Il 04/06/2013 14:24, Peter Maydell ha scritto:
> On 4 June 2013 13:13, Paolo Bonzini <pbonz...@redhat.com> wrote:
>> This matches sysbus_pass_irq in cases where a device is a thin wrapper
>> of another. MMIO regions will keep the subdevice as the owner.
>>
>> Signed-off-by: Paolo Bonzini <pbonz...@redhat.com>
>> ---
>> hw/core/sysbus.c | 12 ++++++++++++
>> hw/cpu/arm11mpcore.c | 2 +-
>> include/hw/sysbus.h | 1 +
>> 3 files changed, 14 insertions(+), 1 deletion(-)
>>
>> diff --git a/hw/core/sysbus.c b/hw/core/sysbus.c
>> index 9004d8c..6dbd1f8 100644
>> --- a/hw/core/sysbus.c
>> +++ b/hw/core/sysbus.c
>> @@ -117,6 +117,18 @@ void sysbus_init_mmio(SysBusDevice *dev, MemoryRegion
>> *memory)
>> dev->mmio[n].memory = memory;
>> }
>>
>> +/* Pass MMIOs from a target device. */
>> +void sysbus_pass_mmio(SysBusDevice *dev, SysBusDevice *target)
>> +{
>> + int i;
>> + assert(dev->num_mmio == 0);
>> + dev->num_mmio = target->num_mmio;
>> + for (i = 0; i < dev->num_mmio; i++) {
>> + assert(target->mmio[i].addr == -1);
>> + dev->mmio[i] = target->mmio[i];
>> + }
>> +}
>
> This is much less flexible than just using sysbus_mmio_get_region(),
> because it only lets you pass the whole set of MMIOs from the
> other device through, not just the ones you want.
How is this different from sysbus_pass_irq?
> Please just make reference counting work properly with passing
> MemoryRegion*s around.
Do you have any idea that doesn't require touch 800 invocation of the
region creation functions? This looks like a solution in search of a
problem to me.
Paolo
