You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We currently do a lot of type switching for the tests and the hint. We should probably make it a bit more generic so that we could support all possible curves for which we have the field emulation.
It would be excellent if we could even support curves over field emulation parameters which are not included in std/math/emulated/emparams, but this would require knowing the NbLimbs and NbBitsPerLimb inside hint, but the hints cannot be parametric as the solver is not able to find it then.
Maybe we could have a very generic *big.Int based scalar multiplication where we only define curve coefficients and modulus for using inside hint. It's cost would probably be insignificant compared to rest of the solver.
We currently do a lot of type switching for the tests and the hint. We should probably make it a bit more generic so that we could support all possible curves for which we have the field emulation.
It would be excellent if we could even support curves over field emulation parameters which are not included in
std/math/emulated/emparams
, but this would require knowing the NbLimbs and NbBitsPerLimb inside hint, but the hints cannot be parametric as the solver is not able to find it then.Maybe we could have a very generic
*big.Int
based scalar multiplication where we only define curve coefficients and modulus for using inside hint. It's cost would probably be insignificant compared to rest of the solver.cc @yelhousni
The text was updated successfully, but these errors were encountered: