Re: [PATCH v1 03/11] KVM: x86: dynamic kvm_apic_map

[Paolo Bonzini <pbonzini@redhat.com>]

On 01/07/2016 16:38, Radim Krčmář wrote:
>> > Should it?
> Yes, x2APIC ID cannot be changed in hardware and is initialized to the
> intitial APIC ID.
> Letting LAPIC_SET change x2APIC ID would allow scenarios where userspace
> reuses old VMs instead of building new ones after reconfiguration.
> I don't think it's a sensible use case and it it is currently broken,
> because we don't exit to userspace when changing APIC mode, so KVM would
> just set APIC ID to VCPU ID on any transition and userspace couldn't
> amend it.
> 
>> >             According to QEMU if you have e.g. 3 cores per socket one
>> > socket take 4 APIC IDs.  For Knights Landing the "worst" prime factor in
>> > 288 is 3^2 so you need APIC IDs up to 288 * (4/3)^2 = 512.
> The topology can result in sparse APIC ID and APIC ID is initialized
> from VCPU ID, so userspace has to pick VCPU ID accordingly.

Right, I was confusing KVM_MAX_VCPUS and KVM_MAX_VCPU_ID.  So the
overflow case cannot happen and neither can the 32GB allocation.  On the
other hand, I suspect you need to bump KVM_MAX_VCPU_ID beyond its
current default setting (which is equal to KVM_MAX_VCPUS), up to 511 or
1023.

Thanks,

Paolo