summaryrefslogtreecommitdiffstats
path: root/src/vnet
AgeCommit message (Collapse)AuthorFilesLines
2018-05-04Harmonize vec/pool_get_aligned object sizes and alignment requestsDave Barach35-12/+146
Object sizes must evenly divide alignment requests, or vice versa. Otherwise, only the first object will be aligned as requested. Three choices: add CLIB_CACHE_LINE_ALIGN_MARK(align_me) at the end of structures, manually pad to an even divisor or multiple of the alignment request, or use plain vectors/pools. static assert for enforcement. Change-Id: I41aa6ff1a58267301d32aaf4b9cd24678ac1c147 Signed-off-by: Dave Barach <dbarach@cisco.com>
2018-05-03MPLS prefetch fixesNeale Ranns2-15/+15
Change-Id: I565d79af410825c72f291ab40178883b1bc6f1df Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-05-03tcp: fix echo server rx retries counter vec allocFlorin Coras1-0/+2
If sessions are not preallocated, the rx retries counters are not correctly validated/initialized Change-Id: Iaf7456f3a0e2181fcea0c370613d694f8e98276d Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-05-02Fix CentOS 7 build issue with vxlanMatthew Smith1-2/+4
Two vxlan union/struct initializations caused gcc on CentOS 7 to puke. Modified them to make the build work again. Change-Id: Iad667444b86cfde5ee4329993b520028d3b593ad Signed-off-by: Matthew Smith <mgsmith@netgate.com>
2018-05-02vxlan:encap - use vnet rewriteeyal bari3-53/+50
moving the rewrite into the tunnel struct Change-Id: Iec74b48e13456d32957e826cffb5ea35a8ebd1a0 Signed-off-by: Eyal Bari <ebari@cisco.com>
2018-04-30tcp/session: debug improvements/fixesFlorin Coras6-30/+25
Change-Id: I906e58b4f9827a79a6ab673f8fa2e03036c69820 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-30FIB: elide cover walk for insert of host routeNeale Ranns3-3/+25
Change-Id: I2d39e56ff605e3a24927d6330d65d0406f588381 Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-27IPv6 NS/RS; do not vec_validate global structs in the DPNeale Ranns1-12/+16
Change-Id: I5cab31639e7819c9ab7e9c9159d931b25161d00b Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-27IPv6 NS: use the mcast rewrite node to fill in the destination MACNeale Ranns2-263/+275
- move the IPv6 incomplete and glean node to ip6_neighbour.c (so it has access to ip6_neighbour_main_t) - use the RA info config on the interface to find the multicast adj to use Change-Id: I835e419072abe54fb09dafb0e7eb0a9e50eba1af Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-27TAP memory leaks:Neale Ranns4-28/+22
1 - use bit-map to re-use ID values and thus VLIB nodes 2 - free vrings 3 - free hw_address on HW interface delete (a HW * struct is memset on pool_get) 4 - free temporary node names during TX node setup Change-Id: Id114c8bb9c844fd4ceb02fbbeb4b511ecfeb61ce Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-27A bit of buffer metadata reshuffling to accommodate flow_idDamjan Marion7-24/+24
Change-Id: I2794384557c6272fe217269b14a9db09eda19220 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-04-25Carry recorded QOS bits in the outer VXLAN IP headerIgor Mikhailov (imichail)1-0/+35
Currently for VXLAN IPv4. Change-Id: Id4b8bc0d9f6ab043810e4d1b9f28e01c27ce0660 Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
2018-04-25span: crash in span_mirror [VPP-1254]Steven1-2/+7
It is possible for span-input to get call with sw_if_index which is greater than sm->interfaces and crashes in span_mirror () in the following line span_interface_t *si0 = vec_elt_at_index (sm->interfaces, sw_if_index0); For example, span-input mirrors a main interface as source, it may actually get call for traffic coming in from the subinterface and crashes. The fix is simply to check if sw_if_index >= vec_len (sm->interfaces) and punt if it is. Change-Id: I8312eb321d638518e14ba2326fffd1a7919646ca Signed-off-by: Steven <sluong@cisco.com> (cherry picked from commit 516d63ff2c6671f3b0dc641511a50017a9804179)
2018-04-25GBPv6: NAT66 actions for GBPNeale Ranns1-1/+1
