Re: [ovs-dev] [PATCH net-next 00/22 v2] Lightweight flow based encapsulation
On 7/22/15, 1:58 AM, thomas.mo...@orange.com wrote: Hi Thomas, This looks promising. One question: will this approach allow MPLS-in-GRE and MPLS-in-UDP ? The current series was focused on IP to MPLS tunnels. But, the infrastructure allows associating encap state with routes and calling the respective encapsulation output handlers. For MPLS-in-GRE and MPLS-in-UDP, it appears that MPLS LSP routes (af_mpls.c) could support associating GRE and UDP encap info (RTA_ENCAP and RTA_ENCAP_TYPE) similar to IP routes and the re-direction from MPLS to GRE or MPLS to UDP can be achieved in a similar way using the same infrastructure. Thanks, Roopa -- To unsubscribe from this list: send the line unsubscribe netdev in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ovs-dev] [PATCH net-next 00/22 v2] Lightweight flow based encapsulation
Hi Thomas, This looks promising. One question: will this approach allow MPLS-in-GRE and MPLS-in-UDP ? -Thomas 2015-07-21, Thomas Graf: This series combines the work previously posted by Roopa, Robert and myself. It's according to what we discussed at NFWS. The motivation of this series is to: * Consolidate code between OVS and the rest of the kernel and get rid of OVS vports and instead represent them as pure net_devices. * Introduce a lightweight tunneling mechanism which enables flow based encapsulation to improve scalability on both RX and TX. * Do the above in an encapsulation unspecific way so that the encapsulation type is eventually abstracted away from the user. * Use the same forwarding decision for both native forwarding and encapsulation thus allowing to switch between native IPv6 and UDP encapsulation based on endpoint without requiring additional logic The fundamental changes introduces in this series are: * A new RTA_ENCAP Netlink attribute for routes carrying encapsulation instructions. Depending on the specified type, the instructions apply to UDP encapsulations, MPLS and possible other in the future. * Depending on the encapsulation type, the output function of the dst is directly overwritten or the dst merely attaches metadata and relies on a subsequent net_device to apply it to the packet. The latter is typically used if an inner and outer IP header exist which require two subsequent routing lookups to be performed. * A new metadata_dst structure which can be attached to skbs to carry metadata in between subsystems. This new metadata transport is used to provide a single interface for VXLAN, routing and OVS to communicate through metadata. The OVS interfaces remain as-is but will transparently create a real VXLAN net_device in the background. iproute2 is extended with a new use cases: VXLAN: ip route add 40.1.1.1/32 encap vxlan id 10 dst 50.1.1.2 dev vxlan0 MPLS: ip route add 10.1.1.0/30 encap mpls 200 via inet 10.1.1.1 dev swp1 Performance implications: The additional memory allocation in the receive path should have performance implications although it is not observable in standard throughput tests if GRO is properly done. The correct net_device model outweights the additional cost of the allocation. Furthermore, this implication can be relaxed by reintroducing a direct unqueued path from a software device to a consumer like bridge or OVS if needed. $ netperf -t TCP_STREAM -H 15.1.1.201 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 15.1.1.201 (15.1.1.201) port 0 AF_INET : demo Recv SendSend Socket Socket Message Elapsed Size SizeSize Time Throughput bytes bytes bytessecs.10^6bits/sec 87380 16384 1638410.009118.17 Changes since v1: * Properly initialize tun_id as reported by Julian * Drop dupliate netif_keep_dst() as reported by Alexei Roopa Prabhu (9): rtnetlink: introduce new RTA_ENCAP_TYPE and RTA_ENCAP attributes lwtunnel: infrastructure for handling light weight tunnels like mpls ipv4: support for fib route lwtunnel encap attributes ipv6: support for fib route lwtunnel encap attributes lwtunnel: support dst output redirect function ipv4: redirect dst output to lwtunnel output ipv6: rt6_info output redirect to tunnel output mpls: export mpls functions for use by mpls iptunnels mpls: ip tunnel support Thomas Graf (13): ip_tunnel: Make ovs_tunnel_info and ovs_key_ipv4_tunnel generic icmp: Don't leak original dst into ip_route_input() dst: Metadata destinations arp: Inherit metadata dst when creating ARP requests vxlan: Flow based tunneling route: Extend flow representation with tunnel key route: Per route IP tunnel metadata via lightweight tunnel fib: Add fib rule match on tunnel id vxlan: Factor out device configuration openvswitch: Make tunnel set action attach a metadata dst openvswitch: Move dev pointer into vport itself openvswitch: Abstract vport name through ovs_vport_name() openvswitch: Use regular VXLAN net_device device drivers/net/vxlan.c | 672 +-- include/linux/lwtunnel.h | 6 + include/linux/mpls_iptunnel.h| 6 + include/linux/skbuff.h | 1 + include/net/dst.h| 6 +- include/net/dst_metadata.h | 55 +++ include/net/fib_rules.h | 1 + include/net/flow.h | 8 + include/net/ip6_fib.h| 3 + include/net/ip_fib.h | 5 +- include/net/ip_tunnels.h | 95 - include/net/lwtunnel.h | 144 include/net/mpls_iptunnel.h | 29 ++ include/net/route.h | 1 + include/net/rtnetlink.h | 1 +
Re: [PATCH net-next 00/22 v2] Lightweight flow based encapsulation
