freebsd security - Jan 2018 - Intel hardware bug

Hi,

On Wed, 03 Jan 2018 16:29:12 -0800
"Ronald F. Guilmette" <rfg at tristatelogic.com> wrote:
> In message <02563ce4-437c-ab96-54bb-a8b591900ba0 at FreeBSD.org>, 
> Eric van Gyzen <vangyzen at FreeBSD.org> wrote:
> 
> Obviously, the enemy is what it has always been... complexity.  All
I disagree. The problem started when Intel published the handbooks for
the 8086 with many thing 'reserved for future use' but nobody cared.

Intel used segments to separate things everybody hated.

Intel introduced later the rings, everybody ignored.

Instead of keeping the things separated - as suggested by Intel's
design - people used shortcuts whenever possible.

People would have had to think a bit more those days.

But Darwin was not sleeping and people have to think now.
> this stuff is just so hellishly complex nowadays that no single human
> can grasp and hold even a significant fraction of these things in
Because it is all mixed up as a huge mess instead of being separated
like Intel suggested in the Seventies and Eighties.
> P.S.  Right about now, I'd like to have a job working for whichever
> big ad agency has the AMD account.  The ad copy for AMD's next
> marketing campaign practically writes itself... "Performance without
> penality!"
Did you claim copyright?

Erich

Erich Dollansky <freebsd.ed.lists at sumeritec.com>
writes:
> Intel used segments to separate things everybody hated.
Everybody hated segment-level memory protection, but the i386 also
introduced page-level memory protection, which was widely used and has
since been expanded to provide features that were never available at the
segment level.
> Intel introduced later the rings, everybody ignored.
Not at all.  They just don't use all four.  Unless you start looking at
hardware virtualization extensions, which introduce additional
protection levels.
> Instead of keeping the things separated - as suggested by Intel's
> design - people used shortcuts whenever possible.
This is irrelevant.  We are talking about timing-based side-channel
attacks.  The attacker is not able to access protected memory directly,
but is able to deduce its contents by repeatedly performing illegal
memory accesses and then checking how they affect the cache.

DES
-- 
Dag-Erling Sm?rgrav - des at des.no