Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

6 files changed, 2862 insertions, 0 deletions

diff --git a/fs/nfs/blocklayout/Makefile b/fs/nfs/blocklayout/Makefile
new file mode 100644
index 000000000..7668a1bfb
--- /dev/null
+++ b/fs/nfs/blocklayout/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile for the pNFS block layout driver kernel module
+#
+obj-$(CONFIG_PNFS_BLOCK) += blocklayoutdriver.o
+
+blocklayoutdriver-y += blocklayout.o dev.o extent_tree.o rpc_pipefs.o
diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
new file mode 100644
index 000000000..d54a141a8
--- /dev/null
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -0,0 +1,1047 @@
+/*
+ *  linux/fs/nfs/blocklayout/blocklayout.c
+ *
+ *  Module for the NFSv4.1 pNFS block layout driver.
+ *
+ *  Copyright (c) 2006 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/bio.h>		/* struct bio */
+#include <linux/prefetch.h>
+
+#include "../pnfs.h"
+#include "../nfs4session.h"
+#include "../internal.h"
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY	NFSDBG_PNFS_LD
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Andy Adamson <andros@citi.umich.edu>");
+MODULE_DESCRIPTION("The NFSv4.1 pNFS Block layout driver");
+
+static bool is_hole(struct pnfs_block_extent *be)
+{
+	switch (be->be_state) {
+	case PNFS_BLOCK_NONE_DATA:
+		return true;
+	case PNFS_BLOCK_INVALID_DATA:
+		return be->be_tag ? false : true;
+	default:
+		return false;
+	}
+}
+
+/* The data we are handed might be spread across several bios.  We need
+ * to track when the last one is finished.
+ */
+struct parallel_io {
+	struct kref refcnt;
+	void (*pnfs_callback) (void *data);
+	void *data;
+};
+
+static inline struct parallel_io *alloc_parallel(void *data)
+{
+	struct parallel_io *rv;
+
+	rv = kmalloc_obj(*rv, GFP_NOFS);
+	if (rv) {
+		rv->data = data;
+		kref_init(&rv->refcnt);
+	}
+	return rv;
+}
+
+static inline void get_parallel(struct parallel_io *p)
+{
+	kref_get(&p->refcnt);
+}
+
+static void destroy_parallel(struct kref *kref)
+{
+	struct parallel_io *p = container_of(kref, struct parallel_io, refcnt);
+
+	dprintk("%s enter\n", __func__);
+	p->pnfs_callback(p->data);
+	kfree(p);
+}
+
+static inline void put_parallel(struct parallel_io *p)
+{
+	kref_put(&p->refcnt, destroy_parallel);
+}
+
+static struct bio *
+bl_submit_bio(struct bio *bio)
+{
+	if (bio) {
+		get_parallel(bio->bi_private);
+		dprintk("%s submitting %s bio %u@%llu\n", __func__,
+			bio_op(bio) == READ ? "read" : "write",
+			bio->bi_iter.bi_size,
+			(unsigned long long)bio->bi_iter.bi_sector);
+		submit_bio(bio);
+	}
+	return NULL;
+}
+
+static bool offset_in_map(u64 offset, struct pnfs_block_dev_map *map)
+{
+	return offset >= map->start && offset < map->start + map->len;
+}
+
+static struct bio *
+do_add_page_to_bio(struct bio *bio, int npg, enum req_op op, sector_t isect,
+		struct page *page, struct pnfs_block_dev_map *map,
+		struct pnfs_block_extent *be, bio_end_io_t end_io,
+		struct parallel_io *par, unsigned int offset, int *len)
+{
+	struct pnfs_block_dev *dev =
+		container_of(be->be_device, struct pnfs_block_dev, node);
+	u64 disk_addr, end;
+
+	dprintk("%s: npg %d rw %d isect %llu offset %u len %d\n", __func__,
+		npg, (__force u32)op, (unsigned long long)isect, offset, *len);
+
+	/* translate to device offset */
+	isect += be->be_v_offset;
+	isect -= be->be_f_offset;
+
+	/* translate to physical disk offset */
+	disk_addr = (u64)isect << SECTOR_SHIFT;
+	if (!offset_in_map(disk_addr, map)) {
+		if (!dev->map(dev, disk_addr, map) || !offset_in_map(disk_addr, map))
+			return ERR_PTR(-EIO);
+		bio = bl_submit_bio(bio);
+	}
+	disk_addr += map->disk_offset;
+	disk_addr -= map->start;
+
+	/* limit length to what the device mapping allows */
+	end = disk_addr + *len;
+	if (end >= map->disk_offset + map->len)
+		*len = map->disk_offset + map->len - disk_addr;
+
+retry:
+	if (!bio) {
+		bio = bio_alloc(map->bdev, bio_max_segs(npg), op, GFP_NOIO);
+		bio->bi_iter.bi_sector = disk_addr >> SECTOR_SHIFT;
+		bio->bi_end_io = end_io;
+		bio->bi_private = par;
+	}
+	if (bio_add_page(bio, page, *len, offset) < *len) {
+		bio = bl_submit_bio(bio);
+		goto retry;
+	}
+	return bio;
+}
+
+static void bl_mark_devices_unavailable(struct nfs_pgio_header *header, bool rw)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	size_t bytes_left = header->args.count;
+	sector_t isect, extent_length = 0;
+	struct pnfs_block_extent be;
+
+	isect = header->args.offset >> SECTOR_SHIFT;
+	bytes_left += header->args.offset - (isect << SECTOR_SHIFT);
+
+	while (bytes_left > 0) {
+		if (!ext_tree_lookup(bl, isect, &be, rw))
+				return;
+		extent_length = be.be_length - (isect - be.be_f_offset);
+		nfs4_mark_deviceid_unavailable(be.be_device);
+		isect += extent_length;
+		if (bytes_left > extent_length << SECTOR_SHIFT)
+			bytes_left -= extent_length << SECTOR_SHIFT;
+		else
+			bytes_left = 0;
+	}
+}
+
+static void bl_end_io_read(struct bio *bio)
+{
+	struct parallel_io *par = bio->bi_private;
+
+	if (bio->bi_status) {
+		struct nfs_pgio_header *header = par->data;
+
+		if (!header->pnfs_error)
+			header->pnfs_error = -EIO;
+		pnfs_set_lo_fail(header->lseg);
+		bl_mark_devices_unavailable(header, false);
+	}
+
+	bio_put(bio);
+	put_parallel(par);
+}
+
+static void bl_read_cleanup(struct work_struct *work)
+{
+	struct rpc_task *task;
+	struct nfs_pgio_header *hdr;
+	dprintk("%s enter\n", __func__);
+	task = container_of(work, struct rpc_task, u.tk_work);
+	hdr = container_of(task, struct nfs_pgio_header, task);
+	pnfs_ld_read_done(hdr);
+}
+
+static void
+bl_end_par_io_read(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	hdr->task.tk_status = hdr->pnfs_error;
+	INIT_WORK(&hdr->task.u.tk_work, bl_read_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
+}
+
+static enum pnfs_try_status
+bl_read_pagelist(struct nfs_pgio_header *header)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 };
+	struct bio *bio = NULL;
+	struct pnfs_block_extent be;
+	sector_t isect, extent_length = 0;
+	struct parallel_io *par;
+	loff_t f_offset = header->args.offset;
+	size_t bytes_left = header->args.count;
+	unsigned int pg_offset = header->args.pgbase, pg_len;
+	struct page **pages = header->args.pages;
+	int pg_index = header->args.pgbase >> PAGE_SHIFT;
+	const bool is_dio = (header->dreq != NULL);
+	struct blk_plug plug;
+	int i;
+
+	dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__,
+		header->page_array.npages, f_offset,
+		(unsigned int)header->args.count);
+
+	par = alloc_parallel(header);
+	if (!par)
+		return PNFS_NOT_ATTEMPTED;
+	par->pnfs_callback = bl_end_par_io_read;
+
+	blk_start_plug(&plug);
+
+	isect = (sector_t) (f_offset >> SECTOR_SHIFT);
+	/* Code assumes extents are page-aligned */
+	for (i = pg_index; i < header->page_array.npages; i++) {
+		if (extent_length <= 0) {
+			/* We've used up the previous extent */
+			bio = bl_submit_bio(bio);
+
+			/* Get the next one */
+			if (!ext_tree_lookup(bl, isect, &be, false)) {
+				header->pnfs_error = -EIO;
+				goto out;
+			}
+			extent_length = be.be_length - (isect - be.be_f_offset);
+		}
+
+		if (is_dio) {
+			if (pg_offset + bytes_left > PAGE_SIZE)
+				pg_len = PAGE_SIZE - pg_offset;
+			else
+				pg_len = bytes_left;
+		} else {
+			BUG_ON(pg_offset != 0);
+			pg_len = PAGE_SIZE;
+		}
+
+		if (is_hole(&be)) {
+			bio = bl_submit_bio(bio);
+			/* Fill hole w/ zeroes w/o accessing device */
+			dprintk("%s Zeroing page for hole\n", __func__);
+			zero_user_segment(pages[i], pg_offset, pg_len);
+
+			/* invalidate map */
+			map.start = NFS4_MAX_UINT64;
+		} else {
+			bio = do_add_page_to_bio(bio,
+						 header->page_array.npages - i,
+						 REQ_OP_READ,
+						 isect, pages[i], &map, &be,
+						 bl_end_io_read, par,
+						 pg_offset, &pg_len);
+			if (IS_ERR(bio)) {
+				header->pnfs_error = PTR_ERR(bio);
+				bio = NULL;
+				goto out;
+			}
+		}
+		isect += (pg_len >> SECTOR_SHIFT);
+		extent_length -= (pg_len >> SECTOR_SHIFT);
+		f_offset += pg_len;
+		bytes_left -= pg_len;
+		pg_offset = 0;
+	}
+	if ((isect << SECTOR_SHIFT) >= header->inode->i_size) {
+		header->res.eof = 1;
+		header->res.count = header->inode->i_size - header->args.offset;
+	} else {
+		header->res.count = (isect << SECTOR_SHIFT) - header->args.offset;
+	}
+out:
+	bl_submit_bio(bio);
+	blk_finish_plug(&plug);
+	put_parallel(par);
+	return PNFS_ATTEMPTED;
+}
+
+static void bl_end_io_write(struct bio *bio)
+{
+	struct parallel_io *par = bio->bi_private;
+	struct nfs_pgio_header *header = par->data;
+
+	if (bio->bi_status) {
+		if (!header->pnfs_error)
+			header->pnfs_error = -EIO;
+		pnfs_set_lo_fail(header->lseg);
+		bl_mark_devices_unavailable(header, true);
+	}
+	bio_put(bio);
+	put_parallel(par);
+}
+
+/* Function scheduled for call during bl_end_par_io_write,
+ * it marks sectors as written and extends the commitlist.
