Re: [PATCH] PM: runtime: Simplify __rpm_get_callback()

[Bjorn Helgaas]

On Tue, Jan 24, 2023 at 03:45:50PM +0100, Rafael J. Wysocki wrote:
> On Tue, Jan 24, 2023 at 3:37 PM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> > On Tue, Jan 24, 2023 at 3:18 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
> > > On Tue, Jan 24, 2023 at 12:20 PM Geert Uytterhoeven
> > > <geert@xxxxxxxxxxxxxx> wrote:
> > > > On Thu, Dec 15, 2022 at 7:23 PM Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote:
> > > > > From: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > > > >
> > > > > Simplify __rpm_get_callback() slightly by returning as soon as the return
> > > > > value is known.  No functional change intended.
> > > > >
> > > > > Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > > >
> > > > Thanks for your patch, which is now commit 650bdddb6b311705 ("PM:
> > > > runtime: Simplify __rpm_get_callback()") in pm/linux-next.
> > > >
> > > > > --- a/drivers/base/power/runtime.c
> > > > > +++ b/drivers/base/power/runtime.c
> > > > > @@ -20,8 +20,7 @@ typedef int (*pm_callback_t)(struct device *);
> > > > >
> > > > >  static pm_callback_t __rpm_get_callback(struct device *dev, size_t cb_offset)
> > > > >  {
> > > > > -       pm_callback_t cb;
> > > > > -       const struct dev_pm_ops *ops;
> > > > > +       const struct dev_pm_ops *ops = NULL;
> > > > >
> > > > >         if (dev->pm_domain)
> > > > >                 ops = &dev->pm_domain->ops;
> > > > > @@ -31,18 +30,14 @@ static pm_callback_t __rpm_get_callback(struct device *dev, size_t cb_offset)
> > > > >                 ops = dev->class->pm;
> > > > >         else if (dev->bus && dev->bus->pm)
> > > > >                 ops = dev->bus->pm;
> > > > > -       else
> > > > > -               ops = NULL;
> > > > >
> > > > >         if (ops)
> > > > > -               cb = *(pm_callback_t *)((void *)ops + cb_offset);
> > > > > -       else
> > > > > -               cb = NULL;
> > > > > +               return *(pm_callback_t *)((void *)ops + cb_offset);
> > > >
> > > > This is a change in behavior in case the callback turns out to be NULL:
> > > >   - before, it would fall back to the driver-specific callback below,
> > > >   - after, it always returns NULL.
> > >
> > > Good point and sorry for missing this!
> > >
> > > > >
> > > > > -       if (!cb && dev->driver && dev->driver->pm)
> > > > > -               cb = *(pm_callback_t *)((void *)dev->driver->pm + cb_offset);
> > > > > +       if (dev->driver && dev->driver->pm)
> > > > > +               return *(pm_callback_t *)((void *)dev->driver->pm + cb_offset);
> > > > >
> > > > > -       return cb;
> > > > > +       return NULL;
> > > > >  }
> > > >
> > >
> > > Something like the patch below (modulo gmail-induced whitespace
> > > breakage) should restore the previous behavior if I'm not mistaken:
> > >
> > > ---
> > >  drivers/base/power/runtime.c |    9 +++++++--
> > >  1 file changed, 7 insertions(+), 2 deletions(-)
> > >
> > > Index: linux-pm/drivers/base/power/runtime.c
> > > ===================================================================
> > > --- linux-pm.orig/drivers/base/power/runtime.c
> > > +++ linux-pm/drivers/base/power/runtime.c
> > > @@ -31,8 +31,13 @@ static pm_callback_t __rpm_get_callback(
> > >      else if (dev->bus && dev->bus->pm)
> > >          ops = dev->bus->pm;
> > >
> > > -    if (ops)
> > > -        return *(pm_callback_t *)((void *)ops + cb_offset);
> > > +    if (ops) {
> > > +        pm_callback_t cb;
> > > +
> > > +        cb = *(pm_callback_t *)((void *)ops + cb_offset);
> > > +        if (cb)
> > > +            return cb;
> > > +    }
> > >
> > >      if (dev->driver && dev->driver->pm)
> > >          return *(pm_callback_t *)((void *)dev->driver->pm + cb_offset);
> >
> > Which is now more complex than the original?
> 
> Arguably so.
> 
> OK, I'll drop the commit in question then, sorry Bjorn.

Really sorry about this.  Think I'm all out of brown paper bags :(
Thanks for catching this, Geert, and sorry for all the time you had to
spend debugging it.

Bjorn