From: Sunil Kumar Kori
When user requests to configure a device which is already in
configured state then first device gets resets to default and
then reconfigured with latest parameters.
While resetting the device, MAC address table is left stale which
causes entry update in later state.
Hence
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Tuesday, 8 April 2025 19.19
>
> Latest DPDK releases no longer officially support RHEL 7 as a target
> platform, so remove the note about pkg-config not working on those
> systems.
>
> Signed-off-by: Bruce Richardson
> ---
> d
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Tuesday, 8 April 2025 19.19
>
> Rather than specifying a minimum compiler version, update the docs to
> recommend a particular compiler version. In other words, earlier
> versions may work, but are not officially supported, or ma
On Wed, 9 Apr 2025 09:07:11 +0800
Kaiwen Deng wrote:
> The xstats name needs to be changed correctly
> tx_size_1023_to_max_packets should be tx_size_1024_to_max_packets.
>
> Fixes: e6defdfddc3b ("net/igc: enable statistics")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Kaiwen Deng
> ---
> driver
The 4.19 kernel version was end of life in December 2024.
Update to 5.4 and also add better description of what "not supported"
means in this case.
Signed-off-by: Stephen Hemminger
Acked-by: Morten Brørup
---
doc/guides/linux_gsg/sys_reqs.rst | 17 -
1 file changed, 12 insertion
Use the base code build handling logic in the drivers/meson.build file,
rather than re-implementing it in the driver itself.
Signed-off-by: Bruce Richardson
---
drivers/net/thunderx/base/meson.build | 10 +-
drivers/net/thunderx/meson.build | 3 ---
2 files changed, 1 insertion(+),
On Mon, Apr 07, 2025 at 04:28:57PM +0100, Bruce Richardson wrote:
> For SVE, as for AVX2 and AVX-512, support building files for these ISAs
> from the top-level drivers/meson.build file, rather than having each
> driver re-implement it.
>
> This removes the remaining build task for drivers in DPDK
using static_assert(RTE_BUILD_BUG_ON) replaced BUILD_BUG_ON.
Fixes: 9c3a123f9e08 ("net/zxdh: optimize msg processing ops and modify some
issues")
Signed-off-by: Junlong Wang
---
drivers/net/zxdh/zxdh_common.h | 9 ++---
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/drivers/
The xstats name needs to be changed correctly.
tx_size_1023_to_max_packets should be tx_size_1024_to_max_packets.
Fixes: 38552317dcef ("igb: add extended stats")
Cc: sta...@dpdk.org
Signed-off-by: Kaiwen Deng
---
drivers/net/intel/e1000/igb_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 del
Now that the issues preventing this lib from being compiled with MSVC
are fixed it can be included in the compilation.
The "net" library will automatically get compiled as well as it
has a dependency on "mbuf" which will now get fulfilled.
Signed-off-by: Andre Muezerie
Acked-by: Bruce Richardson
Compiling with MSVC logs the warnings below, which result in
build error:
../app/test/test_hash_readwrite.c(73): warning C4476: 'printf' :
unknown type field character ''' in format specifier
../app/test/test_hash_readwrite.c(75): warning C4474: 'printf' :
too many arguments passed for for
It's common to use %' in the printf format specifier to make large numbers
more easily readable by having the thousands grouped. However, this
grouping does not work on Windows. Therefore, a function is needed to make
uint64_t numbers more easily readable. There are at least two tests that
can bene
Compiling with MSVC logs the warnings below, which result in
build error:
../app/test/test_hash_multiwriter.c(71): warning C4476: 'printf' :
unknown type field character ''' in format specifier
../app/test/test_hash_multiwriter.c(73): warning C4474: 'printf' :
too many arguments passed for
v5:
- Rebased and updated to conform to new function versioning mechanism.
v4:
- Added parameter "unit", which allows rte_size_to_str() to be smarter
about the need to append a space after the number. This keeps the
function easy to use and avoids the need for complex explanations
about
On Wed, Mar 12, 2025 at 09:22:42PM +0530, Shaiq Wani wrote:
> In case some CPUs don't support AVX512. Enable AVX2 for them to
> get better per-core performance.
