14 files changed, 294 insertions, 265 deletions

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 281ae95932c9..48cfac31f64c 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -90,9 +90,9 @@ static int is_writable(struct pnfs_block_extent *be, sector_t isect)
  */
 struct parallel_io {
 	struct kref refcnt;
-	struct rpc_call_ops call_ops;
-	void (*pnfs_callback) (void *data);
+	void (*pnfs_callback) (void *data, int num_se);
 	void *data;
+	int bse_count;
 };
 
 static inline struct parallel_io *alloc_parallel(void *data)
@@ -103,6 +103,7 @@ static inline struct parallel_io *alloc_parallel(void *data)
 	if (rv) {
 		rv->data = data;
 		kref_init(&rv->refcnt);
+		rv->bse_count = 0;
 	}
 	return rv;
 }
@@ -117,7 +118,7 @@ static void destroy_parallel(struct kref *kref)
 	struct parallel_io *p = container_of(kref, struct parallel_io, refcnt);
 
 	dprintk("%s enter\n", __func__);
-	p->pnfs_callback(p->data);
+	p->pnfs_callback(p->data, p->bse_count);
 	kfree(p);
 }
 
@@ -146,14 +147,19 @@ static struct bio *bl_alloc_init_bio(int npg, sector_t isect,
 {
 	struct bio *bio;
 
+	npg = min(npg, BIO_MAX_PAGES);
 	bio = bio_alloc(GFP_NOIO, npg);
-	if (!bio)
-		return NULL;
+	if (!bio && (current->flags & PF_MEMALLOC)) {
+		while (!bio && (npg /= 2))
+			bio = bio_alloc(GFP_NOIO, npg);
+	}
 
-	bio->bi_sector = isect - be->be_f_offset + be->be_v_offset;
-	bio->bi_bdev = be->be_mdev;
-	bio->bi_end_io = end_io;
-	bio->bi_private = par;
+	if (bio) {
+		bio->bi_sector = isect - be->be_f_offset + be->be_v_offset;
+		bio->bi_bdev = be->be_mdev;
+		bio->bi_end_io = end_io;
+		bio->bi_private = par;
+	}
 	return bio;
 }
 
@@ -212,22 +218,15 @@ static void bl_read_cleanup(struct work_struct *work)
 }
 
 static void
-bl_end_par_io_read(void *data)
+bl_end_par_io_read(void *data, int unused)
 {
 	struct nfs_read_data *rdata = data;
 
+	rdata->task.tk_status = rdata->pnfs_error;
 	INIT_WORK(&rdata->task.u.tk_work, bl_read_cleanup);
 	schedule_work(&rdata->task.u.tk_work);
 }
 
-/* We don't want normal .rpc_call_done callback used, so we replace it
- * with this stub.
- */
-static void bl_rpc_do_nothing(struct rpc_task *task, void *calldata)
-{
-	return;
-}
-
 static enum pnfs_try_status
 bl_read_pagelist(struct nfs_read_data *rdata)
 {
@@ -247,8 +246,6 @@ bl_read_pagelist(struct nfs_read_data *rdata)
 	par = alloc_parallel(rdata);
 	if (!par)
 		goto use_mds;
-	par->call_ops = *rdata->mds_ops;
-	par->call_ops.rpc_call_done = bl_rpc_do_nothing;
 	par->pnfs_callback = bl_end_par_io_read;
 	/* At this point, we can no longer jump to use_mds */
 
@@ -322,6 +319,7 @@ static void mark_extents_written(struct pnfs_block_layout *bl,
 {
 	sector_t isect, end;
 	struct pnfs_block_extent *be;
+	struct pnfs_block_short_extent *se;
 
 	dprintk("%s(%llu, %u)\n", __func__, offset, count);
 	if (count == 0)
@@ -334,8 +332,11 @@ static void mark_extents_written(struct pnfs_block_layout *bl,
 		be = bl_find_get_extent(bl, isect, NULL);
 		BUG_ON(!be); /* FIXME */
 		len = min(end, be->be_f_offset + be->be_length) - isect;
-		if (be->be_state == PNFS_BLOCK_INVALID_DATA)
-			bl_mark_for_commit(be, isect, len); /* What if fails? */
+		if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
+			se = bl_pop_one_short_extent(be->be_inval);
+			BUG_ON(!se);
+			bl_mark_for_commit(be, isect, len, se);
+		}
 		isect += len;
 		bl_put_extent(be);
 	}
@@ -357,7 +358,8 @@ static void bl_end_io_write_zero(struct bio *bio, int err)
 		end_page_writeback(page);
 		page_cache_release(page);
 	} while (bvec >= bio->bi_io_vec);
-	if (!uptodate) {
+
+	if (unlikely(!uptodate)) {
 		if (!wdata->pnfs_error)
 			wdata->pnfs_error = -EIO;
 		pnfs_set_lo_fail(wdata->lseg);
@@ -366,7 +368,6 @@ static void bl_end_io_write_zero(struct bio *bio, int err)
 	put_parallel(par);
 }
 
