aboutsummaryrefslogtreecommitdiffstats
path: root/src/vnet
AgeCommit message (Collapse)AuthorFilesLines
2017-09-07LISP: Add APIs for enable/disable xTR/P-ITR/P-ETR modesFilip Tehlar1-0/+60
Change-Id: Ieeb3b7eaabb568180320fe54d3eae2d26f3e4704 Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2017-09-07Devices: Set interface rx-mode may cause SIGSEGV with nonexistent queueSteven3-3/+17
When I type in set interface rx-mode with a nonexistent queue, I got a crash with the following traceback. It looks like the vm is NULL when vlib_node_get_runtime is called. DBGvpp# sh int rx Thread 0 (vpp_main): node dpdk-input: TenGigabitEthernet5/0/0 queue 0 (polling) TenGigabitEthernet5/0/1 queue 0 (polling) TenGigabitEthernet7/0/0 queue 0 (polling) TenGigabitEthernet7/0/1 queue 0 (polling) node vhost-user-input: VirtualEthernet0/0/2 queue 0 (adaptive) DBGvpp# set interface rx-mode VirtualEthernet0/0/2 queue 1 polling Thread 1 "vpp_main" received signal SIGSEGV, Segmentation fault. 0x00007ffff6d4e0bc in vlib_node_get_runtime (vm=0x0, node_index=125) at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:92 92 vlib_node_t *n = vec_elt (nm->nodes, node_index); (gdb) where at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:92 at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:112 vnm=0x6f0fa0 <vnet_main>, hw_if_index=7, queue_id=1, mode=0x7fffb62099e8) at /home/sluong/vpp/build-data/../src/vnet/devices/devices.c:307 hw_if_index=7, queue_id=1, mode=VNET_HW_INTERFACE_RX_MODE_POLLING) at /home/sluong/vpp/build-data/../src/vnet/interface_cli.c:1192 vm=0x7ffff7b9d440 <vlib_global_main>, input=0x7fffb6209ef0, cmd=0x7fffb61d5d14) at /home/sluong/vpp/build-data/../src/vnet/interface_cli.c:1288 vm=0x7ffff7b9d440 <vlib_global_main>, cm=0x7ffff7b9d630 <vlib_global_main+496>, input=0x7fffb6209ef0, parent_command_index=18) at /home/sluong/vpp/build-data/../src/vlib/cli.c:588 vm=0x7ffff7b9d440 <vlib_global_main>, cm=0x7ffff7b9d630 <vlib_global_main+496>, input=0x7fffb6209ef0, parent_command_index=12) The fix is to add a check for vec_len(hw->input_node_thread_index_by_queue) and vec_len (hw->rx_mode_by_queue) to reject the command if the queue_id is out of bound. While at it, I notice inputting queue_id=-1 is being interpreted as all queues. An easy fix is to not overload the queue_id variable with -1 to mean something else. Change-Id: Id70ec3e7d06ccc67635e6d28ef53420bdac4a988 Signed-off-by: Steven <sluong@cisco.com>
2017-09-07Fix session connect_* api message handling.Dave Wallace3-46/+47
Change-Id: If7fd125989c90240de12953658d10007b9eb4f07 Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2017-09-07Support for bridge domain free text tagJerome Tollet4-2/+66
Change-Id: I9a75fdafd0c1d87b6f071fda5b77ff5f6b79deb7 Signed-off-by: Jerome Tollet <jtollet@cisco.com>
2017-09-06Improve "show segment-manager segments"Dave Barach4-26/+74
Clean up private-segment fifo preallocation Change-Id: I53c630ed81d48f2832a204992d73635536926983 Signed-off-by: Dave Barach <dave@barachs.net>
2017-09-06Fixes for issues raised by Coverity (VPP-972)Chris Luke1-3/+0
Change-Id: I4b1f27b95d67d48b7a13750ff8754c344ed7afa7 Signed-off-by: Chris Luke <chrisy@flirble.org>
2017-09-05PUNT socket: External control plane processes connected via UNIX domain sockets.Ole Troan3-11/+555
An external (to VPP) process can register (over the VPP binary API) to receive control plane packets over a UNIX domain socket. The packets are prepended with a packet descriptor containing meta-data (if_index of interface, etc). Currently only UDP is supported. The socket supports sending of packets/frames as well. The sent packet is prepended with a descriptor, telling VPP to route the packet (via ip4-lookup, ip6-lookup) or as an pre-formed Ethernet frame that is sent directly to interface-output. The intended use case for this is for an external DHCP client or a RIP implementation. New configuration option: punt { socket <socket path> } To register use the punt_socket API message. TODO: - Add support for pre-routing. I.e send L3 packet to given TX interface, but do ARP/ND (ip[46]-rewrite) - Add test scripts - Support for abstract names (starting with \0) - Add rate limiting (COP) - Support for other protocols, e.g. IPv6 ND Change-Id: I4a0afc8020deebb3d9d74686dde694ee5bcb8d0f Signed-off-by: Ole Troan <ot@cisco.com>