>
> The single queue model processes all packets in order while
> the split queue model separates packet data and metadata into
> differe
From: Long Li
To prepare for supporting to set hyperv event from secondary process
when the channel has monitoring disable, introduce a private data
region for storing the vmbus device. The secondary process will get
access to its vmbus device in case it needs to signal the host.
Signed-off-by:
From: Long Li
For vmbus channels without monitoring support, use kernel UIO interface
to indicate packet through interrupt page and UIO file handle.
Signed-off-by: Long Li
---
Changes:
v4: replace RTE_ATOMIC(uint32_t) * with RTE_ATOMIC(uint32_t *) for declaring
variable
drivers/bus/vmbus/vmb
> Subject: [EXTERNAL] Re: [patch v4 6/6] bus/vmbus: set event for channel
> without
> monitoring support
>
> On Mon, 7 Apr 2025 15:45:04 -0700
> lon...@linuxonhyperv.com wrote:
>
> > From: Long Li
> >
> > For vmbus channels without monitoring support, use kernel UIO
> > interface to indicate p
From: Long Li
Hyperv host may offer channels without monitor enabled. The max monitor
ID it supports is 128. Over those channels without monitor enabled,
Hyperv does not send or receive large amount of data traffic and almost all
the data traffic is going over the VF.
Change the code to not fail
From: Long Li
Introduce a function get the vmbus device from hn_data. For secondary
process, the vmbus device is in eth_dev's private region.
Signed-off-by: Long Li
---
drivers/net/netvsc/hn_nvs.c | 15 +++
drivers/net/netvsc/hn_nvs.h | 2 ++
2 files changed, 17 insertions(+)
dif
From: Long Li
The secondary process can access its vmbus device through device private
region. Add and pass it on all call chains leading to vmbus code.
Signed-off-by: Long Li
---
Change log.
v5: rebased to the latest net-next
drivers/bus/vmbus/linux/vmbus_uio.c | 2 +-
drivers/bus/vmbus/pri
From: Long Li
Secondary process will get access to vmbus device and this UIO fd for
signaling hyperv host on channels without monitoring support.
Signed-off-by: Long Li
---
drivers/bus/vmbus/vmbus_common_uio.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers
From: Long Li
Hyperv may expose VMBUS channels without monitoring enabled. In this case,
it programs almost all the data traffic to VF.
This patchset enabled vmbus/netvsc to use channels without monitoring
enabled.
This needs to change the APIs exposed by drivers/bus/vmbus. Because those
APIs a
Now that the issues preventing this lib from being compiled with MSVC
are fixed it can be included in the compilation.
v4:
- removed all symbol-related changes, leaving only the change to enable
MSVC compilation.
- only adding cflags '-mpclmul' and '-maes' when the compiler used
supports them.
On Mon, Apr 07, 2025 at 03:53:06PM +0200, David Marchand wrote:
> Hello Andre,
>
> On Thu, Feb 20, 2025 at 9:32 PM Andre Muezerie
> wrote:
> >
> > Now that the issues preventing this lib from be compiled with MSVC
> > are fixed it can be included in the compilation.
> >
> > The "net" library will
Revert "app/testpmd: add port attach/detach for multiple process"
This reverts commit 994635edb2c038e64617bcf2790a8cd326c3e8e0.
This commit breaks using pdump and other secondary processes that
create there own devices. The patch makes testpmd grab any new
hotplug device and configure it.
It may
https://bugs.dpdk.org/show_bug.cgi?id=1695
Bug ID: 1695
Summary: pdump regression caused by test-pmd change
Product: DPDK
Version: 25.03
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: major
On Tue, 8 Apr 2025 11:30:32 -0700
Stephen Hemminger wrote:
> In early days of virtio, the x86 I/O instructions were used
> to ring the doorbell for host access. To support these DPDK
> needed to enable IO access. This is no longer necessary and
> the vestiage can be marked deprecated.
