Feature/bn254 neg trait #1630
Feature/bn254 neg trait #1630
Conversation
anupsdf
requested review from
jayz22 and
sisuresh
and removed request for
jayz22
soroban-sdk/src/crypto/bn254.rs
Outdated
|
|
||
| impl_bytesn_repr!(G1Affine, G1_SERIALIZED_SIZE); | ||
| impl_bytesn_repr!(G2Affine, G2_SERIALIZED_SIZE); | ||
| impl_bytesn_repr!(Fq, FP_SERIALIZED_SIZE); |
There was a problem hiding this comment.
If Fq and FP the same things or do we need a size var for Fq?
There was a problem hiding this comment.
It's just a naming convention. Technically, in bn254 they are called Fq and not Fp. I forgot to change FP_SERIALIZED_SIZE in FQ_SERIALIZED_SIZE
There was a problem hiding this comment.
I'm not too familiar with outside conventions, but I think we should use Fp just for consistency. (honestly I found the switching really confusing and never understood why)
Anyways we haven't introduced Fq anywhere (in the CAP or in the env), and the Fp here is just an internal type to facilitate negation. I don't feel it's worth to switch the naming (to match some outside convention).
soroban-sdk/src/crypto/bn254.rs
Outdated
| #[cfg(not(target_family = "wasm"))] | ||
| use crate::xdr::ScVal; | ||
| use crate::{ | ||
| crypto::bls12_381::BigInt, |
There was a problem hiding this comment.
We should move this to a shared util file. I made the change here that you can cherry-pick, or if you'd like, I can pick up this PR. Please let me know.
4 participants
Implement
Negtrait forG1Affine. This is useful for verifying Groth16 proofs.What
Fqnewtype wrappingBytesN<FP_SERIALIZED_SIZE>as the BN254 base field elementNegforFqNegforG1Affineby:Why
Known limitations