ts->tv_nsec;
+}
+
+#ifdef CONFIG_CRC32
+/**
+ * nfs_fhandle_hash - calculate the crc32 hash for the filehandle
+ * @fh - pointer to filehandle
+ *
+ * returns a crc32 hash for the filehandle that is compatible with
+ * the one displayed by "wireshark".
+ */
+static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
+{
+	return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size);
+}
+#else
+static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
+{
+	return 0;
+}
+#endif

fs/nfs/iostat.h

@@ -0,0 +1,71 @@
+/*
+ *  linux/fs/nfs/iostat.h
+ *
+ *  Declarations for NFS client per-mount statistics
+ *
+ *  Copyright (C) 2005, 2006 Chuck Lever <cel@netapp.com>
+ *
+ */
+
+#ifndef _NFS_IOSTAT
+#define _NFS_IOSTAT
+
+#include <linux/percpu.h>
+#include <linux/cache.h>
+#include <linux/nfs_iostat.h>
+
+struct nfs_iostats {
+	unsigned long long	bytes[__NFSIOS_BYTESMAX];
+#ifdef CONFIG_NFS_FSCACHE
+	unsigned long long	fscache[__NFSIOS_FSCACHEMAX];
+#endif
+	unsigned long		events[__NFSIOS_COUNTSMAX];
+} ____cacheline_aligned;
+
+static inline void nfs_inc_server_stats(const struct nfs_server *server,
+					enum nfs_stat_eventcounters stat)
+{
+	this_cpu_inc(server->io_stats->events[stat]);
+}
+
+static inline void nfs_inc_stats(const struct inode *inode,
+				 enum nfs_stat_eventcounters stat)
+{
+	nfs_inc_server_stats(NFS_SERVER(inode), stat);
+}
+
+static inline void nfs_add_server_stats(const struct nfs_server *server,
+					enum nfs_stat_bytecounters stat,
+					long addend)
+{
+	this_cpu_add(server->io_stats->bytes[stat], addend);
+}
+
+static inline void nfs_add_stats(const struct inode *inode,
+				 enum nfs_stat_bytecounters stat,
+				 long addend)
+{
+	nfs_add_server_stats(NFS_SERVER(inode), stat, addend);
+}
+
+#ifdef CONFIG_NFS_FSCACHE
+static inline void nfs_add_fscache_stats(struct inode *inode,
+					 enum nfs_stat_fscachecounters stat,
+					 long addend)
+{
+	this_cpu_add(NFS_SERVER(inode)->io_stats->fscache[stat], addend);
+}
+#endif
+
+static inline struct nfs_iostats __percpu *nfs_alloc_iostats(void)
+{
+	return alloc_percpu(struct nfs_iostats);
+}
+
+static inline void nfs_free_iostats(struct nfs_iostats __percpu *stats)
+{
+	if (stats != NULL)
+		free_percpu(stats);
+}
+
+#endif /* _NFS_IOSTAT */

fs/nfs/mount_clnt.c

@@ -0,0 +1,535 @@
+/*
+ * In-kernel MOUNT protocol client
+ *
+ * Copyright (C) 1997, Olaf Kirch <okir@monad.swb.de>
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/uio.h>
+#include <linux/net.h>
+#include <linux/in.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/nfs_fs.h>
+#include "internal.h"
+
+#ifdef NFS_DEBUG
+# define NFSDBG_FACILITY	NFSDBG_MOUNT
+#endif
+
+/*
+ * Defined by RFC 1094, section A.3; and RFC 1813, section 5.1.4
+ */
+#define MNTPATHLEN		(1024)
+
+/*
+ * XDR data type sizes
+ */
+#define encode_dirpath_sz	(1 + XDR_QUADLEN(MNTPATHLEN))
+#define MNT_status_sz		(1)
+#define MNT_fhs_status_sz	(1)
+#define MNT_fhandle_sz		XDR_QUADLEN(NFS2_FHSIZE)
+#define MNT_fhandle3_sz		(1 + XDR_QUADLEN(NFS3_FHSIZE))
+#define MNT_authflav3_sz	(1 + NFS_MAX_SECFLAVORS)
+
+/*
+ * XDR argument and result sizes
+ */
+#define MNT_enc_dirpath_sz	encode_dirpath_sz
+#define MNT_dec_mountres_sz	(MNT_status_sz + MNT_fhandle_sz)
+#define MNT_dec_mountres3_sz	(MNT_status_sz + MNT_fhandle_sz + \
+				 MNT_authflav3_sz)
+
+/*
+ * Defined by RFC 1094, section A.5
+ */
+enum {
+	MOUNTPROC_NULL		= 0,
+	MOUNTPROC_MNT		= 1,
+	MOUNTPROC_DUMP		= 2,
+	MOUNTPROC_UMNT		= 3,
+	MOUNTPROC_UMNTALL	= 4,
+	MOUNTPROC_EXPORT	= 5,
+};
+
+/*
+ * Defined by RFC 1813, section 5.2
+ */
+enum {
+	MOUNTPROC3_NULL		= 0,
+	MOUNTPROC3_MNT		= 1,
+	MOUNTPROC3_DUMP		= 2,
+	MOUNTPROC3_UMNT		= 3,
+	MOUNTPROC3_UMNTALL	= 4,
+	MOUNTPROC3_EXPORT	= 5,
+};
+
+static const struct rpc_program mnt_program;
+
+/*
+ * Defined by OpenGroup XNFS Version 3W, chapter 8
+ */
+enum mountstat {
+	MNT_OK			= 0,
+	MNT_EPERM		= 1,
+	MNT_ENOENT		= 2,
+	MNT_EACCES		= 13,
+	MNT_EINVAL		= 22,
+};
+
+static struct {
+	u32 status;
+	int errno;
+} mnt_errtbl[] = {
+	{ .status = MNT_OK,			.errno = 0,		},
+	{ .status = MNT_EPERM,			.errno = -EPERM,	},
+	{ .status = MNT_ENOENT,			.errno = -ENOENT,	},
+	{ .status = MNT_EACCES,			.errno = -EACCES,	},
+	{ .status = MNT_EINVAL,			.errno = -EINVAL,	},
+};
+
+/*
+ * Defined by RFC 1813, section 5.1.5
+ */
+enum mountstat3 {
+	MNT3_OK			= 0,		/* no error */
+	MNT3ERR_PERM		= 1,		/* Not owner */
+	MNT3ERR_NOENT		= 2,		/* No such file or directory */
+	MNT3ERR_IO		= 5,		/* I/O error */
+	MNT3ERR_ACCES		= 13,		/* Permission denied */
+	MNT3ERR_NOTDIR		= 20,		/* Not a directory */
+	MNT3ERR_INVAL		= 22,		/* Invalid argument */
+	MNT3ERR_NAMETOOLONG	= 63,		/* Filename too long */
+	MNT3ERR_NOTSUPP		= 10004,	/* Operation not supported */
+	MNT3ERR_SERVERFAULT	= 10006,	/* A failure on the server */
+};
+
+static struct {
+	u32 status;
+	int errno;
+} mnt3_errtbl[] = {
+	{ .status = MNT3_OK,			.errno = 0,		},
+	{ .status = MNT3ERR_PERM,		.errno = -EPERM,	},
+	{ .status = MNT3ERR_NOENT,		.errno = -ENOENT,	},
+	{ .status = MNT3ERR_IO,			.errno = -EIO,		},
+	{ .status = MNT3ERR_ACCES,		.errno = -EACCES,	},
+	{ .status = MNT3ERR_NOTDIR,		.errno = -ENOTDIR,	},
+	{ .status = MNT3ERR_INVAL,		.errno = -EINVAL,	},
+	{ .status = MNT3ERR_NAMETOOLONG,	.errno = -ENAMETOOLONG,	},
+	{ .status = MNT3ERR_NOTSUPP,		.errno = -ENOTSUPP,	},
+	{ .status = MNT3ERR_SERVERFAULT,	.errno = -EREMOTEIO,	},
+};
+
+struct mountres {
+	int errno;
+	struct nfs_fh *fh;
+	unsigned int *auth_count;
+	rpc_authflavor_t *auth_flavors;
+};
+
+struct mnt_fhstatus {
+	u32 status;
+	struct nfs_fh *fh;
+};
+
+/**
+ * nfs_mount - Obtain an NFS file handle for the given host and path
+ * @info: pointer to mount request arguments
+ *
+ * Uses default timeout parameters specified by underlying transport. On
+ * successful return, the auth_flavs list and auth_flav_len will be populated
+ * with the list from the server or a faked-up list if the server didn't
+ * provide one.
+ */
+int nfs_mount(struct nfs_mount_request *info)
+{
+	struct mountres	result = {
+		.fh		= info->fh,
+		.auth_count	= info->auth_flav_len,
+		.auth_flavors	= info->auth_flavs,
+	};
+	struct rpc_message msg	= {
+		.rpc_argp	= info->dirpath,
+		.rpc_resp	= &result,
+	};
+	struct rpc_create_args args = {
+		.net		= info->net,
+		.protocol	= info->protocol,
+		.address	= info->sap,
+		.addrsize	= info->salen,
+		.servername	= info->hostname,
+		.program	= &mnt_program,
+		.version	= info->version,
+		.authflavor	= RPC_AUTH_UNIX,
+	};
+	struct rpc_clnt		*mnt_clnt;
+	int			status;
+
+	dprintk("NFS: sending MNT request for %s:%s\n",
+		(info->hostname ? info->hostname : "server"),
+			info->dirpath);
+
+	if (strlen(info->dirpath) > MNTPATHLEN)
+		return -ENAMETOOLONG;
+
+	if (info->noresvport)
+		args.flags |= RPC_CLNT_CREATE_NONPRIVPORT;
+
+	mnt_clnt = rpc_create(&args);
+	if (IS_ERR(mnt_clnt))
+		goto out_clnt_err;
+
+	if (info->version == NFS_MNT3_VERSION)
+		msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC3_MNT];
+	else
+		msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC_MNT];
+
+	status = rpc_call_sync(mnt_clnt, &msg, RPC_TASK_SOFT|RPC_TASK_TIMEOUT);
+	rpc_shutdown_client(mnt_clnt);
+
+	if (status < 0)
+		goto out_call_err;
+	if (result.errno != 0)
+		goto out_mnt_err;
+
+	dprintk("NFS: MNT request succeeded\n");
+	status = 0;
+
+	/*
+	 * If the server didn't provide a flavor list, allow the
+	 * client to try any flavor.
+	 */
+	if (info->version != NFS_MNT3_VERSION || *info->auth_flav_len == 0) {
+		dprintk("NFS: Faking up auth_flavs list\n");
+		info->auth_flavs[0] = RPC_AUTH_NULL;
+		*info->auth_flav_len = 1;
+	}
+out:
+	return status;
+
+out_clnt_err:
+	status = PTR_ERR(mnt_clnt);
+	dprintk("NFS: failed to create MNT RPC client, status=%d\n", status);
+	goto out;
+
+out_call_err:
+	dprintk("NFS: MNT request failed, status=%d\n", status);
+	goto out;
+
+out_mnt_err:
+	dprintk("NFS: MNT server returned result %d\n", result.errno);
+	status = result.errno;
+	goto out;
+}
+
+/**
+ * nfs_umount - Notify a server that we have unmounted this export
+ * @info: pointer to umount request arguments
+ *
+ * MOUNTPROC_UMNT is advisory, so we set a short timeout, and always
+ * use UDP.
+ */
+void nfs_umount(const struct nfs_mount_request *info)
+{
+	static const struct rpc_timeout nfs_umnt_timeout = {
+		.to_initval = 1 * HZ,
+		.to_maxval = 3 * HZ,
+		.to_retries = 2,
+	};
+	struct rpc_create_args args = {
+		.net		= info->net,
+		.protocol	= IPPROTO_UDP,
+		.address	= info->sap,
+		.addrsize	= info->salen,
+		.timeout	= &nfs_umnt_timeout,
+		.servername	= info->hostname,
+		.program	= &mnt_program,
+		.version	= info->version,
+		.authflavor	= RPC_AUTH_UNIX,
+		.flags		= RPC_CLNT_CREATE_NOPING,
+	};
+	struct rpc_message msg	= {
+		.rpc_argp	= info->dirpath,
+	};
+	struct rpc_clnt *clnt;
+	int status;
+
+	if (strlen(info->dirpath) > MNTPATHLEN)
+		return;
+
+	if (info->noresvport)
+		args.flags |= RPC_CLNT_CREATE_NONPRIVPORT;
+
+	clnt = rpc_create(&args);
+	if (IS_ERR(clnt))
+		goto out_clnt_err;
+
+	dprintk("NFS: sending UMNT request for %s:%s\n",
+		(info->hostname ? info->hostname : "server"), info->dirpath);
+
+	if (info->version == NFS_MNT3_VERSION)
+		msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC3_UMNT];
+	else
+		msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC_UMNT];
+
+	status = rpc_call_sync(clnt, &msg, 0);
+	rpc_shutdown_client(clnt);
+
+	if (unlikely(status < 0))
+		goto out_call_err;
+
+	return;
+
+out_clnt_err:
+	dprintk("NFS: failed to create UMNT RPC client, status=%ld\n",
+			PTR_ERR(clnt));
+	return;
+
+out_call_err:
+	dprintk("NFS: UMNT request failed, status=%d\n", status);
+}
+
+/*
+ * XDR encode/decode functions for MOUNT
+ */
+
+static void encode_mntdirpath(struct xdr_stream *xdr, const char *pathname)
+{
+	const u32 pathname_len = strlen(pathname);
+	__be32 *p;
+
+	p = xdr_reserve_space(xdr, 4 + pathname_len);
+	xdr_encode_opaque(p, pathname, pathname_len);
+}
+
+static void mnt_xdr_enc_dirpath(struct rpc_rqst *req, struct xdr_stream *xdr,
+				const char *dirpath)
+{
+	encode_mntdirpath(xdr, dirpath);
+}
+
+/*
+ * RFC 1094: "A non-zero status indicates some sort of error.  In this
+ * case, the status is a UNIX error number."  This can be problematic
+ * if the server and client use different errno values for the same
+ * error.
+ *
+ * However, the OpenGroup XNFS spec provides a simple mapping that is
+ * independent of local errno values on the server and the client.
+ */
+static int decode_status(struct xdr_stream *xdr, struct mountres *res)
+{
+	unsigned int i;
+	u32 status;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(p == NULL))
+		return -EIO;
+	status = be32_to_cpup(p);
+
+	for (i = 0; i < ARRAY_SIZE(mnt_errtbl); i++) {
+		if (mnt_errtbl[i].status == status) {
+			res->errno = mnt_errtbl[i].errno;
+			return 0;
+		}
+	}
+
+	dprintk("NFS: unrecognized MNT status code: %u\n", status);
+	res->errno = -EACCES;
+	return 0;
+}
+
+static int decode_fhandle(struct xdr_stream *xdr, struct mountres *res)
+{
+	struct nfs_fh *fh = res->fh;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, NFS2_FHSIZE);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	fh->size = NFS2_FHSIZE;
+	memcpy(fh->data, p, NFS2_FHSIZE);
+	return 0;
+}
+
+static int mnt_xdr_dec_mountres(struct rpc_rqst *req,
+				struct xdr_stream *xdr,
+				struct mountres *res)
+{
+	int status;
+
+	status = decode_status(xdr, res);
+	if (unlikely(status != 0 || res->errno != 0))
+		return status;
+	return decode_fhandle(xdr, res);
+}
+
+static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res)
+{
+	unsigned int i;
+	u32 status;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(p == NULL))
+		return -EIO;
+	status = be32_to_cpup(p);
+
+	for (i = 0; i < ARRAY_SIZE(mnt3_errtbl); i++) {
+		if (mnt3_errtbl[i].status == status) {
+			res->errno = mnt3_errtbl[i].errno;
+			return 0;
+		}
+	}
+
+	dprintk("NFS: unrecognized MNT3 status code: %u\n", status);
+	res->errno = -EACCES;
+	return 0;
+}
+
+static int decode_fhandle3(struct xdr_stream *xdr, struct mountres *res)
+{
+	struct nfs_fh *fh = res->fh;
+	u32 size;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	size = be32_to_cpup(p);
+	if (size > NFS3_FHSIZE || size == 0)
+		return -EIO;
+
+	p = xdr_inline_decode(xdr, size);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	fh->size = size;
+	memcpy(fh->data, p, size);
+	return 0;
+}
+
+static int decode_auth_flavors(struct xdr_stream *xdr, struct mountres *res)
+{
+	rpc_authflavor_t *flavors = res->auth_flavors;
+	unsigned int *count = res->auth_count;
+	u32 entries, i;
+	__be32 *p;
+
+	if (*count == 0)
+		return 0;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(p == NULL))
+		return -EIO;
+	entries = be32_to_cpup(p);
+	dprintk("NFS: received %u auth flavors\n", entries);
+	if (entries > NFS_MAX_SECFLAVORS)
+		entries = NFS_MAX_SECFLAVORS;
+
+	p = xdr_inline_decode(xdr, 4 * entries);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	if (entries > *count)
+		entries = *count;
+
+	for (i = 0; i < entries; i++) {
+		flavors[i] = be32_to_cpup(p++);
+		dprintk("NFS:   auth flavor[%u]: %d\n", i, flavors[i]);
+	}
+	*count = i;
+
+	return 0;
+}
+
+static int mnt_xdr_dec_mountres3(struct rpc_rqst *req,
+				 struct xdr_stream *xdr,
+				 struct mountres *res)
+{
+	int status;
+
+	status = decode_fhs_status(xdr, res);
+	if (unlikely(status != 0 || res->errno != 0))
+		return status;
+	status = decode_fhandle3(xdr, res);
+	if (unlikely(status != 0)) {
+		res->errno = -EBADHANDLE;
+		return 0;
+	}
+	return decode_auth_flavors(xdr, res);
+}
+
+static struct rpc_procinfo mnt_procedures[] = {
+	[MOUNTPROC_MNT] = {
+		.p_proc		= MOUNTPROC_MNT,
+		.p_encode	= (kxdreproc_t)mnt_xdr_enc_dirpath,
+		.p_decode	= (kxdrdproc_t)mnt_xdr_dec_mountres,
+		.p_arglen	= MNT_enc_dirpath_sz,
+		.p_replen	= MNT_dec_mountres_sz,
+		.p_statidx	= MOUNTPROC_MNT,
+		.p_name		= "MOUNT",
+	},
+	[MOUNTPROC_UMNT] = {
+		.p_proc		= MOUNTPROC_UMNT,
+		.p_encode	= (kxdreproc_t)mnt_xdr_enc_dirpath,
+		.p_arglen	= MNT_enc_dirpath_sz,
+		.p_statidx	= MOUNTPROC_UMNT,
+		.p_name		= "UMOUNT",
+	},
+};
+
+static struct rpc_procinfo mnt3_procedures[] = {
+	[MOUNTPROC3_MNT] = {
+		.p_proc		= MOUNTPROC3_MNT,
+		.p_encode	= (kxdreproc_t)mnt_xdr_enc_dirpath,
+		.p_decode	= (kxdrdproc_t)mnt_xdr_dec_mountres3,
+		.p_arglen	= MNT_enc_dirpath_sz,
+		.p_replen	= MNT_dec_mountres3_sz,
+		.p_statidx	= MOUNTPROC3_MNT,
+		.p_name		= "MOUNT",
+	},
+	[MOUNTPROC3_UMNT] = {
+		.p_proc		= MOUNTPROC3_UMNT,
+		.p_encode	= (kxdreproc_t)mnt_xdr_enc_dirpath,
+		.p_arglen	= MNT_enc_dirpath_sz,
+		.p_statidx	= MOUNTPROC3_UMNT,
+		.p_name		= "UMOUNT",
+	},
+};
+
+
+static const struct rpc_version mnt_version1 = {
+	.number		= 1,
+	.nrprocs	= ARRAY_SIZE(mnt_procedures),
+	.procs		= mnt_procedures,
+};
+
+static const struct rpc_version mnt_version3 = {
+	.number		= 3,
+	.nrprocs	= ARRAY_SIZE(mnt3_procedures),
+	.procs		= mnt3_procedures,
+};
+
+static const struct rpc_version *mnt_version[] = {
+	NULL,
+	&mnt_version1,
+	NULL,
+	&mnt_version3,
+};
+
+static struct rpc_stat mnt_stats;
+
+static const struct rpc_program mnt_program = {
+	.name		= "mount",
+	.number		= NFS_MNT_PROGRAM,
+	.nrvers		= ARRAY_SIZE(mnt_version),
+	.version	= mnt_version,
+	.stats		= &mnt_stats,
+};

fs/nfs/namespace.c

@@ -0,0 +1,289 @@
+/*
+ * linux/fs/nfs/namespace.c
+ *
+ * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com>
+ * - Modified by David Howells <dhowells@redhat.com>
+ *
+ * NFS namespace
+ */
+
+#include <linux/module.h>
+#include <linux/dcache.h>
+#include <linux/gfp.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/nfs_fs.h>
+#include <linux/string.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/vfs.h>
+#include <linux/sunrpc/gss_api.h>
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_VFS
+
+static void nfs_expire_automounts(struct work_struct *work);
+
+static LIST_HEAD(nfs_automount_list);
+static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts);
+int nfs_mountpoint_expiry_timeout = 500 * HZ;
+
+/*
+ * nfs_path - reconstruct the path given an arbitrary dentry
+ * @base - used to return pointer to the end of devname part of path
+ * @dentry - pointer to dentry
+ * @buffer - result buffer
+ * @buflen - length of buffer
+ * @flags - options (see below)
+ *
+ * Helper function for constructing the server pathname
+ * by arbitrary hashed dentry.
+ *
+ * This is mainly for use in figuring out the path on the
+ * server side when automounting on top of an existing partition
+ * and in generating /proc/mounts and friends.
+ *
+ * Supported flags:
+ * NFS_PATH_CANONICAL: ensure there is exactly one slash after
+ *		       the original device (export) name
+ *		       (if unset, the original name is returned verbatim)
+ */
+char *nfs_path(char **p, struct dentry *dentry, char *buffer, ssize_t buflen,
+	       unsigned flags)
+{
+	char *end;
+	int namelen;
+	unsigned seq;
+	const char *base;
+
+rename_retry:
+	end = buffer+buflen;
+	*--end = '\0';
+	buflen--;
+
+	seq = read_seqbegin(&rename_lock);
+	rcu_read_lock();
+	while (1) {
+		spin_lock(&dentry->d_lock);
+		if (IS_ROOT(dentry))
+			break;
+		namelen = dentry->d_name.len;
+		buflen -= namelen + 1;
+		if (buflen < 0)
+			goto Elong_unlock;
+		end -= namelen;
+		memcpy(end, dentry->d_name.name, namelen);
+		*--end = '/';
+		spin_unlock(&dentry->d_lock);
+		dentry = dentry->d_parent;
+	}
+	if (read_seqretry(&rename_lock, seq)) {
+		spin_unlock(&dentry->d_lock);
+		rcu_read_unlock();
+		goto rename_retry;
+	}
+	if ((flags & NFS_PATH_CANONICAL) && *end != '/') {
+		if (--buflen < 0) {
+			spin_unlock(&dentry->d_lock);
+			rcu_read_unlock();
+			goto Elong;
+		}
+		*--end = '/';
+	}
+	*p = end;
+	base = dentry->d_fsdata;
+	if (!base) {
+		spin_unlock(&dentry->d_lock);
+		rcu_read_unlock();
+		WARN_ON(1);
+		return end;
+	}
+	namelen = strlen(base);
+	if (flags & NFS_PATH_CANONICAL) {
+		/* Strip off excess slashes in base string */
+		while (namelen > 0 && base[namelen - 1] == '/')
+			namelen--;
+	}
+	buflen -= namelen;
+	if (buflen < 0) {
+		spin_unlock(&dentry->d_lock);
+		rcu_read_unlock();
+		goto Elong;
+	}
+	end -= namelen;
+	memcpy(end, base, namelen);
+	spin_unlock(&dentry->d_lock);
+	rcu_read_unlock();
+	return end;
+Elong_unlock:
+	spin_unlock(&dentry->d_lock);
+	rcu_read_unlock();
+	if (read_seqretry(&rename_lock, seq))
+		goto rename_retry;
+Elong:
+	return ERR_PTR(-ENAMETOOLONG);
+}
+EXPORT_SYMBOL_GPL(nfs_path);
+
+/*
+ * nfs_d_automount - Handle crossing a mountpoint on the server
+ * @path - The mountpoint
+ *
+ * When we encounter a mountpoint on the server, we want to set up
+ * a mountpoint on the client too, to prevent inode numbers from
+ * colliding, and to allow "df" to work properly.
+ * On NFSv4, we also want to allow for the fact that different
+ * filesystems may be migrated to different servers in a failover
+ * situation, and that different filesystems may want to use
+ * different security flavours.
+ */
+struct vfsmount *nfs_d_automount(struct path *path)
+{
+	struct vfsmount *mnt;
+	struct nfs_server *server = NFS_SERVER(path->dentry->d_inode);
+	struct nfs_fh *fh = NULL;
+	struct nfs_fattr *fattr = NULL;
+
+	dprintk("--> nfs_d_automount()\n");
+
+	mnt = ERR_PTR(-ESTALE);
+	if (IS_ROOT(path->dentry))
+		goto out_nofree;
+
+	mnt = ERR_PTR(-ENOMEM);
+	fh = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	if (fh == NULL || fattr == NULL)
+		goto out;
+
+	dprintk("%s: enter\n", __func__);
+
+	mnt = server->nfs_client->rpc_ops->submount(server, path->dentry, fh, fattr);
+	if (IS_ERR(mnt))
+		goto out;
+
+	dprintk("%s: done, success\n", __func__);
+	mntget(mnt); /* prevent immediate expiration */
+	mnt_set_expiry(mnt, &nfs_automount_list);
+	schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+
+out:
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fh);
+out_nofree:
+	if (IS_ERR(mnt))
+		dprintk("<-- %s(): error %ld\n", __func__, PTR_ERR(mnt));
+	else
+		dprintk("<-- %s() = %p\n", __func__, mnt);
+	return mnt;
+}
+
+static int
+nfs_namespace_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
+{
+	if (NFS_FH(dentry->d_inode)->size != 0)
+		return nfs_getattr(mnt, dentry, stat);
+	generic_fillattr(dentry->d_inode, stat);
+	return 0;
+}
+
+static int
+nfs_namespace_setattr(struct dentry *dentry, struct iattr *attr)
+{
+	if (NFS_FH(dentry->d_inode)->size != 0)
+		return nfs_setattr(dentry, attr);
+	return -EACCES;
+}
+
+const struct inode_operations nfs_mountpoint_inode_operations = {
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+};
+
+const struct inode_operations nfs_referral_inode_operations = {
+	.getattr	= nfs_namespace_getattr,
+	.setattr	= nfs_namespace_setattr,
+};
+
+static void nfs_expire_automounts(struct work_struct *work)
+{
+	struct list_head *list = &nfs_automount_list;
+
+	mark_mounts_for_expiry(list);
+	if (!list_empty(list))
+		schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+}
+
+void nfs_release_automount_timer(void)
+{
+	if (list_empty(&nfs_automount_list))
+		cancel_delayed_work(&nfs_automount_task);
+}
+
+/*
+ * Clone a mountpoint of the appropriate type
+ */
+static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
+					   const char *devname,
+					   struct nfs_clone_mount *mountdata)
+{
+	return vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata);
+}
+
+/**
+ * nfs_do_submount - set up mountpoint when crossing a filesystem boundary
+ * @dentry - parent directory
+ * @fh - filehandle for new root dentry
+ * @fattr - attributes for new root inode
+ * @authflavor - security flavor to use when performing the mount
+ *
+ */
+struct vfsmount *nfs_do_submount(struct dentry *dentry, struct nfs_fh *fh,
+				 struct nfs_fattr *fattr, rpc_authflavor_t authflavor)
+{
+	struct nfs_clone_mount mountdata = {
+		.sb = dentry->d_sb,
+		.dentry = dentry,
+		.fh = fh,
+		.fattr = fattr,
+		.authflavor = authflavor,
+	};
+	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
+	char *page = (char *) __get_free_page(GFP_USER);
+	char *devname;
+
+	dprintk("--> nfs_do_submount()\n");
+
+	dprintk("%s: submounting on %pd2\n", __func__,
+			dentry);
+	if (page == NULL)
+		goto out;
+	devname = nfs_devname(dentry, page, PAGE_SIZE);
+	mnt = (struct vfsmount *)devname;
+	if (IS_ERR(devname))
+		goto free_page;
+	mnt = nfs_do_clone_mount(NFS_SB(dentry->d_sb), devname, &mountdata);
+free_page:
+	free_page((unsigned long)page);
+out:
+	dprintk("%s: done\n", __func__);
+
+	dprintk("<-- nfs_do_submount() = %p\n", mnt);
+	return mnt;
+}
+EXPORT_SYMBOL_GPL(nfs_do_submount);
+
+struct vfsmount *nfs_submount(struct nfs_server *server, struct dentry *dentry,
+			      struct nfs_fh *fh, struct nfs_fattr *fattr)
+{
+	int err;
+	struct dentry *parent = dget_parent(dentry);
+
+	/* Look it up again to get its attributes */
+	err = server->nfs_client->rpc_ops->lookup(parent->d_inode, &dentry->d_name, fh, fattr, NULL);
+	dput(parent);
+	if (err != 0)
+		return ERR_PTR(err);
+
+	return nfs_do_submount(dentry, fh, fattr, server->client->cl_auth->au_flavor);
+}
+EXPORT_SYMBOL_GPL(nfs_submount);

fs/nfs/netns.h

@@ -0,0 +1,40 @@
+/*
+ * NFS-private data for each "struct net".  Accessed with net_generic().
+ */
+
+#ifndef __NFS_NETNS_H__
+#define __NFS_NETNS_H__
+
+#include <linux/nfs4.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+
+struct bl_dev_msg {
+	int32_t status;
+	uint32_t major, minor;
+};
+
+struct nfs_net {
+	struct cache_detail *nfs_dns_resolve;
+	struct rpc_pipe *bl_device_pipe;
+	struct bl_dev_msg bl_mount_reply;
+	wait_queue_head_t bl_wq;
+	struct mutex bl_mutex;
+	struct list_head nfs_client_list;
+	struct list_head nfs_volume_list;
+#if IS_ENABLED(CONFIG_NFS_V4)
+	struct idr cb_ident_idr; /* Protected by nfs_client_lock */
+	unsigned short nfs_callback_tcpport;
+	unsigned short nfs_callback_tcpport6;
+	int cb_users[NFS4_MAX_MINOR_VERSION + 1];
+#endif
+	spinlock_t nfs_client_lock;
+	struct timespec boot_time;
+#ifdef CONFIG_PROC_FS
+	struct proc_dir_entry *proc_nfsfs;
+#endif
+};
+
+extern int nfs_net_id;
+
+#endif

fs/nfs/nfs.h

@@ -0,0 +1,29 @@
+/*
+ * Copyright (c) 2012 Netapp, Inc. All rights reserved.
+ *
+ * Function and structures exported by the NFS module
+ * for use by NFS version-specific modules.
+ */
+#ifndef __LINUX_INTERNAL_NFS_H
+#define __LINUX_INTERNAL_NFS_H
+
+#include <linux/fs.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/nfs_xdr.h>
+
+struct nfs_subversion {
+	struct module *owner;	/* THIS_MODULE pointer */
+	struct file_system_type *nfs_fs;	/* NFS filesystem type */
+	const struct rpc_version *rpc_vers;	/* NFS version information */
+	const struct nfs_rpc_ops *rpc_ops;	/* NFS operations */
+	const struct super_operations *sops;	/* NFS Super operations */
+	const struct xattr_handler **xattr;	/* NFS xattr handlers */
+	struct list_head list;		/* List of NFS versions */
+};
+
+struct nfs_subversion *get_nfs_version(unsigned int);
+void put_nfs_version(struct nfs_subversion *);
+void register_nfs_version(struct nfs_subversion *);
+void unregister_nfs_version(struct nfs_subversion *);
+
+#endif /* __LINUX_INTERNAL_NFS_H */

fs/nfs/nfs2super.c

@@ -0,0 +1,31 @@
+/*
+ * Copyright (c) 2012 Netapp, Inc. All rights reserved.
+ */
+#include <linux/module.h>
+#include <linux/nfs_fs.h>
+#include "internal.h"
+#include "nfs.h"
+
+static struct nfs_subversion nfs_v2 = {
+	.owner = THIS_MODULE,
+	.nfs_fs   = &nfs_fs_type,
+	.rpc_vers = &nfs_version2,
+	.rpc_ops  = &nfs_v2_clientops,
+	.sops     = &nfs_sops,
+};
+
+static int __init init_nfs_v2(void)
+{
+	register_nfs_version(&nfs_v2);
+	return 0;
+}
+
+static void __exit exit_nfs_v2(void)
+{
+	unregister_nfs_version(&nfs_v2);
+}
+
+MODULE_LICENSE("GPL");
+
+module_init(init_nfs_v2);
+module_exit(exit_nfs_v2);

fs/nfs/nfs3_fs.h

@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2014 Anna Schumaker.
+ *
+ * NFSv3-specific filesystem definitions and declarations
+ */
+#ifndef __LINUX_FS_NFS_NFS3_FS_H
+#define __LINUX_FS_NFS_NFS3_FS_H
+
+/*
+ * nfs3acl.c
+ */
+#ifdef CONFIG_NFS_V3_ACL
+extern struct posix_acl *nfs3_get_acl(struct inode *inode, int type);
+extern int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type);
+extern int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
+		struct posix_acl *dfacl);
+extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t);
+extern const struct xattr_handler *nfs3_xattr_handlers[];
+#else
+static inline int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
+		struct posix_acl *dfacl)
+{
+	return 0;
+}
+#define nfs3_listxattr NULL
+#endif /* CONFIG_NFS_V3_ACL */
+
+/* nfs3client.c */
+struct nfs_server *nfs3_create_server(struct nfs_mount_info *, struct nfs_subversion *);
+struct nfs_server *nfs3_clone_server(struct nfs_server *, struct nfs_fh *,
+				     struct nfs_fattr *, rpc_authflavor_t);
+
+
+#endif /* __LINUX_FS_NFS_NFS3_FS_H */

fs/nfs/nfs3acl.c

@@ -0,0 +1,296 @@
+#include <linux/fs.h>
+#include <linux/gfp.h>
+#include <linux/nfs.h>
+#include <linux/nfs3.h>
+#include <linux/nfs_fs.h>
+#include <linux/posix_acl_xattr.h>
+#include <linux/nfsacl.h>
+
+#include "internal.h"
+#include "nfs3_fs.h"
+
+#define NFSDBG_FACILITY	NFSDBG_PROC
+
+struct posix_acl *nfs3_get_acl(struct inode *inode, int type)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct page *pages[NFSACL_MAXPAGES] = { };
+	struct nfs3_getaclargs args = {
+		.fh = NFS_FH(inode),
+		/* The xdr layer may allocate pages here. */
+		.pages = pages,
+	};
+	struct nfs3_getaclres res = {
+		NULL,
+	};
+	struct rpc_message msg = {
+		.rpc_argp	= &args,
+		.rpc_resp	= &res,
+	};
+	int status, count;
+
+	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	status = nfs_revalidate_inode(server, inode);
+	if (status < 0)
+		return ERR_PTR(status);
+
+	/*
+	 * Only get the access acl when explicitly requested: We don't
+	 * need it for access decisions, and only some applications use
+	 * it. Applications which request the access acl first are not
+	 * penalized from this optimization.
+	 */
+	if (type == ACL_TYPE_ACCESS)
+		args.mask |= NFS_ACLCNT|NFS_ACL;
+	if (S_ISDIR(inode->i_mode))
+		args.mask |= NFS_DFACLCNT|NFS_DFACL;
+	if (args.mask == 0)
+		return NULL;
+
+	dprintk("NFS call getacl\n");
+	msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_GETACL];
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	status = rpc_call_sync(server->client_acl, &msg, 0);
+	dprintk("NFS reply getacl: %d\n", status);
+
+	/* pages may have been allocated at the xdr layer. */
+	for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++)
+		__free_page(args.pages[count]);
+
+	switch (status) {
+		case 0:
+			status = nfs_refresh_inode(inode, res.fattr);
+			break;
+		case -EPFNOSUPPORT:
+		case -EPROTONOSUPPORT:
+			dprintk("NFS_V3_ACL extension not supported; disabling\n");
+			server->caps &= ~NFS_CAP_ACLS;
+		case -ENOTSUPP:
+			status = -EOPNOTSUPP;
+		default:
+			goto getout;
+	}
+	if ((args.mask & res.mask) != args.mask) {
+		status = -EIO;
+		goto getout;
+	}
+
+	if (res.acl_access != NULL) {
+		if ((posix_acl_equiv_mode(res.acl_access, NULL) == 0) ||
+		    res.acl_access->a_count == 0) {
+			posix_acl_release(res.acl_access);
+			res.acl_access = NULL;
+		}
+	}
+
+	if (res.mask & NFS_ACL)
+		set_cached_acl(inode, ACL_TYPE_ACCESS, res.acl_access);
+	else
+		forget_cached_acl(inode, ACL_TYPE_ACCESS);
+
+	if (res.mask & NFS_DFACL)
+		set_cached_acl(inode, ACL_TYPE_DEFAULT, res.acl_default);
+	else
+		forget_cached_acl(inode, ACL_TYPE_DEFAULT);
+
+	nfs_free_fattr(res.fattr);
+	if (type == ACL_TYPE_ACCESS) {
+		posix_acl_release(res.acl_default);
+		return res.acl_access;
+	} else {
+		posix_acl_release(res.acl_access);
+		return res.acl_default;
+	}
+
+getout:
+	posix_acl_release(res.acl_access);
+	posix_acl_release(res.acl_default);
+	nfs_free_fattr(res.fattr);
+	return ERR_PTR(status);
+}
+
+static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
+		struct posix_acl *dfacl)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct nfs_fattr *fattr;
+	struct page *pages[NFSACL_MAXPAGES];
+	struct nfs3_setaclargs args = {
+		.inode = inode,
+		.mask = NFS_ACL,
+		.acl_access = acl,
+		.pages = pages,
+	};
+	struct rpc_message msg = {
+		.rpc_argp	= &args,
+		.rpc_resp	= &fattr,
+	};
+	int status = 0;
+
+	if (acl == NULL && (!S_ISDIR(inode->i_mode) || dfacl == NULL))
+		goto out;
+
+	status = -EOPNOTSUPP;
+	if (!nfs_server_capable(inode, NFS_CAP_ACLS))
+		goto out;
+
+	/* We are doing this here because XDR marshalling does not
+	 * return any results, it BUGs. */
+	status = -ENOSPC;
+	if (acl != NULL && acl->a_count > NFS_ACL_MAX_ENTRIES)
+		goto out;
+	if (dfacl != NULL && dfacl->a_count > NFS_ACL_MAX_ENTRIES)
+		goto out;
+	if (S_ISDIR(inode->i_mode)) {
+		args.mask |= NFS_DFACL;
+		args.acl_default = dfacl;
+		args.len = nfsacl_size(acl, dfacl);
+	} else
+		args.len = nfsacl_size(acl, NULL);
+
+	if (args.len > NFS_ACL_INLINE_BUFSIZE) {
+		unsigned int npages = 1 + ((args.len - 1) >> PAGE_SHIFT);
+
+		status = -ENOMEM;
+		do {
+			args.pages[args.npages] = alloc_page(GFP_KERNEL);
+			if (args.pages[args.npages] == NULL)
+				goto out_freepages;
+			args.npages++;
+		} while (args.npages < npages);
+	}
+
+	dprintk("NFS call setacl\n");
+	status = -ENOMEM;
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out_freepages;
+
+	msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_SETACL];
+	msg.rpc_resp = fattr;
+	status = rpc_call_sync(server->client_acl, &msg, 0);
+	nfs_access_zap_cache(inode);
+	nfs_zap_acl_cache(inode);
+	dprintk("NFS reply setacl: %d\n", status);
+
+	switch (status) {
+		case 0:
+			status = nfs_refresh_inode(inode, fattr);
+			set_cached_acl(inode, ACL_TYPE_ACCESS, acl);
+			set_cached_acl(inode, ACL_TYPE_DEFAULT, dfacl);
+			break;
+		case -EPFNOSUPPORT:
+		case -EPROTONOSUPPORT:
+			dprintk("NFS_V3_ACL SETACL RPC not supported"
+					"(will not retry)\n");
+			server->caps &= ~NFS_CAP_ACLS;
+		case -ENOTSUPP:
+			status = -EOPNOTSUPP;
+	}
+	nfs_free_fattr(fattr);
+out_freepages:
+	while (args.npages != 0) {
+		args.npages--;
+		__free_page(args.pages[args.npages]);
+	}
+out:
+	return status;
+}
+
+int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
+		struct posix_acl *dfacl)
+{
+	int ret;
+	ret = __nfs3_proc_setacls(inode, acl, dfacl);
+	return (ret == -EOPNOTSUPP) ? 0 : ret;
+
+}
+
+int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+{
+	struct posix_acl *alloc = NULL, *dfacl = NULL;
+	int status;
+
+	if (S_ISDIR(inode->i_mode)) {
+		switch(type) {
+		case ACL_TYPE_ACCESS:
+			alloc = dfacl = get_acl(inode, ACL_TYPE_DEFAULT);
+			if (IS_ERR(alloc))
+				goto fail;
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			dfacl = acl;
+			alloc = acl = get_acl(inode, ACL_TYPE_ACCESS);
+			if (IS_ERR(alloc))
+				goto fail;
+			break;
+		}
+	}
+
+	if (acl == NULL) {
+		alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
+		if (IS_ERR(alloc))
+			goto fail;
+	}
+	status = __nfs3_proc_setacls(inode, acl, dfacl);
+	posix_acl_release(alloc);
+	return status;
+
+fail:
+	return PTR_ERR(alloc);
+}
+
+const struct xattr_handler *nfs3_xattr_handlers[] = {
+	&posix_acl_access_xattr_handler,
+	&posix_acl_default_xattr_handler,
+	NULL,
+};
+
+static int
+nfs3_list_one_acl(struct inode *inode, int type, const char *name, void *data,
+		size_t size, ssize_t *result)
+{
+	struct posix_acl *acl;
+	char *p = data + *result;
+
+	acl = get_acl(inode, type);
+	if (IS_ERR_OR_NULL(acl))
+		return 0;
+
+	posix_acl_release(acl);
+
+	*result += strlen(name);
+	*result += 1;
+	if (!size)
+		return 0;
+	if (*result > size)
+		return -ERANGE;
+
+	strcpy(p, name);
+	return 0;
+}
+
+ssize_t
+nfs3_listxattr(struct dentry *dentry, char *data, size_t size)
+{
+	struct inode *inode = dentry->d_inode;
+	ssize_t result = 0;
+	int error;
+
+	error = nfs3_list_one_acl(inode, ACL_TYPE_ACCESS,
+			POSIX_ACL_XATTR_ACCESS, data, size, &result);
+	if (error)
+		return error;
+
+	error = nfs3_list_one_acl(inode, ACL_TYPE_DEFAULT,
+			POSIX_ACL_XATTR_DEFAULT, data, size, &result);
+	if (error)
+		return error;
+	return result;
+}

