For AES-CBC, the IV must be unpredictable (see NIST SP800-38a Appendix
C). Chaining IVs like is done by ipsecmb and native backends for the
VNET_CRYPTO_OP_FLAG_INIT_IV is fully predictable.
Encrypt a counter as part of the message, making the (predictable)
counter-generated IV unpredictable.
Fixes: VPP-2037
Type: fix
Change-Id: If4f192d62bf97dda553e7573331c75efa11822ae
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Fix an issue where multiple VPP instances with DPDK starting at the
same time would not initialize VFs properly. This is done by using the
iavf PMD (where the issue can't be reproduced) instead of the i40evf
PMD.
Type: fix
Ticket: VPP-1943
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
Change-Id: Idcc48088c7d66a76da2b4675c02c7c115706c8b3
Type: test
Ticket: VPP-1903
Change-Id: I7fab6931833d6e253b7b921172825387302d8f70
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
(cherry picked from commit 027d813a50)
- In a new centos-8 installation, vpp-ext-deps fails on missing
ssl.h header file after 'make install-deps'.
Type: fix
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: I521d817dd1f1e21aff427d98b9832ea7c7b89339
The list of plugins is outdated.
This change introduces a dynamically
generated list of the plugins along with their descriptions,
extracted directly from the sources.
Type: docs
Change-Id: Icb7b65e6b45289e257d71a1c18d10f62ced59cbe
Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
(cherry picked from commit 630ca994e0)
Fix in nat44 hairpinning code to check if anything was actually
changed in the snat_hairpinning() routine, and return 0 if nothing
changed. This helps avoid an infinite loop repeating the three
nodes nat44-hairpinning-->ip4-lookup-->ip4-local in case there
was no change. Also add a corresponding test case.
This is essentially a cherry-pick of change 30284 but the
automatic cherry-picking did not work because of some filename
changes.
Type: fix
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
Change-Id: I21a59ae7423f40abeff9fc0411330da58b3011f0
A simple counter is a two dimensional array by threads and
counter index. 28017 introduced an error missing the first
dimension.
If a vector is updated at the same time as a client reads,
an invalid pointer my result. This will be caught by the
optimistic locking after copying out the data, but if
following a pointer outside of the stat segment then
the stat client would crash. Add suitable boundary checks
for access to stat memory segment.
Fixes: 7d29e320fb
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: I94f124ec71d98218c4eda5d124ac5594743d93d6
(cherry picked from commit 65c56c83ce)
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
IPSec punting to IKEv2 is valid only for NAT-T in IPv4.
Fix coverity CID 214915.
Type: fix
Change-Id: I6f2db38abf179565316f50c5d47c78acce3a0d01
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit d9ed0b6786)
This helps to resolve sporadic failures in unit tests.
Type: fix
Change-Id: I3abd77ed74310f9729a841e8569eafe6d7758dcb
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
(cherry picked from commit 761f8f0eaa)
Type: fix
Ticket: VPP-1900
This fixes a crash when initiating IKE connection using interface
without any IP address.
It also ensures that the IKE connection is automatically retried once the
interface obtains an address.
Signed-off-by: jan_cavojsky <Jan.Cavojsky@pantheon.tech>
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Change-Id: Ia1919c349e64b3a0a4198365e075e177e3ba3de5
(cherry picked from commit 6960da5284)
When looking for existing SA connection to clean up search all per
thread data, not only current one.
Type: fix
Change-Id: I59312e08a07ca1f474b6389999e59320c5128e7d
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
(cherry picked from commit e7c8396982)
Using vlib_main of another thread is prohibited.
Type: fix
Change-Id: I7ae294dfaf2526738e91408c9b4865ef9f801b8a
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 7483a7272d)
We should not use main thread vlib_main in workers.
Type: fix
Change-Id: I58c0a8cadf2dc7f768b20ac90e7ec7921e2e8ca4
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 0a2fde105a)
dns cache should no longer be unlocked by caller.
Type: fix
Fixes: 84a563ae40
Change-Id: I3708718ae8f00e4e4f4e04381caa0095c8494b82
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 2113c7f28b)
crypto_native/main.h is being built as default,
and crypto_native_main is initialized with a size of 64 bytes.
crypto_native/aes_gcm.c and crypto_native/aes_cbc.c are march
variants, their ICL variants are expecting crypto_native_main
to be 256 bytes.
Type: fix
Signed-off-by: Georgii Tkachuk <georgii.tkachuk@intel.com>
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Change-Id: I4cddb75b712ea83c9cfca621887605d7bae104ec
When removing an interface we must reset all per-interface per-feature
arc data to ensure we do not get wrong feature arc config data when the
sw_if_index is recycled.
Type: fix
Change-Id: I8c9d850d7c62b7b77193da4258ab5fb9bdda85a6
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 6178bdafa6)
NUMA node parsing with vlib_get_thread_core_numa() can failed on single
socket systems. Use clib_get_current_numa_node() instead as we already
pinned the main thread to the requested core.
Type: fix
Change-Id: I22339516d0305689a58584c92ded7c96eb53be39
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 33ce5e568f)
ipsec_{crypto,integ}_alg_t are packed and smaller than u32. Callers are
using those enums so unformat functions should too instead of u32 to
not overflow the stack.
Type: fix
Change-Id: Ifc86366f1928ca6352f06f390a88ac64668289d5
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit f6422ffbc8)
- fix fifo initialization overflowing chunk size
- stick to the default base virtual address to initialize fifo. ASAN
can be picky about address space
Type: fix
Change-Id: If9a29138d2c207859d72845e928290c808c4a982
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 07b9455808)
Chunks can be allocated from another process. We need to manually
mark them as accessible for ASAN.
Type: fix
Change-Id: Ifbeef3346e9cee2c1231f80cbcf7f9673b5b54be
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit df601ae2d1)
This change mitigates software faults issued by some versions of the
linux kernel vfio-pci driver when VF PCI BARs are used without setting
the memory enable bit in the PCI configuration. This problem is
mentionned in https://lkml.org/lkml/2020/6/25/628
Change-Id: Idc177be4a5adb6ee467b4dd8f055f133ff267fe1
Type: improvement
Signed-off-by: Mohammed Hawari <mohammed@hawari.fr>
(cherry picked from commit 70fc36f268)
When GSO is enabled, vhost clears the checksum field prior to transmitting
the packet. Some newer kernel version does not like that and complains
about checksum error. This was seen with ip6 traffic.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I7c6f2a6148f4a30107bfa8b078f5990e64300cf1
(cherry picked from commit ac0f536388)
- Generation of patched original files (eg. dpdk-devbind.py.orig)
causes build breakage of 'make install-ext-deps' on CentOS-8.
Note: this only occurs on builds using 'make' for dpdk instead
of 'meson'. Thus only applies to stable/2009.
Type: fix
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: I3f1f70781b7a5564cd38e8876644d817e2eb4aad
- Installation vpp-plugin-core from packagecloud.io/master
& packagecloud.io/2009 breaks due to invalid dependencies
on newer versions of libmbedtls & libmbedcrypto
Type: fix
Fixes: 641467406
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: If736dabcc4a91a04b46515620dd87662b7b14260
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
- On Centos-8, 'make install-ext-deps' fails due to the
bpr-mangle-shebangs script failing the dpdk usertools
python scripts containing '#!/usr/bin/env python'.
Backport usertools section of the DPDK patch which
fixes this issue:
http://git.dpdk.org/dpdk/commit/?id=3f6f83626cf4967a99382a6518a614a1bf3d2c20
- Also fix README to reflect name change of dpdk master
branch to 'main'.
- On stable/2009, the dpdk build using make fails due to rpmbuild
defaulting to using /usr/bin/python (version 2.7) which
fails to build. The build on master uses meson which somehow
resolves this issue. For this cherry-pick, fix the build by
fixing the rpm vpp-ext-deps spec to force the use of python3.
Type: fix
Change-Id: I487b1ff2da786a4a3fd8fb0f859436b0e1885f1b
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
.. when associated profile is deleted.
Type: fix
Change-Id: Ib05831d79b3b58664ee0a930960513fd465373bf
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
(cherry picked from commit 6614df5350)
1. add libelf-dev to default deb deps
2. Also use libffi7 instead of libffi6 for debian-testing
Type: fix
Signed-off-by: Chuan Han <chuan.han.comm@gmail.com>
Change-Id: I9f13955812877422ecb8aac3dd34c5828b9c4607
The RPM build ends up with "vXX.YY" to vstring,
which is not what we'd expect - so fix it up.
Change-Id: I0af68e69b1e40fc49ade759bb2f0ed9f47614217
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
QUEUE_SELECT and QUEUE_NOTIFY_OFF registers are shared between all
workers operating on the same device and operations are not atomic
Type: fix
Change-Id: Ie017b1bfc7e3b6b4e59029f45db78eeffd9f3aeb
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
(cherry picked from commit 162a296756)
Initialize txq lock only if some txq are shared and check if another
worker is already operating on the txq before processing gro timeouts
in input node.
Type: fix
Change-Id: I89dab6c0e6eb6a7aa621fa1548b0a2c76e6c7581
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit b6b484d01a)
sanitizer complains about strlen on hi->name in tap_dump_ifs.
hi->name is a vector which is not null-terminated, so use vec_len.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: Icdd5f65369bb51b0c4a9cd86c24899e6febd837c
(cherry picked from commit 84f3d9fba4)
This reverts commit 73903d7e8a.
After discussion with Damjan, the change is more risky than it seemed,
so revert.
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Change-Id: Idee82806cd4a12a92540f904397d259b531770d1
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
This reverts commit 7a76017951.
After discussion with Damjan, the change is more risky than it seemed,
so revert.
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Change-Id: I67defb01a0bebb8ff4231c27690b694f608fcb8e
Thanks to Martin Sustrik for spotting the bug introduced by a316744
and submitting the fix.
Type: fix
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I4984fc32503b0c7b6db3543834dfbbfed2a1f23c
All timer and control plane functions happen from main thread
Type: fix
Change-Id: I4fc333c644485cd17e6f426493feef91688d9b24
Signed-off-by: Artem Glazychev <artem.glazychev@xored.com>
(cherry picked from commit 8eb69407c8)
If libbsd is detected by the DPDK build system, DPDK does not provide
implementations for strlcpy and dynamically link against the one
provided by libbsd. When the DPDK plugin is loaded by VPP, a crash
occurs because libbsd is not loaded by VPP.
