On Wed, Feb 13, 2019 at 12:11 PM David Ahern wrote:
>
> On 2/13/19 12:57 PM, Peter Oskolkov wrote:
> > Thanks, David! I was not able to reproduce the leak, but based on your
> > suggestion and similar code elsewhere I made a change in v11 to explicitly
> > release a dst with error.
>
> ok. Did you
On 2/13/19 12:57 PM, Peter Oskolkov wrote:
> Thanks, David! I was not able to reproduce the leak, but based on your
> suggestion and similar code elsewhere I made a change in v11 to explicitly
> release a dst with error.
ok. Did you run the test with a debug kernel - checking refcount, use
after f
On Tue, Feb 12, 2019 at 6:58 PM David Ahern wrote:
>
> On 2/12/19 10:32 AM, Peter Oskolkov wrote:
> > @@ -148,6 +174,87 @@ static int xmit_check_hhlen(struct sk_buff *skb)
> > return 0;
> > }
> >
> > +static int bpf_lwt_xmit_reroute(struct sk_buff *skb)
> > +{
> > + struct net_device *l
On 2/12/19 10:32 AM, Peter Oskolkov wrote:
> @@ -148,6 +174,87 @@ static int xmit_check_hhlen(struct sk_buff *skb)
> return 0;
> }
>
> +static int bpf_lwt_xmit_reroute(struct sk_buff *skb)
> +{
> + struct net_device *l3mdev = l3mdev_master_dev_rcu(skb_dst(skb)->dev);
> + int oif =
This patch builds on top of the previous patch in the patchset,
which added BPF_LWT_ENCAP_IP mode to bpf_lwt_push_encap. As the
encapping can result in the skb needing to go via a different
interface/route/dst, bpf programs can indicate this by returning
BPF_LWT_REROUTE, which triggers a new route
