Merge branch 'master' into for-2.6.35

Conflicts:
	fs/block_dev.c

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

70 files changed, 753 insertions, 476 deletions

diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index 5c5bc8480070..f8b86e92cd66 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -238,6 +238,13 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 		return ERR_PTR(-ENOMEM);
 	}
 
+	rc = bdi_setup_and_register(&v9ses->bdi, "9p", BDI_CAP_MAP_COPY);
+	if (rc) {
+		__putname(v9ses->aname);
+		__putname(v9ses->uname);
+		return ERR_PTR(rc);
+	}
+
 	spin_lock(&v9fs_sessionlist_lock);
 	list_add(&v9ses->slist, &v9fs_sessionlist);
 	spin_unlock(&v9fs_sessionlist_lock);
@@ -301,6 +308,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 	return fid;
 
 error:
+	bdi_destroy(&v9ses->bdi);
 	return ERR_PTR(retval);
 }
 
@@ -326,6 +334,8 @@ void v9fs_session_close(struct v9fs_session_info *v9ses)
 	__putname(v9ses->uname);
 	__putname(v9ses->aname);
 
+	bdi_destroy(&v9ses->bdi);
+
 	spin_lock(&v9fs_sessionlist_lock);
 	list_del(&v9ses->slist);
 	spin_unlock(&v9fs_sessionlist_lock);
diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h
index a0a8d3dd1361..bec4d0bcb458 100644
--- a/fs/9p/v9fs.h
+++ b/fs/9p/v9fs.h
@@ -20,6 +20,7 @@
  *  Boston, MA  02111-1301  USA
  *
  */
+#include <linux/backing-dev.h>
 
 /**
  * enum p9_session_flags - option flags for each 9P session
@@ -102,6 +103,7 @@ struct v9fs_session_info {
 	u32 uid;		/* if ACCESS_SINGLE, the uid that has access */
 	struct p9_client *clnt;	/* 9p client */
 	struct list_head slist; /* list of sessions registered with v9fs */
+	struct backing_dev_info bdi;
 };
 
 struct p9_fid *v9fs_session_init(struct v9fs_session_info *, const char *,
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index 491108bd6e0d..806da5d3b3a0 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -77,6 +77,7 @@ v9fs_fill_super(struct super_block *sb, struct v9fs_session_info *v9ses,
 	sb->s_blocksize = 1 << sb->s_blocksize_bits;
 	sb->s_magic = V9FS_MAGIC;
 	sb->s_op = &v9fs_super_ops;
+	sb->s_bdi = &v9ses->bdi;
 
 	sb->s_flags = flags | MS_ACTIVE | MS_SYNCHRONOUS | MS_DIRSYNC |
 	    MS_NOATIME;
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index c54dad4e6063..a10f2582844f 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -19,6 +19,7 @@
 #include <linux/workqueue.h>
 #include <linux/sched.h>
 #include <linux/fscache.h>
+#include <linux/backing-dev.h>
 
 #include "afs.h"
 #include "afs_vl.h"
@@ -313,6 +314,7 @@ struct afs_volume {
 	unsigned short		rjservers;	/* number of servers discarded due to -ENOMEDIUM */
 	struct afs_server	*servers[8];	/* servers on which volume resides (ordered) */
 	struct rw_semaphore	server_sem;	/* lock for accessing current server */
+	struct backing_dev_info	bdi;
 };
 
 /*
diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c
index 5e813a816ce4..b3feddc4f7d6 100644
--- a/fs/afs/mntpt.c
+++ b/fs/afs/mntpt.c
@@ -138,9 +138,9 @@ static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt)
 {
 	struct afs_super_info *super;
 	struct vfsmount *mnt;
-	struct page *page = NULL;
+	struct page *page;
 	size_t size;
-	char *buf, *devname = NULL, *options = NULL;
+	char *buf, *devname, *options;
 	int ret;
 
 	_enter("{%s}", mntpt->d_name.name);
@@ -150,22 +150,22 @@ static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt)
 	ret = -EINVAL;
 	size = mntpt->d_inode->i_size;
 	if (size > PAGE_SIZE - 1)
-		goto error;
+		goto error_no_devname;
 
 	ret = -ENOMEM;
 	devname = (char *) get_zeroed_page(GFP_KERNEL);
 	if (!devname)
-		goto error;
+		goto error_no_devname;
 
 	options = (char *) get_zeroed_page(GFP_KERNEL);
 	if (!options)
-		goto error;
+		goto error_no_options;
 
 	/* read the contents of the AFS special symlink */
 	page = read_mapping_page(mntpt->d_inode->i_mapping, 0, NULL);
 	if (IS_ERR(page)) {
 		ret = PTR_ERR(page);
-		goto error;
+		goto error_no_page;
 	}
 
 	ret = -EIO;
@@ -196,12 +196,12 @@ static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt)
 	return mnt;
 
 error:
-	if (page)
-		page_cache_release(page);
-	if (devname)
-		free_page((unsigned long) devname);
-	if (options)
-		free_page((unsigned long) options);
+	page_cache_release(page);
+error_no_page:
+	free_page((unsigned long) options);
+error_no_options:
+	free_page((unsigned long) devname);
+error_no_devname:
 	_leave(" = %d", ret);
 	return ERR_PTR(ret);
 }
diff --git a/fs/afs/super.c b/fs/afs/super.c
index 14f6431598ad..e932e5a3a0c1 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -311,6 +311,7 @@ static int afs_fill_super(struct super_block *sb, void *data)
 	sb->s_magic		= AFS_FS_MAGIC;
 	sb->s_op		= &afs_super_ops;
 	sb->s_fs_info		= as;
+	sb->s_bdi		= &as->volume->bdi;
 
 	/* allocate the root inode and dentry */
 	fid.vid		= as->volume->vid;
diff --git a/fs/afs/volume.c b/fs/afs/volume.c
index a353e69e2391..401eeb21869f 100644
--- a/fs/afs/volume.c
+++ b/fs/afs/volume.c
@@ -106,6 +106,10 @@ struct afs_volume *afs_volume_lookup(struct afs_mount_params *params)
 	volume->cell		= params->cell;
 	volume->vid		= vlocation->vldb.vid[params->type];
 
+	ret = bdi_setup_and_register(&volume->bdi, "afs", BDI_CAP_MAP_COPY);
+	if (ret)
+		goto error_bdi;
+
 	init_rwsem(&volume->server_sem);
 
 	/* look up all the applicable server records */
@@ -151,6 +155,8 @@ error:
 	return ERR_PTR(ret);
 
 error_discard:
+	bdi_destroy(&volume->bdi);
+error_bdi:
 	up_write(&params->cell->vl_sem);
 
 	for (loop = volume->nservers - 1; loop >= 0; loop--)
@@ -200,6 +206,7 @@ void afs_put_volume(struct afs_volume *volume)
 	for (loop = volume->nservers - 1; loop >= 0; loop--)
 		afs_put_server(volume->servers[loop]);
 
+	bdi_destroy(&volume->bdi);
 	kfree(volume);
 
 	_leave(" [destroyed]");
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 7ab23e006e4c..2c5f9a0e5d72 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -1005,15 +1005,8 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(
 				}
 			} else if (!mm->start_data) {
 				mm->start_data = seg->addr;
-#ifndef CONFIG_MMU
 				mm->end_data = seg->addr + phdr->p_memsz;
-#endif
 			}
-
-#ifdef CONFIG_MMU
-			if (seg->addr + phdr->p_memsz > mm->end_data)
-				mm->end_data = seg->addr + phdr->p_memsz;
-#endif
 		}
 
 		seg++;
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index e0e769bdca59..49566c1687d8 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -355,7 +355,7 @@ calc_reloc(unsigned long r, struct lib_info *p, int curid, int internalp)
 
 	if (!flat_reloc_valid(r, start_brk - start_data + text_len)) {
 		printk("BINFMT_FLAT: reloc outside program 0x%x (0 - 0x%x/0x%x)",
-		       (int) r,(int)(start_brk-start_code),(int)text_len);
+		       (int) r,(int)(start_brk-start_data+text_len),(int)text_len);
 		goto failed;
 	}
 
diff --git a/fs/block_dev.c b/fs/block_dev.c
index dd769304382e..55dcb7884f4d 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -406,17 +406,23 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin)
  
 int blkdev_fsync(struct file *filp, struct dentry *dentry, int datasync)
 {
-	struct block_device *bdev = I_BDEV(filp->f_mapping->host);
+	struct inode *bd_inode = filp->f_mapping->host;
+	struct block_device *bdev = I_BDEV(bd_inode);
 	int error;
 
-	error = sync_blockdev(bdev);
-	if (error)
-		return error;
-	
-	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL,
-				(BLKDEV_IFL_WAIT));
+	/*
+	 * There is no need to serialise calls to blkdev_issue_flush with
+	 * i_mutex and doing so causes performance issues with concurrent
+	 * O_SYNC writers to a block device.
+	 */
+	mutex_unlock(&bd_inode->i_mutex);
+
+	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL, BLKDEV_IFL_WAIT);
 	if (error == -EOPNOTSUPP)
 		error = 0;
+
+	mutex_lock(&bd_inode->i_mutex);
+
 	return error;
 }
 EXPORT_SYMBOL(blkdev_fsync);
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index e7b8f2c89ccb..feca04197d02 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -44,8 +44,6 @@ static struct extent_io_ops btree_extent_io_ops;
 static void end_workqueue_fn(struct btrfs_work *work);
 static void free_fs_root(struct btrfs_root *root);
 
-static atomic_t btrfs_bdi_num = ATOMIC_INIT(0);
-
 /*
  * end_io_wq structs are used to do processing in task context when an IO is
  * complete.  This is used during reads to verify checksums, and it is used
@@ -1375,19 +1373,11 @@ static int setup_bdi(struct btrfs_fs_info *info, struct backing_dev_info *bdi)
 {
 	int err;
 
-	bdi->name = "btrfs";
 	bdi->capabilities = BDI_CAP_MAP_COPY;
-	err = bdi_init(bdi);
+	err = bdi_setup_and_register(bdi, "btrfs", BDI_CAP_MAP_COPY);
 	if (err)
 		return err;
 
-	err = bdi_register(bdi, NULL, "btrfs-%d",
-				atomic_inc_return(&btrfs_bdi_num));
-	if (err) {
-		bdi_destroy(bdi);
-		return err;
-	}
-
 	bdi->ra_pages	= default_backing_dev_info.ra_pages;
 	bdi->unplug_io_fn	= btrfs_unplug_io_fn;
 	bdi->unplug_io_data	= info;
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index aa3cd7cc3e40..412593703d1e 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -337,16 +337,15 @@ out:
 /*
  * Get ref for the oldest snapc for an inode with dirty data... that is, the
  * only snap context we are allowed to write back.
- *
- * Caller holds i_lock.
  */
-static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
-						      u64 *snap_size)
+static struct ceph_snap_context *get_oldest_context(struct inode *inode,
+						    u64 *snap_size)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_snap_context *snapc = NULL;
 	struct ceph_cap_snap *capsnap = NULL;
 
+	spin_lock(&inode->i_lock);
 	list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 		dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
 		     capsnap->context, capsnap->dirty_pages);
@@ -357,21 +356,11 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
 			break;
 		}
 	}
-	if (!snapc && ci->i_snap_realm) {
-		snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
+	if (!snapc && ci->i_head_snapc) {
+		snapc = ceph_get_snap_context(ci->i_head_snapc);
 		dout(" head snapc %p has %d dirty pages\n",
 		     snapc, ci->i_wrbuffer_ref_head);
 	}
-	return snapc;
-}
-
-static struct ceph_snap_context *get_oldest_context(struct inode *inode,
-						    u64 *snap_size)
-{
-	struct ceph_snap_context *snapc = NULL;
-
-	spin_lock(&inode->i_lock);
-	snapc = __get_oldest_context(inode, snap_size);
 	spin_unlock(&inode->i_lock);
 	return snapc;
 }
@@ -392,7 +381,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	int len = PAGE_CACHE_SIZE;
 	loff_t i_size;
 	int err = 0;
-	struct ceph_snap_context *snapc;
+	struct ceph_snap_context *snapc, *oldest;
 	u64 snap_size = 0;
 	long writeback_stat;
 
@@ -413,13 +402,16 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 		dout("writepage %p page %p not dirty?\n", inode, page);
 		goto out;
 	}
-	if (snapc != get_oldest_context(inode, &snap_size)) {
+	oldest = get_oldest_context(inode, &snap_size);
+	if (snapc->seq > oldest->seq) {
 		dout("writepage %p page %p snapc %p not writeable - noop\n",
 		     inode, page, (void *)page->private);
 		/* we should only noop if called by kswapd */
 		WARN_ON((current->flags & PF_MEMALLOC) == 0);
+		ceph_put_snap_context(oldest);
 		goto out;
 	}
+	ceph_put_snap_context(oldest);
 
 	/* is this a partial page at end of file? */
 	if (snap_size)
@@ -458,7 +450,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	ClearPagePrivate(page);
 	end_page_writeback(page);
 	ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
-	ceph_put_snap_context(snapc);
+	ceph_put_snap_context(snapc);  /* page's reference */
 out:
 	return err;
 }
@@ -558,9 +550,9 @@ static void writepages_finish(struct ceph_osd_request *req,
 			dout("inode %p skipping page %p\n", inode, page);
 			wbc->pages_skipped++;
 		}
+		ceph_put_snap_context((void *)page->private);
 		page->private = 0;
 		ClearPagePrivate(page);
-		ceph_put_snap_context(snapc);
 		dout("unlocking %d %p\n", i, page);
 		end_page_writeback(page);
 
@@ -618,7 +610,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 	int range_whole = 0;
 	int should_loop = 1;
 	pgoff_t max_pages = 0, max_pages_ever = 0;
-	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL;
+	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
 	struct pagevec pvec;
 	int done = 0;
 	int rc = 0;
@@ -770,9 +762,10 @@ get_more_pages:
 			}
 
 			/* only if matching snap context */
-			if (snapc != (void *)page->private) {
-				dout("page snapc %p != oldest %p\n",
-				     (void *)page->private, snapc);
+			pgsnapc = (void *)page->private;
+			if (pgsnapc->seq > snapc->seq) {
+				dout("page snapc %p %lld > oldest %p %lld\n",
+				     pgsnapc, pgsnapc->seq, snapc, snapc->seq);
 				unlock_page(page);
 				if (!locked_pages)
 					continue; /* keep looking for snap */
@@ -914,7 +907,10 @@ static int context_is_writeable_or_written(struct inode *inode,
 					   struct ceph_snap_context *snapc)
 {
 	struct ceph_snap_context *oldest = get_oldest_context(inode, NULL);
-	return !oldest || snapc->seq <= oldest->seq;
+	int ret = !oldest || snapc->seq <= oldest->seq;
+
+	ceph_put_snap_context(oldest);
+	return ret;
 }
 
