Merge tag 'nfs-for-3.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

  Stable fixes:
   - fix an NFSv4.1 state renewal regression
   - fix open/lock state recovery error handling
   - fix lock recovery when CREATE_SESSION/SETCLIENTID_CONFIRM fails
   - fix statd when reconnection fails
   - don't wake tasks during connection abort
   - don't start reboot recovery if lease check fails
   - fix duplicate proc entries

  Features:
  - pNFS block driver fixes and clean ups from Christoph
  - More code cleanups from Anna
  - Improve mmap() writeback performance
  - Replace use of PF_TRANS with a more generic mechanism for avoiding
    deadlocks in nfs_release_page"

* tag 'nfs-for-3.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (66 commits)
  NFSv4.1: Fix an NFSv4.1 state renewal regression
  NFSv4: fix open/lock state recovery error handling
  NFSv4: Fix lock recovery when CREATE_SESSION/SETCLIENTID_CONFIRM fails
  NFS: Fabricate fscache server index key correctly
  SUNRPC: Add missing support for RPC_CLNT_CREATE_NO_RETRANS_TIMEOUT
  NFSv3: Fix missing includes of nfs3_fs.h
  NFS/SUNRPC: Remove other deadlock-avoidance mechanisms in nfs_release_page()
  NFS: avoid waiting at all in nfs_release_page when congested.
  NFS: avoid deadlocks with loop-back mounted NFS filesystems.
  MM: export page_wakeup functions
  SCHED: add some "wait..on_bit...timeout()" interfaces.
  NFS: don't use STABLE writes during writeback.
  NFSv4: use exponential retry on NFS4ERR_DELAY for async requests.
  rpc: Add -EPERM processing for xs_udp_send_request()
  rpc: return sent and err from xs_sendpages()
  lockd: Try to reconnect if statd has moved
  SUNRPC: Don't wake tasks during connection abort
  Fixing lease renewal
  nfs: fix duplicate proc entries
  pnfs/blocklayout: Fix a 64-bit division/remainder issue in bl_map_stripe
  ...

fs/lockd/mon.c

@@ -159,6 +159,12 @@ static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res,
 
 	msg.rpc_proc = &clnt->cl_procinfo[proc];
 	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFTCONN);
+	if (status == -ECONNREFUSED) {
+		dprintk("lockd:	NSM upcall RPC failed, status=%d, forcing rebind\n",
+				status);
+		rpc_force_rebind(clnt);
+		status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFTCONN);
+	}
 	if (status < 0)
 		dprintk("lockd: NSM upcall RPC failed, status=%d\n",
 				status);

fs/nfs/blocklayout/Makefile

@@ -2,4 +2,5 @@
 # Makefile for the pNFS block layout driver kernel module
 #
 obj-$(CONFIG_PNFS_BLOCK) += blocklayoutdriver.o
-blocklayoutdriver-objs := blocklayout.o extents.o blocklayoutdev.o blocklayoutdm.o
+
+blocklayoutdriver-y += blocklayout.o dev.o extent_tree.o rpc_pipefs.o

fs/nfs/blocklayout/blocklayout.h

@@ -44,105 +44,112 @@
 #define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)
 #define SECTOR_SIZE (1 << SECTOR_SHIFT)
 
-struct block_mount_id {
-	spinlock_t			bm_lock;    /* protects list */
-	struct list_head		bm_devlist; /* holds pnfs_block_dev */
-};
+struct pnfs_block_dev;
 
-struct pnfs_block_dev {
-	struct list_head		bm_node;
-	struct nfs4_deviceid		bm_mdevid;    /* associated devid */
-	struct block_device		*bm_mdev;     /* meta device itself */
-	struct net			*net;
+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,
 };
 
-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 */
+#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;
+	};
 };
 
-#define MY_MAX_TAGS (15) /* tag bitnums used must be less than this */
+struct pnfs_block_dev_map {
+	sector_t			start;
+	sector_t			len;
 