>
> Signe
On Tue, 8 Apr 2025 at 19:31, Ali Alnubani wrote:
>
> > -Original Message-
> > From: luca.bocca...@gmail.com
> > Sent: Wednesday, March 26, 2025 1:51 PM
> > To: sta...@dpdk.org
> > Cc: dev@dpdk.org; Ali Alnubani ; David Christensen
> > ; Hemant Agrawal ; Ian
> > Stokes ; Jerin Jacob ; John
Release status meeting minutes 2025-04-08
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* ARM
* Debian
* Intel
* Marvell
* Red Hat
Release Dates
-
The following are the proposed working dates for 27.03:
> -Original Message-
> From: luca.bocca...@gmail.com
> Sent: Wednesday, March 26, 2025 1:51 PM
> To: sta...@dpdk.org
> Cc: dev@dpdk.org; Ali Alnubani ; David Christensen
> ; Hemant Agrawal ; Ian
> Stokes ; Jerin Jacob ; John
> McNamara ; Kevin Traynor
> ; Luca Boccassi ; Pei Zhang
> ; Rasl
In early days of virtio, the x86 I/O instructions were used
to ring the doorbell for host access. To support these DPDK
needed to enable IO access. This is no longer necessary and
the vestiage can be marked deprecated.
Signed-off-by: Stephen Hemminger
---
doc/guides/rel_notes/deprecation.rst |
Latest DPDK releases no longer officially support RHEL 7 as a target
platform, so remove the note about pkg-config not working on those
systems.
Signed-off-by: Bruce Richardson
---
doc/guides/linux_gsg/sys_reqs.rst | 7 ---
1 file changed, 7 deletions(-)
diff --git a/doc/guides/linux_gsg/sy
> -Original Message-
> From: David Marchand
> Sent: Monday, April 7, 2025 10:50 AM
> To: Bruce Richardson
> Cc: dev@dpdk.org; Jie Hai
> Subject: Re: [RFC PATCH] drivers: add generic build of SVE files
>
> On Mon, Apr 7, 2025 at 5:29 PM Bruce Richardson
> wrote:
> >
> > For SVE, as fo
On Thu, Apr 03, 2025 at 04:54:25PM +0530, Shaiq Wani wrote:
> reworked the drivers to use the common functions
> and structures from drivers/net/intel/common.
>
> Shaiq Wani (4):
> net/intel: align Tx queue struct field names
> net/intel: use common Tx entry structure
> net/intel: use common
On Thu, Apr 03, 2025 at 04:54:29PM +0530, Shaiq Wani wrote:
> Switch the idpf driver to use the common Tx free function for
> AVX2 and AVX512
>
> Signed-off-by: Shaiq Wani
> ---
> .../net/intel/idpf/idpf_common_rxtx_avx2.c| 61 +
> .../net/intel/idpf/idpf_common_rxtx_avx512.c | 230 +--
On Thu, Apr 03, 2025 at 04:54:28PM +0530, Shaiq Wani wrote:
> Merge in additional fields used by the idpf driver and then convert it
> over to using the common Tx queue structure
>
> Signed-off-by: Shaiq Wani
> ---
> drivers/net/intel/common/tx.h | 18 ++
> drivers/net/intel/
On Thu, Apr 03, 2025 at 04:54:26PM +0530, Shaiq Wani wrote:
> Align the Tx queue struct field names in idpf and
> cpfl driver with the common Tx queue struct.
>
> Signed-off-by: Shaiq Wani
> ---
Acked-by: Bruce Richardson
On Tue, Apr 08, 2025 at 05:16:11PM +0100, Bruce Richardson wrote:
> On Wed, Mar 12, 2025 at 09:22:42PM +0530, Shaiq Wani wrote:
> > In case some CPUs don't support AVX512. Enable AVX2 for them to
> > get better per-core performance.
> >
> > The single queue model processes all packets in order whi
The determination of whether or not the buffer size is less than the
packet MTU is made in the Rx queue start function in the ice driver.
This means that we cannot choose the Rx function before we have started
the queues. Move the Rx function section down in the code to fix this.
Fixes: 3c79a3d91d
On Tue, Apr 08, 2025 at 03:58:38PM +0100, Bruce Richardson wrote:
> Add a Cargo.toml file in the root folder and a number of other scripts
> and rust-related files into buildtools/rust, which then enables DPDK to
> be cloned and built as a rust crate - all-be-it one with only two
> functions: rte_e
On Wed, 26 Mar 2025 at 11:51, wrote:
>
> Hi all,
>
> Here is a list of patches targeted for stable release 22.11.8.