+ */
+static void bl_write_cleanup(struct work_struct *work)
+{
+	struct rpc_task *task = container_of(work, struct rpc_task, u.tk_work);
+	struct nfs_pgio_header *hdr =
+			container_of(task, struct nfs_pgio_header, task);
+
+	dprintk("%s enter\n", __func__);
+
+	if (likely(!hdr->pnfs_error)) {
+		struct pnfs_block_layout *bl = BLK_LSEG2EXT(hdr->lseg);
+		u64 start = hdr->args.offset & (loff_t)PAGE_MASK;
+		u64 end = (hdr->args.offset + hdr->args.count +
+			PAGE_SIZE - 1) & (loff_t)PAGE_MASK;
+		u64 lwb = hdr->args.offset + hdr->args.count;
+
+		ext_tree_mark_written(bl, start >> SECTOR_SHIFT,
+					(end - start) >> SECTOR_SHIFT, lwb);
+	}
+
+	pnfs_ld_write_done(hdr);
+}
+
+/* Called when last of bios associated with a bl_write_pagelist call finishes */
+static void bl_end_par_io_write(void *data)
+{
+	struct nfs_pgio_header *hdr = data;
+
+	hdr->task.tk_status = hdr->pnfs_error;
+	hdr->verf.committed = NFS_FILE_SYNC;
+	INIT_WORK(&hdr->task.u.tk_work, bl_write_cleanup);
+	schedule_work(&hdr->task.u.tk_work);
+}
+
+static enum pnfs_try_status
+bl_write_pagelist(struct nfs_pgio_header *header, int sync)
+{
+	struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg);
+	struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 };
+	struct bio *bio = NULL;
+	struct pnfs_block_extent be;
+	sector_t isect, extent_length = 0;
+	struct parallel_io *par = NULL;
+	loff_t offset = header->args.offset;
+	struct page **pages = header->args.pages;
+	int pg_index = header->args.pgbase >> PAGE_SHIFT;
+	unsigned int pg_len;
+	struct blk_plug plug;
+	int i;
+
+	dprintk("%s enter, %u@%lld\n", __func__, header->args.count, offset);
+
+	/* At this point, header->page_aray is a (sequential) list of nfs_pages.
+	 * We want to write each, and if there is an error set pnfs_error
+	 * to have it redone using nfs.
+	 */
+	par = alloc_parallel(header);
+	if (!par)
+		return PNFS_NOT_ATTEMPTED;
+	par->pnfs_callback = bl_end_par_io_write;
+
+	blk_start_plug(&plug);
+
+	/* we always write out the whole page */
+	offset = offset & (loff_t)PAGE_MASK;
+	isect = offset >> SECTOR_SHIFT;
+
+	for (i = pg_index; i < header->page_array.npages; i++) {
+		if (extent_length <= 0) {
+			/* We've used up the previous extent */
+			bio = bl_submit_bio(bio);
+			/* Get the next one */
+			if (!ext_tree_lookup(bl, isect, &be, true)) {
+				header->pnfs_error = -EINVAL;
+				goto out;
+			}
+
+			extent_length = be.be_length - (isect - be.be_f_offset);
+		}
+
+		pg_len = PAGE_SIZE;
+		bio = do_add_page_to_bio(bio, header->page_array.npages - i,
+					 REQ_OP_WRITE, isect, pages[i], &map,
+					 &be, bl_end_io_write, par, 0, &pg_len);
+		if (IS_ERR(bio)) {
+			header->pnfs_error = PTR_ERR(bio);
+			bio = NULL;
+			goto out;
+		}
+
+		offset += pg_len;
+		isect += (pg_len >> SECTOR_SHIFT);
+		extent_length -= (pg_len >> SECTOR_SHIFT);
+	}
+
+	header->res.count = header->args.count;
+out:
+	bl_submit_bio(bio);
+	blk_finish_plug(&plug);
+	put_parallel(par);
+	return PNFS_ATTEMPTED;
+}
+
+static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	int err;
+
+	dprintk("%s enter\n", __func__);
+
+	err = ext_tree_remove(bl, true, 0, LLONG_MAX);
+	WARN_ON(err);
+
+	kfree_rcu(bl, bl_layout.plh_rcu);
+}
+
+static struct pnfs_layout_hdr *__bl_alloc_layout_hdr(struct inode *inode,
+		gfp_t gfp_flags, bool is_scsi_layout)
+{
+	struct pnfs_block_layout *bl;
+
+	dprintk("%s enter\n", __func__);
+	bl = kzalloc_obj(*bl, gfp_flags);
+	if (!bl)
+		return NULL;
+
+	bl->bl_ext_rw = RB_ROOT;
+	bl->bl_ext_ro = RB_ROOT;
+	spin_lock_init(&bl->bl_ext_lock);
+
+	bl->bl_scsi_layout = is_scsi_layout;
+	return &bl->bl_layout;
+}
+
+static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode,
+						   gfp_t gfp_flags)
+{
+	return __bl_alloc_layout_hdr(inode, gfp_flags, false);
+}
+
+static struct pnfs_layout_hdr *sl_alloc_layout_hdr(struct inode *inode,
+						   gfp_t gfp_flags)
+{
+	return __bl_alloc_layout_hdr(inode, gfp_flags, true);
+}
+
+static void bl_free_lseg(struct pnfs_layout_segment *lseg)
+{
+	dprintk("%s enter\n", __func__);
+	kfree(lseg);
+}
+
+/* Tracks info needed to ensure extents in layout obey constraints of spec */
+struct layout_verification {
+	u32 mode;	/* R or RW */
+	u64 start;	/* Expected start of next non-COW extent */
+	u64 inval;	/* Start of INVAL coverage */
+	u64 cowread;	/* End of COW read coverage */
+};
+
+/* Verify the extent meets the layout requirements of the pnfs-block draft,
+ * section 2.3.1.
+ */
+static int verify_extent(struct pnfs_block_extent *be,
+			 struct layout_verification *lv)
+{
+	if (lv->mode == IOMODE_READ) {
+		if (be->be_state == PNFS_BLOCK_READWRITE_DATA ||
+		    be->be_state == PNFS_BLOCK_INVALID_DATA)
+			return -EIO;
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		return 0;
+	}
+	/* lv->mode == IOMODE_RW */
+	if (be->be_state == PNFS_BLOCK_READWRITE_DATA) {
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		if (lv->cowread > lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		lv->inval = lv->start;
+		return 0;
+	} else if (be->be_state == PNFS_BLOCK_INVALID_DATA) {
+		if (be->be_f_offset != lv->start)
+			return -EIO;
+		lv->start += be->be_length;
+		return 0;
+	} else if (be->be_state == PNFS_BLOCK_READ_DATA) {
+		if (be->be_f_offset > lv->start)
+			return -EIO;
+		if (be->be_f_offset < lv->inval)
+			return -EIO;
+		if (be->be_f_offset < lv->cowread)
+			return -EIO;
+		/* It looks like you might want to min this with lv->start,
+		 * but you really don't.
+		 */
+		lv->inval = lv->inval + be->be_length;
+		lv->cowread = be->be_f_offset + be->be_length;
+		return 0;
+	} else
+		return -EIO;
+}
+
+static int decode_sector_number(__be32 **rp, sector_t *sp)
+{
+	uint64_t s;
+
+	*rp = xdr_decode_hyper(*rp, &s);
+	if (s & 0x1ff) {
+		printk(KERN_WARNING "NFS: %s: sector not aligned\n", __func__);
+		return -1;
+	}
+	*sp = s >> SECTOR_SHIFT;
+	return 0;
+}
+
+static struct nfs4_deviceid_node *
+bl_find_get_deviceid(struct nfs_server *server,
+		const struct nfs4_deviceid *id, const struct cred *cred,
+		gfp_t gfp_mask)
+{
+	struct nfs4_deviceid_node *node;
+	int err = -ENODEV;
+
+retry:
+	node = nfs4_find_get_deviceid(server, id, cred, gfp_mask);
+	if (!node)
+		return ERR_PTR(-ENODEV);
+
+	/*
+	 * Devices that are marked unavailable are left in the cache with a
+	 * timeout to avoid sending GETDEVINFO after every LAYOUTGET, or
+	 * constantly attempting to register the device.  Once marked as
+	 * unavailable they must be deleted and never reused.
+	 */
+	if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) {
+		unsigned long end = jiffies;
+		unsigned long start = end - PNFS_DEVICE_RETRY_TIMEOUT;
+
+		if (!time_in_range(node->timestamp_unavailable, start, end)) {
+			/* Uncork subsequent GETDEVINFO operations for this device */
+			nfs4_delete_deviceid(node->ld, node->nfs_client, id);
+			goto retry;
+		}
+		goto out_put;
+	}
+
+	if (!bl_register_dev(container_of(node, struct pnfs_block_dev, node))) {
+		/*
+		 * If we cannot register, treat this device as transient:
+		 * Make a negative cache entry for the device
+		 */
+		nfs4_mark_deviceid_unavailable(node);
+		goto out_put;
+	}
+
+	return node;
+
+out_put:
+	nfs4_put_deviceid_node(node);
+	return ERR_PTR(err);
+}
+
+static int
+bl_alloc_extent(struct xdr_stream *xdr, struct pnfs_layout_hdr *lo,
+		struct layout_verification *lv, struct list_head *extents,
+		gfp_t gfp_mask)
+{
+	struct pnfs_block_extent *be;
+	struct nfs4_deviceid id;
+	int error;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, 28 + NFS4_DEVICEID4_SIZE);
+	if (!p)
+		return -EIO;
+
+	be = kzalloc_obj(*be, GFP_NOFS);
+	if (!be)
+		return -ENOMEM;
+
+	memcpy(&id, p, NFS4_DEVICEID4_SIZE);
+	p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE);
+
+	be->be_device = bl_find_get_deviceid(NFS_SERVER(lo->plh_inode), &id,
+						lo->plh_lc_cred, gfp_mask);
+	if (IS_ERR(be->be_device)) {
+		error = PTR_ERR(be->be_device);
+		goto out_free_be;
+	}
+
+	/*
+	 * The next three values are read in as bytes, but stored in the
+	 * extent structure in 512-byte granularity.
+	 */
+	error = -EIO;
+	if (decode_sector_number(&p, &be->be_f_offset) < 0)
+		goto out_put_deviceid;
+	if (decode_sector_number(&p, &be->be_length) < 0)
+		goto out_put_deviceid;
+	if (decode_sector_number(&p, &be->be_v_offset) < 0)
+		goto out_put_deviceid;
+	be->be_state = be32_to_cpup(p++);
+
+	error = verify_extent(be, lv);
+	if (error) {
+		dprintk("%s: extent verification failed\n", __func__);
+		goto out_put_deviceid;
+	}
+
+	list_add_tail(&be->be_list, extents);
+	return 0;
+
+out_put_deviceid:
+	nfs4_put_deviceid_node(be->be_device);
+out_free_be:
+	kfree(be);
+	return error;
+}
+
+static struct pnfs_layout_segment *
+bl_alloc_lseg(struct pnfs_layout_hdr *lo, struct nfs4_layoutget_res *lgr,
+		gfp_t gfp_mask)
+{
+	struct layout_verification lv = {
+		.mode = lgr->range.iomode,
+		.start = lgr->range.offset >> SECTOR_SHIFT,
+		.inval = lgr->range.offset >> SECTOR_SHIFT,
+		.cowread = lgr->range.offset >> SECTOR_SHIFT,
+	};
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	struct pnfs_layout_segment *lseg;
+	struct xdr_buf buf;
+	struct xdr_stream xdr;
+	struct folio *scratch;
+	int status, i;
+	uint32_t count;
+	__be32 *p;
+	LIST_HEAD(extents);
+
+	dprintk("---> %s\n", __func__);
+
+	lseg = kzalloc_obj(*lseg, gfp_mask);
+	if (!lseg)
+		return ERR_PTR(-ENOMEM);
+
+	status = -ENOMEM;
+	scratch = folio_alloc(gfp_mask, 0);
+	if (!scratch)
+		goto out;
+
+	xdr_init_decode_pages(&xdr, &buf,
+			lgr->layoutp->pages, lgr->layoutp->len);
+	xdr_set_scratch_folio(&xdr, scratch);
+
+	status = -EIO;
+	p = xdr_inline_decode(&xdr, 4);
+	if (unlikely(!p))
+		goto out_free_scratch;
+
+	count = be32_to_cpup(p++);
+	dprintk("%s: number of extents %d\n", __func__, count);
+
+	/*
+	 * Decode individual extents, putting them in temporary staging area
+	 * until whole layout is decoded to make error recovery easier.
