Hi Adrien,
Thanks for the prompt reply!
You are right, I didn't go via the DPDK route, in the hope that Mellanox
will provide the exact source and configuration.
DPDK 16.07 from dpdk.org works like a charm and my NIC is in PMD mode,
thanks a lot for your guidance!
Best regards,
Georgios
On Thu,
Hi George,
On Thu, Aug 18, 2016 at 05:41:38PM +0200, george.dit at gmail.com wrote:
> Hi,
>
> I have a single port Mellanox ConnectX 4 100GbE NIC and I want to test its
> Rx/Tx capabilites using DPDK.
> My system runs a Linux kernel 4.4 compiled from sources.
>
> I found the PMD driver for
Hi,
When I try to add the virtio interface to dpdk enabled ovs bridge I am
getting below error message.
"Unable to add port vnet1 to OVS bridge ovsbr0"
I have added created the ovs bridge and added ports to the same as below.
$OVS_DIR/utilities/ovs-vsctl add-port ovsbr0 vhost-user1 -- set
Hi,
I have a single port Mellanox ConnectX 4 100GbE NIC and I want to test its
Rx/Tx capabilites using DPDK.
My system runs a Linux kernel 4.4 compiled from sources.
I found the PMD driver for this NIC as provided by Mellanox here
Existing cntvct_el0 based rte_rdtsc() provides portable
means to get wall clock counter at user space. Typically
it runs at <= 100MHz.
The alternative method to enable rte_rdtsc() for high resolution
wall clock counter is through armv8 PMU subsystem.
The PMU cycle counter runs at CPU frequency,
No need to use a pointer to store/retrieve features.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/vhost_user.c | 20
1 file changed, 8 insertions(+), 12 deletions(-)
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index ef4a0c1..eee99e9 100644
Invoke get_device() at the beginning of vhost_user_msg_handler, so that
we could check the return value once. Which could save tons of duplicate
get-and-check device.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/vhost_user.c | 160 +-
1 file changed,
Some functions are with prefix "user_", while others with "vhost_".
Making them all starting with "vhost_user_" to unify the function names.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/vhost_user.c | 60 +--
1 file changed, 30 insertions(+), 30
Due to history reason (that we have 2 vhost implementations), some
messages are handled in two calls: vhost specific implementation
handles it first and then invoke the common one to do another handling.
We have one implementation only now, we could write one method for
each message. Here fold
The code structure is a bit messy now. For example, vhost-user message
handling is spread to three different files:
vhost-net-user.c virtio-net.c virtio-net-user.c
Where, vhost-net-user.c is the entrance to handle all those messages
and then invoke the right method for a specific message.
We now have one vhost implementation; no sub source dir is needed.
Rmove it by move them to upper dir.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/Makefile | 6 +++---
lib/librte_vhost/{vhost_user => }/fd_man.c | 0
lib/librte_vhost/{vhost_user =>
remove vhost-cuse code, including the eventfd_link kernel module that
is for vhost-cuse only.
The lib/virt/qemu-wrap.py is also removed, as it's mainly for vhost-cuse
usage.
As we have one vhost implementation now, one vhost config option is
needed only. Thus, CONFIG_RTE_LIBRTE_VHOST_USER is
The first patch removes the vhost-cuse (see following link for the
deprecate note)
http://dpdk.org/ml/archives/dev/2016-July/044080.html
After the removal, there is no reason to keep the vhost_user sub source
dir any more. This also brings a chance to rename all those files in a
more proper
On Thu, Aug 18, 2016 at 10:27:55AM +0200, Maxime Coquelin wrote:
>
>
> On 08/16/2016 06:14 PM, Jiayu Hu wrote:
> >When examples/vhost runs in client mode, only one QEMU can be connected.
