Add RX/TX function prototypes for further datapath development.
Signed-off-by: Howard Wang
---
drivers/net/r8169/meson.build| 1 +
drivers/net/r8169/r8169_ethdev.c | 20 +--
drivers/net/r8169/r8169_ethdev.h | 3 ++
drivers/net/r8169/r8169_rxtx.c | 57 +
Signed-off-by: Minggang Li(Gavin)
Acked-by: Alex Vesker
---
drivers/net/mlx5/hws/mlx5dr_send.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/hws/mlx5dr_send.c
b/drivers/net/mlx5/hws/mlx5dr_send.c
index 3022c50260..e9abf3dddb 100644
--- a/drivers/ne
Signed-off-by: Minggang Li(Gavin)
Acked-by: Alex Vesker
---
drivers/net/mlx5/hws/mlx5dr_rule.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/mlx5/hws/mlx5dr_rule.c
b/drivers/net/mlx5/hws/mlx5dr_rule.c
index 1edb7eac74..5d66d81ea5 100644
--- a/drivers/net/mlx5/hws/mlx5dr_
rte_errno should be set if anything wrong happened in under layer so that
user can figure out what's going on.
There were some cases that did not set it when ipool allocation failed. To
fix the issue, set rte_errno to ENOMEM if mlx5_ipool_malloc failed to
allocate ID.
Fixes: c40c061a02 ("net/mlx5
This patch set is to improve error handling in pmd and under layer.
Gavin Li (3):
net/mlx5: set rte errno if malloc failed
---
changelog:
v0->v1
- Fix typo in commit message
v1->v2
- Fix signoff warning
---
net/mlx5/hws: add log for failing to create rule in HWS
---
changelog:
v1->v2
- Fix sig
This patch refactors the mempool cache to fix two bugs:
1. When a mempool is created with a cache size of N objects, the cache was
actually created with a size of 1.5 * N objects.
2. The mempool cache field names did not reflect their purpose;
the "flushthresh" field held the size, and the "size" f
> On 6/19/2024 11:11 AM, Chaoyong He wrote:
> > From: Long Wu
> >
> > The Rx packet type offload feature may affect the performance, so add
> > a control flag for applications to turn it on or off.
> >
> > Signed-off-by: Long Wu
> > ---
> > lib/ethdev/rte_ethdev.h | 1 +
> > 1 file changed, 1 in
Hi, Ferruh
Could you please provide feedback on the patch I submitted?
Any suggestions for improvement would be appreciated.
Thanks
This patch enables VDUSE reconnection support making use of
the newly introduced reconnection mechanism in Vhost
library.
At DPDK VDUSE device creation time, there are two
possibilities:
1. The Kernel VDUSE device does not exist:
a. A reconnection file named after the VDUSE device name
is
This patch introduces a way for backend to keep track
of the needed information to be able to reconnect without
frontend cooperation.
It will be used for VDUSE, which does not provide interface
for the backend to save and later recover local virtqueues
metadata needed to reconnect.
Vhost-user sup
Hi Morten,
Apologies for delayed response.
> Maybe a combination, returning the lowest end of the two versions of the list,
> would work...
>
> --
> Common header file (rte_common.h):
> --
>
> /* Add at end of enum list in the heade
This series adds support for VDUSE reconnection.
First patch introduces the reconnection file layout and
track the virtqueues available index updates in the
datapath and control queue.
Second patch adds VDUSE reconnect intialization and some
sanity checks to prevent incompatible reconnections.
C
On 9/23/24 17:42, David Marchand wrote:
On Fri, Sep 20, 2024 at 11:09 AM Maxime Coquelin
wrote:
This patch introduces a way for backend to keep track
of the needed information to be able to reconnect without
frontend cooperation.
It will be used for VDUSE, which does not provide interface
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
>
> The scatter Rx offload capability is needed for the pmd_buffer_scatter
> test suite. The command that retrieves the capability is:
> show port rx_offload capabilities
>
> The command also retrieves a lot of other capabilities (RX_OFFLOAD_*)
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
>
> Add support for marking test cases as requiring a certain topology. The
> default topology is a two link topology and the other supported
> topologies are one link and no link topologies.
>
> The TestProtocol of test suites and cases is exte
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
>
> Add parsing for the show rxq info tespmd command
> and add support for the Scattered Rx capability.
