On Fri, Apr 12, 2024 at 02:00:43PM +0100, Bruce Richardson wrote:
> On Thu, Mar 14, 2024 at 11:30:21PM -0700, Tyler Retzlaff wrote:
> > MSVC is the only compiler that can produce usable shared libraries for
> > DPDK on Windows because of the use of exported TLS variables.
> >
> > Disable building
On Fri, Apr 12, 2024 at 03:09:12PM +0100, Bruce Richardson wrote:
> On Thu, Mar 14, 2024 at 11:30:22PM -0700, Tyler Retzlaff wrote:
> > Use static deps when default_library=static and use shared deps when
> > using default_library=shared.
> >
> > Signed-off-by: Tyler Retzlaff
> > ---
> > buildto
Use the value of default_library and use it when building deps instead
of always using static deps so we get the correct static or shared deps.
Signed-off-by: Tyler Retzlaff
---
buildtools/chkincs/meson.build | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/buildtools/chk
MSVC is the only compiler that can produce usable shared libraries for
DPDK on Windows because of the use of exported TLS variables.
Disable building of shared libraries with LLVM and MinGW so that
remaining __declspec macros needed for the functional libraries built by
MSVC can be used without tr
MSVC is the only compiler that can produce usable shared libraries for
DPDK on Windows because of the use of exported TLS variables.
Disable building of shared libraries with LLVM and MinGW so that
remaining __declspec macros needed for the functional libraries built by
MSVC can be used without tr
On 3/28/2022 4:16 PM, Gaoxiang Liu wrote:
> The following log message may appear after a slave is idle(or nearly
> idle)
> for a few minutes:"PMD: Failed to allocate LACP packet from pool".
> And bond mode 4 negotiation may fail.
>
> Problem:When bond mode 4 has been chosed and delicated queue has
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, April 12, 2024 6:23 AM
> To: Alan Elder ; Long Li ;
> Andrew Rybchenko
> Cc: dev@dpdk.org; stephen
> Subject: Re: [EXTERNAL] Re: [PATCH v3] net/netvsc: fix number Tx queues > Rx
> queues
>
> On 4/11/2024 9:45 PM, Alan Elder wr
> > > > > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx
> checksum
> > > offload
> > > > > > > > > examples.
> > > > > > > >
> > > > > > > > I strongly disagree with this change!
> > > > > > > >
> > > > > > > > It will cause a huge performance degradation for shaping
> > > applications
>
> > > > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum
> > offload
> > > > > > > > examples.
> > > > > > >
> > > > > > > I strongly disagree with this change!
> > > > > > >
> > > > > > > It will cause a huge performance degradation for shaping
> > applications:
> > > > >
> > > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum
> offload
> > > > > > > examples.
> > > > > >
> > > > > > I strongly disagree with this change!
> > > > > >
> > > > > > It will cause a huge performance degradation for shaping
> applications:
> > > > > >
> > > > > > A packet
On Thu, Mar 14, 2024 at 11:30:22PM -0700, Tyler Retzlaff wrote:
> Use static deps when default_library=static and use shared deps when
> using default_library=shared.
>
> Signed-off-by: Tyler Retzlaff
> ---
> buildtools/chkincs/meson.build | 18 +-
> 1 file changed, 13 insertions
On Thu, Mar 14, 2024 at 11:30:21PM -0700, Tyler Retzlaff wrote:
> MSVC is the only compiler that can produce usable shared libraries for
> DPDK on Windows because of the use of exported TLS variables.
>
> Disable building of shared libraries with LLVM and MinGW so that
> remaining __declspec macro
The conversion to bool is not needed, remove it.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_phy_impl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/axgbe/axgbe_phy_impl.c
b/drivers/net/axgbe/axgbe_phy_impl.c
index 5acccf9146..b5796b13e0 10064
The existing implementation for non-Autonegotiation 10G speed modes does
not enable RX adaptation in the Driver and FW. The RX Equalization
settings (AFE settings alone) are manually configured and the existing
link-up sequence in the driver does not perform rx adaptation process as
mentioned in th
MAC version 21H supports the 10Mbps speed. So, extend support to
platforms that support it.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_phy_impl.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/axgbe/axgbe_phy_impl.c
b/drivers/net/axgbe/axgb
In the event of a change in AXGBE mode, the current auto-negotiation
needs to be reset and the AN cycle needs to be re-triggerred. However,
the current code ignores the return value of axgbe_set_mode(), leading to
false information as the link is declared without checking the status
register.
