Bug 200879

Summary: Poor network performance using CX-5 Mellanox card
Product: Networking Reporter: kolga
Component: IPV4Assignee: Stephen Hemminger (stephen)
Status: NEW ---    
Severity: normal CC: josh
Priority: P1    
Hardware: Intel   
OS: Linux   
Kernel Version: 4.18 Subsystem:
Regression: No Bisected commit-id:

Description kolga 2018-08-21 18:37:01 UTC 
I'm having issues with the latest kernel (4.18) and using Mellanox CX-5 cards. Tested using a direct connection between the machines and via a 40G link. There is an asymmetric throughput.

I have tested some kernels in between 4.15-rc4 and 4.18. What I notice is that I have asymmetric flow performance where "good" direction gets 20+G and the bad
direction gets 2+G. I measure performance over doing multiple runs (10). I have noticed that while 4.15-rc4 was not perfect at getting symmetric 20+G performance all the time (details below), the poor performance is more prevalent starting from 4.16 kernel.

In 4.15-rc4 6 out of 10 runs show good performance 20+G (in the bad
direction). performance in other direction is mostly 28+G (7 out of 10
runs. where 3 runs it goes down to 15+G)
In 4.16 3 out of 10 runs show good performance 20+G. performance in
other direction is mostly 20+G (7 out of 10 runs where 3 runs goes
down to 15G)
In 4.17 0 out of 10 runs show good performance. performance in other
direction is mostly 20+G (7 out of 10 runs where 3 runs it goes down
to 5G)
In 4.18 0 out of 10 runs show good performance in the "bad
direction". the "good direction" is now also pretty bad where 7 out of
10 runs had throughput between 7-10G and 3 runs had 11-17G.

Let me try to do with a description of the setup. I have 2machines:
sti-rx200-231 and sti-rx200-232. I have been running iperf3 testing
bandwidth where at first the server was on -231 and then when the
server was on -232. The "bad" direction is when -232 is the server and
-231 is the client.

> can you share some configuration details:
> CPU numa and affinity details:

[kolga@sti-rx200-232 ~]$ numactl -H
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 12 13 14 15 16 17
node 0 size: 32155 MB
node 0 free: 30394 MB
node 1 cpus: 6 7 8 9 10 11 18 19 20 21 22 23
node 1 size: 32231 MB
node 1 free: 30486 MB
node distances:
node   0   1
  0:  10  20
  1:  20  10

[kolga@sti-rx200-231 ~]$ numactl -H
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 12 13 14 15 16 17
node 0 size: 32161 MB
node 0 free: 31100 MB
node 1 cpus: 6 7 8 9 10 11 18 19 20 21 22 23
node 1 size: 32231 MB
node 1 free: 29774 MB
node distances:
node   0   1
  0:  10  20
  1:  20  10

> ethtool -l
[kolga@sti-rx200-231 ~]$ sudo ethtool -l enp4s0
Channel parameters for enp4s0:
Pre-set maximums:
RX:             0
TX:             0
Other:          0
Combined:       24
Current hardware settings:
RX:             0
TX:             0
Other:          0
Combined:       24

(same for the other machine)

> ethtool -g
[kolga@sti-rx200-231 ~]$ sudo ethtool -g enp4s0
Ring parameters for enp4s0:
Pre-set maximums:
RX:             8192
RX Mini:        0
RX Jumbo:       0
TX:             8192
Current hardware settings:
RX:             1024
RX Mini:        0
RX Jumbo:       0
TX:             1024

(same for the other machine)

