summaryrefslogtreecommitdiffstats
path: root/src/vnet
AgeCommit message (Collapse)AuthorFilesLines
2019-02-19VPP-1568:when entry src cover change, recursive-loop fib still can not work.mu.duojiao1-0/+9
Change-Id: I90762b59f94175f278380c95776471a30bc94d34 Signed-off-by: mu.duojiao <mu.duojiao@zte.com.cn>
2019-02-19tap gso: experimental supportAndrew Yourtchenko14-59/+786
This commit adds a "gso" parameter to existing "create tap..." CLI, and a "no-gso" parameter for the compatibility with the future, when/if defaults change. It makes use of the lowest bit of the "tap_flags" field in the API call in order to allow creation of GSO interfaces via API as well. It does the necessary syscalls to enable the GSO and checksum offload support on the kernel side and sets two flags on the interface: virtio-specific virtio_if_t.gso_enabled, and vnet_hw_interface_t.flags & VNET_HW_INTERFACE_FLAG_SUPPORTS_GSO. The first one, if enabled, triggers the marking of the GSO-encapsulated packets on ingress with VNET_BUFFER_F_GSO flag, and setting vnet_buffer2(b)->gso_size to the desired L4 payload size. VNET_HW_INTERFACE_FLAG_SUPPORTS_GSO determines the egress packet processing in interface-output for such packets: When the flag is set, they are sent out almost as usual (just taking care to set the vnet header for virtio). When the flag is not enabled (the case for most interfaces), the egress path performs the re-segmentation such that the L4 payload of the transmitted packets equals gso_size. The operations in the datapath are enabled only when there is at least one GSO-compatible interface in the system - this is done by tracking the count in interface_main.gso_interface_count. This way the impact of conditional checks for the setups that do not use GSO is minimized. "show tap" CLI shows the state of the GSO flag on the interface, and the total count of GSO-enabled interfaces (which is used to enable the GSO-related processing in the packet path). This commit lacks IPv6 extension header traversal support of any kind - the L4 payload is assumed to follow the IPv6 header. Also it performs the offloads only for TCP (TSO - TCP segmentation offload). The UDP fragmentation offload (UFO) is not part of it. For debug purposes it also adds the debug CLI: "set tap gso {<interface> | sw_if_index <sw_idx>} <enable|disable>" Change-Id: Ifd562db89adcc2208094b3d1032cee8c307aaef9 Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2019-02-19VPP-1573 fix crash in ip6 reassemblyKlement Sekera1-1/+1
Change-Id: I3a3076c7d87446b5ec2a02e70d3b6d05f1875875 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2019-02-19ip6-local: fix uninitialized variable errorDamjan Marion1-1/+1
Change-Id: I245a8cc8f237242efadcf10d47b76222a6497e89 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-18tls: fix openssl/mbedtls use of app_wrk indexFlorin Coras2-15/+15
Change-Id: I7ccc948357d815a1bd4279a7079cf4db2949183c Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-18session: move fifo allocation logic to app workerFlorin Coras8-175/+158
Change-Id: I1662ec4b5503cb64a8a86a2441842311d959b3a6 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-18tcp: harden for high scale scenariosFlorin Coras9-35/+82
- Better handle buffer starvation scenarios - Handle case when both peers enter recovery due to packet loss. - Fix passive open establish cleanup Change-Id: I2f28baa2ff0383bb8f5f6d2452b49aa38ce69bce Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-18DHCP: trace flags copied from old to new not vice-versaNeale Ranns2-2/+2
Change-Id: I6c13af7e97753da013cdd5b212302d008cbd616d Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-18Explicit dual-loop in ip6-localBenoît Ganne1-72/+134
Makes ip6-local node dual-loop explicit. This is only a style change. Change-Id: Ic8e7cecb3f51e98b8a069b501f5c338156934a6d Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-02-18IPSEC: move SA counters into the stats segmentNeale Ranns11-18/+66
1) stats are accessed via the stat segment which is more condusive to monitoring 2) stats are accurate in the presence of multiple threads. There's no guarantee that an SA is access from only one worker. Change-Id: Id5e217ea253ddfc9480aaedb0d008dea031b1148 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-18tcp: fix handling of retransmitted synsFlorin Coras2-16/+26
Change-Id: I88fdeb0ca56d3a5d9c0f36def13c6de8142f59f0 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-15Optimize ip6-localBenoît Ganne1-246/+187
Optimize IPv6 ip6-local node by rewriting the dual/single loop with prefetch and simpler unrolling. My local, unrepresentative tests for GRE4 termination over IPv6 show a performance improvement of ~40% for ip6-local node alone and ~5% globally. Change-Id: I11e1e86d3838dd3c081aa6be5e25dae16ed6e2d8 Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-02-15Fix connexion segfault in session api affecting tcp_echoNathan Skrzypczak1-1/+2
Change-Id: If6d45f7f6e6dab70eecd6ec7241126968a34ad6b Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