-struct my_tree {
-	sector_t		mtt_step_size;	/* Internal sector alignment */
-	struct list_head	mtt_stub; /* Should be a radix tree */
+	sector_t			disk_offset;
+	struct block_device		*bdev;
 };
 
-struct pnfs_inval_markings {
-	spinlock_t	im_lock;
-	struct my_tree	im_tree;	/* Sectors that need LAYOUTCOMMIT */
-	sector_t	im_block_size;	/* Server blocksize in sectors */
-	struct list_head im_extents;	/* Short extents for INVAL->RW conversion */
+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);
 };
 
-struct pnfs_inval_tracking {
-	struct list_head it_link;
-	int		 it_sector;
-	int		 it_tags;
+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 {
-	struct kref	be_refcnt;
-	struct list_head be_node;	/* link into lseg list */
-	struct nfs4_deviceid be_devid;  /* FIXME: could use device cache instead */
-	struct block_device *be_mdev;
+	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 */
-	struct pnfs_inval_markings *be_inval; /* tracks INVAL->RW transition */
+#define EXTENT_WRITTEN		1
+#define EXTENT_COMMITTING	2
+	unsigned int	be_tag;
 };
 
-/* Shortened extent used by LAYOUTCOMMIT */
-struct pnfs_block_short_extent {
-	struct list_head bse_node;
-	struct nfs4_deviceid bse_devid;
-	struct block_device *bse_mdev;
-	sector_t	bse_f_offset;	/* the starting offset in the file */
-	sector_t	bse_length;	/* the size of the extent */
-};
-
-static inline void
-BL_INIT_INVAL_MARKS(struct pnfs_inval_markings *marks, sector_t blocksize)
-{
-	spin_lock_init(&marks->im_lock);
-	INIT_LIST_HEAD(&marks->im_tree.mtt_stub);
-	INIT_LIST_HEAD(&marks->im_extents);
-	marks->im_block_size = blocksize;
-	marks->im_tree.mtt_step_size = min((sector_t)PAGE_CACHE_SECTORS,
-					   blocksize);
-}
-
-enum extentclass4 {
-	RW_EXTENT       = 0, /* READWRTE and INVAL */
-	RO_EXTENT       = 1, /* READ and NONE */
-	EXTENT_LISTS    = 2,
-};
-
-static inline int bl_choose_list(enum exstate4 state)
-{
-	if (state == PNFS_BLOCK_READ_DATA || state == PNFS_BLOCK_NONE_DATA)
-		return RO_EXTENT;
-	else
-		return RW_EXTENT;
-}
+/* 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 pnfs_inval_markings bl_inval; /* tracks INVAL->RW transition */
+	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 */
-	struct list_head	bl_extents[EXTENT_LISTS]; /* R and RW extents */
-	struct list_head	bl_commit;	/* Needs layout commit */
-	struct list_head	bl_committing;	/* Layout committing */
-	unsigned int		bl_count;	/* entries in bl_commit */
-	sector_t		bl_blocksize;  /* Server blocksize in sectors */
 };
 
-#define BLK_ID(lo) ((struct block_mount_id *)(NFS_SERVER(lo->plh_inode)->pnfs_ld_data))
-
 static inline struct pnfs_block_layout *
 BLK_LO2EXT(struct pnfs_layout_hdr *lo)
 {
@@ -171,41 +178,27 @@ struct bl_msg_hdr {
 #define BL_DEVICE_REQUEST_PROC         0x1 /* User level process succeeds */
 #define BL_DEVICE_REQUEST_ERR          0x2 /* User level process fails */
 
