constantine VS constant-time

Percival cache attacks on Hyperthreading.

I go over some examples here: https://github.com/mratsim/constantine/issues/358#issuecomme...

`when myCondition():` instead of `if myCondition:` is done at compile-time.

Alternatively you can use a `static:` code block to force compile time evaluation. Or tag a function {.compileTime.} or tag function inputs with `static` modifier.

It is possible to create a compiler or an assembler running fully in Nim macros as well:

- https://github.com/mratsim/constantine/blob/master/constanti... (all that file runs at compile-time)

You can also implement Continuation-Passing-Style transformation at compile-time:

At a glance this sounds like a re-discovery of addition chains and using them to construct Pippenger algorithm. But applied to matrices instead of group elements.

See: https://github.com/mratsim/constantine/issues/37

I usually explain extension fields as similar to complex numbers with regards to reals.

I've collected a lot of extension fields references while working on my own implementation: https://github.com/mratsim/constantine/tree/master/constanti...

The best likely being

- Arithmetic of Finite Fields

Regarding your first question, you don't need to attack the hard-drive, for non constant-time crypto you can read power consumption or electromagnetic traces when the secret key is used to reconstruct it: - https://github.com/mratsim/constantine/wiki/Constant-time-arithmetics

Unfortunate name collision with my constant-time pairing-based cryptography library :/.

https://github.com/mratsim/constantine

KZG for sure yes, I actually already started implementing them: https://github.com/mratsim/constantine/tree/c2d716b/research/kzg_poly_commit

I have a small write-up on various details of elliptic curve crypto implementation here: https://github.com/mratsim/constantine/tree/master/constantine/elliptic

I noticed in July of 2022 that Go did exactly the vulnerable example and reported it to the security team.

https://github.com/golang/go/issues/53849

It was fixed as of Go 1.21 https://go.dev/doc/go1.21

---

The article cites JavaScript, which is not constant time. There's no sure way to do constant time operations in JavaScript and thus no secure way to do crypto directly in Javascript. Browsers like Firefox depend on low level calls which should be implemented in languages that are constant time capable.

JavaScript needs something like constant time WASM in order to do crypto securely, but seeing the only constant time WASM project on GitHub has only 16 stars and the last commit was 2 years ago, it doesn't appear to have much interest. https://github.com/WebAssembly/constant-time

However, for JavaScript, I recommend Paul's library Noble which is "hardened to be algorithmically constant time". It is by far the best library available for JavaScript. https://github.com/paulmillr/noble-secp256k1

Also see the constant time spec: https://github.com/WebAssembly/constant-time

I'm waiting for constant time WASM. https://github.com/WebAssembly/constant-time/blob/main/propo...

Also: We're excited for constant time Wasm.