+	 */
+	for (i = 0; i < count; i++) {
+		status = bl_alloc_extent(&xdr, lo, &lv, &extents, gfp_mask);
+		if (status)
+			goto process_extents;
+	}
+
+	if (lgr->range.offset + lgr->range.length !=
+			lv.start << SECTOR_SHIFT) {
+		dprintk("%s Final length mismatch\n", __func__);
+		status = -EIO;
+		goto process_extents;
+	}
+
+	if (lv.start < lv.cowread) {
+		dprintk("%s Final uncovered COW extent\n", __func__);
+		status = -EIO;
+	}
+
+process_extents:
+	while (!list_empty(&extents)) {
+		struct pnfs_block_extent *be =
+			list_first_entry(&extents, struct pnfs_block_extent,
+					 be_list);
+		list_del(&be->be_list);
+
+		if (!status)
+			status = ext_tree_insert(bl, be);
+
+		if (status) {
+			nfs4_put_deviceid_node(be->be_device);
+			kfree(be);
+		}
+	}
+
+out_free_scratch:
+	folio_put(scratch);
+out:
+	dprintk("%s returns %d\n", __func__, status);
+	switch (status) {
+	case -ENODEV:
+		/* Our extent block devices are unavailable */
+		set_bit(NFS_LSEG_UNAVAILABLE, &lseg->pls_flags);
+		fallthrough;
+	case 0:
+		return lseg;
+	default:
+		kfree(lseg);
+		return ERR_PTR(status);
+	}
+}
+
+static void
+bl_return_range(struct pnfs_layout_hdr *lo,
+		struct pnfs_layout_range *range)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(lo);
+	sector_t offset = range->offset >> SECTOR_SHIFT, end;
+
+	if (range->offset % 8) {
+		dprintk("%s: offset %lld not block size aligned\n",
+			__func__, range->offset);
+		return;
+	}
+
+	if (range->length != NFS4_MAX_UINT64) {
+		if (range->length % 8) {
+			dprintk("%s: length %lld not block size aligned\n",
+				__func__, range->length);
+			return;
+		}
+
+		end = offset + (range->length >> SECTOR_SHIFT);
+	} else {
+		end = round_down(NFS4_MAX_UINT64, PAGE_SIZE);
+	}
+
+	ext_tree_remove(bl, range->iomode & IOMODE_RW, offset, end);
+}
+
+static int
+bl_prepare_layoutcommit(struct nfs4_layoutcommit_args *arg)
+{
+	return ext_tree_prepare_commit(arg);
+}
+
+static void
+bl_cleanup_layoutcommit(struct nfs4_layoutcommit_data *lcdata)
+{
+	ext_tree_mark_committed(&lcdata->args, lcdata->res.status);
+}
+
+static int
+bl_set_layoutdriver(struct nfs_server *server, const struct nfs_fh *fh)
+{
+	dprintk("%s enter\n", __func__);
+
+	if (server->pnfs_blksize == 0) {
+		dprintk("%s Server did not return blksize\n", __func__);
+		return -EINVAL;
+	}
+	if (server->pnfs_blksize > PAGE_SIZE) {
+		printk(KERN_ERR "%s: pNFS blksize %d not supported.\n",
+			__func__, server->pnfs_blksize);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static bool
+is_aligned_req(struct nfs_pageio_descriptor *pgio,
+		struct nfs_page *req, unsigned int alignment, bool is_write)
+{
+	/*
+	 * Always accept buffered writes, higher layers take care of the
+	 * right alignment.
+	 */
+	if (pgio->pg_dreq == NULL)
+		return true;
+
+	if (!IS_ALIGNED(req->wb_offset, alignment))
+		return false;
+
+	if (IS_ALIGNED(req->wb_bytes, alignment))
+		return true;
+
+	if (is_write &&
+	    (req_offset(req) + req->wb_bytes == i_size_read(pgio->pg_inode))) {
+		/*
+		 * If the write goes up to the inode size, just write
+		 * the full page.  Data past the inode size is
+		 * guaranteed to be zeroed by the higher level client
+		 * code, and this behaviour is mandated by RFC 5663
+		 * section 2.3.2.
+		 */
+		return true;
+	}
+
+	return false;
+}
+
+static void
+bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE, false)) {
+		nfs_pageio_reset_read_mds(pgio);
+		return;
+	}
+
+	pnfs_generic_pg_init_read(pgio, req);
+
+	if (pgio->pg_lseg &&
+		test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) {
+		pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg);
+		pnfs_set_lo_fail(pgio->pg_lseg);
+		nfs_pageio_reset_read_mds(pgio);
+	}
+}
+
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t
+bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+		struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, SECTOR_SIZE, false))
+		return 0;
+	return pnfs_generic_pg_test(pgio, prev, req);
+}
+
+/*
+ * Return the number of contiguous bytes for a given inode
+ * starting at page frame idx.
+ */
+static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx)
+{
+	struct address_space *mapping = inode->i_mapping;
+	pgoff_t end;
+
+	/* Optimize common case that writes from 0 to end of file */
+	end = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE);
+	if (end != inode->i_mapping->nrpages) {
+		rcu_read_lock();
+		end = page_cache_next_miss(mapping, idx + 1, ULONG_MAX);
+		rcu_read_unlock();
+	}
+
+	if (!end)
+		return i_size_read(inode) - (idx << PAGE_SHIFT);
+	else
+		return (end - idx) << PAGE_SHIFT;
+}
+
+static void
+bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	u64 wb_size;
+
+	if (!is_aligned_req(pgio, req, PAGE_SIZE, true)) {
+		nfs_pageio_reset_write_mds(pgio);
+		return;
+	}
+
+	if (pgio->pg_dreq == NULL)
+		wb_size = pnfs_num_cont_bytes(pgio->pg_inode, req->wb_index);
+	else
+		wb_size = nfs_dreq_bytes_left(pgio->pg_dreq, req_offset(req));
+
+	pnfs_generic_pg_init_write(pgio, req, wb_size);
+
+	if (pgio->pg_lseg &&
+		test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) {
+
+		pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg);
+		pnfs_set_lo_fail(pgio->pg_lseg);
+		nfs_pageio_reset_write_mds(pgio);
+	}
+}
+
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t
+bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+		 struct nfs_page *req)
+{
+	if (!is_aligned_req(pgio, req, PAGE_SIZE, true))
+		return 0;
+	return pnfs_generic_pg_test(pgio, prev, req);
+}
+
+static const struct nfs_pageio_ops bl_pg_read_ops = {
+	.pg_init = bl_pg_init_read,
+	.pg_test = bl_pg_test_read,
+	.pg_doio = pnfs_generic_pg_readpages,
+	.pg_cleanup = pnfs_generic_pg_cleanup,
+};
+
+static const struct nfs_pageio_ops bl_pg_write_ops = {
+	.pg_init = bl_pg_init_write,
+	.pg_test = bl_pg_test_write,
+	.pg_doio = pnfs_generic_pg_writepages,
+	.pg_cleanup = pnfs_generic_pg_cleanup,
+};
+
+static struct pnfs_layoutdriver_type blocklayout_type = {
+	.id				= LAYOUT_BLOCK_VOLUME,
+	.name				= "LAYOUT_BLOCK_VOLUME",
+	.owner				= THIS_MODULE,
+	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_LAYOUTRET_ON_ERROR |
+					  PNFS_READ_WHOLE_PAGE,
+	.read_pagelist			= bl_read_pagelist,
+	.write_pagelist			= bl_write_pagelist,
+	.alloc_layout_hdr		= bl_alloc_layout_hdr,
+	.free_layout_hdr		= bl_free_layout_hdr,
+	.alloc_lseg			= bl_alloc_lseg,
+	.free_lseg			= bl_free_lseg,
+	.return_range			= bl_return_range,
+	.prepare_layoutcommit		= bl_prepare_layoutcommit,
+	.cleanup_layoutcommit		= bl_cleanup_layoutcommit,
+	.set_layoutdriver		= bl_set_layoutdriver,
+	.alloc_deviceid_node		= bl_alloc_deviceid_node,
+	.free_deviceid_node		= bl_free_deviceid_node,
+	.pg_read_ops			= &bl_pg_read_ops,
+	.pg_write_ops			= &bl_pg_write_ops,
+	.sync				= pnfs_generic_sync,
+};
+
+static struct pnfs_layoutdriver_type scsilayout_type = {
+	.id				= LAYOUT_SCSI,
+	.name				= "LAYOUT_SCSI",
+	.owner				= THIS_MODULE,
+	.flags				= PNFS_LAYOUTRET_ON_SETATTR |
+					  PNFS_LAYOUTRET_ON_ERROR |
+					  PNFS_READ_WHOLE_PAGE,
+	.read_pagelist			= bl_read_pagelist,
+	.write_pagelist			= bl_write_pagelist,
+	.alloc_layout_hdr		= sl_alloc_layout_hdr,
+	.free_layout_hdr		= bl_free_layout_hdr,
+	.alloc_lseg			= bl_alloc_lseg,
+	.free_lseg			= bl_free_lseg,
+	.return_range			= bl_return_range,
+	.prepare_layoutcommit		= bl_prepare_layoutcommit,
+	.cleanup_layoutcommit		= bl_cleanup_layoutcommit,
+	.set_layoutdriver		= bl_set_layoutdriver,
+	.alloc_deviceid_node		= bl_alloc_deviceid_node,
+	.free_deviceid_node		= bl_free_deviceid_node,
+	.pg_read_ops			= &bl_pg_read_ops,
+	.pg_write_ops			= &bl_pg_write_ops,
+	.sync				= pnfs_generic_sync,
+};
+
+
+static int __init nfs4blocklayout_init(void)
+{
+	int ret;
+
+	dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__);
+
+	ret = bl_init_pipefs();
+	if (ret)
+		goto out;
+
+	ret = pnfs_register_layoutdriver(&blocklayout_type);
+	if (ret)
+		goto out_cleanup_pipe;
+
+	ret = pnfs_register_layoutdriver(&scsilayout_type);
+	if (ret)
+		goto out_unregister_block;
+	return 0;
+
+out_unregister_block:
+	pnfs_unregister_layoutdriver(&blocklayout_type);
+out_cleanup_pipe:
+	bl_cleanup_pipefs();
+out:
+	return ret;
+}
+
+static void __exit nfs4blocklayout_exit(void)
+{
+	dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n",
+	       __func__);
+
+	pnfs_unregister_layoutdriver(&scsilayout_type);
+	pnfs_unregister_layoutdriver(&blocklayout_type);
+	bl_cleanup_pipefs();
+}
+
+MODULE_ALIAS("nfs-layouttype4-3");
+MODULE_ALIAS("nfs-layouttype4-5");
+
+module_init(nfs4blocklayout_init);
+module_exit(nfs4blocklayout_exit);
diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h
new file mode 100644
index 000000000..6da40ca19
--- /dev/null
+++ b/fs/nfs/blocklayout/blocklayout.h
@@ -0,0 +1,204 @@
+/*
+ *  linux/fs/nfs/blocklayout/blocklayout.h
+ *
+ *  Module for the NFSv4.1 pNFS block layout driver.
