Re: [PATCH v3] hw/core/qdev: Increase qdev_realize() kindness

[Philippe Mathieu-Daudé]

On 7/7/20 5:33 AM, Philippe Mathieu-Daudé wrote:
> Since commit 510ef98dca5, qdev_realize() aborts if bus-less device
> is realized on a bus. While commits 514db7710b..007d1dbf72 took
> care of converting all mainstream uses, QEMU forks weren't. These

I guess I missed "weren't [converted]".

> forks are usually maintained by hobbyist with interest in following
> mainstream development, but with limited time, so usually rebase
> from time to time. To avoid them to spend time on debugging and
> reading git-log history, display a kind hint about what is wrong.
> 
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> ---
> Since v2:
> - scratch __func__ (armbru)
> - reword to justify this is not an impossible case (armbru)
> ---
>  hw/core/qdev.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/hw/core/qdev.c b/hw/core/qdev.c
> index 2131c7f951..a16f1270f1 100644
> --- a/hw/core/qdev.c
> +++ b/hw/core/qdev.c
> @@ -392,8 +392,11 @@ bool qdev_realize(DeviceState *dev, BusState *bus, Error 
> **errp)
>  
>      if (bus) {
>          qdev_set_parent_bus(dev, bus);
> -    } else {
> -        assert(!DEVICE_GET_CLASS(dev)->bus_type);
> +    } else if (DEVICE_GET_CLASS(dev)->bus_type) {
> +        error_setg(errp, "Unexpected bus '%s' for bus-less device '%s'",
> +                   DEVICE_GET_CLASS(dev)->bus_type,
> +                   object_get_typename(OBJECT(dev)));
> +        return false;
>      }
>  
>      object_property_set_bool(OBJECT(dev), true, "realized", &err);
>