[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110529232337.GD5856@verge.net.au>
Date: Mon, 30 May 2011 08:23:37 +0900
From: Simon Horman <horms@...ge.net.au>
To: Jesper Juhl <jj@...osbits.net>
Cc: linux-kernel@...r.kernel.org, netfilter@...r.kernel.org,
coreteam@...filter.org, netfilter-devel@...r.kernel.org,
lvs-devel@...r.kernel.org, netdev@...r.kernel.org,
"David S. Miller" <davem@...emloft.net>,
Patrick McHardy <kaber@...sh.net>,
Julian Anastasov <ja@....bg>,
Wensong Zhang <wensong@...ux-vs.org>,
Pablo Neira Ayuso <pablo@...filter.org>
Subject: Re: [PATCH] netfilter, ipvs: Avoid undefined order of evaluation in
assignments to struct nf_conn *
On Sun, May 29, 2011 at 08:22:56PM +0200, Jesper Juhl wrote:
> In net/netfilter/ipvs/ip_vs_nfct.c::ip_vs_update_conntrack(),
> net/netfilter/ipvs/ip_vs_xmit.c::ip_vs_nat_xmit(),
> net/netfilter/ipvs/ip_vs_xmit.c::ip_vs_nat_xmit_v6(),
> net/netfilter/ipvs/ip_vs_xmit.c::ip_vs_icmp_xmit)()
> net/netfilter/ipvs/ip_vs_xmit.c::and ip_vs_icmp_xmit_v6() we do this:
> ...
> struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo);
> ...
>
> Since '=' is not a sequence point the order of these assignments happening
> is undefined. Luckily it's easy to avoid by just doing what is obviously
> the intended thing:
> struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
>
> Signed-off-by: Jesper Juhl <jj@...osbits.net>
Acked-by: Simon Horman <horms@...ge.net.au>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists