> Add support for LZ4 compression algorithm for both API and app.
>
> In addition, enable app to test de/comp only. This option already
> exists in both app doc and arguments but isn't implemented inside the
> application.
>
> LZ4: https://urldefense.proofpoint.com/v2/url?u=https-
> 3A__github.co
On 2/14/2023 11:47 PM, Ferruh Yigit wrote:
On 2/14/2023 11:03 AM, Singh, Aman Deep wrote:
On 1/24/2023 4:17 PM, David Marchand wrote:
Reduce code duplication by introducing a helper that takes care of
transmitting, retrying if enabled and incrementing tx counter.
Signed-off-by: David Marchan
From: Long Wu
Sending LACP control packets depends on calling the bonding port's
sending function if we disable dedicated queue. In some cases app
would not call the bonding port's sending function if there are
only LACP control packets and the negotiation between the two
bonding ports will fail.
App may not support the LACP negotiation in some cases.
This patch series solves this problem and add logics to
testpmd app to support the forward of bonding port in
mode 4 with the disabled dedicated queue.
---
v2:
* Export symbol to solve the link problem.
---
Long Wu (2):
net/bonding: add in
From: Long Wu
If bonding port is mode4 with disabling dedicated queue and there
are no other packets, forward loop will not call port's TX function
and bonding port will not send LACP packets.
Add sending LACP packets periodically in forward loop to avoid
LACP negotiation failed.
Signed-off-by:
> -Original Message-
> From: Chengwen Feng
> Sent: Tuesday, February 14, 2023 3:12 PM
> To: tho...@monjalon.net; ferruh.yi...@amd.com; Ruifeng Wang
> ;
> Dongdong Liu ; Yisen Zhuang
> Cc: dev@dpdk.org
> Subject: [PATCH] net/hns3: support disable IOVA as PA mode
>
> Claim PMD supports pm
On Thu, Feb 16, 2023 at 09:28:17AM +0800, fengchengwen wrote:
> add cc maintainer
>
> On 2023/2/13 20:44, Chengwen Feng wrote:
> > The spinlock should unlock when clock_gettime() failed. This patch
> > fixes it by invoking clock_gettime() before lock.
> >
> > Fixes: 26021a715067 ("eal/bsd: suppor
On Thu, Feb 16, 2023 at 09:24:38AM +0800, fengchengwen wrote:
> On 2023/2/15 19:57, Bruce Richardson wrote:
> > On Wed, Feb 15, 2023 at 09:59:06AM +0800, fengchengwen wrote:
> >> On 2023/1/17 1:37, Bruce Richardson wrote:
> >>> Validate device operation when a device is stopped or restarted.
> >>>
On 15-Feb-23 8:46 PM, Thomas Monjalon wrote:
15/02/2023 11:29, Hemant Agrawal:
This patch replaces simple printf with fprintf for debug dump
related functions for various NXP dpaaX related drivers.
Why replacing with fprintf(stdout)?
Would it be better to provide a FILE* parameter to the fun
Hi Bili,
OK. I’m not familiar with this library, but the other function assumes 8 byte
aligned data, so this should be OK too.
And with the required function signature, I now understand why you use endian
conversions this way.
Acked-by: Morten Brørup
Med venlig hilsen / Kind r
On 2/16/2023 8:01 AM, Singh, Aman Deep wrote:
>
> On 2/14/2023 11:47 PM, Ferruh Yigit wrote:
>> On 2/14/2023 11:03 AM, Singh, Aman Deep wrote:
>>> On 1/24/2023 4:17 PM, David Marchand wrote:
Reduce code duplication by introducing a helper that takes care of
transmitting, retrying if enab
On 15/02/2023 18:28, Ferruh Yigit wrote:
On 2/15/2023 5:47 PM, Niklas Söderlund wrote:
Hi Ferruh,
Thanks for your continues effort in dealing with NFP patches.
On 2023-02-15 13:42:01 +, Ferruh Yigit wrote:
On 2/8/2023 9:15 AM, Chaoyong He wrote:
From: Peng Zhang
48-bit DMA address is s
Hi Kevin,
Thanks for your input.
On 2023-02-16 10:28:34 +, Kevin Traynor wrote:
> On 15/02/2023 18:28, Ferruh Yigit wrote:
> > On 2/15/2023 5:47 PM, Niklas Söderlund wrote:
> > > Hi Ferruh,
> > >
> > > Thanks for your continues effort in dealing with NFP patches.
> > >
> > > On 2023-02-15 1
On Wed, Feb 15, 2023 at 11:10:25AM +0100, Thomas Monjalon wrote:
> Looking for reviewers please.
I will help on this.
>
> 10/02/2023 07:30, Fengnan Chang:
> > Here is a simple test case:
> > "
> > uint64_t entry_time, time;
> > size_t size = 4096;
> > unsigned align = 4096;
> > for (int j = 0; j
> -Original Message-
> From: Niklas Soderlund
> Sent: Thursday, February 16, 2023 6:37 PM
> To: Kevin Traynor
> Cc: Ferruh Yigit ; Xueming(Steven) Li
> ; Chaoyong He ;
> dev@dpdk.org; Luca Boccassi ; oss-drivers driv...@corigine.com>; Nole Zhang ; Kevin Liu
> ; sta...@dpdk.org
> Subje
There are some devices supporting 400G speed,
and it is well standardized in IEEE.
Signed-off-by: Thomas Monjalon
---
app/test-pmd/cmdline.c | 2 ++
app/test-pmd/config.c | 2 ++
app/test-pmd/parameters.c | 3 +++
app/test/test_ethdev_link.c| 3 ++-
driv
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Thursday, 16 February 2023 11.44
>
> There are some devices supporting 400G speed,
> and it is well standardized in IEEE.
>
> Signed-off-by: Thomas Monjalon
> ---
Plain and simple, easy to...
Reviewed-by: Morten Brørup
Hello again,
On 2023-02-16 11:41:13 +0100, Chaoyong He wrote:
>
>
> > -Original Message-
> > From: Niklas Soderlund
> > Sent: Thursday, February 16, 2023 6:37 PM
> > To: Kevin Traynor
> > Cc: Ferruh Yigit ; Xueming(Steven) Li
> > ; Chaoyong He ;
> > dev@dpdk.org; Luca Boccassi ; oss-dr
On 2/16/2023 10:41 AM, Chaoyong He wrote:
>
>
>> -Original Message-
>> From: Niklas Soderlund
>> Sent: Thursday, February 16, 2023 6:37 PM
>> To: Kevin Traynor
>> Cc: Ferruh Yigit ; Xueming(Steven) Li
>> ; Chaoyong He ;
>> dev@dpdk.org; Luca Boccassi ; oss-drivers > driv...@corigine.com
The HW DMA devices supported by IOAT driver do not transition to
the "active" state until the first operation is started by the HW.
Therefore, if the user calls "rte_dma_stop()" on a device without
triggering any operations, the sequence of commands to be sent to
the HW is different, as is the fina
This patchset fixes a couple of problems with stopping and restarting an
ioat DMA device. Following the three fixes, a series of improvements are
made to the dmadev unit tests to properly validate that dmadevs work
correctly as they are started and stopped, and ensure that no other or
future driver
As part of the process of restarting a dma instance, the IOAT driver
will reset the HW addresses and state values. The read and write
indexes for SW use need to be similarly reset to keep HW and SW in
sync.
Fixes: 583f046dd404 ("dma/ioat: add start and stop")
Cc: conor.wa...@intel.com
Cc: sta...@d
When the DMA device was stopped and restarted by the driver, the control
register specifying the behaviour on error was not getting correctly
reset. This caused unit tests to fail as explicitly introduced errors
were got getting reported back.
Fix by moving the setting of the register to the start
The DMA device stop API can return an error value so check that return
value when running dmadev unit tests.
Signed-off-by: Bruce Richardson
Reviewed-by: Conor Walsh
Acked-by: Kevin Laatz
---
app/test/test_dmadev.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/app/tes
The copy tests for dmadev had separate blocks in the test function for
single copy and burst copies. Separate out the single-copy block to its
own function so that it can be re-used if necessary.
