Re: The SpedStep crash problem ...
Martin Skøtt [martin@z3n.dk] wrote:
>
> OK, we are now a couple of people here on the list who are struggling with
> random crashes of our laptops. The standard solution is to disable SpeedStep
> in the BIOS. Until now this haven't helped on my machine, but I have many
> SpeedStep settings and I still haven't tried all of them.
> Disabling SpeedStep is a workaround, but it's not acceptible for me as a
> private laptop buyer. What is the problem with Linux and SpeedStep? Does it
> affect all machines (judgeing from this list, no) and what is that happen? In
> my own case it has always happened while running on AC power, but a good
> explanation to this is that I run on AC 99% of the time. My model of machine[1]
> can be delivered with Linux pre-installed by IBM and they don't mention any
> problems nor do they mention that SpeedStep has been disabled so they must have
> a workaround.
>
erm a small thing to mention. Speedstep = bollocks (in a word). The reason
it came about was because macrocroft refused to create a patch that called
the cpu cooling instruction (HLT) when the machine was idle. This ment
although your machine is not 100% busy it is 100% active. The HLT command
tells the CPU to turn off power for a short time to most of the active parts
on the chip. This increases battery life. Is this *unacceptable*?
The solution that the computer world came up with was that if the CPU
*detected* that it was idle (although not calling the HLT instruction) it
would clock itsself down and saves power at a lower clock speed, however
this is *not* as efficient as the HLT instruction (I have found through
experimentation).
Now when this happens it heavily messes up the Bogo MIP rating in the kernel
and things go bad from there! This is what the problem is, a timing loop!
Now as I said before and others have said (with damn good reason) turn *OFF*
speedstep and use HLT instead. <Mr_T> Damn Fool </Mr_T>. If this is *still*
unacceptable then ask Intel for the nice documentation and write the damn
support into the kernel yourself. Otherwise learn to use Google and Deja and
use the linux utility to control speedstep under linux (look under
alt.comp.portable.linux , if I remember correctly for the thread, it was
about 2-4 weeks ago).
>From a *good* laptop you should get about 3-4 hours battery life from a
single charge on a single battery. If this doesn't happen get a proper
laptop and stop treating a laptop like a desktop, you paid an extra $500 or
so it was small, not so it would be a cosmetic toy that became *unacceptable*
because you cannot use Google or Deja (aka groups.google.com).
Alex
Reply to: