Jack Lloyd

Add EC_Scalar and EC_AffinePoint types

5

This is the other side of #3979 where we push the interface that's exposed to the rest of the library to a non-BigInt oriented model. #4027

Bridge pcurves into the main elliptic curve arithmetic

3

This results in roughly a 2x speedup for all elliptic curve algorithms, when using a supported curve.

Somewhat regular failures of gcc sanitizer tls_proxy cli test

8

Nothing but > ERROR: Unexpected return code from tls_proxy 1 https://github.com/randombit/botan/actions/runs/9439667258/job/25999341641?pr=4042 I've seen this several times, always on this specific build (not for instance clang) Probably better logging from the...

Replace BigInt based elliptic curve library

10

### Botan 3.5.0 In this release pcurves is really just used for hash to curve * [x] Initial pcurves (point arithmetic, fixed curve params) - that's #3979 * [x] Deprecate...

Add faster reduction for secp256k1 pcurves

5

This takes advantage of the modulus being near a power of 2. Rough idea of improvements. Seems to vary a lot between compilers and CPUs: * Sandybridge/GCC: ECDH and ECDSA...

We aren't necessarily going to promise this is stable to end users, at least for all algorithms, but such a test would have caught #4110

I'm not sure how this happened but there is just a single secp521r1 ECDSA test in `ecdsa_prob.vec`. No Brainpool. Etc

We're currently using a window of 64 but BoringSSL in https://github.com/google/boringssl/commit/f94f3ed3965ea033001fb9ae006084eee408b861 increased their limit to 256 and we need to have a window at least as big in order to...