On Fri 18 Jan 09:39 PST 2019, Stephen Boyd wrote:

> Quoting Rajendra Nayak (2019-01-17 20:48:01)
> >  drivers/soc/qcom/Kconfig | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig
> > index a5d5167c3f16..1ee298f6bf17 100644
> > --- a/drivers/soc/qcom/Kconfig
> > +++ b/drivers/soc/qcom/Kconfig
> > @@ -109,7 +109,7 @@ config QCOM_RPMHPD
> >  
> >  config QCOM_RPMPD
> >         bool "Qualcomm RPM Power domain driver"
> 
> Just curious, does it need to be bool for some reason?
> 

It's unfortunately not possible to have any genpd, iommu or pinctrl
drivers compiled as modules, because once you pass lateinit probe
deferral is purposefully broken. See
driver_deferred_probe_check_state().

This also means that if you're unlucky and your kernel reached lateinit
before the SMD communication with RPM is established and has brought up
the rpmpd, you are left with a completely broken system. Unfortunately
this isn't that hard to reproduce with a minimal defconfig.

Regards,
Bjorn

Reply via email to