WHAT?

EVERY chip ever is inherently vulnerable to Spectre/Meltdown? And there's no way to fix it? >Google’s shocking discovery is that they exploit a foundational flaw in the way information processors work. And because of this, security experts may never be able to protect these devices—even in principle. Article link: https://www.technologyreview.com/s/612989/chips-may-be-inherently-vulnerable-to-spectre-and-meltdown-attacks/ https://pastebin.com/b8busZ5Z Study: https://arxiv.org/abs/1902.05178

not the old hardware.. i think that this implanted vuln appears a little after pentium series for pcs, and for laptop there is a much more large spectre of models. anyway this vuln is not going to be patch. the new hardware is soykaf, and this is known from atleast 10 years ago.

>Inb4 This is another nothing burger and nobody will ever exploit this because it's too difficult to

>>7607 Did you even read the article? > The Google team say the threat affects all chipmakers, including Intel, ARM, AMD, MIPS, IBM, and Oracle. “This class of flaws are deeper and more widely distributed than perhaps any security flaw in history, affecting billions of CPUs in production across all device classes,” say McIlroy and co.

they make grandiose claims that this flaw is foundational to computation but the reality is that it is foundational to the way optimizations are done in practice. There is *absolutely* no reason why spectre should work, it's just a result of lazy microarchitecture design.

>>7609 Did you even read the sentence you're quoting?

>>7610 productivity vs security. Security is a cost not a feature to most people until they get slammed hard.

>>7608 This. Nobody ever got pwned for the "EVERY INTEL PROCESSOR IS RUNNING MINIX!!!" thingie.

>>7614 The idea that you would hear about it is ridicilous.

>>7606 The premise of the article is that if you can read the actual electrical pulses while the computer is running, you can piece together what it's doing This should be obvious, but what should be more obvious is that any program that is able to do that remotely should be detectable, and the only way to do it without a computer being able to detect that it's being done is to have physical access. What does this mean practically? Don't keep valuable information on a laptop, and make sure your desktop or server requires a password before any critical or sensitive information can be loaded.

>>7616 The actual paper is restricted to a formal setting and does not deal with "bit bleeding". It assumes the architecture is idealized and performs common optimizations, and builds attacks such as Spectre based on those optimizations.

Just turn off javascript, keep it off and don't run anything you don't 100% trust.

>>7609 >>7610 >>7612 All I can think of is that Uriel was fuarrrking right.

>>7614 >neurosuggesting you can detect whether you've been compromised