Re: [PATCH 1/1] xfrm: nat_keepalive: avoid double free on send error

[Steffen Klassert <steffen.klassert@secunet.com>]

On Fri, Jun 19, 2026 at 05:06:40PM +0800, Qianyu Luo wrote:
> On Fri, Jun 19, 2026 at 1:21 AM Eyal Birger <eyal.birger@gmail.com> wrote:
> >
> > On Thu, Jun 18, 2026 at 9:36 AM Ren Wei <n05ec@lzu.edu.cn> wrote:
> > >
> > > From: Qianyu Luo <qianyuluo3@gmail.com>
> > >
> > >         skb_dst_set(skb, &rt->dst);
> > >
> > > @@ -100,6 +102,7 @@ static int nat_keepalive_send_ipv6(struct sk_buff *skb,
> > >         sock_net_set(sk, net);
> > >         dst = ip6_dst_lookup_flow(net, sk, &fl6, NULL);
> > >         if (IS_ERR(dst)) {
> > > +               kfree_skb(skb);
> > >                 local_unlock_nested_bh(&nat_keepalive_sk_ipv6.bh_lock);
> >
> > Any reason to do the kfree under lock?
> >
> 
> Thank you for your reply! I did this without particular reason.
> Just to keep the pre-handoff cleanup in the same error path.
> kfree_skb() does not need the bh lock there, so I can move it after the
> unlock in v2 if you need!

Then move it after the unlock to make it clear it is not needed.