Re: [PATCH 2/12] PCI / PM: Propagate wake-up enable for PCIe devices too

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

On Tuesday 05 January 2010, Jesse Barnes wrote:
> On Sun, 27 Dec 2009 21:00:33 +0100
> "Rafael J. Wysocki" <rjw@sisk.pl> wrote:
> 
> > From: Rafael J. Wysocki <rjw@sisk.pl>
> > 
> > Having read the PM part of the PCIe 2.0 specification more carefully
> > I think that it was a mistake to restrict the wake-up enable
> > propagation to non-PCIe devices, because if we do not request
> > control of the root ports' PME registers via OSC, PCIe PME is
> > supposed to be handled by the platform, just like the non-PCIe PME.
> > Even if we do that, the wake-up propagation is done to allow the
> > devices to wake up the system from sleep states which involves the
> > platform anyway, so it won't hurt.
> > 
> > Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
> > ---
> >  drivers/pci/pci-acpi.c |   10 ++--------
> >  1 file changed, 2 insertions(+), 8 deletions(-)
> > 
> > Index: linux-2.6/drivers/pci/pci-acpi.c
> > ===================================================================
> > --- linux-2.6.orig/drivers/pci/pci-acpi.c
> > +++ linux-2.6/drivers/pci/pci-acpi.c
> > @@ -112,11 +112,7 @@ static bool acpi_pci_can_wakeup(struct p
> >  static void acpi_pci_propagate_wakeup_enable(struct pci_bus *bus,
> > bool enable) {
> >  	while (bus->parent) {
> > -		struct pci_dev *bridge = bus->self;
> > -		int ret;
> > -
> > -		ret = acpi_pm_device_sleep_wake(&bridge->dev,
> > enable);
> > -		if (!ret || pci_is_pcie(bridge))
> > +		if (!acpi_pm_device_sleep_wake(&bus->self->dev,
> > enable)) return;
> >  		bus = bus->parent;
> >  	}
> > @@ -131,9 +127,7 @@ static int acpi_pci_sleep_wake(struct pc
> >  	if (acpi_pci_can_wakeup(dev))
> >  		return acpi_pm_device_sleep_wake(&dev->dev, enable);
> >  
> > -	if (!pci_is_pcie(dev))
> > -		acpi_pci_propagate_wakeup_enable(dev->bus, enable);
> > -
> > +	acpi_pci_propagate_wakeup_enable(dev->bus, enable);
> >  	return 0;
> >  }
> 
> Seems ok to try it out, but as usual with this sort of thing, the real
> question is, "What does Windows do?", since wakeup enable ACPI support
> for PCIe devices is likely broken unless Windows uses it.

Well, that's in the Linus' tree already, so I guess we're going to learn. ;-)

Seriously, I think that Windows actually uses ACPI wake-up support for PCIe
devices.  On my Toshiba Portege R500 the BIOS doesn't even allow us to use
the native PCIe PME mechanism directly, so the patch is correct IMO.

Rafael