Merge tag 'fuse-update-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:

 - Allow mounting an active fuse device. Previously the fuse device
   would always be mounted during initialization, and sharing a fuse
   superblock was only possible through mount or namespace cloning

 - Fix data flushing in syncfs (virtiofs only)

 - Fix data flushing in copy_file_range()

 - Fix a possible deadlock in atomic O_TRUNC

 - Misc fixes and cleanups

* tag 'fuse-update-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: remove unused arg in fuse_write_file_get()
  fuse: wait for writepages in syncfs
  fuse: flush extending writes
  fuse: truncate pagecache on atomic_o_trunc
  fuse: allow sharing existing sb
  fuse: move fget() to fuse_get_tree()
  fuse: move option checking into fuse_fill_super()
  fuse: name fs_context consistently
  fuse: fix use after free in fuse_read_interrupt()

fs/fuse/control.c

@@ -328,7 +328,7 @@ void fuse_ctl_remove_conn(struct fuse_conn *fc)
 	drop_nlink(d_inode(fuse_control_sb->s_root));
 }
 
-static int fuse_ctl_fill_super(struct super_block *sb, struct fs_context *fctx)
+static int fuse_ctl_fill_super(struct super_block *sb, struct fs_context *fsc)
 {
 	static const struct tree_descr empty_descr = {""};
 	struct fuse_conn *fc;
@@ -354,18 +354,18 @@ static int fuse_ctl_fill_super(struct super_block *sb, struct fs_context *fctx)
 	return 0;
 }
 
-static int fuse_ctl_get_tree(struct fs_context *fc)
+static int fuse_ctl_get_tree(struct fs_context *fsc)
 {
-	return get_tree_single(fc, fuse_ctl_fill_super);
+	return get_tree_single(fsc, fuse_ctl_fill_super);
 }
 
 static const struct fs_context_operations fuse_ctl_context_ops = {
 	.get_tree	= fuse_ctl_get_tree,
 };
 
-static int fuse_ctl_init_fs_context(struct fs_context *fc)
+static int fuse_ctl_init_fs_context(struct fs_context *fsc)
 {
-	fc->ops = &fuse_ctl_context_ops;
+	fsc->ops = &fuse_ctl_context_ops;
 	return 0;
 }
 

fs/fuse/dev.c

@@ -288,10 +288,10 @@ void fuse_request_end(struct fuse_req *req)
 
 	/*
 	 * test_and_set_bit() implies smp_mb() between bit
-	 * changing and below intr_entry check. Pairs with
+	 * changing and below FR_INTERRUPTED check. Pairs with
 	 * smp_mb() from queue_interrupt().
 	 */
-	if (!list_empty(&req->intr_entry)) {
+	if (test_bit(FR_INTERRUPTED, &req->flags)) {
 		spin_lock(&fiq->lock);
 		list_del_init(&req->intr_entry);
 		spin_unlock(&fiq->lock);

fs/fuse/file.c

@@ -198,12 +198,11 @@ void fuse_finish_open(struct inode *inode, struct file *file)
 	struct fuse_file *ff = file->private_data;
 	struct fuse_conn *fc = get_fuse_conn(inode);
 
-	if (!(ff->open_flags & FOPEN_KEEP_CACHE))
-		invalidate_inode_pages2(inode->i_mapping);
 	if (ff->open_flags & FOPEN_STREAM)
 		stream_open(inode, file);
 	else if (ff->open_flags & FOPEN_NONSEEKABLE)
 		nonseekable_open(inode, file);
+
 	if (fc->atomic_o_trunc && (file->f_flags & O_TRUNC)) {
 		struct fuse_inode *fi = get_fuse_inode(inode);
 
