summaryrefslogtreecommitdiffstats
path: root/src/vppinfra
AgeCommit message (Collapse)AuthorFilesLines
2021-04-21buffers: vlib_get_buffers() with 512-bit SIMDDamjan Marion1-0/+1
Type: improvement Change-Id: Id8ce3ffc1299a38171b82a7082454412c840a40c Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-21vppinfra: more avx512 inlines (compress, expand, from, is_equal_mask)Damjan Marion2-30/+80
Type: improvement Change-Id: I4cb86cafba92ae70cea160b9bf45f28a916ab6db Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-20pmalloc: tolerate move_pages failure on numa node #0Klement Sekera1-1/+2
Type: fix Change-Id: I65dbc4f3db47c745a0187e015845e07b3b0568e5 Signed-off-by: Klement Sekera <ksekera@cisco.com>
2021-04-19vppinfra: momve clib_sysfs_list_to_bitmap to sysfs.cDamjan Marion2-0/+29
Type: refactor Change-Id: Ia5a7ebd5c8492326f5bf2f1264f1c5ef38ccb4dc Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-18vppinfra: remove outdated and unused unix format functionsDamjan Marion2-486/+0
Type: refactor Change-Id: If1772980dbdbde1fbf3d1989daa40599e9f23e8c Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-18vppinfra: remove linux/syscall.hDamjan Marion9-136/+51
For portabiliy reasons it is better to have all wrapped in clib code. I.e. instead of using getcpu() we have clib_get_current_numa_node () and clib_get_current_cpu_id(). Type: refactor Change-Id: I29b52d7f29bc7f93873402c4070561f564b71c63 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-15vppinfra: move bitmap format functions to .c file, add format_bitmap_listDamjan Marion3-120/+172
Type: improvement Change-Id: I9baa845ecab8655e0623453666092d2dbc674b0f Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-04-15vppinfra: correct intrinsic called by u16x16_from_u8x16Lijian.Zhang1-2/+2
u16x16_from_u8x16() and i16x16_from_i8x16() call intrisics _mm256_cvtepu8_epi64 and _mm256_cvtepi8_epi64. But they are not seems doing the right data conversion from the name of the wrappers. The correct intrinsics been called should be _mm256_cvtepu8_epi16 and _mm256_cvtepi8_epi16. Type: fix Change-Id: Id71de6ae1a266a370f11c33a46684202be766c43 Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
2021-03-25tests: move vppinfra tests to src/vppinfra/testDave Wallace1-0/+40
- Refactor make test code to be co-located with the vpp feature source code Type: test Signed-off-by: Dave Wallace <dwallacelf@gmail.com> Change-Id: Iceaeab769c08e541b31546663d0e402ced862e5b
2021-03-11vlib: refactor node function variantsDamjan Marion1-19/+22
It allows default variant selection from startup.conf Type: improvement Change-Id: Idff95e12dd0c105dab7c905089548b05a6e974e0 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-07dispatch-trace: move dispatch trace pcap code to pluginDamjan Marion1-39/+0
Type: refactor Change-Id: I02a527f57853ebff797f0d85761b71127916d6ce Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-03-04vppinfra: fix compiling error due to incompatible udphdr field namesJieqiang Wang1-6/+3
Compiling VPP on CentOS 7 will fail shown as below. The root cause is that uh_sport/uh_dport field names for struct udphdr are chosen only if macro __FAVOR_BSD in /usr/include/netinet/udp.h is defined for glibc version less than 2.19. Fix this issue by using source and dest field names in struct udphdr for compatibility reasons. FAILED: vppinfra/CMakeFiles/vppinfra.dir/unix-formats.c.o ccache /opt/rh/devtoolset-9/root/bin/cc -Dvppinfra_EXPORTS -I/vpp/src -I. -Iinclude -Wno-address-of-packed-member -g -fPIC -Werror -Wall -march=corei7 -mtune=corei7-avx -O2 -fstack-protector -D_FORTIFY_SOURCE=2 -fno-common -flto -fno-fat-lto-objects -fPIC -fvisibility=hidden -ffunction-sections -fdata-sections -MD -MT vppinfra/CMakeFiles/vppinfra.dir/unix-formats.c.o -MF vppinfra/CMakeFiles/vppinfra.dir/unix-formats.c.o.d -o vppinfra/CMakeFiles/vppinfra.dir/unix-formats.c.o -c /vpp/src/vppinfra/unix-formats.c /vpp/src/vppinfra/unix-formats.c: In function 'format_udp4_packet': /vpp/src/vppinfra/unix-formats.c:319:19: error: 'struct udphdr' has no member named 'uh_sport' 319 | u16 source = udp->uh_sport; | ^~ /vpp/src/vppinfra/unix-formats.c:320:17: error: 'struct udphdr' has no member named 'uh_dport' 320 | u16 dest = udp->uh_dport; Type: fix Change-Id: Ifc99c7286ea3fac463096152267033ac0518c230 Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com> Reviewed-by: Lijian Zhang <lijian.zhang@arm.com> Reviewed-by: Tianyu Li <tianyu.li@arm.com>
2021-02-26vppinfra: mem leak in show memory main-heapFlorin Coras1-1/+5
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I23d1dda86c781ac077dbee7cb0e1ddeaa328c660
2021-02-19session vppinfra: asan fixesFlorin Coras1-4/+16
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Ie709d76438542783cbc8c6174b5e712ef18a6276
2021-02-16cjson: upgrade to new versionOle Troan1-14/+33
See if this fixes the coverity issues. Now at 324a6ac9a9b285ff7a5a3e5b2071e3624b94f2db Type: fix Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: I2cd281ebaeda69e214e6dc93a84888298741d0ee Signed-off-by: Ole Troan <ot@cisco.com>
2021-02-15vppinfra: fix memcpy undefined behaviourBenoît Ganne6-80/+123
Calling mem{cpy,move} with NULL pointers results in undefined behaviour. This in turns is exploited by GCC. For example, the sequence: memcpy (dst, src, n); if (!src) return; src[0] = 0xcafe; will be optimized as memcpy (dst, src, n); src[0] = 0xcafe; IOW the test for NULL is gone. vec_*() functions sometime call memcpy with NULL pointers and 0 length, triggering this optimization. For example, the sequence: vec_append(v1, v2); len = vec_len(v2); will crash if v2 is NULL, because the test for NULL pointer in vec_len() has been optimized out. This commit fixes occurrences of such undefined behaviour, and also introduces a memcpy wrapper to catch those in debug mode. Type: fix Change-Id: I175e2dd726a883f97cf7de3b15f66d4b237ddefd Signed-off-by: Benoît Ganne <bganne@cisco.com>
2021-02-11vppinfra: fix membulk coverity warningsFlorin Coras1-2/+2
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I4830656ad852de62211e31574a0e0afb62d00f37
2021-02-10vppinfra: fix bihash test-debug aarch64 sporadic failureDave Barach1-0/+3
Turns out that both the multi-core and single-core add / del tests run for just under 5 seconds (aarch64, debug image). Increase the vapi => debug CLI timeout to avoid spurious failures. Type: test Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: Ia64bffa82a908172a21c34f3ae4402ad774af28a
2021-02-08virtio: add atomic call for kickingMohsin Kazmi1-0/+3
Type: fix Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com> Change-Id: I41faa2ca249ff75e564a732af896e6b5d76bf665
2021-02-04vppinfra: fix mem map mt issue due to snprintfFlorin Coras1-1/+1
