Re: [PATCH] uv/x2apic: update for change in pci bridge handling.

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Thu, 9 Jun 2011 13:43:52 +1000
Dave Airlie <airlied@gmail.com> wrote:

> On Wed, May 25, 2011 at 2:00 PM, Dave Airlie <airlied@gmail.com> wrote:
> > From: Dave Airlie <airlied@redhat.com>
> >
> > I forgot about the special uv handling code for this, so this
> > patch fixes it up.
> 
> ping?
> 
> This fixes a warning and bug in Linus tree,
> 
>   CC      arch/x86/kernel/apic/x2apic_uv_x.o
> /home/airlied/git/linux-2.6/arch/x86/kernel/apic/x2apic_uv_x.c: In
> function ‘uv_system_init’:
> /home/airlied/git/linux-2.6/arch/x86/kernel/apic/x2apic_uv_x.c:882:2:
> warning: passing argument 1 of ‘pci_register_set_vga_state’ from
> incompatible pointer type [enabled by default]
> /home/airlied/git/linux-2.6/include/linux/pci.h:1094:13: note:
> expected ‘arch_set_vga_state_t’ but argument is of type ‘int
> (*)(struct pci_dev *, bool,  unsigned int,  bool)’
> 
> 
> Dave.
> 
> >
> > Signed-off-by: Dave Airlie <airlied@redhat.com>
> > ---
> >  arch/x86/kernel/apic/x2apic_uv_x.c |    8 ++++----
> >  drivers/pci/pci.c                  |    4 ++--
> >  2 files changed, 6 insertions(+), 6 deletions(-)
> >
> > diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
> > index 33b10a0..185cd1e 100644
> > --- a/arch/x86/kernel/apic/x2apic_uv_x.c
> > +++ b/arch/x86/kernel/apic/x2apic_uv_x.c
> > @@ -599,14 +599,14 @@ late_initcall(uv_init_heartbeat);
> >
> >  /* Direct Legacy VGA I/O traffic to designated IOH */
> >  int uv_set_vga_state(struct pci_dev *pdev, bool decode,
> > -                     unsigned int command_bits, bool change_bridge)
> > +                     unsigned int command_bits, u32 flags)
> >  {
> >        int domain, bus, rc;
> >
> > -       PR_DEVEL("devfn %x decode %d cmd %x chg_brdg %d\n",
> > -                       pdev->devfn, decode, command_bits, change_bridge);
> > +       PR_DEVEL("devfn %x decode %d cmd %x flags %d\n",
> > +                       pdev->devfn, decode, command_bits, flags);
> >
> > -       if (!change_bridge)
> > +       if (!(flags & PCI_VGA_STATE_CHANGE_BRIDGE))
> >                return 0;
> >
> >        if ((command_bits & PCI_COMMAND_IO) == 0)
> > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> > index a339237..4528ee7 100644
> > --- a/drivers/pci/pci.c
> > +++ b/drivers/pci/pci.c
> > @@ -2862,11 +2862,11 @@ void __init pci_register_set_vga_state(arch_set_vga_state_t func)
> >  }
> >
> >  static int pci_set_vga_state_arch(struct pci_dev *dev, bool decode,
> > -                     unsigned int command_bits, bool change_bridge)
> > +                     unsigned int command_bits, u32 flags)
> >  {
> >        if (arch_set_vga_state)
> >                return arch_set_vga_state(dev, decode, command_bits,
> > -                                               change_bridge);
> > +                                               flags);
> >        return 0;
> >  }

Yeah, this is fine with me, you can push it through your tree.

Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>

-- 
Jesse Barnes, Intel Open Source Technology Center