[ovs-dev] [PATCH net-next v2] openvswitch: add TTL decrement action

bindiya Kurle bindiyakurle at gmail.com
Tue Dec 24 23:16:24 UTC 2019


Hi Tonghao,
Once this kernel patch is integrated . I will be submitting patch to
ovs-dpdk  for implementing dec_ttl action on dpdk datapath.
Regards,
Bindiya

On Tue, Dec 24, 2019 at 2:12 PM Tonghao Zhang <xiangxia.m.yue at gmail.com>
wrote:

> Hi Matteo,
> Did you have plan to implement the TTL decrement action in userspace
> datapath(with dpdk),
> I am doing some research offloading about TTL decrement action, and
> may sent patch TTL decrement offload action,
> using dpdk rte_flow.
>
> On Fri, Dec 20, 2019 at 8:37 PM Matteo Croce <mcroce at redhat.com> wrote:
> >
> > On Tue, Dec 17, 2019 at 5:30 PM Nikolay Aleksandrov
> > <nikolay at cumulusnetworks.com> wrote:
> > >
> > > On 17/12/2019 17:51, Matteo Croce wrote:
> > > > New action to decrement TTL instead of setting it to a fixed value.
> > > > This action will decrement the TTL and, in case of expired TTL, drop
> it
> > > > or execute an action passed via a nested attribute.
> > > > The default TTL expired action is to drop the packet.
> > > >
> > > > Supports both IPv4 and IPv6 via the ttl and hop_limit fields,
> respectively.
> > > >
> > > > Tested with a corresponding change in the userspace:
> > > >
> > > >     # ovs-dpctl dump-flows
> > > >     in_port(2),eth(),eth_type(0x0800), packets:0, bytes:0,
> used:never, actions:dec_ttl{ttl<=1 action:(drop)},1,1
> > > >     in_port(1),eth(),eth_type(0x0800), packets:0, bytes:0,
> used:never, actions:dec_ttl{ttl<=1 action:(drop)},1,2
> > > >     in_port(1),eth(),eth_type(0x0806), packets:0, bytes:0,
> used:never, actions:2
> > > >     in_port(2),eth(),eth_type(0x0806), packets:0, bytes:0,
> used:never, actions:1
> > > >
> > > >     # ping -c1 192.168.0.2 -t 42
> > > >     IP (tos 0x0, ttl 41, id 61647, offset 0, flags [DF], proto ICMP
> (1), length 84)
> > > >         192.168.0.1 > 192.168.0.2: ICMP echo request, id 386, seq
> 1, length 64
> > > >     # ping -c1 192.168.0.2 -t 120
> > > >     IP (tos 0x0, ttl 119, id 62070, offset 0, flags [DF], proto ICMP
> (1), length 84)
> > > >         192.168.0.1 > 192.168.0.2: ICMP echo request, id 388, seq
> 1, length 64
> > > >     # ping -c1 192.168.0.2 -t 1
> > > >     #
> > > >
> > > > Co-authored-by: Bindiya Kurle <bindiyakurle at gmail.com>
> > > > Signed-off-by: Bindiya Kurle <bindiyakurle at gmail.com>
> > > > Signed-off-by: Matteo Croce <mcroce at redhat.com>
> > > > ---
> > > >  include/uapi/linux/openvswitch.h |  22 +++++++
> > > >  net/openvswitch/actions.c        |  71 +++++++++++++++++++++
> > > >  net/openvswitch/flow_netlink.c   | 105
> +++++++++++++++++++++++++++++++
> > > >  3 files changed, 198 insertions(+)
> > > >
> > >
> > > Hi Matteo,
> > >
> > > [snip]
> > > > +}
> > > > +
> > > >  /* When 'last' is true, sample() should always consume the 'skb'.
> > > >   * Otherwise, sample() should keep 'skb' intact regardless what
> > > >   * actions are executed within sample().
> > > > @@ -1176,6 +1201,44 @@ static int execute_check_pkt_len(struct
> datapath *dp, struct sk_buff *skb,
> > > >                            nla_len(actions), last, clone_flow_key);
> > > >  }
> > > >
> > > > +static int execute_dec_ttl(struct sk_buff *skb, struct sw_flow_key
> *key)
> > > > +{
> > > > +     int err;
> > > > +
> > > > +     if (skb->protocol == htons(ETH_P_IPV6)) {
> > > > +             struct ipv6hdr *nh = ipv6_hdr(skb);
> > > > +
> > > > +             err = skb_ensure_writable(skb, skb_network_offset(skb)
> +
> > > > +                                       sizeof(*nh));
> > >
> > > skb_ensure_writable() calls pskb_may_pull() which may reallocate so nh
> might become invalid.
> > > It seems the IPv4 version below is ok as the ptr is reloaded.
> > >
> >
> > Right
> >
> > > One q as I don't know ovs that much - can this action be called only
> with
> > > skb->protocol ==  ETH_P_IP/IPV6 ? I.e. Are we sure that if it's not
> v6, then it must be v4 ?
> > >
> >
> > I'm adding a check in validate_and_copy_dec_ttl() so only ipv4/ipv6
> > packet will pass.
> >
> > Thanks,
> >
> > --
> > Matteo Croce
> > per aspera ad upstream
> >
> > _______________________________________________
> > dev mailing list
> > dev at openvswitch.org
> > https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>


More information about the dev mailing list