Re: [PATCH v1 4/9] PM: Move pm_runtime_force_suspend/resume() under CONFIG_PM_SLEEP

[Rafael J. Wysocki]

On Thu, Jun 26, 2025 at 12:05 PM Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
>
> On Thu, 26 Jun 2025 at 11:41, Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
> >
> > On Thu, Jun 26, 2025 at 11:38 AM Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
> > >
> > > On Wed, 25 Jun 2025 at 21:25, Rafael J. Wysocki <rjw@xxxxxxxxxxxxx> wrote:
> > > >
> > > > From: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > > >
> > > > Since pm_runtime_force_suspend/resume() and pm_runtime_need_not_resume()
> > > > are only used during system-wide PM transitions, there is no reason to
> > > > compile them in if CONFIG_PM_SLEEP is unset.
> > > >
> > > > Accordingly, move them all under CONFIG_PM_SLEEP and make the static
> > > > inline stubs for pm_runtime_force_suspend/resume() return an error
> > > > to indicate that they should not be used outside CONFIG_PM_SLEEP.
> > > >
> > >
> > > Just realized that there seems to be some drivers that actually make
> > > use of pm_runtime_force_suspend() from their ->remove() callbacks.
> > >
> > > To not break them, we probably need to leave this code to stay under CONFIG_PM.
> >
> > OK, pm_runtime_force_suspend() need not be under CONFIG_PM_SLEEP.
> > That's not the case for the other two functions though AFAICS.
>
> Right, but maybe better to keep them to avoid confusion?

There really is no point holding pm_runtime_need_not_resume() outside
CONFIG_PM_SLEEP and pm_runtime_force_resume() really should not be
used anywhere outside system resume flows.

> At least the corresponding flag is needed.

What flag do you mean?  If pm_runtime_force_suspend() does not go
under CONFIG_PM_SLEEP, needs_force_resume will not go under it either
(so I'll drop the next patch altogether).

Cheers, Rafael