linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Al Viro <viro@zeniv.linux.org.uk>
To: linux-fsdevel@vger.kernel.org
Cc: brauner@kernel.org, ebiederm@xmission.com, jack@suse.cz,
	torvalds@linux-foundation.org
Subject: [PATCH 09/26] clone_mnt(): simplify the propagation-related logics
Date: Tue, 10 Jun 2025 09:21:31 +0100	[thread overview]
Message-ID: <20250610082148.1127550-9-viro@zeniv.linux.org.uk> (raw)
In-Reply-To: <20250610082148.1127550-1-viro@zeniv.linux.org.uk>

The underlying rules are simple:
	* MNT_SHARED should be set iff ->mnt_group_id of new mount ends up
non-zero.
	* mounts should be on the same ->mnt_share cyclic list iff they have
the same non-zero ->mnt_group_id value.
	* CL_PRIVATE is mutually exclusive with MNT_SHARED, MNT_SLAVE,
MNT_SHARED_TO_SLAVE and MNT_EXPIRE; the whole point of that thing is to
get a clone of old mount that would *not* be on any namespace-related
lists.

The above allows to make the logics more straightforward; what's more,
it makes the proof that invariants are maintained much simpler.
The variant in mainline is safe (aside of a very narrow race with
unsafe modification of mnt_flags right after we had the mount exposed
in superblock's ->s_mounts; theoretically it can race with ro remount
of the original, but it's not easy to hit), but proof of its correctness
is really unpleasant.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namespace.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index d6c81eab6a11..02e9f37f49b9 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1337,6 +1337,9 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 	if (!mnt)
 		return ERR_PTR(-ENOMEM);
 
+	mnt->mnt.mnt_flags = READ_ONCE(old->mnt.mnt_flags) &
+			     ~MNT_INTERNAL_FLAGS;
+
 	if (flag & (CL_SLAVE | CL_PRIVATE | CL_SHARED_TO_SLAVE))
 		mnt->mnt_group_id = 0; /* not a peer of original */
 	else
@@ -1348,8 +1351,8 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 			goto out_free;
 	}
 
-	mnt->mnt.mnt_flags = old->mnt.mnt_flags;
-	mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL|MNT_LOCKED);
+	if (mnt->mnt_group_id)
+		set_mnt_shared(mnt);
 
 	atomic_inc(&sb->s_active);
 	mnt->mnt.mnt_idmap = mnt_idmap_get(mnt_idmap(&old->mnt));
@@ -1362,22 +1365,20 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 	list_add_tail(&mnt->mnt_instance, &sb->s_mounts);
 	unlock_mount_hash();
 
+	if (flag & CL_PRIVATE)	// we are done with it
+		return mnt;
+
+	if (peers(mnt, old))
+		list_add(&mnt->mnt_share, &old->mnt_share);
+
 	if ((flag & CL_SLAVE) ||
 	    ((flag & CL_SHARED_TO_SLAVE) && IS_MNT_SHARED(old))) {
 		list_add(&mnt->mnt_slave, &old->mnt_slave_list);
 		mnt->mnt_master = old;
-		CLEAR_MNT_SHARED(mnt);
-	} else if (!(flag & CL_PRIVATE)) {
-		if ((flag & CL_MAKE_SHARED) || IS_MNT_SHARED(old))
-			list_add(&mnt->mnt_share, &old->mnt_share);
-		if (IS_MNT_SLAVE(old))
-			list_add(&mnt->mnt_slave, &old->mnt_slave);
+	} else if (IS_MNT_SLAVE(old)) {
+		list_add(&mnt->mnt_slave, &old->mnt_slave);
 		mnt->mnt_master = old->mnt_master;
-	} else {
-		CLEAR_MNT_SHARED(mnt);
 	}
-	if (flag & CL_MAKE_SHARED)
-		set_mnt_shared(mnt);
 
 	/* stick the duplicate mount on the same expiry list
 	 * as the original if that was on one */
@@ -1385,7 +1386,6 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 		if (!list_empty(&old->mnt_expire))
 			list_add(&mnt->mnt_expire, &old->mnt_expire);
 	}
-
 	return mnt;
 
  out_free:
-- 
2.39.5


  parent reply	other threads:[~2025-06-10  8:21 UTC|newest]

Thread overview: 175+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-10  8:17 [PATCHES][RFC][CFR] mount-related stuff Al Viro
2025-06-10  8:21 ` [PATCH 01/26] copy_tree(): don't set ->mnt_mountpoint on the root of copy Al Viro
2025-06-10  8:21   ` [PATCH 02/26] constify mnt_has_parent() Al Viro
2025-06-11 10:26     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 03/26] pnode: lift peers() into pnode.h Al Viro
2025-06-11 10:29     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 04/26] new predicate: mount_is_ancestor() Al Viro
2025-06-11 10:32     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 05/26] constify is_local_mountpoint() Al Viro
2025-06-11 10:32     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 06/26] new predicate: anon_ns_root(mount) Al Viro
2025-06-11 10:39     ` Christian Brauner
2025-06-11 17:57       ` Al Viro
2025-06-10  8:21   ` [PATCH 07/26] dissolve_on_fput(): use anon_ns_root() Al Viro
2025-06-11 10:41     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 08/26] don't set MNT_LOCKED on parentless mounts Al Viro
2025-06-11 10:49     ` Christian Brauner
2025-06-10  8:21   ` Al Viro [this message]
2025-06-11 10:53     ` [PATCH 09/26] clone_mnt(): simplify the propagation-related logics Christian Brauner
2025-06-10  8:21   ` [PATCH 10/26] do_umount(): simplify the "is it still mounted" checks Al Viro
2025-06-11 10:54     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 11/26] sanitize handling of long-term internal mounts Al Viro
2025-06-11 10:56     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 12/26] Rewrite of propagate_umount() Al Viro
2025-06-11 10:56     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 13/26] attach_mnt(): expand in attach_recursive_mnt(), then lose the flag argument Al Viro
2025-06-11 10:59     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 14/26] do_move_mount(): take dropping the old mountpoint into attach_recursive_mnt() Al Viro
2025-06-11 10:59     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 15/26] get rid of mnt_set_mountpoint_beneath() Al Viro
2025-06-11 11:01     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 16/26] make commit_tree() usable in same-namespace move case Al Viro
2025-06-11 11:03     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 17/26] attach_recursive_mnt(): unify the mnt_change_mountpoint() logics Al Viro
2025-06-11 11:05     ` Christian Brauner
2025-06-11 18:12       ` Al Viro
2025-06-12 12:08         ` Christian Brauner
2025-06-10  8:21   ` [PATCH 18/26] attach_recursive_mnt(): pass destination mount in all cases Al Viro
2025-06-11 11:07     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 19/26] attach_recursive_mnt(): get rid of flags entirely Al Viro
2025-06-11 11:08     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 20/26] do_move_mount(): get rid of 'attached' flag Al Viro
2025-06-11 11:08     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 21/26] attach_recursive_mnt(): remove from expiry list on move Al Viro
2025-06-11 11:09     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 22/26] take ->mnt_expire handling under mount_lock [read_seqlock_excl] Al Viro
2025-06-11 11:11     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 23/26] pivot_root(): reorder tree surgeries, collapse unhash_mnt() and put_mountpoint() Al Viro
2025-06-11 11:11     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 24/26] combine __put_mountpoint() with unhash_mnt() Al Viro
2025-06-11 11:12     ` Christian Brauner
2025-06-10  8:21   ` [PATCH 25/26] get rid of mountpoint->m_count Al Viro
2025-06-11 11:19     ` Christian Brauner
2025-06-11 18:47       ` Al Viro
2025-06-16 20:38         ` Al Viro
2025-06-16 21:52           ` Linus Torvalds
2025-06-10  8:21   ` [PATCH 26/26] don't have mounts pin their parents Al Viro
2025-06-11 11:22     ` Christian Brauner
2025-06-16  2:50     ` Ian Kent
2025-06-10 22:30   ` [PATCH 01/26] copy_tree(): don't set ->mnt_mountpoint on the root of copy Eric W. Biederman
2025-06-10 23:14     ` Al Viro
2025-06-11 10:31 ` [PATCHES][RFC][CFR] mount-related stuff Christian Brauner
2025-06-11 17:51   ` Al Viro
2025-06-12 12:09     ` Christian Brauner
2025-06-23  4:49 ` [PATCHES v2][RFC][CFR] " Al Viro
2025-06-23  4:53   ` [PATCH v2 01/35] replace collect_mounts()/drop_collected_mounts() with a safer variant Al Viro
2025-06-23  4:53     ` [PATCH v2 02/35] attach_recursive_mnt(): do not lock the covering tree when sliding something under it Al Viro
2025-06-23  4:53     ` [PATCH v2 03/35] attach_mnt(): expand in attach_recursive_mnt(), then lose the flag argument Al Viro
2025-06-23  4:53     ` [PATCH v2 04/35] get rid of mnt_set_mountpoint_beneath() Al Viro
2025-06-23  4:53     ` [PATCH v2 05/35] prevent mount hash conflicts Al Viro
2025-06-23  4:53     ` [PATCH v2 06/35] copy_tree(): don't set ->mnt_mountpoint on the root of copy Al Viro
2025-06-23  4:54     ` [PATCH v2 07/35] constify mnt_has_parent() Al Viro
2025-06-23  4:54     ` [PATCH v2 08/35] pnode: lift peers() into pnode.h Al Viro
2025-06-23  4:54     ` [PATCH v2 09/35] new predicate: mount_is_ancestor() Al Viro
2025-06-23  4:54     ` [PATCH v2 10/35] constify is_local_mountpoint() Al Viro
2025-06-23  4:54     ` [PATCH v2 11/35] new predicate: anon_ns_root(mount) Al Viro
2025-06-23  4:54     ` [PATCH v2 12/35] dissolve_on_fput(): use anon_ns_root() Al Viro
2025-06-23  4:54     ` [PATCH v2 13/35] __attach_mnt(): lose the second argument Al Viro
2025-06-23  4:54     ` [PATCH v2 14/35] don't set MNT_LOCKED on parentless mounts Al Viro
2025-06-23  4:54     ` [PATCH v2 15/35] clone_mnt(): simplify the propagation-related logics Al Viro
2025-06-23  4:54     ` [PATCH v2 16/35] do_umount(): simplify the "is it still mounted" checks Al Viro
2025-06-23  4:54     ` [PATCH v2 17/35] sanitize handling of long-term internal mounts Al Viro
2025-06-23 16:18       ` Linus Torvalds
2025-06-23 17:03         ` Al Viro
2025-06-23 18:21           ` Linus Torvalds
2025-06-28  7:58           ` [RFC] vfs_parse_fs_string() calling conventions change (was Re: [PATCH v2 17/35] sanitize handling of long-term internal mounts) Al Viro
2025-06-28 16:28             ` Al Viro
2025-06-29 17:47               ` Al Viro
2025-06-28 17:41             ` Linus Torvalds
2025-06-30 15:19           ` David Howells
2025-06-30 16:55             ` Al Viro
2025-06-30 17:04               ` Linus Torvalds
2025-06-23  4:54     ` [PATCH v2 18/35] Rewrite of propagate_umount() Al Viro
2025-06-23  4:54     ` [PATCH v2 19/35] make commit_tree() usable in same-namespace move case Al Viro
2025-06-23  4:54     ` [PATCH v2 20/35] attach_recursive_mnt(): unify the mnt_change_mountpoint() logics Al Viro
2025-06-23  4:54     ` [PATCH v2 21/35] attach_recursive_mnt(): pass destination mount in all cases Al Viro
2025-06-23  4:54     ` [PATCH v2 22/35] attach_recursive_mnt(): get rid of flags entirely Al Viro
2025-06-23  4:54     ` [PATCH v2 23/35] do_move_mount(): take dropping the old mountpoint into attach_recursive_mnt() Al Viro
