Re: [RFC 0/3] x86: fix cpu hotplug with secure boot

[Laszlo Ersek]

On 07/10/20 18:17, Igor Mammedov wrote:
> CPU hotplug with Secure Boot was not really supported and firmware wasn't 
> aware
> of hotplugged CPUs (which might lead to guest crashes). During 4.2 we 
> introduced
> locked SMI handler RAM arrea to make sure that guest OS wasn't able to inject
> its own SMI handler and OVMF added initial CPU hotplug support.
> 
> This series is QEMU part of that support [1] which lets QMVF tell QEMU that
> CPU hotplug with SMI broadcast enabled is supported so that QEMU would be able
> to prevent hotplug in case it's not supported and trigger SMI on hotplug when
> it's necessary. 
> 
> 1) CPU hotplug negotiation part was introduced later so it might not be
> in upstream OVMF yet or I might have missed the patch on edk2-devel
> (Laszlo will point out to it/post formal patch)

I'll post it later, after testing it with this patch series.

Thanks!
Laszlo

> 
> Igor Mammedov (3):
>   x86: lpc9: let firmware negotiate CPU hotplug SMI feature
>   x86: cphp: prevent guest crash on CPU hotplug when broadcast SMI is in
>     use
>   x68: acpi: trigger SMI before scanning for hotplugged CPUs
> 
>  include/hw/acpi/cpu.h  |  1 +
>  include/hw/i386/ich9.h |  1 +
>  hw/acpi/cpu.c          |  6 ++++++
>  hw/acpi/ich9.c         | 12 +++++++++++-
>  hw/i386/acpi-build.c   | 33 ++++++++++++++++++++++++++++++++-
>  hw/i386/pc.c           | 15 ++++++++++++++-
>  hw/isa/lpc_ich9.c      | 10 ++++++++++
>  7 files changed, 75 insertions(+), 3 deletions(-)
>