Merge tag 'fuse-update-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:

 - Support directly accessing host page cache from virtiofs. This can
   improve I/O performance for various workloads, as well as reducing
   the memory requirement by eliminating double caching. Thanks to Vivek
   Goyal for doing most of the work on this.

 - Allow automatic submounting inside virtiofs. This allows unique
   st_dev/ st_ino values to be assigned inside the guest to files
   residing on different filesystems on the host. Thanks to Max Reitz
   for the patches.

 - Fix an old use after free bug found by Pradeep P V K.

* tag 'fuse-update-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse: (25 commits)
  virtiofs: calculate number of scatter-gather elements accurately
  fuse: connection remove fix
  fuse: implement crossmounts
  fuse: Allow fuse_fill_super_common() for submounts
  fuse: split fuse_mount off of fuse_conn
  fuse: drop fuse_conn parameter where possible
  fuse: store fuse_conn in fuse_req
  fuse: add submount support to <uapi/linux/fuse.h>
  fuse: fix page dereference after free
  virtiofs: add logic to free up a memory range
  virtiofs: maintain a list of busy elements
  virtiofs: serialize truncate/punch_hole and dax fault path
  virtiofs: define dax address space operations
  virtiofs: add DAX mmap support
  virtiofs: implement dax read/write operations
  virtiofs: introduce setupmapping/removemapping commands
  virtiofs: implement FUSE_INIT map_alignment field
  virtiofs: keep a list of free dax memory ranges
  virtiofs: add a mount option to enable dax
  virtiofs: set up virtio_fs dax_device
  ...

Documentation/filesystems/fuse.rst

@@ -47,7 +47,7 @@ filesystems.  A good example is sshfs: a secure network filesystem
 using the sftp protocol.
 
 The userspace library and utilities are available from the
-`FUSE homepage: <http://fuse.sourceforge.net/>`_
+`FUSE homepage: <https://github.com/libfuse/>`_
 
 Filesystem type
 ===============

MAINTAINERS

@@ -7238,7 +7238,7 @@ FUSE: FILESYSTEM IN USERSPACE
 M:	Miklos Szeredi <miklos@szeredi.hu>
 L:	linux-fsdevel@vger.kernel.org
 S:	Maintained
-W:	http://fuse.sourceforge.net/
+W:	https://github.com/libfuse/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse.git
 F:	Documentation/filesystems/fuse.rst
 F:	fs/fuse/

drivers/dax/super.c

@@ -46,7 +46,8 @@ EXPORT_SYMBOL_GPL(dax_read_unlock);
 int bdev_dax_pgoff(struct block_device *bdev, sector_t sector, size_t size,
 		pgoff_t *pgoff)
 {
-	phys_addr_t phys_off = (get_start_sect(bdev) + sector) * 512;
+	sector_t start_sect = bdev ? get_start_sect(bdev) : 0;
+	phys_addr_t phys_off = (start_sect + sector) * 512;
 
 	if (pgoff)
 		*pgoff = PHYS_PFN(phys_off);

fs/dax.c

@@ -559,8 +559,11 @@ static void *grab_mapping_entry(struct xa_state *xas,
 }
 
 /**
- * dax_layout_busy_page - find first pinned page in @mapping
+ * dax_layout_busy_page_range - find first pinned page in @mapping
  * @mapping: address space to scan for a page with ref count > 1
+ * @start: Starting offset. Page containing 'start' is included.
+ * @end: End offset. Page containing 'end' is included. If 'end' is LLONG_MAX,
+ *       pages from 'start' till the end of file are included.
  *
  * DAX requires ZONE_DEVICE mapped pages. These pages are never
  * 'onlined' to the page allocator so they are considered idle when
@@ -573,12 +576,15 @@ static void *grab_mapping_entry(struct xa_state *xas,
  * to be able to run unmap_mapping_range() and subsequently not race
  * mapping_mapped() becoming true.
  */
-struct page *dax_layout_busy_page(struct address_space *mapping)
+struct page *dax_layout_busy_page_range(struct address_space *mapping,
+					loff_t start, loff_t end)
 {
-	XA_STATE(xas, &mapping->i_pages, 0);
 	void *entry;
 	unsigned int scanned = 0;
 	struct page *page = NULL;
+	pgoff_t start_idx = start >> PAGE_SHIFT;
+	pgoff_t end_idx;
+	XA_STATE(xas, &mapping->i_pages, start_idx);
 
 	/*
 	 * In the 'limited' case get_user_pages() for dax is disabled.
@@ -589,6 +595,11 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
 	if (!dax_mapping(mapping) || !mapping_mapped(mapping))
 		return NULL;
 
+	/* If end == LLONG_MAX, all pages from start to till end of file */
+	if (end == LLONG_MAX)
+		end_idx = ULONG_MAX;
+	else
+		end_idx = end >> PAGE_SHIFT;
 	/*
 	 * If we race get_user_pages_fast() here either we'll see the
 	 * elevated page count in the iteration and wait, or
@@ -596,15 +607,15 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
 	 * against is no longer mapped in the page tables and bail to the
 	 * get_user_pages() slow path.  The slow path is protected by
 	 * pte_lock() and pmd_lock(). New references are not taken without
-	 * holding those locks, and unmap_mapping_range() will not zero the
+	 * holding those locks, and unmap_mapping_pages() will not zero the
 	 * pte or pmd without holding the respective lock, so we are
 	 * guaranteed to either see new references or prevent new
 	 * references from being established.
 	 */
-	unmap_mapping_range(mapping, 0, 0, 0);
+	unmap_mapping_pages(mapping, start_idx, end_idx - start_idx + 1, 0);
 
 	xas_lock_irq(&xas);
-	xas_for_each(&xas, entry, ULONG_MAX) {
+	xas_for_each(&xas, entry, end_idx) {
 		if (WARN_ON_ONCE(!xa_is_value(entry)))
 			continue;
 		if (unlikely(dax_is_locked(entry)))
@@ -625,6 +636,12 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
 	xas_unlock_irq(&xas);
 	return page;
 }
+EXPORT_SYMBOL_GPL(dax_layout_busy_page_range);
+
+struct page *dax_layout_busy_page(struct address_space *mapping)
+{
+	return dax_layout_busy_page_range(mapping, 0, LLONG_MAX);
+}
 EXPORT_SYMBOL_GPL(dax_layout_busy_page);
 
 static int __dax_invalidate_entry(struct address_space *mapping,

fs/fuse/Kconfig

@@ -8,7 +8,7 @@ config FUSE_FS
 
 	  There's also a companion library: libfuse2.  This library is available
 	  from the FUSE homepage:
-	  <http://fuse.sourceforge.net/>
+	  <https://github.com/libfuse/>
 	  although chances are your distribution already has that library
 	  installed if you've installed the "fuse" package itself.
 
@@ -38,3 +38,17 @@ config VIRTIO_FS
 
 	  If you want to share files between guests or with the host, answer Y
 	  or M.
+
+config FUSE_DAX
+	bool "Virtio Filesystem Direct Host Memory Access support"
+	default y
+	select INTERVAL_TREE
+	depends on VIRTIO_FS
+	depends on FS_DAX
+	depends on DAX_DRIVER
+	help
+	  This allows bypassing guest page cache and allows mapping host page
+	  cache directly in guest address space.
+
+	  If you want to allow mounting a Virtio Filesystem with the "dax"
+	  option, answer Y.

fs/fuse/Makefile

@@ -7,5 +7,7 @@ obj-$(CONFIG_FUSE_FS) += fuse.o
 obj-$(CONFIG_CUSE) += cuse.o
 obj-$(CONFIG_VIRTIO_FS) += virtiofs.o
 
-fuse-objs := dev.o dir.o file.o inode.o control.o xattr.o acl.o readdir.o
-virtiofs-y += virtio_fs.o
+fuse-y := dev.o dir.o file.o inode.o control.o xattr.o acl.o readdir.o
+fuse-$(CONFIG_FUSE_DAX) += dax.o
+
+virtiofs-y := virtio_fs.o

fs/fuse/control.c

@@ -164,6 +164,7 @@ static ssize_t fuse_conn_congestion_threshold_write(struct file *file,
 {
 	unsigned val;
 	struct fuse_conn *fc;
+	struct fuse_mount *fm;
 	ssize_t ret;
 
 	ret = fuse_conn_limit_write(file, buf, count, ppos, &val,
@@ -174,18 +175,27 @@ static ssize_t fuse_conn_congestion_threshold_write(struct file *file,
 	if (!fc)
 		goto out;
 
+	down_read(&fc->killsb);
 	spin_lock(&fc->bg_lock);
 	fc->congestion_threshold = val;
-	if (fc->sb) {
+
+	/*
+	 * Get any fuse_mount belonging to this fuse_conn; s_bdi is
+	 * shared between all of them
+	 */
+
+	if (!list_empty(&fc->mounts)) {
+		fm = list_first_entry(&fc->mounts, struct fuse_mount, fc_entry);
 		if (fc->num_background < fc->congestion_threshold) {
-			clear_bdi_congested(fc->sb->s_bdi, BLK_RW_SYNC);
-			clear_bdi_congested(fc->sb->s_bdi, BLK_RW_ASYNC);
+			clear_bdi_congested(fm->sb->s_bdi, BLK_RW_SYNC);
+			clear_bdi_congested(fm->sb->s_bdi, BLK_RW_ASYNC);
 		} else {
-			set_bdi_congested(fc->sb->s_bdi, BLK_RW_SYNC);
-			set_bdi_congested(fc->sb->s_bdi, BLK_RW_ASYNC);
+			set_bdi_congested(fm->sb->s_bdi, BLK_RW_SYNC);
+			set_bdi_congested(fm->sb->s_bdi, BLK_RW_ASYNC);
 		}
 	}
 	spin_unlock(&fc->bg_lock);
+	up_read(&fc->killsb);
 	fuse_conn_put(fc);
 out:
 	return ret;

fs/fuse/cuse.c

@@ -57,6 +57,7 @@
 
 struct cuse_conn {
 	struct list_head	list;	/* linked on cuse_conntbl */
+	struct fuse_mount	fm;	/* Dummy mount referencing fc */
 	struct fuse_conn	fc;	/* fuse connection */
 	struct cdev		*cdev;	/* associated character device */
 	struct device		*dev;	/* device representing @cdev */
@@ -134,7 +135,7 @@ static int cuse_open(struct inode *inode, struct file *file)
 	 * Generic permission check is already done against the chrdev
 	 * file, proceed to open.
 	 */
-	rc = fuse_do_open(&cc->fc, 0, file, 0);
+	rc = fuse_do_open(&cc->fm, 0, file, 0);
 	if (rc)
 		fuse_conn_put(&cc->fc);
 	return rc;
@@ -143,10 +144,10 @@ static int cuse_open(struct inode *inode, struct file *file)
 static int cuse_release(struct inode *inode, struct file *file)
 {
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 
 	fuse_sync_release(NULL, ff, file->f_flags);
-	fuse_conn_put(fc);
+	fuse_conn_put(fm->fc);
 
 	return 0;
 }
@@ -155,7 +156,7 @@ static long cuse_file_ioctl(struct file *file, unsigned int cmd,
 			    unsigned long arg)
 {
 	struct fuse_file *ff = file->private_data;
-	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	struct cuse_conn *cc = fc_to_cc(ff->fm->fc);
 	unsigned int flags = 0;
 
 	if (cc->unrestricted_ioctl)
@@ -168,7 +169,7 @@ static long cuse_file_compat_ioctl(struct file *file, unsigned int cmd,
 				   unsigned long arg)
 {
 	struct fuse_file *ff = file->private_data;
-	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	struct cuse_conn *cc = fc_to_cc(ff->fm->fc);
 	unsigned int flags = FUSE_IOCTL_COMPAT;
 
 	if (cc->unrestricted_ioctl)
@@ -313,9 +314,10 @@ struct cuse_init_args {
  * required data structures for it.  Please read the comment at the
  * top of this file for high level overview.
  */
-static void cuse_process_init_reply(struct fuse_conn *fc,
+static void cuse_process_init_reply(struct fuse_mount *fm,
 				    struct fuse_args *args, int error)
 {
+	struct fuse_conn *fc = fm->fc;
 	struct cuse_init_args *ia = container_of(args, typeof(*ia), ap.args);
 	struct fuse_args_pages *ap = &ia->ap;
 	struct cuse_conn *cc = fc_to_cc(fc), *pos;
@@ -424,7 +426,7 @@ static int cuse_send_init(struct cuse_conn *cc)
 {
 	int rc;
 	struct page *page;
-	struct fuse_conn *fc = &cc->fc;
+	struct fuse_mount *fm = &cc->fm;
 	struct cuse_init_args *ia;
 	struct fuse_args_pages *ap;
 
@@ -460,7 +462,7 @@ static int cuse_send_init(struct cuse_conn *cc)
 	ia->desc.length = ap->args.out_args[1].size;
 	ap->args.end = cuse_process_init_reply;
 
-	rc = fuse_simple_background(fc, &ap->args, GFP_KERNEL);
+	rc = fuse_simple_background(fm, &ap->args, GFP_KERNEL);
 	if (rc) {
 		kfree(ia);
 err_free_page:
@@ -506,7 +508,8 @@ static int cuse_channel_open(struct inode *inode, struct file *file)
 	 * Limit the cuse channel to requests that can
 	 * be represented in file->f_cred->user_ns.
 	 */
-	fuse_conn_init(&cc->fc, file->f_cred->user_ns, &fuse_dev_fiq_ops, NULL);
+	fuse_conn_init(&cc->fc, &cc->fm, file->f_cred->user_ns,
+		       &fuse_dev_fiq_ops, NULL);
 
