>> >>> This RFC introduces a new API,
>> >>> rte_event_eth_rx_adapter_queues_add(),
>> >>> designed to enhance the flexibility of configuring multiple Rx
>> >>> queues in eventdev Rx adapter.
>> >>>
>> >>> The existing rte_event_eth_rx_adapter_queue_add() API supports
>> >>> adding multiple queues
在 2025/1/15 19:31, Thomas Monjalon 写道:
15/01/2025 04:41, Huisong Li:
If application verify the validity of the port id or configure this port in
the new event callback, application may happen to the port id is invalid.
Actually, when application receive a new event from one port, the port is
在 2025/1/15 19:36, Thomas Monjalon 写道:
15/01/2025 04:41, Huisong Li:
The rte_eth_dev_socket_id() and rte_eth_dev_owner_*() can be used after
allocating an ethdev. So this patch relaxes the conditions for using them.
You should be more explicit:
"during probing, before it becomes generally ava
Thanks, applied to next-dts with the git fixup.
Thanks, applied to next-dts.
Hi Dean. This looks good to me, though I agree Luca's point about
show_port_info_all() is compelling. We are enforcing the paired topology
today but may not always.
I don't view it as something which should block the testsuite, but this is
worth tracking. Can you make a Bugzilla ticket for this, a
On Wed, Jan 15, 2025 at 11:30 AM Dean Marx wrote:
> +def disable_promisc_setup(self, testpmd: TestPmdShell, port_id: int)
> -> TestPmdShell:
> +"""Sets up testpmd shell config for cases where promisc mode is
> disabled.
> +
> +Args:
> +testpmd: Testpmd session that
provided dev simple rx implementations.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 1 +
doc/guides/nics/zxdh.rst | 1 +
drivers/net/zxdh/zxdh_ethdev.c| 1 +
drivers/net/zxdh/zxdh_rxtx.c | 313 ++
drivers/net/zxdh/zxdh_r
provided rss hash config/update, reta update/get ops.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 3 +
doc/guides/nics/zxdh.rst | 1 +
drivers/net/zxdh/zxdh_ethdev.c | 52
drivers/net/zxdh/zxdh_ethdev.h | 3 +
drivers/net/zxdh/zxdh_ethdev_op
provided vlan filter, vlan offload ops.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 3 +
doc/guides/nics/zxdh.rst | 3 +
drivers/net/zxdh/zxdh_ethdev.c | 40 +-
drivers/net/zxdh/zxdh_ethdev_ops.c | 223 +
drivers/net/zxd
provided promiscuous/allmulticast ops.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 2 +
doc/guides/nics/zxdh.rst | 2 +
drivers/net/zxdh/zxdh_ethdev.c | 21 ++-
drivers/net/zxdh/zxdh_ethdev.h | 2 +
drivers/net/zxdh/zxdh_ethdev_ops.c | 128 ++
provided mac set/add/remove ops.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 2 +
doc/guides/nics/zxdh.rst | 2 +
drivers/net/zxdh/zxdh_common.c | 24 +++
drivers/net/zxdh/zxdh_common.h | 1 +
drivers/net/zxdh/zxdh_ethdev.c | 33 -
drive
provided link info update, set link up /down,
and link intr.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 2 +
doc/guides/nics/zxdh.rst | 3 +
drivers/net/zxdh/meson.build | 1 +
drivers/net/zxdh/zxdh_ethdev.c | 22 +++-
drivers/net/zxdh/zxdh_et
provided dev simple tx implementations.
Signed-off-by: Junlong Wang
---
drivers/net/zxdh/meson.build | 1 +
drivers/net/zxdh/zxdh_ethdev.c | 22 ++
drivers/net/zxdh/zxdh_queue.h | 26 ++-
drivers/net/zxdh/zxdh_rxtx.c | 396 +
drivers/net/zxdh/zxdh_rxtx.h
dev start/stop implementations, start/stop the rx/tx queues.