Change-Id: I379150a88f2d53d6281be41e8bad6fc4f4e88a71 Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
2018-04-25vxlan:remove single bucket load-balance dpo'sEyal Bari1-9/+23
from encap path Change-Id: I62a8d13495355ad5e687f13b86c2a5d360bb2b7f Signed-off-by: Eyal Bari <ebari@cisco.com>
2018-04-25Fix some build warnings about "Old Style VLA"Juraj Sloboda2-4/+9
Change-Id: I69fee1dcf07a4d2eed69a59f0a36e63e3741ed4e Signed-off-by: Juraj Sloboda <jsloboda@cisco.com>
2018-04-25Adjacency walk more scalableNeale Ranns2-50/+30
When walking all adjacencies for a given {next-hop,interface} instead of walking all the adjacencies on that interface and matching the next-hop (which is O(n) in the number of adjacencies on that link, find all instances of an adjacency with any link-type and wtih that {next-hop,interfacE} pair: this is O(1). Change-Id: Ic80399fc9e93c8df111379c039e592d8cafbab18 Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-24lacp: deleting the bond subinterface may cause lacp to lose the partner ↵Steven3-23/+2
[VPP-1251] Problem: When the bond subinterface is removed, it was observed that we lost the lacp partner. Show hardware shows rx counter goes up, but show interface does not for the slave interfaces. Cause: We reset the interface promiscuous mode when the bond subinterface is deleted. This causes dpdk not to accept any packet. Leave the interface in promiscuous mode fixes the problem. Other fixes: There are few places we use hw_if_index as if they are sw_if_index. But they don't necessarily have the same value. As soon as a subinterface is created, they start to diverge. The fix is to use the correct API for the hw_if_index and sw_if_index. Change-Id: I1e6b8bca0a4aae396d217a141271cbf968500c91 Signed-off-by: Steven <sluong@cisco.com> (cherry picked from commit 42c6599bf3057a7e8f4f00f5b6a9dd72af48d283)
2018-04-24memory leakKingwel Xie1-0/+3
obvious leak of parent_indices Change-Id: I572b33de1756c8062a87c754117d990622fe12fe Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2018-04-24Support QOS_SOURCE_IP recording from L2 input node.Igor Mikhailov (imichail)2-5/+77
Some scenarios not involving ip[4,6]-input paths might benefit from IP header QOS fields recorded and applied. An example: L2 (overlay) traffic being encapsulated by VPP in VXLAN and transmitted on another (underlay) interface might want the QOS information carried over in the outer IP header. Change-Id: I4d9462c47ae6ba97680edb1e53340b17cfd7845b Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
2018-04-23MPLS VPLS CLI fixNeale Ranns1-0/+1
set the IS_RX flag based on CLI 'l2-input-on' Change-Id: I53d9129a7c09f605c1eb55753426f392b1b480bc Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
2018-04-20tcp: make newreno byte instead of acks dependentFlorin Coras4-3/+28
Should be more resilient to ack losses Change-Id: Icec3b93c1d290dec437fcc4e6fe5171906c9ba8a Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-20tcp: improve statsFlorin Coras3-128/+177
Change-Id: I9ab11ba9f958c679112eb22c8db39cb269a29dc7 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-19Allow an IPsec tunnel interface to be renumberedMatthew Smith5-29/+42
When creating an IPsec tunnel interface, allow a numeric identifier to be set for use in the interface's name in place of the dev instance. Default to using the dev instance if no value is explicitly set. When an IPsec tunnel is deleted, the interface is deleted now instead of being kept in a pool of available hw interfaces. Otherwise there was the possibility of conflicting tx node names between deleted tunnels and newly created ones. Change-Id: Ic525466622a0dec38a845fa5871c084f6d9da380 Signed-off-by: Matthew Smith <mgsmith@netgate.com>
2018-04-19Minor optimzation/cleanup to ethernet-input nodeJohn Lo1-5/+5
Change-Id: Ibf3ef82950f50b746394a731cd2e7cba1cd16ec4 Signed-off-by: John Lo <loj@cisco.com>
2018-04-18typo fix: UNKOWN -> UNKNOWNAndrey "Zed" Zaikin5-12/+12
