Re: kapm-idled shows 90+% cpu usage when idle

Mike Fedyk (mfedyk@matchmail.com)
Fri, 10 Aug 2001 10:36:49 -0700


> > On Thu, 9 Aug 2001 19:33:42 -0400, safemode <safemode@speakeasy.net> wrote:
> > >Is this a true usage reading or just some quirk that's supposed to happen?
> > >I really doubt that this kernel daemon should really be using cpu. It
> > > seems to respond with a higher cpu usage when i'm idle. It immediately
> > > goes away when something else uses cpu. If you need any more info just
> > > ask. I'm
> >

> On Thursday 09 August 2001 20:36, Mikael Pettersson wrote:
> > 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
>

On Thu, Aug 09, 2001 at 10:01:55PM -0400, safemode wrote:
> I've been told by others that this is exactly what's supposed to happen. It
> acts like it's using cpu when it's idle and does it job that way. I see no
> difference either way. I'm using a KA7 motherboard and it says it supports
> apm and lspci shows what i pasted in the original post. Oh well, it's not
> causing the cpu to generate more heat than it would be idle.

If you don't see any benefit, I'd disable it just because of the modified
results from top...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/