summaryrefslogtreecommitdiffstats
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2021-04-06session: basic support for interrupt modeFlorin Coras4-4/+176
Experimental support for session layer interrupt mode. When enabled (use-private-rx-mqs must be set) session queue node switches to interrupt state when lightly loaded, i.e., no events and less than 1 vector/dispatch. Because transport protocols require a periodic time update, when in interrupt state the session queue node workers register a timerfd with the unix-epoll-input node that when triggered signals, i.e., wakes up, the queue node. Under light load, the timer is set to trigger every 1ms whereas if no session is allocated, the worker moves to idle state and the timeout is set to 100ms. Type: feature Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I905b00777fbc025faf9c4074fce4c516cd139387
2021-04-06buffers: performance improvementRadu Nicolau1-7/+21
Initialize the local variables as to prevent first fall through one by one processing; skip prefetching rather than doing one by one when count below 12. Type: improvement Signed-off-by: Radu Nicolau <radu.nicolau@intel.com> Change-Id: If02935da081e24f7cfe408811ccf6922cf982092
2021-04-05tcp: time infra improvementsFlorin Coras12-47/+87
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I998c0686f9f7dc556dda8b28e23bbed127d0aafc
2021-04-05vhost: Crash upon disconnectSteven Luong1-1/+1
Patch 24601 introduced description into template. In vhost_user_process, template is reused inside the while loop to call clib_file_add. But description is allocated outside of the loop only once. As a result, the same storage for description is being referenced by all instances of call to clib_file_add. As long as we don't call clib_file_del, we may be good with multiple fds sharing the same storage for the description. When one of the fds disconnects and frees the description, the other fds is holding onto the free memory pointer. Bad news eventually happens when another fd disconnects and frees the description that was already free previously. The fix is to move the allocation of description inside the loop to avoid sharing. Type: fix Fixes: gerrit.fd.io/r/c/vpp/+/24601 Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: Ie670931acdc2c7b851982d98fd0d837284a19036
2021-04-05arping: coverity woe on COPY_PASTE_ERRORSteven Luong1-1/+1
coverity is trying to outsmart human by guessing on COPY_PASTE_ERROR. Have it your way. Type: fix Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: I0239c9745e21384aebbaf9e977962746b81ded58
2021-04-05cnat: explicit BIHASH_USE_HEAP settingDave Barach1-0/+2
The session db (bihash_40_56_t) was inheriting BIHASH_USE_HEAP=1 from vppinfra/bihash_24_8.h through a convoluted set of #includes. Set BIHASH_USE_HEAP in cnat_bihash.h, to avoid a surprise long after anyone remembers the story. Type: improvement Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: I8aa3be09b306fc99ba0b6baa5096a0a6829a8951
2021-04-05ip6-nd: Solicitation reply only if target is our link-localNeale Ranns1-4/+9
Type: fix The fib source IP6_ND is used for all link-local entries, hence solicitation responses were sent for a peer's address. Constrain the source check to also in clude the LOCAL flag, which indicates that the link-local address is ours. Signed-off-by: Neale Ranns <neale@graphiant.com> Change-Id: Iba7e66049e4d89ee3f36d77aeb09310b978d70de
2021-04-02svm: lock-free fifo chunk list push and popFlorin Coras2-43/+57
This avoids chunk allocation/collection deadlocks if either of the sides crashes. Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I98619e6e035fa8688889ca34db2143c8898732df
2021-04-02build: create _objs target for each libraryDamjan Marion1-3/+10