fs/nfs/nfs3client.c

@@ -0,0 +1,66 @@
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include "internal.h"
+#include "nfs3_fs.h"
+
+#ifdef CONFIG_NFS_V3_ACL
+static struct rpc_stat		nfsacl_rpcstat = { &nfsacl_program };
+static const struct rpc_version *nfsacl_version[] = {
+	[3]			= &nfsacl_version3,
+};
+
+const struct rpc_program nfsacl_program = {
+	.name			= "nfsacl",
+	.number			= NFS_ACL_PROGRAM,
+	.nrvers			= ARRAY_SIZE(nfsacl_version),
+	.version		= nfsacl_version,
+	.stats			= &nfsacl_rpcstat,
+};
+
+/*
+ * Initialise an NFSv3 ACL client connection
+ */
+static void nfs_init_server_aclclient(struct nfs_server *server)
+{
+	if (server->flags & NFS_MOUNT_NOACL)
+		goto out_noacl;
+
+	server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3);
+	if (IS_ERR(server->client_acl))
+		goto out_noacl;
+
+	/* No errors! Assume that Sun nfsacls are supported */
+	server->caps |= NFS_CAP_ACLS;
+	return;
+
+out_noacl:
+	server->caps &= ~NFS_CAP_ACLS;
+}
+#else
+static inline void nfs_init_server_aclclient(struct nfs_server *server)
+{
+	server->flags &= ~NFS_MOUNT_NOACL;
+	server->caps &= ~NFS_CAP_ACLS;
+}
+#endif
+
+struct nfs_server *nfs3_create_server(struct nfs_mount_info *mount_info,
+				      struct nfs_subversion *nfs_mod)
+{
+	struct nfs_server *server = nfs_create_server(mount_info, nfs_mod);
+	/* Create a client RPC handle for the NFS v3 ACL management interface */
+	if (!IS_ERR(server))
+		nfs_init_server_aclclient(server);
+	return server;
+}
+
+struct nfs_server *nfs3_clone_server(struct nfs_server *source,
+				     struct nfs_fh *fh,
+				     struct nfs_fattr *fattr,
+				     rpc_authflavor_t flavor)
+{
+	struct nfs_server *server = nfs_clone_server(source, fh, fattr, flavor);
+	if (!IS_ERR(server) && !IS_ERR(source->client_acl))
+		nfs_init_server_aclclient(server);
+	return server;
+}

fs/nfs/nfs3proc.c

@@ -0,0 +1,965 @@
+/*
+ *  linux/fs/nfs/nfs3proc.c
+ *
+ *  Client-side NFSv3 procedures stubs.
+ *
+ *  Copyright (C) 1997, Olaf Kirch
+ */
+
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/slab.h>
+#include <linux/nfs.h>
+#include <linux/nfs3.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_page.h>
+#include <linux/lockd/bind.h>
+#include <linux/nfs_mount.h>
+#include <linux/freezer.h>
+#include <linux/xattr.h>
+
+#include "iostat.h"
+#include "internal.h"
+#include "nfs3_fs.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PROC
+
+/* A wrapper to handle the EJUKEBOX error messages */
+static int
+nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags)
+{
+	int res;
+	do {
+		res = rpc_call_sync(clnt, msg, flags);
+		if (res != -EJUKEBOX)
+			break;
+		freezable_schedule_timeout_killable_unsafe(NFS_JUKEBOX_RETRY_TIME);
+		res = -ERESTARTSYS;
+	} while (!fatal_signal_pending(current));
+	return res;
+}
+
+#define rpc_call_sync(clnt, msg, flags)	nfs3_rpc_wrapper(clnt, msg, flags)
+
+static int
+nfs3_async_handle_jukebox(struct rpc_task *task, struct inode *inode)
+{
+	if (task->tk_status != -EJUKEBOX)
+		return 0;
+	if (task->tk_status == -EJUKEBOX)
+		nfs_inc_stats(inode, NFSIOS_DELAY);
+	task->tk_status = 0;
+	rpc_restart_call(task);
+	rpc_delay(task, NFS_JUKEBOX_RETRY_TIME);
+	return 1;
+}
+
+static int
+do_proc_get_root(struct rpc_clnt *client, struct nfs_fh *fhandle,
+		 struct nfs_fsinfo *info)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_FSINFO],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= info,
+	};
+	int	status;
+
+	dprintk("%s: call  fsinfo\n", __func__);
+	nfs_fattr_init(info->fattr);
+	status = rpc_call_sync(client, &msg, 0);
+	dprintk("%s: reply fsinfo: %d\n", __func__, status);
+	if (status == 0 && !(info->fattr->valid & NFS_ATTR_FATTR)) {
+		msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR];
+		msg.rpc_resp = info->fattr;
+		status = rpc_call_sync(client, &msg, 0);
+		dprintk("%s: reply getattr: %d\n", __func__, status);
+	}
+	return status;
+}
+
+/*
+ * Bare-bones access to getattr: this is for nfs_get_root/nfs_get_sb
+ */
+static int
+nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
+		   struct nfs_fsinfo *info)
+{
+	int	status;
+
+	status = do_proc_get_root(server->client, fhandle, info);
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = do_proc_get_root(server->nfs_client->cl_rpcclient, fhandle, info);
+	return status;
+}
+
+/*
+ * One function for each procedure in the NFS protocol.
+ */
+static int
+nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
+		struct nfs_fattr *fattr, struct nfs4_label *label)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_GETATTR],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= fattr,
+	};
+	int	status;
+
+	dprintk("NFS call  getattr\n");
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply getattr: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
+			struct iattr *sattr)
+{
+	struct inode *inode = dentry->d_inode;
+	struct nfs3_sattrargs	arg = {
+		.fh		= NFS_FH(inode),
+		.sattr		= sattr,
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_SETATTR],
+		.rpc_argp	= &arg,
+		.rpc_resp	= fattr,
+	};
+	int	status;
+
+	dprintk("NFS call  setattr\n");
+	if (sattr->ia_valid & ATTR_FILE)
+		msg.rpc_cred = nfs_file_cred(sattr->ia_file);
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	if (status == 0)
+		nfs_setattr_update_inode(inode, sattr);
+	dprintk("NFS reply setattr: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_lookup(struct inode *dir, struct qstr *name,
+		 struct nfs_fh *fhandle, struct nfs_fattr *fattr,
+		 struct nfs4_label *label)
+{
+	struct nfs3_diropargs	arg = {
+		.fh		= NFS_FH(dir),
+		.name		= name->name,
+		.len		= name->len
+	};
+	struct nfs3_diropres	res = {
+		.fh		= fhandle,
+		.fattr		= fattr
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_LOOKUP],
+		.rpc_argp	= &arg,
+		.rpc_resp	= &res,
+	};
+	int			status;
+
+	dprintk("NFS call  lookup %s\n", name->name);
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		return -ENOMEM;
+
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_refresh_inode(dir, res.dir_attr);
+	if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) {
+		msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR];
+		msg.rpc_argp = fhandle;
+		msg.rpc_resp = fattr;
+		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	}
+	nfs_free_fattr(res.dir_attr);
+	dprintk("NFS reply lookup: %d\n", status);
+	return status;
+}
+
+static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
+{
+	struct nfs3_accessargs	arg = {
+		.fh		= NFS_FH(inode),
+	};
+	struct nfs3_accessres	res;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_ACCESS],
+		.rpc_argp	= &arg,
+		.rpc_resp	= &res,
+		.rpc_cred	= entry->cred,
+	};
+	int mode = entry->mask;
+	int status = -ENOMEM;
+
+	dprintk("NFS call  access\n");
+
+	if (mode & MAY_READ)
+		arg.access |= NFS3_ACCESS_READ;
+	if (S_ISDIR(inode->i_mode)) {
+		if (mode & MAY_WRITE)
+			arg.access |= NFS3_ACCESS_MODIFY | NFS3_ACCESS_EXTEND | NFS3_ACCESS_DELETE;
+		if (mode & MAY_EXEC)
+			arg.access |= NFS3_ACCESS_LOOKUP;
+	} else {
+		if (mode & MAY_WRITE)
+			arg.access |= NFS3_ACCESS_MODIFY | NFS3_ACCESS_EXTEND;
+		if (mode & MAY_EXEC)
+			arg.access |= NFS3_ACCESS_EXECUTE;
+	}
+
+	res.fattr = nfs_alloc_fattr();
+	if (res.fattr == NULL)
+		goto out;
+
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	nfs_refresh_inode(inode, res.fattr);
+	if (status == 0) {
+		entry->mask = 0;
+		if (res.access & NFS3_ACCESS_READ)
+			entry->mask |= MAY_READ;
+		if (res.access & (NFS3_ACCESS_MODIFY | NFS3_ACCESS_EXTEND | NFS3_ACCESS_DELETE))
+			entry->mask |= MAY_WRITE;
+		if (res.access & (NFS3_ACCESS_LOOKUP|NFS3_ACCESS_EXECUTE))
+			entry->mask |= MAY_EXEC;
+	}
+	nfs_free_fattr(res.fattr);
+out:
+	dprintk("NFS reply access: %d\n", status);
+	return status;
+}
+
+static int nfs3_proc_readlink(struct inode *inode, struct page *page,
+		unsigned int pgbase, unsigned int pglen)
+{
+	struct nfs_fattr	*fattr;
+	struct nfs3_readlinkargs args = {
+		.fh		= NFS_FH(inode),
+		.pgbase		= pgbase,
+		.pglen		= pglen,
+		.pages		= &page
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_READLINK],
+		.rpc_argp	= &args,
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  readlink\n");
+	fattr = nfs_alloc_fattr();
+	if (fattr == NULL)
+		goto out;
+	msg.rpc_resp = fattr;
+
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	nfs_refresh_inode(inode, fattr);
+	nfs_free_fattr(fattr);
+out:
+	dprintk("NFS reply readlink: %d\n", status);
+	return status;
+}
+
+struct nfs3_createdata {
+	struct rpc_message msg;
+	union {
+		struct nfs3_createargs create;
+		struct nfs3_mkdirargs mkdir;
+		struct nfs3_symlinkargs symlink;
+		struct nfs3_mknodargs mknod;
+	} arg;
+	struct nfs3_diropres res;
+	struct nfs_fh fh;
+	struct nfs_fattr fattr;
+	struct nfs_fattr dir_attr;
+};
+
+static struct nfs3_createdata *nfs3_alloc_createdata(void)
+{
+	struct nfs3_createdata *data;
+
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (data != NULL) {
+		data->msg.rpc_argp = &data->arg;
+		data->msg.rpc_resp = &data->res;
+		data->res.fh = &data->fh;
+		data->res.fattr = &data->fattr;
+		data->res.dir_attr = &data->dir_attr;
+		nfs_fattr_init(data->res.fattr);
+		nfs_fattr_init(data->res.dir_attr);
+	}
+	return data;
+}
+
+static int nfs3_do_create(struct inode *dir, struct dentry *dentry, struct nfs3_createdata *data)
+{
+	int status;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &data->msg, 0);
+	nfs_post_op_update_inode(dir, data->res.dir_attr);
+	if (status == 0)
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL);
+	return status;
+}
+
+static void nfs3_free_createdata(struct nfs3_createdata *data)
+{
+	kfree(data);
+}
+
+/*
+ * Create a regular file.
+ */
+static int
+nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
+		 int flags)
+{
+	struct posix_acl *default_acl, *acl;
+	struct nfs3_createdata *data;
+	int status = -ENOMEM;
+
+	dprintk("NFS call  create %pd\n", dentry);
+
+	data = nfs3_alloc_createdata();
+	if (data == NULL)
+		goto out;
+
+	data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_CREATE];
+	data->arg.create.fh = NFS_FH(dir);
+	data->arg.create.name = dentry->d_name.name;
+	data->arg.create.len = dentry->d_name.len;
+	data->arg.create.sattr = sattr;
+
+	data->arg.create.createmode = NFS3_CREATE_UNCHECKED;
+	if (flags & O_EXCL) {
+		data->arg.create.createmode  = NFS3_CREATE_EXCLUSIVE;
+		data->arg.create.verifier[0] = cpu_to_be32(jiffies);
+		data->arg.create.verifier[1] = cpu_to_be32(current->pid);
+	}
+
+	status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl);
+	if (status)
+		goto out;
+
+	for (;;) {
+		status = nfs3_do_create(dir, dentry, data);
+
+		if (status != -ENOTSUPP)
+			break;
+		/* If the server doesn't support the exclusive creation
+		 * semantics, try again with simple 'guarded' mode. */
+		switch (data->arg.create.createmode) {
+			case NFS3_CREATE_EXCLUSIVE:
+				data->arg.create.createmode = NFS3_CREATE_GUARDED;
+				break;
+
+			case NFS3_CREATE_GUARDED:
+				data->arg.create.createmode = NFS3_CREATE_UNCHECKED;
+				break;
+
+			case NFS3_CREATE_UNCHECKED:
+				goto out;
+		}
+		nfs_fattr_init(data->res.dir_attr);
+		nfs_fattr_init(data->res.fattr);
+	}
+
+	if (status != 0)
+		goto out_release_acls;
+
+	/* When we created the file with exclusive semantics, make
+	 * sure we set the attributes afterwards. */
+	if (data->arg.create.createmode == NFS3_CREATE_EXCLUSIVE) {
+		dprintk("NFS call  setattr (post-create)\n");
+
+		if (!(sattr->ia_valid & ATTR_ATIME_SET))
+			sattr->ia_valid |= ATTR_ATIME;
+		if (!(sattr->ia_valid & ATTR_MTIME_SET))
+			sattr->ia_valid |= ATTR_MTIME;
+
+		/* Note: we could use a guarded setattr here, but I'm
+		 * not sure this buys us anything (and I'd have
+		 * to revamp the NFSv3 XDR code) */
+		status = nfs3_proc_setattr(dentry, data->res.fattr, sattr);
+		nfs_post_op_update_inode(dentry->d_inode, data->res.fattr);
+		dprintk("NFS reply setattr (post-create): %d\n", status);
+		if (status != 0)
+			goto out_release_acls;
+	}
+
+	status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl);
+
+out_release_acls:
+	posix_acl_release(acl);
+	posix_acl_release(default_acl);
+out:
+	nfs3_free_createdata(data);
+	dprintk("NFS reply create: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_remove(struct inode *dir, struct qstr *name)
+{
+	struct nfs_removeargs arg = {
+		.fh = NFS_FH(dir),
+		.name = *name,
+	};
+	struct nfs_removeres res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE],
+		.rpc_argp = &arg,
+		.rpc_resp = &res,
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  remove %s\n", name->name);
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		goto out;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_post_op_update_inode(dir, res.dir_attr);
+	nfs_free_fattr(res.dir_attr);
+out:
+	dprintk("NFS reply remove: %d\n", status);
+	return status;
+}
+
+static void
+nfs3_proc_unlink_setup(struct rpc_message *msg, struct inode *dir)
+{
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE];
+}
+
+static void nfs3_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data)
+{
+	rpc_call_start(task);
+}
+
+static int
+nfs3_proc_unlink_done(struct rpc_task *task, struct inode *dir)
+{
+	struct nfs_removeres *res;
+	if (nfs3_async_handle_jukebox(task, dir))
+		return 0;
+	res = task->tk_msg.rpc_resp;
+	nfs_post_op_update_inode(dir, res->dir_attr);
+	return 1;
+}
+
+static void
+nfs3_proc_rename_setup(struct rpc_message *msg, struct inode *dir)
+{
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_RENAME];
+}
+
+static void nfs3_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data)
+{
+	rpc_call_start(task);
+}
+
+static int
+nfs3_proc_rename_done(struct rpc_task *task, struct inode *old_dir,
+		      struct inode *new_dir)
+{
+	struct nfs_renameres *res;
+
+	if (nfs3_async_handle_jukebox(task, old_dir))
+		return 0;
+	res = task->tk_msg.rpc_resp;
+
+	nfs_post_op_update_inode(old_dir, res->old_fattr);
+	nfs_post_op_update_inode(new_dir, res->new_fattr);
+	return 1;
+}
+
+static int
+nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
+{
+	struct nfs3_linkargs	arg = {
+		.fromfh		= NFS_FH(inode),
+		.tofh		= NFS_FH(dir),
+		.toname		= name->name,
+		.tolen		= name->len
+	};
+	struct nfs3_linkres	res;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_LINK],
+		.rpc_argp	= &arg,
+		.rpc_resp	= &res,
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  link %s\n", name->name);
+	res.fattr = nfs_alloc_fattr();
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.fattr == NULL || res.dir_attr == NULL)
+		goto out;
+
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	nfs_post_op_update_inode(dir, res.dir_attr);
+	nfs_post_op_update_inode(inode, res.fattr);
+out:
+	nfs_free_fattr(res.dir_attr);
+	nfs_free_fattr(res.fattr);
+	dprintk("NFS reply link: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
+		  unsigned int len, struct iattr *sattr)
+{
+	struct nfs3_createdata *data;
+	int status = -ENOMEM;
+
+	if (len > NFS3_MAXPATHLEN)
+		return -ENAMETOOLONG;
+
+	dprintk("NFS call  symlink %pd\n", dentry);
+
+	data = nfs3_alloc_createdata();
+	if (data == NULL)
+		goto out;
+	data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_SYMLINK];
+	data->arg.symlink.fromfh = NFS_FH(dir);
+	data->arg.symlink.fromname = dentry->d_name.name;
+	data->arg.symlink.fromlen = dentry->d_name.len;
+	data->arg.symlink.pages = &page;
+	data->arg.symlink.pathlen = len;
+	data->arg.symlink.sattr = sattr;
+
+	status = nfs3_do_create(dir, dentry, data);
+
+	nfs3_free_createdata(data);
+out:
+	dprintk("NFS reply symlink: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
+{
+	struct posix_acl *default_acl, *acl;
+	struct nfs3_createdata *data;
+	int status = -ENOMEM;
+
+	dprintk("NFS call  mkdir %pd\n", dentry);
+
+	data = nfs3_alloc_createdata();
+	if (data == NULL)
+		goto out;
+
+	status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl);
+	if (status)
+		goto out;
+
+	data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKDIR];
+	data->arg.mkdir.fh = NFS_FH(dir);
+	data->arg.mkdir.name = dentry->d_name.name;
+	data->arg.mkdir.len = dentry->d_name.len;
+	data->arg.mkdir.sattr = sattr;
+
+	status = nfs3_do_create(dir, dentry, data);
+	if (status != 0)
+		goto out_release_acls;
+
+	status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl);
+
+out_release_acls:
+	posix_acl_release(acl);
+	posix_acl_release(default_acl);
+out:
+	nfs3_free_createdata(data);
+	dprintk("NFS reply mkdir: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_rmdir(struct inode *dir, struct qstr *name)
+{
+	struct nfs_fattr	*dir_attr;
+	struct nfs3_diropargs	arg = {
+		.fh		= NFS_FH(dir),
+		.name		= name->name,
+		.len		= name->len
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_RMDIR],
+		.rpc_argp	= &arg,
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  rmdir %s\n", name->name);
+	dir_attr = nfs_alloc_fattr();
+	if (dir_attr == NULL)
+		goto out;
+
+	msg.rpc_resp = dir_attr;
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_post_op_update_inode(dir, dir_attr);
+	nfs_free_fattr(dir_attr);
+out:
+	dprintk("NFS reply rmdir: %d\n", status);
+	return status;
+}
+
+/*
+ * The READDIR implementation is somewhat hackish - we pass the user buffer
+ * to the encode function, which installs it in the receive iovec.
+ * The decode function itself doesn't perform any decoding, it just makes
+ * sure the reply is syntactically correct.
+ *
+ * Also note that this implementation handles both plain readdir and
+ * readdirplus.
+ */
+static int
+nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
+		  u64 cookie, struct page **pages, unsigned int count, int plus)
+{
+	struct inode		*dir = dentry->d_inode;
+	__be32			*verf = NFS_I(dir)->cookieverf;
+	struct nfs3_readdirargs	arg = {
+		.fh		= NFS_FH(dir),
+		.cookie		= cookie,
+		.verf		= {verf[0], verf[1]},
+		.plus		= plus,
+		.count		= count,
+		.pages		= pages
+	};
+	struct nfs3_readdirres	res = {
+		.verf		= verf,
+		.plus		= plus
+	};
+	struct rpc_message	msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_READDIR],
+		.rpc_argp	= &arg,
+		.rpc_resp	= &res,
+		.rpc_cred	= cred
+	};
+	int status = -ENOMEM;
+
+	if (plus)
+		msg.rpc_proc = &nfs3_procedures[NFS3PROC_READDIRPLUS];
+
+	dprintk("NFS call  readdir%s %d\n",
+			plus? "plus" : "", (unsigned int) cookie);
+
+	res.dir_attr = nfs_alloc_fattr();
+	if (res.dir_attr == NULL)
+		goto out;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+
+	nfs_invalidate_atime(dir);
+	nfs_refresh_inode(dir, res.dir_attr);
+
+	nfs_free_fattr(res.dir_attr);
+out:
+	dprintk("NFS reply readdir%s: %d\n",
+			plus? "plus" : "", status);
+	return status;
+}
+
+static int
+nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
+		dev_t rdev)
+{
+	struct posix_acl *default_acl, *acl;
+	struct nfs3_createdata *data;
+	int status = -ENOMEM;
+
+	dprintk("NFS call  mknod %pd %u:%u\n", dentry,
+			MAJOR(rdev), MINOR(rdev));
+
+	data = nfs3_alloc_createdata();
+	if (data == NULL)
+		goto out;
+
+	status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl);
+	if (status)
+		goto out;
+
+	data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKNOD];
+	data->arg.mknod.fh = NFS_FH(dir);
+	data->arg.mknod.name = dentry->d_name.name;
+	data->arg.mknod.len = dentry->d_name.len;
+	data->arg.mknod.sattr = sattr;
+	data->arg.mknod.rdev = rdev;
+
+	switch (sattr->ia_mode & S_IFMT) {
+	case S_IFBLK:
+		data->arg.mknod.type = NF3BLK;
+		break;
+	case S_IFCHR:
+		data->arg.mknod.type = NF3CHR;
+		break;
+	case S_IFIFO:
+		data->arg.mknod.type = NF3FIFO;
+		break;
+	case S_IFSOCK:
+		data->arg.mknod.type = NF3SOCK;
+		break;
+	default:
+		status = -EINVAL;
+		goto out;
+	}
+
+	status = nfs3_do_create(dir, dentry, data);
+	if (status != 0)
+		goto out_release_acls;
+
+	status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl);
+
+out_release_acls:
+	posix_acl_release(acl);
+	posix_acl_release(default_acl);
+out:
+	nfs3_free_createdata(data);
+	dprintk("NFS reply mknod: %d\n", status);
+	return status;
+}
+
+static int
+nfs3_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
+		 struct nfs_fsstat *stat)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_FSSTAT],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= stat,
+	};
+	int	status;
+
+	dprintk("NFS call  fsstat\n");
+	nfs_fattr_init(stat->fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply fsstat: %d\n", status);
+	return status;
+}
+
+static int
+do_proc_fsinfo(struct rpc_clnt *client, struct nfs_fh *fhandle,
+		 struct nfs_fsinfo *info)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_FSINFO],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= info,
+	};
+	int	status;
+
+	dprintk("NFS call  fsinfo\n");
+	nfs_fattr_init(info->fattr);
+	status = rpc_call_sync(client, &msg, 0);
+	dprintk("NFS reply fsinfo: %d\n", status);
+	return status;
+}
+
+/*
+ * Bare-bones access to fsinfo: this is for nfs_get_root/nfs_get_sb via
+ * nfs_create_server
+ */
+static int
+nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
+		   struct nfs_fsinfo *info)
+{
+	int	status;
+
+	status = do_proc_fsinfo(server->client, fhandle, info);
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = do_proc_fsinfo(server->nfs_client->cl_rpcclient, fhandle, info);
+	return status;
+}
+
+static int
+nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
+		   struct nfs_pathconf *info)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs3_procedures[NFS3PROC_PATHCONF],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= info,
+	};
+	int	status;
+
+	dprintk("NFS call  pathconf\n");
+	nfs_fattr_init(info->fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply pathconf: %d\n", status);
+	return status;
+}
+
+static int nfs3_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
+{
+	struct inode *inode = hdr->inode;
+
+	if (nfs3_async_handle_jukebox(task, inode))
+		return -EAGAIN;
+
+	nfs_invalidate_atime(inode);
+	nfs_refresh_inode(inode, &hdr->fattr);
+	return 0;
+}
+
+static void nfs3_proc_read_setup(struct nfs_pgio_header *hdr,
+				 struct rpc_message *msg)
+{
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ];
+}
+
+static int nfs3_proc_pgio_rpc_prepare(struct rpc_task *task,
+				      struct nfs_pgio_header *hdr)
+{
+	rpc_call_start(task);
+	return 0;
+}
+
+static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
+{
+	struct inode *inode = hdr->inode;
+
+	if (nfs3_async_handle_jukebox(task, inode))
+		return -EAGAIN;
+	if (task->tk_status >= 0)
+		nfs_post_op_update_inode_force_wcc(inode, hdr->res.fattr);
+	return 0;
+}
+
+static void nfs3_proc_write_setup(struct nfs_pgio_header *hdr,
+				  struct rpc_message *msg)
+{
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE];
+}
+
+static void nfs3_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data)
+{
+	rpc_call_start(task);
+}
+
+static int nfs3_commit_done(struct rpc_task *task, struct nfs_commit_data *data)
+{
+	if (nfs3_async_handle_jukebox(task, data->inode))
+		return -EAGAIN;
+	nfs_refresh_inode(data->inode, data->res.fattr);
+	return 0;
+}
+
+static void nfs3_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg)
+{
+	msg->rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT];
+}
+
+static int
+nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
+{
+	struct inode *inode = file_inode(filp);
+
+	return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl);
+}
+
+static int nfs3_have_delegation(struct inode *inode, fmode_t flags)
+{
+	return 0;
+}
+
+static int nfs3_return_delegation(struct inode *inode)
+{
+	nfs_wb_all(inode);
+	return 0;
+}
+
+static const struct inode_operations nfs3_dir_inode_operations = {
+	.create		= nfs_create,
+	.lookup		= nfs_lookup,
+	.link		= nfs_link,
+	.unlink		= nfs_unlink,
+	.symlink	= nfs_symlink,
+	.mkdir		= nfs_mkdir,
+	.rmdir		= nfs_rmdir,
+	.mknod		= nfs_mknod,
+	.rename		= nfs_rename,
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+#ifdef CONFIG_NFS_V3_ACL
+	.listxattr	= nfs3_listxattr,
+	.getxattr	= generic_getxattr,
+	.setxattr	= generic_setxattr,
+	.removexattr	= generic_removexattr,
+	.get_acl	= nfs3_get_acl,
+	.set_acl	= nfs3_set_acl,
+#endif
+};
+
+static const struct inode_operations nfs3_file_inode_operations = {
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+#ifdef CONFIG_NFS_V3_ACL
+	.listxattr	= nfs3_listxattr,
+	.getxattr	= generic_getxattr,
+	.setxattr	= generic_setxattr,
+	.removexattr	= generic_removexattr,
+	.get_acl	= nfs3_get_acl,
+	.set_acl	= nfs3_set_acl,
+#endif
+};
+
+const struct nfs_rpc_ops nfs_v3_clientops = {
+	.version	= 3,			/* protocol version */
+	.dentry_ops	= &nfs_dentry_operations,
+	.dir_inode_ops	= &nfs3_dir_inode_operations,
+	.file_inode_ops	= &nfs3_file_inode_operations,
+	.file_ops	= &nfs_file_operations,
+	.getroot	= nfs3_proc_get_root,
+	.submount	= nfs_submount,
+	.try_mount	= nfs_try_mount,
+	.getattr	= nfs3_proc_getattr,
+	.setattr	= nfs3_proc_setattr,
+	.lookup		= nfs3_proc_lookup,
+	.access		= nfs3_proc_access,
+	.readlink	= nfs3_proc_readlink,
+	.create		= nfs3_proc_create,
+	.remove		= nfs3_proc_remove,
+	.unlink_setup	= nfs3_proc_unlink_setup,
+	.unlink_rpc_prepare = nfs3_proc_unlink_rpc_prepare,
+	.unlink_done	= nfs3_proc_unlink_done,
+	.rename_setup	= nfs3_proc_rename_setup,
+	.rename_rpc_prepare = nfs3_proc_rename_rpc_prepare,
+	.rename_done	= nfs3_proc_rename_done,
+	.link		= nfs3_proc_link,
+	.symlink	= nfs3_proc_symlink,
+	.mkdir		= nfs3_proc_mkdir,
+	.rmdir		= nfs3_proc_rmdir,
+	.readdir	= nfs3_proc_readdir,
+	.mknod		= nfs3_proc_mknod,
+	.statfs		= nfs3_proc_statfs,
+	.fsinfo		= nfs3_proc_fsinfo,
+	.pathconf	= nfs3_proc_pathconf,
+	.decode_dirent	= nfs3_decode_dirent,
+	.pgio_rpc_prepare = nfs3_proc_pgio_rpc_prepare,
+	.read_setup	= nfs3_proc_read_setup,
+	.read_done	= nfs3_read_done,
+	.write_setup	= nfs3_proc_write_setup,
+	.write_done	= nfs3_write_done,
+	.commit_setup	= nfs3_proc_commit_setup,
+	.commit_rpc_prepare = nfs3_proc_commit_rpc_prepare,
+	.commit_done	= nfs3_commit_done,
+	.lock		= nfs3_proc_lock,
+	.clear_acl_cache = forget_all_cached_acls,
+	.close_context	= nfs_close_context,
+	.have_delegation = nfs3_have_delegation,
+	.return_delegation = nfs3_return_delegation,
+	.alloc_client	= nfs_alloc_client,
+	.init_client	= nfs_init_client,
+	.free_client	= nfs_free_client,
+	.create_server	= nfs3_create_server,
+	.clone_server	= nfs3_clone_server,
+};

fs/nfs/nfs3super.c

@@ -0,0 +1,35 @@
+/*
+ * Copyright (c) 2012 Netapp, Inc. All rights reserved.
+ */
+#include <linux/module.h>
+#include <linux/nfs_fs.h>
+#include "internal.h"
+#include "nfs3_fs.h"
+#include "nfs.h"
+
+static struct nfs_subversion nfs_v3 = {
+	.owner = THIS_MODULE,
+	.nfs_fs   = &nfs_fs_type,
+	.rpc_vers = &nfs_version3,
+	.rpc_ops  = &nfs_v3_clientops,
+	.sops     = &nfs_sops,
+#ifdef CONFIG_NFS_V3_ACL
+	.xattr    = nfs3_xattr_handlers,
+#endif
+};
+
+static int __init init_nfs_v3(void)
+{
+	register_nfs_version(&nfs_v3);
+	return 0;
+}
+
+static void __exit exit_nfs_v3(void)
+{
+	unregister_nfs_version(&nfs_v3);
+}
+
+MODULE_LICENSE("GPL");
+
+module_init(init_nfs_v3);
+module_exit(exit_nfs_v3);

fs/nfs/nfs42.h

@@ -0,0 +1,14 @@
+/*
+ * Copyright (c) 2014 Anna Schumaker <Anna.Schumaker@Netapp.com>
+ */
+
+#ifndef __LINUX_FS_NFS_NFS4_2_H
+#define __LINUX_FS_NFS_NFS4_2_H
+
+/* nfs4.2proc.c */
+loff_t nfs42_proc_llseek(struct file *, loff_t, int);
+
+/* nfs4.2xdr.h */
+extern struct rpc_procinfo nfs4_2_procedures[];
+
+#endif /* __LINUX_FS_NFS_NFS4_2_H */

fs/nfs/nfs42proc.c

@@ -0,0 +1,69 @@
+/*
+ * Copyright (c) 2014 Anna Schumaker <Anna.Schumaker@Netapp.com>
+ */
+#include <linux/fs.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/nfs.h>
+#include <linux/nfs3.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_xdr.h>
+#include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
+#include "nfs42.h"
+
+static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
+				fmode_t fmode)
+{
+	struct nfs_open_context *open;
+	struct nfs_lock_context *lock;
+	int ret;
+
+	open = get_nfs_open_context(nfs_file_open_context(file));
+	lock = nfs_get_lock_context(open);
+	if (IS_ERR(lock)) {
+		put_nfs_open_context(open);
+		return PTR_ERR(lock);
+	}
+
+	ret = nfs4_set_rw_stateid(dst, open, lock, fmode);
+
+	nfs_put_lock_context(lock);
+	put_nfs_open_context(open);
+	return ret;
+}
+
+loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+{
+	struct inode *inode = file_inode(filep);
+	struct nfs42_seek_args args = {
+		.sa_fh		= NFS_FH(inode),
+		.sa_offset	= offset,
+		.sa_what	= (whence == SEEK_HOLE) ?
+					NFS4_CONTENT_HOLE : NFS4_CONTENT_DATA,
+	};
+	struct nfs42_seek_res res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEEK],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	struct nfs_server *server = NFS_SERVER(inode);
+	int status;
+
+	if (!(server->caps & NFS_CAP_SEEK))
+		return -ENOTSUPP;
+
+	status = nfs42_set_rw_stateid(&args.sa_stateid, filep, FMODE_READ);
+	if (status)
+		return status;
+
+	nfs_wb_all(inode);
+	status = nfs4_call_sync(server->client, server, &msg,
+				&args.seq_args, &res.seq_res, 0);
+	if (status == -ENOTSUPP)
+		server->caps &= ~NFS_CAP_SEEK;
+	if (status)
+		return status;
+
+	return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes);
+}

fs/nfs/nfs42xdr.c

@@ -0,0 +1,98 @@
+/*
+ * Copyright (c) 2014 Anna Schumaker <Anna.Schumaker@Netapp.com>
+ */
+#ifndef __LINUX_FS_NFS_NFS4_2XDR_H
+#define __LINUX_FS_NFS_NFS4_2XDR_H
+
+#define encode_seek_maxsz		(op_encode_hdr_maxsz + \
+					 encode_stateid_maxsz + \
+					 2 /* offset */ + \
+					 1 /* whence */)
+#define decode_seek_maxsz		(op_decode_hdr_maxsz + \
+					 1 /* eof */ + \
+					 1 /* whence */ + \
+					 2 /* offset */ + \
+					 2 /* length */)
+
+#define NFS4_enc_seek_sz		(compound_encode_hdr_maxsz + \
+					 encode_putfh_maxsz + \
+					 encode_seek_maxsz)
+#define NFS4_dec_seek_sz		(compound_decode_hdr_maxsz + \
+					 decode_putfh_maxsz + \
+					 decode_seek_maxsz)
+
+
+static void encode_seek(struct xdr_stream *xdr,
+			struct nfs42_seek_args *args,
+			struct compound_hdr *hdr)
+{
+	encode_op_hdr(xdr, OP_SEEK, decode_seek_maxsz, hdr);
+	encode_nfs4_stateid(xdr, &args->sa_stateid);
+	encode_uint64(xdr, args->sa_offset);
+	encode_uint32(xdr, args->sa_what);
+}
+
+/*
+ * Encode SEEK request
+ */
+static void nfs4_xdr_enc_seek(struct rpc_rqst *req,
+			      struct xdr_stream *xdr,
+			      struct nfs42_seek_args *args)
+{
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_putfh(xdr, args->sa_fh, &hdr);
+	encode_seek(xdr, args, &hdr);
+	encode_nops(&hdr);
+}
+
+static int decode_seek(struct xdr_stream *xdr, struct nfs42_seek_res *res)
+{
+	int status;
+	__be32 *p;
+
+	status = decode_op_hdr(xdr, OP_SEEK);
+	if (status)
+		return status;
+
+	p = xdr_inline_decode(xdr, 4 + 8);
+	if (unlikely(!p))
+		goto out_overflow;
+
+	res->sr_eof = be32_to_cpup(p++);
+	p = xdr_decode_hyper(p, &res->sr_offset);
+	return 0;
+
+out_overflow:
+	print_overflow_msg(__func__, xdr);
+	return -EIO;
+}
+
+/*
+ * Decode SEEK request
+ */
+static int nfs4_xdr_dec_seek(struct rpc_rqst *rqstp,
+			     struct xdr_stream *xdr,
+			     struct nfs42_seek_res *res)
+{
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_putfh(xdr);
+	if (status)
+		goto out;
+	status = decode_seek(xdr, res);
+out:
+	return status;
+}
+#endif /* __LINUX_FS_NFS_NFS4_2XDR_H */

