
Phew! That was a close call, but I'm pretty sure my 6502 is not vulnerable.
Googlers have lately found not one but two more security vulnerabilities in Intel and AMD processors that can be exploited to steal sensitive data from a vulnerable computer's memory. Specifically, there's one flaw in Intel components, and one in AMD. Both can be abused by malware running on a system, or a rogue logged-in user …
It's definitely just a happy little accident.
How do I know? These vulnerabilities are the result of a standard technique to improve computer performance that has been around for ages, out-of-order executiion. This was introduced with the IBM System/360 Model 91, which only handled floating-point instructions with OoO. That computer was succeeded with the Model 195, which added cache memory, another very successful performance-increasing innovation from the System/360 Model 85.
So fast-forward decades later, after Moore's Law finally let that level of complexity be put on a microchip - and out came the Pentium Pro and the Pentium II, which, just like the 360/195, had cache on the chip, and OoO execution for the floating-point section. (And an advanced division algorithm, although not quite as fast as the one the 360/195 used.)
So they were trying to make the best chip they could, using a proven way to increase performance... but which, sadly, had a weakness that could finally be uncovered in today's more hostile computing environment.
The problem is not out of order execution at all. The problem is branch prediction and speculative execution. During speculative execution the old contents of registers that get modified must be stored _and protected against overwriting_ so that speculative execution can be undone. And that’s what both AMD with Zenbleed and now zingelt got wrong.
"They allow an attacker to violate the software-hardware boundary established in modern processors."
503 Service Unavailable: No server is available to handle this request.