Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 29 Sep 2022 12:54:52 +0300
From:   Leon Romanovsky <leon@kernel.org>
To:     Christian Langrock <christian.langrock@secunet.com>
Cc:     Steffen Klassert <steffen.klassert@secunet.com>,
        herbert@gondor.apana.org.au, davem@davemloft.net,
        edumazet@google.com, kuba@kernel.org, pabeni@redhat.com,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH ipsec v3] xfrm: replay: Fix ESN wrap around for GSO
Message-ID: <YzVrbHw9LMoTjYzI@unreal>
References: <02b5650c-29f4-568f-b3be-689594dfacc2@secunet.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <02b5650c-29f4-568f-b3be-689594dfacc2@secunet.com>
Precedence: bulk

On Thu, Sep 29, 2022 at 07:59:31AM +0200, Christian Langrock wrote:
> When using GSO it can happen that the wrong seq_hi is used for the last
> packets before the wrap around. This can lead to double usage of a
> sequence number. To avoid this, we should serialize this last GSO
> packet.
> 
> Changes in v3:
> - fix build
> - remove wrapper function
> 
> Changes in v2:
> - switch to bool as return value
> - remove switch case in wrapper function
> 
> Fixes: d7dbefc45cf5 ("xfrm: Add xfrm_replay_overflow functions for...")
> Signed-off-by: Christian Langrock <christian.langrock@secunet.com>
> ---

Please put changelog after "---" trailer.

>  include/net/xfrm.h     |  1 +
>  net/xfrm/xfrm_output.c |  2 +-
>  net/xfrm/xfrm_replay.c | 26 ++++++++++++++++++++++++++
>  3 files changed, 28 insertions(+), 1 deletion(-)
> 
> diff --git a/include/net/xfrm.h b/include/net/xfrm.h
> index 6e8fa98f786f..b845f911767c 100644
> --- a/include/net/xfrm.h
> +++ b/include/net/xfrm.h
> @@ -1749,6 +1749,7 @@ void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq);
>  int xfrm_replay_check(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq);
>  void xfrm_replay_notify(struct xfrm_state *x, int event);
>  int xfrm_replay_overflow(struct xfrm_state *x, struct sk_buff *skb);
> +bool xfrm_replay_overflow_check(struct xfrm_state *x, struct sk_buff *skb);
>  int xfrm_replay_recheck(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq);
>  
>  static inline int xfrm_aevent_is_on(struct net *net)
> diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
> index 9a5e79a38c67..c470a68d9c88 100644
> --- a/net/xfrm/xfrm_output.c
> +++ b/net/xfrm/xfrm_output.c
> @@ -738,7 +738,7 @@ int xfrm_output(struct sock *sk, struct sk_buff *skb)
>  		skb->encapsulation = 1;
>  
>  		if (skb_is_gso(skb)) {
> -			if (skb->inner_protocol)
> +			if (skb->inner_protocol || xfrm_replay_overflow_check(x, skb))
>  				return xfrm_output_gso(net, sk, skb);
>  
>  			skb_shinfo(skb)->gso_type |= SKB_GSO_ESP;
> diff --git a/net/xfrm/xfrm_replay.c b/net/xfrm/xfrm_replay.c
> index 9277d81b344c..23858eb5eab4 100644
> --- a/net/xfrm/xfrm_replay.c
> +++ b/net/xfrm/xfrm_replay.c
> @@ -750,6 +750,27 @@ int xfrm_replay_overflow(struct xfrm_state *x, struct sk_buff *skb)
>  
>  	return xfrm_replay_overflow_offload(x, skb);
>  }
> +
> +static bool xfrm_replay_overflow_check(struct xfrm_state *x, struct sk_buff *skb)
> +{
> +	struct xfrm_replay_state_esn *replay_esn = x->replay_esn;
> +	__u32 oseq = replay_esn->oseq;
> +
> +	/* We assume that this function is called with
> +	 * skb_is_gso(skb) == true
> +	 */
> +
> +	if (x->repl_mode == XFRM_REPLAY_MODE_ESN) {
> +		if (x->type->flags & XFRM_TYPE_REPLAY_PROT) {

It can be one if( ... && ...), but not critical.

Once you fix commit message, feel free to add my tag.

Thanks,
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>