From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH net] ipv6: fix udpv6 sendmsg crash caused by too small
 MTU
Date: Wed, 10 Jan 2018 09:21:51 -0800
Message-ID: <1515604911.131759.23.camel@gmail.com>
References: <20180110171013.129868-1-maloneykernel@gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Cc: Mike Maloney <maloney@google.com>
To: Mike Maloney <maloneykernel@gmail.com>, netdev@vger.kernel.org,
        davem@davemloft.net
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pg0-f65.google.com ([74.125.83.65]:36414 "EHLO
        mail-pg0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751205AbeAJRVy (ORCPT
        <rfc822;netdev@vger.kernel.org>); Wed, 10 Jan 2018 12:21:54 -0500
Received: by mail-pg0-f65.google.com with SMTP id j2so10119346pgv.3
        for <netdev@vger.kernel.org>; Wed, 10 Jan 2018 09:21:53 -0800 (PST)
In-Reply-To: <20180110171013.129868-1-maloneykernel@gmail.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Wed, 2018-01-10 at 12:10 -0500, Mike Maloney wrote:
> From: Mike Maloney <maloney@google.com>
> 
> The logic in __ip6_append_data() assumes that the MTU is at least large
> enough for the headers.  A device's MTU may be adjusted after being
> added while sendmsg() is processing data, resulting in
> __ip6_append_data() seeing any MTU.  For an mtu smaller than the size of
> the fragmentation header, the math results in a negative 'maxfraglen',
> which causes problems when refragmenting any previous skb in the
> skb_write_queue, leaving it possibly malformed.
> 
> Instead sendmsg returns EINVAL when the mtu is calculated to be less
> than IPV6_MIN_MTU.
> 

You forgot your SOB

> Reported-by: syzbot <syzkaller@googlegroups.com>
> ---

Also please add after this '---' marker that your patch depends on my
prior fix ( https://patchwork.ozlabs.org/patch/858234/ 
ipv6: fix possible mem leaks in ipv6_make_skb() )

( Or we leak a dst reference )

We probably should have sent a patch series.

Thanks.

>  net/ipv6/ip6_output.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
> index f7dd51c42314..f309ce7120d0 100644
> --- a/net/ipv6/ip6_output.c
> +++ b/net/ipv6/ip6_output.c
> @@ -1206,14 +1206,16 @@ static int ip6_setup_cork(struct sock *sk, struct inet_cork_full *cork,
>  	v6_cork->tclass = ipc6->tclass;
>  	if (rt->dst.flags & DST_XFRM_TUNNEL)
>  		mtu = np->pmtudisc >= IPV6_PMTUDISC_PROBE ?
> -		      rt->dst.dev->mtu : dst_mtu(&rt->dst);
> +		      READ_ONCE(rt->dst.dev->mtu) : dst_mtu(&rt->dst);
>  	else
>  		mtu = np->pmtudisc >= IPV6_PMTUDISC_PROBE ?
> -		      rt->dst.dev->mtu : dst_mtu(rt->dst.path);
> +		      READ_ONCE(rt->dst.dev->mtu) : dst_mtu(rt->dst.path);
>  	if (np->frag_size < mtu) {
>  		if (np->frag_size)
>  			mtu = np->frag_size;
>  	}
> +	if (mtu < IPV6_MIN_MTU)
> +		return -EINVAL;
>  	cork->base.fragsize = mtu;
>  	if (dst_allfrag(rt->dst.path))
>  		cork->base.flags |= IPCORK_ALLFRAG;