[Bug 173541] High (0.60+, 1.00) idle load averages

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=173541

Mark Millard <marklmi26-fbsd@yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |marklmi26-fbsd@yahoo.com

--- Comment #36 from Mark Millard <marklmi26-fbsd@yahoo.com> ---
Context: aarch64 HoneyComb (so: 16 cores), main non-debug build,
no video hardware, serial console and ssh access in use.

I see the likes of: load averages:   0.20,   0.22,   0.25
in top -SIaocpu . The list of non-idle, non-zero WCPU% columns
is usually something like:

0.05% top -SIaocpu
0.03% [clock]
0.01% [intr]
0.01% [usb]

Sometimes there is also something like:

0.03% [acpi_thermal]
or:
0.01% [kernel]

Less often (varies more widely but is non-zero less often):

0.15% [zfskern]

There are also a number of processes listed with 0.00%,
such as (expect some overlaps with the above):

0.00% /usr/sbin/mountd -r /etc/exports /etc/zfs/exports
0.00% sshd: markmi@pts/1 (sshd)
0.00% /usr/sbin/ntpd -p /var/db/ntp/ntpd.pid -c /etc/ntp.conf -f
/var/db/ntp/ntpd.drift -g
0.00% [intr]
0.00% [pagedaemon]
0.00% [bufdaemon]
0.00% [rand_harvestq]
0.00% [zfskern]
0.00% [syncer]
0.00% [vnlru]
0.00% [acpi_cooling0]

Truely zero CPU usage processes are not listed.

Definitely doing somewhat more than nothing, as is normal for
so-called "idle" as far as I know.

What about this would make the load averages unreasonable?
(I'm also not aware of any sort of technical accuracy
criteria being claimed for the load averages. More general
judgments are involved about good-enough instead.)

For reference:

# uname -apKU # long output line split for readability
FreeBSD CA72_16Gp_ZFS 14.0-CURRENT FreeBSD 14.0-CURRENT #75
main-n259950-5723e5ac6d76-dirty: Fri Jan  6 01:50:53 PST 2023
root@CA72_16Gp_ZFS:/usr/obj/BUILDs/main-CA72-nodbg-clang/usr/main-src/arm64.aarch64/sys/GENERIC-NODBG-CA72
arm64 aarch64 1400077 1400077

-- 
You are receiving this mail because:
You are the assignee for the bug.