-/* This is basically copied from mpage_end_io_read */
 static void bl_end_io_write(struct bio *bio, int err)
 {
 	struct parallel_io *par = bio->bi_private;
@@ -392,7 +393,7 @@ static void bl_write_cleanup(struct work_struct *work)
 	dprintk("%s enter\n", __func__);
 	task = container_of(work, struct rpc_task, u.tk_work);
 	wdata = container_of(task, struct nfs_write_data, task);
-	if (!wdata->pnfs_error) {
+	if (likely(!wdata->pnfs_error)) {
 		/* Marks for LAYOUTCOMMIT */
 		mark_extents_written(BLK_LSEG2EXT(wdata->lseg),
 				     wdata->args.offset, wdata->args.count);
@@ -401,11 +402,16 @@ static void bl_write_cleanup(struct work_struct *work)
 }
 
 /* Called when last of bios associated with a bl_write_pagelist call finishes */
-static void bl_end_par_io_write(void *data)
+static void bl_end_par_io_write(void *data, int num_se)
 {
 	struct nfs_write_data *wdata = data;
 
-	wdata->task.tk_status = 0;
+	if (unlikely(wdata->pnfs_error)) {
+		bl_free_short_extents(&BLK_LSEG2EXT(wdata->lseg)->bl_inval,
+					num_se);
+	}
+
+	wdata->task.tk_status = wdata->pnfs_error;
 	wdata->verf.committed = NFS_FILE_SYNC;
 	INIT_WORK(&wdata->task.u.tk_work, bl_write_cleanup);
 	schedule_work(&wdata->task.u.tk_work);
@@ -484,6 +490,55 @@ cleanup:
 	return ret;
 }
 
+/* Find or create a zeroing page marked being writeback.
+ * Return ERR_PTR on error, NULL to indicate skip this page and page itself
+ * to indicate write out.
+ */
+static struct page *
+bl_find_get_zeroing_page(struct inode *inode, pgoff_t index,
+			struct pnfs_block_extent *cow_read)
+{
+	struct page *page;
+	int locked = 0;
+	page = find_get_page(inode->i_mapping, index);
+	if (page)
+		goto check_page;
+
+	page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
+	if (unlikely(!page)) {
+		dprintk("%s oom\n", __func__);
+		return ERR_PTR(-ENOMEM);
+	}
+	locked = 1;
+
+check_page:
+	/* PageDirty: Other will write this out
+	 * PageWriteback: Other is writing this out
+	 * PageUptodate: It was read before
+	 */
+	if (PageDirty(page) || PageWriteback(page)) {
+		print_page(page);
+		if (locked)
+			unlock_page(page);
+		page_cache_release(page);
+		return NULL;
+	}
+
+	if (!locked) {
+		lock_page(page);
+		locked = 1;
+		goto check_page;
+	}
+	if (!PageUptodate(page)) {
+		/* New page, readin or zero it */
+		init_page_for_write(page, cow_read);
+	}
+	set_page_writeback(page);
+	unlock_page(page);
+
+	return page;
+}
+
 static enum pnfs_try_status
 bl_write_pagelist(struct nfs_write_data *wdata, int sync)
 {
@@ -508,9 +563,7 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync)
 	 */
 	par = alloc_parallel(wdata);
 	if (!par)
-		return PNFS_NOT_ATTEMPTED;
-	par->call_ops = *wdata->mds_ops;
-	par->call_ops.rpc_call_done = bl_rpc_do_nothing;
+		goto out_mds;
 	par->pnfs_callback = bl_end_par_io_write;
 	/* At this point, have to be more careful with error handling */
 
@@ -518,12 +571,15 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync)
 	be = bl_find_get_extent(BLK_LSEG2EXT(wdata->lseg), isect, &cow_read);
 	if (!be || !is_writable(be, isect)) {
 		dprintk("%s no matching extents!\n", __func__);
-		wdata->pnfs_error = -EINVAL;
-		goto out;
+		goto out_mds;
 	}
 
 	/* First page inside INVALID extent */
 	if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
+		if (likely(!bl_push_one_short_extent(be->be_inval)))
+			par->bse_count++;
+		else
+			goto out_mds;
 		temp = offset >> PAGE_CACHE_SHIFT;
 		npg_zero = do_div(temp, npg_per_block);
 		isect = (sector_t) (((offset - npg_zero * PAGE_CACHE_SIZE) &
@@ -543,36 +599,16 @@ fill_invalid_ext:
 			dprintk("%s zero %dth page: index %lu isect %llu\n",
 				__func__, npg_zero, index,
 				(unsigned long long)isect);
-			page =
-			    find_or_create_page(wdata->inode->i_mapping, index,
-						GFP_NOFS);
-			if (!page) {
-				dprintk("%s oom\n", __func__);
-				wdata->pnfs_error = -ENOMEM;
+			page = bl_find_get_zeroing_page(wdata->inode, index,
+							cow_read);
+			if (unlikely(IS_ERR(page))) {
+				wdata->pnfs_error = PTR_ERR(page);
 				goto out;
-			}
-
-			/* PageDirty: Other will write this out
-			 * PageWriteback: Other is writing this out
-			 * PageUptodate: It was read before
-			 * sector_initialized: already written out
-			 */
-			if (PageDirty(page) || PageWriteback(page)) {
-				print_page(page);
-				unlock_page(page);
-				page_cache_release(page);
+			} else if (page == NULL)
 				goto next_page;
-			}
-			if (!PageUptodate(page)) {
-				/* New page, readin or zero it */
-				init_page_for_write(page, cow_read);
-			}
-			set_page_writeback(page);
-			unlock_page(page);
 
