> -Original Message-
> From: Huang, Wei
> Sent: Monday, February 21, 2022 3:52 PM
> To: dev@dpdk.org; Xu, Rosen ; Zhang, Qi Z
> ; nipun.gu...@nxp.com; hemant.agra...@nxp.com
> Cc: sta...@dpdk.org; Zhang, Tianfei ; Yigit, Ferruh
> ; Huang, Wei
> Subject: [PATCH v5] raw/ifpga: initialize
Previously, each time a burst of packets is received, SW reads HW
register and assembles it and the timestamp from descriptor together to
get the complete 64 bits timestamp.
This patch optimizes the algorithm. The SW only needs to check the
monotonicity of the low 32bits timestamp to avoid crossin
> -Original Message-
> From: Bruce Richardson
> Sent: Tuesday, February 22, 2022 5:41 PM
> To: Ma, WenwuX
> Cc: Burakov, Anatoly ; dev@dpdk.org; Hu,
> Jiayu ; Wang, Yinan ; He,
> Xingguang
> Subject: Re: [PATCH v3] examples/multi_process: reconfigure port when rss
> or csum isn't support
Trusted VF is needed to offload rules with rte_flow to a group
that is bigger than 0.
The configuration is done in two parts: driver and FW.
This patch adds the needed steps to configure a VF to be trusted.
Signed-off-by: Asaf Penso
Reviewed-by: Raslan Darawsheh
---
doc/guides/nics/mlx5.rst |
The AGE action can be implemented by either counters or ASO mechanism.
ASO is more efficient than generating counters just for the purpose of
aging, so when ASO is supported its use is preferable. On the other
hand, when there is count in the list of actions, the counter is already
generated, and i
Fix mistakes done in floe_dv_validate() for count and AGE actions
combination checking.
Michael Baum (2):
net/mlx5: fix overridden flag in flow validation
net/mlx5: fix insufficient check in count action validation
drivers/net/mlx5/mlx5_flow_dv.c | 43 -
1 fil
The AGE action can be implemented by either counters or ASO mechanism.
When user ask count action in the flow rule, AGE action is implemented
by the same counter. However, if user ask indirect count action, it
cannot be used for AGE.
The flow_dv_validate() function has a flag named "shared_count"
[Public]
Hi Roy Fan,
-Original Message-
From: Zhang, Roy Fan
Sent: Friday, February 25, 2022 7:21 PM
To: Akhil Goyal ; Ji, Kai ; dev@dpdk.org
Cc: Namburu, Chandu-babu
Subject: RE: [EXT] [dpdk-dev v5] crypto/openssl: openssl 3.0 support on sym
crypto routine
Hi Akhil,
> -Original
From: Satheesh Paul
Added support for RTE_FLOW_ITEM_TYPE_MARK. This item type
can be used to create ingress flow rules to match packets
from CPT(second pass packets).
Signed-off-by: Satheesh Paul
---
doc/guides/nics/cnxk.rst | 7 +++
doc/guides/nics/features/cnxk.ini | 1 +
From: Satheesh Paul
Added support to create flow rules to match packets
from CPT(second pass packets). With this change, ingress
rules will be created with bits 10 and 11 of channel field
in the MCAM ignored by default. For rules specific to
second pass packets, the CPT channel bits will be set
i
The octeontx2 9xxx SoC family support is added.
Signed-off-by: Mahipal Challa
---
drivers/compress/octeontx/include/zip_regs.h | 12
drivers/compress/octeontx/otx_zip.c | 6 +-
drivers/compress/octeontx/otx_zip.h | 1 +
drivers/compress/octeontx/otx_zip_pmd.c
Testpmd forwards packets in checksum mode that it needs to calculate
the checksum of each layer's protocol.
In process_inner_cksums, when parsing tunnel packets, inner L4 offset should be
outer_l2_len + outer_l3_len + l2_len + l3_len.
