On Thu, Jul 03, 2025 at 03:59:14PM +0200, Morten Brørup wrote:
> For TX mbuf fast release offload, I propose to add the mbuf mempool
> pointer to the ethdev tx queue configuration structure,
> so the ethdev TX burst operation doesn't need to fetch it from the
> first mbuf of each burst being fast f
> -Original Message-
> From: David Marchand
> Sent: Thursday 3 July 2025 15:10
>
> On Thu, Jun 26, 2025 at 11:53 AM Marat Khalili
> wrote:
> > > @@ -15,7 +15,7 @@ call_all_telemetry() {
> > > telemetry_script=$rootdir/usertools/dpdk-telemetry.py
> > > echo >$tmpoutput
> > >
The SSE rx and tx paths will be removed from the i40e, iavf and ice
drivers in the 25.11 release. Each of these drivers have faster vector
paths (AVX2 and AVX-512) which have feature parity with the soon to be
removed SSE paths. In environments where AVX2 or AVX-512 are not
supported, the scalar pa
On Wed, Jul 2, 2025 at 12:23 PM Dean Marx wrote:
> Add virtual functions to DTS framework, along with
> a field for specifying VF test runs in the config file.
>
> Signed-off-by: Patrick Robb
> Signed-off-by: Dean Marx
> ---
> dts/framework/config/test_run.py | 2 +
> dts/framewor
On Wed, Jul 2, 2025 at 12:23 PM Dean Marx wrote:
> Add a physical function NIC capability that checks
> if the test run is using PFs. Add PF capability
> requirement to all suites/cases that do not
> run without error on virtual functions.
>
> Signed-off-by: Dean Marx
> ---
> dts/framework/remo
Reviewed-by: Patrick Robb
Tested-by: Patrick Robb
On Wed, Jul 2, 2025 at 12:23 PM Dean Marx wrote:
> Remove unnecessary verification step in the start method
> of TestPmdShell which checks the link status.
>
> Signed-off-by: Dean Marx
> ---
> dts/framework/remote_session/testpmd_shell.py | 7 ---
> 1 file changed, 7 deletions(-)
>
> diff -
And FYI VF results for ConnectX-5 virtual functions for a testrun I just ran
test_suites: ERROR
vlan: PASS
test_vlan_header_insertion: PASS
test_vlan_no_receipt: PASS
test_vlan_receipt_no_stripping: PASS
test_vlan_receipt_stripping: PASS
rte_flow: FAIL
test_drop_action_ETH:
Please can you compare which libraries are built before/after your patch?
02/07/2025 21:30, Dylan Schneider:
> Hey Thomas,
> I retried a build when libpcap was absent and was able to reproduce the same
> errors as you did.
>
> Error: cannot find librte_pcapng.so.25.1 in
> dpdk-build/build-arm6
On Thu, 3 Jul 2025 10:12:10 +0200
David Marchand wrote:
> On Thu, Jul 3, 2025 at 10:01 AM Bruce Richardson
> wrote:
> >
> > On Wed, Jul 02, 2025 at 11:22:42AM -0700, Stephen Hemminger wrote:
> > > On Wed, 2 Jul 2025 17:00:45 +0100
> > > Bruce Richardson wrote:
> > >
> > > > Some older comp
On Thu, 3 Jul 2025 17:29:03 +0530
madhuker.myt...@oracle.com wrote:
> From: Madhuker Mythri
>
> In bonding PMD, member_start() function checks whether RSS(mq_mode) is
> enabled and then calling the RSS rte_eth_dev_rss_reta_update() API, which
> is returning error in-case of device configured in
https://bugs.dpdk.org/show_bug.cgi?id=1738
Bug ID: 1738
Summary: bus/pci: Secondary process initialization kills
non-DPDK network interfaces
Product: DPDK
Version: 25.03
Hardware: x86
OS: Linux
S
https://bugs.dpdk.org/show_bug.cgi?id=1739
Bug ID: 1739
Summary: net/cnxk: identincal code in branches
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
When changing the order of the Tx queues startup, the depth of the
queue is compared. If not equal to the current big log2 value, next
queue will be checked and the current one will be skipped for the
next iteration.
The mlx5_txq_get() will increase the reference count number, and the
size check n
I did a clean build on system where libpcap was NOT installed.
My changes make rte_pcapng lib depend on libpcap (this is for libpcaps DLT_USER
Macros).
So now, in turn, everything that depends on rte_pcapng also depends on libpcap.
Dumpcap, rte_graph and rte_pdump all depend on rte_pcapng, so I
03/07/2025 18:29, Dylan Schneider:
> I did a clean build on system where libpcap was NOT installed.