 			ret = bl_mark_sectors_init(be->be_inval, isect,
-						       PAGE_CACHE_SECTORS,
-						       NULL);
+						       PAGE_CACHE_SECTORS);
 			if (unlikely(ret)) {
 				dprintk("%s bl_mark_sectors_init fail %d\n",
 					__func__, ret);
@@ -581,6 +617,19 @@ fill_invalid_ext:
 				wdata->pnfs_error = ret;
 				goto out;
 			}
+			if (likely(!bl_push_one_short_extent(be->be_inval)))
+				par->bse_count++;
+			else {
+				end_page_writeback(page);
+				page_cache_release(page);
+				wdata->pnfs_error = -ENOMEM;
+				goto out;
+			}
+			/* FIXME: This should be done in bi_end_io */
+			mark_extents_written(BLK_LSEG2EXT(wdata->lseg),
+					     page->index << PAGE_CACHE_SHIFT,
+					     PAGE_CACHE_SIZE);
+
 			bio = bl_add_page_to_bio(bio, npg_zero, WRITE,
 						 isect, page, be,
 						 bl_end_io_write_zero, par);
@@ -589,10 +638,6 @@ fill_invalid_ext:
 				bio = NULL;
 				goto out;
 			}
-			/* FIXME: This should be done in bi_end_io */
-			mark_extents_written(BLK_LSEG2EXT(wdata->lseg),
-					     page->index << PAGE_CACHE_SHIFT,
-					     PAGE_CACHE_SIZE);
 next_page:
 			isect += PAGE_CACHE_SECTORS;
 			extent_length -= PAGE_CACHE_SECTORS;
@@ -616,13 +661,21 @@ next_page:
 				wdata->pnfs_error = -EINVAL;
 				goto out;
 			}
+			if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
+				if (likely(!bl_push_one_short_extent(
+								be->be_inval)))
+					par->bse_count++;
+				else {
+					wdata->pnfs_error = -ENOMEM;
+					goto out;
+				}
+			}
 			extent_length = be->be_length -
 			    (isect - be->be_f_offset);
 		}
 		if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
 			ret = bl_mark_sectors_init(be->be_inval, isect,
-						       PAGE_CACHE_SECTORS,
-						       NULL);
+						       PAGE_CACHE_SECTORS);
 			if (unlikely(ret)) {
 				dprintk("%s bl_mark_sectors_init fail %d\n",
 					__func__, ret);
@@ -664,6 +717,10 @@ out:
 	bl_submit_bio(WRITE, bio);
 	put_parallel(par);
 	return PNFS_ATTEMPTED;
+out_mds:
+	bl_put_extent(be);
+	kfree(par);
+	return PNFS_NOT_ATTEMPTED;
 }
 
 /* FIXME - range ignored */
@@ -690,11 +747,17 @@ static void
 release_inval_marks(struct pnfs_inval_markings *marks)
 {
 	struct pnfs_inval_tracking *pos, *temp;
+	struct pnfs_block_short_extent *se, *stemp;
 
 	list_for_each_entry_safe(pos, temp, &marks->im_tree.mtt_stub, it_link) {
 		list_del(&pos->it_link);
 		kfree(pos);
 	}
+
+	list_for_each_entry_safe(se, stemp, &marks->im_extents, bse_node) {
+		list_del(&se->bse_node);
+		kfree(se);
+	}
 	return;
 }
 
@@ -779,16 +842,13 @@ bl_cleanup_layoutcommit(struct nfs4_layoutcommit_data *lcdata)
 static void free_blk_mountid(struct block_mount_id *mid)
 {
 	if (mid) {
-		struct pnfs_block_dev *dev;
-		spin_lock(&mid->bm_lock);
-		while (!list_empty(&mid->bm_devlist)) {
-			dev = list_first_entry(&mid->bm_devlist,
-					       struct pnfs_block_dev,
-					       bm_node);
+		struct pnfs_block_dev *dev, *tmp;
+
+		/* No need to take bm_lock as we are last user freeing bm_devlist */
+		list_for_each_entry_safe(dev, tmp, &mid->bm_devlist, bm_node) {
 			list_del(&dev->bm_node);
 			bl_free_block_dev(dev);
 		}
-		spin_unlock(&mid->bm_lock);
 		kfree(mid);
 	}
 }
diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h
index 42acf7ef5992..e31a2df28e70 100644
--- a/fs/nfs/blocklayout/blocklayout.h
+++ b/fs/nfs/blocklayout/blocklayout.h
@@ -70,6 +70,7 @@ struct pnfs_inval_markings {
 	spinlock_t	im_lock;
 	struct my_tree	im_tree;	/* Sectors that need LAYOUTCOMMIT */
 	sector_t	im_block_size;	/* Server blocksize in sectors */
+	struct list_head im_extents;	/* Short extents for INVAL->RW conversion */
 };
 
