Researcher To Demonstrate Attack Code for Intel Chips

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Security researcher Kris Kaspersky says he has plans to demonstrate how an attacker can exploit a flaw in Intel processors to remotely attack a computer using JavaScript or TCP/IP packets, regardless of what operating system the computer is running.

Kaspersky will demonstrate how such an attack can be made in a presentation at the upcoming Hack In The Box (HITB) Security Conference in Kuala Lumpur, Malaysia, during October. The proof-of-concept attacks will show how processor bugs, called errata, can be exploited using certain instruction sequences and a knowledge of how Java compilers work, allowing an attacker to take control of the compiler.
 
Let's hope it can be patched up through an OS and/or BIOS update. Sooner rather than later.

I already run with java script disabled most of the time, TCP/IP on the other hand, is something I can't really do without.
 
A BIOS update to protect from processor flaws. That is going to be hell for corporations to patch. This may be the business driver that causes Windows based BIOS updates to become available.
 
A BIOS update to protect from processor flaws. That is going to be hell for corporations to patch. This may be the business driver that causes Windows based BIOS updates to become available.

Yes, I can't imagine having to go around to all 3000 machines and patch each one. The good news is that this will not affect all processor types, as Netburst processors may not be affected. In that case, my company may have dodged a bullet. We have less than 500 Core architecture machines, not including servers.
 
Yes, I can't imagine having to go around to all 3000 machines and patch each one. The good news is that this will not affect all processor types, as Netburst processors may not be affected. In that case, my company may have dodged a bullet. We have less than 500 Core architecture machines, not including servers.

still 500 you have to do..

what i hate about these announcements is that there is 3 months until he will show people..

people, after hearing about this, are probably trying to develop how to do this now, and perhaps an actual attack via this exploit comes before its even been demonstrated..

however unlikely it may be...
 
They can just wait because he is going tos how them how. Then when something cant be done about it in time, the whole world is fucked in the name of making sure Intel/MS knows that this can be done and making them work faster to fix it. This guy is doing nothing more than making it even worse. He is a tool to think that any computer can be 100% secure.
 
A BIOS update to protect from processor flaws. That is going to be hell for corporations to patch. This may be the business driver that causes Windows based BIOS updates to become available.

A reliable and consistent API to update BIOS from the OS? We're talking about PC hardware, not MAC ;)
 
This opens a whole new world to attacking a computer.

What can the end user do if there is no antivirus in the world that can stop the attack?

All you have to do is log onto the wrong website, read the wrong packet, boom, everything abotu you is stolen or destroyed.


I think AMD's stock may be going up here soon :D
 
This opens a whole new world to attacking a computer.

What can the end user do if there is no antivirus in the world that can stop the attack?

All you have to do is log onto the wrong website, read the wrong packet, boom, everything abotu you is stolen or destroyed.

i think this is a little overdramatic. most likely it will require a bios update to repair it...which is going to send industry into a tizzy for a while...and make life suck really bad for those of us who are in IT for a while... i just did bios upgrades on the 120 or so systems we have here...and it's a pain.
 
They can just wait because he is going tos how them how. Then when something cant be done about it in time, the whole world is fucked in the name of making sure Intel/MS knows that this can be done and making them work faster to fix it. This guy is doing nothing more than making it even worse. He is a tool to think that any computer can be 100% secure.


No computer or network is 100% secure. However, the key to making a system as secure as possible is to remain vigilant; attacks and security flaws will always happen but it'll be fairly secure if they're dealt with quickly to minimize risk.

The guy probably thinks that if it's not publicized then the flaw will never be fixed - even if Intel already knows about it. Someone's going to discover it eventually; better for one of the 'good' guys to discover it first.
 
see, this is why I like Intel, they are linux-friendly

They release ucode for uploading in userland THUS you don't have to wait for yr mobo maker to update your BIOS :D
 
Heh, the last Intel I had was a P3 400. I was just about to buy another. I guess it's wait time.
 
No computer or network is 100% secure. However, the key to making a system as secure as possible is to remain vigilant; attacks and security flaws will always happen but it'll be fairly secure if they're dealt with quickly to minimize risk.

The guy probably thinks that if it's not publicized then the flaw will never be fixed - even if Intel already knows about it. Someone's going to discover it eventually; better for one of the 'good' guys to discover it first.

QFT

Also to ensure layers of security. HTTP and email security services (ie proxies) could still drop potentially hostile connections so long as they are not vulnerable themselves. The real issue is when high priority Internet facing systems become vulnerable.
 
He's not demonstrating it until October. That's one way to light a fire under Intel. :p

IMO it's probably overblown anyways. You'll probably need to do dumb things x, y and z in order for the exploit(s) to work.
 
its a long way from a damn jvm to the actual hardware. a LONG way.

it's java OR TCP packets....

That said, I'd imagine this could be worked around within the various IP stacks that are out there.
 
its a long way from a damn jvm to the actual hardware. a LONG way.

QFT. I'm having a hard time believing that some TCP/IP packets can exploit a flaw in Intel without anything being able to stop them along the way.
 
TCP/IP packets are easily stopped with good external firewall policies if your network team actually knows whats going on.
 
TCP/IP attack regardless of OS tells me this is bull sh*t

If this is the case then the nic driver or OS update for the stack could be modified to stop it, or a firewall/IDS could detect and block the attack

Java is sitting up at the application layer, TCP is much lower in the network stack, while java can communicate and pass data through the stack, it's not directly talking to the hardware, drivers and the OS handle that communication.
 
Back
Top