Hello Everyone
We are trying out the l3fwd example application using lpm. However, the
packets being sent are not being forwarded to the required port. We're not
sure if the lpm routes added to the table are wrong, or if the
configuration of the setup is wrong. How do we proceed?
Regards
Tarun
On Fri, Mar 6, 2020 at 10:35 AM Gavin Hu wrote:
>
> Tx mbuf free is a hotspot for i40e on aarch64, as there are no
> inter-loop dependencies, it is safe to enable auto-vectorization
> to speed up.
>
> This patch showed 2~3% performance lift on ThunderX2 and no degradation
> on Arm N1SDP. The test
Hi Morten,
Thanks for the comments.
> -Original Message-
> From: Morten Brørup
> Sent: Thursday, March 5, 2020 5:52 PM
> To: Suanming Mou ; Olivier Matz
> ; Xueming(Steven) Li
> Cc: Andrew Rybchenko ; dev@dpdk.org; Asaf
> Penso ; Ori Kam
> Subject: RE: [dpdk-dev] [RFC] mempool: introdu
test app to measures the performance of libipsec
api rte_ipsec_pkt_crypto and rte_ipsec_pkt_process.
Signed-off-by: Savinay Dharmappa
Acked-by: Konstantin Ananyev
---
MAINTAINERS| 2 +
app/test/Makefile | 2 +-
app/test/meson.build | 2 +
app/test/test_ipsec
Hi Luca,
Testing with dpdk v19.11.1-rc1 from Red Hat looks good.
We cover below 13 scenarios and and all get PASS on RHEL8 testing:
(1)Guest with device assignment(PF) throughput testing(1G hugepage size): PASS
(2)Guest with device assignment(PF) throughput testing(2M hugepage size) : PASS
(3)
> -Original Message-
> From: Ori Kam [mailto:or...@mellanox.com]
> Sent: Tuesday, March 3, 2020 4:49 PM
> To: Zhang, Xiao ; dev@dpdk.org
> Cc: arybche...@solarflare.com; Yigit, Ferruh
> Subject: RE: [dpdk-dev] ethdev: add PFCP header to flow API
>
> Hi Xiao,
>
> Sorry for delayed resp
This patch adds the new flow item RTE_FLOW_ITEM_TYPE_PFCP to flow API to
match a PFCP header.
Add sample PFCP rules for testpmd guide. Since Session Endpoint
Identifier (SEID) only will be present in PFCP Session header and PFCP
Session headers shall be identified when the S field is equal to 1, wh
Hi Mattias,
Have a question on this fix. I understand you wanting a certain number of
events before making a decision to migrate (in the above fix).
However, suppose there are fewer events over a few flows (even if not many)
and yet your core is heavily loaded -- indicating may be they are one or
m
Modified Secuirty gateway application to support load/unload
esp-ah ddp package on i40e NIC from ipsec-secgw application.
Signed-off-by: Praveen Shetty
---
examples/ipsec-secgw/ipsec-secgw.c | 67 +++-
examples/ipsec-secgw/ipsec.c | 158 +
examples/ipse
restrict pointer aliasing to optimize the code generated.
The patch showed ~3% performnace uplift on Arm N1SDP platform, and no
degradation on ThunderX2. The tet case is RFC2544 zero-loss L2
forwarding running testpmd.
[1] https://gcc.gnu.org/onlinedocs/gcc-4.8.5/gcc/Restricted-Pointers.html
Sig
Tx mbuf free is a hotspot for i40e on aarch64, as there are no
inter-loop dependencies, it is safe to enable auto-vectorization
to speed up.
This patch showed 2~3% performance lift on ThunderX2 and no degradation
on Arm N1SDP. The test case is single core RFC2544 zero-loss test.
Signed-off-by: Ga
This series is to optimize the i40e vPMD performance on aarch64.
The patches were benchmarked by running the following command on Marvell
ThunderX2 and Arm N1SDP and showed positive performance results.
sudo ./build/app/testpmd -l 1,3 -w 0001:01:00.0 -w 0001:01:00.1
--master-lcore 1 -- -i --rxq=4
To keep ordering of mixed accesses, rte_cio is sufficient.
