Constant load of 1 on a recent 12-STABLE

[Gordon Bergling]

Hi Ian,

On Wed, Jun 03, 2020 at 02:40:07PM -0600, Ian Lepore wrote:
> On Wed, 2020-06-03 at 22:29 +0200, Gordon Bergling via freebsd-hackers wrote:
> > [...]
> > 
> > The only key performance indicator that is relatively high IMHO, for a 
> > non-busy system, are the context switches, that vmstat has reported.
> > 
> > -------------------------------------------------------------------------------
> > procs  memory       page                    disks     faults         cpu
> > r b w  avm   fre   flt  re  pi  po    fr   sr da0 da1   in    sy    cs us sy id
> > 0 0 0 514G  444M  7877   2   7   0  9595  171   0   0    0  4347 43322 17  2 81
> > 0 0 0 514G  444M     1   0   0   0     0   44   0   0    0   121 40876  0  0 100
> > 0 0 0 514G  444M     0   0   0   0     0   40   0   0    0   133 42520  0  0 100
> > 0 0 0 514G  444M     0   0   0   0     0   40   0   0    0   120 43830  0  0 100
> > 0 0 0 514G  444M     0   0   0   0     0   40   0   0    0   132 42917  0  0 100
> > --------------------------------------------------------------------------------
> > 
> > Any other ideas what could generate that load?
> 
> Interrupts/second consistently zero is the strange part of that vmstat
> output.  That makes me think the problem has something to do with
> timekeeping and is a problem with the statistics used by top rather
> than having some "stuck process" actually consuming time.
> 
> Are there any other signs of timekeeping trouble on the machine, like
> ntpd repeatedly stepping the clock?  What's the output for
> 
>   sysctl kern.timecounter
>   sysctl kern.eventtimer

first, sorry for the late response. The e-mail went into my spam folder.

There aren't any timekeeping problemes on that machine. The requested output
of the request sysctl trees are the following:

-------------------------------------------------------------------------------
$ sysctl kern.timecounter
kern.timecounter.tsc_shift: 1
kern.timecounter.smp_tsc_adjust: 0
kern.timecounter.smp_tsc: 0
kern.timecounter.invariant_tsc: 0
kern.timecounter.fast_gettime: 1
kern.timecounter.tick: 1
kern.timecounter.choice: ACPI-fast(900) Hyper-V-TSC(3000) TSC-low(-100) Hyper-V(2000) dummy(-1000000)
kern.timecounter.hardware: Hyper-V-TSC
kern.timecounter.alloweddeviation: 5
kern.timecounter.timehands_count: 2
kern.timecounter.stepwarnings: 0
kern.timecounter.tc.ACPI-fast.quality: 900
kern.timecounter.tc.ACPI-fast.frequency: 3579545
kern.timecounter.tc.ACPI-fast.counter: 483840041
kern.timecounter.tc.ACPI-fast.mask: 4294967295
kern.timecounter.tc.Hyper-V-TSC.quality: 3000
kern.timecounter.tc.Hyper-V-TSC.frequency: 10000000
kern.timecounter.tc.Hyper-V-TSC.counter: 1283654243
kern.timecounter.tc.Hyper-V-TSC.mask: 4294967295
kern.timecounter.tc.TSC-low.quality: -100
kern.timecounter.tc.TSC-low.frequency: 1600955785
kern.timecounter.tc.TSC-low.counter: 1215462580
kern.timecounter.tc.TSC-low.mask: 4294967295
kern.timecounter.tc.Hyper-V.quality: 2000
kern.timecounter.tc.Hyper-V.frequency: 10000000
kern.timecounter.tc.Hyper-V.counter: 1283862164
kern.timecounter.tc.Hyper-V.mask: 4294967295
-------------------------------------------------------------------------------

and

-------------------------------------------------------------------------------
$ sysctl kern.eventtimer
kern.eventtimer.periodic: 0
kern.eventtimer.timer: LAPIC
kern.eventtimer.idletick: 0
kern.eventtimer.singlemul: 4
kern.eventtimer.choice: Hyper-V(1000) LAPIC(100) RTC(0)
kern.eventtimer.et.Hyper-V.quality: 1000
kern.eventtimer.et.Hyper-V.frequency: 10000000
kern.eventtimer.et.Hyper-V.flags: 6
kern.eventtimer.et.RTC.quality: 0
kern.eventtimer.et.RTC.frequency: 32768
kern.eventtimer.et.RTC.flags: 17
kern.eventtimer.et.LAPIC.quality: 100
kern.eventtimer.et.LAPIC.frequency: 100426660
kern.eventtimer.et.LAPIC.flags: 15
-------------------------------------------------------------------------------

I hope you see something unusual that could be changed to solve the problem.

Kind regards,

Gordon

-- 
Gordon Bergling
Mobile: +49 170 23 10 948
Web: https://www.gordons-perspective.com/
Mail: gbergling at gmail.com

Think before you print!