Change-Id: I008a4d7ad7160d1f07e7ceef712a5318a9368308 Signed-off-by: Andrey "Zed" Zaikin <zed.0xff@gmail.com>
2018-04-18Mcast rewrite optimisationsNeale Ranns8-45/+59
hard code the address mask offsets. This are protocol specific and only used on ethernet when used at all. Change-Id: Ib1f6f33682f53254ffbb5a241a1583e65420e0c7 Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-18session: coverity warningsFlorin Coras2-3/+3
Change-Id: I53bb1e289ba16ee6d9aed7285b18437c996077a4 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-18SCTP: coverity warningsMarco Varlese1-5/+2
This patch addresses the warning 182994 Change-Id: I071f3cbdf9965c7b57d444f79430ee75c66c856e Signed-off-by: Marco Varlese <marco.varlese@suse.com>
2018-04-18udp/session: refactor to support dgram modeFlorin Coras18-274/+778
- adds session layer support for datagram based protocols - updates udp to work in pure connectionless and datagram mode. The existing connected mode is now 'accessible' for apps as a dummy UDPC, as in, connected udp, protocol. - updates udp_echo, echo client, echo server code to work in datagram mode. Change-Id: I2960c0d2d246cb166005f545794ec31fe0d546dd Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-18Fix the ip header offset counting in vnet/ipsecSzymon Sliwa1-5/+4
IP header is set as data + sizeof(ethernet_header_t), what does not need to be true. The solution is to use l3_hdr_offset. Change-Id: I5d9f41599ba8d8eb14ce2d9d523f82ea6e0fd10d Signed-off-by: Szymon Sliwa <szs@semihalf.com>
2018-04-17Add logging supportDamjan Marion5-52/+136
Change-Id: Ieac9cf50156dbbb4962411e900d59256441915ef Signed-off-by: Damjan Marion <damarion@cisco.com> Signed-off-by: Klement Sekera <ksekera@cisco.com>
2018-04-17IP mcast: allow unicast address as a next-hopNeale Ranns8-41/+306
Change-Id: I5e679f2601e37688f2768620479dc2efb7d19ca3 Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-17VPP-1243: fix reassembly CLIKlement Sekera1-5/+6
Change-Id: Ie20aaf0eb1a5a338a54f0de4d6da661431be5163 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2018-04-17vhost: show vhost-user <interface> may crash [VPP-1242]Steven1-0/+7
show vhost-user <interface> may cause a crash if interface is semi-bogus. Semi-bogus means it is a known vpp interface which has a hw_if_index, but it is bogus because it is not a vhost-user interface. The fix is to add a check to reject non vhost-user interface for the command. Change-Id: I63f1e8bfbf46f5ec4c30f9fb3546982b63cd7cc5 Signed-off-by: Steven <sluong@cisco.com>
2018-04-17SCTP: coverity scan warningsMarco Varlese3-30/+39
This patch addresses 182346, 182347, 182353. Change-Id: I59c3d74bd3a2122b836c93e22ddb632fa032f7fe Signed-off-by: Marco Varlese <marco.varlese@suse.com>
2018-04-17vxlan:remove counters writeback cacheEyal Bari2-519/+242
+refactor decap loop to remove repetitions and goto's slightly improves performance in scale (3k-4k tunnels) tests (7-9 clocks) slightly deteriorates performance in single tunnel tests (3-4 clocks) Change-Id: I1a64ed0279c00481b61a162296c5a30f58bf29c4 Signed-off-by: Eyal Bari <ebari@cisco.com>
2018-04-16Implement ip_probe_neighbor APIJohn Lo2-9/+63
Add API support similar to VPP CLI "ip probe-neighbor" except API call is asynch and will not wait, as the CLI does, for address resolution of probed neighbor. The API client can use the APIs want_ip4_arp_events or want_ip6_nd_events to get notified of the desired address resolution event. Change-Id: Ieab58abe75b5cc7f5185b3b91418b6362f8992d3 Signed-off-by: John Lo <loj@cisco.com>
2018-04-16session: use generic session pool for listenersFlorin Coras8-103/+42
Change-Id: I825f3822910a6ee368b9004e39ca16ac6833e30c Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-04-13Revert "MTU: Setting of MTU on software interface (instead of hardware ↵Damjan Marion19-232/+221
