Re: [PATCH v13 10/13] x86/sgx: Add sgx_einit() for initializing enclaves

From: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
To: "Dr. Greg" <greg@enjellic.com>
Cc: Sean Christopherson <sean.j.christopherson@intel.com>,
	"Huang, Kai" <kai.huang@intel.com>,
	"platform-driver-x86@vger.kernel.org" 
	<platform-driver-x86@vger.kernel.org>,
	"x86@kernel.org" <x86@kernel.org>,
	"nhorman@redhat.com" <nhorman@redhat.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"tglx@linutronix.de" <tglx@linutronix.de>,
	"suresh.b.siddha@intel.com" <suresh.b.siddha@intel.com>,
	"Ayoun, Serge" <serge.ayoun@intel.com>,
	"hpa@zytor.com" <hpa@zytor.com>,
	"npmccallum@redhat.com" <npmccallum@redhat.com>,
	"mingo@redhat.com" <mingo@redhat.com>,
	"linux-sgx@vger.kernel.org" <linux-sgx@vger.kernel.org>,
	"Hansen, Dave" <dave.hansen@intel.com>
Subject: Re: [PATCH v13 10/13] x86/sgx: Add sgx_einit() for initializing enclaves
Date: Mon, 3 Sep 2018 22:27:54 +0300	[thread overview]
Message-ID: <20180903192754.GD13497@linux.intel.com> (raw)
In-Reply-To: <20180831213445.GA4098@wind.enjellic.com>

On Fri, Aug 31, 2018 at 04:34:45PM -0500, Dr. Greg wrote:
> On Fri, Aug 31, 2018 at 10:43:30AM -0700, Sean Christopherson wrote:
> 
> Good afternoon to everyone.
> 
> > > Sorry I missed this one. To be honest I don't know. I checked the
> > > SDM and all I can find is:
> > >
> > > "On reset, the default value is the digest of Intel's signing key."
> 
> > I confirmed the MSRs are reset any time the EPC is lost.  Not sure
> > what happens if the MSRs contained a non-Intel value but feature
> > control is locked with SGX launch control disabled.  I'll post an
> > update when I have an answer.
> 
> It was our interpretation from the SDM that the identity modulus
> signature MSR's are 'trap-door' registers.  If flexible launch control
> (FLC) is enabled the platform has one opportunity to write a new
> signature value, after which the registers are locked from
> modification until the next platform reset.

In the driver we support only MSRs that are left writable by the BIOS
before locking the feature control.

> From a security architecture perspective it seemed that an FLC based
> SGX implementation would use a modified version of TBOOT to securely
> write that register once per platform boot/reset.  The architecture
> that is being discussed where there is a need to continually check
> whether or not the correct root signing key is loaded sounds a bit
> clunky at best.
> 
> At worst it has potential security implications since it is the
> reponsibility of the enclave launch control infrastructure to control
> which enclaves are allowed to have the PROVISION_KEY attribute bit
> set.

Based on the previous feedback supporting read-only MSRs in the driver
is an unwanted feature i.e. the kernel must be able to decide what gets
lauched (i.e. no launch enclave).

> Have a good weekend.
> 
> Dr. Greg
> 
> As always,
> Dr. G.W. Wettstein, Ph.D.   Enjellic Systems Development, LLC.
> 4206 N. 19th Ave.           Specializing in information infra-structure
> Fargo, ND  58102            development.
> PH: 701-281-1686
> FAX: 701-281-3949           EMAIL: greg@enjellic.com
> ------------------------------------------------------------------------------
> "Extensive interviews show that not one alcoholic has ever actually seen
>  a pink elephant."
>                                 -- Yale University
>                                    Center of Alcohol Studies

/Jarkko