 /*
@@ -936,8 +932,8 @@ static int ceph_update_writeable_page(struct file *file,
 	int pos_in_page = pos & ~PAGE_CACHE_MASK;
 	int end_in_page = pos_in_page + len;
 	loff_t i_size;
-	struct ceph_snap_context *snapc;
 	int r;
+	struct ceph_snap_context *snapc, *oldest;
 
 retry_locked:
 	/* writepages currently holds page lock, but if we change that later, */
@@ -947,23 +943,24 @@ retry_locked:
 	BUG_ON(!ci->i_snap_realm);
 	down_read(&mdsc->snap_rwsem);
 	BUG_ON(!ci->i_snap_realm->cached_context);
-	if (page->private &&
-	    (void *)page->private != ci->i_snap_realm->cached_context) {
+	snapc = (void *)page->private;
+	if (snapc && snapc != ci->i_head_snapc) {
 		/*
 		 * this page is already dirty in another (older) snap
 		 * context!  is it writeable now?
 		 */
-		snapc = get_oldest_context(inode, NULL);
+		oldest = get_oldest_context(inode, NULL);
 		up_read(&mdsc->snap_rwsem);
 
-		if (snapc != (void *)page->private) {
+		if (snapc->seq > oldest->seq) {
+			ceph_put_snap_context(oldest);
 			dout(" page %p snapc %p not current or oldest\n",
-			     page, (void *)page->private);
+			     page, snapc);
 			/*
 			 * queue for writeback, and wait for snapc to
 			 * be writeable or written
 			 */
-			snapc = ceph_get_snap_context((void *)page->private);
+			snapc = ceph_get_snap_context(snapc);
 			unlock_page(page);
 			ceph_queue_writeback(inode);
 			r = wait_event_interruptible(ci->i_cap_wq,
@@ -973,6 +970,7 @@ retry_locked:
 				return r;
 			return -EAGAIN;
 		}
+		ceph_put_snap_context(oldest);
 
 		/* yay, writeable, do it now (without dropping page lock) */
 		dout(" page %p snapc %p not current, but oldest\n",
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 3710e077a857..aa2239fa9a3b 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1205,6 +1205,12 @@ retry:
 		if (capsnap->dirty_pages || capsnap->writing)
 			continue;
 
+		/*
+		 * if cap writeback already occurred, we should have dropped
+		 * the capsnap in ceph_put_wrbuffer_cap_refs.
+		 */
+		BUG_ON(capsnap->dirty == 0);
+
 		/* pick mds, take s_mutex */
 		mds = __ceph_get_cap_mds(ci, &mseq);
 		if (session && session->s_mds != mds) {
@@ -2118,8 +2124,8 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 		}
 	spin_unlock(&inode->i_lock);
 
-	dout("put_cap_refs %p had %s %s\n", inode, ceph_cap_string(had),
-	     last ? "last" : "");
+	dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
+	     last ? " last" : "", put ? " put" : "");
 
 	if (last && !flushsnaps)
 		ceph_check_caps(ci, 0, NULL);
@@ -2143,7 +2149,8 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 {
 	struct inode *inode = &ci->vfs_inode;
 	int last = 0;
-	int last_snap = 0;
+	int complete_capsnap = 0;
+	int drop_capsnap = 0;
 	int found = 0;
 	struct ceph_cap_snap *capsnap = NULL;
 
@@ -2166,19 +2173,32 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 		list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 			if (capsnap->context == snapc) {
 				found = 1;
-				capsnap->dirty_pages -= nr;
-				last_snap = !capsnap->dirty_pages;
 				break;
 			}
 		}
 		BUG_ON(!found);
+		capsnap->dirty_pages -= nr;
+		if (capsnap->dirty_pages == 0) {
+			complete_capsnap = 1;
+			if (capsnap->dirty == 0)
+				/* cap writeback completed before we created
+				 * the cap_snap; no FLUSHSNAP is needed */
+				drop_capsnap = 1;
+		}
 		dout("put_wrbuffer_cap_refs on %p cap_snap %p "
-		     " snap %lld %d/%d -> %d/%d %s%s\n",
+		     " snap %lld %d/%d -> %d/%d %s%s%s\n",
 		     inode, capsnap, capsnap->context->seq,
 		     ci->i_wrbuffer_ref+nr, capsnap->dirty_pages + nr,
 		     ci->i_wrbuffer_ref, capsnap->dirty_pages,
 		     last ? " (wrbuffer last)" : "",
-		     last_snap ? " (capsnap last)" : "");
+		     complete_capsnap ? " (complete capsnap)" : "",
+		     drop_capsnap ? " (drop capsnap)" : "");
+		if (drop_capsnap) {
+			ceph_put_snap_context(capsnap->context);
+			list_del(&capsnap->ci_item);
+			list_del(&capsnap->flushing_item);
+			ceph_put_cap_snap(capsnap);
+		}
 	}
 
 	spin_unlock(&inode->i_lock);
@@ -2186,10 +2206,12 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 	if (last) {
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
 		iput(inode);
-	} else if (last_snap) {
+	} else if (complete_capsnap) {
 		ceph_flush_snaps(ci);
 		wake_up(&ci->i_cap_wq);
 	}
+	if (drop_capsnap)
+		iput(inode);
 }
 
 /*
@@ -2465,8 +2487,8 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
 				break;
 			}
 			WARN_ON(capsnap->dirty_pages || capsnap->writing);
-			dout(" removing cap_snap %p follows %lld\n",
-			     capsnap, follows);
+			dout(" removing %p cap_snap %p follows %lld\n",
+			     inode, capsnap, follows);
 			ceph_put_snap_context(capsnap->context);
 			list_del(&capsnap->ci_item);
 			list_del(&capsnap->flushing_item);
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 7261dc6c2ead..ea8ee2e526aa 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -171,11 +171,11 @@ more:
 	spin_lock(&inode->i_lock);
 	spin_lock(&dcache_lock);
 
+	last = dentry;
+
 	if (err < 0)
 		goto out_unlock;
 
-	last = dentry;
-
 	p = p->prev;
 	filp->f_pos++;
 
@@ -312,7 +312,7 @@ more:
 		req->r_readdir_offset = fi->next_offset;
 		req->r_args.readdir.frag = cpu_to_le32(frag);
 		req->r_args.readdir.max_entries = cpu_to_le32(max_entries);
-		req->r_num_caps = max_entries;
+		req->r_num_caps = max_entries + 1;
 		err = ceph_mdsc_do_request(mdsc, NULL, req);
 		if (err < 0) {
 			ceph_mdsc_put_request(req);
@@ -489,6 +489,7 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 		struct inode *inode = ceph_get_snapdir(parent);
 		dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
 		     dentry, dentry->d_name.len, dentry->d_name.name, inode);
+		BUG_ON(!d_unhashed(dentry));
 		d_add(dentry, inode);
 		err = 0;
 	}
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index aca82d55cc53..26f883c275e8 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -886,6 +886,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	struct inode *in = NULL;
 	struct ceph_mds_reply_inode *ininfo;
 	struct ceph_vino vino;
+	struct ceph_client *client = ceph_sb_to_client(sb);
 	int i = 0;
 	int err = 0;
 
@@ -949,7 +950,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 			return err;
 	}
 
-	if (rinfo->head->is_dentry && !req->r_aborted) {
+	/*
+	 * ignore null lease/binding on snapdir ENOENT, or else we
+	 * will have trouble splicing in the virtual snapdir later
+	 */
+	if (rinfo->head->is_dentry && !req->r_aborted &&
+	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
+					       client->mount_args->snapdir_name,
+					       req->r_dentry->d_name.len))) {
 		/*
 		 * lookup link rename   : null -> possibly existing inode
 		 * mknod symlink mkdir  : null -> new inode
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index 8f1715ffbe4b..cdaaa131add3 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -30,6 +30,10 @@ static char tag_msg = CEPH_MSGR_TAG_MSG;
 static char tag_ack = CEPH_MSGR_TAG_ACK;
 static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
 
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key socket_class;
+#endif
+
 
 static void queue_con(struct ceph_connection *con);
 static void con_work(struct work_struct *);
@@ -228,6 +232,10 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
 	con->sock = sock;
 	sock->sk->sk_allocation = GFP_NOFS;
 
+#ifdef CONFIG_LOCKDEP
+	lockdep_set_class(&sock->sk->sk_lock, &socket_class);
+#endif
+
 	set_sock_callbacks(sock, con);
 
 	dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
@@ -333,6 +341,7 @@ static void reset_connection(struct ceph_connection *con)
 		con->out_msg = NULL;
 	}
 	con->in_seq = 0;
+	con->in_seq_acked = 0;
 }
 
 /*
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index 21c6623c4b07..2e2c15eed82a 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -314,71 +314,6 @@ bad:
 	return ERR_PTR(err);
 }
 
-
-/*
- * osd map
- */
-void ceph_osdmap_destroy(struct ceph_osdmap *map)
-{
-	dout("osdmap_destroy %p\n", map);
-	if (map->crush)
-		crush_destroy(map->crush);
-	while (!RB_EMPTY_ROOT(&map->pg_temp)) {
-		struct ceph_pg_mapping *pg =
-			rb_entry(rb_first(&map->pg_temp),
-				 struct ceph_pg_mapping, node);
-		rb_erase(&pg->node, &map->pg_temp);
-		kfree(pg);
-	}
-	while (!RB_EMPTY_ROOT(&map->pg_pools)) {
-		struct ceph_pg_pool_info *pi =
-			rb_entry(rb_first(&map->pg_pools),
-				 struct ceph_pg_pool_info, node);
-		rb_erase(&pi->node, &map->pg_pools);
-		kfree(pi);
-	}
-	kfree(map->osd_state);
-	kfree(map->osd_weight);
-	kfree(map->osd_addr);
-	kfree(map);
-}
-
-/*
- * adjust max osd value.  reallocate arrays.
- */
-static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
-{
-	u8 *state;
-	struct ceph_entity_addr *addr;
-	u32 *weight;
-
-	state = kcalloc(max, sizeof(*state), GFP_NOFS);
-	addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
-	weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
-	if (state == NULL || addr == NULL || weight == NULL) {
-		kfree(state);
-		kfree(addr);
-		kfree(weight);
-		return -ENOMEM;
-	}
-
-	/* copy old? */
-	if (map->osd_state) {
-		memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
-		memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
-		memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
-		kfree(map->osd_state);
-		kfree(map->osd_addr);
-		kfree(map->osd_weight);
-	}
-
-	map->osd_state = state;
-	map->osd_weight = weight;
-	map->osd_addr = addr;
-	map->max_osd = max;
-	return 0;
-}
-
 /*
  * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid
  * to a set of osds)
@@ -482,6 +417,13 @@ static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
 	return NULL;
 }
 
+static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
+{
+	rb_erase(&pi->node, root);
+	kfree(pi->name);
+	kfree(pi);
+}
+
 void __decode_pool(void **p, struct ceph_pg_pool_info *pi)
 {
 	ceph_decode_copy(p, &pi->v, sizeof(pi->v));
@@ -490,6 +432,98 @@ void __decode_pool(void **p, struct ceph_pg_pool_info *pi)
 	*p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2;
 }
 
+static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
+{
+	struct ceph_pg_pool_info *pi;
+	u32 num, len, pool;
+
+	ceph_decode_32_safe(p, end, num, bad);
+	dout(" %d pool names\n", num);
+	while (num--) {
+		ceph_decode_32_safe(p, end, pool, bad);
+		ceph_decode_32_safe(p, end, len, bad);
+		dout("  pool %d len %d\n", pool, len);
+		pi = __lookup_pg_pool(&map->pg_pools, pool);
+		if (pi) {
+			kfree(pi->name);
+			pi->name = kmalloc(len + 1, GFP_NOFS);
+			if (pi->name) {
+				memcpy(pi->name, *p, len);
+				pi->name[len] = '\0';
+				dout("  name is %s\n", pi->name);
+			}
+		}
+		*p += len;
+	}
+	return 0;
+
+bad:
+	return -EINVAL;
+}
+
+/*
+ * osd map
+ */
+void ceph_osdmap_destroy(struct ceph_osdmap *map)
+{
+	dout("osdmap_destroy %p\n", map);
+	if (map->crush)
+		crush_destroy(map->crush);
+	while (!RB_EMPTY_ROOT(&map->pg_temp)) {
+		struct ceph_pg_mapping *pg =
+			rb_entry(rb_first(&map->pg_temp),
+				 struct ceph_pg_mapping, node);
+		rb_erase(&pg->node, &map->pg_temp);
+		kfree(pg);
+	}
+	while (!RB_EMPTY_ROOT(&map->pg_pools)) {
+		struct ceph_pg_pool_info *pi =
+			rb_entry(rb_first(&map->pg_pools),
+				 struct ceph_pg_pool_info, node);
+		__remove_pg_pool(&map->pg_pools, pi);
+	}
+	kfree(map->osd_state);
+	kfree(map->osd_weight);
+	kfree(map->osd_addr);
+	kfree(map);
+}
+
+/*
+ * adjust max osd value.  reallocate arrays.
+ */
+static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
+{
+	u8 *state;
+	struct ceph_entity_addr *addr;
+	u32 *weight;
+
+	state = kcalloc(max, sizeof(*state), GFP_NOFS);
+	addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
+	weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
+	if (state == NULL || addr == NULL || weight == NULL) {
+		kfree(state);
+		kfree(addr);
+		kfree(weight);
+		return -ENOMEM;
+	}
+
+	/* copy old? */
+	if (map->osd_state) {
+		memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
+		memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
+		memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
+		kfree(map->osd_state);
+		kfree(map->osd_addr);
+		kfree(map->osd_weight);
+	}
+
+	map->osd_state = state;
+	map->osd_weight = weight;
+	map->osd_addr = addr;
+	map->max_osd = max;
+	return 0;
+}
+
 /*
  * decode a full map.
  */
@@ -526,7 +560,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 	ceph_decode_32_safe(p, end, max, bad);
 	while (max--) {
 		ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad);
-		pi = kmalloc(sizeof(*pi), GFP_NOFS);
+		pi = kzalloc(sizeof(*pi), GFP_NOFS);
 		if (!pi)
 			goto bad;
 		pi->id = ceph_decode_32(p);
@@ -539,6 +573,10 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 		__decode_pool(p, pi);
 		__insert_pg_pool(&map->pg_pools, pi);
 	}
+
+	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+		goto bad;
+
 	ceph_decode_32_safe(p, end, map->pool_max, bad);
 
 	ceph_decode_32_safe(p, end, map->flags, bad);
@@ -712,7 +750,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 		}
 		pi = __lookup_pg_pool(&map->pg_pools, pool);
 		if (!pi) {
-			pi = kmalloc(sizeof(*pi), GFP_NOFS);
+			pi = kzalloc(sizeof(*pi), GFP_NOFS);
 			if (!pi) {
 				err = -ENOMEM;
 				goto bad;
@@ -722,6 +760,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 		}
 		__decode_pool(p, pi);
 	}
+	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+		goto bad;
 
 	/* old_pool */
 	ceph_decode_32_safe(p, end, len, bad);
@@ -730,10 +770,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 
 		ceph_decode_32_safe(p, end, pool, bad);
 		pi = __lookup_pg_pool(&map->pg_pools, pool);
-		if (pi) {
-			rb_erase(&pi->node, &map->pg_pools);
-			kfree(pi);
-		}
+		if (pi)
+			__remove_pg_pool(&map->pg_pools, pi);
 	}
 
