Re: [PATCH] drivers: base: add support to skip power management in device/driver model

[Rafael J. Wysocki]

On Tue, Feb 12, 2019 at 6:49 PM Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
>
> On Mon, Feb 11, 2019 at 05:20:20PM +0100, Ulf Hansson wrote:
> > On Thu, 7 Feb 2019 at 16:29, Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
> > >
> > > On Thu, Feb 07, 2019 at 04:18:57PM +0100, Ulf Hansson wrote:
> > > > On Thu, 7 Feb 2019 at 16:06, Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
> > >
> > > [...]
> > >
> > > > > Indeed, I was ignoring knowing that it's harmless. But more people
> > > > > started to complain, and Rafael suggested this which I agree as we
> > > > > have several pseudo devices created in the kernel that we can bypass
> > > > > some of these pm handling knowing we won't need it.
> > > >
> > > > Okay, I see.
> > > >
> > > > Anyway, I will likely need to restore part of this change, via my
> > > > cluster idling series then. As from that point, the cpu device that
> > > > you call device_set_pm_not_required() for, starts to be used from both
> > > > PM core and runtime PM point of view. But I guess that's okay then.
> > > >
> > >
> > > Ah I see. I can drop for CPU devices then. Since I didn't see any use for
> > > them, I set the flag, but I can drop it now or you can do that as part
> > > of that series.
> >
> > Well, I prefer if you drop it for CPU devices, as least for now.
> >
> > > There are quite a few devices(especially the ones
> > > registered under system subsys can set this but I would take it separate
> > > once we settle on this). Also Rafael may have seen use for few more
> > > devices when he suggested this.
> >
> > Yep, let's find another first user of this.
> >
> > Additionally, it seems like we should drop the print in device_pm_add().
> >
>
> Hi Rafael,
>
> Do you prefer to drop the error message or retain it as is ?

Do you mean the one in device_pm_add()?

> With this patch, we don't have to. I will repost v2 dropping this flags for cpus
> and just retaining the cache nodes for now.

OK