Post a Comment
Tweek, you're responding to the headline, which does not reflect the article content: while she does say that anti-virus software is ineffective, it mainly discusses attacks based on moving the running OS into a virtual machine, and installing the malware as the hypervisor, which is a really neat trick.
But practically impossible and definitely always discoverable according to the Xen hacker. Seems that AMD has designed its virtualisation to make discovery of this sort of thing easy.
He states, "I wouldn't lose a bit of sleep over this particular threat. I don't feel there is any new risk here at all."
Also it is interesting that the technology built into AMD virtualisation technology allows for "attestation" which could make anti-virus software a thing of the past. The idea is that you can detect at any point in time whether software is running that should not be running.
He states, "Currently, anti-malware software has to look specifically for known threats. Attestation lets you do something much stronger. Attestation allows you to validate that there is no unknown threats.
Imagine anti-virus software that doesn't need to be updated--ever. With attestation, there is no such thing as zero-day threats."
Seems to me that the new technology AMD is bundling is making secure systems proveable rather than allowing things to run hidden without permission.
This is what I'm talking about.
This type of technique is in the wild, I've run in to it twice now ... on Macintoshes.
Yes, you read right.
In '97 on a 68k (Quadra), using "inits", and last year, '05 on Xserves and G4s, G5s, using raw disk access and OpenFirmware, and "inits" from Classic.
It gives "owned" an all together new meaning, it fights back. I've, at times resorted to screenshots with an external camera.
(Nothing like a _KillPicture in your picts to bounce you in to MacsBug)
It is all about (hardware) disks, disk drivers and ports, but not the way you think of them.
For the last 12 months now people have told me "that's impossible", and I show them the print outs, the evidence, the theory/facts (that Ms. Rutkowska just illustrated, rather expertly). The archived disks I have (SCSI, IDE etc.) have, buried in "bad boot blocks" microcode, networking commands and the such.
I've been through the wringer with ... (name) experts, telling me how "absolutely impossible" this is to implement. Covering their asses just long enough to say "if ... please call me", "well, then you'll be vindicated".
I'm not a hardware programmer, but I'm in over my head and know it, but not so much as to see what is obvious.
Why is this so far off the radar?
Denial.
I won't get in to a pissing match here.
(clueless users and their cheap shots usually follow).
I mean, really, last year if some people read this interview (with a GIRL, no less) the name calling would be unbelievable.
See:
An Open Challenge to David Maynor and Jon Ellch
http://daringfireball.net/2006/09/open_challenge
Yeah, they're faking this, they'll pull the wool over everyones eyes and they use 1337 cracker names, so they can prof ... oh, wait.
That's their real names?
H D Moore
http://kernelfun.blogspot.com/2006/11/mokb-starts-mokb-01-11-2006-a...
Ain't hardware great?
I'm just putting this out so folks will look a bit closer and I might get some help, maybe flush out someone else.
Yeah, I NEED this kind of attention.
I got nothing to sell.
I want an expert discussion (not here), with a real expert.
Neat trick?
O-M-G
Ask yourself what the ramifications are (now).
ring0 / EFI
Further:
http://developers.slashdot.org/comments.pl?sid=204213&cid=16690789
_GetClue
hylas



