Support asymmetric crypto operations.
Signed-off-by: Gowrishankar Muthukrishnan
Acked-by: Akhil Goyal
---
v5:
- doc update
---
doc/guides/sample_app_ug/vhost_crypto.rst | 5 +++
examples/vhost_crypto/main.c | 50 +--
2 files changed, 43 insertions(+), 12 delet
Hi,
From: Junfeng Guo
Sent: Thursday, February 20, 2025 9:08 AM
To: dev@dpdk.org
Cc: Dariusz Sosnowski; Slava Ovsiienko; Bing Zhao; Ori Kam; Suanming Mou; Matan
Azrad; Minggang(Gavin) Li; Jiawei(Jonny) Wang; sta...@dpdk.org
Subject: [PATCH] net/mlx5: fix err message overwrite for actions transla
>
> 在 2025/2/20 17:41, Konstantin Ananyev 写道:
> > Hi
> >
> >> Hi all,
> >>
> >> Kindly ping for review.
> >>
> >>
> >> 在 2024/12/19 15:53, Huisong Li 写道:
> >>> The queue_list_entry structure data is used in rx_callback of io path
> >>> when enable PMD Power Management. However its memory is curr
From: Shai Brandes
This patch enables the configuration of a mutable RSS table size, allowing
it to adapt to the device's capabilities instead of relying on a fixed 128
hardcoded value.
Signed-off-by: Yosef Raisman
Signed-off-by: Shai Brandes
Reviewed-by: Amit Bernstein
---
doc/guides/rel_no
From: Shai Brandes
This update upgrades the ena-com module to the most recent version,
which includes various cosmetic changes and the removal of obsolete
routines.
Signed-off-by: Shai Brandes
---
drivers/net/ena/base/ena_com.c| 24 ++-
drivers/net/ena/base/ena_
From: Shai Brandes
Dear DPDK maintainers and contributors,
We would appreciate your review of the ENA driver 2.12.0 patchset,
which includes readiness for the upcoming hardware generations.
Your feedback and insights on the patchset would be highly appreciated.
Thank you in advance for your time
From: Shai Brandes
Upgrade ENA driver version to 2.12.0.
Signed-off-by: Shai Brandes
---
drivers/net/ena/ena_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 705b0b3a85..038b10eec2 100644
--- a/driver
From: Shai Brandes
Remove deprecated asynchronous event notifications that may arrive
from the device. The removal ensures that only supported event
handling mechanisms are in use.
Signed-off-by: Wajeeh Atrash
Signed-off-by: Shai Brandes
---
drivers/net/ena/ena_ethdev.c | 51 -
在 2025/2/24 19:12, Konstantin Ananyev 写道:
在 2025/2/20 17:41, Konstantin Ananyev 写道:
Hi
Hi all,
Kindly ping for review.
在 2024/12/19 15:53, Huisong Li 写道:
The queue_list_entry structure data is used in rx_callback of io path
when enable PMD Power Management. However its memory is curren
Add a facility to create and ues a temporary runtime directory on the
remote nodes for each test run.
This requires adding setup and teardown steps for Node. To ensure that
setup is not performed twice in case of a loopback setup where the same
node is used both as SUT and TG, an internal state ab
Hi there,
sending in a solution that could enable DTS to bind drivers for
DPDK-based apps, like traffic generators, that lack the DPDK build to
gather dpdk-devbind.py from.
Best,
Luca
Luca Vizzarro (6):
dts: ensure runtime working directory
dts: use topology in smoke tests
dts: add tmp dir
Allow the DPDKRuntimeEnvironment to work without an associated DPDK
build. This is useful when executing any program that is based on DPDK,
so that the common runtime functionality can be shared.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/context.py
The smoke tests where wrongly using all the node's ports for the smoke
tests. Ensure only the ports specified in the current topology are used.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/tests/TestSuite_smoke_tests.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
The setup and teardown was previously missed when reworking the
execution internals into states. Add back the traffic generator setup
and teardown respectively in TestRunSetup and TestRunTeardown.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/test_run.py
Add a facility to bind drivers for DPDK on runtimes where the DPDK build
is lacking. This is needed to execute any program that is based on DPDK,
but is unrelated to the testing.
Bugzilla ID: 1420
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/remote_session/dpdk.p
After loading the user's settings and supplied configuration files, make
the dts folder the working directory for the runner.