interface)" This reverts commit 70083ee74c3141bbefb185525315f1b34497dcaa. Reverting as this patch is causing following crash: 0: /home/damarion/cisco/vpp3/build-data/../src/vnet/devices/devices.h:131 (vnet_get_device_input_thread_index) assertion `queue_id < vec_len (hw->input_node_thread_index_by_queue)' fails Aborted Change-Id: Ie2a365032110b1f67be7a9d832885b9899813d39 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-04-13MTU: Setting of MTU on software interface (instead of hardware interface)Ole Troan19-221/+232
Change-Id: I98bd454a761a1032738a21edeb0fe847e801f901 Signed-off-by: Ole Troan <ot@cisco.com>
2018-04-13GBP V2Neale Ranns10-7/+73
update the GBP plugin to implement the full NAT feature set of opflex agent Change-Id: Ic06a039c889445ed0b9087fa1f292634192b0f8d Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
2018-04-13bond: ping fails between l2 BD [VPP-1238]Steven4-15/+91
In dpdk based bonding, when the bond interface is configured for l2, it automatically sets the bond interface to promiscuous mode and sets rx redirect to ethernet-input. This allows traffic to be bridged to non compute node facing interface when it is received from the compute node interface. For native vpp bonding, we need to do similar things. When the bond interface is configured for l2, we set the slave interfaces to promiscuous mode and set rx redirect to ethernet-input because dpdk does not know anything about the bond interface. Likewise, when a new interface is enslaved, we also need to do the same thing if the bond interface has already been configured for l2. Change-Id: I7e168008e8a4221be74929b2a20e6db0ce8f3110 Signed-off-by: Steven <sluong@cisco.com>
2018-04-12Fixes for 'make UNATTENDED=yes CC=clang CXX=clang verify'Neale Ranns12-95/+42
Change-Id: I994649761fe2e66e12ae0e49a84fb1d0a966ddfb Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-04-12bond: 1 packet/frame == bad performance [VPP-1236]Steven1-6/+8
While https://gerrit.fd.io/r/#/c/11316/ took care of 1 packet/frame for most of the bonding modes, it missed the broadcast mode. This patch is to fix the 1 packet/frame for the broadcast mode. Change-Id: Iac48a2977c7f702f341479cc712a6448090dbc60 Signed-off-by: Steven <sluong@cisco.com>
2018-04-10Make IPsec tunnel intf work with IPv4 output featuresMatthew Smith2-0/+13
With no IPv4 output features on an IPsec tunnel inferface, when packets are forwarded to that interface, they reach the ipsec-if-output node via the output_node_index on the hw interface and they are handled correctly. When an IPv4 output feature (e.g. output ACL, outbound NAT) is enabled on an IPsec tunnel interface, outbound IPsec stops working for that interface. The last node in the ip4-output feature arc is interface-output. From there a packet is sent to ipsec<N>-output, and then ipsec<N>-tx. The tx function for an IPsec tunnel interface that is called by ipsec<N>-tx is a dummy that doesn't do anything except write a warning message. Enable a feature on the interface-output feature arc for an IPsec tunnel interface so the ipsec-if-output node is reached from the interface-output node. Change-Id: Ia9c73d3932f5930ec7ce0791a0375b1d37148b01 Signed-off-by: Matthew Smith <mgsmith@netgate.com>
2018-04-09features: don't break linked list, create separate one for arcDamjan Marion3-5/+5
We need to keep original linked list so destructire can remove entries. Change-Id: I5ff5ca0e1a417d88707255207725bba46433c943 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-04-09L2: no-flood interface type in the Bridge-DomainNeale Ranns5-11/+42
Change-Id: I50ff0cacf88182f8e0be19840c50f4954de586e2 Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
2018-04-09plugins: unload plugin if early init failsDamjan Marion2-0/+57
Change-Id: I32f68e2ee8f5d32962acdefb0193583f71d342b3 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-04-09DVR: save the rewrite length in packet meta-data for featuresNeale Ranns1-3/+9
Change-Id: I0efd03bdb84bc9ff2334d398bfdb82486228114a Signed-off-by: Neale Ranns <neale.ranns@cisco.com>