From: Thomas Graf tg...@suug.ch Date: Tue, 21 Jul 2015 10:43:44 +0200 This series combines the work previously posted by Roopa, Robert and myself. It's according to what we discussed at NFWS. The motivation of this series is to: * Consolidate code between OVS and the rest of the kernel and get rid of OVS vports and instead represent them as pure net_devices. * Introduce a lightweight tunneling mechanism which enables flow based encapsulation to improve scalability on both RX and TX. * Do the above in an encapsulation unspecific way so that the encapsulation type is eventually abstracted away from the user. * Use the same forwarding decision for both native forwarding and encapsulation thus allowing to switch between native IPv6 and UDP encapsulation based on endpoint without requiring additional logic The fundamental changes introduces in this series are: * A new RTA_ENCAP Netlink attribute for routes carrying encapsulation instructions. Depending on the specified type, the instructions apply to UDP encapsulations, MPLS and possible other in the future. * Depending on the encapsulation type, the output function of the dst is directly overwritten or the dst merely attaches metadata and relies on a subsequent net_device to apply it to the packet. The latter is typically used if an inner and outer IP header exist which require two subsequent routing lookups to be performed. * A new metadata_dst structure which can be attached to skbs to carry metadata in between subsystems. This new metadata transport is used to provide a single interface for VXLAN, routing and OVS to communicate through metadata. Series applied, but please take Alexei's endianness feedback into consideration. Thanks! -- To unsubscribe from this list: send the line unsubscribe netdev in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH net-next 00/22 v2] Lightweight flow based encapsulation
This series combines the work previously posted by Roopa, Robert and myself. It's according to what we discussed at NFWS. The motivation of this series is to: * Consolidate code between OVS and the rest of the kernel and get rid of OVS vports and instead represent them as pure net_devices. * Introduce a lightweight tunneling mechanism which enables flow based encapsulation to improve scalability on both RX and TX. * Do the above in an encapsulation unspecific way so that the encapsulation type is eventually abstracted away from the user. * Use the same forwarding decision for both native forwarding and encapsulation thus allowing to switch between native IPv6 and UDP encapsulation based on endpoint without requiring additional logic The fundamental changes introduces in this series are: * A new RTA_ENCAP Netlink attribute for routes carrying encapsulation instructions. Depending on the specified type, the instructions apply to UDP encapsulations, MPLS and possible other in the future. * Depending on the encapsulation type, the output function of the dst is directly overwritten or the dst merely attaches metadata and relies on a subsequent net_device to apply it to the packet. The latter is typically used if an inner and outer IP header exist which require two subsequent routing lookups to be performed. * A new metadata_dst structure which can be attached to skbs to carry metadata in between subsystems. This new metadata transport is used to provide a single interface for VXLAN, routing and OVS to communicate through metadata. The OVS interfaces remain as-is but will transparently create a real VXLAN net_device in the background. iproute2 is extended with a new use cases: VXLAN: ip route add 40.1.1.1/32 encap vxlan id 10 dst 50.1.1.2 dev vxlan0 MPLS: ip route add 10.1.1.0/30 encap mpls 200 via inet 10.1.1.1 dev swp1 Performance implications: The additional memory allocation in the receive path should have performance implications although it is not observable in standard throughput tests if GRO is properly done. The correct net_device model outweights the additional cost of the allocation. Furthermore, this implication can be relaxed by reintroducing a direct unqueued path from a software device to a consumer like bridge or OVS if needed. $ netperf -t TCP_STREAM -H 15.1.1.201 MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 15.1.1.201 (15.1.1.201) port 0 AF_INET : demo Recv SendSend Socket Socket Message Elapsed Size SizeSize Time Throughput bytes bytes bytessecs.10^6bits/sec 87380 16384 1638410.009118.17 Changes since v1: * Properly initialize tun_id as reported by Julian * Drop dupliate netif_keep_dst() as reported by Alexei Roopa Prabhu (9): rtnetlink: introduce new RTA_ENCAP_TYPE and RTA_ENCAP attributes lwtunnel: infrastructure for handling light weight tunnels like mpls ipv4: support for fib route lwtunnel encap attributes ipv6: support for fib route lwtunnel encap attributes lwtunnel: support dst output redirect function ipv4: redirect dst output to lwtunnel output ipv6: rt6_info output redirect to tunnel output mpls: export mpls functions for use by mpls iptunnels mpls: ip tunnel support Thomas Graf (13): ip_tunnel: Make ovs_tunnel_info and ovs_key_ipv4_tunnel generic icmp: Don't leak original dst into ip_route_input() dst: Metadata destinations arp: Inherit metadata dst when creating ARP requests vxlan: Flow based tunneling route: Extend flow representation with tunnel key route: Per route IP tunnel metadata via lightweight tunnel fib: Add fib rule match on tunnel id vxlan: Factor out device configuration openvswitch: Make tunnel set action attach a metadata dst openvswitch: Move dev pointer into vport itself openvswitch: Abstract vport name through ovs_vport_name() openvswitch: Use regular VXLAN net_device device drivers/net/vxlan.c | 672 +-- include/linux/lwtunnel.h | 6 + include/linux/mpls_iptunnel.h| 6 + include/linux/skbuff.h | 1 + include/net/dst.h| 6 +- include/net/dst_metadata.h | 55 +++ include/net/fib_rules.h | 1 + include/net/flow.h | 8 + include/net/ip6_fib.h| 3 + include/net/ip_fib.h | 5 +- include/net/ip_tunnels.h | 95 - include/net/lwtunnel.h | 144 include/net/mpls_iptunnel.h | 29 ++ include/net/route.h | 1 + include/net/rtnetlink.h | 1 + include/net/vxlan.h | 85 - include/uapi/linux/fib_rules.h | 2 +- include/uapi/linux/if_link.h | 1 + include/uapi/linux/lwtunnel.h| 16 + include/uapi/linux/mpls_iptunnel.h | 28 ++
