There are new Ethernet device events - NEW and DESTROY, and new option
to register all ports by one call.
Adjust application to aforementioned changes.
Signed-off-by: Matan Azrad
---
app/test-pmd/parameters.c | 4
app/test-pmd/testpmd.c| 30
Add option to register event callback for all ports by one call to
rte_eth_dev_callback_register using port_id=RTE_ETH_ALL.
In this case the callback is also registered to invalid ports.
Signed-off-by: Matan Azrad
---
lib/librte_ether/rte_ethdev.c | 130 +
The ethdev name is taken from the "name" parameter from the helper
function rte_eth_dev_allocate(name). The name is given by the caller,
thus, there is no way to guarantee all the callers will follow the
same syntax, leaving us the port name is not standardized.
For example, for all ports probed b
Signed-off-by: Pavan Nikhilesh
---
.../tools/img/eventdev_perf_pipeline_test.svg | 3292
doc/guides/tools/testeventdev.rst | 99 +
2 files changed, 3391 insertions(+)
create mode 100644 doc/guides/tools/img/eventdev_perf_pipeline_test.svg
diff --git
Add logs for packet distribution across worker cores to be printed
along with the test results.
Signed-off-by: Pavan Nikhilesh
---
app/test-eventdev/evt_main.c | 3 +++
app/test-eventdev/test_perf_common.c | 12
2 files changed, 15 insertions(+)
diff --git a/app/test-event
This is a performance test case that aims at testing the following:
1. Measure the end-to-end performance of an event dev with a ethernet dev.
2. Maintain packet ordering from Rx to Tx.
The perf pipeline test configures the eventdev with Q queues and P ports,
where Q is nb_ethdev * nb_stages and P
Modify ethernet dev setup to support packet transmission and probe
through ethdev tx capabilities.
Signed-off-by: Pavan Nikhilesh
---
This patch set depends on : http://dpdk.org/dev/patchwork/patch/30520
app/test-eventdev/test_perf_common.c | 56
app/test-e
> -Original Message-
> From: Zhang, Qi Z
> Sent: Wednesday, November 29, 2017 6:28 PM
> To: Xing, Beilei
> Cc: dev@dpdk.org; Zhang, Qi Z ; sta...@dpdk.org
> Subject: [PATCH] net/i40e: fix VLAN offload setting
>
> Vlan offload flag "rx_mode.hw_vlan_extend" should be considerred at
> rte_e
On 11/29/2017 4:01 AM, Ferruh Yigit wrote:
On 5/3/2017 4:21 AM, Hemant Agrawal wrote:
This patch adds following:
1. option to configure the mac address during create
2. inform usespace, if mac address is being changed in linux
Signed-off-by: Hemant Agrawal
Hi Hemant,
Hi Ferruh,
Thanks for
Failed to create two FDIR rules with different vlan id for
the same PCTYPE. Root cause is that wrong hash key length
is used.
Fixes: 4149825bbdb9 ("net/i40e: finish integration FDIR with generic flow API")
Cc: sta...@dpdk.org
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 2 +-
otherwise the build fails with
In file included from contigmem.c:57:
/usr/srcs/head/src/sys/vm/vm_phys.h:122:10: error: use of undeclared
identifier 'vm_cnt'
return (vm_cnt.v_free_count += adj);
^
1 error generated.
*** Error code 1
Signed-off-by: Kefu Chai
---
lib/librt
This commit modifies the sw PMD credit scheme such that credits are
consumed when enqueueing a NEW event and released when an event is
released -- typically, the beginning and end of a pipeline. Workers that
simply forward events do not interact with the credit pool.
Signed-off-by: Gage Eads
---
This commit introduces a capability for disabling the "implicit" release
functionality for a port, which prevents the eventdev PMD from issuing
outstanding releases for previously dequeued events when dequeuing a new
batch of events.
If a PMD does not support this capability, the application will
This patch set introduces an event device capability for disabling implicit
release -- in which rte_event_dequeue_burst() releases any unreleased,
previously dequeued events -- on a per-port basis. The port configuration
struct is likewise extended with a disable_implicit_release option, for PMDs
t
Hi Ananda
It is quite easy to find out which NIC has been enabled on DPDK, by checking
its device ID.
For ixgbe, please refer to drivers/net/ixgbe/base/ixgbe_type.h. Those two
device IDs you listed below have been included at least in 17.11.
You can check if it has been enabled in other release
Ajith, Ferruh,
The device is BRCM 631 NIC. Let me get details about the failure.
16.11.3 is what I have compiled and the NIC bind did not work.
Thanks,
Sandeep.
From: Ajit Kumar Khaparde [mailto:ajit.khapa...@broadcom.com]
Sent: Thursday, November 30, 2017 3:11 AM
To: Ferruh Yigit
Cc: Raman, San
This commit introduces dynamically-sized IQs, by switching the underlying
data structure from a fixed-size ring to a linked list of queue 'chunks.'
This has a number of benefits:
- Certain corner cases were observed in which all of a pipeline's flows
could be pinned to one port for extended perio
This commit reinitializes a queue before it is reconfigured, such that
reorder buffer memory is not leaked.
This bug masked a few other problems, which this commit corrects as well:
- sw_port_link() allowed a port to link to a queue twice, such that the
port could then successfully unlink the qu
On Fri, 24 Nov 2017 03:03:59 -0800
Xiao Wang wrote:
> @@ -184,8 +186,10 @@ struct rte_virtio_xstats_name_off {
> "vq->hw->cvq = %p vq = %p",
> vq->vq_desc_head_idx, status, vq->hw->cvq, vq);
>
> - if ((vq->vq_free_cnt < ((uint32_t)pkt_num + 2)) || (pkt_num < 1))
From: Chas Williams
The IOMMU in some machines report that they can only support
limited widths. IOVA virtual addresses may exceed this width
making the use of IOVA virtual addresses difficult. The option
CONFIG_RTE_LIBRTE_USE_PHYS_IOVA can used to force IOVA physical
address usage.
Signed-off
From: Chas Williams
The IOMMU in some machines report that they can only support
limited widths. IOVA virtual addresses may exceed this width
making the use of IOVA virtual addresses difficult. The option
CONFIG_RTE_LIBRTE_USE_PHYS_IOVA can used to force IOVA physical
address usage.
Signed-off
From: Chas Williams
The IOMMU in some machines report that they can only support
limited widths. IOVA virtual addresses may exceed this width
making the use of IOVA virtual addresses difficult. The option
CONFIG_RTE_LIBRTE_USE_PHYS_IOVA can used to force IOVA physical
address usage.
Signed-off
> -Original Message-
> From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> Sent: Monday, November 27, 2017 8:49 PM
> To: Wang, Xiao W
> Cc: dev@dpdk.org
> Subject: Re: [PATCH 2/2] net/virtio: support GUEST ANNOUNCE
>
> On Fri, Nov 24, 2017 at 03:04:00AM -0800, Xiao Wang wrote:
> > When li
> -Original Message-
> From: Bie, Tiwei
> Sent: Friday, November 24, 2017 2:05 PM
> To: Wang, Xiao W
> Cc: dev@dpdk.org; y...@fridaylinux.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] net/virtio: support GUEST ANNOUNCE
>
> Hi,
>
> Some quick comments. Will go through the whole patch later.
> -Original Message-
> From: Bie, Tiwei
> Sent: Friday, November 24, 2017 1:39 PM
> To: Wang, Xiao W
> Cc: dev@dpdk.org; y...@fridaylinux.org
> Subject: Re: [dpdk-dev] [PATCH 1/2] net/virtio: make control queue thread-
> safe
>
> On Fri, Nov 24, 2017 at 03:03:59AM -0800, Xiao Wang wrote
OK, I found a machine in our local collection that couldn't get by with
this patch, so I will submitting a follow up that forces
rte_eval_iova_mode() into PA mode.
On Mon, Nov 27, 2017 at 7:58 AM, Chas Williams <3ch...@gmail.com> wrote:
>
>
> On Mon, Nov 27, 2017 at 4:30 AM, Zhang, Qi Z wrote:
>
Hi, Hanoh
Yes, that's true.
Thanks
Jingjing
From: Hanoch Haim (hhaim) [mailto:hh...@cisco.com]
Sent: Tuesday, November 28, 2017 11:07 PM
To: Wu, Jingjing
Cc: dev@dpdk.org
Subject: RE: [dpdk-dev] net/i40e: latency issue due fix interrupt throttling
setting in PF
Hi Jinjing,
So to verify,
Befor
From: Stephen Hemminger
This patch adds support for an additional bus type VMBUS.
Virtual Machine BUS is used on Microsoft Hyper-V in
Windows 10, Windows Server 2016 and Azure.
Most of this code was extracted from FreeBSD and
some of this is from earlier code donated by Brocade.
Only Linux is su
20/11/2017 11:27, Eelco Chaudron:
> On 17/11/17 17:19, Timothy Redaelli wrote:
> > testpmd is often used inside a VM to test the OVS PVP scenario.
> > This commit makes testpmd to link to virtio PMD when DPDK is built as
> > shared libraries too.
> >
> > Signed-off-by: Timothy Redaelli
>
> Acked-
Thanks Ferruh.
Raman,
What device are you trying this on?
Can you send the device id?
Thanks
Ajit
On Nov 29, 2017, at 11:19, Ferruh Yigit wrote:
On 11/29/2017 7:59 AM, Raman, Sandeep wrote:
Hi,
With 17.11 DPDK, Broadcom NIC binds successfully. However with 16.11 it
fails.
Is there a way to
From: Stephen Hemminger
This patch adds support for an additional bus type VMBUS.
Virtual Machine BUS is used on Microsoft Hyper-V in
Windows 10, Windows Server 2016 and Azure.
Most of this code was extracted from FreeBSD and
some of this is from earlier code donated by Brocade.
Only Linux is su
From: Stephen Hemminger
This patch adds support for an additional bus type VMBUS.
Virtual Machine BUS is used on Microsoft Hyper-V in
Windows 10, Windows Server 2016 and Azure.
Most of this code was extracted from FreeBSD and
some of this is from earlier code donated by Brocade.
Only Linux is su
On 10/23/2017 1:36 AM, Gaëtan Rivet wrote:
> On Mon, Oct 23, 2017 at 07:17:41AM +, Ophir Munk wrote:
>> Hi Gaetan,
>> Thanks for your quick reply. Please see comments inline.
>>
>> Regards,
>> Ophir
>>
>>> -Original Message-
>>> From: Gaëtan Rivet [mailto:gaetan.ri...@6wind.com]
>>> Sen
On 8/28/2017 10:08 PM, Rongqiang XIE wrote:
> In function cmd_show_bonding_config_parsed() used number represent
> the bond type,in order more detailed,add bond type description
> otherwise we may confused about the number type.
> And also,the primary port just use in mode active backup and tlb,
>
On 9/4/2017 4:54 AM, Radu Nicolau wrote:
> Hi,
>
> Wouldn't be possible to treat the section of code that segfaults as a
> critical one, i.e. use the lock/unlock instead of triggering alarms?
Hi Sha,
Any update? Is this patch still valid?
Thanks,
ferruh
>
>
> On 7/26/2017 11:13 AM, zhangsha
Signed-off-by: Xueming Li
---
lib/librte_ether/rte_eth_ctrl.h | 4 +++-
lib/librte_ether/rte_ethdev.h | 2 ++
lib/librte_ether/rte_flow.c | 1 +
lib/librte_ether/rte_flow.h | 7 +++
4 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/lib/librte_ether/rte_eth_ctrl.h b/lib/
Signed-off-by: Xueming Li
---
app/test-pmd/cmdline_flow.c | 14 +-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index df16d2a..9402eb7 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
Signed-off-by: Xueming Li
---
app/test-pmd/cmdline_flow.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 9402eb7..762ab2c 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -160,6 +160,7 @@ en
Some tunnel enhancements:
1. support GRE tunnel type
2. support L3 VXLAN tunnel type - no inner L2 header
3. introduce RSS tunnel level into rte_flow_action_rss
RSS on inner or outer tunnel headers
4. implementation of rss tunnel level parsing in testpmd
Xueming Li (4):
ethdev: support rss l
There was no RSS hash fields level definition on tunnel, implementations
default RSS on tunnel to outer or inner. Adding rss level enable users
to specifiy the tunnel level of RSS hash fields.
0: outer most,
1: inner,
-1: inner most(PMD auto detection if nested tunnel specified in fields)
Signe
On 11/29/2017 7:59 AM, Raman, Sandeep wrote:
> Hi,
>
> With 17.11 DPDK, Broadcom NIC binds successfully. However with 16.11 it fails.
> Is there a way to backport the bnxt library from 17.11 to 16.11?
>
> Any other ways to get this to work with 16.11.
Can Long term stable tree works for you (DPD
On 29 Nov 04:56, Jerin Jacob wrote:
> -Original Message-
> > Date: Wed, 29 Nov 2017 12:19:54 +
> > From: "Ma, Liang"
> > To: Jerin Jacob
> > CC: dev@dpdk.org, "Van Haaren, Harry" ,
> > "Richardson, Bruce" , "Jain, Deepak K"
> > , "Mccarthy, Peter"
> > Subject: Re: [RFC PATCH 0/7] R
Add support to AES-CCM, for 128, 192 and 256-bit keys.
Signed-off-by: Fan Zhang
---
v2:
* Fixed a few bugs
doc/guides/cryptodevs/aesni_mb.rst | 4 +
doc/guides/cryptodevs/features/aesni_mb.ini| 3 +
doc/guides/rel_notes/release_18_02.rst | 5 +
drivers/
Apologies for not clearing up what was in the clean up. In the cleanups
i have removed "rte_" from the file
names and also variable names as the issue was that many APIs were
exposed in the documentation.
In doing so it leaves rte_power.h as the only header file that's installed.
On 28/11/2017
As i saw that many other variables were changed from unsigned to
unsigned int
and since i was cleaning up the code i thought it would be beneficial
to change it now not to cause checkpatch problems or other problems down
the line.
Marko
On 28/11/2017 14:16, Bruce Richardson wrote:
On Tue, Nov
Hi,
With 17.11 DPDK, Broadcom NIC binds successfully. However with 16.11 it fails.
Is there a way to backport the bnxt library from 17.11 to 16.11?
Any other ways to get this to work with 16.11.
Thanks,
Sandeep.
Some devices needs more time to initialize and bring interface up. When
link is down the link properties are not valid, e.g. link_speed is
reported as 0 and this is not a valid speed for slave as well as for whole
bonding.
During NIC (and bonding) initialization there's concurrency between
updatin
Some devices needs more time to initialize and bring interface up. When
link is down the link properties are not valid, e.g. link_speed is
reported as 0 and this is not a valid speed for slave as well as for whole
bonding.
During NIC (and bonding) initialization there's concurrency between
updatin
Wednesday, November 29, 2017 2:14 PM, Nelio Laranjeiro:
> On Wed, Nov 29, 2017 at 01:29:07PM +0200, Shahaf Shuler wrote:
> > Update the guide with more details on the different statistics query
> > possible with MLX5 PMD.
> >
> > Signed-off-by: Shahaf Shuler
> > ---
> > doc/guides/nics/mlx5.rst |
-Original Message-
> Date: Wed, 29 Nov 2017 12:19:54 +
> From: "Ma, Liang"
> To: Jerin Jacob
> CC: dev@dpdk.org, "Van Haaren, Harry" ,
> "Richardson, Bruce" , "Jain, Deepak K"
> , "Mccarthy, Peter"
> Subject: Re: [RFC PATCH 0/7] RFC:EventDev OPDL PMD
> User-Agent: Mutt/1.5.20 (2009
Function i40e_GLQF_reg_init() overwrites global register for
flexible payload, forcing extraction of first 16 bytes of L2/L3/L4
payload to the field vector even if flexible payload is not used by
an application. Such unconditional turn on of flexible payload
effectively disables ability to use oute
Hi Anoob,
On Wed, Nov 29, 2017 at 06:00:38PM +0530, Anoob wrote:
>Hi Nelio,
>
>Since support of RSS with inline crypto/protocol is hardware
>implementation dependent, it would be better if there is some sort of
>capability check before setting the flow parameters in the applicatio
-Original Message-
> Date: Mon, 27 Nov 2017 15:49:45 +0800
> From: Herbert Guan
> To: jerin.ja...@caviumnetworks.com, jianbo@arm.com, dev@dpdk.org
> CC: Herbert Guan
> Subject: [PATCH] arch/arm: optimization for memcpy on AArch64
> X-Mailer: git-send-email 1.8.3.1
> +
> +/
Hi Nelio,
Since support of RSS with inline crypto/protocol is hardware
implementation dependent, it would be better if there is some sort of
capability check before setting the flow parameters in the application.
If the hardware doesn't support RSS with inline processing, then the RSS
flow a
Hi Jerin,
Many thanks for your comments. Please check my comment below.
On 25 Nov 02:25, Jerin Jacob wrote:
> -Original Message-
> > Date: Fri, 24 Nov 2017 11:23:45 +
> > From: liang.j...@intel.com
> > To: jerin.ja...@caviumnetworks.com
> > CC: dev@dpdk.org, harry.van.haa...@intel.
On Wed, Nov 29, 2017 at 01:29:07PM +0200, Shahaf Shuler wrote:
> Update the guide with more details on the different statistics query
> possible with MLX5 PMD.
>
> Signed-off-by: Shahaf Shuler
> ---
> doc/guides/nics/mlx5.rst | 13 +++--
> 1 file changed, 11 insertions(+), 2 deletions(-)
-Original Message-
> Date: Thu, 9 Nov 2017 12:24:13 +0530
> From: Abhinandan Gujjar
> To: jerin.ja...@caviumnetworks.com
> CC: dev@dpdk.org, narender.vang...@intel.com, Abhinandan Gujjar
> , Nikhil Rao , Gage
> Eads
> Subject: [RFC] eventdev: add crypto adapter API header
> X-Mailer: gi
Update the guide with more details on the different statistics query
possible with MLX5 PMD.
Signed-off-by: Shahaf Shuler
---
doc/guides/nics/mlx5.rst | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index f955
Extend the PMD extended statistics with more counters on the physical
port.
Signed-off-by: Shahaf Shuler
---
drivers/net/mlx5/mlx5_stats.c | 16
1 file changed, 16 insertions(+)
diff --git a/drivers/net/mlx5/mlx5_stats.c b/drivers/net/mlx5/mlx5_stats.c
index 5e225d374..ae55edaa
On Mon, Nov 27, 2017 at 06:07:35PM +0800, Sam wrote:
> I'm working in ovs 2.4 version, and I use "dpdkb" type netdev. Then I
> call `ovs-ofctl mod br dpdkbond DOWN` to down this device, at last, my
> code call "rte_eth_dev_stop(dev->port_id);".
>
> But counter shows the port is still rx/tx, why?
On Tue, Nov 28, 2017 at 11:40:07AM -0600, Erik Gabriel Carrillo wrote:
> If an eventdev PMD does not wish to provide event timer adapter ops
> definitions, the library will fall back to a default software
> implementation whose entry points are added by this commit.