Fix
Add support to the driver to fully recognize and enable 2.5GbE speed in
10GBaseT mode.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_phy_impl.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/axgbe/axgbe_phy_impl.c
b/drivers/net/axgbe/axgbe
The xgbe MDIO bus driver can perform both C22 and C45 transfers, when
using its MDIO bus hardware. The SFP I2C mdio bus driver only supports
C22. Create separate functions for each and register the C45 versions
using the new API calls where appropriate.
Signed-off-by: Venkat Kumar Ande
---
drive
The axgbe driver reuses MII_ADDR_C45 for its own purpose. The values
derived with it are never passed to phylib or a linux MDIO bus driver.
In order that MII_ADDR_C45 can be removed, add an AXGBE specific
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_common.h | 4 ++--
drivers/net
AN restart triggered during KR training not only aborts the KR training
process but also move the HW to unstable state. Driver has to wait up to
500ms or until the KR training is completed before restarting AN cycle.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_ethdev.h | 2 ++
There is difference in the TX Flow Control registers (TFCR) between the
revisions of the hardware. The older revisions of hardware used to have
single register per queue. Whereas, the newer revision of hardware (from
ver 30H onwards) have one register per priority.
Signed-off-by: Venkat Kumar Ande
There are cables that exist that can support speeds in excess of 10GbE.
The driver, however, restricts the EEPROM advertised nominal bitrate to
a specific range, which can prevent usage of cables that can support,
for example, up to 25GbE.
Rather than checking that an active or passive cable suppo
Add the necessary changes to support 10 Mbps speed for BaseT and SFP
port modes. This is supported in MAC ver >= 30H.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_dev.c | 3 ++
drivers/net/axgbe/axgbe_ethdev.h | 2 +
drivers/net/axgbe/axgbe_mdio.c | 22 +
dri
SFP+ active and passive cables are copper cables with fixed SFP+ end
connectors. Due to a misinterpretation of this, SFP+ active cables could
end up not being recognized, causing the driver to fail to establish a
connection.
Introduce a new enum in SFP+ cable types, XGBE_SFP_CABLE_FIBER, that is
t
The current AXGBE code assumes that offset 6 of EEPROM SFP DAC (passive)
cables is NULL. However, some cables (the 5 meter and 7 meter Molex
passive cables) have non-zero data at offset 6. Fix the logic by moving
the passive cable check above the active checks, so as not to be
improperly identified
PLL control setting(RRC) is needed only in fixed PHY configuration to
fix the peer-peer issues. Without the PLL control setting, the link up
takes longer time in a fixed phy configuration.
Driver implements SW RRC for Autoneg On configuration, hence PLL control
setting (RRC) is not needed for AN O
Instead of using hardcoded values, use enumerations for mailbox command
and sub commands.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_ethdev.h | 25 +
drivers/net/axgbe/axgbe_phy_impl.c | 25 +++--
2 files changed, 40 insertions(+),
Based on the IOMMU configuration, the current cache control settings can
result in possible coherency issues. The hardware team has recommended
new settings for the PCI device path to eliminate the issue.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_dev.c | 20 +--
Link stability issues are noticed on Yellow carp platforms when Receiver
Reset Cycle is issued. Since the CDR workaround is disabled on these
platforms, the Receiver Reset Cycle is not needed.
So, avoid issuing rrc on Yellow carp platforms.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe
Hardware interrupts are enabled during the init, however,
they are not disabled during close.
Disable all hardware interrupts during close operation to avoid any
issues.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_ethdev.c | 5 +
1 file changed, 5 insertions(+)
diff --git
Frequent link up/down events can happen when a Bel Fuse SFP part is
connected to the amd-xgbe device. Try to avoid the frequent link
issues by resetting the PHY as documented in Bel Fuse SFP datasheets.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_phy_impl.c | 3 +++
1 file chang
Normally, auto negotiation and reconnect should be automatically done by
the hardware. But there seems to be an issue where auto negotiation has
to be restarted manually. This happens because of link training and so
even though still connected to the partner the link never "comes back".
This needs
The XGBE hardware has support for performing MDIO operations using an
MDIO command request. The driver mistakenly uses the mdio port address
as the MDIO command request device address instead of the MDIO command
request port address. Additionally, the driver does not properly check
for and create a
After changing speed to 100Mbps as a result of auto-negotiation (AN),
some 10/100/1000Mbps SFPs indicate a successful link (no faults or loss
of signal), but cannot successfully transmit or receive data. These
SFPs required an extra auto-negotiation (AN) after the speed change in
order to operate
The driver currently sets an indication of whether the SFP supports, and
that the driver can obtain, diagnostics data. This isn't currently used
by the driver. Remove this field and the logic supporting it.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/axgbe_phy_impl.c | 10 --
The comm_owned field can hide logic where double locking is attempted
and prevent multiple threads for the same device from accessing the
mutex properly. Remove the comm_owned field and use the mutex API
exclusively for gaining ownership.
