On 21/08/2015 15:54, Eduardo Habkost wrote:
> > + if (tcg_enabled()) {
> > + memory_region_add_subregion_overlap(cpu->cpu_as_root,
> > + apic->apicbase &
> > + MSR_IA32_APICBASE_BASE,
> > + &apic->io_memory,
> > + 0x1000);
>
> Why exactly is this necessary? If this is necessary, why don't we need
> to do this for non-TCG accelerators?At least KVM and qtest do not support per-CPU address spaces. I'm not sure about Xen, it looks like it could but it would be slower. Paolo
