summaryrefslogtreecommitdiffstats
path: root/src/vnet
AgeCommit message (Collapse)AuthorFilesLines
2017-07-20Fix coverity issue introduced with IP checksum offload commitDamjan Marion1-3/+3
Change-Id: Ib5ad1369678389534426902ce698bccf7a6d9ec2 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-07-20Prevent double freeing of first segment manager.Dave Wallace1-1/+2
Change-Id: I27d7cb4c4e142f29a0c3c3ff85dd1f581970abfc Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2017-07-20L2FWD:move vec_validate out of access functionEyal Bari2-2/+9
Change-Id: Id9737b6aa2b6fe3032f4627dfdbd2ea728cc3fb1 Signed-off-by: Eyal Bari <ebari@cisco.com>
2017-07-19Add a bihash prefetchable bucket-level cacheDave Barach3-7/+7
According to Maciek, the easiest way to leverage the csit "performance trend" job is to actually merge the patch once verified. Manual testing indicates that the patch improves l2 path performance. Other use-cases are TBD. It's possible that we'll need to back out the patch depending on what happens. Change-Id: Ic0a0363de35ef9be953ad7709c57c3936b73fd5a Signed-off-by: Dave Barach <dave@barachs.net>
2017-07-19Fix IP checksum offload metadata setupDave Barach1-2/+12
Change-Id: Idf4f8b02337551670d6b7ce1a03799de9c53d19a Signed-off-by: Dave Barach <dave@barachs.net>
2017-07-18FIB path preferenceNeale Ranns13-62/+647
Paths are given a preference, lowest value is 'best'. Only paths that are up are up contribute to fprwarding - that's unchanged. What's new is that of the path's that re up only those that have the best preference contribute. A poor man's primary and backup. It's not true primary/backup function because the FIB must converge before the lower preference paths are used. Change-Id: Ie4453c4a7b1094c6c2b51fe1594b8302103bb68e Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-07-18DHCP client option 61 "client_id"Neale Ranns4-2/+19
the existing seeting of client_id to a VPP version number was unused and so overridden Change-Id: If9ebea936336f1fcca8d07e67186c95f8f8f0ccd Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-07-18DHCP client - remove interface address when DHCP de-configuredNeale Ranns1-0/+1
Change-Id: I63c59e3c13859b51999d283774f7783ef0a6a5ed Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-07-18TCP/UDP checksum offload APIDave Barach6-13/+121
Change-Id: I2cb6ce4e29813f6602b14e6e61713fb381fbcef8 Signed-off-by: Dave Barach <dave@barachs.net>
2017-07-16Adjust replication context to match vnet_buffer(b)Dave Barach1-2/+2
Fixes a coverity warning, albeit not a serious issue. Change-Id: I0fa28342ac4b222c1ba7f7f2befde455fd42d682 Signed-off-by: Dave Barach <dave@barachs.net>
2017-07-15Fixes and improved tcp/session debuggingFlorin Coras9-91/+960
- Fix rx sack option parsing - Add session sack scoreboard tracing and replaying - Add svm fifo tracing and replaying - Scoreboard/svm fifo ooo segment reception fixes - Improved overall debugging Change-Id: Ieae07eba355e66f5935253232bb00f2dfb7ece00 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-07-14vnet_buffer_t flags cleanupDamjan Marion13-103/+108
Change-Id: I123eccea98abafeb31f25d2a162501e2eded60d4 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-07-14Introduce l{2,3,4}_hdr_offset fields in the buffer metadataDamjan Marion12-53/+36
To save space in the first cacheline following is changed: - total_length_not_including_first_buffer moved to the 2nd cacheline. This field is used only when VLIB_BUFFER_TOTAL_LENGTH_VALID and VLIB_BUFFER_NEXT_PRESENT are both set. - free_list_index is now stored in 4bits inside flags, which allows up to 16 free lists. In case we need more we can store index in the 2nd cachelin Change-Id: Ic8521350819391af470d31d3fa1013e67ecb7681 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-07-13Fix typo for non x86 platforms.Christophe Fontaine1-1/+1
Change-Id: Ic31b388cb972fb5f8a3fc42a5412401c3ee3e487 Signed-off-by: Christophe Fontaine <christophe.fontaine@enea.com>
