Merge branch 'v2.6.36-rc8' into for-2.6.37/barrier

Conflicts:
	block/blk-core.c
	drivers/block/loop.c
	mm/swapfile.c

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>

115 files changed, 1740 insertions, 996 deletions

diff --git a/fs/9p/fid.c b/fs/9p/fid.c
index 358563689064..6406f896bf95 100644
--- a/fs/9p/fid.c
+++ b/fs/9p/fid.c
@@ -242,7 +242,8 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry)
 	}
 	kfree(wnames);
 fid_out:
-	v9fs_fid_add(dentry, fid);
+	if (!IS_ERR(fid))
+		v9fs_fid_add(dentry, fid);
 err_out:
 	up_read(&v9ses->rename_sem);
 	return fid;
diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c
index 16c8a2a98c1b..899f168fd19c 100644
--- a/fs/9p/vfs_dir.c
+++ b/fs/9p/vfs_dir.c
@@ -292,9 +292,11 @@ int v9fs_dir_release(struct inode *inode, struct file *filp)
 
 	fid = filp->private_data;
 	P9_DPRINTK(P9_DEBUG_VFS,
-			"inode: %p filp: %p fid: %d\n", inode, filp, fid->fid);
+			"v9fs_dir_release: inode: %p filp: %p fid: %d\n",
+			inode, filp, fid ? fid->fid : -1);
 	filemap_write_and_wait(inode->i_mapping);
-	p9_client_clunk(fid);
+	if (fid)
+		p9_client_clunk(fid);
 	return 0;
 }
 
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index c7c23eab9440..9e670d527646 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -730,7 +730,10 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int mode,
 		P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err);
 		goto error;
 	}
-	dentry->d_op = &v9fs_cached_dentry_operations;
+	if (v9ses->cache)
+		dentry->d_op = &v9fs_cached_dentry_operations;
+	else
+		dentry->d_op = &v9fs_dentry_operations;
 	d_instantiate(dentry, inode);
 	err = v9fs_fid_add(dentry, fid);
 	if (err < 0)
@@ -1128,6 +1131,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	v9fs_stat2inode(st, dentry->d_inode, dentry->d_inode->i_sb);
 		generic_fillattr(dentry->d_inode, stat);
 
+	p9stat_free(st);
 	kfree(st);
 	return 0;
 }
@@ -1489,6 +1493,7 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen)
 
 	retval = strnlen(buffer, buflen);
 done:
+	p9stat_free(st);
 	kfree(st);
 	return retval;
 }
@@ -1942,7 +1947,7 @@ static const struct inode_operations v9fs_dir_inode_operations_dotu = {
 	.unlink = v9fs_vfs_unlink,
 	.mkdir = v9fs_vfs_mkdir,
 	.rmdir = v9fs_vfs_rmdir,
-	.mknod = v9fs_vfs_mknod_dotl,
+	.mknod = v9fs_vfs_mknod,
 	.rename = v9fs_vfs_rename,
 	.getattr = v9fs_vfs_getattr,
 	.setattr = v9fs_vfs_setattr,
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index f9311077de68..1d12ba0ed3db 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -122,6 +122,10 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
 	fid = v9fs_session_init(v9ses, dev_name, data);
 	if (IS_ERR(fid)) {
 		retval = PTR_ERR(fid);
+		/*
+		 * we need to call session_close to tear down some
+		 * of the data structure setup by session_init
+		 */
 		goto close_session;
 	}
 
@@ -144,7 +148,6 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
 		retval = -ENOMEM;
 		goto release_sb;
 	}
-
 	sb->s_root = root;
 
 	if (v9fs_proto_dotl(v9ses)) {
@@ -152,7 +155,7 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
 		st = p9_client_getattr_dotl(fid, P9_STATS_BASIC);
 		if (IS_ERR(st)) {
 			retval = PTR_ERR(st);
-			goto clunk_fid;
+			goto release_sb;
 		}
 
 		v9fs_stat2inode_dotl(st, root->d_inode);
@@ -162,7 +165,7 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
 		st = p9_client_stat(fid);
 		if (IS_ERR(st)) {
 			retval = PTR_ERR(st);
-			goto clunk_fid;
+			goto release_sb;
 		}
 
 		root->d_inode->i_ino = v9fs_qid2ino(&st->qid);
@@ -174,19 +177,24 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags,
 
 	v9fs_fid_add(root, fid);
 
-P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n");
+	P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n");
 	simple_set_mnt(mnt, sb);
 	return 0;
 
 clunk_fid:
 	p9_client_clunk(fid);
-
 close_session:
 	v9fs_session_close(v9ses);
 	kfree(v9ses);
 	return retval;
-
 release_sb:
+	/*
+	 * we will do the session_close and root dentry release
+	 * in the below call. But we need to clunk fid, because we haven't
+	 * attached the fid to dentry so it won't get clunked
+	 * automatically.
+	 */
+	p9_client_clunk(fid);
 	deactivate_locked_super(sb);
 	return retval;
 }
diff --git a/fs/aio.c b/fs/aio.c
index 3006b5bc33d6..250b0a73c8a8 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -712,8 +712,16 @@ static ssize_t aio_run_iocb(struct kiocb *iocb)
 	 */
 	ret = retry(iocb);
 
-	if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED)
+	if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED) {
+		/*
+		 * There's no easy way to restart the syscall since other AIO's
+		 * may be already running. Just fail this IO with EINTR.
+		 */
+		if (unlikely(ret == -ERESTARTSYS || ret == -ERESTARTNOINTR ||
+			     ret == -ERESTARTNOHAND || ret == -ERESTART_RESTARTBLOCK))
+			ret = -EINTR;
 		aio_complete(iocb, ret, 0);
+	}
 out:
 	spin_lock_irq(&ctx->ctx_lock);
 
@@ -1659,6 +1667,9 @@ long do_io_submit(aio_context_t ctx_id, long nr,
 	if (unlikely(nr < 0))
 		return -EINVAL;
 
+	if (unlikely(nr > LONG_MAX/sizeof(*iocbpp)))
+		nr = LONG_MAX/sizeof(*iocbpp);
+
 	if (unlikely(!access_ok(VERIFY_READ, iocbpp, (nr*sizeof(*iocbpp)))))
 		return -EFAULT;
 
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index f96eff04e11a..a6395bdb26ae 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -134,10 +134,6 @@ static int aout_core_dump(struct coredump_params *cprm)
 		if (!dump_write(file, dump_start, dump_size))
 			goto end_coredump;
 	}
-/* Finally dump the task struct.  Not be used by gdb, but could be useful */
-	set_fs(KERNEL_DS);
-	if (!dump_write(file, current, sizeof(*current)))
-		goto end_coredump;
 end_coredump:
 	set_fs(fs);
 	return has_dumped;
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index a7528b913936..fd0cc0bf9a40 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -724,7 +724,7 @@ static int __init init_misc_binfmt(void)
 {
 	int err = register_filesystem(&bm_fs_type);
 	if (!err) {
-		err = register_binfmt(&misc_format);
+		err = insert_binfmt(&misc_format);
 		if (err)
 			unregister_filesystem(&bm_fs_type);
 	}
diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c
index 612a5c38d3c1..4d0ff5ee27b8 100644
--- a/fs/bio-integrity.c
+++ b/fs/bio-integrity.c
@@ -413,10 +413,10 @@ int bio_integrity_prep(struct bio *bio)
 
 	/* Allocate kernel buffer for protection data */
 	len = sectors * blk_integrity_tuple_size(bi);
-	buf = kmalloc(len, GFP_NOIO | __GFP_NOFAIL | q->bounce_gfp);
+	buf = kmalloc(len, GFP_NOIO | q->bounce_gfp);
 	if (unlikely(buf == NULL)) {
 		printk(KERN_ERR "could not allocate integrity buffer\n");
-		return -EIO;
+		return -ENOMEM;
 	}
 
 	end = (((unsigned long) buf) + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index bc87b9c1d27e..0fcd2640c23f 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -3,6 +3,7 @@ config CEPH_FS
 	depends on INET && EXPERIMENTAL
 	select LIBCRC32C
 	select CRYPTO_AES
+	select CRYPTO
 	help
 	  Choose Y or M here to include support for mounting the
 	  experimental Ceph distributed file system.  Ceph is an extremely
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 5598a0d02295..efbc604001c8 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -87,7 +87,7 @@ static int ceph_set_page_dirty(struct page *page)
 
 	/* dirty the head */
 	spin_lock(&inode->i_lock);
-	if (ci->i_wrbuffer_ref_head == 0)
+	if (ci->i_head_snapc == NULL)
 		ci->i_head_snapc = ceph_get_snap_context(snapc);
 	++ci->i_wrbuffer_ref_head;
 	if (ci->i_wrbuffer_ref == 0)
@@ -105,13 +105,7 @@ static int ceph_set_page_dirty(struct page *page)
 	spin_lock_irq(&mapping->tree_lock);
 	if (page->mapping) {	/* Race with truncate? */
 		WARN_ON_ONCE(!PageUptodate(page));
-
-		if (mapping_cap_account_dirty(mapping)) {
-			__inc_zone_page_state(page, NR_FILE_DIRTY);
-			__inc_bdi_stat(mapping->backing_dev_info,
-					BDI_RECLAIMABLE);
-			task_io_account_write(PAGE_CACHE_SIZE);
-		}
+		account_page_dirtied(page, page->mapping);
 		radix_tree_tag_set(&mapping->page_tree,
 				page_index(page), PAGECACHE_TAG_DIRTY);
 
@@ -352,7 +346,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode,
 			break;
 		}
 	}
-	if (!snapc && ci->i_head_snapc) {
+	if (!snapc && ci->i_wrbuffer_ref_head) {
 		snapc = ceph_get_snap_context(ci->i_head_snapc);
 		dout(" head snapc %p has %d dirty pages\n",
 		     snapc, ci->i_wrbuffer_ref_head);
@@ -417,8 +411,8 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	if (i_size < page_off + len)
 		len = i_size - page_off;
 
-	dout("writepage %p page %p index %lu on %llu~%u\n",
-	     inode, page, page->index, page_off, len);
+	dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
+	     inode, page, page->index, page_off, len, snapc);
 
 	writeback_stat = atomic_long_inc_return(&client->writeback_count);
 	if (writeback_stat >
@@ -772,7 +766,8 @@ get_more_pages:
 			/* ok */
 			if (locked_pages == 0) {
 				/* prepare async write request */
-				offset = page->index << PAGE_CACHE_SHIFT;
+				offset = (unsigned long long)page->index
+					<< PAGE_CACHE_SHIFT;
 				len = wsize;
 				req = ceph_osdc_new_request(&client->osdc,
 					    &ci->i_layout,
diff --git a/fs/ceph/auth_x.c b/fs/ceph/auth_x.c
index 582e0b2caf8a..a2d002cbdec2 100644
--- a/fs/ceph/auth_x.c
+++ b/fs/ceph/auth_x.c
@@ -376,7 +376,7 @@ static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed)
 
 		th = get_ticket_handler(ac, service);
 
-		if (!th) {
+		if (IS_ERR(th)) {
 			*pneed |= service;
 			continue;
 		}
@@ -399,6 +399,9 @@ static int ceph_x_build_request(struct ceph_auth_client *ac,
 	struct ceph_x_ticket_handler *th =
 		get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
 
+	if (IS_ERR(th))
+		return PTR_ERR(th);
+
 	ceph_x_validate_tickets(ac, &need);
 
 	dout("build_request want %x have %x need %x\n",
@@ -450,7 +453,6 @@ static int ceph_x_build_request(struct ceph_auth_client *ac,
 			return -ERANGE;
 		head->op = cpu_to_le16(CEPHX_GET_PRINCIPAL_SESSION_KEY);
 
-		BUG_ON(!th);
 		ret = ceph_x_build_authorizer(ac, th, &xi->auth_authorizer);
 		if (ret)
 			return ret;
@@ -505,7 +507,8 @@ static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result,
 
 	case CEPHX_GET_PRINCIPAL_SESSION_KEY:
 		th = get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
-		BUG_ON(!th);
+		if (IS_ERR(th))
+			return PTR_ERR(th);
 		ret = ceph_x_proc_ticket_reply(ac, &th->session_key,
 					       buf + sizeof(*head), end);
 		break;
@@ -563,8 +566,8 @@ static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac,
 	void *end = p + sizeof(au->reply_buf);
 
 	th = get_ticket_handler(ac, au->service);
-	if (!th)
-		return -EIO;  /* hrm! */
+	if (IS_ERR(th))
+		return PTR_ERR(th);
 	ret = ceph_x_decrypt(&th->session_key, &p, end, &reply, sizeof(reply));
 	if (ret < 0)
 		return ret;
@@ -626,7 +629,7 @@ static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac,
 	struct ceph_x_ticket_handler *th;
 
 	th = get_ticket_handler(ac, peer_type);
-	if (th && !IS_ERR(th))
+	if (!IS_ERR(th))
 		remove_ticket_handler(ac, th);
 }
 
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 7bf182b03973..5e9da996a151 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -814,7 +814,7 @@ int __ceph_caps_used(struct ceph_inode_info *ci)
 		used |= CEPH_CAP_PIN;
 	if (ci->i_rd_ref)
 		used |= CEPH_CAP_FILE_RD;
-	if (ci->i_rdcache_ref || ci->i_rdcache_gen)
+	if (ci->i_rdcache_ref || ci->vfs_inode.i_data.nrpages)
 		used |= CEPH_CAP_FILE_CACHE;
 	if (ci->i_wr_ref)
 		used |= CEPH_CAP_FILE_WR;
@@ -1082,6 +1082,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	gid_t gid;
 	struct ceph_mds_session *session;
 	u64 xattr_version = 0;
+	struct ceph_buffer *xattr_blob = NULL;
 	int delayed = 0;
 	u64 flush_tid = 0;
 	int i;
@@ -1142,6 +1143,10 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 		for (i = 0; i < CEPH_CAP_BITS; i++)
 			if (flushing & (1 << i))
 				ci->i_cap_flush_tid[i] = flush_tid;
+
+		follows = ci->i_head_snapc->seq;
+	} else {
+		follows = 0;
 	}
 
 	keep = cap->implemented;
@@ -1155,14 +1160,14 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	mtime = inode->i_mtime;
 	atime = inode->i_atime;
 	time_warp_seq = ci->i_time_warp_seq;
-	follows = ci->i_snap_realm->cached_context->seq;
 	uid = inode->i_uid;
 	gid = inode->i_gid;
 	mode = inode->i_mode;
 
-	if (dropping & CEPH_CAP_XATTR_EXCL) {
+	if (flushing & CEPH_CAP_XATTR_EXCL) {
 		__ceph_build_xattrs_blob(ci);
-		xattr_version = ci->i_xattrs.version + 1;
+		xattr_blob = ci->i_xattrs.blob;
+		xattr_version = ci->i_xattrs.version;
 	}
 
 	spin_unlock(&inode->i_lock);
@@ -1170,9 +1175,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	ret = send_cap_msg(session, ceph_vino(inode).ino, cap_id,
 		op, keep, want, flushing, seq, flush_tid, issue_seq, mseq,
 		size, max_size, &mtime, &atime, time_warp_seq,
-		uid, gid, mode,
-		xattr_version,
-		(flushing & CEPH_CAP_XATTR_EXCL) ? ci->i_xattrs.blob : NULL,
+		uid, gid, mode, xattr_version, xattr_blob,
 		follows);
 	if (ret < 0) {
 		dout("error sending cap msg, must requeue %p\n", inode);
@@ -1192,10 +1195,14 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
  * asynchronously back to the MDS once sync writes complete and dirty
  * data is written out.
  *
+ * Unless @again is true, skip cap_snaps that were already sent to
+ * the MDS (i.e., during this session).
+ *
  * Called under i_lock.  Takes s_mutex as needed.
  */
 void __ceph_flush_snaps(struct ceph_inode_info *ci,
-			struct ceph_mds_session **psession)
+			struct ceph_mds_session **psession,
+			int again)
 		__releases(ci->vfs_inode->i_lock)
 		__acquires(ci->vfs_inode->i_lock)
 {
@@ -1224,7 +1231,7 @@ retry:
 		 * pages to be written out.
 		 */
 		if (capsnap->dirty_pages || capsnap->writing)
-			continue;
+			break;
 
 		/*
 		 * if cap writeback already occurred, we should have dropped
@@ -1237,6 +1244,13 @@ retry:
 			dout("no auth cap (migrating?), doing nothing\n");
 			goto out;
 		}
+
+		/* only flush each capsnap once */
+		if (!again && !list_empty(&capsnap->flushing_item)) {
+			dout("already flushed %p, skipping\n", capsnap);
+			continue;
+		}
+
 		mds = ci->i_auth_cap->session->s_mds;
 		mseq = ci->i_auth_cap->mseq;
 
@@ -1273,8 +1287,8 @@ retry:
 			      &session->s_cap_snaps_flushing);
 		spin_unlock(&inode->i_lock);
 
-		dout("flush_snaps %p cap_snap %p follows %lld size %llu\n",
-		     inode, capsnap, next_follows, capsnap->size);
+		dout("flush_snaps %p cap_snap %p follows %lld tid %llu\n",
+		     inode, capsnap, capsnap->follows, capsnap->flush_tid);
 		send_cap_msg(session, ceph_vino(inode).ino, 0,
 			     CEPH_CAP_OP_FLUSHSNAP, capsnap->issued, 0,
 			     capsnap->dirty, 0, capsnap->flush_tid, 0, mseq,
@@ -1282,7 +1296,7 @@ retry:
 			     &capsnap->mtime, &capsnap->atime,
 			     capsnap->time_warp_seq,
 			     capsnap->uid, capsnap->gid, capsnap->mode,
-			     0, NULL,
+			     capsnap->xattr_version, capsnap->xattr_blob,
 			     capsnap->follows);
 
 		next_follows = capsnap->follows + 1;
@@ -1311,7 +1325,7 @@ static void ceph_flush_snaps(struct ceph_inode_info *ci)
 	struct inode *inode = &ci->vfs_inode;
 
 	spin_lock(&inode->i_lock);
-	__ceph_flush_snaps(ci, NULL);
+	__ceph_flush_snaps(ci, NULL, 0);
 	spin_unlock(&inode->i_lock);
 }
 
@@ -1332,7 +1346,11 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 	     ceph_cap_string(was | mask));
 	ci->i_dirty_caps |= mask;
 	if (was == 0) {
-		dout(" inode %p now dirty\n", &ci->vfs_inode);
+		if (!ci->i_head_snapc)
+			ci->i_head_snapc = ceph_get_snap_context(
+				ci->i_snap_realm->cached_context);
+		dout(" inode %p now dirty snapc %p\n", &ci->vfs_inode,
+			ci->i_head_snapc);
 		BUG_ON(!list_empty(&ci->i_dirty_item));
 		spin_lock(&mdsc->cap_dirty_lock);
 		list_add(&ci->i_dirty_item, &mdsc->cap_dirty);
@@ -1470,7 +1488,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 
 	/* flush snaps first time around only */
 	if (!list_empty(&ci->i_cap_snaps))
-		__ceph_flush_snaps(ci, &session);
+		__ceph_flush_snaps(ci, &session, 0);
 	goto retry_locked;
 retry:
 	spin_lock(&inode->i_lock);
@@ -1887,7 +1905,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc,
 		if (cap && cap->session == session) {
 			dout("kick_flushing_caps %p cap %p capsnap %p\n", inode,
 			     cap, capsnap);
-			__ceph_flush_snaps(ci, &session);
+			__ceph_flush_snaps(ci, &session, 1);
 		} else {
 			pr_err("%p auth cap %p not mds%d ???\n", inode,
 			       cap, session->s_mds);
@@ -2190,7 +2208,9 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 
 	if (ci->i_head_snapc == snapc) {
 		ci->i_wrbuffer_ref_head -= nr;
-		if (!ci->i_wrbuffer_ref_head) {
+		if (ci->i_wrbuffer_ref_head == 0 &&
+		    ci->i_dirty_caps == 0 && ci->i_flushing_caps == 0) {
+			BUG_ON(!ci->i_head_snapc);
 			ceph_put_snap_context(ci->i_head_snapc);
 			ci->i_head_snapc = NULL;
 		}
@@ -2263,7 +2283,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int mds = session->s_mds;
-	int seq = le32_to_cpu(grant->seq);
+	unsigned seq = le32_to_cpu(grant->seq);
+	unsigned issue_seq = le32_to_cpu(grant->issue_seq);
 	int newcaps = le32_to_cpu(grant->caps);
 	int issued, implemented, used, wanted, dirty;
 	u64 size = le64_to_cpu(grant->size);
@@ -2275,8 +2296,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
 	int revoked_rdcache = 0;
 	int queue_invalidate = 0;
 
-	dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n",
-	     inode, cap, mds, seq, ceph_cap_string(newcaps));
+	dout("handle_cap_grant inode %p cap %p mds%d seq %u/%u %s\n",
+	     inode, cap, mds, seq, issue_seq, ceph_cap_string(newcaps));
 	dout(" size %llu max_size %llu, i_size %llu\n", size, max_size,
 		inode->i_size);
 
@@ -2372,6 +2393,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
 	}
 
 	cap->seq = seq;
+	cap->issue_seq = issue_seq;
 
 	/* file layout may have changed */
 	ci->i_layout = grant->layout;
@@ -2483,6 +2505,11 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
 			dout(" inode %p now clean\n", inode);
 			BUG_ON(!list_empty(&ci->i_dirty_item));
 			drop = 1;
+			if (ci->i_wrbuffer_ref_head == 0) {
+				BUG_ON(!ci->i_head_snapc);
+				ceph_put_snap_context(ci->i_head_snapc);
+				ci->i_head_snapc = NULL;
+			}
 		} else {
 			BUG_ON(list_empty(&ci->i_dirty_item));
 		}
@@ -2749,15 +2776,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		if (op == CEPH_CAP_OP_IMPORT)
 			__queue_cap_release(session, vino.ino, cap_id,
 					    mseq, seq);
-
-		/*
-		 * send any full release message to try to move things
-		 * along for the mds (who clearly thinks we still have this
-		 * cap).
-		 */
-		ceph_add_cap_releases(mdsc, session);
-		ceph_send_cap_releases(mdsc, session);
-		goto done;
+		goto flush_cap_releases;
 	}
 
 	/* these will work even if we don't have a cap yet */
@@ -2785,7 +2804,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		dout(" no cap on %p ino %llx.%llx from mds%d\n",
 		     inode, ceph_ino(inode), ceph_snap(inode), mds);
 		spin_unlock(&inode->i_lock);
-		goto done;
+		goto flush_cap_releases;
 	}
 
 	/* note that each of these drops i_lock for us */
@@ -2809,6 +2828,17 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		       ceph_cap_op_name(op));
 	}
 
+	goto done;
+
+flush_cap_releases:
+	/*
+	 * send any full release message to try to move things
+	 * along for the mds (who clearly thinks we still have this
+	 * cap).
+	 */
+	ceph_add_cap_releases(mdsc, session);
+	ceph_send_cap_releases(mdsc, session);
+
 done:
 	mutex_unlock(&session->s_mutex);
 done_unlocked:
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 360c4f22718d..6fd8b20a8611 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -171,6 +171,8 @@ static int mdsc_show(struct seq_file *s, void *p)
 		} else if (req->r_dentry) {
 			path = ceph_mdsc_build_path(req->r_dentry, &pathlen,
 						    &pathbase, 0);
+			if (IS_ERR(path))
+				path = NULL;
 			spin_lock(&req->r_dentry->d_lock);
 			seq_printf(s, " #%llx/%.*s (%s)",
 				   ceph_ino(req->r_dentry->d_parent->d_inode),
@@ -187,6 +189,8 @@ static int mdsc_show(struct seq_file *s, void *p)
 		if (req->r_old_dentry) {
 			path = ceph_mdsc_build_path(req->r_old_dentry, &pathlen,
 						    &pathbase, 0);
+			if (IS_ERR(path))
+				path = NULL;
 			spin_lock(&req->r_old_dentry->d_lock);
 			seq_printf(s, " #%llx/%.*s (%s)",
 			   ceph_ino(req->r_old_dentry->d_parent->d_inode),
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 67bbb41d5526..a1986eb52045 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -46,7 +46,7 @@ int ceph_init_dentry(struct dentry *dentry)
 	else
 		dentry->d_op = &ceph_snap_dentry_ops;
 
-	di = kmem_cache_alloc(ceph_dentry_cachep, GFP_NOFS);
+	di = kmem_cache_alloc(ceph_dentry_cachep, GFP_NOFS | __GFP_ZERO);
 	if (!di)
 		return -ENOMEM;          /* oh well */
 
@@ -1021,11 +1021,15 @@ out_touch:
 static void ceph_dentry_release(struct dentry *dentry)
 {
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
-	struct inode *parent_inode = dentry->d_parent->d_inode;
-	u64 snapid = ceph_snap(parent_inode);
+	struct inode *parent_inode = NULL;
+	u64 snapid = CEPH_NOSNAP;
 
+	if (!IS_ROOT(dentry)) {
+		parent_inode = dentry->d_parent->d_inode;
+		if (parent_inode)
+			snapid = ceph_snap(parent_inode);
+	}
 	dout("dentry_release %p parent %p\n", dentry, parent_inode);
-
 	if (parent_inode && snapid != CEPH_SNAPDIR) {
 		struct ceph_inode_info *ci = ceph_inode(parent_inode);
 
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index 4480cb1c63e7..e38423e82f2e 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -42,32 +42,37 @@ struct ceph_nfs_confh {
 static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
 			  int connectable)
 {
+	int type;
 	struct ceph_nfs_fh *fh = (void *)rawfh;
 	struct ceph_nfs_confh *cfh = (void *)rawfh;
 	struct dentry *parent = dentry->d_parent;
 	struct inode *inode = dentry->d_inode;
-	int type;
+	int connected_handle_length = sizeof(*cfh)/4;
+	int handle_length = sizeof(*fh)/4;
 
 	/* don't re-export snaps */
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EINVAL;
 