Type: fix
Change-Id: Ib691bbe27edcf0f6f0a3d39952e439027cef72cb
Signed-off-by: Mohammed Hawari <mohammed@hawari.fr>
(cherry picked from commit 1f799bc3cb)
The format for deprecation is "option deprecated" now,
so harmonize the in-progress marking to logically be
"option in_progress"
At the same time recognize the legacy/erroneous
types of marking, print the warning.
Change-Id: If418dfadd69ffb112550164d63d13420e51cefd7
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit 6a3d4cc9a1)
Historically, there was only rc1/rc2 tags, fix in case we tag
rc3/rc4 during longer release windows.
Type: fix
Change-Id: I4243a1c03663a877f96a06d647e89adb74abd977
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit bc31247c92)
Type: fix
The GBP plugin that uses this module is also in-ptogress, hence so is
this module.
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: I3cb5dd124afac05da013d92d67b2abf6cdf9b769
(cherry picked from commit b468773aa4)
- VPP on opensuse has not been supported
for several releases.
Type: fix
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: I2b5316ad5c20a843b8936f4ceb473f932a5338d9
(cherry picked from commit bc35f469c8)
- The vpp build on the ubuntu-20.04 executor failed
due to the package 'dh-python' not getting
installed by 'make install-dep'
Type: fix
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: Id9307ad1b4e34c413d90258c6bde2aa5afafec63
(cherry picked from commit 5cc11b1210)
outbound:
wrong header len computation
gso size and header length need to be set in the first segment of the
chain
inbound:
EOP may have zero length descriptor to terminate the chain
missing endian conversion for ethertype
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Iaa003c0e9af3ead4df6c6c0d5772a179d2ff15c4
(cherry picked from commit 007abe751f)
in-progress messages do not give any API stability guarantees,
by design, to allow easy iteration. Provide an easy way
to know which messages are in-progress.
If as a user you see "in-progress" message that you want
to use in production, please contact the feature owner
and discuss the path to removing the "in-progress" status
before that.
Change-Id: I27729995e26a70af373e642b871dbb5cc5526959
Type: improvement
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit 8b0cd69d31)
Avoid situations where promisc mode is chaged while avf process
is suspended in the middle of adminq operation.
Type: fix
Change-Id: Ia1fc6551e83218b5938630ad3a15d4f3f0ceceff
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit 160a2a9a8c)
device pool my grow during suspemd which will cause crash in avf process
after it exits from suspend.
Type: fix
Change-Id: I51fec90088c909cfbaaca6c245272a28c0827ca0
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit 171d6aceb0)
The syntax of the deprecated flag has evolved.
Clean up usage to be "option deprecated;".
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: If2b639f275eb8db58b36c457f9245fe35a4d8cb1
(cherry picked from commit f916414b38)
Report if the messages were marked as deprecated,
but not yet deleted.
Useful for building the release notes and comparing
between the releases.
Also, put the dict_compare() call into the report(),
since latter always consumes the output of the former.
Change-Id: Iceab3e94ff66da931a4669b612026bd162dd5d1a
Type: improvement
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit 62bd50de97)
Type: fix
add UT for sneding handshale init and transport packets
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: Iab1ed8864c666d5a0ae0b2364a9ca4de3c8770dc
(cherry picked from commit d75a2d12c4)
VAT crashes when the one_add_del_local_eid api is invoked with mac
address as eid.
Type: fix
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: I29e246f6cad4b350fec52d54e94dbed586d488c4
(cherry picked from commit 5f473c0efc)
The lisp|one show eid-table command's help msg does not display the
available options. This patch fixes that.
show lisp eid-table [local|remote|eid <eid>]
show one eid-table [local|remote|eid <eid>]
Type: fix
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: Id39148db2ff291a7fe859830c1488b69ccd15c05
(cherry picked from commit b418c397dc)
In the vat help msg for one_add_del_l2_arp_entry the IP address option
is misspelled as "ip4 <ip4>" when it should have been "ip <ip4>".
Type: fix
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: Id4058a3ddfdb78b840d7e5a3c330e67b393f5d3b
(cherry picked from commit 8c6ba2957c)
The local|remote options to vat's lisp|one_eid_table_dump api command
does not print the eid details instead it produces the following error
messages:
Filter error, unknown filter: 1
Filter error, unknown filter: 2
Type: fix
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: I000c290b400dbf39bd883d57115923167092c9bd
(cherry picked from commit 2237cc8ce1)
It may happen that process node is suspended while it waits for response
from adminq and during that time CLI or API process can call
avf_delete_if. When avf process node resumes, it may happen that device
is not there anymeore.
This patch delegates interface deletion to process node, so CLI/API
process just sends signal instead of deleting device instance itself.
Type: fix
Change-Id: I7f12e12df3071650f6e60ad7eb5af23b7acfe335
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit 66bb7dd64e)
Multiple (> 1) workers leads to handoff node being enabled.
This node pops next feature index to nat.arc_next to make sure
that packet will be pushed to the next feature in the arc.
But node nat44-ed-in2out-output also pops next feature and changes
arc_next. So actual next feature will be skipped in that case.
It leads to all nat44-ed-in2out packets being dropped if we have
multiple workers (handoff node enabled).
To resolve this a new node was added (nat-pre-in2out-output) to fill
arc_next in single worker case and multiple worker case is already
handled by handoff node.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: I9dfba68f00164d2d5ab867224871811bef4411ed
(cherry picked from commit 8fb4d10dc2)
With packet trace on, VPP crashes when an arp packet arrives. This patch
fixes the crash and also ensures that the packet trace displays the eid
info.
Type: fix
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: Iaad09a5e2b33e931ab9bd7bc3d4573b5ed5e4bfd
(cherry picked from commit a3960a8b74)
Type: fix
Static analysis identified a possible null pointer dereference. It
was introduced by a recent patch which expanded the DMAC comparison
on inbound packets on a BVI interface to include any secondary MAC
addresses which were added to an interface.
Check if the pointer is null before dereferencing.
Change-Id: Ic2afe2b062eda32977e05bf3f98d82c1fe64620c
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
(cherry picked from commit 78681def21)
- make sure everything is freed on cleanup
- reuse already allocated vectors where possible
Type: fix
Change-Id: Ibd8da1edb37126522dc2d525596521d32dceb73a
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 730cec8c06)
vpp-20.05 on up-to-date Centos 7.8 host with enforcing SELinux fails to
create a host-interface due to two missing SELinux-permissions:
vpp_t self:packet_socket { read write }
This simple patch adds these two permissions. Tested successfully on
local installation.
The steps to reproduce:
$ ip link add vpeer-host type veth peer name vpeer-vpp
vpp# create host-interface name vpeer-vpp
create host-interface: Permission denied (errno 13)
[...]
$ semodule -i vpp-packet-socket.pp
vpp# create host-interface name vpeer-vpp
host-vpeer-vpp
Type: fix
Ticket: VPP-1931
Change-Id: I2b3d92b27b9a9f26aa1c85af2946b15e83e27944
Signed-off-by: Martin Millnert <martin@millnert.se>
(cherry picked from commit 68849350c5)
Type: fix
VRRP cannot be used on a BVI interface currently because packets sent
to the virtual mac address of the VR fail the destination mac check in
l2_to_bvi().
Apparently people want to use VRRP on BVI interfaces, so update the
check in l2_to_bvi() so that it will check any secondary mac addresses
which have been added to the ethernet interface if the destination mac
address does not match the primary mac address for the interface.
An equivalent check is already done in ethernet_input_inline() for L3
interfaces which are in promiscuous mode.
Change-Id: I7c5bf624dafda8744fea236c704e8e17e5f53b35
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
also, remove a duplicate 17.01.1 section and wrong page tag for 20.09
Change-Id: Ia2cc77faa75cccab38972ad46517762cb7456466
Type: docs
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Type: feature
This patch updateds cryptodev engine uses new DPDK Cryptodev
API planned to be upstreamed in DPDK 20.11.
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: Piotr Bronowski <piotrX.bronowski@intel.com>
Change-Id: I8dd1a8ac643f1e952deb787e466b76ea7aa5f420
This patch adds the RSS steering queues set interface, and it's
implementation in DPDK device:
/* Interface to set rss queues of the interface */
typedef clib_error_t *(vnet_interface_rss_queues_set_t)
(struct vnet_main_t * vnm, struct vnet_hw_interface_t * hi,
clib_bitmap_t *bitmap);
This patch also introduces a command line to set the RSS queues:
set interface rss queues <interface> <list <queue-list>>
To display the rss queues, use "show hardware-interfaces"
Below is the example to configure rss queues for interface Gig0:
vpp# set interface rss queues Gig0 list 0,2,4-7
vpp# show hardware-interfaces brief
Name Idx Link Hardware
VirtualFunctionEthernet18/1/0 1 down VirtualFunctionEthernet18/1/0
Link speed: unknown
RSS queues: 0 2 4 5 6 7
local0 0 down local0
Link speed: unknown
vpp#
Users can also configure the rss queues on a dpdk interface in
startup.conf:
dpdk {
dev 0000:18:01.0 {
rss-queues 0,2,5-7
}
}
Type: feature
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: I1835595a1c54016a84eabee9fd62ce137935385d
- These were displaying blank, apparently dpdk extended stat strings
must be within the heap so they are identified as vectors by
format_c_identifier even though they are not.
Type: fix
Change-Id: I2b153b100203b9856ce3af6d5ecb2daae410fb5b
Signed-off-by: Christian Hopps <chopps@labn.net>
Was seeing imissed counter become negative. Reuse the RX_ERROR code for all
three error counters to avoid the problem.
Type: fix
Change-Id: I99a69c8816326682745785ecd30e18a131ac2969
Signed-off-by: Christian Hopps <chopps@labn.net>
Type: fix
- Document that ooo dequeues with ooo lookups cannot be done in
combination with in order dequeues.
- Added assert to capture this scenario and de-initialized rbtrees for
cut-through tx fifo
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ic40d020b3f0391fcf022ea3c906b86121744144f
Not all ESP crypto algorithms require padding/alignment to be the same
as AES block/IV size. CCM, CTR and GCM all have no padding/alignment
requirements, and the RFCs indicate that no padding (beyond ESPs 4 octet
alignment requirement) should be used unless TFC (traffic flow
confidentiality) has been requested.
