DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: Eelco Chaudron <echaudro@redhat.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] VF of a X520 card does not process VLAN packets
Date: Wed, 23 Jan 2019 14:09:10 +0000	[thread overview]
Message-ID: <2601191342CEEE43887BDE71AB977258010D9070EB@irsmsx105.ger.corp.intel.com> (raw)
In-Reply-To: <2601191342CEEE43887BDE71AB977258010D9070D3@irsmsx105.ger.corp.intel.com>

Hi Eelco,

> 
> 
> Hi Konstantin,
> Thanks for your reply…
> Yes adding the VLAN to the VF from the kernel side will work, however, this is a different behaviour than the XL710.
> The XL710 will allow all VLANs to pass through once the PMD takes control of the VF.

It is not about who controls VF.
On ixgbe VLAN filter are configured and controlled by PF.
In your case PF is controlled by kernel ixgbe driver.
AFAIK, in SRIOV mode ixgbe kernel driver always enables VLAN filtering,
and there is not possible to revert it (at least I don't know how).
Konstantin 

> The real use case is using this VF interface trough Open vSwitch which needs to see all packets regardless of the VLAN.
> It does not make sense to create 4K VLANs through the kernel to get this working.
> Looking forward to your reply…
> Cheers,
> Eelco
> 
> On 18 Jan 2019, at 12:25, Ananyev, Konstantin wrote:
> Hi
> It’s been while since I looked at it last time,
> but shouldn’t you assign desired vlan tag to the VF first:
> ip link set <device> vf <num> vlan <num>
> ?
> Konstantin
> 
> From: Eelco Chaudron [mailto:echaudro@redhat.com]
> Sent: Thursday, January 3, 2019 4:42 PM
> To: Lu, Wenzhuo <wenzhuo.lu@intel.com>; Ananyev, Konstantin <konstantin.ananyev@intel.com>
> Cc: dev <dev@dpdk.org>
> Subject: Re: [dpdk-dev] VF of a X520 card does not process VLAN packets
> 
> Hi maintainers, any feedback on the below?
> Thanks,
> Eelco
> On 18 Dec 2018, at 12:06, Eelco Chaudron wrote:
> Hi,
> I’m assigning a VF of an X520 card for DPDK/testpmd/OVS but it's not accepting tagged VLAN packets (it does accept tag 0). Is this a known
> bug/limitation of the 82599ES chipset?
> This is how I've tested it:
> $ echo 1 > /sys/bus/pci/devices/0000:05:00.0/sriov_numvfs
> $driverctl -v set-override 0000:05:10.0 vfio-pci
> $./testpmd -c 7 -n 4 --socket-mem 2048,0 -w 0000:05:10.0 -- \
> --burst 64 -i --rxq=1 --txq=1 --rxd=4096 --txd=1024 \
> —coremask=6 --auto-start --port-topology=chained
> EAL: Detected 28 lcore(s)
> EAL: Detected 1 NUMA nodes
> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
> ...
> testpmd>
> I’m sending broadcast ARP packets:
> Ethernet II, Src: 04:f4:bc:43:e1:00 (04:f4:bc:43:e1:00), Dst: ff:ff:ff:ff:ff:ff (ff:ff:ff:ff:ff:ff)
> Destination: ff:ff:ff:ff:ff:ff (ff:ff:ff:ff:ff:ff)
> Address: ff:ff:ff:ff:ff:ff (ff:ff:ff:ff:ff:ff)
> .... ..1. .... .... .... .... = LG bit: Locally administered address (this is NOT the factory default)
> .... ...1 .... .... .... .... = IG bit: Group address (multicast/broadcast)
> Source: 04:f4:bc:43:e1:00 (04:f4:bc:43:e1:00)
> Address: 04:f4:bc:43:e1:00 (04:f4:bc:43:e1:00)
> .... ..0. .... .... .... .... = LG bit: Globally unique address (factory default)
> .... ...0 .... .... .... .... = IG bit: Individual address (unicast)
> Type: 802.1Q Virtual LAN (0x8100)
> 802.1Q Virtual LAN, PRI: 0, CFI: 0, ID: 10
> 000. .... .... .... = Priority: Best Effort (default) (0)
> ...0 .... .... .... = CFI: Canonical (0)
> .... 0000 0000 1010 = ID: 10
> Type: ARP (0x0806)
> Padding: 2e2f303132333435363738393a3b
> Trailer: 3c3d3e3f404142434445464748494a4b4c4d4e4f50515253...
> Address Resolution Protocol (request/gratuitous ARP)
> Hardware type: Ethernet (1)
> Protocol type: IP (0x0800)
> Hardware size: 6
> Protocol size: 4
> Opcode: request (1)
> [Is gratuitous: True]
> Sender MAC address: 00:00:00:00:00:00 (00:00:00:00:00:00)
> Sender IP address: 0.0.0.0 (0.0.0.0)
> Target MAC address: 00:00:00:00:00:00 (00:00:00:00:00:00)
> Target IP address: 0.0.0.0 (0.0.0.0)
> No traffic is received, if you replace tag 10 with 0, packets are received.
> Note that this is a simple reproducer with testpmd, but it's reported as part of an OVS integration. This works fine on the physical port, or
> with a VF on an XL710 card.
> Any input appreciated.
> Thanks,
> Eelco

       reply	other threads:[~2019-01-23 14:09 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <2601191342CEEE43887BDE71AB977258010D9070D3@irsmsx105.ger.corp.intel.com>
2019-01-23 14:09 ` Ananyev, Konstantin [this message]
2019-01-29  9:30   ` Zhao1, Wei
2018-12-18 11:06 Eelco Chaudron
2019-01-03 16:42 ` Eelco Chaudron
2019-01-18 11:25   ` Ananyev, Konstantin
2019-01-22 13:33     ` Eelco Chaudron

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=2601191342CEEE43887BDE71AB977258010D9070EB@irsmsx105.ger.corp.intel.com \
    --to=konstantin.ananyev@intel.com \
    --cc=dev@dpdk.org \
    --cc=echaudro@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).