-	if (*max_len >= sizeof(*cfh)) {
+	if (*max_len >= connected_handle_length) {
 		dout("encode_fh %p connectable\n", dentry);
 		cfh->ino = ceph_ino(dentry->d_inode);
 		cfh->parent_ino = ceph_ino(parent->d_inode);
 		cfh->parent_name_hash = parent->d_name.hash;
-		*max_len = sizeof(*cfh);
+		*max_len = connected_handle_length;
 		type = 2;
-	} else if (*max_len > sizeof(*fh)) {
-		if (connectable)
-			return -ENOSPC;
+	} else if (*max_len >= handle_length) {
+		if (connectable) {
+			*max_len = connected_handle_length;
+			return 255;
+		}
 		dout("encode_fh %p\n", dentry);
 		fh->ino = ceph_ino(dentry->d_inode);
-		*max_len = sizeof(*fh);
+		*max_len = handle_length;
 		type = 1;
 	} else {
-		return -ENOSPC;
+		*max_len = handle_length;
+		return 255;
 	}
 	return type;
 }
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 8c044a4f0457..66e4da6dba22 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -697,7 +697,7 @@ more:
 			 * start_request so that a tid has been assigned.
 			 */
 			spin_lock(&ci->i_unsafe_lock);
-			list_add(&ci->i_unsafe_writes, &req->r_unsafe_item);
+			list_add(&req->r_unsafe_item, &ci->i_unsafe_writes);
 			spin_unlock(&ci->i_unsafe_lock);
 			ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
 		}
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 5d893d31e399..62377ec37edf 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -677,6 +677,7 @@ static int fill_inode(struct inode *inode,
 		if (ci->i_files == 0 && ci->i_subdirs == 0 &&
 		    ceph_snap(inode) == CEPH_NOSNAP &&
 		    (le32_to_cpu(info->cap.caps) & CEPH_CAP_FILE_SHARED) &&
+		    (issued & CEPH_CAP_FILE_EXCL) == 0 &&
 		    (ci->i_ceph_flags & CEPH_I_COMPLETE) == 0) {
 			dout(" marking %p complete (empty)\n", inode);
 			ci->i_ceph_flags |= CEPH_I_COMPLETE;
@@ -844,7 +845,7 @@ static void ceph_set_dentry_offset(struct dentry *dn)
  * the caller) if we fail.
  */
 static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
-				    bool *prehash)
+				    bool *prehash, bool set_offset)
 {
 	struct dentry *realdn;
 
@@ -876,7 +877,8 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
 	}
 	if ((!prehash || *prehash) && d_unhashed(dn))
 		d_rehash(dn);
-	ceph_set_dentry_offset(dn);
+	if (set_offset)
+		ceph_set_dentry_offset(dn);
 out:
 	return dn;
 }
@@ -1061,7 +1063,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 				d_delete(dn);
 				goto done;
 			}
-			dn = splice_dentry(dn, in, &have_lease);
+			dn = splice_dentry(dn, in, &have_lease, true);
 			if (IS_ERR(dn)) {
 				err = PTR_ERR(dn);
 				goto done;
@@ -1104,7 +1106,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 			goto done;
 		}
 		dout(" linking snapped dir %p to dn %p\n", in, dn);
-		dn = splice_dentry(dn, in, NULL);
+		dn = splice_dentry(dn, in, NULL, true);
 		if (IS_ERR(dn)) {
 			err = PTR_ERR(dn);
 			goto done;
@@ -1229,14 +1231,14 @@ retry_lookup:
 			in = dn->d_inode;
 		} else {
 			in = ceph_get_inode(parent->d_sb, vino);
-			if (in == NULL) {
+			if (IS_ERR(in)) {
 				dout("new_inode badness\n");
 				d_delete(dn);
 				dput(dn);
-				err = -ENOMEM;
+				err = PTR_ERR(in);
 				goto out;
 			}
-			dn = splice_dentry(dn, in, NULL);
+			dn = splice_dentry(dn, in, NULL, false);
 			if (IS_ERR(dn))
 				dn = NULL;
 		}
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index ae85af06454f..ff4e753aae92 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -82,7 +82,8 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 		length = fl->fl_end - fl->fl_start + 1;
 
 	err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
-				(u64)fl->fl_pid, (u64)fl->fl_nspid,
+				(u64)fl->fl_pid,
+				(u64)(unsigned long)fl->fl_nspid,
 				lock_cmd, fl->fl_start,
 				length, wait);
 	if (!err) {
@@ -92,7 +93,8 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 			/* undo! This should only happen if the kernel detects
 			 * local deadlock. */
 			ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
-					  (u64)fl->fl_pid, (u64)fl->fl_nspid,
+					  (u64)fl->fl_pid,
+					  (u64)(unsigned long)fl->fl_nspid,
 					  CEPH_LOCK_UNLOCK, fl->fl_start,
 					  length, 0);
 			dout("got %d on posix_lock_file, undid lock", err);
@@ -132,7 +134,8 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
 		length = fl->fl_end - fl->fl_start + 1;
 
 	err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK,
-				file, (u64)fl->fl_pid, (u64)fl->fl_nspid,
+				file, (u64)fl->fl_pid,
+				(u64)(unsigned long)fl->fl_nspid,
 				lock_cmd, fl->fl_start,
 				length, wait);
 	if (!err) {
@@ -141,7 +144,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
 			ceph_lock_message(CEPH_LOCK_FLOCK,
 					  CEPH_MDS_OP_SETFILELOCK,
 					  file, (u64)fl->fl_pid,
-					  (u64)fl->fl_nspid,
+					  (u64)(unsigned long)fl->fl_nspid,
 					  CEPH_LOCK_UNLOCK, fl->fl_start,
 					  length, 0);
 			dout("got %d on flock_lock_file_wait, undid lock", err);
@@ -235,7 +238,8 @@ int lock_to_ceph_filelock(struct file_lock *lock,
 	cephlock->length = cpu_to_le64(lock->fl_end - lock->fl_start + 1);
 	cephlock->client = cpu_to_le64(0);
 	cephlock->pid = cpu_to_le64(lock->fl_pid);
-	cephlock->pid_namespace = cpu_to_le64((u64)lock->fl_nspid);
+	cephlock->pid_namespace =
+	        cpu_to_le64((u64)(unsigned long)lock->fl_nspid);
 
 	switch (lock->fl_type) {
 	case F_RDLCK:
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index a75ddbf9fe37..fad95f8f2608 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -560,6 +560,13 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
  *
  * Called under mdsc->mutex.
  */
+struct dentry *get_nonsnap_parent(struct dentry *dentry)
+{
+	while (!IS_ROOT(dentry) && ceph_snap(dentry->d_inode) != CEPH_NOSNAP)
+		dentry = dentry->d_parent;
+	return dentry;
+}
+
 static int __choose_mds(struct ceph_mds_client *mdsc,
 			struct ceph_mds_request *req)
 {
@@ -590,14 +597,29 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 	if (req->r_inode) {
 		inode = req->r_inode;
 	} else if (req->r_dentry) {
-		if (req->r_dentry->d_inode) {
+		struct inode *dir = req->r_dentry->d_parent->d_inode;
+
+		if (dir->i_sb != mdsc->client->sb) {
+			/* not this fs! */
+			inode = req->r_dentry->d_inode;
+		} else if (ceph_snap(dir) != CEPH_NOSNAP) {
+			/* direct snapped/virtual snapdir requests
+			 * based on parent dir inode */
+			struct dentry *dn =
+				get_nonsnap_parent(req->r_dentry->d_parent);
+			inode = dn->d_inode;
+			dout("__choose_mds using nonsnap parent %p\n", inode);
+		} else if (req->r_dentry->d_inode) {
+			/* dentry target */
 			inode = req->r_dentry->d_inode;
 		} else {
-			inode = req->r_dentry->d_parent->d_inode;
+			/* dir + name */
+			inode = dir;
 			hash = req->r_dentry->d_name.hash;
 			is_hash = true;
 		}
 	}
+
 	dout("__choose_mds %p is_hash=%d (%d) mode %d\n", inode, (int)is_hash,
 	     (int)hash, mode);
 	if (!inode)
@@ -2208,7 +2230,7 @@ static void handle_session(struct ceph_mds_session *session,
 			pr_info("mds%d reconnect denied\n", session->s_mds);
 		remove_session_caps(session);
 		wake = 1; /* for good measure */
-		complete_all(&mdsc->session_close_waiters);
+		wake_up_all(&mdsc->session_close_wq);
 		kick_requests(mdsc, mds);
 		break;
 
@@ -2302,7 +2324,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 		path = ceph_mdsc_build_path(dentry, &pathlen, &pathbase, 0);
 		if (IS_ERR(path)) {
 			err = PTR_ERR(path);
-			BUG_ON(err);
+			goto out_dput;
 		}
 	} else {
 		path = NULL;
@@ -2310,7 +2332,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 	}
 	err = ceph_pagelist_encode_string(pagelist, path, pathlen);
 	if (err)
-		goto out;
+		goto out_free;
 
 	spin_lock(&inode->i_lock);
 	cap->seq = 0;        /* reset cap seq */
@@ -2352,10 +2374,13 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 						num_fcntl_locks,
 						num_flock_locks);
 		unlock_kernel();
+	} else {
+		err = ceph_pagelist_append(pagelist, &rec, reclen);
 	}
 
-out:
+out_free:
 	kfree(path);
+out_dput:
 	dput(dentry);
 	return err;
 }
@@ -2876,7 +2901,7 @@ int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 		return -ENOMEM;
 
 	init_completion(&mdsc->safe_umount_waiters);
-	init_completion(&mdsc->session_close_waiters);
+	init_waitqueue_head(&mdsc->session_close_wq);
 	INIT_LIST_HEAD(&mdsc->waiting_for_map);
 	mdsc->sessions = NULL;
 	mdsc->max_sessions = 0;
@@ -3021,6 +3046,23 @@ void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
 	wait_event(mdsc->cap_flushing_wq, check_cap_flush(mdsc, want_flush));
 }
 
+/*
+ * true if all sessions are closed, or we force unmount
+ */
+bool done_closing_sessions(struct ceph_mds_client *mdsc)
+{
+	int i, n = 0;
+
+	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+		return true;
+
+	mutex_lock(&mdsc->mutex);
+	for (i = 0; i < mdsc->max_sessions; i++)
+		if (mdsc->sessions[i])
+			n++;
+	mutex_unlock(&mdsc->mutex);
+	return n == 0;
+}
 
 /*
  * called after sb is ro.
@@ -3029,45 +3071,32 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_session *session;
 	int i;
-	int n;
 	struct ceph_client *client = mdsc->client;
-	unsigned long started, timeout = client->mount_args->mount_timeout * HZ;
+	unsigned long timeout = client->mount_args->mount_timeout * HZ;
 
 	dout("close_sessions\n");
 
-	mutex_lock(&mdsc->mutex);
-
 	/* close sessions */
-	started = jiffies;
-	while (time_before(jiffies, started + timeout)) {
-		dout("closing sessions\n");
-		n = 0;
-		for (i = 0; i < mdsc->max_sessions; i++) {
-			session = __ceph_lookup_mds_session(mdsc, i);
-			if (!session)
-				continue;
-			mutex_unlock(&mdsc->mutex);
-			mutex_lock(&session->s_mutex);
-			__close_session(mdsc, session);
-			mutex_unlock(&session->s_mutex);
-			ceph_put_mds_session(session);
-			mutex_lock(&mdsc->mutex);
-			n++;
-		}
-		if (n == 0)
-			break;
-
-		if (client->mount_state == CEPH_MOUNT_SHUTDOWN)
-			break;
-
-		dout("waiting for sessions to close\n");
+	mutex_lock(&mdsc->mutex);
+	for (i = 0; i < mdsc->max_sessions; i++) {
+		session = __ceph_lookup_mds_session(mdsc, i);
+		if (!session)
+			continue;
 		mutex_unlock(&mdsc->mutex);
-		wait_for_completion_timeout(&mdsc->session_close_waiters,
-					    timeout);
+		mutex_lock(&session->s_mutex);
+		__close_session(mdsc, session);
+		mutex_unlock(&session->s_mutex);
+		ceph_put_mds_session(session);
 		mutex_lock(&mdsc->mutex);
 	}
+	mutex_unlock(&mdsc->mutex);
+
+	dout("waiting for sessions to close\n");
+	wait_event_timeout(mdsc->session_close_wq, done_closing_sessions(mdsc),
+			   timeout);
 
 	/* tear down remaining sessions */
+	mutex_lock(&mdsc->mutex);
 	for (i = 0; i < mdsc->max_sessions; i++) {
 		if (mdsc->sessions[i]) {
 			session = get_session(mdsc->sessions[i]);
@@ -3080,9 +3109,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 			mutex_lock(&mdsc->mutex);
 		}
 	}
-
 	WARN_ON(!list_empty(&mdsc->cap_delay_list));
-
 	mutex_unlock(&mdsc->mutex);
 
 	ceph_cleanup_empty_realms(mdsc);
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index ab7e89f5e344..c98267ce6d2a 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -234,7 +234,8 @@ struct ceph_mds_client {
 	struct mutex            mutex;         /* all nested structures */
 
 	struct ceph_mdsmap      *mdsmap;
-	struct completion       safe_umount_waiters, session_close_waiters;
+	struct completion       safe_umount_waiters;
+	wait_queue_head_t       session_close_wq;
 	struct list_head        waiting_for_map;
 
 	struct ceph_mds_session **sessions;    /* NULL for mds if no session */
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
index bed6391e52c7..3b5571b8ce22 100644
--- a/fs/ceph/osd_client.c
+++ b/fs/ceph/osd_client.c
@@ -549,7 +549,7 @@ static void __unregister_request(struct ceph_osd_client *osdc,
  */
 static void __cancel_request(struct ceph_osd_request *req)
 {
-	if (req->r_sent) {
+	if (req->r_sent && req->r_osd) {
 		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
 		req->r_sent = 0;
 	}
@@ -661,7 +661,7 @@ static int __send_request(struct ceph_osd_client *osdc,
 	reqhead->reassert_version = req->r_reassert_version;
 
 	req->r_stamp = jiffies;
-	list_move_tail(&osdc->req_lru, &req->r_req_lru_item);
+	list_move_tail(&req->r_req_lru_item, &osdc->req_lru);
 
 	ceph_msg_get(req->r_request); /* send consumes a ref */
 	ceph_con_send(&req->r_osd->o_con, req->r_request);
diff --git a/fs/ceph/pagelist.c b/fs/ceph/pagelist.c
index b6859f47d364..46a368b6dce5 100644
--- a/fs/ceph/pagelist.c
+++ b/fs/ceph/pagelist.c
@@ -5,10 +5,18 @@
 
 #include "pagelist.h"
 
+static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
+{
+	struct page *page = list_entry(pl->head.prev, struct page,
+				       lru);
+	kunmap(page);
+}
+
 int ceph_pagelist_release(struct ceph_pagelist *pl)
 {
 	if (pl->mapped_tail)
-		kunmap(pl->mapped_tail);
+		ceph_pagelist_unmap_tail(pl);
+
 	while (!list_empty(&pl->head)) {
 		struct page *page = list_first_entry(&pl->head, struct page,
 						     lru);
@@ -26,7 +34,7 @@ static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
 	pl->room += PAGE_SIZE;
 	list_add_tail(&page->lru, &pl->head);
 	if (pl->mapped_tail)
-		kunmap(pl->mapped_tail);
+		ceph_pagelist_unmap_tail(pl);
 	pl->mapped_tail = kmap(page);
 	return 0;
 }
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index c0b26b6badba..190b6c4a6f2b 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -119,6 +119,7 @@ static struct ceph_snap_realm *ceph_create_snap_realm(
 	INIT_LIST_HEAD(&realm->children);
 	INIT_LIST_HEAD(&realm->child_item);
 	INIT_LIST_HEAD(&realm->empty_item);
+	INIT_LIST_HEAD(&realm->dirty_item);
 	INIT_LIST_HEAD(&realm->inodes_with_caps);
 	spin_lock_init(&realm->inodes_with_caps_lock);
 	__insert_snap_realm(&mdsc->snap_realms, realm);
@@ -435,7 +436,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
 {
 	struct inode *inode = &ci->vfs_inode;
 	struct ceph_cap_snap *capsnap;
-	int used;
+	int used, dirty;
 
 	capsnap = kzalloc(sizeof(*capsnap), GFP_NOFS);
 	if (!capsnap) {
@@ -445,6 +446,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
 
 	spin_lock(&inode->i_lock);
 	used = __ceph_caps_used(ci);
+	dirty = __ceph_caps_dirty(ci);
 	if (__ceph_have_pending_cap_snap(ci)) {
 		/* there is no point in queuing multiple "pending" cap_snaps,
 		   as no new writes are allowed to start when pending, so any
@@ -452,27 +454,37 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
 		   cap_snap.  lucky us. */
 		dout("queue_cap_snap %p already pending\n", inode);
 		kfree(capsnap);
-	} else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR)) {
+	} else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR) ||
+		   (dirty & (CEPH_CAP_AUTH_EXCL|CEPH_CAP_XATTR_EXCL|
+			     CEPH_CAP_FILE_EXCL|CEPH_CAP_FILE_WR))) {
 		struct ceph_snap_context *snapc = ci->i_head_snapc;
 
+		dout("queue_cap_snap %p cap_snap %p queuing under %p\n", inode,
+		     capsnap, snapc);
 		igrab(inode);
-
+		
 		atomic_set(&capsnap->nref, 1);
 		capsnap->ci = ci;
 		INIT_LIST_HEAD(&capsnap->ci_item);
 		INIT_LIST_HEAD(&capsnap->flushing_item);
 
-		capsnap->follows = snapc->seq - 1;
+		capsnap->follows = snapc->seq;
 		capsnap->issued = __ceph_caps_issued(ci, NULL);
-		capsnap->dirty = __ceph_caps_dirty(ci);
+		capsnap->dirty = dirty;
 
 		capsnap->mode = inode->i_mode;
 		capsnap->uid = inode->i_uid;
 		capsnap->gid = inode->i_gid;
 
-		/* fixme? */
-		capsnap->xattr_blob = NULL;
-		capsnap->xattr_len = 0;
+		if (dirty & CEPH_CAP_XATTR_EXCL) {
+			__ceph_build_xattrs_blob(ci);
+			capsnap->xattr_blob =
+				ceph_buffer_get(ci->i_xattrs.blob);
+			capsnap->xattr_version = ci->i_xattrs.version;
+		} else {
+			capsnap->xattr_blob = NULL;
+			capsnap->xattr_version = 0;
+		}
 
 		/* dirty page count moved from _head to this cap_snap;
 		   all subsequent writes page dirties occur _after_ this
@@ -480,7 +492,9 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
 		capsnap->dirty_pages = ci->i_wrbuffer_ref_head;
 		ci->i_wrbuffer_ref_head = 0;
 		capsnap->context = snapc;
-		ci->i_head_snapc = NULL;
+		ci->i_head_snapc =
+			ceph_get_snap_context(ci->i_snap_realm->cached_context);
+		dout(" new snapc is %p\n", ci->i_head_snapc);
 		list_add_tail(&capsnap->ci_item, &ci->i_cap_snaps);
 
 		if (used & CEPH_CAP_FILE_WR) {
@@ -539,6 +553,41 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 	return 1;  /* caller may want to ceph_flush_snaps */
 }
 
+/*
+ * Queue cap_snaps for snap writeback for this realm and its children.
+ * Called under snap_rwsem, so realm topology won't change.
+ */
+static void queue_realm_cap_snaps(struct ceph_snap_realm *realm)
+{
+	struct ceph_inode_info *ci;
+	struct inode *lastinode = NULL;
+	struct ceph_snap_realm *child;
+
+	dout("queue_realm_cap_snaps %p %llx inodes\n", realm, realm->ino);
+
+	spin_lock(&realm->inodes_with_caps_lock);
+	list_for_each_entry(ci, &realm->inodes_with_caps,
+			    i_snap_realm_item) {
+		struct inode *inode = igrab(&ci->vfs_inode);
+		if (!inode)
+			continue;
+		spin_unlock(&realm->inodes_with_caps_lock);
+		if (lastinode)
+			iput(lastinode);
+		lastinode = inode;
+		ceph_queue_cap_snap(ci);
+		spin_lock(&realm->inodes_with_caps_lock);
+	}
+	spin_unlock(&realm->inodes_with_caps_lock);
+	if (lastinode)
+		iput(lastinode);
+
+	dout("queue_realm_cap_snaps %p %llx children\n", realm, realm->ino);
+	list_for_each_entry(child, &realm->children, child_item)
+		queue_realm_cap_snaps(child);
+
+	dout("queue_realm_cap_snaps %p %llx done\n", realm, realm->ino);
+}
 
 /*
  * Parse and apply a snapblob "snap trace" from the MDS.  This specifies
@@ -556,6 +605,7 @@ int ceph_update_snap_trace(struct ceph_mds_client *mdsc,
 	struct ceph_snap_realm *realm;
 	int invalidate = 0;
 	int err = -ENOMEM;
+	LIST_HEAD(dirty_realms);
 
 	dout("update_snap_trace deletion=%d\n", deletion);
 more:
@@ -578,45 +628,6 @@ more:
 		}
 	}
 
-	if (le64_to_cpu(ri->seq) > realm->seq) {
-		dout("update_snap_trace updating %llx %p %lld -> %lld\n",
-		     realm->ino, realm, realm->seq, le64_to_cpu(ri->seq));
-		/*
-		 * if the realm seq has changed, queue a cap_snap for every
-		 * inode with open caps.  we do this _before_ we update
-		 * the realm info so that we prepare for writeback under the
-		 * _previous_ snap context.
-		 *
-		 * ...unless it's a snap deletion!
-		 */
-		if (!deletion) {
-			struct ceph_inode_info *ci;
-			struct inode *lastinode = NULL;
-
-			spin_lock(&realm->inodes_with_caps_lock);
-			list_for_each_entry(ci, &realm->inodes_with_caps,
-					    i_snap_realm_item) {
-				struct inode *inode = igrab(&ci->vfs_inode);
-				if (!inode)
-					continue;
-				spin_unlock(&realm->inodes_with_caps_lock);
-				if (lastinode)
-					iput(lastinode);
-				lastinode = inode;
-				ceph_queue_cap_snap(ci);
-				spin_lock(&realm->inodes_with_caps_lock);
-			}
-			spin_unlock(&realm->inodes_with_caps_lock);
-			if (lastinode)
-				iput(lastinode);
-			dout("update_snap_trace cap_snaps queued\n");
-		}
-
-	} else {
-		dout("update_snap_trace %llx %p seq %lld unchanged\n",
-		     realm->ino, realm, realm->seq);
-	}
-
 	/* ensure the parent is correct */
 	err = adjust_snap_realm_parent(mdsc, realm, le64_to_cpu(ri->parent));
 	if (err < 0)
@@ -624,6 +635,8 @@ more:
 	invalidate += err;
 
 	if (le64_to_cpu(ri->seq) > realm->seq) {
+		dout("update_snap_trace updating %llx %p %lld -> %lld\n",
+		     realm->ino, realm, realm->seq, le64_to_cpu(ri->seq));
 		/* update realm parameters, snap lists */
 		realm->seq = le64_to_cpu(ri->seq);
 		realm->created = le64_to_cpu(ri->created);
@@ -641,9 +654,17 @@ more:
 		if (err < 0)
 			goto fail;
 
+		/* queue realm for cap_snap creation */
+		list_add(&realm->dirty_item, &dirty_realms);
+
 		invalidate = 1;
 	} else if (!realm->cached_context) {
+		dout("update_snap_trace %llx %p seq %lld new\n",
+		     realm->ino, realm, realm->seq);
 		invalidate = 1;
+	} else {
+		dout("update_snap_trace %llx %p seq %lld unchanged\n",
+		     realm->ino, realm, realm->seq);
 	}
 
 	dout("done with %llx %p, invalidated=%d, %p %p\n", realm->ino,
@@ -656,6 +677,14 @@ more:
 	if (invalidate)
 		rebuild_snap_realms(realm);
 
+	/*
+	 * queue cap snaps _after_ we've built the new snap contexts,
+	 * so that i_head_snapc can be set appropriately.
+	 */
+	list_for_each_entry(realm, &dirty_realms, dirty_item) {
+		queue_realm_cap_snaps(realm);
+	}
+
 	__cleanup_empty_realms(mdsc);
 	return 0;
 
@@ -688,7 +717,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc)
 		igrab(inode);
 		spin_unlock(&mdsc->snap_flush_lock);
 		spin_lock(&inode->i_lock);
-		__ceph_flush_snaps(ci, &session);
+		__ceph_flush_snaps(ci, &session, 0);
 		spin_unlock(&inode->i_lock);
 		iput(inode);
 		spin_lock(&mdsc->snap_flush_lock);
@@ -789,6 +818,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
 			};
 			struct inode *inode = ceph_find_inode(sb, vino);
 			struct ceph_inode_info *ci;
+			struct ceph_snap_realm *oldrealm;
 
 			if (!inode)
 				continue;
@@ -814,18 +844,19 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
 			dout(" will move %p to split realm %llx %p\n",
 			     inode, realm->ino, realm);
 			/*
-			 * Remove the inode from the realm's inode
-			 * list, but don't add it to the new realm
-			 * yet.  We don't want the cap_snap to be
-			 * queued (again) by ceph_update_snap_trace()
-			 * below.  Queue it _now_, under the old context.
+			 * Move the inode to the new realm
 			 */
 			spin_lock(&realm->inodes_with_caps_lock);
 			list_del_init(&ci->i_snap_realm_item);
+			list_add(&ci->i_snap_realm_item,
+				 &realm->inodes_with_caps);
+			oldrealm = ci->i_snap_realm;
+			ci->i_snap_realm = realm;
 			spin_unlock(&realm->inodes_with_caps_lock);
 			spin_unlock(&inode->i_lock);
 
-			ceph_queue_cap_snap(ci);
+			ceph_get_snap_realm(mdsc, realm);
+			ceph_put_snap_realm(mdsc, oldrealm);
 
 			iput(inode);
 			continue;
@@ -853,43 +884,9 @@ skip_inode:
 	ceph_update_snap_trace(mdsc, p, e,
 			       op == CEPH_SNAP_OP_DESTROY);
 
-	if (op == CEPH_SNAP_OP_SPLIT) {
-		/*
-		 * ok, _now_ add the inodes into the new realm.
-		 */
-		for (i = 0; i < num_split_inos; i++) {
-			struct ceph_vino vino = {
-				.ino = le64_to_cpu(split_inos[i]),
-				.snap = CEPH_NOSNAP,
-			};
-			struct inode *inode = ceph_find_inode(sb, vino);
-			struct ceph_inode_info *ci;
-
-			if (!inode)
-				continue;
-			ci = ceph_inode(inode);
-			spin_lock(&inode->i_lock);
-			if (list_empty(&ci->i_snap_realm_item)) {
-				struct ceph_snap_realm *oldrealm =
-					ci->i_snap_realm;
-
-				dout(" moving %p to split realm %llx %p\n",
-				     inode, realm->ino, realm);
-				spin_lock(&realm->inodes_with_caps_lock);
-				list_add(&ci->i_snap_realm_item,
-					 &realm->inodes_with_caps);
-				ci->i_snap_realm = realm;
-				spin_unlock(&realm->inodes_with_caps_lock);
-				ceph_get_snap_realm(mdsc, realm);
-				ceph_put_snap_realm(mdsc, oldrealm);
-			}
-			spin_unlock(&inode->i_lock);
-			iput(inode);
-		}
-
+	if (op == CEPH_SNAP_OP_SPLIT)
 		/* we took a reference when we created the realm, above */
 		ceph_put_snap_realm(mdsc, realm);
-	}
 
 	__cleanup_empty_realms(mdsc);
 
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 2482d696f0de..b87638e84c4b 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -216,8 +216,7 @@ struct ceph_cap_snap {
 	uid_t uid;
 	gid_t gid;
 
