Re:Tor in a safer language: Network team update from Amsterdam
by wyager on Saturday April 01, 2017 @03:22PM
(#14013669)
It's unlikely that the cost of whole-program side channel resistance is worth it for an application like Tor. Side-channel-resistant cryptography is almost certainly sufficient for most reasonable threat/cost models.
[ Parent
| on HN ]
Re:Tor in a safer language: Network team update from Amsterdam
by nickpsecurity on Saturday April 01, 2017 @03:42PM
(#14013768)
You can limit it to the protocol layer. High-assurance security in military systems just used fixed-speed, fixed-timing transmission with covert channel analysis of setup, crypto operations, and error handling. Any problems at all violating key invariants cause fail-safe. Tor could do something similar.
Still need a language that allows such analysis at least for those components being analyzed.