> -Original Message-
> From: Shijith Thotton
> Sent: Thursday, February 6, 2025 8:03 PM
> To: Jerin Jacob
> Cc: dev@dpdk.org; Pavan Nikhilesh Bhagavatula
> ; Shijith Thotton ;
> sta...@dpdk.org; Naga Harish K S V
> Subject: [PATCH] test/event: fix number of queues in eventdev conf
>
v2:
- Added RSS capabilities check.
- Added testsuite that tests RSS hashes.
- Added testsuite that tests updating the RSS hash key.
- Added testsuite that verifies the RETA size.
Alex Chapman (5):
dts: add RSS functions to testpmd
dts: add utils for PMD RSS testsuites
dts: add PMD RSS hash
From: Alex Chapman
This patch adds the required functionality for the RSS key_update,
RETA, and hash test suites. This includes:
The setting of custom RETA values for routing packets to specific
queues.
The setting of the RSS mode on all ports, to specify how to hash
the packets.
The updating of
From: Alex Chapman
To reduce the amount of maintenance and code duplication,
common functionality between the rss_key_update, pmd_rss_reta
and pmd_rss_hash test suites has been collated into a single
file called rss_utils.
It contains 3 main functions:
1. verify that a packets RSS hash correctly
On Tue, Feb 25, 2025 at 05:21:10PM +, Bruce Richardson wrote:
> On Tue, Feb 25, 2025 at 08:37:27AM -0800, Andre Muezerie wrote:
> > On Tue, Feb 25, 2025 at 09:03:37AM +, Bruce Richardson wrote:
> > > On Mon, Feb 24, 2025 at 01:01:18PM -0800, Andre Muezerie wrote:
> > > > Top level 'cc_avx2_
On Tue, Feb 25, 2025 at 09:03:37AM +, Bruce Richardson wrote:
> On Mon, Feb 24, 2025 at 01:01:18PM -0800, Andre Muezerie wrote:
> > Top level 'cc_avx2_flags' was created and holds the correct flags
> > depending on the compiler used.
> >
> > File meson.build was updated to handle the correct A
> -Original Message-
> From: Bing Zhao
> Sent: Tuesday, February 25, 2025 17:03
> To: Slava Ovsiienko ; dev@dpdk.org; Raslan
> Darawsheh
> Cc: Ori Kam ; Dariusz Sosnowski ;
> Suanming Mou ; Matan Azrad ;
> Hamdan Agbariya
> Subject: [PATCH] net/mlx5: add eCPRI support
>
> Support eCP
On Tue, Feb 25, 2025 at 09:06:26AM +, Bruce Richardson wrote:
> On Mon, Feb 24, 2025 at 01:01:17PM -0800, Andre Muezerie wrote:
> > Top level 'cc_avx2_flags' was created and holds the correct flags
> > depending on the compiler used.
> >
> > Signed-off-by: Andre Muezerie
> > ---
> > drivers/
On Mon, Feb 24, 2025 at 02:43:07PM +0100, Burakov, Anatoly wrote:
> On 21/02/2025 17:04, Bruce Richardson wrote:
> > The mbuf cleanup function is not being called correctly on close of the
> > driver, due to the vector_tx flag not being set correctly on each Tx
> > queue. Normally, this causes no v
On Tue, Feb 25, 2025 at 08:37:27AM -0800, Andre Muezerie wrote:
> On Tue, Feb 25, 2025 at 09:03:37AM +, Bruce Richardson wrote:
> > On Mon, Feb 24, 2025 at 01:01:18PM -0800, Andre Muezerie wrote:
> > > Top level 'cc_avx2_flags' was created and holds the correct flags
> > > depending on the comp
Updates are added to enable releasing of unused credits by threads
to avoid deadlock and starving of other active threads needing credits.
It also adds compile time options to enable/disable credit checks.