Signed-off-by: Junlong Wang
---
doc/guides/nics/features/zxdh.ini | 2 +
doc/guides/nics/zxdh.rst | 2 +
drivers/net/zxdh/zxdh_ethdev.c| 71
drivers/net/zxdh/zxdh_pci.c | 21 +++
driver
rx/tx queue setup and intr enable implementations.
Signed-off-by: Junlong Wang
---
drivers/net/zxdh/zxdh_ethdev.c | 4 +
drivers/net/zxdh/zxdh_queue.c | 149 +
drivers/net/zxdh/zxdh_queue.h | 33
3 files changed, 186 insertions(+)
diff --git a/drive
delete port tables in host.
Signed-off-by: Junlong Wang
---
drivers/net/zxdh/zxdh_ethdev.c | 18 ++
drivers/net/zxdh/zxdh_msg.h| 1 +
drivers/net/zxdh/zxdh_np.c | 103 +
drivers/net/zxdh/zxdh_np.h | 9 +++
drivers/net/zxdh/zxdh_tables.c | 33
(np)network processor release resources in host.
Signed-off-by: Junlong Wang
---
drivers/net/zxdh/zxdh_ethdev.c | 48
drivers/net/zxdh/zxdh_np.c | 470 +
drivers/net/zxdh/zxdh_np.h | 107
3 files changed, 625 insertions(+)
diff --git a/driv
V7:
- resolved warning '-Waddress-of-packed-member'
in function 'zxdh_dev_rss_reta_update'.
V6:
- Remove unnecessary __rte_packed in the virtqueue structure and others.
- Remove Some blank before or after log message,
and remove some end with period in log message.
V5:
- Simplify
It was a common pattern to have "GCC diagnostic ignored" pragmas
sprinkled over the code and only activate these pragmas for certain
compilers (gcc and clang). Clang supports GCC's pragma for
compatibility with existing source code, so #pragma GCC diagnostic
and #pragma clang diagnostic are synonym
It was a common pattern to have "GCC diagnostic ignored" pragmas
sprinkled over the code and only activate these pragmas for certain
compilers (gcc and clang). Clang supports GCC's pragma for
compatibility with existing source code, so #pragma GCC diagnostic
and #pragma clang diagnostic are synonym
It was a common pattern to have "GCC diagnostic ignored" pragmas
sprinkled over the code and only activate these pragmas for certain
compilers (gcc and clang). Clang supports GCC's pragma for
compatibility with existing source code, so #pragma GCC diagnostic
and #pragma clang diagnostic are synonym
It was a common pattern to have "GCC diagnostic ignored" pragmas
sprinkled over the code and only activate these pragmas for certain
compilers (gcc and clang). Clang supports GCC's pragma for
compatibility with existing source code, so #pragma GCC diagnostic
and #pragma clang diagnostic are synonym
Members Attending
-
Aaron Conole
Bruce Richardson
Hemant Agrawal
Kevin Traynor
Konstantin Ananyev (chair)
Maxime Coquelin
Morten Brørup
Stephen Hemminger
Thomas Monjalon
NOTE: The technical board meetings are on every second Wednesday at 3pm
UTC. Meetings are public, and
Port over unified packet type flag testing suite from old DTS.
According to DPDK documentation, each Poll Mode Driver should reserve 32
bits of packet headers for unified packet type flags. These flags serve
as an identifier for user applications, and are divided into
subcategories: L2, L3, L4, tu
Port over unified packet testing suite from old DTS. This suite
tests the ability of the PMD to recognize valid or invalid packet flags.
Signed-off-by: Dean Marx
Reviewed-by: Jeremy Spewock
---
dts/tests/TestSuite_uni_pkt.py | 300 +
1 file changed, 300 insertion
Add rx_vxlan_port add/rm method to testpmd shell for adding
or removing a vxlan id to the specified port filter list.
