lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-Id: <20251230-fix-use-after-free-gfs2-v2-1-7b2760be547c@gmail.com>
Date: Tue, 30 Dec 2025 03:08:46 +0900
From: Ryota Sakamoto <sakamo.ryota@...il.com>
To: Andreas Gruenbacher <agruenba@...hat.com>
Cc: Markus Elfring <Markus.Elfring@....de>, gfs2@...ts.linux.dev, 
 linux-kernel@...r.kernel.org, 
 syzbot+4cb0d0336db6bc6930e9@...kaller.appspotmail.com, 
 stable@...r.kernel.org, Ryota Sakamoto <sakamo.ryota@...il.com>
Subject: [PATCH v2] gfs2: Fix use-after-free in gfs2_fill_super()

The issue occurs when gfs2_freeze_lock_shared() fails in
gfs2_fill_super(). If !sb_rdonly(sb), threads for the quotad and logd
were started, however, in the error path for gfs2_freeze_lock_shared(),
the threads are not stopped by gfs2_destroy_threads() before jumping to
fail_per_node.

Introduce fail_threads to handle stopping the threads if the threads were
started.

Reported-by: syzbot+4cb0d0336db6bc6930e9@...kaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=4cb0d0336db6bc6930e9
Fixes: a28dc123fa66 ("gfs2: init system threads before freeze lock")
Cc: stable@...r.kernel.org
Signed-off-by: Ryota Sakamoto <sakamo.ryota@...il.com>
---
Changes in v2:
- Fix commit message style (imperative mood) as suggested by Markus Elfring.
- Add parentheses to function name in subject as suggested by Markus Elfring.
- Link to v1: https://lore.kernel.org/r/20251230-fix-use-after-free-gfs2-v1-1-ef0e46db6ec9@gmail.com
---
 fs/gfs2/ops_fstype.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index e7a88b717991ae3647c1da039636daef7005a7f0..4b5ac1a7050f1fd34e10be4100a2bc381f49c83d 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1269,21 +1269,23 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 
 	error = gfs2_freeze_lock_shared(sdp);
 	if (error)
-		goto fail_per_node;
+		goto fail_threads;
 
 	if (!sb_rdonly(sb))
 		error = gfs2_make_fs_rw(sdp);
 
 	if (error) {
 		gfs2_freeze_unlock(sdp);
-		gfs2_destroy_threads(sdp);
 		fs_err(sdp, "can't make FS RW: %d\n", error);
-		goto fail_per_node;
+		goto fail_threads;
 	}
 	gfs2_glock_dq_uninit(&mount_gh);
 	gfs2_online_uevent(sdp);
 	return 0;
 
+fail_threads:
+	if (!sb_rdonly(sb))
+		gfs2_destroy_threads(sdp);
 fail_per_node:
 	init_per_node(sdp, UNDO);
 fail_inodes:

---
base-commit: 7839932417dd53bb09eb5a585a7a92781dfd7cb2
change-id: 20251230-fix-use-after-free-gfs2-66cfbe23baa8

Best regards,
-- 
Ryota Sakamoto <sakamo.ryota@...il.com>


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