This allows DTS to easily access the local DPDK tree.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/main.py | 6 ++
1 file changed, 6 insertions
On 21/02/2025 17:04, Bruce Richardson wrote:
The mbuf cleanup function is not being called correctly on close of the
driver, due to the vector_tx flag not being set correctly on each Tx
queue. Normally, this causes no visible problems, because the default is
to have the vector flag unset, which m
The RFC states that the first 96 bits should be compared, not the first
32 bits.
Add unit tests to ensure this works.
Cc: sta...@dpdk.org
Fixes: 3d6d85f58c1c ("net: add utilities for well known IPv6 address types")
Reported-by: Junxiao Shi
Closes: https://bugs.dpdk.org/show_bug.cgi?id=1664
Signe
Some small cleanups and reformatting of the iavf driver documentation
to make it easier to read. Also add an additional note on large VF
support.
Bruce Richardson (3):
doc/nics: improve structure of Intel VF section
doc/nics: rework notes for iavf use
doc/nics: document how to create large V
The Intel VF section in the NIC guide document had descriptions of the
various drivers at a very deep level of the document hierarchy. Improve
the structure of the doc by giving those per-driver subsections their
own top-level section. Also create a new top-level section for details
of hypervisor u
When describing the iavf driver, there was a lot of text in a single
note block which could be clarified by splitting it up into separate
sections and appropriate bulletted lists.
Signed-off-by: Bruce Richardson
---
doc/guides/nics/intel_vf.rst | 60 +---
1 file c
Document the additional step necessary to create large VFs for iavf when
using ice kernel driver v1.13.7 or later.
Signed-off-by: Bruce Richardson
---
doc/guides/nics/intel_vf.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/nics/intel_vf.rst b/doc/guides/nics/intel_vf.rst
On Mon, 24 Feb 2025 14:10:39 +0800
Junlong Wang wrote:
> Hi maintainer,
> Do these warnings need to be resolved?
These warnings do need to be resolved, but not necessarily a blocker for
inclusing now.
>
> The usage of this code is also supported by the standard C language,
No VLA is a
On Sun, 23 Feb 2025 22:41:23 +0100
Ariel Otilibili wrote:
> rte_pcapng_close() might dereference a null pointer; as example,
> PVS-Studio gives its usage in test_pcapng.c: indeed, that call to
> rte_pcapng_close() might receive a null pointer.
>
> In that case, rte_errno is set to EINVAL. The AP
On Mon, 24 Feb 2025 13:40:08 +0200
wrote:
> @@ -768,8 +768,6 @@ int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, u32
> mtu);
> *
> * @return: 0 on Success and negative value otherwise.
> */
> -int ena_com_get_offload_settings(struct ena_com_dev *ena_dev,
> -
On Fri, Feb 21, 2025 at 09:58:28PM +0100, Morten Brørup wrote:
> > From: Andre Muezerie [mailto:andre...@linux.microsoft.com]
> > Sent: Friday, 21 February 2025 20.53
> >
> > Compiling with MSVC results in the error below:
> >
> > app/test/test_ring_perf.c(197): error C7712: address argument to a
Compiling with MSVC results in warnings like the one below:
app/test-pmd/csumonly.c(1085): warning C4477: 'printf' : format string
'%d' requires an argument of type 'int',
but variadic argument 1 has type 'uint64_t'
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
---
app/tes
Compiling with MSVC results in the warning below:
app/test/test_alarm.c(54): warning C4090: 'function':
different '_Atomic' qualifiers
The fix is to use a macro to explicitly drop the qualifier.
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
---
app/test/test_alarm.c | 12
There's no MSVC equivalent for compiler extension __builtin_constant_p,
but the same result can be obtained through a clever expression using
_Generic.
This patch redefines the macro __rte_constant when msvc is used and uses
it as a replacement for __builtin_constant_p.
Signed-off-by: Andre Mueze
Compiling with MSVC results in warnings like below:
app/test-pmd/cmdline.c(9023): warning C5101: use of preprocessor
directive in function-like macro argument list is undefined behavior
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
---
app/test-pmd/cmdline.c | 12
Compiling with MSVC results in warnings like the one below:
app/test-pmd/util.c(201): warning C4334: '<<': result of 32-bit shift
implicitly converted to 64 bits (was 64-bit shift intended?)
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
Acked-by: Bruce Richardson
---
app/test
Enabled "app" directory to be compiled with MSVC along with all its
contents.
Removed flag Wno-deprecated-declarations which is not needed anymore.
