public inbox for linux-security-module@vger.kernel.org
 help / color / mirror / Atom feed
From: Al Viro <viro@zeniv.linux.org.uk>
To: alexjlzheng@gmail.com
Cc: paul@paul-moore.com, jmorris@namei.org, serge@hallyn.com,
	greg@kroah.com, chrisw@osdl.org,
	linux-security-module@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	Jinliang Zheng <alexjlzheng@tencent.com>
Subject: [PATCH 2/8] securityfs: pin filesystem only for objects directly in root
Date: Fri, 9 May 2025 05:38:47 +0100	[thread overview]
Message-ID: <20250509043847.GM2023217@ZenIV> (raw)
In-Reply-To: <20250509032326.GJ2023217@ZenIV>

From 23947dd21a91b95208863c74d4dbc67d0027f43a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 14 May 2024 08:46:00 -0600
Subject: [PATCH 2/8] securityfs: pin filesystem only for objects directly in
 root

Nothing on securityfs ever changes parents, so we don't need
to pin the internal mount if it's already pinned for parent.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 security/inode.c | 21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/security/inode.c b/security/inode.c
index 58cc60c50498..19ab99feeee3 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -112,18 +112,20 @@ static struct dentry *securityfs_create_dentry(const char *name, umode_t mode,
 	struct dentry *dentry;
 	struct inode *dir, *inode;
 	int error;
+	bool pinned = false;
 
 	if (!(mode & S_IFMT))
 		mode = (mode & S_IALLUGO) | S_IFREG;
 
 	pr_debug("securityfs: creating file '%s'\n",name);
 
-	error = simple_pin_fs(&fs_type, &mount, &mount_count);
-	if (error)
-		return ERR_PTR(error);
-
-	if (!parent)
+	if (!parent) {
+		error = simple_pin_fs(&fs_type, &mount, &mount_count);
+		if (error)
+			return ERR_PTR(error);
+		pinned = true;
 		parent = mount->mnt_root;
+	}
 
 	dir = d_inode(parent);
 
@@ -167,7 +169,8 @@ static struct dentry *securityfs_create_dentry(const char *name, umode_t mode,
 	dentry = ERR_PTR(error);
 out:
 	inode_unlock(dir);
-	simple_release_fs(&mount, &mount_count);
+	if (pinned)
+		simple_release_fs(&mount, &mount_count);
 	return dentry;
 }
 
@@ -307,13 +310,15 @@ void securityfs_remove(struct dentry *dentry)
 			simple_unlink(dir, dentry);
 	}
 	inode_unlock(dir);
-	simple_release_fs(&mount, &mount_count);
+	if (dir != dir->i_sb->s_root->d_inode)
+		simple_release_fs(&mount, &mount_count);
 }
 EXPORT_SYMBOL_GPL(securityfs_remove);
 
 static void remove_one(struct dentry *victim)
 {
-	simple_release_fs(&mount, &mount_count);
+	if (victim->d_parent != victim->d_sb->s_root)
+		simple_release_fs(&mount, &mount_count);
 }
 
 /**
-- 
2.39.5


  parent reply	other threads:[~2025-05-09  4:38 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-08 14:04 [PATCH v3] securityfs: fix missing of d_delete() in securityfs_remove() alexjlzheng
2025-05-09  1:55 ` Fan Wu
2025-05-09  2:45   ` Jinliang Zheng
2025-05-09  3:23 ` Al Viro
2025-05-09  4:37   ` Al Viro
2025-05-09  4:46     ` Al Viro
2025-05-12 21:19       ` Paul Moore
2025-05-12 22:24         ` Al Viro
2025-05-13  0:10         ` Fan Wu
2025-05-09  4:37   ` [PATCH 1/8] securityfs: don't pin dentries twice, once is enough Al Viro
2025-05-13 23:13     ` Paul Moore
2025-05-09  4:38   ` Al Viro [this message]
2025-05-09  4:39   ` [PATCH 3/8] fix locking in efi_secret_unlink() Al Viro
2025-05-09  4:39   ` [PATCH 4/8] make securityfs_remove() remove the entire subtree Al Viro
2025-05-09  4:40   ` [PATCH 5/8] efi_secret: clean securityfs use up Al Viro
2025-05-09  4:40   ` [PATCH 6/8] ima_fs: don't bother with removal of files in directory we'll be removing Al Viro
2025-05-09  4:41   ` [PATCH 7/8] ima_fs: get rid of lookup-by-dentry stuff Al Viro
2025-05-09  4:41   ` [PATCH 8/8] evm_secfs: clear securityfs interactions 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=20250509043847.GM2023217@ZenIV \
    --to=viro@zeniv.linux.org.uk \
    --cc=alexjlzheng@gmail.com \
    --cc=alexjlzheng@tencent.com \
    --cc=chrisw@osdl.org \
    --cc=greg@kroah.com \
    --cc=jmorris@namei.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=paul@paul-moore.com \
    --cc=serge@hallyn.com \
    /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