CTR: https://tools.ietf.org/html/rfc3686#section-3.2
GCM: https://tools.ietf.org/html/rfc4106#section-3.2
CCM: https://tools.ietf.org/html/rfc4309#section-3.2
- VPP is incorrectly using the IV/AES block size to pad CTR and GCM.
These modes do not require padding (beyond ESPs 4 octet requirement), as
a result packets will have unnecessary padding, which will waste
bandwidth at least and possibly fail certain network configurations that
have finely tuned MTU configurations at worst.
Fix this as well as changing the field names from ".*block_size" to
".*block_align" to better represent their actual (and only) use. Rename
"block_sz" in esp_encrypt to "esp_align" and set it correctly as well.
test: ipsec: Add unit-test to test for RFC correct padding/alignment
test: patch scapy to not incorrectly pad ccm, ctr, gcm modes as well
- Scapy is also incorrectly using the AES block size of 16 to pad CCM,
CTR, and GCM cipher modes. A bug report has been opened with the
and acknowledged with the upstream scapy project as well:
https://github.com/secdev/scapy/issues/2322
Ticket: VPP-1928
Type: fix
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: Iaa4d6a325a2e99fdcb2c375a3395bcfe7947770e
Add the ability to configure the pp2 rx and tx queue sizes in the CLI.
Type: improvement
Signed-off-by: Christian E. Hopps <chopps@chopps.org>
Change-Id: I6a824f92e22fa47fec3d84525cc2d82524ddf639
Type: fix
Calling vlib_get_node_by_name via the VPE api
doesn't work due to hash weirdness. Haven't
gotten around the real cause of this. But this
fixes it.
Change-Id: I89f95dba2bcd9573b8f1f435e063e9dd57f9ca93
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
Otherwise, the debug CLI command is unusable in a script because it
will eat (and complain about) subsequent lines in the script. Missing
this guitar lick, etc:
/* Get a line of input. */
if (!unformat_user (input, unformat_line_input, line_input))
return 0;
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Id328e6f1cc4d2e1672c3946db3865ab5a1a3af8d
Type: refactor
tap, virtio and vhost use virtio/vhost header files from linux
kernel. Different features are supported on different kernel
versions, making it difficult to use those in VPP. This patch
removes virtio/vhost based header dependencies to local header
files.
Change-Id: I064a8adb5cd9753c986b6f224bb075200b3856af
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Type: fix
The ARP/ND feature nodes reply to requests for a VR virtual IP address
when a VR is in the master state. If the VR is in the backup state, the
request is passed to the next node on the feature arc.
This can cause an incorrect response to be sent. If some other feature
(e.g. NAT) causes a virtual IP address to be configured as a "local"
address on the system, a later node on the feature arc may respond to
an ARP/ND request with the real MAC address of the interface.
RFC 5798 says that a router must respond to ARP/ND requests for VR
virtual IP addresses with the VR virtual MAC address. And it says a
router must not respond to ARP/ND requests for VR virtual IP addresses
when the VR is in the backup state. Ensure that ARP/ND requests for
VR virtual IP addresses are dropped when in the backup state rather
than allowing them to continue on the feature arc where another node
may end up responding.
In order to do this, enable/disable the feature nodes when leaving
or entering the init state instead of the master state.
Change-Id: I416f83e125cbf91deb90c3b6eb00ba3207de24ad
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Adjust buffer allocation so it always have odd number of cache lines.
That should result in better distribution of cachelines among cache sets.
Type: improvement
Change-Id: I0d39d4cf01cff36ad6f70a700730823a96448c22
Signed-off-by: Damjan Marion <damarion@cisco.com>
Use VLIB_MAIN_LOOP_ENTER_FUNCTION to do post init initialization for
dpdk crypto rather than create a one-time process to do the same.
Type: fix
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: I06e480b028c8e1fc1b0024a66b2338eb21a797ca
Fix the shown crypto inflight counts which were reversed. Also improve a
couple error descriptions to tell them apart when viewed.
Type: fix
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: I6d4054c64aa842658cfcde8969c7aa48f6d21207
The issue is not easily hit. When GRE_teb packets are received the post
crypto processing adjusts the l2.l2_len value in the vnet_buffer opaque
data. This is overwriting the ipsec opaque data. Later the trace code
fetches the sa_index from the ipsec opaque data. It's just an accident
that this currently works, if the ipsec data is changed so that the
sa_index moves around it will be overwritten by the l2_len modification.
Indeed, this was found b/c local development changes had moved the
sa_index so it was over-lapping with the l2_len memory space, and the UT
failed.
Type: fix
Change-Id: Iaecfa750cf0b36653fd9e75b4d799f323a14d932
Signed-off-by: Christian Hopps <chopps@labn.net>
This patch aims to improve decap performance by reducing expensive
hash_get callings as less as possible using AVX512 on XEON.
e.g. vxlan, vxlan_gpe, geneve, gtpu.
For the existing code, if vtep4 of the current packet match the last
vtep4_key_t well, expensive hash computation can be avoided and the
code returns directly.
This patch improves tunnel decap multiple flows case greatly by
leveraging 512bit vector register on XEON accommodating 8 vtep4_keys.
It enhances the possiblity of avoiding unnecessary hash computing
once hash key of the current packet hits any one of 8 in the 512bit
cache.
The oldest element in vtep4_cache_t is updated in round-robin order.
vlib_get_buffers is also leveraged in the meanwhile.
Type: improvement
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Signed-off-by: Junfeng Wang <drenfong.wang@intel.com>
Change-Id: I313103202bd76f2dd638cd942554721b37ddad60
The protocol value was changed to 50 and 51 (rather than 0 and 1), but
the custom_dump function wasn't updated to reflect this. Also the is_add
value wasn't being shown. Fix both these issues.
Type: fix
Change-Id: I429b4616d6c7937f73308b644154370fab32eaae
Signed-off-by: Christian Hopps <chopps@labn.net>
"clear counters" is not appropriate for a protocol to own. Change
to "clear l2tp counters" (and "test l2tp counter").
Type: fix
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: I3faac3907c4697c1c95df34ac7d31e48063869a8
Scenarios where SIGHUP is sent would include the user closing an xterm
while in interactive/nodaemon mode, or similarly when running vpp in the
same mode during testing (e.g., using ssh to run VPP on a DUT). VPP
should exit in these cases; however, generating a core is unwanted.
Type: fix
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: Ibccfe5e676547e913c8a205ff16ab56d9abb1c82
Type: feature
This adds api calls for the following CLIs:
* set sw_scheuduler worker <N> crypto on|off
* set crypto async dispatch polling|interrupt
* set crypto handler
* set crypto async handler
Change-Id: Ic701d149c440e42ea4575da42b9f69e4c8759602
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
Type: feature
This patch adds new sw_scheduler async crypto engine.
The engine transforms async frames info sync crypto ops and
delegates them to active sync engines. With the patch it
is possible to increase the single worker crypto throughput
by offloading the crypto workload to multiple workers.
By default all workers in the system will attend the crypto
workload processing. However a worker's available cycles
are limited. To avail more cycles to one worker to process
other workload (e.g. the worker core that handles the RX/TX
and IPSec stack processing), a useful cli command is added
to remove itself (or add it back later) from the heavy
crypto workload but only let other workers to process the
crypto. The command is:
- set sw_scheduler worker <idx> crypto <on|off>
It also adds new interrupt mode to async crypto dispatch node.
This mode signals the node when new frames are enqueued
as opposed to polling mode that continuously calls dispatch node.
New cli commands:
- set crypto async dispatch [polling|interrupt]
- show crypto async status (displays mode and nodes' states)
Signed-off-by: PiotrX Kleski <piotrx.kleski@intel.com>
Signed-off-by: DariuszX Kazimierski <dariuszx.kazimierski@intel.com>
Reviewed-by: Fan Zhang <roy.fan.zhang@intel.com>
Change-Id: I332655f347bb9e3bc9c64166e86e393e911bdb39
Let twice-nat static mapping pick specific
address from the twice-nat pool.
Type: improvement
Change-Id: Iadaa036af2fa3b0e6e9a68ff6e68b4bbe1650eb1
Signed-off-by: Filip Varga <fivarga@cisco.com>
Type: fix
adding routes should be MP safe. When new prefixes with differrent
prefix lengths are added, adjust the sorted list in an MP safe way.
Change-Id: Ib73a3c84d01eb86d17f8e79ea2bd2505dd9afb3d
Signed-off-by: Neale Ranns <nranns@cisco.com>
gso option is available for the debug CLI version of bond create.
This patch is to create a new API to have the corresponding option in
the binary API. The old binary API bond_create is marked deprecated.
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Id9501b8e6d267ae09e2b411957f181343da459c0
enum bar_enum {
BAR1 = 0,
BAR2,
BAR3 [backwards_compatible],
BAR4 = 9 [backwards_compatible],
};
This allows adding backwards compatible (as guaranteed by the developer) enums.
The enums marked backwards compatible are not considered in the CRC calculation.
Type: improvement
Change-Id: I6fc0c21b19e1a02cff7f5d279a0f3a32d2f8b630
Signed-off-by: Ole Troan <ot@cisco.com>
In one's complement, there are two representations of zero: the all
zero and the all one bit values, often referred to as +0 and -0. See
RFC 1624 section 3 for more details.
This used to be taken care of in ip4_header_checksum(), but it is no
longer the case. The check ip->checksum == ip4_header_checksum (ip) is
no longer correct in the -0 case.
Always use ip4_header_checksum_is_valid() instead (which behaves
correctly since 9a79a1ab93).
Type: fix
Fixes: e5f0050c7a
Change-Id: Iacc6b60645a834287b085aecb9e3fdb4554cf0cf
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Adjacencies are only defined for IPv4 and IPv6.
Type: fix
Fixes: 20aec3db44
Change-Id: I19b2b7f6958da49f41c6eabc9f248840769acbbb
Signed-off-by: Benoît Ganne <bganne@cisco.com>
It is cheaper to include checksum field in calculation and simply
compare result with zero.
Type: improvement
Change-Id: I6f77632c0a4d2f2c632d044d3a5d2fcf2b5bac62
Signed-off-by: Damjan Marion <damarion@cisco.com>
Translation memory size is internally a uword, but in api it was u32,
resulting in the returned value being 0 all the time.
Fix the "incorrect" API reply to return a u32 capped to 0xffffffff if
the u64 is larger than that, introduce the message with
the correct type, deprecate the message with the incorrect type.
Also, while we are updating the message definition,
add the max translations / max users per worker thread
into the new message.
Type: fix
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I92e38a6a2bcb70fc8d1b129bbe416bf7f9e54280
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
There is a number of TERMs with big length, such as
'screen.konsole-256color' (23). These TERMs can not
be processed properly by vpp because maximum telnet
byte stream supported is 24 and we need 6 more service
bytes to send TTYPE.
So I extended maximum depth guard to 32.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: I9ca506996a97e9567d06483c5f020d6cc394329c
Make it shorter and same format when converting to biggor or smaller
types.
Type: refactor
Change-Id: I443d67e18ae65d779b4d9a0dce5406f7d9f0e4ac
Signed-off-by: Damjan Marion <damarion@cisco.com>
Node variant names have changed to skx and icl.
Type: fix
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Change-Id: I5be832aa59e4e440df53869077d801c011e3defa
Building VPP using gcc-10 fails because of the array bounds check
error for function ip4_header_checksum(), with option field in IPv4
header exceeding the ip4_header_t bound. Fix this error by turning
off the array bounds check option for function ip4_header_checksum().
Change-Id: I68cc241ae9e403d35ac2e320549506dc6565a0b6
Type: fix
Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com>
This is the code refactor for vnet/flow infra and the dpdk_plugin flow
implementation. The main works of the refactor are:
1. Added two base flow type: VNET_FLOW_TYPE_IP4 and VNET_FLOW_TYPE_IP6
as the base the flow type
2. All the other flows are derived from the base flow types
3. Removed some flow types that are not currently supported by
the hardware, and VPP won't leverage them either:
IP4_GTPU_IP4, IP4_GTPU_IP6, IP6_GTPC, IP6_GTPU,
IP6_GTPU_IP4, IP6_GTPU_IP6
4. Re-implemented the vnet/flow cli as well as the dpdk_plugin
implementation
5. refine cli prompt
6. refine display info in command "show flow entry"
Type: refactor
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: Ica5e61c5881adc73b28335fd83e36ec1cb420c96
Also clib_mem_destroy() to destroy the current mspace.
Handy when an application wants to make a memory allocation arena
disappear.
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I020db902fbe2473545506fecbc230c2b048992f8
When we have both format blacklisted devices like:
blacklist 1234:5678
blacklist 1234:56:78.0
unformat with fmt=%x:%x matches for both strings
and the rest 78.0 substring is kept in input
and it can't be parsed for init args
This patch checks first if device format matches PCI address and
just then if it matches Vendor and Product
Type: fix
Change-Id: If111762c0e0a424b052e4f6dc0f67731bf89dc2a
Signed-off-by: Vladimir Ratnikov <vratnikov@netgate.com>
Some of the tests are time-sensitive, and at present require a non-trivial
modification in order to run at high concurrency.
Without these modifications, they intermittently fail, and require
the test retries.
Rather than setting them to the extended tests and forgetting
about them, put them into a "solo" set, which gets run in a
single-threaded mode after the rest of the tests are done.
Mark a few of the tests that showed errors during TEST_JOBS=48
as forced-solo.
Also, give a better diagnostic if the testcase misses a docstring
needed to represent it in the diagnostic outputs.
Type: fix
Change-Id: I33fe62eb17edc1885bd2c3523892051d52da6546
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
threads vcl workers scenario
Type: fix
In multiple threads vcl workers scenario, multiple vcl workes can create
sessions with same index. Because only one vls worker created, key of
session_index_to_vlsh_table is duplicated.
Signed-off-by: hanlin <hanlin_wang@163.com>
Change-Id: I7e1f5bc471adc5378194452aef85e611f5d0df1d
Make sure session is in non-closed state after non-blocking connect to
ensure cleanup is performed, if needed.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ic1533b9864828f875c840dfdfeed08bf32d10651
This patch adds the API for vnet/flow infra.
Currently this API supports the below flow types:
VNET_FLOW_TYPE_IP4_N_TUPLE
VNET_FLOW_TYPE_IP6_N_TUPLE
VNET_FLOW_TYPE_IP4_N_TUPLE_TAGGED
VNET_FLOW_TYPE_IP6_N_TUPLE_TAGGED
VNET_FLOW_TYPE_IP4_L2TPV3OIP
VNET_FLOW_TYPE_IP4_IPSEC_ESP
VNET_FLOW_TYPE_IP4_IPSEC_AH
VNET_FLOW_TYPE_IP4_GTPU
All the above flows are tested with Intel E810/X710 NIC
Type: feature
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: Icb8ae20cab9bdad6b120dddc3bd4fb1d85634f3f
Type: fix
In ip6_neighbor_probe(), if the interface is not enabled for ip6,
return NULL and skip trying to build a packet.
If the interface is not enabled, its mcast adjacency will be set to
~0 and a seg fault will ensue.
Change-Id: I825c9f40a0d5b2a77f788ac8dbd618138706383d
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Endian handlers was not registered for reply messages.
Causing endian-neutral handlers to crash.
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Id14173300ee1ab2601e92d58c70d2fa260814b69
- check input sw_if_index to make sure it is sane. Coverity actually
complains about it.
- return rv. Some of the APIs handlers were not passing back the rv.
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I8378ea948af561ba2bd9b02fb10bf4f9df2a2cd2
- Display the interface name upon successfully creating the interface.
- Don't go silent when error occurs. Be explicit about it.
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ie4fdaf323f8b3833ad4a7af3872290141204aba8
To avoid ridiculous pointer gymnastics in golang
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Ic5a641c18281970e2b9bc3013a54cfc3ae92e372
Type: fix
pci_types.api lives in vlib/pci which wasn't included
in the generate_json.py script
Change-Id: I15649786282584b7aefb9ace83c5154eb697b50b
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
Make sure payload_proto variable is set properly in vnet_ip_mroute_cmd()
function. This avoids using an uninitialized payload_proto value which
could lead to assertion failure when using e.g. the ip mroute add
command.
Type: fix
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
Change-Id: I8b1d1df02e80150836b7b0448814d8f99747eeed
'type' is a keyword in golang, so s/type/event_type/ in elog.h and
elsewhere.
Add vec_len_not_inline(...), elog_write_file_not_inline(...) and
elog_read_file_not_inline(...) since the inline forms aren't usable.
More such tweaks may follow.
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I9a80a6afa635f5cdedee554ee9abe400fafc1cb6
This addresses problem with incorrect sorting of artifacts in the per-patch packagecloud repositories
for throttle branches, by slightly changing the version string for the non-released (per-build) artifacts:
1) the rc1 and rc2 artifacts now will have the count of
commits since tag (0) and the explicit commit hash
2) the builds after version XX.YY but before XX.YY.1
are named as if the version was "XX.YY.0"
Type: fix
Change-Id: Ib69a59ee98aa296c5cb24b183bba807b6cb3c05c
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
There can be a race condition in the case
a thread tries to do a bihash_search while
another instantiates the bihash.
Type: fix
Change-Id: Ic61b590763beb409e112957c43a5a66cd10afb28
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
Generate a semver define for API modules in the VPP side C generator.
E.g in map.api_types.h
\#define VL_API_MAP_API_VERSION_MAJOR 4
\#define VL_API_MAP_API_VERSION_MINOR 2
\#define VL_API_MAP_API_VERSION_PATCH 1
Type: improvement
Change-Id: I2ab86154e85e26fb075c4e91a01ff3692640654b
Signed-off-by: Ole Troan <ot@cisco.com>
fib table removal would leave lingering sessions in vpp
this patch is aimed at solving this issue by grouping
sessions by source and destionation fib. if one of the
fibs gets removed this grouping is tagged as expired
and session won't be passed to non existing fib table
Ticket: VPPSUPP-93
Type: improvement
Change-Id: I45b1205a8b58d91f174e6feb862554ec2f6cffad
Signed-off-by: Filip Varga <fivarga@cisco.com>
Add deq request on ct fifo if present and clear event when session is
removed from epoll group.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I8a20342879b3aac2f8fe8171556baaf9c992363b
vrf table may be dynamically added or deleted. When the table is deleted,
clients who use the corresponding vrf table may need a callback to
do the clean up. The mechanism added here is cloned from
VNET_SW_INTERFACE_ADD_DEL_FUNCTION.
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I08635c715cd7361a6c359b90890dd3545b0da94c
The included image was pulling in figure 2, change to the correct
figure 11.
Type: docs
Signed-off-by: Christian Hopps <chopps@labn.net>
Change-Id: I3679916364753fc682d1d1713fce952de398e626
Some statistics counters were implemented as error counters. Move them
to stat segment, where they belong.
Type: improvement
Change-Id: I5600bec1b4e0496282297374ec1e79d909cdaf8a
Signed-off-by: Klement Sekera <ksekera@cisco.com>
The "ACL as a service" lookup infra is shared, so a global heap must
be used.
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Change-Id: I86894254b737392c1968b6a581b5a37590376428
Represent pointers directly in shared memory and require clients to adjust for
shared memory segment being mapped at different base address.
Deprecated: stat_segment_pointer() / stat_segment_offset()
Added: stat_segment_adjust()
Bumped the stat segment version to 2.
Type: refactor
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: I33e756187b8903b45dcd353e6c1a101b7a4acb79
Currently, mutlti-threads only support one dispatch thread and multiple
worker threads, eventually only dispatch thread is a vcl worker and can
interact with epoll.
This patch will register all threads as vcl worker, and then each
thread can interact with epoll now. Moreover, session migration also
supported, such as socket created in thread A and used (bind, connect
and etc.) in thread B.
Type: feature
Signed-off-by: hanlin <hanlin_wang@163.com>
Change-Id: Iab0b43a33466968c1423d7d20faf1460c8589d91
In unions all representations of the union are decoded.
Which means trying to decode something that isn't an address
might have invalid address family types.
Type: fix
Ticket: VPP-1884
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Id3381ef8cc885952c1eb488ebc70e276eaceb366
add new api stat_segment_set_timeout_nsec to limit time waiting for vpp
in_progress state.
