Age | Commit message (Collapse) | Author | Files | Lines |
|
- hash is great. But it is a bit too slow for the DP. Use direct array indexing
to quickly retrieve the slave interface.
- the algorithm used by flow hash is great. But it is a bit too slow for the DP.
Use l2_hash_hash() extracted from lb_hash.h which ECMP is using. It makes use
of intrinsic crc32 instruction set.
- shortcut modulo arithmetic when the operand is 2**x (where x up to 4) to
avoid division instruction.
- special case for link count == 1 in bond_tx_fn()
- use clib_mem_unaligned to access data for the packet to avoid alignment error
- Fix some typos for packet tracing.
Change-Id: I8eae3ad497061c5473aa675ba894ee0211120d25
Signed-off-by: Steven <sluong@cisco.com>
(cherry picked from commit 0d88301a576191a0e330e539cf1dcb3837ee1bf6)
|
|
Object sizes must evenly divide alignment requests, or vice
versa. Otherwise, only the first object will be aligned as
requested.
Three choices: add CLIB_CACHE_LINE_ALIGN_MARK(align_me) at
the end of structures, manually pad to an even divisor or multiple of
the alignment request, or use plain vectors/pools.
static assert for enforcement.
Change-Id: I41aa6ff1a58267301d32aaf4b9cd24678ac1c147
Signed-off-by: Dave Barach <dbarach@cisco.com>
|
|
Change-Id: Ide8bf41e24a427643a3a17b1c9089993790c12a6
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Removed the sole use of it from ip6_neighbor.c
Change-Id: Ie53cb3b6a3a41ec0917ec2042e5006d0cfaefc01
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Use (u64) cast to ensure proper address calculations.
Change-Id: I6bad50010b140189f1b0af177e55da0045bd7a93
Signed-off-by: Lee Roberts <lee.roberts@hpe.com>
|
|
With glibc 2.27 the memfd_create has been added to the devel libraries.
That's causing the internally defined static function to clash with the
system wide one. This patch addresses that issue on systems with latest
glibc libraries.
Change-Id: I788bf49b23d5b5f1cb1c0374e243d8a429178a71
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
Change-Id: I6d049c0875b91f67f008dc04ae7efe2f8ddc276e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
platform. VPP-1129
Change-Id: I954acb56d901e42976e71534317f38d7c4359bcf
Signed-off-by: Adrian Oanca <adrian.oanca@enea.com>
|
|
Nintin
Change-Id: I88aabd34ef385d620695ac17ec3fe2f4a5177ada
Signed-off-by: Adrian Oanca <adrian.oanca@enea.com>
|
|
Change-Id: I2e9d01ccba5288e89b886464436097d3cb7d2d18
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Directly allocate and carve cache-line-aligned chunks of virtual
memory. To a first approximation, bihash wasn't using
clib_mem_free(...).
We eliminate mheap object header/trailers, which improves space
efficiency. We also eliminate the 4gb bihash table size limit. An 8_8
bihash w/ 100 million random entries uses 3.8 Gbytes.
Change-Id: Icf925fdf99bce7d6ac407ac4edd30560b8f04808
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This change only affects Aarch64 where previously we were using 128
bytes.
Change-Id: I52a3f2f3ff8c06abe8ae3933bc0d7a2a7749dd8a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I71fac34b071a07b4331cb5c900c3b8667c1fc114
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I5045e0f3ac4698e820b69ad46b96763e404e6fe4
Signed-off-by: Adrian Oanca <adrian.oanca@enea.com>
|
|
Change-Id: Ie9f611fa6a962b0937245f5cc949571ba11c5604
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Otherwise, in a "catch-up / multiple tick" case, the code will repeatedly
hand previously-processed expired timer handles to the user callback.
Change-Id: Idef4f242279ea41cb557bb6cff5984de02a6503d
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I6511110d0472203498a4f8741781eeeeb4f90844
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Also it removes ethernet_frame_is_any_taged implemebntation
which seems to be equally costly compared to two
invocations of ethernet_frame_is_tagged.