-/* blocklayoutdev.c */
-ssize_t bl_pipe_downcall(struct file *, const char __user *, size_t);
-void bl_pipe_destroy_msg(struct rpc_pipe_msg *);
-void nfs4_blkdev_put(struct block_device *bdev);
-struct pnfs_block_dev *nfs4_blk_decode_device(struct nfs_server *server,
-						struct pnfs_device *dev);
-int nfs4_blk_process_layoutget(struct pnfs_layout_hdr *lo,
-				struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);
-
-/* blocklayoutdm.c */
-void bl_free_block_dev(struct pnfs_block_dev *bdev);
-
-/* extents.c */
-struct pnfs_block_extent *
-bl_find_get_extent(struct pnfs_block_layout *bl, sector_t isect,
-		struct pnfs_block_extent **cow_read);
-int bl_mark_sectors_init(struct pnfs_inval_markings *marks,
-			     sector_t offset, sector_t length);
-void bl_put_extent(struct pnfs_block_extent *be);
-struct pnfs_block_extent *bl_alloc_extent(void);
-int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect);
-int encode_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
-				   struct xdr_stream *xdr,
-				   const struct nfs4_layoutcommit_args *arg);
-void clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
-				   const struct nfs4_layoutcommit_args *arg,
-				   int status);
-int bl_add_merge_extent(struct pnfs_block_layout *bl,
-			 struct pnfs_block_extent *new);
-int bl_mark_for_commit(struct pnfs_block_extent *be,
-			sector_t offset, sector_t length,
-			struct pnfs_block_short_extent *new);
-int bl_push_one_short_extent(struct pnfs_inval_markings *marks);
-struct pnfs_block_short_extent *
-bl_pop_one_short_extent(struct pnfs_inval_markings *marks);
-void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free);
+/* 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/blocklayoutdev.c

-/*
- *  linux/fs/nfs/blocklayout/blocklayoutdev.c
- *
- *  Device operations for the pnfs nfs4 file 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/buffer_head.h> /* __bread */
-
-#include <linux/genhd.h>
-#include <linux/blkdev.h>
-#include <linux/hash.h>
-
-#include "blocklayout.h"
-
-#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
-
-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;
-}
-
-/*
- * Release the block device
- */
-void nfs4_blkdev_put(struct block_device *bdev)
-{
-	dprintk("%s for device %d:%d\n", __func__, MAJOR(bdev->bd_dev),
-			MINOR(bdev->bd_dev));
-	blkdev_put(bdev, FMODE_READ);
-}
-
-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;
-}
-
-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);
-}
-
-/*
- * Decodes pnfs_block_deviceaddr4 which is XDR encoded in dev->dev_addr_buf.
- */
-struct pnfs_block_dev *
-nfs4_blk_decode_device(struct nfs_server *server,
-		       struct pnfs_device *dev)
-{
-	struct pnfs_block_dev *rv;
-	struct block_device *bd = NULL;
-	struct bl_pipe_msg bl_pipe_msg;
-	struct rpc_pipe_msg *msg = &bl_pipe_msg.msg;
-	struct bl_msg_hdr bl_msg = {
-		.type = BL_DEVICE_MOUNT,
-		.totallen = dev->mincount,
-	};
-	uint8_t *dataptr;
-	DECLARE_WAITQUEUE(wq, current);
-	int offset, len, i, rc;
-	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;
-
-	dprintk("%s CREATING PIPEFS MESSAGE\n", __func__);
-	dprintk("%s: deviceid: %s, mincount: %d\n", __func__, dev->dev_id.data,
-		dev->mincount);
-
-	bl_pipe_msg.bl_wq = &nn->bl_wq;
-	memset(msg, 0, sizeof(*msg));
-	msg->data = kzalloc(sizeof(bl_msg) + dev->mincount, GFP_NOFS);
-	if (!msg->data) {
-		rv = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
-	memcpy(msg->data, &bl_msg, sizeof(bl_msg));
-	dataptr = (uint8_t *) msg->data;
-	len = dev->mincount;
-	offset = sizeof(bl_msg);
-	for (i = 0; len > 0; i++) {
-		memcpy(&dataptr[offset], page_address(dev->pages[i]),
-				len < PAGE_CACHE_SIZE ? len : PAGE_CACHE_SIZE);
-		len -= PAGE_CACHE_SIZE;
-		offset += PAGE_CACHE_SIZE;
-	}
-	msg->len = sizeof(bl_msg) + dev->mincount;
-
-	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);
-		rv = ERR_PTR(rc);
-		goto out;
-	}
-
-	set_current_state(TASK_UNINTERRUPTIBLE);
-	schedule();
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&nn->bl_wq, &wq);
-
-	if (reply->status != BL_DEVICE_REQUEST_PROC) {
-		dprintk("%s failed to open device: %d\n",
-			__func__, reply->status);
-		rv = ERR_PTR(-EINVAL);
-		goto out;
-	}
-
-	bd = blkdev_get_by_dev(MKDEV(reply->major, reply->minor),
-			       FMODE_READ, NULL);
-	if (IS_ERR(bd)) {
-		dprintk("%s failed to open device : %ld\n", __func__,
-			PTR_ERR(bd));
-		rv = ERR_CAST(bd);
-		goto out;
-	}
-
-	rv = kzalloc(sizeof(*rv), GFP_NOFS);
-	if (!rv) {
-		rv = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
-	rv->bm_mdev = bd;
-	memcpy(&rv->bm_mdevid, &dev->dev_id, sizeof(struct nfs4_deviceid));
-	rv->net = net;
-	dprintk("%s Created device %s with bd_block_size %u\n",
-		__func__,
-		bd->bd_disk->disk_name,
-		bd->bd_block_size);
-
-out:
-	kfree(msg->data);
-	return rv;
-}
-
-/* Map deviceid returned by the server to constructed block_device */
-static struct block_device *translate_devid(struct pnfs_layout_hdr *lo,
-					    struct nfs4_deviceid *id)
-{
-	struct block_device *rv = NULL;
-	struct block_mount_id *mid;
-	struct pnfs_block_dev *dev;
-
-	dprintk("%s enter, lo=%p, id=%p\n", __func__, lo, id);
-	mid = BLK_ID(lo);
-	spin_lock(&mid->bm_lock);
-	list_for_each_entry(dev, &mid->bm_devlist, bm_node) {
-		if (memcmp(id->data, dev->bm_mdevid.data,
-			   NFS4_DEVICEID4_SIZE) == 0) {
-			rv = dev->bm_mdev;
-			goto out;
-		}
-	}
- out:
-	spin_unlock(&mid->bm_lock);
-	dprintk("%s returning %p\n", __func__, rv);
-	return rv;
-}
-
-/* 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;
-}
-
-/* XDR decode pnfs_block_layout4 structure */
-int
-nfs4_blk_process_layoutget(struct pnfs_layout_hdr *lo,
-			   struct nfs4_layoutget_res *lgr, gfp_t gfp_flags)
-{
-	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
-	int i, status = -EIO;
-	uint32_t count;
-	struct pnfs_block_extent *be = NULL, *save;
-	struct xdr_stream stream;
-	struct xdr_buf buf;
-	struct page *scratch;
-	__be32 *p;
-	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,
-	};
-	LIST_HEAD(extents);
-
-	dprintk("---> %s\n", __func__);
-
-	scratch = alloc_page(gfp_flags);
-	if (!scratch)
-		return -ENOMEM;
-
-	xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, lgr->layoutp->len);
-	xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);
-
-	p = xdr_inline_decode(&stream, 4);
-	if (unlikely(!p))
-		goto out_err;
-
-	count = be32_to_cpup(p++);
-
-	dprintk("%s enter, number of extents %i\n", __func__, count);
-	p = xdr_inline_decode(&stream, (28 + NFS4_DEVICEID4_SIZE) * count);
-	if (unlikely(!p))
-		goto out_err;
-
-	/* 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++) {
-		be = bl_alloc_extent();
-		if (!be) {
-			status = -ENOMEM;
-			goto out_err;
-		}
-		memcpy(&be->be_devid, p, NFS4_DEVICEID4_SIZE);
-		p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE);
-		be->be_mdev = translate_devid(lo, &be->be_devid);
-		if (!be->be_mdev)
-			goto out_err;
-
-		/* The next three values are read in as bytes,
-		 * but stored as 512-byte sector lengths
-		 */
-		if (decode_sector_number(&p, &be->be_f_offset) < 0)
-			goto out_err;
-		if (decode_sector_number(&p, &be->be_length) < 0)
-			goto out_err;
-		if (decode_sector_number(&p, &be->be_v_offset) < 0)
-			goto out_err;
-		be->be_state = be32_to_cpup(p++);
-		if (be->be_state == PNFS_BLOCK_INVALID_DATA)
-			be->be_inval = &bl->bl_inval;
-		if (verify_extent(be, &lv)) {
-			dprintk("%s verify failed\n", __func__);
-			goto out_err;
-		}
-		list_add_tail(&be->be_node, &extents);
-	}
-	if (lgr->range.offset + lgr->range.length !=
-			lv.start << SECTOR_SHIFT) {
-		dprintk("%s Final length mismatch\n", __func__);
-		be = NULL;
-		goto out_err;
-	}
-	if (lv.start < lv.cowread) {
-		dprintk("%s Final uncovered COW extent\n", __func__);
-		be = NULL;
-		goto out_err;
-	}
-	/* Extents decoded properly, now try to merge them in to
-	 * existing layout extents.
-	 */
-	spin_lock(&bl->bl_ext_lock);
-	list_for_each_entry_safe(be, save, &extents, be_node) {
-		list_del(&be->be_node);
-		status = bl_add_merge_extent(bl, be);
-		if (status) {
-			spin_unlock(&bl->bl_ext_lock);
-			/* This is a fairly catastrophic error, as the
-			 * entire layout extent lists are now corrupted.
-			 * We should have some way to distinguish this.
-			 */
-			be = NULL;
-			goto out_err;
-		}
-	}
-	spin_unlock(&bl->bl_ext_lock);
-	status = 0;
- out:
-	__free_page(scratch);
-	dprintk("%s returns %i\n", __func__, status);
-	return status;
-
- out_err:
-	bl_put_extent(be);
-	while (!list_empty(&extents)) {
-		be = list_first_entry(&extents, struct pnfs_block_extent,
-				      be_node);
-		list_del(&be->be_node);
-		bl_put_extent(be);
-	}
-	goto out;
-}

fs/nfs/blocklayout/blocklayoutdm.c

-/*
- *  linux/fs/nfs/blocklayout/blocklayoutdm.c
- *
- *  Module for the NFSv4.1 pNFS block layout driver.
- *
- *  Copyright (c) 2007 The Regents of the University of Michigan.
- *  All rights reserved.
- *
- *  Fred Isaman <iisaman@umich.edu>
- *  Andy Adamson <andros@citi.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/genhd.h> /* gendisk - used in a dprintk*/
-#include <linux/sched.h>
-#include <linux/hash.h>
-
-#include "blocklayout.h"
-
-#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
-
-static void dev_remove(struct net *net, dev_t dev)
-{
-	struct bl_pipe_msg bl_pipe_msg;
-	struct rpc_pipe_msg *msg = &bl_pipe_msg.msg;
-	struct bl_dev_msg bl_umount_request;
-	struct bl_msg_hdr bl_msg = {
-		.type = BL_DEVICE_UMOUNT,
-		.totallen = sizeof(bl_umount_request),
-	};
-	uint8_t *dataptr;
-	DECLARE_WAITQUEUE(wq, current);
-	struct nfs_net *nn = net_generic(net, nfs_net_id);
-
-	dprintk("Entering %s\n", __func__);
-
-	bl_pipe_msg.bl_wq = &nn->bl_wq;
-	memset(msg, 0, sizeof(*msg));
-	msg->len = sizeof(bl_msg) + bl_msg.totallen;
-	msg->data = kzalloc(msg->len, GFP_NOFS);
-	if (!msg->data)
-		goto out;
-
-	memset(&bl_umount_request, 0, sizeof(bl_umount_request));
-	bl_umount_request.major = MAJOR(dev);
-	bl_umount_request.minor = MINOR(dev);
-
-	memcpy(msg->data, &bl_msg, sizeof(bl_msg));
-	dataptr = (uint8_t *) msg->data;
-	memcpy(&dataptr[sizeof(bl_msg)], &bl_umount_request, sizeof(bl_umount_request));
-
-	add_wait_queue(&nn->bl_wq, &wq);
-	if (rpc_queue_upcall(nn->bl_device_pipe, msg) < 0) {
-		remove_wait_queue(&nn->bl_wq, &wq);
-		goto out;
-	}
-
-	set_current_state(TASK_UNINTERRUPTIBLE);
-	schedule();
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&nn->bl_wq, &wq);
-
-out:
-	kfree(msg->data);
-}
-
-/*
- * Release meta device
- */
-static void nfs4_blk_metadev_release(struct pnfs_block_dev *bdev)
-{
-	dprintk("%s Releasing\n", __func__);
-	nfs4_blkdev_put(bdev->bm_mdev);
-	dev_remove(bdev->net, bdev->bm_mdev->bd_dev);
-}
-
-void bl_free_block_dev(struct pnfs_block_dev *bdev)
-{
-	if (bdev) {
-		if (bdev->bm_mdev) {
-			dprintk("%s Removing DM device: %d:%d\n",
-				__func__,
-				MAJOR(bdev->bm_mdev->bd_dev),
-				MINOR(bdev->bm_mdev->bd_dev));
-			nfs4_blk_metadev_release(bdev);
-		}
-		kfree(bdev);
-	}
-}

