switch means hardware switch attach this PC, not ovs-dpdk
2017-09-27 11:57 GMT+08:00 Sam :
> Hi all,
>
> I'm running ovs-dpdk using ixgbe driver with sriov mode. My topology is:
> vf(ens16) and pf(eth0) is from netdevice 1; vf(ens161) and pf(eth1) is from
> netdevice 2. Then linux kernel driver b
Hi all,
I'm running ovs-dpdk using ixgbe driver with sriov mode. My topology is:
vf(ens16) and pf(eth0) is from netdevice 1; vf(ens161) and pf(eth1) is from
netdevice 2. Then linux kernel driver bond ens16 and ens161 into bond1,
ovs-dpdk bond eth0 and eth1 into dpdkb2 as ovs-dpdk port.
I found th
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Thursday, September 14, 2017 10:55 PM
> To: Thomas Monjalon
> Cc: dev@dpdk.org; Yigit, Ferruh ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] ethdev: fix LIBABIVER
>
> Fixes: c33ade1227a5 ("do
The assignment at initialization is overwritten immediately. Drop the
assignment.
Signed-off-by: Aaron Conole
---
drivers/net/enic/base/vnic_dev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/enic/base/vnic_dev.c b/drivers/net/enic/base/vnic_dev.c
index a4b7f4
The functions here aren't called anywhere in code, at least according to
both the compiler, and some greps.
Signed-off-by: Aaron Conole
---
drivers/net/enic/base/vnic_cq.c | 10
drivers/net/enic/base/vnic_dev.c | 102 ---
drivers/net/enic/base/vnic_rq.c
As it stands, the existing assingment to mbuf has no effect outside of
the function. Prior to this change, the mbuf argument would contain
an invalid address, but it would not be null. After this change, the
caller gets a null mbuf back.
Signed-off-by: Aaron Conole
---
drivers/net/enic/enic_ma
These still need some testing, and I will cc to stable
with the official patch.
I'm especially interested in the comments on base.
Aaron Conole (4):
enic: update format string to match arg types
enic: fix assignment
enic: remove unused code
enic: remove anscillary assignment
drivers/net
The argument `index` (and unique_id) is unsigned, but the format
string type used was for signed types.
Signed-off-by: Aaron Conole
---
drivers/net/enic/base/vnic_cq.c | 4 ++--
drivers/net/enic/base/vnic_dev.c | 6 +++---
drivers/net/enic/base/vnic_rq.c | 4 ++--
drivers/net/enic/base/vnic_wq
Hello Team
I'm installing OVS with DPDK-v16.11 on ubuntu 16.04.3 and getting
following error on running below command..
cd /opt/stack/ovs/
make -j 8 'CFLAGS=-O3 -march=native -fPIC'
/bin/bash ./libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I.
-I ./include -I ./include -I ./lib -I ./l
From: Shachar Beiser
This removes the dependency on specific Mellanox OFED libraries by
using the upstream rdma-core and linux upstream community code.
Both rdma-core upstream and Mellanox OFED are Linux user-space packages:
1. Rdma-core is Linux upstream user-space package.(Generic)
2. Mella
Hi Cristian,
No problem, it doesn't sounds like it will be too difficult.
Unfortunately I'm in the middle of another piece of work, so it will probably
be a couple of weeks before I get around to it.
Regards
Alan
-Original Message-
From: Dumitrescu, Cristian [mailto:cristian.dumitre.
On Tue, Sep 26, 2017 at 11:39:58AM +0200, Tomasz Duszynski wrote:
> Since MRVL NET PMD needs librte_cfgfile to parse QoS configuration file
> link it as the whole library.
>
> Signed-off-by: Jacek Siuda
> Signed-off-by: Tomasz Duszynski
> ---
Can you clarify a bit more why this is needed? For a
> -Original Message-
> From: Hunt, David
> Sent: Monday, September 25, 2017 8:27 PM
> To: dev@dpdk.org
> Cc: Ananyev, Konstantin ; Wu, Jingjing
> ; Sexton, Rory ; Marjanovic,
> Nemanja
> ; Hunt, David
> Subject: [PATCH v2 1/8] net/i40e: add API to convert VF MAC to VSI index
>
VSI inde
> -Original Message-
> From: Yong Wang [mailto:wang.yon...@zte.com.cn]
> Sent: Friday, September 22, 2017 3:18 PM
> To: Wu, Jingjing
> Cc: dev@dpdk.org; Yong Wang
> Subject: [PATCH] net/i40e: fix uninitialized variable
>
> In func i40evf_dev_link_update(), "new_link.link_autoneg" is us
> -Original Message-
> From: Dai, Wei
> Sent: Monday, September 25, 2017 2:37 PM
> To: Wu, Jingjing ; Xing, Beilei
> ; Ananyev,
> Konstantin
> Cc: dev@dpdk.org; Dai, Wei ; sta...@dpdk.org
> Subject: [PATCH v6] net/i40e: fix mirror rule reset when port is closed
>
> When an i40e PF port
Hello Keith and thank you for your answer,
The goal is indeed to generate as much traffic per machine as possible
(we use pktgen-dpdk to benchmark datacenter routers before putting them
on production).
For this we use all available CPU power to send packets.
Following your suggestion, I modi
Hello and thank you for your answer,
We use the following versions:
MLNX DPDK version 16.11, revision 3.0
pktgen-dpdk version 3.3.8
mlnx_ofed version 4.1-1.0.2.0
fw revision: 12.18.1000
Regards,
On 2017-09-25 08:46 PM, Yongseok Koh wrote:
Hi, Damien
Can you please let me know the version
On Tuesday 19 September 2017 07:27 PM, Shreyansh Jain wrote:
On Tuesday 19 September 2017 07:10 PM, Ferruh Yigit wrote:
On 9/19/2017 2:18 PM, Shreyansh Jain wrote:
On Monday 18 September 2017 08:19 PM, Ferruh Yigit wrote:
On 9/9/2017 12:20 PM, Shreyansh Jain wrote:
From: Hemant Agrawal
Link
On Monday 25 September 2017 08:41 PM, Ferruh Yigit wrote:
On 9/25/2017 3:32 PM, Shreyansh Jain wrote:
On Tuesday 19 September 2017 06:44 PM, Shreyansh Jain wrote:
Hello Ferruh,
On Monday 18 September 2017 08:17 PM, Ferruh Yigit wrote:
On 9/9/2017 12:20 PM, Shreyansh Jain wrote:
Signed-off-by
Hi Martin,
We will look into this issue and will try to reproduce.
Will update you as soon as we have any news.
Can you please send kernel crush stack that you are seeing.
Best Regards,
Olga
Olga Shern
SW Director DPDK
Mellanox
Hi Olivier,
Thanks for the feedback. I'll submit the fixes for our V2 of this series.
Some comments inline.
Olivier Matz wrote:
>
> Hi Boris,
>
> Some comments inline.
>
> On Mon, Sep 18, 2017 at 07:54:03AM +, Boris Pismenny wrote:
> > Hi Olivier,
> >
> > On 9/14/2017 11:27 AM, Akhil Goyal
Hi all,
I'm running dpdk application, and this application is always hanged after
running for a long time, so I want to see if there is memory leak or
something.
I think the memory allocate by dpdk api such as "rte_rxmbuf_alloc" is
located in memory of a core, not system memory, is it right?
If
Add unit tests for MRVL CRYPTO PMD driver.
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
test/test/test_cryptodev.c | 168
test/test/test_cryptodev.h | 1 +
test/test/test_cryptodev_aes_test_vectors.h | 72 ++
Add documentation for the MRVL CRYPTO PMD driver.
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
doc/guides/cryptodevs/features/mrvl.ini | 42 +++
doc/guides/cryptodevs/index.rst | 1 +
doc/guides/cryptodevs/mrvl.rst | 198
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
MAINTAINERS | 10 ++
1 file changed, 10 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index d4810cf..93800d2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -561,6 +561,16 @@ F: drivers/crypto/dpaa2_sec/
F: doc/guides/
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
MAINTAINERS | 10 ++
1 file changed, 10 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index a0cd75e..d4810cf 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -393,6 +393,16 @@ F: drivers/net/mlx5/
F: doc/guides/nics/mlx
Add support for the Marvell Security Crypto Accelerator EIP197.
Driver is based on external, publicly available, Marvell MUSDK
library that provides access to the hardware with minimum overhead
and high performance.
