Skip to contentSkip to conversation

Intel SGX Driver

PATCH Dr. Greg Wettstein 2 participants 2 messages Jan 4, 2018

lore
Share & tools
lore
#0 lore raw

On Jan 3, 3:00am, Christian Stroetmann wrote:
} Subject: Re: [PATCH v6 00/11] Intel SGX Driver

Good morning, I hope the day is going well for you.

Quoted 8 lines
Quoted 4 lines

Philosophically we take the approach that for high security assurance environments that it is virtually impossible to allow any untrusted code to run on a platform. Which is why we focus on autonomous introspection for these environments.

Interesting. I would like to hear more about this autonomous introspection concept, specifically what it monitors and how it reacts on an issue.

AI as we call it.. :-), at a high level, is the notion that an operating system platform can be modeled for its intended behavior, just as is done for buildings, bridges and airplanes. The model is based on a precise mathematical description of the set of desired information exchange events that the operating system will be allowed to mediate.

Autonomous introspection is the notion of the platform using this model to monitor and self-regulate its behavior.

Quoted 2 lines

Maybe you have a summary or short introduction of the concept or/and a link to a document.

The initial concept and implementation were described by the following Linux Security Summit paper:

http://kernsec.org/files/lss2015/idfusion-iso-identity-paper.pdf

The work has matured a great deal since then.

Most notably the modeling and control engine have been lifted out of the kernel into an SGX enclave. In this architecture the kernel cannot act without permission from the enclave based engine, which requires the integrity of the enclave to be violated in order to compromise the kernel.

A virtuous circle.

Quoted 2 lines

Regards
Christian Stroetmann

We can engage further in private e-mail as this may or may not be of general interest.

Thanks for the interest.

Have a good day.

Greg

}-- End of excerpt from Christian Stroetmann

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: [email protected]
------------------------------------------------------------------------------
"Four year projections, are you kidding me. We don't know what we are
supposed to be doing at the 4 o'clock meeting this afternoon."
-- Terry Wieland
Resurrection

#1 lore raw

On the 04.Jan.2018 09:02, Dr. Greg Wettstein wrote:
Aloha

First of all, I would like to appologize that there was an unusual issue with my email client, so that my email did not reach the mailing list linux-fsdevel. Hopefully, this does not happen once again with this email.

Thank you for answering but let me ask some more (simple) questions below.

Quoted 12 lines · Christian Stroetmann

} Subject: Re: [PATCH v6 00/11] Intel SGX Driver

Good morning, I hope the day is going well for you.

Quoted 7 lines
Quoted 4 lines

Philosophically we take the approach that for high security assurance environments that it is virtually impossible to allow any untrusted code to run on a platform. Which is why we focus on autonomous introspection for these environments.

Interesting. I would like to hear more about this autonomous introspection concept, specifically what it monitors and how it reacts on an issue.

Quoted 6 lines

AI as we call it.. :-), at a high level, is the notion that an operating system platform can be modeled for its intended behavior, just as is done for buildings, bridges and airplanes. The model is based on a precise mathematical description of the set of desired information exchange events that the operating system will be allowed to mediate.

So you have artificial intelligence in the operating system respectively kernel?

What is with machine learning?

Also, you need to ensure that your precise mathematical model is correct and that your code implements the model in a correct way.

Quoted 3 lines

Autonomous introspection is the notion of the platform using this model to monitor and self-regulate its behavior.

This sounds like somekind of a reflective functionality.

Quoted 8 lines
Quoted 3 lines

Maybe you have a summary or short introduction of the concept or/and a link to a document.

The initial concept and implementation were described by the following Linux Security Summit paper:

http://kernsec.org/files/lss2015/idfusion-iso-identity-paper.pdf

Quoted 7 lines

The work has matured a great deal since then.

Most notably the modeling and control engine have been lifted out of the kernel into an SGX enclave. In this architecture the kernel cannot act without permission from the enclave based engine, which requires the integrity of the enclave to be violated in order to compromise the kernel.

This would be ideal with a microkernel based operating system.

Quoted 9 lines

A virtuous circle.

Quoted 2 lines

Regards
Christian Stroetmann

We can engage further in private e-mail as this may or may not be of general interest.

Thanks for the interest.

I think that the others are also interested in this subject matter.

Regards
Christian Stroetmann

Quoted 6 lines

Have a good day.

Greg

}-- End of excerpt from Christian Stroetmann

[snip]