From: Thomas Renninger <trenn@xxxxxxx>
Date: Thu, Mar 25, 2010 at 09:59:31PM +0100
> On Thursday 25 March 2010 08:55:19 pm Valdis.Kletnieks@xxxxxx wrote:
> > On Thu, 25 Mar 2010 10:43:04 BST, Thomas Renninger said:
> > > > + if (c->cpuid_level >= 6) {
> > >
> > > and remove this (arch/x86/kernel/cpu/intel.c):
> > > if (c->cpuid_level > 6) {
> >
> > So is > or >= the correct comparator here?
>
> This one: >= is correct (for both).
> I double checked, there is one Intel CPU type
> having a cpuid_level of 6, but this would not support aperf/mperf, thus
> above is still fine.
Agreed. ">" won't work on machines which have base cpuidlevel == 6 (like
AMD, f.e.)
> The remaining question is what Borislav said:
> are there other x86 CPU vendors who could use this differently.
> I very much expect there are not.
>
> IMO you should resubmit this one or the whole series with this change
Ok, will do tomorrow.
> and Dave should just push this in his tree and queue it up, there was enough
> time to object.
The problem here is that the patches touch both cpufreq and x86 code.
Hmmm...
--
Regards/Gruss,
Boris.
--
Advanced Micro Devices, Inc.
Operating Systems Research Center
--
To unsubscribe from this list: send the line "unsubscribe cpufreq" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