 struct pnfs_inval_tracking {
@@ -105,6 +106,7 @@ BL_INIT_INVAL_MARKS(struct pnfs_inval_markings *marks, sector_t blocksize)
 {
 	spin_lock_init(&marks->im_lock);
 	INIT_LIST_HEAD(&marks->im_tree.mtt_stub);
+	INIT_LIST_HEAD(&marks->im_extents);
 	marks->im_block_size = blocksize;
 	marks->im_tree.mtt_step_size = min((sector_t)PAGE_CACHE_SECTORS,
 					   blocksize);
@@ -186,8 +188,7 @@ struct pnfs_block_extent *
 bl_find_get_extent(struct pnfs_block_layout *bl, sector_t isect,
 		struct pnfs_block_extent **cow_read);
 int bl_mark_sectors_init(struct pnfs_inval_markings *marks,
-			     sector_t offset, sector_t length,
-			     sector_t **pages);
+			     sector_t offset, sector_t length);
 void bl_put_extent(struct pnfs_block_extent *be);
 struct pnfs_block_extent *bl_alloc_extent(void);
 int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect);
@@ -200,6 +201,11 @@ void clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
 int bl_add_merge_extent(struct pnfs_block_layout *bl,
 			 struct pnfs_block_extent *new);
 int bl_mark_for_commit(struct pnfs_block_extent *be,
-			sector_t offset, sector_t length);
+			sector_t offset, sector_t length,
+			struct pnfs_block_short_extent *new);
+int bl_push_one_short_extent(struct pnfs_inval_markings *marks);
+struct pnfs_block_short_extent *
+bl_pop_one_short_extent(struct pnfs_inval_markings *marks);
+void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free);
 
 #endif /* FS_NFS_NFS4BLOCKLAYOUT_H */
diff --git a/fs/nfs/blocklayout/extents.c b/fs/nfs/blocklayout/extents.c
index 19fa7b0b8c00..1abac09f7cd5 100644
--- a/fs/nfs/blocklayout/extents.c
+++ b/fs/nfs/blocklayout/extents.c
@@ -110,13 +110,7 @@ static int _add_entry(struct my_tree *tree, u64 s, int32_t tag,
 		return 0;
 	} else {
 		struct pnfs_inval_tracking *new;
-		if (storage)
-			new = storage;
-		else {
-			new = kmalloc(sizeof(*new), GFP_NOFS);
-			if (!new)
-				return -ENOMEM;
-		}
+		new = storage;
 		new->it_sector = s;
 		new->it_tags = (1 << tag);
 		list_add(&new->it_link, &pos->it_link);
@@ -139,11 +133,13 @@ static int _set_range(struct my_tree *tree, int32_t tag, u64 s, u64 length)
 }
 
 /* Ensure that future operations on given range of tree will not malloc */
-static int _preload_range(struct my_tree *tree, u64 offset, u64 length)
+static int _preload_range(struct pnfs_inval_markings *marks,
+		u64 offset, u64 length)
 {
 	u64 start, end, s;
 	int count, i, used = 0, status = -ENOMEM;
 	struct pnfs_inval_tracking **storage;
+	struct my_tree  *tree = &marks->im_tree;
 
 	dprintk("%s(%llu, %llu) enter\n", __func__, offset, length);
 	start = normalize(offset, tree->mtt_step_size);
@@ -161,12 +157,11 @@ static int _preload_range(struct my_tree *tree, u64 offset, u64 length)
 			goto out_cleanup;
 	}
 
-	/* Now need lock - HOW??? */
-
+	spin_lock_bh(&marks->im_lock);
 	for (s = start; s < end; s += tree->mtt_step_size)
 		used += _add_entry(tree, s, INTERNAL_EXISTS, storage[used]);
+	spin_unlock_bh(&marks->im_lock);
 
-	/* Unlock - HOW??? */
 	status = 0;
 
  out_cleanup:
@@ -179,41 +174,14 @@ static int _preload_range(struct my_tree *tree, u64 offset, u64 length)
 	return status;
 }
 
-static void set_needs_init(sector_t *array, sector_t offset)
-{
-	sector_t *p = array;
-
-	dprintk("%s enter\n", __func__);
-	if (!p)
-		return;
-	while (*p < offset)
-		p++;
-	if (*p == offset)
-		return;
-	else if (*p == ~0) {
-		*p++ = offset;
-		*p = ~0;
-		return;
-	} else {
-		sector_t *save = p;
-		dprintk("%s Adding %llu\n", __func__, (u64)offset);
-		while (*p != ~0)
-			p++;
-		p++;
-		memmove(save + 1, save, (char *)p - (char *)save);
-		*save = offset;
-		return;
-	}
-}
-
 /* We are relying on page lock to serialize this */
 int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect)
 {
 	int rv;
 
-	spin_lock(&marks->im_lock);
+	spin_lock_bh(&marks->im_lock);
 	rv = _has_tag(&marks->im_tree, isect, EXTENT_INITIALIZED);
-	spin_unlock(&marks->im_lock);
+	spin_unlock_bh(&marks->im_lock);
 	return rv;
 }
 
