summaryrefslogtreecommitdiffstats
path: root/src/vnet/pg/input.c
AgeCommit message (Collapse)AuthorFilesLines
2021-05-06vlib: fix the offload flags sizeMohsin Kazmi1-1/+1
Type: fix Change-Id: I433fe3799975fe3ba00fa30226f6e8dae34e88fc Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2021-02-15vlib: refactor checksum offload supportMohsin Kazmi1-13/+18
Type: refactor This patch refactors the offload flags in vlib_buffer_t. There are two main reasons behind this refactoring. First, offload flags are insufficient to represent outer and inner headers offloads. Second, room for these flags in first cacheline of vlib_buffer_t is also limited. This patch introduces a generic offload flag in first cacheline. And detailed offload flags in 2nd cacheline of the structure for performance optimization. Change-Id: Icc363a142fb9208ec7113ab5bbfc8230181f6004 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2020-12-14misc: refactor clib_bitmap_foreach macroDamjan Marion1-2/+2
Type: refactor Change-Id: I077110e1a422722e20aa546a6f3224c06ab0cde5 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-11-13pg: fix uninitialized variable in pg mac filter CLIAndrew Yourtchenko1-1/+1
Fixes coverity issue 214887 Change-Id: I81bfc009faabcb74f950a94715a0395fac264ee9 Type: fix Fixes: 21fb4f71ee3824c8f177045f21fea258ece602a9 Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2020-11-09vlib: fix trace number accountingBenoît Ganne1-10/+17
When using classifier to filter traces, not all packets will be traced. In that case, we should only count traced packets. Type: fix Change-Id: I87d1e217b580ebff8c6ade7860eb43950420ae78 Signed-off-by: Benoît Ganne <bganne@cisco.com>
2020-10-08pg: Choose the input interface from the stream's rxNeale Ranns1-1/+3
Type: fix Signed-off-by: Neale Ranns <nranns@cisco.com> Change-Id: I656918a417d33ec6bea30054805e03ae19c38f2d
2020-10-08fib: Register multicast MAC with interface for accepting interfacesNeale Ranns1-0/+118
Type: fix Signed-off-by: Neale Ranns <nranns@cisco.com> Change-Id: Ic6c76b65e2dcc08916373153944507a297c962c0
2020-08-14gso: packet coalesce libraryMohsin Kazmi1-0/+4
Type: feature Change-Id: Ia19d3611e596d9ec47509889b34e8fe793a0ccc3 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2020-06-16gso: fix the udp checksum in testMohsin Kazmi1-12/+7
Type: fix Change-Id: I7005e4763e73f3679c40f94bdab26d439cd23188 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2020-06-08vxlan: Fixed checksum caclculation offsetVladimir Isaev1-10/+19
VXLAN uses csum_offload for IPv6 packets. But without gso node we have csum calculated only for inner packet. This patch adds support for outer header csum calculation. Checksum for inner packet should be calculated before interface-output node (for example in vxlan node). Type: fix Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com> Signed-off-by: Vladimir Isaev <visaev@netgate.com> Change-Id: Ica68429ede4426293769207cd83c791ebe72fe56
2020-04-15pg: set vnet buffer flags in pg streamsDave Barach1-1/+4
See .../extras/pg/checksum_offload.pg for a nontrivial example, which deliberately sets bogus ip and udp checksums in the generated packets, then fixes the mess with (software emulated) hardware checksum offload. Validated via "pcap dispatch trace on max 1000 buffer-trace pg-input 100". Packets stuffed into loop1-output have the configured bogus ip and udp checksums. vnet_calc_checksums_inline(...) fixes the checksums, which are correct when packets visit loop1-tx. The packet generator is a dumb robot in this regard. If you ask for a ridiculous flag combination - example: ip4 and ip6 - your mileage may vary. Type: feature Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: I6d9e790f373bcd9e445a89113ca8e4c8f9aa9419
2019-12-06gso: fix the tap/virtio driver for header offsetMohsin Kazmi1-1/+24
Type: fix Change-Id: Ied34466907fa8ad44f997c600dbf481be4d22027 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-12-05gso: add protocol header parserMohsin Kazmi1-29/+9
Type: feature Change-Id: I7c6be2b96d19f82be237f6159944f3164ea512d0 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-10-03misc: add "maxframe" and "rate" to packet-generator cli.Christian E. Hopps1-2/+2