> >This is because that examples/vhost just supports one socket file. This
> >patch is to add multiple
On Thu, Aug 18, 2016 at 10:22:38AM +0200, Maxime Coquelin wrote:
> Hi Jiayu,
>
> On 08/16/2016 06:14 PM, Jiayu Hu wrote:
> >In examples/vhost, "dev-basename" is a program option, which is to set
> >the vhost-net socket used by vhost-user, or the character device used
> >by vhost-cuse. Since
On Wed, Aug 17, 2016 at 03:54:21PM +0530, Pankaj Chauhan wrote:
> My use case is that my machine/board which is not sitting as end node of
> network but somewhere in between like an router. So the traffic looks
> something like this:
>
> Physical port 1 -> Enter VM(s) through virtio -> exit from
On Tue, Aug 16, 2016 at 12:14:39PM -0400, Jiayu Hu wrote:
> +/*
> + * This function is used to unregister drivers.
> + */
> +static void
> +unregister_drivers(int socket_num)
> +{
Redundant comment. The function name already explains it well.
> /* Register vhost user driver to handle
On Tue, Jul 05, 2016 at 06:19:24PM +0530, Jerin Jacob wrote:
> Split out SSE instruction based virtio simple Rx
> implementation to a separate file
>
> Signed-off-by: Jerin Jacob
Hi,
I was about to apply this set. I then did some build test and found a
weird issue: it breaks the build with
add test for vf vlan anti spoof
add test for vf mac anti spoof
add test for vf ping
add test for vf vlan strip
add test for vf vlan insert
add test for tx loopback
add test for all queues drop enable bit
add test for vf split drop enable bit
add test for vf mac address
add new API's to the testpmd
Add new functions to configure and manage VF's on a Niantic NIC.
add ixgbe_vf_ping function.
add ixgbe_set_vf_vlan_anti_spoof function.
add ixgbe_set_vf_mac_anti_spoof function.
Signed-off-by: azelezniak
add ixgbe_set_vf_vlan_strip function
add ixgbe_set_vf_vlan_insert function.
add
Add new API functions to configure and manage VF's on a NIC.
add rte_eth_dev_vf_ping function.
add rte_eth_dev_set_vf_vlan_anti_spoof function.
add rte_eth_dev_set_vf_mac_anti_spoof function.
Signed-off-by: azelezniak
add rte_eth_dev_set_vf_vlan_strip function.
add
call _rte_eth_dev_callback_process_vf from ixgbe_rcv_msg_from_vf function.
The callback asks the user application if it is allowed to perform
the function.
If the cb_param.retval is RTE_ETH_MB_EVENT_PROCEED then continue,
if 0, do nothing and send ACK to VF
if > 1, do nothing and send NAK to VF.
add _rte_eth_dev_callback_process_vf function.
add _rte_eth_dev_callback_process_generic function
Adding a callback to the user application on VF to PF mailbox message,
allows passing information to the application controlling the PF
when a VF mailbox event message is received, such as VF reset.
This RFC patchset contains new DPDK API's requested by AT for use
with the Virtual Function Daemon (VFD).
The need to configure and manage VF's on a NIC has grown to the
point where AT have devloped a DPDK based tool, VFD, to do this.
This RFC proposes to add the following API extensions to
Hi,
I am trying to get tagged packets to work in SRIOV mode. I am using dpdk 2.1
version with an application on KVM.
The setup is as below: The same configuration works for untagged packets.
Guest VM (Virtual Function/Created tagged kni interfaces)--- > KVM host (PF/no
tag on the VF )
Added NULL algorithm to test file for Intel(R) QuickAssist
Technology Driver
Signed-off-by: Deepak Kumar Jain
---
app/test/test_cryptodev.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 8553759..67ca912 100644
---
enabled NULL crypto for Intel(R) QuickAssist Technology
Signed-off-by: Deepak Kumar Jain
---
doc/guides/cryptodevs/qat.rst| 3 ++-
drivers/crypto/qat/qat_adf/qat_algs_build_desc.c | 2 ++
drivers/crypto/qat/qat_crypto.c | 4
3 files changed, 8
This patchset adds support of NULL crypto in Intel(R) QuickAssist Technology
driver.
