Re: [PATCH] byte-addressable grub_pci_make_address

[Seth Goldberg]

 What's the basis for these writes, btw?  I askesd a while ago but no one 
seemed to know.  Feel free to point me to a chipset datasheet, but right now 
this code truly a black box :).

 --S

Quoting Robert Millan, who wrote the following on Thu, 24 Dec 2009:

> On Tue, Dec 22, 2009 at 03:12:09PM +0100, Vladimir 'φ-coder/phcoder' Serbinenko 
> wrote:
> > === modified file 'commands/efi/loadbios.c'
> > --- commands/efi/loadbios.c     2009-12-21 22:06:04 +0000
> > +++ commands/efi/loadbios.c     2009-12-22 12:03:12 +0000
> > @@ -50,7 +50,7 @@
> >        return 0;
> >      }
> >
> > -  addr = grub_pci_make_address (dev, 36);
> > +  addr = grub_pci_make_address (dev, 144);
> >    grub_pci_write_byte (addr++, 0x30);
> >    grub_pci_write_byte (addr++, 0x33);
> >    grub_pci_write_byte (addr++, 0x33);
> > @@ -76,7 +76,7 @@
> >    grub_pci_address_t addr;
> >    grub_pci_device_t dev = { .bus = 0, .device = 0, .function = 0};
> >
> > -  addr = grub_pci_make_address (dev, 36);
> > +  addr = grub_pci_make_address (dev, 144);
>
> These look like black magic.  Could they be macroified instead?
>
> --
> Robert Millan
>
>  "Be the change you want to see in the world" -- Gandhi
>
>
> _______________________________________________
> Grub-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/grub-devel