Re: [PATCH 02/25] x86/cpufeatures: Add SGX1 and SGX2 sub-features

[Sean Christopherson]

On Tue, Mar 02, 2021, Kai Huang wrote:
> On Mon, 2021-03-01 at 12:32 +0100, Borislav Petkov wrote:
> > On Tue, Mar 02, 2021 at 12:28:27AM +1300, Kai Huang wrote:
> > > I think some script can utilize /proc/cpuinfo. For instance, admin can have
> > > automation tool/script to deploy enclave (with sgx2) apps, and that script can check
> > > whether platform supports sgx2 or not, before it can deploy those enclave apps. Or
> > > enclave author may just want to check /proc/cpuinfo to know whether the machine can
> > > be used for testing sgx2 enclave or not.
> > 
> > This doesn't sound like a concrete use of this. So you can hide it
> > initially with "" until you guys have a use case. Exposing it later is
> > always easy vs exposing it now and then not being able to change it
> > anymore.
> > 
> 
> Hi Haitao, Jarkko,
> 
> Do you have more concrete use case of needing "sgx2" in /proc/cpuinfo?

The KVM use case is to query /proc/cpuinfo to see if sgx2 can be enabled in a
guest.

The counter-argument to that is we might want sgx2 in /proc/cpuinfo to mean sgx2
is enabled in hardware _and_ supported by the kernel.  Userspace can grep for
sgx in /proc/cpuinfo, and use cpuid to discover sgx2, so it's not a blocker.

That being said, adding some form of capability/versioning to SGX seems
inevitable, not sure it's worth witholding sgx2 from /proc/cpuinfo.