In secondary processes, directly accessing 'dev->data->dev_private' can
cause a segmentation fault if the primary process has exited or if the
shared memory is no longer accessible. If a secondary process attempts
to query device information (e.g., via testpmd), a NULL dereference
may occur due to
In secondary processes, directly accessing 'dev->data->dev_private' can
cause a segmentation fault if the primary process has exited or if the
shared memory is no longer accessible.
The secondary process not only crashes on device close but also segfaults
when executing commands like "show device
In secondary processes, directly accessing 'dev->data->dev_private' can
cause a segmentation fault if the primary process has exited or if the
shared memory is no longer accessible. If a secondary process attempts
to query device information (e.g., via testpmd), a NULL dereference
may occur due to
Hi Stephen,
Many Thanks for the feedback.
I have addressed all the comments on the patch, I have removed the unnecessary
check in ethdev, the only one check in rte_ethdev seems to be a neccessary one,
as without it, it can still generate segfault while secondary exists after
primary.
Best rega
Hi Stephen,
I will take care of it and resubmit.
Thanks
Sunil Kumar Kori
From: Stephen Hemminger
Sent: Tuesday, July 29, 2025 9:59 AM
To: Sunil Kumar Kori
Cc: Shepard Siegel ; Ed Czeck
; John Miller ; Igor
Russkikh ; Ajit Khaparde ;
Somnath Kotur ; Nithin Kumar Dabilpuram
; Kiran Kumar Kok
Hi Ivan,
No, it does not breaking testpmd or any other applications.
Yes, the structure is correct in rte_net.h and comment as well.
You can run testpmd in rxonly mode and set verbose to 3, send the following
packet from remote and you will observe the output as given below:
pkt=Ether(src="0
On Wed, 2 Jul 2025 15:21:39 +0530
Gagandeep Singh wrote:
> V1-logs:
> This patch series introduces enhancements and fixes to the
> NXP DPAA2 Ethernet driver.
> It includes support for
> - software taildrop on ordered queues.
> - setup speed capabilities.
> - DPAA2 resource version.
> - MAC l
On Mon, 12 May 2025 20:37:19 +0530
wrote:
> From: Sunil Kumar Kori
>
> As rte_eth_rx_queue_count() returns signed value to represent
> the error cases but internally invoked callback is returning
> unsigned value. Hence unnecessary type conversion is done.
>
> To avoid this typecasting from si
On Mon, 7 Jul 2025 12:17:51 +0100
Bruce Richardson wrote:
> Two small improvements for the cmdline file processing in testpmd.
>
> * Now that we support multiple files, change the prompt to indicate what
> file is currently being processed, and just print an EOF message when
> done.
> * Whe
On Tue, 15 Jul 2025 12:53:30 +
Bruce Richardson wrote:
> For testing purposes it can be desirable to enable RSS even when only a
> single queue per port is in use. Add an "enable-rss" flag to testpmd to
> match the existing "disable-rss" flag.
>
> Signed-off-by: Bruce Richardson
> ---
Appl
On Thu, 8 May 2025 10:36:38 +0800
Sunyang Wu wrote:
> Support testing the flow tables query rate.
>
> Signed-off-by: Sunyang Wu
>
> ---
Applied to next-net for 25.11
Add new commands to enable testing of port mirror functionality.
Signed-off-by: Stephen Hemminger
---
app/test-pmd/cmdline.c | 13 ++
app/test-pmd/cmdline_mirror.c | 194
app/test-pmd/cmdline_mirror.h | 13 ++
app/test-pmd/m
Use the new port mirror API instead of pdump.
Since filtering is now done in dumpcap process (not primary).
For portability use PCAP to do packet filtering.
This removes dependency on DPDK implementation of BPF.
Filtering will be slower (no JIT) but is portable to Windows.
Can revisit later when B
In new model, the packet was already copied, only need
to wrap it in pcapng format.