fs/nfs/nfs4_fs.h

@@ -0,0 +1,519 @@
+/*
+ * linux/fs/nfs/nfs4_fs.h
+ *
+ * Copyright (C) 2005 Trond Myklebust
+ *
+ * NFSv4-specific filesystem definitions and declarations
+ */
+
+#ifndef __LINUX_FS_NFS_NFS4_FS_H
+#define __LINUX_FS_NFS_NFS4_FS_H
+
+#if defined(CONFIG_NFS_V4_2)
+#define NFS4_MAX_MINOR_VERSION 2
+#elif defined(CONFIG_NFS_V4_1)
+#define NFS4_MAX_MINOR_VERSION 1
+#else
+#define NFS4_MAX_MINOR_VERSION 0
+#endif
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+
+#define NFS4_MAX_LOOP_ON_RECOVER (10)
+
+#include <linux/seqlock.h>
+
+struct idmap;
+
+enum nfs4_client_state {
+	NFS4CLNT_MANAGER_RUNNING  = 0,
+	NFS4CLNT_CHECK_LEASE,
+	NFS4CLNT_LEASE_EXPIRED,
+	NFS4CLNT_RECLAIM_REBOOT,
+	NFS4CLNT_RECLAIM_NOGRACE,
+	NFS4CLNT_DELEGRETURN,
+	NFS4CLNT_SESSION_RESET,
+	NFS4CLNT_LEASE_CONFIRM,
+	NFS4CLNT_SERVER_SCOPE_MISMATCH,
+	NFS4CLNT_PURGE_STATE,
+	NFS4CLNT_BIND_CONN_TO_SESSION,
+	NFS4CLNT_MOVED,
+	NFS4CLNT_LEASE_MOVED,
+};
+
+#define NFS4_RENEW_TIMEOUT		0x01
+#define NFS4_RENEW_DELEGATION_CB	0x02
+
+struct nfs4_minor_version_ops {
+	u32	minor_version;
+	unsigned init_caps;
+
+	int	(*init_client)(struct nfs_client *);
+	void	(*shutdown_client)(struct nfs_client *);
+	bool	(*match_stateid)(const nfs4_stateid *,
+			const nfs4_stateid *);
+	int	(*find_root_sec)(struct nfs_server *, struct nfs_fh *,
+			struct nfs_fsinfo *);
+	void	(*free_lock_state)(struct nfs_server *,
+			struct nfs4_lock_state *);
+	const struct rpc_call_ops *call_sync_ops;
+	const struct nfs4_state_recovery_ops *reboot_recovery_ops;
+	const struct nfs4_state_recovery_ops *nograce_recovery_ops;
+	const struct nfs4_state_maintenance_ops *state_renewal_ops;
+	const struct nfs4_mig_recovery_ops *mig_recovery_ops;
+};
+
+#define NFS_SEQID_CONFIRMED 1
+struct nfs_seqid_counter {
+	ktime_t create_time;
+	int owner_id;
+	int flags;
+	u32 counter;
+	spinlock_t lock;		/* Protects the list */
+	struct list_head list;		/* Defines sequence of RPC calls */
+	struct rpc_wait_queue	wait;	/* RPC call delay queue */
+};
+
+struct nfs_seqid {
+	struct nfs_seqid_counter *sequence;
+	struct list_head list;
+	struct rpc_task *task;
+};
+
+static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status)
+{
+	if (seqid_mutating_err(-status))
+		seqid->flags |= NFS_SEQID_CONFIRMED;
+}
+
+/*
+ * NFS4 state_owners and lock_owners are simply labels for ordered
+ * sequences of RPC calls. Their sole purpose is to provide once-only
+ * semantics by allowing the server to identify replayed requests.
+ */
+struct nfs4_state_owner {
+	struct nfs_server    *so_server;
+	struct list_head     so_lru;
+	unsigned long        so_expires;
+	struct rb_node	     so_server_node;
+
+	struct rpc_cred	     *so_cred;	 /* Associated cred */
+
+	spinlock_t	     so_lock;
+	atomic_t	     so_count;
+	unsigned long	     so_flags;
+	struct list_head     so_states;
+	struct nfs_seqid_counter so_seqid;
+	seqcount_t	     so_reclaim_seqcount;
+	struct mutex	     so_delegreturn_mutex;
+};
+
+enum {
+	NFS_OWNER_RECLAIM_REBOOT,
+	NFS_OWNER_RECLAIM_NOGRACE
+};
+
+#define NFS_LOCK_NEW		0
+#define NFS_LOCK_RECLAIM	1
+#define NFS_LOCK_EXPIRED	2
+
+/*
+ * struct nfs4_state maintains the client-side state for a given
+ * (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
+ *
+ * OPEN:
+ * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server,
+ * we need to know how many files are open for reading or writing on a
+ * given inode. This information too is stored here.
+ *
+ * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN)
+ */
+
+struct nfs4_lock_state {
+	struct list_head	ls_locks;	/* Other lock stateids */
+	struct nfs4_state *	ls_state;	/* Pointer to open state */
+#define NFS_LOCK_INITIALIZED 0
+#define NFS_LOCK_LOST        1
+	unsigned long		ls_flags;
+	struct nfs_seqid_counter	ls_seqid;
+	nfs4_stateid		ls_stateid;
+	atomic_t		ls_count;
+	fl_owner_t		ls_owner;
+};
+
+/* bits for nfs4_state->flags */
+enum {
+	LK_STATE_IN_USE,
+	NFS_DELEGATED_STATE,		/* Current stateid is delegation */
+	NFS_OPEN_STATE,			/* OPEN stateid is set */
+	NFS_O_RDONLY_STATE,		/* OPEN stateid has read-only state */
+	NFS_O_WRONLY_STATE,		/* OPEN stateid has write-only state */
+	NFS_O_RDWR_STATE,		/* OPEN stateid has read/write state */
+	NFS_STATE_RECLAIM_REBOOT,	/* OPEN stateid server rebooted */
+	NFS_STATE_RECLAIM_NOGRACE,	/* OPEN stateid needs to recover state */
+	NFS_STATE_POSIX_LOCKS,		/* Posix locks are supported */
+	NFS_STATE_RECOVERY_FAILED,	/* OPEN stateid state recovery failed */
+};
+
+struct nfs4_state {
+	struct list_head open_states;	/* List of states for the same state_owner */
+	struct list_head inode_states;	/* List of states for the same inode */
+	struct list_head lock_states;	/* List of subservient lock stateids */
+
+	struct nfs4_state_owner *owner;	/* Pointer to the open owner */
+	struct inode *inode;		/* Pointer to the inode */
+
+	unsigned long flags;		/* Do we hold any locks? */
+	spinlock_t state_lock;		/* Protects the lock_states list */
+
+	seqlock_t seqlock;		/* Protects the stateid/open_stateid */
+	nfs4_stateid stateid;		/* Current stateid: may be delegation */
+	nfs4_stateid open_stateid;	/* OPEN stateid */
+
+	/* The following 3 fields are protected by owner->so_lock */
+	unsigned int n_rdonly;		/* Number of read-only references */
+	unsigned int n_wronly;		/* Number of write-only references */
+	unsigned int n_rdwr;		/* Number of read/write references */
+	fmode_t state;			/* State on the server (R,W, or RW) */
+	atomic_t count;
+};
+
+
+struct nfs4_exception {
+	long timeout;
+	int retry;
+	struct nfs4_state *state;
+	struct inode *inode;
+};
+
+struct nfs4_state_recovery_ops {
+	int owner_flag_bit;
+	int state_flag_bit;
+	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
+	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
+	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
+	int (*reclaim_complete)(struct nfs_client *, struct rpc_cred *);
+	int (*detect_trunking)(struct nfs_client *, struct nfs_client **,
+		struct rpc_cred *);
+};
+
+struct nfs4_state_maintenance_ops {
+	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *, unsigned);
+	struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *);
+	int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
+};
+
+struct nfs4_mig_recovery_ops {
+	int (*get_locations)(struct inode *, struct nfs4_fs_locations *,
+		struct page *, struct rpc_cred *);
+	int (*fsid_present)(struct inode *, struct rpc_cred *);
+};
+
+extern const struct dentry_operations nfs4_dentry_operations;
+
+/* dir.c */
+int nfs_atomic_open(struct inode *, struct dentry *, struct file *,
+		    unsigned, umode_t, int *);
+
+/* super.c */
+extern struct file_system_type nfs4_fs_type;
+
+/* nfs4namespace.c */
+struct rpc_clnt *nfs4_negotiate_security(struct rpc_clnt *, struct inode *, struct qstr *);
+struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *,
+			       struct nfs_fh *, struct nfs_fattr *);
+int nfs4_replace_transport(struct nfs_server *server,
+				const struct nfs4_fs_locations *locations);
+
+/* nfs4proc.c */
+extern int nfs4_call_sync(struct rpc_clnt *, struct nfs_server *,
+			  struct rpc_message *, struct nfs4_sequence_args *,
+			  struct nfs4_sequence_res *, int);
+extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *);
+extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *);
+extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *, bool);
+extern int nfs4_proc_bind_conn_to_session(struct nfs_client *, struct rpc_cred *cred);
+extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
+extern int nfs4_destroy_clientid(struct nfs_client *clp);
+extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
+extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait);
+extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle);
+extern int nfs4_proc_fs_locations(struct rpc_clnt *, struct inode *, const struct qstr *,
+				  struct nfs4_fs_locations *, struct page *);
+extern int nfs4_proc_get_locations(struct inode *, struct nfs4_fs_locations *,
+		struct page *page, struct rpc_cred *);
+extern int nfs4_proc_fsid_present(struct inode *, struct rpc_cred *);
+extern struct rpc_clnt *nfs4_proc_lookup_mountpoint(struct inode *, struct qstr *,
+			    struct nfs_fh *, struct nfs_fattr *);
+extern int nfs4_proc_secinfo(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *);
+extern const struct xattr_handler *nfs4_xattr_handlers[];
+extern int nfs4_set_rw_stateid(nfs4_stateid *stateid,
+		const struct nfs_open_context *ctx,
+		const struct nfs_lock_context *l_ctx,
+		fmode_t fmode);
+
+#if defined(CONFIG_NFS_V4_1)
+static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
+{
+	return server->nfs_client->cl_session;
+}
+
+extern int nfs41_setup_sequence(struct nfs4_session *session,
+		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+		struct rpc_task *task);
+extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *);
+extern int nfs4_proc_create_session(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_destroy_session(struct nfs4_session *, struct rpc_cred *);
+extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
+		struct nfs_fsinfo *fsinfo);
+extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data,
+				  bool sync);
+
+static inline bool
+is_ds_only_client(struct nfs_client *clp)
+{
+	return (clp->cl_exchange_flags & EXCHGID4_FLAG_MASK_PNFS) ==
+		EXCHGID4_FLAG_USE_PNFS_DS;
+}
+
+static inline bool
+is_ds_client(struct nfs_client *clp)
+{
+	return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS;
+}
+
+static inline bool
+_nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode,
+		    struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+	struct rpc_cred *newcred = NULL;
+	rpc_authflavor_t flavor;
+
+	if (test_bit(sp4_mode, &clp->cl_sp4_flags)) {
+		spin_lock(&clp->cl_lock);
+		if (clp->cl_machine_cred != NULL)
+			/* don't call get_rpccred on the machine cred -
+			 * a reference will be held for life of clp */
+			newcred = clp->cl_machine_cred;
+		spin_unlock(&clp->cl_lock);
+		msg->rpc_cred = newcred;
+
+		flavor = clp->cl_rpcclient->cl_auth->au_flavor;
+		WARN_ON_ONCE(flavor != RPC_AUTH_GSS_KRB5I &&
+			     flavor != RPC_AUTH_GSS_KRB5P);
+		*clntp = clp->cl_rpcclient;
+
+		return true;
+	}
+	return false;
+}
+
+/*
+ * Function responsible for determining if an rpc_message should use the
+ * machine cred under SP4_MACH_CRED and if so switching the credential and
+ * authflavor (using the nfs_client's rpc_clnt which will be krb5i/p).
+ * Should be called before rpc_call_sync/rpc_call_async.
+ */
+static inline void
+nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode,
+		   struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+	_nfs4_state_protect(clp, sp4_mode, clntp, msg);
+}
+
+/*
+ * Special wrapper to nfs4_state_protect for write.
+ * If WRITE can use machine cred but COMMIT cannot, make sure all writes
+ * that use machine cred use NFS_FILE_SYNC.
+ */
+static inline void
+nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
+			 struct rpc_message *msg, struct nfs_pgio_header *hdr)
+{
+	if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) &&
+	    !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags))
+		hdr->args.stable = NFS_FILE_SYNC;
+}
+#else /* CONFIG_NFS_v4_1 */
+static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
+{
+	return NULL;
+}
+
+static inline bool
+is_ds_only_client(struct nfs_client *clp)
+{
+	return false;
+}
+
+static inline bool
+is_ds_client(struct nfs_client *clp)
+{
+	return false;
+}
+
+static inline void
+nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_flags,
+		   struct rpc_clnt **clntp, struct rpc_message *msg)
+{
+}
+
+static inline void
+nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp,
+			 struct rpc_message *msg, struct nfs_pgio_header *hdr)
+{
+}
+#endif /* CONFIG_NFS_V4_1 */
+
+extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[];
+
+extern const u32 nfs4_fattr_bitmap[3];
+extern const u32 nfs4_statfs_bitmap[3];
+extern const u32 nfs4_pathconf_bitmap[3];
+extern const u32 nfs4_fsinfo_bitmap[3];
+extern const u32 nfs4_fs_locations_bitmap[3];
+
+void nfs40_shutdown_client(struct nfs_client *);
+void nfs41_shutdown_client(struct nfs_client *);
+int nfs40_init_client(struct nfs_client *);
+int nfs41_init_client(struct nfs_client *);
+void nfs4_free_client(struct nfs_client *);
+
+struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *);
+
+/* nfs4renewd.c */
+extern void nfs4_schedule_state_renewal(struct nfs_client *);
+extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
+extern void nfs4_kill_renewd(struct nfs_client *);
+extern void nfs4_renew_state(struct work_struct *);
+
+/* nfs4state.c */
+struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp);
+struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
+struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
+int nfs4_discover_server_trunking(struct nfs_client *clp,
+			struct nfs_client **);
+int nfs40_discover_server_trunking(struct nfs_client *clp,
+			struct nfs_client **, struct rpc_cred *);
+#if defined(CONFIG_NFS_V4_1)
+int nfs41_discover_server_trunking(struct nfs_client *clp,
+			struct nfs_client **, struct rpc_cred *);
+extern void nfs4_schedule_session_recovery(struct nfs4_session *, int);
+extern void nfs41_server_notify_target_slotid_update(struct nfs_client *clp);
+extern void nfs41_server_notify_highest_slotid_update(struct nfs_client *clp);
+
+#else
+static inline void nfs4_schedule_session_recovery(struct nfs4_session *session, int err)
+{
+}
+#endif /* CONFIG_NFS_V4_1 */
+
+extern struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *, gfp_t);
+extern void nfs4_put_state_owner(struct nfs4_state_owner *);
+extern void nfs4_purge_state_owners(struct nfs_server *);
+extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *);
+extern void nfs4_put_open_state(struct nfs4_state *);
+extern void nfs4_close_state(struct nfs4_state *, fmode_t);
+extern void nfs4_close_sync(struct nfs4_state *, fmode_t);
+extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
+extern void nfs_inode_find_state_and_recover(struct inode *inode,
+		const nfs4_stateid *stateid);
+extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *, struct nfs4_state *);
+extern void nfs4_schedule_lease_recovery(struct nfs_client *);
+extern int nfs4_wait_clnt_recover(struct nfs_client *clp);
+extern int nfs4_client_recover_expired_lease(struct nfs_client *clp);
+extern void nfs4_schedule_state_manager(struct nfs_client *);
+extern void nfs4_schedule_path_down_recovery(struct nfs_client *clp);
+extern int nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *);
+extern int nfs4_schedule_migration_recovery(const struct nfs_server *);
+extern void nfs4_schedule_lease_moved_recovery(struct nfs_client *);
+extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
+extern void nfs41_handle_server_scope(struct nfs_client *,
+				      struct nfs41_server_scope **);
+extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
+extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
+extern int nfs4_select_rw_stateid(nfs4_stateid *, struct nfs4_state *,
+		fmode_t, const struct nfs_lockowner *);
+
+extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask);
+extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
+extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_release_seqid(struct nfs_seqid *seqid);
+extern void nfs_free_seqid(struct nfs_seqid *seqid);
+
+extern void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp);
+
+extern const nfs4_stateid zero_stateid;
+
+/* nfs4super.c */
+struct nfs_mount_info;
+extern struct nfs_subversion nfs_v4;
+struct dentry *nfs4_try_mount(int, const char *, struct nfs_mount_info *, struct nfs_subversion *);
+extern bool nfs4_disable_idmapping;
+extern unsigned short max_session_slots;
+extern unsigned short send_implementation_id;
+extern bool recover_lost_locks;
+
+#define NFS4_CLIENT_ID_UNIQ_LEN		(64)
+extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN];
+
+/* nfs4sysctl.c */
+#ifdef CONFIG_SYSCTL
+int nfs4_register_sysctl(void);
+void nfs4_unregister_sysctl(void);
+#else
+static inline int nfs4_register_sysctl(void)
+{
+	return 0;
+}
+
+static inline void nfs4_unregister_sysctl(void)
+{
+}
+#endif
+
+/* nfs4xdr.c */
+extern struct rpc_procinfo nfs4_procedures[];
+
+struct nfs4_mount_data;
+
+/* callback_xdr.c */
+extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
+
+static inline void nfs4_stateid_copy(nfs4_stateid *dst, const nfs4_stateid *src)
+{
+	memcpy(dst, src, sizeof(*dst));
+}
+
+static inline bool nfs4_stateid_match(const nfs4_stateid *dst, const nfs4_stateid *src)
+{
+	return memcmp(dst, src, sizeof(*dst)) == 0;
+}
+
+static inline bool nfs4_stateid_match_other(const nfs4_stateid *dst, const nfs4_stateid *src)
+{
+	return memcmp(dst->other, src->other, NFS4_STATEID_OTHER_SIZE) == 0;
+}
+
+static inline bool nfs4_stateid_is_newer(const nfs4_stateid *s1, const nfs4_stateid *s2)
+{
+	return (s32)(be32_to_cpu(s1->seqid) - be32_to_cpu(s2->seqid)) > 0;
+}
+
+static inline bool nfs4_valid_open_stateid(const struct nfs4_state *state)
+{
+	return test_bit(NFS_STATE_RECOVERY_FAILED, &state->flags) == 0;
+}
+
+#else
+
+#define nfs4_close_state(a, b) do { } while (0)
+#define nfs4_close_sync(a, b) do { } while (0)
+#define nfs4_state_protect(a, b, c, d) do { } while (0)
+#define nfs4_state_protect_write(a, b, c, d) do { } while (0)
+
+#endif /* CONFIG_NFS_V4 */
+#endif /* __LINUX_FS_NFS_NFS4_FS.H */

fs/nfs/nfs4file.c

@@ -0,0 +1,160 @@
+/*
+ *  linux/fs/nfs/file.c
+ *
+ *  Copyright (C) 1992  Rick Sladkey
+ */
+#include <linux/nfs_fs.h>
+#include "internal.h"
+#include "fscache.h"
+#include "pnfs.h"
+
+#ifdef CONFIG_NFS_V4_2
+#include "nfs42.h"
+#endif
+
+#define NFSDBG_FACILITY		NFSDBG_FILE
+
+static int
+nfs4_file_open(struct inode *inode, struct file *filp)
+{
+	struct nfs_open_context *ctx;
+	struct dentry *dentry = filp->f_path.dentry;
+	struct dentry *parent = NULL;
+	struct inode *dir;
+	unsigned openflags = filp->f_flags;
+	struct iattr attr;
+	int opened = 0;
+	int err;
+
+	/*
+	 * If no cached dentry exists or if it's negative, NFSv4 handled the
+	 * opens in ->lookup() or ->create().
+	 *
+	 * We only get this far for a cached positive dentry.  We skipped
+	 * revalidation, so handle it here by dropping the dentry and returning
+	 * -EOPENSTALE.  The VFS will retry the lookup/create/open.
+	 */
+
+	dprintk("NFS: open file(%pd2)\n", dentry);
+
+	if ((openflags & O_ACCMODE) == 3)
+		openflags--;
+
+	/* We can't create new files here */
+	openflags &= ~(O_CREAT|O_EXCL);
+
+	parent = dget_parent(dentry);
+	dir = parent->d_inode;
+
+	ctx = alloc_nfs_open_context(filp->f_path.dentry, filp->f_mode);
+	err = PTR_ERR(ctx);
+	if (IS_ERR(ctx))
+		goto out;
+
+	attr.ia_valid = ATTR_OPEN;
+	if (openflags & O_TRUNC) {
+		attr.ia_valid |= ATTR_SIZE;
+		attr.ia_size = 0;
+		nfs_wb_all(inode);
+	}
+
+	inode = NFS_PROTO(dir)->open_context(dir, ctx, openflags, &attr, &opened);
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
+		switch (err) {
+		case -EPERM:
+		case -EACCES:
+		case -EDQUOT:
+		case -ENOSPC:
+		case -EROFS:
+			goto out_put_ctx;
+		default:
+			goto out_drop;
+		}
+	}
+	if (inode != dentry->d_inode)
+		goto out_drop;
+
+	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+	nfs_file_set_open_context(filp, ctx);
+	nfs_fscache_open_file(inode, filp);
+	err = 0;
+
+out_put_ctx:
+	put_nfs_open_context(ctx);
+out:
+	dput(parent);
+	return err;
+
+out_drop:
+	d_drop(dentry);
+	err = -EOPENSTALE;
+	goto out_put_ctx;
+}
+
+static int
+nfs4_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
+{
+	int ret;
+	struct inode *inode = file_inode(file);
+
+	do {
+		ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+		if (ret != 0)
+			break;
+		mutex_lock(&inode->i_mutex);
+		ret = nfs_file_fsync_commit(file, start, end, datasync);
+		if (!ret)
+			ret = pnfs_layoutcommit_inode(inode, true);
+		mutex_unlock(&inode->i_mutex);
+		/*
+		 * If nfs_file_fsync_commit detected a server reboot, then
+		 * resend all dirty pages that might have been covered by
+		 * the NFS_CONTEXT_RESEND_WRITES flag
+		 */
+		start = 0;
+		end = LLONG_MAX;
+	} while (ret == -EAGAIN);
+
+	return ret;
+}
+
+#ifdef CONFIG_NFS_V4_2
+static loff_t nfs4_file_llseek(struct file *filep, loff_t offset, int whence)
+{
+	loff_t ret;
+
+	switch (whence) {
+	case SEEK_HOLE:
+	case SEEK_DATA:
+		ret = nfs42_proc_llseek(filep, offset, whence);
+		if (ret != -ENOTSUPP)
+			return ret;
+	default:
+		return nfs_file_llseek(filep, offset, whence);
+	}
+}
+#endif /* CONFIG_NFS_V4_2 */
+
+const struct file_operations nfs4_file_operations = {
+#ifdef CONFIG_NFS_V4_2
+	.llseek		= nfs4_file_llseek,
+#else
+	.llseek		= nfs_file_llseek,
+#endif
+	.read		= new_sync_read,
+	.write		= new_sync_write,
+	.read_iter	= nfs_file_read,
+	.write_iter	= nfs_file_write,
+	.mmap		= nfs_file_mmap,
+	.open		= nfs4_file_open,
+	.flush		= nfs_file_flush,
+	.release	= nfs_file_release,
+	.fsync		= nfs4_file_fsync,
+	.lock		= nfs_lock,
+	.flock		= nfs_flock,
+	.splice_read	= nfs_file_splice_read,
+	.splice_write	= iter_file_splice_write,
+	.check_flags	= nfs_check_flags,
+	.setlease	= simple_nosetlease,
+};

fs/nfs/nfs4getroot.c

@@ -0,0 +1,50 @@
+/*
+* Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+* Written by David Howells (dhowells@redhat.com)
+*/
+
+#include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_CLIENT
+
+int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe)
+{
+	struct nfs_fsinfo fsinfo;
+	int ret = -ENOMEM;
+
+	dprintk("--> nfs4_get_rootfh()\n");
+
+	fsinfo.fattr = nfs_alloc_fattr();
+	if (fsinfo.fattr == NULL)
+		goto out;
+
+	/* Start by getting the root filehandle from the server */
+	ret = nfs4_proc_get_rootfh(server, mntfh, &fsinfo, auth_probe);
+	if (ret < 0) {
+		dprintk("nfs4_get_rootfh: getroot error = %d\n", -ret);
+		goto out;
+	}
+
+	if (!(fsinfo.fattr->valid & NFS_ATTR_FATTR_TYPE)
+			|| !S_ISDIR(fsinfo.fattr->mode)) {
+		printk(KERN_ERR "nfs4_get_rootfh:"
+		       " getroot encountered non-directory\n");
+		ret = -ENOTDIR;
+		goto out;
+	}
+
+	if (fsinfo.fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		printk(KERN_ERR "nfs4_get_rootfh:"
+		       " getroot obtained referral\n");
+		ret = -EREMOTE;
+		goto out;
+	}
+
+	memcpy(&server->fsid, &fsinfo.fattr->fsid, sizeof(server->fsid));
+out:
+	nfs_free_fattr(fsinfo.fattr);
+	dprintk("<-- nfs4_get_rootfh() = %d\n", ret);
+	return ret;
+}

fs/nfs/nfs4namespace.c

@@ -0,0 +1,523 @@
+/*
+ * linux/fs/nfs/nfs4namespace.c
+ *
+ * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com>
+ * - Modified by David Howells <dhowells@redhat.com>
+ *
+ * NFSv4 namespace
+ */
+
+#include <linux/dcache.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/vfs.h>
+#include <linux/inet.h>
+#include "internal.h"
+#include "nfs4_fs.h"
+#include "dns_resolve.h"
+
+#define NFSDBG_FACILITY		NFSDBG_VFS
+
+/*
+ * Convert the NFSv4 pathname components into a standard posix path.
+ *
+ * Note that the resulting string will be placed at the end of the buffer
+ */
+static inline char *nfs4_pathname_string(const struct nfs4_pathname *pathname,
+					 char *buffer, ssize_t buflen)
+{
+	char *end = buffer + buflen;
+	int n;
+
+	*--end = '\0';
+	buflen--;
+
+	n = pathname->ncomponents;
+	while (--n >= 0) {
+		const struct nfs4_string *component = &pathname->components[n];
+		buflen -= component->len + 1;
+		if (buflen < 0)
+			goto Elong;
+		end -= component->len;
+		memcpy(end, component->data, component->len);
+		*--end = '/';
+	}
+	return end;
+Elong:
+	return ERR_PTR(-ENAMETOOLONG);
+}
+
+/*
+ * return the path component of "<server>:<path>"
+ *  nfspath - the "<server>:<path>" string
+ *  end - one past the last char that could contain "<server>:"
+ * returns NULL on failure
+ */
+static char *nfs_path_component(const char *nfspath, const char *end)
+{
+	char *p;
+
+	if (*nfspath == '[') {
+		/* parse [] escaped IPv6 addrs */
+		p = strchr(nfspath, ']');
+		if (p != NULL && ++p < end && *p == ':')
+			return p + 1;
+	} else {
+		/* otherwise split on first colon */
+		p = strchr(nfspath, ':');
+		if (p != NULL && p < end)
+			return p + 1;
+	}
+	return NULL;
+}
+
+/*
+ * Determine the mount path as a string
+ */
+static char *nfs4_path(struct dentry *dentry, char *buffer, ssize_t buflen)
+{
+	char *limit;
+	char *path = nfs_path(&limit, dentry, buffer, buflen,
+			      NFS_PATH_CANONICAL);
+	if (!IS_ERR(path)) {
+		char *path_component = nfs_path_component(path, limit);
+		if (path_component)
+			return path_component;
+	}
+	return path;
+}
+
+/*
+ * Check that fs_locations::fs_root [RFC3530 6.3] is a prefix for what we
+ * believe to be the server path to this dentry
+ */
+static int nfs4_validate_fspath(struct dentry *dentry,
+				const struct nfs4_fs_locations *locations,
+				char *page, char *page2)
+{
+	const char *path, *fs_path;
+
+	path = nfs4_path(dentry, page, PAGE_SIZE);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	fs_path = nfs4_pathname_string(&locations->fs_path, page2, PAGE_SIZE);
+	if (IS_ERR(fs_path))
+		return PTR_ERR(fs_path);
+
+	if (strncmp(path, fs_path, strlen(fs_path)) != 0) {
+		dprintk("%s: path %s does not begin with fsroot %s\n",
+			__func__, path, fs_path);
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
+static size_t nfs_parse_server_name(char *string, size_t len,
+		struct sockaddr *sa, size_t salen, struct net *net)
+{
+	ssize_t ret;
+
+	ret = rpc_pton(net, string, len, sa, salen);
+	if (ret == 0) {
+		ret = nfs_dns_resolve_name(net, string, len, sa, salen);
+		if (ret < 0)
+			ret = 0;
+	}
+	return ret;
+}
+
+/**
+ * nfs_find_best_sec - Find a security mechanism supported locally
+ * @server: NFS server struct
+ * @flavors: List of security tuples returned by SECINFO procedure
+ *
+ * Return an rpc client that uses the first security mechanism in
+ * "flavors" that is locally supported.  The "flavors" array
+ * is searched in the order returned from the server, per RFC 3530
+ * recommendation and each flavor is checked for membership in the
+ * sec= mount option list if it exists.
+ *
+ * Return -EPERM if no matching flavor is found in the array.
+ *
+ * Please call rpc_shutdown_client() when you are done with this rpc client.
+ *
+ */
+static struct rpc_clnt *nfs_find_best_sec(struct rpc_clnt *clnt,
+					  struct nfs_server *server,
+					  struct nfs4_secinfo_flavors *flavors)
+{
+	rpc_authflavor_t pflavor;
+	struct nfs4_secinfo4 *secinfo;
+	unsigned int i;
+
+	for (i = 0; i < flavors->num_flavors; i++) {
+		secinfo = &flavors->flavors[i];
+
+		switch (secinfo->flavor) {
+		case RPC_AUTH_NULL:
+		case RPC_AUTH_UNIX:
+		case RPC_AUTH_GSS:
+			pflavor = rpcauth_get_pseudoflavor(secinfo->flavor,
+							&secinfo->flavor_info);
+			/* does the pseudoflavor match a sec= mount opt? */
+			if (pflavor != RPC_AUTH_MAXFLAVOR &&
+			    nfs_auth_info_match(&server->auth_info, pflavor)) {
+				struct rpc_clnt *new;
+				struct rpc_cred *cred;
+
+				/* Cloning creates an rpc_auth for the flavor */
+				new = rpc_clone_client_set_auth(clnt, pflavor);
+				if (IS_ERR(new))
+					continue;
+				/**
+				* Check that the user actually can use the
+				* flavor. This is mostly for RPC_AUTH_GSS
+				* where cr_init obtains a gss context
+				*/
+				cred = rpcauth_lookupcred(new->cl_auth, 0);
+				if (IS_ERR(cred)) {
+					rpc_shutdown_client(new);
+					continue;
+				}
+				put_rpccred(cred);
+				return new;
+			}
+		}
+	}
+	return ERR_PTR(-EPERM);
+}
+
+/**
+ * nfs4_negotiate_security - in response to an NFS4ERR_WRONGSEC on lookup,
+ * return an rpc_clnt that uses the best available security flavor with
+ * respect to the secinfo flavor list and the sec= mount options.
+ *
+ * @clnt: RPC client to clone
+ * @inode: directory inode
+ * @name: lookup name
+ *
+ * Please call rpc_shutdown_client() when you are done with this rpc client.
+ */
+struct rpc_clnt *
+nfs4_negotiate_security(struct rpc_clnt *clnt, struct inode *inode,
+					struct qstr *name)
+{
+	struct page *page;
+	struct nfs4_secinfo_flavors *flavors;
+	struct rpc_clnt *new;
+	int err;
+
+	page = alloc_page(GFP_KERNEL);
+	if (!page)
+		return ERR_PTR(-ENOMEM);
+
+	flavors = page_address(page);
+
+	err = nfs4_proc_secinfo(inode, name, flavors);
+	if (err < 0) {
+		new = ERR_PTR(err);
+		goto out;
+	}
+
+	new = nfs_find_best_sec(clnt, NFS_SERVER(inode), flavors);
+
+out:
+	put_page(page);
+	return new;
+}
+
+static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
+				     char *page, char *page2,
+				     const struct nfs4_fs_location *location)
+{
+	const size_t addr_bufsize = sizeof(struct sockaddr_storage);
+	struct net *net = rpc_net_ns(NFS_SB(mountdata->sb)->client);
+	struct vfsmount *mnt = ERR_PTR(-ENOENT);
+	char *mnt_path;
+	unsigned int maxbuflen;
+	unsigned int s;
+
+	mnt_path = nfs4_pathname_string(&location->rootpath, page2, PAGE_SIZE);
+	if (IS_ERR(mnt_path))
+		return ERR_CAST(mnt_path);
+	mountdata->mnt_path = mnt_path;
+	maxbuflen = mnt_path - 1 - page2;
+
+	mountdata->addr = kmalloc(addr_bufsize, GFP_KERNEL);
+	if (mountdata->addr == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	for (s = 0; s < location->nservers; s++) {
+		const struct nfs4_string *buf = &location->servers[s];
+
+		if (buf->len <= 0 || buf->len >= maxbuflen)
+			continue;
+
+		if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len))
+			continue;
+
+		mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len,
+				mountdata->addr, addr_bufsize, net);
+		if (mountdata->addrlen == 0)
+			continue;
+
+		rpc_set_port(mountdata->addr, NFS_PORT);
+
+		memcpy(page2, buf->data, buf->len);
+		page2[buf->len] = '\0';
+		mountdata->hostname = page2;
+
+		snprintf(page, PAGE_SIZE, "%s:%s",
+				mountdata->hostname,
+				mountdata->mnt_path);
+
+		mnt = vfs_kern_mount(&nfs4_referral_fs_type, 0, page, mountdata);
+		if (!IS_ERR(mnt))
+			break;
+	}
+	kfree(mountdata->addr);
+	return mnt;
+}
+
+/**
+ * nfs_follow_referral - set up mountpoint when hitting a referral on moved error
+ * @dentry - parent directory
+ * @locations - array of NFSv4 server location information
+ *
+ */
+static struct vfsmount *nfs_follow_referral(struct dentry *dentry,
+					    const struct nfs4_fs_locations *locations)
+{
+	struct vfsmount *mnt = ERR_PTR(-ENOENT);
+	struct nfs_clone_mount mountdata = {
+		.sb = dentry->d_sb,
+		.dentry = dentry,
+		.authflavor = NFS_SB(dentry->d_sb)->client->cl_auth->au_flavor,
+	};
+	char *page = NULL, *page2 = NULL;
+	int loc, error;
+
+	if (locations == NULL || locations->nlocations <= 0)
+		goto out;
+
+	dprintk("%s: referral at %pd2\n", __func__, dentry);
+
+	page = (char *) __get_free_page(GFP_USER);
+	if (!page)
+		goto out;
+
+	page2 = (char *) __get_free_page(GFP_USER);
+	if (!page2)
+		goto out;
+
+	/* Ensure fs path is a prefix of current dentry path */
+	error = nfs4_validate_fspath(dentry, locations, page, page2);
+	if (error < 0) {
+		mnt = ERR_PTR(error);
+		goto out;
+	}
+
+	for (loc = 0; loc < locations->nlocations; loc++) {
+		const struct nfs4_fs_location *location = &locations->locations[loc];
+
+		if (location == NULL || location->nservers <= 0 ||
+		    location->rootpath.ncomponents == 0)
+			continue;
+
+		mnt = try_location(&mountdata, page, page2, location);
+		if (!IS_ERR(mnt))
+			break;
+	}
+
+out:
+	free_page((unsigned long) page);
+	free_page((unsigned long) page2);
+	dprintk("%s: done\n", __func__);
+	return mnt;
+}
+
+/*
+ * nfs_do_refmount - handle crossing a referral on server
+ * @dentry - dentry of referral
+ *
+ */
+static struct vfsmount *nfs_do_refmount(struct rpc_clnt *client, struct dentry *dentry)
+{
+	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
+	struct dentry *parent;
+	struct nfs4_fs_locations *fs_locations = NULL;
+	struct page *page;
+	int err;
+
+	/* BUG_ON(IS_ROOT(dentry)); */
+	dprintk("%s: enter\n", __func__);
+
+	page = alloc_page(GFP_KERNEL);
+	if (page == NULL)
+		goto out;
+
+	fs_locations = kmalloc(sizeof(struct nfs4_fs_locations), GFP_KERNEL);
+	if (fs_locations == NULL)
+		goto out_free;
+
+	/* Get locations */
+	mnt = ERR_PTR(-ENOENT);
+
+	parent = dget_parent(dentry);
+	dprintk("%s: getting locations for %pd2\n",
+		__func__, dentry);
+
+	err = nfs4_proc_fs_locations(client, parent->d_inode, &dentry->d_name, fs_locations, page);
+	dput(parent);
+	if (err != 0 ||
+	    fs_locations->nlocations <= 0 ||
+	    fs_locations->fs_path.ncomponents <= 0)
+		goto out_free;
+
+	mnt = nfs_follow_referral(dentry, fs_locations);
+out_free:
+	__free_page(page);
+	kfree(fs_locations);
+out:
+	dprintk("%s: done\n", __func__);
+	return mnt;
+}
+
+struct vfsmount *nfs4_submount(struct nfs_server *server, struct dentry *dentry,
+			       struct nfs_fh *fh, struct nfs_fattr *fattr)
+{
+	rpc_authflavor_t flavor = server->client->cl_auth->au_flavor;
+	struct dentry *parent = dget_parent(dentry);
+	struct inode *dir = parent->d_inode;
+	struct qstr *name = &dentry->d_name;
+	struct rpc_clnt *client;
+	struct vfsmount *mnt;
+
+	/* Look it up again to get its attributes and sec flavor */
+	client = nfs4_proc_lookup_mountpoint(dir, name, fh, fattr);
+	dput(parent);
+	if (IS_ERR(client))
+		return ERR_CAST(client);
+
+	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		mnt = nfs_do_refmount(client, dentry);
+		goto out;
+	}
+
+	if (client->cl_auth->au_flavor != flavor)
+		flavor = client->cl_auth->au_flavor;
+	mnt = nfs_do_submount(dentry, fh, fattr, flavor);
+out:
+	rpc_shutdown_client(client);
+	return mnt;
+}
+
+/*
+ * Try one location from the fs_locations array.
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+static int nfs4_try_replacing_one_location(struct nfs_server *server,
+		char *page, char *page2,
+		const struct nfs4_fs_location *location)
+{
+	const size_t addr_bufsize = sizeof(struct sockaddr_storage);
+	struct net *net = rpc_net_ns(server->client);
+	struct sockaddr *sap;
+	unsigned int s;
+	size_t salen;
+	int error;
+
+	sap = kmalloc(addr_bufsize, GFP_KERNEL);
+	if (sap == NULL)
+		return -ENOMEM;
+
+	error = -ENOENT;
+	for (s = 0; s < location->nservers; s++) {
+		const struct nfs4_string *buf = &location->servers[s];
+		char *hostname;
+
+		if (buf->len <= 0 || buf->len > PAGE_SIZE)
+			continue;
+
+		if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len) != NULL)
+			continue;
+
+		salen = nfs_parse_server_name(buf->data, buf->len,
+						sap, addr_bufsize, net);
+		if (salen == 0)
+			continue;
+		rpc_set_port(sap, NFS_PORT);
+
+		error = -ENOMEM;
+		hostname = kstrndup(buf->data, buf->len, GFP_KERNEL);
+		if (hostname == NULL)
+			break;
+
+		error = nfs4_update_server(server, hostname, sap, salen, net);
+		kfree(hostname);
+		if (error == 0)
+			break;
+	}
+
+	kfree(sap);
+	return error;
+}
+
+/**
+ * nfs4_replace_transport - set up transport to destination server
+ *
+ * @server: export being migrated
+ * @locations: fs_locations array
+ *
+ * Returns zero on success, or a negative errno value.
+ *
+ * The client tries all the entries in the "locations" array, in the
+ * order returned by the server, until one works or the end of the
+ * array is reached.
+ */
+int nfs4_replace_transport(struct nfs_server *server,
+			   const struct nfs4_fs_locations *locations)
+{
+	char *page = NULL, *page2 = NULL;
+	int loc, error;
+
+	error = -ENOENT;
+	if (locations == NULL || locations->nlocations <= 0)
+		goto out;
+
+	error = -ENOMEM;
+	page = (char *) __get_free_page(GFP_USER);
+	if (!page)
+		goto out;
+	page2 = (char *) __get_free_page(GFP_USER);
+	if (!page2)
+		goto out;
+
+	for (loc = 0; loc < locations->nlocations; loc++) {
+		const struct nfs4_fs_location *location =
+						&locations->locations[loc];
+
+		if (location == NULL || location->nservers <= 0 ||
+		    location->rootpath.ncomponents == 0)
+			continue;
+
+		error = nfs4_try_replacing_one_location(server, page,
+							page2, location);
+		if (error == 0)
+			break;
+	}
+
+out:
+	free_page((unsigned long)page);
+	free_page((unsigned long)page2);
+	return error;
+}

fs/nfs/nfs4renewd.c

@@ -0,0 +1,143 @@
+/*
+ *  fs/nfs/nfs4renewd.c
+ *
+ *  Copyright (c) 2002 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Kendrick Smith <kmsmith@umich.edu>
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the University 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 ``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 REGENTS 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.
+ *
+ * Implementation of the NFSv4 "renew daemon", which wakes up periodically to
+ * send a RENEW, to keep state alive on the server.  The daemon is implemented
+ * as an rpc_task, not a real kernel thread, so it always runs in rpciod's
+ * context.  There is one renewd per nfs_server.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/clnt.h>
+
+#include <linux/nfs.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
+#include "delegation.h"
+
+#define NFSDBG_FACILITY		NFSDBG_STATE
+
+void
+nfs4_renew_state(struct work_struct *work)
+{
+	const struct nfs4_state_maintenance_ops *ops;
+	struct nfs_client *clp =
+		container_of(work, struct nfs_client, cl_renewd.work);
+	struct rpc_cred *cred;
+	long lease;
+	unsigned long last, now;
+	unsigned renew_flags = 0;
+
+	ops = clp->cl_mvops->state_renewal_ops;
+	dprintk("%s: start\n", __func__);
+
+	if (test_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state))
+		goto out;
+
+	spin_lock(&clp->cl_lock);
+	lease = clp->cl_lease_time;
+	last = clp->cl_last_renewal;
+	now = jiffies;
+	/* Are we close to a lease timeout? */
+	if (time_after(now, last + lease/3))
+		renew_flags |= NFS4_RENEW_TIMEOUT;
+	if (nfs_delegations_present(clp))
+		renew_flags |= NFS4_RENEW_DELEGATION_CB;
+
+	if (renew_flags != 0) {
+		cred = ops->get_state_renewal_cred_locked(clp);
+		spin_unlock(&clp->cl_lock);
+		if (cred == NULL) {
+			if (!(renew_flags & NFS4_RENEW_DELEGATION_CB)) {
+				set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+				goto out;
+			}
+			nfs_expire_all_delegations(clp);
+		} else {
+			int ret;
+
+			/* Queue an asynchronous RENEW. */
+			ret = ops->sched_state_renewal(clp, cred, renew_flags);
+			put_rpccred(cred);
+			switch (ret) {
+			default:
+				goto out_exp;
+			case -EAGAIN:
+			case -ENOMEM:
+				break;
+			}
+		}
+	} else {
+		dprintk("%s: failed to call renewd. Reason: lease not expired \n",
+				__func__);
+		spin_unlock(&clp->cl_lock);
+	}
+	nfs4_schedule_state_renewal(clp);
+out_exp:
+	nfs_expire_unreferenced_delegations(clp);
+out:
+	dprintk("%s: done\n", __func__);
+}
+
+void
+nfs4_schedule_state_renewal(struct nfs_client *clp)
+{
+	long timeout;
+
+	spin_lock(&clp->cl_lock);
+	timeout = (2 * clp->cl_lease_time) / 3 + (long)clp->cl_last_renewal
+		- (long)jiffies;
+	if (timeout < 5 * HZ)
+		timeout = 5 * HZ;
+	dprintk("%s: requeueing work. Lease period = %ld\n",
+			__func__, (timeout + HZ - 1) / HZ);
+	mod_delayed_work(system_wq, &clp->cl_renewd, timeout);
+	set_bit(NFS_CS_RENEWD, &clp->cl_res_state);
+	spin_unlock(&clp->cl_lock);
+}
+
+void
+nfs4_kill_renewd(struct nfs_client *clp)
+{
+	cancel_delayed_work_sync(&clp->cl_renewd);
+}
+
+/*
+ * Local variables:
+ *   c-basic-offset: 8
+ * End:
+ */