 	fud = fuse_dev_alloc_install(&cc->fc);
 	if (!fud) {

fs/fuse/dax.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * dax: direct host memory access
+ * Copyright (C) 2020 Red Hat, Inc.
+ */
+
+#include "fuse_i.h"
+
+#include <linux/delay.h>
+#include <linux/dax.h>
+#include <linux/uio.h>
+#include <linux/pfn_t.h>
+#include <linux/iomap.h>
+#include <linux/interval_tree.h>
+
+/*
+ * Default memory range size.  A power of 2 so it agrees with common FUSE_INIT
+ * map_alignment values 4KB and 64KB.
+ */
+#define FUSE_DAX_SHIFT	21
+#define FUSE_DAX_SZ	(1 << FUSE_DAX_SHIFT)
+#define FUSE_DAX_PAGES	(FUSE_DAX_SZ / PAGE_SIZE)
+
+/* Number of ranges reclaimer will try to free in one invocation */
+#define FUSE_DAX_RECLAIM_CHUNK		(10)
+
+/*
+ * Dax memory reclaim threshold in percetage of total ranges. When free
+ * number of free ranges drops below this threshold, reclaim can trigger
+ * Default is 20%
+ */
+#define FUSE_DAX_RECLAIM_THRESHOLD	(20)
+
+/** Translation information for file offsets to DAX window offsets */
+struct fuse_dax_mapping {
+	/* Pointer to inode where this memory range is mapped */
+	struct inode *inode;
+
+	/* Will connect in fcd->free_ranges to keep track of free memory */
+	struct list_head list;
+
+	/* For interval tree in file/inode */
+	struct interval_tree_node itn;
+
+	/* Will connect in fc->busy_ranges to keep track busy memory */
+	struct list_head busy_list;
+
+	/** Position in DAX window */
+	u64 window_offset;
+
+	/** Length of mapping, in bytes */
+	loff_t length;
+
+	/* Is this mapping read-only or read-write */
+	bool writable;
+
+	/* reference count when the mapping is used by dax iomap. */
+	refcount_t refcnt;
+};
+
+/* Per-inode dax map */
+struct fuse_inode_dax {
+	/* Semaphore to protect modifications to the dmap tree */
+	struct rw_semaphore sem;
+
+	/* Sorted rb tree of struct fuse_dax_mapping elements */
+	struct rb_root_cached tree;
+	unsigned long nr;
+};
+
+struct fuse_conn_dax {
+	/* DAX device */
+	struct dax_device *dev;
+
+	/* Lock protecting accessess to  members of this structure */
+	spinlock_t lock;
+
+	/* List of memory ranges which are busy */
+	unsigned long nr_busy_ranges;
+	struct list_head busy_ranges;
+
+	/* Worker to free up memory ranges */
+	struct delayed_work free_work;
+
+	/* Wait queue for a dax range to become free */
+	wait_queue_head_t range_waitq;
+
+	/* DAX Window Free Ranges */
+	long nr_free_ranges;
+	struct list_head free_ranges;
+
+	unsigned long nr_ranges;
+};
+
+static inline struct fuse_dax_mapping *
+node_to_dmap(struct interval_tree_node *node)
+{
+	if (!node)
+		return NULL;
+
+	return container_of(node, struct fuse_dax_mapping, itn);
+}
+
+static struct fuse_dax_mapping *
+alloc_dax_mapping_reclaim(struct fuse_conn_dax *fcd, struct inode *inode);
+
+static void
+__kick_dmap_free_worker(struct fuse_conn_dax *fcd, unsigned long delay_ms)
+{
+	unsigned long free_threshold;
+
+	/* If number of free ranges are below threshold, start reclaim */
+	free_threshold = max_t(unsigned long, fcd->nr_ranges * FUSE_DAX_RECLAIM_THRESHOLD / 100,
+			     1);
+	if (fcd->nr_free_ranges < free_threshold)
+		queue_delayed_work(system_long_wq, &fcd->free_work,
+				   msecs_to_jiffies(delay_ms));
+}
+
+static void kick_dmap_free_worker(struct fuse_conn_dax *fcd,
+				  unsigned long delay_ms)
+{
+	spin_lock(&fcd->lock);
+	__kick_dmap_free_worker(fcd, delay_ms);
+	spin_unlock(&fcd->lock);
+}
+
+static struct fuse_dax_mapping *alloc_dax_mapping(struct fuse_conn_dax *fcd)
+{
+	struct fuse_dax_mapping *dmap;
+
+	spin_lock(&fcd->lock);
+	dmap = list_first_entry_or_null(&fcd->free_ranges,
+					struct fuse_dax_mapping, list);
+	if (dmap) {
+		list_del_init(&dmap->list);
+		WARN_ON(fcd->nr_free_ranges <= 0);
+		fcd->nr_free_ranges--;
+	}
+	spin_unlock(&fcd->lock);
+
+	kick_dmap_free_worker(fcd, 0);
+	return dmap;
+}
+
+/* This assumes fcd->lock is held */
+static void __dmap_remove_busy_list(struct fuse_conn_dax *fcd,
+				    struct fuse_dax_mapping *dmap)
+{
+	list_del_init(&dmap->busy_list);
+	WARN_ON(fcd->nr_busy_ranges == 0);
+	fcd->nr_busy_ranges--;
+}
+
+static void dmap_remove_busy_list(struct fuse_conn_dax *fcd,
+				  struct fuse_dax_mapping *dmap)
+{
+	spin_lock(&fcd->lock);
+	__dmap_remove_busy_list(fcd, dmap);
+	spin_unlock(&fcd->lock);
+}
+
+/* This assumes fcd->lock is held */
+static void __dmap_add_to_free_pool(struct fuse_conn_dax *fcd,
+				struct fuse_dax_mapping *dmap)
+{
+	list_add_tail(&dmap->list, &fcd->free_ranges);
+	fcd->nr_free_ranges++;
+	wake_up(&fcd->range_waitq);
+}
+
+static void dmap_add_to_free_pool(struct fuse_conn_dax *fcd,
+				struct fuse_dax_mapping *dmap)
+{
+	/* Return fuse_dax_mapping to free list */
+	spin_lock(&fcd->lock);
+	__dmap_add_to_free_pool(fcd, dmap);
+	spin_unlock(&fcd->lock);
+}
+
+static int fuse_setup_one_mapping(struct inode *inode, unsigned long start_idx,
+				  struct fuse_dax_mapping *dmap, bool writable,
+				  bool upgrade)
+{
+	struct fuse_mount *fm = get_fuse_mount(inode);
+	struct fuse_conn_dax *fcd = fm->fc->dax;
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_setupmapping_in inarg;
+	loff_t offset = start_idx << FUSE_DAX_SHIFT;
+	FUSE_ARGS(args);
+	ssize_t err;
+
+	WARN_ON(fcd->nr_free_ranges < 0);
+
+	/* Ask fuse daemon to setup mapping */
+	memset(&inarg, 0, sizeof(inarg));
+	inarg.foffset = offset;
+	inarg.fh = -1;
+	inarg.moffset = dmap->window_offset;
+	inarg.len = FUSE_DAX_SZ;
+	inarg.flags |= FUSE_SETUPMAPPING_FLAG_READ;
+	if (writable)
+		inarg.flags |= FUSE_SETUPMAPPING_FLAG_WRITE;
+	args.opcode = FUSE_SETUPMAPPING;
+	args.nodeid = fi->nodeid;
+	args.in_numargs = 1;
+	args.in_args[0].size = sizeof(inarg);
+	args.in_args[0].value = &inarg;
+	err = fuse_simple_request(fm, &args);
+	if (err < 0)
+		return err;
+	dmap->writable = writable;
+	if (!upgrade) {
+		/*
+		 * We don't take a refernce on inode. inode is valid right now
+		 * and when inode is going away, cleanup logic should first
+		 * cleanup dmap entries.
+		 */
+		dmap->inode = inode;
+		dmap->itn.start = dmap->itn.last = start_idx;
+		/* Protected by fi->dax->sem */
+		interval_tree_insert(&dmap->itn, &fi->dax->tree);
+		fi->dax->nr++;
+		spin_lock(&fcd->lock);
+		list_add_tail(&dmap->busy_list, &fcd->busy_ranges);
+		fcd->nr_busy_ranges++;
+		spin_unlock(&fcd->lock);
+	}
+	return 0;
+}
+
+static int fuse_send_removemapping(struct inode *inode,
+				   struct fuse_removemapping_in *inargp,
+				   struct fuse_removemapping_one *remove_one)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
+	FUSE_ARGS(args);
+
+	args.opcode = FUSE_REMOVEMAPPING;
+	args.nodeid = fi->nodeid;
+	args.in_numargs = 2;
+	args.in_args[0].size = sizeof(*inargp);
+	args.in_args[0].value = inargp;
+	args.in_args[1].size = inargp->count * sizeof(*remove_one);
+	args.in_args[1].value = remove_one;
+	return fuse_simple_request(fm, &args);
+}
+
+static int dmap_removemapping_list(struct inode *inode, unsigned int num,
+				   struct list_head *to_remove)
+{
+	struct fuse_removemapping_one *remove_one, *ptr;
+	struct fuse_removemapping_in inarg;
+	struct fuse_dax_mapping *dmap;
+	int ret, i = 0, nr_alloc;
+
+	nr_alloc = min_t(unsigned int, num, FUSE_REMOVEMAPPING_MAX_ENTRY);
+	remove_one = kmalloc_array(nr_alloc, sizeof(*remove_one), GFP_NOFS);
+	if (!remove_one)
+		return -ENOMEM;
+
+	ptr = remove_one;
+	list_for_each_entry(dmap, to_remove, list) {
+		ptr->moffset = dmap->window_offset;
+		ptr->len = dmap->length;
+		ptr++;
+		i++;
+		num--;
+		if (i >= nr_alloc || num == 0) {
+			memset(&inarg, 0, sizeof(inarg));
+			inarg.count = i;
+			ret = fuse_send_removemapping(inode, &inarg,
+						      remove_one);
+			if (ret)
+				goto out;
+			ptr = remove_one;
+			i = 0;
+		}
+	}
+out:
+	kfree(remove_one);
+	return ret;
+}
+
+/*
+ * Cleanup dmap entry and add back to free list. This should be called with
+ * fcd->lock held.
+ */
+static void dmap_reinit_add_to_free_pool(struct fuse_conn_dax *fcd,
+					    struct fuse_dax_mapping *dmap)
+{
+	pr_debug("fuse: freeing memory range start_idx=0x%lx end_idx=0x%lx window_offset=0x%llx length=0x%llx\n",
+		 dmap->itn.start, dmap->itn.last, dmap->window_offset,
+		 dmap->length);
+	__dmap_remove_busy_list(fcd, dmap);
+	dmap->inode = NULL;
+	dmap->itn.start = dmap->itn.last = 0;
+	__dmap_add_to_free_pool(fcd, dmap);
+}
+
+/*
+ * Free inode dmap entries whose range falls inside [start, end].
+ * Does not take any locks. At this point of time it should only be
+ * called from evict_inode() path where we know all dmap entries can be
+ * reclaimed.
+ */
+static void inode_reclaim_dmap_range(struct fuse_conn_dax *fcd,
+				     struct inode *inode,
+				     loff_t start, loff_t end)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_dax_mapping *dmap, *n;
+	int err, num = 0;
+	LIST_HEAD(to_remove);
+	unsigned long start_idx = start >> FUSE_DAX_SHIFT;
+	unsigned long end_idx = end >> FUSE_DAX_SHIFT;
+	struct interval_tree_node *node;
+
+	while (1) {
+		node = interval_tree_iter_first(&fi->dax->tree, start_idx,
+						end_idx);
+		if (!node)
+			break;
+		dmap = node_to_dmap(node);
+		/* inode is going away. There should not be any users of dmap */
+		WARN_ON(refcount_read(&dmap->refcnt) > 1);
+		interval_tree_remove(&dmap->itn, &fi->dax->tree);
+		num++;
+		list_add(&dmap->list, &to_remove);
+	}
+
+	/* Nothing to remove */
+	if (list_empty(&to_remove))
+		return;
+
+	WARN_ON(fi->dax->nr < num);
+	fi->dax->nr -= num;
+	err = dmap_removemapping_list(inode, num, &to_remove);
+	if (err && err != -ENOTCONN) {
+		pr_warn("Failed to removemappings. start=0x%llx end=0x%llx\n",
+			start, end);
+	}
+	spin_lock(&fcd->lock);
+	list_for_each_entry_safe(dmap, n, &to_remove, list) {
+		list_del_init(&dmap->list);
+		dmap_reinit_add_to_free_pool(fcd, dmap);
+	}
+	spin_unlock(&fcd->lock);
+}
+
+static int dmap_removemapping_one(struct inode *inode,
+				  struct fuse_dax_mapping *dmap)
+{
+	struct fuse_removemapping_one forget_one;
+	struct fuse_removemapping_in inarg;
+
+	memset(&inarg, 0, sizeof(inarg));
+	inarg.count = 1;
+	memset(&forget_one, 0, sizeof(forget_one));
+	forget_one.moffset = dmap->window_offset;
+	forget_one.len = dmap->length;
+
+	return fuse_send_removemapping(inode, &inarg, &forget_one);
+}
+
+/*
+ * It is called from evict_inode() and by that time inode is going away. So
+ * this function does not take any locks like fi->dax->sem for traversing
+ * that fuse inode interval tree. If that lock is taken then lock validator
+ * complains of deadlock situation w.r.t fs_reclaim lock.
+ */
+void fuse_dax_inode_cleanup(struct inode *inode)
+{
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_inode *fi = get_fuse_inode(inode);
+
+	/*
+	 * fuse_evict_inode() has already called truncate_inode_pages_final()
+	 * before we arrive here. So we should not have to worry about any
+	 * pages/exception entries still associated with inode.
+	 */
+	inode_reclaim_dmap_range(fc->dax, inode, 0, -1);
+	WARN_ON(fi->dax->nr);
+}
+
+static void fuse_fill_iomap_hole(struct iomap *iomap, loff_t length)
+{
+	iomap->addr = IOMAP_NULL_ADDR;
+	iomap->length = length;
+	iomap->type = IOMAP_HOLE;
+}
+
+static void fuse_fill_iomap(struct inode *inode, loff_t pos, loff_t length,
+			    struct iomap *iomap, struct fuse_dax_mapping *dmap,
+			    unsigned int flags)
+{
+	loff_t offset, len;
+	loff_t i_size = i_size_read(inode);
+
+	offset = pos - (dmap->itn.start << FUSE_DAX_SHIFT);
+	len = min(length, dmap->length - offset);
+
+	/* If length is beyond end of file, truncate further */
+	if (pos + len > i_size)
+		len = i_size - pos;
+
+	if (len > 0) {
+		iomap->addr = dmap->window_offset + offset;
+		iomap->length = len;
+		if (flags & IOMAP_FAULT)
+			iomap->length = ALIGN(len, PAGE_SIZE);
+		iomap->type = IOMAP_MAPPED;
+		/*
+		 * increace refcnt so that reclaim code knows this dmap is in
+		 * use. This assumes fi->dax->sem mutex is held either
+		 * shared/exclusive.
+		 */
+		refcount_inc(&dmap->refcnt);
+
+		/* iomap->private should be NULL */
+		WARN_ON_ONCE(iomap->private);
+		iomap->private = dmap;
+	} else {
+		/* Mapping beyond end of file is hole */
+		fuse_fill_iomap_hole(iomap, length);
+	}
+}
+
+static int fuse_setup_new_dax_mapping(struct inode *inode, loff_t pos,
+				      loff_t length, unsigned int flags,
+				      struct iomap *iomap)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_conn_dax *fcd = fc->dax;
+	struct fuse_dax_mapping *dmap, *alloc_dmap = NULL;
+	int ret;
+	bool writable = flags & IOMAP_WRITE;
+	unsigned long start_idx = pos >> FUSE_DAX_SHIFT;
+	struct interval_tree_node *node;
+
+	/*
+	 * Can't do inline reclaim in fault path. We call
+	 * dax_layout_busy_page() before we free a range. And
+	 * fuse_wait_dax_page() drops fi->i_mmap_sem lock and requires it.
+	 * In fault path we enter with fi->i_mmap_sem held and can't drop
+	 * it. Also in fault path we hold fi->i_mmap_sem shared and not
+	 * exclusive, so that creates further issues with fuse_wait_dax_page().
+	 * Hence return -EAGAIN and fuse_dax_fault() will wait for a memory
+	 * range to become free and retry.
+	 */
+	if (flags & IOMAP_FAULT) {
+		alloc_dmap = alloc_dax_mapping(fcd);
+		if (!alloc_dmap)
+			return -EAGAIN;
+	} else {
+		alloc_dmap = alloc_dax_mapping_reclaim(fcd, inode);
+		if (IS_ERR(alloc_dmap))
+			return PTR_ERR(alloc_dmap);
+	}
+
+	/* If we are here, we should have memory allocated */
+	if (WARN_ON(!alloc_dmap))
+		return -EIO;
+
+	/*
+	 * Take write lock so that only one caller can try to setup mapping
+	 * and other waits.
+	 */
+	down_write(&fi->dax->sem);
+	/*
+	 * We dropped lock. Check again if somebody else setup
+	 * mapping already.
+	 */
+	node = interval_tree_iter_first(&fi->dax->tree, start_idx, start_idx);
+	if (node) {
+		dmap = node_to_dmap(node);
+		fuse_fill_iomap(inode, pos, length, iomap, dmap, flags);
+		dmap_add_to_free_pool(fcd, alloc_dmap);
+		up_write(&fi->dax->sem);
+		return 0;
+	}
+
+	/* Setup one mapping */
+	ret = fuse_setup_one_mapping(inode, pos >> FUSE_DAX_SHIFT, alloc_dmap,
+				     writable, false);
+	if (ret < 0) {
+		dmap_add_to_free_pool(fcd, alloc_dmap);
+		up_write(&fi->dax->sem);
+		return ret;
+	}
+	fuse_fill_iomap(inode, pos, length, iomap, alloc_dmap, flags);
+	up_write(&fi->dax->sem);
+	return 0;
+}
+
+static int fuse_upgrade_dax_mapping(struct inode *inode, loff_t pos,
+				    loff_t length, unsigned int flags,
+				    struct iomap *iomap)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_dax_mapping *dmap;
+	int ret;
+	unsigned long idx = pos >> FUSE_DAX_SHIFT;
+	struct interval_tree_node *node;
+
+	/*
+	 * Take exclusive lock so that only one caller can try to setup
+	 * mapping and others wait.
+	 */
+	down_write(&fi->dax->sem);
+	node = interval_tree_iter_first(&fi->dax->tree, idx, idx);
+
+	/* We are holding either inode lock or i_mmap_sem, and that should
+	 * ensure that dmap can't be truncated. We are holding a reference
+	 * on dmap and that should make sure it can't be reclaimed. So dmap
+	 * should still be there in tree despite the fact we dropped and
+	 * re-acquired the fi->dax->sem lock.
+	 */
+	ret = -EIO;
+	if (WARN_ON(!node))
+		goto out_err;
+
+	dmap = node_to_dmap(node);
+
+	/* We took an extra reference on dmap to make sure its not reclaimd.
+	 * Now we hold fi->dax->sem lock and that reference is not needed
+	 * anymore. Drop it.
+	 */
+	if (refcount_dec_and_test(&dmap->refcnt)) {
+		/* refcount should not hit 0. This object only goes
+		 * away when fuse connection goes away
+		 */
+		WARN_ON_ONCE(1);
+	}
+
+	/* Maybe another thread already upgraded mapping while we were not
+	 * holding lock.
+	 */
+	if (dmap->writable) {
+		ret = 0;
+		goto out_fill_iomap;
+	}
+
+	ret = fuse_setup_one_mapping(inode, pos >> FUSE_DAX_SHIFT, dmap, true,
+				     true);
+	if (ret < 0)
+		goto out_err;
+out_fill_iomap:
+	fuse_fill_iomap(inode, pos, length, iomap, dmap, flags);
+out_err:
+	up_write(&fi->dax->sem);
+	return ret;
+}
+
+/* This is just for DAX and the mapping is ephemeral, do not use it for other
+ * purposes since there is no block device with a permanent mapping.
+ */
+static int fuse_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
+			    unsigned int flags, struct iomap *iomap,
+			    struct iomap *srcmap)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_dax_mapping *dmap;
+	bool writable = flags & IOMAP_WRITE;
+	unsigned long start_idx = pos >> FUSE_DAX_SHIFT;
+	struct interval_tree_node *node;
+
+	/* We don't support FIEMAP */
+	if (WARN_ON(flags & IOMAP_REPORT))
+		return -EIO;
+
+	iomap->offset = pos;
+	iomap->flags = 0;
+	iomap->bdev = NULL;
+	iomap->dax_dev = fc->dax->dev;
+
+	/*
+	 * Both read/write and mmap path can race here. So we need something
+	 * to make sure if we are setting up mapping, then other path waits
+	 *
+	 * For now, use a semaphore for this. It probably needs to be
+	 * optimized later.
+	 */
+	down_read(&fi->dax->sem);
+	node = interval_tree_iter_first(&fi->dax->tree, start_idx, start_idx);
+	if (node) {
+		dmap = node_to_dmap(node);
+		if (writable && !dmap->writable) {
+			/* Upgrade read-only mapping to read-write. This will
+			 * require exclusive fi->dax->sem lock as we don't want
+			 * two threads to be trying to this simultaneously
+			 * for same dmap. So drop shared lock and acquire
+			 * exclusive lock.
+			 *
+			 * Before dropping fi->dax->sem lock, take reference
+			 * on dmap so that its not freed by range reclaim.
+			 */
+			refcount_inc(&dmap->refcnt);
+			up_read(&fi->dax->sem);
+			pr_debug("%s: Upgrading mapping at offset 0x%llx length 0x%llx\n",
+				 __func__, pos, length);
+			return fuse_upgrade_dax_mapping(inode, pos, length,
+							flags, iomap);
+		} else {
+			fuse_fill_iomap(inode, pos, length, iomap, dmap, flags);
+			up_read(&fi->dax->sem);
+			return 0;
+		}
+	} else {
+		up_read(&fi->dax->sem);
+		pr_debug("%s: no mapping at offset 0x%llx length 0x%llx\n",
+				__func__, pos, length);
+		if (pos >= i_size_read(inode))
+			goto iomap_hole;
+
+		return fuse_setup_new_dax_mapping(inode, pos, length, flags,
+						  iomap);
+	}
+
+	/*
+	 * If read beyond end of file happnes, fs code seems to return
+	 * it as hole
+	 */
+iomap_hole:
+	fuse_fill_iomap_hole(iomap, length);
+	pr_debug("%s returning hole mapping. pos=0x%llx length_asked=0x%llx length_returned=0x%llx\n",
+		 __func__, pos, length, iomap->length);
+	return 0;
+}
+
+static int fuse_iomap_end(struct inode *inode, loff_t pos, loff_t length,
+			  ssize_t written, unsigned int flags,
+			  struct iomap *iomap)
+{
+	struct fuse_dax_mapping *dmap = iomap->private;
+
+	if (dmap) {
+		if (refcount_dec_and_test(&dmap->refcnt)) {
+			/* refcount should not hit 0. This object only goes
+			 * away when fuse connection goes away
+			 */
+			WARN_ON_ONCE(1);
+		}
+	}
+
+	/* DAX writes beyond end-of-file aren't handled using iomap, so the
+	 * file size is unchanged and there is nothing to do here.
+	 */
+	return 0;
+}
+
+static const struct iomap_ops fuse_iomap_ops = {
+	.iomap_begin = fuse_iomap_begin,
+	.iomap_end = fuse_iomap_end,
+};
+
+static void fuse_wait_dax_page(struct inode *inode)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+
+	up_write(&fi->i_mmap_sem);
+	schedule();
+	down_write(&fi->i_mmap_sem);
+}
+
+/* Should be called with fi->i_mmap_sem lock held exclusively */
+static int __fuse_dax_break_layouts(struct inode *inode, bool *retry,
+				    loff_t start, loff_t end)
+{
+	struct page *page;
+
+	page = dax_layout_busy_page_range(inode->i_mapping, start, end);
+	if (!page)
+		return 0;
+
+	*retry = true;
+	return ___wait_var_event(&page->_refcount,
+			atomic_read(&page->_refcount) == 1, TASK_INTERRUPTIBLE,
+			0, 0, fuse_wait_dax_page(inode));
+}
+
+/* dmap_end == 0 leads to unmapping of whole file */
+int fuse_dax_break_layouts(struct inode *inode, u64 dmap_start,
+				  u64 dmap_end)
+{
+	bool	retry;
+	int	ret;
+
+	do {
+		retry = false;
+		ret = __fuse_dax_break_layouts(inode, &retry, dmap_start,
+					       dmap_end);
+	} while (ret == 0 && retry);
+
+	return ret;
+}
+
+ssize_t fuse_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t ret;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock_shared(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock_shared(inode);
+	}
+
+	ret = dax_iomap_rw(iocb, to, &fuse_iomap_ops);
+	inode_unlock_shared(inode);
+
+	/* TODO file_accessed(iocb->f_filp) */
+	return ret;
+}
+
+static bool file_extending_write(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	return (iov_iter_rw(from) == WRITE &&
+		((iocb->ki_pos) >= i_size_read(inode) ||
+		  (iocb->ki_pos + iov_iter_count(from) > i_size_read(inode))));
+}
+
+static ssize_t fuse_dax_direct_write(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct fuse_io_priv io = FUSE_IO_PRIV_SYNC(iocb);
+	ssize_t ret;
+
+	ret = fuse_direct_io(&io, from, &iocb->ki_pos, FUSE_DIO_WRITE);
+	if (ret < 0)
+		return ret;
+
+	fuse_invalidate_attr(inode);
+	fuse_write_update_size(inode, iocb->ki_pos);
+	return ret;
+}
+
+ssize_t fuse_dax_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	ssize_t ret;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock(inode);
+	}
+
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto out;
+
+	ret = file_remove_privs(iocb->ki_filp);
+	if (ret)
+		goto out;
+	/* TODO file_update_time() but we don't want metadata I/O */
+
+	/* Do not use dax for file extending writes as write and on
+	 * disk i_size increase are not atomic otherwise.
+	 */
+	if (file_extending_write(iocb, from))
+		ret = fuse_dax_direct_write(iocb, from);
+	else
+		ret = dax_iomap_rw(iocb, from, &fuse_iomap_ops);
+
+out:
+	inode_unlock(inode);
+
+	if (ret > 0)
+		ret = generic_write_sync(iocb, ret);
+	return ret;
+}
+
+static int fuse_dax_writepages(struct address_space *mapping,
+			       struct writeback_control *wbc)
+{
+
+	struct inode *inode = mapping->host;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+
+	return dax_writeback_mapping_range(mapping, fc->dax->dev, wbc);
+}
+
+static vm_fault_t __fuse_dax_fault(struct vm_fault *vmf,
+				   enum page_entry_size pe_size, bool write)
+{
+	vm_fault_t ret;
+	struct inode *inode = file_inode(vmf->vma->vm_file);
+	struct super_block *sb = inode->i_sb;
+	pfn_t pfn;
+	int error = 0;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_conn_dax *fcd = fc->dax;
+	bool retry = false;
+
+	if (write)
+		sb_start_pagefault(sb);
+retry:
+	if (retry && !(fcd->nr_free_ranges > 0))
+		wait_event(fcd->range_waitq, (fcd->nr_free_ranges > 0));
+
+	/*
+	 * We need to serialize against not only truncate but also against
+	 * fuse dax memory range reclaim. While a range is being reclaimed,
+	 * we do not want any read/write/mmap to make progress and try
+	 * to populate page cache or access memory we are trying to free.
+	 */
+	down_read(&get_fuse_inode(inode)->i_mmap_sem);
+	ret = dax_iomap_fault(vmf, pe_size, &pfn, &error, &fuse_iomap_ops);
+	if ((ret & VM_FAULT_ERROR) && error == -EAGAIN) {
+		error = 0;
+		retry = true;
+		up_read(&get_fuse_inode(inode)->i_mmap_sem);
+		goto retry;
+	}
+
+	if (ret & VM_FAULT_NEEDDSYNC)
+		ret = dax_finish_sync_fault(vmf, pe_size, pfn);
+	up_read(&get_fuse_inode(inode)->i_mmap_sem);
+
+	if (write)
+		sb_end_pagefault(sb);
+
+	return ret;
+}
+
+static vm_fault_t fuse_dax_fault(struct vm_fault *vmf)
+{
+	return __fuse_dax_fault(vmf, PE_SIZE_PTE,
+				vmf->flags & FAULT_FLAG_WRITE);
+}
+
+static vm_fault_t fuse_dax_huge_fault(struct vm_fault *vmf,
+			       enum page_entry_size pe_size)
+{
+	return __fuse_dax_fault(vmf, pe_size, vmf->flags & FAULT_FLAG_WRITE);
+}
+
+static vm_fault_t fuse_dax_page_mkwrite(struct vm_fault *vmf)
+{
+	return __fuse_dax_fault(vmf, PE_SIZE_PTE, true);
+}
+
+static vm_fault_t fuse_dax_pfn_mkwrite(struct vm_fault *vmf)
+{
+	return __fuse_dax_fault(vmf, PE_SIZE_PTE, true);
+}
+
+static const struct vm_operations_struct fuse_dax_vm_ops = {
+	.fault		= fuse_dax_fault,
+	.huge_fault	= fuse_dax_huge_fault,
+	.page_mkwrite	= fuse_dax_page_mkwrite,
+	.pfn_mkwrite	= fuse_dax_pfn_mkwrite,
+};
+
+int fuse_dax_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	file_accessed(file);
+	vma->vm_ops = &fuse_dax_vm_ops;
+	vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+	return 0;
+}
+
+static int dmap_writeback_invalidate(struct inode *inode,
+				     struct fuse_dax_mapping *dmap)
+{
+	int ret;
+	loff_t start_pos = dmap->itn.start << FUSE_DAX_SHIFT;
+	loff_t end_pos = (start_pos + FUSE_DAX_SZ - 1);
+
+	ret = filemap_fdatawrite_range(inode->i_mapping, start_pos, end_pos);
+	if (ret) {
+		pr_debug("fuse: filemap_fdatawrite_range() failed. err=%d start_pos=0x%llx, end_pos=0x%llx\n",
+			 ret, start_pos, end_pos);
+		return ret;
+	}
+
+	ret = invalidate_inode_pages2_range(inode->i_mapping,
+					    start_pos >> PAGE_SHIFT,
+					    end_pos >> PAGE_SHIFT);
+	if (ret)
+		pr_debug("fuse: invalidate_inode_pages2_range() failed err=%d\n",
+			 ret);
+
+	return ret;
+}
+
+static int reclaim_one_dmap_locked(struct inode *inode,
+				   struct fuse_dax_mapping *dmap)
+{
+	int ret;
+	struct fuse_inode *fi = get_fuse_inode(inode);
+
+	/*
+	 * igrab() was done to make sure inode won't go under us, and this
+	 * further avoids the race with evict().
+	 */
+	ret = dmap_writeback_invalidate(inode, dmap);
+	if (ret)
+		return ret;
+
+	/* Remove dax mapping from inode interval tree now */
+	interval_tree_remove(&dmap->itn, &fi->dax->tree);
+	fi->dax->nr--;
+
+	/* It is possible that umount/shutdown has killed the fuse connection
+	 * and worker thread is trying to reclaim memory in parallel.  Don't
+	 * warn in that case.
+	 */
+	ret = dmap_removemapping_one(inode, dmap);
+	if (ret && ret != -ENOTCONN) {
+		pr_warn("Failed to remove mapping. offset=0x%llx len=0x%llx ret=%d\n",
+			dmap->window_offset, dmap->length, ret);
+	}
+	return 0;
+}
+
+/* Find first mapped dmap for an inode and return file offset. Caller needs
+ * to hold fi->dax->sem lock either shared or exclusive.
+ */
+static struct fuse_dax_mapping *inode_lookup_first_dmap(struct inode *inode)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_dax_mapping *dmap;
+	struct interval_tree_node *node;
+
+	for (node = interval_tree_iter_first(&fi->dax->tree, 0, -1); node;
+	     node = interval_tree_iter_next(node, 0, -1)) {
+		dmap = node_to_dmap(node);
+		/* still in use. */
+		if (refcount_read(&dmap->refcnt) > 1)
+			continue;
+
+		return dmap;
+	}
+
+	return NULL;
+}
+
+/*
+ * Find first mapping in the tree and free it and return it. Do not add
+ * it back to free pool.
+ */
+static struct fuse_dax_mapping *
+inode_inline_reclaim_one_dmap(struct fuse_conn_dax *fcd, struct inode *inode,
+			      bool *retry)
+{
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_dax_mapping *dmap;
+	u64 dmap_start, dmap_end;
+	unsigned long start_idx;
+	int ret;
+	struct interval_tree_node *node;
+
+	down_write(&fi->i_mmap_sem);
+
+	/* Lookup a dmap and corresponding file offset to reclaim. */
+	down_read(&fi->dax->sem);
+	dmap = inode_lookup_first_dmap(inode);
+	if (dmap) {
+		start_idx = dmap->itn.start;
+		dmap_start = start_idx << FUSE_DAX_SHIFT;
+		dmap_end = dmap_start + FUSE_DAX_SZ - 1;
+	}
+	up_read(&fi->dax->sem);
+
+	if (!dmap)
+		goto out_mmap_sem;
+	/*
+	 * Make sure there are no references to inode pages using
+	 * get_user_pages()
+	 */
+	ret = fuse_dax_break_layouts(inode, dmap_start, dmap_end);
+	if (ret) {
+		pr_debug("fuse: fuse_dax_break_layouts() failed. err=%d\n",
+			 ret);
+		dmap = ERR_PTR(ret);
+		goto out_mmap_sem;
+	}
+
+	down_write(&fi->dax->sem);
+	node = interval_tree_iter_first(&fi->dax->tree, start_idx, start_idx);
+	/* Range already got reclaimed by somebody else */
+	if (!node) {
+		if (retry)
+			*retry = true;
+		goto out_write_dmap_sem;
+	}
+
+	dmap = node_to_dmap(node);
+	/* still in use. */
+	if (refcount_read(&dmap->refcnt) > 1) {
+		dmap = NULL;
+		if (retry)
+			*retry = true;
+		goto out_write_dmap_sem;
+	}
+
+	ret = reclaim_one_dmap_locked(inode, dmap);
+	if (ret < 0) {
+		dmap = ERR_PTR(ret);
+		goto out_write_dmap_sem;
+	}
+
+	/* Clean up dmap. Do not add back to free list */
+	dmap_remove_busy_list(fcd, dmap);
+	dmap->inode = NULL;
+	dmap->itn.start = dmap->itn.last = 0;
+
+	pr_debug("fuse: %s: inline reclaimed memory range. inode=%p, window_offset=0x%llx, length=0x%llx\n",
+		 __func__, inode, dmap->window_offset, dmap->length);
+
+out_write_dmap_sem:
+	up_write(&fi->dax->sem);
+out_mmap_sem:
+	up_write(&fi->i_mmap_sem);
+	return dmap;
+}
+
+static struct fuse_dax_mapping *
+alloc_dax_mapping_reclaim(struct fuse_conn_dax *fcd, struct inode *inode)
+{
+	struct fuse_dax_mapping *dmap;
+	struct fuse_inode *fi = get_fuse_inode(inode);
+
+	while (1) {
+		bool retry = false;
+
+		dmap = alloc_dax_mapping(fcd);
+		if (dmap)
+			return dmap;
+
+		dmap = inode_inline_reclaim_one_dmap(fcd, inode, &retry);
+		/*
+		 * Either we got a mapping or it is an error, return in both
+		 * the cases.
+		 */
+		if (dmap)
+			return dmap;
+
+		/* If we could not reclaim a mapping because it
+		 * had a reference or some other temporary failure,
+		 * Try again. We want to give up inline reclaim only
+		 * if there is no range assigned to this node. Otherwise
+		 * if a deadlock is possible if we sleep with fi->i_mmap_sem
+		 * held and worker to free memory can't make progress due
+		 * to unavailability of fi->i_mmap_sem lock. So sleep
+		 * only if fi->dax->nr=0
+		 */
+		if (retry)
+			continue;
+		/*
+		 * There are no mappings which can be reclaimed. Wait for one.
+		 * We are not holding fi->dax->sem. So it is possible
+		 * that range gets added now. But as we are not holding
+		 * fi->i_mmap_sem, worker should still be able to free up
+		 * a range and wake us up.
+		 */
+		if (!fi->dax->nr && !(fcd->nr_free_ranges > 0)) {
+			if (wait_event_killable_exclusive(fcd->range_waitq,
+					(fcd->nr_free_ranges > 0))) {
+				return ERR_PTR(-EINTR);
+			}
+		}
+	}
+}
+
+static int lookup_and_reclaim_dmap_locked(struct fuse_conn_dax *fcd,
+					  struct inode *inode,
+					  unsigned long start_idx)
+{
+	int ret;
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	struct fuse_dax_mapping *dmap;
+	struct interval_tree_node *node;
+
+	/* Find fuse dax mapping at file offset inode. */
+	node = interval_tree_iter_first(&fi->dax->tree, start_idx, start_idx);
+
+	/* Range already got cleaned up by somebody else */
+	if (!node)
+		return 0;
+	dmap = node_to_dmap(node);
+
+	/* still in use. */
+	if (refcount_read(&dmap->refcnt) > 1)
+		return 0;
+
+	ret = reclaim_one_dmap_locked(inode, dmap);
+	if (ret < 0)
+		return ret;
+
+	/* Cleanup dmap entry and add back to free list */
+	spin_lock(&fcd->lock);
+	dmap_reinit_add_to_free_pool(fcd, dmap);
+	spin_unlock(&fcd->lock);
+	return ret;
+}
+
+/*
+ * Free a range of memory.
+ * Locking:
+ * 1. Take fi->i_mmap_sem to block dax faults.
+ * 2. Take fi->dax->sem to protect interval tree and also to make sure
+ *    read/write can not reuse a dmap which we might be freeing.
+ */
+static int lookup_and_reclaim_dmap(struct fuse_conn_dax *fcd,
+				   struct inode *inode,
+				   unsigned long start_idx,
+				   unsigned long end_idx)
+{
+	int ret;
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	loff_t dmap_start = start_idx << FUSE_DAX_SHIFT;
+	loff_t dmap_end = (dmap_start + FUSE_DAX_SZ) - 1;
+
+	down_write(&fi->i_mmap_sem);
+	ret = fuse_dax_break_layouts(inode, dmap_start, dmap_end);
+	if (ret) {
+		pr_debug("virtio_fs: fuse_dax_break_layouts() failed. err=%d\n",
+			 ret);
+		goto out_mmap_sem;
+	}
+
+	down_write(&fi->dax->sem);
+	ret = lookup_and_reclaim_dmap_locked(fcd, inode, start_idx);
+	up_write(&fi->dax->sem);
+out_mmap_sem:
+	up_write(&fi->i_mmap_sem);
+	return ret;
+}
+
+static int try_to_free_dmap_chunks(struct fuse_conn_dax *fcd,
+				   unsigned long nr_to_free)
+{
+	struct fuse_dax_mapping *dmap, *pos, *temp;
+	int ret, nr_freed = 0;
+	unsigned long start_idx = 0, end_idx = 0;
+	struct inode *inode = NULL;
+
+	/* Pick first busy range and free it for now*/
+	while (1) {
+		if (nr_freed >= nr_to_free)
+			break;
+
+		dmap = NULL;
+		spin_lock(&fcd->lock);
+
+		if (!fcd->nr_busy_ranges) {
+			spin_unlock(&fcd->lock);
+			return 0;
+		}
+
+		list_for_each_entry_safe(pos, temp, &fcd->busy_ranges,
+						busy_list) {
+			/* skip this range if it's in use. */
+			if (refcount_read(&pos->refcnt) > 1)
+				continue;
+
+			inode = igrab(pos->inode);
+			/*
+			 * This inode is going away. That will free
+			 * up all the ranges anyway, continue to
+			 * next range.
+			 */
+			if (!inode)
+				continue;
+			/*
+			 * Take this element off list and add it tail. If
+			 * this element can't be freed, it will help with
+			 * selecting new element in next iteration of loop.
+			 */
+			dmap = pos;
+			list_move_tail(&dmap->busy_list, &fcd->busy_ranges);
+			start_idx = end_idx = dmap->itn.start;
+			break;
+		}
+		spin_unlock(&fcd->lock);
+		if (!dmap)
+			return 0;
+
+		ret = lookup_and_reclaim_dmap(fcd, inode, start_idx, end_idx);
+		iput(inode);
+		if (ret)
+			return ret;
+		nr_freed++;
+	}
+	return 0;
+}
+
+static void fuse_dax_free_mem_worker(struct work_struct *work)
+{
+	int ret;
+	struct fuse_conn_dax *fcd = container_of(work, struct fuse_conn_dax,
+						 free_work.work);
+	ret = try_to_free_dmap_chunks(fcd, FUSE_DAX_RECLAIM_CHUNK);
+	if (ret) {
+		pr_debug("fuse: try_to_free_dmap_chunks() failed with err=%d\n",
+			 ret);
+	}
+
+	/* If number of free ranges are still below threhold, requeue */
+	kick_dmap_free_worker(fcd, 1);
+}
+
+static void fuse_free_dax_mem_ranges(struct list_head *mem_list)
+{
+	struct fuse_dax_mapping *range, *temp;
+
+	/* Free All allocated elements */
+	list_for_each_entry_safe(range, temp, mem_list, list) {
+		list_del(&range->list);
+		if (!list_empty(&range->busy_list))
+			list_del(&range->busy_list);
+		kfree(range);
+	}
+}
+
+void fuse_dax_conn_free(struct fuse_conn *fc)
+{
+	if (fc->dax) {
+		fuse_free_dax_mem_ranges(&fc->dax->free_ranges);
+		kfree(fc->dax);
+	}
+}
+
+static int fuse_dax_mem_range_init(struct fuse_conn_dax *fcd)
+{
+	long nr_pages, nr_ranges;
+	void *kaddr;
+	pfn_t pfn;
+	struct fuse_dax_mapping *range;
+	int ret, id;
+	size_t dax_size = -1;
+	unsigned long i;
+
+	init_waitqueue_head(&fcd->range_waitq);
+	INIT_LIST_HEAD(&fcd->free_ranges);
+	INIT_LIST_HEAD(&fcd->busy_ranges);
+	INIT_DELAYED_WORK(&fcd->free_work, fuse_dax_free_mem_worker);
+
+	id = dax_read_lock();
+	nr_pages = dax_direct_access(fcd->dev, 0, PHYS_PFN(dax_size), &kaddr,
+				     &pfn);
+	dax_read_unlock(id);
+	if (nr_pages < 0) {
+		pr_debug("dax_direct_access() returned %ld\n", nr_pages);
+		return nr_pages;
+	}
+
+	nr_ranges = nr_pages/FUSE_DAX_PAGES;
+	pr_debug("%s: dax mapped %ld pages. nr_ranges=%ld\n",
+		__func__, nr_pages, nr_ranges);
+
+	for (i = 0; i < nr_ranges; i++) {
+		range = kzalloc(sizeof(struct fuse_dax_mapping), GFP_KERNEL);
+		ret = -ENOMEM;
+		if (!range)
+			goto out_err;
+
+		/* TODO: This offset only works if virtio-fs driver is not
+		 * having some memory hidden at the beginning. This needs
+		 * better handling
+		 */
+		range->window_offset = i * FUSE_DAX_SZ;
+		range->length = FUSE_DAX_SZ;
+		INIT_LIST_HEAD(&range->busy_list);
+		refcount_set(&range->refcnt, 1);
+		list_add_tail(&range->list, &fcd->free_ranges);
+	}
+
+	fcd->nr_free_ranges = nr_ranges;
+	fcd->nr_ranges = nr_ranges;
+	return 0;
+out_err:
+	/* Free All allocated elements */
+	fuse_free_dax_mem_ranges(&fcd->free_ranges);
+	return ret;
+}
+
+int fuse_dax_conn_alloc(struct fuse_conn *fc, struct dax_device *dax_dev)
+{
+	struct fuse_conn_dax *fcd;
+	int err;
+
+	if (!dax_dev)
+		return 0;
+
+	fcd = kzalloc(sizeof(*fcd), GFP_KERNEL);
+	if (!fcd)
+		return -ENOMEM;
+
+	spin_lock_init(&fcd->lock);
+	fcd->dev = dax_dev;
+	err = fuse_dax_mem_range_init(fcd);
+	if (err) {
+		kfree(fcd);
+		return err;
+	}
+
+	fc->dax = fcd;
+	return 0;
+}
+
+bool fuse_dax_inode_alloc(struct super_block *sb, struct fuse_inode *fi)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+
+	fi->dax = NULL;
+	if (fc->dax) {
+		fi->dax = kzalloc(sizeof(*fi->dax), GFP_KERNEL_ACCOUNT);
+		if (!fi->dax)
+			return false;
+
+		init_rwsem(&fi->dax->sem);
+		fi->dax->tree = RB_ROOT_CACHED;
+	}
+
+	return true;
+}
+
+static const struct address_space_operations fuse_dax_file_aops  = {
+	.writepages	= fuse_dax_writepages,
+	.direct_IO	= noop_direct_IO,
+	.set_page_dirty	= noop_set_page_dirty,
+	.invalidatepage	= noop_invalidatepage,
+};
+
+void fuse_dax_inode_init(struct inode *inode)
+{
+	struct fuse_conn *fc = get_fuse_conn(inode);
+
+	if (!fc->dax)
+		return;
+
+	inode->i_flags |= S_DAX;
+	inode->i_data.a_ops = &fuse_dax_file_aops;
+}
+
+bool fuse_dax_check_alignment(struct fuse_conn *fc, unsigned int map_alignment)
+{
+	if (fc->dax && (map_alignment > FUSE_DAX_SHIFT)) {
+		pr_warn("FUSE: map_alignment %u incompatible with dax mem range size %u\n",
+			map_alignment, FUSE_DAX_SZ);
+		return false;
+	}
+	return true;
+}
+
+void fuse_dax_cancel_work(struct fuse_conn *fc)
+{
+	struct fuse_conn_dax *fcd = fc->dax;
+
+	if (fcd)
+		cancel_delayed_work_sync(&fcd->free_work);
+
+}
+EXPORT_SYMBOL_GPL(fuse_dax_cancel_work);

