Request: tester for ipq8064 soc

ok for some reason dsaconfig wasn't running on startup
I made sure there was no bridge for the lan (no br-lan)
ran dsaconfig apply
so far so good
wan port appears to be working and separated

root@OpenWrt:/# dsaconfig show
Switch: switch0
VLAN/ | lan1  | lan2  | lan3  | lan4  | wan   |
Link: | 1000F | down  | down  | down  | 1000F |
1000  |  u*   |  u*   |  u*   |  u*   |       |
2000  |       |       |       |       |  u*   |

[  410.261089] qca8k 37000000.mdio-mii:10 wan: Link is Up - 1Gbps/Full - flow control off
[  410.261171] IPv6: ADDRCONF(NETDEV_CHANGE): wan: link becomes ready
[  410.268401] switch0: port 5(wan) entered blocking state
[  410.274193] switch0: port 5(wan) entered forwarding state

19: switch0.2000@switch0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether xx:xx:xx:xx:xx:xx brd ff:ff:ff:ff:ff:ff
    inet 172.28.8.244/24 brd 172.28.8.255 scope global switch0.2000
       valid_lft forever preferred_lft forever

root@OpenWrt:/# ping 172.28.8.1
PING 172.28.8.1 (172.28.8.1): 56 data bytes
64 bytes from 172.28.8.1: seq=0 ttl=64 time=2.594 ms
64 bytes from 172.28.8.1: seq=1 ttl=64 time=1.549 ms
64 bytes from 172.28.8.1: seq=2 ttl=64 time=2.921 ms
64 bytes from 172.28.8.1: seq=3 ttl=64 time=2.817 ms
64 bytes from 172.28.8.1: seq=4 ttl=64 time=2.908 ms
64 bytes from 172.28.8.1: seq=5 ttl=64 time=2.503 ms
^C
--- 172.28.8.1 ping statistics ---
6 packets transmitted, 6 packets received, 0% packet loss
round-trip min/avg/max = 1.549/2.548/2.921 ms


It seems like as far as tagging goes, there may be a problem when trying to add more than one vlan to a port.

I want port 4 to carry tagged vlan 1000 and tagged vlan 4000 but only the 1000 tag appears to be running at full speed.

root@OpenWrt:/# dsaconfig show
Switch: switch0
VLAN/ | lan1  | lan2  | lan3  | lan4  | wan   |
Link: | down  | down  | 1000F | 1000F | 1000F |
1000  |  u*   |  u*   |       |  t    |       |
2000  |       |       |       |       |  u*   |
3500  |       |       |       |  u*   |       |
4000  |       |       |  u*   |  t    |       |


iperf3 -c 172.28.8.1
Connecting to host 172.28.8.1, port 5201
[  5] local 172.28.8.234 port 60790 connected to 172.28.8.1 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec  2.04 MBytes  17.1 Mbits/sec   49   1.43 KBytes       
[  5]   1.00-2.00   sec  0.00 Bytes  0.00 bits/sec    1   1.43 KBytes       
[  5]   2.00-3.00   sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
[  5]   3.00-4.00   sec  0.00 Bytes  0.00 bits/sec    1   1.43 KBytes       
[  5]   4.00-5.00   sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
[  5]   5.00-6.00   sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
[  5]   6.00-7.00   sec  0.00 Bytes  0.00 bits/sec    1   1.43 KBytes       
[  5]   7.00-8.00   sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
[  5]   8.00-9.00   sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
[  5]   9.00-10.00  sec  0.00 Bytes  0.00 bits/sec    0   1.43 KBytes       
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  2.04 MBytes  1.71 Mbits/sec   52             sender
[  5]   0.00-36.47  sec  1.67 MBytes   383 Kbits/sec                  receiver

iperf Done.

so I can't just generally test it for now it looks like, i'll have a think about if I did something wrong, gonna take a break for now

Can someone test with the patch 100 applied? (and the cpufreq driver)

I tested yesterday the cpufreq and mdio committs on a C2600 and it did not boot. My R7800 booted ok with the same committs.

did you applied the test commits? (the one that removes the regulator-boot-on)

Nope... I thought these test commits would disable the cpufreq related ones and therefore I skipped those. Please tell me which ones I have to use and can test them for you.

you should take this
TEST ipq806x: disable regulator for dsa testing · Ansuel/openwrt@803b788 · GitHub

And keep the patch 100

in theory it should boot and run normally.

Here you are @Ansuel , Logread from a C2600 with all patches applied EXCEPT the last one where you disable the 100 patch: https://1drv.ms/t/s!AlHg8pmWgJJ_gr54p6dRhzd6JQ11RQ?e=hVQENo

Yes, up and running now. Flashed last night, working and stable. Do you want the log?

yes pls...

@Ansuel: I pm'd you the log.

@ansuel - I noticed that wifi roaming is not smooth after I upgraded my C2600 to your DSA version. Do you think it could be related to your build?

shouldn't be related at all... anyway did you notice any crash or random port drop? (check the logread)

No crashes at all. And unless I missed something in the logs as it cycles quickly due to a lot of devices roaming all the time here at home, I didn't see any port drops either. With family mostly working/studying from home and no complaints from them lately, I guess it's safe to say the upgrade is rock solid.

By the way, how's the configuration of the switch with dsa? I had to upgrade without keeping previous settings (otherwise it would not upgrade), and the lan config was then configured as such:

config interface 'lan'
	option type 'bridge'
	option ifname 'lan1 lan2 lan3 lan4'
	option proto 'static'
	option ipaddr '192.168.1.1'
	option netmask '255.255.255.0'

There was no reference to a switch device in the network config file (I compiled without swconfig) and I found inconsistent info on how to setup VLANs and switch devices with DSA enabled firmwares in the openwrt community.

dsa support in openwrt is still wip and not so much documented... there is also a wip luci page that should add dsa support like it was done with swconfig.

can you give me output of these 2 command?
cat /sys/kernel/debug/regulator/regulator_summary and uptime

Here's the output. Unfortunately, I had a momentary loss of power at the house, so the uptime now is only 34min...

1 Like

the regulator system seems to be up and running...
I need to test this more since i notice some instability caused by that... (still don't know if it's random kernel crash or it's caused by the regulators)

Would it help to get the same from another C2600, or do you have all you need for now?

I need to test stability so I would really appreciate it if you can test this
In the next days I will push kernel 5.10 patches so more fun ahahahah

My VR2600 is still up after almost 5 days. I have 7 wired clients (some of them behind another switch), many wireless clients, about 200GB of data transferred (PPPoE) and no error showing in the dmesg.
I am using your branch including the last commit

Tell me if you need more info/logs/tests or do you want me to test your branch without the last commit?

can you build an image with this changes? so that we can test dsa + regulator?

also what i'm interested is port uptime... for example did you notice and disconnection ?