Signed-off-by: Andre Muezerie
Acked-by: Bruce Richardson
Acked-by: Chengwen Feng
---
app/meson.build | 4
app/test-pmd/meson.build
Compiling with MSVC results in the warning below:
app/test-pmd/cmdline_flow.c(13964): warning C4098: 'cmd_set_raw_parsed':
'void' function returning a value
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
Acked-by: Bruce Richardson
---
app/test-pmd/cmdline_flow.c | 12 -
v6:
- remove "volatile" and use rte_atomic_xxx to access the RTE_ATOMIC
qualified lcore_count, which is cleaner than using "volatile".
v5:
- add "volatile" to declaration of lcore_count in test_ring_perf.c.
v4:
- add explanation about the expression used in __rte_constant during v3.
v3:
- use
There's no MSVC equivalent for compiler extension __builtin_constant_p,
so a workaround is needed.
Signed-off-by: Andre Muezerie
Acked-by: Chengwen Feng
---
app/test/test_memcpy_perf.c | 54 ++---
1 file changed, 27 insertions(+), 27 deletions(-)
diff --git a/ap
Compiling with MSVC results in the error below:
app/test/test_ring_perf.c(197): error C7712: address argument to atomic
operation must be a pointer to an atomic integer,
'volatile unsigned int *' is not valid
The fix is to mark lcore_count as atomic.
Signed-off-by: Andre Muezerie
Signed
Compiling with MSVC results in errors like the one below:
app/test-pmd/cmdline_flow.c(8819): error C2099: initializer
is not a constant
Signed-off-by: Andre Muezerie
Signed-off-by: Chengwen Feng
---
app/test-pmd/cmdline_flow.c | 10 +++---
1 file changed, 3 insertions(+), 7 deletions(-
Hello,
This patch addresses a warning found by static analysis; courtesy of Stephen
Hemminger.
Thank you,
---
v3
* silently ignores NULL pointers (Stephen Hemminger)
* changed commit title
v2
(https://inbox.dpdk.org/dev/20250223214123.447579-1-ariel.otilib...@6wind.com/)
* rather change docume
rte_pcapng_close() might dereference a null pointer; as example,
PVS-Studio gives its usage in test_pcapng.c: indeed, that call to
rte_pcapng_close() might receive a null pointer.
Link: https://pvs-studio.com/en/docs/warnings/v522/
Link:
https://github.com/DPDK/dpdk/blob/e5176f23ae8b31437c3e5eb87
Hello Dmitry, hello Stephen;
On Mon, Feb 24, 2025 at 4:42 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> The convention (back from Unix) is that errno is only set on failure.
> Simpler fix would just to silently ignore NULL case.
>
Thanks for your feedback. Here they are addressed,
> -Original Message-
> From: Nithin Dabilpuram
> Sent: Monday, February 24, 2025 3:22 PM
> To: Jerin Jacob ; Nithin Kumar Dabilpuram
> ; Kiran Kumar Kokkilagadda
> ; Sunil Kumar Kori ; Satha
> Koteswara Rao Kottidi ; Harman Kalra
>
> Cc: dev@dpdk.org
> Subject: [PATCH v3 01/33] net/cnx
Added support for symmetric algorithm SM4 GCM in cryptodev.
Signed-off-by: Brian Dooley
---
v5:
Change with GCM mode to in GCM mode
---
doc/guides/cryptodevs/features/default.ini | 2 ++
lib/cryptodev/rte_crypto_sym.h | 4 +++-
lib/cryptodev/rte_cryptodev.c | 3 +
Added SM4-GCM tests for the AESNI MB PMD.
Signed-off-by: Brian Dooley
---
v4:
Remove MAX_AAD_LENGTH array length from test vectors
---
app/test/test_cryptodev.c | 158 +
app/test/test_cryptodev_aead_test_vectors.h | 708
2 files changed, 866 ins
This patch introduces SM4 GCM algorithm support to the AESNI_MB PMD.
SM4 GCM is available in the v2.0 release of Intel IPsec MB.
Signed-off-by: Brian Dooley
---
v2:
Added aad to cpu job params
Added ipsec mb version checks
v3:
Fix naming for patchwork
v4:
Remove cr
> -Original Message-
> From: Vamsi Krishna
> Sent: Tuesday, February 11, 2025 10:36 PM
> To: dev@dpdk.org
> Cc: Jerin Jacob ; Vamsi Krishna Attunuru
>
> Subject: [PATCH v0 1/1] dma/cnxk: support pending count per HW DMA
> channel
>
> From: Vamsi Attunuru
>
> Adds code changes to mai
101 - 145 of 145 matches
Mail list logo