fs/fuse/dev.c

@@ -40,20 +40,21 @@ static struct fuse_dev *fuse_get_dev(struct file *file)
 	return READ_ONCE(file->private_data);
 }
 
-static void fuse_request_init(struct fuse_req *req)
+static void fuse_request_init(struct fuse_mount *fm, struct fuse_req *req)
 {
 	INIT_LIST_HEAD(&req->list);
 	INIT_LIST_HEAD(&req->intr_entry);
 	init_waitqueue_head(&req->waitq);
 	refcount_set(&req->count, 1);
 	__set_bit(FR_PENDING, &req->flags);
+	req->fm = fm;
 }
 
-static struct fuse_req *fuse_request_alloc(gfp_t flags)
+static struct fuse_req *fuse_request_alloc(struct fuse_mount *fm, gfp_t flags)
 {
 	struct fuse_req *req = kmem_cache_zalloc(fuse_req_cachep, flags);
 	if (req)
-		fuse_request_init(req);
+		fuse_request_init(fm, req);
 
 	return req;
 }
@@ -100,10 +101,11 @@ static void fuse_drop_waiting(struct fuse_conn *fc)
 	}
 }
 
-static void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req);
+static void fuse_put_request(struct fuse_req *req);
 
-static struct fuse_req *fuse_get_req(struct fuse_conn *fc, bool for_background)
+static struct fuse_req *fuse_get_req(struct fuse_mount *fm, bool for_background)
 {
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_req *req;
 	int err;
 	atomic_inc(&fc->num_waiting);
@@ -125,7 +127,7 @@ static struct fuse_req *fuse_get_req(struct fuse_conn *fc, bool for_background)
 	if (fc->conn_error)
 		goto out;
 
-	req = fuse_request_alloc(GFP_KERNEL);
+	req = fuse_request_alloc(fm, GFP_KERNEL);
 	err = -ENOMEM;
 	if (!req) {
 		if (for_background)
@@ -143,7 +145,7 @@ static struct fuse_req *fuse_get_req(struct fuse_conn *fc, bool for_background)
 
 	if (unlikely(req->in.h.uid == ((uid_t)-1) ||
 		     req->in.h.gid == ((gid_t)-1))) {
-		fuse_put_request(fc, req);
+		fuse_put_request(req);
 		return ERR_PTR(-EOVERFLOW);
 	}
 	return req;
@@ -153,8 +155,10 @@ static struct fuse_req *fuse_get_req(struct fuse_conn *fc, bool for_background)
 	return ERR_PTR(err);
 }
 
-static void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req)
+static void fuse_put_request(struct fuse_req *req)
 {
+	struct fuse_conn *fc = req->fm->fc;
+
 	if (refcount_dec_and_test(&req->count)) {
 		if (test_bit(FR_BACKGROUND, &req->flags)) {
 			/*
@@ -273,8 +277,10 @@ static void flush_bg_queue(struct fuse_conn *fc)
  * the 'end' callback is called if given, else the reference to the
  * request is released
  */
-void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req)
+void fuse_request_end(struct fuse_req *req)
 {
+	struct fuse_mount *fm = req->fm;
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_iqueue *fiq = &fc->iq;
 
 	if (test_and_set_bit(FR_FINISHED, &req->flags))
@@ -309,9 +315,9 @@ void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req)
 				wake_up(&fc->blocked_waitq);
 		}
 
-		if (fc->num_background == fc->congestion_threshold && fc->sb) {
-			clear_bdi_congested(fc->sb->s_bdi, BLK_RW_SYNC);
-			clear_bdi_congested(fc->sb->s_bdi, BLK_RW_ASYNC);
+		if (fc->num_background == fc->congestion_threshold && fm->sb) {
+			clear_bdi_congested(fm->sb->s_bdi, BLK_RW_SYNC);
+			clear_bdi_congested(fm->sb->s_bdi, BLK_RW_ASYNC);
 		}
 		fc->num_background--;
 		fc->active_background--;
@@ -323,14 +329,16 @@ void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req)
 	}
 
 	if (test_bit(FR_ASYNC, &req->flags))
-		req->args->end(fc, req->args, req->out.h.error);
+		req->args->end(fm, req->args, req->out.h.error);
 put_request:
-	fuse_put_request(fc, req);
+	fuse_put_request(req);
 }
 EXPORT_SYMBOL_GPL(fuse_request_end);
 
-static int queue_interrupt(struct fuse_iqueue *fiq, struct fuse_req *req)
+static int queue_interrupt(struct fuse_req *req)
 {
+	struct fuse_iqueue *fiq = &req->fm->fc->iq;
+
 	spin_lock(&fiq->lock);
 	/* Check for we've sent request to interrupt this req */
 	if (unlikely(!test_bit(FR_INTERRUPTED, &req->flags))) {
@@ -357,8 +365,9 @@ static int queue_interrupt(struct fuse_iqueue *fiq, struct fuse_req *req)
 	return 0;
 }
 
-static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req)
+static void request_wait_answer(struct fuse_req *req)
 {
+	struct fuse_conn *fc = req->fm->fc;
 	struct fuse_iqueue *fiq = &fc->iq;
 	int err;
 
@@ -373,7 +382,7 @@ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req)
 		/* matches barrier in fuse_dev_do_read() */
 		smp_mb__after_atomic();
 		if (test_bit(FR_SENT, &req->flags))
-			queue_interrupt(fiq, req);
+			queue_interrupt(req);
 	}
 
 	if (!test_bit(FR_FORCE, &req->flags)) {
@@ -402,9 +411,9 @@ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req)
 	wait_event(req->waitq, test_bit(FR_FINISHED, &req->flags));
 }
 
-static void __fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
+static void __fuse_request_send(struct fuse_req *req)
 {
-	struct fuse_iqueue *fiq = &fc->iq;
+	struct fuse_iqueue *fiq = &req->fm->fc->iq;
 
 	BUG_ON(test_bit(FR_BACKGROUND, &req->flags));
 	spin_lock(&fiq->lock);
@@ -418,7 +427,7 @@ static void __fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
 		__fuse_get_request(req);
 		queue_request_and_unlock(fiq, req);
 
-		request_wait_answer(fc, req);
+		request_wait_answer(req);
 		/* Pairs with smp_wmb() in fuse_request_end() */
 		smp_rmb();
 	}
@@ -457,8 +466,10 @@ static void fuse_adjust_compat(struct fuse_conn *fc, struct fuse_args *args)
 	}
 }
 
-static void fuse_force_creds(struct fuse_conn *fc, struct fuse_req *req)
+static void fuse_force_creds(struct fuse_req *req)
 {
+	struct fuse_conn *fc = req->fm->fc;
+
 	req->in.h.uid = from_kuid_munged(fc->user_ns, current_fsuid());
 	req->in.h.gid = from_kgid_munged(fc->user_ns, current_fsgid());
 	req->in.h.pid = pid_nr_ns(task_pid(current), fc->pid_ns);
@@ -473,23 +484,24 @@ static void fuse_args_to_req(struct fuse_req *req, struct fuse_args *args)
 		__set_bit(FR_ASYNC, &req->flags);
 }
 
-ssize_t fuse_simple_request(struct fuse_conn *fc, struct fuse_args *args)
+ssize_t fuse_simple_request(struct fuse_mount *fm, struct fuse_args *args)
 {
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_req *req;
 	ssize_t ret;
 
 	if (args->force) {
 		atomic_inc(&fc->num_waiting);
-		req = fuse_request_alloc(GFP_KERNEL | __GFP_NOFAIL);
+		req = fuse_request_alloc(fm, GFP_KERNEL | __GFP_NOFAIL);
 
 		if (!args->nocreds)
-			fuse_force_creds(fc, req);
+			fuse_force_creds(req);
 
 		__set_bit(FR_WAITING, &req->flags);
 		__set_bit(FR_FORCE, &req->flags);
 	} else {
 		WARN_ON(args->nocreds);
-		req = fuse_get_req(fc, false);
+		req = fuse_get_req(fm, false);
 		if (IS_ERR(req))
 			return PTR_ERR(req);
 	}
@@ -500,20 +512,21 @@ ssize_t fuse_simple_request(struct fuse_conn *fc, struct fuse_args *args)
 
 	if (!args->noreply)
 		__set_bit(FR_ISREPLY, &req->flags);
-	__fuse_request_send(fc, req);
+	__fuse_request_send(req);
 	ret = req->out.h.error;
 	if (!ret && args->out_argvar) {
 		BUG_ON(args->out_numargs == 0);
 		ret = args->out_args[args->out_numargs - 1].size;
 	}
-	fuse_put_request(fc, req);
+	fuse_put_request(req);
 
 	return ret;
 }
 
-static bool fuse_request_queue_background(struct fuse_conn *fc,
-					  struct fuse_req *req)
+static bool fuse_request_queue_background(struct fuse_req *req)
 {
+	struct fuse_mount *fm = req->fm;
+	struct fuse_conn *fc = fm->fc;
 	bool queued = false;
 
 	WARN_ON(!test_bit(FR_BACKGROUND, &req->flags));
@@ -527,9 +540,9 @@ static bool fuse_request_queue_background(struct fuse_conn *fc,
 		fc->num_background++;
 		if (fc->num_background == fc->max_background)
 			fc->blocked = 1;
-		if (fc->num_background == fc->congestion_threshold && fc->sb) {
-			set_bdi_congested(fc->sb->s_bdi, BLK_RW_SYNC);
-			set_bdi_congested(fc->sb->s_bdi, BLK_RW_ASYNC);
+		if (fc->num_background == fc->congestion_threshold && fm->sb) {
+			set_bdi_congested(fm->sb->s_bdi, BLK_RW_SYNC);
+			set_bdi_congested(fm->sb->s_bdi, BLK_RW_ASYNC);
 		}
 		list_add_tail(&req->list, &fc->bg_queue);
 		flush_bg_queue(fc);
@@ -540,28 +553,28 @@ static bool fuse_request_queue_background(struct fuse_conn *fc,
 	return queued;
 }
 
-int fuse_simple_background(struct fuse_conn *fc, struct fuse_args *args,
+int fuse_simple_background(struct fuse_mount *fm, struct fuse_args *args,
 			    gfp_t gfp_flags)
 {
 	struct fuse_req *req;
 
 	if (args->force) {
 		WARN_ON(!args->nocreds);
-		req = fuse_request_alloc(gfp_flags);
+		req = fuse_request_alloc(fm, gfp_flags);
 		if (!req)
 			return -ENOMEM;
 		__set_bit(FR_BACKGROUND, &req->flags);
 	} else {
 		WARN_ON(args->nocreds);
-		req = fuse_get_req(fc, true);
+		req = fuse_get_req(fm, true);
 		if (IS_ERR(req))
 			return PTR_ERR(req);
 	}
 
 	fuse_args_to_req(req, args);
 
-	if (!fuse_request_queue_background(fc, req)) {
-		fuse_put_request(fc, req);
+	if (!fuse_request_queue_background(req)) {
+		fuse_put_request(req);
 		return -ENOTCONN;
 	}
 
@@ -569,14 +582,14 @@ int fuse_simple_background(struct fuse_conn *fc, struct fuse_args *args,
 }
 EXPORT_SYMBOL_GPL(fuse_simple_background);
 
-static int fuse_simple_notify_reply(struct fuse_conn *fc,
+static int fuse_simple_notify_reply(struct fuse_mount *fm,
 				    struct fuse_args *args, u64 unique)
 {
 	struct fuse_req *req;
-	struct fuse_iqueue *fiq = &fc->iq;
+	struct fuse_iqueue *fiq = &fm->fc->iq;
 	int err = 0;
 
-	req = fuse_get_req(fc, false);
+	req = fuse_get_req(fm, false);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -591,7 +604,7 @@ static int fuse_simple_notify_reply(struct fuse_conn *fc,
 	} else {
 		err = -ENODEV;
 		spin_unlock(&fiq->lock);
-		fuse_put_request(fc, req);
+		fuse_put_request(req);
 	}
 
 	return err;
@@ -785,15 +798,16 @@ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep)
 	struct page *newpage;
 	struct pipe_buffer *buf = cs->pipebufs;
 
+	get_page(oldpage);
 	err = unlock_request(cs->req);
 	if (err)
-		return err;
+		goto out_put_old;
 
 	fuse_copy_finish(cs);
 
 	err = pipe_buf_confirm(cs->pipe, buf);
 	if (err)
-		return err;
+		goto out_put_old;
 
 	BUG_ON(!cs->nr_segs);
 	cs->currbuf = buf;
@@ -833,7 +847,7 @@ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep)
 	err = replace_page_cache_page(oldpage, newpage, GFP_KERNEL);
 	if (err) {
 		unlock_page(newpage);
-		return err;
+		goto out_put_old;
 	}
 
 	get_page(newpage);
@@ -852,14 +866,19 @@ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep)
 	if (err) {
 		unlock_page(newpage);
 		put_page(newpage);
-		return err;
+		goto out_put_old;
 	}
 
 	unlock_page(oldpage);
+	/* Drop ref for ap->pages[] array */
 	put_page(oldpage);
 	cs->len = 0;
 
-	return 0;
+	err = 0;
+out_put_old:
+	/* Drop ref obtained in this function */
+	put_page(oldpage);
+	return err;
 
 out_fallback_unlock:
 	unlock_page(newpage);
@@ -868,10 +887,10 @@ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep)
 	cs->offset = buf->offset;
 
 	err = lock_request(cs->req);
-	if (err)
-		return err;
+	if (!err)
+		err = 1;
 
-	return 1;
+	goto out_put_old;
 }
 
 static int fuse_ref_page(struct fuse_copy_state *cs, struct page *page,
@@ -883,14 +902,16 @@ static int fuse_ref_page(struct fuse_copy_state *cs, struct page *page,
 	if (cs->nr_segs >= cs->pipe->max_usage)
 		return -EIO;
 
+	get_page(page);
 	err = unlock_request(cs->req);
-	if (err)
+	if (err) {
+		put_page(page);
 		return err;
+	}
 
 	fuse_copy_finish(cs);
 
 	buf = cs->pipebufs;
-	get_page(page);
 	buf->page = page;
 	buf->offset = offset;
 	buf->len = count;
@@ -1250,7 +1271,7 @@ static ssize_t fuse_dev_do_read(struct fuse_dev *fud, struct file *file,
 		/* SETXATTR is special, since it may contain too large data */
 		if (args->opcode == FUSE_SETXATTR)
 			req->out.h.error = -E2BIG;
-		fuse_request_end(fc, req);
+		fuse_request_end(req);
 		goto restart;
 	}
 	spin_lock(&fpq->lock);
@@ -1284,8 +1305,8 @@ static ssize_t fuse_dev_do_read(struct fuse_dev *fud, struct file *file,
 	/* matches barrier in request_wait_answer() */
 	smp_mb__after_atomic();
 	if (test_bit(FR_INTERRUPTED, &req->flags))
-		queue_interrupt(fiq, req);
-	fuse_put_request(fc, req);
+		queue_interrupt(req);
+	fuse_put_request(req);
 
 	return reqsize;
 
@@ -1293,7 +1314,7 @@ static ssize_t fuse_dev_do_read(struct fuse_dev *fud, struct file *file,
 	if (!test_bit(FR_PRIVATE, &req->flags))
 		list_del_init(&req->list);
 	spin_unlock(&fpq->lock);
-	fuse_request_end(fc, req);
+	fuse_request_end(req);
 	return err;
 
  err_unlock:
@@ -1416,11 +1437,8 @@ static int fuse_notify_inval_inode(struct fuse_conn *fc, unsigned int size,
 	fuse_copy_finish(cs);
 
 	down_read(&fc->killsb);
-	err = -ENOENT;
-	if (fc->sb) {
-		err = fuse_reverse_inval_inode(fc->sb, outarg.ino,
+	err = fuse_reverse_inval_inode(fc, outarg.ino,
 				       outarg.off, outarg.len);
-	}
 	up_read(&fc->killsb);
 	return err;
 
@@ -1466,9 +1484,7 @@ static int fuse_notify_inval_entry(struct fuse_conn *fc, unsigned int size,
 	buf[outarg.namelen] = 0;
 
 	down_read(&fc->killsb);
-	err = -ENOENT;
-	if (fc->sb)
-		err = fuse_reverse_inval_entry(fc->sb, outarg.parent, 0, &name);
+	err = fuse_reverse_inval_entry(fc, outarg.parent, 0, &name);
 	up_read(&fc->killsb);
 	kfree(buf);
 	return err;
@@ -1516,10 +1532,7 @@ static int fuse_notify_delete(struct fuse_conn *fc, unsigned int size,
 	buf[outarg.namelen] = 0;
 
 	down_read(&fc->killsb);
-	err = -ENOENT;
-	if (fc->sb)
-		err = fuse_reverse_inval_entry(fc->sb, outarg.parent,
-					       outarg.child, &name);
+	err = fuse_reverse_inval_entry(fc, outarg.parent, outarg.child, &name);
 	up_read(&fc->killsb);
 	kfree(buf);
 	return err;
@@ -1561,10 +1574,7 @@ static int fuse_notify_store(struct fuse_conn *fc, unsigned int size,
 	down_read(&fc->killsb);
 
 	err = -ENOENT;
-	if (!fc->sb)
-		goto out_up_killsb;
-
-	inode = ilookup5(fc->sb, nodeid, fuse_inode_eq, &nodeid);
+	inode = fuse_ilookup(fc, nodeid,  NULL);
 	if (!inode)
 		goto out_up_killsb;
 
@@ -1621,7 +1631,7 @@ struct fuse_retrieve_args {
 	struct fuse_notify_retrieve_in inarg;
 };
 
-static void fuse_retrieve_end(struct fuse_conn *fc, struct fuse_args *args,
+static void fuse_retrieve_end(struct fuse_mount *fm, struct fuse_args *args,
 			      int error)
 {
 	struct fuse_retrieve_args *ra =
@@ -1631,7 +1641,7 @@ static void fuse_retrieve_end(struct fuse_conn *fc, struct fuse_args *args,
 	kfree(ra);
 }
 
-static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode,
+static int fuse_retrieve(struct fuse_mount *fm, struct inode *inode,
 			 struct fuse_notify_retrieve_out *outarg)
 {
 	int err;
@@ -1642,6 +1652,7 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode,
 	unsigned int offset;
 	size_t total_len = 0;
 	unsigned int num_pages;
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_retrieve_args *ra;
 	size_t args_size = sizeof(*ra);
 	struct fuse_args_pages *ap;
@@ -1703,9 +1714,9 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode,
 	args->in_args[0].value = &ra->inarg;
 	args->in_args[1].size = total_len;
 
-	err = fuse_simple_notify_reply(fc, args, outarg->notify_unique);
+	err = fuse_simple_notify_reply(fm, args, outarg->notify_unique);
 	if (err)
-		fuse_retrieve_end(fc, args, err);
+		fuse_retrieve_end(fm, args, err);
 
 	return err;
 }
@@ -1714,7 +1725,9 @@ static int fuse_notify_retrieve(struct fuse_conn *fc, unsigned int size,
 				struct fuse_copy_state *cs)
 {
 	struct fuse_notify_retrieve_out outarg;
+	struct fuse_mount *fm;
 	struct inode *inode;
+	u64 nodeid;
 	int err;
 
 	err = -EINVAL;
@@ -1729,15 +1742,13 @@ static int fuse_notify_retrieve(struct fuse_conn *fc, unsigned int size,
 
 	down_read(&fc->killsb);
 	err = -ENOENT;
-	if (fc->sb) {
-		u64 nodeid = outarg.nodeid;
+	nodeid = outarg.nodeid;
 