2017-09-04LISP: re-fetch mapping before it expiresFilip Tehlar10-138/+330
Change-Id: I0581a1bddad55d8d573c546ec84b0b2760abab3d Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2017-09-03Improve L2FIB PDR/NDR performance (VPP-963)John Lo4-18/+30
1. Limit MAC entry update per l2-learn call to reduce update burst when wall clock advance to the the next minute so all MAC time stamps are behind current time needing update. 2. Optimize l2-learn node fast path code sequence. 3. Invalidate cache_key when update MAC entry. 4. Change L2 learn hit counter to L2 learn hit-update counter. 5. Increase L2FIB table memory size to 512MB to fit 4M entries 6. Set MAC learn limit at 4M entries Change-Id: I3075ee8fb59645a56850126bac2e3e6d341cef4d Signed-off-by: John Lo <loj@cisco.com>
2017-09-01Add fixed-size, preallocated pool supportDave Barach5-48/+73
Simply call pool_init_fixed(...) before using the pool. Note that fixed, preallocated pools live in individually-mmap'ed address segments, except for the free element bitmap. A large fixed pool can exceed 4gb. Fix tcp buffer allocator leak, remove broken assert Change-Id: I4421082e12a77c41c6e20f7747f3150dcd01fc26 Signed-off-by: Dave Barach <dave@barachs.net>
2017-08-31vhost: Disallow interrupt mode config if driver opts out interrupt supportSteven2-3/+9
According to the spec, supporting interrupt mode from the driver is optional, not a must. When interrupt mode is configured on the interface, we should check to make sure that the driver didn't opt out for the kickfd support and reject the configuration if it did. Change-Id: I7d3dbaddde65458e1a6a802754a3768ae8685a0e Signed-off-by: Steven <sluong@cisco.com>
2017-08-31Fix vent_data memory leak in function run_ping_ip46_address (VPP-966)dongjuan1-0/+1
Change-Id: I05d0c8310aa86480a3d283f45e115284c4b4ebd5 Signed-off-by: dongjuan <dong.juan1@zte.com.cn>
2017-08-30tcp: re-enable persist timer if no data available to sendFlorin Coras3-5/+26
Additionally, flush rx fifos for closed sessions. Change-Id: If2cc563fbda0451e7572650e98b15f0a694a0ff9 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-30vhost: Cache qsz_mask instead of qsz in vhost_user_vring_tSteven2-32/+28
In the data path, we grab qsz from vhost_user_vring_t to compute qsz_mask and store it in a stack variable to use on many occasions. We never use qsz for any meaningful purpose. It is more useful to cache qsz_mask in vhost_user_vring_t to avoid the needless computation in the data path. Change-Id: Idf4d94a9754d5c75c899f1f4f59602275b9904a6 Signed-off-by: Steven <sluong@cisco.com>
2017-08-29session: segment manager improvementsFlorin Coras8-164/+346
- cleanup connects segment manager even if first - fix segment manager allocation for listen sessions - improve handling of process private segments (mheaps/main heap) - added segment manager cli Change-Id: Ic2ca97c3622ab2286d5fb5772aeb57680e64f769 Signed-off-by: Florin Coras <fcoras@cisco.com> Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2017-08-29Fix for bootstrap crash when VPP compiled with gcc-7Marco Varlese1-1/+1
See issue VPP-964 Change-Id: I2aed5abdc02e47e7bb5b0ce78cb6589a1ce7302f Signed-off-by: Marco Varlese <marco.varlese@suse.com>
2017-08-28vhost: Remove operation mode in the APISteven1-6/+0
create/delete/modify vhost_user APIs no longer support the operation mode (polling/interrupt/adaptive). They are now done via the generic interface. Change-Id: I9e9bd503f9b56c953ecd2b271b3e2007da20c72a Signed-off-by: Steven <sluong@cisco.com>
2017-08-25TCP horizontal scalingDave Barach3-3/+30
- Remove frame handoff support machinery. We haven't used it in a long time. - Configuration support for the local endpoints bihash table - Drop lookup failure packets in tcp46_syn_sent Change-Id: Icd51e6785f74661c741e76fac23d21c4cc998d17 Signed-off-by: Dave Barach <dave@barachs.net>
2017-08-25tcp: retransmit and multi-buffer segment fixes and improvementsFlorin Coras8-175/+312
- set session state as closed on session manager delete - enable retransmit as opposed to persist timer after persist timer completes - properly discard buffer chain bytes when new data overlaps ooo segments - don't use rxt bytes in snd space estimate used on tx path Change-Id: Id9cab686e532e5fe70c775d5440260e8eb890a9f Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-25Fix crash when TCP client using VPP host stack connects firsttjanciga1-0/+6
