Add test cases in testpmd to test VxLAN Tx Checksum offload, which include
- IPv4 and IPv6 packet
- outer L3, inner L3 and L4 checksum offload for Tx side.
Signed-off-by: Jijiang Liu
---
app/test-pmd/cmdline.c | 13 ++-
app/test-pmd/config.c |6 +-
app/test-pmd/csumonly.c | 194 +
Support VxLAN Tx checksum offload, which include
- outer L3(IP) checksum offload
- inner L3(IP) checksum offload
- inner L4(UDP, TCP and SCTP) checksum offload
Signed-off-by: Jijiang Liu
---
lib/librte_mbuf/rte_mbuf.h |1 +
lib/librte_pmd_i40e/i40e_rxtx.c | 46 ++
Add the "tunnel_filter" command in testpmd to test the API of VxLAN packet
filter.
Signed-off-by: Jijiang Liu
---
app/test-pmd/cmdline.c | 150
1 files changed, 150 insertions(+), 0 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app/test-pm
The filter types supported are listed below for VxLAN:
1. Inner MAC and Inner VLAN ID.
2. Inner MAC address, inner VLAN ID and tenant ID.
3. Inner MAC and tenant ID.
4. Inner MAC address.
5. Outer MAC address, tenant ID and inner MAC address.
Signed-off-by: Jijiang Liu
---
lib/lib
Add definations of the data structures of tunneling packet filter in the
rte_eth_ctrl.h file.
Signed-off-by: Jijiang Liu
---
lib/librte_ether/rte_eth_ctrl.h | 49 +++
1 files changed, 49 insertions(+), 0 deletions(-)
diff --git a/lib/librte_ether/rte_eth_c
Add two commands to test VxLAN packet identification.
The test steps are as follows:
1> use commands to add/delete VxLAN UDP port.
2> use rxonly mode to receive VxLAN packet.
Signed-off-by: Jijiang Liu
---
app/test-pmd/cmdline.c | 65
app/test
Implement the configuration API of VxLAN destination UDP port in
librte_pmd_i40e,
and add new Rx offload flags for supporting VXLAN packet offload.
Signed-off-by: Jijiang Liu
---
lib/librte_mbuf/rte_mbuf.h|2 +
lib/librte_pmd_i40e/i40e_ethdev.c | 157 +++
There are "some" destination UDP port numbers that have unque meaning.
In terms of VxLAN, "IANA has assigned the value 4789 for the VXLAN UDP port,
and this value SHOULD be used by default as the destination UDP port. Some
early implementations of VXLAN have used other values for the destination
Add definations of basic data structures of VxLAN.
Signed-off-by: Jijiang Liu
---
lib/librte_ether/rte_eth_ctrl.h | 12
lib/librte_ether/rte_ethdev.h |8
lib/librte_ether/rte_ether.h| 13 +
3 files changed, 33 insertions(+), 0 deletions(-)
diff --
Replace the "reserved2" field with the "packet_type" field and add the
"inner_l2_l3_len" field in the rte_mbuf structure.
The "packet_type" field is used to indicate ordinary packet format and also
tunneling packet format such as IP in IP, IP in GRE, MAC in GRE and MAC in UDP.
The "inner_l2_len"
The patch set supports VxLAN on Fortville based on latest rte_mbuf structure.
It includes:
- Support VxLAN packet identification by configuring UDP tunneling port.
- Support VxLAN packet filters. It uses MAC and VLAN to point
to a queue. The filter types supported are listed below:
1. Inn
2014-10-23 21:18, Jijiang Liu:
> Add two commands to test VxLAN packet identification.
> The test steps are as follows:
> 1> use commands to add/delete VxLAN UDP port.
> 2> use rxonly mode to receive VxLAN packet.
>
> Signed-off-by: Jijiang Liu
[...]