 	/* new_up */
diff --git a/fs/ceph/osdmap.h b/fs/ceph/osdmap.h
index 1fb55afb2642..8bc9f1e4f562 100644
--- a/fs/ceph/osdmap.h
+++ b/fs/ceph/osdmap.h
@@ -23,6 +23,7 @@ struct ceph_pg_pool_info {
 	int id;
 	struct ceph_pg_pool v;
 	int pg_num_mask, pgp_num_mask, lpg_num_mask, lpgp_num_mask;
+	char *name;
 };
 
 struct ceph_pg_mapping {
diff --git a/fs/ceph/rados.h b/fs/ceph/rados.h
index 26ac8b89a676..a1fc1d017b58 100644
--- a/fs/ceph/rados.h
+++ b/fs/ceph/rados.h
@@ -11,8 +11,10 @@
 /*
  * osdmap encoding versions
  */
-#define CEPH_OSDMAP_INC_VERSION 4
-#define CEPH_OSDMAP_VERSION     4
+#define CEPH_OSDMAP_INC_VERSION     5
+#define CEPH_OSDMAP_INC_VERSION_EXT 5
+#define CEPH_OSDMAP_VERSION         5
+#define CEPH_OSDMAP_VERSION_EXT     5
 
 /*
  * fs id
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index e6f9bc57d472..2b881262ef67 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -431,8 +431,7 @@ static int dup_array(u64 **dst, __le64 *src, int num)
  * Caller must hold snap_rwsem for read (i.e., the realm topology won't
  * change).
  */
-void ceph_queue_cap_snap(struct ceph_inode_info *ci,
-			 struct ceph_snap_context *snapc)
+void ceph_queue_cap_snap(struct ceph_inode_info *ci)
 {
 	struct inode *inode = &ci->vfs_inode;
 	struct ceph_cap_snap *capsnap;
@@ -451,10 +450,11 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
 		   as no new writes are allowed to start when pending, so any
 		   writes in progress now were started before the previous
 		   cap_snap.  lucky us. */
-		dout("queue_cap_snap %p snapc %p seq %llu used %d"
-		     " already pending\n", inode, snapc, snapc->seq, used);
+		dout("queue_cap_snap %p already pending\n", inode);
 		kfree(capsnap);
 	} else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR)) {
+		struct ceph_snap_context *snapc = ci->i_head_snapc;
+
 		igrab(inode);
 
 		atomic_set(&capsnap->nref, 1);
@@ -463,7 +463,6 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
 		INIT_LIST_HEAD(&capsnap->flushing_item);
 
 		capsnap->follows = snapc->seq - 1;
-		capsnap->context = ceph_get_snap_context(snapc);
 		capsnap->issued = __ceph_caps_issued(ci, NULL);
 		capsnap->dirty = __ceph_caps_dirty(ci);
 
@@ -480,7 +479,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
 		   snapshot. */
 		capsnap->dirty_pages = ci->i_wrbuffer_ref_head;
 		ci->i_wrbuffer_ref_head = 0;
-		ceph_put_snap_context(ci->i_head_snapc);
+		capsnap->context = snapc;
 		ci->i_head_snapc = NULL;
 		list_add_tail(&capsnap->ci_item, &ci->i_cap_snaps);
 
@@ -522,15 +521,17 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 	capsnap->ctime = inode->i_ctime;
 	capsnap->time_warp_seq = ci->i_time_warp_seq;
 	if (capsnap->dirty_pages) {
-		dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu "
+		dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu "
 		     "still has %d dirty pages\n", inode, capsnap,
 		     capsnap->context, capsnap->context->seq,
-		     capsnap->size, capsnap->dirty_pages);
+		     ceph_cap_string(capsnap->dirty), capsnap->size,
+		     capsnap->dirty_pages);
 		return 0;
 	}
-	dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu clean\n",
+	dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu\n",
 	     inode, capsnap, capsnap->context,
-	     capsnap->context->seq, capsnap->size);
+	     capsnap->context->seq, ceph_cap_string(capsnap->dirty),
+	     capsnap->size);
 
 	spin_lock(&mdsc->snap_flush_lock);
 	list_add_tail(&ci->i_snap_flush_item, &mdsc->snap_flush_list);
@@ -602,7 +603,7 @@ more:
 				if (lastinode)
 					iput(lastinode);
 				lastinode = inode;
-				ceph_queue_cap_snap(ci, realm->cached_context);
+				ceph_queue_cap_snap(ci);
 				spin_lock(&realm->inodes_with_caps_lock);
 			}
 			spin_unlock(&realm->inodes_with_caps_lock);
@@ -824,8 +825,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
 			spin_unlock(&realm->inodes_with_caps_lock);
 			spin_unlock(&inode->i_lock);
 
-			ceph_queue_cap_snap(ci,
-					    ci->i_snap_realm->cached_context);
+			ceph_queue_cap_snap(ci);
 
 			iput(inode);
 			continue;
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index ca702c67bc66..e30dfbb056c3 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -715,8 +715,7 @@ extern int ceph_update_snap_trace(struct ceph_mds_client *m,
 extern void ceph_handle_snap(struct ceph_mds_client *mdsc,
 			     struct ceph_mds_session *session,
 			     struct ceph_msg *msg);
-extern void ceph_queue_cap_snap(struct ceph_inode_info *ci,
-				struct ceph_snap_context *snapc);
+extern void ceph_queue_cap_snap(struct ceph_inode_info *ci);
 extern int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 				  struct ceph_cap_snap *capsnap);
 extern void ceph_cleanup_empty_realms(struct ceph_mds_client *mdsc);
diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h
index 4797787c6a44..246a167cb913 100644
--- a/fs/cifs/cifs_fs_sb.h
+++ b/fs/cifs/cifs_fs_sb.h
@@ -18,6 +18,8 @@
 #ifndef _CIFS_FS_SB_H
 #define _CIFS_FS_SB_H
 
+#include <linux/backing-dev.h>
+
 #define CIFS_MOUNT_NO_PERM      1 /* do not do client vfs_perm check */
 #define CIFS_MOUNT_SET_UID      2 /* set current's euid in create etc. */
 #define CIFS_MOUNT_SERVER_INUM  4 /* inode numbers from uniqueid from server  */
@@ -50,5 +52,6 @@ struct cifs_sb_info {
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	char   *mountdata; /* mount options received at mount time */
 #endif
+	struct backing_dev_info bdi;
 };
 #endif				/* _CIFS_FS_SB_H */
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index ded66be6597c..ad235d604a0b 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -103,6 +103,12 @@ cifs_read_super(struct super_block *sb, void *data,
 	if (cifs_sb == NULL)
 		return -ENOMEM;
 
+	rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY);
+	if (rc) {
+		kfree(cifs_sb);
+		return rc;
+	}
+
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	/* copy mount params to sb for use in submounts */
 	/* BB: should we move this after the mount so we
@@ -115,6 +121,7 @@ cifs_read_super(struct super_block *sb, void *data,
 		int len = strlen(data);
 		cifs_sb->mountdata = kzalloc(len + 1, GFP_KERNEL);
 		if (cifs_sb->mountdata == NULL) {
+			bdi_destroy(&cifs_sb->bdi);
 			kfree(sb->s_fs_info);
 			sb->s_fs_info = NULL;
 			return -ENOMEM;
@@ -135,6 +142,7 @@ cifs_read_super(struct super_block *sb, void *data,
 
 	sb->s_magic = CIFS_MAGIC_NUMBER;
 	sb->s_op = &cifs_super_ops;
+	sb->s_bdi = &cifs_sb->bdi;
 /*	if (cifs_sb->tcon->ses->server->maxBuf > MAX_CIFS_HDR_SIZE + 512)
 	    sb->s_blocksize =
 		cifs_sb->tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE; */
@@ -183,6 +191,7 @@ out_mount_failed:
 		}
 #endif
 		unload_nls(cifs_sb->local_nls);
+		bdi_destroy(&cifs_sb->bdi);
 		kfree(cifs_sb);
 	}
 	return rc;
@@ -214,6 +223,7 @@ cifs_put_super(struct super_block *sb)
 #endif
 
 	unload_nls(cifs_sb->local_nls);
+	bdi_destroy(&cifs_sb->bdi);
 	kfree(cifs_sb);
 
 	unlock_kernel();
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index a1695dcadd99..d97f9935a028 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -167,6 +167,10 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent)
 		return -EBUSY;
 	}
 
+	error = bdi_setup_and_register(&vc->bdi, "coda", BDI_CAP_MAP_COPY);
+	if (error)
+		goto bdi_err;
+
 	vc->vc_sb = sb;
 
 	sb->s_fs_info = vc;
@@ -175,6 +179,7 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_blocksize_bits = 12;
 	sb->s_magic = CODA_SUPER_MAGIC;
 	sb->s_op = &coda_super_operations;
+	sb->s_bdi = &vc->bdi;
 
 	/* get root fid from Venus: this needs the root inode */
 	error = venus_rootfid(sb, &fid);
@@ -200,6 +205,8 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent)
         return 0;
 
  error:
+	bdi_destroy(&vc->bdi);
+ bdi_err:
 	if (root)
 		iput(root);
 	if (vc)
@@ -210,6 +217,7 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent)
 
 static void coda_put_super(struct super_block *sb)
 {
+	bdi_destroy(&coda_vcp(sb)->bdi);
 	coda_vcp(sb)->vc_sb = NULL;
 	sb->s_fs_info = NULL;
 
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index efb2b9400391..1cc087635a5e 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -382,8 +382,8 @@ out:
 static void ecryptfs_lower_offset_for_extent(loff_t *offset, loff_t extent_num,
 					     struct ecryptfs_crypt_stat *crypt_stat)
 {
-	(*offset) = (crypt_stat->num_header_bytes_at_front
-		     + (crypt_stat->extent_size * extent_num));
+	(*offset) = ecryptfs_lower_header_size(crypt_stat)
+		    + (crypt_stat->extent_size * extent_num);
 }
 
 /**
@@ -835,13 +835,13 @@ void ecryptfs_set_default_sizes(struct ecryptfs_crypt_stat *crypt_stat)
 	set_extent_mask_and_shift(crypt_stat);
 	crypt_stat->iv_bytes = ECRYPTFS_DEFAULT_IV_BYTES;
 	if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
-		crypt_stat->num_header_bytes_at_front = 0;
+		crypt_stat->metadata_size = ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
 	else {
 		if (PAGE_CACHE_SIZE <= ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE)
-			crypt_stat->num_header_bytes_at_front =
+			crypt_stat->metadata_size =
 				ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
 		else
-			crypt_stat->num_header_bytes_at_front =	PAGE_CACHE_SIZE;
+			crypt_stat->metadata_size = PAGE_CACHE_SIZE;
 	}
 }
 
@@ -1108,9 +1108,9 @@ static void write_ecryptfs_marker(char *page_virt, size_t *written)
 	(*written) = MAGIC_ECRYPTFS_MARKER_SIZE_BYTES;
 }
 
-static void
-write_ecryptfs_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat,
-		     size_t *written)
+void ecryptfs_write_crypt_stat_flags(char *page_virt,
+				     struct ecryptfs_crypt_stat *crypt_stat,
+				     size_t *written)
 {
 	u32 flags = 0;
 	int i;
@@ -1238,8 +1238,7 @@ ecryptfs_write_header_metadata(char *virt,
 
 	header_extent_size = (u32)crypt_stat->extent_size;
 	num_header_extents_at_front =
-		(u16)(crypt_stat->num_header_bytes_at_front
-		      / crypt_stat->extent_size);
+		(u16)(crypt_stat->metadata_size / crypt_stat->extent_size);
 	put_unaligned_be32(header_extent_size, virt);
 	virt += 4;
 	put_unaligned_be16(num_header_extents_at_front, virt);
@@ -1292,7 +1291,8 @@ static int ecryptfs_write_headers_virt(char *page_virt, size_t max,
 	offset = ECRYPTFS_FILE_SIZE_BYTES;
 	write_ecryptfs_marker((page_virt + offset), &written);
 	offset += written;
-	write_ecryptfs_flags((page_virt + offset), crypt_stat, &written);
+	ecryptfs_write_crypt_stat_flags((page_virt + offset), crypt_stat,
+					&written);
 	offset += written;
 	ecryptfs_write_header_metadata((page_virt + offset), crypt_stat,
 				       &written);
@@ -1382,7 +1382,7 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry)
 		rc = -EINVAL;
 		goto out;
 	}
-	virt_len = crypt_stat->num_header_bytes_at_front;
+	virt_len = crypt_stat->metadata_size;
 	order = get_order(virt_len);
 	/* Released in this function */
 	virt = (char *)ecryptfs_get_zeroed_pages(GFP_KERNEL, order);
@@ -1428,16 +1428,15 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat,
 	header_extent_size = get_unaligned_be32(virt);
 	virt += sizeof(__be32);
 	num_header_extents_at_front = get_unaligned_be16(virt);
-	crypt_stat->num_header_bytes_at_front =
-		(((size_t)num_header_extents_at_front
-		  * (size_t)header_extent_size));
+	crypt_stat->metadata_size = (((size_t)num_header_extents_at_front
+				     * (size_t)header_extent_size));
 	(*bytes_read) = (sizeof(__be32) + sizeof(__be16));
 	if ((validate_header_size == ECRYPTFS_VALIDATE_HEADER_SIZE)
-	    && (crypt_stat->num_header_bytes_at_front
+	    && (crypt_stat->metadata_size
 		< ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE)) {
 		rc = -EINVAL;
 		printk(KERN_WARNING "Invalid header size: [%zd]\n",
-		       crypt_stat->num_header_bytes_at_front);
+		       crypt_stat->metadata_size);
 	}
 	return rc;
 }
@@ -1452,8 +1451,7 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat,
  */
 static void set_default_header_data(struct ecryptfs_crypt_stat *crypt_stat)
 {
-	crypt_stat->num_header_bytes_at_front =
-		ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
+	crypt_stat->metadata_size = ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
 }
 
 /**
@@ -1607,6 +1605,7 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry)
 						ecryptfs_dentry,
 						ECRYPTFS_VALIDATE_HEADER_SIZE);
 	if (rc) {
+		memset(page_virt, 0, PAGE_CACHE_SIZE);
 		rc = ecryptfs_read_xattr_region(page_virt, ecryptfs_inode);
 		if (rc) {
 			printk(KERN_DEBUG "Valid eCryptfs headers not found in "
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index 542f625312f3..bfc2e0f78f00 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -35,6 +35,7 @@
 #include <linux/scatterlist.h>
 #include <linux/hash.h>
 #include <linux/nsproxy.h>
+#include <linux/backing-dev.h>
 
 /* Version verification for shared data structures w/ userspace */
 #define ECRYPTFS_VERSION_MAJOR 0x00
@@ -273,7 +274,7 @@ struct ecryptfs_crypt_stat {
 	u32 flags;
 	unsigned int file_version;
 	size_t iv_bytes;
-	size_t num_header_bytes_at_front;
+	size_t metadata_size;
 	size_t extent_size; /* Data extent size; default is 4096 */
 	size_t key_size;
 	size_t extent_shift;
@@ -393,6 +394,7 @@ struct ecryptfs_mount_crypt_stat {
 struct ecryptfs_sb_info {
 	struct super_block *wsi_sb;
 	struct ecryptfs_mount_crypt_stat mount_crypt_stat;
+	struct backing_dev_info bdi;
 };
 