The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
This patch fixes by replacing with just sufficient barriers in the
normal PMD and vPMD.
It showed 7% performance uplift on ThunderX2 and 4% on Arm N1SDP.
The test case is th
On 2020/3/6 上午1:14, Alex Williamson wrote:
On Tue, 25 Feb 2020 14:09:07 +0800
Jason Wang wrote:
On 2020/2/25 上午10:33, Tian, Kevin wrote:
From: Alex Williamson
Sent: Thursday, February 20, 2020 2:54 AM
Changes since v1 are primarily to patch 3/7 where the commit log is
rewritten, along with
From: Wang Shougang
This patch unified the Rx ptype table.
Signed-off-by: Wang Shougang
---
drivers/net/iavf/iavf.h | 3 +-
drivers/net/iavf/iavf_ethdev.c| 3 +
drivers/net/iavf/iavf_rxtx.c | 604 +++---
drivers/net/iavf/iavf_rxtx.h
Hi Souvik,
The patch http://patches.dpdk.org/patch/63951/ has been queued to LTS release
18.11.7 by Kevin.
And the release date should be April 2020 according to the release plan on
http://core.dpdk.org/roadmap/.
BR,
Xiao
From: Dey, Souvik [mailto:so...@rbbn.com]
Sent: Thursday, March 5, 2020
For pci devices presented through igb_uio, pcidev->mem_resource[] is
not populated when the device is initialized for secondary process.
Initialize pcidev->mem_resource[] with pci-bar mapped addresses.
Signed-off-by: Vijaya Mohan Guvva
---
drivers/bus/pci/pci_common_uio.c | 1 +
1 file changed,
>
> 05/03/2020 12:45, Ananyev, Konstantin:
> > I think the header need to be "app/test", or "test/ipsec".
>
> It should be "test/ipsec" as it is an IPsec test command
> in the test application.
>
> > Apart from that:
> > Acked-by: Konstantin Ananyev
>
> I wonder why we have a different main
Igor, Evgeny,
Please, review.
On 3/5/20 5:23 PM, Vlad Zolotarov wrote:
There is a funny logic that seems to be outdated which tries to detect
a situation when a user requests a default size of the queue and
configures a device specific default value.
This seems to be not in line with the DPDK c
There is a funny logic that seems to be outdated which tries to detect
a situation when a user requests a default size of the queue and
configures a device specific default value.
This seems to be not in line with the DPDK current design that expects
the user to go via the rte_eth_dev_info_get() i
On Thu, 5 Mar 2020 15:10:48 -0500
Lance Richardson wrote:
> Hi Stephen,
>
> On Thu, Mar 5, 2020 at 1:45 AM Stephen Hemminger
> wrote:
> >
>
> >
> > Make the configuration use the same as other drivers with
> > vector mode: ixge, i40e, ...
> s/ixge/ixgbe/?
>
> >
>
> > This will also mak
On Thu, 27 Feb 2020 18:34:07 +0100
Cornelia Huck wrote:
> On Wed, 19 Feb 2020 11:54:18 -0700
> Alex Williamson wrote:
>
> > The VFIO_DEVICE_FEATURE ioctl is meant to be a general purpose, device
> > agnostic ioctl for setting, retrieving, and probing device features.
> > This implementation pro
Hi Stephen,
On Thu, Mar 5, 2020 at 1:45 AM Stephen Hemminger
wrote:
>
>
> Make the configuration use the same as other drivers with
> vector mode: ixge, i40e, ...
s/ixge/ixgbe/?
>
> This will also make future support of vector mode on other
> architectures possible.
>
> Fixes: bc4a000f2f53 ("n
On Thu, Mar 5, 2020 at 10:22 AM Alex Williamson
wrote:
> On Tue, 25 Feb 2020 03:08:00 +
> "Tian, Kevin" wrote:
>
> > > From: Alex Williamson
> > > Sent: Thursday, February 20, 2020 2:54 AM
> > >
> > > With the VF Token interface we can now expect that a vfio userspace
> > > driver must be in
> Honnappa and Konstantin contributed actively to the ring library
> and volunteered to replace myself as maintainers.
