• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD processors from 2011 to 2019 vulnerable to two new attacks

Soldato
Joined
22 Nov 2009
Posts
13,252
Location
Under the hot sun.
Soldato
Joined
17 May 2004
Posts
4,128
Location
Home
AMD defence force.....assemble :p

Your comment would be mildly amusing if it wasn't for the fact that prior to November 2019 I had Intel hardware. A Core i5 3570k which served me well for years. I moved to AMD because there's no denying their superiority right now. I'd have been a fool to buy Intel. Especially with all of their current security flaws.
 
Soldato
Joined
15 Jun 2005
Posts
2,750
Location
Edinburgh
No need. Because the day this ludicrous paper came out the one bellow was published also.... But somehow was "missed" I wonder why :rolleyes:
You mean the one we discussed two days ago?
https://forums.overclockers.co.uk/posts/33399627/

This paper originates from the same University (and some of the same researchers) who discovered the Meltdown & Spectre vulnerabilities. But this AMD focused paper is deemed "ludicrous" and the work of Intel shills. :rolleyes:
 
Soldato
Joined
1 May 2013
Posts
9,692
Location
M28
No need. Because the day this ludicrous paper came out the one bellow was published also.... But somehow was "missed" I wonder why :rolleyes:

Researchers find new Intel security vulnerability that is “impossible to fix”
https://www.kitguru.net/components/...rity-vulnerability-that-is-impossible-to-fix/

You mean the one we discussed two days ago?
https://forums.overclockers.co.uk/posts/33399627/

This paper originates from the same University (and some of the same researchers) who discovered the Meltdown & Spectre vulnerabilities. But this AMD focused paper is deemed "ludicrous" and the work of Intel shills. :rolleyes:

Panos pwned :p

Also, 'no need'. So it does exist ;)
 
Last edited:
Soldato
Joined
22 Nov 2009
Posts
13,252
Location
Under the hot sun.
Panos pwned :p

Also, 'no need'. So it does exist ;)

You mean the one we discussed two days ago?
https://forums.overclockers.co.uk/posts/33399627/

This paper originates from the same University (and some of the same researchers) who discovered the Meltdown & Spectre vulnerabilities. But this AMD focused paper is deemed "ludicrous" and the work of Intel shills. :rolleyes:

Have you both understood what this vulnerability is on the AMD? They isolated the L1 design and simulated an attack. Ignoring completely the rest of the CPU how would react on such attack. That is why is ludicrous.

Thats is far different than exploiting the CPU as a whole as happened with Intel.

In addition point the coincidence the same day post papers were released. One with a critical flaw (Intel) that cannot be fixed, the other that can never happen.

Ofc it would be my bad, spending my time in a technical forum which average member has no clue of what he reads....
 
Permabanned
OP
Joined
27 Sep 2019
Posts
2,570
Humbug take of your rose coloured glasses, the results are from the same source as the Intel flaws, you cannot choose to believe one set of results and not the other because it fits your agender, Panes post means nothing as it does not matter if this will effect many people or not as the Intel one will not effect home users.

So Panos is now an scientist/engineer who understands CPU/GPU architecture?
 
Soldato
Joined
28 May 2007
Posts
18,200
Humbug take of your rose coloured glasses, the results are from the same source as the Intel flaws, you cannot choose to believe one set of results and not the other because it fits your agender, Panes post means nothing as it does not matter if this will effect many people or not as the Intel one will not effect home users.

So Panos is now an scientist/engineer who understands CPU/GPU architecture?

TBF, you can’t seriously compare this to the Intel report.
 
Permabanned
OP
Joined
27 Sep 2019
Posts
2,570
I never said I was, does it matter what is more an issue than the other both are issues and again will not effect the majority of home users.

Like someone above stated, when its Intel bring out the pitch forks and when its AMD its not a big deal...;)
 
Soldato
Joined
3 Oct 2013
Posts
3,602
I never said I was, does it matter what is more an issue than the other both are issues and again will not effect the majority of home users.

Like someone above stated, when its Intel bring out the pitch forks and when its AMD its not a big deal...;)

To be fair they do say that they have reversed engineered the l1 cache hashes which are undocumented. so it's simulated to the extent of what they have managed to find out.
 
Caporegime
Joined
17 Mar 2012
Posts
47,382
Location
ARC-L1, Stanton System
Humbug take of your rose coloured glasses, the results are from the same source as the Intel flaws, you cannot choose to believe one set of results and not the other because it fits your agender, Panes post means nothing as it does not matter if this will effect many people or not as the Intel one will not effect home users.

So Panos is now an scientist/engineer who understands CPU/GPU architecture?

They didn't attack the CPU, they ran a simulation on the L1 Cache, that's like simulating an attack on a vault by simulating a bypass on the inconvenient vault door.

I can simulative starting your car if a ignore the key.
 
Man of Honour
Joined
13 Oct 2006
Posts
90,821
This paper originates from the same University (and some of the same researchers) who discovered the Meltdown & Spectre vulnerabilities. But this AMD focused paper is deemed "ludicrous" and the work of Intel shills.

LOL.

Have you both understood what this vulnerability is on the AMD? They isolated the L1 design and simulated an attack. Ignoring completely the rest of the CPU how would react on such attack. That is why is ludicrous.

Thats is far different than exploiting the CPU as a whole as happened with Intel.

In addition point the coincidence the same day post papers were released. One with a critical flaw (Intel) that cannot be fixed, the other that can never happen.

Ofc it would be my bad, spending my time in a technical forum which average member has no clue of what he reads....

Funny enough when it was pointed out the other way around with other flaws certain people wouldn't have it! infact I took quite a lot of **** posting from certain people who are posting in this thread for pointing out the same kind of thing and purely from an academic perspective on my part (I even pointed out at the time how it could go on to be a bigger issue).

And a reminder that some of these Intel flaws were built upon baby steps consisting of exactly what you are talking about - not to say that applies the same for AMD necessarily but don't forget that factor.
 
Soldato
Joined
9 Nov 2009
Posts
24,771
Location
Planet Earth
If people are petrified over this,they would be even more petrified about the latest Intel flaw,which is not apparently fixable:
https://arstechnica.com/information...psets-have-a-concerning-flaw-thats-unfixable/

5 years of Intel CPUs and chipsets have a concerning flaw that’s unfixable
Converged Security and Management Engine flaw may jeopardize Intel's root of trust.

Also,Intel also had an L1 cache flaw discovered in January called CacheOut:
https://www.theregister.co.uk/2020/01/28/intel_processor_data_leak/
https://www.pcworld.com/article/351...ntel-processors-with-a-fix-arriving-soon.html

AMD is not vulnerable to that L1 cache flaw though. However,Intel was given prior warning though so they could try and attempt some fixes.
 
Last edited:
Associate
Joined
21 Sep 2018
Posts
895
If people are petrified over this,they would be even more petrified about the latest Intel flaw,which is not apparently fixable:
https://arstechnica.com/information...psets-have-a-concerning-flaw-thats-unfixable/



Also,Intel also had an L1 cache flaw discovered in January called CacheOut:
https://www.theregister.co.uk/2020/01/28/intel_processor_data_leak/
https://www.pcworld.com/article/351...ntel-processors-with-a-fix-arriving-soon.html

AMD is not vulnerable to that L1 cache flaw though. However,Intel was given prior warning though so they could try and attempt some fixes.

Cacheout bypasses mitigations for Fallout and Zombieload. No known fixes for it yet.
 
Back
Top Bottom