Plain text needs to be prehashed for RSA as well in asymmetric
crypto validation.
Fixes: 12ede9ac497f ("examples/fips_validation: support EdDSA")
Signed-off-by: Gowrishankar Muthukrishnan
---
examples/fips_validation/main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/e
Fix EdDSA signature size.
Fixes: 12ede9ac497f ("examples/fips_validation: support EdDSA")
Signed-off-by: Gowrishankar Muthukrishnan
---
examples/fips_validation/fips_validation_eddsa.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/fips_validation/fips_validati
[AMD Official Use Only - AMD Internal Distribution Only]
Hi Huisong,
LGTM!
> -Original Message-
> From: Huisong Li
> Sent: Friday, October 25, 2024 2:49 PM
> To: dev@dpdk.org
> Cc: m...@smartsharesystems.com; tho...@monjalon.net; Yigit, Ferruh
> ; anatoly.bura...@intel.com; david.h...@i
On Tue, 29 Oct 2024 08:56:20 +0800
fengchengwen wrote:
> >>>
> >>> This doesn't need to go to stable. parse_params is always single
> >>> threaded.
> >>
> >> I recommend replacing all, based on:
> >> 1\ almost at no cost.
> >> 2\ reduce analysis costs, if don't we have to analyze the callers
https://bugs.dpdk.org/show_bug.cgi?id=1572
Bug ID: 1572
Summary: The L3FWD_POWER : Cause:
rte_power_ethdev_pmgmt_queue_enable: err=-95, port=0
Product: DPDK
Version: 24.11
Hardware: All
OS: All
S
Arm recommends using -mcpu over -march and march-extensions when the
compiler supports the target CPU (neoverse-n1 etc.). Arm build so far
has been an amalgam of -mcpu and -march. When march is in use, it has
been the case so far to silently fall back to a downgraded march when
the compiler does no
Order SOC configurations and names alphabetically.
Signed-off-by: Wathsala Vithanage
Reviewed-by: Dhruv Tripathi
---
config/arm/meson.build | 72 +-
1 file changed, 36 insertions(+), 36 deletions(-)
diff --git a/config/arm/meson.build b/config/arm/meson
Acked-by: Huisong Li
在 2024/10/11 11:24, Chaoyong He 写道:
From: Long Wu
According to the check rules in the patch check script,
drivers and libraries must use the logging framework.
So standard the log message of bonding driver by using
the logging framework.
Signed-off-by: Long Wu
Reviewed-
Very useful,
Acked-by: Huisong Li
在 2024/10/11 11:24, Chaoyong He 写道:
From: Long Wu
The testpmd application can not modify the value of
dedicated hardware Rx/Tx queue size, and hardcoded
them as (128/512). This will cause the bonding port
start fail if some NIC requires more Rx/Tx descriptors
Reviewed-by: Patrick Robb
Reviewed-by: Patrick Robb
Reviewed-by: Patrick Robb
On Mon, Oct 21, 2024 at 9:50 AM Luca Vizzarro wrote:
> From: Tomáš Ďurovec
>
> Given the whole DPDK tree directory can now be copied to the nodes,
> there is no more need to use the git ref option, as the tree can be
> controlled directly by the user.
>
> Signed-off-by: Tomáš Ďurovec
> Signed-o
On Mon, Oct 21, 2024 at 9:50 AM Luca Vizzarro wrote:
>
> +
> + DPDK Build Options:
> + Arguments in this group (and subgroup) will be applied to a
> DPDKLocation when the DPDK tree, tarball or revision will be provided,
> other arguments
> + like remote source and build dir are optional. A
On Mon, Oct 21, 2024 at 9:50 AM Luca Vizzarro wrote:
> From: Tomáš Ďurovec
>
> Enable the user to use either a DPDK source tree directory or a
> tarball, with and without a pre-built build directory. These can be
> stored on either SUT node or the DTS host. The DPDK build setup or the
> pre-buil
For series,
Acked-by: Huisong Li
在 2024/10/23 16:28, Dengdui Huang 写道:
The function strerror() is insecure in a multi-thread environment.