@@ -253,78 +221,39 @@ static int is_range_written(struct pnfs_inval_markings *marks,
 {
 	int rv;
 
-	spin_lock(&marks->im_lock);
+	spin_lock_bh(&marks->im_lock);
 	rv = _range_has_tag(&marks->im_tree, start, end, EXTENT_WRITTEN);
-	spin_unlock(&marks->im_lock);
+	spin_unlock_bh(&marks->im_lock);
 	return rv;
 }
 
 /* Marks sectors in [offest, offset_length) as having been initialized.
  * All lengths are step-aligned, where step is min(pagesize, blocksize).
- * Notes where partial block is initialized, and helps prepare it for
- * complete initialization later.
+ * Currently assumes offset is page-aligned
  */
-/* Currently assumes offset is page-aligned */
 int bl_mark_sectors_init(struct pnfs_inval_markings *marks,
-			     sector_t offset, sector_t length,
-			     sector_t **pages)
+			     sector_t offset, sector_t length)
 {
-	sector_t s, start, end;
-	sector_t *array = NULL; /* Pages to mark */
+	sector_t start, end;
 
 	dprintk("%s(offset=%llu,len=%llu) enter\n",
 		__func__, (u64)offset, (u64)length);
-	s = max((sector_t) 3,
-		2 * (marks->im_block_size / (PAGE_CACHE_SECTORS)));
-	dprintk("%s set max=%llu\n", __func__, (u64)s);
-	if (pages) {
-		array = kmalloc(s * sizeof(sector_t), GFP_NOFS);
-		if (!array)
-			goto outerr;
-		array[0] = ~0;
-	}
 
 	start = normalize(offset, marks->im_block_size);
 	end = normalize_up(offset + length, marks->im_block_size);
-	if (_preload_range(&marks->im_tree, start, end - start))
+	if (_preload_range(marks, start, end - start))
 		goto outerr;
 
-	spin_lock(&marks->im_lock);
-
-	for (s = normalize_up(start, PAGE_CACHE_SECTORS);
-	     s < offset; s += PAGE_CACHE_SECTORS) {
-		dprintk("%s pre-area pages\n", __func__);
-		/* Portion of used block is not initialized */
-		if (!_has_tag(&marks->im_tree, s, EXTENT_INITIALIZED))
-			set_needs_init(array, s);
-	}
+	spin_lock_bh(&marks->im_lock);
 	if (_set_range(&marks->im_tree, EXTENT_INITIALIZED, offset, length))
 		goto out_unlock;
-	for (s = normalize_up(offset + length, PAGE_CACHE_SECTORS);
-	     s < end; s += PAGE_CACHE_SECTORS) {
-		dprintk("%s post-area pages\n", __func__);
-		if (!_has_tag(&marks->im_tree, s, EXTENT_INITIALIZED))
-			set_needs_init(array, s);
-	}
-
-	spin_unlock(&marks->im_lock);
+	spin_unlock_bh(&marks->im_lock);
 
-	if (pages) {
-		if (array[0] == ~0) {
-			kfree(array);
-			*pages = NULL;
-		} else
-			*pages = array;
-	}
 	return 0;
 
- out_unlock:
-	spin_unlock(&marks->im_lock);
- outerr:
-	if (pages) {
-		kfree(array);
-		*pages = NULL;
-	}
+out_unlock:
+	spin_unlock_bh(&marks->im_lock);
+outerr:
 	return -ENOMEM;
 }
 
@@ -338,9 +267,9 @@ static int mark_written_sectors(struct pnfs_inval_markings *marks,
 
 	dprintk("%s(offset=%llu,len=%llu) enter\n", __func__,
 		(u64)offset, (u64)length);
-	spin_lock(&marks->im_lock);
+	spin_lock_bh(&marks->im_lock);
 	status = _set_range(&marks->im_tree, EXTENT_WRITTEN, offset, length);
-	spin_unlock(&marks->im_lock);
+	spin_unlock_bh(&marks->im_lock);
 	return status;
 }
 
@@ -440,20 +369,18 @@ static void add_to_commitlist(struct pnfs_block_layout *bl,
 
 /* Note the range described by offset, length is guaranteed to be contained
  * within be.
+ * new will be freed, either by this function or add_to_commitlist if they
+ * decide not to use it, or after LAYOUTCOMMIT uses it in the commitlist.
  */
 int bl_mark_for_commit(struct pnfs_block_extent *be,
-		    sector_t offset, sector_t length)
+		    sector_t offset, sector_t length,
+		    struct pnfs_block_short_extent *new)
 {
 	sector_t new_end, end = offset + length;
-	struct pnfs_block_short_extent *new;
 	struct pnfs_block_layout *bl = container_of(be->be_inval,
 						    struct pnfs_block_layout,
 						    bl_inval);
 