Change-Id: If1c95f8267cd34b807ec07e0d675cbd0db2fdf9f
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I9710be2e722d716e22d989b3417fb49d2db0848a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I3908cc112b40d4bb52da18e7c3ac5ae0af455f87
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Group Base Policy (GBP) defines:
- endpoints: typically a VM or container that is connected to the
virtual switch/router (i.e. to VPP)
- endpoint-group: (EPG) a collection of endpoints
- policy: rules determining which traffic can pass between EPGs a.k.a
a 'contract'
Here, policy is implemented via an ACL.
EPG classification for transit packets is determined by:
- source EPG: from the packet's input interface
- destination EPG: from the packet's destination IP address.
Change-Id: I7b983844826b5fc3d49e21353ebda9df9b224e25
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
Enable CLIB_HAVE_VEC128 if both aarch64 and __ARM_NEON
ie. armv8 only, not armv7
Add more neon compare intrinsics wrappers.
I only add simple intrinsics wrappers. More complex ones can be added
later as they are needed, with performance tests on the corresponding
feature to back them up.
Remove wrongly added 128bits definitions defined on both armv7 and armv8
without concern for NEON instructions presence.
Notable correspondinf code activations:
* MHEAP_FLAG_SMALL_OBJECT_CACHE in mheap.c
* ip4 fib mtrie leaves access
* enable ixge plugin compilation for aarch64
(conf still disables it by default)
Change-Id: I99953823627bdff6f222d232c78aa7b655aaf77a
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
|
|
Reference-count the number of entries in each bucket. If the reference
count goes to zero, free the backing store.
Add long-term churn-testing to test_bihash_template.c, thanks to
Andrew Yourtchenko for the initial implementation.
Change-Id: I4fbd9229cacfaba8027a85cbf87b74afdead6e39
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This patch teaches worer threads to sleep and to be waken up by
kernel if there is activity on file desctiptors assigned to that thread.
It also adds counters to epoll file descriptors and new
debug cli 'show unix file'.
Change-Id: Iaf67869f4aa88ff5b0a08982e1c08474013107c4
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
For some files such as hugepages files, ftruncate() fails with the error
"Invalid argument" if the 'length' parameter is not on a page boundary.
Change-Id: I42a9cde98707da15e3c5d1653046e2277fc7a424
Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
|
|
- use valloc as a 'central' segment baseva manager
- use per segment manager segment pools and use rwlocks to guard them
- add session test that exercises segment creation
- embed segment manager properties into application since they're shared
- fix rw locks
Change-Id: I761164c147275d9e8a926f1eda395e090d231f9a
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
This hash table intends to provide an alternative to the widely
used bihash table in places where either:
- Hash entry timeout is required
- The hash table data does not fit in CPU cache
Although the bihash table is very fast, each lookup requires
accessing two cache lines in a serialized fashion. It works fine
when the hash table is in cache, but hits a wall when it does not.
The 'flowhash' table uses a simplified design (at the cost of a
less good bucket auto-scaling) where each access only requires
a single memory lookup (in the absence of collision). The hash
table also uses a reduced number of registers.
In practice, a VPP node implementing a stateful feature would
typically:
- prefetch buffer metadata (in-cache)
- prefetch packet header (in-cache)
- compute hash & prefetch hash bucket (possibly in RAM)
- read/write key and value from bucket
Using this hash table, it is possible to pipeline accesses in a way
that does not exhaust CPU's line field buffers, even when the
requested value is located in RAM (i.e. not in cache).
Measurements showed it was possible to scale to tens of millions
of flows (with a full 5-tuple matching and 32B value, i.e. 1
cache line per flow) with no performance degradation when
the hash table grows to the point it doesn't fit in cache anymore.
I have used this table in a couple of non-open-sourced projects,
but think it might be useful to lb, nat, and possibly other VPP
subsystems.
More information in the .h file.
Change-Id: I2b13dde0eabd868b75da1cedbfca0bf74d705102
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Change-Id: Ibc252d9ed595be955790ec1c97d8730e43ad89b2
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Add some description and cleanup code that uses Arm system counter.