It is better to use rte_strerror() instead of strerror().
In this patchset, only the libs and drivers are modified.
chang log:
v3->v4 fix ci error
v2->v3 d
On Mon, Oct 21, 2024 at 9:49 AM Luca Vizzarro wrote:
>
> +
> +@abstractmethod
> +def remove_remote_file(self, remote_file_path: str | PurePath, force:
> bool = True) -> None:
> +"""Remove remote file, by default remove forcefully.
> +
> +Args:
> +remote_file_pa
On 2024/10/29 7:31, Wathsala Vithanage wrote:
> Arm recommends using -mcpu over -march and march-extensions when the
> compiler supports the target CPU (neoverse-n1 etc.). Arm build so far
> has been an amalgam of -mcpu and -march. When march is in use, it has
> been the case so far to silently fal
On 2024/10/28 23:31, Stephen Hemminger wrote:
> On Mon, 28 Oct 2024 21:04:08 +0800
> fengchengwen wrote:
>
>> On 2024/10/27 11:00, Stephen Hemminger wrote:
>>> On Sat, 26 Oct 2024 18:14:40 +0800
>>> Jie Hai wrote:
>>>
Multiple threads calling the same function may cause condition
ra
Acked-by: Chengwen Feng
On 2024/10/28 21:10, David Marchand wrote:
> Caught in UNH logs for native compilation on ARM:
>
> Program /./dpdk/config/arm/armv8_machine.py found: YES
> (/./dpdk/config/arm/armv8_machine.py)
> WARNING: Project targeting '>= 0.57' but tried to use feature
Order SOC configurations and names alphabetically.
Signed-off-by: Wathsala Vithanage
Reviewed-by: Dhruv Tripathi
---
config/arm/meson.build | 72 +-
1 file changed, 36 insertions(+), 36 deletions(-)
diff --git a/config/arm/meson.build b/config/arm/meson
Arm recommends using -mcpu over -march and march-extensions when the
compiler supports the target CPU (neoverse-n1 etc.). Arm build so far
has been an amalgam of -mcpu and -march. When march is in use, it has
been the case so far to silently fall back to a downgraded march when
the compiler does no
This patch refactors the power management library, addressing uncore
power management. The primary changes involve the creation of dedicated
directories for each driver within 'drivers/power/uncore/*'. The
adjustment of meson.build files enables the selective activation
of individual drivers.
This
This patch renames the existing core power library source files
from rte_power.* to rte_power_cpufreq.* for better clarity
v9:
- documentation update
Signed-off-by: Sivaprasad Tummala
---
app/test/test_power.c| 2 +-
app/test/test_power_cpufreq.c
This patch introduces a comprehensive refactor to the core power
management library. The primary focus is on improving modularity
and organization by relocating specific driver implementations
from the 'lib/power' directory to dedicated directories within
'drivers/power/core/*'. The adjustment of m
Update maintainers for drivers/power/*.
Signed-off-by: Sivaprasad Tummala
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index e25e9465b5..df7a756612 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1753,6 +1753,7 @@ M: Anatoly Burakov
M: David Hunt
This patch introduces driver support for power management of uncore
components in AMD EPYC processors.
v9:
- documentation update
v2:
- fixed typo in comments section.
- added fabric frequency get support for legacy platforms.
Signed-off-by: Sivaprasad Tummala
---
doc/guides/prog_guide/powe
This patchset refactors the power management library, addressing both
core and uncore power management. The primary changes involve the
creation of dedicated directories for each driver within
'drivers/power/core/*' and 'drivers/power/uncore/*'.
This refactor significantly improves code organizati
After refactoring the power library, power management operations are now
consistently supported regardless of the operating environment, making
function pointer checks unnecessary and thus removed from applications.
v2:
- removed function pointer validation in l3fwd-power app.
Signed-off-by: Siv
Enhance build instructions on Windows
Signed-off-by: Andre Muezerie
---
v3:
- Recommended Meson version to use and removed mention about old bug.
