Re: [Qemu-devel] [PATCH 02/34] pci: handle BAR mapping at PCI level

[Blue Swirl]

On Fri, Jul 23, 2010 at 10:40 AM, Isaku Yamahata <address@hidden> wrote:
> On Thu, Jul 22, 2010 at 09:54:46PM +0000, Blue Swirl wrote:
>> diff --git a/hw/pci.c b/hw/pci.c
>> index a98d6f3..49f03fb 100644
>> --- a/hw/pci.c
>> +++ b/hw/pci.c
> ...
>> @@ -817,6 +825,25 @@ void pci_register_bar(PCIDevice *pci_dev, int 
>> region_num,
>>          pci_set_long(pci_dev->wmask + addr, wmask & 0xffffffff);
>>          pci_set_long(pci_dev->cmask + addr, 0xffffffff);
>>      }
>> +    pci_bar_map(pci_dev, region_num, 0, 0, size, -1);
>> +}
>> +
>> +void pci_bar_map(PCIDevice *pci_dev, int region_num, int subregion_num,
>> +                 pcibus_t offset, pcibus_t size, int ix)
>> +{
>> +    PCIIOSubRegion *s;
>> +
>> +    if ((unsigned int)region_num >= PCI_NUM_REGIONS ||
>> +        (unsigned int)subregion_num >= PCI_NUM_SUBREGIONS) {
>> +        return;
>> +    }
>
> abort() or assert()? It's caller's bug.

Yes. I copied this from pci_register_bar(), which should also have
assert() or abort().