Signed-off-by: Bruce Richardson
Acked-by: Kevin Laatz
Acked-by: Chengwen Feng
---
app/test/test_dm
Validate device operation when a device is stopped or restarted.
The only complication - and gap in the dmadev ABI specification - is
what happens to the job ids on restart. Some drivers reset them to 0,
while others continue where things left off. Take account of both
possibilities in the test ca
> -Original Message-
> From: Ferruh Yigit
> Sent: Thursday, February 16, 2023 7:00 PM
> To: Chaoyong He ; Niklas Soderlund
> ; Kevin Traynor
> Cc: Xueming(Steven) Li ; dev@dpdk.org; Luca
> Boccassi ; oss-drivers ; Nole
> Zhang ; Kevin Liu ;
> sta...@dpdk.org
> Subject: Re: [PATCH] net/n
On 2/16/2023 11:11 AM, Nole Zhang wrote:
>
>
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Thursday, February 16, 2023 7:00 PM
>> To: Chaoyong He ; Niklas Soderlund
>> ; Kevin Traynor
>> Cc: Xueming(Steven) Li ; dev@dpdk.org; Luca
>> Boccassi ; oss-drivers ; Nole
>> Zhang ; Kevin
On 2/16/23 13:43, Thomas Monjalon wrote:
There are some devices supporting 400G speed,
and it is well standardized in IEEE.
Signed-off-by: Thomas Monjalon
Acked-by: Andrew Rybchenko
Hi Thomas,
Higher rates are generally implemented with multiple lanes. The lane
configuration is
critical to link negotiation success. The number of lanes configured using the
ethtool
is provided [1].
Currently, the community uses only one u32 bit (link_speed) to indicate a
rate, which
is n
Acked-by: Chengwen Feng
On 2023/2/16 19:09, Bruce Richardson wrote:
> The DMA device stop API can return an error value so check that return
> value when running dmadev unit tests.
>
> Signed-off-by: Bruce Richardson
> Reviewed-by: Conor Walsh
> Acked-by: Kevin Laatz
> ---
> app/test/test_dm
Acked-by: Chengwen Feng
On 2023/2/16 19:09, Bruce Richardson wrote:
> Validate device operation when a device is stopped or restarted.
>
> The only complication - and gap in the dmadev ABI specification - is
> what happens to the job ids on restart. Some drivers reset them to 0,
> while others c
On 2023/2/15 11:19, Dongdong Liu wrote:
> Hi Chengwen
>
> On 2023/2/9 10:32, Chengwen Feng wrote:
>> The xstats reset is useful for debugging, so add it to the ethdev
>> telemetry command lists.
>>
>> Signed-off-by: Chengwen Feng
> This patch looks good, so
> Reviewed-by: Dongdong Liu
>
> A min
On 2/16/2023 11:53 AM, fengchengwen wrote:
> On 2023/2/15 11:19, Dongdong Liu wrote:
>> Hi Chengwen
>>
>> On 2023/2/9 10:32, Chengwen Feng wrote:
>>> The xstats reset is useful for debugging, so add it to the ethdev
>>> telemetry command lists.
>>>
>>> Signed-off-by: Chengwen Feng
>> This patch lo
On 2/15/2023 5:42 PM, Rakesh Kudurumalla wrote:
> added support for testpmd application to accept
> skip_cman action while configuring policy action
>
> Signed-off-by: Rakesh Kudurumalla
A comment seems missed, please check v6 & v8 [1], can you please send a
new version with it?
[1]
https://inb
> -Original Message-
> From: Srikanth Yalavarthi
> Sent: Tuesday, February 7, 2023 9:20 PM
> To: Srikanth Yalavarthi
> Cc: dev@dpdk.org; Shivah Shankar Shankar Narayan Rao
> ; Jerin Jacob Kollanukkaran
> ; Anup Prabhu ; Prince Takkar
> ; Parijat Shukla
> Subject: [PATCH v4 10/12] app/m
On 2/16/2023 10:43 AM, Thomas Monjalon wrote:
> There are some devices supporting 400G speed,
> and it is well standardized in IEEE.
>
> Signed-off-by: Thomas Monjalon
> ---
> app/test-pmd/cmdline.c | 2 ++
> app/test-pmd/config.c | 2 ++
> app/test-pmd/parameters.c
The length of TSO payload could not fit into 16 bits provided by the
IPv4 total length and IPv6 payload length fields. Thus, deduce it
from the length of the packet.
Signed-off-by: Boleslav Stankevich
Reviewed-by: Andrew Rybchenko
---
drivers/net/virtio/virtio_rxtx.c | 25 --
The length of TSO payload could not fit into 16 bits provided by the
IPv4 total length and IPv6 payload length fields. Thus, deduce it
from the length of the packet.
Signed-off-by: Boleslav Stankevich
Reviewed-by: Andrew Rybchenko
---
drivers/net/virtio/virtio_rxtx.c | 25 --
On 2023/2/16 20:06, Ferruh Yigit wrote:
> On 2/16/2023 11:53 AM, fengchengwen wrote:
>> On 2023/2/15 11:19, Dongdong Liu wrote:
>>> Hi Chengwen
>>>
>>> On 2023/2/9 10:32, Chengwen Feng wrote:
The xstats reset is useful for debugging, so add it to the ethdev
telemetry command lists.
>
On Thu, Feb 16, 2023 at 08:42:34PM +0800, fengchengwen wrote:
> On 2023/2/16 20:06, Ferruh Yigit wrote:
> > On 2/16/2023 11:53 AM, fengchengwen wrote:
> >> On 2023/2/15 11:19, Dongdong Liu wrote:
> >>> Hi Chengwen
> >>>
> >>> On 2023/2/9 10:32, Chengwen Feng wrote:
> The xstats reset is useful
On Thu, Feb 16, 2023 at 12:54:20PM +, Bruce Richardson wrote:
> On Thu, Feb 16, 2023 at 08:42:34PM +0800, fengchengwen wrote:
> > On 2023/2/16 20:06, Ferruh Yigit wrote:
> > > On 2/16/2023 11:53 AM, fengchengwen wrote:
> > >> On 2023/2/15 11:19, Dongdong Liu wrote:
> > >>> Hi Chengwen
> > >>>
>
On 2/15/2023 4:30 PM, Shibin Koikkara Reeny wrote:
> Integrate support for the AF_XDP CNI and device plugin [1] so that the
> DPDK AF_XDP PMD can work in an unprivileged container environment.
> Part of the AF_XDP PMD initialization process involves loading
> an eBPF program onto the given netdev.
Add meson build configuration for NVIDIA BlueField-3 platform
with Cortex-A78AE cores.
Signed-off-by: Raslan Darawsheh
---
config/arm/meson.build | 16
1 file changed, 16 insertions(+)
diff --git a/config/arm/meson.build b/config/arm/meson.build
index 6442ec9596..89d8f327f1 100
This update the supproted list of devices to include
missing ConnectX-7 in some drivers and add BlueFeild-3
Signed-off-by: Raslan Darawsheh
---
doc/guides/compressdevs/mlx5.rst | 5 +++--
doc/guides/cryptodevs/mlx5.rst | 12
doc/guides/nics/mlx5.rst | 33 +
This adds the Bluefield-3 and ConnectX-7 device IDs to the
list of supported NVIDIA devices that run the MLX5 compress PMD.
The BlueField-3 device is still in development stage.
Signed-off-by: Raslan Darawsheh
---
drivers/crypto/mlx5/mlx5_crypto.c | 8
1 file changed, 8 insertions(+)
d
This updates the PCI device names for all NVIDIA BlueField
family.
Signed-off-by: Raslan Darawsheh
---
drivers/common/mlx5/mlx5_common.h | 8
drivers/common/mlx5/mlx5_common_pci.c | 2 +-
drivers/compress/mlx5/mlx5_compress.c | 4 ++--
drivers/crypto/mlx5/mlx5_crypto.c | 4 +
On Wed, Feb 15, 2023 at 12:10:23PM +0100, Morten Brørup wrote:
> +CC: Fidaullah Noonari , your name also shows
> up in the git log; perhaps you can help review this patch.
>
>
> I gave up reviewing in depth, because the existing code is not easy to
> quickly understand, and it would take too lo
this series is intentioned to update missing docs, drivers and
config for supporting NVIDIA BlueField-3 device.
---
v2: send cover letter
fix patches order
---
Raslan Darawsheh (4):
mlx: update PCI device IDs names for NVIDIA BlueField NICs
config/arm: add NVIDIA BlueField-3 platform
c
This updates the PCI device names for all NVIDIA BlueField
family.
Signed-off-by: Raslan Darawsheh
---
drivers/common/mlx5/mlx5_common.h | 8
drivers/common/mlx5/mlx5_common_pci.c | 2 +-
drivers/compress/mlx5/mlx5_compress.c | 4 ++--
drivers/crypto/mlx5/mlx5_crypto.c | 2 +
This adds the Bluefield-3 and ConnectX-7 device IDs to the
list of supported NVIDIA devices that run the MLX5 compress PMD.
The BlueField-3 device is still in development stage.
Signed-off-by: Raslan Darawsheh
---
drivers/crypto/mlx5/mlx5_crypto.c | 8
1 file changed, 8 insertions(+)
d
Add meson build configuration for NVIDIA BlueField-3 platform
with Cortex-A78AE cores.
Signed-off-by: Raslan Darawsheh
---
config/arm/meson.build | 16
1 file changed, 16 insertions(+)
diff --git a/config/arm/meson.build b/config/arm/meson.build
index 6442ec9596..89d8f327f1 100
This update the supproted list of devices to include
missing ConnectX-7 in some drivers and add BlueFeild-3
Signed-off-by: Raslan Darawsheh
---
doc/guides/compressdevs/mlx5.rst | 5 +++--
doc/guides/cryptodevs/mlx5.rst | 12
doc/guides/nics/mlx5.rst | 33 +
Currently, cipher IV length is getting used to set auth
xform IV length. Auth IV is needed for AES-GMAC case,
and in all other cases, auth IV should be 0.
Used a separate auth IV length to separate out cipher and auth cases.
Fixes: 9413c3901f31 ("examples/ipsec-secgw: support additional algorithms
Currently, sessions are created without checking the device
capabilities, which may result in failure at a later stage.
Device capabilities are now checked before creating the
security/crypto session.
Signed-off-by: Akhil Goyal
---
examples/ipsec-secgw/ipsec.c | 208
In cases of inline IPsec, the supported ol_flags are
retrieved from security capability of device.
Now that capability checks are added before creating the session,
ol_flags can be retrieved from the same function call.
Signed-off-by: Akhil Goyal
---
examples/ipsec-secgw/ipsec.c | 65 ++-
On Tue, Jan 24, 2023 at 01:24:38PM +0530, Sam Kirubakaran wrote:
> Hi Liang,
>
> Thank you so much for your response.
> I would like to convert our in-house tool into a high performance traffic
> generator and it is written in Python.
> Is it possible to use any of the user space TCP stack by comp
[...]
>> +
>> +if (rte_pmu_init() < 0)
>> +return TEST_FAILED;
>
>Can we return TEST_SKIPPED here?
>On aarch64, this feature requires kernel version >= 5.17. CI setups doesn't
>meet this requirement
>will start to report failure when running fast_tests.
>
Okay. I think that's goo
added support for testpmd application to accept
skip_cman action while configuring policy action
Below is sample command given after configuring meter
profile for which congestion configuration is applied
for red and yellow color packets and bypassed for
green color packets in dpdk-testpmd applica
Dropping of packets is based on action configured
to meter.If both skip_red and drop actions are configured
then tail dropping in invoked else if only drop action is
configured then RED drop is invoked.This action is supported
only when RED is configured using rte_eth_cman_config_set()
Signed-off-
Introduce new flow action to skip congestion management configuration
This feature helps to skip the congestion management processing
based on per flow or the packet color identified by
rte_flow meter object. For example, If one Rx queue configured as
RED congestion and application wants tobypass t
> -Original Message-
> From: Liang Ma
> Sent: Thursday, February 16, 2023 2:40 PM
> To: Sam Kirubakaran
> Cc: dev@dpdk.org
> Subject: Re: TCP stack support on DPDK
>
> On Tue, Jan 24, 2023 at 01:24:38PM +0530, Sam Kirubakaran wrote:
> > Hi Liang,
> >
> > Thank you so much for your respon
On 2/2/2023 1:47 PM, Gregory Etelson wrote:
> Current API allows either query or update indirect flow action.
> If indirect action must be conditionally updated according to it's
> present state application must first issue action query then
> analyze returned data and if needed issue update reques
On 2/16/2023 2:51 PM, Rakesh Kudurumalla wrote:
> Introduce new flow action to skip congestion management configuration
> This feature helps to skip the congestion management processing
> based on per flow or the packet color identified by
> rte_flow meter object. For example, If one Rx queue confi
On 2/16/2023 6:28 AM, Chaoyong He wrote:
> Remove the unneeded check logic of ingress and egress flow attribute.
> After the OVS commit c7da49bc ("netdev-offload-dpdk: Fix transfer flows"),
> the flow can not offload anymore if we don't do this modification.
>
> Signed-off-by: Chaoyong He
> Revie
On 2/16/2023 8:32 AM, Chaoyong He wrote:
> From: Long Wu
>
> If bonding port is mode4 with disabling dedicated queue and there
> are no other packets, forward loop will not call port's TX function
> and bonding port will not send LACP packets.
>
> Add sending LACP packets periodically in forward
On 2/16/2023 6:36 AM, Chaoyong He wrote:
> From: Jin Liu
>
> The flow flush function will destroy all the flow table on all the
> port of NFP nic rather than the provide port. Modify logic, only
> destroy the flow table on the corresponding port.
>
> Fixes: 30ecce522732 ("net/nfp: support flow A
For the title, I suggest
ethdev: add Tx queue mapping of aggregated ports
14/02/2023 16:48, Jiawei Wang:
> When multiple ports are aggregated into a single DPDK port,
> (example: Linux bonding, DPDK bonding, failsafe, etc.),
> we want to know which port use for Tx via a queue.
>
> This patch intr
For the title, I suggest
ethdev: add flow matching of aggregated port
14/02/2023 16:48, Jiawei Wang:
> When multiple ports are aggregated into a single DPDK port,
> (example: Linux bonding, DPDK bonding, failsafe, etc.),
> we want to know which port is used for Rx and Tx.
>
> This patch allows to
This series adds self monitoring support i.e allows to configure and
read performance measurement unit (PMU) counters in runtime without
using perf utility. This has certain advantages when application runs on
isolated cores running dedicated tasks.
Events can be read directly using rte_pmu_read()
Add support for programming PMU counters and reading their values
in runtime bypassing kernel completely.
This is especially useful in cases where CPU cores are isolated
i.e run dedicated tasks. In such cases one cannot use standard
perf utility without sacrificing latency and performance.
Signed
Add support for reading Intel x86_64 PMU events in runtime.
Signed-off-by: Tomasz Duszynski
Acked-by: Morten Brørup
---
app/test/test_pmu.c | 2 ++
lib/pmu/meson.build | 1 +
lib/pmu/rte_pmu.h| 2 ++
lib/pmu/rte_pmu_pmc_x86_64.h | 24
4
Add support for reading ARM PMU events in runtime.
Signed-off-by: Tomasz Duszynski
Acked-by: Morten Brørup
Acked-by: Ruifeng Wang
---
app/test/test_pmu.c | 4 ++
lib/pmu/meson.build | 7 +++
lib/pmu/pmu_arm64.c | 94 +
lib/pmu/rte_p
In order to profile app one needs to store significant amount of samples
somewhere for an analysis latern on. Since trace library supports
storing data in a CTF format lets take adventage of that and add a
dedicated PMU tracepoint.
Signed-off-by: Tomasz Duszynski
Acked-by: Morten Brørup
---
app
On 2/14/2023 3:48 PM, Jiawei Wang wrote:
> When multiple ports are aggregated into a single DPDK port,
> (example: Linux bonding, DPDK bonding, failsafe, etc.),
> we want to know which port use for Tx via a queue.
>
> This patch introduces the new ethdev API
> rte_eth_dev_map_aggr_tx_affinity(), i
> -Original Message-
> From: Tomasz Duszynski
> Sent: Friday, February 17, 2023 1:55 AM
> To: dev@dpdk.org
> Cc: roret...@linux.microsoft.com; Ruifeng Wang ;
> bruce.richard...@intel.com; jer...@marvell.com; mattias.ronnb...@ericsson.com;
> m...@smartsharesystems.com; tho...@monjalon.net;
Google Virtual NIC is not doing IPv4 checksummimg.
Removed that capability from PMD.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/gve_ethdev.c
index 97781f0ed3..e357f16e16 100644
---
Hi Team,
I am trying to set-up 8 Tx-Queues (but only 1 Rx-Queue) and burst traffic
out of different Tx-Queues from the same cpu core on an ixgbe nic (10G).
Although transmitted packets reach the peer, reading the statistics
indicates only the Tx-q[0] has non-zero packets and bytes count, the rest
TX side statistics is updated.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_tx.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/gve/gve_tx.c b/drivers/net/gve/gve_tx.c
index d4e52e3ea5..5c8a3b4379 100644
--- a/drivers/net/gve/gve_tx.c
+++ b/drivers
Google Virtual NIC PMD is enriched with statistics info.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 147 ++-
drivers/net/gve/gve_ethdev.h | 11 +++
drivers/net/gve/gve_rx.c | 18 -
drivers/net/gve/gve_tx.c | 6 ++
4 files change
NIC statistics are enhanced with rx/tx queue errors.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 49 +++-
1 file changed, 43 insertions(+), 6 deletions(-)
diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/gve_ethdev.c
index 735847ede7
Google Virtual NIC is not doing IPv4 checksummimg.
Removed that capability from PMD.
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/gve_ethdev.c
index 06d1b796c8..fef2458a16 100644
---
Google Virtual NIC PMD is enriched with extended statistics info.
eth_dev_ops callback names are also synched with eth_dev_ops field names
Signed-off-by: Levend Sayar
---
drivers/net/gve/gve_ethdev.c | 152 ++-
drivers/net/gve/gve_rx.c | 8 +-
2 files change
On Thu, 16 Feb 2023 23:40:27 +0530
Rajasekhar Pulluru wrote:
> Hi Team,
>
> I am trying to set-up 8 Tx-Queues (but only 1 Rx-Queue) and burst traffic
> out of different Tx-Queues from the same cpu core on an ixgbe nic (10G).
> Although transmitted packets reach the peer, reading the statistics
>
Makes sense. The virtual device only does L4 checksum.
Acked-by: Rushil Gupta
On Thu, Feb 16, 2023 at 10:58 AM Levend Sayar wrote:
> Google Virtual NIC is not doing IPv4 checksummimg.
> Removed that capability from PMD.
>
> Signed-off-by: Levend Sayar
> ---
> drivers/net/gve/gve_ethdev.c | 1
PMU is kind of MSR. Precisely , that's MSR per core.
All MSR reading will lead to IPI(Please reference the kernel
implementation of MSR driver). The IPI will disturb the DPDK application
because the userspace/kernel context switch, which has impact to the
tail latency.
On Thu, Feb 16, 2023 at 03:01:28PM +, Van Haaren, Harry wrote:
> > -Original Message-
> > From: Liang Ma
> > Sent: Thursday, February 16, 2023 2:40 PM
> > To: Sam Kirubakaran
> > Cc: dev@dpdk.org
> > Subject: Re: TCP stack support on DPDK
> >
> > On Tue, Jan 24, 2023 at 01:24:38PM +0
16/02/2023 17:54, Tomasz Duszynski пишет:
Add support for programming PMU counters and reading their values
in runtime bypassing kernel completely.
