| Message ID | 1370502472-7249-6-git-send-email-l.majewski@samsung.com (mailing list archive) |
|---|---|
| State | Changes Requested, archived |
| Headers | show |
On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > +config CPU_FREQ_BOOST > + bool "CPU frequency boost support" > + help > + Switch to enable support for frequency boost > + > + If in doubt, say N. > + This help text is devoid of any useful information. On what platforms ? What's the upside/downside ? Why is it an option ? Dave -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Dave, > On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > > > +config CPU_FREQ_BOOST > > + bool "CPU frequency boost support" > > + help > > + Switch to enable support for frequency boost > > + > > + If in doubt, say N. > > + > > This help text is devoid of any useful information. > > On what platforms ? What's the upside/downside ? Why is it an option ? I shall be more verbose here. The boost option is supposed to provide one solution to control software based (like is is done with Samsung Exynos4 SoC) and hardware based (like Intel's Turbo Boost feature) boost. Support for Intel's boost is already in mainline. Therefore I aimed to extend cpufreq to also manage software based solutions (e.g. Exynos).
On Thu, Jun 06, 2013 at 05:14:31PM +0200, Lukasz Majewski wrote: > Hi Dave, > > > On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > > > > > +config CPU_FREQ_BOOST > > > + bool "CPU frequency boost support" > > > + help > > > + Switch to enable support for frequency boost > > > + > > > + If in doubt, say N. > > > + > > > > This help text is devoid of any useful information. > > > > On what platforms ? What's the upside/downside ? Why is it an option ? > > I shall be more verbose here. > > The boost option is supposed to provide one solution to control > software based (like is is done with Samsung Exynos4 SoC) and hardware > based (like Intel's Turbo Boost feature) boost. > > Support for Intel's boost is already in mainline. Therefore I aimed to > extend cpufreq to also manage software based solutions (e.g. Exynos). Given CPUFREQ is available on more platforms than X86/ARM, this option could also use a depends. Dave -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 6 June 2013 20:51, Dave Jones <davej@redhat.com> wrote: > Given CPUFREQ is available on more platforms than X86/ARM, this option > could also use a depends. It should be disabled by default, so please add default n. But we don't need a depends here as everybody is open to use it. Its a core feature. -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > Enable/disable support for BOOST. New flag - CPU_FREQ_BOOST has been > defined for that. What's the reason for this? Why would you want to not compile-in the boosting code? Also, at least on x86, even with this feature disabled, the CPU will boost by default unless you fiddle with it through the MSRs as root.
Hi, > On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > > Enable/disable support for BOOST. New flag - CPU_FREQ_BOOST has been > > defined for that. > > What's the reason for this? Why would you want to not compile-in the > boosting code? I think that this flag could be removed and the "boost" could be compiled-in permanently. Viresh, what is your opinion? > > Also, at least on x86, even with this feature disabled, the CPU will > boost by default unless you fiddle with it through the MSRs as root. > So this is a good argument for removing this flag from Kconfig.
On 10 June 2013 18:50, Lukasz Majewski <l.majewski@samsung.com> wrote: > Hi, > >> On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: >> > Enable/disable support for BOOST. New flag - CPU_FREQ_BOOST has been >> > defined for that. >> >> What's the reason for this? Why would you want to not compile-in the >> boosting code? > > I think that this flag could be removed and the "boost" could be > compiled-in permanently. > > Viresh, what is your opinion? Yes, Borislav is correct. I haven't actually reviewed these last patches as I saw lots of issues with first one :) -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Viresh, > On 10 June 2013 18:50, Lukasz Majewski <l.majewski@samsung.com> wrote: > > Hi, > > > >> On Thu, Jun 06, 2013 at 09:07:52AM +0200, Lukasz Majewski wrote: > >> > Enable/disable support for BOOST. New flag - CPU_FREQ_BOOST has > >> > been defined for that. > >> > >> What's the reason for this? Why would you want to not compile-in > >> the boosting code? > > > > I think that this flag could be removed and the "boost" could be > > compiled-in permanently. > > > > Viresh, what is your opinion? > > Yes, Borislav is correct. I haven't actually reviewed these last > patches as I saw lots of issues with first one :) Ok, thanks :-)
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index a1488f5..1007c5e 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -23,6 +23,13 @@ config CPU_FREQ_TABLE config CPU_FREQ_GOV_COMMON bool +config CPU_FREQ_BOOST + bool "CPU frequency boost support" + help + Switch to enable support for frequency boost + + If in doubt, say N. + config CPU_FREQ_STAT tristate "CPU frequency translation statistics" select CPU_FREQ_TABLE