fs/nfs/nfs4session.c

@@ -0,0 +1,565 @@
+/*
+ * fs/nfs/nfs4session.c
+ *
+ * Copyright (c) 2012 Trond Myklebust <Trond.Myklebust@netapp.com>
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/printk.h>
+#include <linux/slab.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/bc_xprt.h>
+#include <linux/nfs.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include <linux/module.h>
+
+#include "nfs4_fs.h"
+#include "internal.h"
+#include "nfs4session.h"
+#include "callback.h"
+
+#define NFSDBG_FACILITY		NFSDBG_STATE
+
+static void nfs4_init_slot_table(struct nfs4_slot_table *tbl, const char *queue)
+{
+	tbl->highest_used_slotid = NFS4_NO_SLOT;
+	spin_lock_init(&tbl->slot_tbl_lock);
+	rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, queue);
+	init_completion(&tbl->complete);
+}
+
+/*
+ * nfs4_shrink_slot_table - free retired slots from the slot table
+ */
+static void nfs4_shrink_slot_table(struct nfs4_slot_table  *tbl, u32 newsize)
+{
+	struct nfs4_slot **p;
+	if (newsize >= tbl->max_slots)
+		return;
+
+	p = &tbl->slots;
+	while (newsize--)
+		p = &(*p)->next;
+	while (*p) {
+		struct nfs4_slot *slot = *p;
+
+		*p = slot->next;
+		kfree(slot);
+		tbl->max_slots--;
+	}
+}
+
+/**
+ * nfs4_slot_tbl_drain_complete - wake waiters when drain is complete
+ * @tbl - controlling slot table
+ *
+ */
+void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl)
+{
+	if (nfs4_slot_tbl_draining(tbl))
+		complete(&tbl->complete);
+}
+
+/*
+ * nfs4_free_slot - free a slot and efficiently update slot table.
+ *
+ * freeing a slot is trivially done by clearing its respective bit
+ * in the bitmap.
+ * If the freed slotid equals highest_used_slotid we want to update it
+ * so that the server would be able to size down the slot table if needed,
+ * otherwise we know that the highest_used_slotid is still in use.
+ * When updating highest_used_slotid there may be "holes" in the bitmap
+ * so we need to scan down from highest_used_slotid to 0 looking for the now
+ * highest slotid in use.
+ * If none found, highest_used_slotid is set to NFS4_NO_SLOT.
+ *
+ * Must be called while holding tbl->slot_tbl_lock
+ */
+void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot)
+{
+	u32 slotid = slot->slot_nr;
+
+	/* clear used bit in bitmap */
+	__clear_bit(slotid, tbl->used_slots);
+
+	/* update highest_used_slotid when it is freed */
+	if (slotid == tbl->highest_used_slotid) {
+		u32 new_max = find_last_bit(tbl->used_slots, slotid);
+		if (new_max < slotid)
+			tbl->highest_used_slotid = new_max;
+		else {
+			tbl->highest_used_slotid = NFS4_NO_SLOT;
+			nfs4_slot_tbl_drain_complete(tbl);
+		}
+	}
+	dprintk("%s: slotid %u highest_used_slotid %u\n", __func__,
+		slotid, tbl->highest_used_slotid);
+}
+
+static struct nfs4_slot *nfs4_new_slot(struct nfs4_slot_table  *tbl,
+		u32 slotid, u32 seq_init, gfp_t gfp_mask)
+{
+	struct nfs4_slot *slot;
+
+	slot = kzalloc(sizeof(*slot), gfp_mask);
+	if (slot) {
+		slot->table = tbl;
+		slot->slot_nr = slotid;
+		slot->seq_nr = seq_init;
+	}
+	return slot;
+}
+
+static struct nfs4_slot *nfs4_find_or_create_slot(struct nfs4_slot_table  *tbl,
+		u32 slotid, u32 seq_init, gfp_t gfp_mask)
+{
+	struct nfs4_slot **p, *slot;
+
+	p = &tbl->slots;
+	for (;;) {
+		if (*p == NULL) {
+			*p = nfs4_new_slot(tbl, tbl->max_slots,
+					seq_init, gfp_mask);
+			if (*p == NULL)
+				break;
+			tbl->max_slots++;
+		}
+		slot = *p;
+		if (slot->slot_nr == slotid)
+			return slot;
+		p = &slot->next;
+	}
+	return ERR_PTR(-ENOMEM);
+}
+
+/*
+ * nfs4_alloc_slot - efficiently look for a free slot
+ *
+ * nfs4_alloc_slot looks for an unset bit in the used_slots bitmap.
+ * If found, we mark the slot as used, update the highest_used_slotid,
+ * and respectively set up the sequence operation args.
+ *
+ * Note: must be called with under the slot_tbl_lock.
+ */
+struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl)
+{
+	struct nfs4_slot *ret = ERR_PTR(-EBUSY);
+	u32 slotid;
+
+	dprintk("--> %s used_slots=%04lx highest_used=%u max_slots=%u\n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
+		tbl->max_slotid + 1);
+	slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slotid + 1);
+	if (slotid > tbl->max_slotid)
+		goto out;
+	ret = nfs4_find_or_create_slot(tbl, slotid, 1, GFP_NOWAIT);
+	if (IS_ERR(ret))
+		goto out;
+	__set_bit(slotid, tbl->used_slots);
+	if (slotid > tbl->highest_used_slotid ||
+			tbl->highest_used_slotid == NFS4_NO_SLOT)
+		tbl->highest_used_slotid = slotid;
+	ret->generation = tbl->generation;
+
+out:
+	dprintk("<-- %s used_slots=%04lx highest_used=%u slotid=%u\n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
+		!IS_ERR(ret) ? ret->slot_nr : NFS4_NO_SLOT);
+	return ret;
+}
+
+static int nfs4_grow_slot_table(struct nfs4_slot_table *tbl,
+		 u32 max_reqs, u32 ivalue)
+{
+	if (max_reqs <= tbl->max_slots)
+		return 0;
+	if (!IS_ERR(nfs4_find_or_create_slot(tbl, max_reqs - 1, ivalue, GFP_NOFS)))
+		return 0;
+	return -ENOMEM;
+}
+
+static void nfs4_reset_slot_table(struct nfs4_slot_table *tbl,
+		u32 server_highest_slotid,
+		u32 ivalue)
+{
+	struct nfs4_slot **p;
+
+	nfs4_shrink_slot_table(tbl, server_highest_slotid + 1);
+	p = &tbl->slots;
+	while (*p) {
+		(*p)->seq_nr = ivalue;
+		(*p)->interrupted = 0;
+		p = &(*p)->next;
+	}
+	tbl->highest_used_slotid = NFS4_NO_SLOT;
+	tbl->target_highest_slotid = server_highest_slotid;
+	tbl->server_highest_slotid = server_highest_slotid;
+	tbl->d_target_highest_slotid = 0;
+	tbl->d2_target_highest_slotid = 0;
+	tbl->max_slotid = server_highest_slotid;
+}
+
+/*
+ * (re)Initialise a slot table
+ */
+static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl,
+		u32 max_reqs, u32 ivalue)
+{
+	int ret;
+
+	dprintk("--> %s: max_reqs=%u, tbl->max_slots %u\n", __func__,
+		max_reqs, tbl->max_slots);
+
+	if (max_reqs > NFS4_MAX_SLOT_TABLE)
+		max_reqs = NFS4_MAX_SLOT_TABLE;
+
+	ret = nfs4_grow_slot_table(tbl, max_reqs, ivalue);
+	if (ret)
+		goto out;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	nfs4_reset_slot_table(tbl, max_reqs - 1, ivalue);
+	spin_unlock(&tbl->slot_tbl_lock);
+
+	dprintk("%s: tbl=%p slots=%p max_slots=%u\n", __func__,
+		tbl, tbl->slots, tbl->max_slots);
+out:
+	dprintk("<-- %s: return %d\n", __func__, ret);
+	return ret;
+}
+
+/*
+ * nfs4_release_slot_table - release all slot table entries
+ */
+static void nfs4_release_slot_table(struct nfs4_slot_table *tbl)
+{
+	nfs4_shrink_slot_table(tbl, 0);
+}
+
+/**
+ * nfs4_shutdown_slot_table - release resources attached to a slot table
+ * @tbl: slot table to shut down
+ *
+ */
+void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl)
+{
+	nfs4_release_slot_table(tbl);
+	rpc_destroy_wait_queue(&tbl->slot_tbl_waitq);
+}
+
+/**
+ * nfs4_setup_slot_table - prepare a stand-alone slot table for use
+ * @tbl: slot table to set up
+ * @max_reqs: maximum number of requests allowed
+ * @queue: name to give RPC wait queue
+ *
+ * Returns zero on success, or a negative errno.
+ */
+int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs,
+		const char *queue)
+{
+	nfs4_init_slot_table(tbl, queue);
+	return nfs4_realloc_slot_table(tbl, max_reqs, 0);
+}
+
+static bool nfs41_assign_slot(struct rpc_task *task, void *pslot)
+{
+	struct nfs4_sequence_args *args = task->tk_msg.rpc_argp;
+	struct nfs4_sequence_res *res = task->tk_msg.rpc_resp;
+	struct nfs4_slot *slot = pslot;
+	struct nfs4_slot_table *tbl = slot->table;
+
+	if (nfs4_slot_tbl_draining(tbl) && !args->sa_privileged)
+		return false;
+	slot->generation = tbl->generation;
+	args->sa_slot = slot;
+	res->sr_timestamp = jiffies;
+	res->sr_slot = slot;
+	res->sr_status_flags = 0;
+	res->sr_status = 1;
+	return true;
+}
+
+static bool __nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot)
+{
+	if (rpc_wake_up_first(&tbl->slot_tbl_waitq, nfs41_assign_slot, slot))
+		return true;
+	return false;
+}
+
+bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot)
+{
+	if (slot->slot_nr > tbl->max_slotid)
+		return false;
+	return __nfs41_wake_and_assign_slot(tbl, slot);
+}
+
+static bool nfs41_try_wake_next_slot_table_entry(struct nfs4_slot_table *tbl)
+{
+	struct nfs4_slot *slot = nfs4_alloc_slot(tbl);
+	if (!IS_ERR(slot)) {
+		bool ret = __nfs41_wake_and_assign_slot(tbl, slot);
+		if (ret)
+			return ret;
+		nfs4_free_slot(tbl, slot);
+	}
+	return false;
+}
+
+void nfs41_wake_slot_table(struct nfs4_slot_table *tbl)
+{
+	for (;;) {
+		if (!nfs41_try_wake_next_slot_table_entry(tbl))
+			break;
+	}
+}
+
+#if defined(CONFIG_NFS_V4_1)
+
+static void nfs41_set_max_slotid_locked(struct nfs4_slot_table *tbl,
+		u32 target_highest_slotid)
+{
+	u32 max_slotid;
+
+	max_slotid = min(NFS4_MAX_SLOT_TABLE - 1, target_highest_slotid);
+	if (max_slotid > tbl->server_highest_slotid)
+		max_slotid = tbl->server_highest_slotid;
+	if (max_slotid > tbl->target_highest_slotid)
+		max_slotid = tbl->target_highest_slotid;
+	tbl->max_slotid = max_slotid;
+	nfs41_wake_slot_table(tbl);
+}
+
+/* Update the client's idea of target_highest_slotid */
+static void nfs41_set_target_slotid_locked(struct nfs4_slot_table *tbl,
+		u32 target_highest_slotid)
+{
+	if (tbl->target_highest_slotid == target_highest_slotid)
+		return;
+	tbl->target_highest_slotid = target_highest_slotid;
+	tbl->generation++;
+}
+
+void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
+		u32 target_highest_slotid)
+{
+	spin_lock(&tbl->slot_tbl_lock);
+	nfs41_set_target_slotid_locked(tbl, target_highest_slotid);
+	tbl->d_target_highest_slotid = 0;
+	tbl->d2_target_highest_slotid = 0;
+	nfs41_set_max_slotid_locked(tbl, target_highest_slotid);
+	spin_unlock(&tbl->slot_tbl_lock);
+}
+
+static void nfs41_set_server_slotid_locked(struct nfs4_slot_table *tbl,
+		u32 highest_slotid)
+{
+	if (tbl->server_highest_slotid == highest_slotid)
+		return;
+	if (tbl->highest_used_slotid > highest_slotid)
+		return;
+	/* Deallocate slots */
+	nfs4_shrink_slot_table(tbl, highest_slotid + 1);
+	tbl->server_highest_slotid = highest_slotid;
+}
+
+static s32 nfs41_derivative_target_slotid(s32 s1, s32 s2)
+{
+	s1 -= s2;
+	if (s1 == 0)
+		return 0;
+	if (s1 < 0)
+		return (s1 - 1) >> 1;
+	return (s1 + 1) >> 1;
+}
+
+static int nfs41_sign_s32(s32 s1)
+{
+	if (s1 > 0)
+		return 1;
+	if (s1 < 0)
+		return -1;
+	return 0;
+}
+
+static bool nfs41_same_sign_or_zero_s32(s32 s1, s32 s2)
+{
+	if (!s1 || !s2)
+		return true;
+	return nfs41_sign_s32(s1) == nfs41_sign_s32(s2);
+}
+
+/* Try to eliminate outliers by checking for sharp changes in the
+ * derivatives and second derivatives
+ */
+static bool nfs41_is_outlier_target_slotid(struct nfs4_slot_table *tbl,
+		u32 new_target)
+{
+	s32 d_target, d2_target;
+	bool ret = true;
+
+	d_target = nfs41_derivative_target_slotid(new_target,
+			tbl->target_highest_slotid);
+	d2_target = nfs41_derivative_target_slotid(d_target,
+			tbl->d_target_highest_slotid);
+	/* Is first derivative same sign? */
+	if (nfs41_same_sign_or_zero_s32(d_target, tbl->d_target_highest_slotid))
+		ret = false;
+	/* Is second derivative same sign? */
+	if (nfs41_same_sign_or_zero_s32(d2_target, tbl->d2_target_highest_slotid))
+		ret = false;
+	tbl->d_target_highest_slotid = d_target;
+	tbl->d2_target_highest_slotid = d2_target;
+	return ret;
+}
+
+void nfs41_update_target_slotid(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot,
+		struct nfs4_sequence_res *res)
+{
+	spin_lock(&tbl->slot_tbl_lock);
+	if (!nfs41_is_outlier_target_slotid(tbl, res->sr_target_highest_slotid))
+		nfs41_set_target_slotid_locked(tbl, res->sr_target_highest_slotid);
+	if (tbl->generation == slot->generation)
+		nfs41_set_server_slotid_locked(tbl, res->sr_highest_slotid);
+	nfs41_set_max_slotid_locked(tbl, res->sr_target_highest_slotid);
+	spin_unlock(&tbl->slot_tbl_lock);
+}
+
+static void nfs4_release_session_slot_tables(struct nfs4_session *session)
+{
+	nfs4_release_slot_table(&session->fc_slot_table);
+	nfs4_release_slot_table(&session->bc_slot_table);
+}
+
+/*
+ * Initialize or reset the forechannel and backchannel tables
+ */
+int nfs4_setup_session_slot_tables(struct nfs4_session *ses)
+{
+	struct nfs4_slot_table *tbl;
+	int status;
+
+	dprintk("--> %s\n", __func__);
+	/* Fore channel */
+	tbl = &ses->fc_slot_table;
+	tbl->session = ses;
+	status = nfs4_realloc_slot_table(tbl, ses->fc_attrs.max_reqs, 1);
+	if (status) /* -ENOMEM */
+		return status;
+	/* Back channel */
+	tbl = &ses->bc_slot_table;
+	tbl->session = ses;
+	status = nfs4_realloc_slot_table(tbl, ses->bc_attrs.max_reqs, 0);
+	if (status && tbl->slots == NULL)
+		/* Fore and back channel share a connection so get
+		 * both slot tables or neither */
+		nfs4_release_session_slot_tables(ses);
+	return status;
+}
+
+struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
+{
+	struct nfs4_session *session;
+
+	session = kzalloc(sizeof(struct nfs4_session), GFP_NOFS);
+	if (!session)
+		return NULL;
+
+	nfs4_init_slot_table(&session->fc_slot_table, "ForeChannel Slot table");
+	nfs4_init_slot_table(&session->bc_slot_table, "BackChannel Slot table");
+	session->session_state = 1<<NFS4_SESSION_INITING;
+
+	session->clp = clp;
+	return session;
+}
+
+static void nfs4_destroy_session_slot_tables(struct nfs4_session *session)
+{
+	nfs4_shutdown_slot_table(&session->fc_slot_table);
+	nfs4_shutdown_slot_table(&session->bc_slot_table);
+}
+
+void nfs4_destroy_session(struct nfs4_session *session)
+{
+	struct rpc_xprt *xprt;
+	struct rpc_cred *cred;
+
+	cred = nfs4_get_clid_cred(session->clp);
+	nfs4_proc_destroy_session(session, cred);
+	if (cred)
+		put_rpccred(cred);
+
+	rcu_read_lock();
+	xprt = rcu_dereference(session->clp->cl_rpcclient->cl_xprt);
+	rcu_read_unlock();
+	dprintk("%s Destroy backchannel for xprt %p\n",
+		__func__, xprt);
+	xprt_destroy_backchannel(xprt, NFS41_BC_MIN_CALLBACKS);
+	nfs4_destroy_session_slot_tables(session);
+	kfree(session);
+}
+
+/*
+ * With sessions, the client is not marked ready until after a
+ * successful EXCHANGE_ID and CREATE_SESSION.
+ *
+ * Map errors cl_cons_state errors to EPROTONOSUPPORT to indicate
+ * other versions of NFS can be tried.
+ */
+static int nfs41_check_session_ready(struct nfs_client *clp)
+{
+	int ret;
+	
+	if (clp->cl_cons_state == NFS_CS_SESSION_INITING) {
+		ret = nfs4_client_recover_expired_lease(clp);
+		if (ret)
+			return ret;
+	}
+	if (clp->cl_cons_state < NFS_CS_READY)
+		return -EPROTONOSUPPORT;
+	smp_rmb();
+	return 0;
+}
+
+int nfs4_init_session(struct nfs_client *clp)
+{
+	if (!nfs4_has_session(clp))
+		return 0;
+
+	clear_bit(NFS4_SESSION_INITING, &clp->cl_session->session_state);
+	return nfs41_check_session_ready(clp);
+}
+
+int nfs4_init_ds_session(struct nfs_client *clp, unsigned long lease_time)
+{
+	struct nfs4_session *session = clp->cl_session;
+	int ret;
+
+	spin_lock(&clp->cl_lock);
+	if (test_and_clear_bit(NFS4_SESSION_INITING, &session->session_state)) {
+		/*
+		 * Do not set NFS_CS_CHECK_LEASE_TIME instead set the
+		 * DS lease to be equal to the MDS lease.
+		 */
+		clp->cl_lease_time = lease_time;
+		clp->cl_last_renewal = jiffies;
+	}
+	spin_unlock(&clp->cl_lock);
+
+	ret = nfs41_check_session_ready(clp);
+	if (ret)
+		return ret;
+	/* Test for the DS role */
+	if (!is_ds_client(clp))
+		return -ENODEV;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(nfs4_init_ds_session);
+
+#endif	/* defined(CONFIG_NFS_V4_1) */

fs/nfs/nfs4session.h

@@ -0,0 +1,153 @@
+/*
+ * fs/nfs/nfs4session.h
+ *
+ * Copyright (c) 2012 Trond Myklebust <Trond.Myklebust@netapp.com>
+ *
+ */
+#ifndef __LINUX_FS_NFS_NFS4SESSION_H
+#define __LINUX_FS_NFS_NFS4SESSION_H
+
+/* maximum number of slots to use */
+#define NFS4_DEF_SLOT_TABLE_SIZE (64U)
+#define NFS4_MAX_SLOT_TABLE (1024U)
+#define NFS4_NO_SLOT ((u32)-1)
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+
+/* Sessions slot seqid */
+struct nfs4_slot {
+	struct nfs4_slot_table	*table;
+	struct nfs4_slot	*next;
+	unsigned long		generation;
+	u32			slot_nr;
+	u32		 	seq_nr;
+	unsigned int		interrupted : 1;
+};
+
+/* Sessions */
+enum nfs4_slot_tbl_state {
+	NFS4_SLOT_TBL_DRAINING,
+};
+
+#define SLOT_TABLE_SZ DIV_ROUND_UP(NFS4_MAX_SLOT_TABLE, 8*sizeof(long))
+struct nfs4_slot_table {
+	struct nfs4_session *session;		/* Parent session */
+	struct nfs4_slot *slots;		/* seqid per slot */
+	unsigned long   used_slots[SLOT_TABLE_SZ]; /* used/unused bitmap */
+	spinlock_t	slot_tbl_lock;
+	struct rpc_wait_queue	slot_tbl_waitq;	/* allocators may wait here */
+	u32		max_slots;		/* # slots in table */
+	u32		max_slotid;		/* Max allowed slotid value */
+	u32		highest_used_slotid;	/* sent to server on each SEQ.
+						 * op for dynamic resizing */
+	u32		target_highest_slotid;	/* Server max_slot target */
+	u32		server_highest_slotid;	/* Server highest slotid */
+	s32		d_target_highest_slotid; /* Derivative */
+	s32		d2_target_highest_slotid; /* 2nd derivative */
+	unsigned long	generation;		/* Generation counter for
+						   target_highest_slotid */
+	struct completion complete;
+	unsigned long	slot_tbl_state;
+};
+
+/*
+ * Session related parameters
+ */
+struct nfs4_session {
+	struct nfs4_sessionid		sess_id;
+	u32				flags;
+	unsigned long			session_state;
+	u32				hash_alg;
+	u32				ssv_len;
+
+	/* The fore and back channel */
+	struct nfs4_channel_attrs	fc_attrs;
+	struct nfs4_slot_table		fc_slot_table;
+	struct nfs4_channel_attrs	bc_attrs;
+	struct nfs4_slot_table		bc_slot_table;
+	struct nfs_client		*clp;
+};
+
+enum nfs4_session_state {
+	NFS4_SESSION_INITING,
+};
+
+extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl,
+		unsigned int max_reqs, const char *queue);
+extern void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl);
+extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl);
+extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot);
+extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl);
+bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot);
+void nfs41_wake_slot_table(struct nfs4_slot_table *tbl);
+
+static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl)
+{
+	return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
+}
+
+#if defined(CONFIG_NFS_V4_1)
+extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
+		u32 target_highest_slotid);
+extern void nfs41_update_target_slotid(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *slot,
+		struct nfs4_sequence_res *res);
+
+extern int nfs4_setup_session_slot_tables(struct nfs4_session *ses);
+
+extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+extern void nfs4_destroy_session(struct nfs4_session *session);
+extern int nfs4_init_session(struct nfs_client *clp);
+extern int nfs4_init_ds_session(struct nfs_client *, unsigned long);
+
+/*
+ * Determine if sessions are in use.
+ */
+static inline int nfs4_has_session(const struct nfs_client *clp)
+{
+	if (clp->cl_session)
+		return 1;
+	return 0;
+}
+
+static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
+{
+	if (nfs4_has_session(clp))
+		return (clp->cl_session->flags & SESSION4_PERSIST);
+	return 0;
+}
+
+#ifdef CONFIG_CRC32
+/*
+ * nfs_session_id_hash - calculate the crc32 hash for the session id
+ * @session - pointer to session
+ */
+#define nfs_session_id_hash(sess_id) \
+	(~crc32_le(0xFFFFFFFF, &(sess_id)->data[0], sizeof((sess_id)->data)))
+#else
+#define nfs_session_id_hash(session) (0)
+#endif
+#else /* defined(CONFIG_NFS_V4_1) */
+
+static inline int nfs4_init_session(struct nfs_client *clp)
+{
+	return 0;
+}
+
+/*
+ * Determine if sessions are in use.
+ */
+static inline int nfs4_has_session(const struct nfs_client *clp)
+{
+	return 0;
+}
+
+static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
+{
+	return 0;
+}
+
+#endif /* defined(CONFIG_NFS_V4_1) */
+#endif /* IS_ENABLED(CONFIG_NFS_V4) */
+#endif /* __LINUX_FS_NFS_NFS4SESSION_H */

fs/nfs/nfs4super.c

@@ -0,0 +1,358 @@
+/*
+ * Copyright (c) 2012 Bryan Schumaker <bjschuma@netapp.com>
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/nfs_idmap.h>
+#include <linux/nfs4_mount.h>
+#include <linux/nfs_fs.h>
+#include "delegation.h"
+#include "internal.h"
+#include "nfs4_fs.h"
+#include "dns_resolve.h"
+#include "pnfs.h"
+#include "nfs.h"
+
+#define NFSDBG_FACILITY		NFSDBG_VFS
+
+static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc);
+static void nfs4_evict_inode(struct inode *inode);
+static struct dentry *nfs4_remote_mount(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data);
+static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data);
+static struct dentry *nfs4_remote_referral_mount(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data);
+
+static struct file_system_type nfs4_remote_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "nfs4",
+	.mount		= nfs4_remote_mount,
+	.kill_sb	= nfs_kill_super,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
+};
+
+static struct file_system_type nfs4_remote_referral_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "nfs4",
+	.mount		= nfs4_remote_referral_mount,
+	.kill_sb	= nfs_kill_super,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
+};
+
+struct file_system_type nfs4_referral_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "nfs4",
+	.mount		= nfs4_referral_mount,
+	.kill_sb	= nfs_kill_super,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
+};
+
+static const struct super_operations nfs4_sops = {
+	.alloc_inode	= nfs_alloc_inode,
+	.destroy_inode	= nfs_destroy_inode,
+	.write_inode	= nfs4_write_inode,
+	.drop_inode	= nfs_drop_inode,
+	.put_super	= nfs_put_super,
+	.statfs		= nfs_statfs,
+	.evict_inode	= nfs4_evict_inode,
+	.umount_begin	= nfs_umount_begin,
+	.show_options	= nfs_show_options,
+	.show_devname	= nfs_show_devname,
+	.show_path	= nfs_show_path,
+	.show_stats	= nfs_show_stats,
+	.remount_fs	= nfs_remount,
+};
+
+struct nfs_subversion nfs_v4 = {
+	.owner = THIS_MODULE,
+	.nfs_fs   = &nfs4_fs_type,
+	.rpc_vers = &nfs_version4,
+	.rpc_ops  = &nfs_v4_clientops,
+	.sops     = &nfs4_sops,
+	.xattr    = nfs4_xattr_handlers,
+};
+
+static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc)
+{
+	int ret = nfs_write_inode(inode, wbc);
+
+	if (ret == 0)
+		ret = pnfs_layoutcommit_inode(inode,
+				wbc->sync_mode == WB_SYNC_ALL);
+	return ret;
+}
+
+/*
+ * Clean out any remaining NFSv4 state that might be left over due
+ * to open() calls that passed nfs_atomic_lookup, but failed to call
+ * nfs_open().
+ */
+static void nfs4_evict_inode(struct inode *inode)
+{
+	truncate_inode_pages_final(&inode->i_data);
+	clear_inode(inode);
+	pnfs_return_layout(inode);
+	pnfs_destroy_layout(NFS_I(inode));
+	/* If we are holding a delegation, return it! */
+	nfs_inode_return_delegation_noreclaim(inode);
+	/* First call standard NFS clear_inode() code */
+	nfs_clear_inode(inode);
+}
+
+/*
+ * Get the superblock for the NFS4 root partition
+ */
+static struct dentry *
+nfs4_remote_mount(struct file_system_type *fs_type, int flags,
+		  const char *dev_name, void *info)
+{
+	struct nfs_mount_info *mount_info = info;
+	struct nfs_server *server;
+	struct dentry *mntroot = ERR_PTR(-ENOMEM);
+
+	mount_info->set_security = nfs_set_sb_security;
+
+	/* Get a volume representation */
+	server = nfs4_create_server(mount_info, &nfs_v4);
+	if (IS_ERR(server)) {
+		mntroot = ERR_CAST(server);
+		goto out;
+	}
+
+	mntroot = nfs_fs_mount_common(server, flags, dev_name, mount_info, &nfs_v4);
+
+out:
+	return mntroot;
+}
+
+static struct vfsmount *nfs_do_root_mount(struct file_system_type *fs_type,
+		int flags, void *data, const char *hostname)
+{
+	struct vfsmount *root_mnt;
+	char *root_devname;
+	size_t len;
+
+	len = strlen(hostname) + 5;
+	root_devname = kmalloc(len, GFP_KERNEL);
+	if (root_devname == NULL)
+		return ERR_PTR(-ENOMEM);
+	/* Does hostname needs to be enclosed in brackets? */
+	if (strchr(hostname, ':'))
+		snprintf(root_devname, len, "[%s]:/", hostname);
+	else
+		snprintf(root_devname, len, "%s:/", hostname);
+	root_mnt = vfs_kern_mount(fs_type, flags, root_devname, data);
+	kfree(root_devname);
+	return root_mnt;
+}
+
+struct nfs_referral_count {
+	struct list_head list;
+	const struct task_struct *task;
+	unsigned int referral_count;
+};
+
+static LIST_HEAD(nfs_referral_count_list);
+static DEFINE_SPINLOCK(nfs_referral_count_list_lock);
+
+static struct nfs_referral_count *nfs_find_referral_count(void)
+{
+	struct nfs_referral_count *p;
+
+	list_for_each_entry(p, &nfs_referral_count_list, list) {
+		if (p->task == current)
+			return p;
+	}
+	return NULL;
+}
+
+#define NFS_MAX_NESTED_REFERRALS 2
+
+static int nfs_referral_loop_protect(void)
+{
+	struct nfs_referral_count *p, *new;
+	int ret = -ENOMEM;
+
+	new = kmalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		goto out;
+	new->task = current;
+	new->referral_count = 1;
+
+	ret = 0;
+	spin_lock(&nfs_referral_count_list_lock);
+	p = nfs_find_referral_count();
+	if (p != NULL) {
+		if (p->referral_count >= NFS_MAX_NESTED_REFERRALS)
+			ret = -ELOOP;
+		else
+			p->referral_count++;
+	} else {
+		list_add(&new->list, &nfs_referral_count_list);
+		new = NULL;
+	}
+	spin_unlock(&nfs_referral_count_list_lock);
+	kfree(new);
+out:
+	return ret;
+}
+
+static void nfs_referral_loop_unprotect(void)
+{
+	struct nfs_referral_count *p;
+
+	spin_lock(&nfs_referral_count_list_lock);
+	p = nfs_find_referral_count();
+	p->referral_count--;
+	if (p->referral_count == 0)
+		list_del(&p->list);
+	else
+		p = NULL;
+	spin_unlock(&nfs_referral_count_list_lock);
+	kfree(p);
+}
+
+static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
+		const char *export_path)
+{
+	struct dentry *dentry;
+	int err;
+
+	if (IS_ERR(root_mnt))
+		return ERR_CAST(root_mnt);
+
+	err = nfs_referral_loop_protect();
+	if (err) {
+		mntput(root_mnt);
+		return ERR_PTR(err);
+	}
+
+	dentry = mount_subtree(root_mnt, export_path);
+	nfs_referral_loop_unprotect();
+
+	return dentry;
+}
+
+struct dentry *nfs4_try_mount(int flags, const char *dev_name,
+			      struct nfs_mount_info *mount_info,
+			      struct nfs_subversion *nfs_mod)
+{
+	char *export_path;
+	struct vfsmount *root_mnt;
+	struct dentry *res;
+	struct nfs_parsed_mount_data *data = mount_info->parsed;
+
+	dfprintk(MOUNT, "--> nfs4_try_mount()\n");
+
+	export_path = data->nfs_server.export_path;
+	data->nfs_server.export_path = "/";
+	root_mnt = nfs_do_root_mount(&nfs4_remote_fs_type, flags, mount_info,
+			data->nfs_server.hostname);
+	data->nfs_server.export_path = export_path;
+
+	res = nfs_follow_remote_path(root_mnt, export_path);
+
+	dfprintk(MOUNT, "<-- nfs4_try_mount() = %d%s\n",
+		 PTR_ERR_OR_ZERO(res),
+		 IS_ERR(res) ? " [error]" : "");
+	return res;
+}
+
+static struct dentry *
+nfs4_remote_referral_mount(struct file_system_type *fs_type, int flags,
+			   const char *dev_name, void *raw_data)
+{
+	struct nfs_mount_info mount_info = {
+		.fill_super = nfs_fill_super,
+		.set_security = nfs_clone_sb_security,
+		.cloned = raw_data,
+	};
+	struct nfs_server *server;
+	struct dentry *mntroot = ERR_PTR(-ENOMEM);
+
+	dprintk("--> nfs4_referral_get_sb()\n");
+
+	mount_info.mntfh = nfs_alloc_fhandle();
+	if (mount_info.cloned == NULL || mount_info.mntfh == NULL)
+		goto out;
+
+	/* create a new volume representation */
+	server = nfs4_create_referral_server(mount_info.cloned, mount_info.mntfh);
+	if (IS_ERR(server)) {
+		mntroot = ERR_CAST(server);
+		goto out;
+	}
+
+	mntroot = nfs_fs_mount_common(server, flags, dev_name, &mount_info, &nfs_v4);
+out:
+	nfs_free_fhandle(mount_info.mntfh);
+	return mntroot;
+}
+
+/*
+ * Create an NFS4 server record on referral traversal
+ */
+static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
+		int flags, const char *dev_name, void *raw_data)
+{
+	struct nfs_clone_mount *data = raw_data;
+	char *export_path;
+	struct vfsmount *root_mnt;
+	struct dentry *res;
+
+	dprintk("--> nfs4_referral_mount()\n");
+
+	export_path = data->mnt_path;
+	data->mnt_path = "/";
+
+	root_mnt = nfs_do_root_mount(&nfs4_remote_referral_fs_type,
+			flags, data, data->hostname);
+	data->mnt_path = export_path;
+
+	res = nfs_follow_remote_path(root_mnt, export_path);
+	dprintk("<-- nfs4_referral_mount() = %d%s\n",
+		PTR_ERR_OR_ZERO(res),
+		IS_ERR(res) ? " [error]" : "");
+	return res;
+}
+
+
+static int __init init_nfs_v4(void)
+{
+	int err;
+
+	err = nfs_dns_resolver_init();
+	if (err)
+		goto out;
+
+	err = nfs_idmap_init();
+	if (err)
+		goto out1;
+
+	err = nfs4_register_sysctl();
+	if (err)
+		goto out2;
+
+	register_nfs_version(&nfs_v4);
+	return 0;
+out2:
+	nfs_idmap_quit();
+out1:
+	nfs_dns_resolver_destroy();
+out:
+	return err;
+}
+
+static void __exit exit_nfs_v4(void)
+{
+	unregister_nfs_version(&nfs_v4);
+	nfs4_unregister_sysctl();
+	nfs_idmap_quit();
+	nfs_dns_resolver_destroy();
+}
+
+MODULE_LICENSE("GPL");
+
+module_init(init_nfs_v4);
+module_exit(exit_nfs_v4);

fs/nfs/nfs4sysctl.c

@@ -0,0 +1,69 @@
+/*
+ * linux/fs/nfs/nfs4sysctl.c
+ *
+ * Sysctl interface to NFS v4 parameters
+ *
+ * Copyright (c) 2006 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/sysctl.h>
+#include <linux/nfs_idmap.h>
+#include <linux/nfs_fs.h>
+
+#include "nfs4_fs.h"
+#include "callback.h"
+
+static const int nfs_set_port_min = 0;
+static const int nfs_set_port_max = 65535;
+static struct ctl_table_header *nfs4_callback_sysctl_table;
+
+static struct ctl_table nfs4_cb_sysctls[] = {
+	{
+		.procname = "nfs_callback_tcpport",
+		.data = &nfs_callback_set_tcpport,
+		.maxlen = sizeof(int),
+		.mode = 0644,
+		.proc_handler = proc_dointvec_minmax,
+		.extra1 = (int *)&nfs_set_port_min,
+		.extra2 = (int *)&nfs_set_port_max,
+	},
+	{
+		.procname = "idmap_cache_timeout",
+		.data = &nfs_idmap_cache_timeout,
+		.maxlen = sizeof(int),
+		.mode = 0644,
+		.proc_handler = proc_dointvec_jiffies,
+	},
+	{ }
+};
+
+static struct ctl_table nfs4_cb_sysctl_dir[] = {
+	{
+		.procname = "nfs",
+		.mode = 0555,
+		.child = nfs4_cb_sysctls,
+	},
+	{ }
+};
+
+static struct ctl_table nfs4_cb_sysctl_root[] = {
+	{
+		.procname = "fs",
+		.mode = 0555,
+		.child = nfs4_cb_sysctl_dir,
+	},
+	{ }
+};
+
+int nfs4_register_sysctl(void)
+{
+	nfs4_callback_sysctl_table = register_sysctl_table(nfs4_cb_sysctl_root);
+	if (nfs4_callback_sysctl_table == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+void nfs4_unregister_sysctl(void)
+{
+	unregister_sysctl_table(nfs4_callback_sysctl_table);
+	nfs4_callback_sysctl_table = NULL;
+}

fs/nfs/nfs4trace.c

@@ -0,0 +1,17 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/nfs_fs.h>
+#include "nfs4_fs.h"
+#include "internal.h"
+#include "nfs4session.h"
+#include "callback.h"
+
+#define CREATE_TRACE_POINTS
+#include "nfs4trace.h"
+
+#ifdef CONFIG_NFS_V4_1
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read);
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write);
+EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds);
+#endif

fs/nfs/nfsroot.c

@@ -0,0 +1,309 @@
+/*
+ *  Copyright (C) 1995, 1996  Gero Kuhlmann <gero@gkminix.han.de>
+ *
+ *  Allow an NFS filesystem to be mounted as root. The way this works is:
+ *     (1) Use the IP autoconfig mechanism to set local IP addresses and routes.
+ *     (2) Construct the device string and the options string using DHCP
+ *         option 17 and/or kernel command line options.
+ *     (3) When mount_root() sets up the root file system, pass these strings
+ *         to the NFS client's regular mount interface via sys_mount().
+ *
+ *
+ *	Changes:
+ *
+ *	Alan Cox	:	Removed get_address name clash with FPU.
+ *	Alan Cox	:	Reformatted a bit.
+ *	Gero Kuhlmann	:	Code cleanup
+ *	Michael Rausch  :	Fixed recognition of an incoming RARP answer.
+ *	Martin Mares	: (2.0)	Auto-configuration via BOOTP supported.
+ *	Martin Mares	:	Manual selection of interface & BOOTP/RARP.
+ *	Martin Mares	:	Using network routes instead of host routes,
+ *				allowing the default configuration to be used
+ *				for normal operation of the host.
+ *	Martin Mares	:	Randomized timer with exponential backoff
+ *				installed to minimize network congestion.
+ *	Martin Mares	:	Code cleanup.
+ *	Martin Mares	: (2.1)	BOOTP and RARP made configuration options.
+ *	Martin Mares	:	Server hostname generation fixed.
+ *	Gerd Knorr	:	Fixed wired inode handling
+ *	Martin Mares	: (2.2)	"0.0.0.0" addresses from command line ignored.
+ *	Martin Mares	:	RARP replies not tested for server address.
+ *	Gero Kuhlmann	: (2.3) Some bug fixes and code cleanup again (please
+ *				send me your new patches _before_ bothering
+ *				Linus so that I don' always have to cleanup
+ *				_afterwards_ - thanks)
+ *	Gero Kuhlmann	:	Last changes of Martin Mares undone.
+ *	Gero Kuhlmann	: 	RARP replies are tested for specified server
+ *				again. However, it's now possible to have
+ *				different RARP and NFS servers.
+ *	Gero Kuhlmann	:	"0.0.0.0" addresses from command line are
+ *				now mapped to INADDR_NONE.
+ *	Gero Kuhlmann	:	Fixed a bug which prevented BOOTP path name
+ *				from being used (thanks to Leo Spiekman)
+ *	Andy Walker	:	Allow to specify the NFS server in nfs_root
+ *				without giving a path name
+ *	Swen Thümmler	:	Allow to specify the NFS options in nfs_root
+ *				without giving a path name. Fix BOOTP request
+ *				for domainname (domainname is NIS domain, not
+ *				DNS domain!). Skip dummy devices for BOOTP.
+ *	Jacek Zapala	:	Fixed a bug which prevented server-ip address
+ *				from nfsroot parameter from being used.
+ *	Olaf Kirch	:	Adapted to new NFS code.
+ *	Jakub Jelinek	:	Free used code segment.
+ *	Marko Kohtala	:	Fixed some bugs.
+ *	Martin Mares	:	Debug message cleanup
+ *	Martin Mares	:	Changed to use the new generic IP layer autoconfig
+ *				code. BOOTP and RARP moved there.
+ *	Martin Mares	:	Default path now contains host name instead of
+ *				host IP address (but host name defaults to IP
+ *				address anyway).
+ *	Martin Mares	:	Use root_server_addr appropriately during setup.
+ *	Martin Mares	:	Rewrote parameter parsing, now hopefully giving
+ *				correct overriding.
+ *	Trond Myklebust :	Add in preliminary support for NFSv3 and TCP.
+ *				Fix bug in root_nfs_addr(). nfs_data.namlen
+ *				is NOT for the length of the hostname.
+ *	Hua Qin		:	Support for mounting root file system via
+ *				NFS over TCP.
+ *	Fabian Frederick:	Option parser rebuilt (using parser lib)
+ *	Chuck Lever	:	Use super.c's text-based mount option parsing
+ *	Chuck Lever	:	Add "nfsrootdebug".
+ */
+
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/nfs.h>
+#include <linux/nfs_fs.h>
+#include <linux/utsname.h>
+#include <linux/root_dev.h>
+#include <net/ipconfig.h>
+
+#include "internal.h"
+
+#define NFSDBG_FACILITY NFSDBG_ROOT
+
+/* Default path we try to mount. "%s" gets replaced by our IP address */
+#define NFS_ROOT		"/tftpboot/%s"
+
+/* Default NFSROOT mount options. */
+#define NFS_DEF_OPTIONS		"vers=2,udp,rsize=4096,wsize=4096"
+
+/* Parameters passed from the kernel command line */
+static char nfs_root_parms[256] __initdata = "";
+
+/* Text-based mount options passed to super.c */
+static char nfs_root_options[256] __initdata = NFS_DEF_OPTIONS;
+
+/* Address of NFS server */
+static __be32 servaddr __initdata = htonl(INADDR_NONE);
+
+/* Name of directory to mount */
+static char nfs_export_path[NFS_MAXPATHLEN + 1] __initdata = "";
+
+/* server:export path string passed to super.c */
+static char nfs_root_device[NFS_MAXPATHLEN + 1] __initdata = "";
+
+#ifdef NFS_DEBUG
+/*
+ * When the "nfsrootdebug" kernel command line option is specified,
+ * enable debugging messages for NFSROOT.
+ */
+static int __init nfs_root_debug(char *__unused)
+{
+	nfs_debug |= NFSDBG_ROOT | NFSDBG_MOUNT;
+	return 1;
+}
+
+__setup("nfsrootdebug", nfs_root_debug);
+#endif
+
+/*
+ *  Parse NFS server and directory information passed on the kernel
+ *  command line.
+ *
+ *  nfsroot=[<server-ip>:]<root-dir>[,<nfs-options>]
+ *
+ *  If there is a "%s" token in the <root-dir> string, it is replaced
+ *  by the ASCII-representation of the client's IP address.
+ */
+static int __init nfs_root_setup(char *line)
+{
+	ROOT_DEV = Root_NFS;
+
+	if (line[0] == '/' || line[0] == ',' || (line[0] >= '0' && line[0] <= '9')) {
+		strlcpy(nfs_root_parms, line, sizeof(nfs_root_parms));
+	} else {
+		size_t n = strlen(line) + sizeof(NFS_ROOT) - 1;
+		if (n >= sizeof(nfs_root_parms))
+			line[sizeof(nfs_root_parms) - sizeof(NFS_ROOT) - 2] = '\0';
+		sprintf(nfs_root_parms, NFS_ROOT, line);
+	}
+
+	/*
+	 * Extract the IP address of the NFS server containing our
+	 * root file system, if one was specified.
+	 *
+	 * Note: root_nfs_parse_addr() removes the server-ip from
+	 *	 nfs_root_parms, if it exists.
+	 */
+	root_server_addr = root_nfs_parse_addr(nfs_root_parms);
+
+	return 1;
+}
+
+__setup("nfsroot=", nfs_root_setup);
+
+static int __init root_nfs_copy(char *dest, const char *src,
+				     const size_t destlen)
+{
+	if (strlcpy(dest, src, destlen) > destlen)
+		return -1;
+	return 0;
+}
+
+static int __init root_nfs_cat(char *dest, const char *src,
+			       const size_t destlen)
+{
+	size_t len = strlen(dest);
+
+	if (len && dest[len - 1] != ',')
+		if (strlcat(dest, ",", destlen) > destlen)
+			return -1;
+
+	if (strlcat(dest, src, destlen) > destlen)
+		return -1;
+	return 0;
+}
+
+/*
+ * Parse out root export path and mount options from
+ * passed-in string @incoming.
+ *
+ * Copy the export path into @exppath.
+ */
+static int __init root_nfs_parse_options(char *incoming, char *exppath,
+					 const size_t exppathlen)
+{
+	char *p;
+
+	/*
+	 * Set the NFS remote path
+	 */
+	p = strsep(&incoming, ",");
+	if (*p != '\0' && strcmp(p, "default") != 0)
+		if (root_nfs_copy(exppath, p, exppathlen))
+			return -1;
+
+	/*
+	 * @incoming now points to the rest of the string; if it
+	 * contains something, append it to our root options buffer
+	 */
+	if (incoming != NULL && *incoming != '\0')
+		if (root_nfs_cat(nfs_root_options, incoming,
+						sizeof(nfs_root_options)))
+			return -1;
+	return 0;
+}
+
+/*
+ *  Decode the export directory path name and NFS options from
+ *  the kernel command line.  This has to be done late in order to
+ *  use a dynamically acquired client IP address for the remote
+ *  root directory path.
+ *
+ *  Returns zero if successful; otherwise -1 is returned.
+ */
+static int __init root_nfs_data(char *cmdline)
+{
+	char mand_options[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1];
+	int len, retval = -1;
+	char *tmp = NULL;
+	const size_t tmplen = sizeof(nfs_export_path);
+
+	tmp = kzalloc(tmplen, GFP_KERNEL);
+	if (tmp == NULL)
+		goto out_nomem;
+	strcpy(tmp, NFS_ROOT);
+
+	if (root_server_path[0] != '\0') {
+		dprintk("Root-NFS: DHCPv4 option 17: %s\n",
+			root_server_path);
+		if (root_nfs_parse_options(root_server_path, tmp, tmplen))
+			goto out_optionstoolong;
+	}
+
+	if (cmdline[0] != '\0') {
+		dprintk("Root-NFS: nfsroot=%s\n", cmdline);
+		if (root_nfs_parse_options(cmdline, tmp, tmplen))
+			goto out_optionstoolong;
+	}
+
+	/*
+	 * Append mandatory options for nfsroot so they override
+	 * what has come before
+	 */
+	snprintf(mand_options, sizeof(mand_options), "nolock,addr=%pI4",
+			&servaddr);
+	if (root_nfs_cat(nfs_root_options, mand_options,
+						sizeof(nfs_root_options)))
+		goto out_optionstoolong;
+
+	/*
+	 * Set up nfs_root_device.  For NFS mounts, this looks like
+	 *
+	 *	server:/path
+	 *
+	 * At this point, utsname()->nodename contains our local
+	 * IP address or hostname, set by ipconfig.  If "%s" exists
+	 * in tmp, substitute the nodename, then shovel the whole
+	 * mess into nfs_root_device.
+	 */
+	len = snprintf(nfs_export_path, sizeof(nfs_export_path),
+				tmp, utsname()->nodename);
+	if (len > (int)sizeof(nfs_export_path))
+		goto out_devnametoolong;
+	len = snprintf(nfs_root_device, sizeof(nfs_root_device),
+				"%pI4:%s", &servaddr, nfs_export_path);
+	if (len > (int)sizeof(nfs_root_device))
+		goto out_devnametoolong;
+
+	retval = 0;
+
+out:
+	kfree(tmp);
+	return retval;
+out_nomem:
+	printk(KERN_ERR "Root-NFS: could not allocate memory\n");
+	goto out;
+out_optionstoolong:
+	printk(KERN_ERR "Root-NFS: mount options string too long\n");
+	goto out;
+out_devnametoolong:
+	printk(KERN_ERR "Root-NFS: root device name too long.\n");
+	goto out;
+}
+
+/**
+ * nfs_root_data - Return prepared 'data' for NFSROOT mount
+ * @root_device: OUT: address of string containing NFSROOT device
+ * @root_data: OUT: address of string containing NFSROOT mount options
+ *
+ * Returns zero and sets @root_device and @root_data if successful,
+ * otherwise -1 is returned.
+ */
+int __init nfs_root_data(char **root_device, char **root_data)
+{
+	servaddr = root_server_addr;
+	if (servaddr == htonl(INADDR_NONE)) {
+		printk(KERN_ERR "Root-NFS: no NFS server address\n");
+		return -1;
+	}
+
+	if (root_nfs_data(nfs_root_parms) < 0)
+		return -1;
+
+	*root_device = nfs_root_device;
+	*root_data = nfs_root_options;
+	return 0;
+}

