[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <03f031e5-ed9c-4794-8f08-8a4007c1d704@linux.alibaba.com>
Date: Thu, 16 May 2024 09:41:58 +0800
From: Joseph Qi <joseph.qi@...ux.alibaba.com>
To: Ferry Meng <mengferry@...ux.alibaba.com>, Mark Fasheh <mark@...heh.com>,
Joel Becker <jlbec@...lplan.org>, ocfs2-devel@...ts.linux.dev
Cc: linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] ocfs2: strict bound check before memcmp in
ocfs2_xattr_find_entry()
On 5/15/24 9:29 PM, Ferry Meng wrote:
> xattr in ocfs2 maybe not INLINE, but saved with additional space
> requested. It's better to check if the memory is out of bound before
> memcmp, although this possibility mainly comes from custom poisonous
> images.
Specifically, this only addresses the case non-indexed xattr.
>
> Signed-off-by: Ferry Meng <mengferry@...ux.alibaba.com>
> ---
> fs/ocfs2/xattr.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
> index 37be4a286faf..4ceb0cb4cb71 100644
> --- a/fs/ocfs2/xattr.c
> +++ b/fs/ocfs2/xattr.c
> @@ -1083,10 +1083,15 @@ static int ocfs2_xattr_find_entry(struct inode *inode, void *end,
> cmp = name_index - ocfs2_xattr_get_type(entry);
Or define a local variable 'offset' for le16_to_cpu(entry->xe_name_offset).
Thanks,
Joseph
> if (!cmp)
> cmp = name_len - entry->xe_name_len;
> - if (!cmp)
> + if (!cmp) {
> + if ((xs->base + le16_to_cpu(entry->xe_name_offset) + name_len) > end) {
> + ocfs2_error(inode->i_sb, "corrupted xattr entries");
> + return -EFSCORRUPTED;
> + }
> cmp = memcmp(name, (xs->base +
> le16_to_cpu(entry->xe_name_offset)),
> name_len);
> + }
> if (cmp == 0)
> break;
> entry += 1;
Powered by blists - more mailing lists