-	void *xattr_blob;
-	int xattr_len;
+	struct ceph_buffer *xattr_blob;
 	u64 xattr_version;
 
 	u64 size;
@@ -229,8 +228,11 @@ struct ceph_cap_snap {
 
 static inline void ceph_put_cap_snap(struct ceph_cap_snap *capsnap)
 {
-	if (atomic_dec_and_test(&capsnap->nref))
+	if (atomic_dec_and_test(&capsnap->nref)) {
+		if (capsnap->xattr_blob)
+			ceph_buffer_put(capsnap->xattr_blob);
 		kfree(capsnap);
+	}
 }
 
 /*
@@ -342,7 +344,8 @@ struct ceph_inode_info {
 	unsigned i_cap_exporting_issued;
 	struct ceph_cap_reservation i_cap_migration_resv;
 	struct list_head i_cap_snaps;   /* snapped state pending flush to mds */
-	struct ceph_snap_context *i_head_snapc;  /* set if wr_buffer_head > 0 */
+	struct ceph_snap_context *i_head_snapc;  /* set if wr_buffer_head > 0 or
+						    dirty|flushing caps */
 	unsigned i_snap_caps;           /* cap bits for snapped files */
 
 	int i_nr_by_mode[CEPH_FILE_MODE_NUM];  /* open file counts */
@@ -687,6 +690,8 @@ struct ceph_snap_realm {
 
 	struct list_head empty_item;     /* if i have ref==0 */
 
+	struct list_head dirty_item;     /* if realm needs new context */
+
 	/* the current set of snaps for this realm */
 	struct ceph_snap_context *cached_context;
 
@@ -823,7 +828,8 @@ extern void ceph_put_cap_refs(struct ceph_inode_info *ci, int had);
 extern void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 				       struct ceph_snap_context *snapc);
 extern void __ceph_flush_snaps(struct ceph_inode_info *ci,
-			       struct ceph_mds_session **psession);
+			       struct ceph_mds_session **psession,
+			       int again);
 extern void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 			    struct ceph_mds_session *session);
 extern void ceph_check_delayed_caps(struct ceph_mds_client *mdsc);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 097a2654c00f..9578af610b73 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -485,6 +485,7 @@ void __ceph_build_xattrs_blob(struct ceph_inode_info *ci)
 		ci->i_xattrs.blob = ci->i_xattrs.prealloc_blob;
 		ci->i_xattrs.prealloc_blob = NULL;
 		ci->i_xattrs.dirty = false;
+		ci->i_xattrs.version++;
 	}
 }
 
diff --git a/fs/char_dev.c b/fs/char_dev.c
index f80a4f25123c..143d393881cb 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -40,7 +40,9 @@ struct backing_dev_info directly_mappable_cdev_bdi = {
 #endif
 		/* permit direct mmap, for read, write or exec */
 		BDI_CAP_MAP_DIRECT |
-		BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP),
+		BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP |
+		/* no writeback happens */
+		BDI_CAP_NO_ACCT_AND_WRITEBACK),
 };
 
 static struct kobj_map *cdev_map;
diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h
index 650638275a6f..7fe6b52df507 100644
--- a/fs/cifs/cifs_unicode.h
+++ b/fs/cifs/cifs_unicode.h
@@ -30,6 +30,8 @@
  *     This is a compressed table of upper and lower case conversion.
  *
  */
+#ifndef _CIFS_UNICODE_H
+#define _CIFS_UNICODE_H
 
 #include <asm/byteorder.h>
 #include <linux/types.h>
@@ -67,8 +69,8 @@ extern const struct UniCaseRange CifsUniUpperRange[];
 #endif				/* UNIUPR_NOUPPER */
 
 #ifndef UNIUPR_NOLOWER
-extern signed char UniLowerTable[512];
-extern struct UniCaseRange UniLowerRange[];
+extern signed char CifsUniLowerTable[512];
+extern const struct UniCaseRange CifsUniLowerRange[];
 #endif				/* UNIUPR_NOLOWER */
 
 #ifdef __KERNEL__
@@ -337,15 +339,15 @@ UniStrupr(register wchar_t *upin)
  * UniTolower:  Convert a unicode character to lower case
  */
 static inline wchar_t
-UniTolower(wchar_t uc)
+UniTolower(register wchar_t uc)
 {
-	register struct UniCaseRange *rp;
+	register const struct UniCaseRange *rp;
 
-	if (uc < sizeof(UniLowerTable)) {
+	if (uc < sizeof(CifsUniLowerTable)) {
 		/* Latin characters */
-		return uc + UniLowerTable[uc];	/* Use base tables */
+		return uc + CifsUniLowerTable[uc];	/* Use base tables */
 	} else {
-		rp = UniLowerRange;	/* Use range tables */
+		rp = CifsUniLowerRange;	/* Use range tables */
 		while (rp->start) {
 			if (uc < rp->start)	/* Before start of range */
 				return uc;	/* Uppercase = input */
@@ -374,3 +376,5 @@ UniStrlwr(register wchar_t *upin)
 }
 
 #endif
+
+#endif /* _CIFS_UNICODE_H */
diff --git a/fs/cifs/cifs_uniupr.h b/fs/cifs/cifs_uniupr.h
index 18a9d978e519..0ac7c5a8633a 100644
--- a/fs/cifs/cifs_uniupr.h
+++ b/fs/cifs/cifs_uniupr.h
@@ -140,7 +140,7 @@ const struct UniCaseRange CifsUniUpperRange[] = {
 /*
  * Latin lower case
  */
-static signed char CifsUniLowerTable[512] = {
+signed char CifsUniLowerTable[512] = {
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,	/* 000-00f */
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,	/* 010-01f */
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,	/* 020-02f */
@@ -242,12 +242,12 @@ static signed char UniCaseRangeLff20[27] = {
 /*
  * Lower Case Range
  */
-static const struct UniCaseRange CifsUniLowerRange[] = {
-	0x0380, 0x03ab, UniCaseRangeL0380,
-	0x0400, 0x042f, UniCaseRangeL0400,
-	0x0490, 0x04cb, UniCaseRangeL0490,
-	0x1e00, 0x1ff7, UniCaseRangeL1e00,
-	0xff20, 0xff3a, UniCaseRangeLff20,
-	0, 0, 0
+const struct UniCaseRange CifsUniLowerRange[] = {
+	{0x0380, 0x03ab, UniCaseRangeL0380},
+	{0x0400, 0x042f, UniCaseRangeL0400},
+	{0x0490, 0x04cb, UniCaseRangeL0490},
+	{0x1e00, 0x1ff7, UniCaseRangeL1e00},
+	{0xff20, 0xff3a, UniCaseRangeLff20},
+	{0}
 };
 #endif
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index 847628dfdc44..35042d8f7338 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -223,63 +223,6 @@ int cifs_calculate_mac_key(struct mac_key *key, const char *rn,
 	return 0;
 }
 
-int CalcNTLMv2_partial_mac_key(struct cifsSesInfo *ses,
-			       const struct nls_table *nls_info)
-{
-	char temp_hash[16];
-	struct HMACMD5Context ctx;
-	char *ucase_buf;
-	__le16 *unicode_buf;
-	unsigned int i, user_name_len, dom_name_len;
-
-	if (ses == NULL)
-		return -EINVAL;
-
-	E_md4hash(ses->password, temp_hash);
-
-	hmac_md5_init_limK_to_64(temp_hash, 16, &ctx);
-	user_name_len = strlen(ses->userName);
-	if (user_name_len > MAX_USERNAME_SIZE)
-		return -EINVAL;
-	if (ses->domainName == NULL)
-		return -EINVAL; /* BB should we use CIFS_LINUX_DOM */
-	dom_name_len = strlen(ses->domainName);
-	if (dom_name_len > MAX_USERNAME_SIZE)
-		return -EINVAL;
-
-	ucase_buf = kmalloc((MAX_USERNAME_SIZE+1), GFP_KERNEL);
-	if (ucase_buf == NULL)
-		return -ENOMEM;
-	unicode_buf = kmalloc((MAX_USERNAME_SIZE+1)*4, GFP_KERNEL);
-	if (unicode_buf == NULL) {
-		kfree(ucase_buf);
-		return -ENOMEM;
-	}
-
-	for (i = 0; i < user_name_len; i++)
-		ucase_buf[i] = nls_info->charset2upper[(int)ses->userName[i]];
-	ucase_buf[i] = 0;
-	user_name_len = cifs_strtoUCS(unicode_buf, ucase_buf,
-				      MAX_USERNAME_SIZE*2, nls_info);
-	unicode_buf[user_name_len] = 0;
-	user_name_len++;
-
-	for (i = 0; i < dom_name_len; i++)
-		ucase_buf[i] = nls_info->charset2upper[(int)ses->domainName[i]];
-	ucase_buf[i] = 0;
-	dom_name_len = cifs_strtoUCS(unicode_buf+user_name_len, ucase_buf,
-				     MAX_USERNAME_SIZE*2, nls_info);
-
-	unicode_buf[user_name_len + dom_name_len] = 0;
-	hmac_md5_update((const unsigned char *) unicode_buf,
-		(user_name_len+dom_name_len)*2, &ctx);
-
-	hmac_md5_final(ses->server->ntlmv2_hash, &ctx);
-	kfree(ucase_buf);
-	kfree(unicode_buf);
-	return 0;
-}
-
 #ifdef CONFIG_CIFS_WEAK_PW_HASH
 void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
 			char *lnm_session_key)
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index 1f5450814087..1d60c655e3e0 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -87,8 +87,9 @@ extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
 			struct TCP_Server_Info *server);
 extern int cifs_convert_address(struct sockaddr *dst, const char *src, int len);
+extern int cifs_set_port(struct sockaddr *addr, const unsigned short int port);
 extern int cifs_fill_sockaddr(struct sockaddr *dst, const char *src, int len,
-				unsigned short int port);
+				const unsigned short int port);
 extern int map_smb_to_linux_error(struct smb_hdr *smb, int logErr);
 extern void header_assemble(struct smb_hdr *, char /* command */ ,
 			    const struct cifsTconInfo *, int /* length of
@@ -365,8 +366,6 @@ extern int cifs_verify_signature(struct smb_hdr *,
 				__u32 expected_sequence_number);
 extern int cifs_calculate_mac_key(struct mac_key *key, const char *rn,
 				 const char *pass);
-extern int CalcNTLMv2_partial_mac_key(struct cifsSesInfo *,
-			const struct nls_table *);
 extern void CalcNTLMv2_response(const struct cifsSesInfo *, char *);
 extern void setup_ntlmv2_rsp(struct cifsSesInfo *, char *,
 			     const struct nls_table *);
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index c65c3419dd37..7e83b356cc9e 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -232,7 +232,7 @@ static int
 small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
 		void **request_buf)
 {
-	int rc = 0;
+	int rc;
 
 	rc = cifs_reconnect_tcon(tcon, smb_command);
 	if (rc)
@@ -250,7 +250,7 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
 	if (tcon != NULL)
 		cifs_stats_inc(&tcon->num_smbs_sent);
 
-	return rc;
+	return 0;
 }
 
 int
@@ -281,16 +281,9 @@ small_smb_init_no_tc(const int smb_command, const int wct,
 
 /* If the return code is zero, this function must fill in request_buf pointer */
 static int
-smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
-	 void **request_buf /* returned */ ,
-	 void **response_buf /* returned */ )
+__smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
+			void **request_buf, void **response_buf)
 {
-	int rc = 0;
-
-	rc = cifs_reconnect_tcon(tcon, smb_command);
-	if (rc)
-		return rc;
-
 	*request_buf = cifs_buf_get();
 	if (*request_buf == NULL) {
 		/* BB should we add a retry in here if not a writepage? */
@@ -309,7 +302,31 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
 	if (tcon != NULL)
 		cifs_stats_inc(&tcon->num_smbs_sent);
 
-	return rc;
+	return 0;
+}
+
+/* If the return code is zero, this function must fill in request_buf pointer */
+static int
+smb_init(int smb_command, int wct, struct cifsTconInfo *tcon,
+	 void **request_buf, void **response_buf)
+{
+	int rc;
+
+	rc = cifs_reconnect_tcon(tcon, smb_command);
+	if (rc)
+		return rc;
+
+	return __smb_init(smb_command, wct, tcon, request_buf, response_buf);
+}
+
+static int
+smb_init_no_reconnect(int smb_command, int wct, struct cifsTconInfo *tcon,
+			void **request_buf, void **response_buf)
+{
+	if (tcon->ses->need_reconnect || tcon->need_reconnect)
+		return -EHOSTDOWN;
+
+	return __smb_init(smb_command, wct, tcon, request_buf, response_buf);
 }
 
 static int validate_t2(struct smb_t2_rsp *pSMB)
@@ -4534,8 +4551,8 @@ CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon)
 
 	cFYI(1, "In QFSUnixInfo");
 QFSUnixRetry:
-	rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB,
-		      (void **) &pSMBr);
+	rc = smb_init_no_reconnect(SMB_COM_TRANSACTION2, 15, tcon,
+				   (void **) &pSMB, (void **) &pSMBr);
 	if (rc)
 		return rc;
 
@@ -4604,8 +4621,8 @@ CIFSSMBSetFSUnixInfo(const int xid, struct cifsTconInfo *tcon, __u64 cap)
 	cFYI(1, "In SETFSUnixInfo");
 SETFSUnixRetry:
 	/* BB switch to small buf init to save memory */
-	rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB,
-		      (void **) &pSMBr);
+	rc = smb_init_no_reconnect(SMB_COM_TRANSACTION2, 15, tcon,
+					(void **) &pSMB, (void **) &pSMBr);
 	if (rc)
 		return rc;
 
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 95c2ea67edfb..88c84a38bccb 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -400,7 +400,9 @@ incomplete_rcv:
 			cFYI(1, "call to reconnect done");
 			csocket = server->ssocket;
 			continue;
-		} else if ((length == -ERESTARTSYS) || (length == -EAGAIN)) {
+		} else if (length == -ERESTARTSYS ||
+			   length == -EAGAIN ||
+			   length == -EINTR) {
 			msleep(1); /* minimum sleep to prevent looping
 				allowing socket to clear and app threads to set
 				tcpStatus CifsNeedReconnect if server hung */
@@ -414,18 +416,6 @@ incomplete_rcv:
 			} else
 				continue;
 		} else if (length <= 0) {
-			if (server->tcpStatus == CifsNew) {
-				cFYI(1, "tcp session abend after SMBnegprot");
-				/* some servers kill the TCP session rather than
-				   returning an SMB negprot error, in which
-				   case reconnecting here is not going to help,
-				   and so simply return error to mount */
-				break;
-			}
-			if (!try_to_freeze() && (length == -EINTR)) {
-				cFYI(1, "cifsd thread killed");
-				break;
-			}
 			cFYI(1, "Reconnect after unexpected peek error %d",
 				length);
 			cifs_reconnect(server);
@@ -466,27 +456,19 @@ incomplete_rcv:
 			   an error on SMB negprot response */
 			cFYI(1, "Negative RFC1002 Session Response Error 0x%x)",
 				pdu_length);
-			if (server->tcpStatus == CifsNew) {
-				/* if nack on negprot (rather than
-				ret of smb negprot error) reconnecting
-				not going to help, ret error to mount */
-				break;
-			} else {
-				/* give server a second to
-				clean up before reconnect attempt */
-				msleep(1000);
-				/* always try 445 first on reconnect
-				since we get NACK on some if we ever
-				connected to port 139 (the NACK is
-				since we do not begin with RFC1001
-				session initialize frame) */
-				server->addr.sockAddr.sin_port =
-					htons(CIFS_PORT);
-				cifs_reconnect(server);
-				csocket = server->ssocket;
-				wake_up(&server->response_q);
-				continue;
-			}
+			/* give server a second to clean up  */
+			msleep(1000);
+			/* always try 445 first on reconnect since we get NACK
+			 * on some if we ever connected to port 139 (the NACK
+			 * is since we do not begin with RFC1001 session
+			 * initialize frame)
+			 */
+			cifs_set_port((struct sockaddr *)
+					&server->addr.sockAddr, CIFS_PORT);
+			cifs_reconnect(server);
+			csocket = server->ssocket;
+			wake_up(&server->response_q);
+			continue;
 		} else if (temp != (char) 0) {
 			cERROR(1, "Unknown RFC 1002 frame");
 			cifs_dump_mem(" Received Data: ", (char *)smb_buffer,
@@ -522,8 +504,7 @@ incomplete_rcv:
 		     total_read += length) {
 			length = kernel_recvmsg(csocket, &smb_msg, &iov, 1,
 						pdu_length - total_read, 0);
-			if ((server->tcpStatus == CifsExiting) ||
-			    (length == -EINTR)) {
+			if (server->tcpStatus == CifsExiting) {
 				/* then will exit */
 				reconnect = 2;
 				break;
@@ -534,8 +515,9 @@ incomplete_rcv:
 				/* Now we will reread sock */
 				reconnect = 1;
 				break;
-			} else if ((length == -ERESTARTSYS) ||
-				   (length == -EAGAIN)) {
+			} else if (length == -ERESTARTSYS ||
+				   length == -EAGAIN ||
+				   length == -EINTR) {
 				msleep(1); /* minimum sleep to prevent looping,
 					      allowing socket to clear and app
 					      threads to set tcpStatus
@@ -1673,7 +1655,9 @@ cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol)
 				    MAX_USERNAME_SIZE))
 				continue;
 			if (strlen(vol->username) != 0 &&
-			    strncmp(ses->password, vol->password,
+			    ses->password != NULL &&
+			    strncmp(ses->password,
+				    vol->password ? vol->password : "",
 				    MAX_PASSWORD_SIZE))
 				continue;
 		}
@@ -1722,9 +1706,6 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 	if (ses) {
 		cFYI(1, "Existing smb sess found (status=%d)", ses->status);
 
-		/* existing SMB ses has a server reference already */
-		cifs_put_tcp_session(server);
-
 		mutex_lock(&ses->session_mutex);
 		rc = cifs_negotiate_protocol(xid, ses);
 		if (rc) {
@@ -1747,6 +1728,9 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 			}
 		}
 		mutex_unlock(&ses->session_mutex);
+
+		/* existing SMB ses has a server reference already */
+		cifs_put_tcp_session(server);
 		FreeXid(xid);
 		return ses;
 	}
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 578d88c5b46e..f9ed0751cc12 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -305,8 +305,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
 		rc = -ENOMEM;
-		FreeXid(xid);
-		return rc;
+		goto cifs_create_out;
 	}
 
 	if (oplockEnabled)
@@ -365,9 +364,8 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
 
 	buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
 	if (buf == NULL) {
-		kfree(full_path);
-		FreeXid(xid);
-		return -ENOMEM;
+		rc = -ENOMEM;
+		goto cifs_create_out;
 	}
 
 	/*
@@ -496,6 +494,11 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
 	struct cifsTconInfo *pTcon;
 	char *full_path = NULL;
 	struct inode *newinode = NULL;
+	int oplock = 0;
+	u16 fileHandle;
+	FILE_ALL_INFO *buf = NULL;
+	unsigned int bytes_written;
+	struct win_dev *pdev;
 
 	if (!old_valid_dev(device_number))
 		return -EINVAL;
@@ -506,9 +509,12 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
 	pTcon = cifs_sb->tcon;
 
 	full_path = build_path_from_dentry(direntry);
-	if (full_path == NULL)
+	if (full_path == NULL) {
 		rc = -ENOMEM;
-	else if (pTcon->unix_ext) {
+		goto mknod_out;
+	}
+
+	if (pTcon->unix_ext) {
 		struct cifs_unix_set_info_args args = {
 			.mode	= mode & ~current_umask(),
 			.ctime	= NO_CHANGE_64,
@@ -527,87 +533,78 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
 					    cifs_sb->local_nls,
 					    cifs_sb->mnt_cifs_flags &
 						CIFS_MOUNT_MAP_SPECIAL_CHR);
+		if (rc)
+			goto mknod_out;
 
-		if (!rc) {
-			rc = cifs_get_inode_info_unix(&newinode, full_path,
+		rc = cifs_get_inode_info_unix(&newinode, full_path,
 						inode->i_sb, xid);
-			if (pTcon->nocase)
-				direntry->d_op = &cifs_ci_dentry_ops;
-			else
-				direntry->d_op = &cifs_dentry_ops;
-			if (rc == 0)
-				d_instantiate(direntry, newinode);
-		}
-	} else {
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) {
-			int oplock = 0;
-			u16 fileHandle;
-			FILE_ALL_INFO *buf;
+		if (pTcon->nocase)
+			direntry->d_op = &cifs_ci_dentry_ops;
+		else
+			direntry->d_op = &cifs_dentry_ops;
 
-			cFYI(1, "sfu compat create special file");
+		if (rc == 0)
+			d_instantiate(direntry, newinode);
+		goto mknod_out;
+	}
 
-			buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
-			if (buf == NULL) {
-				kfree(full_path);
-				rc = -ENOMEM;
-				FreeXid(xid);
-				return rc;
-			}
+	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL))
+		goto mknod_out;
 
-			rc = CIFSSMBOpen(xid, pTcon, full_path,
-					 FILE_CREATE, /* fail if exists */
-					 GENERIC_WRITE /* BB would
-					  WRITE_OWNER | WRITE_DAC be better? */,
-					 /* Create a file and set the
-					    file attribute to SYSTEM */
-					 CREATE_NOT_DIR | CREATE_OPTION_SPECIAL,
-					 &fileHandle, &oplock, buf,
-					 cifs_sb->local_nls,
-					 cifs_sb->mnt_cifs_flags &
-					    CIFS_MOUNT_MAP_SPECIAL_CHR);
-
-			/* BB FIXME - add handling for backlevel servers
-			   which need legacy open and check for all
-			   calls to SMBOpen for fallback to SMBLeagcyOpen */
-			if (!rc) {
-				/* BB Do not bother to decode buf since no
-				   local inode yet to put timestamps in,
-				   but we can reuse it safely */
-				unsigned int bytes_written;
-				struct win_dev *pdev;
-				pdev = (struct win_dev *)buf;
-				if (S_ISCHR(mode)) {
-					memcpy(pdev->type, "IntxCHR", 8);
-					pdev->major =
-					      cpu_to_le64(MAJOR(device_number));
-					pdev->minor =
-					      cpu_to_le64(MINOR(device_number));
-					rc = CIFSSMBWrite(xid, pTcon,
-						fileHandle,
-						sizeof(struct win_dev),
-						0, &bytes_written, (char *)pdev,
-						NULL, 0);
-				} else if (S_ISBLK(mode)) {
-					memcpy(pdev->type, "IntxBLK", 8);
-					pdev->major =
-					      cpu_to_le64(MAJOR(device_number));
-					pdev->minor =
-					      cpu_to_le64(MINOR(device_number));
-					rc = CIFSSMBWrite(xid, pTcon,
-						fileHandle,
-						sizeof(struct win_dev),
-						0, &bytes_written, (char *)pdev,
-						NULL, 0);
-				} /* else if(S_ISFIFO */
-				CIFSSMBClose(xid, pTcon, fileHandle);
-				d_drop(direntry);
-			}
-			kfree(buf);
-			/* add code here to set EAs */
-		}
+
+	cFYI(1, "sfu compat create special file");
+
+	buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
+	if (buf == NULL) {
+		kfree(full_path);
+		rc = -ENOMEM;
+		FreeXid(xid);
+		return rc;
 	}
 
+	/* FIXME: would WRITE_OWNER | WRITE_DAC be better? */
+	rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_CREATE,
+			 GENERIC_WRITE, CREATE_NOT_DIR | CREATE_OPTION_SPECIAL,
+			 &fileHandle, &oplock, buf, cifs_sb->local_nls,
+			 cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
+	if (rc)
+		goto mknod_out;
+
+	/* BB Do not bother to decode buf since no local inode yet to put
+	 * timestamps in, but we can reuse it safely */
+
+	pdev = (struct win_dev *)buf;
+	if (S_ISCHR(mode)) {
+		memcpy(pdev->type, "IntxCHR", 8);
+		pdev->major =
+		      cpu_to_le64(MAJOR(device_number));
+		pdev->minor =
+		      cpu_to_le64(MINOR(device_number));
+		rc = CIFSSMBWrite(xid, pTcon,
+			fileHandle,
+			sizeof(struct win_dev),
+			0, &bytes_written, (char *)pdev,
+			NULL, 0);
+	} else if (S_ISBLK(mode)) {
+		memcpy(pdev->type, "IntxBLK", 8);
+		pdev->major =
+		      cpu_to_le64(MAJOR(device_number));
+		pdev->minor =
+		      cpu_to_le64(MINOR(device_number));
+		rc = CIFSSMBWrite(xid, pTcon,
+			fileHandle,
+			sizeof(struct win_dev),
+			0, &bytes_written, (char *)pdev,
+			NULL, 0);
+	} /* else if (S_ISFIFO) */
+	CIFSSMBClose(xid, pTcon, fileHandle);
+	d_drop(direntry);
+
+	/* FIXME: add code here to set EAs */
+
+mknod_out:
 	kfree(full_path);
+	kfree(buf);
 	FreeXid(xid);
 	return rc;
 }
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index db11fdef0e92..de748c652d11 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -242,8 +242,7 @@ int cifs_open(struct inode *inode, struct file *file)
 	full_path = build_path_from_dentry(file->f_path.dentry);
 	if (full_path == NULL) {
 		rc = -ENOMEM;
-		FreeXid(xid);
-		return rc;
+		goto out;
 	}
 
 	cFYI(1, "inode = 0x%p file flags are 0x%x for %s",
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 4bc47e5b5f29..53cce8cc2224 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -801,6 +801,8 @@ retry_iget5_locked:
 			inode->i_flags |= S_NOATIME | S_NOCMTIME;
 		if (inode->i_state & I_NEW) {
 			inode->i_ino = hash;
+			if (S_ISREG(inode->i_mode))
+				inode->i_data.backing_dev_info = sb->s_bdi;
 #ifdef CONFIG_CIFS_FSCACHE
 			/* initialize per-inode cache cookie pointer */
 			CIFS_I(inode)->fscache = NULL;
@@ -834,7 +836,7 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino)
 						xid, NULL);
 
 	if (!inode)
-		return ERR_PTR(-ENOMEM);
+		return ERR_PTR(rc);
 
 #ifdef CONFIG_CIFS_FSCACHE
 	/* populate tcon->resource_id */
@@ -1462,29 +1464,18 @@ int cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
 {
 	char *fromName = NULL;
 	char *toName = NULL;
-	struct cifs_sb_info *cifs_sb_source;
-	struct cifs_sb_info *cifs_sb_target;
+	struct cifs_sb_info *cifs_sb;
 	struct cifsTconInfo *tcon;
 	FILE_UNIX_BASIC_INFO *info_buf_source = NULL;
 	FILE_UNIX_BASIC_INFO *info_buf_target;
 	int xid, rc, tmprc;
 
-	cifs_sb_target = CIFS_SB(target_dir->i_sb);
-	cifs_sb_source = CIFS_SB(source_dir->i_sb);
-	tcon = cifs_sb_source->tcon;
+	cifs_sb = CIFS_SB(source_dir->i_sb);
+	tcon = cifs_sb->tcon;
 
 	xid = GetXid();
 