Signed-off-by: Dean Marx
Reviewed-by: Jeremy Spewock
---
dts/framework/remote_session/testpmd_shell.py | 21 +++
1 file changed, 21 insertions(+)
diff --git a/d
https://bugs.dpdk.org/show_bug.cgi?id=1608
Kevin Traynor (ktray...@redhat.com) changed:
What|Removed |Added
CC||ktray...@redhat.com
https://bugs.dpdk.org/show_bug.cgi?id=1610
Kevin Traynor (ktray...@redhat.com) changed:
What|Removed |Added
Resolution|--- |WONTFIX
Status|
> -Original Message-
> From: Shijith Thotton
> Sent: Monday, January 13, 2025 5:36 PM
> To: Shijith Thotton ; Naga Harish K, S V
> ; dev@dpdk.org
> Cc: Pavan Nikhilesh Bhagavatula ; Pathak,
> Pravin ; Hemant Agrawal
> ; Sachin Saxena ;
> Mattias R_nnblom ; Jerin Jacob
> ; Liang Ma ; Mcca
On Wed, Jan 15, 2025 at 05:13:22PM +0100, David Marchand wrote:
> Hello Andre,
>
> On Fri, Jan 10, 2025 at 11:17 PM Andre Muezerie
> wrote:
> >
> > MSVC struct packing is not compatible with GCC. Different alternatives
> > were considered:
> >
> > 1) Have a macro __RTE_PACKED(decl) to which the s
On Wed, Jan 15, 2025 at 5:42 PM David Marchand
wrote:
>
> On Wed, Jan 15, 2025 at 1:59 PM Maxime Coquelin
> wrote:
> >
> > This patch refactors the error handling in the Vhost async
> > dequeue path to ease its maintenance and readability.
> >
> > Suggested-by: David Marchand
> > Signed-off-by:
On Wed, Jan 15, 2025 at 1:59 PM Maxime Coquelin
wrote:
>
> With previous refactoring, we can now simplify the RARP
> packet injection handling in both the sync and async
> dequeue paths.
>
> Signed-off-by: Maxime Coquelin
> ---
> lib/vhost/virtio_net.c | 42 ++
On Wed, Jan 15, 2025 at 1:59 PM Maxime Coquelin
wrote:
>
> This patch refactors the error handling in the Vhost async
> dequeue path to ease its maintenance and readability.
>
> Suggested-by: David Marchand
> Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
--
David Marchand
On Wed, Jan 15, 2025 at 1:59 PM Maxime Coquelin
wrote:
>
> This patch refactors the error handling in the Vhost
> dequeue path to ease its maintenance and readability.
>
> Suggested-by: David Marchand
> Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
--
David Marchand
On Wed, Jan 15, 2025 at 1:59 PM Maxime Coquelin
wrote:
>
> This patch fixes the rte_vhost_dequeue_burst return value
> when the virtqueue is not ready. Without it, a discrepancy
> between the packet array and its size is faced by the caller
> of this API when the virtqueue is not ready.
>
> Fixes:
lgtm. thanks!
Reviewed-by: Luca Vizzarro
Suite for testing ability of Poll Mode Driver to turn promiscuous
mode on/off, allmulticast mode on/off, and show expected behavior
when sending packets with known, unknown, broadcast, and multicast
destination MAC addresses.
Signed-off-by: Dean Marx
---
dts/tests/TestSuite_dynamic_config.py | 1
Add set multicast function for changing allmulticast mode
within testpmd shell.
Signed-off-by: Dean Marx
---
dts/framework/remote_session/testpmd_shell.py | 24 +++
1 file changed, 24 insertions(+)
diff --git a/dts/framework/remote_session/testpmd_shell.py
b/dts/framework/remot
On 15/01/2025 15:41, Dean Marx wrote:
Also, I noticed after submitting this first version that suites where
some cases are skipped and the rest pass have a 100% pass rate in the
results report. Should this also mean that suites where every case is
skipped have a 100% pass rate? Right now this rep
Hello Andre,
On Fri, Jan 10, 2025 at 11:17 PM Andre Muezerie
wrote:
>
> MSVC struct packing is not compatible with GCC. Different alternatives
> were considered:
>
> 1) Have a macro __RTE_PACKED(decl) to which the struct/union is passed
>and the macro would define the struct/union with the ap
This suite tests the ability of the Poll Mode Driver
to enable and disable Rx/Tx queues on a port.
The verify argument in the deferred start method has been excluded
due to a bug in testpmd, which renders checking port queue info
for deferred start status impossible within a single method.