> ethtool -x
[kolga@sti-rx200-231 ~]$ sudo ethtool -x enp4s0
RX flow hash indirection table for enp4s0 with 24 RX ring(s):
    0:      0     1     2     3     4     5     6     7
    8:      8     9    10    11    12    13    14    15
   16:     16    17    18    19    20    21    22    23
   24:      0     1     2     3     4     5     6     7
   32:      8     9    10    11    12    13    14    15
   40:     16    17    18    19    20    21    22    23
   48:      0     1     2     3     4     5     6     7
   56:      8     9    10    11    12    13    14    15
   64:     16    17    18    19    20    21    22    23
   72:      0     1     2     3     4     5     6     7
   80:      8     9    10    11    12    13    14    15
   88:     16    17    18    19    20    21    22    23
   96:      0     1     2     3     4     5     6     7
  104:      8     9    10    11    12    13    14    15
  112:     16    17    18    19    20    21    22    23
  120:      0     1     2     3     4     5     6     7
RSS hash key:
5e:1c:93:e2:ec:b6:44:b1:e4:ec:b1:20:57:ab:90:f6:0c:1a:46:13:b8:19:66:c8:56:0c:06:b2:d5:53:a6:4d:89:6b:0b:b1:d4:30:90:31

(same for the other machine but the RSS hash key is different)

> ethtool -k
[kolga@sti-rx200-231 ~]$ sudo ethtool -k enp4s0
Features for enp4s0:
Cannot get device udp-fragmentation-offload settings: Operation not supported
rx-checksumming: on
tx-checksumming: on
        tx-checksum-ipv4: on
        tx-checksum-ip-generic: off [fixed]
        tx-checksum-ipv6: on
        tx-checksum-fcoe-crc: off [fixed]
        tx-checksum-sctp: off [fixed]
scatter-gather: on
        tx-scatter-gather: on
        tx-scatter-gather-fraglist: off [fixed]
tcp-segmentation-offload: on
        tx-tcp-segmentation: on
        tx-tcp-ecn-segmentation: off [fixed]
        tx-tcp-mangleid-segmentation: off
        tx-tcp6-segmentation: on
udp-fragmentation-offload: off
generic-segmentation-offload: on
generic-receive-offload: on
large-receive-offload: off
rx-vlan-offload: on
tx-vlan-offload: on
ntuple-filters: off
receive-hashing: on
highdma: on [fixed]
rx-vlan-filter: on
vlan-challenged: off [fixed]
tx-lockless: off [fixed]
netns-local: off [fixed]
tx-gso-robust: off [fixed]
tx-fcoe-segmentation: off [fixed]
tx-gre-segmentation: on
tx-gre-csum-segmentation: on
tx-ipxip4-segmentation: off [fixed]
tx-ipxip6-segmentation: off [fixed]
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-gso-partial: on
tx-sctp-segmentation: off [fixed]
tx-esp-segmentation: off [fixed]
tx-udp-segmentation: off [fixed]
fcoe-mtu: off [fixed]
tx-nocache-copy: off
loopback: off [fixed]
rx-fcs: off
rx-all: off
tx-vlan-stag-hw-insert: on
rx-vlan-stag-hw-parse: off [fixed]
rx-vlan-stag-filter: on [fixed]
l2-fwd-offload: off [fixed]
hw-tc-offload: off
esp-hw-offload: off [fixed]
esp-tx-csum-hw-offload: off [fixed]
rx-udp_tunnel-port-offload: on
tls-hw-tx-offload: off [fixed]
rx-gro-hw: off [fixed]
tls-hw-record: off [fixed]