-		inode = ilookup5(fc->sb, nodeid, fuse_inode_eq, &nodeid);
+	inode = fuse_ilookup(fc, nodeid, &fm);
 	if (inode) {
-			err = fuse_retrieve(fc, inode, &outarg);
+		err = fuse_retrieve(fm, inode, &outarg);
 		iput(inode);
 	}
-	}
 	up_read(&fc->killsb);
 
 	return err;
@@ -1875,9 +1886,9 @@ static ssize_t fuse_dev_do_write(struct fuse_dev *fud,
 		else if (oh.error == -ENOSYS)
 			fc->no_interrupt = 1;
 		else if (oh.error == -EAGAIN)
-			err = queue_interrupt(&fc->iq, req);
+			err = queue_interrupt(req);
 
-		fuse_put_request(fc, req);
+		fuse_put_request(req);
 
 		goto copy_finish;
 	}
@@ -1907,7 +1918,7 @@ static ssize_t fuse_dev_do_write(struct fuse_dev *fud,
 		list_del_init(&req->list);
 	spin_unlock(&fpq->lock);
 
-	fuse_request_end(fc, req);
+	fuse_request_end(req);
 out:
 	return err ? err : nbytes;
 
@@ -2045,7 +2056,7 @@ static __poll_t fuse_dev_poll(struct file *file, poll_table *wait)
 }
 
 /* Abort all requests on the given list (pending or processing) */
-static void end_requests(struct fuse_conn *fc, struct list_head *head)
+static void end_requests(struct list_head *head)
 {
 	while (!list_empty(head)) {
 		struct fuse_req *req;
@@ -2053,7 +2064,7 @@ static void end_requests(struct fuse_conn *fc, struct list_head *head)
 		req->out.h.error = -ECONNABORTED;
 		clear_bit(FR_SENT, &req->flags);
 		list_del_init(&req->list);
-		fuse_request_end(fc, req);
+		fuse_request_end(req);
 	}
 }
 
@@ -2148,7 +2159,7 @@ void fuse_abort_conn(struct fuse_conn *fc)
 		wake_up_all(&fc->blocked_waitq);
 		spin_unlock(&fc->lock);
 
-		end_requests(fc, &to_end);
+		end_requests(&to_end);
 	} else {
 		spin_unlock(&fc->lock);
 	}
@@ -2178,7 +2189,7 @@ int fuse_dev_release(struct inode *inode, struct file *file)
 			list_splice_init(&fpq->processing[i], &to_end);
 		spin_unlock(&fpq->lock);
 
-		end_requests(fc, &to_end);
+		end_requests(&to_end);
 
 		/* Are we the last open device? */
 		if (atomic_dec_and_test(&fc->dev_count)) {

fs/fuse/dir.c

@@ -10,6 +10,7 @@
 
 #include <linux/pagemap.h>
 #include <linux/file.h>
+#include <linux/fs_context.h>
 #include <linux/sched.h>
 #include <linux/namei.h>
 #include <linux/slab.h>
@@ -196,7 +197,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
 {
 	struct inode *inode;
 	struct dentry *parent;
-	struct fuse_conn *fc;
+	struct fuse_mount *fm;
 	struct fuse_inode *fi;
 	int ret;
 
@@ -218,27 +219,29 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
 		if (flags & LOOKUP_RCU)
 			goto out;
 
-		fc = get_fuse_conn(inode);
+		fm = get_fuse_mount(inode);
 
 		forget = fuse_alloc_forget();
 		ret = -ENOMEM;
 		if (!forget)
 			goto out;
 
-		attr_version = fuse_get_attr_version(fc);
+		attr_version = fuse_get_attr_version(fm->fc);
 
 		parent = dget_parent(entry);
-		fuse_lookup_init(fc, &args, get_node_id(d_inode(parent)),
+		fuse_lookup_init(fm->fc, &args, get_node_id(d_inode(parent)),
 				 &entry->d_name, &outarg);
-		ret = fuse_simple_request(fc, &args);
+		ret = fuse_simple_request(fm, &args);
 		dput(parent);
 		/* Zero nodeid is same as -ENOENT */
 		if (!ret && !outarg.nodeid)
 			ret = -ENOENT;
 		if (!ret) {
 			fi = get_fuse_inode(inode);
-			if (outarg.nodeid != get_node_id(inode)) {
-				fuse_queue_forget(fc, forget, outarg.nodeid, 1);
+			if (outarg.nodeid != get_node_id(inode) ||
+			    (bool) IS_AUTOMOUNT(inode) != (bool) (outarg.attr.flags & FUSE_ATTR_SUBMOUNT)) {
+				fuse_queue_forget(fm->fc, forget,
+						  outarg.nodeid, 1);
 				goto invalid;
 			}
 			spin_lock(&fi->lock);
@@ -298,6 +301,79 @@ static int fuse_dentry_delete(const struct dentry *dentry)
 	return time_before64(fuse_dentry_time(dentry), get_jiffies_64());
 }
 
+/*
+ * Create a fuse_mount object with a new superblock (with path->dentry
+ * as the root), and return that mount so it can be auto-mounted on
+ * @path.
+ */
+static struct vfsmount *fuse_dentry_automount(struct path *path)
+{
+	struct fs_context *fsc;
+	struct fuse_mount *parent_fm = get_fuse_mount_super(path->mnt->mnt_sb);
+	struct fuse_conn *fc = parent_fm->fc;
+	struct fuse_mount *fm;
+	struct vfsmount *mnt;
+	struct fuse_inode *mp_fi = get_fuse_inode(d_inode(path->dentry));
+	struct super_block *sb;
+	int err;
+
+	fsc = fs_context_for_submount(path->mnt->mnt_sb->s_type, path->dentry);
+	if (IS_ERR(fsc)) {
+		err = PTR_ERR(fsc);
+		goto out;
+	}
+
+	err = -ENOMEM;
+	fm = kzalloc(sizeof(struct fuse_mount), GFP_KERNEL);
+	if (!fm)
+		goto out_put_fsc;
+
+	refcount_set(&fm->count, 1);
+	fsc->s_fs_info = fm;
+	sb = sget_fc(fsc, NULL, set_anon_super_fc);
+	if (IS_ERR(sb)) {
+		err = PTR_ERR(sb);
+		fuse_mount_put(fm);
+		goto out_put_fsc;
+	}
+	fm->fc = fuse_conn_get(fc);
+
+	/* Initialize superblock, making @mp_fi its root */
+	err = fuse_fill_super_submount(sb, mp_fi);
+	if (err)
+		goto out_put_sb;
+
+	sb->s_flags |= SB_ACTIVE;
+	fsc->root = dget(sb->s_root);
+	/* We are done configuring the superblock, so unlock it */
+	up_write(&sb->s_umount);
+
+	down_write(&fc->killsb);
+	list_add_tail(&fm->fc_entry, &fc->mounts);
+	up_write(&fc->killsb);
+
+	/* Create the submount */
+	mnt = vfs_create_mount(fsc);
+	if (IS_ERR(mnt)) {
+		err = PTR_ERR(mnt);
+		goto out_put_fsc;
+	}
+	mntget(mnt);
+	put_fs_context(fsc);
+	return mnt;
+
+out_put_sb:
+	/*
+	 * Only jump here when fsc->root is NULL and sb is still locked
+	 * (otherwise put_fs_context() will put the superblock)
+	 */
+	deactivate_locked_super(sb);
+out_put_fsc:
+	put_fs_context(fsc);
+out:
+	return ERR_PTR(err);
+}
+
 const struct dentry_operations fuse_dentry_operations = {
 	.d_revalidate	= fuse_dentry_revalidate,
 	.d_delete	= fuse_dentry_delete,
@@ -305,6 +381,7 @@ const struct dentry_operations fuse_dentry_operations = {
 	.d_init		= fuse_dentry_init,
 	.d_release	= fuse_dentry_release,
 #endif
+	.d_automount	= fuse_dentry_automount,
 };
 
 const struct dentry_operations fuse_root_dentry_operations = {
@@ -329,7 +406,7 @@ bool fuse_invalid_attr(struct fuse_attr *attr)
 int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name,
 		     struct fuse_entry_out *outarg, struct inode **inode)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
 	FUSE_ARGS(args);
 	struct fuse_forget_link *forget;
 	u64 attr_version;
@@ -346,10 +423,10 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name
 	if (!forget)
 		goto out;
 
-	attr_version = fuse_get_attr_version(fc);
+	attr_version = fuse_get_attr_version(fm->fc);
 
-	fuse_lookup_init(fc, &args, nodeid, name, outarg);
-	err = fuse_simple_request(fc, &args);
+	fuse_lookup_init(fm->fc, &args, nodeid, name, outarg);
+	err = fuse_simple_request(fm, &args);
 	/* Zero nodeid is same as -ENOENT, but with valid timeout */
 	if (err || !outarg->nodeid)
 		goto out_put_forget;
@@ -365,7 +442,7 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name
 			   attr_version);
 	err = -ENOMEM;
 	if (!*inode) {
-		fuse_queue_forget(fc, forget, outarg->nodeid, 1);
+		fuse_queue_forget(fm->fc, forget, outarg->nodeid, 1);
 		goto out;
 	}
 	err = 0;
@@ -434,7 +511,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 {
 	int err;
 	struct inode *inode;
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 	struct fuse_forget_link *forget;
 	struct fuse_create_in inarg;
@@ -452,11 +529,11 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 		goto out_err;
 
 	err = -ENOMEM;
-	ff = fuse_file_alloc(fc);
+	ff = fuse_file_alloc(fm);
 	if (!ff)
 		goto out_put_forget_req;
 
-	if (!fc->dont_mask)
+	if (!fm->fc->dont_mask)
 		mode &= ~current_umask();
 
 	flags &= ~O_NOCTTY;
@@ -477,7 +554,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 	args.out_args[0].value = &outentry;
 	args.out_args[1].size = sizeof(outopen);
 	args.out_args[1].value = &outopen;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err)
 		goto out_free_ff;
 
@@ -494,7 +571,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 	if (!inode) {
 		flags &= ~(O_CREAT | O_EXCL | O_TRUNC);
 		fuse_sync_release(NULL, ff, flags);
-		fuse_queue_forget(fc, forget, outentry.nodeid, 1);
+		fuse_queue_forget(fm->fc, forget, outentry.nodeid, 1);
 		err = -ENOMEM;
 		goto out_err;
 	}
@@ -567,7 +644,7 @@ static int fuse_atomic_open(struct inode *dir, struct dentry *entry,
 /*
  * Code shared between mknod, mkdir, symlink and link
  */
-static int create_new_entry(struct fuse_conn *fc, struct fuse_args *args,
+static int create_new_entry(struct fuse_mount *fm, struct fuse_args *args,
 			    struct inode *dir, struct dentry *entry,
 			    umode_t mode)
 {
@@ -586,7 +663,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_args *args,
 	args->out_numargs = 1;
 	args->out_args[0].size = sizeof(outarg);
 	args->out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, args);
+	err = fuse_simple_request(fm, args);
 	if (err)
 		goto out_put_forget_req;
 
@@ -600,7 +677,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_args *args,
 	inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation,
 			  &outarg.attr, entry_attr_timeout(&outarg), 0);
 	if (!inode) {
-		fuse_queue_forget(fc, forget, outarg.nodeid, 1);
+		fuse_queue_forget(fm->fc, forget, outarg.nodeid, 1);
 		return -ENOMEM;
 	}
 	kfree(forget);
@@ -628,10 +705,10 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, umode_t mode,
 		      dev_t rdev)
 {
 	struct fuse_mknod_in inarg;
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
-	if (!fc->dont_mask)
+	if (!fm->fc->dont_mask)
 		mode &= ~current_umask();
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -644,7 +721,7 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, umode_t mode,
 	args.in_args[0].value = &inarg;
 	args.in_args[1].size = entry->d_name.len + 1;
 	args.in_args[1].value = entry->d_name.name;
-	return create_new_entry(fc, &args, dir, entry, mode);
+	return create_new_entry(fm, &args, dir, entry, mode);
 }
 
 static int fuse_create(struct inode *dir, struct dentry *entry, umode_t mode,
@@ -656,10 +733,10 @@ static int fuse_create(struct inode *dir, struct dentry *entry, umode_t mode,
 static int fuse_mkdir(struct inode *dir, struct dentry *entry, umode_t mode)
 {
 	struct fuse_mkdir_in inarg;
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
-	if (!fc->dont_mask)
+	if (!fm->fc->dont_mask)
 		mode &= ~current_umask();
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -671,13 +748,13 @@ static int fuse_mkdir(struct inode *dir, struct dentry *entry, umode_t mode)
 	args.in_args[0].value = &inarg;
 	args.in_args[1].size = entry->d_name.len + 1;
 	args.in_args[1].value = entry->d_name.name;
-	return create_new_entry(fc, &args, dir, entry, S_IFDIR);
+	return create_new_entry(fm, &args, dir, entry, S_IFDIR);
 }
 
 static int fuse_symlink(struct inode *dir, struct dentry *entry,
 			const char *link)
 {
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	unsigned len = strlen(link) + 1;
 	FUSE_ARGS(args);
 
@@ -687,7 +764,7 @@ static int fuse_symlink(struct inode *dir, struct dentry *entry,
 	args.in_args[0].value = entry->d_name.name;
 	args.in_args[1].size = len;
 	args.in_args[1].value = link;
-	return create_new_entry(fc, &args, dir, entry, S_IFLNK);
+	return create_new_entry(fm, &args, dir, entry, S_IFLNK);
 }
 
 void fuse_update_ctime(struct inode *inode)
@@ -701,7 +778,7 @@ void fuse_update_ctime(struct inode *inode)
 static int fuse_unlink(struct inode *dir, struct dentry *entry)
 {
 	int err;
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
 	args.opcode = FUSE_UNLINK;
@@ -709,13 +786,13 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry)
 	args.in_numargs = 1;
 	args.in_args[0].size = entry->d_name.len + 1;
 	args.in_args[0].value = entry->d_name.name;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err) {
 		struct inode *inode = d_inode(entry);
 		struct fuse_inode *fi = get_fuse_inode(inode);
 
 		spin_lock(&fi->lock);
-		fi->attr_version = atomic64_inc_return(&fc->attr_version);
+		fi->attr_version = atomic64_inc_return(&fm->fc->attr_version);
 		/*
 		 * If i_nlink == 0 then unlink doesn't make sense, yet this can
 		 * happen if userspace filesystem is careless.  It would be
@@ -737,7 +814,7 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry)
 static int fuse_rmdir(struct inode *dir, struct dentry *entry)
 {
 	int err;
-	struct fuse_conn *fc = get_fuse_conn(dir);
+	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
 	args.opcode = FUSE_RMDIR;
@@ -745,7 +822,7 @@ static int fuse_rmdir(struct inode *dir, struct dentry *entry)
 	args.in_numargs = 1;
 	args.in_args[0].size = entry->d_name.len + 1;
 	args.in_args[0].value = entry->d_name.name;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err) {
 		clear_nlink(d_inode(entry));
 		fuse_dir_changed(dir);
@@ -761,7 +838,7 @@ static int fuse_rename_common(struct inode *olddir, struct dentry *oldent,
 {
 	int err;
 	struct fuse_rename2_in inarg;
-	struct fuse_conn *fc = get_fuse_conn(olddir);
+	struct fuse_mount *fm = get_fuse_mount(olddir);
 	FUSE_ARGS(args);
 
 	memset(&inarg, 0, argsize);
@@ -776,7 +853,7 @@ static int fuse_rename_common(struct inode *olddir, struct dentry *oldent,
 	args.in_args[1].value = oldent->d_name.name;
 	args.in_args[2].size = newent->d_name.len + 1;
 	args.in_args[2].value = newent->d_name.name;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err) {
 		/* ctime changes */
 		fuse_invalidate_attr(d_inode(oldent));
@@ -847,7 +924,7 @@ static int fuse_link(struct dentry *entry, struct inode *newdir,
 	int err;
 	struct fuse_link_in inarg;
 	struct inode *inode = d_inode(entry);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -858,7 +935,7 @@ static int fuse_link(struct dentry *entry, struct inode *newdir,
 	args.in_args[0].value = &inarg;
 	args.in_args[1].size = newent->d_name.len + 1;
 	args.in_args[1].value = newent->d_name.name;
-	err = create_new_entry(fc, &args, newdir, newent, inode->i_mode);
+	err = create_new_entry(fm, &args, newdir, newent, inode->i_mode);
 	/* Contrary to "normal" filesystems it can happen that link
 	   makes two "logical" inodes point to the same "physical"
 	   inode.  We invalidate the attributes of the old one, so it
@@ -869,7 +946,7 @@ static int fuse_link(struct dentry *entry, struct inode *newdir,
 		struct fuse_inode *fi = get_fuse_inode(inode);
 
 		spin_lock(&fi->lock);
-		fi->attr_version = atomic64_inc_return(&fc->attr_version);
+		fi->attr_version = atomic64_inc_return(&fm->fc->attr_version);
 		if (likely(inode->i_nlink < UINT_MAX))
 			inc_nlink(inode);
 		spin_unlock(&fi->lock);
@@ -926,11 +1003,11 @@ static int fuse_do_getattr(struct inode *inode, struct kstat *stat,
 	int err;
 	struct fuse_getattr_in inarg;
 	struct fuse_attr_out outarg;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	u64 attr_version;
 
-	attr_version = fuse_get_attr_version(fc);
+	attr_version = fuse_get_attr_version(fm->fc);
 
 	memset(&inarg, 0, sizeof(inarg));
 	memset(&outarg, 0, sizeof(outarg));
@@ -949,7 +1026,7 @@ static int fuse_do_getattr(struct inode *inode, struct kstat *stat,
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err) {
 		if (fuse_invalid_attr(&outarg.attr) ||
 		    (inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
@@ -1002,7 +1079,7 @@ int fuse_update_attributes(struct inode *inode, struct file *file)
 				    STATX_BASIC_STATS & ~STATX_ATIME, 0);
 }
 
-int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
+int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
 			     u64 child_nodeid, struct qstr *name)
 {
 	int err = -ENOTDIR;
@@ -1010,7 +1087,7 @@ int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
 	struct dentry *dir;
 	struct dentry *entry;
 
-	parent = ilookup5(sb, parent_nodeid, fuse_inode_eq, &parent_nodeid);
+	parent = fuse_ilookup(fc, parent_nodeid, NULL);
 	if (!parent)
 		return -ENOENT;
 
@@ -1102,14 +1179,14 @@ int fuse_allow_current_process(struct fuse_conn *fc)
 
 static int fuse_access(struct inode *inode, int mask)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_access_in inarg;
 	int err;
 
 	BUG_ON(mask & MAY_NOT_BLOCK);
 
-	if (fc->no_access)
+	if (fm->fc->no_access)
 		return 0;
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -1119,9 +1196,9 @@ static int fuse_access(struct inode *inode, int mask)
 	args.in_numargs = 1;
 	args.in_args[0].size = sizeof(inarg);
 	args.in_args[0].value = &inarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
-		fc->no_access = 1;
+		fm->fc->no_access = 1;
 		err = 0;
 	}
 	return err;
@@ -1209,7 +1286,7 @@ static int fuse_permission(struct inode *inode, int mask)
 
 static int fuse_readlink_page(struct inode *inode, struct page *page)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_page_desc desc = { .length = PAGE_SIZE - 1 };
 	struct fuse_args_pages ap = {
 		.num_pages = 1,
@@ -1226,7 +1303,7 @@ static int fuse_readlink_page(struct inode *inode, struct page *page)
 	ap.args.page_zeroing = true;
 	ap.args.out_numargs = 1;
 	ap.args.out_args[0].size = desc.length;
-	res = fuse_simple_request(fc, &ap.args);
+	res = fuse_simple_request(fm, &ap.args);
 
 	fuse_invalidate_atime(inode);
 
@@ -1454,7 +1531,7 @@ static void fuse_setattr_fill(struct fuse_conn *fc, struct fuse_args *args,
  */
 int fuse_flush_times(struct inode *inode, struct fuse_file *ff)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_setattr_in inarg;
 	struct fuse_attr_out outarg;
@@ -1465,7 +1542,7 @@ int fuse_flush_times(struct inode *inode, struct fuse_file *ff)
 	inarg.valid = FATTR_MTIME;
 	inarg.mtime = inode->i_mtime.tv_sec;
 	inarg.mtimensec = inode->i_mtime.tv_nsec;
-	if (fc->minor >= 23) {
+	if (fm->fc->minor >= 23) {
 		inarg.valid |= FATTR_CTIME;
 		inarg.ctime = inode->i_ctime.tv_sec;
 		inarg.ctimensec = inode->i_ctime.tv_nsec;
@@ -1474,9 +1551,9 @@ int fuse_flush_times(struct inode *inode, struct fuse_file *ff)
 		inarg.valid |= FATTR_FH;
 		inarg.fh = ff->fh;
 	}
-	fuse_setattr_fill(fc, &args, inode, &inarg, &outarg);
+	fuse_setattr_fill(fm->fc, &args, inode, &inarg, &outarg);
 
-	return fuse_simple_request(fc, &args);
+	return fuse_simple_request(fm, &args);
 }
 
 /*
@@ -1491,7 +1568,8 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 		    struct file *file)
 {
 	struct inode *inode = d_inode(dentry);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_inode *fi = get_fuse_inode(inode);
 	FUSE_ARGS(args);
 	struct fuse_setattr_in inarg;
@@ -1501,6 +1579,7 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 	loff_t oldsize;
 	int err;
 	bool trust_local_cmtime = is_wb && S_ISREG(inode->i_mode);
+	bool fault_blocked = false;
 
 	if (!fc->default_permissions)
 		attr->ia_valid |= ATTR_FORCE;
@@ -1509,6 +1588,22 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 	if (err)
 		return err;
 
+	if (attr->ia_valid & ATTR_SIZE) {
+		if (WARN_ON(!S_ISREG(inode->i_mode)))
+			return -EIO;
+		is_truncate = true;
+	}
+
+	if (FUSE_IS_DAX(inode) && is_truncate) {
+		down_write(&fi->i_mmap_sem);
+		fault_blocked = true;
+		err = fuse_dax_break_layouts(inode, 0, 0);
+		if (err) {
+			up_write(&fi->i_mmap_sem);
+			return err;
+		}
+	}
+
 	if (attr->ia_valid & ATTR_OPEN) {
 		/* This is coming from open(..., ... | O_TRUNC); */
 		WARN_ON(!(attr->ia_valid & ATTR_SIZE));
@@ -1521,17 +1616,11 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 			 */
 			i_size_write(inode, 0);
 			truncate_pagecache(inode, 0);
-			return 0;
+			goto out;
 		}
 		file = NULL;
 	}
 
-	if (attr->ia_valid & ATTR_SIZE) {
-		if (WARN_ON(!S_ISREG(inode->i_mode)))
-			return -EIO;
-		is_truncate = true;
-	}
-
 	/* Flush dirty data/metadata before non-truncate SETATTR */
 	if (is_wb && S_ISREG(inode->i_mode) &&
 	    attr->ia_valid &
@@ -1566,7 +1655,7 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 		inarg.lock_owner = fuse_lock_owner_id(fc, current->files);
 	}
 	fuse_setattr_fill(fc, &args, inode, &inarg, &outarg);
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err) {
 		if (err == -EINTR)
 			fuse_invalidate_attr(inode);
@@ -1614,6 +1703,10 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 	}
 
 	clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
+out:
+	if (fault_blocked)
+		up_write(&fi->i_mmap_sem);
+
 	return 0;
 
 error:
@@ -1621,6 +1714,9 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 		fuse_release_nowrite(inode);
 
 	clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
+
+	if (fault_blocked)
+		up_write(&fi->i_mmap_sem);
 	return err;
 }
 

fs/fuse/file.c

@@ -32,7 +32,7 @@ static struct page **fuse_pages_alloc(unsigned int npages, gfp_t flags,
 	return pages;
 }
 
-static int fuse_send_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+static int fuse_send_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
 			  int opcode, struct fuse_open_out *outargp)
 {
 	struct fuse_open_in inarg;
@@ -40,7 +40,7 @@ static int fuse_send_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
 
 	memset(&inarg, 0, sizeof(inarg));
 	inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
-	if (!fc->atomic_o_trunc)
+	if (!fm->fc->atomic_o_trunc)
 		inarg.flags &= ~O_TRUNC;
 	args.opcode = opcode;
 	args.nodeid = nodeid;
@@ -51,7 +51,7 @@ static int fuse_send_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
 	args.out_args[0].size = sizeof(*outargp);
 	args.out_args[0].value = outargp;
 
-	return fuse_simple_request(fc, &args);
+	return fuse_simple_request(fm, &args);
 }
 
 struct fuse_release_args {
@@ -60,7 +60,7 @@ struct fuse_release_args {
 	struct inode *inode;
 };
 
-struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
+struct fuse_file *fuse_file_alloc(struct fuse_mount *fm)
 {
 	struct fuse_file *ff;
 
@@ -68,7 +68,7 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
 	if (unlikely(!ff))
 		return NULL;
 
-	ff->fc = fc;
+	ff->fm = fm;
 	ff->release_args = kzalloc(sizeof(*ff->release_args),
 				   GFP_KERNEL_ACCOUNT);
 	if (!ff->release_args) {
@@ -82,7 +82,7 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
 	RB_CLEAR_NODE(&ff->polled_node);
 	init_waitqueue_head(&ff->poll_wait);
 
-	ff->kh = atomic64_inc_return(&fc->khctr);
+	ff->kh = atomic64_inc_return(&fm->fc->khctr);
 
 	return ff;
 }
@@ -100,7 +100,7 @@ static struct fuse_file *fuse_file_get(struct fuse_file *ff)
 	return ff;
 }
 
-static void fuse_release_end(struct fuse_conn *fc, struct fuse_args *args,
+static void fuse_release_end(struct fuse_mount *fm, struct fuse_args *args,
 			     int error)
 {
 	struct fuse_release_args *ra = container_of(args, typeof(*ra), args);
@@ -114,29 +114,30 @@ static void fuse_file_put(struct fuse_file *ff, bool sync, bool isdir)
 	if (refcount_dec_and_test(&ff->count)) {
 		struct fuse_args *args = &ff->release_args->args;
 