+ *
+ *  Copyright (c) 2006 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+#ifndef FS_NFS_NFS4BLOCKLAYOUT_H
+#define FS_NFS_NFS4BLOCKLAYOUT_H
+
+#include <linux/device-mapper.h>
+#include <linux/nfs_fs.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+
+#include "../nfs4_fs.h"
+#include "../pnfs.h"
+#include "../netns.h"
+
+#define PAGE_CACHE_SECTORS (PAGE_SIZE >> SECTOR_SHIFT)
+#define PAGE_CACHE_SECTOR_SHIFT (PAGE_SHIFT - SECTOR_SHIFT)
+#define SECTOR_SIZE (1 << SECTOR_SHIFT)
+
+struct pnfs_block_dev;
+
+#define PNFS_BLOCK_MAX_UUIDS	4
+#define PNFS_BLOCK_MAX_DEVICES	64
+
+/*
+ * Random upper cap for the uuid length to avoid unbounded allocation.
+ * Not actually limited by the protocol.
+ */
+#define PNFS_BLOCK_UUID_LEN	128
+
+struct pnfs_block_volume {
+	enum pnfs_block_volume_type	type;
+	union {
+		struct {
+			int		len;
+			int		nr_sigs;
+			struct {
+				u64		offset;
+				u32		sig_len;
+				u8		sig[PNFS_BLOCK_UUID_LEN];
+			} sigs[PNFS_BLOCK_MAX_UUIDS];
+		} simple;
+		struct {
+			u64		start;
+			u64		len;
+			u32		volume;
+		} slice;
+		struct {
+			u32		volumes_count;
+			u32		volumes[PNFS_BLOCK_MAX_DEVICES];
+		} concat;
+		struct {
+			u64		chunk_size;
+			u32		volumes_count;
+			u32		volumes[PNFS_BLOCK_MAX_DEVICES];
+		} stripe;
+		struct {
+			enum scsi_code_set		code_set;
+			enum scsi_designator_type	designator_type;
+			int				designator_len;
+			u8				designator[256];
+			u64				pr_key;
+		} scsi;
+	};
+};
+
+struct pnfs_block_dev_map {
+	u64			start;
+	u64			len;
+	u64			disk_offset;
+	struct block_device		*bdev;
+};
+
+struct pnfs_block_dev {
+	struct nfs4_deviceid_node	node;
+
+	u64				start;
+	u64				len;
+
+	enum pnfs_block_volume_type	type;
+	u32				nr_children;
+	struct pnfs_block_dev		*children;
+	u64				chunk_size;
+
+	struct file			*bdev_file;
+	u64				disk_offset;
+	unsigned long			flags;
+
+	u64				pr_key;
+
+	bool (*map)(struct pnfs_block_dev *dev, u64 offset,
+			struct pnfs_block_dev_map *map);
+};
+
+/* pnfs_block_dev flag bits */
+enum {
+	PNFS_BDEV_REGISTERED = 0,
+};
+
+/* sector_t fields are all in 512-byte sectors */
+struct pnfs_block_extent {
+	union {
+		struct rb_node	be_node;
+		struct list_head be_list;
+	};
+	struct nfs4_deviceid_node *be_device;
+	sector_t	be_f_offset;	/* the starting offset in the file */
+	sector_t	be_length;	/* the size of the extent */
+	sector_t	be_v_offset;	/* the starting offset in the volume */
+	enum pnfs_block_extent_state be_state;	/* the state of this extent */
+#define EXTENT_WRITTEN		1
+#define EXTENT_COMMITTING	2
+	unsigned int	be_tag;
+};
+
+struct pnfs_block_layout {
+	struct pnfs_layout_hdr	bl_layout;
+	struct rb_root		bl_ext_rw;
+	struct rb_root		bl_ext_ro;
+	spinlock_t		bl_ext_lock;   /* Protects list manipulation */
+	bool			bl_scsi_layout;
+	u64			bl_lwb;
+};
+
+static inline struct pnfs_block_layout *
+BLK_LO2EXT(struct pnfs_layout_hdr *lo)
+{
+	return container_of(lo, struct pnfs_block_layout, bl_layout);
+}
+
+static inline struct pnfs_block_layout *
+BLK_LSEG2EXT(struct pnfs_layout_segment *lseg)
+{
+	return BLK_LO2EXT(lseg->pls_layout);
+}
+
+struct bl_pipe_msg {
+	struct rpc_pipe_msg msg;
+	wait_queue_head_t *bl_wq;
+};
+
+struct bl_msg_hdr {
+	u8  type;
+	u16 totallen; /* length of entire message, including hdr itself */
+};
+
+#define BL_DEVICE_UMOUNT               0x0 /* Umount--delete devices */
+#define BL_DEVICE_MOUNT                0x1 /* Mount--create devices*/
+#define BL_DEVICE_REQUEST_INIT         0x0 /* Start request */
+#define BL_DEVICE_REQUEST_PROC         0x1 /* User level process succeeds */
+#define BL_DEVICE_REQUEST_ERR          0x2 /* User level process fails */
+
+/* dev.c */
+bool bl_register_dev(struct pnfs_block_dev *d);
+struct nfs4_deviceid_node *bl_alloc_deviceid_node(struct nfs_server *server,
+		struct pnfs_device *pdev, gfp_t gfp_mask);
+void bl_free_deviceid_node(struct nfs4_deviceid_node *d);
+
+/* extent_tree.c */
+int ext_tree_insert(struct pnfs_block_layout *bl,
+		struct pnfs_block_extent *new);
+int ext_tree_remove(struct pnfs_block_layout *bl, bool rw, sector_t start,
+		sector_t end);
+int ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
+		sector_t len, u64 lwb);
+bool ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect,
+		struct pnfs_block_extent *ret, bool rw);
+int ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg);
+void ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status);
+
+/* rpc_pipefs.c */
+dev_t bl_resolve_deviceid(struct nfs_server *server,
+		struct pnfs_block_volume *b, gfp_t gfp_mask);
+int __init bl_init_pipefs(void);
+void bl_cleanup_pipefs(void);
+
+#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */
diff --git a/fs/nfs/blocklayout/dev.c b/fs/nfs/blocklayout/dev.c
new file mode 100644
index 000000000..c926b7e43
--- /dev/null
+++ b/fs/nfs/blocklayout/dev.c
@@ -0,0 +1,608 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2014-2016 Christoph Hellwig.
+ */
+#include <linux/sunrpc/svc.h>
+#include <linux/blkdev.h>
+#include <linux/fs_struct.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_xdr.h>
+#include <linux/pr.h>
+
+#include "blocklayout.h"
+#include "../nfs4trace.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS_LD
+
+static void bl_unregister_scsi(struct pnfs_block_dev *dev)
+{
+	struct block_device *bdev = file_bdev(dev->bdev_file);
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	int status;
+
+	status = ops->pr_register(bdev, dev->pr_key, 0, false);
+	if (status)
+		trace_bl_pr_key_unreg_err(bdev, dev->pr_key, status);
+	else
+		trace_bl_pr_key_unreg(bdev, dev->pr_key);
+}
+
+static bool bl_register_scsi(struct pnfs_block_dev *dev)
+{
+	struct block_device *bdev = file_bdev(dev->bdev_file);
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	int status;
+
+	if (test_and_set_bit(PNFS_BDEV_REGISTERED, &dev->flags))
+		return true;
+
+	status = ops->pr_register(bdev, 0, dev->pr_key, true);
+	if (status) {
+		trace_bl_pr_key_reg_err(bdev, dev->pr_key, status);
+		return false;
+	}
+	trace_bl_pr_key_reg(bdev, dev->pr_key);
+	return true;
+}
+
+static void bl_unregister_dev(struct pnfs_block_dev *dev)
+{
+	u32 i;
+
+	if (dev->nr_children) {
+		for (i = 0; i < dev->nr_children; i++)
+			bl_unregister_dev(&dev->children[i]);
+		return;
+	}
+
+	if (dev->type == PNFS_BLOCK_VOLUME_SCSI &&
+		test_and_clear_bit(PNFS_BDEV_REGISTERED, &dev->flags))
+		bl_unregister_scsi(dev);
+}
+
+bool bl_register_dev(struct pnfs_block_dev *dev)
+{
+	u32 i;
+
+	if (dev->nr_children) {
+		for (i = 0; i < dev->nr_children; i++) {
+			if (!bl_register_dev(&dev->children[i])) {
+				while (i > 0)
+					bl_unregister_dev(&dev->children[--i]);
+				return false;
+			}
+		}
+		return true;
+	}
+
+	if (dev->type == PNFS_BLOCK_VOLUME_SCSI)
+		return bl_register_scsi(dev);
+	return true;
+}
+
+static void
+bl_free_device(struct pnfs_block_dev *dev)
+{
+	bl_unregister_dev(dev);
+
+	if (dev->children) {
+		int i;
+
+		for (i = 0; i < dev->nr_children; i++)
+			bl_free_device(&dev->children[i]);
+		kfree(dev->children);
+		dev->children = NULL;
+		dev->nr_children = 0;
+	} else if (dev->bdev_file) {
+		fput(dev->bdev_file);
+		dev->bdev_file = NULL;
+	}
+}
+
+void
+bl_free_deviceid_node(struct nfs4_deviceid_node *d)
+{
+	struct pnfs_block_dev *dev =
+		container_of(d, struct pnfs_block_dev, node);
+
+	bl_free_device(dev);
+	kfree_rcu(dev, node.rcu);
+}
+
+static int
+nfs4_block_decode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b)
+{
+	__be32 *p;
+	int i;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (!p)
+		return -EIO;
+	b->type = be32_to_cpup(p++);