Change-Id: Ic78a97bc5013d67d7e4bbcc4a6f0ef918f9f9b33
Type: improvement
Signed-off-by: Ole Troan <ot@cisco.com>
Type: fix
A struct that is used as a hash key was being initialized in its
declaration. On CentOS 8 this caused some hash lookups to fail.
This seems to be caused by uninitialized padding.
Use clib_memset() to initialize the key with 0's to avoid the issue.
Change-Id: I00555c201a1ab34133971313ba14f20f4e867a30
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
- In a new system (e.g. ubuntu 18.04 docker image), the pip3
installation of jenkins-job-builder fails unless pip3 wheel
is completed on a previous pip3 install command.
Type: fix
Change-Id: I4354c851161b9c470d44951dc8b722ffc87f56f7
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
- switch to using vlib_time as reference for timestamps
- use us precision ticks for tcp but keep using ms precision for
timestamps. As a result, srtt, rttvar and rto are now measured in us
instead of ms. MRTT samples from timestamps are converted from ms to
us (not accurate under ms) while high precision samples are used with us
precision, i.e., they're no longer converted to ms precision samples.
Type: improvement
Change-Id: Ibda559575d9b4fdc85b0985264f7c865ff367e34
Signed-off-by: Florin Coras <fcoras@cisco.com>
Before this patch it was not allowed to replace
a dynamic ARP entry with a static one with the
same mac-address.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: I6cfc0e510ffdf141c61874288f11a60395182374
A predicate function which says whether a given sw_if_index has a
certain feature enabled.
int vnet_feature_is_enabled (const char *arc_name,
const char *feature_node_name,
u32 sw_if_index)
returns 1 if the feature is enabled
returns 0 if the feature is not enabled
returns VNET_API_ERROR_INVALID_VALUE if the arc name is not found
returns VNET_API_ERROR_INVALID_VALUE_2 if the feature node name is not found
returns VNET_API_ERROR_INVALID_SW_IF_INDEX if the sw_if_index is not found
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Id9ed53e2ae9252efc750ae80fd610456b95cefa4
On CentOS-7 aarch64, command of 'debuginfo-install -y glibc openssl-libs mbedtls-devel zlib' in 'make install-deps' fails because it tries to install the corresponding *debuginfo* packages from some inaccessible/unmaintained repos on aarch64, e.g., centos-sclo-rh-debuginfo. The error message shows as below.
Using 'debuginfo-install --enablerepo=xxx' also fails because it will still enable all the repos including the broken repos on aarch64. Using 'debuginfo-install --disablerepo=xxx' (xxx is the broken repo) works fine but we are not centain about that if VPP user will install additional broken repos on aarch64 or not. So to fix this error, we install all the *debuginfo* packages for 'glibc openssl-libs mbedtls-devel zlib' packages using 'yum install' instead.
[root@ ~]# debuginfo-install -y glibc openssl-libs mbedtls-devel zlib
Loaded plugins: auto-update-debuginfo, fastestmirror, ovl
enabling epel-debuginfo
enabling base-debuginfo
enabling centos-sclo-rh-debuginfo
Loading mirror speeds from cached hostfile
epel/aarch64/metalink | 8.2 kB 00:00:00
epel-debuginfo/aarch64/metalink | 8.5 kB 00:00:00
* base: mirror.aktkn.sg
* centos-sclo-rh: mirror.aktkn.sg
* epel: mirrors.yun-idc.com
* epel-debuginfo: mirrors.yun-idc.com
* extras: mirror.aktkn.sg
* updates: mirror.xtom.com.hk
http://debuginfo.centos.org/centos/7/sclo/aarch64/repodata/repomd.xml: [Errno 14] HTTP Error 404 - Not Found
Trying other mirror.
To address this issue please refer to the below wiki article
https://wiki.centos.org/yum-errors
If above article doesn't help to resolve this issue please use https://bugs.centos.org/.
failure: repodata/repomd.xml from centos-sclo-rh-debuginfo: [Errno 256] No more mirrors to try.
http://debuginfo.centos.org/centos/7/sclo/aarch64/repodata/repomd.xml: [Errno 14] HTTP Error 404 - Not Found
Type: fix
Change-Id: I017c3b20a167d8035c3ae617b9ad5ae479e52f57
Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com>
For bond interface, we set hardware link state to up initially when
admin state is set to up. Thereafter, if we toggle the admin state to
down and up, we require at least one active member in the bond prior
to bringing up the hardware state which is inconsistent. The fix is to
remove the unnecessary condition.
This is a rework of the original proposed patch which is more complex
and could be tricky to get it right
https://gerrit.fd.io/r/c/vpp/+/27141
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I65f4fee55299b5b4d5645f6e05e1be15b46b73cd
Update the documents for the VPP tutorial.
Type: fix
Signed-off-by: John DeNisco <jdenisco@cisco.com>
Change-Id: I62d0641ff0af7b07a26b4379f19cbdaf285537d7
prefix_len is parsed using '%u' format requiring an integer.
Type: fix
Change-Id: I9f31fa60e80d4f45fe456f3fd6c94f123cfc99ea
Signed-off-by: Benoît Ganne <bganne@cisco.com>
- Replace textual string slave with member except APIs.
- For APIs, mark the existing APIs as deprecated and introduce new APIs
- While introducing sw_bond_interface_dump, add the optional filter by
sw_if_index and enhance the testcases to make use of it.
Type: improvement
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ib6626c514e45350308aeeda0decb70f3aba2f63e
This fixes TCP connection cleanup crash when TCP debugging is enabled.
It could happen if session_stream_accept() returned an error.
Type: fix
Signed-off-by: Ivan Shvedunov <ivan4th@gmail.com>
Change-Id: I0df08969d82bb0f44def81b9e47195bd15502831
The elog track, if enabled, must be initalized before synack is sent.
Type: fix
Signed-off-by: Ivan Shvedunov <ivan4th@gmail.com>
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I0c585f466c9a5cbc13af971ae4951b93f9913c5f
By default, VPP automatically assignes for each tunnel
next available QAT device by order dev_id-que-pair.
In most cases we have more than one device and it can
greatly increase ipsec perfomance without any actions
with configuration from user if we use all the
devices first and first que-pairs
Type: feature
Signed-off-by: Vladimir Ratnikov <vratnikov@netgate.com>
Change-Id: Iac9fe74768775459e22f69bb3706b542090a9375
Copy only exactly the data provided by the user even when it is not a
4-bytes multiple.
Type: fix
Change-Id: I2ef987c37e58523a38b46b09227529db2c26aa55
Signed-off-by: Benoît Ganne <bganne@cisco.com>
"ipv4"/"ipv6" option is not supported in acl_add_replace and
macip_acl_add_replace vat api.
Update its help message per actual api usage.
Type: fix
Change-Id: I8d34fac5f98bd78a46a5e98df05cd35182988dd8
Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
Reviewed-by: Jieqiang Wang <Jieqiang.Wang@arm.com>
Reviewed-by: Govindarajan Mohandoss <Govindarajan.Mohandoss@arm.com>
Also removes delack timer and reuses the u32 for the pending timers
list.
Type: fix
Ticket: VPP-1923
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I4edbb72d5b2aa5e14f87659f49e675af1e834aca
Change the port number selection for new NAT sessions so that it
matches how the thread index is calculated from the port number for
out2in packets. Before this change there was a problem when the
largest port number in the range was used, that resulted in the wrong
thread index being selected when out2in packets arrive for that
session.
Type: fix
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
Change-Id: I936c389eb0d5df6168e18e5e44754de1cdad6ad1
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Do not access data structures based on uninitialized key->alg.
Type: fix
Fixes: f539578bac
Change-Id: I6bfb7e7a51af2c131b8bdf3bca6a38fcf1094760
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Add a callback to clear the per-client packet trace buffer cache.
Save the packet trace dump pg setup script.
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I252be911b5f937ece0da5dca152263ece3d52963
As requested by Neale, mark the ABF API as in-progress
Change-Id: I109a32fa54b1f2a882695d9fd71b235fa46bc6f3
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
As per request from Neale, mark the GBP plugin APIs as in-progress.
Type: fix
Change-Id: I679943edcfff0742ee32c45cd8f97f482c353b9f
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
As requested by Neale, mark the API as in-progress.
Change-Id: Id92cad65c66435e179583507f077816e09e4205b
Type: fix
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
In pppoe_cp_node.c, node->errors[error0] was accessed without
node->errors being initialized.
Found with AFL + ASAN.
Type: fix
Signed-off-by: TimotheeChauvin <timchauv@cisco.com>
Change-Id: Ide8a60021b2d47b5e2fce7062d8f12c7f4d225f7
Callbacks for monitoring and performance measurement:
- Add new callback list type, with context
- Add callbacks for API, CLI, and barrier sync
- Modify node dispatch callback to pass plugin-specific context
- Modify perfmon plugin to keep PMC samples local to the plugin
- Include process nodes in dispatch callback
- Pass dispatch function return value to callback
Type: refactor
Signed-off-by: Tom Seidenberg <tseidenb@cisco.com>
Change-Id: I28b06c58490611e08d76ff5b01b2347ba2109b22
clang tends to force alignment of all sections when compiling for
address sanitizer, confusing VPP plugin infra.
On the contrary, GCC does not support this attribute on sections.
Selectively enable it depending upon the compiler.
Type: fix
Fixes: cea46522e7
Change-Id: I2dd6e52e70e6b4d89c101171cafc813e175ec472
Signed-off-by: Benoît Ganne <bganne@cisco.com>
- Fixes coverity issue #210160.
- Fixes the possible issue in cryptodev when input node does
not update mbuf, such as avf-input.
- Fixes GCM ESN packet incorrect tag.
- Code clean up to reduce binary size.
Type: fix
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: Dariusz Kazimierski <dariuszx.kazimierski@intel.com>
Signed-off-by: Piotr Kleski <piotrx.kleski@intel.com>
Change-Id: Ic05ae29855ac1f7a62e4af5831a4ed9faa8f561a
In case of vector, we must check length before trying to access element.
Also fix wrong DPDK plugin workaround.
Type: fix
Change-Id: I2ecef1c88ebef2362f48cab0d462699aa43cd4b9
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Provide binary API compatibility support for the "cop" APIs until vpp
21.01.
Change the deprecation date in map.api to vpp 21.01.
