On 06/24/2013 08:00 PM, Rafael J. Wysocki wrote: > On Monday, June 24, 2013 04:03:15 PM Viresh Kumar wrote: >> On 24 June 2013 15:50, Rafael J. Wysocki <[email protected]> wrote: >>> My impression was that that new interface wouldn't be useful on all >>> platforms >>> supported by cpufreq. If that's the case, I'd rather not make changes in >>> the >>> core code that are needed only to implement that new interface, at least to >>> start with. >> >> Every platform might not have userspace software that would do some >> actions based on this. But, these are nice and useful stats to have. >> >> They tell you a lot about how the hardware/governor/policies are >> behaving with load. And so will eventually help getting performance based >> on governor tunables for your platform. That's why I asked to put this in >> core as this should be pretty much useful. > > OK, let's see how the code will look like. > > Chanwoo, please prepare a v4 with the changes suggested by Viresh. >
@Viresh, I implement v3 patch for load_table debugfs file. As you suggested, v3 pach include old/new frequency data. But, I don't add additional function to minimize dependency among cpufreq core functions. And then I use CPUFREQ_LOADCHECK and CPUFREQ_POSTCHANGE notification to collect cpu data(time, old/new frequency, CPUs load). Please review v3 patch for load_table debugfs file. Thanks. Best Regards, Chanwoo Choi -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