fs/nfs/nfstrace.c

@@ -0,0 +1,9 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#include <linux/nfs_fs.h>
+#include <linux/namei.h>
+#include "internal.h"
+
+#define CREATE_TRACE_POINTS
+#include "nfstrace.h"

fs/nfs/nfstrace.h

@@ -0,0 +1,730 @@
+/*
+ * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com>
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM nfs
+
+#if !defined(_TRACE_NFS_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NFS_H
+
+#include <linux/tracepoint.h>
+
+#define nfs_show_file_type(ftype) \
+	__print_symbolic(ftype, \
+			{ DT_UNKNOWN, "UNKNOWN" }, \
+			{ DT_FIFO, "FIFO" }, \
+			{ DT_CHR, "CHR" }, \
+			{ DT_DIR, "DIR" }, \
+			{ DT_BLK, "BLK" }, \
+			{ DT_REG, "REG" }, \
+			{ DT_LNK, "LNK" }, \
+			{ DT_SOCK, "SOCK" }, \
+			{ DT_WHT, "WHT" })
+
+#define nfs_show_cache_validity(v) \
+	__print_flags(v, "|", \
+			{ NFS_INO_INVALID_ATTR, "INVALID_ATTR" }, \
+			{ NFS_INO_INVALID_DATA, "INVALID_DATA" }, \
+			{ NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \
+			{ NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \
+			{ NFS_INO_INVALID_ACL, "INVALID_ACL" }, \
+			{ NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \
+			{ NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \
+			{ NFS_INO_INVALID_LABEL, "INVALID_LABEL" })
+
+#define nfs_show_nfsi_flags(v) \
+	__print_flags(v, "|", \
+			{ 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \
+			{ 1 << NFS_INO_STALE, "STALE" }, \
+			{ 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \
+			{ 1 << NFS_INO_FLUSHING, "FLUSHING" }, \
+			{ 1 << NFS_INO_FSCACHE, "FSCACHE" }, \
+			{ 1 << NFS_INO_COMMIT, "COMMIT" }, \
+			{ 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \
+			{ 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" })
+
+DECLARE_EVENT_CLASS(nfs_inode_event,
+		TP_PROTO(
+			const struct inode *inode
+		),
+
+		TP_ARGS(inode),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(u64, version)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->version = inode->i_version;
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu ",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			(unsigned long long)__entry->version
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_inode_event_done,
+		TP_PROTO(
+			const struct inode *inode,
+			int error
+		),
+
+		TP_ARGS(inode, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(unsigned char, type)
+			__field(u64, fileid)
+			__field(u64, version)
+			__field(loff_t, size)
+			__field(unsigned long, nfsi_flags)
+			__field(unsigned long, cache_validity)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->error = error;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->type = nfs_umode_to_dtype(inode->i_mode);
+			__entry->version = inode->i_version;
+			__entry->size = i_size_read(inode);
+			__entry->nfsi_flags = nfsi->flags;
+			__entry->cache_validity = nfsi->cache_validity;
+		),
+
+		TP_printk(
+			"error=%d fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"type=%u (%s) version=%llu size=%lld "
+			"cache_validity=%lu (%s) nfs_flags=%ld (%s)",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->type,
+			nfs_show_file_type(__entry->type),
+			(unsigned long long)__entry->version,
+			(long long)__entry->size,
+			__entry->cache_validity,
+			nfs_show_cache_validity(__entry->cache_validity),
+			__entry->nfsi_flags,
+			nfs_show_nfsi_flags(__entry->nfsi_flags)
+		)
+);
+
+#define DEFINE_NFS_INODE_EVENT(name) \
+	DEFINE_EVENT(nfs_inode_event, name, \
+			TP_PROTO( \
+				const struct inode *inode \
+			), \
+			TP_ARGS(inode))
+#define DEFINE_NFS_INODE_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_inode_event_done, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				int error \
+			), \
+			TP_ARGS(inode, error))
+DEFINE_NFS_INODE_EVENT(nfs_refresh_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_refresh_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_revalidate_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_revalidate_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_invalidate_mapping_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_invalidate_mapping_exit);
+DEFINE_NFS_INODE_EVENT(nfs_getattr_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_getattr_exit);
+DEFINE_NFS_INODE_EVENT(nfs_setattr_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_setattr_exit);
+DEFINE_NFS_INODE_EVENT(nfs_writeback_page_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_page_exit);
+DEFINE_NFS_INODE_EVENT(nfs_writeback_inode_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit);
+DEFINE_NFS_INODE_EVENT(nfs_fsync_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit);
+DEFINE_NFS_INODE_EVENT(nfs_access_enter);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit);
+
+#define show_lookup_flags(flags) \
+	__print_flags((unsigned long)flags, "|", \
+			{ LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \
+			{ LOOKUP_DIRECTORY, "DIRECTORY" }, \
+			{ LOOKUP_OPEN, "OPEN" }, \
+			{ LOOKUP_CREATE, "CREATE" }, \
+			{ LOOKUP_EXCL, "EXCL" })
+
+DECLARE_EVENT_CLASS(nfs_lookup_event,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, dentry, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_lookup_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_LOOKUP_EVENT(name) \
+	DEFINE_EVENT(nfs_lookup_event, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				unsigned int flags \
+			), \
+			TP_ARGS(dir, dentry, flags))
+
+DECLARE_EVENT_CLASS(nfs_lookup_event_done,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_lookup_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_LOOKUP_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_lookup_event_done, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				unsigned int flags, \
+				int error \
+			), \
+			TP_ARGS(dir, dentry, flags, error))
+
+DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_enter);
+DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit);
+DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter);
+DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit);
+
+#define show_open_flags(flags) \
+	__print_flags((unsigned long)flags, "|", \
+		{ O_CREAT, "O_CREAT" }, \
+		{ O_EXCL, "O_EXCL" }, \
+		{ O_TRUNC, "O_TRUNC" }, \
+		{ O_APPEND, "O_APPEND" }, \
+		{ O_DSYNC, "O_DSYNC" }, \
+		{ O_DIRECT, "O_DIRECT" }, \
+		{ O_DIRECTORY, "O_DIRECTORY" })
+
+#define show_fmode_flags(mode) \
+	__print_flags(mode, "|", \
+		{ ((__force unsigned long)FMODE_READ), "READ" }, \
+		{ ((__force unsigned long)FMODE_WRITE), "WRITE" }, \
+		{ ((__force unsigned long)FMODE_EXEC), "EXEC" })
+
+TRACE_EVENT(nfs_atomic_open_enter,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct nfs_open_context *ctx,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, ctx, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(unsigned int, fmode)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, ctx->dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__entry->fmode = (__force unsigned int)ctx->mode;
+			__assign_str(name, ctx->dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) fmode=%s name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			show_fmode_flags(__entry->fmode),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_atomic_open_exit,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct nfs_open_context *ctx,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, ctx, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(unsigned int, fmode)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, ctx->dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__entry->fmode = (__force unsigned int)ctx->mode;
+			__assign_str(name, ctx->dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) fmode=%s "
+			"name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			show_fmode_flags(__entry->fmode),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_create_enter,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags
+		),
+
+		TP_ARGS(dir, dentry, flags),
+
+		TP_STRUCT__entry(
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_create_exit,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			unsigned int flags,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, flags, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(unsigned int, flags)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->flags = flags;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
+			__entry->error,
+			__entry->flags,
+			show_open_flags(__entry->flags),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_directory_event,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry
+		),
+
+		TP_ARGS(dir, dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_DIRECTORY_EVENT(name) \
+	DEFINE_EVENT(nfs_directory_event, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry \
+			), \
+			TP_ARGS(dir, dentry))
+
+DECLARE_EVENT_CLASS(nfs_directory_event_done,
+		TP_PROTO(
+			const struct inode *dir,
+			const struct dentry *dentry,
+			int error
+		),
+
+		TP_ARGS(dir, dentry, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+#define DEFINE_NFS_DIRECTORY_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_directory_event_done, name, \
+			TP_PROTO( \
+				const struct inode *dir, \
+				const struct dentry *dentry, \
+				int error \
+			), \
+			TP_ARGS(dir, dentry, error))
+
+DEFINE_NFS_DIRECTORY_EVENT(nfs_mknod_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mknod_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_mkdir_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mkdir_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_rmdir_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_rmdir_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_remove_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_remove_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_unlink_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_unlink_exit);
+DEFINE_NFS_DIRECTORY_EVENT(nfs_symlink_enter);
+DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_symlink_exit);
+
+TRACE_EVENT(nfs_link_enter,
+		TP_PROTO(
+			const struct inode *inode,
+			const struct inode *dir,
+			const struct dentry *dentry
+		),
+
+		TP_ARGS(inode, dir, dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, fileid)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dir = NFS_FILEID(dir);
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			__entry->fileid,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+TRACE_EVENT(nfs_link_exit,
+		TP_PROTO(
+			const struct inode *inode,
+			const struct inode *dir,
+			const struct dentry *dentry,
+			int error
+		),
+
+		TP_ARGS(inode, dir, dentry, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u64, fileid)
+			__field(u64, dir)
+			__string(name, dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			__assign_str(name, dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			__entry->fileid,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_rename_event,
+		TP_PROTO(
+			const struct inode *old_dir,
+			const struct dentry *old_dentry,
+			const struct inode *new_dir,
+			const struct dentry *new_dentry
+		),
+
+		TP_ARGS(old_dir, old_dentry, new_dir, new_dentry),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u64, old_dir)
+			__field(u64, new_dir)
+			__string(old_name, old_dentry->d_name.name)
+			__string(new_name, new_dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = old_dir->i_sb->s_dev;
+			__entry->old_dir = NFS_FILEID(old_dir);
+			__entry->new_dir = NFS_FILEID(new_dir);
+			__assign_str(old_name, old_dentry->d_name.name);
+			__assign_str(new_name, new_dentry->d_name.name);
+		),
+
+		TP_printk(
+			"old_name=%02x:%02x:%llu/%s new_name=%02x:%02x:%llu/%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->old_dir,
+			__get_str(old_name),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->new_dir,
+			__get_str(new_name)
+		)
+);
+#define DEFINE_NFS_RENAME_EVENT(name) \
+	DEFINE_EVENT(nfs_rename_event, name, \
+			TP_PROTO( \
+				const struct inode *old_dir, \
+				const struct dentry *old_dentry, \
+				const struct inode *new_dir, \
+				const struct dentry *new_dentry \
+			), \
+			TP_ARGS(old_dir, old_dentry, new_dir, new_dentry))
+
+DECLARE_EVENT_CLASS(nfs_rename_event_done,
+		TP_PROTO(
+			const struct inode *old_dir,
+			const struct dentry *old_dentry,
+			const struct inode *new_dir,
+			const struct dentry *new_dentry,
+			int error
+		),
+
+		TP_ARGS(old_dir, old_dentry, new_dir, new_dentry, error),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(int, error)
+			__field(u64, old_dir)
+			__string(old_name, old_dentry->d_name.name)
+			__field(u64, new_dir)
+			__string(new_name, new_dentry->d_name.name)
+		),
+
+		TP_fast_assign(
+			__entry->dev = old_dir->i_sb->s_dev;
+			__entry->old_dir = NFS_FILEID(old_dir);
+			__entry->new_dir = NFS_FILEID(new_dir);
+			__entry->error = error;
+			__assign_str(old_name, old_dentry->d_name.name);
+			__assign_str(new_name, new_dentry->d_name.name);
+		),
+
+		TP_printk(
+			"error=%d old_name=%02x:%02x:%llu/%s "
+			"new_name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->old_dir,
+			__get_str(old_name),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->new_dir,
+			__get_str(new_name)
+		)
+);
+#define DEFINE_NFS_RENAME_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_rename_event_done, name, \
+			TP_PROTO( \
+				const struct inode *old_dir, \
+				const struct dentry *old_dentry, \
+				const struct inode *new_dir, \
+				const struct dentry *new_dentry, \
+				int error \
+			), \
+			TP_ARGS(old_dir, old_dentry, new_dir, \
+				new_dentry, error))
+
+DEFINE_NFS_RENAME_EVENT(nfs_rename_enter);
+DEFINE_NFS_RENAME_EVENT_DONE(nfs_rename_exit);
+
+DEFINE_NFS_RENAME_EVENT_DONE(nfs_sillyrename_rename);
+
+TRACE_EVENT(nfs_sillyrename_unlink,
+		TP_PROTO(
+			const struct nfs_unlinkdata *data,
+			int error
+		),
+
+		TP_ARGS(data, error),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(int, error)
+			__field(u64, dir)
+			__dynamic_array(char, name, data->args.name.len + 1)
+		),
+
+		TP_fast_assign(
+			struct inode *dir = data->dir;
+			size_t len = data->args.name.len;
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->dir = NFS_FILEID(dir);
+			__entry->error = error;
+			memcpy(__get_dynamic_array(name),
+				data->args.name.name, len);
+			((char *)__get_dynamic_array(name))[len] = 0;
+		),
+
+		TP_printk(
+			"error=%d name=%02x:%02x:%llu/%s",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->dir,
+			__get_str(name)
+		)
+);
+#endif /* _TRACE_NFS_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE nfstrace
+/* This part must be outside protection */
+#include <trace/define_trace.h>

fs/nfs/objlayout/Kbuild

@@ -0,0 +1,5 @@
+#
+# Makefile for the pNFS Objects Layout Driver kernel module
+#
+objlayoutdriver-y := objio_osd.o pnfs_osd_xdr_cli.o objlayout.o
+obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayoutdriver.o

fs/nfs/objlayout/objio_osd.c

@@ -0,0 +1,673 @@
+/*
+ *  pNFS Objects layout implementation over open-osd initiator library
+ *
+ *  Copyright (C) 2009 Panasas Inc. [year of first publication]
+ *  All rights reserved.
+ *
+ *  Benny Halevy <bhalevy@panasas.com>
+ *  Boaz Harrosh <ooo@electrozaur.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2
+ *  See the file COPYING included with this distribution for more details.
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the Panasas company 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 ``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 REGENTS 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.
+ */
+
+#include <linux/module.h>
+#include <scsi/osd_ore.h>
+
+#include "objlayout.h"
+#include "../internal.h"
+
+#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
+
+struct objio_dev_ent {
+	struct nfs4_deviceid_node id_node;
+	struct ore_dev od;
+};
+
+static void
+objio_free_deviceid_node(struct nfs4_deviceid_node *d)
+{
+	struct objio_dev_ent *de = container_of(d, struct objio_dev_ent, id_node);
+
+	dprintk("%s: free od=%p\n", __func__, de->od.od);
+	osduld_put_device(de->od.od);
+	kfree(de);
+}
+
+struct objio_segment {
+	struct pnfs_layout_segment lseg;
+
+	struct ore_layout layout;
+	struct ore_components oc;
+};
+
+static inline struct objio_segment *
+OBJIO_LSEG(struct pnfs_layout_segment *lseg)
+{
+	return container_of(lseg, struct objio_segment, lseg);
+}
+
+struct objio_state {
+	/* Generic layer */
+	struct objlayout_io_res oir;
+
+	bool sync;
+	/*FIXME: Support for extra_bytes at ore_get_rw_state() */
+	struct ore_io_state *ios;
+};
+
+/* Send and wait for a get_device_info of devices in the layout,
+   then look them up with the osd_initiator library */
+struct nfs4_deviceid_node *
+objio_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
+			gfp_t gfp_flags)
+{
+	struct pnfs_osd_deviceaddr *deviceaddr;
+	struct objio_dev_ent *ode = NULL;
+	struct osd_dev *od;
+	struct osd_dev_info odi;
+	bool retry_flag = true;
+	__be32 *p;
+	int err;
+
+	deviceaddr = kzalloc(sizeof(*deviceaddr), gfp_flags);
+	if (!deviceaddr)
+		return NULL;
+
+	p = page_address(pdev->pages[0]);
+	pnfs_osd_xdr_decode_deviceaddr(deviceaddr, p);
+
+	odi.systemid_len = deviceaddr->oda_systemid.len;
+	if (odi.systemid_len > sizeof(odi.systemid)) {
+		dprintk("%s: odi.systemid_len > sizeof(systemid=%zd)\n",
+			__func__, sizeof(odi.systemid));
+		err = -EINVAL;
+		goto out;
+	} else if (odi.systemid_len)
+		memcpy(odi.systemid, deviceaddr->oda_systemid.data,
+		       odi.systemid_len);
+	odi.osdname_len	 = deviceaddr->oda_osdname.len;
+	odi.osdname	 = (u8 *)deviceaddr->oda_osdname.data;
+
+	if (!odi.osdname_len && !odi.systemid_len) {
+		dprintk("%s: !odi.osdname_len && !odi.systemid_len\n",
+			__func__);
+		err = -ENODEV;
+		goto out;
+	}
+
+retry_lookup:
+	od = osduld_info_lookup(&odi);
+	if (unlikely(IS_ERR(od))) {
+		err = PTR_ERR(od);
+		dprintk("%s: osduld_info_lookup => %d\n", __func__, err);
+		if (err == -ENODEV && retry_flag) {
+			err = objlayout_autologin(deviceaddr);
+			if (likely(!err)) {
+				retry_flag = false;
+				goto retry_lookup;
+			}
+		}
+		goto out;
+	}
+
+	dprintk("Adding new dev_id(%llx:%llx)\n",
+		_DEVID_LO(&pdev->dev_id), _DEVID_HI(&pdev->dev_id));
+
+	ode = kzalloc(sizeof(*ode), gfp_flags);
+	if (!ode) {
+		dprintk("%s: -ENOMEM od=%p\n", __func__, od);
+		goto out;
+	}
+
+	nfs4_init_deviceid_node(&ode->id_node, server, &pdev->dev_id);
+	kfree(deviceaddr);
+
+	ode->od.od = od;
+	return &ode->id_node;
+
+out:
+	kfree(deviceaddr);
+	return NULL;
+}
+
+static void copy_single_comp(struct ore_components *oc, unsigned c,
+			     struct pnfs_osd_object_cred *src_comp)
+{
+	struct ore_comp *ocomp = &oc->comps[c];
+
+	WARN_ON(src_comp->oc_cap_key.cred_len > 0); /* libosd is NO_SEC only */
+	WARN_ON(src_comp->oc_cap.cred_len > sizeof(ocomp->cred));
+
+	ocomp->obj.partition = src_comp->oc_object_id.oid_partition_id;
+	ocomp->obj.id = src_comp->oc_object_id.oid_object_id;
+
+	memcpy(ocomp->cred, src_comp->oc_cap.cred, sizeof(ocomp->cred));
+}
+
+static int __alloc_objio_seg(unsigned numdevs, gfp_t gfp_flags,
+		       struct objio_segment **pseg)
+{
+/*	This is the in memory structure of the objio_segment
+ *
+ *	struct __alloc_objio_segment {
+ *		struct objio_segment olseg;
+ *		struct ore_dev *ods[numdevs];
+ *		struct ore_comp	comps[numdevs];
+ *	} *aolseg;
+ *	NOTE: The code as above compiles and runs perfectly. It is elegant,
+ *	type safe and compact. At some Past time Linus has decided he does not
+ *	like variable length arrays, For the sake of this principal we uglify
+ *	the code as below.
+ */
+	struct objio_segment *lseg;
+	size_t lseg_size = sizeof(*lseg) +
+			numdevs * sizeof(lseg->oc.ods[0]) +
+			numdevs * sizeof(*lseg->oc.comps);
+
+	lseg = kzalloc(lseg_size, gfp_flags);
+	if (unlikely(!lseg)) {
+		dprintk("%s: Failed allocation numdevs=%d size=%zd\n", __func__,
+			numdevs, lseg_size);
+		return -ENOMEM;
+	}
+
+	lseg->oc.numdevs = numdevs;
+	lseg->oc.single_comp = EC_MULTPLE_COMPS;
+	lseg->oc.ods = (void *)(lseg + 1);
+	lseg->oc.comps = (void *)(lseg->oc.ods + numdevs);
+
+	*pseg = lseg;
+	return 0;
+}
+
+int objio_alloc_lseg(struct pnfs_layout_segment **outp,
+	struct pnfs_layout_hdr *pnfslay,
+	struct pnfs_layout_range *range,
+	struct xdr_stream *xdr,
+	gfp_t gfp_flags)
+{
+	struct nfs_server *server = NFS_SERVER(pnfslay->plh_inode);
+	struct objio_segment *objio_seg;
+	struct pnfs_osd_xdr_decode_layout_iter iter;
+	struct pnfs_osd_layout layout;
+	struct pnfs_osd_object_cred src_comp;
+	unsigned cur_comp;
+	int err;
+
+	err = pnfs_osd_xdr_decode_layout_map(&layout, &iter, xdr);
+	if (unlikely(err))
+		return err;
+
+	err = __alloc_objio_seg(layout.olo_num_comps, gfp_flags, &objio_seg);
+	if (unlikely(err))
+		return err;
+
+	objio_seg->layout.stripe_unit = layout.olo_map.odm_stripe_unit;
+	objio_seg->layout.group_width = layout.olo_map.odm_group_width;
+	objio_seg->layout.group_depth = layout.olo_map.odm_group_depth;
+	objio_seg->layout.mirrors_p1 = layout.olo_map.odm_mirror_cnt + 1;
+	objio_seg->layout.raid_algorithm = layout.olo_map.odm_raid_algorithm;
+
+	err = ore_verify_layout(layout.olo_map.odm_num_comps,
+					  &objio_seg->layout);
+	if (unlikely(err))
+		goto err;
+
+	objio_seg->oc.first_dev = layout.olo_comps_index;
+	cur_comp = 0;
+	while (pnfs_osd_xdr_decode_layout_comp(&src_comp, &iter, xdr, &err)) {
+		struct nfs4_deviceid_node *d;
+		struct objio_dev_ent *ode;
+
+		copy_single_comp(&objio_seg->oc, cur_comp, &src_comp);
+
+		d = nfs4_find_get_deviceid(server,
+				&src_comp.oc_object_id.oid_device_id,
+				pnfslay->plh_lc_cred, gfp_flags);
+		if (!d) {
+			err = -ENXIO;
+			goto err;
+		}
+
+		ode = container_of(d, struct objio_dev_ent, id_node);
+		objio_seg->oc.ods[cur_comp++] = &ode->od;
+	}
+	/* pnfs_osd_xdr_decode_layout_comp returns false on error */
+	if (unlikely(err))
+		goto err;
+
+	*outp = &objio_seg->lseg;
+	return 0;
+
+err:
+	kfree(objio_seg);
+	dprintk("%s: Error: return %d\n", __func__, err);
+	*outp = NULL;
+	return err;
+}
+
+void objio_free_lseg(struct pnfs_layout_segment *lseg)
+{
+	int i;
+	struct objio_segment *objio_seg = OBJIO_LSEG(lseg);
+
+	for (i = 0; i < objio_seg->oc.numdevs; i++) {
+		struct ore_dev *od = objio_seg->oc.ods[i];
+		struct objio_dev_ent *ode;
+
+		if (!od)
+			break;
+		ode = container_of(od, typeof(*ode), od);
+		nfs4_put_deviceid_node(&ode->id_node);
+	}
+	kfree(objio_seg);
+}
+
+static int
+objio_alloc_io_state(struct pnfs_layout_hdr *pnfs_layout_type, bool is_reading,
+	struct pnfs_layout_segment *lseg, struct page **pages, unsigned pgbase,
+	loff_t offset, size_t count, void *rpcdata, gfp_t gfp_flags,
+	struct objio_state **outp)
+{
+	struct objio_segment *objio_seg = OBJIO_LSEG(lseg);
+	struct ore_io_state *ios;
+	int ret;
+	struct __alloc_objio_state {
+		struct objio_state objios;
+		struct pnfs_osd_ioerr ioerrs[objio_seg->oc.numdevs];
+	} *aos;
+
+	aos = kzalloc(sizeof(*aos), gfp_flags);
+	if (unlikely(!aos))
+		return -ENOMEM;
+
+	objlayout_init_ioerrs(&aos->objios.oir, objio_seg->oc.numdevs,
+			aos->ioerrs, rpcdata, pnfs_layout_type);
+
+	ret = ore_get_rw_state(&objio_seg->layout, &objio_seg->oc, is_reading,
+			       offset, count, &ios);
+	if (unlikely(ret)) {
+		kfree(aos);
+		return ret;
+	}
+
+	ios->pages = pages;
+	ios->pgbase = pgbase;
+	ios->private = aos;
+	BUG_ON(ios->nr_pages > (pgbase + count + PAGE_SIZE - 1) >> PAGE_SHIFT);
+
+	aos->objios.sync = 0;
+	aos->objios.ios = ios;
+	*outp = &aos->objios;
+	return 0;
+}
+
+void objio_free_result(struct objlayout_io_res *oir)
+{
+	struct objio_state *objios = container_of(oir, struct objio_state, oir);
+
+	ore_put_io_state(objios->ios);
+	kfree(objios);
+}
+
+static enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep)
+{
+	switch (oep) {
+	case OSD_ERR_PRI_NO_ERROR:
+		return (enum pnfs_osd_errno)0;
+
+	case OSD_ERR_PRI_CLEAR_PAGES:
+		BUG_ON(1);
+		return 0;
+
+	case OSD_ERR_PRI_RESOURCE:
+		return PNFS_OSD_ERR_RESOURCE;
+	case OSD_ERR_PRI_BAD_CRED:
+		return PNFS_OSD_ERR_BAD_CRED;
+	case OSD_ERR_PRI_NO_ACCESS:
+		return PNFS_OSD_ERR_NO_ACCESS;
+	case OSD_ERR_PRI_UNREACHABLE:
+		return PNFS_OSD_ERR_UNREACHABLE;
+	case OSD_ERR_PRI_NOT_FOUND:
+		return PNFS_OSD_ERR_NOT_FOUND;
+	case OSD_ERR_PRI_NO_SPACE:
+		return PNFS_OSD_ERR_NO_SPACE;
+	default:
+		WARN_ON(1);
+		/* fallthrough */
+	case OSD_ERR_PRI_EIO:
+		return PNFS_OSD_ERR_EIO;
+	}
+}
+
+static void __on_dev_error(struct ore_io_state *ios,
+	struct ore_dev *od, unsigned dev_index, enum osd_err_priority oep,
+	u64 dev_offset, u64  dev_len)
+{
+	struct objio_state *objios = ios->private;
+	struct pnfs_osd_objid pooid;
+	struct objio_dev_ent *ode = container_of(od, typeof(*ode), od);
+	/* FIXME: what to do with more-then-one-group layouts. We need to
+	 * translate from ore_io_state index to oc->comps index
+	 */
+	unsigned comp = dev_index;
+
+	pooid.oid_device_id = ode->id_node.deviceid;
+	pooid.oid_partition_id = ios->oc->comps[comp].obj.partition;
+	pooid.oid_object_id = ios->oc->comps[comp].obj.id;
+
+	objlayout_io_set_result(&objios->oir, comp,
+				&pooid, osd_pri_2_pnfs_err(oep),
+				dev_offset, dev_len, !ios->reading);
+}
+
+/*
+ * read
+ */
+static void _read_done(struct ore_io_state *ios, void *private)
+{
+	struct objio_state *objios = private;
+	ssize_t status;
+	int ret = ore_check_io(ios, &__on_dev_error);
+
+	/* FIXME: _io_free(ios) can we dealocate the libosd resources; */
+
+	if (likely(!ret))
+		status = ios->length;
+	else
+		status = ret;
+
+	objlayout_read_done(&objios->oir, status, objios->sync);
+}
+
+int objio_read_pagelist(struct nfs_pgio_header *hdr)
+{
+	struct objio_state *objios;
+	int ret;
+
+	ret = objio_alloc_io_state(NFS_I(hdr->inode)->layout, true,
+			hdr->lseg, hdr->args.pages, hdr->args.pgbase,
+			hdr->args.offset, hdr->args.count, hdr,
+			GFP_KERNEL, &objios);
+	if (unlikely(ret))
+		return ret;
+
+	objios->ios->done = _read_done;
+	dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
+		hdr->args.offset, hdr->args.count);
+	ret = ore_read(objios->ios);
+	if (unlikely(ret))
+		objio_free_result(&objios->oir);
+	return ret;
+}
+
+/*
+ * write
+ */
+static void _write_done(struct ore_io_state *ios, void *private)
+{
+	struct objio_state *objios = private;
+	ssize_t status;
+	int ret = ore_check_io(ios, &__on_dev_error);
+
+	/* FIXME: _io_free(ios) can we dealocate the libosd resources; */
+
+	if (likely(!ret)) {
+		/* FIXME: should be based on the OSD's persistence model
+		 * See OSD2r05 Section 4.13 Data persistence model */
+		objios->oir.committed = NFS_FILE_SYNC;
+		status = ios->length;
+	} else {
+		status = ret;
+	}
+
+	objlayout_write_done(&objios->oir, status, objios->sync);
+}
+
+static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
+{
+	struct objio_state *objios = priv;
+	struct nfs_pgio_header *hdr = objios->oir.rpcdata;
+	struct address_space *mapping = hdr->inode->i_mapping;
+	pgoff_t index = offset / PAGE_SIZE;
+	struct page *page;
+	loff_t i_size = i_size_read(hdr->inode);
+
+	if (offset >= i_size) {
+		*uptodate = true;
+		dprintk("%s: g_zero_page index=0x%lx\n", __func__, index);
+		return ZERO_PAGE(0);
+	}
+
+	page = find_get_page(mapping, index);
+	if (!page) {
+		page = find_or_create_page(mapping, index, GFP_NOFS);
+		if (unlikely(!page)) {
+			dprintk("%s: grab_cache_page Failed index=0x%lx\n",
+				__func__, index);
+			return NULL;
+		}
+		unlock_page(page);
+	}
+	if (PageDirty(page) || PageWriteback(page))
+		*uptodate = true;
+	else
+		*uptodate = PageUptodate(page);
+	dprintk("%s: index=0x%lx uptodate=%d\n", __func__, index, *uptodate);
+	return page;
+}
+
+static void __r4w_put_page(void *priv, struct page *page)
+{
+	dprintk("%s: index=0x%lx\n", __func__,
+		(page == ZERO_PAGE(0)) ? -1UL : page->index);
+	if (ZERO_PAGE(0) != page)
+		page_cache_release(page);
+	return;
+}
+
+static const struct _ore_r4w_op _r4w_op = {
+	.get_page = &__r4w_get_page,
+	.put_page = &__r4w_put_page,
+};
+
+int objio_write_pagelist(struct nfs_pgio_header *hdr, int how)
+{
+	struct objio_state *objios;
+	int ret;
+
+	ret = objio_alloc_io_state(NFS_I(hdr->inode)->layout, false,
+			hdr->lseg, hdr->args.pages, hdr->args.pgbase,
+			hdr->args.offset, hdr->args.count, hdr, GFP_NOFS,
+			&objios);
+	if (unlikely(ret))
+		return ret;
+
+	objios->sync = 0 != (how & FLUSH_SYNC);
+	objios->ios->r4w = &_r4w_op;
+
+	if (!objios->sync)
+		objios->ios->done = _write_done;
+
+	dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
+		hdr->args.offset, hdr->args.count);
+	ret = ore_write(objios->ios);
+	if (unlikely(ret)) {
+		objio_free_result(&objios->oir);
+		return ret;
+	}
+
+	if (objios->sync)
+		_write_done(objios->ios, objios);
+
+	return 0;
+}
+
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t objio_pg_test(struct nfs_pageio_descriptor *pgio,
+			  struct nfs_page *prev, struct nfs_page *req)
+{
+	unsigned int size;
+
+	size = pnfs_generic_pg_test(pgio, prev, req);
+
+	if (!size || pgio->pg_count + req->wb_bytes >
+	    (unsigned long)pgio->pg_layout_private)
+		return 0;
+
+	return min(size, req->wb_bytes);
+}
+
+static void objio_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	pnfs_generic_pg_init_read(pgio, req);
+	if (unlikely(pgio->pg_lseg == NULL))
+		return; /* Not pNFS */
+
+	pgio->pg_layout_private = (void *)
+				OBJIO_LSEG(pgio->pg_lseg)->layout.max_io_length;
+}
+
+static bool aligned_on_raid_stripe(u64 offset, struct ore_layout *layout,
+				   unsigned long *stripe_end)
+{
+	u32 stripe_off;
+	unsigned stripe_size;
+
+	if (layout->raid_algorithm == PNFS_OSD_RAID_0)
+		return true;
+
+	stripe_size = layout->stripe_unit *
+				(layout->group_width - layout->parity);
+
+	div_u64_rem(offset, stripe_size, &stripe_off);
+	if (!stripe_off)
+		return true;
+
+	*stripe_end = stripe_size - stripe_off;
+	return false;
+}
+
+static void objio_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	unsigned long stripe_end = 0;
+	u64 wb_size;
+
+	if (pgio->pg_dreq == NULL)
+		wb_size = i_size_read(pgio->pg_inode) - req_offset(req);
+	else
+		wb_size = nfs_dreq_bytes_left(pgio->pg_dreq);
+
+	pnfs_generic_pg_init_write(pgio, req, wb_size);
+	if (unlikely(pgio->pg_lseg == NULL))
+		return; /* Not pNFS */
+
+	if (req->wb_offset ||
+	    !aligned_on_raid_stripe(req->wb_index * PAGE_SIZE,
+			       &OBJIO_LSEG(pgio->pg_lseg)->layout,
+			       &stripe_end)) {
+		pgio->pg_layout_private = (void *)stripe_end;
+	} else {
+		pgio->pg_layout_private = (void *)
+				OBJIO_LSEG(pgio->pg_lseg)->layout.max_io_length;
+	}
+}
+
+static const struct nfs_pageio_ops objio_pg_read_ops = {
+	.pg_init = objio_init_read,
+	.pg_test = objio_pg_test,
+	.pg_doio = pnfs_generic_pg_readpages,
+};
+
+static const struct nfs_pageio_ops objio_pg_write_ops = {
+	.pg_init = objio_init_write,
+	.pg_test = objio_pg_test,
+	.pg_doio = pnfs_generic_pg_writepages,
+};
+
+static struct pnfs_layoutdriver_type objlayout_type = {
+	.id = LAYOUT_OSD2_OBJECTS,
+	.name = "LAYOUT_OSD2_OBJECTS",
+	.flags                   = PNFS_LAYOUTRET_ON_SETATTR |
+				   PNFS_LAYOUTRET_ON_ERROR,
+
+	.max_deviceinfo_size	 = PAGE_SIZE,
+	.owner		       	 = THIS_MODULE,
+	.alloc_layout_hdr        = objlayout_alloc_layout_hdr,
+	.free_layout_hdr         = objlayout_free_layout_hdr,
+
+	.alloc_lseg              = objlayout_alloc_lseg,
+	.free_lseg               = objlayout_free_lseg,
+
+	.read_pagelist           = objlayout_read_pagelist,
+	.write_pagelist          = objlayout_write_pagelist,
+	.pg_read_ops             = &objio_pg_read_ops,
+	.pg_write_ops            = &objio_pg_write_ops,
+
+	.free_deviceid_node	 = objio_free_deviceid_node,
+
+	.encode_layoutcommit	 = objlayout_encode_layoutcommit,
+	.encode_layoutreturn     = objlayout_encode_layoutreturn,
+};
+
+MODULE_DESCRIPTION("pNFS Layout Driver for OSD2 objects");
+MODULE_AUTHOR("Benny Halevy <bhalevy@panasas.com>");
+MODULE_LICENSE("GPL");
+
+static int __init
+objlayout_init(void)
+{
+	int ret = pnfs_register_layoutdriver(&objlayout_type);
+
+	if (ret)
+		printk(KERN_INFO
+			"NFS: %s: Registering OSD pNFS Layout Driver failed: error=%d\n",
+			__func__, ret);
+	else
+		printk(KERN_INFO "NFS: %s: Registered OSD pNFS Layout Driver\n",
+			__func__);
+	return ret;
+}
+
+static void __exit
+objlayout_exit(void)
+{
+	pnfs_unregister_layoutdriver(&objlayout_type);
+	printk(KERN_INFO "NFS: %s: Unregistered OSD pNFS Layout Driver\n",
+	       __func__);
+}
+
+MODULE_ALIAS("nfs-layouttype4-2");
+
+module_init(objlayout_init);
+module_exit(objlayout_exit);

