Re: [RFC] New Time of day proposal (updated 9/2/04)

[Christoph Lameter]

On Mon, 6 Sep 2004, Ulrich Windl wrote:

> On 3 Sep 2004 at 22:26, Dominik Brodowski wrote:
>
> ...
> > What about removing cpu_freq_khz you have in your patch, adding a per-CPU
> > value, and just use the value of the boot CPU for the other CPUs if
> > !CONFIG_DIFFERENT_CPU_SPEEDS or something like that?
> ...
>
> I would not mention SMP at that stage, but most of the existing code on IA32
> suumes that all CPUs run with the same frequency. I only heared that at least on
> Alphas this is not true. Probably you'll need a per-CPU state regarding time. Most
> likely resulting in the "local CPU's time" and a global concept of time that
> should not be in contradiction with any CPU's time. That might mean that any
> process always has to query the global time, involving extra overhead.

CPUs in SGI Altix systems may also run at different frequencies.
Typically 2 CPUs have a common clock source. For example in a system of
16 CPUs, there may exist 8 clock sources that slightly diverge from one
another.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/