Type: improvement Change-Id: I260580cf2d32ff949f44d80c764937ee6102486c Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-02arping: add arping commandSteven Luong8-0/+1470
Add linux similar arping command to VPP. syntax: arping [gratuitous] <address> <interface> [repeat <count>] [interval <secs>] Type: feature Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: I9267c054235207b8fae8e3f159246777eb0340dd
2021-04-02ip-neighbor: suppress unsolicited clib_warning when sending garpSteven Luong2-5/+19
Replace clib_warning with vlog_log_debug when sending garp. There is nothing to alert/warn the operator about sending garp. vlib_log is more appropriate. Type: fix Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: Ib88a2b46d2ab8b3c35162538ead0dd0983fb7093
2021-04-01session: always clean rx mqs on app detachFlorin Coras1-4/+3
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I47b5b1ad693d131cfc332f6942c86967687a3f2b
2021-04-01avf: disable VLAN stripping on E810 using V2 VLAN APIsDamjan Marion4-14/+201
Type: fix Change-Id: I9bb19a5c9b5b48825f19a4ac124a3628ceaa081d Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-01nat: fix ordering of feature nodesKlement Sekera1-2/+4
Fix ordering of feature nodes so that ACL runs after NAT in in2out direction so it can properly track stuff like ports after NAT rewrite. Type: fix Change-Id: I2c689b64765628e9bc77108914f74c6c801d8ce2 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2021-04-01perfmon: combined set and start command.Ray Kinsella3-47/+36
Original set, start, stop, reset, show etc interface was somewhat cumbersome, we can improve slightly by combining set and start. Type: improvement Signed-off-by: Ray Kinsella <mdr@ashroe.eu> Change-Id: I7b865b2c29d2ab32adbd24d7f8a580da6990bb76
2021-04-01ip: extend show cmd of ip reassembly configurationAnton Nikolaev4-0/+16
Type: feature Add output of field maximum reassembly length with others ip reassembly configuration fields. This field can be configured via vpp API, but there was no output of this field in vpp cli. Output added for ipv4/ipv6 and full/shallow virtual nodes. Signed-off-by: Anton Nikolaev <anikolaev@netgate.com> Change-Id: Ibbd61aab26189d4a93d32e047d8f4e589fe9f8a5
2021-04-01lacp: correctly set is_mp_safe for VL_API_SW_INTERFACE_LACP_DUMPSteven Luong1-2/+1
Need to add msg_id_base to the index Type: fix Signed-off-by: Steven Luong <sluong@cisco.com> Change-Id: Id7ebf168502e8fd811d8f2f0c5538e69e23242fb
2021-04-01perfmon: % power level per nodeRay Kinsella2-0/+58
Show % time spent per graph node in power level 0, 1 and 2. Type: improvement Signed-off-by: Ray Kinsella <mdr@ashroe.eu> Change-Id: I678ee812fa993af39568e9f9dfbf2396fc13ad42
2021-04-01fib: Move the adjacency midchain nodes into a separate fileNeale Ranns5-233/+251
Type: refactor It's only the nodes that require compiling for each arch, Signed-off-by: Neale Ranns <neale@graphiant.com> Change-Id: Ibb2bfc756001735cc10ef0fd8db7e8b8358cf8e2
2021-03-31nat: enable multiworker testsKlement Sekera1-2/+0
Type: fix Signed-off-by: Klement Sekera <ksekera@cisco.com> Change-Id: I4a5d52d54aa2748008e896888bfeabb18a1088bb
2021-03-31dpdk: fix packet offset for GCM crypto opsRobert Shearman1-0/+1
The crypto op data offset passed into DPDK is relative to the mbuf buffer address plus the mbuf data offset, therefore the mbuf data offset needs to be set rather than left at whatever previous value it was at, which is likely to be incorrect and result in the wrong portion of the packet being encrypted/decrypted for GCM. The fe->crypto_start_offset field is relative to the start of the vlib buffer (as opposed to the current data pointer), so set the mbuf data_off field to VLIB_BUFFER_PRE_DATA_SIZE when performing a GCM crypto op enqueue to match the crypto_start_offset semantics. This then matches the behaviour in the non-GCM case. Type: fix Change-Id: I0ac2a44139387158765a3e04cfcaa5ee6f11d395 Signed-off-by: Robert Shearman <robertshearman@gmail.com>