 /* file private data. */
@@ -464,6 +466,14 @@ struct ecryptfs_daemon {
 
 extern struct mutex ecryptfs_daemon_hash_mux;
 
+static inline size_t
+ecryptfs_lower_header_size(struct ecryptfs_crypt_stat *crypt_stat)
+{
+	if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
+		return 0;
+	return crypt_stat->metadata_size;
+}
+
 static inline struct ecryptfs_file_info *
 ecryptfs_file_to_private(struct file *file)
 {
@@ -651,6 +661,9 @@ int ecryptfs_decrypt_page(struct page *page);
 int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry);
 int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry);
 int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry);
+void ecryptfs_write_crypt_stat_flags(char *page_virt,
+				     struct ecryptfs_crypt_stat *crypt_stat,
+				     size_t *written);
 int ecryptfs_read_and_validate_header_region(char *data,
 					     struct inode *ecryptfs_inode);
 int ecryptfs_read_and_validate_xattr_region(char *page_virt,
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index d3362faf3852..e2d4418affac 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -324,6 +324,7 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 	rc = ecryptfs_read_and_validate_header_region(page_virt,
 						      ecryptfs_dentry->d_inode);
 	if (rc) {
+		memset(page_virt, 0, PAGE_CACHE_SIZE);
 		rc = ecryptfs_read_and_validate_xattr_region(page_virt,
 							     ecryptfs_dentry);
 		if (rc) {
@@ -336,7 +337,7 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		ecryptfs_dentry->d_sb)->mount_crypt_stat;
 	if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) {
 		if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
-			file_size = (crypt_stat->num_header_bytes_at_front
+			file_size = (crypt_stat->metadata_size
 				     + i_size_read(lower_dentry->d_inode));
 		else
 			file_size = i_size_read(lower_dentry->d_inode);
@@ -388,9 +389,9 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 	mutex_unlock(&lower_dir_dentry->d_inode->i_mutex);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
-		printk(KERN_ERR "%s: lookup_one_len() returned [%d] on "
-		       "lower_dentry = [%s]\n", __func__, rc,
-		       ecryptfs_dentry->d_name.name);
+		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
+				"[%d] on lower_dentry = [%s]\n", __func__, rc,
+				encrypted_and_encoded_name);
 		goto out_d_drop;
 	}
 	if (lower_dentry->d_inode)
@@ -417,9 +418,9 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 	mutex_unlock(&lower_dir_dentry->d_inode->i_mutex);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
-		printk(KERN_ERR "%s: lookup_one_len() returned [%d] on "
-		       "lower_dentry = [%s]\n", __func__, rc,
-		       encrypted_and_encoded_name);
+		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
+				"[%d] on lower_dentry = [%s]\n", __func__, rc,
+				encrypted_and_encoded_name);
 		goto out_d_drop;
 	}
 lookup_and_interpose:
@@ -456,8 +457,8 @@ static int ecryptfs_link(struct dentry *old_dentry, struct inode *dir,
 	rc = ecryptfs_interpose(lower_new_dentry, new_dentry, dir->i_sb, 0);
 	if (rc)
 		goto out_lock;
-	fsstack_copy_attr_times(dir, lower_new_dentry->d_inode);
-	fsstack_copy_inode_size(dir, lower_new_dentry->d_inode);
+	fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode);
+	fsstack_copy_inode_size(dir, lower_dir_dentry->d_inode);
 	old_dentry->d_inode->i_nlink =
 		ecryptfs_inode_to_lower(old_dentry->d_inode)->i_nlink;
 	i_size_write(new_dentry->d_inode, file_size_save);
@@ -648,38 +649,17 @@ out_lock:
 	return rc;
 }
 
-static int
-ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
+static int ecryptfs_readlink_lower(struct dentry *dentry, char **buf,
+				   size_t *bufsiz)
 {
+	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	char *lower_buf;
-	size_t lower_bufsiz;
-	struct dentry *lower_dentry;
-	struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
-	char *plaintext_name;
-	size_t plaintext_name_size;
+	size_t lower_bufsiz = PATH_MAX;
 	mm_segment_t old_fs;
 	int rc;
 
-	lower_dentry = ecryptfs_dentry_to_lower(dentry);
-	if (!lower_dentry->d_inode->i_op->readlink) {
-		rc = -EINVAL;
-		goto out;
-	}
-	mount_crypt_stat = &ecryptfs_superblock_to_private(
-						dentry->d_sb)->mount_crypt_stat;
-	/*
-	 * If the lower filename is encrypted, it will result in a significantly
-	 * longer name.  If needed, truncate the name after decode and decrypt.
-	 */
-	if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES)
-		lower_bufsiz = PATH_MAX;
-	else
-		lower_bufsiz = bufsiz;
-	/* Released in this function */
 	lower_buf = kmalloc(lower_bufsiz, GFP_KERNEL);
-	if (lower_buf == NULL) {
-		printk(KERN_ERR "%s: Out of memory whilst attempting to "
-		       "kmalloc [%zd] bytes\n", __func__, lower_bufsiz);
+	if (!lower_buf) {
 		rc = -ENOMEM;
 		goto out;
 	}
@@ -689,29 +669,31 @@ ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
 						   (char __user *)lower_buf,
 						   lower_bufsiz);
 	set_fs(old_fs);
-	if (rc >= 0) {
-		rc = ecryptfs_decode_and_decrypt_filename(&plaintext_name,
-							  &plaintext_name_size,
-							  dentry, lower_buf,
-							  rc);
-		if (rc) {
-			printk(KERN_ERR "%s: Error attempting to decode and "
-			       "decrypt filename; rc = [%d]\n", __func__,
-				rc);
-			goto out_free_lower_buf;
-		}
-		/* Check for bufsiz <= 0 done in sys_readlinkat() */
-		rc = copy_to_user(buf, plaintext_name,
-				  min((size_t) bufsiz, plaintext_name_size));
-		if (rc)
-			rc = -EFAULT;
-		else
-			rc = plaintext_name_size;
-		kfree(plaintext_name);
-		fsstack_copy_attr_atime(dentry->d_inode, lower_dentry->d_inode);
-	}
-out_free_lower_buf:
+	if (rc < 0)
+		goto out;
+	lower_bufsiz = rc;
+	rc = ecryptfs_decode_and_decrypt_filename(buf, bufsiz, dentry,
+						  lower_buf, lower_bufsiz);
+out:
 	kfree(lower_buf);
+	return rc;
+}
+
+static int
+ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
+{
+	char *kbuf;
+	size_t kbufsiz, copied;
+	int rc;
+
+	rc = ecryptfs_readlink_lower(dentry, &kbuf, &kbufsiz);
+	if (rc)
+		goto out;
+	copied = min_t(size_t, bufsiz, kbufsiz);
+	rc = copy_to_user(buf, kbuf, copied) ? -EFAULT : copied;
+	kfree(kbuf);
+	fsstack_copy_attr_atime(dentry->d_inode,
+				ecryptfs_dentry_to_lower(dentry)->d_inode);
 out:
 	return rc;
 }
@@ -769,7 +751,7 @@ upper_size_to_lower_size(struct ecryptfs_crypt_stat *crypt_stat,
 {
 	loff_t lower_size;
 
-	lower_size = crypt_stat->num_header_bytes_at_front;
+	lower_size = ecryptfs_lower_header_size(crypt_stat);
 	if (upper_size != 0) {
 		loff_t num_extents;
 
@@ -1016,6 +998,28 @@ out:
 	return rc;
 }
 
+int ecryptfs_getattr_link(struct vfsmount *mnt, struct dentry *dentry,
+			  struct kstat *stat)
+{
+	struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
+	int rc = 0;
+
+	mount_crypt_stat = &ecryptfs_superblock_to_private(
+						dentry->d_sb)->mount_crypt_stat;
+	generic_fillattr(dentry->d_inode, stat);
+	if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) {
+		char *target;
+		size_t targetsiz;
+
+		rc = ecryptfs_readlink_lower(dentry, &target, &targetsiz);
+		if (!rc) {
+			kfree(target);
+			stat->size = targetsiz;
+		}
+	}
+	return rc;
+}
+
 int ecryptfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
 		     struct kstat *stat)
 {
@@ -1040,7 +1044,7 @@ ecryptfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 
 	lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	if (!lower_dentry->d_inode->i_op->setxattr) {
-		rc = -ENOSYS;
+		rc = -EOPNOTSUPP;
 		goto out;
 	}
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
@@ -1058,7 +1062,7 @@ ecryptfs_getxattr_lower(struct dentry *lower_dentry, const char *name,
 	int rc = 0;
 
 	if (!lower_dentry->d_inode->i_op->getxattr) {
-		rc = -ENOSYS;
+		rc = -EOPNOTSUPP;
 		goto out;
 	}
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
@@ -1085,7 +1089,7 @@ ecryptfs_listxattr(struct dentry *dentry, char *list, size_t size)
 
 	lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	if (!lower_dentry->d_inode->i_op->listxattr) {
-		rc = -ENOSYS;
+		rc = -EOPNOTSUPP;
 		goto out;
 	}
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
@@ -1102,7 +1106,7 @@ static int ecryptfs_removexattr(struct dentry *dentry, const char *name)
 
 	lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	if (!lower_dentry->d_inode->i_op->removexattr) {
-		rc = -ENOSYS;
+		rc = -EOPNOTSUPP;
 		goto out;
 	}
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
@@ -1133,6 +1137,7 @@ const struct inode_operations ecryptfs_symlink_iops = {
 	.put_link = ecryptfs_put_link,
 	.permission = ecryptfs_permission,
 	.setattr = ecryptfs_setattr,
+	.getattr = ecryptfs_getattr_link,
 	.setxattr = ecryptfs_setxattr,
 	.getxattr = ecryptfs_getxattr,
 	.listxattr = ecryptfs_listxattr,
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index af1a8f01ebac..760983d0f25e 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -497,17 +497,25 @@ struct kmem_cache *ecryptfs_sb_info_cache;
 static int
 ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent)
 {
+	struct ecryptfs_sb_info *esi;
 	int rc = 0;
 
 	/* Released in ecryptfs_put_super() */
 	ecryptfs_set_superblock_private(sb,
 					kmem_cache_zalloc(ecryptfs_sb_info_cache,
 							 GFP_KERNEL));
-	if (!ecryptfs_superblock_to_private(sb)) {
+	esi = ecryptfs_superblock_to_private(sb);
+	if (!esi) {
 		ecryptfs_printk(KERN_WARNING, "Out of memory\n");
 		rc = -ENOMEM;
 		goto out;
 	}
+
+	rc = bdi_setup_and_register(&esi->bdi, "ecryptfs", BDI_CAP_MAP_COPY);
+	if (rc)
+		goto out;
+
+	sb->s_bdi = &esi->bdi;
 	sb->s_op = &ecryptfs_sops;
 	/* Released through deactivate_super(sb) from get_sb_nodev */
 	sb->s_root = d_alloc(NULL, &(const struct qstr) {
diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c
index d491237c98e7..2ee9a3a7b68c 100644
--- a/fs/ecryptfs/mmap.c
+++ b/fs/ecryptfs/mmap.c
@@ -83,6 +83,19 @@ out:
 	return rc;
 }
 
+static void strip_xattr_flag(char *page_virt,
+			     struct ecryptfs_crypt_stat *crypt_stat)
+{
+	if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) {
+		size_t written;
+
+		crypt_stat->flags &= ~ECRYPTFS_METADATA_IN_XATTR;
+		ecryptfs_write_crypt_stat_flags(page_virt, crypt_stat,
+						&written);
+		crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR;
+	}
+}
+
 /**
  *   Header Extent:
  *     Octets 0-7:        Unencrypted file size (big-endian)
@@ -98,19 +111,6 @@ out:
  *                        (big-endian)
  *     Octet  26:         Begin RFC 2440 authentication token packet set
  */
-static void set_header_info(char *page_virt,
-			    struct ecryptfs_crypt_stat *crypt_stat)
-{
-	size_t written;
-	size_t save_num_header_bytes_at_front =
-		crypt_stat->num_header_bytes_at_front;
-
-	crypt_stat->num_header_bytes_at_front =
-		ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
-	ecryptfs_write_header_metadata(page_virt + 20, crypt_stat, &written);
-	crypt_stat->num_header_bytes_at_front =
-		save_num_header_bytes_at_front;
-}
 
 /**
  * ecryptfs_copy_up_encrypted_with_header
@@ -136,8 +136,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page,
 					   * num_extents_per_page)
 					  + extent_num_in_page);
 		size_t num_header_extents_at_front =
-			(crypt_stat->num_header_bytes_at_front
-			 / crypt_stat->extent_size);
+			(crypt_stat->metadata_size / crypt_stat->extent_size);
 
 		if (view_extent_num < num_header_extents_at_front) {
 			/* This is a header extent */
@@ -147,9 +146,14 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page,
 			memset(page_virt, 0, PAGE_CACHE_SIZE);
 			/* TODO: Support more than one header extent */
 			if (view_extent_num == 0) {
+				size_t written;
+
 				rc = ecryptfs_read_xattr_region(
 					page_virt, page->mapping->host);
-				set_header_info(page_virt, crypt_stat);
+				strip_xattr_flag(page_virt + 16, crypt_stat);
+				ecryptfs_write_header_metadata(page_virt + 20,
+							       crypt_stat,
+							       &written);
 			}
 			kunmap_atomic(page_virt, KM_USER0);
 			flush_dcache_page(page);
@@ -162,7 +166,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page,
 			/* This is an encrypted data extent */
 			loff_t lower_offset =
 				((view_extent_num * crypt_stat->extent_size)
-				 - crypt_stat->num_header_bytes_at_front);
+				 - crypt_stat->metadata_size);
 
 			rc = ecryptfs_read_lower_page_segment(
 				page, (lower_offset >> PAGE_CACHE_SHIFT),
diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
index fcef41c1d2cf..0c0ae491d231 100644
--- a/fs/ecryptfs/super.c
+++ b/fs/ecryptfs/super.c
@@ -86,7 +86,6 @@ static void ecryptfs_destroy_inode(struct inode *inode)
 		if (lower_dentry->d_inode) {
 			fput(inode_info->lower_file);
 			inode_info->lower_file = NULL;
-			d_drop(lower_dentry);
 		}
 	}
 	ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat);
@@ -123,6 +122,7 @@ static void ecryptfs_put_super(struct super_block *sb)
 	lock_kernel();
 
 	ecryptfs_destroy_mount_crypt_stat(&sb_info->mount_crypt_stat);
+	bdi_destroy(&sb_info->bdi);
 	kmem_cache_free(ecryptfs_sb_info_cache, sb_info);
 	ecryptfs_set_superblock_private(sb, NULL);
 
diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h
index 8442e353309f..54373278a353 100644
--- a/fs/exofs/exofs.h
+++ b/fs/exofs/exofs.h
@@ -35,6 +35,7 @@
 
 #include <linux/fs.h>
 #include <linux/time.h>
+#include <linux/backing-dev.h>
 #include "common.h"
 