Signed-off-by: Stephen Hemminger
---
lib/pcapng/rte_pcapng.c | 178 +---
lib/pcapng/rte_pcapng.h | 27 +-
2 files changed, 120 insertions(+), 85 deletions(-)
diff --git a
The queue field is optional in pcapng received packet.
Use UINT16_MAX as flag value.
Signed-off-by: Stephen Hemminger
---
lib/pcapng/rte_pcapng.c | 7 ---
lib/pcapng/rte_pcapng.h | 4 ++--
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/lib/pcapng/rte_pcapng.c b/lib/pcapng/rte
Simple API and packet mirroring standalone tests.
Signed-off-by: Stephen Hemminger
---
app/test/meson.build | 1 +
app/test/test_ethdev_mirror.c | 325 ++
2 files changed, 326 insertions(+)
create mode 100644 app/test/test_ethdev_mirror.c
diff --git a
This adds new feature port mirroring to the ethdev layer.
Signed-off-by: Stephen Hemminger
---
config/rte_config.h | 1 +
lib/ethdev/ethdev_driver.h | 6 +
lib/ethdev/ethdev_private.c | 58 -
lib/ethdev/ethdev_private.h | 3 +
lib/ethdev/ethdev_trace.h
Add names for new port mirror fields.
Signed-off-by: Stephen Hemminger
---
lib/mbuf/rte_mbuf_dyn.h | 9 +
1 file changed, 9 insertions(+)
diff --git a/lib/mbuf/rte_mbuf_dyn.h b/lib/mbuf/rte_mbuf_dyn.h
index 865c90f579..72a31b22b3 100644
--- a/lib/mbuf/rte_mbuf_dyn.h
+++ b/lib/mbuf/rte_m
Need ability to allow process like dumpcap to make a ring ethdev
with a pre-existing ring. Do this via devargs so it can work
with hotplug.
It looked like the API rte_eth_from_ring() would work for this,
but it doesn't do the right thing. That API makes new vdev but
names the ring based on the par
Need to be able have multiple threads all using same
transmit queue when using SPAN.
Signed-off-by: Stephen Hemminger
---
drivers/net/ring/rte_eth_ring.c | 17 +
1 file changed, 17 insertions(+)
diff --git a/drivers/net/ring/rte_eth_ring.c b/drivers/net/ring/rte_eth_ring.c
index
Use null device to exercise ethdev start/stop in secondary process.
Signed-off-by: Stephen Hemminger
---
app/test/test_mp_secondary.c | 51 +---
1 file changed, 48 insertions(+), 3 deletions(-)
diff --git a/app/test/test_mp_secondary.c b/app/test/test_mp_secondar
Best not to depend on other EAL headers to include stuff.
Some of the later changes had build failures on some platforms.
Signed-off-by: Stephen Hemminger
---
lib/ethdev/ethdev_private.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/lib/ethdev/ethdev_private.c b/lib/ethdev/ethdev_
Before this patch if secondary process called start/stop
it would only impact the secondary process, the ethdev on the
primary process was not started.
With this patch, when start/stop is called from secondary,
it calls the primary and does the operation there. The design
is generic, and we can la
This is a rework of how packet capture is done in DPDK.
The existing mechanism using callbacks has a number of problems:
- can't work when packets are sent and received in secondary process
because callbacks only function in the primary process
- requires "opt-in" from application
The new
The node class contains a configure hugepages method to be called
on the DTS nodes if the user has specified a hugepages config value
in their nodes configurations. However, the commit referenced below
removed the code path to this function, making it unreachable. If
the user has configured hugepag
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
> Sent: Monday, 28 July 2025 17.11
>
> On Mon, Jul 28, 2025 at 04:51:30PM +0200, Morten Brørup wrote:
> > > From: Dean Marx [mailto:dm...@iol.unh.edu]
> > > Sent: Friday, 18 July 2025 15.18
> > >
> > > On Fri, Jul 18, 2025 at 4:23 AM Bru
> From: Konstantin Ananyev [mailto:konstantin.anan...@huawei.com]
> Sent: Monday, 28 July 2025 17.42
>
> > > Hi Morten,
> > >
> > > Good patch. Please see below.