fs/nfs/blocklayout/dev.c

+/*
+ * 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

+/*
+ * 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

+/*
+ *  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__);
+
+	bl_pipe_msg.bl_wq = &nn->bl_wq;
+
+	b->simple.len += 4;	/* single volume */
+	if (b->simple.len > PAGE_SIZE)
+		return -EIO;
+
+	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;
+
+	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;
+	}
+
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	schedule();
+	__set_current_state(TASK_RUNNING);
+	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;
+	}
+
+	dev = MKDEV(reply->major, reply->minor);
+out:
+	kfree(msg->data);
+	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;
+
+	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/callback_proc.c

@@ -171,14 +171,26 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 		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))
+					&args->cbl_range)) {
 		rv = NFS4ERR_DELAY;
-	else
-		rv = NFS4ERR_NOMATCHING_LAYOUT;
-	pnfs_set_layout_stateid(lo, &args->cbl_stateid, true);
+		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);
@@ -277,9 +289,6 @@ __be32 nfs4_callback_devicenotify(struct cb_devicenotifyargs *args,
 		}
 
 	found:
-		if (dev->cbd_notify_type == NOTIFY_DEVICEID4_CHANGE)
-			dprintk("%s: NOTIFY_DEVICEID4_CHANGE not supported, "
-				"deleting instead\n", __func__);
 		nfs4_delete_deviceid(server->pnfs_curr_ld, clp, &dev->cbd_dev_id);
 	}
 

fs/nfs/client.c

@@ -1252,6 +1252,7 @@ static int nfs_server_list_open(struct inode *inode, struct file *file)
  * set up the iterator to start reading from the server list and return the first item
  */
 static void *nfs_server_list_start(struct seq_file *m, loff_t *_pos)
+				__acquires(&nn->nfs_client_lock)
 {
 	struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id);
 
@@ -1274,6 +1275,7 @@ static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos)
  * clean up after reading from the transports list
  */
 static void nfs_server_list_stop(struct seq_file *p, void *v)
+				__releases(&nn->nfs_client_lock)
 {
 	struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id);
 
@@ -1318,7 +1320,7 @@ static int nfs_server_list_show(struct seq_file *m, void *v)
  */
 static int nfs_volume_list_open(struct inode *inode, struct file *file)
 {
-	return seq_open_net(inode, file, &nfs_server_list_ops,
+	return seq_open_net(inode, file, &nfs_volume_list_ops,
 			   sizeof(struct seq_net_private));
 }
 
@@ -1326,6 +1328,7 @@ static int nfs_volume_list_open(struct inode *inode, struct file *file)
  * set up the iterator to start reading from the volume list and return the first item
  */
 static void *nfs_volume_list_start(struct seq_file *m, loff_t *_pos)
+				__acquires(&nn->nfs_client_lock)
 {
 	struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id);
 
