Re: [dpdk-users] mellanox connect-x 3 pro dpdk with openvswitch

[Raslan Darawsheh]

I'm not sure about that since as I mentioned before I haven't tried it before,
so give it a try and let me know if it worked for you.

Kindest regards,
Raslan Darawsheh

-Original Message-
From: Tobias Urdin [mailto:tobias.ur...@crystone.com] 
Sent: Monday, June 11, 2018 3:14 PM
To: Raslan Darawsheh 
Cc: users@dpdk.org
Subject: Re: mellanox connect-x 3 pro dpdk with openvswitch

I found this librte-pmd-mlx4 driver but it's for 18.02 and I have dpdk
17.11.1 installed.

It contains files:
./lib/x86_64-linux-gnu/dpdk-18.02-drivers/librte_pmd_mlx4.so.18.02
./lib/x86_64-linux-gnu/librte_pmd_mlx4.so.18.02

I assume I can compile librte-pmd-mlx4 manually for 17.11 instead and place it 
in that folder.
It looks like this:

$ls -l /usr/lib/x86_64-linux-gnu/dpdk-17.11-drivers/ | head -n 5 total 0 
lrwxrwxrwx 1 root root 35 Mar 22 10:23 librte_mempool_octeontx.so.17.11
-> ../librte_mempool_octeontx.so.17.11
lrwxrwxrwx 1 root root 31 Mar 22 10:23 librte_mempool_ring.so.17.11 ->
../librte_mempool_ring.so.17.11
lrwxrwxrwx 1 root root 32 Mar 22 10:23 librte_mempool_stack.so.17.11 ->
../librte_mempool_stack.so.17.11
lrwxrwxrwx 1 root root 32 Mar 22 10:23 librte_pmd_af_packet.so.17.11 ->
../librte_pmd_af_packet.so.17.11
...

Would that be enough or is there anything else I need?
Best regards

On 06/11/2018 02:07 PM, Raslan Darawsheh wrote:
> I think OVS-DPDK has static linking for dpdk and I'm not sure you can 
> do it this way
>
> Kindest regards,
> Raslan Darawsheh
>
> -Original Message-
> From: Tobias Urdin [mailto:tobias.ur...@crystone.com]
> Sent: Monday, June 11, 2018 3:05 PM
> To: Raslan Darawsheh 
> Cc: users@dpdk.org
> Subject: Re: mellanox connect-x 3 pro dpdk with openvswitch
>
> Ok, I'm more wondering if there is any way I can build a librte-pmd-mlx4 
> package that supplies the compiled PMD or something like that instead of 
> having to rebuild all the major parts?
>
> Thanks!
> Best regards
>
> On 06/11/2018 01:51 PM, Raslan Darawsheh wrote:
>> I think you can upgrade the current version that you have as seen in this 
>> article:
>> https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fdoc
>> s 
>> .openvswitch.org%2Fen%2Flatest%2Fintro%2Finstall%2Fgeneral%2F%23hot-u
>> p
>> grading=02%7C01%7Crasland%40mellanox.com%7Cfdaa105976c744dba2f90
>> 8
>> d5cf9394cc%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C1%7C6366431551416
>> 0
>> 5703=VhpqI3M4RdnD5Shpbt6O8YmufIKiyoYK9kxO9TR1Yrk%3D=0
>>
>> but, I haven't tried it before.
>>
>> Kindest regards,
>> Raslan Darawsheh
>>
>> -Original Message-
>> From: Tobias Urdin [mailto:tobias.ur...@crystone.com]
>> Sent: Monday, June 11, 2018 2:45 PM
>> To: Raslan Darawsheh 
>> Cc: users@dpdk.org
>> Subject: Re: mellanox connect-x 3 pro dpdk with openvswitch
>>
>> Is there any to skip compiling DPDK and OVS+OVS-DPDK and instead use the 
>> ones I have installed?
>> Only compiling the Mellanox PMD module/binary or something similar to that?
>>
>> I can see that other drivers supply pmd packages like below.
>>
>> $dpkg -l | grep -E "openvswitch|pmd|dpdk"
>> ii  dpdk  
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (runtime)
>> ii  librte-pmd-af-packet17.11:amd64   
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-af-packet runtime library)
>> ii  librte-pmd-ark17.11:amd64 
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte_pmd_ark runtime library)
>> ii  librte-pmd-avp17.11:amd64 
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte_pmd_avp runtime library)
>> ii  librte-pmd-bnxt17.11:amd64
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-bnxt runtime library)
>> ii  librte-pmd-bond17.11:amd64
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-bond runtime library) ii  
>> librte-pmd-crypto-scheduler17.11:amd64
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte_pmd_crypto_scheduler runtime library)
>> ii  librte-pmd-cxgbe17.11:amd64   
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-cxgbe runtime library)
>> ii  librte-pmd-e1000-17.11:amd64  
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-e1000 runtime library)
>> ii  librte-pmd-ena17.11:amd64 
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-ena runtime library)
>> ii  librte-pmd-enic17.11:amd64
>> 17.11.1-6~cloud0amd64Data Plane
>> Development Kit (librte-pmd-enic runtime library)
>> ii  librte-pmd-failsafe17.11:amd64
>> 17.11.1-6~cloud0   