 /* FIXME: Remove once pnfs hits mainline
@@ -92,6 +93,7 @@ struct exofs_sb_info {
 	struct exofs_layout	layout;		/* Default files layout,
 						 * contains the variable osd_dev
 						 * array. Keep last */
+	struct backing_dev_info bdi;
 	struct osd_dev	*_min_one_dev[1];	/* Place holder for one dev   */
 };
 
diff --git a/fs/exofs/super.c b/fs/exofs/super.c
index 18e57ea1e5b4..03149b9a5178 100644
--- a/fs/exofs/super.c
+++ b/fs/exofs/super.c
@@ -302,6 +302,7 @@ static void exofs_put_super(struct super_block *sb)
 	_exofs_print_device("Unmounting", NULL, sbi->layout.s_ods[0],
 			    sbi->layout.s_pid);
 
+	bdi_destroy(&sbi->bdi);
 	exofs_free_sbi(sbi);
 	sb->s_fs_info = NULL;
 }
@@ -546,6 +547,10 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 	if (!sbi)
 		return -ENOMEM;
 
+	ret = bdi_setup_and_register(&sbi->bdi, "exofs", BDI_CAP_MAP_COPY);
+	if (ret)
+		goto free_bdi;
+
 	/* use mount options to fill superblock */
 	od = osduld_path_lookup(opts->dev_name);
 	if (IS_ERR(od)) {
@@ -612,6 +617,7 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 	}
 
 	/* set up operation vectors */
+	sb->s_bdi = &sbi->bdi;
 	sb->s_fs_info = sbi;
 	sb->s_op = &exofs_sops;
 	sb->s_export_op = &exofs_export_ops;
@@ -643,6 +649,8 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 	return 0;
 
 free_sbi:
+	bdi_destroy(&sbi->bdi);
+free_bdi:
 	EXOFS_ERR("Unable to mount exofs on %s pid=0x%llx err=%d\n",
 		  opts->dev_name, sbi->layout.s_pid, ret);
 	exofs_free_sbi(sbi);
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 94c8ee81f5e1..236b834b4ca8 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3879,6 +3879,7 @@ static int ext4_xattr_fiemap(struct inode *inode,
 		physical += offset;
 		length = EXT4_SB(inode->i_sb)->s_inode_size - offset;
 		flags |= FIEMAP_EXTENT_DATA_INLINE;
+		brelse(iloc.bh);
 	} else { /* external block */
 		physical = EXT4_I(inode)->i_file_acl << blockbits;
 		length = inode->i_sb->s_blocksize;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 5381802d6052..81d605412844 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -5375,7 +5375,7 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
 	} else {
 		struct ext4_iloc iloc;
 
-		err = ext4_get_inode_loc(inode, &iloc);
+		err = __ext4_get_inode_loc(inode, &iloc, 0);
 		if (err)
 			return err;
 		if (wbc->sync_mode == WB_SYNC_ALL)
@@ -5386,6 +5386,7 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
 				   (unsigned long long)iloc.bh->b_blocknr);
 			err = -EIO;
 		}
+		brelse(iloc.bh);
 	}
 	return err;
 }
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index bde9d0b170c2..b423a364dca3 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2535,6 +2535,17 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 		mb_debug(1, "gonna free %u blocks in group %u (0x%p):",
 			 entry->count, entry->group, entry);
 
+		if (test_opt(sb, DISCARD)) {
+			ext4_fsblk_t discard_block;
+
+			discard_block = entry->start_blk +
+				ext4_group_first_block_no(sb, entry->group);
+			trace_ext4_discard_blocks(sb,
+					(unsigned long long)discard_block,
+					entry->count);
+			sb_issue_discard(sb, discard_block, entry->count);
+		}
+
 		err = ext4_mb_load_buddy(sb, entry->group, &e4b);
 		/* we expect to find existing buddy because it's pinned */
 		BUG_ON(err != 0);
@@ -2556,16 +2567,6 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 			page_cache_release(e4b.bd_bitmap_page);
 		}
 		ext4_unlock_group(sb, entry->group);
-		if (test_opt(sb, DISCARD)) {
-			ext4_fsblk_t discard_block;
-
-			discard_block = entry->start_blk +
-				ext4_group_first_block_no(sb, entry->group);
-			trace_ext4_discard_blocks(sb,
-					(unsigned long long)discard_block,
-					entry->count);
-			sb_issue_discard(sb, discard_block, entry->count);
-		}
 		kmem_cache_free(ext4_free_ext_cachep, entry);
 		ext4_mb_release_desc(&e4b);
 	}
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 6c751106c2e5..7faefb4da939 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -228,14 +228,23 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg)
 
 #ifdef CONFIG_BLOCK
 
-#define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits)
-#define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits);
+static inline sector_t logical_to_blk(struct inode *inode, loff_t offset)
+{
+	return (offset >> inode->i_blkbits);
+}
+
+static inline loff_t blk_to_logical(struct inode *inode, sector_t blk)
+{
+	return (blk << inode->i_blkbits);
+}
 
 /**
  * __generic_block_fiemap - FIEMAP for block based inodes (no locking)
- * @inode - the inode to map
- * @arg - the pointer to userspace where we copy everything to
- * @get_block - the fs's get_block function
+ * @inode: the inode to map
+ * @fieinfo: the fiemap info struct that will be passed back to userspace
+ * @start: where to start mapping in the inode
+ * @len: how much space to map
+ * @get_block: the fs's get_block function
  *
  * This does FIEMAP for block based inodes.  Basically it will just loop
  * through get_block until we hit the number of extents we want to map, or we
@@ -250,58 +259,63 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg)
  */
 
 int __generic_block_fiemap(struct inode *inode,
-			   struct fiemap_extent_info *fieinfo, u64 start,
-			   u64 len, get_block_t *get_block)
+			   struct fiemap_extent_info *fieinfo, loff_t start,
+			   loff_t len, get_block_t *get_block)
 {
-	struct buffer_head tmp;
-	unsigned long long start_blk;
-	long long length = 0, map_len = 0;
+	struct buffer_head map_bh;
+	sector_t start_blk, last_blk;
+	loff_t isize = i_size_read(inode);
 	u64 logical = 0, phys = 0, size = 0;
 	u32 flags = FIEMAP_EXTENT_MERGED;
-	int ret = 0, past_eof = 0, whole_file = 0;
+	bool past_eof = false, whole_file = false;
+	int ret = 0;
 
-	if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC)))
+	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
+	if (ret)
 		return ret;
 
-	start_blk = logical_to_blk(inode, start);
-
-	length = (long long)min_t(u64, len, i_size_read(inode));
-	if (length < len)
-		whole_file = 1;
+	/*
+	 * Either the i_mutex or other appropriate locking needs to be held
+	 * since we expect isize to not change at all through the duration of
+	 * this call.
+	 */
+	if (len >= isize) {
+		whole_file = true;
+		len = isize;
+	}
 
-	map_len = length;
+	start_blk = logical_to_blk(inode, start);
+	last_blk = logical_to_blk(inode, start + len - 1);
 
 	do {
 		/*
 		 * we set b_size to the total size we want so it will map as
 		 * many contiguous blocks as possible at once
 		 */
-		memset(&tmp, 0, sizeof(struct buffer_head));
-		tmp.b_size = map_len;
+		memset(&map_bh, 0, sizeof(struct buffer_head));
+		map_bh.b_size = len;
 
-		ret = get_block(inode, start_blk, &tmp, 0);
+		ret = get_block(inode, start_blk, &map_bh, 0);
 		if (ret)
 			break;
 
 		/* HOLE */
-		if (!buffer_mapped(&tmp)) {
-			length -= blk_to_logical(inode, 1);
+		if (!buffer_mapped(&map_bh)) {
 			start_blk++;
 
 			/*
-			 * we want to handle the case where there is an
+			 * We want to handle the case where there is an
 			 * allocated block at the front of the file, and then
 			 * nothing but holes up to the end of the file properly,
 			 * to make sure that extent at the front gets properly
 			 * marked with FIEMAP_EXTENT_LAST
 			 */
 			if (!past_eof &&
-			    blk_to_logical(inode, start_blk) >=
-			    blk_to_logical(inode, 0)+i_size_read(inode))
+			    blk_to_logical(inode, start_blk) >= isize)
 				past_eof = 1;
 
 			/*
-			 * first hole after going past the EOF, this is our
+			 * First hole after going past the EOF, this is our
 			 * last extent
 			 */
 			if (past_eof && size) {
@@ -309,15 +323,18 @@ int __generic_block_fiemap(struct inode *inode,
 				ret = fiemap_fill_next_extent(fieinfo, logical,
 							      phys, size,
 							      flags);
-				break;
+			} else if (size) {
+				ret = fiemap_fill_next_extent(fieinfo, logical,
+							      phys, size, flags);
+				size = 0;
 			}
 
 			/* if we have holes up to/past EOF then we're done */
-			if (length <= 0 || past_eof)
+			if (start_blk > last_blk || past_eof || ret)
 				break;
 		} else {
 			/*
-			 * we have gone over the length of what we wanted to
+			 * We have gone over the length of what we wanted to
 			 * map, and it wasn't the entire file, so add the extent
 			 * we got last time and exit.
 			 *
@@ -331,7 +348,7 @@ int __generic_block_fiemap(struct inode *inode,
 			 * are good to go, just add the extent to the fieinfo
 			 * and break
 			 */
-			if (length <= 0 && !whole_file) {
+			if (start_blk > last_blk && !whole_file) {
 				ret = fiemap_fill_next_extent(fieinfo, logical,
 							      phys, size,
 							      flags);
@@ -351,11 +368,10 @@ int __generic_block_fiemap(struct inode *inode,
 			}
 
 			logical = blk_to_logical(inode, start_blk);
-			phys = blk_to_logical(inode, tmp.b_blocknr);
-			size = tmp.b_size;
+			phys = blk_to_logical(inode, map_bh.b_blocknr);
+			size = map_bh.b_size;
 			flags = FIEMAP_EXTENT_MERGED;
 
-			length -= tmp.b_size;
 			start_blk += logical_to_blk(inode, size);
 
 			/*
@@ -363,15 +379,13 @@ int __generic_block_fiemap(struct inode *inode,
 			 * soon as we find a hole that the last extent we found
 			 * is marked with FIEMAP_EXTENT_LAST
 			 */
-			if (!past_eof &&
-			    logical+size >=
-			    blk_to_logical(inode, 0)+i_size_read(inode))
-				past_eof = 1;
+			if (!past_eof && logical + size >= isize)
+				past_eof = true;
 		}
 		cond_resched();
 	} while (1);
 
-	/* if ret is 1 then we just hit the end of the extent array */
+	/* If ret is 1 then we just hit the end of the extent array */
 	if (ret == 1)
 		ret = 0;
 
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 9dd126276c9f..ed9ba6fe04f5 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -61,7 +61,7 @@ struct inode *jfs_iget(struct super_block *sb, unsigned long ino)
 			inode->i_op = &page_symlink_inode_operations;
 			inode->i_mapping->a_ops = &jfs_aops;
 		} else {
-			inode->i_op = &jfs_symlink_inode_operations;
+			inode->i_op = &jfs_fast_symlink_inode_operations;
 			/*
 			 * The inline data should be null-terminated, but
 			 * don't let on-disk corruption crash the kernel
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index 6c4dfcbf3f55..9e2f6a721668 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -196,7 +196,7 @@ int dbMount(struct inode *ipbmap)
 	bmp->db_maxag = le32_to_cpu(dbmp_le->dn_maxag);
 	bmp->db_agpref = le32_to_cpu(dbmp_le->dn_agpref);
 	bmp->db_aglevel = le32_to_cpu(dbmp_le->dn_aglevel);
-	bmp->db_agheigth = le32_to_cpu(dbmp_le->dn_agheigth);
+	bmp->db_agheight = le32_to_cpu(dbmp_le->dn_agheight);
 	bmp->db_agwidth = le32_to_cpu(dbmp_le->dn_agwidth);
 	bmp->db_agstart = le32_to_cpu(dbmp_le->dn_agstart);
 	bmp->db_agl2size = le32_to_cpu(dbmp_le->dn_agl2size);
@@ -288,7 +288,7 @@ int dbSync(struct inode *ipbmap)
 	dbmp_le->dn_maxag = cpu_to_le32(bmp->db_maxag);
 	dbmp_le->dn_agpref = cpu_to_le32(bmp->db_agpref);
 	dbmp_le->dn_aglevel = cpu_to_le32(bmp->db_aglevel);
-	dbmp_le->dn_agheigth = cpu_to_le32(bmp->db_agheigth);
+	dbmp_le->dn_agheight = cpu_to_le32(bmp->db_agheight);
 	dbmp_le->dn_agwidth = cpu_to_le32(bmp->db_agwidth);
 	dbmp_le->dn_agstart = cpu_to_le32(bmp->db_agstart);
 	dbmp_le->dn_agl2size = cpu_to_le32(bmp->db_agl2size);
@@ -1441,7 +1441,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
 	 * tree index of this allocation group within the control page.
 	 */
 	agperlev =
-	    (1 << (L2LPERCTL - (bmp->db_agheigth << 1))) / bmp->db_agwidth;
+	    (1 << (L2LPERCTL - (bmp->db_agheight << 1))) / bmp->db_agwidth;
 	ti = bmp->db_agstart + bmp->db_agwidth * (agno & (agperlev - 1));
 
 	/* dmap control page trees fan-out by 4 and a single allocation
@@ -1460,7 +1460,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results)
 		 * the subtree to find the leftmost leaf that describes this
 		 * free space.
 		 */
-		for (k = bmp->db_agheigth; k > 0; k--) {
+		for (k = bmp->db_agheight; k > 0; k--) {
 			for (n = 0, m = (ti << 2) + 1; n < 4; n++) {
 				if (l2nb <= dcp->stree[m + n]) {
 					ti = m + n;
@@ -3607,7 +3607,7 @@ void dbFinalizeBmap(struct inode *ipbmap)
 	}
 