Signed-off-by: Venkat Kumar Ande
---
drivers/net/axgbe/a
> > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum offload
> > > > > > examples.
> > > > >
> > > > > I strongly disagree with this change!
> > > > >
> > > > > It will cause a huge performance degradation for shaping applications:
> > > > >
> > > > > A packet will be processe
On Wed, Apr 10, 2024 at 4:07 PM Luca Vizzarro wrote:
>
> On 10/04/2024 14:35, Juraj Linkeš wrote:
> > We should also consider that this would introduce a difference in API
> > between the interactive and non-interactive sessions. Do we want to do
> > this there as well?
>
> Could definitely add it
This commit introduces documentation for the newly introduced link
bonding mode.
Signed-off-by: Kamil Vojanec
---
doc/guides/prog_guide/img/bond-mode-7.svg | 908 ++
.../link_bonding_poll_mode_drv_lib.rst| 13 +
2 files changed, 921 insertions(+)
create mode 100644
This patch adds a new bonding mode: 'direct'. The purpose of this mode
is to redirect output packets to the same member port on which they were
received. In case the requested port is not among the member ports,
round robin TX mode is used as a fallback.
Signed-off-by: Kamil Vojanec
---
app/test
Signed-off-by: Kamil Vojanec
---
doc/guides/prog_guide/img/bond-mode-7.svg | 908 ++
.../link_bonding_poll_mode_drv_lib.rst| 13 +
2 files changed, 921 insertions(+)
create mode 100644 doc/guides/prog_guide/img/bond-mode-7.svg
diff --git a/doc/guides/prog_guide/img/
This patch adds a new bonding mode: 'direct'. The purpose of this mode
is to redirect output packets to the same member port on which they were
received. In case the requested port is not among the member ports,
round robin TX mode is used as a fallback.
Signed-off-by: Kamil Vojanec
---
app/test
Added a test case to verify the new API
rte_cryptodev_qp_depth_used() to get the used
depth of a crypto device queue pair.
Signed-off-by: Akhil Goyal
---
app/test/test_cryptodev.c | 117 ++
1 file changed, 117 insertions(+)
diff --git a/app/test/test_cryptode
Added support to get the used queue pair depth
for a specific queue on cn10k platform.
Signed-off-by: Akhil Goyal
---
drivers/crypto/cnxk/cn10k_cryptodev.c| 1 +
drivers/crypto/cnxk/cn9k_cryptodev.c | 2 ++
drivers/crypto/cnxk/cnxk_cryptodev_ops.c | 16
drivers/crypto/
Added a new fast path API to get used queue pair
descriptors of a specific queue pair of a device.
Applications may monitor the depth used and enqueue
crypto ops accordingly.
Signed-off-by: Akhil Goyal
---
devtools/libabigail.abignore | 3 ++
lib/cryptodev/cryptodev_pmd.c |
Added a new fast path API to get the number of used crypto device
queue pair depth at any given point.
An implementation in cnxk crypto driver is also added along with
a test case in test app.
The addition of new API causes an ABI warning.
This is suppressed as the updated struct rte_crypto_fp_op
On 10/17/2023 5:43 PM, Stephen Hemminger wrote:
> On Tue, 15 Feb 2022 22:16:28 +0530
> Vipul Ashri wrote:
>
>> On 2/14/2022 10:24 PM, Stephen Hemminger wrote:
>>> On Mon, 14 Feb 2022 13:09:19 +
>>> Vipul Ashri wrote:
>>>
PORT 0 supports 16 rx queues and 16 tx queues (driver_name = net
From: Krishna Kanth Reddy
Modified the Configuration file to use the latest ARM Cross-Compiler.
Fixed the linker errors for the undefined references to the APIs
isal_deflate_init, isal_deflate, isal_inflate_init, isal_inflate,
isal_inflate_stateless, isal_deflate_stateless,
isal_deflate_set_huff
From: Alexey Marchuk
mlx5 common library checks if several symbols/definitions
are presented in system header files. If some are not
presented, they will be enabled by mlx5_glue library.
The problem appears with clang and '-Werror' - code
generated by meson is not compiled due to unused variable:
Add a new TestPmdPortStats data structure to represent the output
returned by `show port stats`, which is implemented as part of
TestPmdShell.