> > >
> > > On Sat, 26 Jul 2025, Morten Brørup wrote:
> > >
> > > > Added fast mbuf release, re-using the existing mbuf pool pointer
> >
> > Hi Morten,
> >
> > Good patch. Please see below.
> >
> > On Sat, 26 Jul 2025, Morten Brørup wrote:
> >
> > > Added fast mbuf release, re-using the existing mbuf pool pointer
> > > in the queue structure.
> > >
> > > Signed-off-by: Morten Brørup
> > > ---
> > > v2:
> > > * Also announce the o
All supported linux versions support F_ADD_SEALS (Linux 3.17+) and
MFD_HUGETLB (Linux 4.14+), so no need to have #ifdefs and fallbacks to
support systems which do not support those features.
Signed-off-by: Bruce Richardson
Acked-by: Stephen Hemminger
---
lib/vhost/vhost_user.c | 7 ---
1 fi
All supported linux versions support F_ADD_SEALS (Linux 3.17+) and
MFD_HUGETLB (Linux 4.14+), so no need to have #ifdefs and fallbacks to
support systems which do not support those features.
Signed-off-by: Bruce Richardson
Acked-by: Stephen Hemminger
---
drivers/net/memif/rte_eth_memif.h | 41 -
All supported linux versions support F_ADD_SEALS (Linux 3.17+) and
MFD_HUGETLB (Linux 4.14+), so no need to have #ifdefs and fallbacks to
support systems which do not support those features.
Signed-off-by: Bruce Richardson
Acked-by: Stephen Hemminger
---
lib/eal/linux/eal_memalloc.c | 174 +
We no longer support versions of linux before 4.19[1], therefore
all supported linux versions support F_ADD_SEALS (Linux 3.17+) and
MFD_HUGETLB (Linux 4.14+), so no need to have #ifdefs and fallbacks to
support systems which do not support those features.
[1] https://doc.dpdk.org/guides/linux_gsg/
On Mon, Jul 28, 2025 at 04:51:30PM +0200, Morten Brørup wrote:
> > From: Dean Marx [mailto:dm...@iol.unh.edu]
> > Sent: Friday, 18 July 2025 15.18
> >
> > On Fri, Jul 18, 2025 at 4:23 AM Bruce Richardson
> > wrote:
> > >
> > > On Thu, Jul 17, 2025 at 05:03:13PM -0400, Dean Marx wrote:
> > > > I'v
> From: Dean Marx [mailto:dm...@iol.unh.edu]
> Sent: Friday, 18 July 2025 15.18
>
> On Fri, Jul 18, 2025 at 4:23 AM Bruce Richardson
> wrote:
> >
> > On Thu, Jul 17, 2025 at 05:03:13PM -0400, Dean Marx wrote:
> > > I've created a v1 of a QinQ test suite around the set of test cases
> > > discusse
On Mon, Jul 28, 2025 at 7:31 AM Luca Vizzarro wrote:
> Really good catch Patrick! Thank you for this!
>
> Would it make more sense to place setup_hugepages under Node.setup()?
> In that case the method can also remain private, involving a single line
> change.
>
Yes, that makes more sense. I'll
Hi Morten,
On Mon, 28 Jul 2025, Morten Brørup wrote:
From: Ivan Malov [mailto:ivan.ma...@arknetworks.am]
Sent: Saturday, 26 July 2025 08.15
Hi Morten,
Good patch. Please see below.
On Sat, 26 Jul 2025, Morten Brørup wrote:
Added fast mbuf release, re-using the existing mbuf pool pointer
in
Hi,
On Mon, 28 Jul 2025, Khadem Ullah wrote:
Fix incorrect inner_l2_len values for VXLAN, VXLAN-GPE, GTPU, and Geneve.
These protocols carry full Ethernet frames, so inner_l2_len should be
set to the size of an Ethernet header (14 bytes), not include tunnel or
UDP headers.
Does the bug break
> From: Ivan Malov [mailto:ivan.ma...@arknetworks.am]
> Sent: Saturday, 26 July 2025 08.15
>
> Hi Morten,
>
> Good patch. Please see below.
>
> On Sat, 26 Jul 2025, Morten Brørup wrote:
>
> > Added fast mbuf release, re-using the existing mbuf pool pointer
> > in the queue structure.
> >
> > Si
Fix incorrect inner_l2_len values for VXLAN, VXLAN-GPE, GTPU, and Geneve.
These protocols carry full Ethernet frames, so inner_l2_len should be
set to the size of an Ethernet header (14 bytes), not include tunnel or
UDP headers.
Fixes: 64ed7f854c ('net: add tunnel packet type parsing')
Cc: sta...@
Fix incorrect inner_l2_len values for VXLAN, VXLAN-GPE, GTPU, and Geneve.
These protocols carry full Ethernet frames, so inner_l2_len should be
set to the size of an Ethernet header (14 bytes), not include tunnel or
UDP headers.
Fixes: 64ed7f854c ('net: add tunnel packet type parsing')
Cc: sta...@
Fix incorrect inner_l2_len values for VXLAN, VXLAN-GPE, GTPU, and Geneve.
These protocols carry full Ethernet frames, so inner_l2_len should be
set to the size of an Ethernet header (14 bytes), not include tunnel or
UDP headers.
Fixes: 44fb6c0f51 ('net: initial tunnel type parser')
Cc: sta...@dpdk
Looks good to me.
Reviewed-by: Luca Vizzarro
Really good catch Patrick! Thank you for this!
Would it make more sense to place setup_hugepages under Node.setup()?
In that case the method can also remain private, involving a single line
change.
Reviewed-by: Luca Vizzarro
On Thu, Jun 19, 2025 at 01:36:56PM +, Ciara Loftus wrote:
> Enable Tx QINQ offload if the VF reports support for inserting both an
> outer and inner VLAN tag. The VF capabilities report the locations for
> placing each of the tags - either L2TAG1 in the tx descriptor or L2TAG2
> in the context
On 28/07/2025 10:14, Yanghang Liu wrote:
> I tested below 18 scenarios on RHEL 9.6 host and didn't find any new dpdk
> issues.
>
>- VM with device assignment(PF) throughput testing(1G hugepage size):
>PASS
>- VM with device assignment(PF) throughput testing(2M hugepage size) :
>PAS
On Mon, 28 Jul 2025 at 10:17, Yanghang Liu wrote:
>
> I tested below 18 scenarios on RHEL 9.6 host and didn't find any new dpdk
> issues.
>
> VM with device assignment(PF) throughput testing(1G hugepage size): PASS
> VM with device assignment(PF) throughput testing(2M hugepage size) : PASS
> VM w
I tested below 18 scenarios on RHEL 9.6 host and didn't find any new dpdk
issues.
- VM with device assignment(PF) throughput testing(1G hugepage size):
PASS
- VM with device assignment(PF) throughput testing(2M hugepage size) :
PASS
- VM with device assignment(VF) throughput testing
I tested below 18 scenarios on RHEL 9.6 host and didn't find any new dpdk
issues.
- VM with device assignment(PF) throughput testing(1G hugepage size):
PASS
- VM with device assignment(PF) throughput testing(2M hugepage size) :
PASS
- VM with device assignment(VF) throughput testing
From: Jie Liu
Add simd function.
Signed-off-by: Jie Liu
---
drivers/net/sxe/base/sxe_queue_common.c | 54 +-
drivers/net/sxe/base/sxe_rx_common.c| 128 -
drivers/net/sxe/pf/sxe_vec_common.h | 321
drivers/net/sxe/pf/sxe_vec_neon.c | 602 +++
From: Jie Liu
Add custom cmd led ctrl.
Signed-off-by: Jie Liu
---
drivers/net/sxe/include/sxe/sxe_cli.h | 218
drivers/net/sxe/include/sxe/sxe_ioctl.h | 21 +++
drivers/net/sxe/sxe_testpmd.c | 66 +++
3 files changed, 305 insertions(+)
create mode 10
From: Jie Liu
Add dcb function.
