On Wed, May 13, 2015 at 06:41:54PM +0200, Goffredo Baroncelli wrote: > I think that make sense to add another ioctl instead of relaxing the privilege check > on the basis of the parameters. Yeah, that was another way how to do it. In general I'm more inclined to use the existing ioctls and related structures and avoid single purpose ioctl. Furthermore, we can teach btrfs_ioctl_ino_lookup to verify access rights to the file it finds. The root restriction is not unnecessary in all cases. We've relaxed the ioctl permissions in the past, see FS_INFO for example. I hope we can afford that if that's for sake of better usability and user convenience. > If I read correctly the code, in this case the function > btrfs_search_path_in_tree() is not even called: it is requested to return only > > BTRFS_I(inode)->root->root_key.objectid; Right. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