+
+	switch (b->type) {
+	case PNFS_BLOCK_VOLUME_SIMPLE:
+		p = xdr_inline_decode(xdr, 4);
+		if (!p)
+			return -EIO;
+		b->simple.nr_sigs = be32_to_cpup(p++);
+		if (!b->simple.nr_sigs || b->simple.nr_sigs > PNFS_BLOCK_MAX_UUIDS) {
+			dprintk("Bad signature count: %d\n", b->simple.nr_sigs);
+			return -EIO;
+		}
+
+		b->simple.len = 4 + 4;
+		for (i = 0; i < b->simple.nr_sigs; i++) {
+			p = xdr_inline_decode(xdr, 8 + 4);
+			if (!p)
+				return -EIO;
+			p = xdr_decode_hyper(p, &b->simple.sigs[i].offset);
+			b->simple.sigs[i].sig_len = be32_to_cpup(p++);
+			if (b->simple.sigs[i].sig_len > PNFS_BLOCK_UUID_LEN) {
+				pr_info("signature too long: %d\n",
+					b->simple.sigs[i].sig_len);
+				return -EIO;
+			}
+
+			p = xdr_inline_decode(xdr, b->simple.sigs[i].sig_len);
+			if (!p)
+				return -EIO;
+			memcpy(&b->simple.sigs[i].sig, p,
+				b->simple.sigs[i].sig_len);
+
+			b->simple.len += 8 + 4 + \
+				(XDR_QUADLEN(b->simple.sigs[i].sig_len) << 2);
+		}
+		break;
+	case PNFS_BLOCK_VOLUME_SLICE:
+		p = xdr_inline_decode(xdr, 8 + 8 + 4);
+		if (!p)
+			return -EIO;
+		p = xdr_decode_hyper(p, &b->slice.start);
+		p = xdr_decode_hyper(p, &b->slice.len);
+		b->slice.volume = be32_to_cpup(p++);
+		break;
+	case PNFS_BLOCK_VOLUME_CONCAT:
+		p = xdr_inline_decode(xdr, 4);
+		if (!p)
+			return -EIO;
+
+		b->concat.volumes_count = be32_to_cpup(p++);
+		if (b->concat.volumes_count > PNFS_BLOCK_MAX_DEVICES) {
+			dprintk("Too many volumes: %d\n", b->concat.volumes_count);
+			return -EIO;
+		}
+
+		p = xdr_inline_decode(xdr, b->concat.volumes_count * 4);
+		if (!p)
+			return -EIO;
+		for (i = 0; i < b->concat.volumes_count; i++)
+			b->concat.volumes[i] = be32_to_cpup(p++);
+		break;
+	case PNFS_BLOCK_VOLUME_STRIPE:
+		p = xdr_inline_decode(xdr, 8 + 4);
+		if (!p)
+			return -EIO;
+
+		p = xdr_decode_hyper(p, &b->stripe.chunk_size);
+		b->stripe.volumes_count = be32_to_cpup(p++);
+		if (b->stripe.volumes_count > PNFS_BLOCK_MAX_DEVICES) {
+			dprintk("Too many volumes: %d\n", b->stripe.volumes_count);
+			return -EIO;
+		}
+
+		p = xdr_inline_decode(xdr, b->stripe.volumes_count * 4);
+		if (!p)
+			return -EIO;
+		for (i = 0; i < b->stripe.volumes_count; i++)
+			b->stripe.volumes[i] = be32_to_cpup(p++);
+		break;
+	case PNFS_BLOCK_VOLUME_SCSI:
+		p = xdr_inline_decode(xdr, 4 + 4 + 4);
+		if (!p)
+			return -EIO;
+		b->scsi.code_set = be32_to_cpup(p++);
+		b->scsi.designator_type = be32_to_cpup(p++);
+		b->scsi.designator_len = be32_to_cpup(p++);
+		p = xdr_inline_decode(xdr, b->scsi.designator_len);
+		if (!p)
+			return -EIO;
+		if (b->scsi.designator_len > 256)
+			return -EIO;
+		memcpy(&b->scsi.designator, p, b->scsi.designator_len);
+		p = xdr_inline_decode(xdr, 8);
+		if (!p)
+			return -EIO;
+		p = xdr_decode_hyper(p, &b->scsi.pr_key);
+		break;
+	default:
+		dprintk("unknown volume type!\n");
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static bool bl_map_simple(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	map->start = dev->start;
+	map->len = dev->len;
+	map->disk_offset = dev->disk_offset;
+	map->bdev = file_bdev(dev->bdev_file);
+	return true;
+}
+
+static bool bl_map_concat(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	int i;
+
+	for (i = 0; i < dev->nr_children; i++) {
+		struct pnfs_block_dev *child = &dev->children[i];
+
+		if (child->start > offset ||
+		    child->start + child->len <= offset)
+			continue;
+
+		child->map(child, offset - child->start, map);
+		return true;
+	}
+
+	dprintk("%s: ran off loop!\n", __func__);
+	return false;
+}
+
+static bool bl_map_stripe(struct pnfs_block_dev *dev, u64 offset,
+		struct pnfs_block_dev_map *map)
+{
+	struct pnfs_block_dev *child;
+	u64 chunk;
+	u32 chunk_idx;
+	u64 disk_chunk;
+	u64 disk_offset;
+
+	chunk = div_u64(offset, dev->chunk_size);
+	disk_chunk = div_u64_rem(chunk, dev->nr_children, &chunk_idx);
+
+	if (chunk_idx >= dev->nr_children) {
+		dprintk("%s: invalid chunk idx %d (%lld/%lld)\n",
+			__func__, chunk_idx, offset, dev->chunk_size);
+		/* error, should not happen */
+		return false;
+	}
+
+	/* truncate offset to the beginning of the stripe */
+	offset = chunk * dev->chunk_size;
+
+	/* disk offset of the stripe */
+	disk_offset = disk_chunk * dev->chunk_size;
+
+	child = &dev->children[chunk_idx];
+	child->map(child, disk_offset, map);
+
+	map->start += offset;
+	map->disk_offset += disk_offset;
+	map->len = dev->chunk_size;
+	return true;
+}
+
+static int
+bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask);
+
+
+static int
+bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	struct file *bdev_file;
+	dev_t dev;
+
+	dev = bl_resolve_deviceid(server, v, gfp_mask);
+	if (!dev)
+		return -EIO;
+
+	bdev_file = bdev_file_open_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_WRITE,
+				       NULL, NULL);
+	if (IS_ERR(bdev_file)) {
+		printk(KERN_WARNING "pNFS: failed to open device %d:%d (%ld)\n",
+			MAJOR(dev), MINOR(dev), PTR_ERR(bdev_file));
+		return PTR_ERR(bdev_file);
+	}
+	d->bdev_file = bdev_file;
+	d->len = bdev_nr_bytes(file_bdev(bdev_file));
+	d->map = bl_map_simple;
+
+	printk(KERN_INFO "pNFS: using block device %s\n",
+		file_bdev(bdev_file)->bd_disk->disk_name);
+	return 0;
+}
+
+static bool
+bl_validate_designator(struct pnfs_block_volume *v)
+{
+	switch (v->scsi.designator_type) {
+	case PS_DESIGNATOR_EUI64:
+		if (v->scsi.code_set != PS_CODE_SET_BINARY)
+			return false;
+
+		if (v->scsi.designator_len != 8 &&
+		    v->scsi.designator_len != 10 &&
+		    v->scsi.designator_len != 16)
+			return false;
+
+		return true;
+	case PS_DESIGNATOR_NAA:
+		if (v->scsi.code_set != PS_CODE_SET_BINARY)
+			return false;
+
+		if (v->scsi.designator_len != 8 &&
+		    v->scsi.designator_len != 16)
+			return false;
+
+		return true;
+	case PS_DESIGNATOR_T10:
+	case PS_DESIGNATOR_NAME:
+		pr_err("pNFS: unsupported designator "
+			"(code set %d, type %d, len %d.\n",
+			v->scsi.code_set,
+			v->scsi.designator_type,
+			v->scsi.designator_len);
+		return false;
+	default:
+		pr_err("pNFS: invalid designator "
+			"(code set %d, type %d, len %d.\n",
+			v->scsi.code_set,
+			v->scsi.designator_type,
+			v->scsi.designator_len);
+		return false;
+	}
+}
+
+static struct file *
+bl_open_path(struct pnfs_block_volume *v, const char *prefix)
+{
+	struct file *bdev_file;
+	const char *devname __free(kfree) = NULL;
+
+	devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/%s%*phN",
+			prefix, v->scsi.designator_len, v->scsi.designator);
+	if (!devname)
+		return ERR_PTR(-ENOMEM);
+
+	if (tsk_is_kthread(current)) {
+		scoped_with_init_fs()
+			bdev_file = bdev_file_open_by_path(devname,
+					BLK_OPEN_READ | BLK_OPEN_WRITE,
+					NULL, NULL);
+	} else {
+		bdev_file = bdev_file_open_by_path(devname,
+				BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL);
+	}
+	if (IS_ERR(bdev_file)) {
+		dprintk("failed to open device %s (%ld)\n",
+			devname, PTR_ERR(bdev_file));
+	} else {
+		pr_info("pNFS: using block device %s\n",
+			file_bdev(bdev_file)->bd_disk->disk_name);
+	}
+
+	return bdev_file;
+}
+
+static int
+bl_parse_scsi(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	struct block_device *bdev;
+	const struct pr_ops *ops;
+	struct file *bdev_file;
+	int error;
+
+	if (!bl_validate_designator(v))
+		return -EINVAL;
+
+	/*
+	 * Try to open the RH/Fedora specific dm-mpath udev path first, as the
+	 * wwn- links will only point to the first discovered SCSI device there.
+	 * On other distributions like Debian, the default SCSI by-id path will
+	 * point to the dm-multipath device if one exists.