>
> Signed-off-by: Olivier Matz
> ---
> MAINTAINERS | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index c3785554f..db23
> >
> > > > > +/**
> > > > > + * @internal Reserve ring elements to enqueue several objects on
> > > > > +the ring
> > > > > + *
> > > > > + * @param r
> > > > > + * A pointer to the ring structure.
> > > > > + * @param esize
> > > > > + * The size of ring element, in bytes. It must be a mult
On Tue, 25 Feb 2020 03:08:00 +
"Tian, Kevin" wrote:
> > From: Alex Williamson
> > Sent: Thursday, February 20, 2020 2:54 AM
> >
> > With the VF Token interface we can now expect that a vfio userspace
> > driver must be in collaboration with the PF driver, an unwitting
> > userspace driver wi
On Tue, 25 Feb 2020 02:59:37 +
"Tian, Kevin" wrote:
> > From: Alex Williamson
> > Sent: Thursday, February 20, 2020 2:54 AM
> >
> > If we enable SR-IOV on a vfio-pci owned PF, the resulting VFs are not
> > fully isolated from the PF. The PF can always cause a denial of service
> > to the VF
On 3/5/2020 4:35 PM, Thomas Monjalon wrote:
> 05/03/2020 17:06, Ferruh Yigit:
>> On 3/5/2020 3:11 PM, Thomas Monjalon wrote:
>>> 05/03/2020 15:55, Ferruh Yigit:
FDIR -> Flow Director
>>>
>>> In general I prefer avoiding FDIR for two reasons:
>>> 1/ this is an Intel-only acronym
>>
>> Yes, it
Hi Kevin,
Sorry for the delay, I've been out on PTO...
On Tue, 25 Feb 2020 02:33:27 +
"Tian, Kevin" wrote:
> > From: Alex Williamson
> > Sent: Thursday, February 20, 2020 2:54 AM
> >
> > Changes since v1 are primarily to patch 3/7 where the commit log is
> > rewritten, along with option pa
On Thu, Mar 5, 2020 at 10:17 PM ZY Qiu wrote:
>
> When compiling with -O0,
> the compiler does not optimize two memory accesses into one.
> Leads to accessing a null pointer when queue post Rx burst callback
> removal while traffic is running.
>
> Signed-off-by: ZY Qiu
> ---
> lib/librte_ethdev/
On Tue, 25 Feb 2020 14:09:07 +0800
Jason Wang wrote:
> On 2020/2/25 上午10:33, Tian, Kevin wrote:
> >> From: Alex Williamson
> >> Sent: Thursday, February 20, 2020 2:54 AM
> >>
> >> Changes since v1 are primarily to patch 3/7 where the commit log is
> >> rewritten, along with option parsing and fai
05/03/2020 12:45, Ananyev, Konstantin:
> I think the header need to be "app/test", or "test/ipsec".
It should be "test/ipsec" as it is an IPsec test command
in the test application.
> Apart from that:
> Acked-by: Konstantin Ananyev
I wonder why we have a different maintainer for each IPsec test
Add support for underlying Intel IPSec Multi-buffer library v0.53.
Signed-off-by: Pablo de Lara
---
drivers/crypto/aesni_gcm/aesni_gcm_ops.h | 65 ++---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 130 +-
drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c | 4 +-
.../c
Add support for underlying IPSec Multi-buffer library v0.53.
Signed-off-by: Pablo de Lara
---
drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c| 476 --
.../crypto/aesni_mb/rte_aesni_mb_pmd_ops.c| 205 +---
.../aesni_mb/rte_aesni_mb_pmd_private.h | 30 +-
3 files chan
Link against Intel IPSec Multi-buffer library, which
added support for KASUMI-F8 and KASUMI-F9 from version v0.53,
moving from libSSO KASUMI library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh| 4 +-
doc/guides/cryptodevs/kasumi.rst | 62 +++
Link against Intel IPSec Multi-buffer library, which
added support for SNOW3G-UEA2 and SNOW3G-UIA2 from version v0.53,
moving from libSSO SNOW3G library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh| 4 +-
doc/guides/cryptodevs/snow3g.rst | 58 +++
Link against Intel IPSec Multi-buffer library, which
added support for ZUC-EEA3 and ZUC-EIA3 from version v0.53,
moving from libSSO ZUC library.
Signed-off-by: Pablo de Lara
---
devtools/test-build.sh | 6 +--
doc/guides/cryptodevs/zuc.rst| 52 -
This patchset adds support to the following crypto PMDs to use
Intel IPSec MB v0.53, in DPDK v18.11:
- AESNI MB PMD: had support up to v0.52, extending to v0.53
- AESNI GCM PMD: had support up to v0.52, extending to v0.53
- SNOW3G PMD: linking now to IPSec MB v0.53, instead of libsso
- ZUC PMD: lin
> >
> > Having an API that could be used by parallel hardware does make sense,
> > but the DPDK already has multiple packet processing infrastructure pieces.
> >
> > I would rather the DPDK converge on one widely used, robust and tested
> > packet method. Rather than the current "choose your poison
> -Original Message-
> From: Olivier Matz
> Sent: Thursday, March 5, 2020 11:00 AM
> To: dev@dpdk.org
> Cc: Honnappa Nagarahalli ; Konstantin
> Ananyev
> Subject: [PATCH] maintainers: update for ring library
>
> Honnappa and Konstantin contributed actively to the ring library and
> vo
Honnappa and Konstantin contributed actively to the ring library
and volunteered to replace myself as maintainers.
Signed-off-by: Olivier Matz
---
MAINTAINERS | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index c3785554f..db235c2cc 100644
--- a/M
28/02/2020 12:10, Andrew Rybchenko:
> On 2/26/20 7:32 PM, Jerin Jacob wrote:
> > On Wed, Feb 26, 2020 at 10:00 PM Olivier Matz
> > wrote:
> >>
> >> On Mon, Feb 17, 2020 at 05:46:42PM +0100, Thomas Monjalon wrote:
> >>> Write first lines of the README file about the intent (upstream first)
> >>> a
Hi,
On Thu, Mar 05, 2020 at 04:20:40PM +0800, xiangxia.m@gmail.com wrote:
> From: Tonghao Zhang
>
> The order of mempool initiation affects mempool index in the
> rte_mempool_ops_table. For example, when building APPs with:
>
> $ gcc -lrte_mempool_bucket -lrte_mempool_ring ...
>
> The "buc
When compiling with -O0,
the compiler does not optimize two memory accesses into one.
Leads to accessing a null pointer when queue post Rx burst callback
removal while traffic is running.
Signed-off-by: ZY Qiu
---
lib/librte_ethdev/rte_ethdev.h | 9 +
1 file changed, 5 insertions(+), 4 d
Having taken feedback from the community into account, we would like to propose
some changes to our approach for combining multiple packet-processing functions
into a single operation on a single device, be that an optimized software
library or a hardware accelerator.
The main feedback on the r
05/03/2020 17:06, Ferruh Yigit:
> On 3/5/2020 3:11 PM, Thomas Monjalon wrote:
> > 05/03/2020 15:55, Ferruh Yigit:
> >> FDIR -> Flow Director
> >
> > In general I prefer avoiding FDIR for two reasons:
> > 1/ this is an Intel-only acronym
>
> Yes, it is "Intel Ethernet Flow Director" but still it
On 3/5/2020 3:11 PM, Thomas Monjalon wrote:
> 05/03/2020 15:55, Ferruh Yigit:
>> FDIR -> Flow Director
>
> In general I prefer avoiding FDIR for two reasons:
> 1/ this is an Intel-only acronym
Yes, it is "Intel Ethernet Flow Director" but still it is a valid abbreviation
and we have it in our g
From: Igor Romanov
Call correct cleanup operations on failure in Rx queue start function.