Signed-off-by: Pravin Pathak
Signed-off-by: Tirthendu Sarkar
---
.mailmap
On Mon, Feb 24, 2025 at 01:01:16PM -0800, Andre Muezerie wrote:
> Up to now MSVC has being used with the default mode, which uses SSE2
> instructions for scalar floating-point and vector calculations.
> https://learn.microsoft.com/en-us/cpp/build/reference/arch-x64?view=msvc-170
>
> This patch all
From: Alex Chapman
Port over the rss_key_update test suite from old DTS. This
suite verifies that setting a new hash key when Receive Side
Scaling (RSS) will result in a change in the packets destination
queue. This test suite also verifies that the reported key size
of the NIC is correct.
Signe
From: Alex Chapman
Port over the pmd_rss_hash test suite from old DTS. This
suite verifies that the 4 supported types of hashing
algorithm used in Receive Side Scaling (RSS) function
correctly. Them being DEFAULT, TOEPLITZ
SYMMETRIC_TOEPLITZ and SIMPLE_XOR. This test suite also
verifies the suppo
From: Alex Chapman
Port over the pmd_rss_reta test suite from old DTS. This
suite verifies that Redirection Tables (RETAs) of different
sizes function correctly in Receive Side Scaling (RSS).
This test suite also verifies that the reported reta size
of the NIC is correct.
Signed-off-by: Alex Cha
Added checks for if a nic supports the simple_xor,
symmetric_toeplitz, symmetric_toeplitz_sort,
toeplitz, and default hashing algorithms.
Signed-off-by: Thomas Wilks
Reviewed-by: Paul Szczepanek
---
dts/framework/remote_session/testpmd_shell.py | 146 ++
dts/tests/TestSuite_pmd
This commit queries the FW for the new unified FDB mode and saves it in
mlx5 shared device as fdb_unified_en bit.
Signed-off-by: Ophir Munk
---
drivers/common/mlx5/mlx5_devx_cmds.c | 3 +++
drivers/common/mlx5/mlx5_devx_cmds.h | 1 +
2 files changed, 4 insertions(+)
diff --git a/drivers/common/
Legacy FDB domain is enhanced with optional three sub-domains: FDB_RX
FDB_TX and FDB_UNIFIED. Based on FW capability bit mlx5 will work in
either legacy or unified mode. To work in fdb unified mode the flow
attribute transfer bit must be set and the fdb sub domains are
determined by the table speci
From: Bing Zhao
The "send_to_kernel" action is shared and created during the HWS
configuration stage. By default, the NIC Rx and Tx flags are needed.
There is only one case that the FDB* flags are needed when creating
the action, the same as other shared actions. Only the transfer proxy
port nee
In validating Jump from group A to default miss group B DPDK requests
that both groups types will be identical. In practice jumping from
FDB_UNIFIED to FDB_RX or FDB_TX group types (or vice versa) is allowed.
This commit enables this jumping between the different types.
Fixes: b81f95ca770d ("net/m
On Tue, Feb 25, 2025 at 09:12:36AM +, Mingjin Ye wrote:
> The minimum free packet threshold (tx_free_thresh) and the minimum RS bit
> threshold (tx_rs_thresh) both have a default value of 32. Therefore, the
> default minimum number of ring descriptors value is 64.
>
> For reference, see "Confi
Support eCPRI item matching over L2 (ETHER / VLAN) in HWS, both for
template API and backward compatibility API.