- Set VS developer prompt to produce 64 bit binaries if needed.
v2:
- provide Meson version that works
.mailmap | 1
On Fri, Oct 25, 2024 at 6:35 AM Paul Szczepanek
wrote:
> Alias used to produce the link to the DTS API docs uses config data.
> The required value in config data was being added in the wrong scope.
> This moves the key addition to where the config data object is created.
>
> DTS docs need to be p
Add function attributes that allow detecting use after free,
and calling free on bad pointer at compile time.
Signed-off-by: Stephen Hemminger
---
lib/gpudev/rte_gpudev.h | 46 +
1 file changed, 24 insertions(+), 22 deletions(-)
diff --git a/lib/gpudev/rt
Use function attributes to catch misuse of GPU memory
at compile time.
v2 - remove test case where invalid pointer was being passed.
The test is now caught at compile time, making a runtime test
no longer necessary.
Stephen Hemminger (2):
test-gpudev: avoid use-after-free and free-non
Remove the test intentionally frees a bad pointer in GPU memory,
and GCC sees that at compile time now.
Fix use after free in printf.
Signed-off-by: Stephen Hemminger
---
app/test-gpudev/main.c | 16 +++-
1 file changed, 3 insertions(+), 13 deletions(-)
diff --git a/app/test-gpudev
Wrong CQE is used to get the shared Rx queue port number in
vectorized Rx burst routine. Fix the CQE indexing.
Fixes: 25ed2ebff1 ("net/mlx5: support shared Rx queue port data path")
Signed-off-by: Alexander Kozyrev
---
drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 12 ++--
drivers/net/mlx5
Pydantic models are not treated the same way as dataclasses by autodoc.
As a consequence the docstrings need to be applied directly to each
field. Otherwise the generated API documentation page would present two
entries per each field with each their own differences.
Signed-off-by: Luca Vizzarro
The introduction of TestSuiteSpec adds auto-discovery of test suites,
which are also automatically imported. This causes double imports as the
runner loads the test suites. This changes the behaviour of the runner
to load the imported classes from TestSuiteSpec instead of importing
them again.
Sig
This change brings in pydantic in place of warlock. Pydantic offers
a built-in model validation system in the classes, which allows for
a more resilient and simpler code. As a consequence of this change:
- most validation is now built-in
- further validation is added to verify:
- cross referenci
Add and enable the autodoc-pydantic sphinx extension. Pydantic models
are not correctly recognised by autodoc, causing the generated docs to
lack all the actual model information. The autodoc-pydantic sphinx
extension fixes the original behaviour by correctly formatting them.
Signed-off-by: Luca V
As part of configuration validation and deserialization improvements,
this adds pydantic as a project dependency. Pydantic is a library that
caters to all of the aforementioned needs, while improving the process
and code.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/poetry.
Since pydantic has completely replaced warlock, there is no more need to
keep it as a dependency. This removes it.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/poetry.lock| 227 +
dts/pyproject.toml | 1 -
2 files changed, 1
Hi there,
sending a v4 for the pydantic changes.
v4:
- added autodoc_pydantic due to autodoc warnings
- fixed pydantic models docstrings
- updated docs
- refactored DPDKBuildInfo and NodeInfo which didn't belong in
configuration
v3:
- removed the common FrozenModel and configured each BaseModel
The DPDKBuildInfo and NodeInfo classes, representing information
gathered in runtime, were erroneously placed in the configuration
package. This moves them in more appropriate modules.
NodeInfo, specifically, ia moved to os_session instead of node mostly
as a consequence of circular dependencies.
Currently there is a lack of a definition which identifies all the test
suites available to test. This change intends to simplify the process to
discover all the test suites and idenfity them.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/runner.py
On 10/21/24 10:47 AM, Konstantin Ananyev wrote:
NOTE UPFRONT: this version is still not ready for merging.
Missing items:
- ARM/PPC tests passing
- PG update
SORING tests pass on a P9 PowerNV system. (Results attached for review.)
