Demi Marie Obenour

> Who is actually hotplugging those items outside of the datacentre? Nobody that I am aware of. It’s meant as an example. > If you're changing those you're already doing...

> > CPU model detection applies for live USB/DVD/whatever system too, for example. > > Those mostly load everything into memory. So couldn't one use this theoretical autoconfigure+build system and...

> So in short, I have not really seen any arguments yet that hardware detection at boot time is necessary in general (where boot time = before the kernel give...

> The point is that OS services should not need the FPU This is likely true on some embedded platforms. It is _not_ true in general, as SIMD registers are...

> `kexec` would not make much sense in the model of the verified kernel, but I guess you could add a syscall in an unverified config that is still protected...

> > Could one add “this capability is not used” as a precondition, and prove that kexec() cannot be invoked without authorization? > > That would be possible, but not...

barebones-x509 author here. What was the reason for switching to an alternate library? I wrote barebones-x509 specifically for use in libp2p, and I would like to know what made x509-parser...

> Sure. barebones-x509 was not tested against go implementation (and required several fixes as you remember). There could be other errors so I wrote my own and tested it. If...

@kpp I think one of the remaining tasks would be to get rid of the `UnboundedReceiver`s, for denial-of-service prevention.

> Played around with this pull request once more over the weekend. > > Outcome is basically [mxinden@1230178](https://github.com/mxinden/rust-libp2p/commit/1230178cd660783a57b895d377621abaa66346e7) which removes the unbounded channels. Is it possible to avoid circumventing backpressure?