On Sun, Jul 02, 2000 at 07:21:25PM +0200, Ramon van Handel wrote:
>
> I think the problem is the unreliability of the current timing.
> BogoMIPS calculation relies on reliable timing... if timing
> is very unreliable, it doesn't surprise me that it goes wrong.
>
> Let me elaborate: in order for BogoMIPS calibration to work,
> the system needs to work in such a way that every time a unit
> of (PIT) time elapses, a roughly equal amount of instructions
> has been executed (the same instructions, the BogoMIPS loop
> is tight: see /usr/src/linux/arch/i386/lib/delay.c in
> linux 2.2.x).
Akthough the implementation of BogoMIPS must be relatively robust - I
have an old 486 with one of those turbo buttons on that will slow it
down (by a factor of 8-10). Turning the turbo on and off must mean
that the BogoMIPS is hugely out, but the box doesn't seem to trip up
when you do that (although I haven't ever tried toggling it during the
BogoMIPS calibration). Presumably this has roughly the same effect
from the kernel's point of view as if it were running in a VM with the
load of the host varying rapidly.
James Hollingshead
