Re: [OS] nvOC easy-to-use Linux Nvidia Mining v0019-1.4

Any one has kworker high cpu usage ?
Any idea how to solve it ?

Code:

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1797 root 20 0 0 0 0 S 44.9 0.0 0:53.41 kworker/3:16
6504 root 20 0 0 0 0 R 21.0 0.0 0:40.13 kworker/3:2
6751 m1 20 0 0.170t 1.165g 0.982g S 5.6 15.0 1:27.05 ethminer

My rig with 13 1060 had a very high cpu load. Moved power limit from 75W to 95W and load went down form 6-10 to 1.8-2.5. No perfect but much better.

Good call, Thanks a lot.
Will give it a shot.

Edit
Thanks a lot mate, you saved my day.
So I raised 5 steps at a time, at 85 load average get to 1.5-1.8, went up to 90, 95, 100 and load went up too.
So 85 is the sweet spot for me.

Code:

Load Average: 1.60 1.83 1.83

Happy...
so what was your PL and load average before modification?

Load average was over 4 with power limit of 75 on core i5 6400 cpu

Very rare... my rig has a g3900. I think a i5 should not get that load.

Exactly, That was making me mad.

now :

Code:

load average: 0.68, 1.27, 1.51

Ideal load would be <2? I just checked for my 1060 rig (x10 cards) and it's anywhere from 2.1-3.5 -- Should I bump up my PL (78W now)? Not really too familiar with linux, so what does a high system load lead to (i.e. instability, decreased longevity, etc)?

It depends.. if your cpu has 2 core 2 is high (100%) if it has 4 core 2 is ok (50%). Although this does not mean that there is not an application consuming too many resources.

To check cpu type :

Code:

cat /proc/cpuinfo | grep 'model name' | uniq | cut -d":" -f2

To check number of cores :

Code:

nproc

to check load average:

Code:

cat /proc/loadavg
0.42 0.40 0.36 3/410 588

where:
first number is load average in last minute
second number is load average in last 5 minutes
third number is load average in last 15 minutes
The rest... I don't know Huh

Or use htop to watch in real time
Let me know if you have doubts about this topic