Improve fabric.p4 to reduce pipeline resources and refactor pipeconf impl
This patch affects both the P4 pipeline implementation and the
Java pipeconf.
P4 PIPELINE
- Less tables and smarter use of metadata to reduce inter-tables
dependencies and favor parallel execution of tables.
- Removed unused actions / renamed existing ones to make forwarding
behavior clearer (e.g. ingress_port_vlan table)
- Remove co-existence of simple and hansed table. Hashed should be the
default one, but implementations that do not support action profiles
might compile fabric.p4 to use the simple one.
- Use @name annotations for match fields to make control plane
independent of table implementation.
- Use @hidden to avoid showing actions and table on the p4info that
cannot be controlled at runtime.
- First attempt to support double VLAN cross-connect (xconnect table).
- New design has been tested with "fabric-refactoring" branch of
fabric-p4test:
github.com/opennetworkinglab/fabric-p4test/tree/fabric-refactoring
JAVA PIPECONF
This patch brings a major refactoring that reflects the experience
gathered in the past months of working on fabric.p4 and reasoning on its
pipeconf implementation. Indeed, the FlowObjective API is
under-specified and sometimes ambiguous which makes the process of
creating and maintaining a pipeliner implementation tedious. This
refactoring brings a simplified implementation by removing unused/
unnecessary functionalities and by recognizing commonality when possible
(e.g. by means of abstract and utility classes). It also makes design
patterns more explicit and consistent. Overall, the goal is to reduce
technical debt and to make it easier to support new features as we
evolve fabric.p4
Changes include:
- Changes in pipeliner/interpreter to reflect new pipeline design.
- By default translate objective treatment to PiAction. This favors
debuggability of flow rules in ONOS.
- Support new NextObjective’s NextTreatment class.
- Remove lots of unused/unnecessary code (e.g. async callback handling
for pending objective install status in pipeliner as current
implementation was always returning success)
- Gather commonality in abstract classes and simplify implementation
for objective translator (filtering, forwarding, next)
- New implementation of ForwardingFunctionTypes (FFT) that looks at
criterion instance values along with their types (to avoid relying on
case-specific if-else conditions to recognize variants of an FFT)
- Adaptive translation of NextObjective based on presence of simple or
hashed table.
- Support DENY FilteringObjective
Also:
- Fix onos-p4-gen-constants to avoid generating conflicting
PiMatchFieldId variable names.
- Install Graphviz tools in p4vm to generate p4c graphs
- Generate p4c graphs by default when compiling fabric.p4
- Use more compact Hex string when printing PI values
Change-Id: Ife79e44054dc5bc48833f95d0551a7370150eac5
diff --git a/pipelines/fabric/src/main/resources/include/control/filtering.p4 b/pipelines/fabric/src/main/resources/include/control/filtering.p4
index 3c5bfcb..638d9a1 100644
--- a/pipelines/fabric/src/main/resources/include/control/filtering.p4
+++ b/pipelines/fabric/src/main/resources/include/control/filtering.p4
@@ -18,80 +18,63 @@
#include <v1model.p4>
#include "../header.p4"
-#include "../action.p4"
-control Filtering (
- inout parsed_headers_t hdr,
- inout fabric_metadata_t fabric_metadata,
- inout standard_metadata_t standard_metadata) {
+control Filtering (inout parsed_headers_t hdr,
+ inout fabric_metadata_t fabric_metadata,
+ inout standard_metadata_t standard_metadata) {
/*
* Ingress Port VLAN Table.
- * Process packets for different interfaces (Port number + VLAN).
- * For example, an untagged packet will be tagged when it entered to an
- * interface with untagged VLAN configuration.
+ *
+ * Filter packets based on ingress port and VLAN tag.
*/
direct_counter(CounterType.packets_and_bytes) ingress_port_vlan_counter;
- action drop() {
- mark_to_drop();
+ action deny() {
+ // Packet from unconfigured port. Skip forwarding and next block.
+ // Do ACL table in case we want to punt to cpu.
+ fabric_metadata.skip_forwarding = _TRUE;
+ fabric_metadata.skip_next = _TRUE;
ingress_port_vlan_counter.count();
}
- action set_vlan(vlan_id_t new_vlan_id) {
- hdr.vlan_tag.vlan_id = new_vlan_id;
+ action permit() {
+ // Allow packet as is.
ingress_port_vlan_counter.count();
}
- action push_internal_vlan(vlan_id_t new_vlan_id) {
- // Add internal VLAN header, will be removed before packet emission.
