[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <m2h6c26w7o.fsf@ja-home.int.chopps.org>
Date: Sat, 03 Aug 2024 00:55:46 -0400
From: Christian Hopps <chopps@...pps.org>
To: Florian Westphal <fw@...len.de>
Cc: Christian Hopps <chopps@...pps.org>, devel@...ux-ipsec.org, Steffen
Klassert <steffen.klassert@...unet.com>, netdev@...r.kernel.org, Christian
Hopps <chopps@...n.net>
Subject: Re: [PATCH ipsec-next v7 08/16] xfrm: iptfs: add user packet
(tunnel ingress) handling
Florian Westphal <fw@...len.de> writes:
> Christian Hopps <chopps@...pps.org> wrote:
>> +static void iptfs_output_queued(struct xfrm_state *x, struct sk_buff_head *list)
>> +{
>> + struct xfrm_iptfs_data *xtfs = x->mode_data;
>> + struct sk_buff *skb, *skb2, **nextp;
>> + struct skb_shared_info *shi;
>> +
> [..]
>> + if (skb->dev)
>> + XFRM_INC_STATS(dev_net(skb->dev),
>> + LINUX_MIB_XFRMOUTERROR);
>
> Nit:
> Here and in several places you are using dev_net() helper.
>
> I think that if xfrm_state is available, then xs_net(x) makes more sense.
>
> It also avoids the skb->dev conditional.
Ok, replaced where xfrm_state was available.
Powered by blists - more mailing lists