Fixes: df1bfde4ff0d ("net/sfc: factor out libefx-based Rx datapath")
Fixes: 28944ac098aa ("net/sfc: implement Rx queue start and stop operations")
Cc: sta...@dpdk.org
Signed-off-by: Igor Romanov
Signed-of
On 3/5/20 6:39 PM, Ferruh Yigit wrote:
> On 3/5/2020 3:08 PM, Andrew Rybchenko wrote:
>> On 3/5/20 5:55 PM, Ferruh Yigit wrote:
>>> BAR-> Base Address Register
>>> FDIR -> Flow Director
>>> GENEVE -> Generic Network Virtualization Encapsulation
>>> IO -> Input/Output
>>> MPLS -> Multipr
On 05 Mar 07:19, Ananyev, Konstantin wrote:
>
> > On 05 Mar 11:27, Ananyev, Konstantin wrote:
> > >
> > >
> > > >
> > > > On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> > > > > When compiling with -O0,
> > > > > the compiler does not optimize two memory accesses into one.
> > > > > Lead
On 3/5/2020 3:08 PM, Andrew Rybchenko wrote:
> On 3/5/20 5:55 PM, Ferruh Yigit wrote:
>> BAR-> Base Address Register
>> FDIR -> Flow Director
>> GENEVE -> Generic Network Virtualization Encapsulation
>> IO -> Input/Output
>> MPLS -> Multiprotocol Label Switching
>> NEON
>> null
>> NVGRE
> On 05 Mar 11:27, Ananyev, Konstantin wrote:
> >
> >
> > >
> > > On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> > > > When compiling with -O0,
> > > > the compiler does not optimize two memory accesses into one.
> > > > Leads to accessing a null pointer when queue post Rx burst callbac
05/03/2020 15:55, Ferruh Yigit:
> FDIR -> Flow Director
In general I prefer avoiding FDIR for two reasons:
1/ this is an Intel-only acronym
2/ rte_flow API is superseding it
> OOB-> Out Of Bounds
I don't think it is a good idea to use this acronym. It is not a techno.
I prefer out-of-bound
On 3/5/20 5:55 PM, Ferruh Yigit wrote:
> BAR-> Base Address Register
> FDIR -> Flow Director
> GENEVE -> Generic Network Virtualization Encapsulation
> IO -> Input/Output
> MPLS -> Multiprotocol Label Switching
> NEON
> null
> NVGRE -> Network Virtualization using Generic Routing Encap
On 3/5/2020 10:38 AM, Andrew Rybchenko wrote:
> From: Igor Romanov
>
> NIC deinitialization should not happen after a failed NIC
> initialization.
>
> Fixes: 91831d4068c8 ("net/sfc: estimate available resources")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Igor Romanov
> Signed-off-by: Andrew Ryb
BAR-> Base Address Register
FDIR -> Flow Director
GENEVE -> Generic Network Virtualization Encapsulation
IO -> Input/Output
MPLS -> Multiprotocol Label Switching
NEON
null
NVGRE -> Network Virtualization using Generic Routing Encapsulation
OOB-> Out Of Bounds
RDMA -> Remote Direc
On 05 Mar 11:27, Ananyev, Konstantin wrote:
>
>
> >
> > On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> > > When compiling with -O0,
> > > the compiler does not optimize two memory accesses into one.
> > > Leads to accessing a null pointer when queue post Rx burst callback
> > > remova
Ruifeng Wang writes:
>> -Original Message-
>> From: Aaron Conole
>> Sent: Thursday, March 5, 2020 01:31
>> To: Ruifeng Wang
>> Cc: maicolgabr...@hotmail.com; bruce.richard...@intel.com; dev@dpdk.org;
>> david.march...@redhat.com; Gavin Hu ; Honnappa
>> Nagarahalli ; juraj.lin...@pantheo
> >
> > On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> > > When compiling with -O0,
> > > the compiler does not optimize two memory accesses into one.
> > > Leads to accessing a null pointer when queue post Rx burst callback
> > > removal while traffic is running.
> > > See rte_eth_tx_
On 3/4/2020 2:35 AM, Eduard Serra Miralles wrote:
> When calling to setup RSS on v4 API, ESX will expect
> IPv4/6 TCP RSS to be set/requested mandatorily.