-		if (isdir ? ff->fc->no_opendir : ff->fc->no_open) {
+		if (isdir ? ff->fm->fc->no_opendir : ff->fm->fc->no_open) {
 			/* Do nothing when client does not implement 'open' */
-			fuse_release_end(ff->fc, args, 0);
+			fuse_release_end(ff->fm, args, 0);
 		} else if (sync) {
-			fuse_simple_request(ff->fc, args);
-			fuse_release_end(ff->fc, args, 0);
+			fuse_simple_request(ff->fm, args);
+			fuse_release_end(ff->fm, args, 0);
 		} else {
 			args->end = fuse_release_end;
-			if (fuse_simple_background(ff->fc, args,
+			if (fuse_simple_background(ff->fm, args,
 						   GFP_KERNEL | __GFP_NOFAIL))
-				fuse_release_end(ff->fc, args, -ENOTCONN);
+				fuse_release_end(ff->fm, args, -ENOTCONN);
 		}
 		kfree(ff);
 	}
 }
 
-int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
 		 bool isdir)
 {
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_file *ff;
 	int opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN;
 
-	ff = fuse_file_alloc(fc);
+	ff = fuse_file_alloc(fm);
 	if (!ff)
 		return -ENOMEM;
 
@@ -147,7 +148,7 @@ int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
 		struct fuse_open_out outarg;
 		int err;
 
-		err = fuse_send_open(fc, nodeid, file, opcode, &outarg);
+		err = fuse_send_open(fm, nodeid, file, opcode, &outarg);
 		if (!err) {
 			ff->fh = outarg.fh;
 			ff->open_flags = outarg.open_flags;
@@ -216,27 +217,40 @@ void fuse_finish_open(struct inode *inode, struct file *file)
 
 int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
+	struct fuse_conn *fc = fm->fc;
 	int err;
 	bool is_wb_truncate = (file->f_flags & O_TRUNC) &&
 			  fc->atomic_o_trunc &&
 			  fc->writeback_cache;
+	bool dax_truncate = (file->f_flags & O_TRUNC) &&
+			  fc->atomic_o_trunc && FUSE_IS_DAX(inode);
 
 	err = generic_file_open(inode, file);
 	if (err)
 		return err;
 
-	if (is_wb_truncate) {
+	if (is_wb_truncate || dax_truncate) {
 		inode_lock(inode);
 		fuse_set_nowrite(inode);
 	}
 
-	err = fuse_do_open(fc, get_node_id(inode), file, isdir);
+	if (dax_truncate) {
+		down_write(&get_fuse_inode(inode)->i_mmap_sem);
+		err = fuse_dax_break_layouts(inode, 0, 0);
+		if (err)
+			goto out;
+	}
 
+	err = fuse_do_open(fm, get_node_id(inode), file, isdir);
 	if (!err)
 		fuse_finish_open(inode, file);
 
-	if (is_wb_truncate) {
+out:
+	if (dax_truncate)
+		up_write(&get_fuse_inode(inode)->i_mmap_sem);
+
+	if (is_wb_truncate | dax_truncate) {
 		fuse_release_nowrite(inode);
 		inode_unlock(inode);
 	}
@@ -247,7 +261,7 @@ int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
 static void fuse_prepare_release(struct fuse_inode *fi, struct fuse_file *ff,
 				 int flags, int opcode)
 {
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_conn *fc = ff->fm->fc;
 	struct fuse_release_args *ra = ff->release_args;
 
 	/* Inode is NULL on error path of fuse_create_open() */
@@ -285,7 +299,7 @@ void fuse_release_common(struct file *file, bool isdir)
 
 	if (ff->flock) {
 		ra->inarg.release_flags |= FUSE_RELEASE_FLOCK_UNLOCK;
-		ra->inarg.lock_owner = fuse_lock_owner_id(ff->fc,
+		ra->inarg.lock_owner = fuse_lock_owner_id(ff->fm->fc,
 							  (fl_owner_t) file);
 	}
 	/* Hold inode until release is finished */
@@ -300,7 +314,7 @@ void fuse_release_common(struct file *file, bool isdir)
 	 * synchronous RELEASE is allowed (and desirable) in this case
 	 * because the server can be trusted not to screw up.
 	 */
-	fuse_file_put(ff, ff->fc->destroy, isdir);
+	fuse_file_put(ff, ff->fm->fc->destroy, isdir);
 }
 
 static int fuse_open(struct inode *inode, struct file *file)
@@ -443,7 +457,7 @@ static void fuse_sync_writes(struct inode *inode)
 static int fuse_flush(struct file *file, fl_owner_t id)
 {
 	struct inode *inode = file_inode(file);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_file *ff = file->private_data;
 	struct fuse_flush_in inarg;
 	FUSE_ARGS(args);
@@ -465,12 +479,12 @@ static int fuse_flush(struct file *file, fl_owner_t id)
 		return err;
 
 	err = 0;
-	if (fc->no_flush)
+	if (fm->fc->no_flush)
 		goto inval_attr_out;
 
 	memset(&inarg, 0, sizeof(inarg));
 	inarg.fh = ff->fh;
-	inarg.lock_owner = fuse_lock_owner_id(fc, id);
+	inarg.lock_owner = fuse_lock_owner_id(fm->fc, id);
 	args.opcode = FUSE_FLUSH;
 	args.nodeid = get_node_id(inode);
 	args.in_numargs = 1;
@@ -478,9 +492,9 @@ static int fuse_flush(struct file *file, fl_owner_t id)
 	args.in_args[0].value = &inarg;
 	args.force = true;
 
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
-		fc->no_flush = 1;
+		fm->fc->no_flush = 1;
 		err = 0;
 	}
 
@@ -489,7 +503,7 @@ static int fuse_flush(struct file *file, fl_owner_t id)
 	 * In memory i_blocks is not maintained by fuse, if writeback cache is
 	 * enabled, i_blocks from cached attr may not be accurate.
 	 */
-	if (!err && fc->writeback_cache)
+	if (!err && fm->fc->writeback_cache)
 		fuse_invalidate_attr(inode);
 	return err;
 }
@@ -498,7 +512,7 @@ int fuse_fsync_common(struct file *file, loff_t start, loff_t end,
 		      int datasync, int opcode)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_file *ff = file->private_data;
 	FUSE_ARGS(args);
 	struct fuse_fsync_in inarg;
@@ -511,7 +525,7 @@ int fuse_fsync_common(struct file *file, loff_t start, loff_t end,
 	args.in_numargs = 1;
 	args.in_args[0].size = sizeof(inarg);
 	args.in_args[0].value = &inarg;
-	return fuse_simple_request(fc, &args);
+	return fuse_simple_request(fm, &args);
 }
 
 static int fuse_fsync(struct file *file, loff_t start, loff_t end,
@@ -686,7 +700,7 @@ static void fuse_io_free(struct fuse_io_args *ia)
 	kfree(ia);
 }
 
-static void fuse_aio_complete_req(struct fuse_conn *fc, struct fuse_args *args,
+static void fuse_aio_complete_req(struct fuse_mount *fm, struct fuse_args *args,
 				  int err)
 {
 	struct fuse_io_args *ia = container_of(args, typeof(*ia), ap.args);
@@ -715,7 +729,7 @@ static void fuse_aio_complete_req(struct fuse_conn *fc, struct fuse_args *args,
 	fuse_io_free(ia);
 }
 
-static ssize_t fuse_async_req_send(struct fuse_conn *fc,
+static ssize_t fuse_async_req_send(struct fuse_mount *fm,
 				   struct fuse_io_args *ia, size_t num_bytes)
 {
 	ssize_t err;
@@ -729,9 +743,9 @@ static ssize_t fuse_async_req_send(struct fuse_conn *fc,
 
 	ia->ap.args.end = fuse_aio_complete_req;
 	ia->ap.args.may_block = io->should_dirty;
-	err = fuse_simple_background(fc, &ia->ap.args, GFP_KERNEL);
+	err = fuse_simple_background(fm, &ia->ap.args, GFP_KERNEL);
 	if (err)
-		fuse_aio_complete_req(fc, &ia->ap.args, err);
+		fuse_aio_complete_req(fm, &ia->ap.args, err);
 
 	return num_bytes;
 }
@@ -741,18 +755,18 @@ static ssize_t fuse_send_read(struct fuse_io_args *ia, loff_t pos, size_t count,
 {
 	struct file *file = ia->io->iocb->ki_filp;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 
 	fuse_read_args_fill(ia, file, pos, count, FUSE_READ);
 	if (owner != NULL) {
 		ia->read.in.read_flags |= FUSE_READ_LOCKOWNER;
-		ia->read.in.lock_owner = fuse_lock_owner_id(fc, owner);
+		ia->read.in.lock_owner = fuse_lock_owner_id(fm->fc, owner);
 	}
 
 	if (ia->io->async)
-		return fuse_async_req_send(fc, ia, count);
+		return fuse_async_req_send(fm, ia, count);
 
-	return fuse_simple_request(fc, &ia->ap.args);
+	return fuse_simple_request(fm, &ia->ap.args);
 }
 
 static void fuse_read_update_size(struct inode *inode, loff_t size,
@@ -798,7 +812,7 @@ static void fuse_short_read(struct inode *inode, u64 attr_ver, size_t num_read,
 static int fuse_do_readpage(struct file *file, struct page *page)
 {
 	struct inode *inode = page->mapping->host;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	loff_t pos = page_offset(page);
 	struct fuse_page_desc desc = { .length = PAGE_SIZE };
 	struct fuse_io_args ia = {
@@ -818,14 +832,14 @@ static int fuse_do_readpage(struct file *file, struct page *page)
 	 */
 	fuse_wait_on_page_writeback(inode, page->index);
 
-	attr_ver = fuse_get_attr_version(fc);
+	attr_ver = fuse_get_attr_version(fm->fc);
 
 	/* Don't overflow end offset */
 	if (pos + (desc.length - 1) == LLONG_MAX)
 		desc.length--;
 
 	fuse_read_args_fill(&ia, file, pos, desc.length, FUSE_READ);
-	res = fuse_simple_request(fc, &ia.ap.args);
+	res = fuse_simple_request(fm, &ia.ap.args);
 	if (res < 0)
 		return res;
 	/*
@@ -855,7 +869,7 @@ static int fuse_readpage(struct file *file, struct page *page)
 	return err;
 }
 
-static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_args *args,
+static void fuse_readpages_end(struct fuse_mount *fm, struct fuse_args *args,
 			       int err)
 {
 	int i;
@@ -899,7 +913,7 @@ static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_args *args,
 static void fuse_send_readpages(struct fuse_io_args *ia, struct file *file)
 {
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	struct fuse_args_pages *ap = &ia->ap;
 	loff_t pos = page_offset(ap->pages[0]);
 	size_t count = ap->num_pages << PAGE_SHIFT;
@@ -918,18 +932,18 @@ static void fuse_send_readpages(struct fuse_io_args *ia, struct file *file)
 	WARN_ON((loff_t) (pos + count) < 0);
 
 	fuse_read_args_fill(ia, file, pos, count, FUSE_READ);
-	ia->read.attr_ver = fuse_get_attr_version(fc);
-	if (fc->async_read) {
+	ia->read.attr_ver = fuse_get_attr_version(fm->fc);
+	if (fm->fc->async_read) {
 		ia->ff = fuse_file_get(ff);
 		ap->args.end = fuse_readpages_end;
-		err = fuse_simple_background(fc, &ap->args, GFP_KERNEL);
+		err = fuse_simple_background(fm, &ap->args, GFP_KERNEL);
 		if (!err)
 			return;
 	} else {
-		res = fuse_simple_request(fc, &ap->args);
+		res = fuse_simple_request(fm, &ap->args);
 		err = res < 0 ? res : 0;
 	}
-	fuse_readpages_end(fc, &ap->args, err);
+	fuse_readpages_end(fm, &ap->args, err);
 }
 
 static void fuse_readahead(struct readahead_control *rac)
@@ -1000,7 +1014,7 @@ static void fuse_write_args_fill(struct fuse_io_args *ia, struct fuse_file *ff,
 	args->opcode = FUSE_WRITE;
 	args->nodeid = ff->nodeid;
 	args->in_numargs = 2;
-	if (ff->fc->minor < 9)
+	if (ff->fm->fc->minor < 9)
 		args->in_args[0].size = FUSE_COMPAT_WRITE_IN_SIZE;
 	else
 		args->in_args[0].size = sizeof(ia->write.in);
@@ -1029,7 +1043,7 @@ static ssize_t fuse_send_write(struct fuse_io_args *ia, loff_t pos,
 	struct kiocb *iocb = ia->io->iocb;
 	struct file *file = iocb->ki_filp;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	struct fuse_write_in *inarg = &ia->write.in;
 	ssize_t err;
 
@@ -1037,13 +1051,13 @@ static ssize_t fuse_send_write(struct fuse_io_args *ia, loff_t pos,
 	inarg->flags = fuse_write_flags(iocb);
 	if (owner != NULL) {
 		inarg->write_flags |= FUSE_WRITE_LOCKOWNER;
-		inarg->lock_owner = fuse_lock_owner_id(fc, owner);
+		inarg->lock_owner = fuse_lock_owner_id(fm->fc, owner);
 	}
 
 	if (ia->io->async)
-		return fuse_async_req_send(fc, ia, count);
+		return fuse_async_req_send(fm, ia, count);
 
-	err = fuse_simple_request(fc, &ia->ap.args);
+	err = fuse_simple_request(fm, &ia->ap.args);
 	if (!err && ia->write.out.size > count)
 		err = -EIO;
 
@@ -1074,7 +1088,7 @@ static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
 	struct fuse_args_pages *ap = &ia->ap;
 	struct file *file = iocb->ki_filp;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	unsigned int offset, i;
 	int err;
 
@@ -1084,7 +1098,7 @@ static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
 	fuse_write_args_fill(ia, ff, pos, count);
 	ia->write.in.flags = fuse_write_flags(iocb);
 
-	err = fuse_simple_request(fc, &ap->args);
+	err = fuse_simple_request(fm, &ap->args);
 	if (!err && ia->write.out.size > count)
 		err = -EIO;
 
@@ -1399,7 +1413,7 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
 	struct file *file = io->iocb->ki_filp;
 	struct inode *inode = file->f_mapping->host;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_conn *fc = ff->fm->fc;
 	size_t nmax = write ? fc->max_write : fc->max_read;
 	loff_t pos = *ppos;
 	size_t count = iov_iter_count(iter);
@@ -1539,10 +1553,14 @@ static ssize_t fuse_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	struct file *file = iocb->ki_filp;
 	struct fuse_file *ff = file->private_data;
+	struct inode *inode = file_inode(file);
 
-	if (is_bad_inode(file_inode(file)))
+	if (is_bad_inode(inode))
 		return -EIO;
 
+	if (FUSE_IS_DAX(inode))
+		return fuse_dax_read_iter(iocb, to);
+
 	if (!(ff->open_flags & FOPEN_DIRECT_IO))
 		return fuse_cache_read_iter(iocb, to);
 	else
@@ -1553,10 +1571,14 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct fuse_file *ff = file->private_data;
+	struct inode *inode = file_inode(file);
 
-	if (is_bad_inode(file_inode(file)))
+	if (is_bad_inode(inode))
 		return -EIO;
 
+	if (FUSE_IS_DAX(inode))
+		return fuse_dax_write_iter(iocb, from);
+
 	if (!(ff->open_flags & FOPEN_DIRECT_IO))
 		return fuse_cache_write_iter(iocb, from);
 	else
@@ -1578,7 +1600,7 @@ static void fuse_writepage_free(struct fuse_writepage_args *wpa)
 	kfree(wpa);
 }
 
-static void fuse_writepage_finish(struct fuse_conn *fc,
+static void fuse_writepage_finish(struct fuse_mount *fm,
 				  struct fuse_writepage_args *wpa)
 {
 	struct fuse_args_pages *ap = &wpa->ia.ap;
@@ -1596,7 +1618,7 @@ static void fuse_writepage_finish(struct fuse_conn *fc,
 }
 
 /* Called under fi->lock, may release and reacquire it */
-static void fuse_send_writepage(struct fuse_conn *fc,
+static void fuse_send_writepage(struct fuse_mount *fm,
 				struct fuse_writepage_args *wpa, loff_t size)
 __releases(fi->lock)
 __acquires(fi->lock)
@@ -1622,10 +1644,10 @@ __acquires(fi->lock)
 	args->force = true;
 	args->nocreds = true;
 
-	err = fuse_simple_background(fc, args, GFP_ATOMIC);
+	err = fuse_simple_background(fm, args, GFP_ATOMIC);
 	if (err == -ENOMEM) {
 		spin_unlock(&fi->lock);
-		err = fuse_simple_background(fc, args, GFP_NOFS | __GFP_NOFAIL);
+		err = fuse_simple_background(fm, args, GFP_NOFS | __GFP_NOFAIL);
 		spin_lock(&fi->lock);
 	}
 
@@ -1638,7 +1660,7 @@ __acquires(fi->lock)
  out_free:
 	fi->writectr--;
 	rb_erase(&wpa->writepages_entry, &fi->writepages);
-	fuse_writepage_finish(fc, wpa);
+	fuse_writepage_finish(fm, wpa);
 	spin_unlock(&fi->lock);
 
 	/* After fuse_writepage_finish() aux request list is private */
@@ -1662,7 +1684,7 @@ void fuse_flush_writepages(struct inode *inode)
 __releases(fi->lock)
 __acquires(fi->lock)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_inode *fi = get_fuse_inode(inode);
 	loff_t crop = i_size_read(inode);
 	struct fuse_writepage_args *wpa;
@@ -1671,7 +1693,7 @@ __acquires(fi->lock)
 		wpa = list_entry(fi->queued_writes.next,
 				 struct fuse_writepage_args, queue_entry);
 		list_del_init(&wpa->queue_entry);
-		fuse_send_writepage(fc, wpa, crop);
+		fuse_send_writepage(fm, wpa, crop);
 	}
 }
 
@@ -1712,7 +1734,7 @@ static void tree_insert(struct rb_root *root, struct fuse_writepage_args *wpa)
 	WARN_ON(fuse_insert_writeback(root, wpa));
 }
 
-static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_args *args,
+static void fuse_writepage_end(struct fuse_mount *fm, struct fuse_args *args,
 			       int error)
 {
 	struct fuse_writepage_args *wpa =
@@ -1724,7 +1746,7 @@ static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_args *args,
 	spin_lock(&fi->lock);
 	rb_erase(&wpa->writepages_entry, &fi->writepages);
 	while (wpa->next) {
-		struct fuse_conn *fc = get_fuse_conn(inode);
+		struct fuse_mount *fm = get_fuse_mount(inode);
 		struct fuse_write_in *inarg = &wpa->ia.write.in;
 		struct fuse_writepage_args *next = wpa->next;
 
@@ -1756,10 +1778,10 @@ static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_args *args,
 		 * no invocations of fuse_writepage_end() while we're in
 		 * fuse_set_nowrite..fuse_release_nowrite section.
 		 */
-		fuse_send_writepage(fc, next, inarg->offset + inarg->size);
+		fuse_send_writepage(fm, next, inarg->offset + inarg->size);
 	}
 	fi->writectr--;
-	fuse_writepage_finish(fc, wpa);
+	fuse_writepage_finish(fm, wpa);
 	spin_unlock(&fi->lock);
 	fuse_writepage_free(wpa);
 }
@@ -2317,6 +2339,10 @@ static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct fuse_file *ff = file->private_data;
 
+	/* DAX mmap is superior to direct_io mmap */
+	if (FUSE_IS_DAX(file_inode(file)))
+		return fuse_dax_mmap(file, vma);
+
 	if (ff->open_flags & FOPEN_DIRECT_IO) {
 		/* Can't provide the coherency needed for MAP_SHARED */
 		if (vma->vm_flags & VM_MAYSHARE)
@@ -2395,7 +2421,7 @@ static void fuse_lk_fill(struct fuse_args *args, struct file *file,
 static int fuse_getlk(struct file *file, struct file_lock *fl)
 {
 	struct inode *inode = file_inode(file);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_lk_in inarg;
 	struct fuse_lk_out outarg;
@@ -2405,9 +2431,9 @@ static int fuse_getlk(struct file *file, struct file_lock *fl)
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err)
-		err = convert_fuse_file_lock(fc, &outarg.lk, fl);
+		err = convert_fuse_file_lock(fm->fc, &outarg.lk, fl);
 
 	return err;
 }
@@ -2415,12 +2441,12 @@ static int fuse_getlk(struct file *file, struct file_lock *fl)
 static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
 {
 	struct inode *inode = file_inode(file);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_lk_in inarg;
 	int opcode = (fl->fl_flags & FL_SLEEP) ? FUSE_SETLKW : FUSE_SETLK;
 	struct pid *pid = fl->fl_type != F_UNLCK ? task_tgid(current) : NULL;
-	pid_t pid_nr = pid_nr_ns(pid, fc->pid_ns);
+	pid_t pid_nr = pid_nr_ns(pid, fm->fc->pid_ns);
 	int err;
 
 	if (fl->fl_lmops && fl->fl_lmops->lm_grant) {
@@ -2433,7 +2459,7 @@ static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
 		return 0;
 
 	fuse_lk_fill(&args, file, fl, opcode, pid_nr, flock, &inarg);
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 
 	/* locking is restartable */
 	if (err == -EINTR)
@@ -2487,13 +2513,13 @@ static int fuse_file_flock(struct file *file, int cmd, struct file_lock *fl)
 static sector_t fuse_bmap(struct address_space *mapping, sector_t block)
 {
 	struct inode *inode = mapping->host;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_bmap_in inarg;
 	struct fuse_bmap_out outarg;
 	int err;
 
-	if (!inode->i_sb->s_bdev || fc->no_bmap)
+	if (!inode->i_sb->s_bdev || fm->fc->no_bmap)
 		return 0;
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -2507,9 +2533,9 @@ static sector_t fuse_bmap(struct address_space *mapping, sector_t block)
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS)
-		fc->no_bmap = 1;
+		fm->fc->no_bmap = 1;
 
 	return err ? 0 : outarg.block;
 }
@@ -2517,7 +2543,7 @@ static sector_t fuse_bmap(struct address_space *mapping, sector_t block)
 static loff_t fuse_lseek(struct file *file, loff_t offset, int whence)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_file *ff = file->private_data;
 	FUSE_ARGS(args);
 	struct fuse_lseek_in inarg = {
@@ -2528,7 +2554,7 @@ static loff_t fuse_lseek(struct file *file, loff_t offset, int whence)
 	struct fuse_lseek_out outarg;
 	int err;
 
-	if (fc->no_lseek)
+	if (fm->fc->no_lseek)
 		goto fallback;
 
 	args.opcode = FUSE_LSEEK;
@@ -2539,10 +2565,10 @@ static loff_t fuse_lseek(struct file *file, loff_t offset, int whence)
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err) {
 		if (err == -ENOSYS) {
-			fc->no_lseek = 1;
+			fm->fc->no_lseek = 1;
 			goto fallback;
 		}
 		return err;
@@ -2728,7 +2754,7 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 		   unsigned int flags)
 {
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	struct fuse_ioctl_in inarg = {
 		.fh = ff->fh,
 		.cmd = cmd,
@@ -2761,12 +2787,12 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 	BUILD_BUG_ON(sizeof(struct fuse_ioctl_iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE);
 
 	err = -ENOMEM;
-	ap.pages = fuse_pages_alloc(fc->max_pages, GFP_KERNEL, &ap.descs);
+	ap.pages = fuse_pages_alloc(fm->fc->max_pages, GFP_KERNEL, &ap.descs);
 	iov_page = (struct iovec *) __get_free_page(GFP_KERNEL);
 	if (!ap.pages || !iov_page)
 		goto out;
 
-	fuse_page_descs_length_init(ap.descs, 0, fc->max_pages);
+	fuse_page_descs_length_init(ap.descs, 0, fm->fc->max_pages);
 
 	/*
 	 * If restricted, initialize IO parameters as encoded in @cmd.
@@ -2811,7 +2837,7 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 
 	/* make sure there are enough buffer pages and init request with them */
 	err = -ENOMEM;
-	if (max_pages > fc->max_pages)
+	if (max_pages > fm->fc->max_pages)
 		goto out;
 	while (ap.num_pages < max_pages) {
 		ap.pages[ap.num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
@@ -2848,7 +2874,7 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 	ap.args.out_pages = true;
 	ap.args.out_argvar = true;
 
-	transferred = fuse_simple_request(fc, &ap.args);
+	transferred = fuse_simple_request(fm, &ap.args);
 	err = transferred;
 	if (transferred < 0)
 		goto out;
@@ -2876,7 +2902,7 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 			goto out;
 
 		vaddr = kmap_atomic(ap.pages[0]);
-		err = fuse_copy_ioctl_iovec(fc, iov_page, vaddr,
+		err = fuse_copy_ioctl_iovec(fm->fc, iov_page, vaddr,
 					    transferred, in_iovs + out_iovs,
 					    (flags & FUSE_IOCTL_COMPAT) != 0);
 		kunmap_atomic(vaddr);
@@ -2886,11 +2912,11 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
 		in_iov = iov_page;
 		out_iov = in_iov + in_iovs;
 
-		err = fuse_verify_ioctl_iov(fc, in_iov, in_iovs);
+		err = fuse_verify_ioctl_iov(fm->fc, in_iov, in_iovs);
 		if (err)
 			goto out;
 
-		err = fuse_verify_ioctl_iov(fc, out_iov, out_iovs);
+		err = fuse_verify_ioctl_iov(fm->fc, out_iov, out_iovs);
 		if (err)
 			goto out;
 
@@ -3000,13 +3026,13 @@ static void fuse_register_polled_file(struct fuse_conn *fc,
 __poll_t fuse_file_poll(struct file *file, poll_table *wait)
 {
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	struct fuse_poll_in inarg = { .fh = ff->fh, .kh = ff->kh };
 	struct fuse_poll_out outarg;
 	FUSE_ARGS(args);
 	int err;
 
