[PATCH v12 08/12] PCI: OF: Introduce helper function for retrieving PCI domain numbers

[Liviu Dudau]

On Fri, Sep 26, 2014 at 10:53:26PM +0100, Rob Herring wrote:
> On Fri, Sep 26, 2014 at 4:20 PM,  <liviu at dudau.co.uk> wrote:
> > On Fri, Sep 26, 2014 at 01:20:39PM -0500, Rob Herring wrote:
> >> On Tue, Sep 23, 2014 at 2:01 PM, Liviu Dudau <Liviu.Dudau at arm.com> wrote:
> >> > Add pci_get_new_domain_nr() to allocate a new domain number
> >> > and of_get_pci_domain_nr() to retrieve the PCI domain number
> >> > of a given device from DT. Host bridge drivers or architecture
> >> > specific code can choose to implement their PCI domain number
> >> > policy using these two functions.
> >> >
> >> > Using of_get_pci_domain_nr() guarantees a stable PCI domain
> >> > number on every boot provided that all host bridge controllers
> >> > are assigned a number in the device tree using "linux,pci-domain"
> >> > property. Mix use of pci_get_new_domain_nr() and of_get_pci_domain_nr()
> >> > is not recommended as it can lead to potentially conflicting
> >> > domain numbers being assigned to root busses behind different
> >> > host bridges.
> >> >
> >> > Cc: Bjorn Helgaas <bhelgaas at google.com>
> >> > Cc: Arnd Bergmann <arnd at arndb.de>
> >> > Cc: Grant Likely <grant.likely at linaro.org>
> >> > Cc: Rob Herring <robh+dt at kernel.org>
> >> > Cc: Catalin Marinas <catalin.marinas at arm.com>
> >> > Signed-off-by: Liviu Dudau <Liviu.Dudau at arm.com>
> >>
> >> Looks pretty good, but a couple of comments that can be addressed as
> >> follow-up patches.
> >>
> >> > ---
> >> >  drivers/of/of_pci.c    | 25 +++++++++++++++++++++++++
> >> >  drivers/pci/pci.c      |  9 +++++++++
> >> >  include/linux/of_pci.h |  7 +++++++
> >> >  include/linux/pci.h    |  3 +++
> >> >  4 files changed, 44 insertions(+)
> >> >
> >> > diff --git a/drivers/of/of_pci.c b/drivers/of/of_pci.c
> >> > index 8481996..82d172f 100644
> >> > --- a/drivers/of/of_pci.c
> >> > +++ b/drivers/of/of_pci.c
> >> > @@ -89,6 +89,31 @@ int of_pci_parse_bus_range(struct device_node *node, struct resource *res)
> >> >  }
> >> >  EXPORT_SYMBOL_GPL(of_pci_parse_bus_range);
> >> >
> >> > +/**
> >> > + * This function will try to obtain the host bridge domain number by
> >> > + * finding a property called "linux,pci-domain" of the given device node.
> >> > + *
> >> > + * @node: device tree node with the domain information
> >> > + *
> >> > + * Returns the associated domain number from DT in the range [0-0xffff], or
> >> > + * a negative value if the required property is not found.
> >> > + */
> >> > +int of_get_pci_domain_nr(struct device_node *node)
> >> > +{
> >> > +       const __be32 *value;
> >> > +       int len;
> >> > +       u16 domain;
> >> > +
> >> > +       value = of_get_property(node, "linux,pci-domain", &len);
> >>
> >> This needs to be documented.
> >
> > I would say needs agreed on *and* documented.
> 
> I thought Bjorn already applied the series. Who's agreement are you
> waiting for specifically.

Looks like he only applied about 60% of the series.

I'm not waiting on anyones specifig agreement, just general agreement.
Jason Gunthrope have initially suggested using pci-domain alias which
you don't like, now I'm using "linux,pci-domain" property which you've
ACKed, I wonder if I need to give people more time to catch up to the
fact?

> 
> >>
> >> > +       if (!value || len < sizeof(*value))
> >> > +               return -EINVAL;
> >> > +
> >> > +       domain = (u16)be32_to_cpup(value);
> >>
> >> of_property_read_u32 would not work here?
> >
> > It should, I guess, but I want to limit the domain number range to 16 bits.
> 
> Okay, but it would still be fewer lines to use of_property_read_u32
> and then cast it. You could use the u16 variant and make the property
> be defined to be 16-bit in the documentation.

True. Will do.

Best regards,
Liviu

> 
> Rob
> 

-- 
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ¯\_(ツ)_/¯