RE: Wrong bogomips after plugging in AC power
Jeremy Fitzhardinge (jeremy@goop.org)
Fri, 29 Oct 1999 09:08:47 -0700 (PDT)
On 29-Oct-99 Pavel Machek wrote:
> Hi!
>
> So I've now confirmed that my toshiba can speed-up its cpu under some
> circumstances. When it is powered up while batteries are low (<20%
> will work), it will start with cpu at 150MHz, and bogomips are
> computed accordingly.
>
> If I plug it into AC power, it will speed up to 300MHz, but bogomips
> still have "slow" value. Therefore all udelays are wrong by factor of
> two -- udelay(50) will only wait approx. 25usec. That seems pretty
> dangerous to me. Maybe we need some other source of short loops?
I suspect I'll see a similar problem on the PowerPC. The PPC has a way of
throttling the rate instructions are delivered from the icache to the execution
units, and thereby reducing CPU power consumption/temperature. This slows down
the effective instruction execution speed, but all the clocks are unchanged. In
other words, udelay will break, but timing based on the timebase or decrementer
registers will be unaffected.
J
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/