-	if (fc->no_poll)
+	if (fm->fc->no_poll)
 		return DEFAULT_POLLMASK;
 
 	poll_wait(file, &ff->poll_wait, wait);
@@ -3018,7 +3044,7 @@ __poll_t fuse_file_poll(struct file *file, poll_table *wait)
 	 */
 	if (waitqueue_active(&ff->poll_wait)) {
 		inarg.flags |= FUSE_POLL_SCHEDULE_NOTIFY;
-		fuse_register_polled_file(fc, ff);
+		fuse_register_polled_file(fm->fc, ff);
 	}
 
 	args.opcode = FUSE_POLL;
@@ -3029,12 +3055,12 @@ __poll_t fuse_file_poll(struct file *file, poll_table *wait)
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 
 	if (!err)
 		return demangle_poll(outarg.revents);
 	if (err == -ENOSYS) {
-		fc->no_poll = 1;
+		fm->fc->no_poll = 1;
 		return DEFAULT_POLLMASK;
 	}
 	return EPOLLERR;
@@ -3120,13 +3146,13 @@ fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 	 * By default, we want to optimize all I/Os with async request
 	 * submission to the client filesystem if supported.
 	 */
-	io->async = ff->fc->async_dio;
+	io->async = ff->fm->fc->async_dio;
 	io->iocb = iocb;
 	io->blocking = is_sync_kiocb(iocb);
 
 	/* optimization for short read */
 	if (io->async && !io->write && offset + count > i_size) {
-		iov_iter_truncate(iter, fuse_round_up(ff->fc, i_size - offset));
+		iov_iter_truncate(iter, fuse_round_up(ff->fm->fc, i_size - offset));
 		shortened = count - iov_iter_count(iter);
 		count -= shortened;
 	}
@@ -3196,7 +3222,7 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset,
 	struct fuse_file *ff = file->private_data;
 	struct inode *inode = file_inode(file);
 	struct fuse_inode *fi = get_fuse_inode(inode);
-	struct fuse_conn *fc = ff->fc;
+	struct fuse_mount *fm = ff->fm;
 	FUSE_ARGS(args);
 	struct fuse_fallocate_in inarg = {
 		.fh = ff->fh,
@@ -3208,14 +3234,23 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset,
 	bool lock_inode = !(mode & FALLOC_FL_KEEP_SIZE) ||
 			   (mode & FALLOC_FL_PUNCH_HOLE);
 
+	bool block_faults = FUSE_IS_DAX(inode) && lock_inode;
+
 	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
 		return -EOPNOTSUPP;
 
-	if (fc->no_fallocate)
+	if (fm->fc->no_fallocate)
 		return -EOPNOTSUPP;
 
 	if (lock_inode) {
 		inode_lock(inode);
+		if (block_faults) {
+			down_write(&fi->i_mmap_sem);
+			err = fuse_dax_break_layouts(inode, 0, 0);
+			if (err)
+				goto out;
+		}
+
 		if (mode & FALLOC_FL_PUNCH_HOLE) {
 			loff_t endbyte = offset + length - 1;
 
@@ -3240,9 +3275,9 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset,
 	args.in_numargs = 1;
 	args.in_args[0].size = sizeof(inarg);
 	args.in_args[0].value = &inarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
-		fc->no_fallocate = 1;
+		fm->fc->no_fallocate = 1;
 		err = -EOPNOTSUPP;
 	}
 	if (err)
@@ -3252,7 +3287,7 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset,
 	if (!(mode & FALLOC_FL_KEEP_SIZE)) {
 		bool changed = fuse_write_update_size(inode, offset + length);
 
-		if (changed && fc->writeback_cache)
+		if (changed && fm->fc->writeback_cache)
 			file_update_time(file);
 	}
 
@@ -3265,6 +3300,9 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset,
 	if (!(mode & FALLOC_FL_KEEP_SIZE))
 		clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
 
+	if (block_faults)
+		up_write(&fi->i_mmap_sem);
+
 	if (lock_inode)
 		inode_unlock(inode);
 
@@ -3280,7 +3318,8 @@ static ssize_t __fuse_copy_file_range(struct file *file_in, loff_t pos_in,
 	struct inode *inode_in = file_inode(file_in);
 	struct inode *inode_out = file_inode(file_out);
 	struct fuse_inode *fi_out = get_fuse_inode(inode_out);
-	struct fuse_conn *fc = ff_in->fc;
+	struct fuse_mount *fm = ff_in->fm;
+	struct fuse_conn *fc = fm->fc;
 	FUSE_ARGS(args);
 	struct fuse_copy_file_range_in inarg = {
 		.fh_in = ff_in->fh,
@@ -3349,7 +3388,7 @@ static ssize_t __fuse_copy_file_range(struct file *file_in, loff_t pos_in,
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
 		fc->no_copy_file_range = 1;
 		err = -EOPNOTSUPP;
@@ -3404,6 +3443,7 @@ static const struct file_operations fuse_file_operations = {
 	.release	= fuse_release,
 	.fsync		= fuse_fsync,
 	.lock		= fuse_file_lock,
+	.get_unmapped_area = thp_get_unmapped_area,
 	.flock		= fuse_file_flock,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= iter_file_splice_write,
@@ -3439,4 +3479,7 @@ void fuse_init_file_inode(struct inode *inode)
 	fi->writectr = 0;
 	init_waitqueue_head(&fi->page_waitq);
 	fi->writepages = RB_ROOT;
+
+	if (IS_ENABLED(CONFIG_FUSE_DAX))
+		fuse_dax_inode_init(inode);
 }

fs/fuse/fuse_i.h

@@ -148,6 +148,20 @@ struct fuse_inode {
 
 	/** Lock to protect write related fields */
 	spinlock_t lock;
+
+	/**
+	 * Can't take inode lock in fault path (leads to circular dependency).
+	 * Introduce another semaphore which can be taken in fault path and
+	 * then other filesystem paths can take this to block faults.
+	 */
+	struct rw_semaphore i_mmap_sem;
+
+#ifdef CONFIG_FUSE_DAX
+	/*
+	 * Dax specific inode data
+	 */
+	struct fuse_inode_dax *dax;
+#endif
 };
 
 /** FUSE inode state bits */
@@ -161,12 +175,13 @@ enum {
 };
 
 struct fuse_conn;
+struct fuse_mount;
 struct fuse_release_args;
 
 /** FUSE specific file data */
 struct fuse_file {
 	/** Fuse connection for this file */
-	struct fuse_conn *fc;
+	struct fuse_mount *fm;
 
 	/* Argument space reserved for release */
 	struct fuse_release_args *release_args;
@@ -252,7 +267,7 @@ struct fuse_args {
 	bool may_block:1;
 	struct fuse_in_arg in_args[3];
 	struct fuse_arg out_args[2];
-	void (*end)(struct fuse_conn *fc, struct fuse_args *args, int error);
+	void (*end)(struct fuse_mount *fm, struct fuse_args *args, int error);
 };
 
 struct fuse_args_pages {
@@ -360,6 +375,9 @@ struct fuse_req {
 	/** virtio-fs's physically contiguous buffer for in and out args */
 	void *argbuf;
 #endif
+
+	/** fuse_mount this request belongs to */
+	struct fuse_mount *fm;
 };
 
 struct fuse_iqueue;
@@ -482,11 +500,15 @@ struct fuse_fs_context {
 	bool destroy:1;
 	bool no_control:1;
 	bool no_force_umount:1;
-	bool no_mount_options:1;
+	bool legacy_opts_show:1;
+	bool dax:1;
 	unsigned int max_read;
 	unsigned int blksize;
 	const char *subtype;
 
+	/* DAX device, may be NULL */
+	struct dax_device *dax_dev;
+
 	/* fuse_dev pointer to fill in, should contain NULL on entry */
 	void **fudptr;
 };
@@ -494,9 +516,9 @@ struct fuse_fs_context {
 /**
  * A Fuse connection.
  *
- * This structure is created, when the filesystem is mounted, and is
- * destroyed, when the client device is closed and the filesystem is
- * unmounted.
+ * This structure is created, when the root filesystem is mounted, and
+ * is destroyed, when the client device is closed and the last
+ * fuse_mount is destroyed.
  */
 struct fuse_conn {
 	/** Lock protecting accessess to  members of this structure */
@@ -610,6 +632,9 @@ struct fuse_conn {
 	/** cache READLINK responses in page cache */
 	unsigned cache_symlinks:1;
 
+	/* show legacy mount options */
+	unsigned int legacy_opts_show:1;
+
 	/*
 	 * The following bitfields are only for optimization purposes
 	 * and hence races in setting them will not cause malfunction
@@ -717,8 +742,8 @@ struct fuse_conn {
 	/** Do not allow MNT_FORCE umount */
 	unsigned int no_force_umount:1;
 
-	/* Do not show mount options */
-	unsigned int no_mount_options:1;
+	/* Auto-mount submounts announced by the server */
+	unsigned int auto_submounts:1;
 
 	/** The number of requests waiting for completion */
 	atomic_t num_waiting;
@@ -726,10 +751,10 @@ struct fuse_conn {
 	/** Negotiated minor version */
 	unsigned minor;
 
-	/** Entry on the fuse_conn_list */
+	/** Entry on the fuse_mount_list */
 	struct list_head entry;
 
-	/** Device ID from super block */
+	/** Device ID from the root super block */
 	dev_t dev;
 
 	/** Dentries in the control filesystem */
@@ -747,24 +772,70 @@ struct fuse_conn {
 	/** Called on final put */
 	void (*release)(struct fuse_conn *);
 
-	/** Super block for this connection. */
-	struct super_block *sb;
-
-	/** Read/write semaphore to hold when accessing sb. */
+	/**
+	 * Read/write semaphore to hold when accessing the sb of any
+	 * fuse_mount belonging to this connection
+	 */
 	struct rw_semaphore killsb;
 
 	/** List of device instances belonging to this connection */
 	struct list_head devices;
+
+#ifdef CONFIG_FUSE_DAX
+	/* Dax specific conn data, non-NULL if DAX is enabled */
+	struct fuse_conn_dax *dax;
+#endif
+
+	/** List of filesystems using this connection */
+	struct list_head mounts;
 };
 
-static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb)
+/*
+ * Represents a mounted filesystem, potentially a submount.
+ *
+ * This object allows sharing a fuse_conn between separate mounts to
+ * allow submounts with dedicated superblocks and thus separate device
+ * IDs.
+ */
+struct fuse_mount {
+	/* Underlying (potentially shared) connection to the FUSE server */
+	struct fuse_conn *fc;
+
+	/* Refcount */
+	refcount_t count;
+
+	/*
+	 * Super block for this connection (fc->killsb must be held when
+	 * accessing this).
+	 */
+	struct super_block *sb;
+
+	/* Entry on fc->mounts */
+	struct list_head fc_entry;
+};
+
+static inline struct fuse_mount *get_fuse_mount_super(struct super_block *sb)
 {
 	return sb->s_fs_info;
 }
 
+static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb)
+{
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+
+	return fm ? fm->fc : NULL;
+}
+
+static inline struct fuse_mount *get_fuse_mount(struct inode *inode)
+{
+	return get_fuse_mount_super(inode->i_sb);
+}
+
 static inline struct fuse_conn *get_fuse_conn(struct inode *inode)
 {
-	return get_fuse_conn_super(inode->i_sb);
+	struct fuse_mount *fm = get_fuse_mount(inode);
+
+	return fm ? fm->fc : NULL;
 }
 
 static inline struct fuse_inode *get_fuse_inode(struct inode *inode)
@@ -793,11 +864,6 @@ extern const struct file_operations fuse_dev_operations;
 extern const struct dentry_operations fuse_dentry_operations;
 extern const struct dentry_operations fuse_root_dentry_operations;
 
-/**
- * Inode to nodeid comparison.
- */
-int fuse_inode_eq(struct inode *inode, void *_nodeidp);
-
 /**
  * Get a filled in inode
  */
@@ -848,7 +914,7 @@ void fuse_read_args_fill(struct fuse_io_args *ia, struct file *file, loff_t pos,
  */
 int fuse_open_common(struct inode *inode, struct file *file, bool isdir);
 
-struct fuse_file *fuse_file_alloc(struct fuse_conn *fc);
+struct fuse_file *fuse_file_alloc(struct fuse_mount *fm);
 void fuse_file_free(struct fuse_file *ff);
 void fuse_finish_open(struct inode *inode, struct file *file);
 
@@ -916,14 +982,14 @@ void __exit fuse_ctl_cleanup(void);
 /**
  * Simple request sending that does request allocation and freeing
  */
-ssize_t fuse_simple_request(struct fuse_conn *fc, struct fuse_args *args);
-int fuse_simple_background(struct fuse_conn *fc, struct fuse_args *args,
+ssize_t fuse_simple_request(struct fuse_mount *fm, struct fuse_args *args);
+int fuse_simple_background(struct fuse_mount *fm, struct fuse_args *args,
 			   gfp_t gfp_flags);
 
 /**
  * End a finished request
  */
-void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req);
+void fuse_request_end(struct fuse_req *req);
 
 /* Abort all requests */
 void fuse_abort_conn(struct fuse_conn *fc);
@@ -949,7 +1015,8 @@ struct fuse_conn *fuse_conn_get(struct fuse_conn *fc);
 /**
  * Initialize fuse_conn
  */
-void fuse_conn_init(struct fuse_conn *fc, struct user_namespace *user_ns,
+void fuse_conn_init(struct fuse_conn *fc, struct fuse_mount *fm,
+		    struct user_namespace *user_ns,
 		    const struct fuse_iqueue_ops *fiq_ops, void *fiq_priv);
 
 /**
@@ -957,11 +1024,21 @@ void fuse_conn_init(struct fuse_conn *fc, struct user_namespace *user_ns,
  */
 void fuse_conn_put(struct fuse_conn *fc);
 
+/**
+ * Acquire reference to fuse_mount
+ */
+struct fuse_mount *fuse_mount_get(struct fuse_mount *fm);
+
+/**
+ * Release reference to fuse_mount
+ */
+void fuse_mount_put(struct fuse_mount *fm);
+
 struct fuse_dev *fuse_dev_alloc_install(struct fuse_conn *fc);
 struct fuse_dev *fuse_dev_alloc(void);
 void fuse_dev_install(struct fuse_dev *fud, struct fuse_conn *fc);
 void fuse_dev_free(struct fuse_dev *fud);
-void fuse_send_init(struct fuse_conn *fc);
+void fuse_send_init(struct fuse_mount *fm);
 
 /**
  * Fill in superblock and initialize fuse connection
@@ -970,12 +1047,26 @@ void fuse_send_init(struct fuse_conn *fc);
  */
 int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx);
 
-/**
- * Disassociate fuse connection from superblock and kill the superblock
+/*
+ * Fill in superblock for submounts
+ * @sb: partially-initialized superblock to fill in
+ * @parent_fi: The fuse_inode of the parent filesystem where this submount is
+ * 	       mounted
+ */
+int fuse_fill_super_submount(struct super_block *sb,
+			     struct fuse_inode *parent_fi);
+
+/*
+ * Remove the mount from the connection
  *
- * Calls kill_anon_super(), do not use with bdev mounts.
+ * Returns whether this was the last mount
  */
-void fuse_kill_sb_anon(struct super_block *sb);
+bool fuse_mount_remove(struct fuse_mount *fm);
+
+/*
+ * Shut down the connection (possibly sending DESTROY request).
+ */
+void fuse_conn_destroy(struct fuse_mount *fm);
 
 /**
  * Add connection to control filesystem
@@ -1010,10 +1101,20 @@ void fuse_flush_writepages(struct inode *inode);
 void fuse_set_nowrite(struct inode *inode);
 void fuse_release_nowrite(struct inode *inode);
 
+/**
+ * Scan all fuse_mounts belonging to fc to find the first where
+ * ilookup5() returns a result.  Return that result and the
+ * respective fuse_mount in *fm (unless fm is NULL).
+ *
+ * The caller must hold fc->killsb.
+ */
+struct inode *fuse_ilookup(struct fuse_conn *fc, u64 nodeid,
+			   struct fuse_mount **fm);
+
 /**
  * File-system tells the kernel to invalidate cache for the given node id.
  */
-int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid,
+int fuse_reverse_inval_inode(struct fuse_conn *fc, u64 nodeid,
 			     loff_t offset, loff_t len);
 
 /**
@@ -1026,10 +1127,10 @@ int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid,
  *    - is a file or oan empty directory
  * then the dentry is unhashed (d_delete()).
  */
-int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
+int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
 			     u64 child_nodeid, struct qstr *name);
 
-int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
 		 bool isdir);
 
 /**
@@ -1093,4 +1194,20 @@ unsigned int fuse_len_args(unsigned int numargs, struct fuse_arg *args);
 u64 fuse_get_unique(struct fuse_iqueue *fiq);
 void fuse_free_conn(struct fuse_conn *fc);
 
+/* dax.c */
+
+#define FUSE_IS_DAX(inode) (IS_ENABLED(CONFIG_FUSE_DAX) && IS_DAX(inode))
+
+ssize_t fuse_dax_read_iter(struct kiocb *iocb, struct iov_iter *to);
+ssize_t fuse_dax_write_iter(struct kiocb *iocb, struct iov_iter *from);
+int fuse_dax_mmap(struct file *file, struct vm_area_struct *vma);
+int fuse_dax_break_layouts(struct inode *inode, u64 dmap_start, u64 dmap_end);
+int fuse_dax_conn_alloc(struct fuse_conn *fc, struct dax_device *dax_dev);
+void fuse_dax_conn_free(struct fuse_conn *fc);
+bool fuse_dax_inode_alloc(struct super_block *sb, struct fuse_inode *fi);
+void fuse_dax_inode_init(struct inode *inode);
+void fuse_dax_inode_cleanup(struct inode *inode);
+bool fuse_dax_check_alignment(struct fuse_conn *fc, unsigned int map_alignment);
+void fuse_dax_cancel_work(struct fuse_conn *fc);
+
 #endif /* _FS_FUSE_I_H */

fs/fuse/inode.c

@@ -85,14 +85,22 @@ static struct inode *fuse_alloc_inode(struct super_block *sb)
 	fi->orig_ino = 0;
 	fi->state = 0;
 	mutex_init(&fi->mutex);
+	init_rwsem(&fi->i_mmap_sem);
 	spin_lock_init(&fi->lock);
 	fi->forget = fuse_alloc_forget();
-	if (!fi->forget) {
-		kmem_cache_free(fuse_inode_cachep, fi);
-		return NULL;
-	}
+	if (!fi->forget)
+		goto out_free;
+
+	if (IS_ENABLED(CONFIG_FUSE_DAX) && !fuse_dax_inode_alloc(sb, fi))
+		goto out_free_forget;
 
 	return &fi->inode;
+
+out_free_forget:
+	kfree(fi->forget);
+out_free:
+	kmem_cache_free(fuse_inode_cachep, fi);
+	return NULL;
 }
 
 static void fuse_free_inode(struct inode *inode)
@@ -101,6 +109,9 @@ static void fuse_free_inode(struct inode *inode)
 
 	mutex_destroy(&fi->mutex);
 	kfree(fi->forget);
+#ifdef CONFIG_FUSE_DAX
+	kfree(fi->dax);
+#endif
 	kmem_cache_free(fuse_inode_cachep, fi);
 }
 
@@ -112,9 +123,15 @@ static void fuse_evict_inode(struct inode *inode)
 	clear_inode(inode);
 	if (inode->i_sb->s_flags & SB_ACTIVE) {
 		struct fuse_conn *fc = get_fuse_conn(inode);
-		fuse_queue_forget(fc, fi->forget, fi->nodeid, fi->nlookup);
+
+		if (FUSE_IS_DAX(inode))
+			fuse_dax_inode_cleanup(inode);
+		if (fi->nlookup) {
+			fuse_queue_forget(fc, fi->forget, fi->nodeid,
+					  fi->nlookup);
 			fi->forget = NULL;
 		}
+	}
 	if (S_ISREG(inode->i_mode) && !is_bad_inode(inode)) {
 		WARN_ON(!list_empty(&fi->write_files));
 		WARN_ON(!list_empty(&fi->queued_writes));
@@ -268,7 +285,7 @@ static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr)
 		BUG();
 }
 
-int fuse_inode_eq(struct inode *inode, void *_nodeidp)
+static int fuse_inode_eq(struct inode *inode, void *_nodeidp)
 {
 	u64 nodeid = *(u64 *) _nodeidp;
 	if (get_node_id(inode) == nodeid)
@@ -292,7 +309,26 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid,
 	struct fuse_inode *fi;
 	struct fuse_conn *fc = get_fuse_conn_super(sb);
 
- retry:
+	/*
+	 * Auto mount points get their node id from the submount root, which is
+	 * not a unique identifier within this filesystem.
+	 *
+	 * To avoid conflicts, do not place submount points into the inode hash
+	 * table.
+	 */
+	if (fc->auto_submounts && (attr->flags & FUSE_ATTR_SUBMOUNT) &&
+	    S_ISDIR(attr->mode)) {
+		inode = new_inode(sb);
+		if (!inode)
+			return NULL;
+
+		fuse_init_inode(inode, attr);
+		get_fuse_inode(inode)->nodeid = nodeid;
+		inode->i_flags |= S_AUTOMOUNT;
+		goto done;
+	}
+
+retry:
 	inode = iget5_locked(sb, nodeid, fuse_inode_eq, fuse_inode_set, &nodeid);
 	if (!inode)
 		return NULL;
@@ -310,7 +346,7 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid,
 		iput(inode);
 		goto retry;
 	}
-
+done:
 	fi = get_fuse_inode(inode);
 	spin_lock(&fi->lock);
 	fi->nlookup++;
@@ -320,16 +356,37 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid,
 	return inode;
 }
 
-int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid,
+struct inode *fuse_ilookup(struct fuse_conn *fc, u64 nodeid,
+			   struct fuse_mount **fm)
+{
+	struct fuse_mount *fm_iter;
+	struct inode *inode;
+
+	WARN_ON(!rwsem_is_locked(&fc->killsb));
+	list_for_each_entry(fm_iter, &fc->mounts, fc_entry) {
+		if (!fm_iter->sb)
+			continue;
+
+		inode = ilookup5(fm_iter->sb, nodeid, fuse_inode_eq, &nodeid);
+		if (inode) {
+			if (fm)
+				*fm = fm_iter;
+			return inode;
+		}
+	}
+
+	return NULL;
+}
+
+int fuse_reverse_inval_inode(struct fuse_conn *fc, u64 nodeid,
 			     loff_t offset, loff_t len)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
 	struct fuse_inode *fi;
 	struct inode *inode;
 	pgoff_t pg_start;
 	pgoff_t pg_end;
 
-	inode = ilookup5(sb, nodeid, fuse_inode_eq, &nodeid);
+	inode = fuse_ilookup(fc, nodeid, NULL);
 	if (!inode)
 		return -ENOENT;
 
@@ -379,28 +436,23 @@ static void fuse_umount_begin(struct super_block *sb)
 		fuse_abort_conn(fc);
 }
 
-static void fuse_send_destroy(struct fuse_conn *fc)
+static void fuse_send_destroy(struct fuse_mount *fm)
 {
-	if (fc->conn_init) {
+	if (fm->fc->conn_init) {
 		FUSE_ARGS(args);
 
 		args.opcode = FUSE_DESTROY;
 		args.force = true;
 		args.nocreds = true;
-		fuse_simple_request(fc, &args);
+		fuse_simple_request(fm, &args);
 	}
 }
 
 static void fuse_put_super(struct super_block *sb)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
-
-	mutex_lock(&fuse_mutex);
-	list_del(&fc->entry);
-	fuse_ctl_remove_conn(fc);
-	mutex_unlock(&fuse_mutex);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
 
-	fuse_conn_put(fc);
+	fuse_mount_put(fm);
 }
 
 static void convert_fuse_statfs(struct kstatfs *stbuf, struct fuse_kstatfs *attr)
@@ -420,12 +472,12 @@ static void convert_fuse_statfs(struct kstatfs *stbuf, struct fuse_kstatfs *attr
 static int fuse_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
 	FUSE_ARGS(args);
 	struct fuse_statfs_out outarg;
 	int err;
 
-	if (!fuse_allow_current_process(fc)) {
+	if (!fuse_allow_current_process(fm->fc)) {
 		buf->f_type = FUSE_SUPER_MAGIC;
 		return 0;
 	}
@@ -437,7 +489,7 @@ static int fuse_statfs(struct dentry *dentry, struct kstatfs *buf)
 	args.out_numargs = 1;
 	args.out_args[0].size = sizeof(outarg);
 	args.out_args[0].value = &outarg;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (!err)
 		convert_fuse_statfs(buf, &outarg.st);
 	return err;
@@ -573,11 +625,11 @@ static int fuse_show_options(struct seq_file *m, struct dentry *root)
 	struct super_block *sb = root->d_sb;
 	struct fuse_conn *fc = get_fuse_conn_super(sb);
 
-	if (fc->no_mount_options)
-		return 0;
-
-	seq_printf(m, ",user_id=%u", from_kuid_munged(fc->user_ns, fc->user_id));
-	seq_printf(m, ",group_id=%u", from_kgid_munged(fc->user_ns, fc->group_id));
+	if (fc->legacy_opts_show) {
+		seq_printf(m, ",user_id=%u",
+			   from_kuid_munged(fc->user_ns, fc->user_id));
+		seq_printf(m, ",group_id=%u",
+			   from_kgid_munged(fc->user_ns, fc->group_id));
 		if (fc->default_permissions)
 			seq_puts(m, ",default_permissions");
 		if (fc->allow_other)
@@ -586,6 +638,12 @@ static int fuse_show_options(struct seq_file *m, struct dentry *root)
 			seq_printf(m, ",max_read=%u", fc->max_read);
 		if (sb->s_bdev && sb->s_blocksize != FUSE_DEFAULT_BLKSIZE)
 			seq_printf(m, ",blksize=%lu", sb->s_blocksize);
+	}
+#ifdef CONFIG_FUSE_DAX
+	if (fc->dax)
+		seq_puts(m, ",dax");
+#endif
+
 	return 0;
 }
 
@@ -615,7 +673,8 @@ static void fuse_pqueue_init(struct fuse_pqueue *fpq)
 	fpq->connected = 1;
 }
 
-void fuse_conn_init(struct fuse_conn *fc, struct user_namespace *user_ns,
+void fuse_conn_init(struct fuse_conn *fc, struct fuse_mount *fm,
+		    struct user_namespace *user_ns,
 		    const struct fuse_iqueue_ops *fiq_ops, void *fiq_priv)
 {
 	memset(fc, 0, sizeof(*fc));
@@ -642,6 +701,11 @@ void fuse_conn_init(struct fuse_conn *fc, struct user_namespace *user_ns,
 	fc->pid_ns = get_pid_ns(task_active_pid_ns(current));
 	fc->user_ns = get_user_ns(user_ns);
 	fc->max_pages = FUSE_DEFAULT_MAX_PAGES_PER_REQ;
+
+	INIT_LIST_HEAD(&fc->mounts);
+	list_add(&fm->fc_entry, &fc->mounts);
+	fm->fc = fc;
+	refcount_set(&fm->count, 1);
 }
 EXPORT_SYMBOL_GPL(fuse_conn_init);
 