Original ring_stress_autotest still fails as documented in Bug
The completion queue element size can be independently configured
to report either 64 or 128 bytes CQEs by programming cqe_sz parameter
at CQ creation. This parameter depends on the cache line size and
affects both regular CQEs and error CQEs. But the error handling
assumes that an error CQE is 64
Use the information from the CQE, not from the title packet,
for getting the number of miniCQEs in the compressed CQEs array.
This way we can avoid segfaults in the rxq_cq_decompress_v()
in case of mbuf corruption (due to double mbuf free, for example).
Fixes: 6cb559d67b ("net/mlx5: add vectorized
https://bugs.dpdk.org/show_bug.cgi?id=1313
Ali Alnubani (alia...@nvidia.com) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|-
Acked-by: Viacheslav Ovsiienko
> -Original Message-
> From: Alexander Kozyrev
> Sent: Monday, October 28, 2024 6:43 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava
> Ovsiienko ; Matan Azrad ;
> Dariusz Sosnowski ; Bing Zhao
> ; Suanming Mou
> Subject: [PATCH] net/m
> -Original Message-
> From: Stephen Hemminger
> Sent: Monday, October 28, 2024 5:52 PM
> To: Slava Ovsiienko
> Cc: dev@dpdk.org; Raslan Darawsheh ; Matan Azrad
> ; Suanming Mou
> Subject: Re: [PATCH 1/1] doc: virtual function MTU settings has no meaning
>
> On Mon, 28 Oct 2024 16:45:09
This patch makes use of the imported VDUSE headers.
The VDUSE support is now systematically built on Linux
systems, even if the build system does not support its
ioctl().
Signed-off-by: Maxime Coquelin
Acked-by: Morten Brørup
---
lib/vhost/meson.build | 5 +
lib/vhost/vduse.c | 2 +-
Use the information from the CQE, not from the title packet,
for getting the number of miniCQEs in the compressed CQEs array.
This way we can avoid segfaults in the rxq_cq_decompress_v()
in case of mbuf corruption (due to double mbuf free, for example).
Fixes: 6cb559d67b ("net/mlx5: add vectorized
The struct netlink message structure is part of the internal
TAP device API, not the netlink API itself. Use tap_ prefix
to avoid any confusion with nlmsghdr from netlink.h
Signed-off-by: Stephen Hemminger
---
drivers/net/tap/tap_flow.c| 22 +++---
drivers/net/tap/tap_netlink
The process private data is local to the process and does not
have to come from global shared memory. Using malloc() gets more
coverage with tools and has better caching.
Signed-off-by: Stephen Hemminger
---
drivers/net/tap/rte_eth_tap.c | 21 ++---
1 file changed, 10 insertions(
There is no point in using rte_memcpy to copy Ethernet addresses.
Use rte_ether_addr_copy() instead.
Signed-off-by: Stephen Hemminger
---
drivers/net/tap/rte_eth_tap.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net
The function to convert from internal representation to
netlink was using a void pointer when both caller and callee
were using same structure.
Signed-off-by: Stephen Hemminger
---
drivers/net/tap/tap_flow.c | 47 +-
1 file changed, 16 insertions(+), 31 deleti
These are some minor things in tap device that can be easily fixed.
Not urgent can wait until 25.03 release but submitting now for review.
Stephen Hemminger (4):
net/tap: replace rte_memcpy
net/tap: rename struct nlmsg to tap_nlmsg
net/tap: do not use opaque parameter in convert
net/tap: d
On Mon, Oct 28, 2024 at 6:28 PM Thomas Monjalon wrote:
>
> 24/10/2024 07:26, Ajit Khaparde:
> > On Mon, Oct 21, 2024 at 1:14 AM Sriharsha Basavapatna
> > wrote:
> > >
> > > This patch series introduces TruFlow functionality for Broadcom Thor2 NIC.