 	/*
-	 * BB: this might be allowed if same server, but different share.
-	 * Consider adding support for this
-	 */
-	if (tcon != cifs_sb_target->tcon) {
-		rc = -EXDEV;
-		goto cifs_rename_exit;
-	}
-
-	/*
 	 * we already have the rename sem so we do not need to
 	 * grab it again here to protect the path integrity
 	 */
@@ -1519,17 +1510,16 @@ int cifs_rename(struct inode *source_dir, struct dentry *source_dentry,
 		info_buf_target = info_buf_source + 1;
 		tmprc = CIFSSMBUnixQPathInfo(xid, tcon, fromName,
 					info_buf_source,
-					cifs_sb_source->local_nls,
-					cifs_sb_source->mnt_cifs_flags &
+					cifs_sb->local_nls,
+					cifs_sb->mnt_cifs_flags &
 					CIFS_MOUNT_MAP_SPECIAL_CHR);
 		if (tmprc != 0)
 			goto unlink_target;
 
-		tmprc = CIFSSMBUnixQPathInfo(xid, tcon,
-					toName, info_buf_target,
-					cifs_sb_target->local_nls,
-					/* remap based on source sb */
-					cifs_sb_source->mnt_cifs_flags &
+		tmprc = CIFSSMBUnixQPathInfo(xid, tcon, toName,
+					info_buf_target,
+					cifs_sb->local_nls,
+					cifs_sb->mnt_cifs_flags &
 					CIFS_MOUNT_MAP_SPECIAL_CHR);
 
 		if (tmprc == 0 && (info_buf_source->UniqueId ==
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index f97851119e6c..9aad47a2d62f 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -206,26 +206,30 @@ cifs_convert_address(struct sockaddr *dst, const char *src, int len)
 }
 
 int
-cifs_fill_sockaddr(struct sockaddr *dst, const char *src, int len,
-		   const unsigned short int port)
+cifs_set_port(struct sockaddr *addr, const unsigned short int port)
 {
-	if (!cifs_convert_address(dst, src, len))
-		return 0;
-
-	switch (dst->sa_family) {
+	switch (addr->sa_family) {
 	case AF_INET:
-		((struct sockaddr_in *)dst)->sin_port = htons(port);
+		((struct sockaddr_in *)addr)->sin_port = htons(port);
 		break;
 	case AF_INET6:
-		((struct sockaddr_in6 *)dst)->sin6_port = htons(port);
+		((struct sockaddr_in6 *)addr)->sin6_port = htons(port);
 		break;
 	default:
 		return 0;
 	}
-
 	return 1;
 }
 
+int
+cifs_fill_sockaddr(struct sockaddr *dst, const char *src, int len,
+		   const unsigned short int port)
+{
+	if (!cifs_convert_address(dst, src, len))
+		return 0;
+	return cifs_set_port(dst, port);
+}
+
 /*****************************************************************************
 convert a NT status code to a dos class/code
  *****************************************************************************/
diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c
index de89645777c7..116af7546cf0 100644
--- a/fs/coda/psdev.c
+++ b/fs/coda/psdev.c
@@ -184,8 +184,8 @@ static ssize_t coda_psdev_write(struct file *file, const char __user *buf,
 	}
 
 	/* adjust outsize. is this useful ?? */
-        req->uc_outSize = nbytes;	
-        req->uc_flags |= REQ_WRITE;
+	req->uc_outSize = nbytes;
+	req->uc_flags |= CODA_REQ_WRITE;
 	count = nbytes;
 
 	/* Convert filedescriptor into a file handle */
diff --git a/fs/compat.c b/fs/compat.c
index 718c7062aec1..0644a154672b 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1153,7 +1153,7 @@ static ssize_t compat_do_readv_writev(int type, struct file *file,
 {
 	compat_ssize_t tot_len;
 	struct iovec iovstack[UIO_FASTIOV];
-	struct iovec *iov;
+	struct iovec *iov = iovstack;
 	ssize_t ret;
 	io_fn_t fn;
 	iov_fn_t fnv;
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 51f270b479b6..48d74c7391d1 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -634,7 +634,7 @@ static int dio_send_cur_page(struct dio *dio)
 	int ret = 0;
 
 	if (dio->bio) {
-		loff_t cur_offset = dio->block_in_file << dio->blkbits;
+		loff_t cur_offset = dio->cur_page_fs_offset;
 		loff_t bio_next_offset = dio->logical_offset_in_bio +
 			dio->bio->bi_size;
 
@@ -659,7 +659,7 @@ static int dio_send_cur_page(struct dio *dio)
 		 * Submit now if the underlying fs is about to perform a
 		 * metadata read
 		 */
-		if (dio->boundary)
+		else if (dio->boundary)
 			dio_bio_submit(dio);
 	}
 
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index a2e3b562e65d..cbadc1bee6e7 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -1793,7 +1793,7 @@ struct kmem_cache *ecryptfs_key_tfm_cache;
 static struct list_head key_tfm_list;
 struct mutex key_tfm_list_mutex;
 
-int ecryptfs_init_crypto(void)
+int __init ecryptfs_init_crypto(void)
 {
 	mutex_init(&key_tfm_list_mutex);
 	INIT_LIST_HEAD(&key_tfm_list);
@@ -2169,7 +2169,6 @@ int ecryptfs_encrypt_and_encode_filename(
 				(ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE
 				 + encoded_name_no_prefix_size);
 			(*encoded_name)[(*encoded_name_size)] = '\0';
-			(*encoded_name_size)++;
 		} else {
 			rc = -EOPNOTSUPP;
 		}
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 6c55113e7222..3fbc94203380 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -349,7 +349,7 @@ out:
 
 /**
  * ecryptfs_new_lower_dentry
- * @ename: The name of the new dentry.
+ * @name: The name of the new dentry.
  * @lower_dir_dentry: Parent directory of the new dentry.
  * @nd: nameidata from last lookup.
  *
@@ -386,20 +386,19 @@ ecryptfs_new_lower_dentry(struct qstr *name, struct dentry *lower_dir_dentry,
  * ecryptfs_lookup_one_lower
  * @ecryptfs_dentry: The eCryptfs dentry that we are looking up
  * @lower_dir_dentry: lower parent directory
+ * @name: lower file name
  *
  * Get the lower dentry from vfs. If lower dentry does not exist yet,
  * create it.
  */
 static struct dentry *
 ecryptfs_lookup_one_lower(struct dentry *ecryptfs_dentry,
-			  struct dentry *lower_dir_dentry)
+			  struct dentry *lower_dir_dentry, struct qstr *name)
 {
 	struct nameidata nd;
 	struct vfsmount *lower_mnt;
-	struct qstr *name;
 	int err;
 
-	name = &ecryptfs_dentry->d_name;
 	lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(
 				    ecryptfs_dentry->d_parent));
 	err = vfs_path_lookup(lower_dir_dentry, lower_mnt, name->name , 0, &nd);
@@ -434,6 +433,7 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 	size_t encrypted_and_encoded_name_size;
 	struct ecryptfs_mount_crypt_stat *mount_crypt_stat = NULL;
 	struct dentry *lower_dir_dentry, *lower_dentry;
+	struct qstr lower_name;
 	int rc = 0;
 
 	ecryptfs_dentry->d_op = &ecryptfs_dops;
@@ -444,9 +444,17 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 		goto out_d_drop;
 	}
 	lower_dir_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry->d_parent);
-
+	lower_name.name = ecryptfs_dentry->d_name.name;
+	lower_name.len = ecryptfs_dentry->d_name.len;
+	lower_name.hash = ecryptfs_dentry->d_name.hash;
+	if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) {
+		rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry,
+						    &lower_name);
+		if (rc < 0)
+			goto out_d_drop;
+	}
 	lower_dentry = ecryptfs_lookup_one_lower(ecryptfs_dentry,
-						 lower_dir_dentry);
+						 lower_dir_dentry, &lower_name);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
 		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_lower() returned "
@@ -471,8 +479,17 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 		       "filename; rc = [%d]\n", __func__, rc);
 		goto out_d_drop;
 	}
+	lower_name.name = encrypted_and_encoded_name;
+	lower_name.len = encrypted_and_encoded_name_size;
+	lower_name.hash = full_name_hash(lower_name.name, lower_name.len);
+	if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) {
+		rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry,
+						    &lower_name);
+		if (rc < 0)
+			goto out_d_drop;
+	}
 	lower_dentry = ecryptfs_lookup_one_lower(ecryptfs_dentry,
-						 lower_dir_dentry);
+						 lower_dir_dentry, &lower_name);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
 		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_lower() returned "
diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c
index 89c5476506ef..73811cfa2ea4 100644
--- a/fs/ecryptfs/keystore.c
+++ b/fs/ecryptfs/keystore.c
@@ -515,6 +515,7 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes,
 	if (!s) {
 		printk(KERN_ERR "%s: Out of memory whilst trying to kmalloc "
 		       "[%zd] bytes of kernel memory\n", __func__, sizeof(*s));
+		rc = -ENOMEM;
 		goto out;
 	}
 	s->desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
@@ -806,6 +807,7 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size,
 	if (!s) {
 		printk(KERN_ERR "%s: Out of memory whilst trying to kmalloc "
 		       "[%zd] bytes of kernel memory\n", __func__, sizeof(*s));
+		rc = -ENOMEM;
 		goto out;
 	}
 	s->desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
diff --git a/fs/ecryptfs/kthread.c b/fs/ecryptfs/kthread.c
index d8c3a373aafa..0851ab6980f5 100644
--- a/fs/ecryptfs/kthread.c
+++ b/fs/ecryptfs/kthread.c
@@ -86,7 +86,7 @@ out:
 	return 0;
 }
 
-int ecryptfs_init_kthread(void)
+int __init ecryptfs_init_kthread(void)
 {
 	int rc = 0;
 
diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c
index bcb68c0cb1f0..ab2248090515 100644
--- a/fs/ecryptfs/messaging.c
+++ b/fs/ecryptfs/messaging.c
@@ -473,7 +473,7 @@ sleep:
 	return rc;
 }
 
-int ecryptfs_init_messaging(void)
+int __init ecryptfs_init_messaging(void)
 {
 	int i;
 	int rc = 0;
diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c
index 3745f612bcd4..00208c3d7e92 100644
--- a/fs/ecryptfs/miscdev.c
+++ b/fs/ecryptfs/miscdev.c
@@ -500,7 +500,7 @@ static struct miscdevice ecryptfs_miscdev = {
  *
  * Returns zero on success; non-zero otherwise
  */
-int ecryptfs_init_ecryptfs_miscdev(void)
+int __init ecryptfs_init_ecryptfs_miscdev(void)
 {
 	int rc;
 
diff --git a/fs/exec.c b/fs/exec.c
index 2d9455282744..03278c984ba0 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -376,6 +376,9 @@ static int count(const char __user * const __user * argv, int max)
 			argv++;
 			if (i++ >= max)
 				return -E2BIG;
+
+			if (fatal_signal_pending(current))
+				return -ERESTARTNOHAND;
 			cond_resched();
 		}
 	}
@@ -419,6 +422,12 @@ static int copy_strings(int argc, const char __user *const __user *argv,
 		while (len > 0) {
 			int offset, bytes_to_copy;
 
+			if (fatal_signal_pending(current)) {
+				ret = -ERESTARTNOHAND;
+				goto out;
+			}
+			cond_resched();
+
 			offset = pos % PAGE_SIZE;
 			if (offset == 0)
 				offset = PAGE_SIZE;
@@ -594,6 +603,11 @@ int setup_arg_pages(struct linux_binprm *bprm,
 #else
 	stack_top = arch_align_stack(stack_top);
 	stack_top = PAGE_ALIGN(stack_top);
+
+	if (unlikely(stack_top < mmap_min_addr) ||
+	    unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr))
+		return -ENOMEM;
+
 	stack_shift = vma->vm_end - stack_top;
 
 	bprm->p -= stack_shift;
@@ -2000,3 +2014,41 @@ fail_creds:
 fail:
 	return;
 }
+
+/*
+ * Core dumping helper functions.  These are the only things you should
+ * do on a core-file: use only these functions to write out all the
+ * necessary info.
+ */
+int dump_write(struct file *file, const void *addr, int nr)
+{
+	return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr;
+}
+
+int dump_seek(struct file *file, loff_t off)
+{
+	int ret = 1;
+
+	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
+		if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
+			return 0;
+	} else {
+		char *buf = (char *)get_zeroed_page(GFP_KERNEL);
+
+		if (!buf)
+			return 0;
+		while (off > 0) {
+			unsigned long n = off;
+
+			if (n > PAGE_SIZE)
+				n = PAGE_SIZE;
+			if (!dump_write(file, buf, n)) {
+				ret = 0;
+				break;
+			}
+			off -= n;
+		}
+		free_page((unsigned long)buf);
+	}
+	return ret;
+}
diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index eb7368ebd8cd..3eadd97324b1 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -54,6 +54,9 @@ struct page_collect {
 	unsigned nr_pages;
 	unsigned long length;
 	loff_t pg_first; /* keep 64bit also in 32-arches */
+	bool read_4_write; /* This means two things: that the read is sync
+			    * And the pages should not be unlocked.
+			    */
 };
 
 static void _pcol_init(struct page_collect *pcol, unsigned expected_pages,
@@ -71,6 +74,7 @@ static void _pcol_init(struct page_collect *pcol, unsigned expected_pages,
 	pcol->nr_pages = 0;
 	pcol->length = 0;
 	pcol->pg_first = -1;
+	pcol->read_4_write = false;
 }
 
 static void _pcol_reset(struct page_collect *pcol)
@@ -347,7 +351,8 @@ static int readpage_strip(void *data, struct page *page)
 		if (PageError(page))
 			ClearPageError(page);
 
-		unlock_page(page);
+		if (!pcol->read_4_write)
+			unlock_page(page);
 		EXOFS_DBGMSG("readpage_strip(0x%lx, 0x%lx) empty page,"
 			     " splitting\n", inode->i_ino, page->index);
 
@@ -428,6 +433,7 @@ static int _readpage(struct page *page, bool is_sync)
 	/* readpage_strip might call read_exec(,is_sync==false) at several
 	 * places but not if we have a single page.
 	 */
+	pcol.read_4_write = is_sync;
 	ret = readpage_strip(&pcol, page);
 	if (ret) {
 		EXOFS_ERR("_readpage => %d\n", ret);
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 6769fd0f35b8..f8cc34f542c3 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -769,11 +769,15 @@ EXPORT_SYMBOL(kill_fasync);
 
 static int __init fcntl_init(void)
 {
-	/* please add new bits here to ensure allocation uniqueness */
-	BUILD_BUG_ON(19 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32(
+	/*
+	 * Please add new bits here to ensure allocation uniqueness.
+	 * Exceptions: O_NONBLOCK is a two bit define on parisc; O_NDELAY
+	 * is defined as O_NONBLOCK on some platforms and not on others.
+	 */
+	BUILD_BUG_ON(18 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32(
 		O_RDONLY	| O_WRONLY	| O_RDWR	|
 		O_CREAT		| O_EXCL	| O_NOCTTY	|
-		O_TRUNC		| O_APPEND	| O_NONBLOCK	|
+		O_TRUNC		| O_APPEND	| /* O_NONBLOCK	| */
 		__O_SYNC	| O_DSYNC	| FASYNC	|
 		O_DIRECT	| O_LARGEFILE	| O_DIRECTORY	|
 		O_NOFOLLOW	| O_NOATIME	| O_CLOEXEC	|
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 7d9d06ba184b..ab38fef1c9a1 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -52,8 +52,6 @@ struct wb_writeback_work {
 #define CREATE_TRACE_POINTS
 #include <trace/events/writeback.h>
 
-#define inode_to_bdi(inode)	((inode)->i_mapping->backing_dev_info)
-
 /*
  * We don't actually have pdflush, but this one is exported though /proc...
  */
@@ -71,6 +69,16 @@ int writeback_in_progress(struct backing_dev_info *bdi)
 	return test_bit(BDI_writeback_running, &bdi->state);
 }
 
+static inline struct backing_dev_info *inode_to_bdi(struct inode *inode)
+{
+	struct super_block *sb = inode->i_sb;
+
+	if (strcmp(sb->s_type->name, "bdev") == 0)
+		return inode->i_mapping->backing_dev_info;
+
+	return sb->s_bdi;
+}
+
 static void bdi_queue_work(struct backing_dev_info *bdi,
 		struct wb_writeback_work *work)
 {
@@ -808,7 +816,7 @@ int bdi_writeback_thread(void *data)
 			wb->last_active = jiffies;
 
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (!list_empty(&bdi->work_list)) {
+		if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
 			__set_current_state(TASK_RUNNING);
 			continue;
 		}
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 69ad053ffd78..cde755cca564 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -276,7 +276,7 @@ static void flush_bg_queue(struct fuse_conn *fc)
  * Called with fc->lock, unlocks it
  */
 static void request_end(struct fuse_conn *fc, struct fuse_req *req)
-__releases(&fc->lock)
+__releases(fc->lock)
 {
 	void (*end) (struct fuse_conn *, struct fuse_req *) = req->end;
 	req->end = NULL;
@@ -306,8 +306,8 @@ __releases(&fc->lock)
 
 static void wait_answer_interruptible(struct fuse_conn *fc,
 				      struct fuse_req *req)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	if (signal_pending(current))
 		return;
@@ -325,8 +325,8 @@ static void queue_interrupt(struct fuse_conn *fc, struct fuse_req *req)
 }
 
 static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	if (!fc->no_interrupt) {
 		/* Any signal may interrupt this */
@@ -905,8 +905,8 @@ static int request_pending(struct fuse_conn *fc)
 
 /* Wait until a request is available on the pending list */
 static void request_wait(struct fuse_conn *fc)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	DECLARE_WAITQUEUE(wait, current);
 
@@ -934,7 +934,7 @@ __acquires(&fc->lock)
  */
 static int fuse_read_interrupt(struct fuse_conn *fc, struct fuse_copy_state *cs,
 			       size_t nbytes, struct fuse_req *req)
-__releases(&fc->lock)
+__releases(fc->lock)
 {
 	struct fuse_in_header ih;
 	struct fuse_interrupt_in arg;
@@ -1354,7 +1354,7 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode,
 	loff_t file_size;
 	unsigned int num;
 	unsigned int offset;
-	size_t total_len;
+	size_t total_len = 0;
 
 	req = fuse_get_req(fc);
 	if (IS_ERR(req))
@@ -1720,8 +1720,8 @@ static unsigned fuse_dev_poll(struct file *file, poll_table *wait)
  * This function releases and reacquires fc->lock
  */
 static void end_requests(struct fuse_conn *fc, struct list_head *head)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	while (!list_empty(head)) {
 		struct fuse_req *req;
@@ -1744,8 +1744,8 @@ __acquires(&fc->lock)
  * locked).
  */
 static void end_io_requests(struct fuse_conn *fc)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	while (!list_empty(&fc->io)) {
 		struct fuse_req *req =
@@ -1769,6 +1769,16 @@ __acquires(&fc->lock)
 	}
 }
 
+static void end_queued_requests(struct fuse_conn *fc)
+__releases(fc->lock)
+__acquires(fc->lock)
+{
+	fc->max_background = UINT_MAX;
+	flush_bg_queue(fc);
+	end_requests(fc, &fc->pending);
+	end_requests(fc, &fc->processing);
+}
+
 /*
  * Abort all requests.
  *
@@ -1795,8 +1805,7 @@ void fuse_abort_conn(struct fuse_conn *fc)
 		fc->connected = 0;
 		fc->blocked = 0;
 		end_io_requests(fc);
-		end_requests(fc, &fc->pending);
-		end_requests(fc, &fc->processing);
+		end_queued_requests(fc);
 		wake_up_all(&fc->waitq);
 		wake_up_all(&fc->blocked_waitq);
 		kill_fasync(&fc->fasync, SIGIO, POLL_IN);
@@ -1811,8 +1820,9 @@ int fuse_dev_release(struct inode *inode, struct file *file)
 	if (fc) {
 		spin_lock(&fc->lock);
 		fc->connected = 0;
-		end_requests(fc, &fc->pending);
-		end_requests(fc, &fc->processing);
+		fc->blocked = 0;
+		end_queued_requests(fc);
+		wake_up_all(&fc->blocked_waitq);
 		spin_unlock(&fc->lock);
 		fuse_conn_put(fc);
 	}
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 147c1f71bdb9..c8224587123f 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1144,8 +1144,8 @@ static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req)
 
 /* Called under fc->lock, may release and reacquire it */
 static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	struct fuse_inode *fi = get_fuse_inode(req->inode);
 	loff_t size = i_size_read(req->inode);
@@ -1183,8 +1183,8 @@ __acquires(&fc->lock)
  * Called with fc->lock
  */
 void fuse_flush_writepages(struct inode *inode)
-__releases(&fc->lock)
-__acquires(&fc->lock)
+__releases(fc->lock)
+__acquires(fc->lock)
 {
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_inode *fi = get_fuse_inode(inode);
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 9c65170e932e..eb01f3575e10 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -923,7 +923,7 @@ int gfs2_logd(void *data)
 
 		do {
 			prepare_to_wait(&sdp->sd_logd_waitq, &wait,
-					TASK_UNINTERRUPTIBLE);
+					TASK_INTERRUPTIBLE);
 			if (!gfs2_ail_flush_reqd(sdp) &&
 			    !gfs2_jrnl_flush_reqd(sdp) &&
 			    !kthread_should_stop())
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index e20ee85955d1..f3f3578393a4 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -115,7 +115,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode)
 
 	inode_inc_link_count(dir);
 
-	inode = minix_new_inode(dir, mode, &err);
+	inode = minix_new_inode(dir, S_IFDIR | mode, &err);
 	if (!inode)
 		goto out_dir;
 
diff --git a/fs/namespace.c b/fs/namespace.c
index de402eb6eafb..a72eaabfe8f2 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1484,13 +1484,30 @@ out_unlock:
 }
 
 /*
+ * Sanity check the flags to change_mnt_propagation.
+ */
+
+static int flags_to_propagation_type(int flags)
+{
+	int type = flags & ~MS_REC;
+
+	/* Fail if any non-propagation flags are set */
+	if (type & ~(MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE))
+		return 0;
+	/* Only one propagation flag should be set */
+	if (!is_power_of_2(type))
+		return 0;
+	return type;
+}
+
+/*
  * recursively change the type of the mountpoint.
  */
 static int do_change_type(struct path *path, int flag)
 {
 	struct vfsmount *m, *mnt = path->mnt;
 	int recurse = flag & MS_REC;
-	int type = flag & ~MS_REC;
+	int type;
 	int err = 0;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -1499,6 +1516,10 @@ static int do_change_type(struct path *path, int flag)
 	if (path->dentry != path->mnt->mnt_root)
 		return -EINVAL;
 
+	type = flags_to_propagation_type(flag);
+	if (!type)
+		return -EINVAL;
+
 	down_write(&namespace_sem);
 	if (type == MS_SHARED) {
 		err = invent_group_ids(mnt, recurse);
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 6c2aad49d731..f7e13db613cb 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -63,6 +63,7 @@ config NFS_V3_ACL
 config NFS_V4
 	bool "NFS client support for NFS version 4"
 	depends on NFS_FS
+	select SUNRPC_GSS
 	help
 	  This option enables support for version 4 of the NFS protocol
 	  (RFC 3530) in the kernel's NFS client.
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 4e7df2adb212..e7340729af89 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -275,7 +275,7 @@ static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,
 	    sin1->sin6_scope_id != sin2->sin6_scope_id)
 		return 0;
 
-	return ipv6_addr_equal(&sin1->sin6_addr, &sin1->sin6_addr);
+	return ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr);
 }
 #else	/* !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) */
 static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1,
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index eb51bd6201da..05bf3c0dc751 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -723,10 +723,6 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl)
 		default:
 			BUG();
 	}
-	if (res < 0)
-		dprintk(KERN_WARNING "%s: VFS is out of sync with lock manager"
-			" - error %d!\n",
-				__func__, res);
 	return res;
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index ec3966e4706b..f4cbf0c306c6 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -431,7 +431,15 @@ static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 		goto out_err;
 
 	error = server->nfs_client->rpc_ops->statfs(server, fh, &res);
+	if (unlikely(error == -ESTALE)) {
+		struct dentry *pd_dentry;
 
+		pd_dentry = dget_parent(dentry);
+		if (pd_dentry != NULL) {
+			nfs_zap_caches(pd_dentry->d_inode);
+			dput(pd_dentry);
+		}
+	}
 	nfs_free_fattr(res.fattr);
 	if (error < 0)
 		goto out_err;
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index 95932f523aef..4264377552e2 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -69,6 +69,7 @@ config NFSD_V4
 	depends on NFSD && PROC_FS && EXPERIMENTAL
 	select NFSD_V3
 	select FS_POSIX_ACL
+	select SUNRPC_GSS
 	help
 	  This option enables support in your system's NFS server for
 	  version 4 of the NFS protocol (RFC 3530).
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2e7357104cfd..cf0d2ffb3c84 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -440,7 +440,7 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) {
 
 static int nfs4_access_to_omode(u32 access)
 {
-	switch (access) {
+	switch (access & NFS4_SHARE_ACCESS_BOTH) {
 	case NFS4_SHARE_ACCESS_READ:
 		return O_RDONLY;
 	case NFS4_SHARE_ACCESS_WRITE:
@@ -2450,14 +2450,13 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh,
 static __be32
 nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *cur_fh, struct nfs4_stateid *stp, struct nfsd4_open *open)
 {
-	u32 op_share_access, new_access;
+	u32 op_share_access = open->op_share_access & ~NFS4_SHARE_WANT_MASK;
+	bool new_access;
 	__be32 status;
 
-	set_access(&new_access, stp->st_access_bmap);
-	new_access = (~new_access) & open->op_share_access & ~NFS4_SHARE_WANT_MASK;
-
+	new_access = !test_bit(op_share_access, &stp->st_access_bmap);
 	if (new_access) {
-		status = nfs4_get_vfs_file(rqstp, fp, cur_fh, new_access);
+		status = nfs4_get_vfs_file(rqstp, fp, cur_fh, op_share_access);
 		if (status)
 			return status;
 	}
@@ -2470,7 +2469,6 @@ nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *c
 		return status;
 	}
 	/* remember the open */
-	op_share_access = open->op_share_access & ~NFS4_SHARE_WANT_MASK;
 	__set_bit(op_share_access, &stp->st_access_bmap);
 	__set_bit(open->op_share_deny, &stp->st_deny_bmap);
 
