I am using Mandrake 8.1 on a Duron
Occasionally this process sucks 50% to 60% cpu for several minutes
Anybody know why? (I don't have apmd running)
Occasionally this process sucks 50% to 60% cpu for several minutes
Anybody know why? (I don't have apmd running)
If you had bothered to searchQuote:> I am using Mandrake 8.1 on a Duron
> Occasionally this process sucks 50% to 60% cpu for several minutes
> Anybody know why? (I don't have apmd running)
kapm-idled is a kernel process whose purpose in life is to run when
nothing else is running, and issue HLT instructions to an x86 processor.
(A processor executing a HLT instruction uses much less energy than a
processor executing any other instruction.) Many people have noticed
this, and some of them freaked out. Because this is counterproductive,
recent versions of the kernel (2.4.16, certainly) do not report
kapm-idled as using any CPU at all.
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin /
http://www.brainbench.com / "He is a rhythmic movement of the
-----------------------------/ penguins, is Tux." --MegaHAL
> On Sat, 26 Jan 2002 17:47:35 GMT, Ken Schrock staggered into the Black
> Sun and said:
> > I am using Mandrake 8.1 on a Duron
> > Occasionally this process sucks 50% to 60% cpu for several minutes
> > Anybody know why? (I don't have apmd running)
> If you had bothered to search
> http://groups.google.com/advanced_group_search for keyword "kapm-idle"
> and newsgroups comp.os.linux.* , you would've had an answer very
> quickly.
Replies like this do more damage to Linux and than all other factors
combined
Now I am sure someone as superior as you must knows this, so
IS it that you just don't care, or is Bill paying you?
Quote:> kapm-idled is a kernel process whose purpose in life is to run when
> nothing else is running, and issue HLT instructions to an x86 processor.
> (A processor executing a HLT instruction uses much less energy than a
> processor executing any other instruction.) Many people have noticed
> this, and some of them freaked out. Because this is counterproductive,
> recent versions of the kernel (2.4.16, certainly) do not report
> kapm-idled as using any CPU at all.
> --
> Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
> Brainbench MVP for Linux Admin /
> http://www.brainbench.com / "He is a rhythmic movement of the
> -----------------------------/ penguins, is Tux." --MegaHAL
>> On Sat, 26 Jan 2002 17:47:35 GMT, Ken Schrock staggered into the Black
>> Sun and said:
>> > I am using Mandrake 8.1 on a Duron
>> > Occasionally this process sucks 50% to 60% cpu for several minutes
>> > Anybody know why? (I don't have apmd running)
>> If you had bothered to search
>> http://groups.google.com/advanced_group_search for keyword "kapm-idle"
>> and newsgroups comp.os.linux.* , you would've had an answer very
>> quickly.
> It seems that I got an answer very quickly this way
> And got to allow someone with an ego problem to act superior too
> This way seems like way more fun all the way around
Put that ego back in a bag, and pronounce "thank you", correctly..Quote:> Replies like this do more damage to Linux and than all other factors
2.4.17, which I am using, does show kapm-idled at 50% when I am onQuote:> combined
> Now I am sure someone as superior as you must knows this, so
> IS it that you just don't care, or is Bill paying you?
>> kapm-idled is a kernel process whose purpose in life is to run when
>> nothing else is running, and issue HLT instructions to an x86 processor.
>> (A processor executing a HLT instruction uses much less energy than a
>> processor executing any other instruction.) Many people have noticed
>> this, and some of them freaked out. Because this is counterproductive,
>> recent versions of the kernel (2.4.16, certainly) do not report
>> kapm-idled as using any CPU at all.
I didn't realize that accounting had changed. Thanks (to DWC).
Peter
1. kapm-idled shows 90+% cpu usage when idle
Do you have CONFIG_APM_CPU_IDLE=y in your .config? If so, disable it.
There was a thread about this problem some months ago. I found
that on all of my APM-capable machines, including a Dell laptop,
CONFIG_APM_CPU_IDLE=y had a negative effect. The kernel ended up
in a tight loop performing tons of APM IDLE BIOS calls, since each
BIOS call returned immediately without having idled the CPU.
Leaving CONFIG_APM_CPU_IDLE unset lets the kernel use its own
"HLT when idle" code. On my main development box, idle CPU
temperature dropped >10 degrees C, and kapm-idled now uses 0% CPU.
/Mikael
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
2. replace special character with sed
3. kapm-idled no longer idling CPU?
5. disabling the kapm-idled process
8. help: I can't install FreeBSD 4.3 on Tekram DC-390U3W Card.
11. kapm-idled
12. Kapm-idled doesn't reduce power consumption.