Type: refactor
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I0e60d96de4ae9ae4448f134cf257934126f3b760
- add 64-bit version
- remove byte swaps, as they are actually not needed
Type: improvement
Change-Id: Ia1efe54ee80b4ae7633b9ab36b3154c3357006d7
Signed-off-by: Damjan Marion <damarion@cisco.com>
For some reason clang does not support &((struct foo*)0)->field in
static assertion contrary to gcc.
Use offsetof() macro implementation provided by both compilers instead.
Type: fix
Change-Id: I3311cdd29c5861e45dc0ef92f2bbd66242ca73b8
Signed-off-by: Benoît Ganne <bganne@cisco.com>
This updates the computation of the outer ip6 header payload_length
field in order to take into account the GRE header length.
Change-Id: Ie9f982521aeaef7279a9e329a33272d6fae0a428
Type: fix
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
This saves about 20 clocks/packet in both code paths.
Type: improvement
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: Ib559c74bf8168e3ddd764d51b7e5bcd2a557f591
Previously there's a format_ip4_address in format_ip6...
This patch fixes this typo
Type: fix
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: Ice124db6594720ed35a992d069341f399c331e1d
Also fix session test app name registrations
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I7d365154ab9af83b17b026762ab4f0aea85ce486
This introduces a txq structure mirroring the rxq structure.
This fixes the case when #txq > #rxq, because lock must be per txq.
Type: fix
Fixes: dfb19cabe2
Change-Id: Ic1bce64d2b08b9a98c8242a1ba1bfcdbda322bec
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Type: fix
From kernel 4.20, xdp support has been added in tun_sendmsg.
If sndbuf == INT_MAX, kernel executes xdp data path
for tun driver which assumes packets are ethernet frames.
This patch is avoiding the xdp data path in kernel by setting
the sendbuf value < INT_MAX.
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Change-Id: Ia4aa54b177b96d56a2d513d18d26ca01d5b88929
Tests for AES-GCM and AES-CBC with different key lengths
Type: test
Change-Id: Ie7eeebb0f7e8331a717866475cb4ee00042857ce
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Type: fix
Some debugging declarations were wrapped in an 'ifdef CLIB_DEBUG'.
This seems to always evaluate true because clib.h defines CLIB_DEBUG
to 0 if its not defined.
The result is that if a route table is added and a route is added to
it and then the table is deleted, VPP exits because the debug function
fib_table_assert_empty() gets called whether VPP was started using a
debug build or a release build.
Change the ifdef to 'if CLIB_DEBUG > 0'.
Change-Id: I357dc2c299e81b95244f2f7efaadb8e0de27627a
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Type: fix
The v6 accept mode input feature was being declared with
the node added to ip4-multicast instead of ip6-multicast. Add to
the correct arc.
Change-Id: I08f6e5e7dde84a37687fa0af750a7a16fe537ea6
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
unformat_ip6_mask wasn't accounting for customized field names
when deciding if it managed to parse at least one field.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I26cab4c6828b510e277079628af5115ac43af3ff
Based on the comments in the struct, udp_encap_t_ is meant to span 2
cachelines. Due to the 64 bit alignment of dpo_id_t, the struct spanned
3 cachelines. This caused fetching ue_ip_proto to trigger an additional
cache miss. This patch rearranges the ordering of the struct fields
so that udp_encap_t_ only spans 2 cachelines as intended.
before:
(gdb) print (int)&((struct udp_encap_t_*)0)->cacheline1
$8 = 128
after:
(gdb) print (int)&((struct udp_encap_t_*)0)->cacheline1
$1 = 64
Type: fix
Signed-off-by: Vadym Martsynovskyy <vmartsyn@fb.com>
Change-Id: I066c08654d4a8ef3e2d3954e957d4c5d382b209f
Short Load/Stores combined with prefetching in the beginning of the loop
place too much pressure on AGUs and memory accesses.
The patch interleaves load/store operations with computational operations
to alleviate the pain point.
vlib_get_buffers is also leveraged.
Redefine u8 dst_and_src[12] instead of dst[6] and src[6] in struct
l2input_trace_t in order to merge two copys into one.
Type: improvement
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Change-Id: I7d3df7732c476069235e3019c68f0f53bca9637e
Type: fix
When accept mode is enabled, a backup VR will configure the VR virtual
addresses locally and respond to packets sent to those addresses. This
did not work when the primary VR is the address owner and sends
advertisements using the virtual address as the source address. It
also did not work when NAT was configured on the interface with the
virtual address as the NAT pool address. In both cases, advertisements
from other VRs would arrive and be dropped because they appeared to
be spoofed - the source address would be an address that is
configured as an interface address on the instance receiving it.
When accept mode is enabled for a VR and the VR enters the master state,
add an input feature on ip[46]-multicast for the interface which looks
for VRRP advertisements, figures out whether they are for a VR which
is configured with accept mode and is in the master state and kicks
them straight to the VRRP nodes to avoid dropping them.
Change-Id: I240ba1ee0b3fd6d693de729698c1181dc71bb08b
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
add test_lacp.py to cover basic lacp unit test
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I6f7f3d801956e54106f3c55cedaca186d81dad25
This patch adds the IPSec ESP/AH type flow support
Have tested on E810 with Intel iAVF driver
Type: feature
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: I6ab8e69f67c423cc4e33f3c363881a97cdb98c30
These gdb macros should prove very helpul when poking around in core files.
Pifi (pool_is_free_index) is not straighforward. Best to work it out
once.
Others:
bitmap_get = clib_bitmap_get
vl = vec_len
pe = pool_elts
node_name_from_index, as described
vnet_buffer_opaque, prints the primary buffer opaque
vnet_buffer_opaque2, prints the secondary buffer opaque
Fix vppinfra unit-test compile error
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Id2a2391a47e5a07cf3757f473e3805cc04784161
This ensures we don't recompute the requirements-{2,3}.txt on each make
test run, and skips patching scapy if it is already patched instead of
failing.
Change-Id: I3da57182ae49f3dd04db139d96734a5d145fedff
Type: fix
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
Prevent overflow if input network prefix is too small and crash on
packet #1 due to vector not being allocated/initialized.
Type: fix
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I3494cc62ce889df48cc59cc9340b5dd70338c3a8
Deal with circular macro definitions instead of crashing due to stack
overflow.
Separate macro tables, per CLI session
Add documentation to the Sphinx docs
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I55fc9152bd37ad0c15fa3959f38b07b63100e634
With this change, punt received neighbor advertisements instead of drop
in order to give other elements of the system an opportunity to handle
them.
Type: improvement
Change-Id: Ie6cde9eebc83e42029967141cb32ebc459fd3680
Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
Added external dependancies package. Updated the startup.conf template.
Fix extras/scripts/check_commit_msg so it accepts '_' characters in
feature names.
Type: fix
Signed-off-by: John DeNisco <jdenisco@cisco.com>
Change-Id: I69c33a24b30d31e5b8081021030bff88cbaeace9
Signed-off-by: Dave Barach <dave@barachs.net>
Cover vl_mem_api_dead_client_scan(...), which is not so simple. Spin
up vpp_api_test using pexpect.spawn, make it bootstrap a private
memory segment, kill vpp_api_test, and wait for the dead client
scanner to run a few times so that it declares the client dead.
The test only runs if GCOV_TESTS=yes is set in the environment.
Type: test
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I3c712fa011df2e23761a2e73c1aba0c7e50123a6
All of the pieces have been sitting around for years.
Added several debug CLI commands:
"define <variable-name> <value>"
"undefine <variable-name>"
"show macro [noeval]"
"echo <whatever>"
Macros may refer to other macros. To defer evaluation:
"define foo \$(bar)" or some such.
The macro evaluator is not smart about "define foo \$(foo)" or more
complicated circular definitions, so don't do that.
Environment variables are available, simply use $<name-of-environment-vbl>
The macro expander has a table of (overrideable) builtin names, which
are evaluated by calling functions.
Simple example:
echo $USER
define ip1 192.168.1.1/24
define ip2 192.168.2.1/24
loop create
loop create
set int ip address loop0 $ip1
set int ip address loop1 $ip2
show int addr
show macro
undefine ip1
undefine ip2
Type: feature
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I08a800647bac573d8ae3cfd75c40061d41c5f976
For src/tools/g2, src/vlibapi, and src/vlibmemory
Type: docs
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I800c3e3a1ecb011d5e692d78a015e78342706786
This patch documents the DNS plugin, vlib packet tracing, vnet feature
arcs, and the vppinfra library.
Type: docs
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Idb16ec0593d428ce78fe8404374a823a56b4b420
Change the yaml file matching pattern from FEATURE.yaml to
FEATURE*.yaml, so we can document multiple features in one directory.
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Ic62f9fb4c73330c698724244cd8b362364fce5f9
Use pointed value (ipv6 address) as hash key instead of pointer value
(pointer to ipv6 address on the stack).
Type: fix
Change-Id: I74a8b6d4bb552b5b344bcb50c8e830700c4ce739
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Provide a minimal trace [ip4/ip6 src/dst address] for dropped pkts
when the user specifies "trace add error-drop XXXX", but does not
trace pkts from the original input node.
This is a wireshark dissector problem. Packets thrown at error-drop
may be well-formed, or not. VPP must not crash, no matter what.
The minimal trace capture and decode could be enhanced. Anyone
interested in doing that must consider all of the corner-cases
involved. This version should be at least somewhat useful.
Note that "pcap trace drop ..." - and the packet generator - seem like
the right tools to use when researching more complex issues.
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I961ca133980ffa2a1e5707879a443b21442ed894
Prefer using source port form packet as outside port if possible.
Type: improvement
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I5c25f6a42386f38c9a6cc95bd7dda9f090b49817
Support separate local and remote TEIDs, with local (or RX) one used
for GTPU tunnel RX/decap and remote (or TX) one used encap/TX.
Updated current gtpu API to support seperate RX/TX TEIDs and added new
gtpu_tunnel_update_tteid API to allow changing TX-TEID of an existing
GTPU tunnel. The current tunnel field "teid" is used for RX-TEID and
a new field "tteid" is used for TX-TEID.
Type: improvement
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I549d79750a34bb965036da298c0ca894d15c2c20
The filter sw_if_index was in the API sw_interface_vhost_user_dump. But
it was never implemented in the backend. This patch is to add the backend,
vat, and custom dump support for the filter.
Type: feature
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Iaa41a7e11bfbcbb4c60092375e4b0dcf0950077b
If the id is invalid we cannot check whether we must free the message or
not, free it anyway.