 	/*
-	 * compute db_aglevel, db_agheigth, db_width, db_agstart:
+	 * compute db_aglevel, db_agheight, db_width, db_agstart:
 	 * an ag is covered in aglevel dmapctl summary tree,
 	 * at agheight level height (from leaf) with agwidth number of nodes
 	 * each, which starts at agstart index node of the smmary tree node
@@ -3616,9 +3616,9 @@ void dbFinalizeBmap(struct inode *ipbmap)
 	bmp->db_aglevel = BMAPSZTOLEV(bmp->db_agsize);
 	l2nl =
 	    bmp->db_agl2size - (L2BPERDMAP + bmp->db_aglevel * L2LPERCTL);
-	bmp->db_agheigth = l2nl >> 1;
-	bmp->db_agwidth = 1 << (l2nl - (bmp->db_agheigth << 1));
-	for (i = 5 - bmp->db_agheigth, bmp->db_agstart = 0, n = 1; i > 0;
+	bmp->db_agheight = l2nl >> 1;
+	bmp->db_agwidth = 1 << (l2nl - (bmp->db_agheight << 1));
+	for (i = 5 - bmp->db_agheight, bmp->db_agstart = 0, n = 1; i > 0;
 	     i--) {
 		bmp->db_agstart += n;
 		n <<= 2;
diff --git a/fs/jfs/jfs_dmap.h b/fs/jfs/jfs_dmap.h
index 1a6eb41569bc..6dcb906c55d8 100644
--- a/fs/jfs/jfs_dmap.h
+++ b/fs/jfs/jfs_dmap.h
@@ -210,7 +210,7 @@ struct dbmap_disk {
 	__le32 dn_maxag;	/* 4: max active alloc group number	*/
 	__le32 dn_agpref;	/* 4: preferred alloc group (hint)	*/
 	__le32 dn_aglevel;	/* 4: dmapctl level holding the AG	*/
-	__le32 dn_agheigth;	/* 4: height in dmapctl of the AG	*/
+	__le32 dn_agheight;	/* 4: height in dmapctl of the AG	*/
 	__le32 dn_agwidth;	/* 4: width in dmapctl of the AG	*/
 	__le32 dn_agstart;	/* 4: start tree index at AG height	*/
 	__le32 dn_agl2size;	/* 4: l2 num of blks per alloc group	*/
@@ -229,7 +229,7 @@ struct dbmap {
 	int dn_maxag;		/* max active alloc group number	*/
 	int dn_agpref;		/* preferred alloc group (hint)		*/
 	int dn_aglevel;		/* dmapctl level holding the AG		*/
-	int dn_agheigth;	/* height in dmapctl of the AG		*/
+	int dn_agheight;	/* height in dmapctl of the AG		*/
 	int dn_agwidth;		/* width in dmapctl of the AG		*/
 	int dn_agstart;		/* start tree index at AG height	*/
 	int dn_agl2size;	/* l2 num of blks per alloc group	*/
@@ -255,7 +255,7 @@ struct bmap {
 #define	db_agsize	db_bmap.dn_agsize
 #define	db_agl2size	db_bmap.dn_agl2size
 #define	db_agwidth	db_bmap.dn_agwidth
-#define	db_agheigth	db_bmap.dn_agheigth
+#define	db_agheight	db_bmap.dn_agheight
 #define	db_agstart	db_bmap.dn_agstart
 #define	db_numag	db_bmap.dn_numag
 #define	db_maxlevel	db_bmap.dn_maxlevel
diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h
index 79e2c79661df..9e6bda30a6e8 100644
--- a/fs/jfs/jfs_inode.h
+++ b/fs/jfs/jfs_inode.h
@@ -48,5 +48,6 @@ extern const struct file_operations jfs_dir_operations;
 extern const struct inode_operations jfs_file_inode_operations;
 extern const struct file_operations jfs_file_operations;
 extern const struct inode_operations jfs_symlink_inode_operations;
+extern const struct inode_operations jfs_fast_symlink_inode_operations;
 extern const struct dentry_operations jfs_ci_dentry_operations;
 #endif				/* _H_JFS_INODE */
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 4a3e9f39c21d..a9cf8e8675be 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -956,7 +956,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
 	 */
 
 	if (ssize <= IDATASIZE) {
-		ip->i_op = &jfs_symlink_inode_operations;
+		ip->i_op = &jfs_fast_symlink_inode_operations;
 
 		i_fastsymlink = JFS_IP(ip)->i_inline;
 		memcpy(i_fastsymlink, name, ssize);
@@ -978,7 +978,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
 	else {
 		jfs_info("jfs_symlink: allocate extent ip:0x%p", ip);
 
-		ip->i_op = &page_symlink_inode_operations;
+		ip->i_op = &jfs_symlink_inode_operations;
 		ip->i_mapping->a_ops = &jfs_aops;
 
 		/*
diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c
index 7f24a0bb08ca..1aba0039f1c9 100644
--- a/fs/jfs/resize.c
+++ b/fs/jfs/resize.c
@@ -81,6 +81,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
 	struct inode *iplist[1];
 	struct jfs_superblock *j_sb, *j_sb2;
 	uint old_agsize;
+	int agsizechanged = 0;
 	struct buffer_head *bh, *bh2;
 
 	/* If the volume hasn't grown, get out now */
@@ -333,6 +334,9 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
 	 */
 	if ((rc = dbExtendFS(ipbmap, XAddress, nblocks)))
 		goto error_out;
+
+	agsizechanged |= (bmp->db_agsize != old_agsize);
+
 	/*
 	 * the map now has extended to cover additional nblocks:
 	 * dn_mapsize = oldMapsize + nblocks;
@@ -432,7 +436,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize)
 	 * will correctly identify the new ag);
 	 */
 	/* if new AG size the same as old AG size, done! */
-	if (bmp->db_agsize != old_agsize) {
+	if (agsizechanged) {
 		if ((rc = diExtendFS(ipimap, ipbmap)))
 			goto error_out;
 
diff --git a/fs/jfs/symlink.c b/fs/jfs/symlink.c
index 4af1a05aad0a..205b946d8e0d 100644
--- a/fs/jfs/symlink.c
+++ b/fs/jfs/symlink.c
@@ -29,9 +29,21 @@ static void *jfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	return NULL;
 }
 
-const struct inode_operations jfs_symlink_inode_operations = {
+const struct inode_operations jfs_fast_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.follow_link	= jfs_follow_link,
+	.setattr	= jfs_setattr,
+	.setxattr	= jfs_setxattr,
+	.getxattr	= jfs_getxattr,
+	.listxattr	= jfs_listxattr,
+	.removexattr	= jfs_removexattr,
+};
+
+const struct inode_operations jfs_symlink_inode_operations = {
+	.readlink	= generic_readlink,
+	.follow_link	= page_follow_link_light,
+	.put_link	= page_put_link,
+	.setattr	= jfs_setattr,
 	.setxattr	= jfs_setxattr,
 	.getxattr	= jfs_getxattr,
 	.listxattr	= jfs_listxattr,
diff --git a/fs/logfs/gc.c b/fs/logfs/gc.c
index 84e36f52fe95..76c242fbe1b0 100644
--- a/fs/logfs/gc.c
+++ b/fs/logfs/gc.c
@@ -459,6 +459,14 @@ static void __logfs_gc_pass(struct super_block *sb, int target)
 	struct logfs_block *block;
 	int round, progress, last_progress = 0;
 
+	/*
+	 * Doing too many changes to the segfile at once would result
+	 * in a large number of aliases.  Write the journal before
+	 * things get out of hand.
+	 */
+	if (super->s_shadow_tree.no_shadowed_segments >= MAX_OBJ_ALIASES)
+		logfs_write_anchor(sb);
+
 	if (no_free_segments(sb) >= target &&
 			super->s_no_object_aliases < MAX_OBJ_ALIASES)
 		return;
diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c
index 33bd260b8309..fb0a613f885b 100644
--- a/fs/logfs/journal.c
+++ b/fs/logfs/journal.c
@@ -389,7 +389,10 @@ static void journal_get_erase_count(struct logfs_area *area)
 static int journal_erase_segment(struct logfs_area *area)
 {
 	struct super_block *sb = area->a_sb;
-	struct logfs_segment_header sh;
+	union {
+		struct logfs_segment_header sh;
+		unsigned char c[ALIGN(sizeof(struct logfs_segment_header), 16)];
+	} u;
 	u64 ofs;
 	int err;
 
@@ -397,20 +400,21 @@ static int journal_erase_segment(struct logfs_area *area)
 	if (err)
 		return err;
 
-	sh.pad = 0;
-	sh.type = SEG_JOURNAL;
-	sh.level = 0;
-	sh.segno = cpu_to_be32(area->a_segno);
-	sh.ec = cpu_to_be32(area->a_erase_count);
-	sh.gec = cpu_to_be64(logfs_super(sb)->s_gec);
-	sh.crc = logfs_crc32(&sh, sizeof(sh), 4);
+	memset(&u, 0, sizeof(u));
+	u.sh.pad = 0;
+	u.sh.type = SEG_JOURNAL;
+	u.sh.level = 0;
+	u.sh.segno = cpu_to_be32(area->a_segno);
+	u.sh.ec = cpu_to_be32(area->a_erase_count);
+	u.sh.gec = cpu_to_be64(logfs_super(sb)->s_gec);
+	u.sh.crc = logfs_crc32(&u.sh, sizeof(u.sh), 4);
 
 	/* This causes a bug in segment.c.  Not yet. */
 	//logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, 0);
 
 	ofs = dev_ofs(sb, area->a_segno, 0);
-	area->a_used_bytes = ALIGN(sizeof(sh), 16);
-	logfs_buf_write(area, ofs, &sh, sizeof(sh));
+	area->a_used_bytes = sizeof(u);
+	logfs_buf_write(area, ofs, &u, sizeof(u));
 	return 0;
 }
 
@@ -494,6 +498,8 @@ static void account_shadows(struct super_block *sb)
 
 	btree_grim_visitor64(&tree->new, (unsigned long)sb, account_shadow);
 	btree_grim_visitor64(&tree->old, (unsigned long)sb, account_shadow);
+	btree_grim_visitor32(&tree->segment_map, 0, NULL);
+	tree->no_shadowed_segments = 0;
 
 	if (li->li_block) {
 		/*
@@ -607,9 +613,9 @@ static size_t __logfs_write_je(struct super_block *sb, void *buf, u16 type,
 	if (len == 0)
 		return logfs_write_header(super, header, 0, type);
 
+	BUG_ON(len > sb->s_blocksize);
 	compr_len = logfs_compress(buf, data, len, sb->s_blocksize);
 	if (compr_len < 0 || type == JE_ANCHOR) {
-		BUG_ON(len > sb->s_blocksize);
 		memcpy(data, buf, len);
 		compr_len = len;
 		compr = COMPR_NONE;
@@ -661,6 +667,7 @@ static int logfs_write_je_buf(struct super_block *sb, void *buf, u16 type,
 	if (ofs < 0)
 		return ofs;
 	logfs_buf_write(area, ofs, super->s_compressed_je, len);
+	BUG_ON(super->s_no_je >= MAX_JOURNAL_ENTRIES);
 	super->s_je_array[super->s_no_je++] = cpu_to_be64(ofs);
 	return 0;
 }
diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h
index b84b0eec6024..0a3df1a0c936 100644
--- a/fs/logfs/logfs.h
+++ b/fs/logfs/logfs.h
@@ -257,10 +257,14 @@ struct logfs_shadow {
  * struct shadow_tree
  * @new:			shadows where old_ofs==0, indexed by new_ofs
  * @old:			shadows where old_ofs!=0, indexed by old_ofs
+ * @segment_map:		bitfield of segments containing shadows
+ * @no_shadowed_segment:	number of segments containing shadows
  */
 struct shadow_tree {
 	struct btree_head64 new;
 	struct btree_head64 old;
+	struct btree_head32 segment_map;
+	int no_shadowed_segments;
 };
 
 struct object_alias_item {
@@ -305,13 +309,14 @@ typedef int write_alias_t(struct super_block *sb, u64 ino, u64 bix,
 		level_t level, int child_no, __be64 val);
 struct logfs_block_ops {
 	void	(*write_block)(struct logfs_block *block);
-	gc_level_t	(*block_level)(struct logfs_block *block);
 	void	(*free_block)(struct super_block *sb, struct logfs_block*block);
 	int	(*write_alias)(struct super_block *sb,
 			struct logfs_block *block,
 			write_alias_t *write_one_alias);
 };
 
+#define MAX_JOURNAL_ENTRIES 256
+
 struct logfs_super {
 	struct mtd_info *s_mtd;			/* underlying device */
 	struct block_device *s_bdev;		/* underlying device */
@@ -378,7 +383,7 @@ struct logfs_super {
 	u32	 s_journal_ec[LOGFS_JOURNAL_SEGS]; /* journal erasecounts */
 	u64	 s_last_version;
 	struct logfs_area *s_journal_area;	/* open journal segment */
-	__be64	s_je_array[64];
+	__be64	s_je_array[MAX_JOURNAL_ENTRIES];
 	int	s_no_je;
 
 	int	 s_sum_index;			/* for the 12 summaries */
@@ -722,4 +727,10 @@ static inline struct logfs_area *get_area(struct super_block *sb,
 	return logfs_super(sb)->s_area[(__force u8)gc_level];
 }
 
+static inline void logfs_mempool_destroy(mempool_t *pool)
+{
+	if (pool)
+		mempool_destroy(pool);
+}
+
 #endif
diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c
index bff40253dfb2..3159db6958e5 100644
--- a/fs/logfs/readwrite.c
+++ b/fs/logfs/readwrite.c
@@ -430,25 +430,6 @@ static void inode_write_block(struct logfs_block *block)
 	}
 }
 
-static gc_level_t inode_block_level(struct logfs_block *block)
-{
-	BUG_ON(block->inode->i_ino == LOGFS_INO_MASTER);
-	return GC_LEVEL(LOGFS_MAX_LEVELS);
-}
-
-static gc_level_t indirect_block_level(struct logfs_block *block)
-{
-	struct page *page;
-	struct inode *inode;
-	u64 bix;
-	level_t level;
-
-	page = block->page;
-	inode = page->mapping->host;
-	logfs_unpack_index(page->index, &bix, &level);
-	return expand_level(inode->i_ino, level);
-}
-
 /*
  * This silences a false, yet annoying gcc warning.  I hate it when my editor
  * jumps into bitops.h each time I recompile this file.
@@ -587,14 +568,12 @@ static void indirect_free_block(struct super_block *sb,
 
 static struct logfs_block_ops inode_block_ops = {
 	.write_block = inode_write_block,
-	.block_level = inode_block_level,
 	.free_block = inode_free_block,
 	.write_alias = inode_write_alias,
 };
 
 struct logfs_block_ops indirect_block_ops = {
 	.write_block = indirect_write_block,
-	.block_level = indirect_block_level,
 	.free_block = indirect_free_block,
 	.write_alias = indirect_write_alias,
 };
@@ -1241,6 +1220,18 @@ static void free_shadow(struct inode *inode, struct logfs_shadow *shadow)
 	mempool_free(shadow, super->s_shadow_pool);
 }
 
+static void mark_segment(struct shadow_tree *tree, u32 segno)
+{
+	int err;
+
+	if (!btree_lookup32(&tree->segment_map, segno)) {
+		err = btree_insert32(&tree->segment_map, segno, (void *)1,
+				GFP_NOFS);
+		BUG_ON(err);
+		tree->no_shadowed_segments++;
+	}
+}
+
 /**
  * fill_shadow_tree - Propagate shadow tree changes due to a write
  * @inode:	Inode owning the page
@@ -1288,6 +1279,8 @@ static void fill_shadow_tree(struct inode *inode, struct page *page,
 
 		super->s_dirty_used_bytes += shadow->new_len;
 		super->s_dirty_free_bytes += shadow->old_len;
+		mark_segment(tree, shadow->old_ofs >> super->s_segshift);
+		mark_segment(tree, shadow->new_ofs >> super->s_segshift);
 	}
 }
 
@@ -1845,19 +1838,37 @@ static int __logfs_truncate(struct inode *inode, u64 size)
 	return logfs_truncate_direct(inode, size);
 }
 
-int logfs_truncate(struct inode *inode, u64 size)
+/*
+ * Truncate, by changing the segment file, can consume a fair amount
+ * of resources.  So back off from time to time and do some GC.
+ * 8 or 2048 blocks should be well within safety limits even if
+ * every single block resided in a different segment.
+ */
+#define TRUNCATE_STEP	(8 * 1024 * 1024)
+int logfs_truncate(struct inode *inode, u64 target)
 {
 	struct super_block *sb = inode->i_sb;
-	int err;
+	u64 size = i_size_read(inode);
+	int err = 0;
 