@@ -1348,6 +1351,7 @@ static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos)
  * clean up after reading from the transports list
  */
 static void nfs_volume_list_stop(struct seq_file *p, void *v)
+				__releases(&nn->nfs_client_lock)
 {
 	struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id);
 

fs/nfs/direct.c

@@ -178,7 +178,6 @@ static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq,
 	return memcmp(verfp, &hdr->verf, sizeof(struct nfs_writeverf));
 }
 
-#if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4)
 /*
  * nfs_direct_cmp_commit_data_verf - compare verifier for commit data
  * @dreq - direct request possibly spanning multiple servers
@@ -197,7 +196,6 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq,
 	WARN_ON_ONCE(verfp->committed < 0);
 	return memcmp(verfp, &data->verf, sizeof(struct nfs_writeverf));
 }
-#endif
 
 /**
  * nfs_direct_IO - NFS address space operation for direct I/O
@@ -576,7 +574,6 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter,
 	return result;
 }
 
-#if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4)
 static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 {
 	struct nfs_pageio_descriptor desc;
@@ -700,17 +697,6 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode
 	schedule_work(&dreq->work); /* Calls nfs_direct_write_schedule_work */
 }
 
-#else
-static void nfs_direct_write_schedule_work(struct work_struct *work)
-{
-}
-
-static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode)
-{
-	nfs_direct_complete(dreq, true);
-}
-#endif
-
 static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 {
 	struct nfs_direct_req *dreq = hdr->dreq;

fs/nfs/file.c

@@ -36,6 +36,7 @@
 #include "internal.h"
 #include "iostat.h"
 #include "fscache.h"
+#include "pnfs.h"
 
 #include "nfstrace.h"
 
@@ -327,6 +328,12 @@ static int nfs_want_read_modify_write(struct file *file, struct page *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? */
@@ -468,17 +475,26 @@ static int nfs_release_page(struct page *page, gfp_t gfp)
 
 	dfprintk(PAGECACHE, "NFS: release_page(%p)\n", page);
 
-	/* Only do I/O if gfp is a superset of GFP_KERNEL, and we're not
-	 * doing this memory reclaim for a fs-related allocation.
+	/* 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 && (gfp & GFP_KERNEL) == GFP_KERNEL &&
-	    !(current->flags & PF_FSTRANS)) {
-		int how = FLUSH_SYNC;
-
-		/* Don't let kswapd deadlock waiting for OOM RPC calls */
-		if (current_is_kswapd())
-			how = 0;
-		nfs_commit_inode(mapping->host, how);
+	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))
@@ -539,13 +555,25 @@ static int nfs_launder_page(struct page *page)
 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;
-	return xs_swapper(NFS_CLIENT(file->f_mapping->host)->cl_xprt, 1);
+
+	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)
 {
-	xs_swapper(NFS_CLIENT(file->f_mapping->host)->cl_xprt, 0);
+	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
 

fs/nfs/filelayout/filelayout.c

@@ -265,7 +265,7 @@ filelayout_set_layoutcommit(struct nfs_pgio_header *hdr)
 {
 
 	if (FILELAYOUT_LSEG(hdr->lseg)->commit_through_mds ||
-	    hdr->res.verf->committed == NFS_FILE_SYNC)
+	    hdr->res.verf->committed != NFS_DATA_SYNC)
 		return;
 
 	pnfs_set_layoutcommit(hdr);
@@ -403,6 +403,9 @@ static int filelayout_commit_done_cb(struct rpc_task *task,
 		return -EAGAIN;
 	}
 
+	if (data->verf.committed == NFS_UNSTABLE)
+		pnfs_commit_set_layoutcommit(data);
+
 	return 0;
 }
 
