All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hui Peng <benquike@gmail.com>
To: Damien Le Moal <dlemoal@kernel.org>,
	Christian Brauner <brauner@kernel.org>, Jan Kara <jack@suse.cz>,
	Jeff Layton <jlayton@kernel.org>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	stable@vger.kernel.org, Hui Peng <benquike@gmail.com>
Subject: [PATCH v2 4/6] qnx6: release sb_buf on mmi_fs error path in qnx6_fill_super()
Date: Mon, 21 Sep 2026 04:25:09 +0000	[thread overview]
Message-ID: <20260921042511.1473629-4-benquike@gmail.com> (raw)
In-Reply-To: <20260921042511.1473629-1-benquike@gmail.com>

When mounting with "-o mmi_fs" (QNX6_MOUNT_MMI_FS), qnx6_mmi_fill_super()
allocates the active superblock buffer_head and stores it in qs->sb_buf
(QNX6_SB(s)->sb_buf), while qnx6_fill_super()'s local bh1 and bh2 pointers
remain NULL, and jumps to mmi_success. If a subsequent validation or inode
initialization check after mmi_success fails and jumps to out, out1, out2,
or out3, the error cleanup at out only calls brelse(bh1) and brelse(bh2)
(both NULL), leaking the qs->sb_buf buffer_head reference.

Release qs->sb_buf at out when it is distinct from bh1 and bh2 (using
qs->sb_buf rather than sbi->sb_buf because sbi is not yet initialized if
mmi_success jumps to out on superblock level validation failure).

Tested in QEMU against Linux 7.3.0-rc3 by mounting a crafted QNX6 mmi_fs
image on /dev/loop0 that succeeds in qnx6_mmi_fill_super() (allocating
qs->sb_buf at block 0) and then fails root directory validation, followed
by flushing per-CPU bh_lrus and pagecache via /proc/sys/vm/drop_caches and
POSIX_FADV_DONTNEED: on the unfixed kernel block 0 remains pinned in
bd_mapping by the leaked qs->sb_buf buffer_head (sb_buf_page0_leaked = 1),
whereas on the fixed kernel qs->sb_buf is released and evicted cleanly
(sb_buf_page0_leaked = 0).

Fixes: 5d026c724220 ("fs: initial qnx6fs addition")
Cc: stable@vger.kernel.org
Assisted-by: LLM
Signed-off-by: Hui Peng <benquike@gmail.com>
---
Changes in v2:
- Split the mmi_fs qs->sb_buf leak fix into its own patch (4/6) separate
  from the double-brelse fix (3/6), use qs->sb_buf instead of
  uninitialized sbi->sb_buf, and remove Markdown formatting from the
  commit message, as requested by Damien Le Moal.

 fs/qnx6/inode.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c
index c4a35d81e67c..124b60b47c8e 100644
--- a/fs/qnx6/inode.c
+++ b/fs/qnx6/inode.c
@@ -467,6 +467,8 @@ static int qnx6_fill_super(struct super_block *s, struct fs_context *fc)
 out1:
 	iput(sbi->inodes);
 out:
+	if (qs->sb_buf && qs->sb_buf != bh1 && qs->sb_buf != bh2)
+		brelse(qs->sb_buf);
 	brelse(bh1);
 	brelse(bh2);
 outnobh:
-- 
2.49.0

  parent reply	other threads:[~2026-09-21  4:25 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-19 22:25 [PATCH] qnx6: validate di_filelevels in qnx6_iget() and fix mount error handling Hui Peng
2026-09-20  4:02 ` Damien Le Moal
2026-09-21  4:25   ` [PATCH v2 1/6] qnx6: validate di_filelevels in qnx6_iget() Hui Peng
2026-09-21  4:25     ` [PATCH v2 2/6] qnx6: release buffer_head on error in qnx6_block_map() Hui Peng
2026-09-24  4:31       ` Matthias Goergens
2026-09-24  7:39       ` [PATCH v3 1/6] qnx6: validate di_filelevels in qnx6_iget() before accessing level pointers Hui Peng
2026-09-21  4:25     ` [PATCH v2 3/6] qnx6: avoid double brelse() on error path in qnx6_fill_super() Hui Peng
2026-09-24  4:31       ` Matthias Goergens
2026-09-24  7:39       ` [PATCH v3 2/6] qnx6: release bh on error path in qnx6_block_map() Hui Peng
2026-09-21  4:25     ` Hui Peng [this message]
2026-09-24  4:31       ` [PATCH v2 4/6] qnx6: release sb_buf on mmi_fs error path in qnx6_fill_super() Matthias Goergens
2026-09-24  7:39       ` [PATCH v3 3/6] qnx6: avoid double brelse() on " Hui Peng
2026-09-21  4:25     ` [PATCH v2 5/6] qnx6: abort mount on superblock magic mismatch when silent is set Hui Peng
2026-09-24  4:31       ` Matthias Goergens
2026-09-24  7:39       ` [PATCH v3 4/6] qnx6: release sb_buf on mmi_fs error path in qnx6_fill_super() Hui Peng
2026-09-21  4:25     ` [PATCH v2 6/6] qnx6: validate sb_blocksize before dividing in qnx6_mmi_fill_super() Hui Peng
2026-09-24  4:31       ` Matthias Goergens
2026-09-24  7:39       ` [PATCH v3 5/6] qnx6: abort mount on superblock magic mismatch when silent is set Hui Peng
2026-09-24  4:31     ` [PATCH v2 1/6] qnx6: validate di_filelevels in qnx6_iget() Matthias Goergens
2026-09-24  7:39     ` [PATCH v3 0/6] fs/qnx6: fix buffer head leaks, double free, and inode validation Hui Peng
2026-09-24 10:11       ` Matthias Goergens
2026-09-30  3:15       ` [PATCH v4 " Hui Peng
2026-09-30  7:22         ` Matthias Goergens
2026-09-30  3:15       ` [PATCH v4 1/6] qnx6: validate di_filelevels in qnx6_iget() before accessing level pointers Hui Peng
2026-09-30  7:22         ` Matthias Goergens
2026-09-30  3:16       ` [PATCH v4 2/6] qnx6: release bh on error path in qnx6_block_map() Hui Peng
2026-09-30  7:22         ` Matthias Goergens
2026-09-30  3:16       ` [PATCH v4 3/6] qnx6: avoid double brelse() and fix sb_buf leak on error path in qnx6_fill_super() Hui Peng
2026-09-30  3:16       ` [PATCH v4 4/6] qnx6: release bh2/bh1 on active superblock selection in qnx6_mmi_fill_super() Hui Peng
2026-09-30  3:16       ` [PATCH v4 5/6] qnx6: abort mount on superblock magic mismatch when silent is set " Hui Peng
2026-09-30  7:22         ` Matthias Goergens
2026-09-30  3:16       ` [PATCH v4 6/6] qnx6: validate sb_blocksize before dividing " Hui Peng
2026-09-30  7:22         ` Matthias Goergens
     [not found]     ` <20260921042511.1473629-7-benquike@gmail.com>
2026-09-24  7:39       ` [PATCH v3 " Hui Peng

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=20260921042511.1473629-4-benquike@gmail.com \
    --to=benquike@gmail.com \
    --cc=brauner@kernel.org \
    --cc=dlemoal@kernel.org \
    --cc=jack@suse.cz \
    --cc=jlayton@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.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 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.