In process_outer_cksums, when parsing tunnel packets, outer L4
Testpmd forwards packets in checksum mode that it needs to calculate
the checksum of each layer's protocol.
When setting the hardware calculates the outer UDP checksum and the
software calculates the outer IP checksum, the dev->tx_pkt_burst in
ice_set_tx_function is set to ice_xmit_pkts_vec_avx2.
This patchset fixes two related issues:
* When outer UDP uses HW to calculate the checksum, select basic path to
process the package.
* Correctly modify the calculation method of inner and outer L4 offset.
v2:
- Separate testpmd fix and pmd fix into two patches.
Kevin Liu (2):
net/ice: fix
27/02/2022 22:32, Stephen Hemminger:
> On Sun, 27 Feb 2022 21:27:40 +0100
> Thomas Monjalon wrote:
>
> > 25/02/2022 18:47, Stephen Hemminger:
> > > The text file did not end with newline.
> > >
> > > Signed-off-by: Stephen Hemminger
> >
> > Series applied, thanks for the cleanup.
> >
> > Co
On Sun, 27 Feb 2022 21:27:40 +0100
Thomas Monjalon wrote:
> 25/02/2022 18:47, Stephen Hemminger:
> > The text file did not end with newline.
> >
> > Signed-off-by: Stephen Hemminger
>
> Series applied, thanks for the cleanup.
>
> Could we add a check in checkpatch to avoid it in future?
>
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v22.03-rc2
There are 306 new patches in this snapshot.
Release notes:
https://doc.dpdk.org/guides/rel_notes/release_22_03.html
Highlights of 22.03-rc2:
- ethdev flow API for templates and
20/02/2022 19:21, Stephen Hemminger:
> + * If NULL then, the function does nothing.
I'm not English native, but I thought it should be one of these 2 forms:
- If NULL, the function does nothing.
- If NULL then the function does nothing.
25/02/2022 19:07, Stephen Hemminger:
> Even these are ini files, they should still have new line at
> end of file.
>
> Signed-off-by: Stephen Hemminger
Squashed with your other patch for examples, thanks.
25/02/2022 18:47, Stephen Hemminger:
> The text file did not end with newline.
>
> Signed-off-by: Stephen Hemminger
Series applied, thanks for the cleanup.
Could we add a check in checkpatch to avoid it in future?
+Cc Ferruh
05/12/2021 07:21, Harold Huang:
> When kni driver calls eth_kni_start to start device, some fields such as
> min_mtu and max_mtu of rte_kni_conf are not initialized. It will cause
> kni_ioctl_create create a kni netdevice with a random min_mtu and max_mtu
> value. This isunexpected and
15/02/2022 20:11, Ferruh Yigit:
> On 2/14/2022 6:41 PM, Ferruh Yigit wrote:
> > On 2/9/2022 7:35 AM, Min Hu (Connor) wrote:
> >> From: Huisong Li
> >>
> >> The "kni_dev" is the private data of the "net_device" in kni, and allocated
> >> with the "net_device" by calling "alloc_netdev()". The "net_d
19/01/2022 15:50, Michal Krawczyk:
> Patch changing the way of accessing interrupt handle also changed order
> of the rte_pci_map_device() call and rte_pci_device:driver assignment.
> It was causing issues with Write Combine mapping on the Linux platform
> if it was used with the igb_uio module.
>
15/02/2022 12:20, Madhuker Mythri:
> From: Ferruh Yigit
> >On 2/14/2022 5:08 PM, madhuker.myt...@oracle.com wrote:
> >> From: Madhuker Mythri
> >>
> >> Failsafe pmd started crashing with global devargs syntax as devargs is
> >> not memset to zero. Access it to in rte_devargs_parse() resulted i
On Sun, 27 Feb 2022 18:24:42 +0100
Thomas Monjalon wrote:
> 27/02/2022 17:44, Stephen Hemminger:
> > On Sun, 27 Feb 2022 17:46:17 +0200
> > Asaf Penso wrote:
> >
> > > +#. For each VF PCIe, using the following command to bind the driver::
> > > +
> > > + $ echo ":82:00.2" >> /sys/bus/pc
23/02/2022 12:26, Ferruh Yigit:
> On 2/23/2022 8:49 AM, Steve Yang wrote:
> > 'recv()' fills the 'buf', later 'strlcpy()' used to copy from this buffer.
> > But as coverity warns 'recv()' doesn't guarantee that 'buf' is
> > null-terminated, but 'strlcpy()' requires it.
> >
> > Enlarge 'buf' size t
17/02/2022 16:24, David Hunt:
> Hi Bruce,
>
> On 17/2/2022 3:02 PM, Bruce Richardson wrote:
> > Coverity flags the fact that the tag values used in distributor are
> > 32-bit, which means that when we use bit-manipulation to convert a tag
> > match/no-match to a bit in an array, we need to typecas
25/02/2022 10:27, Pablo de Lara:
> Coverity flags that both elements of efd_online_group_entry
> are used uninitialized. This is OK because this structure
> is initially used for starting values, so any value is OK.
>
> Coverity ID: 375868
> Fixes: 56b6ef874f80 ("efd: new Elastic Flow Distributor
27/02/2022 17:44, Stephen Hemminger:
> On Sun, 27 Feb 2022 17:46:17 +0200
> Asaf Penso wrote:
>
> > +#. For each VF PCIe, using the following command to bind the driver::
> > +
> > + $ echo ":82:00.2" >> /sys/bus/pci/drivers/mlx5_core/bind
> > \ No newline at end of file
>
> Please change
25/02/2022 04:12, eagost...@nvidia.com:
> From: Elena Agostini
>
> To enable the gpudev rte_gpu_mem_cpu_map feature to expose
> GPU memory to the CPU, the GPU CUDA driver library needs
> the GDRCopy library and driver.
>
> If DPDK is built without GDRCopy, the GPU CUDA driver returns
> error if
25/02/2022 04:12, eagost...@nvidia.com:
The features list were missed when introducing the driver.
Fixes: 1306a73b1958 ("gpu/cuda: introduce CUDA driver")
Cc: sta...@dpdk.org
> Signed-off-by: Elena Agostini
> ---
> doc/guides/gpus/features/cuda.ini | 12
> 1 file c
On Sun, 27 Feb 2022 17:46:17 +0200
Asaf Penso wrote:
> +#. For each VF PCIe, using the following command to bind the driver::
> +
> + $ echo ":82:00.2" >> /sys/bus/pci/drivers/mlx5_core/bind
> \ No newline at end of file
Please change your editor and/or git settings so there is a new line
25/02/2022 15:55, Tomasz Duszynski:
> Started device should eventually be stopped.
>
> Fixes: 0e6557b448fa ("raw/cnxk_gpio: add self test")
>
> Signed-off-by: Tomasz Duszynski
> Reviewed-by: Jerin Jacob Kollanukkaran
Applied, thanks.
Trusted VF is needed to offload rules with rte_flow to a group
that is bigger than 0.
The configuration is done in two parts: driver and FW.
This patch adds the needed steps to configure a VF to be trusted.
Signed-off-by: Asaf Penso
Reviewed-by: Raslan Darawsheh
---
doc/guides/nics/mlx5.rst |
When trying to dequeue, an OP may fail due to insufficient space for the
OP output, the compressdev API defines out-of-space for OP status.
The driver can detect out-of-space errors and report them to the user.
Check if hw_error_syndrome specifies out-of-space and set the OP
status accordingly.
Al
26/02/2022 04:48, Zhiheng Chen:
> The original download link is invalid.
>
> Signed-off-by: Zhiheng Chen
> ---
> doc/guides/nics/mlx5.rst | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
> index c3cc0c0f41..625d8a23
36 matches
Mail list logo
