linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Howells <dhowells@redhat.com>
To: viro@zeniv.linux.org.uk
Cc: torvalds@linux-foundation.org, dhowells@redhat.com,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH 03/33] teach move_mount(2) to work with OPEN_TREE_CLONE [ver #11]
Date: Wed, 01 Aug 2018 16:24:24 +0100	[thread overview]
Message-ID: <153313706468.13253.16771852721230651938.stgit@warthog.procyon.org.uk> (raw)
In-Reply-To: <153313703562.13253.5766498657900728120.stgit@warthog.procyon.org.uk>

From: Al Viro <viro@zeniv.linux.org.uk>

Allow a detached tree created by open_tree(..., OPEN_TREE_CLONE) to be
attached by move_mount(2).

If by the time of final fput() of OPEN_TREE_CLONE-opened file its tree is
not detached anymore, it won't be dissolved.  move_mount(2) is adjusted
to handle detached source.

That gives us equivalents of mount --bind and mount --rbind.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
---

 fs/namespace.c |   26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index e2934a4f342b..3981fd7b13f5 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1846,8 +1846,10 @@ void dissolve_on_fput(struct vfsmount *mnt)
 {
 	namespace_lock();
 	lock_mount_hash();
-	mntget(mnt);
-	umount_tree(real_mount(mnt), UMOUNT_SYNC);
+	if (!real_mount(mnt)->mnt_ns) {
+		mntget(mnt);
+		umount_tree(real_mount(mnt), UMOUNT_SYNC);
+	}
 	unlock_mount_hash();
 	namespace_unlock();
 }
@@ -2454,6 +2456,7 @@ static int do_move_mount(struct path *old_path, struct path *new_path)
 	struct mount *old;
 	struct mountpoint *mp;
 	int err;
+	bool attached;
 
 	mp = lock_mount(new_path);
 	err = PTR_ERR(mp);
@@ -2464,10 +2467,19 @@ static int do_move_mount(struct path *old_path, struct path *new_path)
 	p = real_mount(new_path->mnt);
 
 	err = -EINVAL;
-	if (!check_mnt(p) || !check_mnt(old))
+	/* The mountpoint must be in our namespace. */
+	if (!check_mnt(p))
+		goto out1;
+	/* The thing moved should be either ours or completely unattached. */
+	if (old->mnt_ns && !check_mnt(old))
 		goto out1;
 
-	if (!mnt_has_parent(old))
+	attached = mnt_has_parent(old);
+	/*
+	 * We need to allow open_tree(OPEN_TREE_CLONE) followed by
+	 * move_mount(), but mustn't allow "/" to be moved.
+	 */
+	if (old->mnt_ns && !attached)
 		goto out1;
 
 	if (old->mnt.mnt_flags & MNT_LOCKED)
@@ -2482,7 +2494,7 @@ static int do_move_mount(struct path *old_path, struct path *new_path)
 	/*
 	 * Don't move a mount residing in a shared parent.
 	 */
-	if (IS_MNT_SHARED(old->mnt_parent))
+	if (attached && IS_MNT_SHARED(old->mnt_parent))
 		goto out1;
 	/*
 	 * Don't move a mount tree containing unbindable mounts to a destination
@@ -2496,7 +2508,7 @@ static int do_move_mount(struct path *old_path, struct path *new_path)
 			goto out1;
 
 	err = attach_recursive_mnt(old, real_mount(new_path->mnt), mp,
-				   &parent_path);
+				   attached ? &parent_path : NULL);
 	if (err)
 		goto out1;
 
@@ -3182,6 +3194,8 @@ SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name,
 
 /*
  * Move a mount from one place to another.
+ * In combination with open_tree(OPEN_TREE_CLONE [| AT_RECURSIVE]) it can be
+ * used to copy a mount subtree.
  *
  * Note the flags value is a combination of MOVE_MOUNT_* flags.
  */


  parent reply	other threads:[~2018-08-01 15:24 UTC|newest]

