The FDP team is no longer accepting new bugs in Bugzilla. Please report your issues under FDP project in Jira. Thanks.
Bug 1729524 - [mlx5] PF to VF VGT Plus TX not offloaded
Summary: [mlx5] PF to VF VGT Plus TX not offloaded
Keywords:
Status: CLOSED EOL
Alias: None
Product: Red Hat Enterprise Linux Fast Datapath
Classification: Red Hat
Component: openvswitch2.11
Version: RHEL 7.7
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: ---
Assignee: Timothy Redaelli
QA Contact: qding
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2019-07-12 13:33 UTC by Christian Trautman
Modified: 2024-10-08 17:49 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2024-10-08 17:49:14 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker FD-119 0 None None None 2022-05-07 04:47:01 UTC

Description Christian Trautman 2019-07-12 13:33:19 UTC
***NOTE***

Opening this bug because it was not able to be verified in 19.D due to the firmware not being available.  Opening a new bug to target 19.E for verification and support of this fix.

*******************************************

This issue was discovered while testing MLNX_OFED version for Verizon. opening the BZ since same fix should be taken to RHEL-7.

Description of problem:
PF to VF VGT Plus TX not offloaded

Flow:
- Set PRIO_TAG_REQUIRED_EN=1 in the FW ( note that a special FW version is required, as of today it's not available on mellanox website yet).
- Bring up ASAP tunnel based system (VXLAN).
- Create VGT interface over one VF on machine A.
- Set IP address to VF over machine B with the same network

- Add the following rules to machine A:
ovs-ofctl add-flow ovs-sriov1 'in_port=vxlan0, actions=mod_vlan_vid:7, output:enp129s0f0_1'
ovs-ofctl add-flow ovs-sriov1 'in_port=enp129s0f0_1, actions=output:vxlan0'

- Add the following rules to machine B:
ovs-ofctl add-flow ovs-sriov1 'in_port=vxlan0, actions=strip_vlan, output:enp129s0f0_1'
ovs-ofctl add-flow ovs-sriov1 'in_port=enp129s0f0_1, actions=output:vxlan0'

- Strat bidir traffic.

Result:
RX rules are offloaded as expected.
TX rules is NOT offloaded.

ovs-dpctl dump-flows -m

2019-05-15T07:25:27Z|00001|dpif_netlink|INFO|The kernel module does not support meters.
ufid:639c0879-4047-4557-8703-0765039e4457, skb_priority(0/0),tunnel(tun_id=0x5,src=101.196.16.1,dst=101.196.15.1,ttl=0/0,tp_dst=4789,flags(+key)),skb_mark(0/0),in_port(vxlan_sys_4789),packet_type(ns=0/0,id=0/0),eth(src=00:00:00:00:00:00/00:00:00:00:00:00,dst=00:00:00:00:00:00/00:00:00:00:00:00),eth_type(0x8100),vlan(vid=7,pcp=0),encap(eth_type(0x0800),ipv4(src=0.0.0.0/0.0.0.0,dst=0.0.0.0/0.0.0.0,proto=0/0,tos=0/0,ttl=0/0,frag=no)), packets:13, bytes:1326, used:0.240s, offloaded:yes, dp:tc, actions:pop_vlan,enp129s0f0_1
ufid:3e3798b7-0b5e-4698-aa20-ac2a829031bc, skb_priority(0/0),skb_mark(0/0),in_port(enp129s0f0_1),packet_type(ns=0/0,id=0/0),eth(src=00:00:00:00:00:00/00:00:00:00:00:00,dst=00:00:00:00:00:00/00:00:00:00:00:00),eth_type(0x0800),ipv4(src=0.0.0.0/0.0.0.0,dst=0.0.0.0/0.0.0.0,proto=0/0,tos=0/0x3,ttl=0/0,frag=no), packets:0, bytes:0, used:13.300s, offloaded:yes, dp:tc, actions:set(tunnel(tun_id=0x5,src=101.196.15.1,dst=101.196.16.1,tp_dst=4789,flags(key))),vxlan_sys_4789
tcpdump -envi enp129s0f0_1

10:25:38.358111 e4:11:0f:81:00:03 > e4:11:10:81:00:03, ethertype 802.1Q (0x8100), length 102: vlan 0, p 0, ethertype IPv4, (tos 0x0, ttl 64, id 58675, offset 0, flags [DF], proto ICMP (1), length 84)
    20.196.15.1 > 20.196.16.1: ICMP echo request, id 17779, seq 25, length 64
10:25:39.358101 e4:11:0f:81:00:03 > e4:11:10:81:00:03, ethertype 802.1Q (0x8100), length 102: vlan 0, p 0, ethertype IPv4, (tos 0x0, ttl 64, id 59106, offset 0, flags [DF], proto ICMP (1), length 84)
    20.196.15.1 > 20.196.16.1: ICMP echo request, id 17779, seq 26, length 64
10:25:40.358123 e4:11:0f:81:00:03 > e4:11:10:81:00:03, ethertype 802.1Q (0x8100), length 102: vlan 0, p 0, ethertype IPv4, (tos 0x0, ttl 64, id 59500, offset 0, flags [DF], proto ICMP (1), length 84)
    20.196.15.1 > 20.196.16.1: ICMP echo request, id 17779, seq 27, length 64
10:25:41.358100 e4:11:0f:81:00:03 > e4:11:10:81:00:03, ethertype 802.1Q (0x8100), length 102: vlan 0, p 0, ethertype IPv4, (tos 0x0, ttl 64, id 60090, offset 0, flags [DF], proto ICMP (1), length 84)
    20.196.15.1 > 20.196.16.1: ICMP echo request, id 17779, seq 28, length 64

Comment 1 Marcelo Ricardo Leitner 2019-10-18 18:58:44 UTC
Hi Christian,
We should now have all we need to test this.
More info on bz1676558 and bz1758857.

Comment 2 Timothy Redaelli 2019-11-29 16:57:08 UTC
Hi Christian,
any news about that?

Probably it's already fixed since the 2 related bugzillas are already in the kernel

Comment 3 Christian Trautman 2019-12-02 17:55:53 UTC
I will ask Amit if he can test this.

Comment 4 qding 2019-12-10 06:52:24 UTC
Verified with kernel-3.10.0-1111.el7.x86_64 and openvswitch2.11-2.11.0-33.el7fdp.x86_64


ufid:3fa74f52-45f5-4aad-8695-170e40d45c6c, skb_priority(0/0),skb_mark(0/0),in_port(eth0),packet_type(ns=0/0,id=0/0),eth(src=00:de:ad:02:00:01,dst=00:de:ad:01:00:01),eth_type(0x86dd),ipv6(src=::/::,dst=::/::,label=0/0,proto=0/0,tclass=0/0x3,hlimit=0/0,frag=no), packets:10947019, bytes:16573618976, used:0.680s, offloaded:yes, dp:tc, actions:set(tunnel(tun_id=0x7b,dst=192.168.1.1,ttl=64,tp_dst=4789,flags(key))),push_vlan(vid=10,pcp=0),vxlan_sys_4789

ufid:804afd6d-66dd-4410-a5a1-d94448185e92, skb_priority(0/0),skb_mark(0/0),in_port(eth0),packet_type(ns=0/0,id=0/0),eth(src=00:de:ad:02:00:01,dst=00:de:ad:01:00:01),eth_type(0x0800),ipv4(src=0.0.0.0/0.0.0.0,dst=0.0.0.0/0.0.0.0,proto=0/0,tos=0/0x3,ttl=0/0,frag=no), packets:150, bytes:22800, used:0.680s, offloaded:yes, dp:tc, actions:set(tunnel(tun_id=0x7b,dst=192.168.1.1,ttl=64,tp_dst=4789,flags(key))),push_vlan(vid=10,pcp=0),vxlan_sys_4789

ufid:ff946828-6694-4457-96c0-56169cdb0411, skb_priority(0/0),tunnel(tun_id=0x7b,src=192.168.1.1,dst=192.168.1.2,ttl=0/0,tp_dst=4789,flags(+key)),skb_mark(0/0),in_port(vxlan_sys_4789),packet_type(ns=0/0,id=0/0),eth(src=00:de:ad:01:00:01,dst=00:de:ad:02:00:01),eth_type(0x8100),vlan(vid=10,pcp=0),encap(eth_type(0x86dd),ipv6(src=::/::,dst=::/::,label=0/0,proto=0/0,tclass=0/0,hlimit=0/0,frag=no)), packets:228811, bytes:19690610, used:0.680s, offloaded:yes, dp:tc, actions:pop_vlan,eth0

ufid:0428923f-4a75-42a2-8169-bfba635cfcff, skb_priority(0/0),tunnel(tun_id=0x7b,src=192.168.1.1,dst=192.168.1.2,ttl=0/0,tp_dst=4789,flags(+key)),skb_mark(0/0),in_port(vxlan_sys_4789),packet_type(ns=0/0,id=0/0),eth(src=00:de:ad:01:00:01,dst=00:de:ad:02:00:01),eth_type(0x8100),vlan(vid=10,pcp=0),encap(eth_type(0x0800),ipv4(src=0.0.0.0/0.0.0.0,dst=0.0.0.0/0.0.0.0,proto=0/0,tos=0/0,ttl=0/0,frag=no)), packets:150, bytes:14700, used:0.680s, offloaded:yes, dp:tc, actions:pop_vlan,eth0

Comment 6 ovs-bot 2024-10-08 17:49:14 UTC
This bug did not meet the criteria for automatic migration and is being closed.
If the issue remains, please open a new ticket in https://issues.redhat.com/browse/FDP


Note You need to log in before you can comment on or make changes to this bug.