Sqm worse at latest firmware on redmi ax6000

so nothing can be done about it now? where do I report this kernel problem? @moeller0

Hard to say, I am using turrisOS which is lagging behind upstream OpenWrt by two full versions IIRC so I can not really test and expect the same results as you do. However:

Seems to indicate that for eth0 packet steering is activated for all 4 CPUs, yet this does not seem to be the case for wan an ifb4wan, but I am not 100% sure it should.

Could you please do the following:

  1. tc -s qdisc # then post the output here
  2. cat /proc/softirqs # then post the output
  3. a speedtest https://speed.cloudflare.com # then post a screenshot of the results
  4. tc -s qdisc # then post the output here
  5. cat /proc/softirqs # then post the output her

The question is, do we still see signs of all processing being stuck on CPU0?

root@OpenWrt:~# tc -s qdisc
qdisc noqueue 0: dev lo root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc mq 0: dev eth0 root
 Sent 6696301329 bytes 13388493 pkt (dropped 0, overlimits 0 requeues 3551)
 backlog 0b 0p requeues 3551
qdisc fq_codel 0: dev eth0 parent :10 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :f limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :e limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :d limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :c limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :b limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :a limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :9 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :8 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1255428208 bytes 7057553 pkt (dropped 0, overlimits 0 requeues 1864)
 backlog 0b 0p requeues 1864
  maxpacket 15631 drop_overlimit 0 new_flow_count 1662 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :7 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 14464115 bytes 87200 pkt (dropped 0, overlimits 0 requeues 42)
 backlog 0b 0p requeues 42
  maxpacket 1279 drop_overlimit 0 new_flow_count 68 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :6 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 4360361332 bytes 4566338 pkt (dropped 0, overlimits 0 requeues 1314)
 backlog 0b 0p requeues 1314
  maxpacket 15824 drop_overlimit 0 new_flow_count 9842 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :5 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1066045962 bytes 1677386 pkt (dropped 0, overlimits 0 requeues 331)
 backlog 0b 0p requeues 331
  maxpacket 12144 drop_overlimit 0 new_flow_count 153 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :4 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1712 bytes 16 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :3 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :2 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :1 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc noqueue 0: dev lan4 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan3 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan2 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc htb 1: dev wan root refcnt 2 r2q 10 default 0x12 direct_packets_stat 0 direct_qlen 1000
 Sent 1518505705 bytes 7010712 pkt (dropped 0, overlimits 2885 requeues 0)
 backlog 0b 0p requeues 0
qdisc fq_codel 120: dev wan parent 1:12 limit 1001p flows 1024 quantum 300 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1516759860 bytes 7002542 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 61534 drop_overlimit 0 new_flow_count 1339663 ecn_mark 0
  new_flows_len 1 old_flows_len 4
qdisc fq_codel 130: dev wan parent 1:13 limit 1001p flows 1024 quantum 300 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1745845 bytes 8170 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 634 drop_overlimit 0 new_flow_count 1576 ecn_mark 0
  new_flows_len 1 old_flows_len 7
qdisc fq_codel 110: dev wan parent 1:11 limit 1001p flows 1024 quantum 300 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc ingress ffff: dev wan parent ffff:fff1 ----------------
 Sent 25362729920 bytes 21191674 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev br-lan root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc htb 1: dev ifb4wan root refcnt 2 r2q 10 default 0x10 direct_packets_stat 0 direct_qlen 32
 Sent 26151383900 bytes 21190705 pkt (dropped 321, overlimits 445824 requeues 0)
 backlog 0b 0p requeues 0
qdisc fq_codel 110: dev ifb4wan parent 1:10 limit 1001p flows 1024 quantum 1514 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 26151383900 bytes 21190705 pkt (dropped 321, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 58030 drop_overlimit 63 new_flow_count 1133782 ecn_mark 0 drop_overmemory 63
  new_flows_len 1 old_flows_len 17
qdisc noqueue 0: dev phy0-ap0 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev phy1-ap0 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
root@OpenWrt:~# cat /proc/softirqs
                    CPU0       CPU1       CPU2       CPU3
          HI:          0          0          0          0
       TIMER:     891070    1373315    3194082    1841972
      NET_TX:     105399     123910     150869      82319
      NET_RX:   13492677    5315365    7572900   10017819
       BLOCK:          0          0          0          0
    IRQ_POLL:          0          0          0          0
     TASKLET:   50253439   11014473    6993591    3150492
       SCHED:    1997103    2181214    3923276    2247567
     HRTIMER:          0          0          0          0
         RCU:    1110155     985481    1248296     797495

root@OpenWrt:~# tc -s qdisc
qdisc noqueue 0: dev lo root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc mq 0: dev eth0 root
 Sent 12019822249 bytes 18656943 pkt (dropped 0, overlimits 0 requeues 4644)
 backlog 0b 0p requeues 4644
qdisc fq_codel 0: dev eth0 parent :10 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :f limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :e limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :d limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :c limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :b limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :a limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :9 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :8 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 2921618036 bytes 9515233 pkt (dropped 0, overlimits 0 requeues 2300)
 backlog 0b 0p requeues 2300
  maxpacket 17420 drop_overlimit 0 new_flow_count 2006 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :7 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 14532909 bytes 87368 pkt (dropped 0, overlimits 0 requeues 47)
 backlog 0b 0p requeues 47
  maxpacket 1279 drop_overlimit 0 new_flow_count 68 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :6 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 8017623630 bytes 7376940 pkt (dropped 0, overlimits 0 requeues 1966)
 backlog 0b 0p requeues 1966
  maxpacket 15824 drop_overlimit 0 new_flow_count 11802 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :5 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1066045962 bytes 1677386 pkt (dropped 0, overlimits 0 requeues 331)
 backlog 0b 0p requeues 331
  maxpacket 12144 drop_overlimit 0 new_flow_count 153 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :4 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 1712 bytes 16 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :3 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :2 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc fq_codel 0: dev eth0 parent :1 limit 10240p flows 1024 quantum 1518 target 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc noqueue 0: dev lan4 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan3 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan2 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc cake 8023: dev wan root refcnt 2 bandwidth 200Mbit besteffort triple-isolate nonat nowash no-ack-filter split-gso rtt 100ms raw overhead 0
 Sent 587046758 bytes 1039570 pkt (dropped 116, overlimits 438417 requeues 0)
 backlog 0b 0p requeues 0
 memory used: 2091222b of 10000000b
 capacity estimate: 200Mbit
 min/max network layer size:           42 /    1514
 min/max overhead-adjusted size:       42 /    1514
 average network hdr offset:           14

