Re: [Qemu-devel] [PATCH 03/15] sun4u: move ISABus inside of EBusState

[Artyom Tarasenko]

On Fri, Nov 17, 2017 at 2:42 PM, Mark Cave-Ayland
<address@hidden> wrote:
> Since the EBus is effectively a PCI-ISA bridge then the underlying ISA bus
> should be contained within the PCI bridge itself.

While it's like that on the Sabre chipset, the Spitfire chipset (which
I hope to add at some point) has the EBus, but no PCI, so maybe it's
better to model it separately.
On the other hand, the Spitfire has different EBus devices
(particularly different type of the serial ports), so I'm not sure.

> Signed-off-by: Mark Cave-Ayland <address@hidden>
> ---
>  hw/sparc64/sun4u.c |    7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/hw/sparc64/sun4u.c b/hw/sparc64/sun4u.c
> index 394b7d6..63b4aaa 100644
> --- a/hw/sparc64/sun4u.c
> +++ b/hw/sparc64/sun4u.c
> @@ -84,6 +84,7 @@ typedef struct EbusState {
>      /*< private >*/
>      PCIDevice parent_obj;
>
> +    ISABus *isa_bus;
>      MemoryRegion bar0;
>      MemoryRegion bar1;
>  } EbusState;
> @@ -245,8 +246,10 @@ static void ebus_realize(PCIDevice *pci_dev, Error 
> **errp)
>  {
>      EbusState *s = EBUS(pci_dev);
>
> -    if (!isa_bus_new(DEVICE(pci_dev), get_system_memory(),
> -                     pci_address_space_io(pci_dev), errp)) {
> +    s->isa_bus = isa_bus_new(DEVICE(pci_dev), get_system_memory(),
> +                             pci_address_space_io(pci_dev), errp);
> +    if (!s->isa_bus) {
> +        error_setg(errp, "unable to instantiate EBUS ISA bus");
>          return;
>      }
>
> --
> 1.7.10.4
>



-- 
Regards,
Artyom Tarasenko

SPARC and PPC PReP under qemu blog: http://tyom.blogspot.com/search/label/qemu