Re: [dpdk-users] I40evf VLAN stripping disable

[Xing, Beilei]

Hi Souvik,

Yes, if adding vlan filter, the packet will be received with vlan stripped even 
if the vlan_strip is disabled. It’s i40e HW limitation, different from ixgbe.

Best Regards,
Beilei

From: Dey, Souvik [mailto:so...@rbbn.com]
Sent: Tuesday, June 12, 2018 12:26 AM
To: Xing, Beilei ; Wu, Jingjing ; 
d...@dpdk.org
Cc: users@dpdk.org
Subject: RE: I40evf VLAN stripping disable

Hi Xing,
Do we need to do i40evf_add_vlan() to get the vlan packets up 
to the VM like we had in ixgbe_vf?  I see that if I have the vlan added through 
i40evf_add_vlan(), then the packet is received with the vlan stripped even if 
the vlan_strip is disabled in the PF. If I just do vlan_strip disabled and then 
send vlan packets it is all coming up to the VM. But is that not security issue 
as all VLANs even if not configured will be coming up to the VM, which never 
happened in the ixgbe case. I only have rte_eth_allmulticast_enable in my app , 
so expect that unicast/broadcast promisc mode is disabled by default.

--
Regard,
Souvik

From: Xing, Beilei [mailto:beilei.x...@intel.com]
Sent: Monday, June 11, 2018 3:58 AM
To: Dey, Souvik mailto:so...@rbbn.com>>; Wu, Jingjing 
mailto:jingjing...@intel.com>>; 
d...@dpdk.org
Cc: users@dpdk.org
Subject: RE: I40evf VLAN stripping disable


NOTICE: This email was received from an EXTERNAL sender


Hi Souvik,

Could you try with kernel driver (version 2.4.3) first? In my environment, 
disable vlan strip in DPDK works with kernel driver version 2.4.3.

My test steps with testpmd:
>set fwd rxonly
>set verbose 1
>set promisc 0 off
>vlan set strip off 0
Then send a vlan packet with VF mac address.

Best Regards,
Beilei Xing

From: Dey, Souvik [mailto:so...@rbbn.com]
Sent: Friday, June 8, 2018 11:22 PM
To: Xing, Beilei mailto:beilei.x...@intel.com>>; Wu, 
Jingjing mailto:jingjing...@intel.com>>; 
d...@dpdk.org
Cc: users@dpdk.org
Subject: I40evf VLAN stripping disable

Hi i40e/i40evf maintainers,

I was testing VLANs with i40evf pmd and is hitting the below issue.

I have the following configuration:
- host runs with Linux pf i40e driver(version 2.4.6)
- guest runs with DPDK 17.11.2 vf i40e driver

When I am sending a vlan packet from the outside to the guest, on the guest, I 
receive the packet which has the PKT_RX_VLAN_STRIPPED flag set although I'm not 
asking for it.
Even though my DPDK app has the rte_eth_conf.rxmode.hw_vlan_strip set to 0 and 
also verified that the 
i40evf_disable_vlan_strip(VIRTCHNL_OP_DISABLE_VLAN_STRIPPING) function is 
getting called and not returning any error from the linux pf driver.
Is this the default behavior that the VLAN will be always stripped by the PF 
irrespective of the setting ? Should the DPDK version take care of the 
re-adding the tag back to the packet in case hw_vlan_strip is disabled ? What 
should be best way of handling it ? Is it a bug in DPDK or I am missing 
something here.

I do see in the linux i40evf driver we are insert vlan header in the received 
packets in some cases.

--
Regards,
Souvik