@@ -650,6 +714,8 @@ void fuse_conn_put(struct fuse_conn *fc)
 	if (refcount_dec_and_test(&fc->count)) {
 		struct fuse_iqueue *fiq = &fc->iq;
 
+		if (IS_ENABLED(CONFIG_FUSE_DAX))
+			fuse_dax_conn_free(fc);
 		if (fiq->ops->release)
 			fiq->ops->release(fiq);
 		put_pid_ns(fc->pid_ns);
@@ -666,6 +732,23 @@ struct fuse_conn *fuse_conn_get(struct fuse_conn *fc)
 }
 EXPORT_SYMBOL_GPL(fuse_conn_get);
 
+void fuse_mount_put(struct fuse_mount *fm)
+{
+	if (refcount_dec_and_test(&fm->count)) {
+		if (fm->fc)
+			fuse_conn_put(fm->fc);
+		kfree(fm);
+	}
+}
+EXPORT_SYMBOL_GPL(fuse_mount_put);
+
+struct fuse_mount *fuse_mount_get(struct fuse_mount *fm)
+{
+	refcount_inc(&fm->count);
+	return fm;
+}
+EXPORT_SYMBOL_GPL(fuse_mount_get);
+
 static struct inode *fuse_get_root_inode(struct super_block *sb, unsigned mode)
 {
 	struct fuse_attr attr;
@@ -895,14 +978,16 @@ struct fuse_init_args {
 	struct fuse_init_out out;
 };
 
-static void process_init_reply(struct fuse_conn *fc, struct fuse_args *args,
+static void process_init_reply(struct fuse_mount *fm, struct fuse_args *args,
 			       int error)
 {
+	struct fuse_conn *fc = fm->fc;
 	struct fuse_init_args *ia = container_of(args, typeof(*ia), args);
 	struct fuse_init_out *arg = &ia->out;
+	bool ok = true;
 
 	if (error || arg->major != FUSE_KERNEL_VERSION)
-		fc->conn_error = 1;
+		ok = false;
 	else {
 		unsigned long ra_pages;
 
@@ -950,11 +1035,11 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_args *args,
 			if (arg->flags & FUSE_HANDLE_KILLPRIV)
 				fc->handle_killpriv = 1;
 			if (arg->time_gran && arg->time_gran <= 1000000000)
-				fc->sb->s_time_gran = arg->time_gran;
+				fm->sb->s_time_gran = arg->time_gran;
 			if ((arg->flags & FUSE_POSIX_ACL)) {
 				fc->default_permissions = 1;
 				fc->posix_acl = 1;
-				fc->sb->s_xattr = fuse_acl_xattr_handlers;
+				fm->sb->s_xattr = fuse_acl_xattr_handlers;
 			}
 			if (arg->flags & FUSE_CACHE_SYMLINKS)
 				fc->cache_symlinks = 1;
@@ -965,14 +1050,19 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_args *args,
 					min_t(unsigned int, FUSE_MAX_MAX_PAGES,
 					max_t(unsigned int, arg->max_pages, 1));
 			}
+			if (IS_ENABLED(CONFIG_FUSE_DAX) &&
+			    arg->flags & FUSE_MAP_ALIGNMENT &&
+			    !fuse_dax_check_alignment(fc, arg->map_alignment)) {
+				ok = false;
+			}
 		} else {
 			ra_pages = fc->max_read / PAGE_SIZE;
 			fc->no_lock = 1;
 			fc->no_flock = 1;
 		}
 
-		fc->sb->s_bdi->ra_pages =
-				min(fc->sb->s_bdi->ra_pages, ra_pages);
+		fm->sb->s_bdi->ra_pages =
+				min(fm->sb->s_bdi->ra_pages, ra_pages);
 		fc->minor = arg->minor;
 		fc->max_write = arg->minor < 5 ? 4096 : arg->max_write;
 		fc->max_write = max_t(unsigned, 4096, fc->max_write);
@@ -980,11 +1070,16 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_args *args,
 	}
 	kfree(ia);
 
+	if (!ok) {
+		fc->conn_init = 0;
+		fc->conn_error = 1;
+	}
+
 	fuse_set_initialized(fc);
 	wake_up_all(&fc->blocked_waitq);
 }
 
-void fuse_send_init(struct fuse_conn *fc)
+void fuse_send_init(struct fuse_mount *fm)
 {
 	struct fuse_init_args *ia;
 
@@ -992,7 +1087,7 @@ void fuse_send_init(struct fuse_conn *fc)
 
 	ia->in.major = FUSE_KERNEL_VERSION;
 	ia->in.minor = FUSE_KERNEL_MINOR_VERSION;
-	ia->in.max_readahead = fc->sb->s_bdi->ra_pages * PAGE_SIZE;
+	ia->in.max_readahead = fm->sb->s_bdi->ra_pages * PAGE_SIZE;
 	ia->in.flags |=
 		FUSE_ASYNC_READ | FUSE_POSIX_LOCKS | FUSE_ATOMIC_O_TRUNC |
 		FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES | FUSE_DONT_MASK |
@@ -1003,6 +1098,13 @@ void fuse_send_init(struct fuse_conn *fc)
 		FUSE_PARALLEL_DIROPS | FUSE_HANDLE_KILLPRIV | FUSE_POSIX_ACL |
 		FUSE_ABORT_ERROR | FUSE_MAX_PAGES | FUSE_CACHE_SYMLINKS |
 		FUSE_NO_OPENDIR_SUPPORT | FUSE_EXPLICIT_INVAL_DATA;
+#ifdef CONFIG_FUSE_DAX
+	if (fm->fc->dax)
+		ia->in.flags |= FUSE_MAP_ALIGNMENT;
+#endif
+	if (fm->fc->auto_submounts)
+		ia->in.flags |= FUSE_SUBMOUNTS;
+
 	ia->args.opcode = FUSE_INIT;
 	ia->args.in_numargs = 1;
 	ia->args.in_args[0].size = sizeof(ia->in);
@@ -1018,8 +1120,8 @@ void fuse_send_init(struct fuse_conn *fc)
 	ia->args.nocreds = true;
 	ia->args.end = process_init_reply;
 
-	if (fuse_simple_background(fc, &ia->args, GFP_KERNEL) != 0)
-		process_init_reply(fc, &ia->args, -ENOTCONN);
+	if (fuse_simple_background(fm, &ia->args, GFP_KERNEL) != 0)
+		process_init_reply(fm, &ia->args, -ENOTCONN);
 }
 EXPORT_SYMBOL_GPL(fuse_send_init);
 
@@ -1130,10 +1232,92 @@ void fuse_dev_free(struct fuse_dev *fud)
 }
 EXPORT_SYMBOL_GPL(fuse_dev_free);
 
+static void fuse_fill_attr_from_inode(struct fuse_attr *attr,
+				      const struct fuse_inode *fi)
+{
+	*attr = (struct fuse_attr){
+		.ino		= fi->inode.i_ino,
+		.size		= fi->inode.i_size,
+		.blocks		= fi->inode.i_blocks,
+		.atime		= fi->inode.i_atime.tv_sec,
+		.mtime		= fi->inode.i_mtime.tv_sec,
+		.ctime		= fi->inode.i_ctime.tv_sec,
+		.atimensec	= fi->inode.i_atime.tv_nsec,
+		.mtimensec	= fi->inode.i_mtime.tv_nsec,
+		.ctimensec	= fi->inode.i_ctime.tv_nsec,
+		.mode		= fi->inode.i_mode,
+		.nlink		= fi->inode.i_nlink,
+		.uid		= fi->inode.i_uid.val,
+		.gid		= fi->inode.i_gid.val,
+		.rdev		= fi->inode.i_rdev,
+		.blksize	= 1u << fi->inode.i_blkbits,
+	};
+}
+
+static void fuse_sb_defaults(struct super_block *sb)
+{
+	sb->s_magic = FUSE_SUPER_MAGIC;
+	sb->s_op = &fuse_super_operations;
+	sb->s_xattr = fuse_xattr_handlers;
+	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	sb->s_time_gran = 1;
+	sb->s_export_op = &fuse_export_operations;
+	sb->s_iflags |= SB_I_IMA_UNVERIFIABLE_SIGNATURE;
+	if (sb->s_user_ns != &init_user_ns)
+		sb->s_iflags |= SB_I_UNTRUSTED_MOUNTER;
+	sb->s_flags &= ~(SB_NOSEC | SB_I_VERSION);
+
+	/*
+	 * If we are not in the initial user namespace posix
+	 * acls must be translated.
+	 */
+	if (sb->s_user_ns != &init_user_ns)
+		sb->s_xattr = fuse_no_acl_xattr_handlers;
+}
+
+int fuse_fill_super_submount(struct super_block *sb,
+			     struct fuse_inode *parent_fi)
+{
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	struct super_block *parent_sb = parent_fi->inode.i_sb;
+	struct fuse_attr root_attr;
+	struct inode *root;
+
+	fuse_sb_defaults(sb);
+	fm->sb = sb;
+
+	WARN_ON(sb->s_bdi != &noop_backing_dev_info);
+	sb->s_bdi = bdi_get(parent_sb->s_bdi);
+
+	sb->s_xattr = parent_sb->s_xattr;
+	sb->s_time_gran = parent_sb->s_time_gran;
+	sb->s_blocksize = parent_sb->s_blocksize;
+	sb->s_blocksize_bits = parent_sb->s_blocksize_bits;
+	sb->s_subtype = kstrdup(parent_sb->s_subtype, GFP_KERNEL);
+	if (parent_sb->s_subtype && !sb->s_subtype)
+		return -ENOMEM;
+
+	fuse_fill_attr_from_inode(&root_attr, parent_fi);
+	root = fuse_iget(sb, parent_fi->nodeid, 0, &root_attr, 0, 0);
+	/*
+	 * This inode is just a duplicate, so it is not looked up and
+	 * its nlookup should not be incremented.  fuse_iget() does
+	 * that, though, so undo it here.
+	 */
+	get_fuse_inode(root)->nlookup--;
+	sb->s_d_op = &fuse_dentry_operations;
+	sb->s_root = d_make_root(root);
+	if (!sb->s_root)
+		return -ENOMEM;
+
+	return 0;
+}
+
 int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
 {
 	struct fuse_dev *fud = NULL;
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	struct fuse_conn *fc = fm->fc;
 	struct inode *root;
 	struct dentry *root_dentry;
 	int err;
@@ -1142,7 +1326,7 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
 	if (sb->s_flags & SB_MANDLOCK)
 		goto err;
 
-	sb->s_flags &= ~(SB_NOSEC | SB_I_VERSION);
+	fuse_sb_defaults(sb);
 
 	if (ctx->is_bdev) {
 #ifdef CONFIG_BLOCK
@@ -1157,32 +1341,21 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
 
 	sb->s_subtype = ctx->subtype;
 	ctx->subtype = NULL;
-	sb->s_magic = FUSE_SUPER_MAGIC;
-	sb->s_op = &fuse_super_operations;
-	sb->s_xattr = fuse_xattr_handlers;
-	sb->s_maxbytes = MAX_LFS_FILESIZE;
-	sb->s_time_gran = 1;
-	sb->s_export_op = &fuse_export_operations;
-	sb->s_iflags |= SB_I_IMA_UNVERIFIABLE_SIGNATURE;
-	if (sb->s_user_ns != &init_user_ns)
-		sb->s_iflags |= SB_I_UNTRUSTED_MOUNTER;
-
-	/*
-	 * If we are not in the initial user namespace posix
-	 * acls must be translated.
-	 */
-	if (sb->s_user_ns != &init_user_ns)
-		sb->s_xattr = fuse_no_acl_xattr_handlers;
+	if (IS_ENABLED(CONFIG_FUSE_DAX)) {
+		err = fuse_dax_conn_alloc(fc, ctx->dax_dev);
+		if (err)
+			goto err;
+	}
 
 	if (ctx->fudptr) {
 		err = -ENOMEM;
 		fud = fuse_dev_alloc_install(fc);
 		if (!fud)
-			goto err;
+			goto err_free_dax;
 	}
 
 	fc->dev = sb->s_dev;
-	fc->sb = sb;
+	fm->sb = sb;
 	err = fuse_bdi_init(fc, sb);
 	if (err)
 		goto err_dev_free;
@@ -1196,11 +1369,11 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
 	fc->allow_other = ctx->allow_other;
 	fc->user_id = ctx->user_id;
 	fc->group_id = ctx->group_id;
-	fc->max_read = max_t(unsigned, 4096, ctx->max_read);
+	fc->legacy_opts_show = ctx->legacy_opts_show;
+	fc->max_read = max_t(unsigned int, 4096, ctx->max_read);
 	fc->destroy = ctx->destroy;
 	fc->no_control = ctx->no_control;
 	fc->no_force_umount = ctx->no_force_umount;
-	fc->no_mount_options = ctx->no_mount_options;
 
 	err = -ENOMEM;
 	root = fuse_get_root_inode(sb, ctx->rootmode);
@@ -1233,6 +1406,9 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
  err_dev_free:
 	if (fud)
 		fuse_dev_free(fud);
+ err_free_dax:
+	if (IS_ENABLED(CONFIG_FUSE_DAX))
+		fuse_dax_conn_free(fc);
  err:
 	return err;
 }
@@ -1244,6 +1420,7 @@ static int fuse_fill_super(struct super_block *sb, struct fs_context *fsc)
 	struct file *file;
 	int err;
 	struct fuse_conn *fc;
+	struct fuse_mount *fm;
 
 	err = -EINVAL;
 	file = fget(ctx->fd);
@@ -1264,9 +1441,16 @@ static int fuse_fill_super(struct super_block *sb, struct fs_context *fsc)
 	if (!fc)
 		goto err_fput;
 
-	fuse_conn_init(fc, sb->s_user_ns, &fuse_dev_fiq_ops, NULL);
+	fm = kzalloc(sizeof(*fm), GFP_KERNEL);
+	if (!fm) {
+		kfree(fc);
+		goto err_fput;
+	}
+
+	fuse_conn_init(fc, fm, sb->s_user_ns, &fuse_dev_fiq_ops, NULL);
 	fc->release = fuse_free_conn;
-	sb->s_fs_info = fc;
+
+	sb->s_fs_info = fm;
 
 	err = fuse_fill_super_common(sb, ctx);
 	if (err)
@@ -1277,11 +1461,11 @@ static int fuse_fill_super(struct super_block *sb, struct fs_context *fsc)
 	 * CPUs after this
 	 */
 	fput(file);
-	fuse_send_init(get_fuse_conn_super(sb));
+	fuse_send_init(get_fuse_mount_super(sb));
 	return 0;
 
  err_put_conn:
-	fuse_conn_put(fc);
+	fuse_mount_put(fm);
 	sb->s_fs_info = NULL;
  err_fput:
 	fput(file);
@@ -1325,6 +1509,7 @@ static int fuse_init_fs_context(struct fs_context *fc)
 
 	ctx->max_read = ~0;
 	ctx->blksize = FUSE_DEFAULT_BLKSIZE;
+	ctx->legacy_opts_show = true;
 
 #ifdef CONFIG_BLOCK
 	if (fc->fs_type == &fuseblk_fs_type) {
@@ -1338,29 +1523,52 @@ static int fuse_init_fs_context(struct fs_context *fc)
 	return 0;
 }
 
-static void fuse_sb_destroy(struct super_block *sb)
+bool fuse_mount_remove(struct fuse_mount *fm)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_conn *fc = fm->fc;
+	bool last = false;
+
+	down_write(&fc->killsb);
+	list_del_init(&fm->fc_entry);
+	if (list_empty(&fc->mounts))
+		last = true;
+	up_write(&fc->killsb);
+
+	return last;
+}
+EXPORT_SYMBOL_GPL(fuse_mount_remove);
+
+void fuse_conn_destroy(struct fuse_mount *fm)
+{
+	struct fuse_conn *fc = fm->fc;
 
-	if (fc) {
 	if (fc->destroy)
-			fuse_send_destroy(fc);
+		fuse_send_destroy(fm);
 
 	fuse_abort_conn(fc);
 	fuse_wait_aborted(fc);
 
-		down_write(&fc->killsb);
-		fc->sb = NULL;
-		up_write(&fc->killsb);
+	if (!list_empty(&fc->entry)) {
+		mutex_lock(&fuse_mutex);
+		list_del(&fc->entry);
+		fuse_ctl_remove_conn(fc);
+		mutex_unlock(&fuse_mutex);
 	}
 }
+EXPORT_SYMBOL_GPL(fuse_conn_destroy);
 
-void fuse_kill_sb_anon(struct super_block *sb)
+static void fuse_kill_sb_anon(struct super_block *sb)
 {
-	fuse_sb_destroy(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	bool last;
+
+	if (fm) {
+		last = fuse_mount_remove(fm);
+		if (last)
+			fuse_conn_destroy(fm);
+	}
 	kill_anon_super(sb);
 }
-EXPORT_SYMBOL_GPL(fuse_kill_sb_anon);
 
 static struct file_system_type fuse_fs_type = {
 	.owner		= THIS_MODULE,
@@ -1375,7 +1583,14 @@ MODULE_ALIAS_FS("fuse");
 #ifdef CONFIG_BLOCK
 static void fuse_kill_sb_blk(struct super_block *sb)
 {
-	fuse_sb_destroy(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	bool last;
+
+	if (fm) {
+		last = fuse_mount_remove(fm);
+		if (last)
+			fuse_conn_destroy(fm);
+	}
 	kill_block_super(sb);
 }
 

fs/fuse/readdir.c

@@ -252,7 +252,7 @@ static int fuse_direntplus_link(struct file *file,
 static void fuse_force_forget(struct file *file, u64 nodeid)
 {
 	struct inode *inode = file_inode(file);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_forget_in inarg;
 	FUSE_ARGS(args);
 
@@ -266,7 +266,7 @@ static void fuse_force_forget(struct file *file, u64 nodeid)
 	args.force = true;
 	args.noreply = true;
 
-	fuse_simple_request(fc, &args);
+	fuse_simple_request(fm, &args);
 	/* ignore errors */
 }
 
@@ -320,7 +320,7 @@ static int fuse_readdir_uncached(struct file *file, struct dir_context *ctx)
 	ssize_t res;
 	struct page *page;
 	struct inode *inode = file_inode(file);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	struct fuse_io_args ia = {};
 	struct fuse_args_pages *ap = &ia.ap;
 	struct fuse_page_desc desc = { .length = PAGE_SIZE };
@@ -337,7 +337,7 @@ static int fuse_readdir_uncached(struct file *file, struct dir_context *ctx)
 	ap->pages = &page;
 	ap->descs = &desc;
 	if (plus) {
-		attr_version = fuse_get_attr_version(fc);
+		attr_version = fuse_get_attr_version(fm->fc);
 		fuse_read_args_fill(&ia, file, ctx->pos, PAGE_SIZE,
 				    FUSE_READDIRPLUS);
 	} else {
@@ -345,7 +345,7 @@ static int fuse_readdir_uncached(struct file *file, struct dir_context *ctx)
 				    FUSE_READDIR);
 	}
 	locked = fuse_lock_inode(inode);
-	res = fuse_simple_request(fc, &ap->args);
+	res = fuse_simple_request(fm, &ap->args);
 	fuse_unlock_inode(inode, locked);
 	if (res >= 0) {
 		if (!res) {

fs/fuse/virtio_fs.c

@@ -5,12 +5,17 @@
  */
 
 #include <linux/fs.h>
+#include <linux/dax.h>
+#include <linux/pci.h>
+#include <linux/pfn_t.h>
 #include <linux/module.h>
 #include <linux/virtio.h>
 #include <linux/virtio_fs.h>
 #include <linux/delay.h>
 #include <linux/fs_context.h>
+#include <linux/fs_parser.h>
 #include <linux/highmem.h>
+#include <linux/uio.h>
 #include "fuse_i.h"
 
 /* List of virtio-fs device instances and a lock for the list. Also provides
@@ -24,6 +29,8 @@ enum {
 	VQ_REQUEST
 };
 
+#define VQ_NAME_LEN	24
+
 /* Per-virtqueue state */
 struct virtio_fs_vq {
 	spinlock_t lock;
@@ -36,7 +43,7 @@ struct virtio_fs_vq {
 	bool connected;
 	long in_flight;
 	struct completion in_flight_zero; /* No inflight requests */
-	char name[24];
+	char name[VQ_NAME_LEN];
 } ____cacheline_aligned_in_smp;
 
 /* A virtio-fs device instance */
@@ -47,6 +54,12 @@ struct virtio_fs {
 	struct virtio_fs_vq *vqs;
 	unsigned int nvqs;               /* number of virtqueues */
 	unsigned int num_request_queues; /* number of request queues */
+	struct dax_device *dax_dev;
+
+	/* DAX memory window where file contents are mapped */
+	void *window_kaddr;
+	phys_addr_t window_phys_addr;
+	size_t window_len;
 };
 
 struct virtio_fs_forget_req {
@@ -69,6 +82,44 @@ struct virtio_fs_req_work {
 static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
 				 struct fuse_req *req, bool in_flight);
 
+enum {
+	OPT_DAX,
+};
+
+static const struct fs_parameter_spec virtio_fs_parameters[] = {
+	fsparam_flag("dax", OPT_DAX),
+	{}
+};
+
+static int virtio_fs_parse_param(struct fs_context *fc,
+				 struct fs_parameter *param)
+{
+	struct fs_parse_result result;
+	struct fuse_fs_context *ctx = fc->fs_private;
+	int opt;
+
+	opt = fs_parse(fc, virtio_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case OPT_DAX:
+		ctx->dax = 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void virtio_fs_free_fc(struct fs_context *fc)
+{
+	struct fuse_fs_context *ctx = fc->fs_private;
+
+	kfree(ctx);
+}
+
 static inline struct virtio_fs_vq *vq_to_fsvq(struct virtqueue *vq)
 {
 	struct virtio_fs *fs = vq->vdev->priv;
@@ -289,7 +340,6 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
 	struct fuse_req *req;
 	struct virtio_fs_vq *fsvq = container_of(work, struct virtio_fs_vq,
 						 dispatch_work.work);
-	struct fuse_conn *fc = fsvq->fud->fc;
 	int ret;
 
 	pr_debug("virtio-fs: worker %s called.\n", __func__);
@@ -304,7 +354,7 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
 
 		list_del_init(&req->list);
 		spin_unlock(&fsvq->lock);
-		fuse_request_end(fc, req);
+		fuse_request_end(req);
 	}
 
 	/* Dispatch pending requests */
@@ -335,7 +385,7 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
 			spin_unlock(&fsvq->lock);
 			pr_err("virtio-fs: virtio_fs_enqueue_req() failed %d\n",
 			       ret);
-			fuse_request_end(fc, req);
+			fuse_request_end(req);
 		}
 	}
 }
@@ -495,7 +545,6 @@ static void virtio_fs_request_complete(struct fuse_req *req,
 				       struct virtio_fs_vq *fsvq)
 {
 	struct fuse_pqueue *fpq = &fsvq->fud->pq;
-	struct fuse_conn *fc = fsvq->fud->fc;
 	struct fuse_args *args;
 	struct fuse_args_pages *ap;
 	unsigned int len, i, thislen;
@@ -528,7 +577,7 @@ static void virtio_fs_request_complete(struct fuse_req *req,
 	clear_bit(FR_SENT, &req->flags);
 	spin_unlock(&fpq->lock);
 
-	fuse_request_end(fc, req);
+	fuse_request_end(req);
 	spin_lock(&fsvq->lock);
 	dec_in_flight_req(fsvq);
 	spin_unlock(&fsvq->lock);
@@ -596,6 +645,26 @@ static void virtio_fs_vq_done(struct virtqueue *vq)
 	schedule_work(&fsvq->done_work);
 }
 
+static void virtio_fs_init_vq(struct virtio_fs_vq *fsvq, char *name,
+			      int vq_type)
+{
+	strncpy(fsvq->name, name, VQ_NAME_LEN);
+	spin_lock_init(&fsvq->lock);
+	INIT_LIST_HEAD(&fsvq->queued_reqs);
+	INIT_LIST_HEAD(&fsvq->end_reqs);
+	init_completion(&fsvq->in_flight_zero);
+
+	if (vq_type == VQ_REQUEST) {
+		INIT_WORK(&fsvq->done_work, virtio_fs_requests_done_work);
+		INIT_DELAYED_WORK(&fsvq->dispatch_work,
+				  virtio_fs_request_dispatch_work);
+	} else {
+		INIT_WORK(&fsvq->done_work, virtio_fs_hiprio_done_work);
+		INIT_DELAYED_WORK(&fsvq->dispatch_work,
+				  virtio_fs_hiprio_dispatch_work);
+	}
+}
+
 /* Initialize virtqueues */
 static int virtio_fs_setup_vqs(struct virtio_device *vdev,
 			       struct virtio_fs *fs)
@@ -611,7 +680,7 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev,
 	if (fs->num_request_queues == 0)
 		return -EINVAL;
 
-	fs->nvqs = 1 + fs->num_request_queues;
+	fs->nvqs = VQ_REQUEST + fs->num_request_queues;
 	fs->vqs = kcalloc(fs->nvqs, sizeof(fs->vqs[VQ_HIPRIO]), GFP_KERNEL);
 	if (!fs->vqs)
 		return -ENOMEM;
@@ -625,29 +694,17 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev,
 		goto out;
 	}
 
+	/* Initialize the hiprio/forget request virtqueue */
 	callbacks[VQ_HIPRIO] = virtio_fs_vq_done;
-	snprintf(fs->vqs[VQ_HIPRIO].name, sizeof(fs->vqs[VQ_HIPRIO].name),
-			"hiprio");
+	virtio_fs_init_vq(&fs->vqs[VQ_HIPRIO], "hiprio", VQ_HIPRIO);
 	names[VQ_HIPRIO] = fs->vqs[VQ_HIPRIO].name;
-	INIT_WORK(&fs->vqs[VQ_HIPRIO].done_work, virtio_fs_hiprio_done_work);
-	INIT_LIST_HEAD(&fs->vqs[VQ_HIPRIO].queued_reqs);
-	INIT_LIST_HEAD(&fs->vqs[VQ_HIPRIO].end_reqs);
-	INIT_DELAYED_WORK(&fs->vqs[VQ_HIPRIO].dispatch_work,
-			virtio_fs_hiprio_dispatch_work);
-	init_completion(&fs->vqs[VQ_HIPRIO].in_flight_zero);
-	spin_lock_init(&fs->vqs[VQ_HIPRIO].lock);
 
