Re: [RFC][PATCH 2/10] PCI PM: Add suspend counterpart of pci_reenable_device

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Monday 05 January 2009, Pavel Machek wrote:
> Hi!
> 
> > PCI devices without drivers are not disabled during suspend and
> > hibernation, but they are enabled during resume, with the help of
> > pci_reenable_device(), so there is an unbalanced execution of
> > pcibios_enable_device() in the resume code path.
> > 
> > To correct this introduce function pci_disable_enabled_device()
> > that will disable the argument device, if it is enabled when the
> > function is being run, without updating the device's pci_dev
> > structure and use it in the suspend code path to balance the
> > pci_reenable_device() executed during resume.
> 
> > +/**
> > + * pci_disable_enabled_device - Disable device without updating enable_cnt
> > + * @dev: PCI device to disable
> > + *
> > + * NOTE: This function is a backend of PCI power management routines and is
> > + * not supposed to be called drivers.
> 
> "by drivers"?

Yes, thanks.

> > @@ -441,7 +455,10 @@ static int pci_pm_suspend(struct device 
> >  		}
> >  	} else if (pci_has_legacy_pm_support(pci_dev)) {
> >  		error = pci_legacy_suspend(dev, PMSG_SUSPEND);
> > +	} else {
> > +		pci_default_pm_suspend_early(pci_dev);
> >  	}
> > +
> >  	pci_fixup_device(pci_fixup_suspend, pci_dev);
> >  
> >  	return error;
> 
> So tre fixup runs on the disabled device?

Yes, but the device's config space is still available at this point.
 
> Acked-by: Pavel Machek <Pavel@suse.cz>

Thanks,
Rafael