Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <d37a412f244eaa0f1352b394fdf933ce75d9968c.camel@gmail.com>
Subject: memory leak in ext4_multi_mount_protect
From:   Pavel Skripkin <paskripkin@gmail.com>
To:     tytso@mit.edu, adilger.kernel@dilger.ca
Cc:     linux-ext4@vger.kernel.org,
        kernel list <linux-kernel@vger.kernel.org>
Date:   Tue, 13 Apr 2021 16:11:41 +0300
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.38.4 
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi!

I've done debugging on this issue
https://syzkaller.appspot.com/bug?id=420258a304e5d92cfef6b0097f87b42506e1db08
and I want to ask you about 
proper way of fixing it. The problem was in case sbi->s_mmp_tsk hasn’t
started at the time of kthread_stop() call. In that case allocated data
won't be freed.

I wrote fix patch, but I am confused about it, because I didn't find
any kernel code like this. I don't think, that adding new members to
struct super_block is good idea, that's why I came to that decision: 

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index b9693680463a..9c33e97bd5c5 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -5156,8 +5156,10 @@ static int ext4_fill_super(struct super_block
*sb, void *data, int silent)
 failed_mount3:
 	flush_work(&sbi->s_error_work);
 	del_timer_sync(&sbi->s_err_report);
-	if (sbi->s_mmp_tsk)
-		kthread_stop(sbi->s_mmp_tsk);
+	if (sbi->s_mmp_tsk) {
+		if (kthread_stop(sbi->s_mmp_tsk) == -EINTR)
+			kfree(kthread_data(sbi->s_mmp_tsk));
+	}
 failed_mount2:
 	rcu_read_lock();
 	group_desc = rcu_dereference(sbi->s_group_desc);


I look forward to hearing your perspective on this patch :)

With regards,
Pavel Skripkin