> static const char *pkt_rx_flag_names[MAX_
On Thu, Oct 23, 2014 at 5:18 PM, Jay Rolette wrote:
> Tim,
>
> Thanks for sharing this. If nothing else, I wanted to at least provide some
> feedback on the parts that look useful to me for my applications/product.
> Bits that make me interested in the release:
>
>
>
> *> 2.0 (Q1 2015) DPDK Featu
In the case where a userspace reports itself as Ubuntu, but the
kernel isn't providing the expected version signature interface,
turn off Ubuntu specializations.
This situation happens often enough in development environments,
and with multi-distribution build servers (e.g. chroot, containers).
S
Building combined shared libraries results in applications being linked
against separeted/individual and combined libs altogether.
Link only against combined lib when the config option is enabled.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.app.mk | 10 ++
1 file changed, 6 inserti
Building combined shared libs fails if we set EXTRA_CFLAGS=-O0.
/usr/bin/ld: test: hidden symbol `mknod' in
/usr/lib64/libc_nonshared.a(mknod.oS) is referenced by DSO
/usr/bin/ld: final link failed: Bad value
collect2: error: ld returned 1 exit status
Fix: link combined shared lib using CC if LI
Thanks for all the effort!
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, October 23, 2014 4:11 AM
> To: Xie, Huawei
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 00/14] new vhost example
>
> 2014-10-20 12:38, Huawei Xie:
> >
> -Original Message-
> From: Matthew Hall [mailto:mhall at mhcomputing.net]
> Sent: Wednesday, October 22, 2014 7:42 PM
> On Wed, Oct 22, 2014 at 03:20:40PM +, Gonzalez Monroy, Sergio wrote:
>
> > You are not forced to use shared libraries. This module loads
> > successfully with an app
These changes are a conversion of the FreeBSD Getting Started Guide from an
MSWord file to Sphinx rst files.
The following changes since commit 1ab07743b21b785a71fa334641ab58e779532600:
doc: getting started guide for linux (2014-10-23 14:36:42 +0200)
are available in the git repository at:
On 2014/10/23 13:56, Zhang, Jerry wrote:
> Hi lijun,
>
> It seems that one of the PCI BAR mapping failed. Need to identify the
> accurate code line of the mapping error.
>
> Is this a common issue for all 82599 VFs in your environment or just
> one rare case? If it's just one c
On 2014/10/23 13:52, Dong, Binghua wrote:
> Hi all,
>
> It looks like that Lilijun runs the four steps are both in HOST?
>
> Lilijun, you did not startup a VM, pass through the VF to VM, and runs l2fwd
> in VM?
Yes, that's all in HOST. The problem has been fixed.
>
> -Original Message
> Bernard Iremonger (1):
> doc: getting started guide for linux
>
> doc/guides/conf.py | 30 +++
> doc/guides/index.rst | 37 +++
> doc/guides/linux_gsg/build_dpdk.rst| 271 +++
> doc/guides/linux_gsg/build_sampl
Some code style fixes are added.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i40e/i40e_ethdev.c | 20 +++-
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
b/lib/librte_pmd_i40e/i40e_ethdev.c
index 30efbb0..29b17e3 100644
--- a/l
i40e_register_x710_int.h contains all the register definitions
for early hardware only. It should not be included and all the
registers defined in it should not be used anywhere.
Signed-off-by: Helin Zhang
---
lib/librte_pmd_i40e/i40e/i40e_register_x710_int.h | 10712
lib/li
For those registers for early hardware only, they should neither
be defined nor be used. According to this, a header file which
defines all those registers should be deleted at all, and code
lines of accessing those registers should be deleted as well.