Signed-off-by: Bing Zhao
---
drivers/net/mlx5/hws/mlx5dr_definer.c | 86 +++
drivers/net/mlx5/mlx5_flow.h | 3 +
drivers/net/mlx5/mlx5_flow_hw.c
> -Original Message-
> From: Ophir Munk
> Sent: Tuesday, February 25, 2025 13:02
> To: Dariusz Sosnowski ; Slava Ovsiienko
> ; Bing Zhao ; Ori Kam
> ; Suanming Mou ; Matan Azrad
>
> Cc: dev@dpdk.org; Raslan Darawsheh ; Jiawei(Jonny)
> Wang
> Subject: [PATCH V1 3/4] net/mlx5: remove un
> -Original Message-
> From: Ophir Munk
> Sent: Tuesday, February 25, 2025 13:02
> To: Dariusz Sosnowski ; Slava Ovsiienko
> ; Bing Zhao ; Ori Kam
> ; Suanming Mou ; Matan Azrad
>
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH V1 1/4] common/mlx5: support FDB unified capabilit
> -Original Message-
> From: Ophir Munk
> Sent: Tuesday, February 25, 2025 13:02
> To: Dariusz Sosnowski ; Slava Ovsiienko
> ; Bing Zhao ; Ori Kam
> ; Suanming Mou ; Matan Azrad
>
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH V1 2/4] net/mlx5: support FDB unified domain
>
>
> -Original Message-
> From: Ophir Munk
> Sent: Tuesday, February 25, 2025 13:02
> To: Dariusz Sosnowski ; Slava Ovsiienko
> ; Bing Zhao ; Ori Kam
> ; Suanming Mou ; Matan Azrad
>
> Cc: dev@dpdk.org; Raslan Darawsheh
> Subject: [PATCH V1 4/4] net/mlx5/hws: allow different types in mis
> > -Original Message-
> > From: Stephen Hemminger
> > Sent: Thursday, February 20, 2025 10:25 PM
> > To: dev@dpdk.org
> > Cc: Stephen Hemminger ; Morten Brørup
> > ; Richardson, Bruce
> > ; Gujjar, Abhinandan S
> > ; Jerin Jacob ; Amit
> > Prakash Shukla ; Naga Harish K, S V
> >
> > Sub
> -Original Message-
> From: Pavan Nikhilesh Bhagavatula
> Sent: Tuesday, February 25, 2025 6:48 PM
> To: Gujjar, Abhinandan S ; Stephen
> Hemminger ; dev@dpdk.org
> Cc: Morten Brørup ; Richardson, Bruce
> ; Jerin Jacob ; Amit
> Prakash Shukla ; Naga Harish K, S V
>
> Subject: RE: [PATC
Hi,
On 1/27/25 8:41 AM, Kommula Shiva Shankar wrote:
This patch adds virtio hash report feature, which is
supported in packet queue mode with scalar version
Why only restrict to packed ring? isn't it also compatible with split
ring?
Signed-off-by: Kommula Shiva Shankar
---
drivers/net/vi
On Tue, 25 Feb 2025 18:42:52 +0800
Junlong Wang wrote:
> V3:
> - fix warning enable asserts compile.
> - fix warning enable address sanitizer for undefined checks.
> - modify extra memset and cast.
> - remove __rte_unused from function prototypes.
> - fix indentation in mtr get and use
On Tue, 25 Feb 2025 16:41:06 +0800
Wenbo Cao wrote:
> diff --git a/doc/guides/nics/rnp.rst b/doc/guides/nics/rnp.rst
> index 99b96e9b8e..c3547c38b6 100644
> --- a/doc/guides/nics/rnp.rst
> +++ b/doc/guides/nics/rnp.rst
> @@ -71,6 +71,10 @@ Listed below are the rte_eth functions supported:
> * ``
On 2/24/25 11:35 AM, Gowrishankar Muthukrishnan wrote:
Until virtio avail ring is initialized (by VHOST_USER_SET_VRING_ADDR),
worker thread should not try to fetch crypto op, which would lead to
memory fault.
Fixes: 939066d96563 ("vhost/crypto: add public function implementation")
Cc: sta...@
On Tue, 25 Feb 2025 09:34:54 +0800
Chaoyong He wrote:
> From: Long Wu
>
> The 'ipackets'/'opackets' are used to record the number
> of packets on represetor port received/sent. But the
> code does not consider concurrent calculation of
> 'ipackets'/'opackets'. If multiple queues are calculated
101 - 133 of 133 matches
Mail list logo