Thread overview: 116+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-01 15:23 [PATCH 00/33] VFS: Introduce filesystem context [ver #11] David Howells
2018-08-01 15:24 ` [PATCH 01/33] vfs: syscall: Add open_tree(2) to reference or clone a mount " David Howells
2018-08-02 17:31   ` Alan Jenkins
2018-08-02 21:29     ` Al Viro
2018-08-02 21:51   ` David Howells
2018-08-02 23:46     ` Alan Jenkins
2018-08-01 15:24 ` [PATCH 02/33] vfs: syscall: Add move_mount(2) to move mounts around " David Howells
2018-08-01 15:24 ` David Howells [this message]
2018-10-12 14:25   ` [PATCH 03/33] teach move_mount(2) to work with OPEN_TREE_CLONE " Alan Jenkins
2018-08-01 15:24 ` [PATCH 04/33] vfs: Suppress MS_* flag defs within the kernel unless explicitly enabled " David Howells
2018-08-01 15:24 ` [PATCH 05/33] vfs: Introduce the basic header for the new mount API's filesystem context " David Howells
2018-08-01 15:24 ` [PATCH 06/33] vfs: Introduce logging functions " David Howells
2018-08-01 15:24 ` [PATCH 07/33] vfs: Add configuration parser helpers " David Howells
2018-08-01 15:24 ` [PATCH 08/33] vfs: Add LSM hooks for the new mount API " David Howells
2018-08-01 20:50   ` James Morris
2018-08-01 22:53   ` David Howells
2018-08-01 15:25 ` [PATCH 09/33] selinux: Implement the new mount API LSM hooks " David Howells
2018-08-01 15:25 ` [PATCH 10/33] smack: Implement filesystem context security " David Howells
2018-08-01 15:25 ` [PATCH 11/33] apparmor: Implement security hooks for the new mount API " David Howells
2018-08-01 15:25 ` [PATCH 12/33] tomoyo: " David Howells
2018-08-01 15:25 ` [PATCH 13/33] vfs: Separate changing mount flags full remount " David Howells
2018-08-01 15:25 ` [PATCH 14/33] vfs: Implement a filesystem superblock creation/configuration context " David Howells
2018-09-11 17:46   ` Guenter Roeck
2018-09-11 21:52   ` David Howells
2018-09-11 22:07     ` Guenter Roeck
2018-09-11 23:17     ` David Howells
2018-09-11 23:54       ` Guenter Roeck
2018-09-18  9:07         ` Sergey Senozhatsky
2018-09-18  9:40           ` Sergey Senozhatsky
2018-09-18 14:06           ` Guenter Roeck
2018-09-19  1:12             ` Sergey Senozhatsky
2018-09-19  1:26               ` Sergey Senozhatsky
2018-09-18 15:34         ` David Howells
2018-09-18 16:39         ` David Howells
2018-09-19  1:15           ` Sergey Senozhatsky
2018-09-18 17:43         ` David Howells
2018-09-18  9:54   ` Sergey Senozhatsky
2018-09-18 15:28   ` David Howells
2018-08-01 15:25 ` [PATCH 15/33] vfs: Remove unused code after filesystem context changes " David Howells
2018-08-01 15:25 ` [PATCH 16/33] procfs: Move proc_fill_super() to fs/proc/root.c " David Howells
2018-08-01 15:26 ` [PATCH 17/33] proc: Add fs_context support to procfs " David Howells
2018-08-01 15:26 ` [PATCH 18/33] ipc: Convert mqueue fs to fs_context " David Howells
2018-08-01 15:26 ` [PATCH 19/33] cpuset: Use " David Howells
2018-08-01 15:26 ` [PATCH 20/33] kernfs, sysfs, cgroup, intel_rdt: Support " David Howells
2018-08-01 15:26 ` [PATCH 21/33] hugetlbfs: Convert to " David Howells
2018-08-01 15:26 ` [PATCH 22/33] vfs: Remove kern_mount_data() " David Howells
2018-08-01 15:26 ` [PATCH 23/33] vfs: Provide documentation for new mount API " David Howells
2018-08-01 15:26 ` [PATCH 24/33] Make anon_inodes unconditional " David Howells
2018-08-01 15:26 ` [PATCH 25/33] vfs: syscall: Add fsopen() to prepare for superblock creation " David Howells