(i think it's the same for the other machine but just in case here's the output)
[kolga@sti-rx200-232 ~]$ sudo ethtool -k enp4s0
Features for enp4s0:
Cannot get device udp-fragmentation-offload settings: Operation not supported
rx-checksumming: on
tx-checksumming: on
        tx-checksum-ipv4: on
        tx-checksum-ip-generic: off [fixed]
        tx-checksum-ipv6: on
        tx-checksum-fcoe-crc: off [fixed]
        tx-checksum-sctp: off [fixed]
scatter-gather: on
        tx-scatter-gather: on
        tx-scatter-gather-fraglist: off [fixed]
tcp-segmentation-offload: on
        tx-tcp-segmentation: on
        tx-tcp-ecn-segmentation: off [fixed]
        tx-tcp-mangleid-segmentation: off
        tx-tcp6-segmentation: on
udp-fragmentation-offload: off
generic-segmentation-offload: on
generic-receive-offload: on
large-receive-offload: off
rx-vlan-offload: on
tx-vlan-offload: on
ntuple-filters: off
receive-hashing: on
highdma: on [fixed]
rx-vlan-filter: on
vlan-challenged: off [fixed]
tx-lockless: off [fixed]
netns-local: off [fixed]
tx-gso-robust: off [fixed]
tx-fcoe-segmentation: off [fixed]
tx-gre-segmentation: on
tx-gre-csum-segmentation: on
tx-ipxip4-segmentation: off [fixed]
tx-ipxip6-segmentation: off [fixed]
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-gso-partial: on
tx-sctp-segmentation: off [fixed]
tx-esp-segmentation: off [fixed]
tx-udp-segmentation: off [fixed]
fcoe-mtu: off [fixed]
tx-nocache-copy: off
loopback: off [fixed]
rx-fcs: off
rx-all: off
tx-vlan-stag-hw-insert: on
rx-vlan-stag-hw-parse: off [fixed]
rx-vlan-stag-filter: on [fixed]
l2-fwd-offload: off [fixed]
hw-tc-offload: off
esp-hw-offload: off [fixed]
esp-tx-csum-hw-offload: off [fixed]
rx-udp_tunnel-port-offload: on
tls-hw-tx-offload: off [fixed]
rx-gro-hw: off [fixed]
tls-hw-record: off [fixed]

> ethtool --show-priv-flags
[kolga@sti-rx200-231 ~]$ sudo ethtool --show-priv-flags enp4s0
Private flags for enp4s0:
rx_cqe_moder   : on
tx_cqe_moder   : off
rx_cqe_compress: off
rx_striding_rq : on

(same for the other machine)

> ethool -S //before and after the good and bad runs"
> perf report/top while running the test.

This is a run where -232 is a server and -231 is a client.

[kolga@sti-rx200-232 src]$ ./iperf3 -s
-----------------------------------------------------------
Server listening on 5201
-----------------------------------------------------------
Accepted connection from 172.20.35.189, port 37302
[  5] local 172.20.35.191 port 5201 connected to 172.20.35.189 port 37304
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.00   sec   236 MBytes  1.98 Gbits/sec
[  5]   1.00-2.00   sec   233 MBytes  1.95 Gbits/sec
[  5]   2.00-3.00   sec   235 MBytes  1.97 Gbits/sec
[  5]   3.00-4.00   sec   231 MBytes  1.94 Gbits/sec
[  5]   4.00-5.00   sec   243 MBytes  2.04 Gbits/sec
[  5]   5.00-6.00   sec   238 MBytes  1.99 Gbits/sec
[  5]   6.00-7.00   sec   230 MBytes  1.93 Gbits/sec
[  5]   7.00-8.00   sec   232 MBytes  1.94 Gbits/sec
[  5]   8.00-9.00   sec   272 MBytes  2.28 Gbits/sec
[  5]   9.00-10.00  sec   249 MBytes  2.09 Gbits/sec
[  5]  10.00-10.05  sec  10.4 MBytes  1.90 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-10.05  sec  2.35 GBytes  2.01 Gbits/sec                  receiver

[kolga@sti-rx200-231 src]$ sudo ./iperf3 -c 172.20.35.191
Connecting to host 172.20.35.191, port 5201
[  5] local 172.20.35.189 port 37304 connected to 172.20.35.191 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec   249 MBytes  2.09 Gbits/sec    4   1.93 MBytes
[  5]   1.00-2.00   sec   232 MBytes  1.95 Gbits/sec    1   1.69 MBytes
[  5]   2.00-3.00   sec   234 MBytes  1.96 Gbits/sec    0   2.24 MBytes
[  5]   3.00-4.00   sec   232 MBytes  1.95 Gbits/sec    0   2.66 MBytes
[  5]   4.00-5.00   sec   242 MBytes  2.03 Gbits/sec   18   1.32 MBytes
[  5]   5.00-6.00   sec   238 MBytes  1.99 Gbits/sec    1   1.63 MBytes
[  5]   6.00-7.00   sec   230 MBytes  1.93 Gbits/sec    0   2.18 MBytes
[  5]   7.00-8.00   sec   232 MBytes  1.95 Gbits/sec    8   1.38 MBytes
[  5]   8.00-9.00   sec   272 MBytes  2.29 Gbits/sec    2   1.45 MBytes
[  5]   9.00-10.00  sec   248 MBytes  2.08 Gbits/sec    2   1.66 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  2.35 GBytes  2.02 Gbits/sec   36             sender
[  5]   0.00-10.05  sec  2.35 GBytes  2.01 Gbits/sec                  receiver

iperf Done.

Top output from -231
Tasks: 412 total,   1 running, 209 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.0 us,  0.3 sy,  0.0 ni, 99.2 id,  0.0 wa,  0.0 hi,  0.4 si,  0.0 st
KiB Mem : 65938632 total, 63195184 free,  2191948 used,   551500 buff/cache
KiB Swap: 33030140 total, 33030140 free,        0 used. 63105388 avail Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
  162 root      20   0       0      0      0 I   3.3  0.0   0:00.51 kworker/12+
 2530 root      20   0       0      0      0 I   1.7  0.0   0:00.39 kworker/u4+
 2742 root      20   0       0      0      0 I   1.3  0.0   0:00.17 kworker/u4+
 2394 root      20   0       0      0      0 I   1.0  0.0   0:00.16 kworker/17+
 2774 kolga     20   0  158000   4640   3676 R   0.3  0.0   0:00.03 top
    1 root      20   0  191980   6436   3900 S   0.0  0.0   0:26.21 systemd
    2 root      20   0       0      0      0 S   0.0  0.0   0:00.01 kthreadd
    3 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_gp
    4 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_par_gp
    5 root      20   0       0      0      0 I   0.0  0.0   0:00.02 kworker/0:+
    6 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 kworker/0:+
    8 root      20   0       0      0      0 I   0.0  0.0   0:00.00 kworker/u4+
    9 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 mm_percpu_+
   10 root      20   0       0      0      0 S   0.0  0.0   0:00.00 ksoftirqd/0
   11 root      20   0       0      0      0 I   0.0  0.0   0:00.17 rcu_sched
   12 root      20   0       0      0      0 I   0.0  0.0   0:00.00 rcu_bh
   13 root      rt   0       0      0      0 S   0.0  0.0   0:00.06 migration/0

Output from -232
Tasks: 400 total,   2 running, 202 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.1 us,  1.5 sy,  0.0 ni, 97.3 id,  0.0 wa,  0.0 hi,  1.0 si,  0.0 st
KiB Mem : 65932576 total, 63205228 free,  2180492 used,   546856 buff/cache
KiB Swap: 28901372 total, 28901372 free,        0 used. 63114220 avail Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 2467 kolga     20   0   43716   4144   3608 R  37.9  0.0   0:03.32 lt-iperf3
 1199 root      20   0       0      0      0 D   2.7  0.0   0:00.18 kworker/23+
  565 root      20   0       0      0      0 I   1.7  0.0   0:00.23 kworker/2:+
 2360 root      20   0       0      0      0 I   1.3  0.0   0:00.16 kworker/u4+
  153 root      20   0       0      0      0 I   0.7  0.0   0:00.11 kworker/23+
 2273 root      20   0       0      0      0 I   0.7  0.0   0:01.13 kworker/u4+
  691 root      20   0       0      0      0 S   0.3  0.0   0:00.20 xfsaild/dm+
 2448 root      20   0       0      0      0 I   0.3  0.0   0:00.08 kworker/u4+
    1 root      20   0  191820   6108   3804 S   0.0  0.0   0:26.12 systemd
    2 root      20   0       0      0      0 S   0.0  0.0   0:00.01 kthreadd
    3 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_gp
    4 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_par_gp
    6 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 kworker/0:+
    9 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 mm_percpu_+
   10 root      20   0       0      0      0 S   0.0  0.0   0:00.00 ksoftirqd/0
   11 root      20   0       0      0      0 I   0.0  0.0   0:00.13 rcu_sched
   12 root      20   0       0      0      0 I   0.0  0.0   0:00.00 rcu_bh