+	 */
+	bdev_file = bl_open_path(v, "dm-uuid-mpath-0x");
+	if (IS_ERR(bdev_file))
+		bdev_file = bl_open_path(v, "wwn-0x");
+	if (IS_ERR(bdev_file))
+		bdev_file = bl_open_path(v, "nvme-eui.");
+	if (IS_ERR(bdev_file)) {
+		pr_warn("pNFS: no device found for volume %*phN\n",
+			v->scsi.designator_len, v->scsi.designator);
+		return PTR_ERR(bdev_file);
+	}
+	d->bdev_file = bdev_file;
+	bdev = file_bdev(bdev_file);
+
+	d->len = bdev_nr_bytes(bdev);
+	d->map = bl_map_simple;
+	d->pr_key = v->scsi.pr_key;
+
+	if (d->len == 0) {
+		error = -ENODEV;
+		goto out_blkdev_put;
+	}
+
+	ops = bdev->bd_disk->fops->pr_ops;
+	if (!ops) {
+		pr_err("pNFS: block device %s does not support reservations.",
+				bdev->bd_disk->disk_name);
+		error = -EINVAL;
+		goto out_blkdev_put;
+	}
+
+	return 0;
+
+out_blkdev_put:
+	fput(d->bdev_file);
+	d->bdev_file = NULL;
+	return error;
+}
+
+static int
+bl_parse_slice(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	int ret;
+
+	ret = bl_parse_deviceid(server, d, volumes, v->slice.volume, gfp_mask);
+	if (ret)
+		return ret;
+
+	d->disk_offset = v->slice.start;
+	d->len = v->slice.len;
+	return 0;
+}
+
+static int
+bl_parse_concat(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	u64 len = 0;
+	int ret, i;
+
+	d->children = kzalloc_objs(struct pnfs_block_dev,
+				   v->concat.volumes_count, gfp_mask);
+	if (!d->children)
+		return -ENOMEM;
+
+	for (i = 0; i < v->concat.volumes_count; i++) {
+		ret = bl_parse_deviceid(server, &d->children[i],
+				volumes, v->concat.volumes[i], gfp_mask);
+		if (ret) {
+			bl_free_device(&d->children[i]);
+			bl_free_device(d);
+			return ret;
+		}
+
+		d->nr_children++;
+		d->children[i].start += len;
+		len += d->children[i].len;
+	}
+
+	d->len = len;
+	d->map = bl_map_concat;
+	return 0;
+}
+
+static int
+bl_parse_stripe(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	struct pnfs_block_volume *v = &volumes[idx];
+	u64 len = 0;
+	int ret, i;
+
+	d->children = kzalloc_objs(struct pnfs_block_dev,
+				   v->stripe.volumes_count, gfp_mask);
+	if (!d->children)
+		return -ENOMEM;
+
+	for (i = 0; i < v->stripe.volumes_count; i++) {
+		ret = bl_parse_deviceid(server, &d->children[i],
+				volumes, v->stripe.volumes[i], gfp_mask);
+		if (ret) {
+			bl_free_device(&d->children[i]);
+			bl_free_device(d);
+			return ret;
+		}
+
+		d->nr_children++;
+		len += d->children[i].len;
+	}
+
+	d->len = len;
+	d->chunk_size = v->stripe.chunk_size;
+	d->map = bl_map_stripe;
+	return 0;
+}
+
+static int
+bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
+		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
+{
+	d->type = volumes[idx].type;
+
+	switch (d->type) {
+	case PNFS_BLOCK_VOLUME_SIMPLE:
+		return bl_parse_simple(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_SLICE:
+		return bl_parse_slice(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_CONCAT:
+		return bl_parse_concat(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_STRIPE:
+		return bl_parse_stripe(server, d, volumes, idx, gfp_mask);
+	case PNFS_BLOCK_VOLUME_SCSI:
+		return bl_parse_scsi(server, d, volumes, idx, gfp_mask);
+	default:
+		dprintk("unsupported volume type: %d\n", d->type);
+		return -EIO;
+	}
+}
+
+struct nfs4_deviceid_node *
+bl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
+		gfp_t gfp_mask)
+{
+	struct nfs4_deviceid_node *node = NULL;
+	struct pnfs_block_volume *volumes;
+	struct pnfs_block_dev *top;
+	struct xdr_stream xdr;
+	struct xdr_buf buf;
+	struct folio *scratch;
+	int nr_volumes, ret, i;
+	__be32 *p;
+
+	scratch = folio_alloc(gfp_mask, 0);
+	if (!scratch)
+		goto out;
+
+	xdr_init_decode_pages(&xdr, &buf, pdev->pages, pdev->pglen);
+	xdr_set_scratch_folio(&xdr, scratch);
+
+	p = xdr_inline_decode(&xdr, sizeof(__be32));
+	if (!p)
+		goto out_free_scratch;
+	nr_volumes = be32_to_cpup(p++);
+
+	volumes = kzalloc_objs(struct pnfs_block_volume, nr_volumes, gfp_mask);
+	if (!volumes)
+		goto out_free_scratch;
+
+	for (i = 0; i < nr_volumes; i++) {
+		ret = nfs4_block_decode_volume(&xdr, &volumes[i]);
+		if (ret < 0)
+			goto out_free_volumes;
+	}
+
+	top = kzalloc_obj(*top, gfp_mask);
+	if (!top)
+		goto out_free_volumes;
+
+	ret = bl_parse_deviceid(server, top, volumes, nr_volumes - 1, gfp_mask);
+
+	node = &top->node;
+	nfs4_init_deviceid_node(node, server, &pdev->dev_id);
+	if (ret)
+		nfs4_mark_deviceid_unavailable(node);
+
+out_free_volumes:
+	kfree(volumes);
+out_free_scratch:
+	folio_put(scratch);
+out:
+	return node;
+}
diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
new file mode 100644
index 000000000..7393118c4
--- /dev/null
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -0,0 +1,726 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2014-2016 Christoph Hellwig.
+ */
+
+#include <linux/vmalloc.h>
+
+#include "blocklayout.h"
+#include "../nfs4trace.h"
+
+#define NFSDBG_FACILITY		NFSDBG_PNFS_LD
+
+static inline struct pnfs_block_extent *
+ext_node(struct rb_node *node)
+{
+	return rb_entry(node, struct pnfs_block_extent, be_node);
+}
+
+static struct pnfs_block_extent *
+ext_tree_first(struct rb_root *root)
+{
+	struct rb_node *node = rb_first(root);
+	return node ? ext_node(node) : NULL;
+}
+
+static struct pnfs_block_extent *
+ext_tree_prev(struct pnfs_block_extent *be)
+{
+	struct rb_node *node = rb_prev(&be->be_node);
+	return node ? ext_node(node) : NULL;
+}
+
+static struct pnfs_block_extent *
+ext_tree_next(struct pnfs_block_extent *be)
+{
+	struct rb_node *node = rb_next(&be->be_node);
+	return node ? ext_node(node) : NULL;
+}
+
+static inline sector_t
+ext_f_end(struct pnfs_block_extent *be)
+{
+	return be->be_f_offset + be->be_length;
+}
+
+static struct pnfs_block_extent *
+__ext_tree_search(struct rb_root *root, sector_t start)
+{
+	struct rb_node *node = root->rb_node;
+	struct pnfs_block_extent *be = NULL;
+
+	while (node) {
+		be = ext_node(node);
+		if (start < be->be_f_offset)
+			node = node->rb_left;
+		else if (start >= ext_f_end(be))
+			node = node->rb_right;
+		else
+			return be;
+	}
+
+	if (be) {
+		if (start < be->be_f_offset)
+			return be;
+
+		if (start >= ext_f_end(be))
+			return ext_tree_next(be);
+	}
+
+	return NULL;
+}
+
+static bool
+ext_can_merge(struct pnfs_block_extent *be1, struct pnfs_block_extent *be2)
+{
+	if (be1->be_state != be2->be_state)
+		return false;
+	if (be1->be_device != be2->be_device)
+		return false;
+
+	if (be1->be_f_offset + be1->be_length != be2->be_f_offset)
+		return false;
+
+	if (be1->be_state != PNFS_BLOCK_NONE_DATA &&
+	    (be1->be_v_offset + be1->be_length != be2->be_v_offset))
+		return false;
+
+	if (be1->be_state == PNFS_BLOCK_INVALID_DATA &&
+	    be1->be_tag != be2->be_tag)
+		return false;
+
+	return true;
+}
+
+static struct pnfs_block_extent *
+ext_try_to_merge_left(struct rb_root *root, struct pnfs_block_extent *be)
+{
+	struct pnfs_block_extent *left = ext_tree_prev(be);
+
+	if (left && ext_can_merge(left, be)) {
+		left->be_length += be->be_length;
+		rb_erase(&be->be_node, root);
+		nfs4_put_deviceid_node(be->be_device);
+		kfree(be);
+		return left;
+	}
+
+	return be;
+}
+
+static struct pnfs_block_extent *
+ext_try_to_merge_right(struct rb_root *root, struct pnfs_block_extent *be)
+{
+	struct pnfs_block_extent *right = ext_tree_next(be);
+
+	if (right && ext_can_merge(be, right)) {
+		be->be_length += right->be_length;
+		rb_erase(&right->be_node, root);
+		nfs4_put_deviceid_node(right->be_device);
+		kfree(right);
+	}
+
+	return be;
+}
+
+static void __ext_put_deviceids(struct list_head *head)
+{
+	struct pnfs_block_extent *be, *tmp;
+
+	list_for_each_entry_safe(be, tmp, head, be_list) {
+		nfs4_put_deviceid_node(be->be_device);
+		kfree(be);
+	}
+}
+
+static void
+__ext_tree_insert(struct rb_root *root,
+		struct pnfs_block_extent *new, bool merge_ok)
+{
+	struct rb_node **p = &root->rb_node, *parent = NULL;
+	struct pnfs_block_extent *be;
+
+	while (*p) {
+		parent = *p;
+		be = ext_node(parent);
+
+		if (new->be_f_offset < be->be_f_offset) {
+			if (merge_ok && ext_can_merge(new, be)) {
+				be->be_f_offset = new->be_f_offset;
+				if (be->be_state != PNFS_BLOCK_NONE_DATA)
+					be->be_v_offset = new->be_v_offset;
+				be->be_length += new->be_length;
+				be = ext_try_to_merge_left(root, be);
+				goto free_new;
+			}
+			p = &(*p)->rb_left;
+		} else if (new->be_f_offset >= ext_f_end(be)) {
+			if (merge_ok && ext_can_merge(be, new)) {
+				be->be_length += new->be_length;
+				be = ext_try_to_merge_right(root, be);
+				goto free_new;
+			}
+			p = &(*p)->rb_right;
+		} else {
+			BUG();
+		}
+	}
+
+	rb_link_node(&new->be_node, parent, p);
+	rb_insert_color(&new->be_node, root);
+	return;
+free_new:
+	nfs4_put_deviceid_node(new->be_device);
+	kfree(new);
+}
+
+static int
+__ext_tree_remove(struct rb_root *root,
+		sector_t start, sector_t end, struct list_head *tmp)
+{
+	struct pnfs_block_extent *be;
+	sector_t len1 = 0, len2 = 0;
+	sector_t orig_v_offset;
+	sector_t orig_len;
+
+	be = __ext_tree_search(root, start);
+	if (!be)
+		return 0;
+	if (be->be_f_offset >= end)
+		return 0;
+
+	orig_v_offset = be->be_v_offset;
+	orig_len = be->be_length;
+
+	if (start > be->be_f_offset)
+		len1 = start - be->be_f_offset;
+	if (ext_f_end(be) > end)
+		len2 = ext_f_end(be) - end;
+
+	if (len2 > 0) {
+		if (len1 > 0) {
+			struct pnfs_block_extent *new;
+
+			new = kzalloc_obj(*new, GFP_ATOMIC);
+			if (!new)
+				return -ENOMEM;
+
+			be->be_length = len1;
+
+			new->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA) {
+				new->be_v_offset =
+					orig_v_offset + orig_len - len2;
+			}
+			new->be_length = len2;
+			new->be_state = be->be_state;
+			new->be_tag = be->be_tag;
+			new->be_device = nfs4_get_deviceid(be->be_device);
+
+			__ext_tree_insert(root, new, true);
+		} else {
+			be->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA) {
+				be->be_v_offset =
+					orig_v_offset + orig_len - len2;
+			}
+			be->be_length = len2;
+		}
+	} else {
+		if (len1 > 0) {
+			be->be_length = len1;
+			be = ext_tree_next(be);
+		}
+
+		while (be && ext_f_end(be) <= end) {
+			struct pnfs_block_extent *next = ext_tree_next(be);
+
+			rb_erase(&be->be_node, root);
+			list_add_tail(&be->be_list, tmp);
+			be = next;
+		}
+
+		if (be && be->be_f_offset < end) {
+			len1 = ext_f_end(be) - end;
+			be->be_f_offset = end;
+			if (be->be_state != PNFS_BLOCK_NONE_DATA)
+				be->be_v_offset += be->be_length - len1;
+			be->be_length = len1;
+		}
+	}
+
+	return 0;
+}
+
+int
+ext_tree_insert(struct pnfs_block_layout *bl, struct pnfs_block_extent *new)
+{
+	struct pnfs_block_extent *be;
+	struct rb_root *root;
+	int err = 0;
+
+	switch (new->be_state) {
+	case PNFS_BLOCK_READWRITE_DATA:
+	case PNFS_BLOCK_INVALID_DATA:
+		root = &bl->bl_ext_rw;
+		break;
+	case PNFS_BLOCK_READ_DATA:
+	case PNFS_BLOCK_NONE_DATA:
+		root = &bl->bl_ext_ro;
+		break;
+	default:
+		dprintk("invalid extent type\n");
+		return -EINVAL;
+	}
+
+	spin_lock(&bl->bl_ext_lock);
+retry:
+	be = __ext_tree_search(root, new->be_f_offset);
+	if (!be || be->be_f_offset >= ext_f_end(new)) {
+		__ext_tree_insert(root, new, true);