>
I do agree with having sw dri
On Tue, Nov 28, 2017 at 11:40:05AM -0600, Erik Gabriel Carrillo wrote:
> # versioning export map
> EXPORT_MAP := rte_eventdev_version.map
> diff --git a/lib/librte_eventdev/rte_event_timer_adapter.h
> b/lib/librte_eventdev/rte_event_timer_adapter.h
> new file mode 100644
> index 000..14661b5
Hi Jay,
No worries. Welcome to DPDK :-D
Regards,
Fan
> -Original Message-
> From: Jay Zhou [mailto:jianjay.z...@huawei.com]
> Sent: Tuesday, November 28, 2017 1:27 AM
> To: Zhang, Roy Fan ; dev@dpdk.org
> Cc: y...@fridaylinux.org; maxime.coque...@redhat.com;
> arei.gong...@huawei.com; Ze
On Tue, Nov 28, 2017 at 06:13:35PM -0800, Ferruh Yigit wrote:
> On 11/23/2017 8:53 AM, Bruce Richardson wrote:
> > +ifeq ($(CC_AVX2_SUPPORT), 1)
> > + SRCS-$(CONFIG_RTE_LIBRTE_ACL) += i40e_rxtx_vec_avx2.c
>
> Minor issue:
> s/CONFIG_RTE_LIBRTE_ACL/CONFIG_RTE_LIBRTE_I40E_INC_VECTOR/
>
Ooops. Goo
On 11/29/2017 12:38 AM, Ferruh Yigit wrote:
On 11/28/2017 2:17 AM, Andrew Rybchenko wrote:
On 11/27/2017 10:58 PM, Ferruh Yigit wrote:
On 11/16/2017 12:04 AM, Andrew Rybchenko wrote:
From: Mark Spender
What has been fixed here?
If you can provide more details I can amend the commit later.
On 11/27/2017 10:58 PM, Ferruh Yigit wrote:
On 11/16/2017 12:03 AM, Andrew Rybchenko wrote:
From: Andrew Jackson
Signed-off-by: Andrew Jackson
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/siena_flash.h | 12 +---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --
This patch adds command to configure input set for
RSS/flow director/flow director flexible payload.
Signed-off-by: Beilei Xing
---
app/test-pmd/cmdline.c | 232 +
1 file changed, 232 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pm
The patchset adds support RSS/FDIR/FDIR flexible payload input set configuration
for some pctype.
v3 changes:
- Add support reset RSS/FDIR/FDIR flexible payload input set for some pctype.
v2 changes:
- Use 'static inline' to replace 'inline'.
- Add support get status for some filed index of in
This patch supports getting/setting input set info for
RSS/FDIR/FDIR flexible payload.
Also add some helper functions for input set configuration.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/rte_pmd_i40e.c | 141 ++
drivers/net/i40e/rte_pmd_i40e.h
> -Original Message-
> From: Richardson, Bruce
> Sent: Tuesday, November 28, 2017 7:22 PM
> To: Chen, Junjie J
> Cc: Gonzalez Monroy, Sergio ; Mcnamara,
> John ; Tahhan, Maryam
> ; Pattan, Reshma ;
> dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] doc: add a restriction to multi-process
>
This patch add a restriction to multi-process support: secondary
processes should only run alongside primary process with same DPDK
version, so that secondary processes can use the same hugepage mmap
layout as primary process.
Signed-off-by: Junjie Chen
---
v3 changes:
Use 'note' marker to make
This RFC suggests support for registering and waiting for Rx
interrupts in failsafe PMD. This allows applications to wait
for Rx events from the PMD using the DPDK rte_epoll subsystem.
The failsafe PMD presents to the application a facade of a single
device to be handled by the application while in
The metrics example didn't retrieve the metrics' names, and also had
some more minor issues with repetitive error handling code and missing
variable declarations.
Signed-off-by: Mattias Rönnblom
---
doc/guides/prog_guide/metrics_lib.rst | 40 ---
1 file changed, 2
This workaround was needed to properly handle device removal with old
Mellanox OFED releases that are not supported by this PMD anymore.
Starting from rdma-core v16 this removal issue shouldn't happen when
setting MLX4_DEVICE_FATAL_CLEANUP environment variable to 1.
Set the aforementioned variabl
This RFC suggests support for registering and waiting for Rx
interrupts in failsafe PMD. This allows applications to wait
for Rx events from the PMD using the DPDK rte_epoll subsystem.
The failsafe PMD presents to the application a facade of a single device
to be handled by the application while in
75 matches
Mail list logo