fs/nfs/objlayout/objlayout.c

@@ -0,0 +1,706 @@
+/*
+ *  pNFS Objects layout driver high level definitions
+ *
+ *  Copyright (C) 2007 Panasas Inc. [year of first publication]
+ *  All rights reserved.
+ *
+ *  Benny Halevy <bhalevy@panasas.com>
+ *  Boaz Harrosh <ooo@electrozaur.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2
+ *  See the file COPYING included with this distribution for more details.
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the Panasas company 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 ``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 REGENTS 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.
+ */
+
+#include <linux/kmod.h>
+#include <linux/moduleparam.h>
+#include <linux/ratelimit.h>
+#include <scsi/osd_initiator.h>
+#include "objlayout.h"
+
+#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
+/*
+ * Create a objlayout layout structure for the given inode and return it.
+ */
+struct pnfs_layout_hdr *
+objlayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
+{
+	struct objlayout *objlay;
+
+	objlay = kzalloc(sizeof(struct objlayout), gfp_flags);
+	if (!objlay)
+		return NULL;
+	spin_lock_init(&objlay->lock);
+	INIT_LIST_HEAD(&objlay->err_list);
+	dprintk("%s: Return %p\n", __func__, objlay);
+	return &objlay->pnfs_layout;
+}
+
+/*
+ * Free an objlayout layout structure
+ */
+void
+objlayout_free_layout_hdr(struct pnfs_layout_hdr *lo)
+{
+	struct objlayout *objlay = OBJLAYOUT(lo);
+
+	dprintk("%s: objlay %p\n", __func__, objlay);
+
+	WARN_ON(!list_empty(&objlay->err_list));
+	kfree(objlay);
+}
+
+/*
+ * Unmarshall layout and store it in pnfslay.
+ */
+struct pnfs_layout_segment *
+objlayout_alloc_lseg(struct pnfs_layout_hdr *pnfslay,
+		     struct nfs4_layoutget_res *lgr,
+		     gfp_t gfp_flags)
+{
+	int status = -ENOMEM;
+	struct xdr_stream stream;
+	struct xdr_buf buf = {
+		.pages =  lgr->layoutp->pages,
+		.page_len =  lgr->layoutp->len,
+		.buflen =  lgr->layoutp->len,
+		.len = lgr->layoutp->len,
+	};
+	struct page *scratch;
+	struct pnfs_layout_segment *lseg;
+
+	dprintk("%s: Begin pnfslay %p\n", __func__, pnfslay);
+
+	scratch = alloc_page(gfp_flags);
+	if (!scratch)
+		goto err_nofree;
+
+	xdr_init_decode(&stream, &buf, NULL);
+	xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);
+
+	status = objio_alloc_lseg(&lseg, pnfslay, &lgr->range, &stream, gfp_flags);
+	if (unlikely(status)) {
+		dprintk("%s: objio_alloc_lseg Return err %d\n", __func__,
+			status);
+		goto err;
+	}
+
+	__free_page(scratch);
+
+	dprintk("%s: Return %p\n", __func__, lseg);
+	return lseg;
+
+err:
+	__free_page(scratch);
+err_nofree:
+	dprintk("%s: Err Return=>%d\n", __func__, status);
+	return ERR_PTR(status);
+}
+
+/*
+ * Free a layout segement
+ */
+void
+objlayout_free_lseg(struct pnfs_layout_segment *lseg)
+{
+	dprintk("%s: freeing layout segment %p\n", __func__, lseg);
+
+	if (unlikely(!lseg))
+		return;
+
+	objio_free_lseg(lseg);
+}
+
+/*
+ * I/O Operations
+ */
+static inline u64
+end_offset(u64 start, u64 len)
+{
+	u64 end;
+
+	end = start + len;
+	return end >= start ? end : NFS4_MAX_UINT64;
+}
+
+static void _fix_verify_io_params(struct pnfs_layout_segment *lseg,
+			   struct page ***p_pages, unsigned *p_pgbase,
+			   u64 offset, unsigned long count)
+{
+	u64 lseg_end_offset;
+
+	BUG_ON(offset < lseg->pls_range.offset);
+	lseg_end_offset = end_offset(lseg->pls_range.offset,
+				     lseg->pls_range.length);
+	BUG_ON(offset >= lseg_end_offset);
+	WARN_ON(offset + count > lseg_end_offset);
+
+	if (*p_pgbase > PAGE_SIZE) {
+		dprintk("%s: pgbase(0x%x) > PAGE_SIZE\n", __func__, *p_pgbase);
+		*p_pages += *p_pgbase >> PAGE_SHIFT;
+		*p_pgbase &= ~PAGE_MASK;
+	}
+}
+
+/*
+ * I/O done common code
+ */
+static void
+objlayout_iodone(struct objlayout_io_res *oir)
+{
+	if (likely(oir->status >= 0)) {
+		objio_free_result(oir);
+	} else {
+		struct objlayout *objlay = oir->objlay;
+
+		spin_lock(&objlay->lock);
+		objlay->delta_space_valid = OBJ_DSU_INVALID;
+		list_add(&objlay->err_list, &oir->err_list);
+		spin_unlock(&objlay->lock);
+	}
+}
+
+/*
+ * objlayout_io_set_result - Set an osd_error code on a specific osd comp.
+ *
+ * The @index component IO failed (error returned from target). Register
+ * the error for later reporting at layout-return.
+ */
+void
+objlayout_io_set_result(struct objlayout_io_res *oir, unsigned index,
+			struct pnfs_osd_objid *pooid, int osd_error,
+			u64 offset, u64 length, bool is_write)
+{
+	struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[index];
+
+	BUG_ON(index >= oir->num_comps);
+	if (osd_error) {
+		ioerr->oer_component = *pooid;
+		ioerr->oer_comp_offset = offset;
+		ioerr->oer_comp_length = length;
+		ioerr->oer_iswrite = is_write;
+		ioerr->oer_errno = osd_error;
+
+		dprintk("%s: err[%d]: errno=%d is_write=%d dev(%llx:%llx) "
+			"par=0x%llx obj=0x%llx offset=0x%llx length=0x%llx\n",
+			__func__, index, ioerr->oer_errno,
+			ioerr->oer_iswrite,
+			_DEVID_LO(&ioerr->oer_component.oid_device_id),
+			_DEVID_HI(&ioerr->oer_component.oid_device_id),
+			ioerr->oer_component.oid_partition_id,
+			ioerr->oer_component.oid_object_id,
+			ioerr->oer_comp_offset,
+			ioerr->oer_comp_length);
+	} else {
+		/* User need not call if no error is reported */
+		ioerr->oer_errno = 0;
+	}
+}
+
+/* Function scheduled on rpc workqueue to call ->nfs_readlist_complete().
+ * This is because the osd completion is called with ints-off from
+ * the block layer
+ */
+static void _rpc_read_complete(struct work_struct *work)
+{
+	struct rpc_task *task;
+	struct nfs_pgio_header *hdr;
+
+	dprintk("%s enter\n", __func__);
+	task = container_of(work, struct rpc_task, u.tk_work);
+	hdr = container_of(task, struct nfs_pgio_header, task);
+
+	pnfs_ld_read_done(hdr);
+}
+
+void
+objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
+{
+	struct nfs_pgio_header *hdr = oir->rpcdata;
+
+	oir->status = hdr->task.tk_status = status;
+	if (status >= 0)
+		hdr->res.count = status;
+	else
+		hdr->pnfs_error = status;
+	objlayout_iodone(oir);
+	/* must not use oir after this point */
+
+	dprintk("%s: Return status=%zd eof=%d sync=%d\n", __func__,
+		status, hdr->res.eof, sync);
+
+	if (sync)
+		pnfs_ld_read_done(hdr);
+	else {
+		INIT_WORK(&hdr->task.u.tk_work, _rpc_read_complete);
+		schedule_work(&hdr->task.u.tk_work);
+	}
+}
+
+/*
+ * Perform sync or async reads.
+ */
+enum pnfs_try_status
+objlayout_read_pagelist(struct nfs_pgio_header *hdr)
+{
+	struct inode *inode = hdr->inode;
+	loff_t offset = hdr->args.offset;
+	size_t count = hdr->args.count;
+	int err;
+	loff_t eof;
+
+	eof = i_size_read(inode);
+	if (unlikely(offset + count > eof)) {
+		if (offset >= eof) {
+			err = 0;
+			hdr->res.count = 0;
+			hdr->res.eof = 1;
+			/*FIXME: do we need to call pnfs_ld_read_done() */
+			goto out;
+		}
+		count = eof - offset;
+	}
+
+	hdr->res.eof = (offset + count) >= eof;
+	_fix_verify_io_params(hdr->lseg, &hdr->args.pages,
+			      &hdr->args.pgbase,
+			      hdr->args.offset, hdr->args.count);
+
+	dprintk("%s: inode(%lx) offset 0x%llx count 0x%Zx eof=%d\n",
+		__func__, inode->i_ino, offset, count, hdr->res.eof);
+
+	err = objio_read_pagelist(hdr);
+ out:
+	if (unlikely(err)) {
+		hdr->pnfs_error = err;
+		dprintk("%s: Returned Error %d\n", __func__, err);
+		return PNFS_NOT_ATTEMPTED;
+	}
+	return PNFS_ATTEMPTED;
+}
+
+/* Function scheduled on rpc workqueue to call ->nfs_writelist_complete().
+ * This is because the osd completion is called with ints-off from
+ * the block layer
+ */
+static void _rpc_write_complete(struct work_struct *work)
+{
+	struct rpc_task *task;
+	struct nfs_pgio_header *hdr;
+
+	dprintk("%s enter\n", __func__);
+	task = container_of(work, struct rpc_task, u.tk_work);
+	hdr = container_of(task, struct nfs_pgio_header, task);
+
+	pnfs_ld_write_done(hdr);
+}
+
+void
+objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
+{
+	struct nfs_pgio_header *hdr = oir->rpcdata;
+
+	oir->status = hdr->task.tk_status = status;
+	if (status >= 0) {
+		hdr->res.count = status;
+		hdr->verf.committed = oir->committed;
+	} else {
+		hdr->pnfs_error = status;
+	}
+	objlayout_iodone(oir);
+	/* must not use oir after this point */
+
+	dprintk("%s: Return status %zd committed %d sync=%d\n", __func__,
+		status, hdr->verf.committed, sync);
+
+	if (sync)
+		pnfs_ld_write_done(hdr);
+	else {
+		INIT_WORK(&hdr->task.u.tk_work, _rpc_write_complete);
+		schedule_work(&hdr->task.u.tk_work);
+	}
+}
+
+/*
+ * Perform sync or async writes.
+ */
+enum pnfs_try_status
+objlayout_write_pagelist(struct nfs_pgio_header *hdr, int how)
+{
+	int err;
+
+	_fix_verify_io_params(hdr->lseg, &hdr->args.pages,
+			      &hdr->args.pgbase,
+			      hdr->args.offset, hdr->args.count);
+
+	err = objio_write_pagelist(hdr, how);
+	if (unlikely(err)) {
+		hdr->pnfs_error = err;
+		dprintk("%s: Returned Error %d\n", __func__, err);
+		return PNFS_NOT_ATTEMPTED;
+	}
+	return PNFS_ATTEMPTED;
+}
+
+void
+objlayout_encode_layoutcommit(struct pnfs_layout_hdr *pnfslay,
+			      struct xdr_stream *xdr,
+			      const struct nfs4_layoutcommit_args *args)
+{
+	struct objlayout *objlay = OBJLAYOUT(pnfslay);
+	struct pnfs_osd_layoutupdate lou;
+	__be32 *start;
+
+	dprintk("%s: Begin\n", __func__);
+
+	spin_lock(&objlay->lock);
+	lou.dsu_valid = (objlay->delta_space_valid == OBJ_DSU_VALID);
+	lou.dsu_delta = objlay->delta_space_used;
+	objlay->delta_space_used = 0;
+	objlay->delta_space_valid = OBJ_DSU_INIT;
+	lou.olu_ioerr_flag = !list_empty(&objlay->err_list);
+	spin_unlock(&objlay->lock);
+
+	start = xdr_reserve_space(xdr, 4);
+
+	BUG_ON(pnfs_osd_xdr_encode_layoutupdate(xdr, &lou));
+
+	*start = cpu_to_be32((xdr->p - start - 1) * 4);
+
+	dprintk("%s: Return delta_space_used %lld err %d\n", __func__,
+		lou.dsu_delta, lou.olu_ioerr_flag);
+}
+
+static int
+err_prio(u32 oer_errno)
+{
+	switch (oer_errno) {
+	case 0:
+		return 0;
+
+	case PNFS_OSD_ERR_RESOURCE:
+		return OSD_ERR_PRI_RESOURCE;
+	case PNFS_OSD_ERR_BAD_CRED:
+		return OSD_ERR_PRI_BAD_CRED;
+	case PNFS_OSD_ERR_NO_ACCESS:
+		return OSD_ERR_PRI_NO_ACCESS;
+	case PNFS_OSD_ERR_UNREACHABLE:
+		return OSD_ERR_PRI_UNREACHABLE;
+	case PNFS_OSD_ERR_NOT_FOUND:
+		return OSD_ERR_PRI_NOT_FOUND;
+	case PNFS_OSD_ERR_NO_SPACE:
+		return OSD_ERR_PRI_NO_SPACE;
+	default:
+		WARN_ON(1);
+		/* fallthrough */
+	case PNFS_OSD_ERR_EIO:
+		return OSD_ERR_PRI_EIO;
+	}
+}
+
+static void
+merge_ioerr(struct pnfs_osd_ioerr *dest_err,
+	    const struct pnfs_osd_ioerr *src_err)
+{
+	u64 dest_end, src_end;
+
+	if (!dest_err->oer_errno) {
+		*dest_err = *src_err;
+		/* accumulated device must be blank */
+		memset(&dest_err->oer_component.oid_device_id, 0,
+			sizeof(dest_err->oer_component.oid_device_id));
+
+		return;
+	}
+
+	if (dest_err->oer_component.oid_partition_id !=
+				src_err->oer_component.oid_partition_id)
+		dest_err->oer_component.oid_partition_id = 0;
+
+	if (dest_err->oer_component.oid_object_id !=
+				src_err->oer_component.oid_object_id)
+		dest_err->oer_component.oid_object_id = 0;
+
+	if (dest_err->oer_comp_offset > src_err->oer_comp_offset)
+		dest_err->oer_comp_offset = src_err->oer_comp_offset;
+
+	dest_end = end_offset(dest_err->oer_comp_offset,
+			      dest_err->oer_comp_length);
+	src_end =  end_offset(src_err->oer_comp_offset,
+			      src_err->oer_comp_length);
+	if (dest_end < src_end)
+		dest_end = src_end;
+
+	dest_err->oer_comp_length = dest_end - dest_err->oer_comp_offset;
+
+	if ((src_err->oer_iswrite == dest_err->oer_iswrite) &&
+	    (err_prio(src_err->oer_errno) > err_prio(dest_err->oer_errno))) {
+			dest_err->oer_errno = src_err->oer_errno;
+	} else if (src_err->oer_iswrite) {
+		dest_err->oer_iswrite = true;
+		dest_err->oer_errno = src_err->oer_errno;
+	}
+}
+
+static void
+encode_accumulated_error(struct objlayout *objlay, __be32 *p)
+{
+	struct objlayout_io_res *oir, *tmp;
+	struct pnfs_osd_ioerr accumulated_err = {.oer_errno = 0};
+
+	list_for_each_entry_safe(oir, tmp, &objlay->err_list, err_list) {
+		unsigned i;
+
+		for (i = 0; i < oir->num_comps; i++) {
+			struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[i];
+
+			if (!ioerr->oer_errno)
+				continue;
+
+			printk(KERN_ERR "NFS: %s: err[%d]: errno=%d "
+				"is_write=%d dev(%llx:%llx) par=0x%llx "
+				"obj=0x%llx offset=0x%llx length=0x%llx\n",
+				__func__, i, ioerr->oer_errno,
+				ioerr->oer_iswrite,
+				_DEVID_LO(&ioerr->oer_component.oid_device_id),
+				_DEVID_HI(&ioerr->oer_component.oid_device_id),
+				ioerr->oer_component.oid_partition_id,
+				ioerr->oer_component.oid_object_id,
+				ioerr->oer_comp_offset,
+				ioerr->oer_comp_length);
+
+			merge_ioerr(&accumulated_err, ioerr);
+		}
+		list_del(&oir->err_list);
+		objio_free_result(oir);
+	}
+
+	pnfs_osd_xdr_encode_ioerr(p, &accumulated_err);
+}
+
+void
+objlayout_encode_layoutreturn(struct pnfs_layout_hdr *pnfslay,
+			      struct xdr_stream *xdr,
+			      const struct nfs4_layoutreturn_args *args)
+{
+	struct objlayout *objlay = OBJLAYOUT(pnfslay);
+	struct objlayout_io_res *oir, *tmp;
+	__be32 *start;
+
+	dprintk("%s: Begin\n", __func__);
+	start = xdr_reserve_space(xdr, 4);
+	BUG_ON(!start);
+
+	spin_lock(&objlay->lock);
+
+	list_for_each_entry_safe(oir, tmp, &objlay->err_list, err_list) {
+		__be32 *last_xdr = NULL, *p;
+		unsigned i;
+		int res = 0;
+
+		for (i = 0; i < oir->num_comps; i++) {
+			struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[i];
+
+			if (!ioerr->oer_errno)
+				continue;
+
+			dprintk("%s: err[%d]: errno=%d is_write=%d "
+				"dev(%llx:%llx) par=0x%llx obj=0x%llx "
+				"offset=0x%llx length=0x%llx\n",
+				__func__, i, ioerr->oer_errno,
+				ioerr->oer_iswrite,
+				_DEVID_LO(&ioerr->oer_component.oid_device_id),
+				_DEVID_HI(&ioerr->oer_component.oid_device_id),
+				ioerr->oer_component.oid_partition_id,
+				ioerr->oer_component.oid_object_id,
+				ioerr->oer_comp_offset,
+				ioerr->oer_comp_length);
+
+			p = pnfs_osd_xdr_ioerr_reserve_space(xdr);
+			if (unlikely(!p)) {
+				res = -E2BIG;
+				break; /* accumulated_error */
+			}
+
+			last_xdr = p;
+			pnfs_osd_xdr_encode_ioerr(p, &oir->ioerrs[i]);
+		}
+
+		/* TODO: use xdr_write_pages */
+		if (unlikely(res)) {
+			/* no space for even one error descriptor */
+			BUG_ON(!last_xdr);
+
+			/* we've encountered a situation with lots and lots of
+			 * errors and no space to encode them all. Use the last
+			 * available slot to report the union of all the
+			 * remaining errors.
+			 */
+			encode_accumulated_error(objlay, last_xdr);
+			goto loop_done;
+		}
+		list_del(&oir->err_list);
+		objio_free_result(oir);
+	}
+loop_done:
+	spin_unlock(&objlay->lock);
+
+	*start = cpu_to_be32((xdr->p - start - 1) * 4);
+	dprintk("%s: Return\n", __func__);
+}
+
+enum {
+	OBJLAYOUT_MAX_URI_LEN = 256, OBJLAYOUT_MAX_OSDNAME_LEN = 64,
+	OBJLAYOUT_MAX_SYSID_HEX_LEN = OSD_SYSTEMID_LEN * 2 + 1,
+	OSD_LOGIN_UPCALL_PATHLEN  = 256
+};
+
+static char osd_login_prog[OSD_LOGIN_UPCALL_PATHLEN] = "/sbin/osd_login";
+
+module_param_string(osd_login_prog, osd_login_prog, sizeof(osd_login_prog),
+		    0600);
+MODULE_PARM_DESC(osd_login_prog, "Path to the osd_login upcall program");
+
+struct __auto_login {
+	char uri[OBJLAYOUT_MAX_URI_LEN];
+	char osdname[OBJLAYOUT_MAX_OSDNAME_LEN];
+	char systemid_hex[OBJLAYOUT_MAX_SYSID_HEX_LEN];
+};
+
+static int __objlayout_upcall(struct __auto_login *login)
+{
+	static char *envp[] = { "HOME=/",
+		"TERM=linux",
+		"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
+		NULL
+	};
+	char *argv[8];
+	int ret;
+
+	if (unlikely(!osd_login_prog[0])) {
+		dprintk("%s: osd_login_prog is disabled\n", __func__);
+		return -EACCES;
+	}
+
+	dprintk("%s uri: %s\n", __func__, login->uri);
+	dprintk("%s osdname %s\n", __func__, login->osdname);
+	dprintk("%s systemid_hex %s\n", __func__, login->systemid_hex);
+
+	argv[0] = (char *)osd_login_prog;
+	argv[1] = "-u";
+	argv[2] = login->uri;
+	argv[3] = "-o";
+	argv[4] = login->osdname;
+	argv[5] = "-s";
+	argv[6] = login->systemid_hex;
+	argv[7] = NULL;
+
+	ret = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC);
+	/*
+	 * Disable the upcall mechanism if we're getting an ENOENT or
+	 * EACCES error. The admin can re-enable it on the fly by using
+	 * sysfs to set the objlayoutdriver.osd_login_prog module parameter once
+	 * the problem has been fixed.
+	 */
+	if (ret == -ENOENT || ret == -EACCES) {
+		printk(KERN_ERR "PNFS-OBJ: %s was not found please set "
+			"objlayoutdriver.osd_login_prog kernel parameter!\n",
+			osd_login_prog);
+		osd_login_prog[0] = '\0';
+	}
+	dprintk("%s %s return value: %d\n", __func__, osd_login_prog, ret);
+
+	return ret;
+}
+
+/* Assume dest is all zeros */
+static void __copy_nfsS_and_zero_terminate(struct nfs4_string s,
+					   char *dest, int max_len,
+					   const char *var_name)
+{
+	if (!s.len)
+		return;
+
+	if (s.len >= max_len) {
+		pr_warn_ratelimited(
+			"objlayout_autologin: %s: s.len(%d) >= max_len(%d)",
+			var_name, s.len, max_len);
+		s.len = max_len - 1; /* space for null terminator */
+	}
+
+	memcpy(dest, s.data, s.len);
+}
+
+/* Assume sysid is all zeros */
+static void _sysid_2_hex(struct nfs4_string s,
+		  char sysid[OBJLAYOUT_MAX_SYSID_HEX_LEN])
+{
+	int i;
+	char *cur;
+
+	if (!s.len)
+		return;
+
+	if (s.len != OSD_SYSTEMID_LEN) {
+		pr_warn_ratelimited(
+		    "objlayout_autologin: systemid_len(%d) != OSD_SYSTEMID_LEN",
+		    s.len);
+		if (s.len > OSD_SYSTEMID_LEN)
+			s.len = OSD_SYSTEMID_LEN;
+	}
+
+	cur = sysid;
+	for (i = 0; i < s.len; i++)
+		cur = hex_byte_pack(cur, s.data[i]);
+}
+
+int objlayout_autologin(struct pnfs_osd_deviceaddr *deviceaddr)
+{
+	int rc;
+	struct __auto_login login;
+
+	if (!deviceaddr->oda_targetaddr.ota_netaddr.r_addr.len)
+		return -ENODEV;
+
+	memset(&login, 0, sizeof(login));
+	__copy_nfsS_and_zero_terminate(
+		deviceaddr->oda_targetaddr.ota_netaddr.r_addr,
+		login.uri, sizeof(login.uri), "URI");
+
+	__copy_nfsS_and_zero_terminate(
+		deviceaddr->oda_osdname,
+		login.osdname, sizeof(login.osdname), "OSDNAME");
+
+	_sysid_2_hex(deviceaddr->oda_systemid, login.systemid_hex);
+
+	rc = __objlayout_upcall(&login);
+	if (rc > 0) /* script returns positive values */
+		rc = -ENODEV;
+
+	return rc;
+}

fs/nfs/objlayout/objlayout.h

@@ -0,0 +1,184 @@
+/*
+ *  Data types and function declerations for interfacing with the
+ *  pNFS standard object layout driver.
+ *
+ *  Copyright (C) 2007 Panasas Inc. [year of first publication]
+ *  All rights reserved.
+ *
+ *  Benny Halevy <bhalevy@panasas.com>
+ *  Boaz Harrosh <ooo@electrozaur.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2
+ *  See the file COPYING included with this distribution for more details.
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the Panasas company 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 ``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 REGENTS 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 _OBJLAYOUT_H
+#define _OBJLAYOUT_H
+
+#include <linux/nfs_fs.h>
+#include <linux/pnfs_osd_xdr.h>
+#include "../pnfs.h"
+
+/*
+ * per-inode layout
+ */
+struct objlayout {
+	struct pnfs_layout_hdr pnfs_layout;
+
+	 /* for layout_commit */
+	enum osd_delta_space_valid_enum {
+		OBJ_DSU_INIT = 0,
+		OBJ_DSU_VALID,
+		OBJ_DSU_INVALID,
+	} delta_space_valid;
+	s64 delta_space_used;  /* consumed by write ops */
+
+	 /* for layout_return */
+	spinlock_t lock;
+	struct list_head err_list;
+};
+
+static inline struct objlayout *
+OBJLAYOUT(struct pnfs_layout_hdr *lo)
+{
+	return container_of(lo, struct objlayout, pnfs_layout);
+}
+
+/*
+ * per-I/O operation state
+ * embedded in objects provider io_state data structure
+ */
+struct objlayout_io_res {
+	struct objlayout *objlay;
+
+	void *rpcdata;
+	int status;             /* res */
+	int committed;          /* res */
+
+	/* Error reporting (layout_return) */
+	struct list_head err_list;
+	unsigned num_comps;
+	/* Pointer to array of error descriptors of size num_comps.
+	 * It should contain as many entries as devices in the osd_layout
+	 * that participate in the I/O. It is up to the io_engine to allocate
+	 * needed space and set num_comps.
+	 */
+	struct pnfs_osd_ioerr *ioerrs;
+};
+
+static inline
+void objlayout_init_ioerrs(struct objlayout_io_res *oir, unsigned num_comps,
+			struct pnfs_osd_ioerr *ioerrs, void *rpcdata,
+			struct pnfs_layout_hdr *pnfs_layout_type)
+{
+	oir->objlay = OBJLAYOUT(pnfs_layout_type);
+	oir->rpcdata = rpcdata;
+	INIT_LIST_HEAD(&oir->err_list);
+	oir->num_comps = num_comps;
+	oir->ioerrs = ioerrs;
+}
+
+/*
+ * Raid engine I/O API
+ */
+extern int objio_alloc_lseg(struct pnfs_layout_segment **outp,
+	struct pnfs_layout_hdr *pnfslay,
+	struct pnfs_layout_range *range,
+	struct xdr_stream *xdr,
+	gfp_t gfp_flags);
+extern void objio_free_lseg(struct pnfs_layout_segment *lseg);
+
+/* objio_free_result will free these @oir structs received from
+ * objlayout_{read,write}_done
+ */
+extern void objio_free_result(struct objlayout_io_res *oir);
+
+extern int objio_read_pagelist(struct nfs_pgio_header *rdata);
+extern int objio_write_pagelist(struct nfs_pgio_header *wdata, int how);
+
+/*
+ * callback API
+ */
+extern void objlayout_io_set_result(struct objlayout_io_res *oir,
+			unsigned index, struct pnfs_osd_objid *pooid,
+			int osd_error, u64 offset, u64 length, bool is_write);
+
+static inline void
+objlayout_add_delta_space_used(struct objlayout *objlay, s64 space_used)
+{
+	/* If one of the I/Os errored out and the delta_space_used was
+	 * invalid we render the complete report as invalid. Protocol mandate
+	 * the DSU be accurate or not reported.
+	 */
+	spin_lock(&objlay->lock);
+	if (objlay->delta_space_valid != OBJ_DSU_INVALID) {
+		objlay->delta_space_valid = OBJ_DSU_VALID;
+		objlay->delta_space_used += space_used;
+	}
+	spin_unlock(&objlay->lock);
+}
+
+extern void objlayout_read_done(struct objlayout_io_res *oir,
+				ssize_t status, bool sync);
+extern void objlayout_write_done(struct objlayout_io_res *oir,
+				 ssize_t status, bool sync);
+
+/*
+ * exported generic objects function vectors
+ */
+
+extern struct pnfs_layout_hdr *objlayout_alloc_layout_hdr(struct inode *, gfp_t gfp_flags);
+extern void objlayout_free_layout_hdr(struct pnfs_layout_hdr *);
+
+extern struct pnfs_layout_segment *objlayout_alloc_lseg(
+	struct pnfs_layout_hdr *,
+	struct nfs4_layoutget_res *,
+	gfp_t gfp_flags);
+extern void objlayout_free_lseg(struct pnfs_layout_segment *);
+
+extern enum pnfs_try_status objlayout_read_pagelist(
+	struct nfs_pgio_header *);
+
+extern enum pnfs_try_status objlayout_write_pagelist(
+	struct nfs_pgio_header *,
+	int how);
+
+extern void objlayout_encode_layoutcommit(
+	struct pnfs_layout_hdr *,
+	struct xdr_stream *,
+	const struct nfs4_layoutcommit_args *);
+
+extern void objlayout_encode_layoutreturn(
+	struct pnfs_layout_hdr *,
+	struct xdr_stream *,
+	const struct nfs4_layoutreturn_args *);
+
+extern int objlayout_autologin(struct pnfs_osd_deviceaddr *deviceaddr);
+
+#endif /* _OBJLAYOUT_H */

fs/nfs/objlayout/pnfs_osd_xdr_cli.c

@@ -0,0 +1,415 @@
+/*
+ *  Object-Based pNFS Layout XDR layer
+ *
+ *  Copyright (C) 2007 Panasas Inc. [year of first publication]
+ *  All rights reserved.
+ *
+ *  Benny Halevy <bhalevy@panasas.com>
+ *  Boaz Harrosh <ooo@electrozaur.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2
+ *  See the file COPYING included with this distribution for more details.
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. 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.
+ *  3. Neither the name of the Panasas company 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 ``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 REGENTS 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.
+ */
+
+#include <linux/pnfs_osd_xdr.h>
+
+#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
+
+/*
+ * The following implementation is based on RFC5664
+ */
+
+/*
+ * struct pnfs_osd_objid {
+ *	struct nfs4_deviceid	oid_device_id;
+ *	u64			oid_partition_id;
+ *	u64			oid_object_id;
+ * }; // xdr size 32 bytes
+ */
+static __be32 *
+_osd_xdr_decode_objid(__be32 *p, struct pnfs_osd_objid *objid)
+{
+	p = xdr_decode_opaque_fixed(p, objid->oid_device_id.data,
+				    sizeof(objid->oid_device_id.data));
+
+	p = xdr_decode_hyper(p, &objid->oid_partition_id);
+	p = xdr_decode_hyper(p, &objid->oid_object_id);
+	return p;
+}
+/*
+ * struct pnfs_osd_opaque_cred {
+ *	u32 cred_len;
+ *	void *cred;
+ * }; // xdr size [variable]
+ * The return pointers are from the xdr buffer
+ */
+static int
+_osd_xdr_decode_opaque_cred(struct pnfs_osd_opaque_cred *opaque_cred,
+			    struct xdr_stream *xdr)
+{
+	__be32 *p = xdr_inline_decode(xdr, 1);
+
+	if (!p)
+		return -EINVAL;
+
+	opaque_cred->cred_len = be32_to_cpu(*p++);
+
+	p = xdr_inline_decode(xdr, opaque_cred->cred_len);
+	if (!p)
+		return -EINVAL;
+
+	opaque_cred->cred = p;
+	return 0;
+}
+
+/*
+ * struct pnfs_osd_object_cred {
+ *	struct pnfs_osd_objid		oc_object_id;
+ *	u32				oc_osd_version;
+ *	u32				oc_cap_key_sec;
+ *	struct pnfs_osd_opaque_cred	oc_cap_key
+ *	struct pnfs_osd_opaque_cred	oc_cap;
+ * }; // xdr size 32 + 4 + 4 + [variable] + [variable]
+ */
+static int
+_osd_xdr_decode_object_cred(struct pnfs_osd_object_cred *comp,
+			    struct xdr_stream *xdr)
+{
+	__be32 *p = xdr_inline_decode(xdr, 32 + 4 + 4);
+	int ret;
+
+	if (!p)
+		return -EIO;
+
+	p = _osd_xdr_decode_objid(p, &comp->oc_object_id);
+	comp->oc_osd_version = be32_to_cpup(p++);
+	comp->oc_cap_key_sec = be32_to_cpup(p);
+
+	ret = _osd_xdr_decode_opaque_cred(&comp->oc_cap_key, xdr);
+	if (unlikely(ret))
+		return ret;
+
+	ret = _osd_xdr_decode_opaque_cred(&comp->oc_cap, xdr);
+	return ret;
+}
+
+/*
+ * struct pnfs_osd_data_map {
+ *	u32	odm_num_comps;
+ *	u64	odm_stripe_unit;
+ *	u32	odm_group_width;
+ *	u32	odm_group_depth;
+ *	u32	odm_mirror_cnt;
+ *	u32	odm_raid_algorithm;
+ * }; // xdr size 4 + 8 + 4 + 4 + 4 + 4
+ */
+static inline int
+_osd_data_map_xdr_sz(void)
+{
+	return 4 + 8 + 4 + 4 + 4 + 4;
+}
+
+static __be32 *
+_osd_xdr_decode_data_map(__be32 *p, struct pnfs_osd_data_map *data_map)
+{
+	data_map->odm_num_comps = be32_to_cpup(p++);
+	p = xdr_decode_hyper(p, &data_map->odm_stripe_unit);
+	data_map->odm_group_width = be32_to_cpup(p++);
+	data_map->odm_group_depth = be32_to_cpup(p++);
+	data_map->odm_mirror_cnt = be32_to_cpup(p++);
+	data_map->odm_raid_algorithm = be32_to_cpup(p++);
+	dprintk("%s: odm_num_comps=%u odm_stripe_unit=%llu odm_group_width=%u "
+		"odm_group_depth=%u odm_mirror_cnt=%u odm_raid_algorithm=%u\n",
+		__func__,
+		data_map->odm_num_comps,
+		(unsigned long long)data_map->odm_stripe_unit,
+		data_map->odm_group_width,
+		data_map->odm_group_depth,
+		data_map->odm_mirror_cnt,
+		data_map->odm_raid_algorithm);
+	return p;
+}
+
+int pnfs_osd_xdr_decode_layout_map(struct pnfs_osd_layout *layout,
+	struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr)
+{
+	__be32 *p;
+
+	memset(iter, 0, sizeof(*iter));
+
+	p = xdr_inline_decode(xdr, _osd_data_map_xdr_sz() + 4 + 4);
+	if (unlikely(!p))
+		return -EINVAL;
+
+	p = _osd_xdr_decode_data_map(p, &layout->olo_map);
+	layout->olo_comps_index = be32_to_cpup(p++);
+	layout->olo_num_comps = be32_to_cpup(p++);
+	dprintk("%s: olo_comps_index=%d olo_num_comps=%d\n", __func__,
+		layout->olo_comps_index, layout->olo_num_comps);
+
+	iter->total_comps = layout->olo_num_comps;
+	return 0;
+}
+
+bool pnfs_osd_xdr_decode_layout_comp(struct pnfs_osd_object_cred *comp,
+	struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr,
+	int *err)
+{
+	BUG_ON(iter->decoded_comps > iter->total_comps);
+	if (iter->decoded_comps == iter->total_comps)
+		return false;
+
+	*err = _osd_xdr_decode_object_cred(comp, xdr);
+	if (unlikely(*err)) {
+		dprintk("%s: _osd_xdr_decode_object_cred=>%d decoded_comps=%d "
+			"total_comps=%d\n", __func__, *err,
+			iter->decoded_comps, iter->total_comps);
+		return false; /* stop the loop */
+	}
+	dprintk("%s: dev(%llx:%llx) par=0x%llx obj=0x%llx "
+		"key_len=%u cap_len=%u\n",
+		__func__,
+		_DEVID_LO(&comp->oc_object_id.oid_device_id),
+		_DEVID_HI(&comp->oc_object_id.oid_device_id),
+		comp->oc_object_id.oid_partition_id,
+		comp->oc_object_id.oid_object_id,
+		comp->oc_cap_key.cred_len, comp->oc_cap.cred_len);
+
+	iter->decoded_comps++;
+	return true;
+}
+
+/*
+ * Get Device Information Decoding
+ *
+ * Note: since Device Information is currently done synchronously, all
+ *       variable strings fields are left inside the rpc buffer and are only
+ *       pointed to by the pnfs_osd_deviceaddr members. So the read buffer
+ *       should not be freed while the returned information is in use.
+ */
+/*
+ *struct nfs4_string {
+ *	unsigned int len;
+ *	char *data;
+ *}; // size [variable]
+ * NOTE: Returned string points to inside the XDR buffer
+ */
+static __be32 *
+__read_u8_opaque(__be32 *p, struct nfs4_string *str)
+{
+	str->len = be32_to_cpup(p++);
+	str->data = (char *)p;
+
+	p += XDR_QUADLEN(str->len);
+	return p;
+}
+
+/*
+ * struct pnfs_osd_targetid {
+ *	u32			oti_type;
+ *	struct nfs4_string	oti_scsi_device_id;
+ * };// size 4 + [variable]
+ */
+static __be32 *
+__read_targetid(__be32 *p, struct pnfs_osd_targetid* targetid)
+{
+	u32 oti_type;
+
+	oti_type = be32_to_cpup(p++);
+	targetid->oti_type = oti_type;
+
+	switch (oti_type) {
+	case OBJ_TARGET_SCSI_NAME:
+	case OBJ_TARGET_SCSI_DEVICE_ID:
+		p = __read_u8_opaque(p, &targetid->oti_scsi_device_id);
+	}
+
+	return p;
+}
+
+/*
+ * struct pnfs_osd_net_addr {
+ *	struct nfs4_string	r_netid;
+ *	struct nfs4_string	r_addr;
+ * };
+ */
+static __be32 *
+__read_net_addr(__be32 *p, struct pnfs_osd_net_addr* netaddr)
+{
+	p = __read_u8_opaque(p, &netaddr->r_netid);
+	p = __read_u8_opaque(p, &netaddr->r_addr);
+
+	return p;
+}
+
+/*
+ * struct pnfs_osd_targetaddr {
+ *	u32				ota_available;
+ *	struct pnfs_osd_net_addr	ota_netaddr;
+ * };
+ */
+static __be32 *
+__read_targetaddr(__be32 *p, struct pnfs_osd_targetaddr *targetaddr)
+{
+	u32 ota_available;
+
+	ota_available = be32_to_cpup(p++);
+	targetaddr->ota_available = ota_available;
+
+	if (ota_available)
+		p = __read_net_addr(p, &targetaddr->ota_netaddr);
+
+
+	return p;
+}
+
+/*
+ * struct pnfs_osd_deviceaddr {
+ *	struct pnfs_osd_targetid	oda_targetid;
+ *	struct pnfs_osd_targetaddr	oda_targetaddr;
+ *	u8				oda_lun[8];
+ *	struct nfs4_string		oda_systemid;
+ *	struct pnfs_osd_object_cred	oda_root_obj_cred;
+ *	struct nfs4_string		oda_osdname;
+ * };
+ */
+
+/* We need this version for the pnfs_osd_xdr_decode_deviceaddr which does
+ * not have an xdr_stream
+ */
+static __be32 *
+__read_opaque_cred(__be32 *p,
+			      struct pnfs_osd_opaque_cred *opaque_cred)
+{
+	opaque_cred->cred_len = be32_to_cpu(*p++);
+	opaque_cred->cred = p;
+	return p + XDR_QUADLEN(opaque_cred->cred_len);
+}
+
+static __be32 *
+__read_object_cred(__be32 *p, struct pnfs_osd_object_cred *comp)
+{
+	p = _osd_xdr_decode_objid(p, &comp->oc_object_id);
+	comp->oc_osd_version = be32_to_cpup(p++);
+	comp->oc_cap_key_sec = be32_to_cpup(p++);
+
+	p = __read_opaque_cred(p, &comp->oc_cap_key);
+	p = __read_opaque_cred(p, &comp->oc_cap);
+	return p;
+}
+
+void pnfs_osd_xdr_decode_deviceaddr(
+	struct pnfs_osd_deviceaddr *deviceaddr, __be32 *p)
+{
+	p = __read_targetid(p, &deviceaddr->oda_targetid);
+
+	p = __read_targetaddr(p, &deviceaddr->oda_targetaddr);
+
+	p = xdr_decode_opaque_fixed(p, deviceaddr->oda_lun,
+				    sizeof(deviceaddr->oda_lun));
+
+	p = __read_u8_opaque(p, &deviceaddr->oda_systemid);
+
+	p = __read_object_cred(p, &deviceaddr->oda_root_obj_cred);
+
+	p = __read_u8_opaque(p, &deviceaddr->oda_osdname);
+
+	/* libosd likes this terminated in dbg. It's last, so no problems */
+	deviceaddr->oda_osdname.data[deviceaddr->oda_osdname.len] = 0;
+}
+
+/*
+ * struct pnfs_osd_layoutupdate {
+ *	u32	dsu_valid;
+ *	s64	dsu_delta;
+ *	u32	olu_ioerr_flag;
+ * }; xdr size 4 + 8 + 4
+ */
+int
+pnfs_osd_xdr_encode_layoutupdate(struct xdr_stream *xdr,
+				 struct pnfs_osd_layoutupdate *lou)
+{
+	__be32 *p = xdr_reserve_space(xdr,  4 + 8 + 4);
+
+	if (!p)
+		return -E2BIG;
+
+	*p++ = cpu_to_be32(lou->dsu_valid);
+	if (lou->dsu_valid)
+		p = xdr_encode_hyper(p, lou->dsu_delta);
+	*p++ = cpu_to_be32(lou->olu_ioerr_flag);
+	return 0;
+}
+
+/*
+ * struct pnfs_osd_objid {
+ *	struct nfs4_deviceid	oid_device_id;
+ *	u64			oid_partition_id;
+ *	u64			oid_object_id;
+ * }; // xdr size 32 bytes
+ */
+static inline __be32 *
+pnfs_osd_xdr_encode_objid(__be32 *p, struct pnfs_osd_objid *object_id)
+{
+	p = xdr_encode_opaque_fixed(p, &object_id->oid_device_id.data,
+				    sizeof(object_id->oid_device_id.data));
+	p = xdr_encode_hyper(p, object_id->oid_partition_id);
+	p = xdr_encode_hyper(p, object_id->oid_object_id);
+
+	return p;
+}
+
+/*
+ * struct pnfs_osd_ioerr {
+ *	struct pnfs_osd_objid	oer_component;
+ *	u64			oer_comp_offset;
+ *	u64			oer_comp_length;
+ *	u32			oer_iswrite;
+ *	u32			oer_errno;
+ * }; // xdr size 32 + 24 bytes
+ */
+void pnfs_osd_xdr_encode_ioerr(__be32 *p, struct pnfs_osd_ioerr *ioerr)
+{
+	p = pnfs_osd_xdr_encode_objid(p, &ioerr->oer_component);
+	p = xdr_encode_hyper(p, ioerr->oer_comp_offset);
+	p = xdr_encode_hyper(p, ioerr->oer_comp_length);
+	*p++ = cpu_to_be32(ioerr->oer_iswrite);
+	*p   = cpu_to_be32(ioerr->oer_errno);
+}
+
+__be32 *pnfs_osd_xdr_ioerr_reserve_space(struct xdr_stream *xdr)
+{
+	__be32 *p;
+
+	p = xdr_reserve_space(xdr, 32 + 24);
+	if (unlikely(!p))
+		dprintk("%s: out of xdr space\n", __func__);
+
+	return p;
+}