Bugzilla ID: 1407
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/remote_session/testpmd_shell.py | 58 +++
Add a new TestPmdPort data structure to represent the output
returned by `show port info`, which is implemented as part of
TestPmdShell.
The TestPmdPort data structure and its derived classes are modelled
based on the relevant testpmd source code.
This implementation makes extensive use of regula
Adds parsing text into a custom data structure. It provides a new
`TextParser` dataclass to be inherited. This implements the `parse`
method, which combined with the parser functions, it can automatically
parse the value for each field.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
-
The first line of the InteractiveShell send_command method is generally
the command input field. This sometimes is unwanted, therefore this
commit enables the possibility of omitting the first line from the
returned output.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
Reviewed-by: J
When sending a command using an instance of InteractiveShell the output
is meant to filter out the leading shell prompt. The filtering logic is
present but the line is appended anyways.
Bugzilla ID: 1411
Fixes: 88489c0501af ("dts: add smoke tests")
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul
Hello,
As previously mentioned in the DTS meeting, here is the patch series
relating show port info and stats. It also includes a proposal for a
generic parsing utility.
>From my existing testpmd params series I recognise there are quite a few
improvements that can also be added to this series. B
On 4/11/2024 9:45 PM, Alan Elder wrote:
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Thursday, April 11, 2024 7:38 AM
>> To: Alan Elder ; Long Li ;
>> Andrew Rybchenko
>> Cc: dev@dpdk.org; stephen
>> Subject: [EXTERNAL] Re: [PATCH v3] net/netvsc: fix number Tx queues > Rx
>> queu
The tool used to generate DTS API docs is Sphinx, which is already in
use in DPDK. The same configuration is used to preserve style with one
DTS-specific configuration (so that the DPDK docs are unchanged) that
modifies how the sidebar displays the content.
Sphinx generates the documentation from
These sources could be generated with the sphinx-apidoc utility, but
that doesn't give us enough flexibility, such as sorting the order of
modules or changing the headers of the modules.
The sources included in this patch were in fact generated by said
utility, but modified to improve the look of
Sphinx imports every Python module when generating documentation from
docstrings, meaning all DTS dependencies, including Python version,
must be satisfied.
By adding Sphinx to DTS dependencies we provide a convenient way to
generate the DTS API docs which satisfies all dependencies.
Signed-off-by
The generation is done with Sphinx, which DPDK already uses, with
slightly modified configuration of the sidebar present in an if block.
Dependencies are installed using Poetry from the dts directory:
poetry install --with docs
After installing, enter the Poetry shell:
poetry shell
And then ru
On 3/19/2024 7:07 AM, Chaoyong He wrote:
> From: Long Wu
>
> The PF representor port's queue is different from the VF/physical
> representor port. So the release process in close port should
> be different too.
>
> Fixes: a256a1227dbe ("net/nfp: fix resource leak for exit of flower firmware")
>
Fix a problem introduced by a compiler upgrade (from gcc10 to gcc12.3),
which will lead the hns3 NIC can't link up. The root cause is strict
aliasing violation in rte_eth_linkstatus_set() with hns3 driver, see
[1] for more details.
This commit use union to avoid such aliasing violation.
Note: DPD
Hi Stephen,
On 2024/4/11 23:05, Stephen Hemminger wrote:
> On Thu, 11 Apr 2024 03:07:49 +
> Chengwen Feng wrote:
>
>> Fix a problem introduced by a compiler upgrade (from gcc10 to gcc12.3),
>> which will lead the hns3 NIC can't link up. The root cause is strict
>> aliasing violation in rte_e
> 2024年3月21日 17:57，Haoqian He 写道：
>
> We should cleanup vq resubmit info when set_inflight_fd
> before set_vring_kick which will check if there is any
> inflight io waiting for resubmission.
>
> Otherwise, when the vm is rebooting immediately after
> reconnecting to the vhost target (inflight
Signed-off-by: Yutang Jiang
---
config/arm/arm64_ampereonex_linux_gcc | 16
config/arm/meson.build| 19 +++
2 files changed, 35 insertions(+)
create mode 100644 config/arm/arm64_ampereonex_linux_gcc
diff --git a/config/arm/arm64_ampereonex_linux_
> From: fengchengwen [mailto:fengcheng...@huawei.com]
> Sent: Friday, 12 April 2024 05.28
[...]
> >> @@ -1701,12 +1696,8 @@ static inline void
> >> rte_eth_linkstatus_get(const struct rte_eth_dev *dev,
> >> struct rte_eth_link *link)
> >> {
> >> - RTE_ATOMIC(uint64_t) *src =
66 matches
Mail list logo
