Re: [PATCH 10/12] hw/qdev: Ensure parent device is not realized before adding bus

[Thomas Huth]

On 18/10/2023 16.11, Philippe Mathieu-Daudé wrote:
> qbus_new() should not be called on realized device.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
>   hw/core/bus.c | 7 +++++++
>   1 file changed, 7 insertions(+)
>
> diff --git a/hw/core/bus.c b/hw/core/bus.c
> index c7831b5293..c92d07667b 100644
> --- a/hw/core/bus.c
> +++ b/hw/core/bus.c
> @@ -21,6 +21,7 @@
>   #include "hw/qdev-properties.h"
>   #include "qemu/ctype.h"
>   #include "qemu/module.h"
> +#include "qemu/error-report.h"
>   #include "qapi/error.h"
>   
>   void qbus_set_hotplug_handler(BusState *bus, Object *handler)
> @@ -163,6 +164,12 @@ BusState *qbus_new(const char *typename, DeviceState 
> *parent, const char *name)
>   {
>       BusState *bus;
>   
> +    if (parent->realized) {
> +        error_report("qbus_new(type:%s parent:%s, name:%s) but parent 
> realized",
> +                     typename, object_get_typename(OBJECT(parent)), name);
> +        abort();
> +    }
> +
>       bus = BUS(object_new(typename));
>       qbus_init_internal(bus, parent, name);
>   

Sounds like a good idea!

Reviewed-by: Thomas Huth <thuth@redhat.com>