Merge tag 'fuse-update-5.11' of git://git.kernel.org/pub/scm/linux/ke…

…rnel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:

 - Improve performance of virtio-fs in mixed read/write workloads

 - Try to revalidate cache before returning EEXIST on exclusive create

 - Add a couple of miscellaneous bug fixes as well as some code cleanups

* tag 'fuse-update-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: fix bad inode
  fuse: support SB_NOSEC flag to improve write performance
  fuse: add a flag FUSE_OPEN_KILL_SUIDGID for open() request
  fuse: don't send ATTR_MODE to kill suid/sgid for handle_killpriv_v2
  fuse: setattr should set FATTR_KILL_SUIDGID
  fuse: set FUSE_WRITE_KILL_SUIDGID in cached write path
  fuse: rename FUSE_WRITE_KILL_PRIV to FUSE_WRITE_KILL_SUIDGID
  fuse: introduce the notion of FUSE_HANDLE_KILLPRIV_V2
  fuse: always revalidate if exclusive create
  virtiofs: clean up error handling in virtio_fs_get_tree()
  fuse: add fuse_sb_destroy() helper
  fuse: simplify get_fuse_conn*()
  fuse: get rid of fuse_mount refcount
  virtiofs: simplify sb setup
  virtiofs fix leak in setup
  fuse: launder page should wait for page writeback

fs/fuse/acl.c

@@ -19,6 +19,9 @@ struct posix_acl *fuse_get_acl(struct inode *inode, int type)
 	void *value = NULL;
 	struct posix_acl *acl;
 
+	if (fuse_is_bad(inode))
+		return ERR_PTR(-EIO);
+
 	if (!fc->posix_acl || fc->no_getxattr)
 		return NULL;
 
@@ -53,6 +56,9 @@ int fuse_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	const char *name;
 	int ret;
 
+	if (fuse_is_bad(inode))
+		return -EIO;
+
 	if (!fc->posix_acl || fc->no_setxattr)
 		return -EOPNOTSUPP;
 

fs/fuse/dir.c

@@ -202,10 +202,10 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
 	int ret;
 
 	inode = d_inode_rcu(entry);
-	if (inode && is_bad_inode(inode))
+	if (inode && fuse_is_bad(inode))
 		goto invalid;
 	else if (time_before64(fuse_dentry_time(entry), get_jiffies_64()) ||
-		 (flags & LOOKUP_REVAL)) {
+		 (flags & (LOOKUP_EXCL | LOOKUP_REVAL))) {
 		struct fuse_entry_out outarg;
 		FUSE_ARGS(args);
 		struct fuse_forget_link *forget;
@@ -328,12 +328,11 @@ static struct vfsmount *fuse_dentry_automount(struct path *path)
 	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);
+		kfree(fm);
 		goto out_put_fsc;
 	}
 	fm->fc = fuse_conn_get(fc);
@@ -463,6 +462,9 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 	bool outarg_valid = true;
 	bool locked;
 
+	if (fuse_is_bad(dir))
+		return ERR_PTR(-EIO);
+
 	locked = fuse_lock_inode(dir);
 	err = fuse_lookup_name(dir->i_sb, get_node_id(dir), &entry->d_name,
 			       &outarg, &inode);
@@ -542,6 +544,12 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 	inarg.flags = flags;
 	inarg.mode = mode;
 	inarg.umask = current_umask();
+
+	if (fm->fc->handle_killpriv_v2 && (flags & O_TRUNC) &&
+	    !(flags & O_EXCL) && !capable(CAP_FSETID)) {
+		inarg.open_flags |= FUSE_OPEN_KILL_SUIDGID;
+	}
+
 	args.opcode = FUSE_CREATE;
 	args.nodeid = get_node_id(dir);
 	args.in_numargs = 2;
@@ -606,6 +614,9 @@ static int fuse_atomic_open(struct inode *dir, struct dentry *entry,
 	struct fuse_conn *fc = get_fuse_conn(dir);
 	struct dentry *res = NULL;
 
+	if (fuse_is_bad(dir))
+		return -EIO;
+
 	if (d_in_lookup(entry)) {
 		res = fuse_lookup(dir, entry, 0);
 		if (IS_ERR(res))
@@ -654,6 +665,9 @@ static int create_new_entry(struct fuse_mount *fm, struct fuse_args *args,
 	int err;
 	struct fuse_forget_link *forget;
 