@@ -2983,7 +2981,6 @@ nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 				*filpp = find_readable_file(stp->st_file);
 			else
 				*filpp = find_writeable_file(stp->st_file);
-			BUG_ON(!*filpp); /* assured by check_openmode */
 		}
 	}
 	status = nfs_ok;
@@ -3561,7 +3558,8 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	struct nfs4_stateowner *open_sop = NULL;
 	struct nfs4_stateowner *lock_sop = NULL;
 	struct nfs4_stateid *lock_stp;
-	struct file *filp;
+	struct nfs4_file *fp;
+	struct file *filp = NULL;
 	struct file_lock file_lock;
 	struct file_lock conflock;
 	__be32 status = 0;
@@ -3591,7 +3589,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		 * lock stateid.
 		 */
 		struct nfs4_stateid *open_stp = NULL;
-		struct nfs4_file *fp;
 		
 		status = nfserr_stale_clientid;
 		if (!nfsd4_has_session(cstate) &&
@@ -3634,6 +3631,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		if (status)
 			goto out;
 		lock_sop = lock->lk_replay_owner;
+		fp = lock_stp->st_file;
 	}
 	/* lock->lk_replay_owner and lock_stp have been created or found */
 
@@ -3648,13 +3646,19 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	switch (lock->lk_type) {
 		case NFS4_READ_LT:
 		case NFS4_READW_LT:
-			filp = find_readable_file(lock_stp->st_file);
+			if (find_readable_file(lock_stp->st_file)) {
+				nfs4_get_vfs_file(rqstp, fp, &cstate->current_fh, NFS4_SHARE_ACCESS_READ);
+				filp = find_readable_file(lock_stp->st_file);
+			}
 			file_lock.fl_type = F_RDLCK;
 			cmd = F_SETLK;
 		break;
 		case NFS4_WRITE_LT:
 		case NFS4_WRITEW_LT:
-			filp = find_writeable_file(lock_stp->st_file);
+			if (find_writeable_file(lock_stp->st_file)) {
+				nfs4_get_vfs_file(rqstp, fp, &cstate->current_fh, NFS4_SHARE_ACCESS_WRITE);
+				filp = find_writeable_file(lock_stp->st_file);
+			}
 			file_lock.fl_type = F_WRLCK;
 			cmd = F_SETLK;
 		break;
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
index cdfb8c6a4206..c16f8d8331b5 100644
--- a/fs/nfsd/nfsfh.h
+++ b/fs/nfsd/nfsfh.h
@@ -196,8 +196,6 @@ fh_lock(struct svc_fh *fhp)
 static inline void
 fh_unlock(struct svc_fh *fhp)
 {
-	BUG_ON(!fhp->fh_dentry);
-
 	if (fhp->fh_locked) {
 		fill_post_wcc(fhp);
 		mutex_unlock(&fhp->fh_dentry->d_inode->i_mutex);
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 7731a75971dd..322518c88e4b 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -363,23 +363,23 @@ struct nfs4_file {
  * at all? */
 static inline struct file *find_writeable_file(struct nfs4_file *f)
 {
-	if (f->fi_fds[O_RDWR])
-		return f->fi_fds[O_RDWR];
-	return f->fi_fds[O_WRONLY];
+	if (f->fi_fds[O_WRONLY])
+		return f->fi_fds[O_WRONLY];
+	return f->fi_fds[O_RDWR];
 }
 
 static inline struct file *find_readable_file(struct nfs4_file *f)
 {
-	if (f->fi_fds[O_RDWR])
-		return f->fi_fds[O_RDWR];
-	return f->fi_fds[O_RDONLY];
+	if (f->fi_fds[O_RDONLY])
+		return f->fi_fds[O_RDONLY];
+	return f->fi_fds[O_RDWR];
 }
 
 static inline struct file *find_any_file(struct nfs4_file *f)
 {
 	if (f->fi_fds[O_RDWR])
 		return f->fi_fds[O_RDWR];
-	else if (f->fi_fds[O_RDWR])
+	else if (f->fi_fds[O_WRONLY])
 		return f->fi_fds[O_WRONLY];
 	else
 		return f->fi_fds[O_RDONLY];
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 96360a83cb91..661a6cf8e826 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -2033,15 +2033,17 @@ out:
 __be32
 nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat, int access)
 {
-	struct path path = {
-		.mnt	= fhp->fh_export->ex_path.mnt,
-		.dentry	= fhp->fh_dentry,
-	};
 	__be32 err;
 
 	err = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP | access);
-	if (!err && vfs_statfs(&path, stat))
-		err = nfserr_io;
+	if (!err) {
+		struct path path = {
+			.mnt	= fhp->fh_export->ex_path.mnt,
+			.dentry	= fhp->fh_dentry,
+		};
+		if (vfs_statfs(&path, stat))
+			err = nfserr_io;
+	}
 	return err;
 }
 
diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c
index d97310f07bef..d27715103376 100644
--- a/fs/nilfs2/the_nilfs.c
+++ b/fs/nilfs2/the_nilfs.c
@@ -446,6 +446,7 @@ int load_nilfs(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi)
 	nilfs_mdt_destroy(nilfs->ns_cpfile);
 	nilfs_mdt_destroy(nilfs->ns_sufile);
 	nilfs_mdt_destroy(nilfs->ns_dat);
+	nilfs_mdt_destroy(nilfs->ns_gc_dat);
 
  failed:
 	nilfs_clear_recovery_info(&ri);
diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig
index 22c629eedd82..b388443c3a09 100644
--- a/fs/notify/Kconfig
+++ b/fs/notify/Kconfig
@@ -3,4 +3,4 @@ config FSNOTIFY
 
 source "fs/notify/dnotify/Kconfig"
 source "fs/notify/inotify/Kconfig"
-source "fs/notify/fanotify/Kconfig"
+#source "fs/notify/fanotify/Kconfig"
diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 756566fe8449..85366c78cc37 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -165,9 +165,6 @@ static bool fanotify_should_send_event(struct fsnotify_group *group,
 		 "mask=%x data=%p data_type=%d\n", __func__, group, to_tell,
 		 inode_mark, vfsmnt_mark, event_mask, data, data_type);
 
-	pr_debug("%s: group=%p vfsmount_mark=%p inode_mark=%p mask=%x\n",
-		 __func__, group, vfsmnt_mark, inode_mark, event_mask);
-
 	/* sorry, fanotify only gives a damn about files and dirs */
 	if (!S_ISREG(to_tell->i_mode) &&
 	    !S_ISDIR(to_tell->i_mode))
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 032b837fcd11..5ed8e58d7bfc 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -195,6 +195,14 @@ static int prepare_for_access_response(struct fsnotify_group *group,
 	re->fd = fd;
 
 	mutex_lock(&group->fanotify_data.access_mutex);
+
+	if (group->fanotify_data.bypass_perm) {
+		mutex_unlock(&group->fanotify_data.access_mutex);
+		kmem_cache_free(fanotify_response_event_cache, re);
+		event->response = FAN_ALLOW;
+		return 0;
+	}
+		
 	list_add_tail(&re->list, &group->fanotify_data.access_list);
 	mutex_unlock(&group->fanotify_data.access_mutex);
 
@@ -364,9 +372,28 @@ static ssize_t fanotify_write(struct file *file, const char __user *buf, size_t
 static int fanotify_release(struct inode *ignored, struct file *file)
 {
 	struct fsnotify_group *group = file->private_data;
+	struct fanotify_response_event *re, *lre;
 
 	pr_debug("%s: file=%p group=%p\n", __func__, file, group);
 
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	mutex_lock(&group->fanotify_data.access_mutex);
+
+	group->fanotify_data.bypass_perm = true;
+
+	list_for_each_entry_safe(re, lre, &group->fanotify_data.access_list, list) {
+		pr_debug("%s: found group=%p re=%p event=%p\n", __func__, group,
+			 re, re->event);
+
+		list_del_init(&re->list);
+		re->event->response = FAN_ALLOW;
+
+		kmem_cache_free(fanotify_response_event_cache, re);
+	}
+	mutex_unlock(&group->fanotify_data.access_mutex);
+
+	wake_up(&group->fanotify_data.access_waitq);
+#endif
 	/* matches the fanotify_init->fsnotify_alloc_group */
 	fsnotify_put_group(group);
 
@@ -614,7 +641,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 		__func__, flags, event_f_flags);
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
+		return -EPERM;
 
 	if (flags & ~FAN_ALL_INIT_FLAGS)
 		return -EINVAL;
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 3970392b2722..36802420d69a 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -148,13 +148,14 @@ static int send_to_group(struct inode *to_tell, struct vfsmount *mnt,
 			 const unsigned char *file_name,
 			 struct fsnotify_event **event)
 {
-	struct fsnotify_group *group = inode_mark->group;
-	__u32 inode_test_mask = (mask & ~FS_EVENT_ON_CHILD);
-	__u32 vfsmount_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+	struct fsnotify_group *group = NULL;
+	__u32 inode_test_mask = 0;
+	__u32 vfsmount_test_mask = 0;
 
-	pr_debug("%s: group=%p to_tell=%p mnt=%p mark=%p mask=%x data=%p"
-		 " data_is=%d cookie=%d event=%p\n", __func__, group, to_tell,
-		 mnt, inode_mark, mask, data, data_is, cookie, *event);
+	if (unlikely(!inode_mark && !vfsmount_mark)) {
+		BUG();
+		return 0;
+	}
 
 	/* clear ignored on inode modification */
 	if (mask & FS_MODIFY) {
@@ -168,18 +169,29 @@ static int send_to_group(struct inode *to_tell, struct vfsmount *mnt,
 
 	/* does the inode mark tell us to do something? */
 	if (inode_mark) {
+		group = inode_mark->group;
+		inode_test_mask = (mask & ~FS_EVENT_ON_CHILD);
 		inode_test_mask &= inode_mark->mask;
 		inode_test_mask &= ~inode_mark->ignored_mask;
 	}
 
 	/* does the vfsmount_mark tell us to do something? */
 	if (vfsmount_mark) {
+		vfsmount_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+		group = vfsmount_mark->group;
 		vfsmount_test_mask &= vfsmount_mark->mask;
 		vfsmount_test_mask &= ~vfsmount_mark->ignored_mask;
 		if (inode_mark)
 			vfsmount_test_mask &= ~inode_mark->ignored_mask;
 	}
 
+	pr_debug("%s: group=%p to_tell=%p mnt=%p mask=%x inode_mark=%p"
+		 " inode_test_mask=%x vfsmount_mark=%p vfsmount_test_mask=%x"
+		 " data=%p data_is=%d cookie=%d event=%p\n",
+		 __func__, group, to_tell, mnt, mask, inode_mark,
+		 inode_test_mask, vfsmount_mark, vfsmount_test_mask, data,
+		 data_is, cookie, *event);
+
 	if (!inode_test_mask && !vfsmount_test_mask)
 		return 0;
 
@@ -207,13 +219,12 @@ static int send_to_group(struct inode *to_tell, struct vfsmount *mnt,
 int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is,
 	     const unsigned char *file_name, u32 cookie)
 {
-	struct hlist_node *inode_node, *vfsmount_node;
+	struct hlist_node *inode_node = NULL, *vfsmount_node = NULL;
 	struct fsnotify_mark *inode_mark = NULL, *vfsmount_mark = NULL;
 	struct fsnotify_group *inode_group, *vfsmount_group;
 	struct fsnotify_event *event = NULL;
 	struct vfsmount *mnt;
 	int idx, ret = 0;
-	bool used_inode = false, used_vfsmount = false;
 	/* global tests shouldn't care about events on child only the specific event */
 	__u32 test_mask = (mask & ~FS_EVENT_ON_CHILD);
 
@@ -238,57 +249,50 @@ int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is,
 	    (test_mask & to_tell->i_fsnotify_mask))
 		inode_node = srcu_dereference(to_tell->i_fsnotify_marks.first,
 					      &fsnotify_mark_srcu);
-	else
-		inode_node = NULL;
 
-	if (mnt) {
-		if ((mask & FS_MODIFY) ||
-		    (test_mask & mnt->mnt_fsnotify_mask))
-			vfsmount_node = srcu_dereference(mnt->mnt_fsnotify_marks.first,
-							 &fsnotify_mark_srcu);
-		else
-			vfsmount_node = NULL;
-	} else {
-		mnt = NULL;
-		vfsmount_node = NULL;
+	if (mnt && ((mask & FS_MODIFY) ||
+		    (test_mask & mnt->mnt_fsnotify_mask))) {
+		vfsmount_node = srcu_dereference(mnt->mnt_fsnotify_marks.first,
+						 &fsnotify_mark_srcu);
+		inode_node = srcu_dereference(to_tell->i_fsnotify_marks.first,
+					      &fsnotify_mark_srcu);
 	}
 
 	while (inode_node || vfsmount_node) {
+		inode_group = vfsmount_group = NULL;
+
 		if (inode_node) {
 			inode_mark = hlist_entry(srcu_dereference(inode_node, &fsnotify_mark_srcu),
 						 struct fsnotify_mark, i.i_list);
 			inode_group = inode_mark->group;
-		} else
-			inode_group = (void *)-1;
+		}
 
 		if (vfsmount_node) {
 			vfsmount_mark = hlist_entry(srcu_dereference(vfsmount_node, &fsnotify_mark_srcu),
 							struct fsnotify_mark, m.m_list);
 			vfsmount_group = vfsmount_mark->group;
-		} else
-			vfsmount_group = (void *)-1;
+		}
 
-		if (inode_group < vfsmount_group) {
+		if (inode_group > vfsmount_group) {
 			/* handle inode */
 			send_to_group(to_tell, NULL, inode_mark, NULL, mask, data,
 				      data_is, cookie, file_name, &event);
-			used_inode = true;
-		} else if (vfsmount_group < inode_group) {
+			/* we didn't use the vfsmount_mark */
+			vfsmount_group = NULL;
+		} else if (vfsmount_group > inode_group) {
 			send_to_group(to_tell, mnt, NULL, vfsmount_mark, mask, data,
 				      data_is, cookie, file_name, &event);
-			used_vfsmount = true;
+			inode_group = NULL;
 		} else {
 			send_to_group(to_tell, mnt, inode_mark, vfsmount_mark,
 				      mask, data, data_is, cookie, file_name,
 				      &event);
-			used_vfsmount = true;
-			used_inode = true;
 		}
 
-		if (used_inode)
+		if (inode_group)
 			inode_node = srcu_dereference(inode_node->next,
 						      &fsnotify_mark_srcu);
-		if (used_vfsmount)
+		if (vfsmount_group)
 			vfsmount_node = srcu_dereference(vfsmount_node->next,
 							 &fsnotify_mark_srcu);
 	}
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index a76e0aa5cd3f..391915093fe1 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -209,7 +209,10 @@ static int ocfs2_acl_set_mode(struct inode *inode, struct buffer_head *di_bh,
 	}
 
 	inode->i_mode = new_mode;
+	inode->i_ctime = CURRENT_TIME;
 	di->i_mode = cpu_to_le16(inode->i_mode);
+	di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
+	di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
 
 	ocfs2_journal_dirty(handle, di_bh);
 
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 215e12ce1d85..592fae5007d1 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -6672,7 +6672,7 @@ int ocfs2_grab_pages(struct inode *inode, loff_t start, loff_t end,
 	last_page_bytes = PAGE_ALIGN(end);
 	index = start >> PAGE_CACHE_SHIFT;
 	do {
-		pages[numpages] = grab_cache_page(mapping, index);
+		pages[numpages] = find_or_create_page(mapping, index, GFP_NOFS);
 		if (!pages[numpages]) {
 			ret = -ENOMEM;
 			mlog_errno(ret);
diff --git a/fs/ocfs2/blockcheck.c b/fs/ocfs2/blockcheck.c
index ec6d12339593..c7ee03c22226 100644
--- a/fs/ocfs2/blockcheck.c
+++ b/fs/ocfs2/blockcheck.c
@@ -439,7 +439,7 @@ int ocfs2_block_check_validate(void *data, size_t blocksize,
 
 	ocfs2_blockcheck_inc_failure(stats);
 	mlog(ML_ERROR,
-	     "CRC32 failed: stored: %u, computed %u.  Applying ECC.\n",
+	     "CRC32 failed: stored: 0x%x, computed 0x%x. Applying ECC.\n",
 	     (unsigned int)check.bc_crc32e, (unsigned int)crc);
 
 	/* Ok, try ECC fixups */
@@ -453,7 +453,7 @@ int ocfs2_block_check_validate(void *data, size_t blocksize,
 		goto out;
 	}
 
-	mlog(ML_ERROR, "Fixed CRC32 failed: stored: %u, computed %u\n",
+	mlog(ML_ERROR, "Fixed CRC32 failed: stored: 0x%x, computed 0x%x\n",
 	     (unsigned int)check.bc_crc32e, (unsigned int)crc);
 
 	rc = -EIO;
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index 1361997cf205..cbe2f057cc28 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -977,7 +977,7 @@ static int o2net_tx_can_proceed(struct o2net_node *nn,
 int o2net_send_message_vec(u32 msg_type, u32 key, struct kvec *caller_vec,
 			   size_t caller_veclen, u8 target_node, int *status)
 {
-	int ret;
+	int ret = 0;
 	struct o2net_msg *msg = NULL;
 	size_t veclen, caller_bytes = 0;
 	struct kvec *vec = NULL;
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index f04ebcfffc4a..c49f6de0e7ab 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -3931,6 +3931,15 @@ static int ocfs2_dx_dir_rebalance(struct ocfs2_super *osb, struct inode *dir,
 		goto out_commit;
 	}
 
+	cpos = split_hash;
+	ret = ocfs2_dx_dir_new_cluster(dir, &et, cpos, handle,
+				       data_ac, meta_ac, new_dx_leaves,
+				       num_dx_leaves);
+	if (ret) {
+		mlog_errno(ret);
+		goto out_commit;
+	}
+
 	for (i = 0; i < num_dx_leaves; i++) {
 		ret = ocfs2_journal_access_dl(handle, INODE_CACHE(dir),
 					      orig_dx_leaves[i],
@@ -3939,15 +3948,14 @@ static int ocfs2_dx_dir_rebalance(struct ocfs2_super *osb, struct inode *dir,
 			mlog_errno(ret);
 			goto out_commit;
 		}
-	}
 
-	cpos = split_hash;
-	ret = ocfs2_dx_dir_new_cluster(dir, &et, cpos, handle,
-				       data_ac, meta_ac, new_dx_leaves,
-				       num_dx_leaves);
-	if (ret) {
-		mlog_errno(ret);
-		goto out_commit;
+		ret = ocfs2_journal_access_dl(handle, INODE_CACHE(dir),
+					      new_dx_leaves[i],
+					      OCFS2_JOURNAL_ACCESS_WRITE);
+		if (ret) {
+			mlog_errno(ret);
+			goto out_commit;
+		}
 	}
 
 	ocfs2_dx_dir_transfer_leaf(dir, split_hash, handle, tmp_dx_leaf,
diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h
index 4b6ae2c13b47..765298908f1d 100644
--- a/fs/ocfs2/dlm/dlmcommon.h
+++ b/fs/ocfs2/dlm/dlmcommon.h
@@ -1030,6 +1030,7 @@ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm,
 			 struct dlm_lock_resource *res);
 void dlm_clean_master_list(struct dlm_ctxt *dlm,
 			   u8 dead_node);
+void dlm_force_free_mles(struct dlm_ctxt *dlm);
 int dlm_lock_basts_flushed(struct dlm_ctxt *dlm, struct dlm_lock *lock);
 int __dlm_lockres_has_locks(struct dlm_lock_resource *res);
 int __dlm_lockres_unused(struct dlm_lock_resource *res);
diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c
index 5efdd37dfe48..901ca52bf86b 100644
--- a/fs/ocfs2/dlm/dlmdebug.c
+++ b/fs/ocfs2/dlm/dlmdebug.c
@@ -636,8 +636,14 @@ static void *lockres_seq_start(struct seq_file *m, loff_t *pos)
 	spin_lock(&dlm->track_lock);
 	if (oldres)
 		track_list = &oldres->tracking;
-	else
+	else {
 		track_list = &dlm->tracking_list;
+		if (list_empty(track_list)) {
+			dl = NULL;
+			spin_unlock(&dlm->track_lock);
+			goto bail;
+		}
+	}
 
 	list_for_each_entry(res, track_list, tracking) {
 		if (&res->tracking == &dlm->tracking_list)
@@ -660,6 +666,7 @@ static void *lockres_seq_start(struct seq_file *m, loff_t *pos)
 	} else
 		dl = NULL;
 
+bail:
 	/* passed to seq_show */
 	return dl;
 }
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 153abb5abef0..11a5c87fd7f7 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -693,6 +693,7 @@ void dlm_unregister_domain(struct dlm_ctxt *dlm)
 
 		dlm_mark_domain_leaving(dlm);
 		dlm_leave_domain(dlm);
+		dlm_force_free_mles(dlm);
 		dlm_complete_dlm_shutdown(dlm);
 	}
 	dlm_put(dlm);
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index ffb4c68dafa4..f564b0e5f80d 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -3433,3 +3433,43 @@ void dlm_lockres_release_ast(struct dlm_ctxt *dlm,
 	wake_up(&res->wq);
 	wake_up(&dlm->migration_wq);
 }
+
+void dlm_force_free_mles(struct dlm_ctxt *dlm)
+{
+	int i;
+	struct hlist_head *bucket;
+	struct dlm_master_list_entry *mle;
+	struct hlist_node *tmp, *list;
+
+	/*
+	 * We notified all other nodes that we are exiting the domain and
+	 * marked the dlm state to DLM_CTXT_LEAVING. If any mles are still
+	 * around we force free them and wake any processes that are waiting
+	 * on the mles
+	 */
+	spin_lock(&dlm->spinlock);
+	spin_lock(&dlm->master_lock);
+
+	BUG_ON(dlm->dlm_state != DLM_CTXT_LEAVING);
+	BUG_ON((find_next_bit(dlm->domain_map, O2NM_MAX_NODES, 0) < O2NM_MAX_NODES));
+
+	for (i = 0; i < DLM_HASH_BUCKETS; i++) {
+		bucket = dlm_master_hash(dlm, i);
+		hlist_for_each_safe(list, tmp, bucket) {
+			mle = hlist_entry(list, struct dlm_master_list_entry,
+					  master_hash_node);
+			if (mle->type != DLM_MLE_BLOCK) {
+				mlog(ML_ERROR, "bad mle: %p\n", mle);
+				dlm_print_one_mle(mle);
+			}
+			atomic_set(&mle->woken, 1);
+			wake_up(&mle->wq);
+
+			__dlm_unlink_mle(dlm, mle);
+			__dlm_mle_detach_hb_events(dlm, mle);
+			__dlm_put_mle(mle);
+		}
+	}
+	spin_unlock(&dlm->master_lock);
+	spin_unlock(&dlm->spinlock);
+}
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index d1ce48e1b3d6..1d596d8c4a4a 100644
--- a/fs/ocfs2/dlmglue.h
+++ b/fs/ocfs2/dlmglue.h
@@ -84,6 +84,7 @@ enum {
 	OI_LS_PARENT,
 	OI_LS_RENAME1,
 	OI_LS_RENAME2,
+	OI_LS_REFLINK_TARGET,
 };
 
 int ocfs2_dlm_init(struct ocfs2_super *osb);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 81296b4e3646..9a03c151b5ce 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -36,6 +36,7 @@
 #include <linux/writeback.h>
 #include <linux/falloc.h>
 #include <linux/quotaops.h>
+#include <linux/blkdev.h>
 
 #define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
@@ -190,8 +191,16 @@ static int ocfs2_sync_file(struct file *file, int datasync)
 	if (err)
 		goto bail;
 
-	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) {
+		/*
+		 * We still have to flush drive's caches to get data to the
+		 * platter
+		 */
+		if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER)
+			blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL,
+					   NULL, BLKDEV_IFL_WAIT);
 		goto bail;
+	}
 
 	journal = osb->journal->j_journal;
 	err = jbd2_journal_force_commit(journal);
@@ -774,7 +783,7 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from,
 	BUG_ON(abs_to > (((u64)index + 1) << PAGE_CACHE_SHIFT));
 	BUG_ON(abs_from & (inode->i_blkbits - 1));
 
-	page = grab_cache_page(mapping, index);
+	page = find_or_create_page(mapping, index, GFP_NOFS);
 	if (!page) {
 		ret = -ENOMEM;
 		mlog_errno(ret);
@@ -2329,7 +2338,7 @@ out_dio:
 	BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
 
 	if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) ||
-	    ((file->f_flags & O_DIRECT) && has_refcount)) {
+	    ((file->f_flags & O_DIRECT) && !direct_io)) {
 		ret = filemap_fdatawrite_range(file->f_mapping, pos,
 					       pos + count - 1);
 		if (ret < 0)
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 0492464916b1..eece3e05d9d0 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -488,7 +488,11 @@ static int ocfs2_read_locked_inode(struct inode *inode,
 						     OCFS2_BH_IGNORE_CACHE);
 	} else {
 		status = ocfs2_read_blocks_sync(osb, args->fi_blkno, 1, &bh);
-		if (!status)
+		/*
+		 * If buffer is in jbd, then its checksum may not have been
+		 * computed as yet.
+		 */
+		if (!status && !buffer_jbd(bh))
 			status = ocfs2_validate_inode_block(osb->sb, bh);
 	}
 	if (status < 0) {
diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c
index af2b8fe1f139..4c18f4ad93b4 100644
--- a/fs/ocfs2/mmap.c
+++ b/fs/ocfs2/mmap.c
@@ -74,9 +74,11 @@ static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh,
 	/*
 	 * Another node might have truncated while we were waiting on
 	 * cluster locks.
+	 * We don't check size == 0 before the shift. This is borrowed
+	 * from do_generic_file_read.
 	 */
-	last_index = size >> PAGE_CACHE_SHIFT;
-	if (page->index > last_index) {
+	last_index = (size - 1) >> PAGE_CACHE_SHIFT;
+	if (unlikely(!size || page->index > last_index)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -107,7 +109,7 @@ static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh,
 	 * because the "write" would invalidate their data.
 	 */
 	if (page->index == last_index)
-		len = size & ~PAGE_CACHE_MASK;
+		len = ((size - 1) & ~PAGE_CACHE_MASK) + 1;
 
 	ret = ocfs2_write_begin_nolock(mapping, pos, len, 0, &locked_page,
 				       &fsdata, di_bh, page);
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index f171b51a74f7..a00dda2e4f16 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -472,32 +472,23 @@ leave:
 	return status;
 }
 
-static int ocfs2_mknod_locked(struct ocfs2_super *osb,
-			      struct inode *dir,
-			      struct inode *inode,
-			      dev_t dev,
-			      struct buffer_head **new_fe_bh,
-			      struct buffer_head *parent_fe_bh,
-			      handle_t *handle,
-			      struct ocfs2_alloc_context *inode_ac)
+static int __ocfs2_mknod_locked(struct inode *dir,
+				struct inode *inode,
+				dev_t dev,
+				struct buffer_head **new_fe_bh,
+				struct buffer_head *parent_fe_bh,
+				handle_t *handle,
+				struct ocfs2_alloc_context *inode_ac,
+				u64 fe_blkno, u64 suballoc_loc, u16 suballoc_bit)
 {
 	int status = 0;
+	struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
 	struct ocfs2_dinode *fe = NULL;
 	struct ocfs2_extent_list *fel;
-	u64 suballoc_loc, fe_blkno = 0;
-	u16 suballoc_bit;
 	u16 feat;
 
 	*new_fe_bh = NULL;
 
