biggroup_edgecase_handling.hpp

Go to the documentation of this file.

// === AUDIT STATUS ===
// internal:    { status: Complete, auditors: [Suyash], commit: 553c5eb82901955c638b943065acd3e47fc918c0}
// external_1:  { status: not started, auditors: [], commit: }
// external_2:  { status: not started, auditors: [], commit: }
// =====================
 
#pragma once
#include "barretenberg/common/assert.hpp"
#include "barretenberg/ecc/groups/precomputed_generators_bn254_impl.hpp"
#include "barretenberg/ecc/groups/precomputed_generators_secp256k1_impl.hpp"
#include "barretenberg/ecc/groups/precomputed_generators_secp256r1_impl.hpp"
#include "barretenberg/stdlib/primitives/biggroup/biggroup.hpp"
 
namespace bb::stdlib::element_default {
 
template <typename C, class Fq, class Fr, class G>
typename G::affine_element element<C, Fq, Fr, G>::compute_table_offset_generator()
{
    constexpr typename G::affine_element offset_generator =
        get_precomputed_generators<G, "biggroup table offset generator", 1>()[0];
 
    return offset_generator;
}
 
template <typename C, class Fq, class Fr, class G>
std::pair<std::vector<element<C, Fq, Fr, G>>, std::vector<Fr>> element<C, Fq, Fr, G>::mask_points(
    const std::vector<element>& _points, const std::vector<Fr>& _scalars, const Fr& masking_scalar)
{
    std::vector<element> points;
    std::vector<Fr> scalars;
    BB_ASSERT_EQ(_points.size(), _scalars.size());
 
    BB_ASSERT_LTE(uint256_t(masking_scalar.get_value()).get_msb() + 1ULL,
                  128ULL,
                  "biggroup mask_points: masking_scalar must ≤ 128 bits");
 
    // Get the offset generator G_offset in native and in-circuit form
    const typename G::affine_element native_offset_generator = element::compute_table_offset_generator();
    C* builder = validate_context<C>(validate_context<C>(_points), validate_context<C>(_scalars));
    const element offset_generator_element = element::from_witness(builder, native_offset_generator);
    auto empty_tag = OriginTag::constant(); // Disable origin checking during intermediate operations
    offset_generator_element.set_origin_tag(empty_tag);
 
    // Compute initial point to be added: (δ)⋅G_offset
    // TODO(https://github.com/AztecProtocol/barretenberg/issues/1585): do we really need to multiply by δ here?
    element running_point = offset_generator_element.scalar_mul(masking_scalar, 128);
 
    // Start the running scalar at 1
    Fr running_scalar = Fr(1);
    Fr last_scalar = Fr(0);
 
    // For each point and scalar
    for (size_t i = 0; i < _points.size(); i++) {
        scalars.push_back(_scalars[i]);
 
        // Convert point into point + (2ⁱ)⋅(δG_offset)
        points.push_back(_points[i].add_internal(running_point));
 
        // Add 2ⁱ⋅scalar_i to the last scalar
        last_scalar += _scalars[i] * running_scalar;
 
        // Double the running scalar and point for next iteration
        running_scalar += running_scalar;
        running_point = running_point.dbl_internal();
    }
 
    // Add a scalar -(<δ(1, 2, 2²,...,2ⁿ⁻¹),(scalar₀,...,scalarₙ₋₁)> / 2ⁿ)
    const uint32_t n = static_cast<uint32_t>(_points.size());
    const Fr two_power_n = Fr(2).pow(n);
    const Fr two_power_n_inverse = two_power_n.invert();
    last_scalar *= two_power_n_inverse;
    scalars.push_back(-last_scalar);
    if constexpr (Fr::is_composite) {
        scalars.back().self_reduce();
    }
    // Add in-circuit 2ⁿ.(δ.G_offset) to points
    points.push_back(running_point);
 
    return { points, scalars };
}
 
template <typename C, class Fq, class Fr, class G>
std::pair<std::vector<element<C, Fq, Fr, G>>, std::vector<Fr>> element<C, Fq, Fr, G>::handle_points_at_infinity(
    const std::vector<element>& _points, const std::vector<Fr>& _scalars)
{
    C* builder = validate_context<C>(validate_context<C>(_points), validate_context<C>(_scalars));
    std::vector<element> points;
    std::vector<Fr> scalars;
    element one = element::one(builder);
 
    for (auto [_point, _scalar] : zip_view(_points, _scalars)) {
        bool_ct is_point_at_infinity = _point.is_point_at_infinity();
        if (is_point_at_infinity.get_value() && static_cast<bool>(is_point_at_infinity.is_constant())) {
            // if point is at infinity and a circuit constant we can just skip.
            continue;
        }
        if (_scalar.get_value() == 0 && _scalar.is_constant()) {
            // if scalar multiplier is 0 and also a constant, we can skip
            continue;
        }
 
        // Select either the point at infinity or the fixed generator
        element point = _point.conditional_select(one, is_point_at_infinity);
 
        Fr scalar;
        if constexpr (!Fr::is_composite) {
            // For field_t (non-composite), use internal version to avoid premature normalization
            scalar = Fr::conditional_assign_internal(is_point_at_infinity, 0, _scalar);
        } else {
            // For bigfield (composite), conditional_assign doesn't normalize anyway
            scalar = Fr::conditional_assign(is_point_at_infinity, 0, _scalar);
        }
 
        // Push the selected point and scalar to their respective vectors
        points.push_back(point);
        scalars.push_back(scalar);
    }
 
    return { points, scalars };
}
} // namespace bb::stdlib::element_default