>
> My changes make rte_pcapng lib depend on libpcap (this is for libpcaps
> DLT_USER Macros).
> So now, in turn, everything that depends on rte_pcapng also depends on
> libpcap. Dumpcap, rte_grap
On 2025-07-01 23:08, Pathak, Pravin wrote:
-Original Message-
From: Mattias Rönnblom
Sent: Tuesday, July 1, 2025 4:16 AM
To: Pathak, Pravin ; Jerin Jacob
Cc: dev@dpdk.org; jer...@marvell.com; Chen, Mike Ximing
; Richardson, Bruce
; tho...@monjalon.net; Marchand, David
; nipun.gu...@a
The current rte_net_get_ptype() only supports parsing packets with
one 0x8100 VLAN tag or two 0x88a8 VLAN tags. This patch extends it
to support parsing packets with two 0x8100 VLAN tags.
Signed-off-by: Dengdui Huang
---
lib/net/rte_net.c | 34 ++
lib/net/rte_net.
03/07/2025 12:21, Bruce Richardson:
> Some older compilers e.g. gcc 8.5, do not support overriding
> -march=native with another architecture, leading to build warnings such
> as reported in Bugzilla (link below). Add a check for that case, and
> explicitly add the avx512 flags if necessary.
>
> No
On Thu, Jun 26, 2025 at 11:53 AM Marat Khalili wrote:
> > @@ -15,7 +15,7 @@ call_all_telemetry() {
> > telemetry_script=$rootdir/usertools/dpdk-telemetry.py
> > echo >$tmpoutput
> > echo "Telemetry commands log:" >>$tmpoutput
> > -for cmd in $(echo / | $telemetry_script | jq -r
https://bugs.dpdk.org/show_bug.cgi?id=1743
Bug ID: 1743
Summary: malloc: duplicate shadowed variable definition
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
https://bugs.dpdk.org/show_bug.cgi?id=1744
Bug ID: 1744
Summary: bus/flsmc: duplicate variable definition
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
https://bugs.dpdk.org/show_bug.cgi?id=1745
Bug ID: 1745
Summary: bonding: duplicate shadow local variables
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
https://bugs.dpdk.org/show_bug.cgi?id=1746
Bug ID: 1746
Summary: mlx5: lotsof shadow varable warnings
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
The variable 'ret' is defined twice in rte_eth_tx_queue_setup.
Can remove one.
Fixes: 1bb4a528c41f ("ethdev: fix max Rx packet length")
Signed-off-by: Stephen Hemminger
---
config/meson.build | 1 +
lib/ethdev/rte_ethdev.c | 1 -
2 files changed, 1 insertion(+), 1 deletion(-)
diff --git a
On Thu, 3 Jul 2025 10:24:07 -0700
Stephen Hemminger wrote:
> diff --git a/config/meson.build b/config/meson.build
> index f31fef216c..3fc58c53b8 100644
> --- a/config/meson.build
> +++ b/config/meson.build
> @@ -322,6 +322,7 @@ global_cflags = [
> '-Wnested-externs',
> '-Wold-s
Sure Ciara. Thanks.
I will raise a fresh new patch for bug Id: 1735. That would be better.
Thanks
Amiya
On Thu, 3 Jul 2025 at 13:48, Loftus, Ciara wrote:
> > Subject: [PATCH v3] net/iavf: fix VLAN offload strip flag
> >
> > For i40e kernel drivers which support either vlan(v1) or vlan(v2)
> >
On Tue, Jul 1, 2025 at 9:56 AM David Marchand wrote:
>
> On Mon, Jun 30, 2025 at 5:38 PM Bruce Richardson
> wrote:
> >
> > As the number of cores/cpus on platforms has increased over the years,
> > the use of coremasks rather than core-lists for identifying DPDK cores
> > has become more and more
Ensure the correct virtchnl op is called for disabling vlan stripping by
checking
if the device supports either v1 or v2 vlan capability and choosing the op
accordingly.
Bugzilla ID:1735
Fixes: 3bfad066f9b4 ("net/iavf: fix VLAN strip setting after enabling filter")
Cc: sta...@dpdk.org
Signed-off
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Thursday, 3 July 2025 16.14
>
> On Thu, Jul 03, 2025 at 03:59:14PM +0200, Morten Brørup wrote:
> > For TX mbuf fast release offload, I propose to add the mbuf mempool
> > pointer to the ethdev tx queue configuration structure,
>
On Wed, Jun 25, 2025 at 10:43 AM Marat Khalili wrote:
>
> Thank you for doing this.
