- Fixed an issue where containers wouldn't stop and get removed when
a test run is interrupted
- Replaced PIDs with Ginkgo process indexes + PPIDs
- Fixed CPU allocation for envoy and nginx-ldp containers
- All container logs should now get saved properly
Type: test
Change-Id: I4c737c1d326390494c0dda1ec6d3fc1f04f51663
Signed-off-by: Adrian Villin <avillin@cisco.com>
- added checkstyle-go and fixstyle-go
- comments in stats_fs.go were missing a space
Type: make
Change-Id: I520acab5ff61eaf9d0ccfd9425bdc41f74559198
Signed-off-by: Adrian Villin <avillin@cisco.com>
This inline function is introduced to simplify code readability and consolidate AH packet
processing in separate API.
Type: improvement
Change-Id: Id98074d00aba277fed714ea9995655c9e84ec83b
Signed-off-by: vinay tripathi <vinayx.tripathi@intel.com>
'struct.error: required argument is not an integer'
is quite useless itself, so let's raise an error from it at least
saying what was the thing getting packed
Type: improvement
Change-Id: Icb762fbab98446d1e1331315e6c337f789cbba95
Signed-off-by: Klement Sekera <klement.sekera@gmail.com>
- It is now possible to only run tests that are in a certain file
Type: test
Change-Id: I41665dd2bc0942c283be36a5af3e560fd65e9d03
Signed-off-by: Adrian Villin <avillin@cisco.com>
HttpTpsTest now use Gomega's gmeasure package and go internal http
client. With gmeasure you can create "Experiments" which can produce
reports to show the statistical distribution of measurement.
Potentially experiments can also be cached and used to identify
regression in performance.
Type: test
Change-Id: Id049fb0344d8ebed71b15e706b053b5c2a18e0de
Signed-off-by: Matus Fabian <matfabia@cisco.com>
No functional changes, just make it to respect original make binary
name used to start building
Type: improvement
Change-Id: Ic8568237fbb39c6a0d3b7405a9670e9410aeb752
Signed-off-by: Renato Botelho do Couto <renato@netgate.com>
When using gdb_show_traces() in debug mode, don't assert if workers are
not parked, as it is typically called from gdb.
Type: fix
Change-Id: Iabf175d96dc152da4d1abfbce9ccc9020d0b5d61
Signed-off-by: Benoît Ganne <bganne@cisco.com>
This patch adds vnet generic flow type support in OCTEON plugin, which
extends the existing vnet flow types supported. It allows users to
configure additional match patterns like 802.1q tag fields, 802.1ad tag
fields, MPLS fields, IP DSCP etc., if supported by the underlying hardware.
On OCTEON various match patterns including user defined custom protocol
types can be supported depending on the programmable classification profile.
Generic flows operate based on hexadecimal strings representing packet
data bytes and corresponding mask data bytes. The mask data bytes, with
bits set to '1', selectively identify the data bytes used for hardware
flow matching.
To configure generic flow rules, packetforge tool is recommended which
accepts inputs in a user readable and friendly format. This tool is
available in VPP tree under `extras/packetforge`. Detailed instructions
can be found in the documentation under `extras/packetforge`. Additionally
user can use existing vnet flow CLI and binary API interfaces to
configure rules manually.
Type: feature
Change-Id: I8198536cf1fe0a4719542a8b54c599230c7852e9
Signed-off-by: Sriram Vatala <svatala@marvell.com>
This patch fixes counter_start value, as it should hold the cumulative
counter value whenever counter clear routine is called.
Type: fix
Fixes: 38c619115b
Change-Id: I50bf8ddcde419caf1170dfacdea03ff3d93a3327
Signed-off-by: Monendra Singh Kushwaha <kmonendra@marvell.com>
This patch updates trace to use "qid" keywords instead of "q".
Type: feature
Change-Id: Ib0f41a8a1a93cbbbf8c59304924d4e68efff48fd
Signed-off-by: Monendra Singh Kushwaha <kmonendra@marvell.com>
- Fixes errors while obtaining container logs on test failure
(NginxPerf tests), ab/wrk containers are now stopped on test teardown
like other containers.
Type: test
Change-Id: Ic336fcd3ed6bf68dd345d378262cb28eb5efc789
Signed-off-by: Adrian Villin <avillin@cisco.com>
- Remove code from test report that is effectively
untested and categorize based on reason for lack
of testing.
