summaryrefslogtreecommitdiffstats
path: root/src/vnet
AgeCommit message (Collapse)AuthorFilesLines
2019-04-24Rearrange prefetching in ipsec_output_inlineVratko Polak1-6/+6
Change-Id: I6151e57643ebed42f51b795980db2c52084295ab Signed-off-by: Vratko Polak <vrpolak@cisco.com>
2019-04-23API sw_interface_dump: Dump all if index is zeroVratko Polak2-5/+6
This is a temporary measure, to allow CSIT usage of VAT command sw_interface_dump without arguments. Change-Id: Ic40adfcc89d92179e213afc497e4e71bbc0dad83 Signed-off-by: Vratko Polak <vrpolak@cisco.com>
2019-04-23Fix a ipsec command line typoSimon Zhang1-1/+1
Change-Id: Ic75df36e06a77730ff8764f96d3cf53c4e59923b Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
2019-04-23vnet: clean up calc_checksums()Zhiyong Yang1-7/+8
Remove the duplicated code and unnecessary operations. Change-Id: I78005848d29d3156165627926a79015d590d61a6 Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
2019-04-23ipsec4-output: add pkt header and data prefetchingZhiyong Yang1-2/+9
The graph node running IPsec encap in tunnel mode can be saved from 65.8 to 57.3 clocks/pkt on Haswell platform. The graph node can be saved 10 clockes/pkt on DVN as well in the same case. Change-Id: I4804879c4d489465ee56a8f8317596b7e79b9331 Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
2019-04-22GRE: set gre_tunnel_type init value to zero in APIAlexander Chernavin1-1/+1
Change-Id: I9715b0578852a1ed59d78b7a9e28f32fc763ed3c Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
2019-04-19Doxygen cleanup.Dave Wallace2-2/+2
- Add subpages definitions in appropriate section (User or Dev docs) for doc files (*.rst, *.md) that being listed at the top level of the generated doc page. - Generate and add API list to RELEASE doc. - Fix list_api_changes script to use HEAD as the endtag so it doesn't need to be changed every release. Change-Id: Iace7b6433359c6b96869cb1db01facbbcb0ac1e6 Signed-off-by: Dave Wallace <dwallacelf@gmail.com> (cherry picked from commit 11ee93f6abfaddf5bbd56cf0858c0c6ea0384b65)
2019-04-19IPSEC: IPv6 ESP transport mode incorrect packet length and checksum (VPP-1654)Neale Ranns1-7/+14
Change-Id: Ia3474e5bfea5764eae9b2987bf78296535df6778 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-196rd: fix debug cli short-helpBenoît Ganne1-1/+2
Debug cli short help for 6rd tunnel creation was out-of-date. Change-Id: I06e4d28481470825bf225ba0fd371a3aebd889fa Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-04-19DOCS-ONLY: Fix broken doxygen tag in BVIDave Wallace1-1/+1
Change-Id: Ia42e7c93ebe51a36470f1358827451bcb98da433 Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2019-04-19IPSEC: ESP IPv6 transport mode payload length incorrect (VPP-1653)Neale Ranns1-1/+3
Change-Id: I8977100d7a22b50260858bd1ea9db419b53284ff Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-19FIB: recursion casues path reallocNeale Ranns1-1/+12
Change-Id: Ie9c2954eee90ca1a1fc1aa8280f93b2340b544c1 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-18svm: allow indirect fifo data chunksFlorin Coras1-1/+0
Fifos can use multiple memory chunks for simple read/write operations. Adding/removing chunks after assignment not yet supported. Change-Id: I2aceab6aea78059d74e0d3a9993c40d5196d077b Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-18GRE: API updateNeale Ranns4-88/+136
Change-Id: I5010cd34123c6498230dedac6ba8dd774a1085f9 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-18tls: allow engines to customize closeFlorin Coras2-18/+26
Change-Id: I11ac3e4f59206902e5dfc326f815c877c5dd6643 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-18IPSEC: tunnel rekey fix and test (VPP-1652)Neale Ranns1-21/+24
Change-Id: I1c2b3e40c689bedcdcea7887792b6b6b6aeb48d5 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-17session: fix segment manager init conditionFlorin Coras1-1/+1
Change-Id: I0ef3115bd29a11538090c582a4eacdbb7cd86d7a Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-17Make sw_interface_dump more compatible with 2.2.0Vratko Polak2-9/+6
+ As old PAPI clients are likely to put zero as the value for sw_if_index, the behavior should not perform index filtering at least when name filtering is enabled (valid). + interface.api version set to 2.3.0, as the new behavior is backward compatible (at least for PAPI with name filter enabled), but not forward compatible. + Minor whitespace cleanup. Change-Id: I315a0eae4004f9d9b6c5f9ecf0f179e669729118 Signed-off-by: Vratko Polak <vrpolak@cisco.com>