This patchset depends on following patchset:
"crypto/qat: add aes-sha384-hmac capability to Intel QAT driver"
(http://dpdk.org/dev/patchwork/patch/15228/)
Deepak Kumar Jain (2):
crypto/qat: add NULL
From: "Jain, Deepak K"
Added aes-sha384-hmac algorithm to test file for Intel(R) QuickAssist
Technology Driver
Signed-off-by: Deepak Kumar Jain
---
app/test/test_cryptodev_aes.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git
From: "Jain, Deepak K"
enabled support of aes-sha384-hmac in Intel(R) QuickAssist driver
Signed-off-by: Deepak Kumar Jain
---
doc/guides/cryptodevs/qat.rst| 1 +
drivers/crypto/qat/qat_adf/qat_algs_build_desc.c | 33
This patchset adds support of aes-sha384-hmac
in Intel(R) QuickAssist Technology driver.
This patchset depends on following patchset:
"crypto/qat: add aes-sha224-hmac capability to Intel QAT driver"
(http://dpdk.org/dev/patchwork/patch/15226/)
Jain, Deepak K (2):
crypto/qat: add
From: "Jain, Deepak K"
Added aes-sha224-hmac algorithm to test file for Intel(R) QuickAssist
Technology Driver
Signed-off-by: Deepak Kumar Jain
---
app/test/test_cryptodev_aes.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git
From: "Jain, Deepak K"
Added support of aes-sha224-hmac in Intel(R) QuickAssist driver
Signed-off-by: Deepak Kumar Jain
---
doc/guides/cryptodevs/qat.rst| 1 +
drivers/crypto/qat/qat_adf/qat_algs_build_desc.c | 33
This patchset adds support of aes-sha224-hmac
in Intel(R) QuickAssist Technology driver.
This patchset depends on following patchset:
"crypto/qat: add MD5 HMAC capability to Intel QAT driver"
(http://dpdk.org/dev/patchwork/patch/15165/)
Jain, Deepak K (2):
crypto/qat: add aes-sha224-hmac
> Following discussions on the mailing list [1] and since nobody stood up to
> implement the necessary cleanups, here is the ivshmem integration removal.
>
> There is not much to say about this patch, a lot of code is being removed.
> The default configuration file for packet_ordering example is
Thanks Maxime and Yuanhan for your review and suggestions!
Please help review the v2 of this patch.
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan.liu at linux.intel.com]
> Sent: Wednesday, August 17, 2016 5:51 PM
> To: Maxime Coquelin
> Cc: Wang, Zhihong ; dev at dpdk.org
>
Hi all
Is there a function -like rte_eth_rx_queue_count()- to get the number of used
(or free) descriptors in a vHost RX/TX's queue? I used
rte_vhost_avail_entries() but I'm not sure it is correct way.
Thanks in advance.
- Volkan
On Mon, Aug 15, 2016 at 7:37 PM, Yuanhan Liu
wrote:
> On Mon, Aug 15, 2016 at 01:00:24PM -0700, Rich Lane wrote:
> > Concurrent enqueue is an important performance optimization when the
> number
> > of cores used for switching is different than the number of vhost queues.
> > I've observed a 20%
Hi Maxime,
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coquelin at redhat.com]
> Sent: Thursday, August 18, 2016 3:43 PM
> To: Tan, Jianfeng; Yuanhan Liu; Pankaj Chauhan
> Cc: dev at dpdk.org; hemant.agrawal at nxp.com; shreyansh.jain at nxp.com
> Subject: Re: [dpdk-dev]
Hi Maxime,
On 8/17/2016 7:18 PM, Maxime Coquelin wrote:
> Hi Jianfeng,
>
> On 08/17/2016 04:33 AM, Tan, Jianfeng wrote:
>> Hi,
>>
>> Please review below proposal of Pankaj and myself after an offline
>> discussion. (Pankaj, please correct me if I'm going somewhere wrong).
