linux-kernel - [PATCH] jfs: fix array-index-out-of-bounds in dtExtendPage

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: <20260121203121.1867-2-qikeyu2017@gmail.com>
Date: Thu, 22 Jan 2026 04:31:21 +0800
From: Kery Qi <qikeyu2017@...il.com>
To: shaggy@...nel.org
Cc: r.smirnov@....ru,
	zheng.yu@...thwestern.edu,
	duttaditya18@...il.com,
	jfs-discussion@...ts.sourceforge.net,
	linux-kernel@...r.kernel.org,
	Kery Qi <qikeyu2017@...il.com>
Subject: [PATCH] jfs: fix array-index-out-of-bounds in dtExtendPage

Similar to the issue fixed in commit 27e56f59bab5 ("UBSAN:
array-index-out-of-bounds in dtSplitRoot"), the loop that traverses
the freelist in dtExtendPage() uses the condition (fsi != -1) to
terminate. However, if the filesystem image is corrupted or maliciously
crafted, f->next could contain a value less than -1 (e.g., -2), causing
the loop to continue and resulting in an out-of-bounds array access
when indexing sp->slot[fsi].

Fix this by changing the loop termination condition from (fsi != -1)
to (fsi >= 0), so that any negative value will properly terminate the
loop.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kery Qi <qikeyu2017@...il.com>
---
 fs/jfs/jfs_dtree.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c
index 0ab83bb7bbdf..3e365ccfb33b 100644
--- a/fs/jfs/jfs_dtree.c
+++ b/fs/jfs/jfs_dtree.c
@@ -1804,7 +1804,7 @@ static int dtExtendPage(tid_t tid,
 		do {
 			f = &sp->slot[fsi];
 			fsi = f->next;
-		} while (fsi != -1);
+		} while (fsi >= 0);

 		f->next = n;
 	}
-- 
2.34.1