aboutsummaryrefslogtreecommitdiffstats
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2018-06-29memif: minor fixesDamjan Marion2-9/+9
Change-Id: Ib06d9ce0fad48b784fd47db13c7a2f353c845fca Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-29VPP-1324 SIGSEGV vl_msg_api_handler_with_vm_node()Lukasz Majczak2-11/+21
Increase local arrays sizes to the correct number and gather missing statistics. Change-Id: If44d43f258730e9a18f50fb59fc526c214d2be8a Signed-off-by: Lukasz Majczak <lma@semihalf.com>
2018-06-29VPPAPIGEN: use SOURCE_DATE_EPOCH for datestringNirmoy Das1-1/+2
bwiedemann@suse.com pointed me that: For reproducible builds datestring should be generated using SOURCE_DATE_EPOCH which distributions set centrally, and have build tools that consume it to produce reproducible output. https://wiki.debian.org/ReproducibleBuilds/TimestampsProposal#Python Change-Id: Iefa2b93231bbc4c8af49736be4d7a5720c46b28d Signed-off-by: Nirmoy Das <nirmoy.aiemd@gmail.com>
2018-06-29CSIT-1148: VPPAPIGEN 2nd level import of .api ignoredOle Troan1-4/+5
Change-Id: I73dcab0c5ff2405b1fcba25975955e1267a4ac4e Signed-off-by: Ole Troan <ot@cisco.com>
2018-06-29ipsec: fix VPP-1333 - crash in ipsec policy cliKlement Sekera1-0/+6
Change-Id: I843d094b6bbd1cefba82d6026174be005e66d510 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2018-06-29igmp: bugfix and minor improvementsJakub Grajciar5-76/+60
Change-Id: I8d284117a668dc55c06a6d68fe358a3d7e26c738 Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
2018-06-29Removing the static keyword do_fragment routines, so that could be accessed ↵Vijayabhaskar Katamreddy2-3/+8
outside the ip_frag nodes Change-Id: I46d3d10fa763fcf7a579620ec7cf1b204a53bce8 Signed-off-by: Vijayabhaskar Katamreddy <vkatamre@cisco.com>
2018-06-29IP4 Router Alert option handling for IGMPNeale Ranns5-14/+157
and a new ip4-options node, inserted between ip4-input and ip4-punt, that checks for IP-router-alert option + IGMP combination and sends the packet to the ip4-local. This is required because some IGMP packets are sent to the group address and not the all-routers address. All IGMP packets are sent with the router alert option. Change-Id: I01f478d4d98ac9f806e0bcba0f6da6e4e7d26e2a Signed-off-by: Neale Ranns <nranns@cisco.com>
2018-06-29bihash key compare improvementsDamjan Marion3-12/+10
Looks like CPU doesn't like overlaping loads. This new codes in some cases shows 3-4 clock improvements. Change-Id: Ia1b49976ad95140c573f892fdc0a32eebbfa06c8 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-28vcl: move binary api and cfg to separate filesFlorin Coras9-1956/+2081
Change-Id: Ib88d703bb7d4b170059960b0688352c90c5fcc39 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-28Fix mheap_get_aligned() performance jackpotDave Barach2-3/+64
If non-trivial alignment (e.g. 64) requested, and the object size (e.g. 16) is smaller than (alignment_request - MHEAP_ELT_OVERHEAD_BYTES), round up the size request. This avoids creating remainder chunks, which are false-cache-line-sharing bait to begin with. Change-Id: Ie1a21286d29557d125bb346254b1be2def868b1a Signed-off-by: Dave Barach <dave@barachs.net>
2018-06-28vcl: refactor vcl sessionFlorin Coras2-201/+202
Change-Id: Idcba72fd84128547718dd32858e8c728925a6b1d Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-28jvpp: pass tuple of classes to isinstance()Marek Gradzki1-3/+3
Change-Id: I887af26d7d6476d7d9e6fbaa671f923814aa03bb Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2018-06-28reassembly: optimize bihash usageKlement Sekera2-45/+38
Change-Id: I19690db9996ecb8c5e86415d2fd3c7133ce698f8 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2018-06-28ip: vectorized ip checksumDamjan Marion3-69/+147
Change-Id: Ida678e6f31daa8decb18189da712a350336326e2 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-27vcl: refactor debuggingFlorin Coras3-1800/+756