>
> This patch will:
> - Set IPv4/6 TCP RSS when these have not been set. A warning
> message is thrown to make sure we warn the application we
On 3/5/2020 6:49 AM, Gagandeep Singh wrote:
> Hi Ferruh,
>>> diff --git a/drivers/bus/dpaa/rte_bus_dpaa_version.map
>> b/drivers/bus/dpaa/rte_bus_dpaa_version.map
>>> index e6ca4361e..86f5811b0 100644
>>> --- a/drivers/bus/dpaa/rte_bus_dpaa_version.map
>>> +++ b/drivers/bus/dpaa/rte_bus_dpaa_versio
Thanks. I'll try
On Thu, Mar 5, 2020 at 11:23 AM Bruce Richardson
wrote:
> On Wed, Mar 04, 2020 at 08:05:27PM +0200, Avner Taieb wrote:
> > Thanks Lee. I already read it and tried the potential remedy that is
> > suggested i.e. enabling the multi driver support , but it didn't solved
> the
> > p
On Thu, Mar 5, 2020 at 9:04 AM David Marchand wrote:
> On Thu, Mar 5, 2020 at 7:32 AM Hemant Agrawal wrote:
> >
> > Since the introduction of crypto event adapters, event drivers
> > have dependencies on crypto drivers.
> >
> > /usr/bin/ld: cannot find -lrte_pmd_dpaa_sec
> > collect2: error: ld r
Thanks will do that. And will also try to remove the trailer in the patches.
Thanks for pointing it out.
From: dev On Behalf Of Rasesh Mody
Sent: Tuesday, March 3, 2020 6:01 PM
To: Dey, Souvik ; Shahed Shaikh ; Jerin
Jacob Kollanukkaran ; ferruh.yi...@intel.com;
tho...@monjalon.net
Cc: dev@dpd
On 3/5/2020 10:26 AM, Kumar, Ravi1 wrote:
<...>
>> From: Selwin Sebastian
>>
>> Enable scattered rx support and add jumbo packet receive capability
>>
>> Signed-off-by: Selwin Sebastian
>
> Acked-by: Ravi Kumar
>
Applied to dpdk-next-net/master, thanks.
Minutes 5 March 2020
Agenda:
* Release Dates
* Subtrees
* OvS
* Opens
Participants:
* Arm
* Debian/Microsoft
* Intel
* Marvell
* Mellanox
* Red Hat
Release Dates
-
* v20.05 dates:
* Proposal/V1:Wednesday 18 March 2020
* Integration/Merge/RC1
>
> test app to measures the performance of libipsec
> api rte_ipsec_pkt_crypto and rte_ipsec_pkt_process.
>
> Signed-off-by: Savinay Dharmappa
> ---
I think the header need to be "app/test", or "test/ipsec".
Apart from that:
Acked-by: Konstantin Ananyev
> 2.17.1
>
> On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> > When compiling with -O0,
> > the compiler does not optimize two memory accesses into one.
> > Leads to accessing a null pointer when queue post Rx burst callback
> > removal while traffic is running.
> > See rte_eth_tx_burst functi
Avoid reusing recorded events when performing a migration, since this
may make the migration selection logic pick an already-moved flow.
Fixes: f6257b22e767 ("event/dsw: add load balancing")
Cc: sta...@dpdk.org
Reported-by: Venky Venkatesh
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/
Stopping lldp is necessary for DPDK, but it will cause DCB init failed.
For kernel shared code, the prerequisite for successful initialization
of DCB is that LLDP is enabled. This patch starts lldp before DCB init
and stops it when finish init.
Fixes: a8e84b22bd55 ("net/i40e/base: support persiste
From: Ivan Malov
This is needed to reuse pattern processing engine for MAE.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_flow.c | 101 ++---
drivers/net/sfc/sfc_flow.h | 52 +++
2 files changed, 111 insertio
From: Ivan Malov
The flow specification structure will be transformed to a generic one, and
its current contents will be fenced off to form a VNIC-specific parameters.