2025-06-23  4:54     ` [PATCH v2 24/35] do_move_mount(): get rid of 'attached' flag Al Viro
2025-06-23  4:54     ` [PATCH v2 25/35] attach_recursive_mnt(): remove from expiry list on move Al Viro
2025-06-23  4:54     ` [PATCH v2 26/35] take ->mnt_expire handling under mount_lock [read_seqlock_excl] Al Viro
2025-06-23  4:54     ` [PATCH v2 27/35] pivot_root(): reorder tree surgeries, collapse unhash_mnt() and put_mountpoint() Al Viro
2025-06-23  4:54     ` [PATCH v2 28/35] combine __put_mountpoint() with unhash_mnt() Al Viro
2025-06-23  4:54     ` [PATCH v2 29/35] get rid of mountpoint->m_count Al Viro
2025-06-23  4:54     ` [PATCH v2 30/35] don't have mounts pin their parents Al Viro
2025-06-23  4:54     ` [PATCH v2 31/35] copy_tree(): don't link the mounts via mnt_list Al Viro
2025-06-23  4:54     ` [PATCH v2 32/35] mount: separate the flags accessed only under namespace_sem Al Viro
2025-06-23  4:54     ` [PATCH v2 33/35] propagate_one(): get rid of dest_master Al Viro
2025-06-23  4:54     ` [PATCH v2 34/35] propagate_mnt(): get rid of globals Al Viro
2025-06-23  4:54     ` [PATCH v2 35/35] take freeing of emptied mnt_namespace to namespace_unlock() Al Viro
2025-06-23 15:10     ` [PATCH v2 01/35] replace collect_mounts()/drop_collected_mounts() with a safer variant Al Viro
2025-06-23  9:06   ` [PATCHES v2][RFC][CFR] mount-related stuff Ian Kent
2025-06-23 18:55     ` Al Viro
2025-06-24  6:48       ` Ian Kent
2025-06-24  7:05         ` Al Viro
2025-06-24 11:03           ` Ian Kent
2025-06-25  7:57         ` Al Viro
2025-06-25 10:58           ` Ian Kent
2025-06-27  3:03             ` Ian Kent
2025-06-30  2:51   ` [PATCHES v3][RFC][CFR] " Al Viro
2025-06-30  2:52     ` [PATCH v3 01/48] attach_mnt(): expand in attach_recursive_mnt(), then lose the flag argument Al Viro
2025-06-30  2:52       ` [PATCH v3 02/48] get rid of mnt_set_mountpoint_beneath() Al Viro
2025-06-30  2:52       ` [PATCH v3 03/48] prevent mount hash conflicts Al Viro
2025-06-30  2:52       ` [PATCH v3 04/48] copy_tree(): don't set ->mnt_mountpoint on the root of copy Al Viro
2025-06-30  2:52       ` [PATCH v3 05/48] constify mnt_has_parent() Al Viro
2025-06-30  2:52       ` [PATCH v3 06/48] pnode: lift peers() into pnode.h Al Viro
2025-06-30  2:52       ` [PATCH v3 07/48] new predicate: mount_is_ancestor() Al Viro
2025-06-30  2:52       ` [PATCH v3 08/48] constify is_local_mountpoint() Al Viro
2025-06-30  2:52       ` [PATCH v3 09/48] new predicate: anon_ns_root(mount) Al Viro
2025-06-30  2:52       ` [PATCH v3 10/48] dissolve_on_fput(): use anon_ns_root() Al Viro
2025-06-30  2:52       ` [PATCH v3 11/48] __attach_mnt(): lose the second argument Al Viro
2025-06-30  2:52       ` [PATCH v3 12/48] don't set MNT_LOCKED on parentless mounts Al Viro
2025-06-30  2:52       ` [PATCH v3 13/48] clone_mnt(): simplify the propagation-related logics Al Viro
2025-06-30  2:52       ` [PATCH v3 14/48] do_umount(): simplify the "is it still mounted" checks Al Viro
2025-06-30  2:52       ` [PATCH v3 15/48] sanitize handling of long-term internal mounts Al Viro
2025-06-30  2:52       ` [PATCH v3 16/48] Rewrite of propagate_umount() Al Viro
2025-06-30  2:52       ` [PATCH v3 17/48] make commit_tree() usable in same-namespace move case Al Viro
2025-06-30  2:52       ` [PATCH v3 18/48] attach_recursive_mnt(): unify the mnt_change_mountpoint() logics Al Viro
2025-06-30  2:52       ` [PATCH v3 19/48] attach_recursive_mnt(): pass destination mount in all cases Al Viro
2025-06-30  2:52       ` [PATCH v3 20/48] attach_recursive_mnt(): get rid of flags entirely Al Viro