@@ -211,10 +210,14 @@ void fuse_finish_open(struct inode *inode, struct file *file)
 		fi->attr_version = atomic64_inc_return(&fc->attr_version);
 		i_size_write(inode, 0);
 		spin_unlock(&fi->lock);
+		truncate_pagecache(inode, 0);
 		fuse_invalidate_attr(inode);
 		if (fc->writeback_cache)
 			file_update_time(file);
+	} else if (!(ff->open_flags & FOPEN_KEEP_CACHE)) {
+		invalidate_inode_pages2(inode->i_mapping);
 	}
+
 	if ((file->f_mode & FMODE_WRITE) && fc->writeback_cache)
 		fuse_link_write_file(file);
 }
@@ -389,6 +392,7 @@ struct fuse_writepage_args {
 	struct list_head queue_entry;
 	struct fuse_writepage_args *next;
 	struct inode *inode;
+	struct fuse_sync_bucket *bucket;
 };
 
 static struct fuse_writepage_args *fuse_find_writeback(struct fuse_inode *fi,
@@ -1608,6 +1612,9 @@ static void fuse_writepage_free(struct fuse_writepage_args *wpa)
 	struct fuse_args_pages *ap = &wpa->ia.ap;
 	int i;
 
+	if (wpa->bucket)
+		fuse_sync_bucket_dec(wpa->bucket);
+
 	for (i = 0; i < ap->num_pages; i++)
 		__free_page(ap->pages[i]);
 
@@ -1813,8 +1820,7 @@ static void fuse_writepage_end(struct fuse_mount *fm, struct fuse_args *args,
 	fuse_writepage_free(wpa);
 }
 
-static struct fuse_file *__fuse_write_file_get(struct fuse_conn *fc,
-					       struct fuse_inode *fi)
+static struct fuse_file *__fuse_write_file_get(struct fuse_inode *fi)
 {
 	struct fuse_file *ff = NULL;
 
@@ -1829,22 +1835,20 @@ static struct fuse_file *__fuse_write_file_get(struct fuse_conn *fc,
 	return ff;
 }
 
-static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc,
-					     struct fuse_inode *fi)
+static struct fuse_file *fuse_write_file_get(struct fuse_inode *fi)
 {
-	struct fuse_file *ff = __fuse_write_file_get(fc, fi);
+	struct fuse_file *ff = __fuse_write_file_get(fi);
 	WARN_ON(!ff);
 	return ff;
 }
 
 int fuse_write_inode(struct inode *inode, struct writeback_control *wbc)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_inode *fi = get_fuse_inode(inode);
 	struct fuse_file *ff;
 	int err;
 
-	ff = __fuse_write_file_get(fc, fi);
+	ff = __fuse_write_file_get(fi);
 	err = fuse_flush_times(inode, ff);
 	if (ff)
 		fuse_file_put(ff, false, false);
@@ -1871,6 +1875,20 @@ static struct fuse_writepage_args *fuse_writepage_args_alloc(void)
 
 }
 
+static void fuse_writepage_add_to_bucket(struct fuse_conn *fc,
+					 struct fuse_writepage_args *wpa)
+{
+	if (!fc->sync_fs)
+		return;
+
+	rcu_read_lock();
+	/* Prevent resurrection of dead bucket in unlikely race with syncfs */
+	do {
+		wpa->bucket = rcu_dereference(fc->curr_bucket);
+	} while (unlikely(!atomic_inc_not_zero(&wpa->bucket->count)));
+	rcu_read_unlock();
+}
+
 static int fuse_writepage_locked(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
@@ -1894,10 +1912,11 @@ static int fuse_writepage_locked(struct page *page)
 		goto err_free;
 
 	error = -EIO;
-	wpa->ia.ff = fuse_write_file_get(fc, fi);
+	wpa->ia.ff = fuse_write_file_get(fi);
 	if (!wpa->ia.ff)
 		goto err_nofile;
 
