Re: [RFC 0/2] GET_EMULATED_CPUID support with "allow-emulation" option

[gleb@kernel.org]

On Thu, Jun 05, 2014 at 06:26:41PM +0200, Paolo Bonzini wrote:
> Il 05/06/2014 18:24, Alexander Graf ha scritto:
> >
> >On 05.06.14 18:12, Eduardo Habkost wrote:
> >>This implements GET_SUPPORTED_CPUID support using an explicit option
> >>for it:
> >>"allow-emulation". We don't want any emulated feature to be enabled by
> >>accident,
> >>so they will be enabled only if the user explicitly wants to allow them.
> >
> >So is this an all-or-nothing approach? I would really prefer to override
> >individual bits.
> 
> You can still disable them with "cpu foo,-movbe,allow-emulation".
> 
> >Also, I don't think the line "emulated" is the right one to draw. We
> >"emulate" SVM or VMX too, but still enable them by default as soon as we
> >think they're ready enough.
> 
> Well, I disagreed with the whole KVM_GET_EMULATED_CPUID concept for MOVBE
> too for example.  It seemed overengineered to me, sooner or later we might
> graduate MOVBE out of KVM_GET_EMULATED_CPUID as well.
Can you remind me what was your argument against
KVM_GET_EMULATED_CPUID? Where do you want to move MOVBE to? Supported
cpuid? That will make some CPU models unreasonably slow on hosts
that do not support MOVBE natively. KVM is not in a busyness of
instruction emulation, yes sometimes there is no choice (IO/real mode),
sometimes emulating a feature is beneficial (x2apic) and sometimes
guest cannot workaround missing feature, so by emulating it you allow
guest functionality that is impossible otherwise (SVM/VMX). MOVBE (and
MONITOR/MWAIT) is not in any of those categories. By forcing emulation
upon unsuspecting guest you force it to use much slower alternative for
what it can do by other means.

> 
> However, for MONITOR/MWAIT it makes some sense.
> 
> Paolo
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
			Gleb.