>
> The planned date for the final release is 2025/04/10.
>
> Please help with testing and validation of your use cases and report
> any issues/results with reply-all to this mail. F
Add a Cargo.toml file in the root folder and a number of other scripts
and rust-related files into buildtools/rust, which then enables DPDK to
be cloned and built as a rust crate - all-be-it one with only two
functions: rte_eal_init and rte_eal_cleanup.
Signed-off-by: Bruce Richardson
---
This R
https://bugs.dpdk.org/show_bug.cgi?id=1673
Stephen Hemminger (step...@networkplumber.org) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
On Fri, 28 Mar 2025 12:37:31 +0900
郭春霞 wrote:
> Dear DPDK Team,
>
> I hope this email finds you well.
>
> I'm writing to inquire about DPDK's future plans regarding dynamic CPU
> initialization support, particularly in Kubernetes environments where
> exclusive CPU Pod scaling is being enhanc
When the hns3_dfx_reg_fetch_data() function returns from processing
failure, cmd_descs is not freed, which leads to leakage.
This patch fit it.
By the way, this patch uses calloc to apply for heap memory instead
of hugepage memory.
Fixes: ef1fbd355451 ("net/hns3: add more registers to dump")
Cc:
Streamline code for AVX512 and SSE by consolidating the common code and
adding runtime check for selecting appropriate path based on CPU
capability.
Signed-off-by: Tirthendu Sarkar
---
v4:
- Modify some AVX512 instructions to resolve compilation error on
older compliers
v3:
- Simplified cod
When I tried to enable TSan on DPDK using this patch, DPDK cannot be
initialized.
When the DPDK initializes the memseg, a large number of address space is
reserved.
However, The introduction of ThreadSanitizer brings a large amount of memory
overhead.
As a result, the address space is insuffici
ThreadSanitizer is a tool that detects data races. It consists of a
compiler instrumentation module and a run-time library.
Typical slowdown introduced by ThreadSanitizer is about 5x-15x.
Typical memory overhead introduced by ThreadSanitizer is about 5x-10x.
UBSan is integrated with gcc and clang
Implement the logic of get RSS algorithm capability, also add the logic
of print RSS algorithm capability.
Signed-off-by: Chaoyong He
---
drivers/common/nfp/nfp_common.h | 1 +
drivers/common/nfp/nfp_common_ctrl.h | 3 +++
.../net/nfp/flower/nfp_flower_representor.c |
Dear DPDK Team,
I hope this email finds you well.
I'm writing to inquire about DPDK's future plans regarding dynamic CPU initialization support, particularly in Kubernetes environments where exclusive CPU Pod scaling is being enhanced.
Backgroud:With Kubernetes upcoming improvements to In
Update eventdev app to handle SNOW-V correctly.
Signed-off-by: Radu Nicolau
---
app/test-eventdev/evt_options.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/app/test-eventdev/evt_options.c b/app/test-eventdev/evt_options.c
index 323d1e724d..7903b5959c 100644
--- a/app/te
Update l2fwd-crypto to be able to correctly handle SNOW-V
Signed-off-by: Radu Nicolau
---
examples/l2fwd-crypto/main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index a441312f55..f1a6ca8899 100644
--- a/examp
Add suport for SNOW-V and SNOW-V AEAD algorithms.
Signed-off-by: Radu Nicolau
---
app/test-crypto-perf/cperf_ops.c | 6 --
doc/guides/tools/cryptoperf.rst | 2 ++
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/app/test-crypto-perf/cperf_ops.c b/app/test-crypto-perf/cperf_ops
Add SNOW-V and SNOW-V AEAD mode algorithms to cryptodev API.
Signed-off-by: Radu Nicolau
---
lib/cryptodev/rte_crypto_sym.h | 17 -
lib/cryptodev/rte_cryptodev.c | 4 +++-
2 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/lib/cryptodev/rte_crypto_sym.h b/lib/crypt
Acked-by: Chengwen Feng
On 2025/4/3 18:01, lihuisong (C) wrote:
>
> 在 2025/4/2 16:47, Jie Hai 写道:
>> I am moving on to other things and dengdui is going to
>> take over the role of hns3 maintainer. Update the
>> MAINTAINERS accordingly.
