| Message ID | 20171129111227.26798-1-broonie@kernel.org (mailing list archive) |
|---|---|
| State | Mainlined |
| Delegated to: | Rafael Wysocki |
| Headers | show |
On Wed, Nov 29, 2017 at 12:12 PM, Mark Brown <broonie@kernel.org> wrote: > A frequent source of build problems is poor handling of optional PM > support, almost all development is done with the PM options enabled > but they can be turned off. Currently few if any of the build test > services do this as standard as there is no standard config for it and > the use of selects and def_bool means that simply setting CONFIG_PM=n > doesn't do what is expected. To make this easier provide a fragement > that can be used with KCONFIG_ALLCONFIG to force PM off. > > CONFIG_XEN is disabled as Xen uses hibernation callbacks which end up > turning on power management on architectures with Xen. Some cpuidle > implementations on ARM select PM so CONFIG_CPU_IDLE is disabled, and > some ARM architectures unconditionally enable PM so they are also > disabled. > > Signed-off-by: Mark Brown <broonie@kernel.org> I'm going to apply this shortly, sorry for the delay. > --- > MAINTAINERS | 1 + > kernel/configs/nopm.config | 15 +++++++++++++++ > 2 files changed, 16 insertions(+) > create mode 100644 kernel/configs/nopm.config > > diff --git a/MAINTAINERS b/MAINTAINERS > index 07d9067f68fd..539808dc7928 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -10895,6 +10895,7 @@ F: include/linux/pm.h > F: include/linux/pm_* > F: include/linux/powercap.h > F: drivers/powercap/ > +F: kernel/configs/nopm.config > > POWER STATE COORDINATION INTERFACE (PSCI) > M: Mark Rutland <mark.rutland@arm.com> > diff --git a/kernel/configs/nopm.config b/kernel/configs/nopm.config > new file mode 100644 > index 000000000000..81ff07863576 > --- /dev/null > +++ b/kernel/configs/nopm.config > @@ -0,0 +1,15 @@ > +CONFIG_PM=n > +CONFIG_SUSPEND=n > +CONFIG_HIBERNATION=n > + > +# Triggers PM on OMAP > +CONFIG_CPU_IDLE=n > + > +# Triggers enablement via hibernate callbacks > +CONFIG_XEN=n > + > +# ARM/ARM64 architectures that select PM unconditionally > +CONFIG_ARCH_OMAP2PLUS_TYPICAL=n > +CONFIG_ARCH_RENESAS=n > +CONFIG_ARCH_TEGRA=n > +CONFIG_ARCH_VEXPRESS=n > -- > 2.15.0 >
On Wed, Nov 29, 2017 at 03:37:01PM +0100, Rafael J. Wysocki wrote: > On Wed, Nov 29, 2017 at 12:12 PM, Mark Brown <broonie@kernel.org> wrote: > > A frequent source of build problems is poor handling of optional PM > > support, almost all development is done with the PM options enabled > I'm going to apply this shortly, sorry for the delay. Thanks! I sent it just before the merge window so I was expecting to have to resend even when I originally posted it.
On Wednesday, November 29, 2017 12:12:27 PM CET Mark Brown wrote: > A frequent source of build problems is poor handling of optional PM > support, almost all development is done with the PM options enabled > but they can be turned off. Currently few if any of the build test > services do this as standard as there is no standard config for it and > the use of selects and def_bool means that simply setting CONFIG_PM=n > doesn't do what is expected. To make this easier provide a fragement > that can be used with KCONFIG_ALLCONFIG to force PM off. > > CONFIG_XEN is disabled as Xen uses hibernation callbacks which end up > turning on power management on architectures with Xen. Some cpuidle > implementations on ARM select PM so CONFIG_CPU_IDLE is disabled, and > some ARM architectures unconditionally enable PM so they are also > disabled. > > Signed-off-by: Mark Brown <broonie@kernel.org> > --- > MAINTAINERS | 1 + > kernel/configs/nopm.config | 15 +++++++++++++++ > 2 files changed, 16 insertions(+) > create mode 100644 kernel/configs/nopm.config > > diff --git a/MAINTAINERS b/MAINTAINERS > index 07d9067f68fd..539808dc7928 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -10895,6 +10895,7 @@ F: include/linux/pm.h > F: include/linux/pm_* > F: include/linux/powercap.h > F: drivers/powercap/ > +F: kernel/configs/nopm.config > > POWER STATE COORDINATION INTERFACE (PSCI) > M: Mark Rutland <mark.rutland@arm.com> > diff --git a/kernel/configs/nopm.config b/kernel/configs/nopm.config > new file mode 100644 > index 000000000000..81ff07863576 > --- /dev/null > +++ b/kernel/configs/nopm.config > @@ -0,0 +1,15 @@ > +CONFIG_PM=n > +CONFIG_SUSPEND=n > +CONFIG_HIBERNATION=n > + > +# Triggers PM on OMAP > +CONFIG_CPU_IDLE=n > + > +# Triggers enablement via hibernate callbacks > +CONFIG_XEN=n > + > +# ARM/ARM64 architectures that select PM unconditionally > +CONFIG_ARCH_OMAP2PLUS_TYPICAL=n > +CONFIG_ARCH_RENESAS=n > +CONFIG_ARCH_TEGRA=n > +CONFIG_ARCH_VEXPRESS=n > Applied, thanks!
diff --git a/MAINTAINERS b/MAINTAINERS index 07d9067f68fd..539808dc7928 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10895,6 +10895,7 @@ F: include/linux/pm.h F: include/linux/pm_* F: include/linux/powercap.h F: drivers/powercap/ +F: kernel/configs/nopm.config POWER STATE COORDINATION INTERFACE (PSCI) M: Mark Rutland <mark.rutland@arm.com> diff --git a/kernel/configs/nopm.config b/kernel/configs/nopm.config new file mode 100644 index 000000000000..81ff07863576 --- /dev/null +++ b/kernel/configs/nopm.config @@ -0,0 +1,15 @@ +CONFIG_PM=n +CONFIG_SUSPEND=n +CONFIG_HIBERNATION=n + +# Triggers PM on OMAP +CONFIG_CPU_IDLE=n + +# Triggers enablement via hibernate callbacks +CONFIG_XEN=n + +# ARM/ARM64 architectures that select PM unconditionally +CONFIG_ARCH_OMAP2PLUS_TYPICAL=n +CONFIG_ARCH_RENESAS=n +CONFIG_ARCH_TEGRA=n +CONFIG_ARCH_VEXPRESS=n
A frequent source of build problems is poor handling of optional PM support, almost all development is done with the PM options enabled but they can be turned off. Currently few if any of the build test services do this as standard as there is no standard config for it and the use of selects and def_bool means that simply setting CONFIG_PM=n doesn't do what is expected. To make this easier provide a fragement that can be used with KCONFIG_ALLCONFIG to force PM off. CONFIG_XEN is disabled as Xen uses hibernation callbacks which end up turning on power management on architectures with Xen. Some cpuidle implementations on ARM select PM so CONFIG_CPU_IDLE is disabled, and some ARM architectures unconditionally enable PM so they are also disabled. Signed-off-by: Mark Brown <broonie@kernel.org> --- MAINTAINERS | 1 + kernel/configs/nopm.config | 15 +++++++++++++++ 2 files changed, 16 insertions(+) create mode 100644 kernel/configs/nopm.config