Date:   Thu, 25 Nov 2021 20:17:29 -0800
From:   Jakub Kicinski <kuba@kernel.org>
To:     Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Cc:     vijayendra.suman@oracle.com, ramanan.govindarajan@oracle.com,
        george.kennedy@oracle.com, eric.dumazet@gmail.com,
        syzkaller <syzkaller@googlegroups.com>,
        Stephen Hemminger <stephen@networkplumber.org>,
        Jamal Hadi Salim <jhs@mojatatu.com>,
        Cong Wang <xiyou.wangcong@gmail.com>,
        Jiri Pirko <jiri@resnulli.us>,
        "David S. Miller" <davem@davemloft.net>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] net: sched: sch_netem: Fix a divide error in
 netem_enqueue during randomized corruption.
Message-ID: <20211125201729.15728683@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
In-Reply-To: <20211124104716.25882-1-harshit.m.mogalapalli@oracle.com>
References: <20211124104716.25882-1-harshit.m.mogalapalli@oracle.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Wed, 24 Nov 2021 02:47:15 -0800 Harshit Mogalapalli wrote:
> In netem_enqueue function the value of skb_headlen(skb) can be zero
> which leads to a division error during randomized corruption of the
> packet. This fix  adds a check to skb_headlen(skb) to prevent the division
> error.

Empty skbs are not sane. Is it just the fact that netlink_sendmsg()
does not pay much attention to len that's the problem? Can we fix that
instead?