netdev - Re: [PATCH] [netfilter]: Fix skb->csum calculation when netfilter manipulation for NF_NAT_MANIP

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 23 Oct 2019 21:33:37 +0200
From:   Florian Westphal <fw@...len.de>
To:     Praveen Chaudhary <praveen5582@...il.com>
Cc:     davem@...emloft.net, fw@...len.de, kadlec@...filter.org,
        pablo@...filter.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org, Zhenggen Xu <zxu@...kedin.com>,
        Andy Stracner <astracner@...kedin.com>
Subject: Re: [PATCH] [netfilter]: Fix skb->csum calculation when netfilter
 manipulation for NF_NAT_MANIP_SRC\DST is done on IPV6 packet.

Praveen Chaudhary <praveen5582@...il.com> wrote:
> Update skb->csum, when netfilter code updates IPV6 SRC\DST address in IPV6 HEADER due to iptable rule.
> 
> Signed-off-by: Praveen Chaudhary <pchaudhary@...kedin.com>
> Signed-off-by: Zhenggen Xu <zxu@...kedin.com>
> Signed-off-by: Andy Stracner <astracner@...kedin.com>
> ---
>  include/net/checksum.h       |  2 ++
>  net/core/utils.c             | 13 +++++++++++++
>  net/netfilter/nf_nat_proto.c |  2 ++
>  3 files changed, 17 insertions(+)
> 
> diff --git a/include/net/checksum.h b/include/net/checksum.h
> index 97bf488..d7d28b7 100644
> --- a/include/net/checksum.h
> +++ b/include/net/checksum.h
> @@ -145,6 +145,8 @@ void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
>  void inet_proto_csum_replace16(__sum16 *sum, struct sk_buff *skb,
>  			       const __be32 *from, const __be32 *to,
>  			       bool pseudohdr);
> +void inet_proto_skb_csum_replace16(struct sk_buff *skb,
> +			       const __be32 *from, const __be32 *to);
>  void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb,
>  				     __wsum diff, bool pseudohdr);
>  
> diff --git a/net/core/utils.c b/net/core/utils.c
> index 6b6e51d..ab3284b 100644
> --- a/net/core/utils.c
> +++ b/net/core/utils.c
> @@ -458,6 +458,19 @@ void inet_proto_csum_replace16(__sum16 *sum, struct sk_buff *skb,
>  }
>  EXPORT_SYMBOL(inet_proto_csum_replace16);
>  
> +void inet_proto_skb_csum_replace16(struct sk_buff *skb,
> +			       const __be32 *from, const __be32 *to)
> +{
> +	__be32 diff[] = {
> +		~from[0], ~from[1], ~from[2], ~from[3],
> +		to[0], to[1], to[2], to[3],
> +	};
> +	if (skb->ip_summed == CHECKSUM_COMPLETE)
> +		skb->csum = csum_partial(diff, sizeof(diff),
> +				  skb->csum);
> +}
> +EXPORT_SYMBOL(inet_proto_skb_csum_replace16);
> +
>  void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb,
>  				     __wsum diff, bool pseudohdr)
>  {
> diff --git a/net/netfilter/nf_nat_proto.c b/net/netfilter/nf_nat_proto.c
> index 0a59c14..de94590 100644
> --- a/net/netfilter/nf_nat_proto.c
> +++ b/net/netfilter/nf_nat_proto.c
> @@ -467,6 +467,8 @@ static void nf_nat_ipv6_csum_update(struct sk_buff *skb,
>  	}
>  	inet_proto_csum_replace16(check, skb, oldip->s6_addr32,
>  				  newip->s6_addr32, true);
> +	inet_proto_skb_csum_replace16(skb, oldip->s6_addr32,
> +				  newip->s6_addr32);

This is confusing.

You're saying that inet_proto_csum_replace16() is producing a wrong
skb->csum.  So why are you adding a new function to do the
csum calculation instead of fixing inet_proto_csum_replace16() to do
the right thing?