>
> > +static bool
> > +ends_with(const char *str, size_t str_len, const char *tail)
>
> I too think we should have a general ends_with, I for one had to code one
> just this week. However, I do not think it shoul
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Thursday, 3 July 2025 17.21
>
> On Thu, Jul 03, 2025 at 05:12:46PM +0200, Morten Brørup wrote:
> > > From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> > > Sent: Thursday, 3 July 2025 16.14
> > >
> > > On Thu, Jul 03, 2
On Thu, Jul 03, 2025 at 05:29:26PM +0200, Morten Brørup wrote:
> > From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> > Sent: Thursday, 3 July 2025 17.21
> >
> > On Thu, Jul 03, 2025 at 05:12:46PM +0200, Morten Brørup wrote:
> > > > From: Bruce Richardson [mailto:bruce.richard...@intel.c
Reviewed-by: Dean Marx
Reviewed-by: Dean Marx
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Thursday, 3 July 2025 17.36
>
> On Thu, Jul 03, 2025 at 05:29:26PM +0200, Morten Brørup wrote:
> > > From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> > > Sent: Thursday, 3 July 2025 17.21
> > >
> > > On Thu, Jul 03, 2
For TX mbuf fast release offload, I propose to add the mbuf mempool
pointer to the ethdev tx queue configuration structure,
so the ethdev TX burst operation doesn't need to fetch it from the
first mbuf of each burst being fast free'd to the mempool.
This modification of the struct rte_eth_txconf,
On Thu, Jul 03, 2025 at 05:12:46PM +0200, Morten Brørup wrote:
> > From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> > Sent: Thursday, 3 July 2025 16.14
> >
> > On Thu, Jul 03, 2025 at 03:59:14PM +0200, Morten Brørup wrote:
> > > For TX mbuf fast release offload, I propose to add the mb
https://bugs.dpdk.org/show_bug.cgi?id=1741
Bug ID: 1741
Summary: net/intel: duplicate identical branches
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
https://bugs.dpdk.org/show_bug.cgi?id=1740
Bug ID: 1740
Summary: net/bnxt: duplicate code in branches
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
https://bugs.dpdk.org/show_bug.cgi?id=1742
Bug ID: 1742
Summary: trace: duplicate shadow of same variable
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
On Tue, Jun 24, 2025 at 3:34 PM Bruce Richardson
wrote:
>
> For real-world use with dozens if not hundreds of cores, use of core
> masks becomes unwieldy. Core lists are preferred so update the examples
> and DPDK docs to refer to core lists over core masks i.e.
>
> * use "-l" EAL option rather th
On Mon, Jun 30, 2025 at 5:38 PM Bruce Richardson
wrote:
>
> As the number of cores/cpus on platforms has increased over the years,
> the use of coremasks rather than core-lists for identifying DPDK cores
> has become more and more unwieldy. At this point, let's deprecate the
> coremask-based EAL p
On Tue, Jun 17, 2025 at 5:03 PM Stephen Hemminger
wrote:
>
> After investigating latencystat test failures, discovered it poorly
> designed to handle bursts and multiple queues.
>
> v3 - reduce patch changes, break a line that was too long
>
> Stephen Hemminger (2):
> latencystats: fix receive s
> Subject: [PATCH v3] net/iavf: fix VLAN offload strip flag
>
> For i40e kernel drivers which support either vlan(v1) or vlan(v2)
> VIRTCHNL OP,it will set strip on when setting filter on. But dpdk
> side will not change strip flag. To be consistent with dpdk side,
> explicitly disable strip again
If not passing an explicit compiler for C++, meson uses c++ which
defaults to /usr/bin/g++ on a Ubuntu system.
Explicitly choose which compiler to use for C++.
Signed-off-by: David Marchand
---
Changes since v1:
- rebased,
- moved CXX tweaking in the branch enabling headers check,
---
.ci/linux
From: Madhuker Mythri
In bonding PMD, member_start() function checks whether RSS(mq_mode) is
enabled and then calling the RSS rte_eth_dev_rss_reta_update() API, which
is returning error in-case of device configured in flow isolation-mode.
When the device configured with flow isolation mode RSS re
> > > I am talking about different thing:
> > > I think with some extra effort driver can use (in some cases)
> > > rte_mbuf_raw_free_bulk() even when
> RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE
> > > is not specified.
> > > Let say we can make txq->fast_free_mp[] an array with the same size
> as txq-
>
The SSE rx and tx paths will be removed from the i40e, iavf and ice
drivers in the 25.11 release. Each of these drivers have faster vector
paths (AVX2 and AVX-512) which have feature parity with the soon to be
removed SSE paths. In environments where AVX2 or AVX-512 are not
supported, the scalar pa
On Thu, Jul 03, 2025 at 12:16:24PM +, Ciara Loftus wrote:
> The SSE rx and tx paths will be removed from the i40e, iavf and ice
> drivers in the 25.11 release. Each of these drivers have faster vector
> paths (AVX2 and AVX-512) which have feature parity with the soon to be
> removed SSE paths.
