On Tue, Sep 13, 2016 at 05:11:49PM +0300, Andy Shevchenko wrote:
> On Tue, 2016-09-13 at 15:19 +0300, Mika Westerberg wrote:
> > Add a new helper function pci_find_resource() that can be used to find
> > out
> > whether a given resource (for example from a child device) is
> > contained
> > within given PCI device's standard resources.
> >
> > Signed-off-by: Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx>
> > ---
> > drivers/pci/pci.c | 27 +++++++++++++++++++++++++++
> > include/linux/pci.h | 4 ++++
> > 2 files changed, 31 insertions(+)
> >
> > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> > index aab9d5115a5f..491f879f34cb 100644
> > --- a/drivers/pci/pci.c
> > +++ b/drivers/pci/pci.c
> > @@ -480,6 +480,33 @@ struct resource *pci_find_parent_resource(const
> > struct pci_dev *dev,
> > EXPORT_SYMBOL(pci_find_parent_resource);
> >
> > /**
> > + * pci_find_resource - Return matching PCI device resource
> > + * @dev: PCI device to query
> > + * @res: Resource to look for
> > + *
> > + * Goes over standard PCI resources (BARs) and checks if the given
> > resource
> > + * is partially or fully contained in any of them. In that case the
> > + * matching resource is returned, %NULL otherwise.
> > + */
> > +struct resource *pci_find_resource(struct pci_dev *dev, struct
> > resource *res)
> > +{
> > + int i;
> > +
>
> > + if (!res)
> > + return NULL;
>
> Shouldn't it be a problem of caller to supply valid pointer?
> Seems other function(s) has(ve) this assumption.
No, I can drop the check.
> > +
> > + for (i = 0; i < PCI_ROM_RESOURCE; i++) {
> > + struct resource *r = &dev->resource[i];
> > +
> > + if (r->start && resource_contains(r, res))
> > + return r;
>
> I'm not sure what we have to return in case of
> 1) r->start == 0, r->end > 0
> 2) r->start > 0, r->end == 0
>
> assuming that all previous checks are positive.
These are PCI BARs so if they are not populated (r->start == 0) we skip
them. PCI core should have filled those already with correct values (and
resource_contains() should deal with everything that does not match
anyway).
--
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