+	} else if (new->be_f_offset >= be->be_f_offset) {
+		if (ext_f_end(new) <= ext_f_end(be)) {
+			nfs4_put_deviceid_node(new->be_device);
+			kfree(new);
+		} else {
+			sector_t new_len = ext_f_end(new) - ext_f_end(be);
+			sector_t diff = new->be_length - new_len;
+
+			new->be_f_offset += diff;
+			new->be_v_offset += diff;
+			new->be_length = new_len;
+			goto retry;
+		}
+	} else if (ext_f_end(new) <= ext_f_end(be)) {
+		new->be_length = be->be_f_offset - new->be_f_offset;
+		__ext_tree_insert(root, new, true);
+	} else {
+		struct pnfs_block_extent *split;
+		sector_t new_len = ext_f_end(new) - ext_f_end(be);
+		sector_t diff = new->be_length - new_len;
+
+		split = kmemdup(new, sizeof(*new), GFP_ATOMIC);
+		if (!split) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		split->be_length = be->be_f_offset - split->be_f_offset;
+		split->be_device = nfs4_get_deviceid(new->be_device);
+		__ext_tree_insert(root, split, true);
+
+		new->be_f_offset += diff;
+		new->be_v_offset += diff;
+		new->be_length = new_len;
+		goto retry;
+	}
+out:
+	spin_unlock(&bl->bl_ext_lock);
+	return err;
+}
+
+static bool
+__ext_tree_lookup(struct rb_root *root, sector_t isect,
+		struct pnfs_block_extent *ret)
+{
+	struct rb_node *node;
+	struct pnfs_block_extent *be;
+
+	node = root->rb_node;
+	while (node) {
+		be = ext_node(node);
+		if (isect < be->be_f_offset)
+			node = node->rb_left;
+		else if (isect >= ext_f_end(be))
+			node = node->rb_right;
+		else {
+			*ret = *be;
+			return true;
+		}
+	}
+
+	return false;
+}
+
+bool
+ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect,
+	    struct pnfs_block_extent *ret, bool rw)
+{
+	bool found = false;
+
+	spin_lock(&bl->bl_ext_lock);
+	if (!rw)
+		found = __ext_tree_lookup(&bl->bl_ext_ro, isect, ret);
+	if (!found)
+		found = __ext_tree_lookup(&bl->bl_ext_rw, isect, ret);
+	spin_unlock(&bl->bl_ext_lock);
+
+	return found;
+}
+
+int ext_tree_remove(struct pnfs_block_layout *bl, bool rw,
+		sector_t start, sector_t end)
+{
+	int err, err2;
+	LIST_HEAD(tmp);
+
+	spin_lock(&bl->bl_ext_lock);
+	err = __ext_tree_remove(&bl->bl_ext_ro, start, end, &tmp);
+	if (rw) {
+		err2 = __ext_tree_remove(&bl->bl_ext_rw, start, end, &tmp);
+		if (!err)
+			err = err2;
+	}
+	spin_unlock(&bl->bl_ext_lock);
+
+	__ext_put_deviceids(&tmp);
+	return err;
+}
+
+static int
+ext_tree_split(struct rb_root *root, struct pnfs_block_extent *be,
+		sector_t split)
+{
+	struct pnfs_block_extent *new;
+	sector_t orig_len = be->be_length;
+
+	new = kzalloc_obj(*new, GFP_ATOMIC);
+	if (!new)
+		return -ENOMEM;
+
+	be->be_length = split - be->be_f_offset;
+
+	new->be_f_offset = split;
+	if (be->be_state != PNFS_BLOCK_NONE_DATA)
+		new->be_v_offset = be->be_v_offset + be->be_length;
+	new->be_length = orig_len - be->be_length;
+	new->be_state = be->be_state;
+	new->be_tag = be->be_tag;
+	new->be_device = nfs4_get_deviceid(be->be_device);
+
+	__ext_tree_insert(root, new, false);
+	return 0;
+}
+
+int
+ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
+		sector_t len, u64 lwb)
+{
+	struct rb_root *root = &bl->bl_ext_rw;
+	sector_t end = start + len;
+	struct pnfs_block_extent *be;
+	int err = 0;
+	LIST_HEAD(tmp);
+
+	spin_lock(&bl->bl_ext_lock);
+	/*
+	 * First remove all COW extents or holes from written to range.
+	 */
+	err = __ext_tree_remove(&bl->bl_ext_ro, start, end, &tmp);
+	if (err)
+		goto out;
+
+	/*
+	 * Then mark all invalid extents in the range as written to.
+	 */
+	for (be = __ext_tree_search(root, start); be; be = ext_tree_next(be)) {
+		if (be->be_f_offset >= end)
+			break;
+
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA || be->be_tag)
+			continue;
+
+		if (be->be_f_offset < start) {
+			struct pnfs_block_extent *left = ext_tree_prev(be);
+
+			if (left && ext_can_merge(left, be)) {
+				sector_t diff = start - be->be_f_offset;
+
+				left->be_length += diff;
+
+				be->be_f_offset += diff;
+				be->be_v_offset += diff;
+				be->be_length -= diff;
+			} else {
+				err = ext_tree_split(root, be, start);
+				if (err)
+					goto out;
+			}
+		}
+
+		if (ext_f_end(be) > end) {
+			struct pnfs_block_extent *right = ext_tree_next(be);
+
+			if (right && ext_can_merge(be, right)) {
+				sector_t diff = end - be->be_f_offset;
+
+				be->be_length -= diff;
+
+				right->be_f_offset -= diff;
+				right->be_v_offset -= diff;
+				right->be_length += diff;
+			} else {
+				err = ext_tree_split(root, be, end);
+				if (err)
+					goto out;
+			}
+		}
+
+		if (be->be_f_offset >= start && ext_f_end(be) <= end) {
+			be->be_tag = EXTENT_WRITTEN;
+			be = ext_try_to_merge_left(root, be);
+			be = ext_try_to_merge_right(root, be);
+		}
+	}
+out:
+	if (bl->bl_lwb < lwb)
+		bl->bl_lwb = lwb;
+	spin_unlock(&bl->bl_ext_lock);
+
+	__ext_put_deviceids(&tmp);
+	return err;
+}
+
+static size_t ext_tree_layoutupdate_size(struct pnfs_block_layout *bl, size_t count)
+{
+	if (bl->bl_scsi_layout)
+		return sizeof(__be32) + PNFS_SCSI_RANGE_SIZE * count;
+	else
+		return sizeof(__be32) + PNFS_BLOCK_EXTENT_SIZE * count;
+}
+
+static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
+		size_t buffer_size)
+{
+	if (arg->layoutupdate_pages != &arg->layoutupdate_page) {
+		int nr_pages = DIV_ROUND_UP(buffer_size, PAGE_SIZE), i;
+
+		for (i = 0; i < nr_pages; i++)
+			put_page(arg->layoutupdate_pages[i]);
+		vfree(arg->start_p);
+		kfree(arg->layoutupdate_pages);
+	} else {
+		put_page(arg->layoutupdate_page);
+	}
+}
+
+static __be32 *encode_block_extent(struct pnfs_block_extent *be, __be32 *p)
+{
+	p = xdr_encode_opaque_fixed(p, be->be_device->deviceid.data,
+			NFS4_DEVICEID4_SIZE);
+	p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
+	p = xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
+	p = xdr_encode_hyper(p, 0LL);
+	*p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA);
+	return p;
+}
+
+static __be32 *encode_scsi_range(struct pnfs_block_extent *be, __be32 *p)
+{
+	p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT);
+	return xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT);
+}
+
+/**
+ * ext_tree_try_encode_commit - try to encode all extents into the buffer
+ * @bl: pointer to the layout
+ * @p: pointer to the output buffer
+ * @buffer_size: size of the output buffer
+ * @count: output pointer to the number of encoded extents
+ * @lastbyte: output pointer to the last written byte
+ *
+ * Return values:
+ *   %0: Success, all required extents encoded, outputs are valid
+ *   %-ENOSPC: Buffer too small, nothing encoded, outputs are invalid
+ */
+static int
+ext_tree_try_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
+		size_t buffer_size, size_t *count, __u64 *lastbyte)
+{
+	struct pnfs_block_extent *be;
+
+	spin_lock(&bl->bl_ext_lock);
+	for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_WRITTEN)
+			continue;
+
+		(*count)++;
+		if (ext_tree_layoutupdate_size(bl, *count) > buffer_size) {
+			spin_unlock(&bl->bl_ext_lock);
+			return -ENOSPC;
+		}
+	}
+	for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_WRITTEN)
+			continue;
+
+		if (bl->bl_scsi_layout)
+			p = encode_scsi_range(be, p);
+		else
+			p = encode_block_extent(be, p);
+		be->be_tag = EXTENT_COMMITTING;
+	}
+	*lastbyte = (bl->bl_lwb != 0) ? bl->bl_lwb - 1 : U64_MAX;
+	bl->bl_lwb = 0;
+	spin_unlock(&bl->bl_ext_lock);
+
+	return 0;
+}
+
+/**
+ * ext_tree_encode_commit - encode as much as possible extents into the buffer
+ * @bl: pointer to the layout
+ * @p: pointer to the output buffer
+ * @buffer_size: size of the output buffer
+ * @count: output pointer to the number of encoded extents
+ * @lastbyte: output pointer to the last written byte
+ *
+ * Return values:
+ *   %0: Success, all required extents encoded, outputs are valid
+ *   %-ENOSPC: Buffer too small, some extents are encoded, outputs are valid
+ */
+static int
+ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
+		size_t buffer_size, size_t *count, __u64 *lastbyte)
+{
+	struct pnfs_block_extent *be, *be_prev;
+	int ret = 0;
+
+	spin_lock(&bl->bl_ext_lock);
+	for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_WRITTEN)
+			continue;
+
+		(*count)++;
+		if (ext_tree_layoutupdate_size(bl, *count) > buffer_size) {
+			(*count)--;
+			ret = -ENOSPC;
+			break;
+		}
+
+		if (bl->bl_scsi_layout)
+			p = encode_scsi_range(be, p);
+		else
+			p = encode_block_extent(be, p);
+		be->be_tag = EXTENT_COMMITTING;
+		be_prev = be;
+	}
+	if (!ret) {
+		*lastbyte = (bl->bl_lwb != 0) ? bl->bl_lwb - 1 : U64_MAX;
+		bl->bl_lwb = 0;
+	} else {
+		*lastbyte = be_prev->be_f_offset + be_prev->be_length;
+		*lastbyte <<= SECTOR_SHIFT;
+		*lastbyte -= 1;
+	}
+	spin_unlock(&bl->bl_ext_lock);
+
+	return ret;
+}
+
+/**
+ * ext_tree_prepare_commit - encode extents that need to be committed
+ * @arg: layout commit data
+ *
+ * Return values:
+ *   %0: Success, all required extents are encoded
+ *   %-ENOSPC: Some extents are encoded, but not all, due to RPC size limit
+ *   %-ENOMEM: Out of memory, extents not encoded
+ */
+int
+ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout);
+	size_t count = 0, buffer_size = PAGE_SIZE;
+	__be32 *start_p;
+	int ret;
+
+	arg->layoutupdate_page = alloc_page(GFP_NOFS);
+	if (!arg->layoutupdate_page)
+		return -ENOMEM;
+	start_p = page_address(arg->layoutupdate_page);
+	arg->layoutupdate_pages = &arg->layoutupdate_page;
+
+	ret = ext_tree_try_encode_commit(bl, start_p + 1, buffer_size,
+			&count, &arg->lastbytewritten);
+	if (unlikely(ret)) {
+		ext_tree_free_commitdata(arg, buffer_size);
+
+		buffer_size = NFS_SERVER(arg->inode)->wsize;
+		count = 0;
+
+		arg->layoutupdate_pages =
+			kzalloc_objs(struct page *,
+				     DIV_ROUND_UP(buffer_size, PAGE_SIZE),
+				     GFP_NOFS);
+		if (!arg->layoutupdate_pages)
+			return -ENOMEM;
+
+		start_p = __vmalloc(buffer_size, GFP_NOFS);
+		if (!start_p) {
+			kfree(arg->layoutupdate_pages);
+			return -ENOMEM;
+		}
+
+		ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size,
+				&count, &arg->lastbytewritten);
+	}
+
+	*start_p = cpu_to_be32(count);
+	arg->layoutupdate_len = ext_tree_layoutupdate_size(bl, count);
+
+	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
+		void *p = start_p, *end = p + arg->layoutupdate_len;
+		struct page *page = NULL;
+		int i = 0;
+
+		arg->start_p = start_p;
+		for ( ; p < end; p += PAGE_SIZE) {
+			page = vmalloc_to_page(p);
+			arg->layoutupdate_pages[i++] = page;
+			get_page(page);
+		}
+	}
+
+	trace_bl_ext_tree_prepare_commit(ret, count,
+			arg->lastbytewritten, !!ret);
+	return ret;
+}
+
+void
+ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status)
+{
+	struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout);
+	struct rb_root *root = &bl->bl_ext_rw;
+	struct pnfs_block_extent *be;
+
+	dprintk("%s status %d\n", __func__, status);
+
+	ext_tree_free_commitdata(arg, arg->layoutupdate_len);
+
+	spin_lock(&bl->bl_ext_lock);
+	for (be = ext_tree_first(root); be; be = ext_tree_next(be)) {
+		if (be->be_state != PNFS_BLOCK_INVALID_DATA ||
+		    be->be_tag != EXTENT_COMMITTING)
+			continue;
+
+		if (status) {
+			/*
+			 * Mark as written and try again.
