Closed sragss closed 1 month ago
Looks like 50% of the largest layer's Eq evals is unsafe_alloc_zero_vec
so probably not worth doing. Maybe worthwhile for Spartan's sumchecks.
after discussion with Justin we decided that this is not actually possible
During each layer of grand products we compute a progressively larger multilinear lagrange basis polynomial
EqPolynomial_0(r_0), EqPolynomial_1(r_0, r_1), ... EqPolynomial_n(r_0, r_1, ... r_n)
.The polynomial is computed fresh every time but can take as input the previous layer's EqPolynomial.
EqPolynomial_n(EqPolynomial_{n-1})
to accelerate by O(n/2).