Re: Another security vulnerability

On Sun, 24 Mar 2024 18:20:06 +0000
mitchalsup@aol.com (MitchAlsup1) wrote:

Stephen Fuld wrote:
 

https://arstechnica.com/security/2024/03/hackers-can-extract-secret-encryption-keys-from-apples-mac-chips/
 

 

So, is there a way to fix this while maintaining the feature's
performance advantage? 

 
They COULD start by not putting prefetched data into the cache
until after the predicting instruction retires. {{I have a note
from about 20 months ago where this feature was publicized and
the note indicates a potential side-channel.}}
 
An alternative is to notice that [*]cryption instructions are
being processed and turn DMP off during those intervals of time.
{Or both}.
>

Their PoC attacks public key crypto algorithms - RSA-2048, DHKE-2048 and
couple of exotic new algorithms that nobody uses.
I think, neither RSA-2048 nor DHKE-2048 use any special crypto
instructions.
On Intel/AMD it's likely that thise crypto routines use MULX
and ADX instruction much for often than non-crypto code, but that's not
guaranteed.
On ARM64 you don't even have that much, because equivalents of MULX and
of ADX are part of base instruction set.

Principle:: an Architecturally visible unit of data can only become
visible after the causing instruction retires. A high precision timer
makes cache line [dis]placement visible; so either take away the HPT
or don't alter cache visible state too early.
 
And we are off to the races, again.....