2025-06-30  2:52       ` [PATCH v3 21/48] do_move_mount(): take dropping the old mountpoint into attach_recursive_mnt() Al Viro
2025-06-30  2:52       ` [PATCH v3 22/48] do_move_mount(): get rid of 'attached' flag Al Viro
2025-06-30  2:52       ` [PATCH v3 23/48] attach_recursive_mnt(): remove from expiry list on move Al Viro
2025-06-30  2:52       ` [PATCH v3 24/48] take ->mnt_expire handling under mount_lock [read_seqlock_excl] Al Viro
2025-06-30  2:52       ` [PATCH v3 25/48] pivot_root(): reorder tree surgeries, collapse unhash_mnt() and put_mountpoint() Al Viro
2025-06-30  2:52       ` [PATCH v3 26/48] combine __put_mountpoint() with unhash_mnt() Al Viro
2025-06-30  2:52       ` [PATCH v3 27/48] get rid of mountpoint->m_count Al Viro
2025-06-30  2:52       ` [PATCH v3 28/48] don't have mounts pin their parents Al Viro
2025-06-30  2:52       ` [PATCH v3 29/48] mount: separate the flags accessed only under namespace_sem Al Viro
2025-06-30  2:52       ` [PATCH v3 30/48] propagate_one(): get rid of dest_master Al Viro
2025-06-30  2:52       ` [PATCH v3 31/48] propagate_mnt(): handle all peer groups in the same loop Al Viro
2025-06-30  2:52       ` [PATCH v3 32/48] propagate_one(): separate the "do we need secondary here?" logics Al Viro
2025-06-30  2:52       ` [PATCH v3 33/48] propagate_one(): separate the "what should be the master for this copy" part Al Viro
2025-06-30  2:52       ` [PATCH v3 34/48] propagate_one(): fold into the sole caller Al Viro
2025-06-30  2:52       ` [PATCH v3 35/48] fs/pnode.c: get rid of globals Al Viro
2025-06-30  2:52       ` [PATCH v3 36/48] propagate_mnt(): get rid of last_dest Al Viro
2025-06-30  2:52       ` [PATCH v3 37/48] propagate_mnt(): fix comment and convert to kernel-doc, while we are at it Al Viro
2025-06-30  2:52       ` [PATCH v3 38/48] change_mnt_propagation() cleanups, step 1 Al Viro
2025-06-30  2:52       ` [PATCH v3 39/48] change_mnt_propagation(): do_make_slave() is a no-op unless IS_MNT_SHARED() Al Viro
2025-06-30  2:52       ` [PATCH v3 40/48] do_make_slave(): choose new master sanely Al Viro
2025-06-30  2:52       ` [PATCH v3 41/48] turn do_make_slave() into transfer_propagation() Al Viro
2025-06-30  2:52       ` [PATCH v3 42/48] mnt_slave_list/mnt_slave: turn into hlist_head/hlist_node Al Viro
2025-06-30  2:52       ` [PATCH v3 43/48] change_mnt_propagation(): move ->mnt_master assignment into MS_SLAVE case Al Viro
2025-06-30  2:52       ` [PATCH v3 44/48] copy_tree(): don't link the mounts via mnt_list Al Viro
2025-08-13  6:45         ` Lai, Yi
2025-08-13  7:13           ` Al Viro
2025-08-13  7:32             ` Al Viro
2025-08-14 23:21               ` Al Viro
2025-08-14 23:25                 ` Al Viro
2025-08-15  3:19                 ` Lai, Yi
2025-06-30  2:52       ` [PATCH v3 45/48] take freeing of emptied mnt_namespace to namespace_unlock() Al Viro
2025-06-30  2:52       ` [PATCH v3 46/48] get rid of CL_SHARE_TO_SLAVE Al Viro
2025-06-30  2:52       ` [PATCH v3 47/48] invent_group_ids(): zero ->mnt_group_id always implies !IS_MNT_SHARED() Al Viro
2025-06-30  2:52       ` [PATCH v3 48/48] statmount_mnt_basic(): simplify the logics for group id Al Viro
2025-07-02 19:29     ` [PATCHES v3][RFC][CFR] mount-related stuff 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=20250610082148.1127550-9-viro@zeniv.linux.org.uk \
    --to=viro@zeniv.linux.org.uk \
    --cc=brauner@kernel.org \
    --cc=ebiederm@xmission.com \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    /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).