Allow for setting the maximum number of generated packets to be included in the frame passed to next nodes. This is very important for testing code which may be susceptible to multi-frame vs single-frame bugs (e.g., code that is doing re-ordering where packets may be buffered between frames). Update: - remove redundant packet "rate" option. - reduce n_max_frame to u32 as that's what pulled from the CLI. Type: feature Signed-off-by: Christian E. Hopps <chopps@chopps.org> Change-Id: Ie362bbb110b2cf01d9f65c559bbe9101e17b7fdc Signed-off-by: Christian Hopps <chopps@labn.net>
2019-09-24vlib: add flag to explicitelly mark nodes which can init per-node packet traceDamjan Marion1-0/+1
Type: feature Change-Id: I913f08383ee1c24d610c3d2aac07cef402570e2c Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-08-17gso: fix l3 and l4 header offset in case of tagged interfaceMohsin Kazmi1-0/+14
previously, PG and virtio interfaces calculate wrong l3 and l4 header offset. This patch fixes this issue. Type: fix Ticket: VPP-1739 Change-Id: I5ba978e464babeb65e0711e1027320d46b3b9932 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-08-16gso: remove the ip checksum flag in case of ipv6Mohsin Kazmi1-1/+0
Type: fix Ticket: VPP-1727 Change-Id: Icfee35c5ab5e1c65079d1ca7bb514162319113e5 Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-07-28pg: add GSO supportMohsin Kazmi1-2/+71
Type: feature Change-Id: I72676495a85fbecc946aa266a75234cce70c3a5e Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
2019-07-18vlib: convert frame_index into real pointersAndreas Schultz1-1/+1
The fast path almost always has to deal with the real pointers. Deriving the frame pointer from a frame_index requires a load of the 32bit frame_index from memory, another 64bit load of the heap base pointer and some calculations. Lets store the full pointer instead and do a single 64bit load only. This helps avoiding problems when the heap is grown and frames are allocated below vm->heap_aligned_base. Type: refactor Change-Id: Ifa6e6e984aafe1e2755bff80f0a4dfcddee3623c Signed-off-by: Andreas Schultz <andreas.schultz@travelping.com> Signed-off-by: Dave Barach <dave@barachs.net>
2019-06-29pg: format_pg_input_trace - reorder fieldsPaul Vinciguerra1-1/+1
00:00:00:814640: pg-input stream pcap0-sw_if_index-1, 42 bytes, 1 sw_if_index is changed to: 00:00:00:814640: pg-input stream pcap0-sw_if_index-1, 42 bytes, sw_if_index 1 Type: style Change-Id: I9bb32494c9c1d08bc7588f088ed67a60ed3236dd Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
2019-05-09add mactime plugin unit / code coverage testsDave Barach1-4/+3
The unit and code coverage tests are boring. The rest of the patch involves test and packet-generator infra cleanups. Teach the "make test-xxx" family of targets to set the api test plugin path correctly, to make "binary-api <api-message-name> <args>" debug CLI commands work correctly in the "make test" environment. Unfortunately involves both the top-level and test Makefiles. Add a minor pg cli feature, a CLI to manually set s->sw_if_index[VLIB_TX]. Consider the case where one configures an interface with both a device-input and an output feature. To test the output feature using the pg, it's necessary to inject packets into the interface output node with both b->sw_if_index[VLIB_TX] and b->sw_if_index[VLIB_RX] set correctly. For example: packet-generator new { name tx limit 15 size 128-128 interface local0 # rx: device input feature not configured on local0 tx-interface loop0 # tx: output node requires b->sw_if_index[VLIB_TX] node loop0-output data { hex 0x01005e7ffffa000dead0000008000102030405060708090a0b0c0d0e0f0102030405 } } Fix a longstanding bug in the packet generator stream setup. Remove kludges which set b->sw_if_index[VLIB_TX] to ~0 [in multiple places] instead of using the stream value s->sw_if_index[VLIB_TX], and setting THAT datum correctly. Change-Id: I1097a18e8db73661ded6b822c1d718f7e5cf36ed Signed-off-by: Dave Barach <dave@barachs.net>
2019-02-22Add no-append flag to vlib_frame_tDamjan Marion1-0/+1
Change-Id: I01c4f5755d579282773ac227b0bc24f8ddbb2bd1 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-20pg: fix tracing when device-input feature arc is usedDamjan Marion1-6/+4
Change-Id: I82240b43d3a5f3f33ac9ab2de106b3ec0ea31780 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-20pg: fix crash when multiple streams are enqueuing to the same nextDamjan Marion1-1/+1
