Re: [PATCH v4 3/7] acpi-cpufreq: Add support for disabling dynamic overclocking

[Matthew Garrett <mjg@redhat.com>]

On Sat, Jul 23, 2011 at 11:06:29PM -0400, Andy Lutomirski wrote:
> On 07/18/2011 12:37 PM, Matthew Garrett wrote:
> >One feature present in powernow-k8 that isn't present in acpi-cpufreq is
> >support for enabling or disabling AMD's core performance boost technology.
> >This patch adds that support to acpi-cpufreq, but also extends it to allow
> >Intel's dynamic acceleration to be disabled via the same interface. The
> >sysfs entry retains the cpb name for compatibility purposes.
> >
> >Signed-off-by: Matthew Garrett<mjg@redhat.com>
> >---
> >  drivers/cpufreq/acpi-cpufreq.c |  191 ++++++++++++++++++++++++++++++++++++++++
> >  1 files changed, 191 insertions(+), 0 deletions(-)
> >
> 
> How is this different from setting 3400000 vs 3401000 as the max
> freq? (I'm probably missing something about the way that Intel's
> turbo mode works.)

It's not really, it just means that there's a shared interface to 
achieve the same goal.

> >+	unsigned long val = 0;
> >+
> >+	ret = strict_strtoul(buf, 10,&val);
> >+	if (!ret&&  (val == 0 || val == 1)&&  cpb_supported)
> >+		cpb_toggle(val);
> 
> Shouldn't this set the flag only on policy->cpu (as opposed to all
> online cpus?)

It's effectively a per-package MSR rather than a per-thread or per-core 
one. We could make this per-package rather than system-wide, but I don't 
really know of any use cases that would benefit from the extra code. If 
anyone cares enough it could be added, but I'm a great believer in 
letting people come up with arguments for adding something before doing 
it...

> Also, you're missing a space before &&.

Think that's a cut and paste error. I'll fix it up.

-- 
Matthew Garrett | mjg59@srcf.ucam.org