Re: [PATCH] x86/hvm: add HVM-specific Kconfig

[Jan Beulich <jbeulich@xxxxxxxx>]

On 07.02.2025 23:03, dmkhn@xxxxxxxxx wrote:
> Add separate menu for configuring HVM build-time settings to help organizing
> HVM-specific options.
> 
> Signed-off-by: Denis Mukhin <dmukhin@xxxxxxxx>

I think it would be nice if "organizing" was further qualified, to really see
_why_ the change is being made. A particular benefit that Roger points out is
that this way all HVM-dependent options moved here will then appear in a sub-
menu like fashion in the tool.

> --- a/xen/arch/x86/Kconfig
> +++ b/xen/arch/x86/Kconfig
> @@ -30,7 +30,6 @@ config X86
>       select HAS_SCHED_GRANULARITY
>       select HAS_UBSAN
>       select HAS_VMAP
> -     select HAS_VPCI if HVM
>       select NEEDS_LIBELF

As said before, personally I'd prefer if this didn't move.

> --- /dev/null
> +++ b/xen/arch/x86/hvm/Kconfig
> @@ -0,0 +1,73 @@
> +menuconfig HVM
> +     bool "HVM support"
> +     depends on !PV_SHIM_EXCLUSIVE
> +     default !PV_SHIM
> +     select COMPAT
> +     select HAS_VPCI
> +     select IOREQ_SERVER
> +     select MEM_ACCESS_ALWAYS_ON
> +     help
> +       Interfaces to support HVM domains.  HVM domains require hardware
> +       virtualisation extensions (e.g. Intel VT-x, AMD SVM), but can boot
> +       guests which have no specific Xen knowledge.
> +
> +       This option is needed if you want to run HVM or PVH domains.
> +
> +       If unsure, say Y.
> +
> +if HVM
> +
> +config AMD_SVM
> +     bool "AMD-V" if EXPERT
> +     default y
> +     help
> +       Enables virtual machine extensions on platforms that implement the
> +       AMD Virtualization Technology (AMD-V).
> +       If your system includes a processor with AMD-V support, say Y.
> +       If in doubt, say Y.
> +
> +config INTEL_VMX
> +     bool "Intel VT-x" if EXPERT
> +     default y
> +     select ARCH_VCPU_IOREQ_COMPLETION
> +     help
> +       Enables virtual machine extensions on platforms that implement the
> +       Intel Virtualization Technology (Intel VT-x).
> +       If your system includes a processor with Intel VT-x support, say Y.
> +       If in doubt, say Y.

Perhaps not to be done right here, but still: I guess the "default" of these
want to change to use AMD / INTEL respectively. While we permit enabling the
virt extension support separately, generally there's little point doing so
by default when the corresponding CPU support code was disabled.

> +config ALTP2M
> +     bool "Alternate P2M support" if EXPERT
> +     depends on INTEL_VMX
> +     default y
> +     help
> +       Alternate-p2m allows a guest to manage multiple p2m guest physical
> +       "memory views" (as opposed to a single p2m).
> +       Useful for memory introspection.
> +
> +       If unsure, stay with defaults.
> +
> +config MEM_PAGING
> +     bool "Xen memory paging support (UNSUPPORTED)" if UNSUPPORTED
> +
> +config MEM_SHARING
> +     bool "Xen memory sharing support (UNSUPPORTED)" if UNSUPPORTED
> +
> +config HVM_FEP
> +     bool "HVM Forced Emulation Prefix support (UNSUPPORTED)" if UNSUPPORTED
> +     default DEBUG
> +     help
> +       Compiles in a feature that allows HVM guest to arbitrarily
> +       exercise the instruction emulator.
> +
> +       This feature can only be enabled during boot time with
> +       appropriate hypervisor command line option. Please read
> +       hypervisor command line documentation before trying to use
> +       this feature.
> +
> +       This is strictly for testing purposes, and not appropriate
> +       for use in production.
> +
> +       If unsure, say N.

Please can this move ahead of at least the two MEM_* ones, maybe even ahead
of ALTP2M?

Jan