2019-02-15FIB: pass a copy the walk contextNeale Ranns1-1/+3
since it can realloc when new ctx are added. If not we can get some nasty memory corruption. Change-Id: I617709c3013acbcb8aee07dc147894f0de896555 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-14ipsec: memory leak fixupKingwel Xie1-1/+4
Change-Id: Ib2ad196bec1005d6678589d5b5c199b8a541c720 Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2019-02-14Add -fno-common compile optionBenoît Ganne21-16/+28
-fno-common makes sure we do not have multiple declarations of the same global symbol across compilation units. It helps debug nasty linkage bugs by guaranteeing that all reference to a global symbol use the same underlying object. It also helps avoiding benign mistakes such as declaring enum as global objects instead of types in headers (hence the minor fixes scattered across the source). Change-Id: I55c16406dc54ff8a6860238b90ca990fa6b179f1 Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-02-13ip6: convert code to new multiarchDamjan Marion3-97/+87
Change-Id: Idd09b5d0597336e4f2028113cae76c94fd1c5427 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-13IPSEC: restack SAs on backend changeNeale Ranns3-13/+42
Change-Id: I5852ca02d684fa9d59e1690efcaca06371c5faff Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-12ipsec: cli bug fixKingwel Xie1-6/+23
1. unformat_ip46_address must have ip-type specified 2. cannot unformat ip46_address_t with unformat_ip4_address Change-Id: I5f1eecfe71a808302428afb39f910ebf0c7cef71 Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2019-02-11IPSEC-tun: The tunnel DB uses the network order SPI to save the byte swap in ↵Neale Ranns2-8/+11
the DP Change-Id: I78a1c39682d5afd356a3cfe70097fc682e8cb938 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-11session: add config for evt qs segment sizeFlorin Coras2-3/+6
Change-Id: I286d0d53a9ef2d6e28cd301eae25fc119d491f22 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-11session: cleanup application interfaceFlorin Coras15-653/+561
Change-Id: I89d240753b3f3c5e984aa303a7c8fa35fa59bf7f Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-11ipsec: multi-arch, next-node-index cleanupKingwel Xie5-27/+55
1. specify ipsec_xxx_node.c in MULTIARCH_SOURCES 2. cleanup foreach_ipsec_output_next & foreach_ipsec_input_next, as next-nodes are actually added by ipsec_register_xx_backend dynamically thus, ipsec4-input-feature will point to ah4/esp4-encrypt, instead of pointing to ah6/esp6-encrypt 3. remove an unused count and add counter IPSEC_INPUT_ERROR_RX_MATCH_PKTS in ipsec-input Change-Id: Ifcf167812d2cc18187c2cea84b657a52b67e17d4 Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2019-02-09buffers: fix typoDamjan Marion16-22/+21
Change-Id: I4e836244409c98739a13092ee252542a2c5fe259 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-09tls: move test certificates to separate header fileFlorin Coras8-76/+102
Change-Id: I543cb78e268d7a4a7fba590d305351ec79f4e4da Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-09session: refactor listen logicFlorin Coras17-645/+598
Make app-listener the handle for app listens. Consequently transport and local listen sessions are now associated to the app-listener. Change-Id: I9397a26d42cccb100970b6b4794c15bac2e11465 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-07tcp: fix fib_index for v6 ll packetsFlorin Coras1-0/+7
Change-Id: Ie69bdb9860d61f2c599c0c80b0ea80dfdfe178bd Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-07IPSEC: remove unused members of main structNeale Ranns1-4/+0
Change-Id: Ib55deb620f4f58cac07da7cb69418a3a30ff3136 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-07IPSEC: no second lookup after tunnel encapNeale Ranns15-444/+716
in the same maaner as with other tunnel tyeps we use the FIB to cache and track the destination used to reach the tunnel endpoint. Post encap we can then ship the packet straight to this adjacency and thus elide the costly second lookup. - SA add and del function so they can be used both directly from the API and for tunnels. - API change for the SA dump to use the SA type - ipsec_key_t type for convenience (copying, [un]formating) - no matching tunnel counters in ipsec-if-input Change-Id: I9d144a59667f7bf96442f4ca66bef5c1d3c7f1ea Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-07Fix parsing overflow in unformat_mac_address_t()Benoît Ganne1-3/+3
'%x' unformat specifier expects a pointer to a 4-byte object and will overflow when using a pointer to a 1-byte object. Use '%X' instead which allows to pass the size of the object alongside its pointer. The bug was exposed with the following commands: ~# make run DBGvpp# loop create loop0 DBGvpp# set ip6 neigh loop0 3001::2 a:a:a:a:a:a DBGvpp# show ip6 neigh Time Address Flags Link layer Interface 35.7743 ::2 D 0a:0a:0a:0a:0a:0a loop0 ^^^ wrong address: should be 3001::2 Note that the bug impact depends from the parsing order and memory layout. Change-Id: I29ba2eb53ba5a2daf4517215602d027508e2cb9f Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-02-07tcp: fix reset sending in syn-sentFlorin Coras3-6/+7