2017-07-13Fix crash with worker threads on 4K VXLAN/BD setup (VPP-907)John Lo8-173/+51
Cleanup mapping of interface output node for the l2-output node when interface is configured to L2 or L3 modes. The mapping is now always done in the main thread as part of API/CLI processing, instead of initiate mapping in the forwarding path which can be in the worker threads. Change-Id: Ia789493e7d9f5c76d68edfaf34db43f3e3f53506 Signed-off-by: John Lo <loj@cisco.com> (cherry picked from commit bea5ebf205e0bec922bf26c6c1a6a9392b4cad67)
2017-07-12Deprecate support for flattened output nodesDamjan Marion4-276/+3
Change-Id: Id117e219146d9994340fb38c00233ea67db8929b Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-07-11Horizontal (nSessions) scaling draftDave Barach22-225/+679
- Data structure preallocation. - Input state machine fixes for mid-stream 3-way handshake retries. - Batch connections in the builtin_client - Multiple private fifo segment support - Fix elog simultaneous event type registration - Fix sacks when segment hole is added after highest sacked - Add "accepting" session state for sessions pending accept - Add ssvm non-recursive locking - Estimate RTT for syn-ack - Don't init fifo pointers. We're using relative offsets for ooo segments - CLI to dump individual session Change-Id: Ie0598563fd246537bafba4feed7985478ea1d415 Signed-off-by: Dave Barach <dbarach@cisco.com> Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-07-11L2INPUT:fix features mask cailculationEyal Bari1-6/+10
Change-Id: I84cea7530b01302a0adeef95b4924f54dc2e41ec Signed-off-by: Eyal Bari <ebari@cisco.com>
2017-07-10vlib: store buffer memory information in the buffer_mainDamjan Marion3-5/+5
Currently, buffer index is calculated as a offset to the physmem region shifted by log2_cacheline size. When DPDK is used we "hack" physmem data with information taken from dpdk mempool. This makes physmem code not usable with DPDK. This change makes buffer memory start and size independent of physmem basically allowing physmem to be used when DPDK plugin is loaded. Change-Id: Ieb399d398f147583b9baab467152a352d58c9c31 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-07-08lldp packet transmission on a bonded interfaceSteve Shin2-4/+8
LLDP packets are dropped at interface output node if each slave's link is configured as the LLDP interface. The admin state is configured and managed by the bonded interface, so slave link's state is down by default. The checking for the admin state UP should be ignored for the slave link. Change-Id: I06ca250f42fcb8cc50e0ea3a3817a2c5b56865df Signed-off-by: Steve Shin <jonshin@cisco.com>
2017-07-07ETH:optimized double is_tagged checkEyal Bari1-2/+24
a double version of is_tagged, uses "free lanes" in _mm_cmpeq_epi16 to check a second tag this code was not yet tested for performance Change-Id: I640017e1cc75c85a33d196ee911a7e4a512d9849 Signed-off-by: Eyal Bari <ebari@cisco.com>
2017-07-06Unlink /dev/shm files on application detach.Dave Wallace4-5/+36
Change-Id: Ieb5522cd5cb27bcbce808d4cd6df5248716da43c Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2017-07-06DHCP complete event sends mask lengthNeale Ranns3-3/+7
Change-Id: I4a529dfab5d0ce6b0bbc0ccbbd89c6b109dbf917 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-07-06VPP-902: LISP-CP: Wrong size in one_l2_arp_entries_get message.Ole Troan1-1/+1
Change-Id: I56bf6b46527f9465d78ed7c08b6e216e50c135ec Signed-off-by: Ole Troan <ot@cisco.com>
2017-07-06devices: show interface rx-placement displays the wrong information (VPP-894)Steven1-1/+3
