Re: [PATCH] cpufreq: move frequency table helpers to extra module

Dominik Brodowski (linux@brodo.de)
Thu, 13 Feb 2003 11:00:29 +0100


On Thu, Feb 13, 2003 at 09:39:51AM +0000, Dave Jones wrote:
> On Thu, Feb 13, 2003 at 10:11:31AM +0100, Dominik Brodowski wrote:
> > The CPU frequency table helpers can easily be modularized --
> > especially as they are not needed on all architectures, or for
> > all drivers.
>
> As most of the x86 drivers have been converted now, it looks like
> it'd make more sense to conditionalise this on architecture, and
> move the remaining x86 drivers over to the helpers (longrun/longhaul).

Longhaul yes, Longrun no. On longrun you can't set the speed to a specific
frequency[*], only to a _frequency range_. So the frequency table helpers
aren't needed there. Also, for gx-suspmod, which allows really fine-grained
setting of frequencies, the frequency table helpers don't make sense.

> It just strikes me as silly that we have a config option that when
> disabled could end up showing no chip drivers when the conversion
> is complete.

So, even when the conversion is complete, there are two drivers which can be
compiled. Additionally, some users (or distros!) might prefer modularizing
this.

Dominik

[*] you can set policy->max = policy->min, but that really doesn't make
sense on longrun.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/