Helin Zhang (2):
i40e: remove registers for
On 2014/10/23 13:37, Dong, Binghua wrote:
> Hi Lilijun,
>
> I think you need unbind the device 02:10.0 from driver ixgbevf first; In Host
> Linux, you also need ifconfig up the PF eth of this VF;
The script dpdk_nic_bind.py will unbind ixgbevf automatically when doing
dpdk_nic_bind.py -b igb_ui
On 2014/10/23 12:56, Zhang, Helin wrote:
>
>
>> -Original Message-
>> From: Lilijun [mailto:jerry.lilijun at huawei.com]
>> Sent: Thursday, October 23, 2014 12:31 PM
>> To: Zhang, Helin; dev at dpdk.org
>> Subject: Re: [dpdk-dev] Problems when testing l2fwd with ixgbevf pmd dirver
>>
>> O
2014-10-20 12:38, Huawei Xie:
> New vhost example is a transform of old vhost example.
> It calls vhost library API to receive/transmit packets from guest VM.
> zero copy logic is kept in vhost example.
>
> Huawei Xie (14):
> copy old vhost example files main.c/main.h as the base of new vhost ex
On 2014/10/23 11:44, Zhang, Helin wrote:
> Hi
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Lilijun
>> Sent: Thursday, October 23, 2014 11:36 AM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] Problems when testing l2fwd with ixgbevf pmd dirver
>>
>> Hi a
> From: lukego at gmail.com [mailto:lukego at gmail.com] On Behalf Of Luke
> Gorrie
>
> On 22 October 2014 15:48, O'driscoll, Tim wrote:
>
> Single Virtio Driver: Merge existing Virtio drivers into a single
> implementation, incorporating
> the best features from each of the existing drivers.
>
Hi All,
(2014/10/22 22:48), O'driscoll, Tim wrote:
> Single Virtio Driver: Merge existing Virtio drivers into a single
> implementation, incorporating the best features from each of the existing
> drivers.
It's nice plan. We should do it.
In my understanding, the following drivers could be merg
Hi,
(2014/10/22 22:48), O'driscoll, Tim wrote:
> PCI Hot Plug: When you migrate a VM, you need hot plug support as the new VF
> on the new hardware you are running on post-migration needs to be
> initialized. With an emulated NIC migration is seamless as all configuration
> for the NIC is with
2014-10-20 12:38, Huawei Xie:
> new vhost example is ready. Add example makefile
>
> Signed-off-by: Huawei Xie
> ---
> examples/vhost/Makefile | 60
> +
> 1 file changed, 60 insertions(+)
> create mode 100644 examples/vhost/Makefile
A change in
2014-10-20 12:38, Huawei Xie:
> vmdq_rx_q minior change
This patch should be merged with vhost_dev structure definition (patch 4).
--
Thomas
2014-10-20 12:38, Huawei Xie:
> other vhost APIs
> rte_vhost_driver_register: register vhost driver
> rte_vhost_driver_callback_register: register new_device/destroy_device
> callback
> rte_vhost_driver_session_start: start vhost session loop
> rte_vhost_feature_disable: di
2014-10-20 12:38, Huawei Xie:
> remove gpa_to_vva macro and base_index parameter.
gpa_to_vva function should be removed in patch 2
base_index should be removed when implementing registration (next patch)
--
Thomas
Hi all,
I am testing the example l2fwd of dpdk-1.7.0 using intel 82599 VF NICs.
But the VF can't be used when doing probe pmd driver with following error:
EAL: PCI device :02:10.0 on NUMA socket -1
EAL: probe driver: 8086:10ed rte_ixgbevf_pmd
EAL: PCI memory mapped at 0x7f78b181f000
EAL:
> From: Tetsuya Mukawa [mailto:mukawa at igel.co.jp]
>
> (2014/10/22 22:48), O'driscoll, Tim wrote:
> > Single Virtio Driver: Merge existing Virtio drivers into a single
> implementation, incorporating the best features from each of the existing
> drivers.
>
> It's nice plan. We should do it.
> I
On 21/10/14 15:01, Stephen Hemminger wrote:
> On Tue, 21 Oct 2014 11:14:43 +0200
> Marc Sune wrote:
>
>> On 21/10/14 10:46, Thomas Monjalon wrote:
>>> My balance is different because I have a simpler solution for Marc's
>>> problem:
>>> git fetch && git merge $(git tag | grep -v -- -rc | tail
Danny, you are correct in that DPDK shouldn't see any performance
degradation when IOMMU is present.
