New Crosstalk vulnerability can make Intel CPUs leak data from across CPU cores

Clearly we should all buy a new Intel chip every year to get ahead of these vulnerabilities. What other possibility amd what other option could we possibly have?
 
Is it the architectural differences between AMD and Intel that gave Intel such a lead during bulldozer days while sacrificing security or is it that these exploits have been written for the most common CPUs in machine vulnerabilities most likely to return a profit for instigators?
 
Is it the architectural differences between AMD and Intel that gave Intel such a lead during bulldozer days while sacrificing security or is it that these exploits have been written for the most common CPUs in machine vulnerabilities most likely to return a profit for instigators?

In my non expert opinion the core series of cpus has not changed dramatically from it's early days. Sandy bridge came out in 2011 and still shares a lot of commonality with current gen comet lake, and suffers from being immensely popular. As such the idiosyncrasies of the architecture have been explored more because they have remained relevant for longer.

The level of research currently being done on these chips would not be carried out on a pentium 4.

EDIT: Just to expand on this, if the popularity of the zen architecture continues i would not be surprised to see a similar 'phenom'enon in AMD's future lineup of cpus occurring. Intel have traditionally produced very solid products and have more resources to identify potential exploits before release. As such i would wager there is some low hanging fruit of this regard in the zen architecture.
 
Last edited:
In my non expert opinion the core series of cpus has not changed dramatically from it's early days. Sandy bridge came out in 2011 and still shares a lot of commonality with current gen comet lake, and suffers from being immensely popular. As such the idiosyncrasies of the architecture have been explored more because they have remained relevant for longer.

The level of research currently being done on these chips would not be carried out on a pentium 4.

EDIT: Just to expand on this, if the popularity of the zen architecture continues i would not be surprised to see a similar 'phenom'enon in AMD's future lineup of cpus occurring. Intel have traditionally produced very solid products and have more resources to identify potential exploits before release. As such i would wager there is some low hanging fruit of this regard in the zen architecture.

Your post is nearly entirely null of substance. Also, no one researches on a p4 unless looking to collect on property fire insurance claims as most p4 systems active have never had a second thought to TIM degredation over 10 years nor dust cleanup.
 
Your post is nearly entirely null of substance. Also, no one researches on a p4 unless looking to collect on property fire insurance claims as most p4 systems active have never had a second thought to TIM degredation over 10 years nor dust cleanup.

P4 was probably a bad choice of example, but do you care to comment on the rest of the post in regard to its substance?
 
I think Kleptobot is more or less on the money tbf, the likeliness of finding exploits in systems this complex is almost always predominantly down to how much time professional teams are willing to spend looking for them, usually in turn also dictated by the potential value/usefulness of a given exploit. These CPUs often launch with several pages of bugs in the final silicon and it's essentially mathematically impossible for something this complex to be exploit proof, so it's more a matter of "when" than "if" when it comes to popular hardware, and there's no doubt that some recent exploits for Intel CPUs have had much greater value due to the fact they can often impact upto a decades worth of architectures due to the gradual evolution of the core series until Tiger Lake.
 
Last edited:
Back
Top