-	status = ocfs2_claim_new_inode(handle, dir, parent_fe_bh,
-				       inode_ac, &suballoc_loc,
-				       &suballoc_bit, &fe_blkno);
-	if (status < 0) {
-		mlog_errno(status);
-		goto leave;
-	}
-
 	/* populate as many fields early on as possible - many of
 	 * these are used by the support functions here and in
 	 * callers. */
@@ -591,6 +582,34 @@ leave:
 	return status;
 }
 
+static int ocfs2_mknod_locked(struct ocfs2_super *osb,
+			      struct inode *dir,
+			      struct inode *inode,
+			      dev_t dev,
+			      struct buffer_head **new_fe_bh,
+			      struct buffer_head *parent_fe_bh,
+			      handle_t *handle,
+			      struct ocfs2_alloc_context *inode_ac)
+{
+	int status = 0;
+	u64 suballoc_loc, fe_blkno = 0;
+	u16 suballoc_bit;
+
+	*new_fe_bh = NULL;
+
+	status = ocfs2_claim_new_inode(handle, dir, parent_fe_bh,
+				       inode_ac, &suballoc_loc,
+				       &suballoc_bit, &fe_blkno);
+	if (status < 0) {
+		mlog_errno(status);
+		return status;
+	}
+
+	return __ocfs2_mknod_locked(dir, inode, dev, new_fe_bh,
+				    parent_fe_bh, handle, inode_ac,
+				    fe_blkno, suballoc_loc, suballoc_bit);
+}
+
 static int ocfs2_mkdir(struct inode *dir,
 		       struct dentry *dentry,
 		       int mode)
@@ -1852,61 +1871,117 @@ bail:
 	return status;
 }
 
-static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb,
-				    struct inode **ret_orphan_dir,
-				    u64 blkno,
-				    char *name,
-				    struct ocfs2_dir_lookup_result *lookup)
+static int ocfs2_lookup_lock_orphan_dir(struct ocfs2_super *osb,
+					struct inode **ret_orphan_dir,
+					struct buffer_head **ret_orphan_dir_bh)
 {
 	struct inode *orphan_dir_inode;
 	struct buffer_head *orphan_dir_bh = NULL;
-	int status = 0;
-
-	status = ocfs2_blkno_stringify(blkno, name);
-	if (status < 0) {
-		mlog_errno(status);
-		return status;
-	}
+	int ret = 0;
 
 	orphan_dir_inode = ocfs2_get_system_file_inode(osb,
 						       ORPHAN_DIR_SYSTEM_INODE,
 						       osb->slot_num);
 	if (!orphan_dir_inode) {
-		status = -ENOENT;
-		mlog_errno(status);
-		return status;
+		ret = -ENOENT;
+		mlog_errno(ret);
+		return ret;
 	}
 
 	mutex_lock(&orphan_dir_inode->i_mutex);
 
-	status = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1);
-	if (status < 0) {
-		mlog_errno(status);
-		goto leave;
+	ret = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1);
+	if (ret < 0) {
+		mutex_unlock(&orphan_dir_inode->i_mutex);
+		iput(orphan_dir_inode);
+
+		mlog_errno(ret);
+		return ret;
 	}
 
-	status = ocfs2_prepare_dir_for_insert(osb, orphan_dir_inode,
-					      orphan_dir_bh, name,
-					      OCFS2_ORPHAN_NAMELEN, lookup);
-	if (status < 0) {
-		ocfs2_inode_unlock(orphan_dir_inode, 1);
+	*ret_orphan_dir = orphan_dir_inode;
+	*ret_orphan_dir_bh = orphan_dir_bh;
 
-		mlog_errno(status);
-		goto leave;
+	return 0;
+}
+
+static int __ocfs2_prepare_orphan_dir(struct inode *orphan_dir_inode,
+				      struct buffer_head *orphan_dir_bh,
+				      u64 blkno,
+				      char *name,
+				      struct ocfs2_dir_lookup_result *lookup)
+{
+	int ret;
+	struct ocfs2_super *osb = OCFS2_SB(orphan_dir_inode->i_sb);
+
+	ret = ocfs2_blkno_stringify(blkno, name);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+
+	ret = ocfs2_prepare_dir_for_insert(osb, orphan_dir_inode,
+					   orphan_dir_bh, name,
+					   OCFS2_ORPHAN_NAMELEN, lookup);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+/**
+ * ocfs2_prepare_orphan_dir() - Prepare an orphan directory for
+ * insertion of an orphan.
+ * @osb: ocfs2 file system
+ * @ret_orphan_dir: Orphan dir inode - returned locked!
+ * @blkno: Actual block number of the inode to be inserted into orphan dir.
+ * @lookup: dir lookup result, to be passed back into functions like
+ *          ocfs2_orphan_add
+ *
+ * Returns zero on success and the ret_orphan_dir, name and lookup
+ * fields will be populated.
+ *
+ * Returns non-zero on failure. 
+ */
+static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb,
+				    struct inode **ret_orphan_dir,
+				    u64 blkno,
+				    char *name,
+				    struct ocfs2_dir_lookup_result *lookup)
+{
+	struct inode *orphan_dir_inode = NULL;
+	struct buffer_head *orphan_dir_bh = NULL;
+	int ret = 0;
+
+	ret = ocfs2_lookup_lock_orphan_dir(osb, &orphan_dir_inode,
+					   &orphan_dir_bh);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+
+	ret = __ocfs2_prepare_orphan_dir(orphan_dir_inode, orphan_dir_bh,
+					 blkno, name, lookup);
+	if (ret < 0) {
+		mlog_errno(ret);
+		goto out;
 	}
 
 	*ret_orphan_dir = orphan_dir_inode;
 
-leave:
-	if (status) {
+out:
+	brelse(orphan_dir_bh);
+
+	if (ret) {
+		ocfs2_inode_unlock(orphan_dir_inode, 1);
 		mutex_unlock(&orphan_dir_inode->i_mutex);
 		iput(orphan_dir_inode);
 	}
 
-	brelse(orphan_dir_bh);
-
-	mlog_exit(status);
-	return status;
+	mlog_exit(ret);
+	return ret;
 }
 
 static int ocfs2_orphan_add(struct ocfs2_super *osb,
@@ -2053,6 +2128,99 @@ leave:
 	return status;
 }
 
+/**
+ * ocfs2_prep_new_orphaned_file() - Prepare the orphan dir to recieve a newly
+ * allocated file. This is different from the typical 'add to orphan dir'
+ * operation in that the inode does not yet exist. This is a problem because
+ * the orphan dir stringifies the inode block number to come up with it's
+ * dirent. Obviously if the inode does not yet exist we have a chicken and egg
+ * problem. This function works around it by calling deeper into the orphan
+ * and suballoc code than other callers. Use this only by necessity.
+ * @dir: The directory which this inode will ultimately wind up under - not the
+ * orphan dir!
+ * @dir_bh: buffer_head the @dir inode block
+ * @orphan_name: string of length (CFS2_ORPHAN_NAMELEN + 1). Will be filled
+ * with the string to be used for orphan dirent. Pass back to the orphan dir
+ * code.
+ * @ret_orphan_dir: orphan dir inode returned to be passed back into orphan
+ * dir code.
+ * @ret_di_blkno: block number where the new inode will be allocated.
+ * @orphan_insert: Dir insert context to be passed back into orphan dir code.
+ * @ret_inode_ac: Inode alloc context to be passed back to the allocator.
+ *
+ * Returns zero on success and the ret_orphan_dir, name and lookup
+ * fields will be populated.
+ *
+ * Returns non-zero on failure. 
+ */
+static int ocfs2_prep_new_orphaned_file(struct inode *dir,
+					struct buffer_head *dir_bh,
+					char *orphan_name,
+					struct inode **ret_orphan_dir,
+					u64 *ret_di_blkno,
+					struct ocfs2_dir_lookup_result *orphan_insert,
+					struct ocfs2_alloc_context **ret_inode_ac)
+{
+	int ret;
+	u64 di_blkno;
+	struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
+	struct inode *orphan_dir = NULL;
+	struct buffer_head *orphan_dir_bh = NULL;
+	struct ocfs2_alloc_context *inode_ac = NULL;
+
+	ret = ocfs2_lookup_lock_orphan_dir(osb, &orphan_dir, &orphan_dir_bh);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+
+	/* reserve an inode spot */
+	ret = ocfs2_reserve_new_inode(osb, &inode_ac);
+	if (ret < 0) {
+		if (ret != -ENOSPC)
+			mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_find_new_inode_loc(dir, dir_bh, inode_ac,
+				       &di_blkno);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = __ocfs2_prepare_orphan_dir(orphan_dir, orphan_dir_bh,
+					 di_blkno, orphan_name, orphan_insert);
+	if (ret < 0) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+out:
+	if (ret == 0) {
+		*ret_orphan_dir = orphan_dir;
+		*ret_di_blkno = di_blkno;
+		*ret_inode_ac = inode_ac;
+		/*
+		 * orphan_name and orphan_insert are already up to
+		 * date via prepare_orphan_dir
+		 */
+	} else {
+		/* Unroll reserve_new_inode* */
+		if (inode_ac)
+			ocfs2_free_alloc_context(inode_ac);
+
+		/* Unroll orphan dir locking */
+		mutex_unlock(&orphan_dir->i_mutex);
+		ocfs2_inode_unlock(orphan_dir, 1);
+		iput(orphan_dir);
+	}
+
+	brelse(orphan_dir_bh);
+
+	return 0;
+}
+
 int ocfs2_create_inode_in_orphan(struct inode *dir,
 				 int mode,
 				 struct inode **new_inode)
@@ -2068,6 +2236,8 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
 	struct buffer_head *new_di_bh = NULL;
 	struct ocfs2_alloc_context *inode_ac = NULL;
 	struct ocfs2_dir_lookup_result orphan_insert = { NULL, };
+	u64 uninitialized_var(di_blkno), suballoc_loc;
+	u16 suballoc_bit;
 
 	status = ocfs2_inode_lock(dir, &parent_di_bh, 1);
 	if (status < 0) {
@@ -2076,20 +2246,9 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
 		return status;
 	}
 
-	/*
-	 * We give the orphan dir the root blkno to fake an orphan name,
-	 * and allocate enough space for our insertion.
-	 */
-	status = ocfs2_prepare_orphan_dir(osb, &orphan_dir,
-					  osb->root_blkno,
-					  orphan_name, &orphan_insert);
-	if (status < 0) {
-		mlog_errno(status);
-		goto leave;
-	}
-
-	/* reserve an inode spot */
-	status = ocfs2_reserve_new_inode(osb, &inode_ac);
+	status = ocfs2_prep_new_orphaned_file(dir, parent_di_bh,
+					      orphan_name, &orphan_dir,
+					      &di_blkno, &orphan_insert, &inode_ac);
 	if (status < 0) {
 		if (status != -ENOSPC)
 			mlog_errno(status);
@@ -2116,17 +2275,20 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
 		goto leave;
 	did_quota_inode = 1;
 
-	inode->i_nlink = 0;
-	/* do the real work now. */
-	status = ocfs2_mknod_locked(osb, dir, inode,
-				    0, &new_di_bh, parent_di_bh, handle,
-				    inode_ac);
+	status = ocfs2_claim_new_inode_at_loc(handle, dir, inode_ac,
+					      &suballoc_loc,
+					      &suballoc_bit, di_blkno);
 	if (status < 0) {
 		mlog_errno(status);
 		goto leave;
 	}
 
-	status = ocfs2_blkno_stringify(OCFS2_I(inode)->ip_blkno, orphan_name);
+	inode->i_nlink = 0;
+	/* do the real work now. */
+	status = __ocfs2_mknod_locked(dir, inode,
+				      0, &new_di_bh, parent_di_bh, handle,
+				      inode_ac, di_blkno, suballoc_loc,
+				      suballoc_bit);
 	if (status < 0) {
 		mlog_errno(status);
 		goto leave;
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index 33f1c9a8258d..fa31d05e41b7 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -235,18 +235,31 @@
 #define OCFS2_HAS_REFCOUNT_FL   (0x0010)
 
 /* Inode attributes, keep in sync with EXT2 */
-#define OCFS2_SECRM_FL		(0x00000001)	/* Secure deletion */
-#define OCFS2_UNRM_FL		(0x00000002)	/* Undelete */
-#define OCFS2_COMPR_FL		(0x00000004)	/* Compress file */
-#define OCFS2_SYNC_FL		(0x00000008)	/* Synchronous updates */
-#define OCFS2_IMMUTABLE_FL	(0x00000010)	/* Immutable file */
-#define OCFS2_APPEND_FL		(0x00000020)	/* writes to file may only append */
-#define OCFS2_NODUMP_FL		(0x00000040)	/* do not dump file */
-#define OCFS2_NOATIME_FL	(0x00000080)	/* do not update atime */
-#define OCFS2_DIRSYNC_FL	(0x00010000)	/* dirsync behaviour (directories only) */
-
-#define OCFS2_FL_VISIBLE	(0x000100FF)	/* User visible flags */
-#define OCFS2_FL_MODIFIABLE	(0x000100FF)	/* User modifiable flags */
+#define OCFS2_SECRM_FL			FS_SECRM_FL	/* Secure deletion */
+#define OCFS2_UNRM_FL			FS_UNRM_FL	/* Undelete */
+#define OCFS2_COMPR_FL			FS_COMPR_FL	/* Compress file */
+#define OCFS2_SYNC_FL			FS_SYNC_FL	/* Synchronous updates */
+#define OCFS2_IMMUTABLE_FL		FS_IMMUTABLE_FL	/* Immutable file */
+#define OCFS2_APPEND_FL			FS_APPEND_FL	/* writes to file may only append */
+#define OCFS2_NODUMP_FL			FS_NODUMP_FL	/* do not dump file */
+#define OCFS2_NOATIME_FL		FS_NOATIME_FL	/* do not update atime */
+/* Reserved for compression usage... */
+#define OCFS2_DIRTY_FL			FS_DIRTY_FL
+#define OCFS2_COMPRBLK_FL		FS_COMPRBLK_FL	/* One or more compressed clusters */
+#define OCFS2_NOCOMP_FL			FS_NOCOMP_FL	/* Don't compress */
+#define OCFS2_ECOMPR_FL			FS_ECOMPR_FL	/* Compression error */
+/* End compression flags --- maybe not all used */
+#define OCFS2_BTREE_FL			FS_BTREE_FL	/* btree format dir */
+#define OCFS2_INDEX_FL			FS_INDEX_FL	/* hash-indexed directory */
+#define OCFS2_IMAGIC_FL			FS_IMAGIC_FL	/* AFS directory */
+#define OCFS2_JOURNAL_DATA_FL		FS_JOURNAL_DATA_FL /* Reserved for ext3 */
+#define OCFS2_NOTAIL_FL			FS_NOTAIL_FL	/* file tail should not be merged */
+#define OCFS2_DIRSYNC_FL		FS_DIRSYNC_FL	/* dirsync behaviour (directories only) */
+#define OCFS2_TOPDIR_FL			FS_TOPDIR_FL	/* Top of directory hierarchies*/
+#define OCFS2_RESERVED_FL		FS_RESERVED_FL	/* reserved for ext2 lib */
+
+#define OCFS2_FL_VISIBLE		FS_FL_USER_VISIBLE	/* User visible flags */
+#define OCFS2_FL_MODIFIABLE		FS_FL_USER_MODIFIABLE	/* User modifiable flags */
 
 /*
  * Extent record flags (e_node.leaf.flags)
diff --git a/fs/ocfs2/ocfs2_ioctl.h b/fs/ocfs2/ocfs2_ioctl.h
index 2d3420af1a83..5d241505690b 100644
--- a/fs/ocfs2/ocfs2_ioctl.h
+++ b/fs/ocfs2/ocfs2_ioctl.h
@@ -23,10 +23,10 @@
 /*
  * ioctl commands
  */
-#define OCFS2_IOC_GETFLAGS	_IOR('f', 1, long)
-#define OCFS2_IOC_SETFLAGS	_IOW('f', 2, long)
-#define OCFS2_IOC32_GETFLAGS	_IOR('f', 1, int)
-#define OCFS2_IOC32_SETFLAGS	_IOW('f', 2, int)
+#define OCFS2_IOC_GETFLAGS	FS_IOC_GETFLAGS
+#define OCFS2_IOC_SETFLAGS	FS_IOC_SETFLAGS
+#define OCFS2_IOC32_GETFLAGS	FS_IOC32_GETFLAGS
+#define OCFS2_IOC32_SETFLAGS	FS_IOC32_SETFLAGS
 
 /*
  * Space reservation / allocation / free ioctls and argument structure
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 73a11ccfd4c2..efdd75607406 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -2960,7 +2960,7 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle,
 		if (map_end & (PAGE_CACHE_SIZE - 1))
 			to = map_end & (PAGE_CACHE_SIZE - 1);
 
-		page = grab_cache_page(mapping, page_index);
+		page = find_or_create_page(mapping, page_index, GFP_NOFS);
 
 		/*
 		 * In case PAGE_CACHE_SIZE <= CLUSTER_SIZE, This page
@@ -3179,7 +3179,8 @@ static int ocfs2_cow_sync_writeback(struct super_block *sb,
 		if (map_end > end)
 			map_end = end;
 
-		page = grab_cache_page(context->inode->i_mapping, page_index);
+		page = find_or_create_page(context->inode->i_mapping,
+					   page_index, GFP_NOFS);
 		BUG_ON(!page);
 
 		wait_on_page_writeback(page);
@@ -4200,8 +4201,9 @@ static int __ocfs2_reflink(struct dentry *old_dentry,
 		goto out;
 	}
 
-	mutex_lock(&new_inode->i_mutex);
-	ret = ocfs2_inode_lock(new_inode, &new_bh, 1);
+	mutex_lock_nested(&new_inode->i_mutex, I_MUTEX_CHILD);
+	ret = ocfs2_inode_lock_nested(new_inode, &new_bh, 1,
+				      OI_LS_REFLINK_TARGET);
 	if (ret) {
 		mlog_errno(ret);
 		goto out_unlock;
diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c
index d8b6e4259b80..3e78db361bc7 100644
--- a/fs/ocfs2/reservations.c
+++ b/fs/ocfs2/reservations.c
@@ -732,25 +732,23 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
 			   struct ocfs2_alloc_reservation *resv,
 			   int *cstart, int *clen)
 {
-	unsigned int wanted = *clen;
-
 	if (resv == NULL || ocfs2_resmap_disabled(resmap))
 		return -ENOSPC;
 
 	spin_lock(&resv_lock);
 
-	/*
-	 * We don't want to over-allocate for temporary
-	 * windows. Otherwise, we run the risk of fragmenting the
-	 * allocation space.
-	 */
-	wanted = ocfs2_resv_window_bits(resmap, resv);
-	if ((resv->r_flags & OCFS2_RESV_FLAG_TMP) || wanted < *clen)
-		wanted = *clen;
-
 	if (ocfs2_resv_empty(resv)) {
-		mlog(0, "empty reservation, find new window\n");
+		/*
+		 * We don't want to over-allocate for temporary
+		 * windows. Otherwise, we run the risk of fragmenting the
+		 * allocation space.
+		 */
+		unsigned int wanted = ocfs2_resv_window_bits(resmap, resv);
 
+		if ((resv->r_flags & OCFS2_RESV_FLAG_TMP) || wanted < *clen)
+			wanted = *clen;
+
+		mlog(0, "empty reservation, find new window\n");
 		/*
 		 * Try to get a window here. If it works, we must fall
 		 * through and test the bitmap . This avoids some
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index a8e6a95a353f..849c2f0e0a0e 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -57,11 +57,28 @@ struct ocfs2_suballoc_result {
 	u64		sr_bg_blkno;	/* The bg we allocated from.  Set
 					   to 0 when a block group is
 					   contiguous. */
+	u64		sr_bg_stable_blkno; /*
+					     * Doesn't change, always
+					     * set to target block
+					     * group descriptor
+					     * block.
+					     */
 	u64		sr_blkno;	/* The first allocated block */
 	unsigned int	sr_bit_offset;	/* The bit in the bg */
 	unsigned int	sr_bits;	/* How many bits we claimed */
 };
 
+static u64 ocfs2_group_from_res(struct ocfs2_suballoc_result *res)
+{
+	if (res->sr_blkno == 0)
+		return 0;
+
+	if (res->sr_bg_blkno)
+		return res->sr_bg_blkno;
+
+	return ocfs2_which_suballoc_group(res->sr_blkno, res->sr_bit_offset);
+}
+
 static inline void ocfs2_debug_bg(struct ocfs2_group_desc *bg);
 static inline void ocfs2_debug_suballoc_inode(struct ocfs2_dinode *fe);
 static inline u16 ocfs2_find_victim_chain(struct ocfs2_chain_list *cl);
@@ -138,6 +155,10 @@ void ocfs2_free_ac_resource(struct ocfs2_alloc_context *ac)
 	brelse(ac->ac_bh);
 	ac->ac_bh = NULL;
 	ac->ac_resv = NULL;
+	if (ac->ac_find_loc_priv) {
+		kfree(ac->ac_find_loc_priv);
+		ac->ac_find_loc_priv = NULL;
+	}
 }
 
 void ocfs2_free_alloc_context(struct ocfs2_alloc_context *ac)
@@ -336,7 +357,7 @@ out:
 static void ocfs2_bg_discontig_add_extent(struct ocfs2_super *osb,
 					  struct ocfs2_group_desc *bg,
 					  struct ocfs2_chain_list *cl,
-					  u64 p_blkno, u32 clusters)
+					  u64 p_blkno, unsigned int clusters)
 {
 	struct ocfs2_extent_list *el = &bg->bg_list;
 	struct ocfs2_extent_rec *rec;
@@ -348,7 +369,7 @@ static void ocfs2_bg_discontig_add_extent(struct ocfs2_super *osb,
 	rec->e_blkno = cpu_to_le64(p_blkno);
 	rec->e_cpos = cpu_to_le32(le16_to_cpu(bg->bg_bits) /
 				  le16_to_cpu(cl->cl_bpc));
-	rec->e_leaf_clusters = cpu_to_le32(clusters);
+	rec->e_leaf_clusters = cpu_to_le16(clusters);
 	le16_add_cpu(&bg->bg_bits, clusters * le16_to_cpu(cl->cl_bpc));
 	le16_add_cpu(&bg->bg_free_bits_count,
 		     clusters * le16_to_cpu(cl->cl_bpc));
@@ -1678,6 +1699,15 @@ static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac,
 	if (!ret)
 		ocfs2_bg_discontig_fix_result(ac, gd, res);
 
+	/*
+	 * sr_bg_blkno might have been changed by
+	 * ocfs2_bg_discontig_fix_result
+	 */
+	res->sr_bg_stable_blkno = group_bh->b_blocknr;
+
+	if (ac->ac_find_loc_only)
+		goto out_loc_only;
+
 	ret = ocfs2_alloc_dinode_update_counts(alloc_inode, handle, ac->ac_bh,
 					       res->sr_bits,
 					       le16_to_cpu(gd->bg_chain));
@@ -1691,6 +1721,7 @@ static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac,
 	if (ret < 0)
 		mlog_errno(ret);
 
+out_loc_only:
 	*bits_left = le16_to_cpu(gd->bg_free_bits_count);
 
 out:
@@ -1708,7 +1739,6 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 {
 	int status;
 	u16 chain;
-	u32 tmp_used;
 	u64 next_group;
 	struct inode *alloc_inode = ac->ac_inode;
 	struct buffer_head *group_bh = NULL;
@@ -1770,6 +1800,11 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 	if (!status)
 		ocfs2_bg_discontig_fix_result(ac, bg, res);
 
+	/*
+	 * sr_bg_blkno might have been changed by
+	 * ocfs2_bg_discontig_fix_result
+	 */
+	res->sr_bg_stable_blkno = group_bh->b_blocknr;
 
 	/*
 	 * Keep track of previous block descriptor read. When
@@ -1796,22 +1831,17 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 		}
 	}
 
-	/* Ok, claim our bits now: set the info on dinode, chainlist
-	 * and then the group */
-	status = ocfs2_journal_access_di(handle,
-					 INODE_CACHE(alloc_inode),
-					 ac->ac_bh,
-					 OCFS2_JOURNAL_ACCESS_WRITE);
-	if (status < 0) {
+	if (ac->ac_find_loc_only)
+		goto out_loc_only;
+
+	status = ocfs2_alloc_dinode_update_counts(alloc_inode, handle,
+						  ac->ac_bh, res->sr_bits,
+						  chain);
+	if (status) {
 		mlog_errno(status);
 		goto bail;
 	}
 
-	tmp_used = le32_to_cpu(fe->id1.bitmap1.i_used);
-	fe->id1.bitmap1.i_used = cpu_to_le32(res->sr_bits + tmp_used);
-	le32_add_cpu(&cl->cl_recs[chain].c_free, -res->sr_bits);
-	ocfs2_journal_dirty(handle, ac->ac_bh);
-
 	status = ocfs2_block_group_set_bits(handle,
 					    alloc_inode,
 					    bg,
@@ -1826,6 +1856,7 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 	mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits,
 	     (unsigned long long)le64_to_cpu(fe->i_blkno));
 
+out_loc_only:
 	*bits_left = le16_to_cpu(bg->bg_free_bits_count);
 bail:
 	brelse(group_bh);
@@ -1845,6 +1876,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 	int status;
 	u16 victim, i;
 	u16 bits_left = 0;
+	u64 hint = ac->ac_last_group;
 	struct ocfs2_chain_list *cl;
 	struct ocfs2_dinode *fe;
 
@@ -1872,7 +1904,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 		goto bail;
 	}
 
-	res->sr_bg_blkno = ac->ac_last_group;
+	res->sr_bg_blkno = hint;
 	if (res->sr_bg_blkno) {
 		/* Attempt to short-circuit the usual search mechanism
 		 * by jumping straight to the most recently used
@@ -1896,8 +1928,10 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 
 	status = ocfs2_search_chain(ac, handle, bits_wanted, min_bits,
 				    res, &bits_left);
-	if (!status)
+	if (!status) {
+		hint = ocfs2_group_from_res(res);
 		goto set_hint;
+	}
 	if (status < 0 && status != -ENOSPC) {
 		mlog_errno(status);
 		goto bail;
@@ -1920,8 +1954,10 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 		ac->ac_chain = i;
 		status = ocfs2_search_chain(ac, handle, bits_wanted, min_bits,
 					    res, &bits_left);
-		if (!status)
+		if (!status) {
+			hint = ocfs2_group_from_res(res);
 			break;
+		}
 		if (status < 0 && status != -ENOSPC) {
 			mlog_errno(status);
 			goto bail;
@@ -1936,7 +1972,7 @@ set_hint:
 		if (bits_left < min_bits)
 			ac->ac_last_group = 0;
 		else
-			ac->ac_last_group = res->sr_bg_blkno;
+			ac->ac_last_group = hint;
 	}
 
 bail:
@@ -2016,6 +2052,136 @@ static inline void ocfs2_save_inode_ac_group(struct inode *dir,
 	OCFS2_I(dir)->ip_last_used_slot = ac->ac_alloc_slot;
 }
 