-	new = kmalloc(sizeof(*new), GFP_NOFS);
-	if (!new)
-		return -ENOMEM;
-
 	mark_written_sectors(be->be_inval, offset, length);
 	/* We want to add the range to commit list, but it must be
 	 * block-normalized, and verified that the normalized range has
@@ -483,9 +410,6 @@ int bl_mark_for_commit(struct pnfs_block_extent *be,
 	new->bse_mdev = be->be_mdev;
 
 	spin_lock(&bl->bl_ext_lock);
-	/* new will be freed, either by add_to_commitlist if it decides not
-	 * to use it, or after LAYOUTCOMMIT uses it in the commitlist.
-	 */
 	add_to_commitlist(bl, new);
 	spin_unlock(&bl->bl_ext_lock);
 	return 0;
@@ -933,3 +857,53 @@ clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
 		}
 	}
 }
+
+int bl_push_one_short_extent(struct pnfs_inval_markings *marks)
+{
+	struct pnfs_block_short_extent *new;
+
+	new = kmalloc(sizeof(*new), GFP_NOFS);
+	if (unlikely(!new))
+		return -ENOMEM;
+
+	spin_lock_bh(&marks->im_lock);
+	list_add(&new->bse_node, &marks->im_extents);
+	spin_unlock_bh(&marks->im_lock);
+
+	return 0;
+}
+
+struct pnfs_block_short_extent *
+bl_pop_one_short_extent(struct pnfs_inval_markings *marks)
+{
+	struct pnfs_block_short_extent *rv = NULL;
+
+	spin_lock_bh(&marks->im_lock);
+	if (!list_empty(&marks->im_extents)) {
+		rv = list_entry((&marks->im_extents)->next,
+				struct pnfs_block_short_extent, bse_node);
+		list_del_init(&rv->bse_node);
+	}
+	spin_unlock_bh(&marks->im_lock);
+
+	return rv;
+}
+
+void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free)
+{
+	struct pnfs_block_short_extent *se = NULL, *tmp;
+
+	if (num_to_free <= 0)
+		return;
+
+	spin_lock(&marks->im_lock);
+	list_for_each_entry_safe(se, tmp, &marks->im_extents, bse_node) {
+		list_del(&se->bse_node);
+		kfree(se);
+		if (--num_to_free == 0)
+			break;
+	}
+	spin_unlock(&marks->im_lock);
+
+	BUG_ON(num_to_free > 0);
+}
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 07df5f1d85e5..c89d3b9e483c 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -162,7 +162,7 @@ struct cb_layoutrecallargs {
 	};
 };
 
-extern unsigned nfs4_callback_layoutrecall(
+extern __be32 nfs4_callback_layoutrecall(
 	struct cb_layoutrecallargs *args,
 	void *dummy, struct cb_process_state *cps);
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 726e59a9e50f..d50b2742f23b 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -305,6 +305,10 @@ __be32 decode_devicenotify_args(struct svc_rqst *rqstp,
 	n = ntohl(*p++);
 	if (n <= 0)
 		goto out;
+	if (n > ULONG_MAX / sizeof(*args->devs)) {
+		status = htonl(NFS4ERR_BADXDR);
+		goto out;
+	}
 
 	args->devs = kmalloc(n * sizeof(*args->devs), GFP_KERNEL);
 	if (!args->devs) {
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 277dfaf2e99a..31778f74357d 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -84,7 +84,7 @@ retry:
 /*
  * Turn off NFSv4 uid/gid mapping when using AUTH_SYS
  */
-static int nfs4_disable_idmapping = 1;
+static bool nfs4_disable_idmapping = true;
 
 /*
  * RPC cruft for NFS
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index fd9a872fada0..32aa6917265a 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -260,10 +260,10 @@ void nfs_readdir_clear_array(struct page *page)
 	struct nfs_cache_array *array;
 	int i;
 
-	array = kmap_atomic(page, KM_USER0);
+	array = kmap_atomic(page);
 	for (i = 0; i < array->size; i++)
 		kfree(array->array[i].string.name);
-	kunmap_atomic(array, KM_USER0);
+	kunmap_atomic(array);
 }
 
 /*
@@ -1870,11 +1870,11 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
 	if (!page)
 		return -ENOMEM;
 
-	kaddr = kmap_atomic(page, KM_USER0);
+	kaddr = kmap_atomic(page);
 	memcpy(kaddr, symname, pathlen);
 	if (pathlen < PAGE_SIZE)
 		memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen);
-	kunmap_atomic(kaddr, KM_USER0);
+	kunmap_atomic(kaddr);
 
 	error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr);
 	if (error != 0) {
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 25c3bfad7953..f649fba8c384 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -57,7 +57,7 @@
 #define NFS_64_BIT_INODE_NUMBERS_ENABLED	1
 
 /* Default is to see 64-bit inode numbers */
-static int enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED;
+static bool enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED;
 
 static void nfs_invalidate_inode(struct inode *);
 static int nfs_update_inode(struct inode *, struct nfs_fattr *);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 5ee92538b063..8102db9b926c 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -332,7 +332,7 @@ void nfs_commit_release_pages(struct nfs_write_data *data);
 