This is especially useful in cases where CPU cores are isolated
i.e run dedicated tasks. In such cases one cannot use standard
perf utility without
16/02/2023 13:30, Ferruh Yigit:
> On 2/16/2023 10:43 AM, Thomas Monjalon wrote:
> > There are some devices supporting 400G speed,
> > and it is well standardized in IEEE.
> >
> > Signed-off-by: Thomas Monjalon
> > ---
> > app/test-pmd/cmdline.c | 2 ++
> > app/test-pmd/config.c
This patch is to enable outer checksum offload on avx512 Tx path
for tunnel packet by adding Tx path with context descriptor
and adjusting path select logic.
Signed-off-by: Yiding Zhou
Signed-off-by: Zhichao Zeng
---
v3: fix tunneling parameters
---
v2: fix path select logic
---
drivers/net/ia
> -Original Message-
> From: Zeng, ZhichaoX
> Sent: Friday, February 17, 2023 9:49 AM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Zhou, YidingX
> ; Zhang, Qi Z ; Xu, Ke1
> ; Zeng, ZhichaoX ; Wu,
> Jingjing ; Xing, Beilei ;
> Richardson, Bruce ; Konstantin Ananyev
>
> Subject: [PATCH v3] ne
Hi Kamalakshitha,
> -Original Message-
> From: Kamalakshitha Aligeri
> Sent: Tuesday, February 14, 2023 2:18 AM
> To: Zhang, Yuying ; Xing, Beilei
> ; Matz, Olivier ;
> andrew.rybche...@oktetlabs.ru; Richardson, Bruce
> ; m...@smartsharesystems.com;
> konstantin.anan...@huawei.com; honnap
Liang Ma 于2023年2月16日周四 22:04写道：
>
> On Wed, Feb 15, 2023 at 12:10:23PM +0100, Morten Brørup wrote:
> > +CC: Fidaullah Noonari , your name also shows
> > up in the git log; perhaps you can help review this patch.
> >
> >
> > I gave up reviewing in depth, because the existing code is not easy to
>
Follow the advice from community reviewer, we get rid of the use of
RTE log level type and RTE_LOG_*() macro, and also wrap the rte_log()
with our own log macro.
Chaoyong He (3):
net/nfp: add the log source file
net/nfp: get rid of the usage of RTE log level type
net/nfp: get rid of the usag
Prepare for adding more log functionality by moving the existing log
functionality to its own file.
Signed-off-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/meson.build | 1 +
drivers/net/nfp/nfp_common.c | 3 ---
drivers/net/nfp/nfp_logs.c | 10 ++
3 files chan
Register the own RX/TX debug log level type, and get rid of the
usage of RTE_LOGTYPE_*. Then we can control the log by a independent
switch.
Signed-off-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/nfp_logs.c | 10 ++
drivers/net/nfp/nfp_logs.h | 8 ++--
2 files
Replace the usage of RTE_LOG* macro with PMD specific logging.
Signed-off-by: Chaoyong He
Reviewed-by: Niklas Söderlund
---
drivers/net/nfp/flower/nfp_flower.c| 10 ++---
drivers/net/nfp/flower/nfp_flower_ctrl.c | 2 +-
drivers/net/nfp/nfp_cpp_bridge.c | 48 ++--
https://bugs.dpdk.org/show_bug.cgi?id=1161
Bug ID: 1161
Summary: [dpdk-23.03]virtio_user_as_exceptional_path/vhost_exce
ption_path_with_virtio_user: launch testpmd as
virtio-user failed
Product: DPDK
Version: 22.0
Ok Stephen, thanks for the information, I can try that.
One of the problems I see with single Tx Queue mode is that Ixia reports
packet drops, though I confirmed with the help of counters (before invoking
tx burst) that all packets are being sent-out. Dumping HW counters don't
report any drops in
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, February 17, 2023 1:58 AM
> To: Jiawei(Jonny) Wang ; Slava Ovsiienko
> ; Ori Kam ; NBU-Contact-
> Thomas Monjalon (EXTERNAL) ;
> andrew.rybche...@oktetlabs.ru; Aman Singh ;
> Yuying Zhang
> Cc: dev@dpdk.org; Raslan Darawsheh
> S
1 - 100 of 118 matches
Mail list logo
