Re: Kernel is using a lot of CPU (was: Re: test)

> Sent: Monday, January 30, 2023 at 3:37 PM
> From: "Nikos Vassiliadis" <nvass@gmx.com>
> To: "Mike Karels" <mike@karels.net>
> Cc: freebsd-stable@freebsd.org
> Subject: Re: Kernel is using a lot of CPU (was: Re: test)
>
> 
> 
> > Sent: Monday, January 30, 2023 at 3:27 PM
> > From: "Mike Karels" <mike@karels.net>
> > To: "Nikos Vassiliadis" <nvass@gmx.com>
> > Cc: freebsd-stable@freebsd.org
> > Subject: Re: Kernel is using a lot of CPU (was: Re: test)
> >
> > On 30 Jan 2023, at 5:03, Nikos Vassiliadis wrote:
> > 
> > >> Hi,
> > >>
> > >> I just rebooted a system of mine and it seems that the kernel is constantly doing something. How could I debug this?
> > >> I am thinking of rebooting it tonight
> > >>
> > >> root@aurora:~ # top -S -b
> > >> last pid:  2196;  load averages:  1.00,  1.00,  1.00  up 0+01:54:34    12:41:35
> > >> 72 processes:  2 running, 69 sleeping, 1 waiting
> > >> CPU:  0.0% user,  0.0% nice, 24.6% system,  0.0% interrupt, 75.3% idle
> > >> Mem: 46M Active, 171M Inact, 1429M Wired, 14G Free
> > >> ARC: 919M Total, 354M MFU, 487M MRU, 6912K Anon, 8726K Header, 63M Other
> > >>      234M Compressed, 610M Uncompressed, 2.60:1 Ratio
> > >> Swap: 16G Total, 16G Free
> > >>
> > >>   PID USERNAME    THR PRI NICE   SIZE    RES STATE    C   TIME    WCPU COMMAND
> > >>    11 root          4 155 ki31     0B    64K RUN      0 345:06 314.31% idle
> > >>     0 root         49 -16    -     0B   784K swapin   1 112:27  99.85% kernel
> > >>    30 root        319 -16    -     0B  5120K spa->s   3   0:06   0.00% zpool-aurora-os
> > >>  1157 root          1  20    0   176M   150M select   2   0:05   0.00% smbd
> > >>  1154 root          1  20    0   175M   150M select   2   0:05   0.00% smbd
> > >>
> > >> This is a 12.4-RELEASE system.
> > >>
> > >> Thanks for any ideas,
> > >> Nikos
> > >>
> > >>
> > >>
> > >
> > > Sending the email again with the correct subject
> > 
> > You could add -H to the top command, which will show kernel threads within
> > the kernel process.
> > 
> > 		Mike
> > 
> > 
> 
> Thanks Mike,
> 
> A kernel thread named acpi_task_n seems to be the source of the problem. This system is rather old and was ok running 12.2-REL. Maybe some update to the ACPI code?. What could I try next?  
> 
> nik@aurora:~ % top -H -S -b
> last pid:  2529;  load averages:  1.03,  1.03,  1.00  up 0+04:43:45    15:30:46
> 840 threads:   6 running, 820 sleeping, 14 waiting
> CPU:  0.0% user,  0.0% nice, 24.6% system,  0.0% interrupt, 75.3% idle
> Mem: 29M Active, 192M Inact, 1717M Wired, 13G Free
> ARC: 1182M Total, 475M MFU, 608M MRU, 7744K Anon, 12M Header, 79M Other
>      312M Compressed, 782M Uncompressed, 2.51:1 Ratio
> Swap: 16G Total, 16G Free
> 
>   PID USERNAME    PRI NICE   SIZE    RES STATE    C   TIME    WCPU COMMAND
>    11 root        155 ki31     0B    64K CPU3     3 272:53 100.00% idle{idle: cpu3}
>    11 root        155 ki31     0B    64K RUN      0 260:10  81.15% idle{idle: cpu0}
>     0 root          8    -     0B   784K -        3 141:30  76.76% kernel{acpi_task_2}
>    11 root        155 ki31     0B    64K CPU1     1 212:51  74.66% idle{idle: cpu1}
>    11 root        155 ki31     0B    64K RUN      2 108:52  51.86% idle{idle: cpu2}
>     0 root          8    -     0B   784K CPU2     2 116:25  24.27% kernel{acpi_task_0}
>  2324 root         20    0   177M   151M select   3   0:03   0.10% smbd{smbd}
>     0 root          8    -     0B   784K -        0  21:08   0.00% kernel{acpi_task_1}
>     0 root        -16    -     0B   784K swapin   2   0:53   0.00% kernel{swapper}
> 
> 
> 
> 
> 
> 

12.4 autoloaded acpi_wmi with dubious results:

Autoloading module: acpi_wmi
acpi_wmi0: <ACPI-WMI mapping> on acpi0
acpi_wmi0: cannot find EC device
device_attach: acpi_wmi0 attach returned 6
acpi_wmi0: <ACPI-WMI mapping> on acpi0
acpi_wmi0: cannot find EC device
device_attach: acpi_wmi0 attach returned 6

12.2 did not autoload it.