Signed-
Add a test suite to replace hello_world which simply
starts and stops a testpmd session. The user can use
this as a confidence check to verify their configuration.
Signed-off-by: Dean Marx
Reviewed-by: Paul Szczepanek
Reviewed-by: Luca Vizzarro
---
dts/framework/test_suite.py| 10 +
Add condition to results.json pass rate generation
method which returns 0 as the pass rate when the suite
is skipped, rather than causing a divide by 0 error.
Fixes: 9f8a257235ac ("dts: improve test run result statistics")
Signed-off-by: Dean Marx
---
dts/framework/test_result.py | 11 -
Also, I noticed after submitting this first version that suites where
some cases are skipped and the rest pass have a 100% pass rate in the
results report. Should this also mean that suites where every case is
skipped have a 100% pass rate? Right now this reports a 0% when the
entire suite is skipp
Got it, I can implement this asap
Got it, I can implement this asap. I also noticed after submitting
that test suites with skipped cases still report a 100% pass rate if
the non skipped cases still pass, so I'll probably change the pass
rate to 100% for when an entire suite is skipped unless anyone
disagrees.
Makes sense, I'll send out a new version soon
Alright scratch that then. The only thing I'd change Dean is the name of
the method. Saying testsuite is somewhat redundant, I think just log fine.
This has now been reworked and superseded by series-34355.
Release status meeting minutes 2025-01-14
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* ARM
* Intel
* Marvell
* Red Hat
Release Dates
-
The following are the proposed working dates for 25.03:
- Propo
From: Nicholas Pratte
Rework 'lcores' and 'memory_channels' into a new 'dpdk_config'
subsection in an effort to make these attributes SUT specific; the
traffic generator, more often than not, does not need this information.
Ideally, if such information is needed, then it will be listed in the
'tr
From: Nicholas Pratte
The build options configuration contained redundant fields that were not
in use, and there is no future scope for their use.
Bugzilla ID: 1360
Signed-off-by: Nicholas Pratte
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/conf.yaml
To avoid the creation of a big monolithic configuration file, nodes and
test runs are now split into distinct files. This also allows
flexibility to run different test runs on the same nodes.
Since there are now 2 distinct configuration files, there are also 2
command line arguments to specify the
The configuration requires the user to explicitly set the requested test
suites in the files. Sometimes we want to run all the test suites and
don't want to manually specify all of them. It is therefore reasonable
to change the default behaviour to automatically run all the available
test suites if
The current handling of the configuration loading is inconsistent. After
the whole configuration is loaded, if there are any CLI or environment
overrides set, the code forcefully modifies the frozen configuration to
use them.
This changes the handling by passing the environment/CLI settings as
par
From: Nicholas Pratte
The 'arch' attribute in the conf.yaml is unnecessary, as this can be
readily discovered directly from any given node.
Bugzilla ID: 1360
Signed-off-by: Nicholas Pratte
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/conf.yaml
From: Nicholas Pratte
To further the simplification of the user configuration, use_first_core
can be inferred from the lcores. If the user explicitly includes the
core 0 in the lcores range, it will only then be used.
Bugzilla ID: 1360
Signed-off-by: Nicholas Pratte
Signed-off-by: Luca Vizzarr
Hi there,
sending in a v3, which also merges the split nodes and test runs
configuration into this series.
v3:
- rebase
- rework code for Pydantic
- fixed architecture bug for arm
- updated doc pages
- implemented CLI overrides in Pydantic
- changed default behaviour of test suites config field
-
Signed-off-by: Luka Jankovic
---
.../tools/img/eventdev_atomic_queue_test.svg | 1701 +
doc/guides/tools/testeventdev.rst | 92 +
2 files changed, 1793 insertions(+)
create mode 100644 doc/guides/tools/img/eventdev_atomic_queue_test.svg
diff --git a/doc/guides/too
Add an atomic queue test based on the order queue test that exclusively uses
atomic queues.
This makes it compatible with event devices such as the distributed software
eventdev.
The test detects if port maintenance is required.
To verify atomicity, a spinlock is set up for each combination of
With previous refactoring, we can now simplify the RARP
packet injection handling in both the sync and async
dequeue paths.
Signed-off-by: Maxime Coquelin
---
lib/vhost/virtio_net.c | 42 ++
1 file changed, 18 insertions(+), 24 deletions(-)
diff --git a/l
This patch refactors the error handling in the Vhost async
dequeue path to ease its maintenance and readability.
Suggested-by: David Marchand
Signed-off-by: Maxime Coquelin
---
lib/vhost/virtio_net.c | 31 ++-
1 file changed, 14 insertions(+), 17 deletions(-)
diff -
This patch refactors the error handling in the Vhost
dequeue path to ease its maintenance and readability.
Suggested-by: David Marchand
Signed-off-by: Maxime Coquelin
---
lib/vhost/virtio_net.c | 27 ---
1 file changed, 12 insertions(+), 15 deletions(-)
diff --git a/lib
This patch fixes the rte_vhost_dequeue_burst return value
when the virtqueue is not ready. Without it, a discrepancy
between the packet array and its size is faced by the caller
of this API when the virtqueue is not ready.
Fixes: 9fc93a1e2320 ("vhost: fix virtqueue access check in datapath")
Cc: s
This series starts with a fix for a regression in the Vhost
dequeue error path.
The other patches improve the error handling to reduce the
chance of such regressions in the future.
Changes in v2:
==
- Add RARP handling refactoring
Maxime Coquelin (4):
vhost: fix missing packets cou
From: Vamsi Attunuru
In the existing DPI mailbox structure, one of the fields
spans a 64-bit boundary, making it appear unusual and
complicatng extraction using bit operations.
Patch enlarges csize fields to ensure that mailbox fields
are correctly positioned.
Fixes: b6e395692b6d ("common/cnxk:
I think it makes sense to keep logger private and separate. I can see
one reason - we can change the logger in the framework without affecting
the tests. We can also decide to not use the logger to print test suites
messages - for example save per test artifacts.
On 15/01/2025 10:17, Luca Vizzarro
On 2025/1/15 17:38, Morten Brørup wrote:
>> From: huangdengdui [mailto:huangdeng...@huawei.com]
>> Sent: Wednesday, 15 January 2025 07.52
>>
>> On 2025/1/15 0:39, Morten Brørup wrote:
>>> mbuf: add fast free bulk function
>>>
>>> When putting an mbuf back into its mempool, there are certain
>> re
For fixing a warning from devtools/check-git-log.sh
Link: https://inbox.dpdk.org/dev/20250114153419.0a2753ab@hermes.local/
Signed-off-by: Ariel Otilibili
---
.mailmap | 1 +
1 file changed, 1 insertion(+)
diff --git a/.mailmap b/.mailmap
index 818798273fe1..735a635fc7f9 100644
--- a/.mailmap
++
Coverity issue: 38
Fixes: a62ec90522a6 ("net/sfc: add port representors infrastructure")
Signed-off-by: Ariel Otilibili
---
drivers/net/sfc/sfc_repr.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/net/sfc/sfc_repr.c b/drivers/net/sfc/sfc_repr.c
index c2e5d4eb9e28..a0712bf9fb90
Coverity issue: 451221
Fixes: 16158f349000 ("examples/flow_filtering: introduce use cases snippets")
Signed-off-by: Ariel Otilibili
---
examples/flow_filtering/snippets/snippet_match_ipv4.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/examples/flow_filtering/snippets/snippet_match_ipv4.c
b
Hello,
This series clears out the Coverity issues 38 & 451221.
Being one-liners, the series is backported onto stable.
Thank you,
Link:
https://inbox.dpdk.org/dev/cajfav8yeawsx2kdianwukx7zsvtenrvovjnzaoq_ocdzm8z...@mail.gmail.com/
---
v2
* fixed the SHAs given for Fixes
* added a .mailmap
15/01/2025 04:41, Huisong Li:
> The rte_eth_dev_socket_id() and rte_eth_dev_owner_*() can be used after
> allocating an ethdev. So this patch relaxes the conditions for using them.