2019-04-17vcl/session: tx notifications for cut-thru sessionsFlorin Coras3-0/+32
Change-Id: I076c753e419bbb177d2d28609190715e9895b398 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-17api: Add to interface crud - read by sw_if_index.Paul Vinciguerra2-1/+25
Change-Id: I02c857da4cf6da5e0e55c1e48b63716af7ade0a9 Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
2019-04-17Revert "Revert "bond: problem switching from l2 to l3""Steven Luong1-0/+11
This reverts commit 5d0d5494db58422eb528c0f8b39a86ea966505e9. The csit crash was actually due to the test image missing the patch https://gerrit.fd.io/r/#/c/17731/ It was a mistake to revert the original patch https://gerrit.fd.io/r/#/c/15577/ Change-Id: I7fc563981aa13d308d55b25194fee21475ebc57d Signed-off-by: Steven Luong <sluong@cisco.com>
2019-04-17tap: clean-up when linux will delete the tap interfaceMohsin Kazmi2-0/+43
When container is deleted which has tap interface attached, Linux also delete the tap interface leaving the VPP side of tap. This patch does a clean up job to remove that VPP side of tap interface. To produce the behavior: In VPP: create tap On linux: sudo ip netns add ns1 sudo ip link set dev tap0 netns ns1 sudo ip netns del ns1 Change-Id: Iaed1700073a9dc64e626c1d0c449f466c143f3ae Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-04-17IPSEC: Pass the algorithm salt (used in GCM) over the APINeale Ranns6-10/+16
Change-Id: Ia8cea13f7b937294e6a080a55fb2ceff30063acf Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-17Duplicate pcap tx trace fixNeale Ranns1-3/+0
Change-Id: I0657cb44f58942ef281046dd3841bda669b10589 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-16ADJ: crash in format/show (VPP-1648)Neale Ranns1-1/+1
Change-Id: I26279c19b879e59c68fda31426fe42dae62a858d Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-16IPSEC: SA format; don't print keys when there's no algoNeale Ranns1-7/+8
Change-Id: I4d1d22cb24564896264e77c1810804ea3f54cb37 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-16svm_fifo rework to avoid contention on cursizeSirshak Das11-39/+33
Problems Addressed: - Contention of cursize by producer and consumer. - Reduce the no of modulo operations. Changes: - Synchronization between producer and consumer changed from cursize to head and tail indexes Implications: reduces the usable size of fifo by 1. - Using weaker memory ordering C++11 atomics to access head and tail based on producer and consumer role. - Head and tail indexes are unsigned 32 bit integers. Additions and subtraction on them are implicit 32 bit Modulo operation. - Adding weaker memory ordering variants of max_enq, max_deq, is_empty and is_full Using them appropriately in all places. Perfomance improvement (iperf3 via Hoststack): iperf3 Server: Marvell ThunderX2(AArch64) - iperf3 Client: Skylake(x86) ~6%(256 rxd/txd) - ~11%(2048 rxd/txd) Change-Id: I1d484e000e437430fdd5a819657d1c6b62443018 Signed-off-by: Sirshak Das <sirshak.das@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
2019-04-16QUIC: Initial multi stream supportAloys Augustin1-0/+1
To connect a stream, apps should call connect while passing the id of the QUIC connection in the new transport_opts field in session_endpoint_cfg_t. Apps are notified of new streams with their accept callback, which is called each time a peer opens a stream. Change-Id: I0f82ec344db58008d54641553eddec2973768435 Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
2019-04-16Allow sessions to be allocated in app's RX callbacksAloys Augustin1-3/+11
This enables applications to create sessions in their RX callbacks, which can invalidate the session pointer. This is required for the QUIC protocol implementation. Change-Id: I6072c1c368fd9d17a960ec086a788089dd6f54b4 Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
2019-04-16IPSEC: support GCM in ESPNeale Ranns13-13/+127
Change-Id: Id2ddb77b4ec3dd543d6e638bc882923f2bac011d Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-16IPSEC-GRE: return correct API codeNeale Ranns1-1/+1
Change-Id: I4f72989885ecfc6a0b27553194c6611fd71f7e5b Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-16vpp-1647: fix typo in pcap_trace_command_internalDave Barach1-1/+1
pcap rx trace on max 100 intfc tap0 then pcap rx trace status Displays "local0" instead of "tap0" due to a typo in pcap_trace_command_internal(...). Change-Id: Id2de6a24174aac24d9051b7404f01edc806a6573 Signed-off-by: Dave Barach <dave@barachs.net>