Type: fix Apparently snprintf is not thread safe!? Thread 6 "vpp_wk_3" received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7ffeed1a3700 (LWP 819375)] 0x00007ffff58fc2b4 in __GI__IO_default_xsputn (n=<optimized out>, data=<optimized out>, f=<optimized out>) at genops.c:394 394 genops.c: No such file or directory. (gdb) bt 0 0x00007ffff58fc2b4 in __GI__IO_default_xsputn (n=<optimized out>, data=<optimized out>, f=<optimized out>) at genops.c:394 1 __GI__IO_default_xsputn (f=0x7ffeec35e050, data=<optimized out>, n=9) at genops.c:370 2 0x00007ffff58e127c in __vfprintf_internal (s=s@entry=0x7ffeec35e050, format=format@entry=0x7ffff5f833ce "%s", ap=ap@entry=0x7ffeec35e1d0, mode_flags=mode_flags@entry=2) at ../libio/libioP.h:948 3 0x00007ffff58f611a in __vsnprintf_internal (string=0x7ffed57ed018 "", maxlen=<optimized out>, format=0x7ffff5f833ce "%s", args=args@entry=0x7ffeec35e1d0, mode_flags=2) at vsnprintf.c:114 4 0x00007ffff5997fd1 in ___snprintf_chk (s=<optimized out>, maxlen=<optimized out>, flag=<optimized out>, slen=<optimized out>, format=<optimized out>) at snprintf_chk.c:38 5 0x00007ffff5f7fd74 in clib_mem_vm_map_internal (base=<optimized out>, log2_page_sz=<optimized out>, size=<optimized out>, fd=<optimized out>, offset=<optimized out>, name=0x0) at /scratch/fcoras/vpp/src/vppinfra/linux/mem.c:502 Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I8ff7cfed422273dbdc55068772e11f5e550a9b60
2021-02-04vlib: "revert startup multi-arch variant configuration fix for interfaces"Damjan Marion1-20/+0
Type: fix This reverts commit 5a48b3b9d88fa2793793e2bf3db8bf156fe2951f. Change-Id: Ifa91b18bdbbc32bb729abc09d95637d9cdf42c3b Signed-off-by: Damjan Marion <damarion@cisco.com> Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
2021-02-01vppinfra: add defines for bihash add and deletePaul Atkins2-1/+4
When calling the bihash_add_del... functions some callers add a comment beside the value to indicate this is the is_add param. Make the code easier to read by adding defines for add and delete that the callers can use instead of having to use 0 or 1. Type: improvement Signed-off-by: Paul Atkins <patkins@graphiant.com> Change-Id: Iab5f7c8e8df12ac62fc7e726ca1798622dcdb42c
2021-01-25vlib: startup multi-arch variant configuration fix for interfacesRadu Nicolau1-0/+20
Propagate the multi-arch variant selection to interfaces. Type: fix Signed-off-by: Radu Nicolau <radu.nicolau@intel.com> Change-Id: I99c4a7896f172f0d14d2ded22a27383825529a7d
2021-01-21avf: use write combining store for queues tail updateRadu Nicolau1-23/+32
Performance improvement: on supported platforms, currently only Intel Tremont, use a write combining store to update the tail pointers. Also, Tremont node variant is added for all. Type: improvement Signed-off-by: Radu Nicolau <radu.nicolau@intel.com> Change-Id: Ie9606e403b7d9655184f778e3ffee3027c8c9edd
2021-01-21interface: rx queue infra rework, part oneDamjan Marion3-0/+236
Type: improvement Change-Id: I4008cadfd5141f921afbdc09a3ebcd1dcf88eb29 Signed-off-by: Damjan Marion <damarion@cisco.com>
2021-01-21vppinfra: ensure _GNU_SOURCE is presentNathan Moos2-6/+13
Type: fix When building with musl, _GNU_SOURCE is required to be defined in more places. This patch defines it in those required places. Additionally, this patch changes the udphdr field names to the standard names. Change-Id: I0fec54214ccc8458289176488afb4218761e9d51 Signed-off-by: Nathan Moos <nmoos@cisco.com>
2021-01-21vppinfra: add bihash with 32 byte keyMatthew Smith2-0/+108
Type: feature Add header to define bihash_32_8. Change-Id: Ia8a21a1e76cf0c05b8f8a2d1f5dbc0dc586d7801 Signed-off-by: Matthew Smith <mgsmith@netgate.com>
2021-01-20vppinfra: add option to use libexecinfoNathan Moos1-1/+5
Type: feature The musl libc does not provide <execinfo.h> or an implementation of the functions provided. This patch enables the use of libexecinfo as an alternative, which can be used on systems with musl. Change-Id: I76b5744ddf731b66e16144025a6fdedf72925260 Signed-off-by: Nathan Moos <nmoos@cisco.com>
