All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ian Kent <raven@themaw.net>
To: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	David Howells <dhowells@redhat.com>,
	Miklos Szeredi <miklos@szeredi.hu>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: [PATCH 2/3] vfs: add propagate_mount_tree_busy() helper
Date: Mon, 11 Jul 2022 11:37:46 +0800	[thread overview]
Message-ID: <165751066658.210556.1326573473015621909.stgit@donald.themaw.net> (raw)
In-Reply-To: <165751053430.210556.5634228273667507299.stgit@donald.themaw.net>

Now that child mounts are tracked the expire checks need to be able to
use this to check if a mount is in use.

Currently when checking a mount for expiration may_umount_tree() checks
only if the passed in mount is in use. This leads to false positive
callbacks to the automount daemon to umount the mount which fail if any
propagated mounts are in use.

To avoid these unnecessary callbacks may_umount_tree() needs to check
propagated mounts in a similar way to may_umount().

Add a helper that can do this.

Signed-off-by: Ian Kent <raven@themaw.net>
---
 fs/pnode.c |   61 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/pnode.h |    3 +++
 2 files changed, 64 insertions(+)

diff --git a/fs/pnode.c b/fs/pnode.c
index 1106137c747a..e2a906db4324 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -404,6 +404,67 @@ int propagate_mount_busy(struct mount *mnt, int refcnt)
 	return 0;
 }
 
+static int do_mount_in_use_check(struct mount *mnt, int cnt)
+{
+	struct mount *topper;
+
+	/* Is there exactly one mount on the child that covers
+	 * it completely?
+	 */
+	topper = find_topper(mnt);
+	if (topper) {
+		int topper_cnt = topper->mnt_mounts_cnt + 1;
+
+		/* Open file or pwd within singular mount? */
+		if (do_refcount_check(topper, topper_cnt))
+			return 1;
+		/* Account for singular mount on parent */
+		cnt += 1;
+	}
+
+	if (do_refcount_check(mnt, cnt))
+		return 1;
+
+	return 0;
+}
+
+/*
+ * Check if the mount tree at 'mnt' is in use or any of its
+ * propogated mounts are in use.
+ * @mnt: the mount to be checked
+ * @adjust: caller holds an additional reference to mount
+ * Check if mnt or any of its propogated mounts have a reference
+ * count greater than the minimum reference count (ie. are in use).
+ */
+int propagate_mount_tree_busy(struct mount *mnt, unsigned int flags)
+{
+	struct mount *parent = mnt->mnt_parent;
+	struct mount *m, *child;
+	unsigned int referenced = flags & TREE_BUSY_REFERENCED;
+	int cnt;
+
+	/* Check for an elevated refcount on the passed in mount.
+	 * If adjust is true the caller holds a reference to the
+	 * passed in mount.
+	 */
+	cnt = mnt->mnt_mounts_cnt + (referenced ?  2 : 1);
+	if (do_mount_in_use_check(mnt, cnt))
+		return 1;
+
+	for (m = propagation_next(parent, parent); m;
+			m = propagation_next(m, parent)) {
+		child = __lookup_mnt(&m->mnt, mnt->mnt_mountpoint);
+		if (!child)
+			continue;
+
+		cnt = child->mnt_mounts_cnt + 1;
+
+		if (do_mount_in_use_check(child, cnt))
+			return 1;
+	}
+	return 0;
+}
+
 /*
  * Clear MNT_LOCKED when it can be shown to be safe.
  *
diff --git a/fs/pnode.h b/fs/pnode.h
index 988f1aa9b02a..d7b9dddb257b 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -30,6 +30,8 @@
 
 #define CL_COPY_ALL		(CL_COPY_UNBINDABLE | CL_COPY_MNT_NS_FILE)
 
+#define TREE_BUSY_REFERENCED	0x01
+
 static inline void set_mnt_shared(struct mount *mnt)
 {
 	mnt->mnt.mnt_flags &= ~MNT_SHARED_MASK;
@@ -41,6 +43,7 @@ int propagate_mnt(struct mount *, struct mountpoint *, struct mount *,
 		struct hlist_head *);
 int propagate_umount(struct list_head *);
 int propagate_mount_busy(struct mount *, int);
+int propagate_mount_tree_busy(struct mount *, unsigned int);
 void propagate_mount_unlock(struct mount *);
 void mnt_release_group_id(struct mount *);
 int get_dominating_id(struct mount *mnt, const struct path *root);



  parent reply	other threads:[~2022-07-11  3:38 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-11  3:37 [PATCH 0/3] autofs: fix may_umount_tree() Ian Kent
2022-07-11  3:37 ` [PATCH 1/3] vfs: track count of child mounts Ian Kent
2022-07-20  1:50   ` Al Viro
2022-07-20  2:17     ` Ian Kent
2022-07-20  7:26       ` Ian Kent
2022-07-26  5:11       ` Ian Kent
2022-07-26  7:10         ` Ian Kent
2022-07-11  3:37 ` Ian Kent [this message]
2022-07-20  1:54   ` [PATCH 2/3] vfs: add propagate_mount_tree_busy() helper Al Viro
2022-07-20  2:31     ` Ian Kent
2022-07-20  2:39       ` Al Viro
2022-07-11  3:37 ` [PATCH 3/3] vfs: make may_umount_tree() mount namespace aware Ian Kent

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=165751066658.210556.1326573473015621909.stgit@donald.themaw.net \
    --to=raven@themaw.net \
    --cc=akpm@linux-foundation.org \
    --cc=dhowells@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.