Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Zhang Yi <yi.zhang@huawei.com>
To:     <linux-ext4@vger.kernel.org>
CC:     <tytso@mit.edu>, <adilger.kernel@dilger.ca>, <jack@suse.cz>,
        <yi.zhang@huawei.com>, <yukuai3@huawei.com>
Subject: [PATCH v3 2/2] ext4: check and assert if marking an no_delete evicting inode dirty
Date:   Wed, 29 Jun 2022 19:26:47 +0800
Message-ID: <20220629112647.4141034-2-yi.zhang@huawei.com>
In-Reply-To: <20220629112647.4141034-1-yi.zhang@huawei.com>
References: <20220629112647.4141034-1-yi.zhang@huawei.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 7BIT
Content-Type:   text/plain; charset=US-ASCII
Precedence: bulk

In ext4_evict_inode(), if we evicting an inode in the 'no_delete' path,
it cannot be raced by another mark_inode_dirty(). If it happens,
someone else may accidentally dirty it without holding inode refcount
and probably cause use-after-free issues in the writeback procedure.
It's indiscoverable and hard to debug, so add an WARN_ON_ONCE() to
check and detect this issue in advance.

Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
---
v2->v3:
 - Switch to use WARN_ON_ONCE instead of ASSERT.

 fs/ext4/inode.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 702cc208689a..902393373152 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -333,6 +333,12 @@ void ext4_evict_inode(struct inode *inode)
 	ext4_xattr_inode_array_free(ea_inode_array);
 	return;
 no_delete:
+	/*
+	 * Check out some where else accidentally dirty the evicting inode,
+	 * which may probably cause inode use-after-free issues later.
+	 */
+	WARN_ON_ONCE(!list_empty_careful(&inode->i_io_list));
+
 	if (!list_empty(&EXT4_I(inode)->i_fc_list))
 		ext4_fc_mark_ineligible(inode->i_sb, EXT4_FC_REASON_NOMEM, NULL);
 	ext4_clear_inode(inode);	/* We must guarantee clearing of inode... */
-- 
2.31.1