2021-03-31nat: fix test for multi-worker scenarioKlement Sekera1-3/+16
Type: fix Change-Id: I170f9d8c685f59b7ccf6f9ee7fdc1dcdf2a54fe1 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2021-03-31avf: improve capabilities handlingDamjan Marion5-39/+56
Type: improvement Change-Id: If35b40168414807e928cb3026633ae9d2fa703ed Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-31perfmon: add branch mispredictionsRay Kinsella3-0/+76
Add branches, branches taken (a meteric for branchy code), and branch misses. Type: improvement Signed-off-by: Ray Kinsella <mdr@ashroe.eu> Change-Id: If92d4aaf9d0a6e3b99b8c19e6311cc08ca470590
2021-03-31srv6-ad-flow: SRv6 Per-Flow Dynamic ProxyFrancois Clad7-0/+2331
Type: feature Change-Id: Ibfc412200ef5932fa40423075967df6d9156f3ed Signed-off-by: Francois Clad <fclad@cisco.com>
2021-03-31nat: correct thread index usageKlement Sekera3-12/+15
Use actual thread index from hash table instead of doing calculation. Calculated index might be different and could lead to crash. Type: fix Change-Id: Iede9a67fbe541cb44eccd0d6ddf4e2c51eb35423 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2021-03-31nat: avoid creating negative zero checksumKlement Sekera1-1/+3
Type: fix Signed-off-by: Klement Sekera <ksekera@cisco.com> Change-Id: Ibbb72e77c751b7e775620cfb160b52ceff2da86b
2021-03-31nat: NAT44-ED unused value removalFilip Varga1-4/+1
Type: fix Coverity scan issue fix. Change-Id: I9036fade15fd381da9c1e6b575f250fdcb21f876 Signed-off-by: Filip Varga <fivarga@cisco.com>
2021-03-31nat: NAT66 plugin enable&disable calls updateFilip Varga7-119/+263
Type: improvement Adding support for enable&disable calls (dynamic plugin configuration). API (nat66_plugin_enable_disable) and CLI (nat66 plugin enable/nat66 plugin disable) with support for outside_vrf id configuration. Change-Id: I5637ff1621d6662adc3b7c6f7f8176d84a4b492b Signed-off-by: Filip Varga <fivarga@cisco.com>
2021-03-30session: interrupt mode for session node on main threadFlorin Coras1-15/+20
Signal, and implicitly run, main thread session queue node via interrupts instead of the queue process node. Process node still needed for timers. Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I5efd393effc8dad65e4125a3d0ad1be6b7484cb7
2021-03-30interface: fix coverityDamjan Marion1-2/+2
Type: fix Change-Id: Ia7652fe8e8c20f4f7a019ab7ca0649c2180c7e90 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-30stats: python: handle when pattern is not list in lsOle Troan1-0/+9
The reimplementation of python stats module mishandled the case where pattern to ls was not a list. Type: fix Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: I9ba189423a76f2fd4298c4c4b19a0875f705d719
2021-03-30svm: cleanup mq private data on fifo segment cleanupFlorin Coras3-2/+16
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I65373f2cd5a33381da1e51ed7ab7a8b1358cef29
2021-03-30nat: fix HA multi-worker issuesKlement Sekera6-154/+176
Use correct vlib_main() in various code parts. Fix tests. Type: fix Signed-off-by: Klement Sekera <ksekera@cisco.com> Change-Id: Ia379f3b686599532dedaafad2278c4097a3f03f3
2021-03-30avf: fix coverity issuesChenmin Sun1-4/+4
CID #220403 ~ 220406 Type: fix Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: Ibb65f9f2b17b9a9d45c077933b033ca3a65ae42f
2021-03-30map: tune dump/details test for socket transportOle Troan1-1/+1
Type: fix Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: Id3ad783a7b0321a8fa37b4e606c0c4551ca4edd4
2021-03-29svm session vcl: per app rx message queuesFlorin Coras11-104/+518