-	logfs_get_wblocks(sb, NULL, 1);
-	err = __logfs_truncate(inode, size);
-	if (!err)
-		err = __logfs_write_inode(inode, 0);
-	logfs_put_wblocks(sb, NULL, 1);
+	size = ALIGN(size, TRUNCATE_STEP);
+	while (size > target) {
+		if (size > TRUNCATE_STEP)
+			size -= TRUNCATE_STEP;
+		else
+			size = 0;
+		if (size < target)
+			size = target;
+
+		logfs_get_wblocks(sb, NULL, 1);
+		err = __logfs_truncate(inode, target);
+		if (!err)
+			err = __logfs_write_inode(inode, 0);
+		logfs_put_wblocks(sb, NULL, 1);
+	}
 
 	if (!err)
-		err = vmtruncate(inode, size);
+		err = vmtruncate(inode, target);
 
 	/* I don't trust error recovery yet. */
 	WARN_ON(err);
@@ -2251,8 +2262,6 @@ void logfs_cleanup_rw(struct super_block *sb)
 	struct logfs_super *super = logfs_super(sb);
 
 	destroy_meta_inode(super->s_segfile_inode);
-	if (super->s_block_pool)
-		mempool_destroy(super->s_block_pool);
-	if (super->s_shadow_pool)
-		mempool_destroy(super->s_shadow_pool);
+	logfs_mempool_destroy(super->s_block_pool);
+	logfs_mempool_destroy(super->s_shadow_pool);
 }
diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c
index 801a3a141625..f77ce2b470ba 100644
--- a/fs/logfs/segment.c
+++ b/fs/logfs/segment.c
@@ -183,14 +183,8 @@ static int btree_write_alias(struct super_block *sb, struct logfs_block *block,
 	return 0;
 }
 
-static gc_level_t btree_block_level(struct logfs_block *block)
-{
-	return expand_level(block->ino, block->level);
-}
-
 static struct logfs_block_ops btree_block_ops = {
 	.write_block	= btree_write_block,
-	.block_level	= btree_block_level,
 	.free_block	= __free_block,
 	.write_alias	= btree_write_alias,
 };
@@ -919,7 +913,7 @@ err:
 	for (i--; i >= 0; i--)
 		free_area(super->s_area[i]);
 	free_area(super->s_journal_area);
-	mempool_destroy(super->s_alias_pool);
+	logfs_mempool_destroy(super->s_alias_pool);
 	return -ENOMEM;
 }
 
diff --git a/fs/logfs/super.c b/fs/logfs/super.c
index b60bfac3263c..5866ee6e1327 100644
--- a/fs/logfs/super.c
+++ b/fs/logfs/super.c
@@ -12,6 +12,7 @@
 #include "logfs.h"
 #include <linux/bio.h>
 #include <linux/slab.h>
+#include <linux/blkdev.h>
 #include <linux/mtd/mtd.h>
 #include <linux/statfs.h>
 #include <linux/buffer_head.h>
@@ -137,6 +138,10 @@ static int logfs_sb_set(struct super_block *sb, void *_super)
 	sb->s_fs_info = super;
 	sb->s_mtd = super->s_mtd;
 	sb->s_bdev = super->s_bdev;
+	if (sb->s_bdev)
+		sb->s_bdi = &bdev_get_queue(sb->s_bdev)->backing_dev_info;
+	if (sb->s_mtd)
+		sb->s_bdi = sb->s_mtd->backing_dev_info;
 	return 0;
 }
 
@@ -452,6 +457,8 @@ static int logfs_read_sb(struct super_block *sb, int read_only)
 
 	btree_init_mempool64(&super->s_shadow_tree.new, super->s_btree_pool);
 	btree_init_mempool64(&super->s_shadow_tree.old, super->s_btree_pool);
+	btree_init_mempool32(&super->s_shadow_tree.segment_map,
+			super->s_btree_pool);
 
 	ret = logfs_init_mapping(sb);
 	if (ret)
@@ -516,8 +523,8 @@ static void logfs_kill_sb(struct super_block *sb)
 	if (super->s_erase_page)
 		__free_page(super->s_erase_page);
 	super->s_devops->put_device(sb);
-	mempool_destroy(super->s_btree_pool);
-	mempool_destroy(super->s_alias_pool);
+	logfs_mempool_destroy(super->s_btree_pool);
+	logfs_mempool_destroy(super->s_alias_pool);
 	kfree(super);
 	log_super("LogFS: Finished unmounting\n");
 }
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index cf98da1be23e..fa3385154023 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -526,10 +526,15 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 	sb->s_blocksize_bits = 10;
 	sb->s_magic = NCP_SUPER_MAGIC;
 	sb->s_op = &ncp_sops;
+	sb->s_bdi = &server->bdi;
 
 	server = NCP_SBP(sb);
 	memset(server, 0, sizeof(*server));
 
+	error = bdi_setup_and_register(&server->bdi, "ncpfs", BDI_CAP_MAP_COPY);
+	if (error)
+		goto out_bdi;
+
 	server->ncp_filp = ncp_filp;
 	server->ncp_sock = sock;
 	
@@ -719,6 +724,8 @@ out_fput2:
 	if (server->info_filp)
 		fput(server->info_filp);
 out_fput:
+	bdi_destroy(&server->bdi);
+out_bdi:
 	/* 23/12/1998 Marcin Dalecki <dalecki@cs.net.pl>:
 	 * 
 	 * The previously used put_filp(ncp_filp); was bogous, since
@@ -756,6 +763,7 @@ static void ncp_put_super(struct super_block *sb)
 	kill_pid(server->m.wdog_pid, SIGTERM, 1);
 	put_pid(server->m.wdog_pid);
 
+	bdi_destroy(&server->bdi);
 	kfree(server->priv.data);
 	kfree(server->auth.object_name);
 	vfree(server->rxbuf);
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 2a3d352c0bff..a8766c4ef2e0 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1294,7 +1294,8 @@ static int nfs4_init_server(struct nfs_server *server,
 
 	/* Initialise the client representation from the mount data */
 	server->flags = data->flags;
-	server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR;
+	server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR|
+		NFS_CAP_POSIX_LOCK;
 	server->options = data->options;
 
 	/* Get a client record */
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index c6f2750648f4..db3ad849a289 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -837,6 +837,8 @@ out_zap_parent:
 		/* If we have submounts, don't unhash ! */
 		if (have_submounts(dentry))
 			goto out_valid;
+		if (dentry->d_flags & DCACHE_DISCONNECTED)
+			goto out_valid;
 		shrink_dcache_parent(dentry);
 	}
 	d_drop(dentry);
@@ -1025,12 +1027,12 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry
 				res = NULL;
 				goto out;
 			/* This turned out not to be a regular file */
+			case -EISDIR:
 			case -ENOTDIR:
 				goto no_open;
 			case -ELOOP:
 				if (!(nd->intent.open.flags & O_NOFOLLOW))
 					goto no_open;
-			/* case -EISDIR: */
 			/* case -EINVAL: */
 			default:
 				goto out;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 737128f777f3..50a56edca0b5 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -623,10 +623,10 @@ struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_c
 	list_for_each_entry(pos, &nfsi->open_files, list) {
 		if (cred != NULL && pos->cred != cred)
 			continue;
-		if ((pos->mode & mode) == mode) {
-			ctx = get_nfs_open_context(pos);
-			break;
-		}
+		if ((pos->mode & (FMODE_READ|FMODE_WRITE)) != mode)
+			continue;
+		ctx = get_nfs_open_context(pos);
+		break;
 	}
 	spin_unlock(&inode->i_lock);
 	return ctx;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index fe0cd9eb1d4d..638067007c65 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1523,6 +1523,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 		nfs_post_op_update_inode(dir, o_res->dir_attr);
 	} else
 		nfs_refresh_inode(dir, o_res->dir_attr);
+	if ((o_res->rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) == 0)
+		server->caps &= ~NFS_CAP_POSIX_LOCK;
 	if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
 		status = _nfs4_proc_open_confirm(data);
 		if (status != 0)
@@ -1664,7 +1666,7 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in
 	status = PTR_ERR(state);
 	if (IS_ERR(state))
 		goto err_opendata_put;
-	if ((opendata->o_res.rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) != 0)
+	if (server->caps & NFS_CAP_POSIX_LOCK)
 		set_bit(NFS_STATE_POSIX_LOCKS, &state->flags);
 	nfs4_opendata_put(opendata);
 	nfs4_put_state_owner(sp);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 53ff70e23993..de38d63aa920 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -201,6 +201,7 @@ static int nfs_set_page_writeback(struct page *page)
 		struct inode *inode = page->mapping->host;
 		struct nfs_server *nfss = NFS_SERVER(inode);
 
+		page_cache_get(page);
 		if (atomic_long_inc_return(&nfss->writeback) >
 				NFS_CONGESTION_ON_THRESH) {
 			set_bdi_congested(&nfss->backing_dev_info,
@@ -216,6 +217,7 @@ static void nfs_end_page_writeback(struct page *page)
 	struct nfs_server *nfss = NFS_SERVER(inode);
 
 	end_page_writeback(page);
+	page_cache_release(page);
 	if (atomic_long_dec_return(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH)
 		clear_bdi_congested(&nfss->backing_dev_info, BLK_RW_ASYNC);
 }
@@ -421,6 +423,7 @@ static void
 nfs_mark_request_dirty(struct nfs_page *req)
 {
 	__set_page_dirty_nobuffers(req->wb_page);
+	__mark_inode_dirty(req->wb_page->mapping->host, I_DIRTY_DATASYNC);
 }
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
@@ -660,9 +663,11 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page,
 	req = nfs_setup_write_request(ctx, page, offset, count);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
+	nfs_mark_request_dirty(req);
 	/* Update file length */
 	nfs_grow_file(page, offset, count);
 	nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes);
+	nfs_mark_request_dirty(req);
 	nfs_clear_page_tag_locked(req);
 	return 0;
 }
@@ -739,8 +744,6 @@ int nfs_updatepage(struct file *file, struct page *page,
 	status = nfs_writepage_setup(ctx, page, offset, count);
 	if (status < 0)
 		nfs_set_pageerror(page);
-	else
-		__set_page_dirty_nobuffers(page);
 
 	dprintk("NFS:       nfs_updatepage returns %d (isize %lld)\n",
 			status, (long long)i_size_read(inode));
@@ -749,13 +752,12 @@ int nfs_updatepage(struct file *file, struct page *page,
 
 static void nfs_writepage_release(struct nfs_page *req)
 {
+	struct page *page = req->wb_page;
 
-	if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req)) {
-		nfs_end_page_writeback(req->wb_page);
+	if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req))
 		nfs_inode_remove_request(req);
-	} else
-		nfs_end_page_writeback(req->wb_page);
 	nfs_clear_page_tag_locked(req);
+	nfs_end_page_writeback(page);
 }
 
 static int flush_task_priority(int how)
@@ -779,7 +781,6 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
 		int how)
 {
 	struct inode *inode = req->wb_context->path.dentry->d_inode;
-	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
 	int priority = flush_task_priority(how);
 	struct rpc_task *task;
 	struct rpc_message msg = {
@@ -794,9 +795,10 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
 		.callback_ops = call_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = flags,
+		.flags = RPC_TASK_ASYNC,
 		.priority = priority,
 	};
+	int ret = 0;
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
@@ -835,10 +837,18 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
 		(unsigned long long)data->args.offset);
 
 	task = rpc_run_task(&task_setup_data);
-	if (IS_ERR(task))
-		return PTR_ERR(task);
+	if (IS_ERR(task)) {
+		ret = PTR_ERR(task);
+		goto out;
+	}
+	if (how & FLUSH_SYNC) {
+		ret = rpc_wait_for_completion_task(task);
+		if (ret == 0)
+			ret = task->tk_status;
+	}
 	rpc_put_task(task);
-	return 0;
+out:
+	return ret;
 }
 
 /* If a nfs_flush_* function fails, it should remove reqs from @head and
@@ -847,9 +857,11 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
  */
 static void nfs_redirty_request(struct nfs_page *req)
 {
+	struct page *page = req->wb_page;
+
 	nfs_mark_request_dirty(req);
-	nfs_end_page_writeback(req->wb_page);
 	nfs_clear_page_tag_locked(req);
+	nfs_end_page_writeback(page);
 }
 
 /*
@@ -1084,16 +1096,15 @@ static void nfs_writeback_release_full(void *calldata)
 		if (nfs_write_need_commit(data)) {
 			memcpy(&req->wb_verf, &data->verf, sizeof(req->wb_verf));
 			nfs_mark_request_commit(req);
-			nfs_end_page_writeback(page);
 			dprintk(" marked for commit\n");
 			goto next;
 		}
 		dprintk(" OK\n");
 remove_request:
-		nfs_end_page_writeback(page);
 		nfs_inode_remove_request(req);
 	next:
 		nfs_clear_page_tag_locked(req);
+		nfs_end_page_writeback(page);
 	}
 	nfs_writedata_release(calldata);
 }
@@ -1207,7 +1218,6 @@ static int nfs_commit_rpcsetup(struct list_head *head,
 {
 	struct nfs_page *first = nfs_list_entry(head->next);
 	struct inode *inode = first->wb_context->path.dentry->d_inode;
-	int flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
 	int priority = flush_task_priority(how);
 	struct rpc_task *task;
 	struct rpc_message msg = {
@@ -1222,7 +1232,7 @@ static int nfs_commit_rpcsetup(struct list_head *head,
 		.callback_ops = &nfs_commit_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = flags,
+		.flags = RPC_TASK_ASYNC,
 		.priority = priority,
 	};
 
@@ -1252,6 +1262,8 @@ static int nfs_commit_rpcsetup(struct list_head *head,
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
+	if (how & FLUSH_SYNC)
+		rpc_wait_for_completion_task(task);
 	rpc_put_task(task);
 	return 0;
 }
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index e1703175ee28..34ccf815ea8a 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -161,10 +161,10 @@ static __be32 *read_buf(struct nfsd4_compoundargs *argp, u32 nbytes)
 	argp->p = page_address(argp->pagelist[0]);
 	argp->pagelist++;
 	if (argp->pagelen < PAGE_SIZE) {
-		argp->end = p + (argp->pagelen>>2);
+		argp->end = argp->p + (argp->pagelen>>2);
 		argp->pagelen = 0;
 	} else {
-		argp->end = p + (PAGE_SIZE>>2);
+		argp->end = argp->p + (PAGE_SIZE>>2);
 		argp->pagelen -= PAGE_SIZE;
 	}
 	memcpy(((char*)p)+avail, argp->p, (nbytes - avail));
@@ -1426,10 +1426,10 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 			argp->p = page_address(argp->pagelist[0]);
 			argp->pagelist++;
 			if (argp->pagelen < PAGE_SIZE) {
-				argp->end = p + (argp->pagelen>>2);
+				argp->end = argp->p + (argp->pagelen>>2);
 				argp->pagelen = 0;
 			} else {
-				argp->end = p + (PAGE_SIZE>>2);
+				argp->end = argp->p + (PAGE_SIZE>>2);
 				argp->pagelen -= PAGE_SIZE;
 			}
 		}
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 7621db800a74..8418fcc0a6ab 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2909,7 +2909,7 @@ out_no_task:
  */
 static const struct pid_entry tid_base_stuff[] = {
 	DIR("fd",        S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations),
-	DIR("fdinfo",    S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fd_operations),
+	DIR("fdinfo",    S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations),
 	REG("environ",   S_IRUSR, proc_environ_operations),
 	INF("auxv",      S_IRUSR, proc_pid_auxv),
 	ONE("status",    S_IRUGO, proc_pid_status),
diff --git a/fs/quota/Kconfig b/fs/quota/Kconfig
index dad7fb247ddc..3e21b1e2ad3a 100644
--- a/fs/quota/Kconfig
+++ b/fs/quota/Kconfig
@@ -33,6 +33,14 @@ config PRINT_QUOTA_WARNING
 	  Note that this behavior is currently deprecated and may go away in
 	  future. Please use notification via netlink socket instead.
 
