[Cryptography] Intel SGX: Augean stables piled higher & deeper?

[Max Kington]

On 20 May 2015 14:28, "Henry Baker" <hbaker1 at pipeline.com> wrote:
>
> At 06:02 AM 5/20/2015, Peter Gutmann wrote:
> >Henry Baker <hbaker1 at pipeline.com> writes:
> >
> >>Mathematical proof?
> >
> >Because all of the existing security mechanisms we use also have
rigorous mathematical proofs?
>
> No.  But many/most hardware "improvements" come with serious glitches
that are only appreciated in hindsight.
>
> Since SGX is intended as an "improvement" in security, it deserves a much
higher level of scrutiny than do other types of HW improvements.  In
particular, it deserves some sort of proof that this "improvement" didn't
inadvertently (or not: NSA) introduce additional bugs/insecurities.

I disagree. In the environment where it's all fair game, it shouldn't
matter that this is security directed. The risk in this kind of thinking is
that other HW changes should 'get a pass' because they're not security
related (noting the VMM was done to improve the efficiency of memory layout
and to provide large logic address spaces.
>
> We spend endless hours debating random number generators, but
modifications like this get a pass.  How come?

Nobody is saying it gets a pass but the default stance of it being
"bullshit" in the absence of mathematical proof is incongruous.
>
> _______________________________________________
> The cryptography mailing list
> cryptography at metzdowd.com
> http://www.metzdowd.com/mailman/listinfo/cryptography
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20150520/9c88fc2c/attachment.html>