show interface rx-placement somtimes displays the wrong interface names. This happens when there exists subinterfaces in VPP. The problem is due to the function show_interface_rx_placement_fn is calling format_vnet_sw_if_index_name with hw_if_index instead of sw_if_index. VPP has the concept of sw_if_index and hw_if_index. Each serves a different purpose. When there is no subinterfaces, both hw_if_index and sw_if_index may happen to have the same value. But don't count on it. When the API calls for sw_if_index, we must pass the sw_if_index although the hw_if_index has the same type which the compiler does not catch. Passing hw_if_index for an API which requires sw_if_index may have an unpredictable result such as described in the VPP-894 and sometimes it may even crash if the particular index does not exist. Change-Id: I76c4834f79b88a1c20684fcba64f14b2da142d77 Signed-off-by: Steven <sluong@cisco.com> (cherry picked from commit bafa4d048439fdbcc0bd577e43a2784d1b89bfc5)
2017-07-06Send GARP/NA on bonded intf slave up/down if in active-backup modeJohn Lo4-0/+147
If a bonded interface is in active-backup mode and configured with IPv4 and/or IPv6 addresses, on slave interface link up/down, send a GARP packet if configured with an IPv4 address and an unsolcited NA if configured with an IPv6 address. These packets can help with faster route convergence in the next hop router/switch. Change-Id: I68ccb11a4a40cda414704fa08ee0171c952befa2 Signed-off-by: John Lo <loj@cisco.com>
2017-07-05Add API support for LLDP config/interface setSteve Shin8-42/+301
Add API methods to configure LLDP and set interface to enable/disable. Also add port description TLV for LLDP. Change-Id: Ib959d488c2ab8a0069f143558871f41fcc43a5d3 Signed-off-by: Steve Shin <jonshin@cisco.com>
2017-06-30VPP debug image with worker threads hit assert on adding IP route with ↵Neale Ranns1-1/+16
traffic (VPP-892) When stacking DPOs the VLIB graph is also updated to add the edge between the nodes, if this edge does not yet exist. This addition should be done with the workers stopped. Change-Id: I327e4d7d26f0b23eb280f17e4619ff2093ff7940 Signed-off-by: Neale Ranns <nranns@cisco.com> (cherry picked from commit c02bd03ddf5eec9e9c79811360685f13e4ba8ee1)
2017-06-29LLDP: properly parse lldp cmds from startup configKlement Sekera1-3/+1
Change-Id: I0e6c86bd923fcf7cf16f948b9869a5927e6d3745 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2017-06-29Fix SIGBUS on aarch64Christophe Fontaine1-1/+1
A call to 'clib_smp_swap (&((dq)->interrupt_pending), 0)' was creating a SIGBUS. Instead of making dq->interrupt_pending aligned on 64bits, we reduce the size from uword (u64) to u32, as the number of pending interrupts will never go above max of u32. Change-Id: Ifa5a6d3b7adee222329a671be01305cf50853b33 Signed-off-by: Christophe Fontaine <christophe.fontaine@enea.com>
2017-06-28switch vlib process model to tw_timer_template timer implDave Barach1-0/+1
Change-Id: I36bb47faea55a6fea7af7ee58d87d8f6dd28f93d Signed-off-by: Dave Barach <dave@barachs.net>
2017-06-27L2-LEARN:fix l2fib entry seq num not updated on hit (VPP-888)Eyal Bari5-87/+64
fixed instability in l2bd_multi_instnce test - sometimes failing with extra packets captured it appears l2-learn was not updating hit entries but rather a copy of them. if the ager did not have a chance to run before the test was running the learning cycle - entries were not updated with the packet's seq num - causing packets to flood when hitting the stale seq_num in l2-fwd - hence the extra packets fixed handling of filter entries revert workaround for instability in test Change-Id: I16d918e6310a5bf40bad5b7335b2140c2867cb71 Signed-off-by: Eyal Bari <ebari@cisco.com> (cherry picked from commit 25ff2ea3a31e422094f6d91eab46222a29a77c4b)
2017-06-26VPP-889: MAP Stats API/CLI crashes when no domains.Ole Troan2-1/+11
Change-Id: Ib7824bfc08cb3c8f20258379e1a1f2c159c4f687 Signed-off-by: Ole Troan <ot@cisco.com>
2017-06-25VPP crash on creating vxlan gpe interface. VPP-875Hongjun Ni3-6/+11
Change-Id: I6b19634ecb03860a7624d9408e09b52e95f47aef Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
2017-06-23IP4/IP6 FIB: fix crash during interface deletePavel Kotucek4-1/+42
after deleting a sub interface with IP4/IP6 address vpp crash Change-Id: Ie768ca845b9e2394f61e2a8e9722a80a788746e7 Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
