[PATCH v8 1/4] cpufreq: Introduce an optional cpuinfo_avg_freq sysfs entry

Beata Michalska beata.michalska at arm.com
Tue Dec 17 12:10:43 PST 2024


On Tue, Dec 17, 2024 at 09:57:26AM +0530, Viresh Kumar wrote:
> On 16-12-24, 23:15, Beata Michalska wrote:
> > My bad as I must have misinterpreted that message. Although I am not entirely
> > sure why this might be unacceptable as it is not such uncommon approach to use
> > signed int space to cover both: expected positive value as well as potential
> > error code case failure.
> 
> This part is fine. The problem is with handling frequency here. Signed int can
> capture up to 2 GHz of freq, where as unsigned int can capture up to 4 GHz and
> so we would really like to keep it at 4 GHz..
Right, though the arch_freq_get_on_cpu operates on kHz values.

---
BR
Beata
> 
> Maybe we need to move to 64 bits for frequency at some point of time, but at
> least we should try to not break it for now.
> 
> > Enabling the new attribute for all is an option, tough not entirely compelling
> > one as exposing a feature that is known not to be supported seems bit
> > counterintuitive. On the other hand using cpufreq driver flags won't help much
> > as the support for the new attrib is platform-specific, not driver-specific.
> 
> -- 
> viresh



More information about the linux-arm-kernel mailing list