Okay, I should've shouted this earlier. THE AMD VULNERABILITIES THAT JUST CAME OUT HAVE NOTHING WHATSO-FUCKING-EVER TO DO WITH MELTDOWN/SPECTRE IN SCOPE, DANGER, OR APPLICABILITY. ANY COMPARISONS DEMONSTRATE A COMPLETE LACK OF UNDERSTANDING AND A FAILURE AT TECH JOURNALISM.
Same as Spectre. You wrap it in a mispredicted branch (out of bounds access). I'm not aware of any public PoCs but I believe private ones exist.
-
-
In Spectre there's no need for exceptions, you're directly (same process) or indirectly (syscall and HT siblings) mistraining the indirect branch predictor. Meltdown seems quite hard (impossible?) to exploit for JS, but I maybe wrong.
-
Spectre is two flaws, because the morons who named it mixed together two unrelated vulns. I'm talking about GPZ Variant 1, which is about direct branch misprediction. You can use the same technique to speculate through and recover from the exception.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.