>>
>> a. Remove HW
On 08/18/2016 10:35 AM, Yuanhan Liu wrote:
> On Thu, Aug 18, 2016 at 10:22:38AM +0200, Maxime Coquelin wrote:
>> Hi Jiayu,
>>
>> On 08/16/2016 06:14 PM, Jiayu Hu wrote:
>>> In examples/vhost, "dev-basename" is a program option, which is to set
>>> the vhost-net socket used by vhost-user, or the
On 08/16/2016 06:14 PM, Jiayu Hu wrote:
> When examples/vhost runs in client mode, only one QEMU can be connected.
> This is because that examples/vhost just supports one socket file. This
> patch is to add multiple sockets support for examples/vhost.
>
> Signed-off-by: Jiayu Hu
> ---
>
Hi Jiayu,
On 08/16/2016 06:14 PM, Jiayu Hu wrote:
> In examples/vhost, "dev-basename" is a program option, which is to set
> the vhost-net socket used by vhost-user, or the character device used
> by vhost-cuse. Since vhost-cuse should be dropped, and "dev-basename"
> is not a suitable name for
Hi Maxime,
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coquelin at redhat.com]
> Sent: Thursday, August 18, 2016 3:52 PM
> To: Tan, Jianfeng; dev at dpdk.org
> Cc: yuanhan.liu at linux.intel.com
> Subject: Re: [dpdk-dev] [PATCH] examples/vhost: remove VLAN strip option
>
On 08/18/2016 07:46 AM, Jianfeng Tan wrote:
> When VMDQ is enabled, different NICs have different behaviors for
> disabling VLAN strip. In detail, i40e only enables/disables it of
> PF's main vsi; fm10k cannot disable VLAN strip, etc. We now remove
> this option, --vlan-strip, to reduce any
Hi,
On 08/18/2016 04:35 AM, Tan, Jianfeng wrote:
> Hi Maxime,
>
> On 8/17/2016 7:18 PM, Maxime Coquelin wrote:
>> Hi Jianfeng,
>>
>> On 08/17/2016 04:33 AM, Tan, Jianfeng wrote:
>>> Hi,
>>>
>>> Please review below proposal of Pankaj and myself after an offline
>>> discussion. (Pankaj, please
This patch reorders the code to delay virtio header write to optimize cache
access efficiency for cases where the mrg_rxbuf feature is turned on. It
reduces CPU pipeline stall cycles significantly.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost_rxtx.c | 23 ---
1
This patch enables batch update of the used ring for better efficiency.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost-net.h | 4 +++
lib/librte_vhost/vhost_rxtx.c | 68 +--
lib/librte_vhost/virtio-net.c | 15 --
3 files changed, 68
This patch adds descriptor prefetch to hide cache access latency.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost_rxtx.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
index 939957d..7db83d0 100644
---
This patch removes useless volatile attribute to allow compiler
optimization.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost-net.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_vhost/vhost-net.h b/lib/librte_vhost/vhost-net.h
index 38593a2..51fdf3d
This patch removes obsolete functions.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost_rxtx.c | 408 --
1 file changed, 408 deletions(-)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
index 8e6d782..939957d 100644
---
This patch implements the vhost logic from scratch into a single function
designed for high performance and better maintainability.
Signed-off-by: Zhihong Wang
---
lib/librte_vhost/vhost_rxtx.c | 212 --
1 file changed, 205 insertions(+), 7 deletions(-)
This patch set optimizes the vhost enqueue function.
It implements the vhost logic from scratch into a single function designed
for high performance and good maintainability, and improves CPU efficiency
significantly by optimizing cache access, which means:
* For fast frontends (eg. DPDK
53 matches
Mail list logo