2021-01-15vlib: fix counter_will_expand predictionMiklos Tirpak1-0/+15
vlib_validate_combined_counter_will_expand() was calling _vec_resize_will_expand() with wrong arguments, which resulted in false return value. Apart from the initial call, it never indicated a vector resize. The callers relying on this function did not perform a barrier sync because of the wrong prediction even if the vector got extended by a subsequent vlib_validate_combined_counter() call. The fix introduces a new, simplified macro that is easier to call. vec_resize_will_expand() accepts the same arguments as vec_resize(). Type: fix Signed-off-by: Miklos Tirpak <miklos.tirpak@gmail.com> Change-Id: Ib2c2c8afd3e665e0e3d6ae62ff5cfa287acf670f
2020-12-23vppinfra: mem bulk testFlorin Coras1-1/+1
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Icd44ede9604c29839af250a2be93ecf467467aa0
2020-12-23svm: remove fifo segment heapFlorin Coras1-0/+3
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I518e096fe13847759806ff62009e73fd8f7451b7
2020-12-22vppinfra: add bulk memory allocatorDamjan Marion3-0/+249
Small fixed size object memory allocator. Type: improvement Change-Id: I727705d9d4292b6b38d41e239871103b15aa9038 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-12-18misc: deprecate old perfmonDamjan Marion2-47/+0
Type: refactor Change-Id: I1303219f9f2a25d821737665903b0264edd3de32 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-12-16vppinfra: fix vec_max_lenFlorin Coras1-1/+2
Return 0 if the supplied vector pointer is NULL. Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Icb0aab70e0b7c9c1ddcf607c9dfb7e5715f177d1
2020-12-14misc: refactor clib_bitmap_foreach macroDamjan Marion3-9/+15
Type: refactor Change-Id: I077110e1a422722e20aa546a6f3224c06ab0cde5 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-12-14misc: move to new pool_foreach macrosDamjan Marion6-70/+70
Type: refactor Change-Id: Ie67dc579e88132ddb1ee4a34cb69f96920101772 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-12-14vppinfra: simpler and faster pool_foreachDamjan Marion1-15/+31
- reduces number of instructions generated 4 times compared to old code - adds pool_foreach2 which is more friendly to clang-format Type: improvement Change-Id: I51e9c7fb09655c60d883987dadf5b2666c12b3f7 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-12-01session: allow small pacer burstsFlorin Coras1-0/+8
Instead of enforcing a "strict" release of data, which relies on frequent rescheduling of sessions, allow some pacer coalescing, i.e., short bursts, that can minimize load on scheduler/session layer and potentially leverage tso. Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I67e38e5b8dc335bd214113b70c68c27ae92bd6da
2020-11-25vlib: add elog events for vlib log entriesDave Barach2-6/+18
Pretty interesting to see how long plugins take to load. Type: improvement Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: If496eb73b2e1fd9dd714142cd5195fa7ccd93ebd
2020-11-25api: vat2 and json autogeneration for api messagesOle Troan3-0/+3395
