David Roundy
David Roundy
Okay I'm now thinking that this implementation could open users of internment up to hash collision attacks, since it uses a hash with DefaultHasher rather than the hasher for the...
I've confirmed from the docs that DefaultHasher::new() generates a SipHasher with zero for the keys, which means we would be vulnerable to DOS attacks. Probably your shouldn't let potential attackers...
Probably, and that's probably how I'd do this. If you want to give it a shot it might happen sooner. David Roundy On Wed, Nov 24, 2021, 10:20 PM Stepan...
BTW, the reason I'm picky about optimization changes is that I've been bitten in the past by "optimizations" which ended up introducing serious bugs (only discovered months later after they...
I've also been thinking about designing such a filesystem. I've been thinking I'd like a Raft-based set of servers. On Sun, Feb 18, 2018, 1:49 PM Pedro Côrte-Real wrote: >...
Any chance of getting this done?
Well, identifying those crates by name would be a start. I presume there are docs on docs.rs that could also be linked to.
I'd say worst case you have a major revision bump.
What about decreasing the minimum amount of time and increasing the maximum based on R2? That way it would usually run tests even more quickly, but could take longer on...
I was imagining something about that simple, but using the same framework as easybench. But now I'm wondering whether it would be better to make easybench automatically work for slow...