I've found a 4-year old bug in QEMU, allowing to crash it from the guest system. QEMU security team thinks that it's not a vulnerability, so I've sent the PoC and fixing patch to the public ML: https://lists.nongnu.org/archive/html/qemu-devel/2019-07/msg01651.html …
-
-
On a single user system you may still need multiple accounts eg admin and user
-
as different roles for the same actor, accounts are a fine smoke-and-mirrors shim layer. but putting different actors on the same silicon has not yet worked out well in terms of long term security confidence levels.
- Još 1 odgovor
Novi razgovor -
-
-
It really rather depends on the approach. In my opinion, QEMU/KVM is hopelessly complex to prove its security. On the other hand, we have advances in formal verification, which make it plausible for small separation kernels. Example: http://flint.cs.yale.edu/certikos/
-
The above, of course, addresses only the software side of the story. But, while it's certainly not a practical consideration today, I believe there is no fundamental reason why the same approach cannot be extended to hardware design.
- Još 1 odgovor
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.