TDA4VH-Q1: Validate EST functionality under native driver.

Hi,

The "tc filter" is completely software command and it does not have a callback to CPSW driver. Only "taprio qdisc" will make changes to the hardware.

Matthew said:

I have verified that the configuration mentioned above is correct. However, when I replace cpsw2g's eth0 with cpsw9g's eth2, the functionality of tc filter disappears.

Can you explain what do you mean when you say "functionality of tc filter disappears"? Can you still see the tc filter, or are the packets not being tagged with the correct priority?

Matthew said:

“tc filter add dev eth0 egress protocol 802.1Q flower vlan_id 11 action skbedit priority 3”

You are trying to match the vlan id with a priority for this. Is that correct?

Usually, this is not how it goes. With VLAN, the pcp priority field in the VLAN tag are mapped to the queue priority levels. This can be done by ip link command like following:

ip link set eth0.100 type vlan egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7"

Hope this points you in the correct direction.

Regards,
Tanmay

Hi，

Tanmay Patil said:

Can you explain what do you mean when you say "functionality of tc filter disappears"? Can you still see the tc filter, or are the packets not being tagged with the correct priority

yes, I can still see tc filters, but I’m unable to associate packets with priorities, it seems like they are all treated as the default priority.

Tanmay Patil said:

Usually, this is not how it goes. With VLAN, the pcp priority field in the VLAN tag are mapped to the queue priority levels. This can be done by ip link command like following:

ip link set eth0.100 type vlan egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7"

I will try this feature in mac mode, but how can I achieve the same functionality in switch mode?

Regards,

yrq

Hi,

Matthew said:

I will try this feature in mac mode, but how can I achieve the same functionality in switch mode?

It will be the same in switch mode with whatever vlan interface you are using.

Matthew said:

it seems like they are all treated as the default priority.

How are you confirming this?

Regards,
Tanmay

Hi，

Tanmay Patil said:

It will be the same in switch mode with whatever vlan interface you are using

in switch mode，I plan to configure as follows.（The basic commands are omitted.）

ip link add name br0 type bridge
ip link set dev br0 type bridge ageing_time 1000
ip link set dev eth1 master br0
ip link set dev eth2 master br0
ip link set dev eth3 master br0
ip link set dev eth4 master br0
ip link set dev br0 type bridge vlan_filtering 1
bridge vlan add dev br0 vid 1 pvid untagged self
bridge vlan add dev eth2 vid 11 master
bridge vlan add dev eth3 vid 12 master

tc qdisc replace dev eth2 parent root handle 100 taprio \
num_tc 8 \
map 1 0 2 3 4 5 6 7 0 0 0 0 0 0 0 0 \
queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 \
base-time $base \
sched-entry 64 2 250000 \
sched-entry 191 2 4750000 \
flags 2

In the given situation, which command should I use to configure the mapping between VLAN priority and traffic class?

Tanmay Patil said:

How are you confirming this?

I performed some verification based on the connectivity situation depicted in the following diagram.

ifconfig eth1 down 
ifconfig eth2 down
ifconfig eth3 down
ifconfig eth4 down
ethtool -L eth2 tx 4 
ethtool -L eth3 tx 4
ethtool --set-priv-flags eth0 p0-rx-ptype-rrobin off
ethtool --set-priv-flags eth3 p0-rx-ptype-rrobin off
devlink dev param set platform/c000000.ethernet name switch_mode value true cmode runtime
ip link add name br0 type bridge
ip link set dev br0 type bridge ageing_time 1000
ip link set dev eth2 up
ip link set dev eth3 up
ip link set dev eth2 master br0
ip link set dev eth3 master br0
ip link set dev br0 up; ip link set dev br0 type bridge vlan_filtering 1
bridge vlan add dev br0 vid 1 pvid untagged self


# setset gatecontrol
tc qdisc replace dev eth3 parent root handle 100 taprio \
    num_tc 3 map 0 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 queues 1@0 1@1 1@2 base-time 0 \
    sched-entry S 4 400000 sched-entry S 2 200000 sched-entry S 1 100000 flags 2


# set filter
tc qdisc add dev eth3 clsact
tc filter add dev eth3 egress protocol ip prio 1 u32 match ip dport 5003 0xffff action skbedit priority 3

when I send udp port 5003 packet，the tc filter doesn't work，

but when I replaced eth2 with eth0 of cpsw2g, this functionality started working properly.

So, I suspect that it might be because the ports are on the same switch, and they are directly doing MAC forwarding without going through the filtering step.

Regards,

yrq