+			 *
+			 * XXX: some real error handling here wouldn't hurt..
+			 */
+			be->be_tag = EXTENT_WRITTEN;
+		} else {
+			be->be_state = PNFS_BLOCK_READWRITE_DATA;
+			be->be_tag = 0;
+		}
+
+		be = ext_try_to_merge_left(root, be);
+		be = ext_try_to_merge_right(root, be);
+	}
+	spin_unlock(&bl->bl_ext_lock);
+}
diff --git a/fs/nfs/blocklayout/rpc_pipefs.c b/fs/nfs/blocklayout/rpc_pipefs.c
new file mode 100644
index 000000000..d526f5ba7
--- /dev/null
+++ b/fs/nfs/blocklayout/rpc_pipefs.c
@@ -0,0 +1,270 @@
+/*
+ *  Copyright (c) 2006,2007 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Andy Adamson <andros@citi.umich.edu>
+ *  Fred Isaman <iisaman@umich.edu>
+ *
+ * permission is granted to use, copy, create derivative works and
+ * redistribute this software and such derivative works for any purpose,
+ * so long as the name of the university of michigan is not used in
+ * any advertising or publicity pertaining to the use or distribution
+ * of this software without specific, written prior authorization.  if
+ * the above copyright notice or any other identification of the
+ * university of michigan is included in any copy of any portion of
+ * this software, then the disclaimer below must also be included.
+ *
+ * this software is provided as is, without representation from the
+ * university of michigan as to its fitness for any purpose, and without
+ * warranty by the university of michigan of any kind, either express
+ * or implied, including without limitation the implied warranties of
+ * merchantability and fitness for a particular purpose.  the regents
+ * of the university of michigan shall not be liable for any damages,
+ * including special, indirect, incidental, or consequential damages,
+ * with respect to any claim arising out or in connection with the use
+ * of the software, even if it has been or is hereafter advised of the
+ * possibility of such damages.
+ */
+
+#include <linux/module.h>
+#include <linux/blkdev.h>
+
+#include "blocklayout.h"
+
+#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
+
+static void
+nfs4_encode_simple(__be32 *p, struct pnfs_block_volume *b)
+{
+	int i;
+
+	*p++ = cpu_to_be32(1);
+	*p++ = cpu_to_be32(b->type);
+	*p++ = cpu_to_be32(b->simple.nr_sigs);
+	for (i = 0; i < b->simple.nr_sigs; i++) {
+		p = xdr_encode_hyper(p, b->simple.sigs[i].offset);
+		p = xdr_encode_opaque(p, b->simple.sigs[i].sig,
+					 b->simple.sigs[i].sig_len);
+	}
+}
+
+dev_t
+bl_resolve_deviceid(struct nfs_server *server, struct pnfs_block_volume *b,
+		gfp_t gfp_mask)
+{
+	struct net *net = server->nfs_client->cl_net;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	struct bl_dev_msg *reply = &nn->bl_mount_reply;
+	struct bl_pipe_msg bl_pipe_msg;
+	struct rpc_pipe_msg *msg = &bl_pipe_msg.msg;
+	struct bl_msg_hdr *bl_msg;
+	DECLARE_WAITQUEUE(wq, current);
+	dev_t dev = 0;
+	int rc;
+
+	dprintk("%s CREATING PIPEFS MESSAGE\n", __func__);
+
+	mutex_lock(&nn->bl_mutex);
+	bl_pipe_msg.bl_wq = &nn->bl_wq;
+
+	b->simple.len += 4;	/* single volume */
+	if (b->simple.len > PAGE_SIZE)
+		goto out_unlock;
+
+	memset(msg, 0, sizeof(*msg));
+	msg->len = sizeof(*bl_msg) + b->simple.len;
+	msg->data = kzalloc(msg->len, gfp_mask);
+	if (!msg->data)
+		goto out_unlock;
+
+	bl_msg = msg->data;
+	bl_msg->type = BL_DEVICE_MOUNT;
+	bl_msg->totallen = b->simple.len;
+	nfs4_encode_simple(msg->data + sizeof(*bl_msg), b);
+
+	dprintk("%s CALLING USERSPACE DAEMON\n", __func__);
+	add_wait_queue(&nn->bl_wq, &wq);
+	rc = rpc_queue_upcall(nn->bl_device_pipe, msg);
+	if (rc < 0) {
+		remove_wait_queue(&nn->bl_wq, &wq);
+		goto out_free_data;
+	}
+
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	schedule();
+	remove_wait_queue(&nn->bl_wq, &wq);
+
+	if (reply->status != BL_DEVICE_REQUEST_PROC) {
+		printk(KERN_WARNING "%s failed to decode device: %d\n",
+			__func__, reply->status);
+		goto out_free_data;
+	}
+
+	dev = MKDEV(reply->major, reply->minor);
+out_free_data:
+	kfree(msg->data);
+out_unlock:
+	mutex_unlock(&nn->bl_mutex);
+	return dev;
+}
+
+static ssize_t bl_pipe_downcall(struct file *filp, const char __user *src,
+			 size_t mlen)
+{
+	struct nfs_net *nn = net_generic(file_inode(filp)->i_sb->s_fs_info,
+					 nfs_net_id);
+
+	if (mlen != sizeof (struct bl_dev_msg))
+		return -EINVAL;
+
+	if (copy_from_user(&nn->bl_mount_reply, src, mlen) != 0)
+		return -EFAULT;
+
+	wake_up(&nn->bl_wq);
+
+	return mlen;
+}
+
+static void bl_pipe_destroy_msg(struct rpc_pipe_msg *msg)
+{
+	struct bl_pipe_msg *bl_pipe_msg =
+		container_of(msg, struct bl_pipe_msg, msg);
+
+	if (msg->errno >= 0)
+		return;
+	wake_up(bl_pipe_msg->bl_wq);
+}
+
+static const struct rpc_pipe_ops bl_upcall_ops = {
+	.upcall		= rpc_pipe_generic_upcall,
+	.downcall	= bl_pipe_downcall,
+	.destroy_msg	= bl_pipe_destroy_msg,
+};
+
+static int nfs4blocklayout_register_sb(struct super_block *sb,
+					    struct rpc_pipe *pipe)
+{
+	struct dentry *dir;
+	int err;
+
+	dir = rpc_d_lookup_sb(sb, NFS_PIPE_DIRNAME);
+	if (dir == NULL)
+		return -ENOENT;
+	err = rpc_mkpipe_dentry(dir, "blocklayout", NULL, pipe);
+	dput(dir);
+	return err;
+}
+
+static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event,
+			   void *ptr)
+{
+	struct super_block *sb = ptr;
+	struct net *net = sb->s_fs_info;
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	int ret = 0;
+
+	if (!try_module_get(THIS_MODULE))
+		return 0;
+
+	if (nn->bl_device_pipe == NULL) {
+		module_put(THIS_MODULE);
+		return 0;
+	}
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		ret = nfs4blocklayout_register_sb(sb, nn->bl_device_pipe);
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		rpc_unlink(nn->bl_device_pipe);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+static struct notifier_block nfs4blocklayout_block = {
+	.notifier_call = rpc_pipefs_event,
+};
+
+static int nfs4blocklayout_register_net(struct net *net, struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+	int ret;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (!pipefs_sb)
+		return 0;
+	ret = nfs4blocklayout_register_sb(pipefs_sb, pipe);
+	rpc_put_sb_net(net);
+	return ret;
+}
+
+static void nfs4blocklayout_unregister_net(struct net *net,
+					   struct rpc_pipe *pipe)
+{
+	struct super_block *pipefs_sb;
+
+	pipefs_sb = rpc_get_sb_net(net);
+	if (pipefs_sb) {
+		rpc_unlink(pipe);
+		rpc_put_sb_net(net);
+	}
+}
+
+static int nfs4blocklayout_net_init(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+	int err;
+
+	mutex_init(&nn->bl_mutex);
+	init_waitqueue_head(&nn->bl_wq);
+	nn->bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0);
+	if (IS_ERR(nn->bl_device_pipe))
+		return PTR_ERR(nn->bl_device_pipe);
+	err = nfs4blocklayout_register_net(net, nn->bl_device_pipe);
+	if (unlikely(err))
+		rpc_destroy_pipe_data(nn->bl_device_pipe);
+	return err;
+}
+
+static void nfs4blocklayout_net_exit(struct net *net)
+{
+	struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+	nfs4blocklayout_unregister_net(net, nn->bl_device_pipe);
+	rpc_destroy_pipe_data(nn->bl_device_pipe);
+	nn->bl_device_pipe = NULL;
+}
+
+static struct pernet_operations nfs4blocklayout_net_ops = {
+	.init = nfs4blocklayout_net_init,
+	.exit = nfs4blocklayout_net_exit,
+};
+
+int __init bl_init_pipefs(void)
+{
+	int ret;
+
+	ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block);
+	if (ret)
+		goto out;
+	ret = register_pernet_subsys(&nfs4blocklayout_net_ops);
+	if (ret)
+		goto out_unregister_notifier;
+	return 0;
+
+out_unregister_notifier:
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
+out:
+	return ret;
+}
+
+void bl_cleanup_pipefs(void)
+{
+	rpc_pipefs_notifier_unregister(&nfs4blocklayout_block);
+	unregister_pernet_subsys(&nfs4blocklayout_net_ops);
+}