fs/nfs/pnfs.h

@@ -0,0 +1,573 @@
+/*
+ *  pNFS client data structures.
+ *
+ *  Copyright (c) 2002
+ *  The Regents of the University of Michigan
+ *  All Rights Reserved
+ *
+ *  Dean Hildebrand <dhildebz@umich.edu>
+ *
+ *  Permission is granted to use, copy, create derivative works, and
+ *  redistribute this software and such derivative works for any purpose,
+ *  so long as the name of the University of Michigan is not used in
+ *  any advertising or publicity pertaining to the use or distribution
+ *  of this software without specific, written prior authorization. If
+ *  the above copyright notice or any other identification of the
+ *  University of Michigan is included in any copy of any portion of
+ *  this software, then the disclaimer below must also be included.
+ *
+ *  This software is provided as is, without representation or warranty
+ *  of any kind either express or implied, including without limitation
+ *  the implied warranties of merchantability, fitness for a particular
+ *  purpose, or noninfringement.  The Regents of the University of
+ *  Michigan shall not be liable for any damages, including special,
+ *  indirect, incidental, or consequential damages, with respect to any
+ *  claim arising out of or in connection with the use of the software,
+ *  even if it has been or is hereafter advised of the possibility of
+ *  such damages.
+ */
+
+#ifndef FS_NFS_PNFS_H
+#define FS_NFS_PNFS_H
+
+#include <linux/nfs_fs.h>
+#include <linux/nfs_page.h>
+#include <linux/workqueue.h>
+
+enum {
+	NFS_LSEG_VALID = 0,	/* cleared when lseg is recalled/returned */
+	NFS_LSEG_ROC,		/* roc bit received from server */
+	NFS_LSEG_LAYOUTCOMMIT,	/* layoutcommit bit set for layoutcommit */
+};
+
+struct pnfs_layout_segment {
+	struct list_head pls_list;
+	struct list_head pls_lc_list;
+	struct pnfs_layout_range pls_range;
+	atomic_t pls_refcount;
+	unsigned long pls_flags;
+	struct pnfs_layout_hdr *pls_layout;
+	struct work_struct pls_work;
+};
+
+enum pnfs_try_status {
+	PNFS_ATTEMPTED     = 0,
+	PNFS_NOT_ATTEMPTED = 1,
+};
+
+#ifdef CONFIG_NFS_V4_1
+
+#define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4"
+
+enum {
+	NFS_LAYOUT_RO_FAILED = 0,	/* get ro layout failed stop trying */
+	NFS_LAYOUT_RW_FAILED,		/* get rw layout failed stop trying */
+	NFS_LAYOUT_BULK_RECALL,		/* bulk recall affecting layout */
+	NFS_LAYOUT_ROC,			/* some lseg had roc bit set */
+	NFS_LAYOUT_RETURN,		/* Return this layout ASAP */
+	NFS_LAYOUT_INVALID_STID,	/* layout stateid id is invalid */
+};
+
+enum layoutdriver_policy_flags {
+	/* Should the pNFS client commit and return the layout upon truncate to
+	 * a smaller size */
+	PNFS_LAYOUTRET_ON_SETATTR	= 1 << 0,
+	PNFS_LAYOUTRET_ON_ERROR		= 1 << 1,
+	PNFS_READ_WHOLE_PAGE		= 1 << 2,
+};
+
+struct nfs4_deviceid_node;
+
+/* Per-layout driver specific registration structure */
+struct pnfs_layoutdriver_type {
+	struct list_head pnfs_tblid;
+	const u32 id;
+	const char *name;
+	struct module *owner;
+	unsigned flags;
+	unsigned max_deviceinfo_size;
+
+	int (*set_layoutdriver) (struct nfs_server *, const struct nfs_fh *);
+	int (*clear_layoutdriver) (struct nfs_server *);
+
+	struct pnfs_layout_hdr * (*alloc_layout_hdr) (struct inode *inode, gfp_t gfp_flags);
+	void (*free_layout_hdr) (struct pnfs_layout_hdr *);
+
+	struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);
+	void (*free_lseg) (struct pnfs_layout_segment *lseg);
+
+	void (*return_range) (struct pnfs_layout_hdr *lo,
+			      struct pnfs_layout_range *range);
+
+	/* test for nfs page cache coalescing */
+	const struct nfs_pageio_ops *pg_read_ops;
+	const struct nfs_pageio_ops *pg_write_ops;
+
+	struct pnfs_ds_commit_info *(*get_ds_info) (struct inode *inode);
+	void (*mark_request_commit) (struct nfs_page *req,
+				     struct pnfs_layout_segment *lseg,
+				     struct nfs_commit_info *cinfo);
+	void (*clear_request_commit) (struct nfs_page *req,
+				      struct nfs_commit_info *cinfo);
+	int (*scan_commit_lists) (struct nfs_commit_info *cinfo,
+				  int max);
+	void (*recover_commit_reqs) (struct list_head *list,
+				     struct nfs_commit_info *cinfo);
+	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
+						struct page *page);
+	int (*commit_pagelist)(struct inode *inode,
+			       struct list_head *mds_pages,
+			       int how,
+			       struct nfs_commit_info *cinfo);
+
+	/*
+	 * Return PNFS_ATTEMPTED to indicate the layout code has attempted
+	 * I/O, else return PNFS_NOT_ATTEMPTED to fall back to normal NFS
+	 */
+	enum pnfs_try_status (*read_pagelist)(struct nfs_pgio_header *);
+	enum pnfs_try_status (*write_pagelist)(struct nfs_pgio_header *, int);
+
+	void (*free_deviceid_node) (struct nfs4_deviceid_node *);
+	struct nfs4_deviceid_node * (*alloc_deviceid_node)
+			(struct nfs_server *server, struct pnfs_device *pdev,
+			gfp_t gfp_flags);
+
+	void (*encode_layoutreturn) (struct pnfs_layout_hdr *layoutid,
+				     struct xdr_stream *xdr,
+				     const struct nfs4_layoutreturn_args *args);
+
+	void (*cleanup_layoutcommit) (struct nfs4_layoutcommit_data *data);
+	int (*prepare_layoutcommit) (struct nfs4_layoutcommit_args *args);
+	void (*encode_layoutcommit) (struct pnfs_layout_hdr *lo,
+				     struct xdr_stream *xdr,
+				     const struct nfs4_layoutcommit_args *args);
+};
+
+struct pnfs_layout_hdr {
+	atomic_t		plh_refcount;
+	struct list_head	plh_layouts;   /* other client layouts */
+	struct list_head	plh_bulk_destroy;
+	struct list_head	plh_segs;      /* layout segments list */
+	nfs4_stateid		plh_stateid;
+	atomic_t		plh_outstanding; /* number of RPCs out */
+	unsigned long		plh_block_lgets; /* block LAYOUTGET if >0 */
+	u32			plh_barrier; /* ignore lower seqids */
+	unsigned long		plh_retry_timestamp;
+	unsigned long		plh_flags;
+	loff_t			plh_lwb; /* last write byte for layoutcommit */
+	struct rpc_cred		*plh_lc_cred; /* layoutcommit cred */
+	struct inode		*plh_inode;
+};
+
+struct pnfs_device {
+	struct nfs4_deviceid dev_id;
+	unsigned int  layout_type;
+	unsigned int  mincount;
+	unsigned int  maxcount;	/* gdia_maxcount */
+	struct page **pages;
+	unsigned int  pgbase;
+	unsigned int  pglen;	/* reply buffer length */
+};
+
+#define NFS4_PNFS_GETDEVLIST_MAXNUM 16
+
+struct pnfs_devicelist {
+	unsigned int		eof;
+	unsigned int		num_devs;
+	struct nfs4_deviceid	dev_id[NFS4_PNFS_GETDEVLIST_MAXNUM];
+};
+
+extern int pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *);
+extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *);
+
+/* nfs4proc.c */
+extern int nfs4_proc_getdeviceinfo(struct nfs_server *server,
+				   struct pnfs_device *dev,
+				   struct rpc_cred *cred);
+extern struct pnfs_layout_segment* nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags);
+extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp);
+
+/* pnfs.c */
+void pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo);
+void pnfs_put_lseg(struct pnfs_layout_segment *lseg);
+void pnfs_put_lseg_locked(struct pnfs_layout_segment *lseg);
+
+void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, u32);
+void unset_pnfs_layoutdriver(struct nfs_server *);
+void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *);
+int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc);
+void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio,
+			        struct nfs_page *req, u64 wb_size);
+int pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc);
+size_t pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio,
+			    struct nfs_page *prev, struct nfs_page *req);
+void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg);
+struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp);
+void pnfs_free_lseg_list(struct list_head *tmp_list);
+void pnfs_destroy_layout(struct nfs_inode *);
+void pnfs_destroy_all_layouts(struct nfs_client *);
+int pnfs_destroy_layouts_byfsid(struct nfs_client *clp,
+		struct nfs_fsid *fsid,
+		bool is_recall);
+int pnfs_destroy_layouts_byclid(struct nfs_client *clp,
+		bool is_recall);
+void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo);
+void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo,
+			     const nfs4_stateid *new,
+			     bool update_barrier);
+int pnfs_choose_layoutget_stateid(nfs4_stateid *dst,
+				  struct pnfs_layout_hdr *lo,
+				  struct nfs4_state *open_state);
+int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
+				struct list_head *tmp_list,
+				struct pnfs_layout_range *recall_range);
+bool pnfs_roc(struct inode *ino);
+void pnfs_roc_release(struct inode *ino);
+void pnfs_roc_set_barrier(struct inode *ino, u32 barrier);
+bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task);
+void pnfs_set_layoutcommit(struct nfs_pgio_header *);
+void pnfs_commit_set_layoutcommit(struct nfs_commit_data *data);
+void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data);
+int pnfs_layoutcommit_inode(struct inode *inode, bool sync);
+int _pnfs_return_layout(struct inode *);
+int pnfs_commit_and_return_layout(struct inode *);
+void pnfs_ld_write_done(struct nfs_pgio_header *);
+void pnfs_ld_read_done(struct nfs_pgio_header *);
+struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino,
+					       struct nfs_open_context *ctx,
+					       loff_t pos,
+					       u64 count,
+					       enum pnfs_iomode iomode,
+					       gfp_t gfp_flags);
+
+void nfs4_deviceid_mark_client_invalid(struct nfs_client *clp);
+int pnfs_read_done_resend_to_mds(struct nfs_pgio_header *);
+int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *);
+struct nfs4_threshold *pnfs_mdsthreshold_alloc(void);
+
+/* nfs4_deviceid_flags */
+enum {
+	NFS_DEVICEID_INVALID = 0,       /* set when MDS clientid recalled */
+	NFS_DEVICEID_UNAVAILABLE,	/* device temporarily unavailable */
+};
+
+/* pnfs_dev.c */
+struct nfs4_deviceid_node {
+	struct hlist_node		node;
+	struct hlist_node		tmpnode;
+	const struct pnfs_layoutdriver_type *ld;
+	const struct nfs_client		*nfs_client;
+	unsigned long 			flags;
+	unsigned long			timestamp_unavailable;
+	struct nfs4_deviceid		deviceid;
+	atomic_t			ref;
+};
+
+struct nfs4_deviceid_node *
+nfs4_find_get_deviceid(struct nfs_server *server,
+		const struct nfs4_deviceid *id, struct rpc_cred *cred,
+		gfp_t gfp_mask);
+void nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *);
+void nfs4_init_deviceid_node(struct nfs4_deviceid_node *, struct nfs_server *,
+			     const struct nfs4_deviceid *);
+bool nfs4_put_deviceid_node(struct nfs4_deviceid_node *);
+void nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node);
+bool nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node);
+void nfs4_deviceid_purge_client(const struct nfs_client *);
+
+static inline struct nfs4_deviceid_node *
+nfs4_get_deviceid(struct nfs4_deviceid_node *d)
+{
+	atomic_inc(&d->ref);
+	return d;
+}
+
+static inline struct pnfs_layout_segment *
+pnfs_get_lseg(struct pnfs_layout_segment *lseg)
+{
+	if (lseg) {
+		atomic_inc(&lseg->pls_refcount);
+		smp_mb__after_atomic();
+	}
+	return lseg;
+}
+
+/* Return true if a layout driver is being used for this mountpoint */
+static inline int pnfs_enabled_sb(struct nfs_server *nfss)
+{
+	return nfss->pnfs_curr_ld != NULL;
+}
+
+static inline int
+pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how,
+		 struct nfs_commit_info *cinfo)
+{
+	if (cinfo->ds == NULL || cinfo->ds->ncommitting == 0)
+		return PNFS_NOT_ATTEMPTED;
+	return NFS_SERVER(inode)->pnfs_curr_ld->commit_pagelist(inode, mds_pages, how, cinfo);
+}
+
+static inline struct pnfs_ds_commit_info *
+pnfs_get_ds_info(struct inode *inode)
+{
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+
+	if (ld == NULL || ld->get_ds_info == NULL)
+		return NULL;
+	return ld->get_ds_info(inode);
+}
+
+static inline bool
+pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
+			 struct nfs_commit_info *cinfo)
+{
+	struct inode *inode = req->wb_context->dentry->d_inode;
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+
+	if (lseg == NULL || ld->mark_request_commit == NULL)
+		return false;
+	ld->mark_request_commit(req, lseg, cinfo);
+	return true;
+}
+
+static inline bool
+pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo)
+{
+	struct inode *inode = req->wb_context->dentry->d_inode;
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+
+	if (ld == NULL || ld->clear_request_commit == NULL)
+		return false;
+	ld->clear_request_commit(req, cinfo);
+	return true;
+}
+
+static inline int
+pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
+		       int max)
+{
+	if (cinfo->ds == NULL || cinfo->ds->nwritten == 0)
+		return 0;
+	else
+		return NFS_SERVER(inode)->pnfs_curr_ld->scan_commit_lists(cinfo, max);
+}
+
+static inline void
+pnfs_recover_commit_reqs(struct inode *inode, struct list_head *list,
+			 struct nfs_commit_info *cinfo)
+{
+	if (cinfo->ds == NULL || cinfo->ds->nwritten == 0)
+		return;
+	NFS_SERVER(inode)->pnfs_curr_ld->recover_commit_reqs(list, cinfo);
+}
+
+static inline struct nfs_page *
+pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
+			struct page *page)
+{
+	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+
+	if (ld == NULL || ld->search_commit_reqs == NULL)
+		return NULL;
+	return ld->search_commit_reqs(cinfo, page);
+}
+
+/* Should the pNFS client commit and return the layout upon a setattr */
+static inline bool
+pnfs_ld_layoutret_on_setattr(struct inode *inode)
+{
+	if (!pnfs_enabled_sb(NFS_SERVER(inode)))
+		return false;
+	return NFS_SERVER(inode)->pnfs_curr_ld->flags &
+		PNFS_LAYOUTRET_ON_SETATTR;
+}
+
+static inline bool
+pnfs_ld_read_whole_page(struct inode *inode)
+{
+	if (!pnfs_enabled_sb(NFS_SERVER(inode)))
+		return false;
+	return NFS_SERVER(inode)->pnfs_curr_ld->flags & PNFS_READ_WHOLE_PAGE;
+}
+
+static inline bool
+pnfs_layoutcommit_outstanding(struct inode *inode)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+
+	return test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags) != 0 ||
+		test_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags) != 0;
+}
+
+static inline int pnfs_return_layout(struct inode *ino)
+{
+	struct nfs_inode *nfsi = NFS_I(ino);
+	struct nfs_server *nfss = NFS_SERVER(ino);
+
+	if (pnfs_enabled_sb(nfss) && nfsi->layout)
+		return _pnfs_return_layout(ino);
+
+	return 0;
+}
+
+static inline bool
+pnfs_use_threshold(struct nfs4_threshold **dst, struct nfs4_threshold *src,
+		   struct nfs_server *nfss)
+{
+	return (dst && src && src->bm != 0 && nfss->pnfs_curr_ld &&
+					nfss->pnfs_curr_ld->id == src->l_type);
+}
+
+#ifdef NFS_DEBUG
+void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id);
+#else
+static inline void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id)
+{
+}
+#endif /* NFS_DEBUG */
+#else  /* CONFIG_NFS_V4_1 */
+
+static inline void pnfs_destroy_all_layouts(struct nfs_client *clp)
+{
+}
+
+static inline void pnfs_destroy_layout(struct nfs_inode *nfsi)
+{
+}
+
+static inline struct pnfs_layout_segment *
+pnfs_get_lseg(struct pnfs_layout_segment *lseg)
+{
+	return NULL;
+}
+
+static inline void pnfs_put_lseg(struct pnfs_layout_segment *lseg)
+{
+}
+
+static inline int pnfs_return_layout(struct inode *ino)
+{
+	return 0;
+}
+
+static inline int pnfs_commit_and_return_layout(struct inode *inode)
+{
+	return 0;
+}
+
+static inline bool
+pnfs_ld_layoutret_on_setattr(struct inode *inode)
+{
+	return false;
+}
+
+static inline bool
+pnfs_ld_read_whole_page(struct inode *inode)
+{
+	return false;
+}
+
+static inline bool
+pnfs_roc(struct inode *ino)
+{
+	return false;
+}
+
+static inline void
+pnfs_roc_release(struct inode *ino)
+{
+}
+
+static inline void
+pnfs_roc_set_barrier(struct inode *ino, u32 barrier)
+{
+}
+
+static inline bool
+pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task)
+{
+	return false;
+}
+
+static inline void set_pnfs_layoutdriver(struct nfs_server *s,
+					 const struct nfs_fh *mntfh, u32 id)
+{
+}
+
+static inline void unset_pnfs_layoutdriver(struct nfs_server *s)
+{
+}
+
+static inline int
+pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how,
+		 struct nfs_commit_info *cinfo)
+{
+	return PNFS_NOT_ATTEMPTED;
+}
+
+static inline struct pnfs_ds_commit_info *
+pnfs_get_ds_info(struct inode *inode)
+{
+	return NULL;
+}
+
+static inline bool
+pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
+			 struct nfs_commit_info *cinfo)
+{
+	return false;
+}
+
+static inline bool
+pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo)
+{
+	return false;
+}
+
+static inline int
+pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
+		       int max)
+{
+	return 0;
+}
+
+static inline void
+pnfs_recover_commit_reqs(struct inode *inode, struct list_head *list,
+			 struct nfs_commit_info *cinfo)
+{
+}
+
+static inline struct nfs_page *
+pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
+			struct page *page)
+{
+	return NULL;
+}
+
+static inline int pnfs_layoutcommit_inode(struct inode *inode, bool sync)
+{
+	return 0;
+}
+
+static inline bool
+pnfs_use_threshold(struct nfs4_threshold **dst, struct nfs4_threshold *src,
+		   struct nfs_server *nfss)
+{
+	return false;
+}
+
+static inline bool
+pnfs_layoutcommit_outstanding(struct inode *inode)
+{
+	return false;
+}
+
+
+static inline struct nfs4_threshold *pnfs_mdsthreshold_alloc(void)
+{
+	return NULL;
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
+#endif /* FS_NFS_PNFS_H */

fs/nfs/pnfs_dev.c

@@ -0,0 +1,360 @@
+/*
+ *  Device operations for the pnfs client.
+ *
+ *  Copyright (c) 2002
+ *  The Regents of the University of Michigan
+ *  All Rights Reserved
+ *
+ *  Dean Hildebrand <dhildebz@umich.edu>
+ *  Garth Goodson   <Garth.Goodson@netapp.com>
+ *
+ *  Permission is granted to use, copy, create derivative works, and
+ *  redistribute this software and such derivative works for any purpose,
+ *  so long as the name of the University of Michigan is not used in
+ *  any advertising or publicity pertaining to the use or distribution
+ *  of this software without specific, written prior authorization. If
+ *  the above copyright notice or any other identification of the
+ *  University of Michigan is included in any copy of any portion of
+ *  this software, then the disclaimer below must also be included.
+ *
+ *  This software is provided as is, without representation or warranty
+ *  of any kind either express or implied, including without limitation
+ *  the implied warranties of merchantability, fitness for a particular
+ *  purpose, or noninfringement.  The Regents of the University of
+ *  Michigan shall not be liable for any damages, including special,
+ *  indirect, incidental, or consequential damages, with respect to any
+ *  claim arising out of or in connection with the use of the software,
+ *  even if it has been or is hereafter advised of the possibility of
+ *  such damages.
+ */
+
+#include <linux/export.h>
+#include <linux/nfs_fs.h>
+#include "nfs4session.h"
+#include "internal.h"
+#include "pnfs.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS
+
+/*
+ * Device ID RCU cache. A device ID is unique per server and layout type.
+ */
+#define NFS4_DEVICE_ID_HASH_BITS	5
+#define NFS4_DEVICE_ID_HASH_SIZE	(1 << NFS4_DEVICE_ID_HASH_BITS)
+#define NFS4_DEVICE_ID_HASH_MASK	(NFS4_DEVICE_ID_HASH_SIZE - 1)
+
+#define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ)
+
+static struct hlist_head nfs4_deviceid_cache[NFS4_DEVICE_ID_HASH_SIZE];
+static DEFINE_SPINLOCK(nfs4_deviceid_lock);
+
+#ifdef NFS_DEBUG
+void
+nfs4_print_deviceid(const struct nfs4_deviceid *id)
+{
+	u32 *p = (u32 *)id;
+
+	dprintk("%s: device id= [%x%x%x%x]\n", __func__,
+		p[0], p[1], p[2], p[3]);
+}
+EXPORT_SYMBOL_GPL(nfs4_print_deviceid);
+#endif
+
+static inline u32
+nfs4_deviceid_hash(const struct nfs4_deviceid *id)
+{
+	unsigned char *cptr = (unsigned char *)id->data;
+	unsigned int nbytes = NFS4_DEVICEID4_SIZE;
+	u32 x = 0;
+
+	while (nbytes--) {
+		x *= 37;
+		x += *cptr++;
+	}
+	return x & NFS4_DEVICE_ID_HASH_MASK;
+}
+
+static struct nfs4_deviceid_node *
+_lookup_deviceid(const struct pnfs_layoutdriver_type *ld,
+		 const struct nfs_client *clp, const struct nfs4_deviceid *id,
+		 long hash)
+{
+	struct nfs4_deviceid_node *d;
+
+	hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[hash], node)
+		if (d->ld == ld && d->nfs_client == clp &&
+		    !memcmp(&d->deviceid, id, sizeof(*id))) {
+			if (atomic_read(&d->ref))
+				return d;
+			else
+				continue;
+		}
+	return NULL;
+}
+
+static struct nfs4_deviceid_node *
+nfs4_get_device_info(struct nfs_server *server,
+		const struct nfs4_deviceid *dev_id,
+		struct rpc_cred *cred, gfp_t gfp_flags)
+{
+	struct nfs4_deviceid_node *d = NULL;
+	struct pnfs_device *pdev = NULL;
+	struct page **pages = NULL;
+	u32 max_resp_sz;
+	int max_pages;
+	int rc, i;
+
+	/*
+	 * Use the session max response size as the basis for setting
+	 * GETDEVICEINFO's maxcount
+	 */
+	max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
+	if (server->pnfs_curr_ld->max_deviceinfo_size &&
+	    server->pnfs_curr_ld->max_deviceinfo_size < max_resp_sz)
+		max_resp_sz = server->pnfs_curr_ld->max_deviceinfo_size;
+	max_pages = nfs_page_array_len(0, max_resp_sz);
+	dprintk("%s: server %p max_resp_sz %u max_pages %d\n",
+		__func__, server, max_resp_sz, max_pages);
+
+	pdev = kzalloc(sizeof(*pdev), gfp_flags);
+	if (!pdev)
+		return NULL;
+
+	pages = kcalloc(max_pages, sizeof(struct page *), gfp_flags);
+	if (!pages)
+		goto out_free_pdev;
+
+	for (i = 0; i < max_pages; i++) {
+		pages[i] = alloc_page(gfp_flags);
+		if (!pages[i])
+			goto out_free_pages;
+	}
+
+	memcpy(&pdev->dev_id, dev_id, sizeof(*dev_id));
+	pdev->layout_type = server->pnfs_curr_ld->id;
+	pdev->pages = pages;
+	pdev->pgbase = 0;
+	pdev->pglen = max_resp_sz;
+	pdev->mincount = 0;
+	pdev->maxcount = max_resp_sz - nfs41_maxgetdevinfo_overhead;
+
+	rc = nfs4_proc_getdeviceinfo(server, pdev, cred);
+	dprintk("%s getdevice info returns %d\n", __func__, rc);
+	if (rc)
+		goto out_free_pages;
+
+	/*
+	 * Found new device, need to decode it and then add it to the
+	 * list of known devices for this mountpoint.
+	 */
+	d = server->pnfs_curr_ld->alloc_deviceid_node(server, pdev,
+			gfp_flags);
+
+out_free_pages:
+	for (i = 0; i < max_pages; i++)
+		__free_page(pages[i]);
+	kfree(pages);
+out_free_pdev:
+	kfree(pdev);
+	dprintk("<-- %s d %p\n", __func__, d);
+	return d;
+}
+
+/*
+ * Lookup a deviceid in cache and get a reference count on it if found
+ *
+ * @clp nfs_client associated with deviceid
+ * @id deviceid to look up
+ */
+static struct nfs4_deviceid_node *
+__nfs4_find_get_deviceid(struct nfs_server *server,
+		const struct nfs4_deviceid *id, long hash)
+{
+	struct nfs4_deviceid_node *d;
+
+	rcu_read_lock();
+	d = _lookup_deviceid(server->pnfs_curr_ld, server->nfs_client, id,
+			hash);
+	if (d != NULL)
+		atomic_inc(&d->ref);
+	rcu_read_unlock();
+	return d;
+}
+
+struct nfs4_deviceid_node *
+nfs4_find_get_deviceid(struct nfs_server *server,
+		const struct nfs4_deviceid *id, struct rpc_cred *cred,
+		gfp_t gfp_mask)
+{
+	long hash = nfs4_deviceid_hash(id);
+	struct nfs4_deviceid_node *d, *new;
+
+	d = __nfs4_find_get_deviceid(server, id, hash);
+	if (d)
+		return d;
+
+	new = nfs4_get_device_info(server, id, cred, gfp_mask);
+	if (!new)
+		return new;
+
+	spin_lock(&nfs4_deviceid_lock);
+	d = __nfs4_find_get_deviceid(server, id, hash);
+	if (d) {
+		spin_unlock(&nfs4_deviceid_lock);
+		server->pnfs_curr_ld->free_deviceid_node(new);
+		return d;
+	}
+	hlist_add_head_rcu(&new->node, &nfs4_deviceid_cache[hash]);
+	atomic_inc(&new->ref);
+	spin_unlock(&nfs4_deviceid_lock);
+
+	return new;
+}
+EXPORT_SYMBOL_GPL(nfs4_find_get_deviceid);
+
+/*
+ * Remove a deviceid from cache
+ *
+ * @clp nfs_client associated with deviceid
+ * @id the deviceid to unhash
+ *
+ * @ret the unhashed node, if found and dereferenced to zero, NULL otherwise.
+ */
+void
+nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *ld,
+			 const struct nfs_client *clp, const struct nfs4_deviceid *id)
+{
+	struct nfs4_deviceid_node *d;
+
+	spin_lock(&nfs4_deviceid_lock);
+	rcu_read_lock();
+	d = _lookup_deviceid(ld, clp, id, nfs4_deviceid_hash(id));
+	rcu_read_unlock();
+	if (!d) {
+		spin_unlock(&nfs4_deviceid_lock);
+		return;
+	}
+	hlist_del_init_rcu(&d->node);
+	spin_unlock(&nfs4_deviceid_lock);
+	synchronize_rcu();
+
+	/* balance the initial ref set in pnfs_insert_deviceid */
+	if (atomic_dec_and_test(&d->ref))
+		d->ld->free_deviceid_node(d);
+}
+EXPORT_SYMBOL_GPL(nfs4_delete_deviceid);
+
+void
+nfs4_init_deviceid_node(struct nfs4_deviceid_node *d, struct nfs_server *server,
+			const struct nfs4_deviceid *id)
+{
+	INIT_HLIST_NODE(&d->node);
+	INIT_HLIST_NODE(&d->tmpnode);
+	d->ld = server->pnfs_curr_ld;
+	d->nfs_client = server->nfs_client;
+	d->flags = 0;
+	d->deviceid = *id;
+	atomic_set(&d->ref, 1);
+}
+EXPORT_SYMBOL_GPL(nfs4_init_deviceid_node);
+
+/*
+ * Dereference a deviceid node and delete it when its reference count drops
+ * to zero.
+ *
+ * @d deviceid node to put
+ *
+ * return true iff the node was deleted
+ * Note that since the test for d->ref == 0 is sufficient to establish
+ * that the node is no longer hashed in the global device id cache.
+ */
+bool
+nfs4_put_deviceid_node(struct nfs4_deviceid_node *d)
+{
+	if (!atomic_dec_and_test(&d->ref))
+		return false;
+	d->ld->free_deviceid_node(d);
+	return true;
+}
+EXPORT_SYMBOL_GPL(nfs4_put_deviceid_node);
+
+void
+nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node)
+{
+	node->timestamp_unavailable = jiffies;
+	set_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags);
+}
+EXPORT_SYMBOL_GPL(nfs4_mark_deviceid_unavailable);
+
+bool
+nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node)
+{
+	if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) {
+		unsigned long start, end;
+
+		end = jiffies;
+		start = end - PNFS_DEVICE_RETRY_TIMEOUT;
+		if (time_in_range(node->timestamp_unavailable, start, end))
+			return true;
+		clear_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags);
+	}
+	return false;
+}
+EXPORT_SYMBOL_GPL(nfs4_test_deviceid_unavailable);
+
+static void
+_deviceid_purge_client(const struct nfs_client *clp, long hash)
+{
+	struct nfs4_deviceid_node *d;
+	HLIST_HEAD(tmp);
+
+	spin_lock(&nfs4_deviceid_lock);
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[hash], node)
+		if (d->nfs_client == clp && atomic_read(&d->ref)) {
+			hlist_del_init_rcu(&d->node);
+			hlist_add_head(&d->tmpnode, &tmp);
+		}
+	rcu_read_unlock();
+	spin_unlock(&nfs4_deviceid_lock);
+
+	if (hlist_empty(&tmp))
+		return;
+
+	synchronize_rcu();
+	while (!hlist_empty(&tmp)) {
+		d = hlist_entry(tmp.first, struct nfs4_deviceid_node, tmpnode);
+		hlist_del(&d->tmpnode);
+		if (atomic_dec_and_test(&d->ref))
+			d->ld->free_deviceid_node(d);
+	}
+}
+
+void
+nfs4_deviceid_purge_client(const struct nfs_client *clp)
+{
+	long h;
+
+	if (!(clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_MDS))
+		return;
+	for (h = 0; h < NFS4_DEVICE_ID_HASH_SIZE; h++)
+		_deviceid_purge_client(clp, h);
+}
+
+/*
+ * Stop use of all deviceids associated with an nfs_client
+ */
+void
+nfs4_deviceid_mark_client_invalid(struct nfs_client *clp)
+{
+	struct nfs4_deviceid_node *d;
+	int i;
+
+	rcu_read_lock();
+	for (i = 0; i < NFS4_DEVICE_ID_HASH_SIZE; i ++){
+		hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[i], node)
+			if (d->nfs_client == clp)
+				set_bit(NFS_DEVICEID_INVALID, &d->flags);
+	}
+	rcu_read_unlock();
+}

fs/nfs/proc.c

@@ -0,0 +1,751 @@
+/*
+ *  linux/fs/nfs/proc.c
+ *
+ *  Copyright (C) 1992, 1993, 1994  Rick Sladkey
+ *
+ *  OS-independent nfs remote procedure call functions
+ *
+ *  Tuned by Alan Cox <A.Cox@swansea.ac.uk> for >3K buffers
+ *  so at last we can have decent(ish) throughput off a 
+ *  Sun server.
+ *
+ *  Coding optimized and cleaned up by Florian La Roche.
+ *  Note: Error returns are optimized for NFS_OK, which isn't translated via
+ *  nfs_stat_to_errno(), but happens to be already the right return code.
+ *
+ *  Also, the code currently doesn't check the size of the packet, when
+ *  it decodes the packet.
+ *
+ *  Feel free to fix it and mail me the diffs if it worries you.
+ *
+ *  Completely rewritten to support the new RPC call interface;
+ *  rewrote and moved the entire XDR stuff to xdr.c
+ *  --Olaf Kirch June 1996
+ *
+ *  The code below initializes all auto variables explicitly, otherwise
+ *  it will fail to work as a module (gcc generates a memset call for an
+ *  incomplete struct).
+ */
+
+#include <linux/types.h>
+#include <linux/param.h>
+#include <linux/time.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/in.h>
+#include <linux/pagemap.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_page.h>
+#include <linux/lockd/bind.h>
+#include <linux/freezer.h>
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PROC
+
+/*
+ * Bare-bones access to getattr: this is for nfs_read_super.
+ */
+static int
+nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
+		  struct nfs_fsinfo *info)
+{
+	struct nfs_fattr *fattr = info->fattr;
+	struct nfs2_fsstat fsinfo;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_GETATTR],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= fattr,
+	};
+	int status;
+
+	dprintk("%s: call getattr\n", __func__);
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	/* Retry with default authentication if different */
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
+	dprintk("%s: reply getattr: %d\n", __func__, status);
+	if (status)
+		return status;
+	dprintk("%s: call statfs\n", __func__);
+	msg.rpc_proc = &nfs_procedures[NFSPROC_STATFS];
+	msg.rpc_resp = &fsinfo;
+	status = rpc_call_sync(server->client, &msg, 0);
+	/* Retry with default authentication if different */
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
+	dprintk("%s: reply statfs: %d\n", __func__, status);
+	if (status)
+		return status;
+	info->rtmax  = NFS_MAXDATA;
+	info->rtpref = fsinfo.tsize;
+	info->rtmult = fsinfo.bsize;
+	info->wtmax  = NFS_MAXDATA;
+	info->wtpref = fsinfo.tsize;
+	info->wtmult = fsinfo.bsize;
+	info->dtpref = fsinfo.tsize;
+	info->maxfilesize = 0x7FFFFFFF;
+	info->lease_time = 0;
+	return 0;
+}
+
+/*
+ * One function for each procedure in the NFS protocol.
+ */
+static int
+nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
+		struct nfs_fattr *fattr, struct nfs4_label *label)
+{
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_GETATTR],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= fattr,
+	};
+	int	status;
+
+	dprintk("NFS call  getattr\n");
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply getattr: %d\n", status);
+	return status;
+}
+
+static int
+nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
+		 struct iattr *sattr)
+{
+	struct inode *inode = dentry->d_inode;
+	struct nfs_sattrargs	arg = { 
+		.fh	= NFS_FH(inode),
+		.sattr	= sattr
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_SETATTR],
+		.rpc_argp	= &arg,
+		.rpc_resp	= fattr,
+	};
+	int	status;
+
+	/* Mask out the non-modebit related stuff from attr->ia_mode */
+	sattr->ia_mode &= S_IALLUGO;
+
+	dprintk("NFS call  setattr\n");
+	if (sattr->ia_valid & ATTR_FILE)
+		msg.rpc_cred = nfs_file_cred(sattr->ia_file);
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	if (status == 0)
+		nfs_setattr_update_inode(inode, sattr);
+	dprintk("NFS reply setattr: %d\n", status);
+	return status;
+}
+
+static int
+nfs_proc_lookup(struct inode *dir, struct qstr *name,
+		struct nfs_fh *fhandle, struct nfs_fattr *fattr,
+		struct nfs4_label *label)
+{
+	struct nfs_diropargs	arg = {
+		.fh		= NFS_FH(dir),
+		.name		= name->name,
+		.len		= name->len
+	};
+	struct nfs_diropok	res = {
+		.fh		= fhandle,
+		.fattr		= fattr
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_LOOKUP],
+		.rpc_argp	= &arg,
+		.rpc_resp	= &res,
+	};
+	int			status;
+
+	dprintk("NFS call  lookup %s\n", name->name);
+	nfs_fattr_init(fattr);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	dprintk("NFS reply lookup: %d\n", status);
+	return status;
+}
+
+static int nfs_proc_readlink(struct inode *inode, struct page *page,
+		unsigned int pgbase, unsigned int pglen)
+{
+	struct nfs_readlinkargs	args = {
+		.fh		= NFS_FH(inode),
+		.pgbase		= pgbase,
+		.pglen		= pglen,
+		.pages		= &page
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_READLINK],
+		.rpc_argp	= &args,
+	};
+	int			status;
+
+	dprintk("NFS call  readlink\n");
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	dprintk("NFS reply readlink: %d\n", status);
+	return status;
+}
+
+struct nfs_createdata {
+	struct nfs_createargs arg;
+	struct nfs_diropok res;
+	struct nfs_fh fhandle;
+	struct nfs_fattr fattr;
+};
+
+static struct nfs_createdata *nfs_alloc_createdata(struct inode *dir,
+		struct dentry *dentry, struct iattr *sattr)
+{
+	struct nfs_createdata *data;
+
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+
+	if (data != NULL) {
+		data->arg.fh = NFS_FH(dir);
+		data->arg.name = dentry->d_name.name;
+		data->arg.len = dentry->d_name.len;
+		data->arg.sattr = sattr;
+		nfs_fattr_init(&data->fattr);
+		data->fhandle.size = 0;
+		data->res.fh = &data->fhandle;
+		data->res.fattr = &data->fattr;
+	}
+	return data;
+};
+
+static void nfs_free_createdata(const struct nfs_createdata *data)
+{
+	kfree(data);
+}
+
+static int
+nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
+		int flags)
+{
+	struct nfs_createdata *data;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_CREATE],
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  create %pd\n", dentry);
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+	if (status == 0)
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL);
+	nfs_free_createdata(data);
+out:
+	dprintk("NFS reply create: %d\n", status);
+	return status;
+}
+
+/*
+ * In NFSv2, mknod is grafted onto the create call.
+ */
+static int
+nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
+	       dev_t rdev)
+{
+	struct nfs_createdata *data;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_CREATE],
+	};
+	umode_t mode;
+	int status = -ENOMEM;
+
+	dprintk("NFS call  mknod %pd\n", dentry);
+
+	mode = sattr->ia_mode;
+	if (S_ISFIFO(mode)) {
+		sattr->ia_mode = (mode & ~S_IFMT) | S_IFCHR;
+		sattr->ia_valid &= ~ATTR_SIZE;
+	} else if (S_ISCHR(mode) || S_ISBLK(mode)) {
+		sattr->ia_valid |= ATTR_SIZE;
+		sattr->ia_size = new_encode_dev(rdev);/* get out your barf bag */
+	}
+
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+
+	if (status == -EINVAL && S_ISFIFO(mode)) {
+		sattr->ia_mode = mode;
+		nfs_fattr_init(data->res.fattr);
+		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	}
+	if (status == 0)
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL);
+	nfs_free_createdata(data);
+out:
+	dprintk("NFS reply mknod: %d\n", status);
+	return status;
+}
+  
+static int
+nfs_proc_remove(struct inode *dir, struct qstr *name)
+{
+	struct nfs_removeargs arg = {
+		.fh = NFS_FH(dir),
+		.name = *name,
+	};
+	struct rpc_message msg = { 
+		.rpc_proc = &nfs_procedures[NFSPROC_REMOVE],
+		.rpc_argp = &arg,
+	};
+	int			status;
+
+	dprintk("NFS call  remove %s\n", name->name);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+
+	dprintk("NFS reply remove: %d\n", status);
+	return status;
+}
+
+static void
+nfs_proc_unlink_setup(struct rpc_message *msg, struct inode *dir)
+{
+	msg->rpc_proc = &nfs_procedures[NFSPROC_REMOVE];
+}
+
+static void nfs_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data)
+{
+	rpc_call_start(task);
+}
+
+static int nfs_proc_unlink_done(struct rpc_task *task, struct inode *dir)
+{
+	nfs_mark_for_revalidate(dir);
+	return 1;
+}
+
+static void
+nfs_proc_rename_setup(struct rpc_message *msg, struct inode *dir)
+{
+	msg->rpc_proc = &nfs_procedures[NFSPROC_RENAME];
+}
+
+static void nfs_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data)
+{
+	rpc_call_start(task);
+}
+
+static int
+nfs_proc_rename_done(struct rpc_task *task, struct inode *old_dir,
+		     struct inode *new_dir)
+{
+	nfs_mark_for_revalidate(old_dir);
+	nfs_mark_for_revalidate(new_dir);
+	return 1;
+}
+
+static int
+nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
+{
+	struct nfs_linkargs	arg = {
+		.fromfh		= NFS_FH(inode),
+		.tofh		= NFS_FH(dir),
+		.toname		= name->name,
+		.tolen		= name->len
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_LINK],
+		.rpc_argp	= &arg,
+	};
+	int			status;
+
+	dprintk("NFS call  link %s\n", name->name);
+	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	nfs_mark_for_revalidate(inode);
+	nfs_mark_for_revalidate(dir);
+	dprintk("NFS reply link: %d\n", status);
+	return status;
+}
+
+static int
+nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
+		 unsigned int len, struct iattr *sattr)
+{
+	struct nfs_fh *fh;
+	struct nfs_fattr *fattr;
+	struct nfs_symlinkargs	arg = {
+		.fromfh		= NFS_FH(dir),
+		.fromname	= dentry->d_name.name,
+		.fromlen	= dentry->d_name.len,
+		.pages		= &page,
+		.pathlen	= len,
+		.sattr		= sattr
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_SYMLINK],
+		.rpc_argp	= &arg,
+	};
+	int status = -ENAMETOOLONG;
+
+	dprintk("NFS call  symlink %pd\n", dentry);
+
+	if (len > NFS2_MAXPATHLEN)
+		goto out;
+
+	fh = nfs_alloc_fhandle();
+	fattr = nfs_alloc_fattr();
+	status = -ENOMEM;
+	if (fh == NULL || fattr == NULL)
+		goto out_free;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+
+	/*
+	 * V2 SYMLINK requests don't return any attributes.  Setting the
+	 * filehandle size to zero indicates to nfs_instantiate that it
+	 * should fill in the data with a LOOKUP call on the wire.
+	 */
+	if (status == 0)
+		status = nfs_instantiate(dentry, fh, fattr, NULL);
+
+out_free:
+	nfs_free_fattr(fattr);
+	nfs_free_fhandle(fh);
+out:
+	dprintk("NFS reply symlink: %d\n", status);
+	return status;
+}
+
+static int
+nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
+{
+	struct nfs_createdata *data;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_MKDIR],
+	};
+	int status = -ENOMEM;
+
+	dprintk("NFS call  mkdir %pd\n", dentry);
+	data = nfs_alloc_createdata(dir, dentry, sattr);
+	if (data == NULL)
+		goto out;
+	msg.rpc_argp = &data->arg;
+	msg.rpc_resp = &data->res;
+
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+	if (status == 0)
+		status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL);
+	nfs_free_createdata(data);
+out:
+	dprintk("NFS reply mkdir: %d\n", status);
+	return status;
+}
+
+static int
+nfs_proc_rmdir(struct inode *dir, struct qstr *name)
+{
+	struct nfs_diropargs	arg = {
+		.fh		= NFS_FH(dir),
+		.name		= name->name,
+		.len		= name->len
+	};
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_RMDIR],
+		.rpc_argp	= &arg,
+	};
+	int			status;
+
+	dprintk("NFS call  rmdir %s\n", name->name);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	nfs_mark_for_revalidate(dir);
+	dprintk("NFS reply rmdir: %d\n", status);
+	return status;
+}
+
+/*
+ * The READDIR implementation is somewhat hackish - we pass a temporary
+ * buffer to the encode function, which installs it in the receive
+ * the receive iovec. The decode function just parses the reply to make
+ * sure it is syntactically correct; the entries itself are decoded
+ * from nfs_readdir by calling the decode_entry function directly.
+ */
+static int
+nfs_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
+		 u64 cookie, struct page **pages, unsigned int count, int plus)
+{
+	struct inode		*dir = dentry->d_inode;
+	struct nfs_readdirargs	arg = {
+		.fh		= NFS_FH(dir),
+		.cookie		= cookie,
+		.count		= count,
+		.pages		= pages,
+	};
+	struct rpc_message	msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_READDIR],
+		.rpc_argp	= &arg,
+		.rpc_cred	= cred,
+	};
+	int			status;
+
+	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;
+}
+
+static int
+nfs_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
+			struct nfs_fsstat *stat)
+{
+	struct nfs2_fsstat fsinfo;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_STATFS],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= &fsinfo,
+	};
+	int	status;
+
+	dprintk("NFS call  statfs\n");
+	nfs_fattr_init(stat->fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply statfs: %d\n", status);
+	if (status)
+		goto out;
+	stat->tbytes = (u64)fsinfo.blocks * fsinfo.bsize;
+	stat->fbytes = (u64)fsinfo.bfree  * fsinfo.bsize;
+	stat->abytes = (u64)fsinfo.bavail * fsinfo.bsize;
+	stat->tfiles = 0;
+	stat->ffiles = 0;
+	stat->afiles = 0;
+out:
+	return status;
+}
+
+static int
+nfs_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
+			struct nfs_fsinfo *info)
+{
+	struct nfs2_fsstat fsinfo;
+	struct rpc_message msg = {
+		.rpc_proc	= &nfs_procedures[NFSPROC_STATFS],
+		.rpc_argp	= fhandle,
+		.rpc_resp	= &fsinfo,
+	};
+	int	status;
+
+	dprintk("NFS call  fsinfo\n");
+	nfs_fattr_init(info->fattr);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply fsinfo: %d\n", status);
+	if (status)
+		goto out;
+	info->rtmax  = NFS_MAXDATA;
+	info->rtpref = fsinfo.tsize;
+	info->rtmult = fsinfo.bsize;
+	info->wtmax  = NFS_MAXDATA;
+	info->wtpref = fsinfo.tsize;
+	info->wtmult = fsinfo.bsize;
+	info->dtpref = fsinfo.tsize;
+	info->maxfilesize = 0x7FFFFFFF;
+	info->lease_time = 0;
+out:
+	return status;
+}
+
+static int
+nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
+		  struct nfs_pathconf *info)
+{
+	info->max_link = 0;
+	info->max_namelen = NFS2_MAXNAMLEN;
+	return 0;
+}
+
+static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
+{
+	struct inode *inode = hdr->inode;
+
+	nfs_invalidate_atime(inode);
+	if (task->tk_status >= 0) {
+		nfs_refresh_inode(inode, hdr->res.fattr);
+		/* Emulate the eof flag, which isn't normally needed in NFSv2
+		 * as it is guaranteed to always return the file attributes
+		 */
+		if (hdr->args.offset + hdr->res.count >= hdr->res.fattr->size)
+			hdr->res.eof = 1;
+	}
+	return 0;
+}
+
+static void nfs_proc_read_setup(struct nfs_pgio_header *hdr,
+				struct rpc_message *msg)
+{
+	msg->rpc_proc = &nfs_procedures[NFSPROC_READ];
+}
+
+static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task,
+				     struct nfs_pgio_header *hdr)
+{
+	rpc_call_start(task);
+	return 0;
+}
+
+static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
+{
+	struct inode *inode = hdr->inode;
+
+	if (task->tk_status >= 0)
+		nfs_post_op_update_inode_force_wcc(inode, hdr->res.fattr);
+	return 0;
+}
+
+static void nfs_proc_write_setup(struct nfs_pgio_header *hdr,
+				 struct rpc_message *msg)
+{
+	/* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */
+	hdr->args.stable = NFS_FILE_SYNC;
+	msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE];
+}
+
+static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data)
+{
+	BUG();
+}
+
+static void
+nfs_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg)
+{
+	BUG();
+}
+
+static int
+nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
+{
+	struct inode *inode = file_inode(filp);
+
+	return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl);
+}
+
+/* Helper functions for NFS lock bounds checking */
+#define NFS_LOCK32_OFFSET_MAX ((__s32)0x7fffffffUL)
+static int nfs_lock_check_bounds(const struct file_lock *fl)
+{
+	__s32 start, end;
+
+	start = (__s32)fl->fl_start;
+	if ((loff_t)start != fl->fl_start)
+		goto out_einval;
+
+	if (fl->fl_end != OFFSET_MAX) {
+		end = (__s32)fl->fl_end;
+		if ((loff_t)end != fl->fl_end)
+			goto out_einval;
+	} else
+		end = NFS_LOCK32_OFFSET_MAX;
+
+	if (start < 0 || start > end)
+		goto out_einval;
+	return 0;
+out_einval:
+	return -EINVAL;
+}
+
+static int nfs_have_delegation(struct inode *inode, fmode_t flags)
+{
+	return 0;
+}
+
+static int nfs_return_delegation(struct inode *inode)
+{
+	nfs_wb_all(inode);
+	return 0;
+}
+
+static const struct inode_operations nfs_dir_inode_operations = {
+	.create		= nfs_create,
+	.lookup		= nfs_lookup,
+	.link		= nfs_link,
+	.unlink		= nfs_unlink,
+	.symlink	= nfs_symlink,
+	.mkdir		= nfs_mkdir,
+	.rmdir		= nfs_rmdir,
+	.mknod		= nfs_mknod,
+	.rename		= nfs_rename,
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+};
+
+static const struct inode_operations nfs_file_inode_operations = {
+	.permission	= nfs_permission,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+};
+
+const struct nfs_rpc_ops nfs_v2_clientops = {
+	.version	= 2,		       /* protocol version */
+	.dentry_ops	= &nfs_dentry_operations,
+	.dir_inode_ops	= &nfs_dir_inode_operations,
+	.file_inode_ops	= &nfs_file_inode_operations,
+	.file_ops	= &nfs_file_operations,
+	.getroot	= nfs_proc_get_root,
+	.submount	= nfs_submount,
+	.try_mount	= nfs_try_mount,
+	.getattr	= nfs_proc_getattr,
+	.setattr	= nfs_proc_setattr,
+	.lookup		= nfs_proc_lookup,
+	.access		= NULL,		       /* access */
+	.readlink	= nfs_proc_readlink,
+	.create		= nfs_proc_create,
+	.remove		= nfs_proc_remove,
+	.unlink_setup	= nfs_proc_unlink_setup,
+	.unlink_rpc_prepare = nfs_proc_unlink_rpc_prepare,
+	.unlink_done	= nfs_proc_unlink_done,
+	.rename_setup	= nfs_proc_rename_setup,
+	.rename_rpc_prepare = nfs_proc_rename_rpc_prepare,
+	.rename_done	= nfs_proc_rename_done,
+	.link		= nfs_proc_link,
+	.symlink	= nfs_proc_symlink,
+	.mkdir		= nfs_proc_mkdir,
+	.rmdir		= nfs_proc_rmdir,
+	.readdir	= nfs_proc_readdir,
+	.mknod		= nfs_proc_mknod,
+	.statfs		= nfs_proc_statfs,
+	.fsinfo		= nfs_proc_fsinfo,
+	.pathconf	= nfs_proc_pathconf,
+	.decode_dirent	= nfs2_decode_dirent,
+	.pgio_rpc_prepare = nfs_proc_pgio_rpc_prepare,
+	.read_setup	= nfs_proc_read_setup,
+	.read_done	= nfs_read_done,
+	.write_setup	= nfs_proc_write_setup,
+	.write_done	= nfs_write_done,
+	.commit_setup	= nfs_proc_commit_setup,
+	.commit_rpc_prepare = nfs_proc_commit_rpc_prepare,
+	.lock		= nfs_proc_lock,
+	.lock_check_bounds = nfs_lock_check_bounds,
+	.close_context	= nfs_close_context,
+	.have_delegation = nfs_have_delegation,
+	.return_delegation = nfs_return_delegation,
+	.alloc_client	= nfs_alloc_client,
+	.init_client	= nfs_init_client,
+	.free_client	= nfs_free_client,
+	.create_server	= nfs_create_server,
+	.clone_server	= nfs_clone_server,
+};