@@ -646,18 +649,15 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 	}
 
 	/* find and reference the deviceid */
-	d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode)->pnfs_curr_ld,
-				   NFS_SERVER(lo->plh_inode)->nfs_client, id);
-	if (d == NULL) {
-		dsaddr = filelayout_get_device_info(lo->plh_inode, id,
-				lo->plh_lc_cred, gfp_flags);
-		if (dsaddr == NULL)
-			goto out;
-	} else
-		dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node);
+	d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), id,
+			lo->plh_lc_cred, gfp_flags);
+	if (d == NULL)
+		goto out;
+
+	dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node);
 	/* Found deviceid is unavailable */
 	if (filelayout_test_devid_unavailable(&dsaddr->id_node))
-			goto out_put;
+		goto out_put;
 
 	fl->dsaddr = dsaddr;
 
@@ -1368,6 +1368,17 @@ filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 	cinfo->ds->ncommitting = 0;
 	return PNFS_ATTEMPTED;
 }
+static struct nfs4_deviceid_node *
+filelayout_alloc_deviceid_node(struct nfs_server *server,
+		struct pnfs_device *pdev, gfp_t gfp_flags)
+{
+	struct nfs4_file_layout_dsaddr *dsaddr;
+
+	dsaddr = nfs4_fl_alloc_deviceid_node(server, pdev, gfp_flags);
+	if (!dsaddr)
+		return NULL;
+	return &dsaddr->id_node;
+}
 
 static void
 filelayout_free_deveiceid_node(struct nfs4_deviceid_node *d)
@@ -1420,6 +1431,7 @@ static struct pnfs_layoutdriver_type filelayout_type = {
 	.commit_pagelist	= filelayout_commit_pagelist,
 	.read_pagelist		= filelayout_read_pagelist,
 	.write_pagelist		= filelayout_write_pagelist,
+	.alloc_deviceid_node	= filelayout_alloc_deviceid_node,
 	.free_deviceid_node	= filelayout_free_deveiceid_node,
 };
 

fs/nfs/filelayout/filelayout.h

@@ -147,10 +147,11 @@ 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);
-struct nfs4_file_layout_dsaddr *
-filelayout_get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id,
-		struct rpc_cred *cred, gfp_t gfp_flags);
 
 #endif /* FS_NFS_NFS4FILELAYOUT_H */

fs/nfs/fscache-index.c

@@ -74,11 +74,10 @@ static uint16_t nfs_server_get_key(const void *cookie_netfs_data,
 	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;
 
-	memset(key, 0, len);
-
 	switch (clp->cl_addr.ss_family) {
 	case AF_INET:
 		key->port = sin->sin_port;

fs/nfs/inode.c

@@ -505,7 +505,9 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
 		attr->ia_valid &= ~ATTR_MODE;
 
 	if (attr->ia_valid & ATTR_SIZE) {
-		if (!S_ISREG(inode->i_mode) || attr->ia_size == i_size_read(inode))
+		BUG_ON(!S_ISREG(inode->i_mode));
+
+		if (attr->ia_size == i_size_read(inode))
 			attr->ia_valid &= ~ATTR_SIZE;
 	}
 

fs/nfs/internal.h

@@ -218,13 +218,6 @@ static inline void nfs_fs_proc_exit(void)
 int nfs_sockaddr_match_ipaddr(const struct sockaddr *, const struct sockaddr *);
 #endif
 
-/* nfs3client.c */
-#if IS_ENABLED(CONFIG_NFS_V3)
-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
-
 /* callback_xdr.c */
 extern struct svc_version nfs4_callback_version1;
 extern struct svc_version nfs4_callback_version4;

fs/nfs/nfs3_fs.h

+/*
+ * 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

@@ -7,6 +7,7 @@
 #include <linux/nfsacl.h>
 
 #include "internal.h"
+#include "nfs3_fs.h"
 
 #define NFSDBG_FACILITY	NFSDBG_PROC
 

fs/nfs/nfs3client.c

 #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 };

fs/nfs/nfs3proc.c

@@ -22,6 +22,7 @@
 
 #include "iostat.h"
 #include "internal.h"
+#include "nfs3_fs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PROC