VPP is crashing without this fix if there is not any server using VPP host stack and client attemts to connect. Change-Id: I672977f78180bbc5ba9259116b2e165363463ccd Signed-off-by: tjanciga <tomas.janciga@pantheon.tech>
2017-08-25vl_api_sw_interface_set_mtu_t_handler: fix assert in vnet_get_hw_interfaceAndrew Yourtchenko1-3/+10
The handler was calling the routines with sw_if_index instead of hw_if_index, fix that by an extra call to vnet_get_sw_interface, and check that the interface type is VNET_SW_INTERFACE_TYPE_HARDWARE before proceeding. Change-Id: I4a6f65f44e250ecdb2b72d2693c9d7db5a52b966 Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com> (cherry picked from commit 778df28c2a1520f58c293700c578bdd73e61cecf)
2017-08-23Remove unused retval from gpe_native_fwd_rpath type definitionMarek Gradzki1-2/+0
Change-Id: I89a10204454aeea07412289b153ba50d6a15498c Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2017-08-23gpe: add l2 lb countersFlorin Coras1-1/+5
Change-Id: I81b3fc6266ad02239b6f1a216adc283efdcf0dce Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-22Increase default MAC learn limit and check it in learn-update pathJohn Lo5-4/+8
1. Increase default MAC learn limit from 1M to 8M entries. 2. Check MAC learn limit in MAC learning update path. 3. Allow disable of want_l2_macs_events to set MAC learn limit 4. Other minor cleanups Change-Id: I62438440937b5fa455e16f4a2e4d910277753395 Signed-off-by: John Lo <loj@cisco.com>
2017-08-22policer: remove unused codeDamjan Marion1-4/+0
This is causing compiler error with gcc 7. Change-Id: Ia65ee9d80ab263c79e82ad805e5cc5a12af91f85 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-08-22SRv6 testsKris Michielsen3-32/+201
Change-Id: Ib1d2fc5a83d9d007a0468591a73881675f1bec9b Signed-off-by: Kris Michielsen <kmichiel@cisco.com>
2017-08-22ping: remove unnecesary x86 intrinsics #includeDamjan Marion1-1/+0
This include file is present only on x86 so it breaks non-x86 builds. Change-Id: I6406b0b36f207b684535961545e95dbba707d267 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-08-21PPPoE usses a midchain adjacency stack on an interface-tx DPONeale Ranns10-555/+775
1) introduce an interface-tx DPO. This is a simple wrapper around a sw_if_index. enhance DPO stacking functions to allow per-instance next-nodes and hence allow children to stack onto the interface per-instance tx node and not on 'interface-output'. 2) update PPPoE code to use ta midchain stack on a interface-tx DPO of the encap-interface. This remove the need for pppoe_encap node (which is replaced by the adj-midchain-tx) and interface-output node is no longer used (see above). Since PPPoE encap node is no longer needed, the PPPoE seesion does not need to be retrieved in the data-path, hence the cahce misses are removed. Change-Id: Id8b40f53daa14889a9c51d802e14fed7fba4399a Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-19Fixed ip add bug for ip6 with review commentspragash1-4/+4
Change-Id: Ie1fba1a32020775444c7b1ed800e8559b54691ff Signed-off-by: pragash <pxv3620@rit.edu>
2017-08-18session: fix multi-buffer segmentsFlorin Coras3-34/+40
Change-Id: I0e22c85ea570b934b9c78dc5e86d86d690bdae5e Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-18gpe: fix sub-interface hash lookupFlorin Coras2-10/+10
Change-Id: Ice6b3818ee24c7c248bf61e4d6c1ef2a85cb8fb1 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-18feature: exit on misconfigured features (VPP-915)Florin Coras1-3/+7
Change-Id: Ifa11f356d727b69cb121a2bf4cfb9d87883f545d Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-18Use correct msg ID in the sw-interface-event from TAP and VHOSTNeale Ranns2-2/+2
Change-Id: I0124fa264f7f390fc7cd9722da59be03116831c5 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-17TCP source address automationDave Barach5-13/+387
- v6 support - Non-default VRF ID collection - Break up ip source address list into CLI + API-friendly functions - Automate proxy arp / proxy nd configuration - Automate local adjacency insertion - Binary API support Change-Id: Iede31184f65cc1ec8c414447d2d60a1334e3fe15 Signed-off-by: Dave Barach <dave@barachs.net>