+config QUOTA_DEBUG
+	bool "Additional quota sanity checks"
+	depends on QUOTA
+	default n
+	help
+	  If you say Y here, quota subsystem will perform some additional
+	  sanity checks of quota internal structures. If unsure, say N.
+
 # Generic support for tree structured quota files. Selected when needed.
 config QUOTA_TREE
 	 tristate
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index a0a9405b202a..788b5802a7ce 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -80,8 +80,6 @@
 
 #include <asm/uaccess.h>
 
-#define __DQUOT_PARANOIA
-
 /*
  * There are three quota SMP locks. dq_list_lock protects all lists with quotas
  * and quota formats, dqstats structure containing statistics about the lists
@@ -695,7 +693,7 @@ void dqput(struct dquot *dquot)
 
 	if (!dquot)
 		return;
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 	if (!atomic_read(&dquot->dq_count)) {
 		printk("VFS: dqput: trying to free free dquot\n");
 		printk("VFS: device %s, dquot of %s %d\n",
@@ -748,7 +746,7 @@ we_slept:
 		goto we_slept;
 	}
 	atomic_dec(&dquot->dq_count);
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 	/* sanity check */
 	BUG_ON(!list_empty(&dquot->dq_free));
 #endif
@@ -845,7 +843,7 @@ we_slept:
 		dquot = NULL;
 		goto out;
 	}
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 	BUG_ON(!dquot->dq_sb);	/* Has somebody invalidated entry under us? */
 #endif
 out:
@@ -874,7 +872,7 @@ static int dqinit_needed(struct inode *inode, int type)
 static void add_dquot_ref(struct super_block *sb, int type)
 {
 	struct inode *inode, *old_inode = NULL;
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 	int reserved = 0;
 #endif
 
@@ -882,7 +880,7 @@ static void add_dquot_ref(struct super_block *sb, int type)
 	list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
 		if (inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE|I_NEW))
 			continue;
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 		if (unlikely(inode_get_rsv_space(inode) > 0))
 			reserved = 1;
 #endif
@@ -907,7 +905,7 @@ static void add_dquot_ref(struct super_block *sb, int type)
 	spin_unlock(&inode_lock);
 	iput(old_inode);
 
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 	if (reserved) {
 		printk(KERN_WARNING "VFS (%s): Writes happened before quota"
 			" was turned on thus quota information is probably "
@@ -940,7 +938,7 @@ static int remove_inode_dquot_ref(struct inode *inode, int type,
 	inode->i_dquot[type] = NULL;
 	if (dquot) {
 		if (dqput_blocks(dquot)) {
-#ifdef __DQUOT_PARANOIA
+#ifdef CONFIG_QUOTA_DEBUG
 			if (atomic_read(&dquot->dq_count) != 1)
 				printk(KERN_WARNING "VFS: Adding dquot with dq_count %d to dispose list.\n", atomic_read(&dquot->dq_count));
 #endif
diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c
index f8a6075abf50..07930449a958 100644
--- a/fs/reiserfs/dir.c
+++ b/fs/reiserfs/dir.c
@@ -46,8 +46,6 @@ static inline bool is_privroot_deh(struct dentry *dir,
 				   struct reiserfs_de_head *deh)
 {
 	struct dentry *privroot = REISERFS_SB(dir->d_sb)->priv_root;
-	if (reiserfs_expose_privroot(dir->d_sb))
-		return 0;
 	return (dir == dir->d_parent && privroot->d_inode &&
 	        deh->deh_objectid == INODE_PKEY(privroot->d_inode)->k_objectid);
 }
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 4f9586bb7631..e7cc00e636dc 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -554,7 +554,7 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
 	if (!err && new_size < i_size_read(dentry->d_inode)) {
 		struct iattr newattrs = {
 			.ia_ctime = current_fs_time(inode->i_sb),
-			.ia_size = buffer_size,
+			.ia_size = new_size,
 			.ia_valid = ATTR_SIZE | ATTR_CTIME,
 		};
 
@@ -973,21 +973,13 @@ int reiserfs_permission(struct inode *inode, int mask)
 	return generic_permission(inode, mask, NULL);
 }
 
-/* This will catch lookups from the fs root to .reiserfs_priv */
-static int
-xattr_lookup_poison(struct dentry *dentry, struct qstr *q1, struct qstr *name)
+static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
-	struct dentry *priv_root = REISERFS_SB(dentry->d_sb)->priv_root;
-	if (container_of(q1, struct dentry, d_name) == priv_root)
-		return -ENOENT;
-	if (q1->len == name->len &&
-		   !memcmp(q1->name, name->name, name->len))
-		return 0;
-	return 1;
+	return -EPERM;
 }
 
 static const struct dentry_operations xattr_lookup_poison_ops = {
-	.d_compare = xattr_lookup_poison,
+	.d_revalidate = xattr_hide_revalidate,
 };
 
 int reiserfs_lookup_privroot(struct super_block *s)
@@ -1001,8 +993,7 @@ int reiserfs_lookup_privroot(struct super_block *s)
 				strlen(PRIVROOT_NAME));
 	if (!IS_ERR(dentry)) {
 		REISERFS_SB(s)->priv_root = dentry;
-		if (!reiserfs_expose_privroot(s))
-			s->s_root->d_op = &xattr_lookup_poison_ops;
+		dentry->d_op = &xattr_lookup_poison_ops;
 		if (dentry->d_inode)
 			dentry->d_inode->i_flags |= S_PRIVATE;
 	} else
diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c
index 1c4c8f089970..dfa1d67f8fca 100644
--- a/fs/smbfs/inode.c
+++ b/fs/smbfs/inode.c
@@ -479,6 +479,7 @@ smb_put_super(struct super_block *sb)
 	if (server->conn_pid)
 		kill_pid(server->conn_pid, SIGTERM, 1);
 
+	bdi_destroy(&server->bdi);
 	kfree(server->ops);
 	smb_unload_nls(server);
 	sb->s_fs_info = NULL;
@@ -525,6 +526,11 @@ static int smb_fill_super(struct super_block *sb, void *raw_data, int silent)
 	if (!server)
 		goto out_no_server;
 	sb->s_fs_info = server;
+	
+	if (bdi_setup_and_register(&server->bdi, "smbfs", BDI_CAP_MAP_COPY))
+		goto out_bdi;
+
+	sb->s_bdi = &server->bdi;
 
 	server->super_block = sb;
 	server->mnt = NULL;
@@ -624,6 +630,8 @@ out_no_smbiod:
 out_bad_option:
 	kfree(mem);
 out_no_mem:
+	bdi_destroy(&server->bdi);
+out_bdi:
 	if (!server->mnt)
 		printk(KERN_ERR "smb_fill_super: allocation failure\n");
 	sb->s_fs_info = NULL;
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 1cb0d81b164b..653c030eb840 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -87,9 +87,8 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index,
 	u64 cur_index = index >> msblk->devblksize_log2;
 	int bytes, compressed, b = 0, k = 0, page = 0, avail;
 
-
-	bh = kcalloc((msblk->block_size >> msblk->devblksize_log2) + 1,
-				sizeof(*bh), GFP_KERNEL);
+	bh = kcalloc(((srclength + msblk->devblksize - 1)
+		>> msblk->devblksize_log2) + 1, sizeof(*bh), GFP_KERNEL);
 	if (bh == NULL)
 		return -ENOMEM;
 
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index 3550aec2f655..48b6f4a385a6 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -275,7 +275,8 @@ allocate_root:
 
 	err = squashfs_read_inode(root, root_inode);
 	if (err) {
-		iget_failed(root);
+		make_bad_inode(root);
+		iput(root);
 		goto failed_mount;
 	}
 	insert_inode_hash(root);
@@ -353,6 +354,7 @@ static void squashfs_put_super(struct super_block *sb)
 		kfree(sbi->id_table);
 		kfree(sbi->fragment_index);
 		kfree(sbi->meta_index);
+		kfree(sbi->inode_lookup_table);
 		kfree(sb->s_fs_info);
 		sb->s_fs_info = NULL;
 	}
diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c
index 15a03d0fb9f3..7a603874e483 100644
--- a/fs/squashfs/zlib_wrapper.c
+++ b/fs/squashfs/zlib_wrapper.c
@@ -128,8 +128,9 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void **buffer,
 		goto release_mutex;
 	}
 
+	length = stream->total_out;
 	mutex_unlock(&msblk->read_data_mutex);
-	return stream->total_out;
+	return length;
 
 release_mutex:
 	mutex_unlock(&msblk->read_data_mutex);
diff --git a/fs/super.c b/fs/super.c
index f35ac6022109..dc72491a19f9 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -693,6 +693,7 @@ int set_anon_super(struct super_block *s, void *data)
 		return -EMFILE;
 	}
 	s->s_dev = MKDEV(0, dev & MINORMASK);
+	s->s_bdi = &noop_backing_dev_info;
 	return 0;
 }
 
@@ -954,10 +955,11 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void
 	if (error < 0)
 		goto out_free_secdata;
 	BUG_ON(!mnt->mnt_sb);
+	WARN_ON(!mnt->mnt_sb->s_bdi);
 
- 	error = security_sb_kern_mount(mnt->mnt_sb, flags, secdata);
- 	if (error)
- 		goto out_sb;
+	error = security_sb_kern_mount(mnt->mnt_sb, flags, secdata);
+	if (error)
+		goto out_sb;
 
 	/*
 	 * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE
diff --git a/fs/sync.c b/fs/sync.c
index fc5c3d75cf3c..92b228176f7c 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -14,6 +14,7 @@
 #include <linux/pagemap.h>
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
+#include <linux/backing-dev.h>
 #include "internal.h"
 
 #define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
@@ -32,7 +33,7 @@ static int __sync_filesystem(struct super_block *sb, int wait)
 	 * This should be safe, as we require bdi backing to actually
 	 * write out data in the first place
 	 */
-	if (!sb->s_bdi)
+	if (!sb->s_bdi || sb->s_bdi == &noop_backing_dev_info)
 		return 0;
 
 	if (sb->s_qcop && sb->s_qcop->quota_sync)
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 05cd85317f6f..fd9698215759 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -820,10 +820,10 @@ xfs_reclaim_inode(
 	 * call into reclaim to find it in a clean state instead of waiting for
 	 * it now. We also don't return errors here - if the error is transient
 	 * then the next reclaim pass will flush the inode, and if the error
-	 * is permanent then the next sync reclaim will relcaim the inode and
+	 * is permanent then the next sync reclaim will reclaim the inode and
 	 * pass on the error.
 	 */
-	if (error && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+	if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
 		xfs_fs_cmn_err(CE_WARN, ip->i_mount,
 			"inode 0x%llx background reclaim flush failed with %d",
 			(long long)ip->i_ino, error);
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index cd27c9d6c71f..5bba29a07812 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -177,16 +177,26 @@ xfs_swap_extents_check_format(
 	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
 		return EINVAL;
 
-	/* Check root block of temp in btree form to max in target */
+	/*
+	 * If we are in a btree format, check that the temp root block will fit
+	 * in the target and that it has enough extents to be in btree format
+	 * in the target.
+	 *
+	 * Note that we have to be careful to allow btree->extent conversions
+	 * (a common defrag case) which will occur when the temp inode is in
+	 * extent format...
+	 */
 	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
-	    XFS_IFORK_BOFF(ip) &&
-	    tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+	    ((XFS_IFORK_BOFF(ip) &&
+	      tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip)) ||
+	     XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= ip->i_df.if_ext_max))
 		return EINVAL;
 
-	/* Check root block of target in btree form to max in temp */
+	/* Reciprocal target->temp btree format checks */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
-	    XFS_IFORK_BOFF(tip) &&
-	    ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+	    ((XFS_IFORK_BOFF(tip) &&
+	      ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip)) ||
+	     XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= tip->i_df.if_ext_max))
 		return EINVAL;
 
 	return 0;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index e8fba92d7cd9..2be019136287 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -745,9 +745,16 @@ xfs_log_move_tail(xfs_mount_t	*mp,
 
 /*
  * Determine if we have a transaction that has gone to disk
- * that needs to be covered. Log activity needs to be idle (no AIL and
- * nothing in the iclogs). And, we need to be in the right state indicating
- * something has gone out.
+ * that needs to be covered. To begin the transition to the idle state
+ * firstly the log needs to be idle (no AIL and nothing in the iclogs).
+ * If we are then in a state where covering is needed, the caller is informed
+ * that dummy transactions are required to move the log into the idle state.
+ *
+ * Because this is called as part of the sync process, we should also indicate
+ * that dummy transactions should be issued in anything but the covered or
+ * idle states. This ensures that the log tail is accurately reflected in
+ * the log at the end of the sync, hence if a crash occurrs avoids replay
+ * of transactions where the metadata is already on disk.
  */
 int
 xfs_log_need_covered(xfs_mount_t *mp)
@@ -759,17 +766,24 @@ xfs_log_need_covered(xfs_mount_t *mp)
 		return 0;
 
 	spin_lock(&log->l_icloglock);
-	if (((log->l_covered_state == XLOG_STATE_COVER_NEED) ||
-		(log->l_covered_state == XLOG_STATE_COVER_NEED2))
-			&& !xfs_trans_ail_tail(log->l_ailp)
-			&& xlog_iclogs_empty(log)) {
-		if (log->l_covered_state == XLOG_STATE_COVER_NEED)
-			log->l_covered_state = XLOG_STATE_COVER_DONE;
-		else {
-			ASSERT(log->l_covered_state == XLOG_STATE_COVER_NEED2);
-			log->l_covered_state = XLOG_STATE_COVER_DONE2;
+	switch (log->l_covered_state) {
+	case XLOG_STATE_COVER_DONE:
+	case XLOG_STATE_COVER_DONE2:
+	case XLOG_STATE_COVER_IDLE:
+		break;
+	case XLOG_STATE_COVER_NEED:
+	case XLOG_STATE_COVER_NEED2:
+		if (!xfs_trans_ail_tail(log->l_ailp) &&
+		    xlog_iclogs_empty(log)) {
+			if (log->l_covered_state == XLOG_STATE_COVER_NEED)
+				log->l_covered_state = XLOG_STATE_COVER_DONE;
+			else
+				log->l_covered_state = XLOG_STATE_COVER_DONE2;
 		}
+		/* FALLTHRU */
+	default:
 		needed = 1;
+		break;
 	}
 	spin_unlock(&log->l_icloglock);
 	return needed;