Type: fix
Change-Id: Ie4426f601390d1e5e14c739f670e8c1e6e3aaf1e
Signed-off-by: Benoît Ganne <bganne@cisco.com>
The move to cmake obviates the need for this build script (and its terminology).
Type: make
Signed-off-by: Simon Chatterjee <code@chatts.net>
Change-Id: Ie17429bfdf5a4b02ce2b70ba3568a7445c162d38
Add missing dependencies and correct the building to support CentOS8
Type: make
Change-Id: Ie15b9b1174fa9b6d5ae02bace36ebc77e17d770c
Signed-off-by: Yichen Wang <yicwang@cisco.com>
if ((A | B) == false) it means both A and B are false, so for
the following code
if (PREDICT_FALSE (!good_udp1))
{
if ((flags1 & VNET_BUFFER_F_L4_CHECKSUM_COMPUTED) == 0)
{
...
}
}
if ((flags1 & VNET_BUFFER_F_L4_CHECKSUM_COMPUTED) == 0) is always
true if the code run it. Remove it.
Type: improvement
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Change-Id: I6bd1e9340c7a00089fc1c9ae49773add832d309e
When recycling a graph node vnet_register_interface, it is missing an
explicit call to vlib_worker_thread_node_runtime_update(). However,
there is an implicit call to vlib_worker_thread_node_runtime_update()
via vnet_sw_interface_set_flags_helper() if it enables a new feature on
the interface for the first time. But that implicit call is not
guaranteed. For example, if an interface is created, deleted, and
created, then it may skip the implicit call to
vlib_worker_thread_node_runtime_update(). When that happens, the graph
nodes on thread 0 are not sync'ed to the worker threads. So the worker
thread's graph nodes are out of sync momentarily with the main thread's
graph nodes until some other event happens which calls for a sync is
needed. During this window, the worker thread's graph node is
vulnerable and may experience a crash.
When deleting a graph node, we never trigger a sync to the worker
thread. A patch was committed 3 years ago via
https://gerrit.fd.io/r/c/vpp/+/7523 to fix a show run crash. In
hindsight, the approach taken by 7523 is not orthogonal. While at it,
let's fix it right for both issues with a call to
vlib_worker_thread_node_runtime_update() in the appropriate place and
remove 7523.
Type: fix
Ticket: VPPSUPP-86
Fixes: gerrit 7523 / 19e9d954bd
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ic9472bd2d3a212dbfeceb526506ed0400983a142
No change in default behavior. To use htlb pages for the ip4 mtrie,
use the "ip" command-line option "mtrie-hugetlb".
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I5497e426a47200edff2c7e15563ed6a42af12e7f
Type: fix
Anouncing DH group in esp transform proposals will enable PFS which is
not suppored now. This fixes issue during rekey when using strongswan as
responder.
Change-Id: Ib9f586113ae0ab9dc67e6ceadff43f8aac463820
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
This avoids using dirty data from shared memory by client.
Type: fix
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I96eecf655bf344ec29609cedbd8dc891b572e207
Packet tracing performance doesn't justify inlining
vlib_add_trace(...) over 500 times.
It makes a 15% text-segment size difference in a representative use-case:
Inline:
$ size .../vnet_skx.dir/ipsec/ipsec_input.c.o
text data bss dec hex filename
6831 80 0 6911 1aff .../vnet_skx.dir/ipsec/ipsec_input.c.o
Not inline:
$ size .../vnet_skx.dir/ipsec/ipsec_input.c.o
text data bss dec hex filename
5776 80 0 5856 16e0 .../vnet_skx.dir/ipsec/ipsec_input.c.o
Retain the original code as vlib_add_trace_inline, instantiate once as
vlib_add_trace.
Type: refactor
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Iaf431dbf00c4aad03663d86f9dd1322e84d03962
Fix a nit warning: we're not likely to create a vlib process with more
than 4gb of stack.
Type: fix
Ticket: VPP-1888
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I8bc7f64287c2802b0c286ce3d04443ac723a9a33
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
Replace whitespread (mis)use of snat_session_key_t by proper function
arguments where applicable and inline functions to calculate hash keys
instead of using structs for that. Make all hash tables use same network
byte order port so that there is no longer a discrepancy between static
mappings using host byte order while in2out/out2in tables using network
byte order.
Type: improvement
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I80786d2f947c67824c101a13bb608f1fe1080f34
Add dpo_pool_barrier_sync/release, use them to clean up
thread-unsafe pool expansion cases.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I09299124a25f8d541e3bb4b75375568990e9b911
adj_alloc (...) is not thread safe when the adj pool or combined
counter vectors expand.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I55710de6ecc083b7434e11798659cca9250c9131
Add the libssl-dev library for ubuntu 20.04 in Makefile.
Type: fix
Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com>
Change-Id: I4187cb041997e7457734ffdb18bdbec98a051669
load_balance_alloc_i(...) is not thread safe when the
load_balance_pool or combined counter vectors expand.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I7f295ed77350d1df0434d5ff461eedafe79131de
- add ability to pass data together with interrupt
- avoid locking for local interrupts (same thread)
Type: improvement
Change-Id: I73a2ab2e716bb887a1f02c87788ae83e329f9b40
Signed-off-by: Damjan Marion <damarion@cisco.com>
make sure we don't pass a 0 (zero) to log2.
Type: fix
Signed-off-by: Andreas Schultz <andreas.schultz@travelping.com>
Change-Id: Ib3386ad4686835d3836f48294cee4058e897847a
Otherwise, the out2in path will discard return-path traffic with
probability 1.0.
Type: fix
Fixes: gerrit 23963 / f126e746fc
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I621ed99329c04ef358035747dde599c0016b58f5
This reverts commit 9121c415a9.
Reason for revert: causes deadlock with multiple worker threads
Type: fix
Change-Id: Icf3d6d343d99d887abacbaa03acbf04d4628514e
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Type: fix
the hash walk does not give the same guarantees as the bihash so
walk in a safe manner.
Change-Id: Idfe48c3a84ab3a341d887f7d196bc81ba34ae8b0
Signed-off-by: Neale Ranns <nranns@cisco.com>
Fix a warning in sanitizer.rst and publish_docs.sh
Type: fix
Signed-off-by: John DeNisco <jdenisco@cisco.com>
Change-Id: I66d724f52a7b724a97a894e1cd35c1239aeedcdf
Use %U and unformat_udp_port instead of %u for unformat() call for
u16 collector_port number in set_ipfix_exporter_command_fn() to
avoid corruption of other variables which can happen if unformat()
with %u is used with a 16-bit variable. This avoids crash due to
corrupted fib_index value.
Type: fix
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
Change-Id: Id54273fcc458a7f9c5aa4025aa91711f160c1c1a
I had cleaned up the sample.md previously for 20.01, but when later
that was merged to master, the file was renamed. So, fix this issue.
Change-Id: I6347685af216901fbfdd445606735b9bf79f8fe5
Type: docs
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit e2e3c38be6)
Also: permanently solve ordering issues with the vpp builtin vat
plugin loader, by explicitly loading vat plugins once we're sure that
all data plane plugins have registered their APIs / API message
handlers.
Fix compilation / link errors when the vpp builtin vat plugin loader
is disbled by cmake configuration.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Id553c63ae860ebfc196c5ad4b55c19e08fec2c9e
Signed-off-by: Ole Troan <ot@cisco.com>
Type: fix
we can probably do better, but for now this is needed
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: I07161f5ac07a70a0e6db6608ba31659d9bc2c9ed
Fix and optimize DMAC check in ethernet-input node to utilize NIC or
driver which support L3 DMAC-filtering mode so that DMAC check can be
bypassed safely for interfaces/sub-interfaces in L3 mode.
Checking of interface in L3-DMAC-filtering state to avoid DMAC check
require the following:
a) Fix interface driver init sequence for devices which supports L3
DMAC-filtering to indicate its capability and initialize interface
to L3 DMAC-filtering state.
b) Fix ethernet_set_flags() function and its associated callback
flags_change() functions registered by various drivers in interface
infra to provide proper L3 DMAC filtering status.
Maintain interface/sub-interface L3 config count so DMAC checks can be
bypassed if L3 forwarding is not setup on any main/sub-interfaces.
Type: fix
Ticket: VPP-1868
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I204d90459c13e9e486cfcba4e64e3d479bc9f2ae
Looks like MAP_LOCK is not enough, so call mlock(...) instead....
Type: fix
Change-Id: I1bc668a2bf3c861ca1c2d376c0fb6bfea87d4f48
Signed-off-by: Damjan Marion <damarion@cisco.com>
Type: improvement
Bond link state is being maintained inconsistently. It is initially set to
up. If the bond interface admin state is set to down, the link state is
set to down. If the bond interface admin state is set to up, the link
state is only set to up if there are active slave interfaces at that point.
If slaves become active at some later time, it does not get updated. Its
next chance to be updated is the next time the bond interface is set to
admin up.
To address this, do not set the link state to up after creating a bond.
Adjust the link state as slave interfaces are attached or detached
based on whether the bond is getting its first active slave or losing
its last one.
Unit test added to verify correct maintenance of link state.
Change-Id: I31f17321f7f0e727e1ab1e01713423af6566dad9
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Change in snat_ipfix_header_create() to use thread-specific
vlib_main_t *vm pointer to avoid problems with different threads
accessing the same vlib_main_t data structure. This avoids
assertion failure when vlib_time_now() is called with a vm
corresponding to a different thread.
Type: fix
Signed-off-by: Elias Rudberg <elias.rudberg@bahnhof.net>
Change-Id: I2096c1debb5688d3b97e5ed9a0ea78d94053d8b7
Add a new boolean to signal that the API infrastructure should performan any required
endian conversions for the API handler.
am->is_autoendian[mm->msg_id_base + VL_API_MAP_ADD_DOMAIN] = 1;
Similarly add new REPLY_ macros that perform endian conversion.
These changes do not change the on-the-wire encoding of the API messages, and therefore the API CRC is not changed.
Type: feature
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: I7588f8ccb38b2d1e8d85ea17be99bac43f756267
Signed-off-by: Ole Troan <ot@cisco.com>
Type: fix
when the interpose is on an adj-fib and the cover is removed the adj
source will not install. this lead to no path list being found for the
interpose source and a crash. pick a drop path list in this case.