> > >
> > > TruFlow(TF) is the software library t
From: Pavan Nikhilesh
Add CN20K event port profile switch.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 1 +
drivers/event/cnxk/cn20k_worker.c | 11 +++
drivers/event/cnxk/cn20k_worker.h | 1 +
3 files changed, 13 insertions(+)
diff --git a/drivers/ev
From: Pavan Nikhilesh
Add SSO HW info mbox to get hardware capabilities, and reuse
them instead of depending on hardcoded values.
Remove redundant includes.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/roc_mbox.h | 28 ++
drivers/common/cnxk/roc_sso.c
On Mon, 28 Oct 2024 11:18:18 +0200
"Minggang Li(Gavin)" wrote:
> +- ``probe_opt_en`` parameter [int]
> +
> + A non-zero value optimizes the probe process, especially for large scale.
> + PMD will hold the IB device information internally and reuse it.
> +
> + By default, the PMD will set this
From: Pavan Nikhilesh
Add event device stop and close callback functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/drivers/event/cnxk/cn20k_eventdev.c
b/drivers/event/cnxk/cn20k_eventdev.c
On Wed, 23 Oct 2024 19:00:01 +0200
Serhii Iliushyk wrote:
> +
> + rte_memcpy(&tmp_rss_conf.rss_key, rss_conf->rss_key,
> rss_conf->rss_key_len);
Avoid use of rte_memcpy(), it has less checking that memcpy().
The only place it can make sense is in the hot data path where compiler is
From: Pavan Nikhilesh
Add CN20K start function along with few cleanup API's to maintain
sanity.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c | 103 +--
drivers/event/cnxk/cn20k_eventdev.c | 76
drivers/event/cnxk/cnxk_comm
From: Pavan Nikhilesh
Add CN20K event port quiesce function.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 60 +
1 file changed, 60 insertions(+)
diff --git a/drivers/event/cnxk/cn20k_eventdev.c
b/drivers/event/cnxk/cn20k_eventdev.c
inde
From: Pavan Nikhilesh
Update event timer base code to support configuring
HW accelerated timer arm and cancel.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/hw/tim.h| 5 ++
drivers/common/cnxk/roc_mbox.h | 38 -
drivers/common/cnxk/roc_tim.c | 84 ++
From: Pavan Nikhilesh
Add setup and release functions for event queues i.e. SSO HWGRPs.
Allocate buffers in DRAM that hold inflight events.
Register device args to modify inflight event buffer count,
HWGRP QoS and stash.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c |
From: Pavan Nikhilesh
Add CN20K event port preschedule modify and preschedule
functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/hw/ssow.h | 1 +
drivers/event/cnxk/cn20k_eventdev.c | 2 ++
drivers/event/cnxk/cn20k_worker.c | 30 +
drivers/e
Acked-by: Viacheslav Ovsiienko
> -Original Message-
> From: Alexander Kozyrev
> Sent: Thursday, October 3, 2024 11:26 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Raslan Darawsheh ; Slava
> Ovsiienko ; Matan Azrad ;
> step...@networkplumber.org
> Subject: [PATCH v2] drivers: fix error CQ
From: Pavan Nikhilesh
Add CN20K event device configuration that attaches the requested
number of SSO HWS(event ports) and HWGRP(event queues) LFs to
the RVU PF/VF.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 36 +
1 file changed, 36 inse
From: Pavan Nikhilesh
Add configuration APIs for CN20K SSO event
aggregator which allows SSO to generate event
vectors.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/hw/sso.h| 33
drivers/common/cnxk/roc_mbox.h | 21 +++
drivers/common/cnxk/roc_model.h | 13 ++
From: Pavan Nikhilesh
Replace gcc inbuilt __atomic_xxx intrinsics with rte_atomic_xxx API.
Signed-off-by: Pavan Nikhilesh
---
v2 Changes:
- Rebase and remove single dequeue and enqueue functions.
v3 Changes:
- Remove __atomic builtins.
v4 Changes:
- Rebase onto next-event tree.
v5 Changes:
- Re
From: Pavan Nikhilesh
Add CN20K SSO GWS fastpath event device enqueue functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 20 +-
drivers/event/cnxk/cn20k_worker.c | 384
drivers/event/cnxk/cn20k_worker.h | 21 ++
drivers/event/
From: Pavan Nikhilesh
Add SSO HWS a.k.a event port setup, release, link, unlink
functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn10k_eventdev.c | 63 ++-
drivers/event/cnxk/cn20k_eventdev.c | 174
drivers/event/cnxk/cn20k_eventdev.h
From: Pavan Nikhilesh
Add platform specific event device probe and remove, also add
event device info get function.