- // cfi and pri values are dummy.
- hdr.vlan_tag.setValid();
- hdr.vlan_tag.cfi = 0;
- hdr.vlan_tag.pri = 0;
- hdr.vlan_tag.ether_type = hdr.ethernet.ether_type;
- hdr.ethernet.ether_type = ETHERTYPE_VLAN;
- hdr.vlan_tag.vlan_id = new_vlan_id;
-
- // pop internal vlan before packet in
- fabric_metadata.pop_vlan_when_packet_in = _TRUE;
- ingress_port_vlan_counter.count();
- }
-
- action nop_ingress_port_vlan() {
- nop();
+ action permit_with_internal_vlan(vlan_id_t vlan_id) {
+ fabric_metadata.vlan_id = vlan_id;
ingress_port_vlan_counter.count();
}
table ingress_port_vlan {
key = {
- standard_metadata.ingress_port: exact;
- hdr.vlan_tag.isValid(): exact @name("hdr.vlan_tag.is_valid");
- hdr.vlan_tag.vlan_id: ternary;
+ standard_metadata.ingress_port: exact @name("ig_port");
+ hdr.vlan_tag.isValid(): exact @name("vlan_is_valid");
+ hdr.vlan_tag.vlan_id: ternary @name("vlan_id");
}
-
actions = {
- push_internal_vlan;
- set_vlan;
- drop;
- nop_ingress_port_vlan();
+ deny();
+ permit();
+ permit_with_internal_vlan();
}
-
- const default_action = push_internal_vlan(DEFAULT_VLAN_ID);
+ const default_action = deny();
counters = ingress_port_vlan_counter;
}
/*
* Forwarding Classifier.
- * Setup Forwarding Type metadata for Forwarding control block.
+ *
+ * Set which type of forwarding behavior to execute in the next control block.
* There are six types of tables in Forwarding control block:
* - Bridging: default forwarding type
* - MPLS: destination mac address is the router mac and ethernet type is
- * MPLS(0x8847)
+ * MPLS(0x8847)
* - IP Multicast: destination mac address is multicast address and ethernet
- * type is IP(0x0800 or 0x86dd)
+ * type is IP(0x0800 or 0x86dd)
* - IP Unicast: destination mac address is router mac and ethernet type is
- * IP(0x0800 or 0x86dd)
+ * IP(0x0800 or 0x86dd)
*/
direct_counter(CounterType.packets_and_bytes) fwd_classifier_counter;
@@ -102,26 +85,35 @@
table fwd_classifier {
key = {
- standard_metadata.ingress_port: exact;
- hdr.ethernet.dst_addr: ternary;
- hdr.vlan_tag.ether_type: exact;
+ standard_metadata.ingress_port: exact @name("ig_port");
+ hdr.ethernet.dst_addr: ternary @name("eth_dst");
+ fabric_metadata.eth_type: exact @name("eth_type");
}
-
actions = {
set_forwarding_type;
}
-
const default_action = set_forwarding_type(FWD_BRIDGING);
counters = fwd_classifier_counter;
}
apply {
- if (ingress_port_vlan.apply().hit) {
- fwd_classifier.apply();
- } else {
- // Packet from unconfigured port. Skip forwarding processing,
- // except for ACL table in case we want to punt to cpu.
- fabric_metadata.fwd_type = FWD_UNKNOWN;
+ // Initialize lookup metadata. Packets without a VLAN header will be
+ // treated as belonging to a default VLAN ID (see parser).
+ if (hdr.vlan_tag.isValid()) {
+ fabric_metadata.eth_type = hdr.vlan_tag.eth_type;
+ fabric_metadata.vlan_id = hdr.vlan_tag.vlan_id;
+ fabric_metadata.vlan_pri = hdr.vlan_tag.pri;
+ fabric_metadata.vlan_cfi = hdr.vlan_tag.cfi;
}
+ if (!hdr.mpls.isValid()) {
+ // Packets with a valid MPLS header will have
+ // fabric_metadata.mpls_ttl set to the packet's MPLS ttl value (see
+ // parser). In any case, if we are forwarding via MPLS, ttl will be
+ // decremented in egress.
+ fabric_metadata.mpls_ttl = DEFAULT_MPLS_TTL + 1;
+ }
+
+ ingress_port_vlan.apply();
+ fwd_classifier.apply();
}
}