VAT2: A completely auto-generated replacement of VAT. Reads input message in JSON from stdin and outputs received messages in JSON. A VAT2 plugin is automatically built for a .api file. There no longer a need for a separate _test.c. Example: vat2 show_version {} { "_msgname": "show_version_reply", "retval": 0, "program": "vpe", "version": "21.01-rc0~411-gf6eb348a6", "build_date": "2020-11-19T09:49:25", "build_directory": "/vpp/autogen3" } vat2 sw_interface_dump '{"sw_if_index": -1, "name_filter_valid": 0, "name_filter": ""}' [{ "_msgname": "sw_interface_details", "sw_if_index": 0, "sup_sw_if_index": 0, "l2_address": "00:00:00:00:00:00", "flags": "Invalid ENUM", "type": "IF_API_TYPE_HARDWARE", "link_duplex": "LINK_DUPLEX_API_UNKNOWN", "link_speed": 0, "link_mtu": 0, "mtu": [0, 0, 0, 0], "sub_id": 0, "sub_number_of_tags": 0, "sub_outer_vlan_id": 0, "sub_inner_vlan_id": 0, "sub_if_flags": "Invalid ENUM", "vtr_op": 0, "vtr_push_dot1q": 0, "vtr_tag1": 0, "vtr_tag2": 0, "outer_tag": 0, "b_dmac": "00:00:00:00:00:00", "b_smac": "00:00:00:00:00:00", "b_vlanid": 0, "i_sid": 0, "interface_name": "local0", "interface_dev_type": "local", "tag": "" }] This is the first phase and vat2 is not integrated in packaging yet. Type: feature Signed-off-by: Ole Troan <ot@cisco.com> Change-Id: Ib45ddeafb180ea7da8c5dc274a9274d7a4edc876 Signed-off-by: Ole Troan <ot@cisco.com>
2020-11-24buffers: add page-size configNathan Skrzypczak1-31/+18
Type: feature Add a `buffers {page-size}` parameter to specify page size for buffers. This also fixes an issue with the parsing in unformat_log2_page_size. Change-Id: I7d7b1fa0bb7febaa7509cf2c625882f07eeafaad Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
2020-11-19svm: move chunk locks to linked listFlorin Coras1-0/+3
Type: improvement We only need to protect the linked lists. Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Ie1542073f3993acfc66d99096b08bf9ecd10a49b
2020-11-16tests: move bihash tests to src/vppinfra/testDave Wallace1-0/+71
- Refactor make test code to be co-located with the vpp feature source code. Type: test Signed-off-by: Dave Wallace <dwallacelf@gmail.com> Change-Id: I0133a00def595fe1c08ab881633ef9f89148a286
2020-11-10vpp: use vpp heap for libcBenoît Ganne1-5/+7
This makes libc use vpp main heap instead of the default libc heap. This gives better visibility (accounting, tracing) on allocations happening in external libraries called from vpp (eg. OpenSSL). Type: feature Change-Id: I5d8a673472145a4e090bedb443b8c58a967d1cca Signed-off-by: Benoît Ganne <bganne@cisco.com>
2020-11-05vppinfra: mspace_get_aligned should respect MALLOC_ALIGNMENTDamjan Marion1-22/+1
While underlying mspace_malloc() respects MALLOC_ALIGNMENT (16) mspace_get_aligned moves start of data for 4 bytes making final result not aligned to MALLOC_ALIGNED. Type: fix Change-Id: Ib1584f315e3658b3b748ebf4886fb7389ce640fe Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-11-05build: link time optimization for release buildsDamjan Marion1-0/+1
Type: improvement Change-Id: I0e24f1d2ad5c7a11a8bd40125428f86aca867bec Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-10-26vppinfra: clib_mem_vm_{un}map thread safeDamjan Marion2-0/+25
Type: improvement Change-Id: I1ab1b100000b4d7212c58e10312e16e7527bd333 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-10-23vppinfra: add bihash_8/24_16 to headers listjiangxiaoming1-0/+2
Type: improvement Signed-off-by: jiangxiaoming <jiangxiaoming@outlook.com> Change-Id: Ic67977995dbcba97240009373a4b9e9b57c63706
2020-10-22vppinfra: missing __clib_export va_unformatJuraj Linkeš1-1/+1
Fix an issue observed on aarch64 systems when running with dpdk plugin: dpdk_plugin.so: undefined symbol: va_unformat Type: fix Fixes: dae1c7ed38e9f7a86e66f3674de62778cc981331 Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech> Change-Id: I22b7ee842dc606f96effdeb7aa4e161496986aed
2020-10-21vppinfra: missing __clib_exportNathan Skrzypczak1-1/+1
Type: fix Change-Id: I2dbd435d04a09f632286e5d669edc5fafd384f94 Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>