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: Ied217da043926c913657080f5ffb151201225d23
Type: improvement
a bihash per-interface used too much memory.
Change-Id: I447bb66c0907e1632fa5d886a3600e518663c39e
Signed-off-by: Neale Ranns <nranns@cisco.com>
This fixes an issue where the pinned requirements file can be modified
when running the tests.
Change-Id: Ic89d1844d1fd8d00f62211a9b051a26ac34ee316
Type: fix
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
Instead of having to wrap dump/detail calls in control ping, send details messages in between a normal
reply / request pair. As expressed in the below service statement.
Example:
service {
rpc map_domains_gets returns map_domains_get_reply
stream map_domain_details;
};
define map_domains_get
{
u32 client_index;
u32 context;
u32 cursor;
};
define map_domains_get_reply
{
u32 context;
i32 retval;
u32 cursor;
};
To avoid blocking the main thread for too long, the replies are now sent in client message queue size
chunks. The reply message returns VNET_API_ERROR_EAGAIN when there is more to read.
The API handler must also include a "cursor" that is used to the next call to the get function.
API handler example:
REPLY_AND_DETAILS_MACRO (VL_API_MAP_DOMAINS_GET_REPLY, mm->domains,
({
send_domain_details (cursor, rp, mp->context);
}));
The macro starts from cursor and iterates through the pool
until vl_api_process_may_suspend() returns true or the iteration
reaches the end of the list.
Client Example:
cursor = 0
d = []
while True:
rv, details = map_domains_get(cursor=cursor)
d += details
if rv.retval == 0 or rv.retval != -165:
break
cursor = rv.cursor
or the convenience iterator:
for x in vpp.details_iter(vpp.api.map_domains_get):
pass
or
list(details_iter(map_domains_get))
Change-Id: Iad9f6b41b0ef886adb584c97708dd91cf552749e
Type: feature
Signed-off-by: Ole Troan <ot@cisco.com>
Obserbed when VPP is running in k8s container.
Type: fix
Change-Id: Ibbff9c3921bd7f4f97d47cb6f10eed8ed5efe269
Signed-off-by: Damjan Marion <damarion@cisco.com>
Reported as part of "show tcp stats"
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I499ab4e41d13aaf1a0d060e37f09087e63d09567
In case there is no free space in first buffer for ICV and footer,
additional buffer will be added, but esp_encrypt will stay in single
buffer mode.
The issue happens for the following payload sizes:
- TCP packets with payload 1992
- ICMP packets with payload 2004
This fix moves the single/chained buffer ops selection to after
esp_add_footer_and_icv call.
Type: fix
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: PiotrX Kleski <piotrx.kleski@intel.com>
Change-Id: Ic5ceba418f738933f96edb3e489ca2d149033b79
Support VIRTCHNL_VF_CAP_ADV_LINK_SPEED.
Type: improvement
Change-Id: Ie45c9d557d0369464cbb6881ee1259ef3807d7d5
Signed-off-by: Damjan Marion <damarion@cisco.com>
Fix allocation of port per dslite session.
After each session is created per protocol,
when new one should be created with new port,
instead it's trying to create with the same port
and while(1) loop is executed forever and VPP does
not response
Type: fix
Signed-off-by: Vladimir Ratnikov <vratnikov@netgate.com>
Change-Id: Ic91b8b07253498ef9846ca60bcd4c4c76a5fac91
Instead of allocating stack from the main heap, this patch mmaps stack
memory together with guard page.
This aproach reduces main heap usage, and stack memory is prefaulted
on demand, so bigger process stacks will have zero impact on memory
usage as long as stack memory is not needed for real.
In addition, it fixes issue with systems which have bigger default page
size (observed with 65536).
Type: improvement
Change-Id: I593365c603d4702e428967d80fd425fdee2c4a21
Signed-off-by: Damjan Marion <damarion@cisco.com>
Now, libc epfd and vls epfd are independent and can only epoll_wait independently without timeout, then app calling epoll_wait will occupy high CPU. So we nest vcl_mq_epfd into libc epfd when using eventfd with VPP, and then we can only epoll_wait libc epfd with specified timeout.
Type: feature
Signed-off-by: hanlin <hanlin_wang@163.com>
Change-Id: I6b6e0f501c769e186714bfbc187cfaed2533b4c2
Signed-off-by: hanlin <hanlin_wang@163.com>
Document update to reflect merged SRv6 functions of lookup and forwarding for inner IP packet encapsulated with outer IP and GTP-U headers.
Type: docs
Signed-off-by: Satoru Matsushima <satoru.matsushima@gmail.com>
Change-Id: I85c9ddf6bf9fa63f2b8b6e03eff3ecec1e7615ab
vppinfra source files MUST NOT #include <vlib/vlib.h>, <vnet/vnet.h>
or similar. Move mpcap_add_packet(...), mpcap_add_buffer(...) to a new
file: src/vnet/mpcap.h.
Type: refactor
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Id517aef6fe49b618f853ce32940b91ba45a1e60d
Netvsc devices have the port type determined from their link speed.
The link speed between reboots of an Azure VM does not always end
up at the same value, so an interface that was FortyGigabitEthernet0
earlier may be FiftyGigabitEthernet0 now. That makes it difficult
to maintain a persistent store of configurations and apply those at
startup.
Change the port type to be VF so the name will always be generated
as VirtualFunctionEthernetX.
Type: improvement
Change-Id: I58cab852b87c0bcd9f73afe239803f38dab5c159
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
Swap byte order for fields of type vl_api_rx_mode_t.
Ticket: VPP-1871
Type: fix
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Change-Id: Ia1745257b57209d41661d38067e0dd7618f9a9b9
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
(cherry picked from commit aefcd1a357)
cleaned up typos in documetation while doing a read through.
Type: style
Change-Id: Id8abbd8d6297ee10490c12a8e62097224f7ec1e1
Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
The recent introduction of f-strings require >=Python3.6
Fix a couple of pycodestyle warnings.
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Id9e375bb8febb157d7f9d7fa21808724770df6f8
Cleanup L2/L3 mode switch to not redirect to/from ethernet-input node
as it is no longer necessary.
L2 patch should use sw_if_index for device feature enable/disable.
Type: fix
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I0f24161d027b07c188fd1e05276146f94c075710
Identified and removed executable bit from source files in the tree.
find . -perm 755 -name *.[ch] -exec chmod a-x {} \;
Type: improvement
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Change-Id: I00710d59fcc46ce5be5233109af4c8077daff74b
Type: feature
* initiator behind NAT supported
* tested with static NAT mappings
* works only with pre-configured tunnels
The pre-configured tunnel has to be defined as follows:
initiator (i) side: src=ip(i) dst=ip(r)
responder (r) side: src=ip(r) dst=ip(nat)
Change-Id: Ia9f79ddbbcc3f7dc8fde6bbeca2a433e3b784e94
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
Type: fix
Seems clang needs explicit casting to u64 of u64 and f64 multiplication
before truncating to u32
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ib0d7a33d4c5e68577e401e854fc3e55e0723da93
Parallel merges introduced two test clasess with a same name. Rename
latter, so that former is seen (and run) by test runner again.
Type: fix
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I47772b41bb940bfdda4536cdd1f9b5e3768ca18b
When vcl_epoll_wait_handle_mq handles rx events exceeding maxevents, VPP will not signal because cursize > 0, and the remaining rx events cannot be triggered because the eventfd event has been read. Therefore, we should dequeue all events until cursize = 0. And then handle msg up to maxevents with vcl_epoll_wait_handle_mq_event and those beyond with vcl_handle_mq_event.
Type: fix
Signed-off-by: hanlin <hanlin_wang@163.com>
Change-Id: I8a0c87cb41c837deb8284b40f668cc3c7d9d6e56
Signed-off-by: hanlin <hanlin_wang@163.com>
Since the removal of "users" concept in ED-NAT nat44_user_dump API
returns empty array. This brings back previous behaviour at
a considerable runtime cost until a better API is introduced.
Type: improvement
Change-Id: I5a45923cfeb6b8ebe6fc906601264d6567386991
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Removing the comments around eid_type which seem to have been overlooked
by the original patch https://gerrit.fd.io/r/c/vpp/+/24663.
Type: refactor
Signed-off-by: Onong Tayeng <otayeng@cisco.com>
Change-Id: I48e1993cf8869cb32e159d1956f3ec1e5943e33f
If a cli command is run while there are no cli session, then
cm->cli_file_pool will not be initialized and we should not try to
operate on it.
Type: fix
Change-Id: Iaea15a23f7efd5b17fab13e6c1cbb3a9a34080e0
Signed-off-by: Benoît Ganne <bganne@cisco.com>
cxgbe PMD initializes its control channel as part of dev_configure(),
and trying to get link status prior to it will lead to a crash.
DPDK documentation loosely hints that we should not call any device
function before dev_start(), call link_state() only for the relevant
PMDs.
From DPDK API documentation:
The functions exported by the application Ethernet API to setup a device
designated by its port identifier must be invoked in the following
order:
rte_eth_dev_configure()
rte_eth_tx_queue_setup()
rte_eth_rx_queue_setup()
rte_eth_dev_start()
Then, the network application can invoke, in any order, the functions
exported by the Ethernet API to get the MAC address of a given device,
to get the speed and the status of a device physical link, to
receive/transmit [burst of] packets, and so on.
Type: fix
Change-Id: I12d2ab4d84e6bd72a9f695447e86f3222929c804
Signed-off-by: Benoît Ganne <bganne@cisco.com>
By storing thread and session index in hash table we are able to skip
multiple hash lookups in multi-worker scenario, which were used for
handoff before. Also, by storing sesion index in vnet_buffer2, we can
avoid repeating the lookup after handoff.
Type: improvement
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I406fb12f4e2dd8f4a5ca5d83d59dbc37e1af9abf
@@ -11,7 +11,7 @@ code that provides tools that are used in a development environment.
This section covers the following:
* Describes how to manually install VPP Binaries on different OS platforms (Ubuntu, Centos, openSUSE) and then how to configure and use VPP.
* Describes how to manually install VPP Binaries on different OS platforms (Ubuntu, Centos) and then how to configure and use VPP.
* Describes the different types of VPP packages, which are used in both basic and developer installs.
* A VPP tutorial which is a great way to learn VPP basics.
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.