+int ocfs2_find_new_inode_loc(struct inode *dir,
+			     struct buffer_head *parent_fe_bh,
+			     struct ocfs2_alloc_context *ac,
+			     u64 *fe_blkno)
+{
+	int ret;
+	handle_t *handle = NULL;
+	struct ocfs2_suballoc_result *res;
+
+	BUG_ON(!ac);
+	BUG_ON(ac->ac_bits_given != 0);
+	BUG_ON(ac->ac_bits_wanted != 1);
+	BUG_ON(ac->ac_which != OCFS2_AC_USE_INODE);
+
+	res = kzalloc(sizeof(*res), GFP_NOFS);
+	if (res == NULL) {
+		ret = -ENOMEM;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ocfs2_init_inode_ac_group(dir, parent_fe_bh, ac);
+
+	/*
+	 * The handle started here is for chain relink. Alternatively,
+	 * we could just disable relink for these calls.
+	 */
+	handle = ocfs2_start_trans(OCFS2_SB(dir->i_sb), OCFS2_SUBALLOC_ALLOC);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		handle = NULL;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	/*
+	 * This will instruct ocfs2_claim_suballoc_bits and
+	 * ocfs2_search_one_group to search but save actual allocation
+	 * for later.
+	 */
+	ac->ac_find_loc_only = 1;
+
+	ret = ocfs2_claim_suballoc_bits(ac, handle, 1, 1, res);
+	if (ret < 0) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ac->ac_find_loc_priv = res;
+	*fe_blkno = res->sr_blkno;
+
+out:
+	if (handle)
+		ocfs2_commit_trans(OCFS2_SB(dir->i_sb), handle);
+
+	if (ret)
+		kfree(res);
+
+	return ret;
+}
+
+int ocfs2_claim_new_inode_at_loc(handle_t *handle,
+				 struct inode *dir,
+				 struct ocfs2_alloc_context *ac,
+				 u64 *suballoc_loc,
+				 u16 *suballoc_bit,
+				 u64 di_blkno)
+{
+	int ret;
+	u16 chain;
+	struct ocfs2_suballoc_result *res = ac->ac_find_loc_priv;
+	struct buffer_head *bg_bh = NULL;
+	struct ocfs2_group_desc *bg;
+	struct ocfs2_dinode *di = (struct ocfs2_dinode *) ac->ac_bh->b_data;
+
+	/*
+	 * Since di_blkno is being passed back in, we check for any
+	 * inconsistencies which may have happened between
+	 * calls. These are code bugs as di_blkno is not expected to
+	 * change once returned from ocfs2_find_new_inode_loc()
+	 */
+	BUG_ON(res->sr_blkno != di_blkno);
+
+	ret = ocfs2_read_group_descriptor(ac->ac_inode, di,
+					  res->sr_bg_stable_blkno, &bg_bh);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	bg = (struct ocfs2_group_desc *) bg_bh->b_data;
+	chain = le16_to_cpu(bg->bg_chain);
+
+	ret = ocfs2_alloc_dinode_update_counts(ac->ac_inode, handle,
+					       ac->ac_bh, res->sr_bits,
+					       chain);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_block_group_set_bits(handle,
+					 ac->ac_inode,
+					 bg,
+					 bg_bh,
+					 res->sr_bit_offset,
+					 res->sr_bits);
+	if (ret < 0) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits,
+	     (unsigned long long)di_blkno);
+
+	atomic_inc(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs);
+
+	BUG_ON(res->sr_bits != 1);
+
+	*suballoc_loc = res->sr_bg_blkno;
+	*suballoc_bit = res->sr_bit_offset;
+	ac->ac_bits_given++;
+	ocfs2_save_inode_ac_group(dir, ac);
+
+out:
+	brelse(bg_bh);
+
+	return ret;
+}
+
 int ocfs2_claim_new_inode(handle_t *handle,
 			  struct inode *dir,
 			  struct buffer_head *parent_fe_bh,
@@ -2567,7 +2733,8 @@ out:
  * suballoc_bit.
  */
 static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno,
-				       u16 *suballoc_slot, u16 *suballoc_bit)
+				       u16 *suballoc_slot, u64 *group_blkno,
+				       u16 *suballoc_bit)
 {
 	int status;
 	struct buffer_head *inode_bh = NULL;
@@ -2604,6 +2771,8 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno,
 		*suballoc_slot = le16_to_cpu(inode_fe->i_suballoc_slot);
 	if (suballoc_bit)
 		*suballoc_bit = le16_to_cpu(inode_fe->i_suballoc_bit);
+	if (group_blkno)
+		*group_blkno = le64_to_cpu(inode_fe->i_suballoc_loc);
 
 bail:
 	brelse(inode_bh);
@@ -2621,7 +2790,8 @@ bail:
  */
 static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb,
 				   struct inode *suballoc,
-				   struct buffer_head *alloc_bh, u64 blkno,
+				   struct buffer_head *alloc_bh,
+				   u64 group_blkno, u64 blkno,
 				   u16 bit, int *res)
 {
 	struct ocfs2_dinode *alloc_di;
@@ -2642,10 +2812,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb,
 		goto bail;
 	}
 
-	if (alloc_di->i_suballoc_loc)
-		bg_blkno = le64_to_cpu(alloc_di->i_suballoc_loc);
-	else
-		bg_blkno = ocfs2_which_suballoc_group(blkno, bit);
+	bg_blkno = group_blkno ? group_blkno :
+		   ocfs2_which_suballoc_group(blkno, bit);
 	status = ocfs2_read_group_descriptor(suballoc, alloc_di, bg_blkno,
 					     &group_bh);
 	if (status < 0) {
@@ -2680,6 +2848,7 @@ bail:
 int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res)
 {
 	int status;
+	u64 group_blkno = 0;
 	u16 suballoc_bit = 0, suballoc_slot = 0;
 	struct inode *inode_alloc_inode;
 	struct buffer_head *alloc_bh = NULL;
@@ -2687,7 +2856,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res)
 	mlog_entry("blkno: %llu", (unsigned long long)blkno);
 
 	status = ocfs2_get_suballoc_slot_bit(osb, blkno, &suballoc_slot,
-					     &suballoc_bit);
+					     &group_blkno, &suballoc_bit);
 	if (status < 0) {
 		mlog(ML_ERROR, "get alloc slot and bit failed %d\n", status);
 		goto bail;
@@ -2715,7 +2884,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res)
 	}
 
 	status = ocfs2_test_suballoc_bit(osb, inode_alloc_inode, alloc_bh,
-					 blkno, suballoc_bit, res);
+					 group_blkno, blkno, suballoc_bit, res);
 	if (status < 0)
 		mlog(ML_ERROR, "test suballoc bit failed %d\n", status);
 
diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h
index a017dd3ee7d9..b8afabfeede4 100644
--- a/fs/ocfs2/suballoc.h
+++ b/fs/ocfs2/suballoc.h
@@ -56,6 +56,9 @@ struct ocfs2_alloc_context {
 	u64    ac_max_block;  /* Highest block number to allocate. 0 is
 				 is the same as ~0 - unlimited */
 
+	int    ac_find_loc_only;  /* hack for reflink operation ordering */
+	struct ocfs2_suballoc_result *ac_find_loc_priv; /* */
+
 	struct ocfs2_alloc_reservation	*ac_resv;
 };
 
@@ -197,4 +200,22 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_extent_tree *et,
 			  struct ocfs2_alloc_context **meta_ac);
 
 int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res);
+
+
+
+/*
+ * The following two interfaces are for ocfs2_create_inode_in_orphan().
+ */
+int ocfs2_find_new_inode_loc(struct inode *dir,
+			     struct buffer_head *parent_fe_bh,
+			     struct ocfs2_alloc_context *ac,
+			     u64 *fe_blkno);
+
+int ocfs2_claim_new_inode_at_loc(handle_t *handle,
+				 struct inode *dir,
+				 struct ocfs2_alloc_context *ac,
+				 u64 *suballoc_loc,
+				 u16 *suballoc_bit,
+				 u64 di_blkno);
+
 #endif /* _CHAINALLOC_H_ */
diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c
index 32499d213fc4..9975457c981f 100644
--- a/fs/ocfs2/symlink.c
+++ b/fs/ocfs2/symlink.c
@@ -128,7 +128,7 @@ static void *ocfs2_fast_follow_link(struct dentry *dentry,
 	}
 
 	/* Fast symlinks can't be large */
-	len = strlen(target);
+	len = strnlen(target, ocfs2_fast_symlink_chars(inode->i_sb));
 	link = kzalloc(len + 1, GFP_NOFS);
 	if (!link) {
 		status = -ENOMEM;
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index d03469f61801..06fa5e77c40e 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -1286,13 +1286,11 @@ int ocfs2_xattr_get_nolock(struct inode *inode,
 	xis.inode_bh = xbs.inode_bh = di_bh;
 	di = (struct ocfs2_dinode *)di_bh->b_data;
 
-	down_read(&oi->ip_xattr_sem);
 	ret = ocfs2_xattr_ibody_get(inode, name_index, name, buffer,
 				    buffer_size, &xis);
 	if (ret == -ENODATA && di->i_xattr_loc)
 		ret = ocfs2_xattr_block_get(inode, name_index, name, buffer,
 					    buffer_size, &xbs);
-	up_read(&oi->ip_xattr_sem);
 
 	return ret;
 }
@@ -1316,8 +1314,10 @@ static int ocfs2_xattr_get(struct inode *inode,
 		mlog_errno(ret);
 		return ret;
 	}
+	down_read(&OCFS2_I(inode)->ip_xattr_sem);
 	ret = ocfs2_xattr_get_nolock(inode, di_bh, name_index,
 				     name, buffer, buffer_size);
+	up_read(&OCFS2_I(inode)->ip_xattr_sem);
 
 	ocfs2_inode_unlock(inode, 0);
 
diff --git a/fs/proc/base.c b/fs/proc/base.c
index a1c43e7c8a7b..8e4addaa5424 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2675,7 +2675,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 	INF("auxv",       S_IRUSR, proc_pid_auxv),
 	ONE("status",     S_IRUGO, proc_pid_status),
 	ONE("personality", S_IRUSR, proc_pid_personality),
-	INF("limits",	  S_IRUSR, proc_pid_limits),
+	INF("limits",	  S_IRUGO, proc_pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",      S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
@@ -3011,7 +3011,7 @@ static const struct pid_entry tid_base_stuff[] = {
 	INF("auxv",      S_IRUSR, proc_pid_auxv),
 	ONE("status",    S_IRUGO, proc_pid_status),
 	ONE("personality", S_IRUSR, proc_pid_personality),
-	INF("limits",	 S_IRUSR, proc_pid_limits),
+	INF("limits",	 S_IRUGO, proc_pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",     S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
diff --git a/fs/proc/page.c b/fs/proc/page.c
index 180cf5a0bd67..3b8b45660331 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -146,7 +146,7 @@ u64 stable_page_flags(struct page *page)
 	u |= kpf_copy_bit(k, KPF_HWPOISON,	PG_hwpoison);
 #endif
 
-#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR
+#ifdef CONFIG_ARCH_USES_PG_UNCACHED
 	u |= kpf_copy_bit(k, KPF_UNCACHED,	PG_uncached);
 #endif
 
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 439fc1f1c1c4..1dbca4e8cc16 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -224,7 +224,8 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
 	/* We don't show the stack guard page in /proc/maps */
 	start = vma->vm_start;
 	if (vma->vm_flags & VM_GROWSDOWN)
-		start += PAGE_SIZE;
+		if (!vma_stack_continue(vma->vm_prev, vma->vm_start))
+			start += PAGE_SIZE;
 
 	seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
 			start,
@@ -362,13 +363,13 @@ static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 			mss->referenced += PAGE_SIZE;
 		mapcount = page_mapcount(page);
 		if (mapcount >= 2) {
-			if (pte_dirty(ptent))
+			if (pte_dirty(ptent) || PageDirty(page))
 				mss->shared_dirty += PAGE_SIZE;
 			else
 				mss->shared_clean += PAGE_SIZE;
 			mss->pss += (PAGE_SIZE << PSS_SHIFT) / mapcount;
 		} else {
-			if (pte_dirty(ptent))
+			if (pte_dirty(ptent) || PageDirty(page))
 				mss->private_dirty += PAGE_SIZE;
 			else
 				mss->private_clean += PAGE_SIZE;
diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index 91c817ff02c3..2367fb3f70bc 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -163,7 +163,7 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer,
 
 static const struct file_operations proc_vmcore_operations = {
 	.read		= read_vmcore,
-	.llseek		= generic_file_llseek,
+	.llseek		= default_llseek,
 };
 
 static struct vmcore* __init get_new_element(void)
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index f53505de0712..5cbb81e134ac 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -170,6 +170,7 @@ int reiserfs_prepare_write(struct file *f, struct page *page,
 int reiserfs_unpack(struct inode *inode, struct file *filp)
 {
 	int retval = 0;
+	int depth;
 	int index;
 	struct page *page;
 	struct address_space *mapping;
@@ -188,8 +189,8 @@ int reiserfs_unpack(struct inode *inode, struct file *filp)
 	/* we need to make sure nobody is changing the file size beneath
 	 ** us
 	 */
-	mutex_lock(&inode->i_mutex);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb);
+	depth = reiserfs_write_lock_once(inode->i_sb);
 
 	write_from = inode->i_size & (blocksize - 1);
 	/* if we are on a block boundary, we are already unpacked.  */
@@ -224,6 +225,6 @@ int reiserfs_unpack(struct inode *inode, struct file *filp)
 
       out:
 	mutex_unlock(&inode->i_mutex);
-	reiserfs_write_unlock(inode->i_sb);
+	reiserfs_write_unlock_once(inode->i_sb, depth);
 	return retval;
 }
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 1b27b5688f62..da3fefe91a8f 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -340,7 +340,7 @@ static int sysfs_open_file(struct inode *inode, struct file *file)
 	char *p;
 
 	p = d_path(&file->f_path, last_sysfs_file, sizeof(last_sysfs_file));
-	if (p)
+	if (!IS_ERR(p))
 		memmove(last_sysfs_file, p, strlen(p) + 1);
 
 	/* need attr_sd for attr and ops, its parent for kobj */
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 15412fe15c3a..b552f816de15 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -852,8 +852,8 @@ xfs_convert_page(
 		SetPageUptodate(page);
 
 	if (count) {
-		wbc->nr_to_write--;
-		if (wbc->nr_to_write <= 0)
+		if (--wbc->nr_to_write <= 0 &&
+		    wbc->sync_mode == WB_SYNC_NONE)
 			done = 1;
 	}
 	xfs_start_page_writeback(page, !page_dirty, count);
@@ -1068,7 +1068,7 @@ xfs_vm_writepage(
 	 * by themselves.
 	 */
 	if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC)
-		goto out_fail;
+		goto redirty;
 
 	/*
 	 * We need a transaction if there are delalloc or unwritten buffers
@@ -1080,7 +1080,7 @@ xfs_vm_writepage(
 	 */
 	xfs_count_page_state(page, &delalloc, &unwritten);
 	if ((current->flags & PF_FSTRANS) && (delalloc || unwritten))
-		goto out_fail;
+		goto redirty;
 
 	/* Is this page beyond the end of the file? */
 	offset = i_size_read(inode);
@@ -1245,12 +1245,15 @@ error:
 	if (iohead)
 		xfs_cancel_ioend(iohead);
 
+	if (err == -EAGAIN)
+		goto redirty;
+
 	xfs_aops_discard_page(page);
 	ClearPageUptodate(page);
 	unlock_page(page);
 	return err;
 
-out_fail:
+redirty:
 	redirty_page_for_writepage(wbc, page);
 	unlock_page(page);
 	return 0;
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index b93ea3342281..1846a0dd7035 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -440,12 +440,7 @@ _xfs_buf_find(
 		ASSERT(btp == bp->b_target);
 		if (bp->b_file_offset == range_base &&
 		    bp->b_buffer_length == range_length) {
-			/*
-			 * If we look at something, bring it to the
-			 * front of the list for next time.
-			 */
 			atomic_inc(&bp->b_hold);
-			list_move(&bp->b_hash_list, &hash->bh_list);
 			goto found;
 		}
 	}
@@ -1431,8 +1426,7 @@ xfs_alloc_bufhash(
 {
 	unsigned int		i;
 
-	btp->bt_hashshift = external ? 3 : 8;	/* 8 or 256 buckets */
-	btp->bt_hashmask = (1 << btp->bt_hashshift) - 1;
+	btp->bt_hashshift = external ? 3 : 12;	/* 8 or 4096 buckets */
 	btp->bt_hash = kmem_zalloc_large((1 << btp->bt_hashshift) *
 					 sizeof(xfs_bufhash_t));
 	for (i = 0; i < (1 << btp->bt_hashshift); i++) {
@@ -1926,7 +1920,8 @@ xfs_buf_init(void)
 	if (!xfs_buf_zone)
 		goto out;
 
-	xfslogd_workqueue = create_workqueue("xfslogd");
+	xfslogd_workqueue = alloc_workqueue("xfslogd",
+					WQ_RESCUER | WQ_HIGHPRI, 1);
 	if (!xfslogd_workqueue)
 		goto out_free_buf_zone;
 
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index d533d64e2c3e..9d021c73ea52 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -128,7 +128,6 @@ typedef struct xfs_buftarg {
 	size_t			bt_smask;
 
 	/* per device buffer hash table */
-	uint			bt_hashmask;
 	uint			bt_hashshift;
 	xfs_bufhash_t		*bt_hash;
 
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 237f5ffb2ee8..3b9e626f7cd1 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -785,6 +785,8 @@ xfs_ioc_fsgetxattr(
 {
 	struct fsxattr		fa;
 
+	memset(&fa, 0, sizeof(struct fsxattr));
+
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
 	fa.fsx_xflags = xfs_ip2xflags(ip);
 	fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog;
@@ -907,6 +909,13 @@ xfs_ioctl_setattr(
 		return XFS_ERROR(EIO);
 
 	/*
+	 * Disallow 32bit project ids because on-disk structure
+	 * is 16bit only.
+	 */
+	if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1))
+		return XFS_ERROR(EINVAL);
+
+	/*
 	 * If disk quotas is on, we make sure that the dquots do exist on disk,
 	 * before we start any other transactions. Trying to do this later
 	 * is messy. We don't care to take a readlock to look at the ids
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 68be25dcd301..b1fc2a6bfe83 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -664,7 +664,7 @@ xfs_vn_fiemap(
 					fieinfo->fi_extents_max + 1;
 	bm.bmv_count = min_t(__s32, bm.bmv_count,
 			     (PAGE_SIZE * 16 / sizeof(struct getbmapx)));
-	bm.bmv_iflags = BMV_IF_PREALLOC;
+	bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES;
 	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
 		bm.bmv_iflags |= BMV_IF_ATTRFORK;
 	if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC))
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 5fa7a30cc3f0..08fd3102128c 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1225,6 +1225,7 @@ xfs_fs_statfs(
 	struct xfs_inode	*ip = XFS_I(dentry->d_inode);
 	__uint64_t		fakeinos, id;
 	xfs_extlen_t		lsize;
+	__int64_t		ffree;
 
 	statp->f_type = XFS_SB_MAGIC;
 	statp->f_namelen = MAXNAMELEN - 1;
@@ -1248,7 +1249,11 @@ xfs_fs_statfs(
 		statp->f_files = min_t(typeof(statp->f_files),
 					statp->f_files,
 					mp->m_maxicount);
-	statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree);
+
+	/* make sure statp->f_ffree does not underflow */
+	ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree);
+	statp->f_ffree = max_t(__int64_t, ffree, 0);
+
 	spin_unlock(&mp->m_sb_lock);
 
 	if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) ||
@@ -1401,7 +1406,7 @@ xfs_fs_freeze(
 
 	xfs_save_resvblks(mp);
 	xfs_quiesce_attr(mp);
-	return -xfs_fs_log_dummy(mp);
+	return -xfs_fs_log_dummy(mp, SYNC_WAIT);
 }
 
 STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index dfcbd98d1599..81976ffed7d6 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -34,6 +34,7 @@
 #include "xfs_inode_item.h"
 #include "xfs_quota.h"
 #include "xfs_trace.h"
+#include "xfs_fsops.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -341,38 +342,6 @@ xfs_sync_attr(
 }
 
 STATIC int
-xfs_commit_dummy_trans(
-	struct xfs_mount	*mp,
-	uint			flags)
-{
-	struct xfs_inode	*ip = mp->m_rootip;
-	struct xfs_trans	*tp;
-	int			error;
-
-	/*
-	 * Put a dummy transaction in the log to tell recovery
-	 * that all others are OK.
-	 */
-	tp = xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
-	error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
-	if (error) {
-		xfs_trans_cancel(tp, 0);
-		return error;
-	}
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-	xfs_trans_ijoin(tp, ip);
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	error = xfs_trans_commit(tp, 0);
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
-	/* the log force ensures this transaction is pushed to disk */
-	xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0);
-	return error;
-}
-
-STATIC int
 xfs_sync_fsdata(
 	struct xfs_mount	*mp)
 {
@@ -432,7 +401,7 @@ xfs_quiesce_data(
 
 	/* mark the log as covered if needed */
 	if (xfs_log_need_covered(mp))
-		error2 = xfs_commit_dummy_trans(mp, SYNC_WAIT);
+		error2 = xfs_fs_log_dummy(mp, SYNC_WAIT);
 
 	/* flush data-only devices */
 	if (mp->m_rtdev_targp)
@@ -563,7 +532,7 @@ xfs_flush_inodes(
 /*
  * Every sync period we need to unpin all items, reclaim inodes and sync
  * disk quotas.  We might need to cover the log to indicate that the
- * filesystem is idle.
+ * filesystem is idle and not frozen.
  */
 STATIC void
 xfs_sync_worker(
@@ -577,8 +546,9 @@ xfs_sync_worker(
 		xfs_reclaim_inodes(mp, 0);
 		/* dgc: errors ignored here */
 		error = xfs_qm_sync(mp, SYNC_TRYLOCK);
-		if (xfs_log_need_covered(mp))
-			error = xfs_commit_dummy_trans(mp, 0);
+		if (mp->m_super->s_frozen == SB_UNFROZEN &&
+		    xfs_log_need_covered(mp))
+			error = xfs_fs_log_dummy(mp, 0);
 	}
 	mp->m_sync_seq++;
 	wake_up(&mp->m_wait_single_sync_task);
@@ -698,14 +668,11 @@ xfs_inode_set_reclaim_tag(
 	xfs_perag_put(pag);
 }
 
-void
-__xfs_inode_clear_reclaim_tag(
-	xfs_mount_t	*mp,
+STATIC void
+__xfs_inode_clear_reclaim(
 	xfs_perag_t	*pag,
 	xfs_inode_t	*ip)
 {
-	radix_tree_tag_clear(&pag->pag_ici_root,
-			XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
 	pag->pag_ici_reclaimable--;
 	if (!pag->pag_ici_reclaimable) {
 		/* clear the reclaim tag from the perag radix tree */
@@ -719,6 +686,17 @@ __xfs_inode_clear_reclaim_tag(
 	}
 }
 
+void
+__xfs_inode_clear_reclaim_tag(
+	xfs_mount_t	*mp,
+	xfs_perag_t	*pag,
+	xfs_inode_t	*ip)
+{
+	radix_tree_tag_clear(&pag->pag_ici_root,
+			XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+	__xfs_inode_clear_reclaim(pag, ip);
+}
+
 /*
  * Inodes in different states need to be treated differently, and the return
  * value of xfs_iflush is not sufficient to get this right. The following table
@@ -868,6 +846,7 @@ reclaim:
 	if (!radix_tree_delete(&pag->pag_ici_root,
 				XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino)))
 		ASSERT(0);
+	__xfs_inode_clear_reclaim(pag, ip);
 	write_unlock(&pag->pag_ici_lock);
 
 	/*
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 23f14e595c18..f90dadd5a968 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -5533,12 +5533,24 @@ xfs_getbmap(
 					map[i].br_startblock))
 				goto out_free_map;
 
-			nexleft--;
 			bmv->bmv_offset =
 				out[cur_ext].bmv_offset +
 				out[cur_ext].bmv_length;
 			bmv->bmv_length =
 				max_t(__int64_t, 0, bmvend - bmv->bmv_offset);
+
+			/*
+			 * In case we don't want to return the hole,
+			 * don't increase cur_ext so that we can reuse
+			 * it in the next loop.
+			 */
+			if ((iflags & BMV_IF_NO_HOLES) &&
+			    map[i].br_startblock == HOLESTARTBLOCK) {
+				memset(&out[cur_ext], 0, sizeof(out[cur_ext]));
+				continue;
+			}
+
+			nexleft--;
 			bmv->bmv_entries++;
 			cur_ext++;
 		}
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index 7cf7220e7d5f..87c2e9d02288 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -114,8 +114,10 @@ struct getbmapx {
 #define BMV_IF_NO_DMAPI_READ	0x2	/* Do not generate DMAPI read event  */
 #define BMV_IF_PREALLOC		0x4	/* rtn status BMV_OF_PREALLOC if req */
 #define BMV_IF_DELALLOC		0x8	/* rtn status BMV_OF_DELALLOC if req */
+#define BMV_IF_NO_HOLES		0x10	/* Do not return holes */
 #define BMV_IF_VALID	\
-	(BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC|BMV_IF_DELALLOC)
+	(BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC|	\
+	 BMV_IF_DELALLOC|BMV_IF_NO_HOLES)
 
 /*	bmv_oflags values - returned for each non-header segment */
 #define BMV_OF_PREALLOC		0x1	/* segment = unwritten pre-allocation */
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index dbca5f5c37ba..43b1d5699335 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -604,31 +604,36 @@ out:
 	return 0;
 }
 
+/*
+ * Dump a transaction into the log that contains no real change. This is needed
+ * to be able to make the log dirty or stamp the current tail LSN into the log
+ * during the covering operation.
+ *
+ * We cannot use an inode here for this - that will push dirty state back up
+ * into the VFS and then periodic inode flushing will prevent log covering from
+ * making progress. Hence we log a field in the superblock instead.
+ */
 int
 xfs_fs_log_dummy(
-	xfs_mount_t	*mp)
+	xfs_mount_t	*mp,
+	int		flags)
 {
 	xfs_trans_t	*tp;
-	xfs_inode_t	*ip;
 	int		error;
 
 	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP);
-	error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
+	error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
+					XFS_DEFAULT_LOG_COUNT);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
 		return error;
 	}
 
-	ip = mp->m_rootip;
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-	xfs_trans_ijoin(tp, ip);
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	xfs_trans_set_sync(tp);
-	error = xfs_trans_commit(tp, 0);
-
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return error;
+	/* log the UUID because it is an unchanging field */
+	xfs_mod_sb(tp, XFS_SB_UUID);
+	if (flags & SYNC_WAIT)
+		xfs_trans_set_sync(tp);
+	return xfs_trans_commit(tp, 0);
 }
 
 int
diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h
index 88435e0a77c9..a786c5212c1e 100644
--- a/fs/xfs/xfs_fsops.h
+++ b/fs/xfs/xfs_fsops.h
@@ -25,6 +25,6 @@ extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt);
 extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
 				xfs_fsop_resblks_t *outval);
 extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern int xfs_fs_log_dummy(xfs_mount_t *mp);
+extern int xfs_fs_log_dummy(xfs_mount_t *mp, int flags);
 