 #ifdef CONFIG_MIGRATION
 extern int nfs_migrate_page(struct address_space *,
-		struct page *, struct page *);
+		struct page *, struct page *, enum migrate_mode);
 #else
 #define nfs_migrate_page NULL
 #endif
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index ed388aae9689..8ae91908f5aa 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -382,7 +382,7 @@ decode_ds_addr(struct xdr_stream *streamp, gfp_t gfp_flags)
 {
 	struct nfs4_pnfs_ds_addr *da = NULL;
 	char *buf, *portstr;
-	u32 port;
+	__be16 port;
 	int nlen, rlen;
 	int tmp[2];
 	__be32 *p;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 75366dc89686..caf92d05c3a9 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -193,7 +193,7 @@ static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dent
 	 * when talking to the server, we always send cookie 0
 	 * instead of 1 or 2.
 	 */
-	start = p = kmap_atomic(*readdir->pages, KM_USER0);
+	start = p = kmap_atomic(*readdir->pages);
 	
 	if (cookie == 0) {
 		*p++ = xdr_one;                                  /* next */
@@ -221,7 +221,7 @@ static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dent
 
 	readdir->pgbase = (char *)p - (char *)start;
 	readdir->count -= readdir->pgbase;
-	kunmap_atomic(start, KM_USER0);
+	kunmap_atomic(start);
 }
 
 static int nfs4_wait_clnt_recover(struct nfs_client *clp)
@@ -3575,8 +3575,8 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 	}
 	if (npages > 1) {
 		/* for decoding across pages */
-		args.acl_scratch = alloc_page(GFP_KERNEL);
-		if (!args.acl_scratch)
+		res.acl_scratch = alloc_page(GFP_KERNEL);
+		if (!res.acl_scratch)
 			goto out_free;
 	}
 	args.acl_len = npages * PAGE_SIZE;
@@ -3587,7 +3587,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		res.acl_flags |= NFS4_ACL_LEN_REQUEST;
 	resp_buf = page_address(pages[0]);
 
-	dprintk("%s  buf %p buflen %ld npages %d args.acl_len %ld\n",
+	dprintk("%s  buf %p buflen %zu npages %d args.acl_len %zu\n",
 		__func__, buf, buflen, npages, args.acl_len);
 	ret = nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode),
 			     &msg, &args.seq_args, &res.seq_res, 0);
@@ -3612,8 +3612,8 @@ out_free:
 	for (i = 0; i < npages; i++)
 		if (pages[i])
 			__free_page(pages[i]);
-	if (args.acl_scratch)
-		__free_page(args.acl_scratch);
+	if (res.acl_scratch)
+		__free_page(res.acl_scratch);
 	return ret;
 }
 
@@ -4883,8 +4883,10 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 				clp->cl_rpcclient->cl_auth->au_flavor);
 
 	res.server_scope = kzalloc(sizeof(struct server_scope), GFP_KERNEL);
-	if (unlikely(!res.server_scope))
-		return -ENOMEM;
+	if (unlikely(!res.server_scope)) {
+		status = -ENOMEM;
+		goto out;
+	}
 
 	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 	if (!status)
@@ -4901,12 +4903,13 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 			clp->server_scope = NULL;
 		}
 
-		if (!clp->server_scope)
+		if (!clp->server_scope) {
 			clp->server_scope = res.server_scope;
-		else
-			kfree(res.server_scope);
+			goto out;
+		}
 	}
-
+	kfree(res.server_scope);
+out:
 	dprintk("<-- %s status= %d\n", __func__, status);
 	return status;
 }
@@ -5008,37 +5011,53 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
+static struct nfs4_slot *nfs4_alloc_slots(u32 max_slots, gfp_t gfp_flags)
+{
+	return kcalloc(max_slots, sizeof(struct nfs4_slot), gfp_flags);
+}
+
+static void nfs4_add_and_init_slots(struct nfs4_slot_table *tbl,
+		struct nfs4_slot *new,
+		u32 max_slots,
+		u32 ivalue)
+{
+	struct nfs4_slot *old = NULL;
+	u32 i;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	if (new) {
+		old = tbl->slots;
+		tbl->slots = new;
+		tbl->max_slots = max_slots;
+	}
+	tbl->highest_used_slotid = -1;	/* no slot is currently used */
+	for (i = 0; i < tbl->max_slots; i++)
+		tbl->slots[i].seq_nr = ivalue;
+	spin_unlock(&tbl->slot_tbl_lock);
+	kfree(old);
+}
+
 /*
- * Reset a slot table
+ * (re)Initialise a slot table
  */
-static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs,
-				 int ivalue)
+static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs,
+				 u32 ivalue)
 {
 	struct nfs4_slot *new = NULL;
-	int i;
-	int ret = 0;
+	int ret = -ENOMEM;
 
 	dprintk("--> %s: max_reqs=%u, tbl->max_slots %d\n", __func__,
 		max_reqs, tbl->max_slots);
 