From: Madhuker Mythri
In bonding PMD, member_start() function checks whether RSS(mq_mode) is enabled
and then calling the RSS rte_eth_dev_rss_reta_update() API, which is returning
error in-case of device configured in flow isolation-mode. When the device
configured with flow isolation mode RSS re
On Thu, Jul 03, 2025 at 03:50:46PM +0530, Amiya Ranjan Mohakud wrote:
> Ensure the correct virtchnl op is called for disabling vlan stripping by
> checking if the device supports either v1 or v2 vlan capability and
> choosing the op accordingly.
>
> Bugzilla ID:1735
> Cc: sta...@dpdk.org
> Fixes:
Ensure the correct virtchnl op is called for disabling vlan stripping by
checking if the device supports either v1 or v2 vlan capability and
choosing the op accordingly.
Bugzilla ID:1735
Cc: sta...@dpdk.org
Fixes: 3bfad066f9b4 ("net/iavf: fix VLAN strip setting after enabling filter")
Signed-off-
Ensure the correct virtchnl op is called for disabling vlan stripping by
checking if the device supports either v1 or v2 vlan capability and
choosing the op accordingly.
Bugzilla ID:1735
Cc: sta...@dpdk.org
Fixes: 3bfad066f9b4 ("net/iavf: fix VLAN strip setting after enabling filter")
Signed-off-
Some older compilers e.g. gcc 8.5, do not support overriding
-march=native with another architecture, leading to build warnings such
as reported in Bugzilla (link below). Add a check for that case, and
explicitly add the avx512 flags if necessary.
Note: it appears that it is only the "native" flag
>
> Ensure the correct virtchnl op is called for disabling vlan stripping by
> checking
> if the device supports either v1 or v2 vlan capability and choosing the op
> accordingly.
>
> Bugzilla ID:1735
> Fixes: 3bfad066f9b4 ("net/iavf: fix VLAN strip setting after enabling filter")
> Cc: sta...@d
> > I am talking about different thing:
> > I think with some extra effort driver can use (in some cases)
> > rte_mbuf_raw_free_bulk() even when RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE
> > is not specified.
> > Let say we can make txq->fast_free_mp[] an array with the same size as txq-
> > >txep[].
On Thu, Jul 3, 2025 at 10:01 AM Bruce Richardson
wrote:
>
> On Wed, Jul 02, 2025 at 11:22:42AM -0700, Stephen Hemminger wrote:
> > On Wed, 2 Jul 2025 17:00:45 +0100
> > Bruce Richardson wrote:
> >
> > > Some older compilers e.g. gcc 8.5, do not support overriding
> > > -march=native with another
On Wed, Jul 02, 2025 at 11:22:42AM -0700, Stephen Hemminger wrote:
> On Wed, 2 Jul 2025 17:00:45 +0100
> Bruce Richardson wrote:
>
> > Some older compilers e.g. gcc 8.5, do not support overriding
> > -march=native with another architecture, leading to build warnings such
> > as reported in Bugzi
Reviewed-by: Patrick Robb
On Fri, Jun 27, 2025 at 11:12 AM Thomas Wilks wrote:
> From: Luca Vizzarro
>
> The test suite name property was previously returning the class name
> instead of the way that test suite are actually named, e.g.
> TestHelloWorld instead of hello_world.
>
> This change r
On Fri, Jun 27, 2025 at 11:12 AM Thomas Wilks wrote:
> Refactor the DTS result recording system to use a hierarchical tree
> structure based on `ResultNode` and `ResultLeaf`, replacing the prior flat
> model of DTSResult, TestRunResult, and TestSuiteResult. This improves
> clarity, composability,
LGTM, thanks.
As it relates to the TREX patch, yes I will try to refactor towards using
this to handle the TREX foreground server process. I guess the one change I
would need to make is changing the default InteractiveShell timeout from 5
seconds to a larger number (like 20) since TREX can take a
Signed-off-by: Jie Liu
---
MAINTAINERS | 6 +++
doc/guides/nics/features/sxe.ini| 80 +
doc/guides/nics/features/sxe_vf.ini | 39 ++
doc/guides/nics/index.rst | 1 +
doc/guides/nics/sxe.rst | 71 ++
On Wed, Jul 2, 2025 at 12:43 PM Luca Vizzarro wrote:
> Make a distinction between example apps and app/ apps by renaming the
> build app function. Moreover, provide a dedicated helper function to
> retrieve the path to a DPDK app.
>
> Signed-off-by: Luca Vizzarro
> Reviewed-by: Paul Szczepanek
64 matches
Mail list logo