Change-Id: I82e5239239ea24bdc97b1d59946ca0bba739f248 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-27Tune pool_get / pool_putDave Barach2-7/+56
Stop spending cycles repeatedly tail-trimming the pool free element bitmap; possibly at the expense of slightly hurting pool_foreach peformance. Change-Id: I8a7f3e7b26c71d7496ba9393b2a167dc7f538355 Signed-off-by: Dave Barach <dave@barachs.net>
2018-06-27acl: fix for loop initial declarationFlorin Coras1-1/+2
Change-Id: Ie899ccbaae4df7cce4ebbba47ed6c3cce5269bdb Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-27avf: binary API and configurable RX/TX queue sizeJakub Grajciar10-13/+580
Change-Id: Ibd3a8d28d8f1df2bc14c42e48498f6ac26081192 Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
2018-06-27gcc8 and Wstringop-truncationMarco Varlese4-9/+13
gcc8 introduced a new warning (Wstringop-truncation) which in our case is being treated as error. Disabling the warning globally might introduce bugs related to string truncation which are not desired by the developer (e.g. bug). Instead, this patch disables the warning only for those occurences which have been verified to be non-bugs but the desired behaviour as per developer will. Change-Id: I0f04ff6b4fad44061e80a65af633fd7e0148a0c5 Signed-off-by: Marco Varlese <marco.varlese@suse.com>
2018-06-27jvpp: cleanup generation of swap functions for custom typesMarek Gradzki1-8/+5
Change-Id: I2d06890ba25cfae9f96e820086f228286c807b42 Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2018-06-27jvpp: add support for unions (VPP-1322)Marek Gradzki6-50/+277
Change-Id: I2456a9b03bcae43793f9ac29eb74eff81269df7b Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2018-06-27vppinfra: add vector horizontal add and byte swap (SSE4.2 & AVX2)Damjan Marion2-0/+31
Change-Id: I4e0fd487970796f0153a5b16333827d23b57deac Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-27acl-plugin: tm: avoid hash calculation dependency on a memory store operationAndrew Yourtchenko1-1/+9
A small store into a middle of a larger structure that was subsequently loaded for calculating the bihash key was noticeably impacting the performance. Change-Id: If7f33e1b66e8b438ba7cc91abc0ca749850c6e45 Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2018-06-27acl-plugin: tm: add tuplemerge algorithm for relaxing the hashtable masksAndrew Yourtchenko3-5/+625
Slightly refactored from the initial implementation of the TupleMerge [1] algorithm by Valerio Bruschi (valerio.bruschi@telecom-paristech.fr) [1] James Daly, Eric Torng "TupleMerge: Building Online Packet Classifiers by Omitting Bits", In Proc. IEEE ICCCN 2017, pp. 1-10 Also add startup parameters to turn on/off the algorithm ("use tuple merge 1/0"), and a startup parameter to be able to tweak the split threshold ("tuple merge split threshold N"), the default value of the split threshold is 39 as per paper, but some more tuning might be necessary to find the best value. This change, alongside with the optimizations which avoid extra lookups, significantly reduces the slowdown on the ClassBench generated ACLs, which are supposed to resemble realistic ACLs seen in use in the field. Change-Id: I9713e4673970e9a62d4d9e9718365293375fab7b Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2018-06-26tcp/session: tx optimizationsFlorin Coras4-22/+53
- cache and reuse tcp options and rcv_wnd for session layer tx bursts - avoid reading/setting total_length_not_including_first_buffer. It's part of a buffer's second cache line so it comes at a "cost". Change-Id: Id18219c2f7e07cf4c63ee74f9cdd9e5918904036 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-26dpdk: display rx/tx burst function name in "show hardware detail"Damjan Marion1-0/+20