Flow RSS details do not belong to the said specification currently, albeit
being VNIC-specific. This patch addresses this issue a
From: Ivan Malov
Generalise flow attribute parsing function in regard to transfer attribute.
Add a method table and factor out VNIC-specific parsing code as a callback.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_flow.c | 101 +++--
From: Ivan Malov
Add the concept of a flow specification type.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_flow.c | 137 +++--
drivers/net/sfc/sfc_flow.h | 25 ++-
2 files changed, 106 insertions(+), 56 deletions(-)
d
From: Ivan Malov
As a preparation step, generalise flow start and stop path using callbacks.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_flow.c | 113 -
drivers/net/sfc/sfc_flow.h | 6 ++
2 files changed, 81 insertion
From: Ivan Malov
Riverhead boards maintain support for MAE, a low-level Match-Action Engine.
The upcoming patches will bring support for MAE to RTE flow implementation.
A follow-up patch will introduce MAE-specific specification cleanup method.
In order to prepare for the patch, introduce flow a
From: Ivan Malov
A backend which a driver employs to handle flow rules of a given
type depends on the underlying NIC flow engine. The driver in
question in its current state is tailored to support the only
flow engine, VNIC filtering. As the need arises to add support
for transfer rules, the driv
Prepare rte_flow API support code to addition of one more
backend for flow rules handling.
Ivan Malov (7):
net/sfc: make flow RSS details VNIC-specific
net/sfc: make the flow list engine-agnostic
net/sfc: generalise the flow specification structure
net/sfc: introduce flow allocation and fr
From: Igor Romanov
NIC deinitialization should not happen after a failed NIC
initialization.
Fixes: 91831d4068c8 ("net/sfc: estimate available resources")
Cc: sta...@dpdk.org
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc.c | 2 +-
1 file changed, 1 inser
On 2020-02-28 06:57, Venky Venkatesh wrote:
> Hi,
> This is concerning the DSW PMD for eventdev. In our application we put
> some debugs to see that the same ATOMIC flow isn't scheduled to 2
> different cores. Strangely enough we hit it. Not sure if we are
> missing something OR it is a bug.
>
>
Please scratch this. I just now acked the V3.
Regards,
Ravi
>
>
>-Original Message-
>From: Kumar, Ravi1
>Sent: Thursday, March 5, 2020 3:51 PM
>To: Sebastian, Selwin ; 'dev@dpdk.org'
>Subject: RE: [PATCH v1] net/axgbe: add support for Scattered Rx
>
>Acked-by: Ravi Kumar
>
>>From: Selw
[AMD Official Use Only - Internal Distribution Only]
Acked-by: Ravi Kumar
>From: Selwin Sebastian
>
>Enable scattered rx support and add jumbo packet receive capability
>
>Signed-off-by: Selwin Sebastian
>---
> doc/guides/nics/features/axgbe.ini | 1 +
> drivers/net/axgbe/axgbe_common.h |
Acked-by: Ravi Kumar
>From: Selwin Sebastian
>
>Enable scattered rx support and add jumbo packet receive capability
>
>Signed-off-by: Selwin Sebastian
>---
> doc/guides/nics/features/axgbe.ini | 1 +
> drivers/net/axgbe/axgbe_common.h | 2 +
> drivers/net/axgbe/axgbe_ethdev.c | 16 +++-
>
Next meeting Mar 18th 1700 UTC
(Planning to continue discussion on DPDK rte_flow HwOl)
March 4th Minutes
Attendees: Edwin, Keith, Cristian, Aaron, Eli, Ilya, Majd, Ajit, Fouad,
Jingjing, Malvika, Ameer, Harsha, Johann, Oz, David, Ian, Roni, Simon,
Somnath, Thomas, William, Marcelo, Hemal, John Ma
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Suanming Mou
> On 12/26/2019 7:05 PM, Olivier Matz wrote:
> > On Thu, Oct 17, 2019 at 06:55:01AM +, Xueming Li wrote:
> >> Indexed memory pool manages memory entries by index, allocation from
> >> pool returns both memory pointer and index(
Can we add the `volatile` qualifier to the `hwrm_cmd_resp_addr` member (in
drivers/net/bnxt/bnxt.h) to get a stronger guarantee that the compiler won't
insert TOCTOU races in the output?
