Skip to main content
Software Engineering

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

2
  • 1
    "In your example, you seem to assume that accessing the same memory location via different registers and different offsets will confuse the CPU. This is not the case." – One could argue that that is more or less what SPECTRE and friends are, although it is a very specific and very subtle kind of "confusion" and not at all like the OP thinks. (I guess technically it is more the CPU designers getting confused.) Commented Sep 9, 2019 at 1:20
  • 1
    @JörgWMittag Hmm, even with Spectre the CPU does properly roll back the results of memory accesses so that programs continue to execute correctly even with wrong speculation. The issue is that internal caches and predictions are not rolled back, and that this is observable through timings. Spectre is not a correctness bug but the result of an unsuitable security model. Commented Sep 9, 2019 at 5:53