Signed-off-by: Pavan Nikhilesh
---
doc/guides/eventdevs/cnxk.rst | 23 ---
doc/guides/rel_notes/release_24_11.rst | 4 ++
drivers/common/cnxk/roc_sso.c | 1
From: Pavan Nikhilesh
Add CN20K SSO GWS event dequeue fastpath functions.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 5 +
drivers/event/cnxk/cn20k_worker.c | 54 +++
drivers/event/cnxk/cn20k_worker.h | 137 +++-
3 files chang
From: Pavan Nikhilesh
Add event timer adapter support for CN20K platform.
Implement new HWWQE insertion feature supported by CN20K platform.
Signed-off-by: Pavan Nikhilesh
---
drivers/common/cnxk/roc_tim.c| 6 +-
drivers/event/cnxk/cn20k_eventdev.c | 16 ++-
drivers/event/cnxk/cn20
From: Pavan Nikhilesh
Add Event vector support for CN20K Rx/Tx adapter.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 185 ++-
drivers/event/cnxk/cn20k_tx_worker.h | 84 ++
drivers/event/cnxk/cn20k_worker.h| 63
From: Pavan Nikhilesh
Add support for event eth Tx adapter.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 126 +++
drivers/event/cnxk/cn20k_eventdev.h | 4 +
drivers/event/cnxk/cn20k_tx_worker.h | 16
3 files changed, 146 insertions(
From: Pavan Nikhilesh
Add support for event eth Tx adapter fastpath operations.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 29 +++
drivers/event/cnxk/cn20k_tx_worker.h | 176 +++
drivers/event/cnxk/meson.build | 20
From: Pavan Nikhilesh
Add support for event eth Rx adapter fastpath operations.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 122 -
drivers/event/cnxk/cn20k_worker.c | 54 --
drivers/event/cnxk/cn20k_worker.h | 165
From: Pavan Nikhilesh
Add support for CN20K event eth Rx adapter.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/cnxk/cn20k_eventdev.c | 121 +++-
drivers/event/cnxk/cn20k_eventdev.h | 4 +
2 files changed, 124 insertions(+), 1 deletion(-)
diff --git a/drivers/even
From: Pavan Nikhilesh
Add selftest to verify SSO, xstats to get queue specific
stats and add function to dump internal state of SSO.
Signed-off-by: Pavan Nikhilesh
---
app/test/test_eventdev.c| 7 +++
drivers/event/cnxk/cn20k_eventdev.c | 12
drive
On Mon, 28 Oct 2024 16:27:41 +0200
Viacheslav Ovsiienko wrote:
> The mlx5 provides the scheduling send on time capability.
> The check the operating status of this feature the xstats
> counters are provided. This patch adds the counter descriptions
> and provides some meaningful information how t
From: Gavin Li
Print CQE error syndrome and more information in case of queue error.
Signed-off-by: Gavin Li
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/n
On Mon, 28 Oct 2024 11:18:22 +0200
"Minggang Li(Gavin)" wrote:
> +* **Updated NVIDIA mlx5 driver.**
> +
> + Optimized port probe in large scale.
> + In previous release, it would take long time to probe one VF/SF if
> + hundreds of VF/SF were created in the system. With this newly introduced
>
Import VDUSE headers to be included in the Vhost library.
Set header version to v6.10 Linux Kernel release.
Signed-off-by: Maxime Coquelin
Acked-by: Morten Brørup
---
kernel/linux/uapi/linux/vduse.h | 353
kernel/linux/uapi/version | 2 +-
2 files change
Thank you for the patch.
