From patchwork Mon Jun 20 17:21:19 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Eric W. Biederman" X-Patchwork-Id: 9188343 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 32E0360871 for ; Mon, 20 Jun 2016 17:56:47 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 28AA72793A for ; Mon, 20 Jun 2016 17:56:47 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 1A57E27BFC; Mon, 20 Jun 2016 17:56:47 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 6160A2793A for ; Mon, 20 Jun 2016 17:56:46 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754754AbcFTR41 (ORCPT ); Mon, 20 Jun 2016 13:56:27 -0400 Received: from out01.mta.xmission.com ([166.70.13.231]:48338 "EHLO out01.mta.xmission.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754738AbcFTR4X (ORCPT ); Mon, 20 Jun 2016 13:56:23 -0400 Received: from in01.mta.xmission.com ([166.70.13.51]) by out01.mta.xmission.com with esmtps (TLS1.2:RSA_AES_128_CBC_SHA1:128) (Exim 4.82) (envelope-from ) id 1bF35w-0004Q9-HP; Mon, 20 Jun 2016 11:34:44 -0600 Received: from 67-3-204-119.omah.qwest.net ([67.3.204.119] helo=x220.int.ebiederm.org) by in01.mta.xmission.com with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.87) (envelope-from ) id 1bF35v-0001hg-AJ; Mon, 20 Jun 2016 11:34:44 -0600 From: "Eric W. Biederman" To: Linux Containers Cc: linux-fsdevel@vger.kernel.org, Miklos Szeredi , James Bottomley , Djalal Harouni , Seth Forshee , "Serge E. Hallyn" , Andy Lutomirski Date: Mon, 20 Jun 2016 12:21:19 -0500 Message-Id: <20160620172130.15712-2-ebiederm@xmission.com> X-Mailer: git-send-email 2.8.3 In-Reply-To: <20160620172130.15712-1-ebiederm@xmission.com> References: <87fus77pns.fsf@x220.int.ebiederm.org> <20160620172130.15712-1-ebiederm@xmission.com> X-XM-SPF: eid=1bF35v-0001hg-AJ; ; ; mid=<20160620172130.15712-2-ebiederm@xmission.com>; ; ; hst=in01.mta.xmission.com; ; ; ip=67.3.204.119; ; ; frm=ebiederm@xmission.com; ; ; spf=neutral X-XM-AID: U2FsdGVkX19+/E1TDaGhtXXisvGHjko/qzFEf49JEhk= X-SA-Exim-Connect-IP: 67.3.204.119 X-SA-Exim-Mail-From: ebiederm@xmission.com Subject: [PATCH review 02/13] mnt: Refactor fs_fully_visible into mount_too_revealing X-SA-Exim-Version: 4.2.1 (built Thu, 05 May 2016 13:38:54 -0600) X-SA-Exim-Scanned: Yes (on in01.mta.xmission.com) Sender: linux-fsdevel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Replace the call of fs_fully_visible in do_new_mount from before the new superblock is allocated with a call of mount_too_revealing after the superblock is allocated. This winds up being a much better location for maintainability of the code. The first change this enables is the replacement of FS_USERNS_VISIBLE with SB_I_USERNS_VISIBLE. Moving the flag from struct filesystem_type to sb_iflags on the superblock. Unfortunately mount_too_revealing fundamentally needs to touch mnt_flags adding several MNT_LOCKED_XXX flags at the appropriate times. If the mnt_flags did not need to be touched the code could be easily moved into the filesystem specific mount code. Signed-off-by: "Eric W. Biederman" Acked-by: Seth Forshee --- fs/namespace.c | 38 +++++++++++++++++++++++++------------- fs/proc/inode.c | 1 + fs/proc/root.c | 2 +- fs/sysfs/mount.c | 4 ++-- include/linux/fs.h | 4 +++- 5 files changed, 32 insertions(+), 17 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index 783004af5707..1a69aa786975 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2375,7 +2375,7 @@ unlock: return err; } -static bool fs_fully_visible(struct file_system_type *fs_type, int *new_mnt_flags); +static bool mount_too_revealing(struct vfsmount *mnt, int *new_mnt_flags); /* * create a new mount for userspace and request it to be added into the @@ -2408,12 +2408,6 @@ static int do_new_mount(struct path *path, const char *fstype, int flags, flags |= MS_NODEV; mnt_flags |= MNT_NODEV | MNT_LOCK_NODEV; } - if (type->fs_flags & FS_USERNS_VISIBLE) { - if (!fs_fully_visible(type, &mnt_flags)) { - put_filesystem(type); - return -EPERM; - } - } } mnt = vfs_kern_mount(type, flags, name, data); @@ -2425,6 +2419,11 @@ static int do_new_mount(struct path *path, const char *fstype, int flags, if (IS_ERR(mnt)) return PTR_ERR(mnt); + if (mount_too_revealing(mnt, &mnt_flags)) { + mntput(mnt); + return -EPERM; + } + err = do_add_mount(real_mount(mnt), path, mnt_flags); if (err) mntput(mnt); @@ -3216,22 +3215,19 @@ bool current_chrooted(void) return chrooted; } -static bool fs_fully_visible(struct file_system_type *type, int *new_mnt_flags) +static bool mnt_already_visible(struct mnt_namespace *ns, struct vfsmount *new, + int *new_mnt_flags) { - struct mnt_namespace *ns = current->nsproxy->mnt_ns; int new_flags = *new_mnt_flags; struct mount *mnt; bool visible = false; - if (unlikely(!ns)) - return false; - down_read(&namespace_sem); list_for_each_entry(mnt, &ns->list, mnt_list) { struct mount *child; int mnt_flags; - if (mnt->mnt.mnt_sb->s_type != type) + if (mnt->mnt.mnt_sb->s_type != new->mnt_sb->s_type) continue; /* This mount is not fully visible if it's root directory @@ -3298,6 +3294,22 @@ found: return visible; } +static bool mount_too_revealing(struct vfsmount *mnt, int *new_mnt_flags) +{ + struct mnt_namespace *ns = current->nsproxy->mnt_ns; + unsigned long s_iflags; + + if (ns->user_ns == &init_user_ns) + return false; + + /* Can this filesystem be too revealing? */ + s_iflags = mnt->mnt_sb->s_iflags; + if (!(s_iflags & SB_I_USERNS_VISIBLE)) + return false; + + return !mnt_already_visible(ns, mnt, new_mnt_flags); +} + static struct ns_common *mntns_get(struct task_struct *task) { struct ns_common *ns = NULL; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 42305ddcbaa0..78fa452d65ed 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -462,6 +462,7 @@ int proc_fill_super(struct super_block *s) struct inode *root_inode; int ret; + s->s_iflags |= SB_I_USERNS_VISIBLE; s->s_flags |= MS_NODIRATIME | MS_NOSUID | MS_NOEXEC; s->s_blocksize = 1024; s->s_blocksize_bits = 10; diff --git a/fs/proc/root.c b/fs/proc/root.c index 55bc7d6c8aac..a1b2860fec62 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -158,7 +158,7 @@ static struct file_system_type proc_fs_type = { .name = "proc", .mount = proc_mount, .kill_sb = proc_kill_sb, - .fs_flags = FS_USERNS_VISIBLE | FS_USERNS_MOUNT, + .fs_flags = FS_USERNS_MOUNT, }; void __init proc_root_init(void) diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index f3db82071cfb..f31e36994dfb 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -42,7 +42,7 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, kobj_ns_drop(KOBJ_NS_TYPE_NET, ns); else if (new_sb) /* Userspace would break if executables appear on sysfs */ - root->d_sb->s_iflags |= SB_I_NOEXEC; + root->d_sb->s_iflags |= SB_I_USERNS_VISIBLE | SB_I_NOEXEC; return root; } @@ -59,7 +59,7 @@ static struct file_system_type sysfs_fs_type = { .name = "sysfs", .mount = sysfs_mount, .kill_sb = sysfs_kill_sb, - .fs_flags = FS_USERNS_VISIBLE | FS_USERNS_MOUNT, + .fs_flags = FS_USERNS_MOUNT, }; int __init sysfs_init(void) diff --git a/include/linux/fs.h b/include/linux/fs.h index dd288148a6b1..71988dd3af95 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1328,6 +1328,9 @@ struct mm_struct; #define SB_I_CGROUPWB 0x00000001 /* cgroup-aware writeback enabled */ #define SB_I_NOEXEC 0x00000002 /* Ignore executables on this fs */ +/* sb->s_iflags to limit user namespace mounts */ +#define SB_I_USERNS_VISIBLE 0x00000010 /* fstype already mounted */ + /* Possible states of 'frozen' field */ enum { SB_UNFROZEN = 0, /* FS is unfrozen */ @@ -2011,7 +2014,6 @@ struct file_system_type { #define FS_HAS_SUBTYPE 4 #define FS_USERNS_MOUNT 8 /* Can be mounted by userns root */ #define FS_USERNS_DEV_MOUNT 16 /* A userns mount does not imply MNT_NODEV */ -#define FS_USERNS_VISIBLE 32 /* FS must already be visible */ #define FS_RENAME_DOES_D_MOVE 32768 /* FS will handle d_move() during rename() internally. */ struct dentry *(*mount) (struct file_system_type *, int, const char *, void *);