+	if (fuse_is_bad(dir))
+		return -EIO;
+
 	forget = fuse_alloc_forget();
 	if (!forget)
 		return -ENOMEM;
@@ -781,6 +795,9 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry)
 	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
+	if (fuse_is_bad(dir))
+		return -EIO;
+
 	args.opcode = FUSE_UNLINK;
 	args.nodeid = get_node_id(dir);
 	args.in_numargs = 1;
@@ -817,6 +834,9 @@ static int fuse_rmdir(struct inode *dir, struct dentry *entry)
 	struct fuse_mount *fm = get_fuse_mount(dir);
 	FUSE_ARGS(args);
 
+	if (fuse_is_bad(dir))
+		return -EIO;
+
 	args.opcode = FUSE_RMDIR;
 	args.nodeid = get_node_id(dir);
 	args.in_numargs = 1;
@@ -895,6 +915,9 @@ static int fuse_rename2(struct inode *olddir, struct dentry *oldent,
 	struct fuse_conn *fc = get_fuse_conn(olddir);
 	int err;
 
+	if (fuse_is_bad(olddir))
+		return -EIO;
+
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
 		return -EINVAL;
 
@@ -1030,7 +1053,7 @@ static int fuse_do_getattr(struct inode *inode, struct kstat *stat,
 	if (!err) {
 		if (fuse_invalid_attr(&outarg.attr) ||
 		    (inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
-			make_bad_inode(inode);
+			fuse_make_bad(inode);
 			err = -EIO;
 		} else {
 			fuse_change_attributes(inode, &outarg.attr,
@@ -1232,6 +1255,9 @@ static int fuse_permission(struct inode *inode, int mask)
 	bool refreshed = false;
 	int err = 0;
 
+	if (fuse_is_bad(inode))
+		return -EIO;
+
 	if (!fuse_allow_current_process(fc))
 		return -EACCES;
 
@@ -1327,7 +1353,7 @@ static const char *fuse_get_link(struct dentry *dentry, struct inode *inode,
 	int err;
 
 	err = -EIO;
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		goto out_err;
 
 	if (fc->cache_symlinks)
@@ -1375,7 +1401,7 @@ static int fuse_dir_fsync(struct file *file, loff_t start, loff_t end,
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	int err;
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	if (fc->no_fsyncdir)
@@ -1649,10 +1675,20 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 		inarg.valid |= FATTR_FH;
 		inarg.fh = ff->fh;
 	}
+
+	/* Kill suid/sgid for non-directory chown unconditionally */
+	if (fc->handle_killpriv_v2 && !S_ISDIR(inode->i_mode) &&
+	    attr->ia_valid & (ATTR_UID | ATTR_GID))
+		inarg.valid |= FATTR_KILL_SUIDGID;
+
 	if (attr->ia_valid & ATTR_SIZE) {
 		/* For mandatory locking in truncate */
 		inarg.valid |= FATTR_LOCKOWNER;
 		inarg.lock_owner = fuse_lock_owner_id(fc, current->files);
+
+		/* Kill suid/sgid for truncate only if no CAP_FSETID */
+		if (fc->handle_killpriv_v2 && !capable(CAP_FSETID))
+			inarg.valid |= FATTR_KILL_SUIDGID;
 	}
 	fuse_setattr_fill(fc, &args, inode, &inarg, &outarg);
 	err = fuse_simple_request(fm, &args);
@@ -1664,7 +1700,7 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 
 	if (fuse_invalid_attr(&outarg.attr) ||
 	    (inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
-		make_bad_inode(inode);
+		fuse_make_bad(inode);
 		err = -EIO;
 		goto error;
 	}
@@ -1727,6 +1763,9 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr)
 	struct file *file = (attr->ia_valid & ATTR_FILE) ? attr->ia_file : NULL;
 	int ret;
 
+	if (fuse_is_bad(inode))
+		return -EIO;
+
 	if (!fuse_allow_current_process(get_fuse_conn(inode)))
 		return -EACCES;
 
@@ -1740,7 +1779,7 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr)
 		 *
 		 * This should be done on write(), truncate() and chown().
 		 */
-		if (!fc->handle_killpriv) {
+		if (!fc->handle_killpriv && !fc->handle_killpriv_v2) {
 			/*
 			 * ia_mode calculation may have used stale i_mode.
 			 * Refresh and recalculate.
@@ -1785,6 +1824,9 @@ static int fuse_getattr(const struct path *path, struct kstat *stat,
 	struct inode *inode = d_inode(path->dentry);
 	struct fuse_conn *fc = get_fuse_conn(inode);
 
