Date:   Thu, 2 Mar 2023 19:26:36 +0000
From:   Al Viro <viro@zeniv.linux.org.uk>
To:     Jan Kara <jack@suse.cz>
Cc:     "Fabio M. De Francesco" <fmdefrancesco@gmail.com>,
        Jan Kara <jack@suse.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org
Subject: Re: [git pull] vfs.git sysv pile
Message-ID: <ZAD4bNDEIZf/VAnh@ZenIV>
References: <Y/gugbqq858QXJBY@ZenIV>
 <13214812.uLZWGnKmhe@suse>
 <20230301130018.yqds5yvqj7q26f7e@quack3>
 <Y/9duET0Mt5hPu2L@ZenIV>
 <20230302095931.jwyrlgtxcke7iwuu@quack3>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230302095931.jwyrlgtxcke7iwuu@quack3>
Sender: Al Viro <viro@ftp.linux.org.uk>
Precedence: bulk

On Thu, Mar 02, 2023 at 10:59:31AM +0100, Jan Kara wrote:
> OK, I think your changes to ext2_rename() in PATCH 1 leak a reference and
> mapping of old_page

In which case?  ext2_delete_entry() failing?

-       ext2_delete_entry(old_de, old_page, old_page_addr);
+       err = ext2_delete_entry(old_de, old_page, old_page_addr);
+       if (err)
+               goto out_dir;

and on out_dir: we have
out_dir:
        if (dir_de)
                ext2_put_page(dir_page, dir_page_addr);
out_old:
        ext2_put_page(old_page, old_page_addr);
out:
        return err;

How is the old_page leaked here?