 #endif	/* __XFS_FSOPS_H__ */
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index abf80ae1e95b..5371d2dc360e 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -1213,7 +1213,6 @@ xfs_imap_lookup(
 	struct xfs_inobt_rec_incore rec;
 	struct xfs_btree_cur	*cur;
 	struct xfs_buf		*agbp;
-	xfs_agino_t		startino;
 	int			error;
 	int			i;
 
@@ -1227,13 +1226,13 @@ xfs_imap_lookup(
 	}
 
 	/*
-	 * derive and lookup the exact inode record for the given agino. If the
-	 * record cannot be found, then it's an invalid inode number and we
-	 * should abort.
+	 * Lookup the inode record for the given agino. If the record cannot be
+	 * found, then it's an invalid inode number and we should abort. Once
+	 * we have a record, we need to ensure it contains the inode number
+	 * we are looking up.
 	 */
 	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
-	startino = agino & ~(XFS_IALLOC_INODES(mp) - 1);
-	error = xfs_inobt_lookup(cur, startino, XFS_LOOKUP_EQ, &i);
+	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
 			error = xfs_inobt_get_rec(cur, &rec, &i);
@@ -1246,6 +1245,11 @@ xfs_imap_lookup(
 	if (error)
 		return error;
 
+	/* check that the returned record contains the required inode */
+	if (rec.ir_startino > agino ||
+	    rec.ir_startino + XFS_IALLOC_INODES(mp) <= agino)
+		return EINVAL;
+
 	/* for untrusted inodes check it is allocated first */
 	if ((flags & XFS_IGET_UNTRUSTED) &&
 	    (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 68415cb4f23c..34798f391c49 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1914,6 +1914,11 @@ xfs_iunlink_remove(
 	return 0;
 }
 
+/*
+ * A big issue when freeing the inode cluster is is that we _cannot_ skip any
+ * inodes that are in memory - they all must be marked stale and attached to
+ * the cluster buffer.
+ */
 STATIC void
 xfs_ifree_cluster(
 	xfs_inode_t	*free_ip,
@@ -1945,8 +1950,6 @@ xfs_ifree_cluster(
 	}
 
 	for (j = 0; j < nbufs; j++, inum += ninodes) {
-		int	found = 0;
-
 		blkno = XFS_AGB_TO_DADDR(mp, XFS_INO_TO_AGNO(mp, inum),
 					 XFS_INO_TO_AGBNO(mp, inum));
 
@@ -1965,7 +1968,9 @@ xfs_ifree_cluster(
 		/*
 		 * Walk the inodes already attached to the buffer and mark them
 		 * stale. These will all have the flush locks held, so an
-		 * in-memory inode walk can't lock them.
+		 * in-memory inode walk can't lock them. By marking them all
+		 * stale first, we will not attempt to lock them in the loop
+		 * below as the XFS_ISTALE flag will be set.
 		 */
 		lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
 		while (lip) {
@@ -1977,11 +1982,11 @@ xfs_ifree_cluster(
 							&iip->ili_flush_lsn,
 							&iip->ili_item.li_lsn);
 				xfs_iflags_set(iip->ili_inode, XFS_ISTALE);
-				found++;
 			}
 			lip = lip->li_bio_list;
 		}
 
+
 		/*
 		 * For each inode in memory attempt to add it to the inode
 		 * buffer and set it up for being staled on buffer IO
@@ -1993,6 +1998,7 @@ xfs_ifree_cluster(
 		 * even trying to lock them.
 		 */
 		for (i = 0; i < ninodes; i++) {
+retry:
 			read_lock(&pag->pag_ici_lock);
 			ip = radix_tree_lookup(&pag->pag_ici_root,
 					XFS_INO_TO_AGINO(mp, (inum + i)));
@@ -2003,38 +2009,36 @@ xfs_ifree_cluster(
 				continue;
 			}
 
-			/* don't try to lock/unlock the current inode */
+			/*
+			 * Don't try to lock/unlock the current inode, but we
+			 * _cannot_ skip the other inodes that we did not find
+			 * in the list attached to the buffer and are not
+			 * already marked stale. If we can't lock it, back off
+			 * and retry.
+			 */
 			if (ip != free_ip &&
 			    !xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
 				read_unlock(&pag->pag_ici_lock);
-				continue;
+				delay(1);
+				goto retry;
 			}
 			read_unlock(&pag->pag_ici_lock);
 
-			if (!xfs_iflock_nowait(ip)) {
-				if (ip != free_ip)
-					xfs_iunlock(ip, XFS_ILOCK_EXCL);
-				continue;
-			}
-
+			xfs_iflock(ip);
 			xfs_iflags_set(ip, XFS_ISTALE);
-			if (xfs_inode_clean(ip)) {
-				ASSERT(ip != free_ip);
-				xfs_ifunlock(ip);
-				xfs_iunlock(ip, XFS_ILOCK_EXCL);
-				continue;
-			}
 
+			/*
+			 * we don't need to attach clean inodes or those only
+			 * with unlogged changes (which we throw away, anyway).
+			 */
 			iip = ip->i_itemp;
-			if (!iip) {
-				/* inode with unlogged changes only */
+			if (!iip || xfs_inode_clean(ip)) {
 				ASSERT(ip != free_ip);
 				ip->i_update_core = 0;
 				xfs_ifunlock(ip);
 				xfs_iunlock(ip, XFS_ILOCK_EXCL);
 				continue;
 			}
-			found++;
 
 			iip->ili_last_fields = iip->ili_format.ilf_fields;
 			iip->ili_format.ilf_fields = 0;
@@ -2049,8 +2053,7 @@ xfs_ifree_cluster(
 				xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		}
 
-		if (found)
-			xfs_trans_stale_inode_buf(tp, bp);
+		xfs_trans_stale_inode_buf(tp, bp);
 		xfs_trans_binval(tp, bp);
 	}
 
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 430a8fc02c1f..ba8e36e0b4e7 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -3002,7 +3002,8 @@ _xfs_log_force(
 
 	XFS_STATS_INC(xs_log_force);
 
-	xlog_cil_push(log, 1);
+	if (log->l_cilp)
+		xlog_cil_force(log);
 
 	spin_lock(&log->l_icloglock);
 
@@ -3154,7 +3155,7 @@ _xfs_log_force_lsn(
 	XFS_STATS_INC(xs_log_force);
 
 	if (log->l_cilp) {
-		lsn = xlog_cil_push_lsn(log, lsn);
+		lsn = xlog_cil_force_lsn(log, lsn);
 		if (lsn == NULLCOMMITLSN)
 			return 0;
 	}
@@ -3711,7 +3712,7 @@ xfs_log_force_umount(
 	 * call below.
 	 */
 	if (!logerror && (mp->m_flags & XFS_MOUNT_DELAYLOG))
-		xlog_cil_push(log, 1);
+		xlog_cil_force(log);
 
 	/*
 	 * We must hold both the GRANT lock and the LOG lock,
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index 31e4ea2d19ac..7e206fc1fa36 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -68,6 +68,7 @@ xlog_cil_init(
 	ctx->sequence = 1;
 	ctx->cil = cil;
 	cil->xc_ctx = ctx;
+	cil->xc_current_sequence = ctx->sequence;
 
 	cil->xc_log = log;
 	log->l_cilp = cil;
@@ -269,15 +270,10 @@ xlog_cil_insert(
 static void
 xlog_cil_format_items(
 	struct log		*log,
-	struct xfs_log_vec	*log_vector,
-	struct xlog_ticket	*ticket,
-	xfs_lsn_t		*start_lsn)
+	struct xfs_log_vec	*log_vector)
 {
 	struct xfs_log_vec *lv;
 
-	if (start_lsn)
-		*start_lsn = log->l_cilp->xc_ctx->sequence;
-
 	ASSERT(log_vector);
 	for (lv = log_vector; lv; lv = lv->lv_next) {
 		void	*ptr;
@@ -301,9 +297,24 @@ xlog_cil_format_items(
 			ptr += vec->i_len;
 		}
 		ASSERT(ptr == lv->lv_buf + lv->lv_buf_len);
+	}
+}
 
+static void
+xlog_cil_insert_items(
+	struct log		*log,
+	struct xfs_log_vec	*log_vector,
+	struct xlog_ticket	*ticket,
+	xfs_lsn_t		*start_lsn)
+{
+	struct xfs_log_vec *lv;
+
+	if (start_lsn)
+		*start_lsn = log->l_cilp->xc_ctx->sequence;
+
+	ASSERT(log_vector);
+	for (lv = log_vector; lv; lv = lv->lv_next)
 		xlog_cil_insert(log, ticket, lv->lv_item, lv);
-	}
 }
 
 static void
@@ -321,80 +332,6 @@ xlog_cil_free_logvec(
 }
 
 /*
- * Commit a transaction with the given vector to the Committed Item List.
- *
- * To do this, we need to format the item, pin it in memory if required and
- * account for the space used by the transaction. Once we have done that we
- * need to release the unused reservation for the transaction, attach the
- * transaction to the checkpoint context so we carry the busy extents through
- * to checkpoint completion, and then unlock all the items in the transaction.
- *
- * For more specific information about the order of operations in
- * xfs_log_commit_cil() please refer to the comments in
- * xfs_trans_commit_iclog().
- *
- * Called with the context lock already held in read mode to lock out
- * background commit, returns without it held once background commits are
- * allowed again.
- */
-int
-xfs_log_commit_cil(
-	struct xfs_mount	*mp,
-	struct xfs_trans	*tp,
-	struct xfs_log_vec	*log_vector,
-	xfs_lsn_t		*commit_lsn,
-	int			flags)
-{
-	struct log		*log = mp->m_log;
-	int			log_flags = 0;
-	int			push = 0;
-
-	if (flags & XFS_TRANS_RELEASE_LOG_RES)
-		log_flags = XFS_LOG_REL_PERM_RESERV;
-
-	if (XLOG_FORCED_SHUTDOWN(log)) {
-		xlog_cil_free_logvec(log_vector);
-		return XFS_ERROR(EIO);
-	}
-
-	/* lock out background commit */
-	down_read(&log->l_cilp->xc_ctx_lock);
-	xlog_cil_format_items(log, log_vector, tp->t_ticket, commit_lsn);
-
-	/* check we didn't blow the reservation */
-	if (tp->t_ticket->t_curr_res < 0)
-		xlog_print_tic_res(log->l_mp, tp->t_ticket);
-
-	/* attach the transaction to the CIL if it has any busy extents */
-	if (!list_empty(&tp->t_busy)) {
-		spin_lock(&log->l_cilp->xc_cil_lock);
-		list_splice_init(&tp->t_busy,
-					&log->l_cilp->xc_ctx->busy_extents);
-		spin_unlock(&log->l_cilp->xc_cil_lock);
-	}
-
-	tp->t_commit_lsn = *commit_lsn;
-	xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
-	xfs_trans_unreserve_and_mod_sb(tp);
-
-	/* check for background commit before unlock */
-	if (log->l_cilp->xc_ctx->space_used > XLOG_CIL_SPACE_LIMIT(log))
-		push = 1;
-	up_read(&log->l_cilp->xc_ctx_lock);
-
-	/*
-	 * We need to push CIL every so often so we don't cache more than we
-	 * can fit in the log. The limit really is that a checkpoint can't be
-	 * more than half the log (the current checkpoint is not allowed to
-	 * overwrite the previous checkpoint), but commit latency and memory
-	 * usage limit this to a smaller size in most cases.
-	 */
-	if (push)
-		xlog_cil_push(log, 0);
-	return 0;
-}
-
-/*
  * Mark all items committed and clear busy extents. We free the log vector
  * chains in a separate pass so that we unpin the log items as quickly as
  * possible.
@@ -427,13 +364,23 @@ xlog_cil_committed(
 }
 
 /*
- * Push the Committed Item List to the log. If the push_now flag is not set,
- * then it is a background flush and so we can chose to ignore it.
+ * Push the Committed Item List to the log. If @push_seq flag is zero, then it
+ * is a background flush and so we can chose to ignore it. Otherwise, if the
+ * current sequence is the same as @push_seq we need to do a flush. If
+ * @push_seq is less than the current sequence, then it has already been
+ * flushed and we don't need to do anything - the caller will wait for it to
+ * complete if necessary.
+ *
+ * @push_seq is a value rather than a flag because that allows us to do an
+ * unlocked check of the sequence number for a match. Hence we can allows log
+ * forces to run racily and not issue pushes for the same sequence twice. If we
+ * get a race between multiple pushes for the same sequence they will block on
+ * the first one and then abort, hence avoiding needless pushes.
  */
-int
+STATIC int
 xlog_cil_push(
 	struct log		*log,
-	int			push_now)
+	xfs_lsn_t		push_seq)
 {
 	struct xfs_cil		*cil = log->l_cilp;
 	struct xfs_log_vec	*lv;
@@ -453,12 +400,20 @@ xlog_cil_push(
 	if (!cil)
 		return 0;
 
+	ASSERT(!push_seq || push_seq <= cil->xc_ctx->sequence);
+
 	new_ctx = kmem_zalloc(sizeof(*new_ctx), KM_SLEEP|KM_NOFS);
 	new_ctx->ticket = xlog_cil_ticket_alloc(log);
 
-	/* lock out transaction commit, but don't block on background push */
+	/*
+	 * Lock out transaction commit, but don't block for background pushes
+	 * unless we are well over the CIL space limit. See the definition of
+	 * XLOG_CIL_HARD_SPACE_LIMIT() for the full explanation of the logic
+	 * used here.
+	 */
 	if (!down_write_trylock(&cil->xc_ctx_lock)) {
-		if (!push_now)
+		if (!push_seq &&
+		    cil->xc_ctx->space_used < XLOG_CIL_HARD_SPACE_LIMIT(log))
 			goto out_free_ticket;
 		down_write(&cil->xc_ctx_lock);
 	}
@@ -469,7 +424,11 @@ xlog_cil_push(
 		goto out_skip;
 
 	/* check for spurious background flush */
-	if (!push_now && cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+	if (!push_seq && cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+		goto out_skip;
+
+	/* check for a previously pushed seqeunce */
+	if (push_seq && push_seq < cil->xc_ctx->sequence)
 		goto out_skip;
 
 	/*
@@ -515,6 +474,13 @@ xlog_cil_push(
 	cil->xc_ctx = new_ctx;
 
 	/*
+	 * mirror the new sequence into the cil structure so that we can do
+	 * unlocked checks against the current sequence in log forces without
+	 * risking deferencing a freed context pointer.
+	 */
+	cil->xc_current_sequence = new_ctx->sequence;
+
+	/*
 	 * The switch is now done, so we can drop the context lock and move out
 	 * of a shared context. We can't just go straight to the commit record,
 	 * though - we need to synchronise with previous and future commits so
@@ -626,6 +592,102 @@ out_abort:
 }
 
 /*
+ * Commit a transaction with the given vector to the Committed Item List.
+ *
+ * To do this, we need to format the item, pin it in memory if required and
+ * account for the space used by the transaction. Once we have done that we
+ * need to release the unused reservation for the transaction, attach the
+ * transaction to the checkpoint context so we carry the busy extents through
+ * to checkpoint completion, and then unlock all the items in the transaction.
+ *
+ * For more specific information about the order of operations in
+ * xfs_log_commit_cil() please refer to the comments in
+ * xfs_trans_commit_iclog().
+ *
+ * Called with the context lock already held in read mode to lock out
+ * background commit, returns without it held once background commits are
+ * allowed again.
+ */
+int
+xfs_log_commit_cil(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	struct xfs_log_vec	*log_vector,
+	xfs_lsn_t		*commit_lsn,
+	int			flags)
+{
+	struct log		*log = mp->m_log;
+	int			log_flags = 0;
+	int			push = 0;
+
+	if (flags & XFS_TRANS_RELEASE_LOG_RES)
+		log_flags = XFS_LOG_REL_PERM_RESERV;
+
+	if (XLOG_FORCED_SHUTDOWN(log)) {
+		xlog_cil_free_logvec(log_vector);
+		return XFS_ERROR(EIO);
+	}
+
+	/*
+	 * do all the hard work of formatting items (including memory
+	 * allocation) outside the CIL context lock. This prevents stalling CIL
+	 * pushes when we are low on memory and a transaction commit spends a
+	 * lot of time in memory reclaim.
+	 */
+	xlog_cil_format_items(log, log_vector);
+
+	/* lock out background commit */
+	down_read(&log->l_cilp->xc_ctx_lock);
+	xlog_cil_insert_items(log, log_vector, tp->t_ticket, commit_lsn);
+
+	/* check we didn't blow the reservation */
+	if (tp->t_ticket->t_curr_res < 0)
+		xlog_print_tic_res(log->l_mp, tp->t_ticket);
+
+	/* attach the transaction to the CIL if it has any busy extents */
+	if (!list_empty(&tp->t_busy)) {
+		spin_lock(&log->l_cilp->xc_cil_lock);
+		list_splice_init(&tp->t_busy,
+					&log->l_cilp->xc_ctx->busy_extents);
+		spin_unlock(&log->l_cilp->xc_cil_lock);
+	}
+
+	tp->t_commit_lsn = *commit_lsn;
+	xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
+	xfs_trans_unreserve_and_mod_sb(tp);
+
+	/*
+	 * Once all the items of the transaction have been copied to the CIL,
+	 * the items can be unlocked and freed.
+	 *
+	 * This needs to be done before we drop the CIL context lock because we
+	 * have to update state in the log items and unlock them before they go
+	 * to disk. If we don't, then the CIL checkpoint can race with us and
+	 * we can run checkpoint completion before we've updated and unlocked
+	 * the log items. This affects (at least) processing of stale buffers,
+	 * inodes and EFIs.
+	 */
+	xfs_trans_free_items(tp, *commit_lsn, 0);
+
+	/* check for background commit before unlock */
+	if (log->l_cilp->xc_ctx->space_used > XLOG_CIL_SPACE_LIMIT(log))
+		push = 1;
+
+	up_read(&log->l_cilp->xc_ctx_lock);
+
+	/*
+	 * We need to push CIL every so often so we don't cache more than we
+	 * can fit in the log. The limit really is that a checkpoint can't be
+	 * more than half the log (the current checkpoint is not allowed to
+	 * overwrite the previous checkpoint), but commit latency and memory
+	 * usage limit this to a smaller size in most cases.
+	 */
+	if (push)
+		xlog_cil_push(log, 0);
+	return 0;
+}
+
+/*
  * Conditionally push the CIL based on the sequence passed in.
  *
  * We only need to push if we haven't already pushed the sequence
@@ -639,39 +701,34 @@ out_abort:
  * commit lsn is there. It'll be empty, so this is broken for now.
  */
 xfs_lsn_t
-xlog_cil_push_lsn(
+xlog_cil_force_lsn(
 	struct log	*log,
-	xfs_lsn_t	push_seq)
+	xfs_lsn_t	sequence)
 {
 	struct xfs_cil		*cil = log->l_cilp;
 	struct xfs_cil_ctx	*ctx;
 	xfs_lsn_t		commit_lsn = NULLCOMMITLSN;
 
-restart:
-	down_write(&cil->xc_ctx_lock);
-	ASSERT(push_seq <= cil->xc_ctx->sequence);
-
-	/* check to see if we need to force out the current context */
-	if (push_seq == cil->xc_ctx->sequence) {
-		up_write(&cil->xc_ctx_lock);
-		xlog_cil_push(log, 1);
-		goto restart;
-	}
+	ASSERT(sequence <= cil->xc_current_sequence);
+
+	/*
+	 * check to see if we need to force out the current context.
+	 * xlog_cil_push() handles racing pushes for the same sequence,
+	 * so no need to deal with it here.
+	 */
+	if (sequence == cil->xc_current_sequence)
+		xlog_cil_push(log, sequence);
 
 	/*
 	 * See if we can find a previous sequence still committing.
-	 * We can drop the flush lock as soon as we have the cil lock
-	 * because we are now only comparing contexts protected by
-	 * the cil lock.
-	 *
 	 * We need to wait for all previous sequence commits to complete
 	 * before allowing the force of push_seq to go ahead. Hence block
 	 * on commits for those as well.
 	 */
+restart:
 	spin_lock(&cil->xc_cil_lock);
-	up_write(&cil->xc_ctx_lock);
 	list_for_each_entry(ctx, &cil->xc_committing, committing) {
-		if (ctx->sequence > push_seq)
+		if (ctx->sequence > sequence)
 			continue;
 		if (!ctx->commit_lsn) {
 			/*
@@ -681,7 +738,7 @@ restart:
 			sv_wait(&cil->xc_commit_wait, 0, &cil->xc_cil_lock, 0);
 			goto restart;
 		}
-		if (ctx->sequence != push_seq)
+		if (ctx->sequence != sequence)
 			continue;
 		/* found it! */
 		commit_lsn = ctx->commit_lsn;
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 8c072618965c..edcdfe01617f 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -422,16 +422,17 @@ struct xfs_cil {
 	struct rw_semaphore	xc_ctx_lock;
 	struct list_head	xc_committing;
 	sv_t			xc_commit_wait;
+	xfs_lsn_t		xc_current_sequence;
 };
 
 /*
- * The amount of log space we should the CIL to aggregate is difficult to size.
- * Whatever we chose we have to make we can get a reservation for the log space
- * effectively, that it is large enough to capture sufficient relogging to
- * reduce log buffer IO significantly, but it is not too large for the log or
- * induces too much latency when writing out through the iclogs. We track both
- * space consumed and the number of vectors in the checkpoint context, so we
- * need to decide which to use for limiting.
+ * The amount of log space we allow the CIL to aggregate is difficult to size.
+ * Whatever we choose, we have to make sure we can get a reservation for the
+ * log space effectively, that it is large enough to capture sufficient
+ * relogging to reduce log buffer IO significantly, but it is not too large for
+ * the log or induces too much latency when writing out through the iclogs. We
+ * track both space consumed and the number of vectors in the checkpoint
+ * context, so we need to decide which to use for limiting.
  *
  * Every log buffer we write out during a push needs a header reserved, which
  * is at least one sector and more for v2 logs. Hence we need a reservation of
@@ -458,16 +459,21 @@ struct xfs_cil {
  * checkpoint transaction ticket is specific to the checkpoint context, rather
  * than the CIL itself.
  *
- * With dynamic reservations, we can basically make up arbitrary limits for the
- * checkpoint size so long as they don't violate any other size rules.  Hence
- * the initial maximum size for the checkpoint transaction will be set to a
- * quarter of the log or 8MB, which ever is smaller. 8MB is an arbitrary limit
- * right now based on the latency of writing out a large amount of data through
- * the circular iclog buffers.
+ * With dynamic reservations, we can effectively make up arbitrary limits for
+ * the checkpoint size so long as they don't violate any other size rules.
+ * Recovery imposes a rule that no transaction exceed half the log, so we are
+ * limited by that.  Furthermore, the log transaction reservation subsystem
+ * tries to keep 25% of the log free, so we need to keep below that limit or we
+ * risk running out of free log space to start any new transactions.
+ *
+ * In order to keep background CIL push efficient, we will set a lower
+ * threshold at which background pushing is attempted without blocking current
+ * transaction commits.  A separate, higher bound defines when CIL pushes are
+ * enforced to ensure we stay within our maximum checkpoint size bounds.
+ * threshold, yet give us plenty of space for aggregation on large logs.
  */
-
-#define XLOG_CIL_SPACE_LIMIT(log)	\
-	(min((log->l_logsize >> 2), (8 * 1024 * 1024)))
+#define XLOG_CIL_SPACE_LIMIT(log)	(log->l_logsize >> 3)
+#define XLOG_CIL_HARD_SPACE_LIMIT(log)	(3 * (log->l_logsize >> 4))
 
 /*
  * The reservation head lsn is not made up of a cycle number and block number.
@@ -562,8 +568,16 @@ int	xlog_cil_init(struct log *log);
 void	xlog_cil_init_post_recovery(struct log *log);
 void	xlog_cil_destroy(struct log *log);
 
-int	xlog_cil_push(struct log *log, int push_now);
-xfs_lsn_t xlog_cil_push_lsn(struct log *log, xfs_lsn_t push_sequence);
+/*
+ * CIL force routines
+ */
+xfs_lsn_t xlog_cil_force_lsn(struct log *log, xfs_lsn_t sequence);
+
+static inline void
+xlog_cil_force(struct log *log)
+{
+	xlog_cil_force_lsn(log, log->l_cilp->xc_current_sequence);
+}
 
 /*
  * Unmount record type is used as a pseudo transaction type for the ticket.
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index fdca7416c754..1c47edaea0d2 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -1167,7 +1167,7 @@ xfs_trans_del_item(
  * Unlock all of the items of a transaction and free all the descriptors
  * of that transaction.
  */
-STATIC void
+void
 xfs_trans_free_items(
 	struct xfs_trans	*tp,
 	xfs_lsn_t		commit_lsn,
@@ -1653,9 +1653,6 @@ xfs_trans_commit_cil(
 		return error;
 
 	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
-
-	/* xfs_trans_free_items() unlocks them first */
-	xfs_trans_free_items(tp, *commit_lsn, 0);
 	xfs_trans_free(tp);
 	return 0;
 }
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index e2d93d8ead7b..62da86c90de5 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -25,7 +25,8 @@ struct xfs_trans;
 
 void	xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *);
 void	xfs_trans_del_item(struct xfs_log_item *);
-
+void	xfs_trans_free_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn,
+				int flags);
 void	xfs_trans_item_committed(struct xfs_log_item *lip,
 				xfs_lsn_t commit_lsn, int aborted);
 void	xfs_trans_unreserve_and_mod_sb(struct xfs_trans *tp);
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 66d585c6917c..4c7c7bfb2b2f 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -2299,15 +2299,22 @@ xfs_alloc_file_space(
 			e = allocatesize_fsb;
 		}
 
+		/*
+		 * The transaction reservation is limited to a 32-bit block
+		 * count, hence we need to limit the number of blocks we are
+		 * trying to reserve to avoid an overflow. We can't allocate
+		 * more than @nimaps extents, and an extent is limited on disk
+		 * to MAXEXTLEN (21 bits), so use that to enforce the limit.
+		 */
+		resblks = min_t(xfs_fileoff_t, (e - s), (MAXEXTLEN * nimaps));
 		if (unlikely(rt)) {
-			resrtextents = qblocks = (uint)(e - s);
+			resrtextents = qblocks = resblks;
 			resrtextents /= mp->m_sb.sb_rextsize;
 			resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
 			quota_flag = XFS_QMOPT_RES_RTBLKS;
 		} else {
 			resrtextents = 0;
-			resblks = qblocks = \
-				XFS_DIOSTRAT_SPACE_RES(mp, (uint)(e - s));
+			resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks);
 			quota_flag = XFS_QMOPT_RES_REGBLKS;
 		}