Hi Konstantin,
I just got results on running the ipsec-secgw on NXP hardware.
it seems there is a drop of around 10% for lookaside proto with normal
command(i.e. without -l option)
with -l option, I got a seg fault while running traffic. gdb suggest
that pkt_func is not filled up and is NULL.
On Thu, Dec 27, 2018 at 02:52:17PM +0900, Hideyuki Yamashita wrote:
> Q1.
> I CCed my patch to sta...@dpdk.org
> Is that mean it will be reflected into 18.11 stable release?
Yes.
> The reason why I ask this is I think it is better that bug should
> be reflected into LTS because I and SPP users m
This patch adds glue functions for operations:
- dv_open_device.
- devx object create, destroy, query and modify.
- devx general command
The new operations depend on HAVE_IBV_DEVX_OBJ.
Signed-off-by: Moti Haimovsky
---
v2:
* Modifications according to review inputs.
see message Id: 154574
This series of commits add support for creating, allocating, querying
and destroying flow counters in mlx5 PMD using the devx interface.
Moti Haimovsky (3):
net/mlx5: fix shared counter allocation logic
net/mlx5: add devx functions to glue
net/mlx5: support flow counters using devx
drivers
This commit fixes the logic for searching and allocating a shared
counter in mlx5_flow_verbs.
Now only the shared counters in the counters list are checked for
a match and not all the counters as before.
Fixes: 84c406e74524 ("net/mlx5: add flow translate function")
Cc: sta...@dpdk.org
Signed-off-
This commit adds counters support when creating flows via direct
verbs. The implementation uses devx interface in order to create
query and delete the counters.
This support requires MLNX_OFED_LINUX-4.5-0.1.0.1 installation.
Signed-off-by: Moti Haimovsky
---
v2:
* Modifications according to code
On 12/28/2018 9:03 PM, Konstantin Ananyev wrote:
> In some cases crypto-ops could never be dequeued from the crypto-device.
> The easiest way to reproduce:
> start ipsec-secgw with crypto-dev and send to it less then 32 packets.
> none packets will be forwarded.
> Reason for that is that the appl
Hi,
From: David Marchand [mailto:david.march...@redhat.com]
Sent: Friday, December 21, 2018 2:58 PM
To: Burakov, Anatoly ; Vemula, Hari KumarX
Cc: dev@dpdk.org; Pattan, Reshma ; Yigit, Ferruh
; sta...@dpdk.org
Subject: Re: [dpdk-dev] [PATCH] eal: fix core number validation
On Fri, Dec 21, 20
> Suggested-by: Gavin Hu
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> Reviewed-by: Honnappa Nagarahalli
> Reviewed-by: Dharmik Thakkar
> Reviewed-by: Ola Liljedahl
> Reviewed-by: Gavin Hu
Small comment about the email addresses,
Please do not use uppercase so we can match/grep
02/01/2019 01:55, Gavin Hu:
> V4: Include the ring perf test case enhancement patch in the series.
>
> V3: Allow experimental API for meson build
>
> V2: Fix the coding style issue(commit message line too long)
>
> V1: To flush a ring not in use, dequeue one by one is wasting cpu cycles.
> The p
Hi Akhil,
> Hi Konstantin,
>
> I just got results on running the ipsec-secgw on NXP hardware.
Thanks for doing that.
We don't have NXP HW, so would need more help from you.
>
> with -l option, I got a seg fault while running traffic. gdb suggest
> that pkt_func is not filled up and is NULL.
On 12/28/2018 9:03 PM, Konstantin Ananyev wrote:
> Right now ipsec-secgw always enables TX offloads
> (DEV_TX_OFFLOAD_MULTI_SEGS, DEV_TX_OFFLOAD_SECURITY),
> even when they are not requested by the config.
> That causes many PMD to choose full-featured TX function,
> which in many cases is much s
>
> On 12/28/2018 9:03 PM, Konstantin Ananyev wrote:
> > In some cases crypto-ops could never be dequeued from the crypto-device.
> > The easiest way to reproduce:
> > start ipsec-secgw with crypto-dev and send to it less then 32 packets.
> > none packets will be forwarded.
> > Reason for that i
On 1/2/2019 7:13 PM, Ananyev, Konstantin wrote:
>
>> On 12/28/2018 9:03 PM, Konstantin Ananyev wrote:
>>> In some cases crypto-ops could never be dequeued from the crypto-device.
>>> The easiest way to reproduce:
>>> start ipsec-secgw with crypto-dev and send to it less then 32 packets.
>>> none
Added assert check for rte_bbdev_*_op_alloc_bulk in bbdev test app
Coverity issue: 328516
Coverity issue: 328525
Fixes: f714a18885a6 ("app/testbbdev: add test application for bbdev")
Signed-off-by: Amr Mokhtar
---
app/test-bbdev/test_bbdev_perf.c | 10 --
1 file changed, 8 insertions(+)
On Thu, Dec 20, 2018 at 11:01 AM Hari Kumar Vemula <
hari.kumarx.vem...@intel.com> wrote:
> When incorrect core value or range provided,
> as part of -l command line option, a crash occurs.
>
> Added valid range checks to fix the crash.
>
> Fixes: d888cb8b9613 ("eal: add core list input format")
>
On 1/2/2019 6:31 PM, Ananyev, Konstantin wrote:
> Hi Akhil,
>
>> Hi Konstantin,
>>
>> I just got results on running the ipsec-secgw on NXP hardware.
> Thanks for doing that.
> We don't have NXP HW, so would need more help from you.
>
>> with -l option, I got a seg fault while running traffic. gdb
Hello Jeff,
On Sat, Dec 29, 2018 at 5:06 AM Jeff Guo wrote:
> On 12/17/2018 6:15 PM, David Marchand wrote:
>
>
> On Fri, Dec 14, 2018 at 8:41 AM Jeff Guo wrote:
>
>> This command-line option will enable hotplug event detecting and enable
>> hotplug handling for device hotplug.
>>
>> Signed-off-
On 12/25/2018 6:30 AM, Dekel Peled wrote:
> PSB.
>
>> -Original Message-
>> From: Ori Kam
>> Sent: Tuesday, December 25, 2018 5:25 AM
>> To: Dekel Peled
>> Cc: dev@dpdk.org; Dekel Peled
>> Subject: RE: [dpdk-dev] [PATCH] examples/flow_filtering: fix example
>> documentation
>>
>>
>>
>>>
On 12/23/2018 12:43 PM, zhuangyan wrote:
> Hello,
>
> Is there any way to disable the generation of “vmovdqa64” instruction during
> the compilation of dpdk library on Intel Skylake CPU?
>
> My dev VM uses Intel Skylake CPU while my test machine uses Intel uses
> Intel(R) Xeon(R) CPU E5-2630 v4
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Wednesday, January 2, 2019 1:51 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 03/10] examples/ipsec-secgw: fix crypto-op
> might never get dequeued
>
>
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Wednesday, January 2, 2019 2:29 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 00/10] examples/ipsec-secgw: make app to
> use ipsec library
>
>
>
> On 1/2/2019 6:31 PM, Anany
Previous commit adding multiprocess support broke VF support.
When VFs, the PMD does not set the link up or down.
Fixes: ef28aa96e53b ("net/nfp: support multiprocess")
Signed-off-by: Alejandro Lucero
---
drivers/net/nfp/nfp_net.c | 28
1 file changed, 16 insertions(
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Wednesday, January 2, 2019 1:43 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Horton, Remy
> Subject: Re: [dpdk-dev] [PATCH v5 01/10] examples/ipsec-secgw: allow user to
> disable some RX/TX offloads
>
>
Hi,
I am using DPDK 17.11.4 version . Do anybody have idea that DPDK is using
benefit of Transparent huge-pages in case of Madvise.
Thanks,
Chetan Bhasin
If the port has received less than ``pkt_per_port`` packets (for
example, the port has missed some packets), the test is in an infinite
loop.
Instead of expecting a number of packet to receive, let the port to be
drained by itself. If no more packets are received, the test can
continue.
Fixes: 00
Fixes: 6f22f2f67268 ("net/fm10k: redefine link status semantics")
Cc: sta...@dpdk.org
Signed-off-by: Julien Meunier
---
drivers/net/fm10k/fm10k_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 85fb6c5..caf4d1b
Loopback mode is also supported on X540 and X550 NICs, according to
their datasheet (section 15.2). The way to set it up is a little
different of the 82599.
Signed-off-by: Julien Meunier
---
drivers/net/ixgbe/ixgbe_ethdev.c | 10 ++---
drivers/net/ixgbe/ixgbe_ethdev.h | 5 ++---
drivers/net
On 12/23/2018 6:06 AM, Shahaf Shuler wrote:
> Ferruh,
>
> I share the same thoughts as Tom here.
>
>
>
>>Ferruh Yigit wrote :
>
>>> Is this a common enough feature to include into ethdev abstraction layer?
>>> Or a
>
>>> feature for a single vendor?
>
>>
>
>>I found reference to mbuf’s t
On 12/22/2018 12:41 PM, Haiyang Tan wrote:
> The uninitialized field 'extra_flag' of hash parameter may enable
> certain feature silently. Typically, if bit0 of 'extra_flag' set, the
> hardware transactional memory support will be enabled unexpectedly.
>
> Signed-off-by: Haiyang Tan
> ---
> driv
Thanks Ruifeng for the patch. I have one question inline.
Jerin/Hemant,
It would be good if you could test this on your platforms, since this
is being made default.
Thanks,
Honnappa
> -Original Message-
> From: Ruifeng Wang
> Sent: Tuesday, January 1, 2019 11:28 PM
> To: dev@dp
>
> 02/01/2019 01:55, Gavin Hu:
> > V4: Include the ring perf test case enhancement patch in the series.
This change is not related to this patch. Should be a separate patch?
There were comments provided:
http://mails.dpdk.org/archives/dev/2018-December/121893.html
http://mails.dpdk.org/archives/d
Hi,
We need to gather inputs about the pros/cons of the C compilers
available for Windows.