Change-Id: Ie1fe00e3e4b5d98867617b7b0184ac526e333c53
Signed-off-by: Brian Brooks <brian.brooks@arm.com>
|
|
Change-Id: I75e6c7d1a6ff1fcebc81ec10bd86b79f2bf3dc22
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I606fd89c410369cbd9ce9dcaaaa9dc58796e7c0e
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
- update segment manager and session api to work with both flavors of
ssvm segments
- added generic ssvm slave/master init and del functions
- cleanup/refactor tcp_echo
- fixed uses of svm fifo pool as vector
Change-Id: Ieee8b163faa407da6e77e657a2322de213a9d2a0
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
when verbose option is used
Change-Id: Ib63ead4525332f897b8a1d8a4cf5a0eb1da1e7f3
Signed-off-by: Vijayabhaskar Katamreddy <vkatamre@cisco.com>
|
|
On deleting sub-interfaces, functions vnet_delete_sub_interface()
and vnet_delete_hw_interface() are not cleaning up sub-interface
related hash tables and memory properly.
Change-Id: I17c7c4b2078c062c77bfe48889beb677610035ca
Signed-off-by: John Lo <loj@cisco.com>
(cherry picked from commit 7f5bec647c9dc743c015d461d040e63a77fd0a08)
|
|
Change-Id: I67648dbed3c7ed291b3e1ce617d83a776d3623bb
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I056598a1818a39c2da73e252600c14585e5aae83
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
- separate client/server code for both memory and socket apis
- separate memory api code from generic vlib api code
- move unix_shared_memory_fifo to svm and rename to svm_fifo_t
- overall declutter
Change-Id: I90cdd98ff74d0787d58825b914b0f1eafcfa4dc2
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Combine implementer, part, variant, and revision into one cpu
description line.
For example : ARM (Cortex-A57 PASS 1.2)
* get infos from /proc/cpuinfo
* only recognize armv8 processors
* add all given cavium processors
* Cavium starts counting variants from 1 instead of 0
Change-Id: I4f3820fb13a6bd2a0dc59e28fbe6f48a5b0ceb25
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
|
|
- add function to sock client that bootstraps shm api
- allow sock clients to request custom shm ring configs
Change-Id: Iabc1dd4f0dc8bbf8ba24de37f4966339fcf86107
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Move the functions hash_set_key_copy() and hash_unset_key_free()
which are dupilicated in various tunnel support code modules to
hash.h as hash_set_mem_alloc() and hash_unset_mem_free() to be
used by all.
Change-Id: I40723cabe29072ab7feb1804c221f28606d8e4fe
Signed-off-by: John Lo <loj@cisco.com>
|
|
This allows arm platforms to also take advantage of crc32 hardware
acceleration.
* add a wrapper for crc32_u64. It's the only one really used. Using it
instead of a call to clib_crc32c() eases building symmetrical hash
functions.
* replace #ifdef on SSE4 by a test on clib_crc32c_uses_intrinsics.
Note: keep the test on i386
* fix typo in lb test log
Change-Id: I03a0897b70f6c1717e6901d93cf0fe024d5facb5
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
|
|
Change-Id: Ic9c1c70e06b953538ed43fc91ed26b6be82ce812
Signed-off-by: Kevin Wang <kevin.wang@arm.com>
|
|
Taken from DPDK, also AVX2 variant updated to be in sync with DPDK
version.
Change-Id: I8a42e4141a5a1a8cfbee328b07bd0c9b38a9eb05
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
"This time, for sure..."
Change-Id: Ie981003842d37c5eb6a0b2fe3abe974a93b86df8
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Ic551af286aa84293deb260560c12def430449598
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I0a0f8c9aad1530d18c70c962e729e84948a074ee
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
It looks like different compiler versions produce different results for
expressions like "(cast) ptr + inc".
Use parenthesis to avoid such issues.
Change-Id: I93a9883bf5fc05ae462df5b004817775f0739405
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
Change-Id: I63d720378b92813993525f80fee90fc79df27fba
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Move elog_sample.c to src/examples/vlib
Change-Id: I7d32c83c424b9ca4a057372c7fc6a6e2b7dab034
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
clib_maplog_process(...): handle logs which weren't closed properly.
It will happen.
Change-Id: Ibcf9c9ea7a09991e6294050e7d2979a0d3f965cf
Signed-off-by: Dave Barach <dave@barachs.net>
|