Here's a run where -231 is the server and -232 is the client. A "good" direction
[kolga@sti-rx200-231 src]$ ./iperf3 -s
-----------------------------------------------------------
Server listening on 5201
-----------------------------------------------------------
Accepted connection from 172.20.35.191, port 35060
[  5] local 172.20.35.189 port 5201 connected to 172.20.35.191 port 35062
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.00   sec  1.81 GBytes  15.5 Gbits/sec
[  5]   1.00-2.00   sec  1.90 GBytes  16.3 Gbits/sec
[  5]   2.00-3.00   sec  1.25 GBytes  10.8 Gbits/sec
[  5]   3.00-4.00   sec   826 MBytes  6.93 Gbits/sec
[  5]   4.00-5.00   sec   819 MBytes  6.87 Gbits/sec
[  5]   5.00-6.00   sec  1.47 GBytes  12.6 Gbits/sec
[  5]   6.00-7.00   sec  1.79 GBytes  15.4 Gbits/sec
[  5]   7.00-8.00   sec  1.14 GBytes  9.75 Gbits/sec
[  5]   8.00-9.00   sec  1.81 GBytes  15.6 Gbits/sec
[  5]   9.00-10.00  sec  1.77 GBytes  15.2 Gbits/sec
[  5]  10.00-10.04  sec  78.2 MBytes  16.3 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-10.04  sec  14.6 GBytes  12.5 Gbits/sec                  receiver
-----------------------------------------------------------
Server listening on 5201
-----------------------------------------------------------

./iperf3 -c 172.20.35.189
Connecting to host 172.20.35.189, port 5201
[  5] local 172.20.35.191 port 35062 connected to 172.20.35.189 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec  1.89 GBytes  16.2 Gbits/sec  100   1.45 MBytes
[  5]   1.00-2.00   sec  1.90 GBytes  16.4 Gbits/sec   86    638 KBytes
[  5]   2.00-3.00   sec  1.21 GBytes  10.4 Gbits/sec   57   1.15 MBytes
[  5]   3.00-4.00   sec   830 MBytes  6.96 Gbits/sec    7   1.54 MBytes
[  5]   4.00-5.00   sec   815 MBytes  6.84 Gbits/sec   15   1.10 MBytes
[  5]   5.00-6.00   sec  1.51 GBytes  12.9 Gbits/sec  362    690 KBytes
[  5]   6.00-7.00   sec  1.72 GBytes  14.8 Gbits/sec  665    690 KBytes
[  5]   7.00-8.00   sec  1.20 GBytes  10.4 Gbits/sec  639    778 KBytes
[  5]   8.00-9.00   sec  1.81 GBytes  15.6 Gbits/sec  879    708 KBytes
[  5]   9.00-10.00  sec  1.77 GBytes  15.2 Gbits/sec  865    577 KBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  14.6 GBytes  12.6 Gbits/sec  3675             sender
[  5]   0.00-10.04  sec  14.6 GBytes  12.5 Gbits/sec                  receiver

iperf Done.