Signed-off-by: Jie Liu
---
drivers/net/sxe/meson.build | 28 +
drivers/net/sxe/pf/rte_pmd_sxe.h| 2 +
drivers/net/sxe/pf/sxe_dcb.c| 946
drivers/net/sxe/pf/sxe_dcb.h| 99 +++
drivers/net/sx
From: Jie Liu
Add ptp module.
Signed-off-by: Jie Liu
---
drivers/net/sxe/pf/sxe_ptp.c | 209 +++
drivers/net/sxe/pf/sxe_ptp.h | 26 +
2 files changed, 235 insertions(+)
create mode 100644 drivers/net/sxe/pf/sxe_ptp.c
create mode 100644 drivers/net/sxe/pf/
From: Jie Liu
Add xstats function.
Signed-off-by: Jie Liu
---
drivers/net/sxe/pf/sxe_main.c | 8 +
drivers/net/sxe/pf/sxe_stats.c | 589 +
drivers/net/sxe/pf/sxe_stats.h | 68
3 files changed, 665 insertions(+)
create mode 100644 drivers/net/sxe/pf/sxe
From: Jie Liu
Support rss offload.
Signed-off-by: Jie Liu
---
drivers/net/sxe/base/sxe_offload_common.c | 11 +-
drivers/net/sxe/pf/sxe_offload.c | 299 ++
drivers/net/sxe/pf/sxe_offload.h | 33 +++
3 files changed, 338 insertions(+), 5 deletions(-)
dif
From: Jie Liu
Add filter function.
Signed-off-by: Jie Liu
---
drivers/net/sxe/pf/sxe_filter.c | 191
drivers/net/sxe/pf/sxe_filter.h | 29 +
drivers/net/sxe/pf/sxe_main.c | 1 +
3 files changed, 221 insertions(+)
diff --git a/drivers/net/sxe/pf/sxe_fi
From: Jie Liu
Support vlan filter.
Signed-off-by: Jie Liu
---
drivers/net/sxe/base/sxe_offload_common.c | 4 +
drivers/net/sxe/pf/sxe_filter.c | 277 ++
drivers/net/sxe/pf/sxe_filter.h | 13 +
drivers/net/sxe/pf/sxe_main.c | 2 +
4 files
From: Jie Liu
Add link, flow ctrl, mac ops, mtu ops function.
Signed-off-by: Jie Liu
---
drivers/net/sxe/meson.build| 19 +-
drivers/net/sxe/pf/sxe.h | 36 +-
drivers/net/sxe/pf/sxe_ethdev.c| 638 +++-
drivers/net/sxe/pf/sxe_filter.c| 278 +
From: Jie Liu
Adding a minimum maintainable directory structure for the
network driver and request maintenance of the sxe driver.
Signed-off-by: Jie Liu
---
MAINTAINERS | 6 +++
doc/guides/nics/features/sxe.ini| 55 ++
doc/guides/nics/features/s
Hi,
> Hi, Konstantin
>
> To illustrate this issue, I made the following modifications to the test
> cases:
> Modify the “test_data” content of the function "test_acl.c:
> test_build_ports_range":
> delete the second test data, and leave only one test message data. This way,
> when
> the funct
[AMD Official Use Only - AMD Internal Distribution Only]
Snipped
> > [Public]
> >
> > Hi Morten,
> >
> > We have tested the effect of the patch using func-latency and PPs via
> > testpmd.
> > Please find our observations below
> >
> > - DPDK tag: 25.07-rc1
> > - compiler: gcc 14.2
> > - platfo
The test dummy vector data is incorrect for AEAD decryption, so we should
only run when passed a valid test vector file. This patch also moves the
check for correct test vector data to after the vector has been
populated by test vector file or dummy data.
Signed-off-by: Emma Finn
---
app/test-cr
61 matches
Mail list logo
