aboutsummaryrefslogtreecommitdiffstats
path: root/src/plugins/dpdk/device
AgeCommit message (Collapse)AuthorFilesLines
2017-04-11dpdk: use common interface placement infraDamjan Marion6-301/+64
This pathch deprecates "show dpdk placement" and "set dpdk placement" CLI commands. Change-Id: I4e052ec3e8b8e6c54b4816e1e689e5b7a24892db Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-04-07Fix init of ENIC driver to allow receiving of jumbo packetsJohn Lo1-49/+27
Set rxmode.enable_scatter field in DPDK port config parameter so ENIC driver will use multiple mbuf's for receiving jumbo packets. Also remove ENIC driver check to disable setting ENIC MTU as this capability is now working with the new ENIC driver, subject to 9002B limit. Change-Id: I563976201c4968d4538c0759505cef2de876934a Signed-off-by: John Lo <loj@cisco.com>
2017-04-07MPLS McastNeale Ranns1-1/+1
1 - interface-DPO Used in the Data-plane to change a packet's input interface 2 - MPLS multicast FIB entry Same as a unicast entry but it links to a replicate not a load-balance DPO 3 - Multicast MPLS tunnel Update MPLS tunnels to use a FIB path-list to describe the endpoint[s]. Use the path-list to generate the forwarding chain (DPOs) to link to . 4 - Resolve a path via a local label (of an mLDP LSP) For IP multicast entries to use an LSP in the replication list, we need to decribe the 'resolve-via-label' where the label is that of a multicast LSP. 5 - MPLS disposition path sets RPF-ID For a interface-less LSP (i.e. mLDP not RSVP-TE) at the tail of the LSP we still need to perform an RPF check. An MPLS disposition DPO performs the MPLS pop validation checks and sets the RPF-ID in the packet. 6 - RPF check with per-entry RPF-ID An RPF-ID is used instead of a real interface SW if index in the case the IP traffic arrives from an LSP that does not have an associated interface. Change-Id: Ib92e177be919147bafeb599729abf3d1abc2f4b3 Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-04-06Use thread local storage for thread indexDamjan Marion4-24/+26
This patch deprecates stack-based thread identification, Also removes requirement that thread stacks are adjacent. Finally, possibly annoying for some folks, it renames all occurences of cpu_index and cpu_number with thread index. Using word "cpu" is misleading here as thread can be migrated ti different CPU, and also it is not related to linux cpu index. Change-Id: I68cdaf661e701d2336fc953dcb9978d10a70f7c1 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-03-31dpdk: add support for Mellanox ConnectX-5 devicesDamjan Marion1-1/+4
Change-Id: I3ed2834a326eac50a7cb4faa592f42fd06325d5a Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-03-22Update MAC address kept for Bonded Interface and its SlavesJohn Lo1-23/+32
Change-Id: I9f7f9f840c3c1aad5e8c9a4fa1ba7a58a85cfd9e Signed-off-by: John Lo <loj@cisco.com>
2017-03-18Fix bonded interface name to use instance number starting from 0John Lo3-5/+7
With DPDK 17.02, bonded interface device_index does not start from 0 and may vary depends on the existence of other interfaces. Implement instance number for bonded interface so the interface name can make use of instance number starting from 0. Change-Id: Ia8eb7bd530446faa02ee7c7d1f6abdc22ac60b62 Signed-off-by: John Lo <loj@cisco.com>
2017-03-17Fix bonded ethernet PMD name for DPDK 17.02Alexander Popovsky (apopovsk)2-7/+24
In DPDK 17.02 bonded PMD name is changed from rte_bond_pmd to net_bonding. Change-Id: I1a57a16b0ae68b5fa56a561a4f75981112228572 Signed-off-by: Alexander Popovsky (apopovsk) <apopovsk@cisco.com>
2017-03-14Fix packet trace output: dpdk_rx_trace must decode the actual ethertypeDave Barach1-1/+31
As opposed to rubbish (typically) 14 octets past the ethertype. Also fix buffer error code setup in dpdk-input node single loop. Change-Id: Ide7c4097d3bb91e62749ed4e1d69a7b4b90225de Signed-off-by: Dave Barach <dbarach@cisco.com> Signed-off-by: John Lo <loj@cisco.com>
2017-03-13VPP-279: Document changes for vnet/vnet/devicesBilly McFall1-0/+27
Add doxygen documentation for DPDK crypto CLI command. The move of DPDK to a plugin invalidated the local dir.dox. So moved dpdk/dir.dox to dpdk/device/dir.dox to fix. Change-Id: I229e2921c4b0cdd380021adb520cd2089a376afa Signed-off-by: Billy McFall <bmcfall@redhat.com>
2017-03-08dpdk: dpdk-input optimizations and fixesDamjan Marion3-96/+122
- fix issue caused by assumption that ethertype starts at 0 - intoduce buffer templates to speed-up vlib_buffer_t metadata initialization - avoid check for multiseg buffers inside loop if multiseg is disabled - interleave prefetches to reduce load on L1 cache Change-Id: I3b76e6d3e1e15ed28f01625edb7fbe9f38112e03 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-03-08Change dpdk input node fixed-sleep units to usecDave Barach3-5/+5
Change-Id: I94e8737cad9222d24602db4ad03bf43c12c62b79 Signed-off-by: Dave Barach <dave@barachs.net>
2017-03-08dpdk: Fixed QAT devices binding, bind the VFs onlyRadu Nicolau1-3/+7
Change-Id: I9ff64b6c583e1c513f91f90798298b55d0793f06 Signed-off-by: Radu Nicolau <radu.nicolau@intel.com>
2017-03-06features: take device-input buffer advance value directlyDamjan Marion1-4/+2
Change-Id: Ifac7d9134d03d79164ce6f06ae9413279bbaadb3 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-03-06dpdk: init.c should be under device/Damjan Marion1-0/+2074
Change-Id: I80831cee062a38a0f5ab1f1e56c2dc6dcd512b9d Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-03-02Determine pkt type in dpdk-input node using ethertype only (VPP-647)John Lo1-52/+22
Remove reliance on DPDK driver provided mbuf packet type in dpdk-input node as some NIC driver provide misleading information. Now using ethertype from the packet itself to determine packet type for next node. Change-Id: Ie7b514a984f9382c29f1a1e3eb423d68f817c064 Signed-off-by: John Lo <loj@cisco.com>
2017-03-01dpdk: be a pluginDamjan Marion6-0/+4984
Change-Id: I238258cdeb77035adc5e88903d824593d0a1da90 Signed-off-by: Damjan Marion <damarion@cisco.com>