BogoMIPS (Re: [plug] Bogus Sound Card Problems)

[bburns at erggroup.com]

Matt Kemner eloquently stated :

> On Fri, 4 Aug 2000 bburns at erggroup.com wrote:

> > G'day Pluggers,
> >
> > First a question:  What is a Bogomip - specifically, how is it calculated
and
> > what is it meant to represent?  It seemed to jump a whole heap (5 times)
after
> > installing a later kernel on the same machine.  Is it Bogus?
>
> Yes :)
>
> It is just an internal timing loop, which can fluctuate between different
> kernel versions, and will fluctuate between different CPUs.
>
> >From the BogoMIPS mini-HOWTO:
> (which will be available somewhere on http://www.linux.org.au/LDP/HOWTO/
>  when someone kicks the webserver back into gear)
>
>  `MIPS is short for Millions of Instructions Per Second.  It
>       is a measure for the computation speed of a program.  Like
>       most such measures, it is more often abused than used prop
>       erly (it is very difficult to justly compare MIPS for dif
.       ferent kinds of computers).
>
>       BogoMips are Linus's invention. The kernel (or was it a
>       device driver?) needs a timing loop (the time is too short
>       and/or needs to be too exact for a non-busy-loop method of
>       waiting), which must be calibrated to the processor speed of
>       the machine. Hence, the kernel measures at boot time how
>       fast a certain kind of busy loop runs on a computer. "Bogo"
>       comes from "bogus", i.e, something which is a fake. Hence,
>       the BogoMips value gives some indication of the processor
>       speed, but it is way too unscientific to be called anything
>       but BogoMips.
>
>       The reasons (there are two) it is printed during bootup is
>       that a) it is slightly useful for debugging and for checking
>       that the computers caches and turbo button work, and b)
>       Linus loves to chuckle when he sees confused people on the
>       news.'

Seems that if it can go from 66 to 330, just by upgrading the kernel (and not
ending up with 5 times speed increase for real applications), that it is a
"little" unreliable.

BUT:  Is it running through the same timing loop, implying that at least for
THIS calculation - there is a 5 times speed increase??

Cheers,
Ben