2017-08-16ip4: cleanup ip localFlorin Coras1-153/+85
Change-Id: I4f751b0181f25ec6c4abd3a9c7b0386d80743b3a Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-16tcp: fix v6 sessionsroot7-81/+90
Change-Id: Ia6dd5e948b17b2f3866fe70838eabb09e35415e1 Signed-off-by: Dave Barach <dbarach@cisco.com> Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-16No context in SW interface eventNeale Ranns2-3/+8
At this pointin the game, events do not have an associated context. they have the client_index and pid instead. Change-Id: I5052cab78f710bba630a61d2390acbaa1b3813ab Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-16tcp: improve builtin http serverFlorin Coras5-197/+108
Additionally: - remove opaques from stream_session_t - ensure first segment manager is only used once per app. Change-Id: I143d1fdb8effc88815ef969b78122ba3ac29e06e Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-16Add missing counters for P2P ethernet interfacesPavel Kotucek2-0/+40
Change-Id: I9d172d70f645fea5005abced73ca2700e640263a Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
2017-08-16NAT64: Fallback to 3-tuple key for non TCP/UDP sessions (VPP-884)Matus Fabian2-0/+123
Change-Id: I4cafc8291725feb499355092bd429433e649b5b2 Signed-off-by: Matus Fabian <matfabia@cisco.com>
2017-08-15Support proxy ARP on mirrored TAP interfacesNeale Ranns1-4/+16
When VPP has an interface whose address is also applied to a TAP interface on the host, then VPP's TAP interface will be unnumbered to the 'real' interface and do proxy ARP from the host. the curious aspect of this setup is that ARP requests from the host will come from the VPP's own address. Change-Id: Ia238790e1034ba3cd3facdab29387b65a31525f2 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-15IPv6 local-address for P2P ethernet interfacesPavel Kotucek1-1/+2
Change-Id: I87d3cb5754843f38593e92f60a39af2e5241d1cf Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
2017-08-15VPP-939: Update CLI Helptext for rx-placement commandsBilly McFall2-94/+293
The CLI commands were change from 'set/show interface placement' to 'set/show interface rx-placement', but the associated help text was not updated. On the 'set interface rx-placement', some of the parameters were renamed, which was confusing when looking at the documentation for help. While in the file interface_cli.c, cleaned up several other CLI commands. There are still other commands in the file that need addressing. Change-Id: Ifdc357c73ad9f3362133c495217b5a6b3a411eab Signed-off-by: Billy McFall <bmcfall@redhat.com>
2017-08-15tcp: state machine improvementsFlorin Coras7-23/+40
- Add SYN_RCVD timeout - Fix FIN_WAIT_1 to CLOSING transition Change-Id: I42ca7fc087f6fdfae15bd7a6175dd3226ed341c7 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-15coverity: remove dependency on uint32_t in fib_test.cMarek Gradzki1-1/+1
Fixes make build-coverity Change-Id: I5cf6992224f18b8e8770d365c42b3e759dc93f73 Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2017-08-14TCP: Update time_now once per burstFlorin Coras4-3/+15
Change-Id: I58089d7a9867ede9d8a36b2aea62edef04cb5b81 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-14FIB table add/delete API onlyNeale Ranns4-0/+59
commit only the addition of the .api definition and their invocation through VAT so CSIT can use it Change-Id: Id510f14b1ce007fe5e92120507ea34100652fc64 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-14TCP: update builtin server/client cli helpFlorin Coras2-5/+19
Change-Id: I224ef3ce1c4ac2ade5ba733a42c78b7ba5c0040e Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-08-13default update adjacency function deos not return multicast adjacencyNeale Ranns1-6/+40
by not returning a multicast adjacency type when requested, but instead returning a nbr type, the mcast adj was never correctly deleted. hence when reused the adjacency object was not realocated from the pool and when it was freed a second time a crash occured. Change-Id: Ia74ae3e889db0dfba8ec3c6a0cccfef215587ff6 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-08-12LISP: fix fid nsh address formattingFlorin Coras1-1/+8
Change-Id: I912fa53c02c720901c9fb253550790829107de39 Signed-off-by: Florin Coras <fcoras@cisco.com>