Currently, IPsec tunnel traffic goes to ip4-input/ip6-input after esp-encrypt.
It is not necessary to check that the new IP header is valid (if it is not
valid then we have otehr issues).
Instead, just send packets straight to ip4-lookup/ip6-lookup after esp-encrypt.
Change-Id: I5e35d500cb0f33f418f8554ed1f4390f02b6647d
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
This change shows around 7% improvement in ip4 path
when ip4-output features are not used.
Change-Id: I81c1c4c38c99fff0ad8270d3c0df907acf1d8c58
Signed-off-by: Damjan Marion <damarion@cisco.com>
Improve handling of timeouts. Add a workaround for when timing wheel
fires an event a tiny amount of time before it should actually be
fired. Don't delete unneeded events at all from timing wheel, instead
ignoring unexpected events. Enable the skipped BFD test, which passes
now.
Change-Id: I6ffd4fc0ba7a049ffe63bb0e5290641a7300dd6f
Signed-off-by: Klement Sekera <ksekera@cisco.com>
1. sr-rewrite - SR insertion in v6 : SR RH is inserted immediately after v6 header
but if hbh header is present as per RFC2460 it should immediately follow
v6 header. This is fixed.
2. sr-local : v6 packet destined to a sr segment is received if hbh is present
it is not handed over to sr-local for processing. fixed ip6-local handling to
skip hbh as there is no register handler for hbh for now.
3. sr-replicate - update in dual of sr_rewrite to handle replicate, fixes in
sr-replicate to handle presence of hbh header
Change-Id: I034523a42d2fedf97134761f956ab534babb8b36
Signed-off-by: Shwetha <shwethab@cisco.com>
The vpp control plane thread generates different types of packets such as ARP,
DHCP client, IPv6 RA, etc. which need to go through HQoS thread associated
with the output interfaces. Therefore, this patch enables vpp main thread
connection with the HQoS thread through dedicated SWQ.
Change-Id: Idcb759546d4bcd9b40beedaec4f12b29baae3e40
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Properly wakeup the main process loop only when necessary to avoid
missed events. Improve BFD unit test reliability and speed. Add
timestamps to logs and replace Queue with more efficient deque.
Change-Id: I01255a02057d3146917b43b669ccdc74f31ef0c8
Signed-off-by: Klement Sekera <ksekera@cisco.com>
- On VXLAN packet decap, validate its DIP against VXLAN tunnel.
- Add extra logic to validate and handle creation of multicast
VXLAN tunnels.
Change-Id: I6abdddd7be4cd9f1bcfc88d9970ba681fdd72f7c
Signed-off-by: John Lo <loj@cisco.com>
See Jira ticket for more details
- New plugins created to (a) Add VxLAN-GPE as transport (b) Provide export infra for
VxLAN-GPE.
Change-Id: Ife50c7434f53d17a4783062310f73d063d53494c
Signed-off-by: Vengada Govindan <venggovi@cisco.com>
Instead of storing a pointer, store an offset within vlib_buffer marking
the start of the ip header (similar to storing start of ethernet
header). Handle negative timeout in the bfd process main loop.
Change-Id: I05a1ff3ac41da5bdc3b2ac6f9e03b3241994688b
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Sometimes NICs are flagging checksum 0 as bad, which is not
always true.
Change-Id: Ibb0147c94f436966e6afa472025bdf92c41f201e
Signed-off-by: Damjan Marion <damarion@cisco.com>
This is a work-in-progress basic BFD session handling. Only
asynchronous mode is supported at the moment. Setting the session flags
doesn't work.
Change-Id: Idba27f721b5c35be5a66a6d202a63d23ff7ecf6f
Signed-off-by: Klement Sekera <ksekera@cisco.com>
The re-entrant walks were caused when the walk from an IP adj updated a fib_netry with an MPLS adj which in turn triggers a walk of the IP adj. Re-entrant walks do unnecessary work.
also fixed a walk merge issue where the encountered walk should only be checked for equivalence woth the most recent alk, not any in the list. Otherwise an UO,DOWN,UP beceoms (2*)UP,DOWN
Change-Id: Ib8b27f055dc6c1366d33740276d1c26cd314220a
Signed-off-by: Neale Ranns <nranns@cisco.com>
- s/l2/span in span_api.c in foreach macro
- Not installing .json from Makefile.am
Change-Id: I2469fe01138d62e044b8262cdb5a6e3b009bb43e
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Ib6dd290085e6f9a434499af8d19f346220dc8428
Signed-off-by: Damjan Marion <damarion@cisco.com>
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
This patch significantly improves vhost Tx and Rx
performance (20% Mpps improvement in PVP scenario).
Most significant change consists in the split of
virtio ring parsing and memory copies.
Other minor changes mostly consist in getting
rid of unnecessary variables.
Change-Id: Ia94e12ab9a6c114e3f7fef376a9343823d051209
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
After a bug was found, I decided to review old vhost code.
This change cleans some of it by:
- Use unix file private data rather than hash table
- Use vnet interface lookups rather than hash table
- Keep track of unix_fd index rather than fd value
- Change some indent
- Store interfaces in a clib pool rather than hand-made pool
The overall change reduces the number of lines as well as
the amount of state kept in vhost structures.
Change-Id: I6b9b670b9c7e1fb85981bcf2099f7140b0251fe0
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
added API to dump ipv4/ipv6 neighboors (added by ip_neighbor_add_del).
Change-Id: I33209a3d06beba64d68465c0892a9f4c65657334
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
Add a flag to the vnet_buffer to indicate a packet is locally originated. Then in the rewrite nodes we can check this flag and not perform the TTL decrement. The switch path cost is expected to be fractions of a clock - the flags will be hot in the cache.
The cehcks are necessary to due the requirements that VPP must be able to emit an IP packet with TTL=255.
Change-Id: Ieb9cf06e34df54fd5c950293de8b665016295c51
Signed-off-by: Neale Ranns <nranns@cisco.com>
Add ICMP echo reply punt logic in ip[46]-icmp-echo-reply (ping) nodes.
ICMP echo reply packets corresponding to the locally (VPP) originated
ICMP echo requests are still dropped as before. Rest of the (unknown)
ICMP echo reply packets are pushed to the “error-punt” node.
Also added ICMP echo packet (id/seq) trace information
Change-Id: I998198430dedc9b4d771b6aff2a97f18598663f9
Signed-off-by: Alexander Popovsky <apopovsk@cisco.com>
In DPDK 16.11 bonded PMD is still using old naming ‘rte_bond_pmd’.
See: dpdk-16.11/drivers/net/bonding/rte_eth_bond_api.c
Fixed crash when VPP is generating a name for ‘unknown’ non-PCI interface
Change-Id: I59e15f009e1a11949a19abf5234d09425a81c6ce
Signed-off-by: Alexander Popovsky <apopovsk@cisco.com>
Flatten variant was used only by old native drivers.
Change-Id: If206c6c8b4f10843d60f908c3acb4e8db83ce03d
Signed-off-by: Damjan Marion <damarion@cisco.com>
PatchSet 9: add range check for decap_next_index
PatchSet 6: delete runtime range check
Change-Id: I415e156d05d09d2ff34f6578924f0b243058f464
Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
Change I4ad40f398f7b5cf introduced a bug with chained buffers.
This patch should fix it.
Change-Id: Ia409d5eb90fcd6d7ff3c7da0836edfd2eb25998b
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
This change allows us to keep feature path disabled
until 1st feature is enabled. Enabling 1st feature also means
that end feature node is enabled helping feature arc tenants
to reach the end.
Change-Id: Idbd64e681bd2b42e7c67612074649e5ae51f46e6
Signed-off-by: Damjan Marion <damarion@cisco.com>
