On Wed, Nov 06, 2013 at 04:28:48PM -0800, Yinghai Lu wrote:
> On Wed, Nov 6, 2013 at 12:46 PM, Bjorn Helgaas <bhelgaas@xxxxxxxxxx> wrote:
>
> > Oh, I see. The problem is when VF is on a virtual bus and the
> > corresponding PF is on a root bus. Then we have:
> >
> > pci_is_root_bus(VF) == false
> > pci_upstream_bridge(VF) == pci_upstream_bridge(PF)
> > pci_uptream_bridge(PF) == NULL
> >
> > I guess that's what your suggestion about "caching the return from
> > pci_upstream_bridge()" was about. Like this:
>
> Yes.
>
> For the two patches,
>
> Acked-by: Yinghai Lu <yinghai@xxxxxxxxxx>
Thanks, I applied these two with your acks to my pci/misc branch for
v3.13.
Bjorn
> >
> > commit 21ea57bf1311f7a8d1b755d355322a1f077fccb7
> > Author: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > Date: Wed Nov 6 10:11:48 2013 -0700
> >
> > PCI: Add pci_upstream_bridge()
> >
> > This adds a pci_upstream_bridge() interface to find the PCI-to-PCI bridge
> > upstream from a device. This is typically just "dev->bus->self", but in
> > the case of a VF on a virtual bus, we have to start from the corresponding
> > PF. Returns NULL if there is no upstream PCI bridge, i.e., if the device
> > is on a root bus.
> >
> > Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> >
> > diff --git a/include/linux/pci.h b/include/linux/pci.h
> > index d3a888a..835ec7b 100644
> > --- a/include/linux/pci.h
> > +++ b/include/linux/pci.h
> > @@ -480,6 +480,15 @@ static inline bool pci_is_root_bus(struct pci_bus *pbus)
> > return !(pbus->parent);
> > }
> >
> > +static inline struct pci_dev *pci_upstream_bridge(struct pci_dev *dev)
> > +{
> > + dev = pci_physfn(dev);
> > + if (pci_is_root_bus(dev->bus))
> > + return NULL;
> > +
> > + return dev->bus->self;
> > +}
> > +
> > #ifdef CONFIG_PCI_MSI
> > static inline bool pci_dev_msi_enabled(struct pci_dev *pci_dev)
> > {
> > commit bca690ab580d5c0c4a0c7201f3c42057288add8b
> > Author: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > Date: Wed Nov 6 10:00:51 2013 -0700
> >
> > PCI: Enable upstream bridges even for VFs on virtual buses
> >
> > Previously we enabled the upstream PCI-to-PCI bridge only when
> > "dev->bus->self != NULL". In the case of a VF on a virtual bus, where
> > "bus->self == NULL", we didn't enable the upstream bridge.
> >
> > This fixes that by enabling the upstream bridge of the PF corresponding to
> > the VF.
> >
> > Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> >
> > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> > index ac40f90..d3ed931 100644
> > --- a/drivers/pci/pci.c
> > +++ b/drivers/pci/pci.c
> > @@ -1148,12 +1148,12 @@ int pci_reenable_device(struct pci_dev *dev)
> >
> > static void pci_enable_bridge(struct pci_dev *dev)
> > {
> > + struct pci_dev *bridge;
> > int retval;
> >
> > - if (!dev)
> > - return;
> > -
> > - pci_enable_bridge(dev->bus->self);
> > + bridge = pci_upstream_bridge(dev);
> > + if (bridge)
> > + pci_enable_bridge(bridge);
> >
> > if (pci_is_enabled(dev)) {
> > if (!dev->is_busmaster)
> > @@ -1170,6 +1170,7 @@ static void pci_enable_bridge(struct pci_dev *dev)
> >
> > static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags)
> > {
> > + struct pci_dev *bridge;
> > int err;
> > int i, bars = 0;
> >
> > @@ -1188,7 +1189,9 @@ static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags)
> > if (atomic_inc_return(&dev->enable_cnt) > 1)
> > return 0; /* already enabled */
> >
> > - pci_enable_bridge(dev->bus->self);
> > + bridge = pci_upstream_bridge(dev);
> > + if (bridge)
> > + pci_enable_bridge(bridge);
> >
> > /* only skip sriov related */
> > for (i = 0; i <= PCI_ROM_RESOURCE; i++)
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