                  Tin 0
  thresh        200Mbit
  target            5ms
  interval        100ms
  pk_delay        443us
  av_delay        121us
  sp_delay          1us
  backlog            0b
  pkts          1039686
  bytes       587222062
  way_inds       146856
  way_miss          584
  way_cols            0
  drops             116
  marks               0
  ack_drop            0
  sp_flows            1
  bk_flows            1
  un_flows            0
  max_len         45888
  quantum          1514

qdisc ingress ffff: dev wan parent ffff:fff1 ----------------
 Sent 1661234244 bytes 1274374 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev br-lan root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev phy0-ap0 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev phy1-ap0 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc cake 8024: dev ifb4wan root refcnt 2 bandwidth 400Mbit besteffort triple-isolate nonat wash no-ack-filter split-gso rtt 100ms raw overhead 0
 Sent 1719637398 bytes 1271799 pkt (dropped 2575, overlimits 1384908 requeues 0)
 backlog 0b 0p requeues 0
 memory used: 2136832b of 15140Kb
 capacity estimate: 400Mbit
 min/max network layer size:           60 /    1514
 min/max overhead-adjusted size:       60 /    1514
 average network hdr offset:           14

                  Tin 0
  thresh        400Mbit
  target            5ms
  interval        100ms
  pk_delay        164us
  av_delay         46us
  sp_delay          1us
  backlog            0b
  pkts          1274374
  bytes      1723528828
  way_inds        13209
  way_miss          546
  way_cols            0
  drops            2575
  marks               0
  ack_drop            0
  sp_flows            2
  bk_flows            1
  un_flows            0
  max_len         43020
  quantum          1514
root@OpenWrt:~# cat /proc/softirqs
                    CPU0       CPU1       CPU2       CPU3
          HI:          0          0          0          0
       TIMER:     897099    1380656    3215848    1852740
      NET_TX:    1183085    1082337     618193     386641
      NET_RX:   14513342    6164830    8215635   10530253
       BLOCK:          0          0          0          0
    IRQ_POLL:          0          0          0          0
     TASKLET:   51224756   11827874    7286389    3460186
       SCHED:    2010125    2195933    3949574    2261858
     HRTIMER:          0          0          0          0
         RCU:    1115357     991743    1255128     802494

Thanks, the first tc -s qdisc was taken with someting else than cake (I see two HTB instances and fq_codels so this might have been simplest.qos/fq_codel or simple.qos/fq_codel).
That brings up the question, did you change to cake before or after the speedtest?

What we need is cake (and /proc/softirqs) from before and after a speedtest...

That said, the speedtest does not look terrible, sure a bit below the specified 400/200 rates.
If that speedtest was with simple[st].qos then I would guess 400/200 is simply too much for your router in its current configuration...

Before, I left it in fq_codel before because it was better.

Before...

root@OpenWrt:~# cat /proc/softirqs
                    CPU0       CPU1       CPU2       CPU3
          HI:          0          0          0          0
       TIMER:     930077    1426319    3320479    1910095
      NET_TX:    1465088    1783251     719446    1240059
      NET_RX:   15183930    6847239    8577322   11365757
       BLOCK:          0          0          0          0
    IRQ_POLL:          0          0          0          0
     TASKLET:   53158607   12726327    7648190    4423831
       SCHED:    2086500    2264958    4087600    2336348
     HRTIMER:          0          0          0          0
         RCU:    1158662    1024969    1307280     834664

after...

root@OpenWrt:~# cat /proc/softirqs
                    CPU0       CPU1       CPU2       CPU3
          HI:          0          0          0          0
       TIMER:     931485    1428878    3326868    1913206
      NET_TX:    1796406    1805071     740925    1363383
      NET_RX:   15391618    6892965    8635267   11549969
       BLOCK:          0          0          0          0
    IRQ_POLL:          0          0          0          0
     TASKLET:   53566763   12757229    7683851    4457431
       SCHED:    2090419    2269202    4095833    2340808
     HRTIMER:          0          0          0          0
         RCU:    1160861    1026882    1310285     836633

This still does not look at that bad, at least latency under load seems decently in check.

BTW, sqm shaper rates are gross rates so for IPv6 you can expect at most:

400 * ((1500 - 40 - 20) / (1500 +14)) = 380.45 Mbps
200 * ((1500 - 40 - 20) / (1500 +14)) = 190.22 Mbps

So what is your verdict when you actually use your network as you planned to do?

BTW, I think we can improve isolation of different local machines a bit more, but maybe try to test with your actual normal load to figure out whether sqm works as expected or not....

I just want to avoid bufferbloat completely, in the original firmware everything seemed fine in the test.

with oem firmware last year

now with fq_codel simple

Download results are between +1 and +3 with each test. I think it's improved even though it's not perfect.