On Thu, May 27, 2021 at 01:36:37PM +0200, Claudio Fontana wrote:
> Just to check whether this is actually the issue we are talking about,
> Sid et al, could you try this?
>
>
> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
> index c496bfa1c2..810c46281b 100644
> --- a/target/i386/cpu.c
> +++ b/target/i386/cpu.c
> @@ -4252,6 +4252,7 @@ static void max_x86_cpu_initfn(Object *obj)
> object_property_set_str(OBJECT(cpu), "model-id",
> "QEMU TCG CPU version " QEMU_HW_VERSION,
> &error_abort);
> + accel_cpu_instance_init(CPU(cpu));
> }
>
> static const TypeInfo max_x86_cpu_type_info = {
> ------------------------------------------------------------------------------------------
>
> Does this band-aid happen to cover-up the issue?
Yes it does fix the issue for me. Thanks.
~ Sid.
> I need to think about the proper fix though, any suggestions Paolo,
> Eduardo?
>
> The pickle here is that we'd need to call the accelerator specific
> initialization of the x86 accel-cpu only after the x86 cpu subclass
> initfn, otherwise the accel-specific cpu initialization code has no
> chance to see the subclass (max) trying to set ->max_features.
>
> C.
Amazon Development Center Germany GmbH
Krausenstr. 38
10117 Berlin
Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss
Eingetragen am Amtsgericht Charlottenburg unter HRB 149173 B
Sitz: Berlin
Ust-ID: DE 289 237 879
