Re: [PATCH v3 25/29] hw/arm/virt: Let virt support pci hotplug/unplug GED event

[Jonathan Cameron via <qemu-devel@nongnu.org>]

On Mon, 16 Jun 2025 11:46:54 +0200
Eric Auger <eric.auger@redhat.com> wrote:

> Set up the IO registers used to communicate between QEMU
> and ACPI.
> 
> Signed-off-by: Eric Auger <eric.auger@redhat.com>
> 
> ---
> v2 -> v3:
> - remove acpi_ged_state->pcihp_state.use_acpi_hotplug_bridge = true;
> - use sysbus_mmio_map_name for all regs (Igor)

I'm very much in favor of this change but maybe break them out to
a separate patch - perhaps even one that can run ahead of the rest
of this series?

Mind you if this is going to land shortly perhaps not worth the bother.

Either way

Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>


> - create_pcie left at its original place
> 
> v1 -> v2:
> - use ACPI_PCIHP_REGION_NAME
> ---
>  include/hw/acpi/generic_event_device.h |  1 +
>  hw/arm/virt.c                          | 19 +++++++++++++++++--
>  2 files changed, 18 insertions(+), 2 deletions(-)
> 
> diff --git a/include/hw/acpi/generic_event_device.h b/include/hw/acpi/generic_event_device.h
> index ec8e1abe0a..8f5d903146 100644
> --- a/include/hw/acpi/generic_event_device.h
> +++ b/include/hw/acpi/generic_event_device.h
> @@ -111,6 +111,7 @@ typedef struct GEDState {
>  } GEDState;
>  
>  #define ACPI_PCIHP_REGION_NAME "pcihp container"
> +#define ACPI_MEMHP_REGION_NAME "memhp container"
>  
>  struct AcpiGedState {
>      SysBusDevice parent_obj;
> diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> index 41be8f6dbb..8c882e0794 100644
> --- a/hw/arm/virt.c
> +++ b/hw/arm/virt.c
> @@ -684,6 +684,8 @@ static inline DeviceState *create_acpi_ged(VirtMachineState *vms)
>      DeviceState *dev;
>      MachineState *ms = MACHINE(vms);
>      SysBusDevice *sbdev;
> +    AcpiGedState *acpi_ged_state;
> +    AcpiPciHpState *pcihp_state;
>      int irq = vms->irqmap[VIRT_ACPI_GED];
>      uint32_t event = ACPI_GED_PWR_DOWN_EVT;
>  
> @@ -696,13 +698,26 @@ static inline DeviceState *create_acpi_ged(VirtMachineState *vms)
>      }
>  
>      dev = qdev_new(TYPE_ACPI_GED);
> +    acpi_ged_state = ACPI_GED(dev);
> +    pcihp_state = &acpi_ged_state->pcihp_state;
> +    if (pcihp_state->use_acpi_hotplug_bridge) {
> +        event |= ACPI_GED_PCI_HOTPLUG_EVT;
> +    }
>      qdev_prop_set_uint32(dev, "ged-event", event);
>      object_property_set_link(OBJECT(dev), "bus", OBJECT(vms->bus), &error_abort);
>      sbdev = SYS_BUS_DEVICE(dev);
>      sysbus_realize_and_unref(sbdev, &error_fatal);
>  
> -    sysbus_mmio_map(sbdev, 0, vms->memmap[VIRT_ACPI_GED].base);
> -    sysbus_mmio_map(sbdev, 1, vms->memmap[VIRT_PCDIMM_ACPI].base);
> +    sysbus_mmio_map_name(sbdev, TYPE_ACPI_GED, vms->memmap[VIRT_ACPI_GED].base);
> +    sysbus_mmio_map_name(sbdev, ACPI_MEMHP_REGION_NAME,
> +                         vms->memmap[VIRT_PCDIMM_ACPI].base);
> +    if (pcihp_state->use_acpi_hotplug_bridge) {
> +        int pcihp_region_index;
> +
> +        pcihp_region_index = sysbus_mmio_map_name(sbdev, ACPI_PCIHP_REGION_NAME,
> +                                                  vms->memmap[VIRT_ACPI_PCIHP].base);
> +        assert(pcihp_region_index >= 0);
> +    }
>      sysbus_connect_irq(sbdev, 0, qdev_get_gpio_in(vms->gic, irq));
>  
>      return dev;