I was rather presenting an existing issue with kernel drivers
(bnx2x,ixgbe,mlx4, etc...) that create new RX/TX descriptors for each
packet and perform lots of map/unmap operations.
On Wed, Oct 22,
> From: Tetsuya Mukawa [mailto:mukawa at igel.co.jp]
>
> (2014/10/22 22:48), O'driscoll, Tim wrote:
>
> > PCI Hot Plug: When you migrate a VM, you need hot plug support as the
> new VF on the new hardware you are running on post-migration needs to be
> initialized. With an emulated NIC migration
2014-10-23 07:27, Zhang, Helin:
> > > The first patch is missing.
> >
> > It was blocked in moderation queue because of its size.
>
> Yes, the first patch includes removing a head file of about 10,000 lines of
> code.
> I need to do nothing, right?
No problem, it's unblocked now and the patch w
> The first patch is missing.
It was blocked in moderation queue because of its size.
--
Thomas
Hi Helin,
The first patch is missing.
--
Thomas
Tim,
Thanks for sharing this. If nothing else, I wanted to at least provide some
feedback on the parts that look useful to me for my applications/product.
Bits that make me interested in the release:
*> 2.0 (Q1 2015) DPDK Features:> Bifurcated Driver: With the Bifurcated
Driver, the kernel will
For me these defines make perfect sense - tunnelling filters require
combinations of different tunnel components, but not all combinations are
valid. So defining valid combinations separately helps.
Regards,
Andrey
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at
Hi,
in this code, pointer of local variable (mb_def) is returned by your changes.
mb_def should be static for each thread.
2014-10-22 19:55 GMT+09:00 Bruce Richardson :
> Remove the "static" prefix to the template mbuf variable in
> ixgbe_rxq_vec_setup function. This will then allow different
> t
Hi Thomas
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Thursday, October 23, 2014 3:25 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/2] remove registers for early hardware only
>
> > The first patch is missing.
Hi lijun,
It seems that one of the PCI BAR mapping failed. Need to identify the
accurate code line of the mapping error.
Is this a common issue for all 82599 VFs in your environment or just one
rare case? If it's just one case, maybe it is caused by configure problem and
please try
Hi all,
It looks like that Lilijun runs the four steps are both in HOST?
Lilijun, you did not startup a VM, pass through the VF to VM, and runs l2fwd in
VM?
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Lilijun
Sent: Thursday, October 23, 2014 12:31 PM
To: Zhan
Hi Lilijun,
I think you need unbind the device 02:10.0 from driver ixgbevf first; In Host
Linux, you also need ifconfig up the PF eth of this VF;
Please ref my script to unbind a VF device and pass into VM, such as:
PCI_eth=":02:10.0"
ID_eth=" "
PCIX_eth=" -device pci-assign,host="${P
> -Original Message-
> From: Lilijun [mailto:jerry.lilijun at huawei.com]
> Sent: Thursday, October 23, 2014 12:31 PM
> To: Zhang, Helin; dev at dpdk.org
> Subject: Re: [dpdk-dev] Problems when testing l2fwd with ixgbevf pmd dirver
>
> On 2014/10/23 11:44, Zhang, Helin wrote:
> > Hi
> >
Hi
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Lilijun
> Sent: Thursday, October 23, 2014 11:36 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] Problems when testing l2fwd with ixgbevf pmd dirver
>
> Hi all,
>
> I am testing the example l2fwd of dpdk-1.7
Hi
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Wednesday, October 22, 2014 4:46 PM
> To: Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v6 1/9] librte_mbuf:the rte_mbuf structure
> changes
>
> 2014-10-21 14:14
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson
> Sent: Wednesday, October 22, 2014 6:56 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 0/2] ixgbe: vector pmd fixes
>
> This patch set contains small fixes for issues with the vecto
55 matches
Mail list logo