 	/* Does the newly negotiated max_reqs match the existing slot table? */
 	if (max_reqs != tbl->max_slots) {
-		ret = -ENOMEM;
-		new = kmalloc(max_reqs * sizeof(struct nfs4_slot),
-			      GFP_NOFS);
+		new = nfs4_alloc_slots(max_reqs, GFP_NOFS);
 		if (!new)
 			goto out;
-		ret = 0;
-		kfree(tbl->slots);
 	}
-	spin_lock(&tbl->slot_tbl_lock);
-	if (new) {
-		tbl->slots = new;
-		tbl->max_slots = max_reqs;
-	}
-	for (i = 0; i < tbl->max_slots; ++i)
-		tbl->slots[i].seq_nr = ivalue;
-	spin_unlock(&tbl->slot_tbl_lock);
+	ret = 0;
+
+	nfs4_add_and_init_slots(tbl, new, max_reqs, ivalue);
 	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
 		tbl, tbl->slots, tbl->max_slots);
 out:
@@ -5061,36 +5080,6 @@ static void nfs4_destroy_slot_tables(struct nfs4_session *session)
 }
 
 /*
- * Initialize slot table
- */
-static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
-		int max_slots, int ivalue)
-{
-	struct nfs4_slot *slot;
-	int ret = -ENOMEM;
-
-	BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE);
-
-	dprintk("--> %s: max_reqs=%u\n", __func__, max_slots);
-
-	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_NOFS);
-	if (!slot)
-		goto out;
-	ret = 0;
-
-	spin_lock(&tbl->slot_tbl_lock);
-	tbl->max_slots = max_slots;
-	tbl->slots = slot;
-	tbl->highest_used_slotid = -1;  /* no slot is currently used */
-	spin_unlock(&tbl->slot_tbl_lock);
-	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
-		tbl, tbl->slots, tbl->max_slots);
-out:
-	dprintk("<-- %s: return %d\n", __func__, ret);
-	return ret;
-}
-
-/*
  * Initialize or reset the forechannel and backchannel tables
  */
 static int nfs4_setup_session_slot_tables(struct nfs4_session *ses)
@@ -5101,25 +5090,16 @@ static int nfs4_setup_session_slot_tables(struct nfs4_session *ses)
 	dprintk("--> %s\n", __func__);
 	/* Fore channel */
 	tbl = &ses->fc_slot_table;
-	if (tbl->slots == NULL) {
-		status = nfs4_init_slot_table(tbl, ses->fc_attrs.max_reqs, 1);
-		if (status) /* -ENOMEM */
-			return status;
-	} else {
-		status = nfs4_reset_slot_table(tbl, ses->fc_attrs.max_reqs, 1);
-		if (status)
-			return status;
-	}
+	status = nfs4_realloc_slot_table(tbl, ses->fc_attrs.max_reqs, 1);
+	if (status) /* -ENOMEM */
+		return status;
 	/* Back channel */
 	tbl = &ses->bc_slot_table;
-	if (tbl->slots == NULL) {
-		status = nfs4_init_slot_table(tbl, ses->bc_attrs.max_reqs, 0);
-		if (status)
-			/* Fore and back channel share a connection so get
-			 * both slot tables or neither */
-			nfs4_destroy_slot_tables(ses);
-	} else
-		status = nfs4_reset_slot_table(tbl, ses->bc_attrs.max_reqs, 0);
+	status = nfs4_realloc_slot_table(tbl, ses->bc_attrs.max_reqs, 0);
+	if (status && tbl->slots == NULL)
+		/* Fore and back channel share a connection so get
+		 * both slot tables or neither */
+		nfs4_destroy_slot_tables(ses);
 	return status;
 }
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index a53f33b4ac3a..45392032e7bd 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1132,6 +1132,8 @@ void nfs4_schedule_stateid_recovery(const struct nfs_server *server, struct nfs4
 {
 	struct nfs_client *clp = server->nfs_client;
 
+	if (test_and_clear_bit(NFS_DELEGATED_STATE, &state->flags))
+		nfs_async_inode_return_delegation(state->inode, &state->stateid);
 	nfs4_state_mark_reclaim_nograce(clp, state);
 	nfs4_schedule_state_manager(clp);
 }
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 95e92e438407..33bd8d0f745d 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -2522,7 +2522,6 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr,
 
 	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
 		args->acl_pages, args->acl_pgbase, args->acl_len);
-	xdr_set_scratch_buffer(xdr, page_address(args->acl_scratch), PAGE_SIZE);
 
 	encode_nops(&hdr);
 }
@@ -6032,6 +6031,10 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, struct xdr_stream *xdr,
 	struct compound_hdr hdr;
 	int status;
 
+	if (res->acl_scratch != NULL) {
+		void *p = page_address(res->acl_scratch);
+		xdr_set_scratch_buffer(xdr, p, PAGE_SIZE);
+	}
 	status = decode_compound_hdr(xdr, &hdr);
 	if (status)
 		goto out;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 0c3885255f97..834f0fe96f89 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1688,7 +1688,7 @@ out_error:
 
 #ifdef CONFIG_MIGRATION
 int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
-		struct page *page)
+		struct page *page, enum migrate_mode mode)
 {
 	/*
 	 * If PagePrivate is set, then the page is currently associated with
@@ -1703,7 +1703,7 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 
 	nfs_fscache_release_page(page, GFP_KERNEL);
 
-	return migrate_page(mapping, newpage, page);
+	return migrate_page(mapping, newpage, page, mode);
 }
 #endif