>>
>> Signed-off-by: Jie Hai
> Thanks for your contributio
Hello Nitin,
On Mon, Apr 7, 2025 at 9:48 AM Nitin Saxena wrote:
>
> Currently each rte_node registers separate mbuf dynamic fields for their
> own purpose. This leads to wastage of mbuf space as once mbuf get passed
> a particular node, the registered dynamic field(by that node) is no
> longer us
Hi David,
Thanks,
Nitin
On Tue, Apr 8, 2025 at 1:05 PM David Marchand wrote:
>
> Hello Nitin,
>
> On Mon, Apr 7, 2025 at 9:48 AM Nitin Saxena wrote:
> >
> > Currently each rte_node registers separate mbuf dynamic fields for their
> > own purpose. This leads to wastage of mbuf space as once mbuf
When the port is started, if the Tx queue fails to be started,
the map interrupt should be rolled back.
Fixes: fdfde7a4a0f8 ("net/hns3: fix mbuf leakage")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_ethdev.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
The number of interrupt vector may exceed the range of uint8_t.
So hns3_unmap_rx_interrupt() should use uint16_t for 'vec' variable.
Fixes: 02a7b55657b2 ("net/hns3: support Rx interrupt")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_common.c | 4 ++--
1 file change
Some resources, like, unmapping Rx interrupt, doesn't perform
when execute dev_stop on reset. This will lead to other issues.
Fixes: 2790c6464725 ("net/hns3: support device reset")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_ethdev.c | 19 +++
1 fi
Currently, when the application creates an indirect action,
the hns3 driver allocates a memory for the structure
rte_flow_action_handle and returns this structure pointer to
application. When the application invokes the destroy function
to destroy the indirect action, the driver releases the memory
Driver may encounter divide-by-zero if the total_tqps_num
and rss_size_max in hw structure from firmware are zero.
So add some verification to them.
Fixes: d51867db65c1 ("net/hns3: add initialization")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
---
drivers/net/hns3/hns3_ethdev.c | 10
If the link auto-negotiation of the NIC is disabled,
or the flow control auto-negotiation is not supported,
it's unnecessary to wait for link establishment.
Fixes: 1f411e31a826 ("net/hns3: support flow control autoneg for copper port")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
---
driver
This patchset fixes some bugs.
Dengdui Huang (7):
net/hns3: fix variable overflow
net/hns3: fix memory leakage in failure path
net/hns3: fix extra waiting for link up
net/hns3: fix memory leakage for indirect action
net/hns3: fix incorrect failed rollback
net/hns3: fix divide by zero
When using the Meson option "ibverbs_link=dlopen",
a glue library is built.
After the recent change requiring eal_export.h,
the build stopped because the internal EAL include directory
was unknown to this glue library:
drivers/common/mlx5/linux/mlx5_glue.c:19:10: fatal error:
eal_
08/04/2025 09:57, David Marchand:
> On Tue, Apr 8, 2025 at 9:47 AM Thomas Monjalon wrote:
> >
> > When using the Meson option "ibverbs_link=dlopen",
> > a glue library is built.
> >
> > After the recent change requiring eal_export.h,
> > the build stopped because the internal EAL include directory
On Tue, Apr 8, 2025 at 9:47 AM Thomas Monjalon wrote:
>
> When using the Meson option "ibverbs_link=dlopen",
> a glue library is built.
>
> After the recent change requiring eal_export.h,
> the build stopped because the internal EAL include directory
> was unknown to this glue library:
>
>
Implement the logic of get RSS algorithm configuration, also do little
refactor to related functions.
Signed-off-by: Chaoyong He
---
drivers/net/nfp/nfp_net_common.c | 84 ++--
1 file changed, 57 insertions(+), 27 deletions(-)
diff --git a/drivers/net/nfp/nfp_net_com
On 4/2/25 8:53 AM, David Marchand wrote:
The idx field of a virtqueue available ring is increased by the driver
regardless of the ring size. It is for the device to mask this index
modulo the ring size (2.7.6 of the virtio 1.3 specification).
The same applies to the used ring.
Failing to mask
70 matches
Mail list logo