fs/nfs/read.c

@@ -0,0 +1,424 @@
+/*
+ * linux/fs/nfs/read.c
+ *
+ * Block I/O for NFS
+ *
+ * Partial copy of Linus' read cache modifications to fs/nfs/file.c
+ * modified for async RPC by okir@monad.swb.de
+ */
+
+#include <linux/time.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fcntl.h>
+#include <linux/stat.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/pagemap.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_page.h>
+#include <linux/module.h>
+
+#include "nfs4_fs.h"
+#include "internal.h"
+#include "iostat.h"
+#include "fscache.h"
+#include "pnfs.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PAGECACHE
+
+static const struct nfs_pgio_completion_ops nfs_async_read_completion_ops;
+static const struct nfs_rw_ops nfs_rw_read_ops;
+
+static struct kmem_cache *nfs_rdata_cachep;
+
+static struct nfs_pgio_header *nfs_readhdr_alloc(void)
+{
+	return kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL);
+}
+
+static void nfs_readhdr_free(struct nfs_pgio_header *rhdr)
+{
+	kmem_cache_free(nfs_rdata_cachep, rhdr);
+}
+
+static
+int nfs_return_empty_page(struct page *page)
+{
+	zero_user(page, 0, PAGE_CACHE_SIZE);
+	SetPageUptodate(page);
+	unlock_page(page);
+	return 0;
+}
+
+void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio,
+			      struct inode *inode, bool force_mds,
+			      const struct nfs_pgio_completion_ops *compl_ops)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops;
+
+#ifdef CONFIG_NFS_V4_1
+	if (server->pnfs_curr_ld && !force_mds)
+		pg_ops = server->pnfs_curr_ld->pg_read_ops;
+#endif
+	nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_read_ops,
+			server->rsize, 0);
+}
+EXPORT_SYMBOL_GPL(nfs_pageio_init_read);
+
+void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio)
+{
+	pgio->pg_ops = &nfs_pgio_rw_ops;
+	pgio->pg_bsize = NFS_SERVER(pgio->pg_inode)->rsize;
+}
+EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds);
+
+int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
+		       struct page *page)
+{
+	struct nfs_page	*new;
+	unsigned int len;
+	struct nfs_pageio_descriptor pgio;
+
+	len = nfs_page_length(page);
+	if (len == 0)
+		return nfs_return_empty_page(page);
+	new = nfs_create_request(ctx, page, NULL, 0, len);
+	if (IS_ERR(new)) {
+		unlock_page(page);
+		return PTR_ERR(new);
+	}
+	if (len < PAGE_CACHE_SIZE)
+		zero_user_segment(page, len, PAGE_CACHE_SIZE);
+
+	nfs_pageio_init_read(&pgio, inode, false,
+			     &nfs_async_read_completion_ops);
+	nfs_pageio_add_request(&pgio, new);
+	nfs_pageio_complete(&pgio);
+	NFS_I(inode)->read_io += pgio.pg_bytes_written;
+	return 0;
+}
+
+static void nfs_readpage_release(struct nfs_page *req)
+{
+	struct inode *d_inode = req->wb_context->dentry->d_inode;
+
+	dprintk("NFS: read done (%s/%llu %d@%lld)\n", d_inode->i_sb->s_id,
+		(unsigned long long)NFS_FILEID(d_inode), req->wb_bytes,
+		(long long)req_offset(req));
+
+	if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) {
+		if (PageUptodate(req->wb_page))
+			nfs_readpage_to_fscache(d_inode, req->wb_page, 0);
+
+		unlock_page(req->wb_page);
+	}
+	nfs_release_request(req);
+}
+
+static void nfs_page_group_set_uptodate(struct nfs_page *req)
+{
+	if (nfs_page_group_sync_on_bit(req, PG_UPTODATE))
+		SetPageUptodate(req->wb_page);
+}
+
+static void nfs_read_completion(struct nfs_pgio_header *hdr)
+{
+	unsigned long bytes = 0;
+
+	if (test_bit(NFS_IOHDR_REDO, &hdr->flags))
+		goto out;
+	while (!list_empty(&hdr->pages)) {
+		struct nfs_page *req = nfs_list_entry(hdr->pages.next);
+		struct page *page = req->wb_page;
+		unsigned long start = req->wb_pgbase;
+		unsigned long end = req->wb_pgbase + req->wb_bytes;
+
+		if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) {
+			/* note: regions of the page not covered by a
+			 * request are zeroed in nfs_readpage_async /
+			 * readpage_async_filler */
+			if (bytes > hdr->good_bytes) {
+				/* nothing in this request was good, so zero
+				 * the full extent of the request */
+				zero_user_segment(page, start, end);
+
+			} else if (hdr->good_bytes - bytes < req->wb_bytes) {
+				/* part of this request has good bytes, but
+				 * not all. zero the bad bytes */
+				start += hdr->good_bytes - bytes;
+				WARN_ON(start < req->wb_pgbase);
+				zero_user_segment(page, start, end);
+			}
+		}
+		bytes += req->wb_bytes;
+		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) {
+			if (bytes <= hdr->good_bytes)
+				nfs_page_group_set_uptodate(req);
+		} else
+			nfs_page_group_set_uptodate(req);
+		nfs_list_remove_request(req);
+		nfs_readpage_release(req);
+	}
+out:
+	hdr->release(hdr);
+}
+
+static void nfs_initiate_read(struct nfs_pgio_header *hdr,
+			      struct rpc_message *msg,
+			      struct rpc_task_setup *task_setup_data, int how)
+{
+	struct inode *inode = hdr->inode;
+	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
+
+	task_setup_data->flags |= swap_flags;
+	NFS_PROTO(inode)->read_setup(hdr, msg);
+}
+
+static void
+nfs_async_read_error(struct list_head *head)
+{
+	struct nfs_page	*req;
+
+	while (!list_empty(head)) {
+		req = nfs_list_entry(head->next);
+		nfs_list_remove_request(req);
+		nfs_readpage_release(req);
+	}
+}
+
+static const struct nfs_pgio_completion_ops nfs_async_read_completion_ops = {
+	.error_cleanup = nfs_async_read_error,
+	.completion = nfs_read_completion,
+};
+
+/*
+ * This is the callback from RPC telling us whether a reply was
+ * received or some error occurred (timeout or socket shutdown).
+ */
+static int nfs_readpage_done(struct rpc_task *task,
+			     struct nfs_pgio_header *hdr,
+			     struct inode *inode)
+{
+	int status = NFS_PROTO(inode)->read_done(task, hdr);
+	if (status != 0)
+		return status;
+
+	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count);
+
+	if (task->tk_status == -ESTALE) {
+		set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+		nfs_mark_for_revalidate(inode);
+	}
+	return 0;
+}
+
+static void nfs_readpage_retry(struct rpc_task *task,
+			       struct nfs_pgio_header *hdr)
+{
+	struct nfs_pgio_args *argp = &hdr->args;
+	struct nfs_pgio_res  *resp = &hdr->res;
+
+	/* This is a short read! */
+	nfs_inc_stats(hdr->inode, NFSIOS_SHORTREAD);
+	/* Has the server at least made some progress? */
+	if (resp->count == 0) {
+		nfs_set_pgio_error(hdr, -EIO, argp->offset);
+		return;
+	}
+	/* Yes, so retry the read at the end of the hdr */
+	hdr->mds_offset += resp->count;
+	argp->offset += resp->count;
+	argp->pgbase += resp->count;
+	argp->count -= resp->count;
+	rpc_restart_call_prepare(task);
+}
+
+static void nfs_readpage_result(struct rpc_task *task,
+				struct nfs_pgio_header *hdr)
+{
+	if (hdr->res.eof) {
+		loff_t bound;
+
+		bound = hdr->args.offset + hdr->res.count;
+		spin_lock(&hdr->lock);
+		if (bound < hdr->io_start + hdr->good_bytes) {
+			set_bit(NFS_IOHDR_EOF, &hdr->flags);
+			clear_bit(NFS_IOHDR_ERROR, &hdr->flags);
+			hdr->good_bytes = bound - hdr->io_start;
+		}
+		spin_unlock(&hdr->lock);
+	} else if (hdr->res.count != hdr->args.count)
+		nfs_readpage_retry(task, hdr);
+}
+
+/*
+ * Read a page over NFS.
+ * We read the page synchronously in the following case:
+ *  -	The error flag is set for this page. This happens only when a
+ *	previous async read operation failed.
+ */
+int nfs_readpage(struct file *file, struct page *page)
+{
+	struct nfs_open_context *ctx;
+	struct inode *inode = page_file_mapping(page)->host;
+	int		error;
+
+	dprintk("NFS: nfs_readpage (%p %ld@%lu)\n",
+		page, PAGE_CACHE_SIZE, page_file_index(page));
+	nfs_inc_stats(inode, NFSIOS_VFSREADPAGE);
+	nfs_add_stats(inode, NFSIOS_READPAGES, 1);
+
+	/*
+	 * Try to flush any pending writes to the file..
+	 *
+	 * NOTE! Because we own the page lock, there cannot
+	 * be any new pending writes generated at this point
+	 * for this page (other pages can be written to).
+	 */
+	error = nfs_wb_page(inode, page);
+	if (error)
+		goto out_unlock;
+	if (PageUptodate(page))
+		goto out_unlock;
+
+	error = -ESTALE;
+	if (NFS_STALE(inode))
+		goto out_unlock;
+
+	if (file == NULL) {
+		error = -EBADF;
+		ctx = nfs_find_open_context(inode, NULL, FMODE_READ);
+		if (ctx == NULL)
+			goto out_unlock;
+	} else
+		ctx = get_nfs_open_context(nfs_file_open_context(file));
+
+	if (!IS_SYNC(inode)) {
+		error = nfs_readpage_from_fscache(ctx, inode, page);
+		if (error == 0)
+			goto out;
+	}
+
+	error = nfs_readpage_async(ctx, inode, page);
+
+out:
+	put_nfs_open_context(ctx);
+	return error;
+out_unlock:
+	unlock_page(page);
+	return error;
+}
+
+struct nfs_readdesc {
+	struct nfs_pageio_descriptor *pgio;
+	struct nfs_open_context *ctx;
+};
+
+static int
+readpage_async_filler(void *data, struct page *page)
+{
+	struct nfs_readdesc *desc = (struct nfs_readdesc *)data;
+	struct nfs_page *new;
+	unsigned int len;
+	int error;
+
+	len = nfs_page_length(page);
+	if (len == 0)
+		return nfs_return_empty_page(page);
+
+	new = nfs_create_request(desc->ctx, page, NULL, 0, len);
+	if (IS_ERR(new))
+		goto out_error;
+
+	if (len < PAGE_CACHE_SIZE)
+		zero_user_segment(page, len, PAGE_CACHE_SIZE);
+	if (!nfs_pageio_add_request(desc->pgio, new)) {
+		error = desc->pgio->pg_error;
+		goto out_unlock;
+	}
+	return 0;
+out_error:
+	error = PTR_ERR(new);
+out_unlock:
+	unlock_page(page);
+	return error;
+}
+
+int nfs_readpages(struct file *filp, struct address_space *mapping,
+		struct list_head *pages, unsigned nr_pages)
+{
+	struct nfs_pageio_descriptor pgio;
+	struct nfs_readdesc desc = {
+		.pgio = &pgio,
+	};
+	struct inode *inode = mapping->host;
+	unsigned long npages;
+	int ret = -ESTALE;
+
+	dprintk("NFS: nfs_readpages (%s/%Lu %d)\n",
+			inode->i_sb->s_id,
+			(unsigned long long)NFS_FILEID(inode),
+			nr_pages);
+	nfs_inc_stats(inode, NFSIOS_VFSREADPAGES);
+
+	if (NFS_STALE(inode))
+		goto out;
+
+	if (filp == NULL) {
+		desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ);
+		if (desc.ctx == NULL)
+			return -EBADF;
+	} else
+		desc.ctx = get_nfs_open_context(nfs_file_open_context(filp));
+
+	/* attempt to read as many of the pages as possible from the cache
+	 * - this returns -ENOBUFS immediately if the cookie is negative
+	 */
+	ret = nfs_readpages_from_fscache(desc.ctx, inode, mapping,
+					 pages, &nr_pages);
+	if (ret == 0)
+		goto read_complete; /* all pages were read */
+
+	nfs_pageio_init_read(&pgio, inode, false,
+			     &nfs_async_read_completion_ops);
+
+	ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);
+
+	nfs_pageio_complete(&pgio);
+	NFS_I(inode)->read_io += pgio.pg_bytes_written;
+	npages = (pgio.pg_bytes_written + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	nfs_add_stats(inode, NFSIOS_READPAGES, npages);
+read_complete:
+	put_nfs_open_context(desc.ctx);
+out:
+	return ret;
+}
+
+int __init nfs_init_readpagecache(void)
+{
+	nfs_rdata_cachep = kmem_cache_create("nfs_read_data",
+					     sizeof(struct nfs_pgio_header),
+					     0, SLAB_HWCACHE_ALIGN,
+					     NULL);
+	if (nfs_rdata_cachep == NULL)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void nfs_destroy_readpagecache(void)
+{
+	kmem_cache_destroy(nfs_rdata_cachep);
+}
+
+static const struct nfs_rw_ops nfs_rw_read_ops = {
+	.rw_mode		= FMODE_READ,
+	.rw_alloc_header	= nfs_readhdr_alloc,
+	.rw_free_header		= nfs_readhdr_free,
+	.rw_done		= nfs_readpage_done,
+	.rw_result		= nfs_readpage_result,
+	.rw_initiate		= nfs_initiate_read,
+};

fs/nfs/symlink.c

@@ -0,0 +1,78 @@
+/*
+ *  linux/fs/nfs/symlink.c
+ *
+ *  Copyright (C) 1992  Rick Sladkey
+ *
+ *  Optimization changes Copyright (C) 1994 Florian La Roche
+ *
+ *  Jun 7 1999, cache symlink lookups in the page cache.  -DaveM
+ *
+ *  nfs symlink handling code
+ */
+
+#include <linux/time.h>
+#include <linux/errno.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/nfs.h>
+#include <linux/nfs2.h>
+#include <linux/nfs_fs.h>
+#include <linux/pagemap.h>
+#include <linux/stat.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/namei.h>
+
+/* Symlink caching in the page cache is even more simplistic
+ * and straight-forward than readdir caching.
+ */
+
+static int nfs_symlink_filler(struct inode *inode, struct page *page)
+{
+	int error;
+
+	error = NFS_PROTO(inode)->readlink(inode, page, 0, PAGE_SIZE);
+	if (error < 0)
+		goto error;
+	SetPageUptodate(page);
+	unlock_page(page);
+	return 0;
+
+error:
+	SetPageError(page);
+	unlock_page(page);
+	return -EIO;
+}
+
+static void *nfs_follow_link(struct dentry *dentry, struct nameidata *nd)
+{
+	struct inode *inode = dentry->d_inode;
+	struct page *page;
+	void *err;
+
+	err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping));
+	if (err)
+		goto read_failed;
+	page = read_cache_page(&inode->i_data, 0,
+				(filler_t *)nfs_symlink_filler, inode);
+	if (IS_ERR(page)) {
+		err = page;
+		goto read_failed;
+	}
+	nd_set_link(nd, kmap(page));
+	return page;
+
+read_failed:
+	nd_set_link(nd, err);
+	return NULL;
+}
+
+/*
+ * symlinks can't do much...
+ */
+const struct inode_operations nfs_symlink_inode_operations = {
+	.readlink	= generic_readlink,
+	.follow_link	= nfs_follow_link,
+	.put_link	= page_put_link,
+	.getattr	= nfs_getattr,
+	.setattr	= nfs_setattr,
+};

fs/nfs/sysctl.c

@@ -0,0 +1,64 @@
+/*
+ * linux/fs/nfs/sysctl.c
+ *
+ * Sysctl interface to NFS parameters
+ */
+#include <linux/types.h>
+#include <linux/linkage.h>
+#include <linux/ctype.h>
+#include <linux/fs.h>
+#include <linux/sysctl.h>
+#include <linux/module.h>
+#include <linux/nfs_fs.h>
+
+static struct ctl_table_header *nfs_callback_sysctl_table;
+
+static struct ctl_table nfs_cb_sysctls[] = {
+	{
+		.procname	= "nfs_mountpoint_timeout",
+		.data		= &nfs_mountpoint_expiry_timeout,
+		.maxlen		= sizeof(nfs_mountpoint_expiry_timeout),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.procname	= "nfs_congestion_kb",
+		.data		= &nfs_congestion_kb,
+		.maxlen		= sizeof(nfs_congestion_kb),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{ }
+};
+
+static struct ctl_table nfs_cb_sysctl_dir[] = {
+	{
+		.procname = "nfs",
+		.mode = 0555,
+		.child = nfs_cb_sysctls,
+	},
+	{ }
+};
+
+static struct ctl_table nfs_cb_sysctl_root[] = {
+	{
+		.procname = "fs",
+		.mode = 0555,
+		.child = nfs_cb_sysctl_dir,
+	},
+	{ }
+};
+
+int nfs_register_sysctl(void)
+{
+	nfs_callback_sysctl_table = register_sysctl_table(nfs_cb_sysctl_root);
+	if (nfs_callback_sysctl_table == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+void nfs_unregister_sysctl(void)
+{
+	unregister_sysctl_table(nfs_callback_sysctl_table);
+	nfs_callback_sysctl_table = NULL;
+}

fs/nfs/unlink.c

@@ -0,0 +1,604 @@
+/*
+ *  linux/fs/nfs/unlink.c
+ *
+ * nfs sillydelete handling
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/dcache.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/nfs_fs.h>
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/namei.h>
+#include <linux/fsnotify.h>
+
+#include "internal.h"
+#include "nfs4_fs.h"
+#include "iostat.h"
+#include "delegation.h"
+
+#include "nfstrace.h"
+
+/**
+ * nfs_free_unlinkdata - release data from a sillydelete operation.
+ * @data: pointer to unlink structure.
+ */
+static void
+nfs_free_unlinkdata(struct nfs_unlinkdata *data)
+{
+	iput(data->dir);
+	put_rpccred(data->cred);
+	kfree(data->args.name.name);
+	kfree(data);
+}
+
+#define NAME_ALLOC_LEN(len)	((len+16) & ~15)
+/**
+ * nfs_copy_dname - copy dentry name to data structure
+ * @dentry: pointer to dentry
+ * @data: nfs_unlinkdata
+ */
+static int nfs_copy_dname(struct dentry *dentry, struct nfs_unlinkdata *data)
+{
+	char		*str;
+	int		len = dentry->d_name.len;
+
+	str = kmemdup(dentry->d_name.name, NAME_ALLOC_LEN(len), GFP_KERNEL);
+	if (!str)
+		return -ENOMEM;
+	data->args.name.len = len;
+	data->args.name.name = str;
+	return 0;
+}
+
+static void nfs_free_dname(struct nfs_unlinkdata *data)
+{
+	kfree(data->args.name.name);
+	data->args.name.name = NULL;
+	data->args.name.len = 0;
+}
+
+static void nfs_dec_sillycount(struct inode *dir)
+{
+	struct nfs_inode *nfsi = NFS_I(dir);
+	if (atomic_dec_return(&nfsi->silly_count) == 1)
+		wake_up(&nfsi->waitqueue);
+}
+
+/**
+ * nfs_async_unlink_done - Sillydelete post-processing
+ * @task: rpc_task of the sillydelete
+ *
+ * Do the directory attribute update.
+ */
+static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs_unlinkdata *data = calldata;
+	struct inode *dir = data->dir;
+
+	trace_nfs_sillyrename_unlink(data, task->tk_status);
+	if (!NFS_PROTO(dir)->unlink_done(task, dir))
+		rpc_restart_call_prepare(task);
+}
+
+/**
+ * nfs_async_unlink_release - Release the sillydelete data.
+ * @task: rpc_task of the sillydelete
+ *
+ * We need to call nfs_put_unlinkdata as a 'tk_release' task since the
+ * rpc_task would be freed too.
+ */
+static void nfs_async_unlink_release(void *calldata)
+{
+	struct nfs_unlinkdata	*data = calldata;
+	struct super_block *sb = data->dir->i_sb;
+
+	nfs_dec_sillycount(data->dir);
+	nfs_free_unlinkdata(data);
+	nfs_sb_deactive(sb);
+}
+
+static void nfs_unlink_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_unlinkdata *data = calldata;
+	NFS_PROTO(data->dir)->unlink_rpc_prepare(task, data);
+}
+
+static const struct rpc_call_ops nfs_unlink_ops = {
+	.rpc_call_done = nfs_async_unlink_done,
+	.rpc_release = nfs_async_unlink_release,
+	.rpc_call_prepare = nfs_unlink_prepare,
+};
+
+static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data)
+{
+	struct rpc_message msg = {
+		.rpc_argp = &data->args,
+		.rpc_resp = &data->res,
+		.rpc_cred = data->cred,
+	};
+	struct rpc_task_setup task_setup_data = {
+		.rpc_message = &msg,
+		.callback_ops = &nfs_unlink_ops,
+		.callback_data = data,
+		.workqueue = nfsiod_workqueue,
+		.flags = RPC_TASK_ASYNC,
+	};
+	struct rpc_task *task;
+	struct dentry *alias;
+
+	alias = d_lookup(parent, &data->args.name);
+	if (alias != NULL) {
+		int ret;
+		void *devname_garbage = NULL;
+
+		/*
+		 * Hey, we raced with lookup... See if we need to transfer
+		 * the sillyrename information to the aliased dentry.
+		 */
+		nfs_free_dname(data);
+		ret = nfs_copy_dname(alias, data);
+		spin_lock(&alias->d_lock);
+		if (ret == 0 && alias->d_inode != NULL &&
+		    !(alias->d_flags & DCACHE_NFSFS_RENAMED)) {
+			devname_garbage = alias->d_fsdata;
+			alias->d_fsdata = data;
+			alias->d_flags |= DCACHE_NFSFS_RENAMED;
+			ret = 1;
+		} else
+			ret = 0;
+		spin_unlock(&alias->d_lock);
+		nfs_dec_sillycount(dir);
+		dput(alias);
+		/*
+		 * If we'd displaced old cached devname, free it.  At that
+		 * point dentry is definitely not a root, so we won't need
+		 * that anymore.
+		 */
+		kfree(devname_garbage);
+		return ret;
+	}
+	data->dir = igrab(dir);
+	if (!data->dir) {
+		nfs_dec_sillycount(dir);
+		return 0;
+	}
+	nfs_sb_active(dir->i_sb);
+	data->args.fh = NFS_FH(dir);
+	nfs_fattr_init(data->res.dir_attr);
+
+	NFS_PROTO(dir)->unlink_setup(&msg, dir);
+
+	task_setup_data.rpc_client = NFS_CLIENT(dir);
+	task = rpc_run_task(&task_setup_data);
+	if (!IS_ERR(task))
+		rpc_put_task_async(task);
+	return 1;
+}
+
+static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data)
+{
+	struct dentry *parent;
+	struct inode *dir;
+	int ret = 0;
+
+
+	parent = dget_parent(dentry);
+	if (parent == NULL)
+		goto out_free;
+	dir = parent->d_inode;
+	/* Non-exclusive lock protects against concurrent lookup() calls */
+	spin_lock(&dir->i_lock);
+	if (atomic_inc_not_zero(&NFS_I(dir)->silly_count) == 0) {
+		/* Deferred delete */
+		hlist_add_head(&data->list, &NFS_I(dir)->silly_list);
+		spin_unlock(&dir->i_lock);
+		ret = 1;
+		goto out_dput;
+	}
+	spin_unlock(&dir->i_lock);
+	ret = nfs_do_call_unlink(parent, dir, data);
+out_dput:
+	dput(parent);
+out_free:
+	return ret;
+}
+
+void nfs_wait_on_sillyrename(struct dentry *dentry)
+{
+	struct nfs_inode *nfsi = NFS_I(dentry->d_inode);
+
+	wait_event(nfsi->waitqueue, atomic_read(&nfsi->silly_count) <= 1);
+}
+
+void nfs_block_sillyrename(struct dentry *dentry)
+{
+	struct nfs_inode *nfsi = NFS_I(dentry->d_inode);
+
+	wait_event(nfsi->waitqueue, atomic_cmpxchg(&nfsi->silly_count, 1, 0) == 1);
+}
+
+void nfs_unblock_sillyrename(struct dentry *dentry)
+{
+	struct inode *dir = dentry->d_inode;
+	struct nfs_inode *nfsi = NFS_I(dir);
+	struct nfs_unlinkdata *data;
+
+	atomic_inc(&nfsi->silly_count);
+	spin_lock(&dir->i_lock);
+	while (!hlist_empty(&nfsi->silly_list)) {
+		if (!atomic_inc_not_zero(&nfsi->silly_count))
+			break;
+		data = hlist_entry(nfsi->silly_list.first, struct nfs_unlinkdata, list);
+		hlist_del(&data->list);
+		spin_unlock(&dir->i_lock);
+		if (nfs_do_call_unlink(dentry, dir, data) == 0)
+			nfs_free_unlinkdata(data);
+		spin_lock(&dir->i_lock);
+	}
+	spin_unlock(&dir->i_lock);
+}
+
+/**
+ * nfs_async_unlink - asynchronous unlinking of a file
+ * @dir: parent directory of dentry
+ * @dentry: dentry to unlink
+ */
+static int
+nfs_async_unlink(struct inode *dir, struct dentry *dentry)
+{
+	struct nfs_unlinkdata *data;
+	int status = -ENOMEM;
+	void *devname_garbage = NULL;
+
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (data == NULL)
+		goto out;
+
+	data->cred = rpc_lookup_cred();
+	if (IS_ERR(data->cred)) {
+		status = PTR_ERR(data->cred);
+		goto out_free;
+	}
+	data->res.dir_attr = &data->dir_attr;
+
+	status = -EBUSY;
+	spin_lock(&dentry->d_lock);
+	if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
+		goto out_unlock;
+	dentry->d_flags |= DCACHE_NFSFS_RENAMED;
+	devname_garbage = dentry->d_fsdata;
+	dentry->d_fsdata = data;
+	spin_unlock(&dentry->d_lock);
+	/*
+	 * If we'd displaced old cached devname, free it.  At that
+	 * point dentry is definitely not a root, so we won't need
+	 * that anymore.
+	 */
+	kfree(devname_garbage);
+	return 0;
+out_unlock:
+	spin_unlock(&dentry->d_lock);
+	put_rpccred(data->cred);
+out_free:
+	kfree(data);
+out:
+	return status;
+}
+
+/**
+ * nfs_complete_unlink - Initialize completion of the sillydelete
+ * @dentry: dentry to delete
+ * @inode: inode
+ *
+ * Since we're most likely to be called by dentry_iput(), we
+ * only use the dentry to find the sillydelete. We then copy the name
+ * into the qstr.
+ */
+void
+nfs_complete_unlink(struct dentry *dentry, struct inode *inode)
+{
+	struct nfs_unlinkdata	*data = NULL;
+
+	spin_lock(&dentry->d_lock);
+	if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
+		dentry->d_flags &= ~DCACHE_NFSFS_RENAMED;
+		data = dentry->d_fsdata;
+		dentry->d_fsdata = NULL;
+	}
+	spin_unlock(&dentry->d_lock);
+
+	if (data != NULL && (NFS_STALE(inode) || !nfs_call_unlink(dentry, data)))
+		nfs_free_unlinkdata(data);
+}
+
+/* Cancel a queued async unlink. Called when a sillyrename run fails. */
+static void
+nfs_cancel_async_unlink(struct dentry *dentry)
+{
+	spin_lock(&dentry->d_lock);
+	if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
+		struct nfs_unlinkdata *data = dentry->d_fsdata;
+
+		dentry->d_flags &= ~DCACHE_NFSFS_RENAMED;
+		dentry->d_fsdata = NULL;
+		spin_unlock(&dentry->d_lock);
+		nfs_free_unlinkdata(data);
+		return;
+	}
+	spin_unlock(&dentry->d_lock);
+}
+
+/**
+ * nfs_async_rename_done - Sillyrename post-processing
+ * @task: rpc_task of the sillyrename
+ * @calldata: nfs_renamedata for the sillyrename
+ *
+ * Do the directory attribute updates and the d_move
+ */
+static void nfs_async_rename_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs_renamedata *data = calldata;
+	struct inode *old_dir = data->old_dir;
+	struct inode *new_dir = data->new_dir;
+	struct dentry *old_dentry = data->old_dentry;
+
+	trace_nfs_sillyrename_rename(old_dir, old_dentry,
+			new_dir, data->new_dentry, task->tk_status);
+	if (!NFS_PROTO(old_dir)->rename_done(task, old_dir, new_dir)) {
+		rpc_restart_call_prepare(task);
+		return;
+	}
+
+	if (data->complete)
+		data->complete(task, data);
+}
+
+/**
+ * nfs_async_rename_release - Release the sillyrename data.
+ * @calldata: the struct nfs_renamedata to be released
+ */
+static void nfs_async_rename_release(void *calldata)
+{
+	struct nfs_renamedata	*data = calldata;
+	struct super_block *sb = data->old_dir->i_sb;
+
+	if (data->old_dentry->d_inode)
+		nfs_mark_for_revalidate(data->old_dentry->d_inode);
+
+	dput(data->old_dentry);
+	dput(data->new_dentry);
+	iput(data->old_dir);
+	iput(data->new_dir);
+	nfs_sb_deactive(sb);
+	put_rpccred(data->cred);
+	kfree(data);
+}
+
+static void nfs_rename_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_renamedata *data = calldata;
+	NFS_PROTO(data->old_dir)->rename_rpc_prepare(task, data);
+}
+
+static const struct rpc_call_ops nfs_rename_ops = {
+	.rpc_call_done = nfs_async_rename_done,
+	.rpc_release = nfs_async_rename_release,
+	.rpc_call_prepare = nfs_rename_prepare,
+};
+
+/**
+ * nfs_async_rename - perform an asynchronous rename operation
+ * @old_dir: directory that currently holds the dentry to be renamed
+ * @new_dir: target directory for the rename
+ * @old_dentry: original dentry to be renamed
+ * @new_dentry: dentry to which the old_dentry should be renamed
+ *
+ * It's expected that valid references to the dentries and inodes are held
+ */
+struct rpc_task *
+nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
+		 struct dentry *old_dentry, struct dentry *new_dentry,
+		 void (*complete)(struct rpc_task *, struct nfs_renamedata *))
+{
+	struct nfs_renamedata *data;
+	struct rpc_message msg = { };
+	struct rpc_task_setup task_setup_data = {
+		.rpc_message = &msg,
+		.callback_ops = &nfs_rename_ops,
+		.workqueue = nfsiod_workqueue,
+		.rpc_client = NFS_CLIENT(old_dir),
+		.flags = RPC_TASK_ASYNC,
+	};
+
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (data == NULL)
+		return ERR_PTR(-ENOMEM);
+	task_setup_data.callback_data = data;
+
+	data->cred = rpc_lookup_cred();
+	if (IS_ERR(data->cred)) {
+		struct rpc_task *task = ERR_CAST(data->cred);
+		kfree(data);
+		return task;
+	}
+
+	msg.rpc_argp = &data->args;
+	msg.rpc_resp = &data->res;
+	msg.rpc_cred = data->cred;
+
+	/* set up nfs_renamedata */
+	data->old_dir = old_dir;
+	ihold(old_dir);
+	data->new_dir = new_dir;
+	ihold(new_dir);
+	data->old_dentry = dget(old_dentry);
+	data->new_dentry = dget(new_dentry);
+	nfs_fattr_init(&data->old_fattr);
+	nfs_fattr_init(&data->new_fattr);
+	data->complete = complete;
+
+	/* set up nfs_renameargs */
+	data->args.old_dir = NFS_FH(old_dir);
+	data->args.old_name = &old_dentry->d_name;
+	data->args.new_dir = NFS_FH(new_dir);
+	data->args.new_name = &new_dentry->d_name;
+
+	/* set up nfs_renameres */
+	data->res.old_fattr = &data->old_fattr;
+	data->res.new_fattr = &data->new_fattr;
+
+	nfs_sb_active(old_dir->i_sb);
+
+	NFS_PROTO(data->old_dir)->rename_setup(&msg, old_dir);
+
+	return rpc_run_task(&task_setup_data);
+}
+
+/*
+ * Perform tasks needed when a sillyrename is done such as cancelling the
+ * queued async unlink if it failed.
+ */
+static void
+nfs_complete_sillyrename(struct rpc_task *task, struct nfs_renamedata *data)
+{
+	struct dentry *dentry = data->old_dentry;
+
+	if (task->tk_status != 0) {
+		nfs_cancel_async_unlink(dentry);
+		return;
+	}
+
+	/*
+	 * vfs_unlink and the like do not issue this when a file is
+	 * sillyrenamed, so do it here.
+	 */
+	fsnotify_nameremove(dentry, 0);
+}
+
+#define SILLYNAME_PREFIX ".nfs"
+#define SILLYNAME_PREFIX_LEN ((unsigned)sizeof(SILLYNAME_PREFIX) - 1)
+#define SILLYNAME_FILEID_LEN ((unsigned)sizeof(u64) << 1)
+#define SILLYNAME_COUNTER_LEN ((unsigned)sizeof(unsigned int) << 1)
+#define SILLYNAME_LEN (SILLYNAME_PREFIX_LEN + \
+		SILLYNAME_FILEID_LEN + \
+		SILLYNAME_COUNTER_LEN)
+
+/**
+ * nfs_sillyrename - Perform a silly-rename of a dentry
+ * @dir: inode of directory that contains dentry
+ * @dentry: dentry to be sillyrenamed
+ *
+ * NFSv2/3 is stateless and the server doesn't know when the client is
+ * holding a file open. To prevent application problems when a file is
+ * unlinked while it's still open, the client performs a "silly-rename".
+ * That is, it renames the file to a hidden file in the same directory,
+ * and only performs the unlink once the last reference to it is put.
+ *
+ * The final cleanup is done during dentry_iput.
+ *
+ * (Note: NFSv4 is stateful, and has opens, so in theory an NFSv4 server
+ * could take responsibility for keeping open files referenced.  The server
+ * would also need to ensure that opened-but-deleted files were kept over
+ * reboots.  However, we may not assume a server does so.  (RFC 5661
+ * does provide an OPEN4_RESULT_PRESERVE_UNLINKED flag that a server can
+ * use to advertise that it does this; some day we may take advantage of
+ * it.))
+ */
+int
+nfs_sillyrename(struct inode *dir, struct dentry *dentry)
+{
+	static unsigned int sillycounter;
+	unsigned char silly[SILLYNAME_LEN + 1];
+	unsigned long long fileid;
+	struct dentry *sdentry;
+	struct rpc_task *task;
+	int            error = -EBUSY;
+
+	dfprintk(VFS, "NFS: silly-rename(%pd2, ct=%d)\n",
+		dentry, d_count(dentry));
+	nfs_inc_stats(dir, NFSIOS_SILLYRENAME);
+
+	/*
+	 * We don't allow a dentry to be silly-renamed twice.
+	 */
+	if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
+		goto out;
+
+	fileid = NFS_FILEID(dentry->d_inode);
+
+	/* Return delegation in anticipation of the rename */
+	NFS_PROTO(dentry->d_inode)->return_delegation(dentry->d_inode);
+
+	sdentry = NULL;
+	do {
+		int slen;
+		dput(sdentry);
+		sillycounter++;
+		slen = scnprintf(silly, sizeof(silly),
+				SILLYNAME_PREFIX "%0*llx%0*x",
+				SILLYNAME_FILEID_LEN, fileid,
+				SILLYNAME_COUNTER_LEN, sillycounter);
+
+		dfprintk(VFS, "NFS: trying to rename %pd to %s\n",
+				dentry, silly);
+
+		sdentry = lookup_one_len(silly, dentry->d_parent, slen);
+		/*
+		 * N.B. Better to return EBUSY here ... it could be
+		 * dangerous to delete the file while it's in use.
+		 */
+		if (IS_ERR(sdentry))
+			goto out;
+	} while (sdentry->d_inode != NULL); /* need negative lookup */
+
+	/* queue unlink first. Can't do this from rpc_release as it
+	 * has to allocate memory
+	 */
+	error = nfs_async_unlink(dir, dentry);
+	if (error)
+		goto out_dput;
+
+	/* populate unlinkdata with the right dname */
+	error = nfs_copy_dname(sdentry,
+				(struct nfs_unlinkdata *)dentry->d_fsdata);
+	if (error) {
+		nfs_cancel_async_unlink(dentry);
+		goto out_dput;
+	}
+
+	/* run the rename task, undo unlink if it fails */
+	task = nfs_async_rename(dir, dir, dentry, sdentry,
+					nfs_complete_sillyrename);
+	if (IS_ERR(task)) {
+		error = -EBUSY;
+		nfs_cancel_async_unlink(dentry);
+		goto out_dput;
+	}
+
+	/* wait for the RPC task to complete, unless a SIGKILL intervenes */
+	error = rpc_wait_for_completion_task(task);
+	if (error == 0)
+		error = task->tk_status;
+	switch (error) {
+	case 0:
+		/* The rename succeeded */
+		nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+		d_move(dentry, sdentry);
+		break;
+	case -ERESTARTSYS:
+		/* The result of the rename is unknown. Play it safe by
+		 * forcing a new lookup */
+		d_drop(dentry);
+		d_drop(sdentry);
+	}
+	rpc_put_task(task);
+out_dput:
+	dput(sdentry);
+out:
+	return error;
+}