Only with debug image and when next node is not ethernet-input... Change-Id: Iaa404b5d35d5c04996ff48cd16877858092b78d7 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-09buffers: fix typoDamjan Marion1-1/+1
Change-Id: I4e836244409c98739a13092ee252542a2c5fe259 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-02-06buffers: make buffer data size configurable from startup configDamjan Marion1-3/+3
Example: buffers { default data-size 1536 } Change-Id: I5b4436850ca18025c9fdcfc7ed648c2c2732d660 Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-01-18Add vlib_buffer_copy_indices inline functionDamjan Marion1-4/+3
This reverts commit 1e59f9ddbdda14591967e1d66eab8623f9ba58e4. Change-Id: Iae1d372b887e170d28cac2fe4c61325ee5a5894a Signed-off-by: Damjan Marion <damarion@cisco.com>
2019-01-06pktgen: support mixed multi/single chunk pcap replayDave Barach1-71/+158
Change-Id: I3d387b5e2b17f89ed688ea6cfee3fb6d782fe326 Signed-off-by: Dave Barach <dave@barachs.net>
2019-01-01buffers: remove unused codeDamjan Marion1-7/+5
Change-Id: If2bbfbc52994f5de0879763e0b7a7864498debb6 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-11-20ethernet-input optimizationsDamjan Marion1-1/+19
Change-Id: I4ec7750ef58363bd8966a16a2baeec6db18b7e9e Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-11-14Remove c-11 memcpy checks from perf-critical codeDave Barach1-13/+16
Change-Id: Id4f37f5d4a03160572954a416efa1ef9b3d79ad1 Signed-off-by: Dave Barach <dave@barachs.net>
2018-07-19Cleanup of handoff codeDamjan Marion1-3/+0
- removed handoff-dispatch node - removed some unused buffer metadata fields - enqueue to thread logic moved to inline function Change-Id: I7361e1d88f8cce74cd4fcec90d172eade1855cbd Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-05-17Packet generator: preserve pcap file timestampsDave Barach1-0/+1
Set vnet_buffer2(b0)->pg_replay_timestamp, for use when desired. Fix a memory leak in pg_stream_free(...), which wasn't freeing the replay packet templates. Change-Id: I01822a9e91a52de4774d2b95cf0c2ee254a915e9 Signed-off-by: Dave Barach <dave@barachs.net>
2018-04-27A bit of buffer metadata reshuffling to accommodate flow_idDamjan Marion1-1/+1
Change-Id: I2794384557c6272fe217269b14a9db09eda19220 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-03-07packet-generator: use default free listDamjan Marion1-46/+6
Change-Id: I2a34546f9b32edc9bfb86b5492dde34aaef49ccc Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-11-21dpdk: add l2_hdr_offset and l3_hdr_offset in vlib_buffer_tDamjan Marion1-2/+2
Change-Id: I0a6d1257e391c3b6f7da6498bd5f7d4c545d17e9 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-10-04[aarch64] Fixes CLI crashes on dpaa2 platform.Christophe Fontaine1-1/+1
- always use 'va_args' as pointer in all format_* functions - u32 for all 'indent' params as it's declaration was inconsistent Change-Id: Ic5799309a6b104c9b50fec309cba789c8da99e79 Signed-off-by: Christophe Fontaine <christophe.fontaine@enea.com>
2017-07-10vlib: store buffer memory information in the buffer_mainDamjan Marion1-3/+3
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>
2017-04-06Use thread local storage for thread indexDamjan Marion1-2/+2
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>
2017-03-07DHCP Multiple Servers (VPP-602, VPP-605)Neale Ranns1-0/+6
Multiple DHCP (4 and/or 6) servers can be added and removed through multiple calls to the 'set dhcp server' API. All 4/6/ discover/solicit messages will then be replicated to all servers in the list. The expectation is that the servers/system is configured in such a way that this is viable. If VSS information is providied for the clinet VRF which also has multiple servers configured, then the same VSS information is sent to each server. Likewise the source address of packets sent to from VPP to each server is the same. Change-Id: I3287cb084c84b3f612b78bc69cfcb5b9c1f8934d Signed-off-by: Neale Ranns <nranns@cisco.com>
2017-03-01dpdk: be a pluginDamjan Marion1-5/+6
Change-Id: I238258cdeb77035adc5e88903d824593d0a1da90 Signed-off-by: Damjan Marion <damarion@cisco.com>
2016-12-28Reorganize source tree to use single autotools instanceDamjan Marion1-0/+1667
Change-Id: I7b51f88292e057c6443b12224486f2d0c9f8ae23 Signed-off-by: Damjan Marion <damarion@cisco.com>