You should be more explicit:
"during probing, before it becomes generally available and considered as valid".
Should
15/01/2025 04:41, Huisong Li:
> If application verify the validity of the port id or configure this port in
> the new event callback, application may happen to the port id is invalid.
>
> Actually, when application receive a new event from one port, the port is
> not fully probed and is just in al
On Tue, Jan 14, 2025 at 08:27:17PM -0800, Andre Muezerie wrote:
> It was a common pattern to have "GCC diagnostic ignored" pragmas
> sprinkled over the code and only activate these pragmas for certain
> compilers (gcc and clang). Clang supports GCC's pragma for
> compatibility with existing source
On Tue, Jan 14, 2025 at 11:20:05AM -0800, Andre Muezerie wrote:
> On Wed, Jan 08, 2025 at 09:20:27AM +, Bruce Richardson wrote:
> > On Tue, Jan 07, 2025 at 06:46:48PM -0800, Andre Muezerie wrote:
> > > On Mon, Jan 06, 2025 at 11:00:15AM +, Bruce Richardson wrote:
> > > > On Fri, Jan 03, 202
15/01/2025 00:34, Stephen Hemminger:
> Also, please send a patch to add your address to the .mailmap file.
> Right now, check-git-log.sh script is complaining.
In general we don't add mailmap entries as a separate commit.
You (author or tree maintainer) can add it in the first patch.
This patch refactors the error handling in the Vhost async
dequeue path to ease its maintenance and readability.
Suggested-by: David Marchand
Signed-off-by: Maxime Coquelin
---
lib/vhost/virtio_net.c | 31 ++-
1 file changed, 14 insertions(+), 17 deletions(-)
diff -
This patch refactors the error handling in the Vhost
dequeue path to ease its maintenance and readability.
Suggested-by: David Marchand
Signed-off-by: Maxime Coquelin
---
lib/vhost/virtio_net.c | 27 ---
1 file changed, 12 insertions(+), 15 deletions(-)
diff --git a/lib
This patch fixes the rte_vhost_dequeue_burst return value
when the virtqueue is not ready. Without it, a discrepancy
between the packet array and its size is faced by the caller
of this API when the virtqueue is not ready.
Fixes: 9fc93a1e2320 ("vhost: fix virtqueue access check in datapath")
Cc: s
This series starts with a fix for a regression in the Vhost
dequeue error path.
The other patches improve the error handling to reduce the
chance of such regressions in the future.
Maxime Coquelin (3):
vhost: fix missing packets count reset when not ready
vhost: rework dequeue path error hand
Sorry for the late reply Dean. This is a bit redundant in my opinion,
there is no reason not to give full access to the test suite's own
logger. I'd just change TestSuite._logger into TestSuite.logger to
indicate that it's a public member, and do what you initially did.
Looks good.
Reviewed-by: Luca Vizzarro
Hi Dean,
good timing! I actually noticed this issue last week and was going to
tackle it soon.
On 13/01/2025 21:52, Dean Marx wrote:
@@ -324,13 +324,15 @@ def generate_pass_rate_dict(self, test_run_summary) ->
dict[str, float]:
Returns:
A dictionary with the PASS/FAI
> From: huangdengdui [mailto:huangdeng...@huawei.com]
> Sent: Wednesday, 15 January 2025 07.52
>
> On 2025/1/15 0:39, Morten Brørup wrote:
> > mbuf: add fast free bulk function
> >
> > When putting an mbuf back into its mempool, there are certain
> requirements
> > to the mbuf. Specifically, some
> +/*
> + * Macro to ignore whenever a pointer is cast so as to remove a type
> + * qualifier from the target type.
> + */
This description could be better, something like the push/pop description:
"Macro to disable compiler warnings about ..."
> +#if !defined __INTEL_COMPILER && !defined RTE_TOO
From: Jun Zhang
When there are at least two VFs on a single adapter and both are used in
the same VM, each of them will register its own PTP clock. However, every
E810/E822 adapter has only one PHC clock that we use. In order to register
only one PTP clock, VFs need to identify HW and make sure t
88 matches
Mail list logo