2019-04-15tap: fix the crashMohsin Kazmi1-0/+3
Crash will happen when someone will try to setup a tap interface in host namespace without providing the host side of tap interface custom name. This patch fixes the problem by using the default name in this case. Change-Id: Ic1eaea5abd01bc6c766d0e0fcacae29ab7a7ec45 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-04-15IPSEC: crypto overflowNeale Ranns1-1/+1
decrypting too many bytes. Change-Id: I4663e70271d9734eda7f9a127967b9224c0e5efc Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-14ip4: don't format upper layer if it's a fragmentfaicker.mo1-0/+3
Parsing ipv4 upper layer is not meaningful if it's a fragment packet except the first. Change-Id: I442fb7ec01244fde8c4f7656a8ba633d0aa0f97e Signed-off-by: Faicker Mo <faicker.mo@ucloud.cn>
2019-04-14crypto: add support for AES-CTR cipherfituldo1-1/+4
Change-Id: I7d84bab7768421ed37813702c0413e52167f41ab Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2019-04-12session: drop lock on segment allocation errorFlorin Coras1-12/+9
Change-Id: Ib346570daa3e40f4f53100a05e9355ce60d533a4 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-11IPSEC: ESP with ESN tests and fixesNeale Ranns2-4/+4
Change-Id: Ie42b26e6d5cdb7b23f370ea2933c65079e8d1089 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-11IPSEC-MB: Use random & non-repeating IV (VPP-1642)Neale Ranns3-7/+6
hard code IV and key lengths based on cipher. Init IV from random data, use AES instruction to rotate. Change-Id: I13a6507d12267b823c528660a903787baeba47a0 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-11Stats: Spelling mistake: s/tx-unicast-miss/tx-unicast/Ole Troan1-1/+1
Change-Id: Ie71235e5a0ff84222d8a35f5d4e99e9b20447d61 Signed-off-by: Ole Troan <ot@cisco.com>
2019-04-11crypto: add performace test to unittest pluginDamjan Marion2-0/+22
Change-Id: I49c710c5ace24a4c1f083120fd4c2972566a1695 Signed-off-by: Damjan Marion <damarion@cisco.com> Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2019-04-10udp: fix copyright typoFlorin Coras1-1/+1
Change-Id: I74a61e941176321da0767affff0fb023f4c54c35 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-10crypto: Intel IPSEC-MB engineNeale Ranns2-2/+13
A plugin to use Intel IPSec MB library as a VPP crypto engine This changes uses concepts from: https://gerrit.fd.io/r/#/c/17301/ hence that author's work is acknowledge below Change-Id: I2bf3beeb10f3c9706fa5efbdc9bc023e310f5a92 Signed-off-by: Neale Ranns <nranns@cisco.com> Signed-off-by: Klement Sekera <ksekera@cisco.com>
2019-04-10ethernet: fix packet tracingBenoît Ganne1-1/+1
Node tracing condition was wrongly reversed by commit "5ecd5a5d15 Move pcap rx/tx trace code out of the dpdk plugin". This prevented packet tracing in ethernet-input node and also impacted performance in the no tracing case. Change-Id: I345a11191d027c6c4ec474a2901995338050680a Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-04-10IPSEC: for each engine and algorithm testsNeale Ranns1-1/+1
refactor the IPSEC tests a bit so we can parameterise the setup. Change-Id: I777e5eb8f29ca1dce3dd273ebd05dae5846790af Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-10session: binary api app names as vectorsFlorin Coras1-2/+2
Change-Id: Iae358365de8ccbc0441b14f21ba6b365cbfec09a Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-10API: Fix shared memory only action handlers.Ole Troan6-195/+86
Some API action handlers called vl_msg_ai_send_shmem() directly. That breaks Unix domain socket API transport. A couple (bond / vhost) also tried to send a sw_interface_event directly, but did not send the message to all that had registred interest. That scheme never worked correctly. Refactored and improved the interface event code. Change-Id: Idb90edfd8703c6ae593b36b4eeb4d3ed7da5c808 Signed-off-by: Ole Troan <ot@cisco.com>
2019-04-10IPSEC: remove double byte swap of IP addressesNeale Ranns1-18/+10
Change-Id: I8c03c4aa90fb0056e11e0f234999c25d7839d759 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-04-10Make tcp/udp/icmp compute checksum safer for buffer-chain caseJohn Lo2-2/+2
Change-Id: I046e481a67fbeffdaa8504c8d77d232b986a61ee Signed-off-by: John Lo <loj@cisco.com>
2019-04-09session: fix session flagsFlorin Coras2-5/+2
Change-Id: I681169b82c661b7f0bf19f09d07d76ac1d3ed173 Signed-off-by: Aloys Augustin <aloaugus@cisco.com> Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com> Signed-off-by: Florin Coras <fcoras@cisco.com>