On Thu, 13 Feb 2020 12:17:32 -0600
Babu Moger <babu.mo...@amd.com> wrote:
> Introduce model specific apicid functions inside X86MachineState.
> These functions will be loaded from X86CPUDefinition.
>
> Signed-off-by: Babu Moger <babu.mo...@amd.com>
> ---
> include/hw/i386/x86.h | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/include/hw/i386/x86.h b/include/hw/i386/x86.h
> index 38c2d27910..75c2462770 100644
> --- a/include/hw/i386/x86.h
> +++ b/include/hw/i386/x86.h
> @@ -79,6 +79,15 @@ typedef struct {
> uint16_t boot_cpus;
> unsigned smp_dies;
>
> + /* Apic id specific handlers */
> + uint32_t (*apicid_from_cpu_idx)(X86CPUTopoInfo *topo_info,
> + unsigned cpu_index);
> + void (*topo_ids_from_apicid)(apic_id_t apicid, X86CPUTopoInfo *topo_info,
> + X86CPUTopoIDs *topo_ids);
> + apic_id_t (*apicid_from_topo_ids)(X86CPUTopoInfo *topo_info,
> + const X86CPUTopoIDs *topo_ids);
> + uint32_t (*apicid_pkg_offset)(X86CPUTopoInfo *topo_info);
> +
> /*
> * Address space used by IOAPIC device. All IOAPIC interrupts
> * will be translated to MSI messages in the address space.
>
>
Suggest to move defaults into this patch as well. i.e.
move parts that replace direct calls with hooks from 12/16
and set default hooks (i.e. non EPYC ones) in x86's machine_class_init().
This way It's immediately obvious what you are changing and then
followup EPYC related patch won't be cluttered by non related defaults
conversion.