Christopher Ertl | MSRC Vulnerabilities & Mitigations | Microsoft Limited | +44
7773976589 | christopher.e..
On Wed, Mar 04, 2020 at 08:05:27PM +0200, Avner Taieb wrote:
> Thanks Lee. I already read it and tried the potential remedy that is
> suggested i.e. enabling the multi driver support , but it didn't solved the
> problem.
> Are there more setting documented anywhere ?
>
Did you try using igb_uio d
On Thu, Mar 05, 2020 at 01:33:49AM +0800, ZY Qiu wrote:
> When compiling with -O0,
> the compiler does not optimize two memory accesses into one.
> Leads to accessing a null pointer when queue post Rx burst callback
> removal while traffic is running.
> See rte_eth_tx_burst function.
>
> Signed-of
Hi David,
> On Thu, Mar 5, 2020 at 10:06 AM Hemant Agrawal (OSS)
> wrote:
> >
> > Hi David,
> > > On Mon, Mar 2, 2020 at 10:26 AM Hemant Agrawal
> > > wrote:
> > > >
> > > > This patch series add various patches for enhancing and fixing NXP
> > > > fslmc bus, dpaa bus, and dpaax.
> > > >
> > > >
On Thu, Mar 5, 2020 at 10:06 AM Hemant Agrawal (OSS)
wrote:
>
> Hi David,
> > On Mon, Mar 2, 2020 at 10:26 AM Hemant Agrawal
> > wrote:
> > >
> > > This patch series add various patches for enhancing and fixing NXP
> > > fslmc bus, dpaa bus, and dpaax.
> > >
> > > - the main change is support to
Hi David,
> On Mon, Mar 2, 2020 at 10:26 AM Hemant Agrawal
> wrote:
> >
> > This patch series add various patches for enhancing and fixing NXP
> > fslmc bus, dpaa bus, and dpaax.
> >
> > - the main change is support to allow thread migration across lcores
> > - improving the multi-process support
> -Original Message-
> From: dev On Behalf Of Shougang Wang
> Sent: Thursday, March 5, 2020 4:41 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Xing, Beilei
> ; Wang, ShougangX ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/i40e: fix DCB init issue
>
> Stopping lldp is necessary for
On 3/5/2020 5:25 AM, sseba...@amd.com wrote:
> From: Selwin Sebastian
>
> Enable scattered rx support and add jumbo packet receive capability
>
> Signed-off-by: Selwin Sebastian
Hi Selwin,
Looks good to me, I will wait Ravi's ack/review for merge.
Thanks,
ferruh
Minutes of Technical Board meeting, 2020-02-26
Members Attending: 9/11
- Ferruh Yigit
- Hemant Agrawal
- Honnappa Nagarahalli
- Jerin Jacob
- Kevin Traynor
- Konstantin Ananyev
- Olivier Matz
- Stephen Hemminger
- Thomas Monjalon (Chair)
NOTE: Technical Board meetings happen eve
Stopping lldp is necessary for DPDK, but it will cause DCB init failed.
For kernel shared code, the prerequisite for successful initialization
of DCB is that LLDP is enabled. This patch starts lldp before DCB init
and stops it when finish init.
Fixes: a8e84b22bd55 ("net/i40e/base: support persiste
From: Tonghao Zhang
The order of mempool initiation affects mempool index in the
rte_mempool_ops_table. For example, when building APPs with:
$ gcc -lrte_mempool_bucket -lrte_mempool_ring ...
The "bucket" mempool will be registered firstly, and its index
in table is 0 while the index of "ring"
On Thu, Mar 5, 2020 at 7:32 AM Hemant Agrawal wrote:
>
> Since the introduction of crypto event adapters, event drivers
> have dependencies on crypto drivers.
>
> /usr/bin/ld: cannot find -lrte_pmd_dpaa_sec
> collect2: error: ld returned 1 exit status
> make[9]: *** [.../mk/rte.lib.mk:100:
>
98 matches
Mail list logo
