Re: [E] Re: Patch "tunnels: Don't apply GRO to multiple layers of encapsulation" (CVE-2016-8666) is missing in 4.1 and 3.18 stable tree

[Greg KH <gregkh@linuxfoundation.org>]

On Fri, Jan 20, 2017 at 08:14:23AM +0900, Seung-Woo Kim wrote:
> Hi,
> 
> On Fri, Jan 13, 2017 at 2:51 AM,  <alexander.levin@verizon.com> wrote:
> > On Tue, Jan 10, 2017 at 07:45:48AM +0100, Greg KH wrote:
> >> On Tue, Jan 10, 2017 at 03:09:28AM +0100, Thomas Deutschmann wrote:
> >> > Hi,
> >> >
> >> > the following patch was backported to the following LTS kernels
> >> >
> >> > - >=4.4.29
> >> > - >=3.16.35
> >> >
> >> >
> >> > however it is missing from LTS kernels
> >> >
> >> > - linux-4.1
> >> > - linux-3.18
> >>
> >> 3.18 is (or should be), end-of-life now (Sasha, can we just kill it?
> >> Please?)
> >>
> >> 4.1 is a bit slower release cycle, and again, Sasha will speak to that.
> >
> > I'll add commit to both, and will do final release for 3.18 in a couple of days.
> 
> The commit is applied linux-4.1.y and linux-3.18.y, but it causes
> build warning about unused function sit_gro_receive.
> 
> It seems that the following part of mainline commit is missed.
> 
>  static int ipv6_gro_complete(struct sk_buff *skb, int nhoff)
>  {
>   const struct net_offload *ops;
> @@ -302,7 +315,7 @@ static struct packet_offload ipv6_packet
>  static const struct net_offload sit_offload = {
>   .callbacks = {
>   .gso_segment = ipv6_gso_segment,
> - .gro_receive    = ipv6_gro_receive,
> + .gro_receive    = sit_gro_receive,
>   .gro_complete   = sit_gro_complete,
>   },
>  };

Ah, nice catch.

Sasha, want to add this to the trees and do another release?  Or I can
do it if you don't want to, for 3.18.y

thanks,

greg k-h