Re: [PATCH 06/16] x86/tdx: Override PV calibration routines with CPUID-based calibration

[Nikunj A Dadhania <nikunj@amd.com>]

Sean Christopherson <seanjc@google.com> writes:

> On Tue, Feb 04, 2025, Nikunj A Dadhania wrote:
>> Sean Christopherson <seanjc@google.com> writes:
>> 
>> > When running as a TDX guest, explicitly override the TSC frequency
>> > calibration routine with CPUID-based calibration instead of potentially
>> > relying on a hypervisor-controlled PV routine.  For TDX guests, CPUID.0x15
>> > is always emulated by the TDX-Module, i.e. the information from CPUID is
>> > more trustworthy than the information provided by the hypervisor.
>> >
>> > To maintain backwards compatibility with TDX guest kernels that use native
>> > calibration, and because it's the least awful option, retain
>> > native_calibrate_tsc()'s stuffing of the local APIC bus period using the
>> > core crystal frequency.  While it's entirely possible for the hypervisor
>> > to emulate the APIC timer at a different frequency than the core crystal
>> > frequency, the commonly accepted interpretation of Intel's SDM is that APIC
>> > timer runs at the core crystal frequency when that latter is enumerated via
>> > CPUID:
>> >
>> >   The APIC timer frequency will be the processor’s bus clock or core
>> >   crystal clock frequency (when TSC/core crystal clock ratio is enumerated
>> >   in CPUID leaf 0x15).
>> >
>> > If the hypervisor is malicious and deliberately runs the APIC timer at the
>> > wrong frequency, nothing would stop the hypervisor from modifying the
>> > frequency at any time, i.e. attempting to manually calibrate the frequency
>> > out of paranoia would be futile.
>> >
>> > Deliberately leave the CPU frequency calibration routine as is, since the
>> > TDX-Module doesn't provide any guarantees with respect to CPUID.0x16.
>> 
>> Does TDX use kvmclock?
>
> A TDX guest can.  That's up to the host (expose kvmclock) and the guest (enable
> kvmclock).
>
>> If yes, kvmclock would have registered the CPU frequency calibration routine:
>> 
>> 	tsc_register_calibration_routines(kvm_get_tsc_khz, kvm_get_cpu_khz,
>>  					  tsc_properties);
>> 
>> so TDX will use kvm_get_cpu_khz(), which will either use CPUID.0x16 or
>> PV clock, is this on the expected line ?
>
> What do you mean by "is this on the expected line"?  If you are asking "is this
> intended",

Yes, that is what I meant.

> then the answer is "yes, working as intended".  As above, the TDX-Module
> doesn't emulate CPUID.0x16, so no matter what, the guest is relying on the untrusted
> hypervisor to get the CPU frequency.  If someone thinks that TDX guests should
> assume the CPU runs as the same frequency as the TSC, a la SNP's Secure TSC, then
> they are welcome to propose such a change.

Ok, that makes sense.

Regards
Nikunj