Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Rafael J. Wysocki
On Monday, February 20, 2017 02:49:18 PM Rafael J. Wysocki wrote: > On Monday, February 20, 2017 03:28:03 PM Viresh Kumar wrote: > > On 17-02-17, 13:48, Peter Zijlstra wrote: > > > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > > > On Thursday, February 16, 2017 01:36:05 PM

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Rafael J. Wysocki
On Monday, February 20, 2017 02:49:18 PM Rafael J. Wysocki wrote: > On Monday, February 20, 2017 03:28:03 PM Viresh Kumar wrote: > > On 17-02-17, 13:48, Peter Zijlstra wrote: > > > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > > > On Thursday, February 16, 2017 01:36:05 PM

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Rafael J. Wysocki
On Monday, February 20, 2017 03:28:03 PM Viresh Kumar wrote: > On 17-02-17, 13:48, Peter Zijlstra wrote: > > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > > > On Thu, Feb 16, 2017 at 03:42:10PM +0530,

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Rafael J. Wysocki
On Monday, February 20, 2017 03:28:03 PM Viresh Kumar wrote: > On 17-02-17, 13:48, Peter Zijlstra wrote: > > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > > > On Thu, Feb 16, 2017 at 03:42:10PM +0530,

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Viresh Kumar
On 17-02-17, 13:48, Peter Zijlstra wrote: > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > > > But when I discussed this with Vincent, he

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-20 Thread Viresh Kumar
On 17-02-17, 13:48, Peter Zijlstra wrote: > On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > > > But when I discussed this with Vincent, he

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-17 Thread Peter Zijlstra
On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > > But when I discussed this with Vincent, he suggested that it may not be > > > required > >

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-17 Thread Peter Zijlstra
On Fri, Feb 17, 2017 at 01:15:56PM +0100, Rafael J. Wysocki wrote: > On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > > But when I discussed this with Vincent, he suggested that it may not be > > > required > >

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-17 Thread Rafael J. Wysocki
On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > But when I discussed this with Vincent, he suggested that it may not be > > required > > at all as the scheduler (with the helped of "decayed") doesn't call into >

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-17 Thread Rafael J. Wysocki
On Thursday, February 16, 2017 01:36:05 PM Peter Zijlstra wrote: > On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > > But when I discussed this with Vincent, he suggested that it may not be > > required > > at all as the scheduler (with the helped of "decayed") doesn't call into >

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-16 Thread Peter Zijlstra
On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > But when I discussed this with Vincent, he suggested that it may not be > required > at all as the scheduler (with the helped of "decayed") doesn't call into > schedutil too often, i.e. at least 1 ms. And if the CPUs are stable

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-16 Thread Peter Zijlstra
On Thu, Feb 16, 2017 at 03:42:10PM +0530, Viresh Kumar wrote: > But when I discussed this with Vincent, he suggested that it may not be > required > at all as the scheduler (with the helped of "decayed") doesn't call into > schedutil too often, i.e. at least 1 ms. And if the CPUs are stable

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-16 Thread Viresh Kumar
On 16-02-17, 01:02, Rafael J. Wysocki wrote: > More precisely, while the governor computations are less costly than updating > the CPU state, they are not zero-cost, so do we really want to run them on > every governor callback invocation until the CPU state is updated? > > We may end up running

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-16 Thread Viresh Kumar
On 16-02-17, 01:02, Rafael J. Wysocki wrote: > More precisely, while the governor computations are less costly than updating > the CPU state, they are not zero-cost, so do we really want to run them on > every governor callback invocation until the CPU state is updated? > > We may end up running

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Viresh Kumar
On 15-02-17, 23:35, Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > First of all, [RFC] pretty please on things like this. Sure. > > Normally, the time it takes to reevaluate the frequency is negligible > > compared to the time it takes to change

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Viresh Kumar
On 15-02-17, 23:35, Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > First of all, [RFC] pretty please on things like this. Sure. > > Normally, the time it takes to reevaluate the frequency is negligible > > compared to the time it takes to change

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 11:52:49 PM Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 11:35:29 PM Rafael J. Wysocki wrote: > > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > > > First of all, [RFC] pretty please on things like this. > > > > > For an ideal

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 11:52:49 PM Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 11:35:29 PM Rafael J. Wysocki wrote: > > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > > > First of all, [RFC] pretty please on things like this. > > > > > For an ideal

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 11:35:29 PM Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > First of all, [RFC] pretty please on things like this. > > > For an ideal system (where frequency change doesn't incur any penalty) > > we would like to

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 11:35:29 PM Rafael J. Wysocki wrote: > On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: > > First of all, [RFC] pretty please on things like this. > > > For an ideal system (where frequency change doesn't incur any penalty) > > we would like to

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: First of all, [RFC] pretty please on things like this. > For an ideal system (where frequency change doesn't incur any penalty) > we would like to change the frequency as soon as the load changes for a > CPU. But the systems we have

Re: [PATCH] cpufreq: schedutil: govern how frequently we change frequency with rate_limit

2017-02-15 Thread Rafael J. Wysocki
On Wednesday, February 15, 2017 10:45:47 PM Viresh Kumar wrote: First of all, [RFC] pretty please on things like this. > For an ideal system (where frequency change doesn't incur any penalty) > we would like to change the frequency as soon as the load changes for a > CPU. But the systems we have