+	fuse_writepage_add_to_bucket(fc, wpa);
 	fuse_write_args_fill(&wpa->ia, wpa->ia.ff, page_offset(page), 0);
 
 	copy_highpage(tmp_page, page);
@@ -2113,7 +2132,7 @@ static int fuse_writepages_fill(struct page *page,
 
 	if (!data->ff) {
 		err = -EIO;
-		data->ff = fuse_write_file_get(fc, fi);
+		data->ff = fuse_write_file_get(fi);
 		if (!data->ff)
 			goto out_unlock;
 	}
@@ -2148,6 +2167,8 @@ static int fuse_writepages_fill(struct page *page,
 			__free_page(tmp_page);
 			goto out_unlock;
 		}
+		fuse_writepage_add_to_bucket(fc, wpa);
+
 		data->max_pages = 1;
 
 		ap = &wpa->ia.ap;
@@ -2881,7 +2902,7 @@ fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 
 static int fuse_writeback_range(struct inode *inode, loff_t start, loff_t end)
 {
-	int err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	int err = filemap_write_and_wait_range(inode->i_mapping, start, -1);
 
 	if (!err)
 		fuse_sync_writes(inode);

fs/fuse/fuse_i.h

@@ -482,6 +482,7 @@ struct fuse_dev {
 
 struct fuse_fs_context {
 	int fd;
+	struct file *file;
 	unsigned int rootmode;
 	kuid_t user_id;
 	kgid_t group_id;
@@ -508,6 +509,13 @@ struct fuse_fs_context {
 	void **fudptr;
 };
 
+struct fuse_sync_bucket {
+	/* count is a possible scalability bottleneck */
+	atomic_t count;
+	wait_queue_head_t waitq;
+	struct rcu_head rcu;
+};
+
 /**
  * A Fuse connection.
  *
@@ -800,6 +808,9 @@ struct fuse_conn {
 
 	/** List of filesystems using this connection */
 	struct list_head mounts;
+
+	/* New writepages go into this bucket */
+	struct fuse_sync_bucket __rcu *curr_bucket;
 };
 
 /*
@@ -903,6 +914,15 @@ static inline void fuse_page_descs_length_init(struct fuse_page_desc *descs,
 		descs[i].length = PAGE_SIZE - descs[i].offset;
 }
 
+static inline void fuse_sync_bucket_dec(struct fuse_sync_bucket *bucket)
+{
+	/* Need RCU protection to prevent use after free after the decrement */
+	rcu_read_lock();
+	if (atomic_dec_and_test(&bucket->count))
+		wake_up(&bucket->waitq);
+	rcu_read_unlock();
+}
+
 /** Device operations */
 extern const struct file_operations fuse_dev_operations;
 

fs/fuse/virtio_fs.c

@@ -97,14 +97,14 @@ static const struct fs_parameter_spec virtio_fs_parameters[] = {
 	{}
 };
 
-static int virtio_fs_parse_param(struct fs_context *fc,
+static int virtio_fs_parse_param(struct fs_context *fsc,
 				 struct fs_parameter *param)
 {
 	struct fs_parse_result result;
-	struct fuse_fs_context *ctx = fc->fs_private;
+	struct fuse_fs_context *ctx = fsc->fs_private;
 	int opt;
 
-	opt = fs_parse(fc, virtio_fs_parameters, param, &result);
+	opt = fs_parse(fsc, virtio_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
@@ -119,9 +119,9 @@ static int virtio_fs_parse_param(struct fs_context *fc,
 	return 0;
 }
 
-static void virtio_fs_free_fc(struct fs_context *fc)
+static void virtio_fs_free_fsc(struct fs_context *fsc)
 {
-	struct fuse_fs_context *ctx = fc->fs_private;
+	struct fuse_fs_context *ctx = fsc->fs_private;
 
 	kfree(ctx);
 }
@@ -1488,7 +1488,7 @@ 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,
+	.free		= virtio_fs_free_fsc,
 	.parse_param	= virtio_fs_parse_param,
 	.get_tree	= virtio_fs_get_tree,
 };