Add option to use per app private segments for app to vpp message queues, as opposed to exposing internal message queues segment. When so configured, internal message queues are still polled by the session queue node but external app message queues are handled by a new input node (appsl-rx-mqs-input) that runs in interrupt state. Signaling of the node, when mqs receive new messages, is done through eventfds epolled by worker epoll input nodes. Type: feature Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Iffe8ce5a9944a56a14e6d0f492a850cb9e392d16
2021-03-29tls: fix app_worker_connect_notify param erroryedg1-1/+1
Type: fix Signed-off-by: yedongggang <yedg@wangsu.com> Change-Id: I5ec690e71fa9119a7bf1cfcb6e36263aa732eff0
2021-03-29tests: move vxlan tests to src/vnet/vxlan/testsDave Wallace6-0/+1464
- Refactor make test code to be co-located with the vpp feature source code Type: test Signed-off-by: Dave Wallace <dwallacelf@gmail.com> Change-Id: I66379dfe671628e39dfc9685c4fd70fa0e566f6f
2021-03-28vcl: fix only first epoll listen session be registerednandfan1-14/+2
If configure listen port more than 1 with epoll in nginx, only first listen port can be registered successed in nginx worker process. Type: fix Signed-off-by: nandfan <fanyufei521@outlook.com> Change-Id: Iec2e058ee972a2cd5b2d1d7d20d373dc17084ce1
2021-03-26vlib: convert foreach_vlib_main macro to be more gdb and clang-format friendlyDamjan Marion12-350/+314
Type: improvement Change-Id: I1152e58d7bfcb3c4347147f87a834d45ad51cdfe Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-26vlib: split vlib_main_t into global and per-threadDamjan Marion24-456/+469
Type: refactor Change-Id: I8b273bc3bf16aa360f031f1b2692f766e5fc4613 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-26vlib: introduce vlib_get_elog_main()Damjan Marion15-85/+88
Type: improvement Change-Id: I73383eb15186021cd6527d112da8443a0082f129 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-26vlib: introduce vlib_get_main_by_index(), vlib_get_n_threads()Damjan Marion43-137/+166
Type: improvement Change-Id: If3da7d4338470912f37ff1794620418d928fb77f Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-26tests: move ping tests to src/plugin/ping/testDave Wallace1-0/+176
- Refactor make test code to be co-located with the vpp plugin source code Type: test Signed-off-by: Dave Wallace <dwallacelf@gmail.com> Change-Id: I09d55301665449b3f183191537749435f806fb15
2021-03-26avf: flow support enhancementChenmin Sun1-49/+219
This patch enables ip4, ip6, ip6-n-tuple, l2tpv3 esp ah flows in avf plugin Type: improvement Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I92c6d4217a6cce4085ee83e8beb4b037c8d7da5f
2021-03-26avf: handle flow id in data pathChenmin Sun3-27/+122
Type: improvement Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I24fa52fee88dabad8f0b84446f796ee055819617
2021-03-26avf: add avf flow frameworkChenmin Sun7-34/+2430
This patch adds the flow support to VPP avf_plugin. Currently only IP4-N-TUPLE and GTPU flow are supported as a reference. More flows will be added later. Type: feature Signed-off-by: Chenmin Sun <chenmin.sun@intel.com> Change-Id: I563dbfacecd3eba31e8c3c7769142ea6adba54b6
2021-03-25svm: add producer wait functionFlorin Coras3-14/+47
Type: refactor Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I9488ad7e045c908b60b5821d9c48583f6d513c2f
2021-03-25acl: fix the integer overflow bug in API message length validation logicAndrew Yourtchenko1-4/+4
Sending the bogus acl_add_replace message with count=~0 will result in an overflow of "expected_len" field which is a u32, thus the message will pass the validation when it should not. Solution - make the expected_len a u64 to avoid overflow. The bug was found while experimenting with libfuzzer as part of https://gerrit.fd.io/r/c/vpp/+/31763 Type: fix Change-Id: I4a866d48f2418148236f1b1d77c487b869c7c43d Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>