2017-06-22Update lisp map record default ttl to 24hv17.10-rc0Florin Coras1-2/+2
Change-Id: Ib8c72f8e08e89357b64f2f69ab70d60d3a7ec506 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-06-22Improve svm fifo and tcp tx path performance (VPP-846)Florin Coras6-25/+137
- multiarch on svm fifo - avoid ip lookup on tx Change-Id: Iab0d85204a710979417bca1d692cc47877131203 Signed-off-by: Florin Coras <fcoras@cisco.com> Signed-off-by: Dave Barach <dbarach@cisco.com>
2017-06-22VNET:explicitly pad l2_classifyEyal Bari1-10/+4
Change-Id: I77412aa8c17b45b1533604e7bfe8fe052ed0f80a Signed-off-by: Eyal Bari <ebari@cisco.com>
2017-06-21Introduce default rx mode for device driversDamjan Marion4-6/+18
If interface is down and queues are not configured then we are not able to change rx-mode. This change introducess default mode which is stored per interface and applied if driver wants. Change-Id: I70149c21c1530eafc148d5e4aa03fbee53dec62f Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-06-21ARP: ignore non-connected routes and non-interface sources when determing if ↵Neale Ranns2-33/+104
source is connected Change-Id: I39fb0ec44cc322eaa12c0ff0700fc405d3982bfc Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-06-20vlib: make runtime_data handling thread-localIgor Mikhailov (imichail)1-16/+26
Change-Id: Ic2f2dc234199a5f882846880cbacff20fc8d477b Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
2017-06-20VPP-879 MAP: s/u32 is_add/u8 is_add in map.apiOle Troan1-1/+1
Change-Id: If35171005e409f77bed4cc16eccb66a85aae5dfb Signed-off-by: Ole Troan <ot@cisco.com>
2017-06-19L2FWD:fix seq_num overwritten + validate l2fib entries when forwardingEyal Bari8-56/+99
l2_classify memeber table_index was overlaid over l2.l2fib_seq_num which over written when table_index gets initialized in l2_input_classify solved by overlaying both table_index and opaque_index as only one is used seperated l2fib seq num from l2_input configs for better handling of theoretical ABA issue where an entry for a deleted interface is considered valid by the ager because a different interface with same sw_if_index and seq_num was created before the ager got a chance to delete Change-Id: I7b0eeded971627406f1c80834d7e02c0ebe62136 Signed-off-by: Eyal Bari <ebari@cisco.com>
2017-06-19Overall tcp performance improvements (VPP-846)Florin Coras17-426/+664
- limit minimum rto per connection - cleanup sack scoreboard - switched svm fifo out-of-order data handling from absolute offsets to relative offsets. - improve cwnd handling when using sacks - add cc event debug stats - improved uri tcp test client/server: bugfixes and added half-duplex mode - expanded builtin client/server - updated uri socket client/server code to work in half-duplex - ensure session node unsets fifo event for empty fifo - fix session detach Change-Id: Ia446972340e32a65e0694ee2844355167d0c170d Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-06-19NAT64: Hairpinning (VPP-699)Matus Fabian2-28/+49
Change-Id: I83a6c277fa211ac2c2ca2d603650c992886af0a7 Signed-off-by: Matus Fabian <matfabia@cisco.com>
2017-06-17Fix map-notify processing with multiple workersFlorin Coras2-13/+44
Change-Id: Id160346ebf533ee5f55bd735803624a75ed997b9 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-06-16LISP-GPE: add test CLI for NSHFilip Tehlar2-0/+132
Change-Id: I9999474c1a4b744f5d5880ee99a0293c576f2819 Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
2017-06-16Add missing barrier sync to rx placement infra codeDamjan Marion2-6/+18
Change-Id: I25ccf8260dbe7e1550aee3904a688fc135ce1f03 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-06-16Fix vni/dp_table endianness for gpe iface addition (VPP-882)Florin Coras1-6/+8
Change-Id: I2b78dad740b67fc05b0e2cf9c180809bc0962cd5 Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-06-16Fix gpe_native_fwd_rpaths_get (VPP-883)Florin Coras1-1/+3
Change-Id: Iab2aa5fd92b9e95049f55fce4177e236a482723c Signed-off-by: Florin Coras <fcoras@cisco.com>