+	if (fuse_is_bad(inode))
+		return -EIO;
+
 	if (!fuse_allow_current_process(fc)) {
 		if (!request_mask) {
 			/*

fs/fuse/file.c

@@ -42,6 +42,12 @@ static int fuse_send_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
 	inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
 	if (!fm->fc->atomic_o_trunc)
 		inarg.flags &= ~O_TRUNC;
+
+	if (fm->fc->handle_killpriv_v2 &&
+	    (inarg.flags & O_TRUNC) && !capable(CAP_FSETID)) {
+		inarg.open_flags |= FUSE_OPEN_KILL_SUIDGID;
+	}
+
 	args.opcode = opcode;
 	args.nodeid = nodeid;
 	args.in_numargs = 1;
@@ -226,6 +232,9 @@ int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
 	bool dax_truncate = (file->f_flags & O_TRUNC) &&
 			  fc->atomic_o_trunc && FUSE_IS_DAX(inode);
 
+	if (fuse_is_bad(inode))
+		return -EIO;
+
 	err = generic_file_open(inode, file);
 	if (err)
 		return err;
@@ -463,7 +472,7 @@ static int fuse_flush(struct file *file, fl_owner_t id)
 	FUSE_ARGS(args);
 	int err;
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	err = write_inode_now(inode, 1);
@@ -535,7 +544,7 @@ static int fuse_fsync(struct file *file, loff_t start, loff_t end,
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	int err;
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	inode_lock(inode);
@@ -859,7 +868,7 @@ static int fuse_readpage(struct file *file, struct page *page)
 	int err;
 
 	err = -EIO;
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		goto out;
 
 	err = fuse_do_readpage(file, page);
@@ -952,7 +961,7 @@ static void fuse_readahead(struct readahead_control *rac)
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	unsigned int i, max_pages, nr_pages = 0;
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return;
 
 	max_pages = min_t(unsigned int, fc->max_pages,
@@ -1097,6 +1106,8 @@ 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);
+	if (fm->fc->handle_killpriv_v2 && !capable(CAP_FSETID))
+		ia->write.in.write_flags |= FUSE_WRITE_KILL_SUIDGID;
 
 	err = fuse_simple_request(fm, &ap->args);
 	if (!err && ia->write.out.size > count)
@@ -1260,17 +1271,24 @@ static ssize_t fuse_cache_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	ssize_t written_buffered = 0;
 	struct inode *inode = mapping->host;
 	ssize_t err;
+	struct fuse_conn *fc = get_fuse_conn(inode);
 	loff_t endbyte = 0;
 
-	if (get_fuse_conn(inode)->writeback_cache) {
+	if (fc->writeback_cache) {
 		/* Update size (EOF optimization) and mode (SUID clearing) */
 		err = fuse_update_attributes(mapping->host, file);
 		if (err)
 			return err;
 
+		if (fc->handle_killpriv_v2 &&
+		    should_remove_suid(file_dentry(file))) {
+			goto writethrough;
+		}
+
 		return generic_file_write_iter(iocb, from);
 	}
 
+writethrough:
 	inode_lock(inode);
 
 	/* We can write back this queue in page reclaim */
@@ -1451,7 +1469,7 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
 
 		if (write) {
 			if (!capable(CAP_FSETID))
-				ia->write.in.write_flags |= FUSE_WRITE_KILL_PRIV;
+				ia->write.in.write_flags |= FUSE_WRITE_KILL_SUIDGID;
 
 			nres = fuse_send_write(ia, pos, nbytes, owner);
 		} else {
@@ -1555,7 +1573,7 @@ static ssize_t fuse_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	struct fuse_file *ff = file->private_data;
 	struct inode *inode = file_inode(file);
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	if (FUSE_IS_DAX(inode))
@@ -1573,7 +1591,7 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	struct fuse_file *ff = file->private_data;
 	struct inode *inode = file_inode(file);
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	if (FUSE_IS_DAX(inode))
@@ -2172,7 +2190,7 @@ static int fuse_writepages(struct address_space *mapping,
 	int err;
 
 	err = -EIO;
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		goto out;
 
 	data.inode = inode;
@@ -2281,6 +2299,9 @@ static int fuse_launder_page(struct page *page)
 	int err = 0;
 	if (clear_page_dirty_for_io(page)) {
 		struct inode *inode = page->mapping->host;
+
+		/* Serialize with pending writeback for the same page */
+		fuse_wait_on_page_writeback(inode, page->index);
 		err = fuse_writepage_locked(page);
 		if (!err)
 			fuse_wait_on_page_writeback(inode, page->index);
@@ -2954,7 +2975,7 @@ long fuse_ioctl_common(struct file *file, unsigned int cmd,
 	if (!fuse_allow_current_process(fc))
 		return -EACCES;
 
-	if (is_bad_inode(inode))
+	if (fuse_is_bad(inode))
 		return -EIO;
 
 	return fuse_do_ioctl(file, cmd, arg, flags);