Change-Id: I6fa4c6bf9c4e96ba4502a06907bdecc654ace665 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-26Fix load_unaligned undefined and other possible build failuresSirshak Das1-26/+40
Add aarch64 neon intrinsics to fix build failures similar to this: error: implicit declaration of function ‘u64x2_load_unaligned’ Change-Id: I6178504a48242742df3f7d75abdaf108796cf73f Signed-off-by: Sirshak Das <sirshak.das@arm.com>
2018-06-26Fix assert issue in ip_csum_add_even()Hongjun Ni1-1/+2
ASSERT (ip_csum_with_carry (d, x) == c) will raise assert if d equals to zero while x not equals to zero. Change-Id: Ia9ccdbf801ae565eaadd49f04569d13bfc31cba8 Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
2018-06-26tcp: avoid doing work in tcp_rcv_sacks for no sacksFlorin Coras1-2/+3
Change-Id: I00a0d7f57dc144d338d5ad45b0a6e3720c32c400 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-26NAT44: fix nat44_ed_not_translate_output_feature (VPP-1329)Matus Fabian1-0/+5
Change-Id: Iddb0b848c53da03116524e203c7112c82b401ac5 Signed-off-by: Matus Fabian <matfabia@cisco.com>
2018-06-26node functions cannot be always_inlineDamjan Marion7-18/+18
Thanks to gcc-8 for highlighting this... Change-Id: I53bfab631a40fd1b680c76a48b0307a33fa2b154 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-26We don't have (yet) 128-bit unaligned load/store on ARMDamjan Marion1-2/+2
Change-Id: I16395bbf843e338cdd366d85bb4df3de95d9b265 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-26add backtrace in unix_signal_handlerKingwel Xie2-30/+63
crash stack backtrace will be directed to syslog 1. make use of glic backtrace in execinfo.h. the old clib_backtrace is removed 2. install SIGABRT in signal handler, but have to remove it when backtrace is done. reason is to capture stack trace caused by SIGABRT. vPP ASSERT always call os_exit then abort(). we definitely want to know the trace of this situation. It is a little tricky to avoid SIGABRT infinite loop 3. always load symbols by calling clib_elf_main_init () in main(). Otherwise, PC addresses instead of symbols will be displayed. Change-Id: I150e15b94a4620b2ea4f08c73dc3e6ad1856de1e Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
2018-06-26L3DSR fix ip checksum issue and add testHongjun Ni1-9/+12
Change-Id: Iedebbac71d3e694b915d6a126c80ecc3b5473a4a Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
2018-06-26acl-plugin: tm: optimize multi-lookups and prepare to add tuplemergeAndrew Yourtchenko6-252/+456
- instantiate the per-use mask type entry for a given hash ACE this prepares to adding tuplemerge where the applied ACE may have a different mask type due to relaxing of the tuples - store the vector of the colliding rules for linear lookups rather than traversing the linked list. - store the lowest rule index for a given mask type inside the structure. This allows to skip looking up at the later mask types if we already matched an entry that is in front of the very first entry in the new candidate mask type, thus saving a worthless hash table lookup. - use a vector of mask type indices rather than bitmap, in the sorted order (by construction) of ascending lowest rule index - this allows to terminate the lookups early. - adapt the debug cli outputs accordingly to show the data - propagate the is_ip6 into the inner calls Change-Id: I7a67b271e66785c6eab738b632b432d5886a0a8a Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2018-06-26Fix api trace replay of handler to pass vm parameterJohn Lo1-2/+2
I suppose most API handlers do not use vm parameter so it has not been a problem so far. Now vl_api_vnet_set_ip6_ethernet_neighbor() was crashing when called from api trace replay because either of ip_neighbor_add_del_t_handler() vnet_arp_set_ip4_over_ethernet() need vm to be correct when it calls vlin_time_now() to update the neighbor timestamp. Change-Id: Iffb2084a7c90f92c4b86b339ea11800dd41117eb Signed-off-by: John Lo <loj@cisco.com>
2018-06-26tcp: cleanup functionsFlorin Coras4-263/+258
- sprinkle statics for functions - move some inlines from header files to corresponding .c files - replace some always_inlines with statics where inlining is not performance critical Change-Id: I371dbf63431ce7e27e4ebbbdd844a9546a1f1849 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-26SIMD optimized linear search in clib_bitmap_first_setDamjan Marion1-2/+23
Change-Id: Ib3a55598a83cc99485b40e38e7c406ecb126fd42 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-06-25Add checks for bad ARP request packets in arp-termination nodeJohn Lo1-12/+9
Check ARP request packet with bad requester MAC/IP address and drop these packets. Also removed useless VRRP check in the node. Change-Id: I2c8a774d291928eb623b3a515f1edf7e338fa760 Signed-off-by: John Lo <loj@cisco.com>
2018-06-25tw: add light weight timer update functionFlorin Coras6-41/+183
Because it avoids pool putting/getting the timer, this function is somewhat faster than stopping and restarting a timer. Change-Id: Id99ed9d356b0b1f7e12facfe8da193e1cd30b3ec Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-25dpdk: Enhancement to call crypto start api at initializationSachin Saxena1-0/+4
- Some crypto devices rely on rte_cryptodev_start() API to be called by application to enable a pre-configured H/W Crypto device. - NXP dpaa2 is one of the example. Change-Id: I2ad8ca0060604fb4e0541161e91bdebc6642f4da Signed-off-by: Sachin Saxena <sachin.saxena@nxp.com>
2018-06-25acl-plugin: remove the noisy debug outputAndrew Yourtchenko1-1/+1
Change-Id: I6a3cfcb24f5027ec0f2cd2ec21ea47a01fef331b Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2018-06-25MAP: Move MAP-E/T to a plugin.Ole Troan26-469/+675
Only remaining traces of MAP in the src/vnet is now in buffer.h. Awaiting a new buffer opaque API (hint, hint). Change-Id: Ie165561484731f1d7ed6e0f604b43624e06db3f0 Signed-off-by: Ole Troan <ot@cisco.com>
2018-06-25VCL: Refactor VCL test (part 1)Dave Wallace6-338/+2171
Change-Id: I6a326e24ed953b1cef63bd4010a3bedd6c4a7b1c Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
2018-06-24Revert "Revert "ipsec: VPP-1316 calculate IP/TCP/UDP inner checksums""Klement Sekera6-195/+180
This reverts commit e0d2bd6bd7fc59c0c6ac48195d7f825dc99bfd91. Change-Id: If491e16f9ea66b2493a6a7c7f3c684ed585f8f51 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2018-06-23Tx feature node, for accounting purposesDave Barach3-136/+204
Switch to combined allow/drop counters Show matching ip4 neighbor address if known Add static-allow mactime entries for unknown mac addresses Add the "clear mactime" command Change-Id: Ib963981438dfb8a123df1b3c023bd5fcc27f888f Signed-off-by: Dave Barach <dbarach@cisco.com>
2018-06-22Revert "ipsec: VPP-1316 calculate IP/TCP/UDP inner checksums"Ole Troan6-180/+195
This reverts commit a98346f664aae148d26a8e158008b773d73db96f. Change-Id: Iee5b3a5ddff0e8fd3a30fe5973cee24de434fe12 Signed-off-by: Ole Troan <ot@cisco.com>
2018-06-22jvpp: add support for enums (VPP-1153)Marek Gradzki5-4/+244
Change-Id: I2a1b946a71419e1fb3c5d70567c54a6a7d841f10 Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
2018-06-22session: improve session lookup speedsFlorin Coras1-57/+24
It would seem that this avoids using the stack for computing the subsequent crc32 by storing the 2 x u64 ip4 5-tuple key into two registers. Probably the cast to the "convenience" data structure "masks" the fact that the key is 16B. Probably ... :-) "sh run" on an ip4 tcp sender reports that the number of clocks spent in tcp_input, where a session lookup is performed, drops ~25% (from ~100 clocks/packet to ~75 clocks/packet) Change-Id: I50d647115ac72f8407bff7e3b73328ba42f0ddb5 Signed-off-by: Florin Coras <fcoras@cisco.com>