On 11/10/2020 20:03, Philippe Mathieu-Daudé wrote:
> The Grackle PCI host model expects the interrupt controller
> being set, but does not verify it is present. Add a check to
> help developers using this model.
>
> Signed-off-by: Philippe Mathieu-Daudé <f4...@amsat.org>
> ---
> hw/pci-host/grackle.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/hw/pci-host/grackle.c b/hw/pci-host/grackle.c
> index 57c29b20afb..20361d215ca 100644
> --- a/hw/pci-host/grackle.c
> +++ b/hw/pci-host/grackle.c
> @@ -76,6 +76,10 @@ static void grackle_realize(DeviceState *dev, Error **errp)
> GrackleState *s = GRACKLE_PCI_HOST_BRIDGE(dev);
> PCIHostState *phb = PCI_HOST_BRIDGE(dev);
>
> + if (!s->pic) {
> + error_setg(errp, TYPE_GRACKLE_PCI_HOST_BRIDGE ": 'pic' link not
> set");
> + return;
> + }
> phb->bus = pci_register_root_bus(dev, NULL,
> pci_grackle_set_irq,
> pci_grackle_map_irq,
Reviewing this now, my feeling is that both grackle and uninorth are doing the
wrong
thing here by passing in a link to the PIC - certainly if I had written this
more
recently than I originally did, I would simply use qdev_init_gpio_out() for the
IRQ
lines and do the wiring during machine init.
I've got a few other things to look at first, but I'll post a patchset later
which I
think will improve this for both Mac machines.
ATB,
Mark.