Interesting criterias could be:
- ease of use
- availability
- standards compliance
- performance
When the comparison will be complete, we should publish it
in the doc/ direc
Hi Honnappa,
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Thursday, January 3, 2019 2:23
> To: Ruifeng Wang (Arm Technology China) ;
> dev@dpdk.org
> Cc: tho...@monjalon.net; jer...@marvell.com; hemant.agra...@nxp.com;
> bruce.richard...@intel.com; chao...@linux.vnet.ibm.com;
Hi Julien,
> -Original Message-
> From: Julien Meunier [mailto:julien.meun...@nokia.com]
> Sent: Wednesday, January 2, 2019 11:58 PM
> To: Zhang, Qi Z ; Wang, Xiao W
>
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH] net/fm10k: initialize sm_down variable
>
> Fixes: 6f22f2f67268 ("
From: Gavin Hu
Currently, the flush is done by dequeuing the ring in a while loop. It is
much simpler to flush the queue by resetting the head and tail indices.
Signed-off-by: gavin hu
Reviewed-by: ruifeng wang
Reviewed-by: honnappa nagarahalli
---
lib/librte_ring/rte_ring.h | 20 +
From: Gavin Hu
Within rte_hash_reset, calling a while loop to dequeue one by
one from the ring, while not using them at all, is wasting cycles,
The patch just flush the ring by resetting the indices can save cpu
cycles.
Fixes: b26473ff8f4a ("hash: add reset function")
Fixes: 75706568a7eb ("hash:
V5:
1. Commit message tweaking for ring test case enhancement patch
2. Upper to lower for mails to make match/grep more easily
V4:
1. Include the ring perf test case enhancement patch in the series.
2. Replace ARRAY_SIZE with RTE_DIM.
3. Call memset to avoid clang compling complains.
V3: Allow e
From: Joyce Kong
Run ring perf test on all available cores to really verify MPMC operations.
The old way of running on a pair of cores is not enough for MPMC rings.
Suggested-by: gavin hu
Signed-off-by: joyce kong
Reviewed-by: ruifeng wang
Reviewed-by: honnappa nagarahalli
Reviewed-by: dharm
> -Original Message-
> From: Thomas Monjalon
> Sent: Wednesday, January 2, 2019 8:49 PM
> To: Gavin Hu (Arm Technology China)
> Cc: dev@dpdk.org; jer...@marvell.com; hemant.agra...@nxp.com;
> bruce.richard...@intel.com; chao...@linux.vnet.ibm.com; Honnappa
> Nagarahalli ; nd ;
> olivie
Currently virtio-user doesn't support event idx.
Fixes: aea29aa5d37b ("net/virtio: enable packed virtqueues by default")
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_user/virtio_user_dev.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/virtio/virtio_
Add the RING_PACKED feature to dev->unsupported_features
when it's disabled, and add the missing packed vq param
string. And also revert the unexpected change to MAC option
introduced when adding packed vq option.
Fixes: 34f3966c7f81 ("net/virtio-user: add option to use packed queues")
Signed-off
Tiwei Bie (2):
net/virtio-user: fix packed vq option parsing
net/virtio-user: fix supported features list
drivers/net/virtio/virtio_user/virtio_user_dev.c | 14 +-
drivers/net/virtio/virtio_user_ethdev.c | 7 ---
2 files changed, 9 insertions(+), 12 deletions(-)
--
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Thursday, January 3, 2019 2:41 AM
> To: Thomas Monjalon ; Gavin Hu (Arm Technology
> China)
> Cc: dev@dpdk.org; jer...@marvell.com; hemant.agra...@nxp.com;
> bruce.richard...@intel.com; chao...@linux.vnet.ibm.com; nd
> ; olivier
On Wed, 2019-01-02 at 13:28 +0800, Ruifeng Wang wrote:
> ---
> ---
> Compile option for hash_multi_lookup was broken, and caused feature
> cannot be enabled on Arm.
> This patch sets hash_multi_lookup method as default, and sequential
What about Gcc under the WSL thing (ie Linux emulation in Windows).
Much better than Cygwin type stuff.
-Original Message-
From: Thomas Monjalon
Sent: Wednesday, January 2, 2019 2:45 PM
To: Jason Messer ; Harini Ramakrishnan
; Omar Cardona ;
Ranjit Menon
Cc: Mattias Rönnblom ; Jeff Sha
Hi Ferruh, PSB
> -Original Message-
> From: Ferruh Yigit
> Sent: Wednesday, January 2, 2019 4:53 PM
> To: Dekel Peled ; Ori Kam
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] examples/flow_filtering: fix example
> documentation
>
> On 12/25/2018 6:30 AM, Dekel Peled wrote:
> > PSB
03/01/2019 03:38, gavin hu:
> From: Joyce Kong
>
> Run ring perf test on all available cores to really verify MPMC operations.
> The old way of running on a pair of cores is not enough for MPMC rings.
>
> Suggested-by: gavin hu
> Signed-off-by: joyce kong
> Reviewed-by: ruifeng wang
> Reviewe
48 matches
Mail list logo