 	/* Initialize the requests virtqueues */
 	for (i = VQ_REQUEST; i < fs->nvqs; i++) {
-		spin_lock_init(&fs->vqs[i].lock);
-		INIT_WORK(&fs->vqs[i].done_work, virtio_fs_requests_done_work);
-		INIT_DELAYED_WORK(&fs->vqs[i].dispatch_work,
-				  virtio_fs_request_dispatch_work);
-		INIT_LIST_HEAD(&fs->vqs[i].queued_reqs);
-		INIT_LIST_HEAD(&fs->vqs[i].end_reqs);
-		init_completion(&fs->vqs[i].in_flight_zero);
-		snprintf(fs->vqs[i].name, sizeof(fs->vqs[i].name),
-			 "requests.%u", i - VQ_REQUEST);
+		char vq_name[VQ_NAME_LEN];
+
+		snprintf(vq_name, VQ_NAME_LEN, "requests.%u", i - VQ_REQUEST);
+		virtio_fs_init_vq(&fs->vqs[i], vq_name, VQ_REQUEST);
 		callbacks[i] = virtio_fs_vq_done;
 		names[i] = fs->vqs[i].name;
 	}
@@ -676,6 +733,130 @@ static void virtio_fs_cleanup_vqs(struct virtio_device *vdev,
 	vdev->config->del_vqs(vdev);
 }
 
+/* Map a window offset to a page frame number.  The window offset will have
+ * been produced by .iomap_begin(), which maps a file offset to a window
+ * offset.
+ */
+static long virtio_fs_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
+				    long nr_pages, void **kaddr, pfn_t *pfn)
+{
+	struct virtio_fs *fs = dax_get_private(dax_dev);
+	phys_addr_t offset = PFN_PHYS(pgoff);
+	size_t max_nr_pages = fs->window_len/PAGE_SIZE - pgoff;
+
+	if (kaddr)
+		*kaddr = fs->window_kaddr + offset;
+	if (pfn)
+		*pfn = phys_to_pfn_t(fs->window_phys_addr + offset,
+					PFN_DEV | PFN_MAP);
+	return nr_pages > max_nr_pages ? max_nr_pages : nr_pages;
+}
+
+static size_t virtio_fs_copy_from_iter(struct dax_device *dax_dev,
+				       pgoff_t pgoff, void *addr,
+				       size_t bytes, struct iov_iter *i)
+{
+	return copy_from_iter(addr, bytes, i);
+}
+
+static size_t virtio_fs_copy_to_iter(struct dax_device *dax_dev,
+				       pgoff_t pgoff, void *addr,
+				       size_t bytes, struct iov_iter *i)
+{
+	return copy_to_iter(addr, bytes, i);
+}
+
+static int virtio_fs_zero_page_range(struct dax_device *dax_dev,
+				     pgoff_t pgoff, size_t nr_pages)
+{
+	long rc;
+	void *kaddr;
+
+	rc = dax_direct_access(dax_dev, pgoff, nr_pages, &kaddr, NULL);
+	if (rc < 0)
+		return rc;
+	memset(kaddr, 0, nr_pages << PAGE_SHIFT);
+	dax_flush(dax_dev, kaddr, nr_pages << PAGE_SHIFT);
+	return 0;
+}
+
+static const struct dax_operations virtio_fs_dax_ops = {
+	.direct_access = virtio_fs_direct_access,
+	.copy_from_iter = virtio_fs_copy_from_iter,
+	.copy_to_iter = virtio_fs_copy_to_iter,
+	.zero_page_range = virtio_fs_zero_page_range,
+};
+
+static void virtio_fs_cleanup_dax(void *data)
+{
+	struct dax_device *dax_dev = data;
+
+	kill_dax(dax_dev);
+	put_dax(dax_dev);
+}
+
+static int virtio_fs_setup_dax(struct virtio_device *vdev, struct virtio_fs *fs)
+{
+	struct virtio_shm_region cache_reg;
+	struct dev_pagemap *pgmap;
+	bool have_cache;
+
+	if (!IS_ENABLED(CONFIG_FUSE_DAX))
+		return 0;
+
+	/* Get cache region */
+	have_cache = virtio_get_shm_region(vdev, &cache_reg,
+					   (u8)VIRTIO_FS_SHMCAP_ID_CACHE);
+	if (!have_cache) {
+		dev_notice(&vdev->dev, "%s: No cache capability\n", __func__);
+		return 0;
+	}
+
+	if (!devm_request_mem_region(&vdev->dev, cache_reg.addr, cache_reg.len,
+				     dev_name(&vdev->dev))) {
+		dev_warn(&vdev->dev, "could not reserve region addr=0x%llx len=0x%llx\n",
+			 cache_reg.addr, cache_reg.len);
+		return -EBUSY;
+	}
+
+	dev_notice(&vdev->dev, "Cache len: 0x%llx @ 0x%llx\n", cache_reg.len,
+		   cache_reg.addr);
+
+	pgmap = devm_kzalloc(&vdev->dev, sizeof(*pgmap), GFP_KERNEL);
+	if (!pgmap)
+		return -ENOMEM;
+
+	pgmap->type = MEMORY_DEVICE_FS_DAX;
+
+	/* Ideally we would directly use the PCI BAR resource but
+	 * devm_memremap_pages() wants its own copy in pgmap.  So
+	 * initialize a struct resource from scratch (only the start
+	 * and end fields will be used).
+	 */
+	pgmap->range = (struct range) {
+		.start = (phys_addr_t) cache_reg.addr,
+		.end = (phys_addr_t) cache_reg.addr + cache_reg.len - 1,
+	};
+	pgmap->nr_range = 1;
+
+	fs->window_kaddr = devm_memremap_pages(&vdev->dev, pgmap);
+	if (IS_ERR(fs->window_kaddr))
+		return PTR_ERR(fs->window_kaddr);
+
+	fs->window_phys_addr = (phys_addr_t) cache_reg.addr;
+	fs->window_len = (phys_addr_t) cache_reg.len;
+
+	dev_dbg(&vdev->dev, "%s: window kaddr 0x%px phys_addr 0x%llx len 0x%llx\n",
+		__func__, fs->window_kaddr, cache_reg.addr, cache_reg.len);
+
+	fs->dax_dev = alloc_dax(fs, NULL, &virtio_fs_dax_ops, 0);
+	if (IS_ERR(fs->dax_dev))
+		return PTR_ERR(fs->dax_dev);
+
+	return devm_add_action_or_reset(&vdev->dev, virtio_fs_cleanup_dax,
+					fs->dax_dev);
+}
+
 static int virtio_fs_probe(struct virtio_device *vdev)
 {
 	struct virtio_fs *fs;
@@ -697,6 +878,10 @@ static int virtio_fs_probe(struct virtio_device *vdev)
 
 	/* TODO vq affinity */
 
+	ret = virtio_fs_setup_dax(vdev, fs);
+	if (ret < 0)
+		goto out_vqs;
+
 	/* Bring the device online in case the filesystem is mounted and
 	 * requests need to be sent before we return.
 	 */
@@ -833,18 +1018,37 @@ __releases(fiq->lock)
 	spin_unlock(&fiq->lock);
 }
 
+/* Count number of scatter-gather elements required */
+static unsigned int sg_count_fuse_pages(struct fuse_page_desc *page_descs,
+				       unsigned int num_pages,
+				       unsigned int total_len)
+{
+	unsigned int i;
+	unsigned int this_len;
+
+	for (i = 0; i < num_pages && total_len; i++) {
+		this_len =  min(page_descs[i].length, total_len);
+		total_len -= this_len;
+	}
+
+	return i;
+}
+
 /* Return the number of scatter-gather list elements required */
 static unsigned int sg_count_fuse_req(struct fuse_req *req)
 {
 	struct fuse_args *args = req->args;
 	struct fuse_args_pages *ap = container_of(args, typeof(*ap), args);
-	unsigned int total_sgs = 1 /* fuse_in_header */;
+	unsigned int size, total_sgs = 1 /* fuse_in_header */;
 
 	if (args->in_numargs - args->in_pages)
 		total_sgs += 1;
 
-	if (args->in_pages)
-		total_sgs += ap->num_pages;
+	if (args->in_pages) {
+		size = args->in_args[args->in_numargs - 1].size;
+		total_sgs += sg_count_fuse_pages(ap->descs, ap->num_pages,
+						 size);
+	}
 
 	if (!test_bit(FR_ISREPLY, &req->flags))
 		return total_sgs;
@@ -854,8 +1058,11 @@ static unsigned int sg_count_fuse_req(struct fuse_req *req)
 	if (args->out_numargs - args->out_pages)
 		total_sgs += 1;
 
-	if (args->out_pages)
-		total_sgs += ap->num_pages;
+	if (args->out_pages) {
+		size = args->out_args[args->out_numargs - 1].size;
+		total_sgs += sg_count_fuse_pages(ap->descs, ap->num_pages,
+						 size);
+	}
 
 	return total_sgs;
 }
@@ -1071,24 +1278,28 @@ static const struct fuse_iqueue_ops virtio_fs_fiq_ops = {
 	.release			= virtio_fs_fiq_release,
 };
 
-static int virtio_fs_fill_super(struct super_block *sb)
+static inline void virtio_fs_ctx_set_defaults(struct fuse_fs_context *ctx)
+{
+	ctx->rootmode = S_IFDIR;
+	ctx->default_permissions = 1;
+	ctx->allow_other = 1;
+	ctx->max_read = UINT_MAX;
+	ctx->blksize = 512;
+	ctx->destroy = true;
+	ctx->no_control = true;
+	ctx->no_force_umount = true;
+}
+
+static int virtio_fs_fill_super(struct super_block *sb, struct fs_context *fsc)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	struct fuse_conn *fc = fm->fc;
 	struct virtio_fs *fs = fc->iq.priv;
+	struct fuse_fs_context *ctx = fsc->fs_private;
 	unsigned int i;
 	int err;
-	struct fuse_fs_context ctx = {
-		.rootmode = S_IFDIR,
-		.default_permissions = 1,
-		.allow_other = 1,
-		.max_read = UINT_MAX,
-		.blksize = 512,
-		.destroy = true,
-		.no_control = true,
-		.no_force_umount = true,
-		.no_mount_options = true,
-	};
 
+	virtio_fs_ctx_set_defaults(ctx);
 	mutex_lock(&virtio_fs_mutex);
 
 	/* After holding mutex, make sure virtiofs device is still there.
@@ -1112,8 +1323,10 @@ static int virtio_fs_fill_super(struct super_block *sb)
 	}
 
 	/* virtiofs allocates and installs its own fuse devices */
-	ctx.fudptr = NULL;
-	err = fuse_fill_super_common(sb, &ctx);
+	ctx->fudptr = NULL;
+	if (ctx->dax)
+		ctx->dax_dev = fs->dax_dev;
+	err = fuse_fill_super_common(sb, ctx);
 	if (err < 0)
 		goto err_free_fuse_devs;
 
@@ -1125,7 +1338,7 @@ static int virtio_fs_fill_super(struct super_block *sb)
 
 	/* Previous unmount will stop all queues. Start these again */
 	virtio_fs_start_all_queues(fs);
-	fuse_send_init(fc);
+	fuse_send_init(fm);
 	mutex_unlock(&virtio_fs_mutex);
 	return 0;
 
@@ -1136,18 +1349,17 @@ static int virtio_fs_fill_super(struct super_block *sb)
 	return err;
 }
 
-static void virtio_kill_sb(struct super_block *sb)
+static void virtio_fs_conn_destroy(struct fuse_mount *fm)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
-	struct virtio_fs *vfs;
-	struct virtio_fs_vq *fsvq;
+	struct fuse_conn *fc = fm->fc;
+	struct virtio_fs *vfs = fc->iq.priv;
+	struct virtio_fs_vq *fsvq = &vfs->vqs[VQ_HIPRIO];
 
-	/* If mount failed, we can still be called without any fc */
-	if (!fc)
-		return fuse_kill_sb_anon(sb);
-
-	vfs = fc->iq.priv;
-	fsvq = &vfs->vqs[VQ_HIPRIO];
+	/* Stop dax worker. Soon evict_inodes() will be called which
+	 * will free all memory ranges belonging to all inodes.
+	 */
+	if (IS_ENABLED(CONFIG_FUSE_DAX))
+		fuse_dax_cancel_work(fc);
 
 	/* Stop forget queue. Soon destroy will be sent */
 	spin_lock(&fsvq->lock);
@@ -1155,9 +1367,9 @@ static void virtio_kill_sb(struct super_block *sb)
 	spin_unlock(&fsvq->lock);
 	virtio_fs_drain_all_queues(vfs);
 
-	fuse_kill_sb_anon(sb);
+	fuse_conn_destroy(fm);
 
-	/* fuse_kill_sb_anon() must have sent destroy. Stop all queues
+	/* fuse_conn_destroy() must have sent destroy. Stop all queues
 	 * and drain one more time and free fuse devices. Freeing fuse
 	 * devices will drop their reference on fuse_conn and that in
 	 * turn will drop its reference on virtio_fs object.
@@ -1167,12 +1379,27 @@ static void virtio_kill_sb(struct super_block *sb)
 	virtio_fs_free_devs(vfs);
 }
 
+static void virtio_kill_sb(struct super_block *sb)
+{
+	struct fuse_mount *fm = get_fuse_mount_super(sb);
+	bool last;
+
+	/* If mount failed, we can still be called without any fc */
+	if (fm) {
+		last = fuse_mount_remove(fm);
+		if (last)
+			virtio_fs_conn_destroy(fm);
+	}
+	kill_anon_super(sb);
+}
+
 static int virtio_fs_test_super(struct super_block *sb,
 				struct fs_context *fsc)
 {
-	struct fuse_conn *fc = fsc->s_fs_info;
+	struct fuse_mount *fsc_fm = fsc->s_fs_info;
+	struct fuse_mount *sb_fm = get_fuse_mount_super(sb);
 
-	return fc->iq.priv == get_fuse_conn_super(sb)->iq.priv;
+	return fsc_fm->fc->iq.priv == sb_fm->fc->iq.priv;
 }
 
 static int virtio_fs_set_super(struct super_block *sb,
@@ -1182,7 +1409,7 @@ static int virtio_fs_set_super(struct super_block *sb,
 
 	err = get_anon_bdev(&sb->s_dev);
 	if (!err)
-		fuse_conn_get(fsc->s_fs_info);
+		fuse_mount_get(fsc->s_fs_info);
 
 	return err;
 }
@@ -1192,6 +1419,7 @@ static int virtio_fs_get_tree(struct fs_context *fsc)
 	struct virtio_fs *fs;
 	struct super_block *sb;
 	struct fuse_conn *fc;
+	struct fuse_mount *fm;
 	int err;
 
 	/* This gets a reference on virtio_fs object. This ptr gets installed
@@ -1212,19 +1440,29 @@ static int virtio_fs_get_tree(struct fs_context *fsc)
 		return -ENOMEM;
 	}
 
-	fuse_conn_init(fc, get_user_ns(current_user_ns()), &virtio_fs_fiq_ops,
-		       fs);
+	fm = kzalloc(sizeof(struct fuse_mount), GFP_KERNEL);
+	if (!fm) {
+		mutex_lock(&virtio_fs_mutex);
+		virtio_fs_put(fs);
+		mutex_unlock(&virtio_fs_mutex);
+		kfree(fc);
+		return -ENOMEM;
+	}
+
+	fuse_conn_init(fc, fm, get_user_ns(current_user_ns()),
+		       &virtio_fs_fiq_ops, fs);
 	fc->release = fuse_free_conn;
 	fc->delete_stale = true;
+	fc->auto_submounts = true;
 
-	fsc->s_fs_info = fc;
+	fsc->s_fs_info = fm;
 	sb = sget_fc(fsc, virtio_fs_test_super, virtio_fs_set_super);
-	fuse_conn_put(fc);
+	fuse_mount_put(fm);
 	if (IS_ERR(sb))
 		return PTR_ERR(sb);
 
 	if (!sb->s_root) {
-		err = virtio_fs_fill_super(sb);
+		err = virtio_fs_fill_super(sb, fsc);
 		if (err) {
 			deactivate_locked_super(sb);
 			return err;
@@ -1239,11 +1477,19 @@ static int virtio_fs_get_tree(struct fs_context *fsc)
 }
 
 static const struct fs_context_operations virtio_fs_context_ops = {
+	.free		= virtio_fs_free_fc,
+	.parse_param	= virtio_fs_parse_param,
 	.get_tree	= virtio_fs_get_tree,
 };
 
 static int virtio_fs_init_fs_context(struct fs_context *fsc)
 {
+	struct fuse_fs_context *ctx;
+
+	ctx = kzalloc(sizeof(struct fuse_fs_context), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	fsc->fs_private = ctx;
 	fsc->ops = &virtio_fs_context_ops;
 	return 0;
 }

fs/fuse/xattr.c

@@ -14,12 +14,12 @@
 int fuse_setxattr(struct inode *inode, const char *name, const void *value,
 		  size_t size, int flags)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_setxattr_in inarg;
 	int err;
 
-	if (fc->no_setxattr)
+	if (fm->fc->no_setxattr)
 		return -EOPNOTSUPP;
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -34,9 +34,9 @@ int fuse_setxattr(struct inode *inode, const char *name, const void *value,
 	args.in_args[1].value = name;
 	args.in_args[2].size = size;
 	args.in_args[2].value = value;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
-		fc->no_setxattr = 1;
+		fm->fc->no_setxattr = 1;
 		err = -EOPNOTSUPP;
 	}
 	if (!err) {
@@ -49,13 +49,13 @@ int fuse_setxattr(struct inode *inode, const char *name, const void *value,
 ssize_t fuse_getxattr(struct inode *inode, const char *name, void *value,
 		      size_t size)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_getxattr_in inarg;
 	struct fuse_getxattr_out outarg;
 	ssize_t ret;
 
-	if (fc->no_getxattr)
+	if (fm->fc->no_getxattr)
 		return -EOPNOTSUPP;
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -77,11 +77,11 @@ ssize_t fuse_getxattr(struct inode *inode, const char *name, void *value,
 		args.out_args[0].size = sizeof(outarg);
 		args.out_args[0].value = &outarg;
 	}
-	ret = fuse_simple_request(fc, &args);
+	ret = fuse_simple_request(fm, &args);
 	if (!ret && !size)
 		ret = min_t(ssize_t, outarg.size, XATTR_SIZE_MAX);
 	if (ret == -ENOSYS) {
-		fc->no_getxattr = 1;
+		fm->fc->no_getxattr = 1;
 		ret = -EOPNOTSUPP;
 	}
 	return ret;
@@ -107,16 +107,16 @@ static int fuse_verify_xattr_list(char *list, size_t size)
 ssize_t fuse_listxattr(struct dentry *entry, char *list, size_t size)
 {
 	struct inode *inode = d_inode(entry);
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	struct fuse_getxattr_in inarg;
 	struct fuse_getxattr_out outarg;
 	ssize_t ret;
 
-	if (!fuse_allow_current_process(fc))
+	if (!fuse_allow_current_process(fm->fc))
 		return -EACCES;
 
-	if (fc->no_listxattr)
+	if (fm->fc->no_listxattr)
 		return -EOPNOTSUPP;
 
 	memset(&inarg, 0, sizeof(inarg));
@@ -136,13 +136,13 @@ ssize_t fuse_listxattr(struct dentry *entry, char *list, size_t size)
 		args.out_args[0].size = sizeof(outarg);
 		args.out_args[0].value = &outarg;
 	}
-	ret = fuse_simple_request(fc, &args);
+	ret = fuse_simple_request(fm, &args);
 	if (!ret && !size)
 		ret = min_t(ssize_t, outarg.size, XATTR_LIST_MAX);
 	if (ret > 0 && size)
 		ret = fuse_verify_xattr_list(list, ret);
 	if (ret == -ENOSYS) {
-		fc->no_listxattr = 1;
+		fm->fc->no_listxattr = 1;
 		ret = -EOPNOTSUPP;
 	}
 	return ret;
@@ -150,11 +150,11 @@ ssize_t fuse_listxattr(struct dentry *entry, char *list, size_t size)
 
 int fuse_removexattr(struct inode *inode, const char *name)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_mount *fm = get_fuse_mount(inode);
 	FUSE_ARGS(args);
 	int err;
 
-	if (fc->no_removexattr)
+	if (fm->fc->no_removexattr)
 		return -EOPNOTSUPP;
 
 	args.opcode = FUSE_REMOVEXATTR;
@@ -162,9 +162,9 @@ int fuse_removexattr(struct inode *inode, const char *name)
 	args.in_numargs = 1;
 	args.in_args[0].size = strlen(name) + 1;
 	args.in_args[0].value = name;
-	err = fuse_simple_request(fc, &args);
+	err = fuse_simple_request(fm, &args);
 	if (err == -ENOSYS) {
-		fc->no_removexattr = 1;
+		fm->fc->no_removexattr = 1;
 		err = -EOPNOTSUPP;
 	}
 	if (!err) {

include/linux/dax.h

@@ -149,6 +149,7 @@ int dax_writeback_mapping_range(struct address_space *mapping,
 		struct dax_device *dax_dev, struct writeback_control *wbc);
 
 struct page *dax_layout_busy_page(struct address_space *mapping);
+struct page *dax_layout_busy_page_range(struct address_space *mapping, loff_t start, loff_t end);
 dax_entry_t dax_lock_page(struct page *page);
 void dax_unlock_page(struct page *page, dax_entry_t cookie);
 #else
@@ -179,6 +180,11 @@ static inline struct page *dax_layout_busy_page(struct address_space *mapping)
 	return NULL;
 }
 
+static inline struct page *dax_layout_busy_page_range(struct address_space *mapping, pgoff_t start, pgoff_t nr_pages)
+{
+	return NULL;
+}
+
 static inline int dax_writeback_mapping_range(struct address_space *mapping,
 		struct dax_device *dax_dev, struct writeback_control *wbc)
 {

include/uapi/linux/fuse.h

@@ -172,6 +172,9 @@
  *  - add FUSE_WRITE_KILL_PRIV flag
  *  - add FUSE_SETUPMAPPING and FUSE_REMOVEMAPPING
  *  - add map_alignment to fuse_init_out, add FUSE_MAP_ALIGNMENT flag
+ *
+ *  7.32
+ *  - add flags to fuse_attr, add FUSE_ATTR_SUBMOUNT, add FUSE_SUBMOUNTS
  */
 
 #ifndef _LINUX_FUSE_H
@@ -207,7 +210,7 @@
 #define FUSE_KERNEL_VERSION 7
 
 /** Minor version number of this interface */
-#define FUSE_KERNEL_MINOR_VERSION 31
+#define FUSE_KERNEL_MINOR_VERSION 32
 
 /** The node ID of the root inode */
 #define FUSE_ROOT_ID 1
@@ -231,7 +234,7 @@ struct fuse_attr {
 	uint32_t	gid;
 	uint32_t	rdev;
 	uint32_t	blksize;
-	uint32_t	padding;
+	uint32_t	flags;
 };
 
 struct fuse_kstatfs {
@@ -313,7 +316,10 @@ struct fuse_file_lock {
  * FUSE_CACHE_SYMLINKS: cache READLINK responses
  * FUSE_NO_OPENDIR_SUPPORT: kernel supports zero-message opendir
  * FUSE_EXPLICIT_INVAL_DATA: only invalidate cached pages on explicit request
- * FUSE_MAP_ALIGNMENT: map_alignment field is valid
+ * FUSE_MAP_ALIGNMENT: init_out.map_alignment contains log2(byte alignment) for
+ *		       foffset and moffset fields in struct
+ *		       fuse_setupmapping_out and fuse_removemapping_one.
+ * FUSE_SUBMOUNTS: kernel supports auto-mounting directory submounts
  */
 #define FUSE_ASYNC_READ		(1 << 0)
 #define FUSE_POSIX_LOCKS	(1 << 1)
@@ -342,6 +348,7 @@ struct fuse_file_lock {
 #define FUSE_NO_OPENDIR_SUPPORT (1 << 24)
 #define FUSE_EXPLICIT_INVAL_DATA (1 << 25)
 #define FUSE_MAP_ALIGNMENT	(1 << 26)
+#define FUSE_SUBMOUNTS		(1 << 27)
 
 /**
  * CUSE INIT request/reply flags
@@ -417,6 +424,13 @@ struct fuse_file_lock {
  */
 #define FUSE_FSYNC_FDATASYNC	(1 << 0)
 
+/**
+ * fuse_attr flags
+ *
+ * FUSE_ATTR_SUBMOUNT: Object is a submount root
+ */
+#define FUSE_ATTR_SUBMOUNT      (1 << 0)
+
 enum fuse_opcode {
 	FUSE_LOOKUP		= 1,
 	FUSE_FORGET		= 2,  /* no reply */
@@ -892,4 +906,34 @@ struct fuse_copy_file_range_in {
 	uint64_t	flags;
 };
 
+#define FUSE_SETUPMAPPING_FLAG_WRITE (1ull << 0)
+#define FUSE_SETUPMAPPING_FLAG_READ (1ull << 1)
+struct fuse_setupmapping_in {
+	/* An already open handle */
+	uint64_t	fh;
+	/* Offset into the file to start the mapping */
+	uint64_t	foffset;
+	/* Length of mapping required */
+	uint64_t	len;
+	/* Flags, FUSE_SETUPMAPPING_FLAG_* */
+	uint64_t	flags;
+	/* Offset in Memory Window */
+	uint64_t	moffset;
+};
+
+struct fuse_removemapping_in {
+	/* number of fuse_removemapping_one follows */
+	uint32_t        count;
+};
+
+struct fuse_removemapping_one {
+	/* Offset into the dax window start the unmapping */
+	uint64_t        moffset;
+	/* Length of mapping required */
+	uint64_t	len;
+};
+
+#define FUSE_REMOVEMAPPING_MAX_ENTRY   \
+		(PAGE_SIZE / sizeof(struct fuse_removemapping_one))
+
 #endif /* _LINUX_FUSE_H */

include/uapi/linux/virtio_fs.h

@@ -16,4 +16,7 @@ struct virtio_fs_config {
 	__le32 num_request_queues;
 } __attribute__((packed));
 
+/* For the id field in virtio_pci_shm_cap */
+#define VIRTIO_FS_SHMCAP_ID_CACHE 0
+
 #endif /* _UAPI_LINUX_VIRTIO_FS_H */