On Wed, Sep 08, 2021 at 12:03:56PM +0200, Paolo Bonzini wrote:
> From: Sean Christopherson <sean.j.christopher...@intel.com>
>
> Because SGX EPC is enumerated through CPUID, EPC "devices" need to be
> realized prior to realizing the vCPUs themselves, i.e. long before
> generic devices are parsed and realized. From a virtualization
> perspective, the CPUID aspect also means that EPC sections cannot be
> hotplugged without paravirtualizing the guest kernel (hardware does
> not support hotplugging as EPC sections must be locked down during
> pre-boot to provide EPC's security properties).
>
> qapi/machine.json | 26 +++++++++++++++
> qemu-options.hx | 10 ++++--
> 9 files changed, 166 insertions(+), 8 deletions(-)
> create mode 100644 hw/i386/sgx.c
...
> +++ b/qapi/machine.json
> @@ -1194,6 +1194,32 @@
> }
> }
>
> +##
> +# @SgxEPC:
> +#
> +# Sgx EPC cmdline information
> +#
> +# @memdev: memory backend linked with device
> +#
> +# Since: 6.1
Another instance where we'll want the followup patch to correct things
to 6.2.
> +##
> +{ 'struct': 'SgxEPC',
> + 'data': { 'memdev': 'str' }
> +}
> +
> +##
> +# @SgxEPCProperties:
> +#
> +# Properties for SgxEPC objects.
> +#
> +# @sgx-epc: sgx epc section properties.
> +#
> +# Since: 6.1
> +##
> +{ 'struct': 'SgxEPCProperties',
> + 'data': { 'sgx-epc': ['SgxEPC'] }
> +}
> +
> ##
> # @MemoryDeviceInfo:
> #
> diff --git a/qemu-options.hx b/qemu-options.hx
> index 8f603cc7e6..ceca52818a 100644
> --- a/qemu-options.hx
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3266
Virtualization: qemu.org | libvirt.org