>
> Signed-off-by: Juraj Linkeš
> Reviewed-by: Dean Marx
Reviewed-by: Jeremy Spewock
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
>
> Some test cases or suites may be testing a NIC feature that is not
> supported on all NICs, so add support for marking test cases or suites
> as requiring NIC capabilities.
>
> The marking is done with a decorator, which populates the intern
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
>
> If a test case is not relevant to the testing environment (such as when
> a NIC doesn't support a tested feature), the framework should skip it.
> The mechanism is a skeleton without actual logic that would set a test
> case or suite to be s
One super nit-pick comment below, even without that though I still
think this looks good.
Reviewed-by: Jeremy Spewock
On Mon, Sep 23, 2024 at 11:02 AM Juraj Linkeš
wrote:
> +def is_test_case(function: Callable) -> bool:
> +if inspect.isfunction(function):
> +
From: Jeremy Spewock
In order for test suites to create virtual functions there has to be
functions in the API that developers can use. This patch adds the
ability to create virtual functions to the Node API so that they are
reachable within test suites.
Bugzilla ID: 1500
Depends-on: patch-14431
From: Jeremy Spewock
Virtual functions in the framework are created using SR-IOV. The process
for doing this can vary depending on the operating system, so the
commands to create VFs have to be abstracted into different classes
based on the operating system. This patch adds the stubs for methods
From: Jeremy Spewock
In DPDK applications virtual functions are treated the same as ports,
but within the framework there are benefits to differentiating the two
in order to add more metadata to VFs about where they originate from.
For this reason this patch adds a new class for handling virtual
From: Jeremy Spewock
Previously in the DTS framework the helper methods in the TestSuite
class designated ports as either ingress or egress ports and would wrap
the methods of the traffic generator to allow packets to only flow to
those designated ingress or egress ports. This is undesirable in s
From: Jeremy Spewock
Previously the DTS framework only included methods that bind all ports
that the test run was aware of to either the DPDK driver or the OS
driver. There are however some cases, like creating virtual functions,
where you would want some ports bound to the OS driver and others b
From: Jeremy Spewock
v4:
* apply to next-dts
Jeremy Spewock (5):
dts: allow binding only a single port to a different driver
dts: parameterize what ports the TG sends packets to
dts: add class for virtual functions
dts: add OS abstractions for creating virtual functions
dts: add funct
From: Jeremy Spewock
This patch ports over the port_control test suite from the Old DTS
framework and adapts the functionality to fit with the current testing
framework. The test suite provides validation of basic port control
functions such as starting, stopping, and closing ports. It should
be
From: Jeremy Spewock
v2:
* apply to next-dts
Jeremy Spewock (2):
dts: add method for closing ports to testpmd
dts: add port control testing suite
dts/framework/config/conf_yaml_schema.json| 3 +-
dts/framework/remote_session/testpmd_shell.py | 18 +
dts/tests/TestSuite_port_contr
From: Jeremy Spewock
Closing ports is a standard configuration feature that is available in
testpmd but the framework lacks the ability to access this command
through the Testpmd API. This patch adds a method that performs this
action and verifies the results of sending the command to allow
devel
On Fri, Sep 20, 2024 at 11:09 AM Maxime Coquelin
wrote:
>
> This patch enables VDUSE reconnection support making use of
> the newly introduced reconnection mechanism in Vhost
> library.
>
> At DPDK VDUSE device creation time, there are two
> possibilities:
> 1. The Kernel VDUSE device does not ex
Thanks to all of you. Much appreciated!
On Sat, Sep 21, 2024 at 9:07 PM Ferruh Yigit wrote:
> On 9/19/2024 5:37 PM, Jerin Jacob wrote:
> > On Thu, Sep 19, 2024 at 12:16 AM Adel Belkhiri
> wrote:
> >>
> >> From: Adel Belkhiri
> >>
> >> Split the tracepoints rte_ethdev_trace_rx_burst and
> >> rt
Addressed a specific overflow issue in the eth_dev_adjust_nb_desc()
function where the uint16_t variable nb_desc would overflow when its
value was greater than (2^16 - nb_align). This overflow caused nb_desc
to incorrectly wrap around between 0 and nb_align-1, leading to the
function setting nb_des
Apologies, please disregard this patch, I sent out a v3 with a fixed
subject line.
From: Jeremy Spewock
This patch adds a new test suite to DTS that validates the accuracy of
the port statistics using testpmd. The functionality is tested by
sending a packet of a fixed side to the SUT and verifying that the
statistic for packets received, received bytes, packets sent, and sent
b
From: Jeremy Spewock
Methods currently exist for querying the statistics of a port in
testpmd, but there weren't methods added for clearing the current
statistics on a port. This patch adds methods that allow you to clear
the statistics of a single port or all ports to account for situations
wher
From: Jeremy Spewock
v3:
* fix name of the first patch
Jeremy Spewock (2):
dts: add clearing port stats to testpmd shell
dts: add port stats checks test suite
dts/framework/config/conf_yaml_schema.json| 3 +-
dts/framework/remote_session/testpmd_shell.py | 49 +-
dts/tests/Tes
On Thu, Sep 19, 2024 at 11:33 PM wrote:
>
> From: Pavan Nikhilesh
>
> Unregister SSO device and NPA IRQs before resizing
> them.
>
> Signed-off-by: Pavan Nikhilesh
# Add fixes: tag
# Change commit message to common/cnxk: fix...
> ---
> v2 Changes:
> - Reorder npa interrupt un-registration.
>
On Tue, Sep 10, 2024 at 2:54 PM Nithin Dabilpuram
wrote:
>
> This series adds support for Marvell cn20k SOC for mempool and
> net PMD's.
>
> This series also adds few net/cnxk PMD updates to expose IPsec
> features supported by HW that are very custom in nature and
> some enhancements for cn10k.
From: Jeremy Spewock
This patch adds a new test suite to DTS that validates the accuracy of
the port statistics using testpmd. The functionality is tested by
sending a packet of a fixed side to the SUT and verifying that the
statistic for packets received, received bytes, packets sent, and sent
b
From: Jeremy Spewock
Methods currently exist for querying the statistics of a port in
testpmd, but there weren't methods added for clearing the current
statistics on a port. This patch adds methods that allow you to clear
the statistics of a single port or all ports to account for situations
wher
From: Jeremy Spewock
v2:
* apply on next-dts
* add raw testpmd output to show_port_stats_all to make test method
consistent
* add dependency on VLAN suite for set_verbose method
Jeremy Spewock (2):
dts: add clearing port stats and verbose mode to testpmd
dts: add port stats checks test
On Fri, Sep 20, 2024 at 11:09 AM Maxime Coquelin
wrote:
>
> This patch introduces a way for backend to keep track
> of the needed information to be able to reconnect without
> frontend cooperation.
>
> It will be used for VDUSE, which does not provide interface
> for the backend to save and later
On Sun, Sep 8, 2024 at 1:28 AM Akhil Goyal wrote:
>
> CNXK product families can have a use case to allow PF and VF
> applications to communicate using mailboxes and also get notified
> of any interrupt that may occur on the device.
> Hence, a new raw device driver is added for such RVU LF devices.
https://bugs.dpdk.org/show_bug.cgi?id=1548
Bug ID: 1548
Summary: Add asynchronous output collector for testpmd verbose
information
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status:
On Sun, Sep 8, 2024 at 1:03 AM Akhil Goyal wrote:
>
> Added an internal API for PMDs to get raw device pointer
> from a device id.
>
> Signed-off-by: Akhil Goyal
> ---
> lib/rawdev/rte_rawdev_pmd.h | 24
# This patch needs to go through the main tree. So please send this
Add the capabilities advertised by the testpmd command "show port info"
so that test cases may be marked as requiring those capabilities:
RUNTIME_RX_QUEUE_SETUP
RUNTIME_TX_QUEUE_SETUP
RXQ_SHARE
FLOW_RULE_KEEP
FLOW_SHARED_OBJECT_KEEP
These names are copy pasted from the existing DeviceCapabilitiesF
Add new files to generate DTS API documentation from.
Signed-off-by: Juraj Linkeš
Reviewed-by: Jeremy Spewock
Reviewed-by: Dean Marx
---
doc/api/dts/framework.testbed_model.capability.rst | 6 ++
doc/api/dts/framework.testbed_model.rst| 2 ++
doc/api/dts/framework.testbed_model
The scatter Rx offload capability is needed for the pmd_buffer_scatter
test suite. The command that retrieves the capability is:
show port rx_offload capabilities
The command also retrieves a lot of other capabilities (RX_OFFLOAD_*)
which are all added into a Flag. The Flag members correspond to
Add support for marking test cases as requiring a certain topology. The
default topology is a two link topology and the other supported
topologies are one link and no link topologies.
The TestProtocol of test suites and cases is extended with the topology
type each test suite or case requires. Eac
Add parsing for the show rxq info tespmd command
and add support for the Scattered Rx capability.
Signed-off-by: Juraj Linkeš
Reviewed-by: Dean Marx
---
dts/framework/remote_session/testpmd_shell.py | 137 +-
dts/framework/testbed_model/capability.py | 12 ++
dts/tests/Te
Some test cases or suites may be testing a NIC feature that is not
supported on all NICs, so add support for marking test cases or suites
as requiring NIC capabilities.
The marking is done with a decorator, which populates the internal
required_capabilities attribute of TestProtocol. The NIC capab
If a test case is not relevant to the testing environment (such as when
a NIC doesn't support a tested feature), the framework should skip it.
The mechanism is a skeleton without actual logic that would set a test
case or suite to be skipped.
The mechanism uses a protocol to extend test suites and
A test case or suite may require certain capabilities to be present in
the tested environment. Add the basic infrastructure for checking the
support status of capabilities:
* The Capability ABC defining the common capability API
* Extension of the TestProtocol with required capabilities (each test
We currently assume there are two links between the SUT and TG nodes,
but that's too strict, even for some of the already existing test cases.
Add support for topologies with less than two links.
For topologies with no links, dummy ports are used. The expectation is
that test suites or cases that
Add decorators for functional and performance test cases. These
decorators add attributes to the decorated test cases.
With the addition of decorators, we change the test case discovery
mechanism from looking at test case names according to a regex to simply
checking an attribute of the function a
Regular Python enumerations create only one instance for members with
the same value, such as:
class MyEnum(Enum):
foo = 1
bar = 1
MyEnum.foo and MyEnum.bar are aliases that return the same instance.
DTS needs to return different instances in the above scenario so that we
can map capabili
The question is when the exception would be raised, or, in other words,
what should we do when hasattr(cls, name) is False. If I understand this
correctly, is it's False, then name is not among the flags and that
means testpmd returned an unsupported flag, which shouldn't happen, but
if it doe
Members Attending: 11/11
- Aaron Conole
- Bruce Richardson
- Hemant Agrawal
- Honnappa Nagarahalli
- Jerin Jacob
- Kevin Traynor
- Konstantin Ananyev
- Maxime Coquelin
- Morten Brørup
- Stephen Hemminger
- Thoma
On 9/23/24 12:26, Niall Meade wrote:
Addressed a specific overflow issue in the eth_dev_adjust_nb_desc()
function where the uint16_t variable nb_desc would overflow when its
value was greater than (2^16 - nb_align). This overflow caused nb_desc
to incorrectly wrap around between 0 and nb_align-1,
On 2024-09-19 15:13, Pavan Nikhilesh Bhagavatula wrote:
From: pbhagavat...@marvell.com
Sent: Tuesday, September 17, 2024 3:11 AM
To: jer...@marvell.com; sthot...@marvell.com; Sevincer, Abdullah
; hemant.agra...@nxp.com;
sachin.sax...@oss.nxp.com; Van Haaren, Harry
;
mattias.ronnb...@ericsson.c
On Fri, Sep 20, 2024 at 2:38 PM Morten Brørup
wrote:
>
> Some applications want to omit the trace feature.
> Either to reduce the memory footprint, to reduce the exposed attack
> surface, or for other reasons.
>
> This patch adds an option in rte_config.h to include or omit trace in the
> build.
> On Sep 20, 2024, at 23:09, Maxime Coquelin wrote:
>
> External email: Use caution opening links or attachments
>
>
> This patch enables VDUSE reconnection support making use of
> the newly introduced reconnection mechanism in Vhost
> library.
>
> At DPDK VDUSE device creation time, there a
> On Sep 20, 2024, at 23:09, Maxime Coquelin wrote:
>
> External email: Use caution opening links or attachments
>
>
> This patch introduces a way for backend to keep track
> of the needed information to be able to reconnect without
> frontend cooperation.
>
> It will be used for VDUSE, whic
62 matches
Mail list logo