Change-Id: I468b1015bca5f95b162fb49e904aa4e4d38db75e Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-06buffers: make buffer data size configurable from startup configDamjan Marion16-21/+24
Example: buffers { default data-size 1536 } Change-Id: I5b4436850ca18025c9fdcfc7ed648c2c2732d660 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-06virtio: enable msix interrupt modeMohsin Kazmi4-44/+114
Change-Id: Idd560f3afde1dd03bc3d6fbb2070096146865f50 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-02-06virtio: Use new buffer optimizationMohsin Kazmi5-2/+24
Change-Id: Ifc98373371b967c49a75989eac415ddda1dcf15f Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-02-06transport: cleanupFlorin Coras19-544/+571
- move transport specific types to transport_types - add transport wrapper functions for interaction with transport protocol vfts Change-Id: I93f70d884585fc2f41c4a605e310c80e8a8972f2 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-05session: cleanup part 3Florin Coras7-234/+245
Aggregate session specific types, getters and setters under session_types.h Change-Id: Ib205337502654969d60c72d1800c90247e1a9068 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-05session: cleanup part 2Florin Coras8-1191/+1228
Move app worker logic to app_worker.c Change-Id: Ic5e5735b2884f006c064d023f491aa6888114810 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-05IPSEC: SPD counters in the stats sgementNeale Ranns12-505/+411
- return the stats_index of each SPD in the create API call - no ip_any in the API as this creates 2 SPD entries. client must add both v4 and v6 explicitly - only one pool of SPD entries (rhter than one per-SPD) to support this - no packets/bytes in the dump API. Polling the stats segment is much more efficient (if the SA lifetime is based on packet/bytes) - emit the policy index in the packet trace and CLI commands. Change-Id: I7eaf52c9d0495fa24450facf55229941279b8569 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-02-04session: cleanup part 1Florin Coras37-497/+482
Rename core data structures. This will break compatibility for out of tree builtin apps. - stream_session_t to session_t - server_rx/tx_fifo to rx/tx_fifo - stream_session.h to session_types.h - update copyright Change-Id: I414097c6e28bcbea866fbf13b8773c7db3f49325 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-04ipsec: cli 'create ipsec tunnel' more optionsKingwel Xie1-3/+7
use-esp use-anti-replay Change-Id: I977b65eee926adaded0cb923e14feb0ee90fc32c Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2019-02-02session: improve show session cliFlorin Coras3-43/+64
Change-Id: If30cea80058752cedcc6ecc38e90bef8732c83d3 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-02Deprecate old mutliarch code, phase 1Damjan Marion6-56/+9
It is causing compilation sloness with gcc-7 so removing it before it was originally planned. So far macros are left in the tree so we can know which nodes to convert to new multiarch code. Change-Id: Idb14622ca61fdce1eba59723b20d98715b7971e6 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-02tcp: remove internal buffer cacheFlorin Coras2-81/+32
Change-Id: I3051d49aa9b9c38dcd8069ea1748f74ab20ac0f6 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-02session: remove local buffer cacheFlorin Coras1-33/+14
Change-Id: I71ccb39c56962fe0160d3f4c83453e72d2dd3030 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-02-01IPSEC: minor refactorNeale Ranns11-754/+1002
No function change. Only breaking the monster ipsec.[hc] into smaller constituent parts Change-Id: I3fd4d2d041673db5865d46a4002f6bd383f378af Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-01-31IPSEC: API modernisationNeale Ranns3-257/+396
- use enums to enumerate the algoritms and protocols that are supported - use address_t types to simplify encode/deocde - use typedefs of entry objects to get consistency between add/del API and dump Change-Id: I7e7c58c06a150e2439633ba9dca58bc1049677ee Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-01-30virtio: fix the device order (legacy or modern)Mohsin Kazmi1-3/+3
Change-Id: I60f88d50f062b004e6dea487bd627d303d0a5e75 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-01-30Use IP and MAC API types for neighborsNeale Ranns23-597/+685
use address_t and mac_address_t for IPv6 and ARP entries and all other API calls in ip.api aprat from the route ones, that will follow in a separate commit Change-Id: I67161737c2184d3f8fc1e79ebd2b55121c5b0191 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-01-30MPLS tunnel; fix crash when deleting non-existant pathNeale Ranns1-0/+3
in the case the tunnel is the only user of the shared path list then removing its dependency removes the path list. hence lock the list Change-Id: I18318441698ceac16715b1826266a7d19dcd76e1 Signed-off-by: Neale Ranns <nranns@cisco.com>
2019-01-30ipsec: fix check support functionsKlement Sekera2-5/+17
Change-Id: If94c57fbb07a7376a9f2873e1489c00b28152620 Signed-off-by: Klement Sekera <ksekera@cisco.com>