-232 top
Tasks: 391 total,   1 running, 202 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.0 us,  1.4 sy,  0.0 ni, 97.2 id,  0.0 wa,  0.0 hi,  1.4 si,  0.0 st
KiB Mem : 65932576 total, 63203164 free,  2181920 used,   547492 buff/cache
KiB Swap: 28901372 total, 28901372 free,        0 used. 63111764 avail Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 2637 kolga     20   0   43716   4476   3944 S  23.6  0.0   0:02.09 lt-iperf3
 2510 root      20   0       0      0      0 I   4.3  0.0   0:00.46 kworker/10+
 2653 root      20   0       0      0      0 I   3.0  0.0   0:00.16 kworker/1:+
 2630 root      20   0       0      0      0 I   1.7  0.0   0:00.14 kworker/u4+
  565 root      20   0       0      0      0 I   1.0  0.0   0:00.32 kworker/2:+
 2609 root      20   0       0      0      0 I   1.0  0.0   0:00.48 kworker/u4+
 2273 root      20   0       0      0      0 I   0.7  0.0   0:01.67 kworker/u4+
   25 root      20   0       0      0      0 S   0.3  0.0   0:00.01 ksoftirqd/2
   74 root      20   0       0      0      0 S   0.3  0.0   0:00.03 ksoftirqd/+
 2651 kolga     20   0  158000   4568   3640 R   0.3  0.0   0:00.03 top
    1 root      20   0  191820   6108   3804 S   0.0  0.0   0:26.12 systemd
    2 root      20   0       0      0      0 S   0.0  0.0   0:00.01 kthreadd
    3 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_gp
    4 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_par_gp
    6 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 kworker/0:+
    9 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 mm_percpu_+
   10 root      20   0       0      0      0 S   0.0  0.0   0:00.00 ksoftirqd/0

-231 top
Tasks: 413 total,   3 running, 209 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.1 us,  3.1 sy,  0.0 ni, 94.7 id,  0.0 wa,  0.0 hi,  2.1 si,  0.0 st
KiB Mem : 65938632 total, 63202896 free,  2183572 used,   552164 buff/cache
KiB Swap: 33030140 total, 33030140 free,        0 used. 63113436 avail Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 2859 kolga     20   0   43716   4108   3572 R  70.8  0.0   0:05.75 lt-iperf3
 2875 root      20   0       0      0      0 I   7.3  0.0   0:00.36 kworker/9:+
  185 root      20   0       0      0      0 I   6.6  0.0   0:00.47 kworker/9:+
   68 root      20   0       0      0      0 S   5.0  0.0   0:00.25 ksoftirqd/9
 2421 root      20   0       0      0      0 I   3.0  0.0   0:00.10 kworker/13+
 2832 root      20   0       0      0      0 I   1.7  0.0   0:00.28 kworker/u4+
 2742 root      20   0       0      0      0 I   1.0  0.0   0:00.63 kworker/u4+
 2530 root      20   0       0      0      0 I   0.7  0.0   0:01.41 kworker/u4+
 2877 kolga     20   0  158000   4712   3740 R   0.3  0.0   0:00.02 top
    1 root      20   0  191980   6440   3900 S   0.0  0.0   0:26.23 systemd
    2 root      20   0       0      0      0 S   0.0  0.0   0:00.01 kthreadd
    3 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_gp
    4 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 rcu_par_gp
    5 root      20   0       0      0      0 I   0.0  0.0   0:00.02 kworker/0:+
    6 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 kworker/0:+
    8 root      20   0       0      0      0 I   0.0  0.0   0:00.00 kworker/u4+
    9 root       0 -20       0      0      0 I   0.0  0.0   0:00.00 mm_percpu_+

I could attach outputs from ethtool -S (too long to include) 

From google about CPU states and disabling them
[kolga@sti-rx200-232 ~]$ sudo cat
/sys/module/intel_idle/parameters/max_cstate  9

adding processor.max_cstate=0 and intel_idle.max_cstate=0 to the
kernel boot parameters made is so max_cstate stated at 0.

I re-did the re-experiments and it made no difference.