Driver comes with support for the following features:
* Symmetric crypto
* Sym op
Add documentation for the MRVL NET PMD driver.
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
doc/guides/nics/features/mrvl.ini | 24 ++
doc/guides/nics/index.rst | 1 +
doc/guides/nics/mrvl.rst | 151 ++
3 files change
Add support for the Marvell PPv2 (Packet Processor v2) 1/10 Gbps adapter.
Driver is based on external, publicly available, light-weight Marvell
MUSDK library that provides access to network packet processor.
Driver comes with support for the following features:
* Speed capabilities
* Link status
Hello,
This patch series introduces net/crypto drivers for Marvell Armada 7k/8k
SoCs along with documentation and crypto driver pmd tests.
Below you can find the list of features which net/crypto pmds support.
Net pmd features:
* Speed capabilities
* Link status
* Queue start/stop
* MTU update
*
Since MRVL NET PMD needs librte_cfgfile to parse QoS configuration file
link it as the whole library.
Signed-off-by: Jacek Siuda
Signed-off-by: Tomasz Duszynski
---
mk/rte.app.mk | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index c25fdd9..945
Hi,
we are currently testing the Mellanox ConnectX-5 100G NIC with DPDK
17.08 as well as dpdk-net-next and are
experiencing mbuf leaks as well as crashes (and in some instances even
kernel panics in a mlx5 module) under
certain load conditions.
We initially saw these issues only in our own DPDK-b
On 9/22/2017 1:25 PM, Pablo de Lara wrote:
In order to improve memory utilization, a single mempool
is created, containing the crypto operation and mbufs
(one if operation is in-place, two if out-of-place).
This way, a single object is allocated and freed
per operation, reducing the amount of mem
Hi, Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, September 20, 2017 6:36 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 1/2] net/i40e: queue region set and flush
>
> On 9/15/2017 4:13 AM, Wei Zhao wrote:
> > This feature enable queu
Hi Pablo,
On 9/22/2017 1:25 PM, Pablo de Lara wrote:
Add parameter "qps" in crypto performance app,
to create multiple queue pairs per device.
This new parameter is useful to have multiple logical
cores using a single crypto device, without needing
to initialize a crypto device per core.
Signed
Watch live now
https://www.pscp.tv/w/1LyxBEoZAdzJN
Hi, Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Monday, September 25, 2017 5:32 PM
> To: Zhao1, Wei
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 1/2] net/i40e: queue region set and flush
>
> On 9/25/2017 8:40 AM, Zhao1, Wei wrote:
> > Hi, Ferruh
> >
> >> -Or
This patch dynamically selects x86 EFD functions at run-time.
This patch uses function pointer and binds it to the relative
function based on CPU flags at constructor time.
Signed-off-by: Xiaoyun Li
---
lib/librte_efd/rte_efd_x86.h | 41 ++---
1 file changed,
This patchset dynamically selects functions at run-time based on CPU flags
that current machine supports. This patchset modifies mempcy, memcpy perf
test and x86 EFD, using function pointers and bind them at constructor time.
Then in the cloud environment, users can compiler once for the minimum ta
This patch modifies assignment of alignment unit from build-time
to run-time based on CPU flags that machine supports.
Signed-off-by: Xiaoyun Li
---
test/test/test_memcpy_perf.c | 40 +++-
1 file changed, 27 insertions(+), 13 deletions(-)
diff --git a/test/te
This patch dynamically selects functions of memcpy at run-time based
on CPU flags that current machine supports. This patch uses function
pointers which are bind to the relative functions at constrctor time.
In addition, AVX512 instructions set would be compiled only if users
config it enabled and
The DPDK needs to stay up to date with current LTS Linux kernel support.
If the kernel is older than LTS it is likely to be insecure and buggy.
Therefore only require DPDK to work on oldest LTS kernel.
If distribution vendors want to support DPDK on older kernels, that is
their choice. But the ups
Hi Ferruh, Thomas,
> > @@ -40,7 +40,7 @@ LIB = librte_pmd_af_packet.a
> >
> > EXPORT_MAP := rte_pmd_af_packet_version.map
> >
> > -LIBABIVER := 1
> > +LIBABIVER := 2
>
> I have just recognized this one. This shouldn't be updated. Only LIABIVER of
> the
> libraries that their API modified should
43 matches
Mail list logo