2018-08-01 15:27 ` [PATCH 26/33] vfs: Implement logging through fs_context " David Howells
2018-08-01 15:27 ` [PATCH 27/33] vfs: Add some logging to the core users of the fs_context log " David Howells
2018-08-01 15:27 ` [PATCH 28/33] vfs: syscall: Add fsconfig() for configuring and managing a context " David Howells
2018-08-06 17:28   ` Eric W. Biederman
2018-08-09 14:14   ` David Howells
2018-08-09 14:24   ` David Howells
2018-08-09 14:35     ` Miklos Szeredi
2018-08-09 15:32     ` Eric W. Biederman
2018-08-09 16:33     ` David Howells
2018-08-11 20:20     ` David Howells
2018-08-11 23:26       ` Andy Lutomirski
2018-08-01 15:27 ` [PATCH 29/33] vfs: syscall: Add fsmount() to create a mount for a superblock " David Howells
2018-08-01 15:27 ` [PATCH 30/33] vfs: syscall: Add fspick() to select a superblock for reconfiguration " David Howells
2018-08-24 14:51   ` Miklos Szeredi
2018-08-24 14:54     ` Andy Lutomirski
2018-08-01 15:27 ` [PATCH 31/33] afs: Add fs_context support " David Howells
2018-08-01 15:27 ` [PATCH 32/33] afs: Use fs_context to pass parameters over automount " David Howells
2018-08-01 15:27 ` [PATCH 33/33] vfs: Add a sample program for the new mount API " David Howells
2018-08-10 14:05 ` BUG: Mount ignores mount options Eric W. Biederman
2018-08-10 14:36   ` Andy Lutomirski
2018-08-10 15:17     ` Eric W. Biederman
2018-08-10 15:24     ` Al Viro
2018-08-10 15:11   ` Tetsuo Handa
2018-08-10 15:13   ` David Howells
2018-08-10 15:16   ` Al Viro
2018-08-11  1:05     ` Eric W. Biederman
2018-08-11  1:46       ` Theodore Y. Ts'o
2018-08-11  4:48         ` Eric W. Biederman
2018-08-11 17:47           ` Casey Schaufler
2018-08-15  4:03             ` Eric W. Biederman
2018-08-11  1:58       ` Al Viro
2018-08-11  2:17         ` Al Viro
2018-08-11  4:43           ` Eric W. Biederman
2018-08-13 12:54         ` Miklos Szeredi
2018-08-10 15:11 ` David Howells
2018-08-10 15:39   ` Theodore Y. Ts'o
2018-08-10 15:55     ` Casey Schaufler
2018-08-10 16:11     ` David Howells
2018-08-10 18:00     ` Eric W. Biederman
2018-08-10 15:53   ` David Howells
2018-08-10 16:14     ` Theodore Y. Ts'o
2018-08-10 20:06       ` Andy Lutomirski
2018-08-10 20:46         ` Theodore Y. Ts'o
2018-08-10 22:12           ` Darrick J. Wong
2018-08-10 23:54             ` Theodore Y. Ts'o
2018-08-11  0:38               ` Darrick J. Wong
2018-08-11  1:32                 ` Eric W. Biederman
2018-08-13 16:35         ` Alan Cox
2018-08-13 16:48           ` Andy Lutomirski
2018-08-13 17:29             ` Al Viro
2018-08-13 19:00               ` James Morris
2018-08-13 19:20                 ` Casey Schaufler
2018-08-15 23:29                 ` Serge E. Hallyn
2018-08-11  0:28       ` Eric W. Biederman
2018-08-11  1:19   ` Eric W. Biederman
2018-08-11  7:29   ` David Howells
2018-08-11 16:31     ` Andy Lutomirski
2018-08-11 16:51       ` Al Viro
2018-08-15 16:31 ` Should we split the network filesystem setup into two phases? David Howells
2018-08-15 16:51   ` Andy Lutomirski
2018-08-16  3:51   ` Steve French
2018-08-16  5:06   ` Eric W. Biederman
2018-08-16 16:24     ` Steve French
2018-08-16 17:21       ` Eric W. Biederman
2018-08-16 17:23       ` Aurélien Aptel
2018-08-16 18:36         ` Steve French
2018-08-17 23:11     ` Al Viro

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=153313706468.13253.16771852721230651938.stgit@warthog.procyon.org.uk \
    --to=dhowells@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).