Acked-by: Viacheslav Ovsiienko
> -Original Message-
> From: Igor Gutorov
> Sent: Wednesday, August 7, 2024 11:44 PM
> To: dev@dpdk.org
> Cc: Igor Gutorov
> Subject: [PATCH v3 2/2] common/mlx5: reduce HCA attribute type sizes
>
> A number of `log_max_*` fields'
This patch introduces uAPI headers import into the DPDK
repository. This import is possible thanks to Linux Kernel
licence exception for syscalls:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/LICENSES/exceptions/Linux-syscall-note
Header files have to be explicitly impo
This series enables importing Linux Kernel uAPI headers
into the DPDK repository. It aims at solving alignment
issues between the build system and the system applications
linked ot DPDK libraries are run on.
It can also help simplify spaghetti code done to support
different versions of the Linux K
Hi,
I have some notes regarding the patch:
- patch does not fix the root cause of SQ state setting FW call failure
(https://bugs.dpdk.org/show_bug.cgi?id=1449 )
- MLX5_ASSERT is engaged only in debug mode and should not be used in release
- removing "return " prevents the tx_burst from retrying o
Thank you for the patch.
Acked-by: Viacheslav Ovsiienko
> -Original Message-
> From: Igor Gutorov
> Sent: Wednesday, August 7, 2024 11:44 PM
> To: dev@dpdk.org
> Cc: Igor Gutorov ; sta...@dpdk.org
> Subject: [PATCH v3 1/2] net/mlx5: fix reported Rx/Tx desc limits
>
> Currently, `rte_eth
We will send the fix once we sync internally hopefully tomorrow.
Sorry for the delay.
Jaroslaw Ilgiewicz
From: David Marchand
Sent: Friday, October 25, 2024 1:45 PM
To: Burakov, Anatoly ; Richardson, Bruce
; Ilgiewicz, Jaroslaw
Cc: dev@dpdk.org ; Thomas Monjalon
On Fri, Oct 25, 2024 at 4:28 PM Hanumanth Pothula wrote:
>
> The last argument passed to ssovf_parsekv() is an
> unsigned char*, but it is accessed as an integer.
> This can lead to an integer overflow.
>
> Hence, make ensure the argument is accessed as a char
> and for better error handling use s
https://bugs.dpdk.org/show_bug.cgi?id=1571
Bug ID: 1571
Summary: Documentation of non root table
Product: DPDK
Version: 24.07
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
Priority:
There is the mlx5 NIC limitations - configuring MTU
for PCI Virtual Function has no meaning. The actual maximal
packet size in VF's receiving is limited by MTU configured
on the related PCI Physical Function, the DPDK datapath
running over VF should be prepared to handle packets
of this maximal siz
The mlx5 provides the scheduling send on time capability.
The check the operating status of this feature the xstats
counters are provided. This patch adds the counter descriptions
and provides some meaningful information how to interpret
the counter values in runtime.
Signed-off-by: Viacheslav Ovs
Hi,
From: Itamar Gozlan
Sent: Sunday, October 27, 2024 6:14 PM
To: Itamar Gozlan; Hamdan Agbariya; NBU-Contact-Thomas Monjalon (EXTERNAL);
Suanming Mou; Dariusz Sosnowski; Slava Ovsiienko; Bing Zhao; Ori Kam; Matan
Azrad; Erez Shitrit
Cc: dev@dpdk.org; sta...@dpdk.org
Subject: [v2] net/mlx5/hws
The issue only appeared with hardware-timestamping enabled
(RTE_ETH_RX_OFFLOAD_TIMESTAMP).
The length of the prepended hardware timestamp was not subtracted from
the data length so that received packets were 16 bytes longer than
expected.
In scatter-gather mode only the first mbuf has a timesta
It seems unlikely that there are other users of contigmem on FreeBSD,
especially since it's not necessary for other applications.
On FreeBSD, use of large and huge pages is automatic; you just call mmap() with
a large size and it automatically tries to use the largest physical pages
possible.
1 - 100 of 182 matches
Mail list logo