Type: test
Change-Id: I6ca5444055b3a81a4880945b6845afc867556277
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Currently, buffers-per-numa is used for all NUMA domains, and there's
no way to allocate different amounts for each domain.
Therefore, this adds a per-NUMA-domain buffer config section, which
currently has a single option to set the number of buffers.
If it's not specified or set to zero, the code falls back to
buffers-per-numa as before.
Type: improvement
Change-Id: If35d7a9eff6f8d1d78063ea7873dbf50780d0ec3
Signed-off-by: Lukas Stockner <lstockner@genesiscloud.com>
Type: fix
This patch fixes the missing UDP port check in IPsec NAT-T
case. As of RFC3948 UDP encapped ESP traffic should have
destination port ID of 4500, which was missing.
The related tests are updated with this port ID, too.
Change-Id: I73ecc6a93de8d0f4b642313b0f4d9c2f214a7790
Signed-off-by: Fan Zhang <fanzhang.oss@gmail.com>
Type: fix
fib_api_path_decode() is utilized by the IP route API call
to translate the path from the API to the fib_route_path_t
structure. The ip_route_add_del_handler_t function initializes
the fib_route_path_t structure to zeros, consequently setting
the sw_if_index value to 0, which is a valid value in VPP.
Typically, the default VRF (Virtual Routing and Forwarding)
has a local interface at index 0, leading to normal functionality.
However, a custom VRF table without any interface will result
in a crash.
The issue arises because the DROP route in fib_api_path_decode()
does not override the sw_if_index value with the one provided
in vl_api_fib_path_t. Subsequently, when this sw_if_index is
attempted to be resolved in the VRF table where the interface
does not exist, it leads to a crash.
This patch addresses the problem by setting the sw_if_index of
fib_route_path_t to the sw_if_index value of the API path.
To reproduce the issue, please remove the fix and run the following command:
make test-debug TEST=test_ip4.TestIPv4RouteLookup.test_exact_match
Change-Id: I5d72e91e5c701e749a92873941bee7b7b5eabd41
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
- Libunwind was missing in Dockerfile.nginx, causing some tests to fail.
Tests that were temporarily disabled because of that issue
are now re-enabled.
Type: test
Change-Id: I3f544be483784e8a7a1f22737cafca615b9f836b
Signed-off-by: Adrian Villin <avillin@cisco.com>
- Some tests are broken in the CI, skipping them
for now
Type: test
Change-Id: I3d4efeee63b819956e5ffa1b3920e81962a2fcc9
Signed-off-by: Adrian Villin <avillin@cisco.com>
- Reduced the amount of error messages while obtaining container logs
when a test fails by keeping track of started containers. NginxPerf
tests still have those error messages, because wrk/ab containers shut
themselves down before the test ends.
Type: test
Change-Id: I40a193345e5b46aec1834774f23aebc822eee885
Signed-off-by: Adrian Villin <avillin@cisco.com>
- missing from https://gerrit.fd.io/r/c/vpp/+/40929 due to hst not running in CI yet.
Type: fix
Change-Id: Ib5d9b877725624ef7de9f2be5e517609aec7c5a1
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
- use libunwrap which seems to be industry standard
- display traceback on console if running interactive or with syslog
disabled (color output unless nocolor specified)
- print hexdump of offending code
- print library filename for each stack frame
Type: improvement
Change-Id: I61d3056251b87076be0578ccda300aa311c222ef
Signed-off-by: Damjan Marion <damarion@cisco.com>
Type: feature
If 'main-core' is not specified, the main thread is pinned
by default to the cpu it is running on during initialization.
This change does not impact manual core-pinning, which
requires the 'main-core' argument
e.g. 'cpu {main-core x corelist-workers n}.
Change-Id: I874034591bf0acf4d71b231dfbbb0f6de8fe6060
Signed-off-by: hsandid <halsandi@cisco.com>
In production network, often the host and the vm are managed by different
groups. The host statistics may not be readily available to the folks who
manage the VM. Having the RX queue full statistics in VPP can inform
that there might possibly be drops in the host.
Type: improvement
Change-Id: I43206647ac0d8092968c4187236d9696ae0acccd
Signed-off-by: Steven Luong <sluong@cisco.com>
On non-glibc systems, execinfo is the only option available, but the lib
is old and can crash when unwinding. We now can use libunwind to unroll
it instead of using execinfo.h.
Type: improvement
Signed-off-by: Guillaume Solignac <gsoligna@cisco.com>
Change-Id: I8b55f7aca97261a2efb5dca998889d5e0645939a
